Chris@10: Sat Nov 24 22:37:54 EST 2012 stevenj@fftw.org Chris@10: * fixed deadlock bug caused by bogosity flag getting out of synch between processes; thanks to Michael Pippig for the bug report Chris@10: Chris@10: M ./kernel/ifftw.h +1 Chris@10: M ./kernel/planner.c -3 +6 Chris@10: M ./mpi/api.c +12 Chris@10: Chris@10: Wed Nov 21 18:34:29 EST 2012 athena@fftw.org Chris@10: * Updated NEWS Chris@10: Chris@10: M ./NEWS -2 +7 Chris@10: Chris@10: Wed Nov 21 18:33:15 EST 2012 athena@fftw.org Chris@10: * use 2x2 AVX transposition instead of individual stores. Chris@10: Chris@10: This seems to improve single-precision AVX on Sandy Bridge machines. Chris@10: Chris@10: Chris@10: M ./simd-support/simd-avx.h -2 +14 Chris@10: Chris@10: Tue Nov 20 12:18:00 EST 2012 stevenj@fftw.org Chris@10: * revert part of Taylor patch to acx_mpi.m4: do not link -lmpi if mpicc works without libraries, as -lmpi may be some completely different MPI implementation Chris@10: Chris@10: M ./m4/acx_mpi.m4 -3 +3 Chris@10: Chris@10: Tue Nov 20 11:44:57 EST 2012 stevenj@fftw.org Chris@10: * fix deadlock bug (thanks to Michael Pippig for the bug report and patch, and to Graham Dennis for the bug report) in which some processes called MPI_Alltoall and some called MPI_Alltoallv Chris@10: Chris@10: M ./mpi/transpose-alltoall.c -3 +2 Chris@10: Chris@10: Mon Oct 29 15:20:01 EDT 2012 athena@fftw.org Chris@10: * fix texinfo quirk Chris@10: Chris@10: M ./doc/tutorial.texi -2 +2 Chris@10: Chris@10: Mon Oct 29 09:16:43 EDT 2012 athena@fftw.org Chris@10: * clarify that padding only applies to in-place transforms Chris@10: Chris@10: M ./doc/tutorial.texi -5 +10 Chris@10: Chris@10: Sun Oct 28 18:42:48 EDT 2012 athena@fftw.org Chris@10: * make the index-computation logic less paranoid Chris@10: Chris@10: The problem is that for each K and for each expression of the form P[I Chris@10: + STRIDE * K] in a loop, most compilers will try to lift an induction Chris@10: variable PK := &P[I + STRIDE * K]. In large codelets we have many Chris@10: such values of K. For example, a codelet of size 32 with 4 input Chris@10: pointers will generate O(128) induction variables, which will likely Chris@10: overflow the register set, which is likely worse than doing the index Chris@10: computation in the first place. Chris@10: Chris@10: In the past we (wisely and correctly) assumed that compilers will do Chris@10: the wrong thing, and consequently we disabled the induction-variable Chris@10: "optimization" altogether by setting STRIDE ^= ZERO, where ZERO is a Chris@10: value guaranteed to be 0. Since the compiler does not know that Chris@10: ZERO=0, it cannot perform its "optimization" and it is forced to Chris@10: behave sensibly. Chris@10: Chris@10: With this patch, FFTW is a little bit less paranoid. FFTW now Chris@10: disables the induction-variable optimization" only when we estimate Chris@10: that the codelet uses more than ESTIMATED_AVAILABLE_INDEX_REGISTERS Chris@10: induction variables. Chris@10: Chris@10: Currently we set ESTIMATED_AVAILABLE_INDEX_REGISTERS=16. 16 registers ought Chris@10: to be enough for anybody (or so the amd64 and ARM ISA's seem to imply). Chris@10: Chris@10: Chris@10: M ./genfft/gen_hc2c.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft_c.ml -1 +1 Chris@10: M ./genfft/gen_hc2hc.ml -1 +1 Chris@10: M ./genfft/gen_notw.ml -2 +2 Chris@10: M ./genfft/gen_notw_c.ml -2 +2 Chris@10: M ./genfft/gen_r2cb.ml -3 +3 Chris@10: M ./genfft/gen_r2cf.ml -3 +3 Chris@10: M ./genfft/gen_r2r.ml -2 +2 Chris@10: M ./genfft/gen_twiddle.ml -1 +1 Chris@10: M ./genfft/gen_twiddle_c.ml -1 +1 Chris@10: M ./genfft/gen_twidsq.ml -2 +2 Chris@10: M ./genfft/gen_twidsq_c.ml -2 +2 Chris@10: M ./genfft/genutil.ml -1 +2 Chris@10: M ./kernel/ifftw.h -3 +20 Chris@10: Chris@10: Sun Oct 28 18:33:24 EDT 2012 athena@fftw.org Chris@10: * silence warnings Chris@10: Chris@10: M ./kernel/buffered.c +1 Chris@10: M ./rdft/rank0.c +1 Chris@10: Chris@10: Sat Oct 27 09:58:49 EDT 2012 athena@fftw.org Chris@10: * bump version to 3.3.3 Chris@10: Chris@10: M ./NEWS +7 Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Oct 27 09:55:15 EDT 2012 athena@fftw.org Chris@10: * evaluate plans for >1ms when using gettimeofday() Chris@10: Chris@10: The previous limit 10ms was too paranoid, and it made life difficult Chris@10: on machines without an "official" cycle counter, such as ARM. Chris@10: Chris@10: M ./kernel/timer.c -1 +1 Chris@10: Chris@10: Sat Oct 27 09:46:04 EDT 2012 athena@fftw.org Chris@10: * use 4-way NEON SIMD instead of 2-way Chris@10: Chris@10: Kai-Uwe Bloem tried to warn me a year ago that 128-bit NEON was better Chris@10: than 64-bit NEON even on machines with a 64-bit pipe, but I foolishly Chris@10: did not listen. Now that 128-bit NEON pipes are starting to appear on Chris@10: the market it is definitely time to switch. Chris@10: Chris@10: Chris@10: M ./simd-support/simd-neon.h -55 +100 Chris@10: Chris@10: Wed Sep 26 14:21:12 EDT 2012 athena@fftw.org Chris@10: * Note that fftw-3.3 includes MPI support Chris@10: Chris@10: M ./doc/intro.texi -5 +4 Chris@10: Chris@10: Wed Jul 18 11:25:40 EDT 2012 athena@fftw.org Chris@10: * remove obsolete unused function Chris@10: Chris@10: M ./dft/bluestein.c -14 Chris@10: Chris@10: Fri Jun 29 15:57:14 EDT 2012 stevenj@fftw.org Chris@10: * whoops, call omp_get_max_threads; thanks to Hanno Rein for the bug report Chris@10: Chris@10: M ./doc/threads.texi -1 +1 Chris@10: Chris@10: Sat Apr 28 10:55:09 EDT 2012 athena@fftw.org Chris@10: * Fix libfftw3/libfftw3_threads chicken-egg problem Chris@10: Chris@10: On most systems we want to build libfftw3 first, so that Chris@10: libfftw3_threads can depend upon libfftw3. When producing a single Chris@10: combined-thread library (e.g. on Windows) we want the opposite, Chris@10: so that libfftw3 can include libfftw3_threads. Chris@10: Chris@10: Chris@10: M ./Makefile.am -1 +18 Chris@10: M ./threads/Makefile.am +5 Chris@10: Chris@10: Sat Apr 28 10:11:28 EDT 2012 athena@fftw.org Chris@10: * updated NEWS for 3.3.2 Chris@10: Chris@10: M ./NEWS +15 Chris@10: Chris@10: Thu Apr 26 19:36:11 EDT 2012 athena@fftw.org Chris@10: * change revision to 3.3.2 Chris@10: Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Thu Apr 26 19:31:02 EDT 2012 athena@fftw.org Chris@10: * Remove old aligned_main() hack. Chris@10: Chris@10: On i386, in our benchmark program we used to manually aligned the Chris@10: stack to 16-byte boundary via asm trickery. This was a good idea in Chris@10: 1999 (and it was actually necessary to make things work) but the hack Chris@10: is now obsolete and it seems to break gcc-4.7. So the hack is now Chris@10: gone. Chris@10: Chris@10: Chris@10: M ./libbench2/Makefile.am -6 +6 Chris@10: R ./libbench2/aligned-main.c Chris@10: M ./libbench2/main.c -1 +1 Chris@10: Chris@10: Thu Mar 29 16:26:16 EDT 2012 athena@fftw.org Chris@10: * Bugfix: a couple of uninitialized values in the benchmark program Chris@10: Chris@10: M ./libbench2/problem.c -1 +1 Chris@10: Chris@10: Tue Mar 20 19:03:47 EDT 2012 athena@fftw.org Chris@10: * make libfftw{threads,mpi} depend upon libfftw for libtool purposes Chris@10: Chris@10: Thanks Julian Taylor for the patch Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./m4/acx_mpi.m4 -3 +3 Chris@10: M ./mpi/Makefile.am +1 Chris@10: M ./threads/Makefile.am +1 Chris@10: Chris@10: Tue Mar 6 04:44:00 EST 2012 stevenj@fftw.org Chris@10: * formatting tweak Chris@10: Chris@10: M ./doc/reference.texi -1 +1 Chris@10: Chris@10: Mon Mar 5 21:05:27 EST 2012 athena@fftw.org Chris@10: * destroying => overwriting Chris@10: Chris@10: M ./doc/tutorial.texi -1 +1 Chris@10: Chris@10: Fri Mar 2 10:31:20 EST 2012 stevenj@fftw.org Chris@10: * note that WISDOM_ONLY is a documented flag Chris@10: Chris@10: M ./api/fftw3.h -1 +1 Chris@10: Chris@10: Fri Mar 2 10:27:08 EST 2012 stevenj@fftw.org Chris@10: * check for icc pretending to be gcc before including quad-precision decls; thanks to Michael Anselmi for the bug report Chris@10: Chris@10: M ./api/fftw3.h +1 Chris@10: Chris@10: Fri Mar 2 10:23:19 EST 2012 stevenj@fftw.org Chris@10: * foo_CFLAGS needs to manually include AM_CFLAGS; thanks to Henry Gomersall for the Windows bug report Chris@10: Chris@10: M ./threads/Makefile.am -2 +2 Chris@10: Chris@10: Sat Feb 25 15:21:39 EST 2012 athena@fftw.org Chris@10: * update for latest mingw Chris@10: Chris@10: M ./support/BUILD-MINGW32.sh -3 +3 Chris@10: M ./support/BUILD-MINGW64.sh -2 +7 Chris@10: Chris@10: Sat Feb 25 15:21:17 EST 2012 athena@fftw.org Chris@10: tagged fftw-3.3.1 Chris@10: Chris@10: Chris@10: Mon Feb 20 23:06:13 EST 2012 stevenj@fftw.org Chris@10: * added Fortran NEWS Chris@10: Chris@10: M ./NEWS +6 Chris@10: Chris@10: Mon Feb 20 23:00:13 EST 2012 stevenj@fftw.org Chris@10: * move non-portable extended/quad precision F03 interfaces into separate .f03 files (while keeping double/single in fftw3.f03 for minimal ABI breakage) Chris@10: Chris@10: M ./api/Makefile.am -3 +10 Chris@10: M ./api/f03api.sh -2 +4 Chris@10: M ./api/genf03.pl -5 +5 Chris@10: M ./doc/modern-fortran.texi -2 +39 Chris@10: M ./doc/mpi.texi -2 +2 Chris@10: M ./mpi/Makefile.am -5 +11 Chris@10: M ./mpi/f03api.sh -2 +4 Chris@10: Chris@10: Mon Feb 20 11:21:57 EST 2012 athena@fftw.org Chris@10: * rm mpi/fftw3-mpi.f03 at make clean time; thanks Tyler Luchko for the bug report. Chris@10: Chris@10: M ./mpi/Makefile.am +1 Chris@10: Chris@10: Mon Feb 20 11:18:24 EST 2012 athena@fftw.org Chris@10: * Disable a Visual Studion warning that was obnoxious enough for Sebastian Schuberth to send us a patch. Chris@10: Chris@10: M ./simd-support/simd-sse2.h +14 Chris@10: Chris@10: Mon Feb 20 11:18:06 EST 2012 athena@fftw.org Chris@10: * Change version to 3.3.1 Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Mon Feb 20 11:03:15 EST 2012 athena@fftw.org Chris@10: * Integrated Visual Studio AVX patches by Carsten Steger Chris@10: Chris@10: M ./NEWS +3 Chris@10: M ./simd-support/amd64-cpuid.h -10 +40 Chris@10: M ./simd-support/simd-avx.h +6 Chris@10: M ./simd-support/simd-sse2.h +8 Chris@10: M ./simd-support/x86-cpuid.h +7 Chris@10: Chris@10: Wed Nov 9 10:13:32 EST 2011 stevenj@fftw.org Chris@10: * typo Chris@10: Chris@10: M ./doc/modern-fortran.texi -2 +2 Chris@10: Chris@10: Tue Nov 8 22:45:09 EST 2011 stevenj@fftw.org Chris@10: * add missing F77 set_timelimit function; thanks to Martin Diehl for the bug repory Chris@10: Chris@10: M ./api/f77funcs.h +5 Chris@10: Chris@10: Sun Sep 25 10:54:56 EDT 2011 athena@fftw.org Chris@10: * note requirement of /machine:x64 in windows x64 README Chris@10: Chris@10: M ./support/BUILD-MINGW64.sh +7 Chris@10: Chris@10: Sun Sep 18 09:28:20 EDT 2011 athena@fftw.org Chris@10: * AVX detection for MSVC Chris@10: Chris@10: M ./simd-support/amd64-cpuid.h -2 +10 Chris@10: Chris@10: Tue Sep 13 14:58:29 EDT 2011 athena@fftw.org Chris@10: * compile with C89 Chris@10: Chris@10: M ./dft/bluestein.c -1 +1 Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Sat Sep 3 16:25:50 EDT 2011 athena@fftw.org Chris@10: * use the same search pruning heuristics for threaded plans as for nonthreaded plans Chris@10: Chris@10: M ./threads/ct.c -1 +3 Chris@10: M ./threads/hc2hc.c -1 +3 Chris@10: Chris@10: Sat Sep 3 16:12:11 EDT 2011 athena@fftw.org Chris@10: * shorten ESTIMATE planning time for certain weird sizes Chris@10: Chris@10: FFTW includes a collection of "solvers" that apply to a subset of Chris@10: "problems". Assume for simplicity that a "problem" is a single 1D Chris@10: complex transform of size N, even though real "problems" are much more Chris@10: general than that. FFTW includes three "prime" solvers called Chris@10: "generic", "bluestein", and "rader", which implement different Chris@10: algorithms for prime sizes. Chris@10: Chris@10: Now, for a "problem" of size 13 (say) FFTW also includes special code Chris@10: that handles that size at high speed. It would be a waste of time to Chris@10: measure the execution time of the prime solvers, since we know that Chris@10: the special code is way faster. However, FFTW is modular and one may Chris@10: or may not include the special code for size 13, in which case we must Chris@10: resort to one of the "prime" solvers. To address this issue, the Chris@10: "prime" solvers (and others) are proclaimed to be SLOW". When Chris@10: planning, FFTW first tries to produce a plan ignoring all the SLOW Chris@10: solvers, and if this fails FFTW tries again allowing SLOW solvers. Chris@10: Chris@10: This heuristic works ok unless the sizes are too large. For example Chris@10: for 1044000=2*2*2*2*2*3*3*5*5*5*29 FFTW explores a huge search tree of Chris@10: all zillion factorizations of 1044000/29, failing every time because Chris@10: 29 is SLOW; then it finally allows SLOW solvers and finds a solution Chris@10: immediately. Chris@10: Chris@10: This patch proclaims solvers to be SLOW only for small values of N. Chris@10: For example, the "generic" solver implements an O(n^2) DFT algorithm; Chris@10: we say that it is SLOW only for N<=16. Chris@10: Chris@10: The side effects of this choice are as follows. If one modifies FFTW to Chris@10: include a fast solver of size 17, then planning for N=17*K will be Chris@10: slower than today, because FFTW till try both the fast solver and the Chris@10: generic solver (which is SLOW today and therefore not tried, but is no Chris@10: longer SLOW after the patch). If one removes a fast solver, of size say Chris@10: 13, then he may still fall into the current exponential-search behavior Chris@10: for "problems" of size 13*HIGHLY_FACTORIZABLE_N. Chris@10: Chris@10: If somebody had compleined about transforms of size 1044000 ten years Chris@10: ago, "don't do that" would have been an acceptable answer. I guess the Chris@10: bar is higher today, so I am going to include this patch in our 3.3.1 Chris@10: release despite their side-effects for people who want to modify FFTW. Chris@10: Chris@10: Chris@10: M ./dft/bluestein.c -5 +14 Chris@10: M ./dft/generic.c -15 +6 Chris@10: M ./dft/rader.c -7 +7 Chris@10: M ./kernel/ifftw.h +8 Chris@10: M ./kernel/primes.c +6 Chris@10: M ./rdft/dht-rader.c -7 +8 Chris@10: M ./rdft/generic.c -14 +4 Chris@10: Chris@10: Sat Aug 27 13:55:24 EDT 2011 athena@fftw.org Chris@10: * Fix typo fftw_execute_dft_r2r => fftw_execute_r2r Chris@10: Chris@10: Thanks KIU Shueng Chuan for the bug report. Chris@10: Chris@10: M ./doc/reference.texi -1 +1 Chris@10: Chris@10: Fri Aug 26 06:13:55 EDT 2011 athena@fftw.org Chris@10: * In Rader's algorithm, compute the generator lazily. Chris@10: Chris@10: The planner was spending a lot of time computing generators for Chris@10: plans that were immediately discarded. Now we compute generators Chris@10: only when absolutely needed. Chris@10: Chris@10: M ./NEWS +5 Chris@10: M ./dft/rader.c -3 +4 Chris@10: M ./rdft/dht-rader.c -3 +4 Chris@10: Chris@10: Sun Aug 21 16:27:31 EDT 2011 athena@fftw.org Chris@10: tagged fftw-3.3.1-beta1 Chris@10: Chris@10: Chris@10: Sun Aug 21 16:16:38 EDT 2011 athena@fftw.org Chris@10: * Release notes for 3.3.1-beta1 Chris@10: Chris@10: M ./NEWS +4 Chris@10: M ./configure.ac -1 +1 Chris@10: M ./doc/install.texi -8 +18 Chris@10: M ./doc/other.texi -2 +3 Chris@10: Chris@10: Fri Aug 19 19:59:17 EDT 2011 stevenj@alum.mit.edu Chris@10: * make fftw_mpi_block routine 10x faster, since it is being called zillions of times (thanks to Tom Vacek for the profiling) Chris@10: Chris@10: M ./mpi/block.c -6 +2 Chris@10: Chris@10: Thu Aug 18 14:19:36 EDT 2011 athena@fftw.org Chris@10: * Implement autodetection of NEON extensions Chris@10: Chris@10: M ./simd-support/neon.c -4 +51 Chris@10: Chris@10: Sun Aug 14 14:12:29 EDT 2011 athena@fftw.org Chris@10: * Update the FSF address. Chris@10: Chris@10: The FSF moved downtown. Chris@10: Chris@10: M ./COPYRIGHT -1 +1 Chris@10: M ./api/api.h -1 +1 Chris@10: M ./api/apiplan.c -1 +1 Chris@10: M ./api/configure.c -1 +1 Chris@10: M ./api/execute-dft-c2r.c -1 +1 Chris@10: M ./api/execute-dft-r2c.c -1 +1 Chris@10: M ./api/execute-dft.c -1 +1 Chris@10: M ./api/execute-r2r.c -1 +1 Chris@10: M ./api/execute-split-dft-c2r.c -1 +1 Chris@10: M ./api/execute-split-dft-r2c.c -1 +1 Chris@10: M ./api/execute-split-dft.c -1 +1 Chris@10: M ./api/execute.c -1 +1 Chris@10: M ./api/export-wisdom-to-file.c -1 +1 Chris@10: M ./api/export-wisdom-to-string.c -1 +1 Chris@10: M ./api/export-wisdom.c -1 +1 Chris@10: M ./api/f77api.c -1 +1 Chris@10: M ./api/f77funcs.h -1 +1 Chris@10: M ./api/flops.c -1 +1 Chris@10: M ./api/forget-wisdom.c -1 +1 Chris@10: M ./api/import-system-wisdom.c -1 +1 Chris@10: M ./api/import-wisdom-from-file.c -1 +1 Chris@10: M ./api/import-wisdom-from-string.c -1 +1 Chris@10: M ./api/import-wisdom.c -1 +1 Chris@10: M ./api/malloc.c -1 +1 Chris@10: M ./api/map-r2r-kind.c -1 +1 Chris@10: M ./api/mapflags.c -1 +1 Chris@10: M ./api/mkprinter-file.c -1 +1 Chris@10: M ./api/mktensor-iodims.h -1 +1 Chris@10: M ./api/mktensor-rowmajor.c -1 +1 Chris@10: M ./api/plan-dft-1d.c -1 +1 Chris@10: M ./api/plan-dft-2d.c -1 +1 Chris@10: M ./api/plan-dft-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-1d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-2d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r.c -1 +1 Chris@10: M ./api/plan-dft-r2c-1d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-2d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-3d.c -1 +1 Chris@10: M ./api/plan-dft-r2c.c -1 +1 Chris@10: M ./api/plan-dft.c -1 +1 Chris@10: M ./api/plan-guru-dft-c2r.h -1 +1 Chris@10: M ./api/plan-guru-dft-r2c.h -1 +1 Chris@10: M ./api/plan-guru-dft.h -1 +1 Chris@10: M ./api/plan-guru-r2r.h -1 +1 Chris@10: M ./api/plan-guru-split-dft-c2r.h -1 +1 Chris@10: M ./api/plan-guru-split-dft-r2c.h -1 +1 Chris@10: M ./api/plan-guru-split-dft.h -1 +1 Chris@10: M ./api/plan-many-dft-c2r.c -1 +1 Chris@10: M ./api/plan-many-dft-r2c.c -1 +1 Chris@10: M ./api/plan-many-dft.c -1 +1 Chris@10: M ./api/plan-many-r2r.c -1 +1 Chris@10: M ./api/plan-r2r-1d.c -1 +1 Chris@10: M ./api/plan-r2r-2d.c -1 +1 Chris@10: M ./api/plan-r2r-3d.c -1 +1 Chris@10: M ./api/plan-r2r.c -1 +1 Chris@10: M ./api/print-plan.c -1 +1 Chris@10: M ./api/rdft2-pad.c -1 +1 Chris@10: M ./api/the-planner.c -1 +1 Chris@10: M ./api/version.c -1 +1 Chris@10: M ./api/x77.h -1 +1 Chris@10: M ./commercialize.sh -2 +2 Chris@10: M ./dft/bluestein.c -1 +1 Chris@10: M ./dft/buffered.c -1 +1 Chris@10: M ./dft/codelet-dft.h -1 +1 Chris@10: M ./dft/conf.c -1 +1 Chris@10: M ./dft/ct.c -1 +1 Chris@10: M ./dft/ct.h -1 +1 Chris@10: M ./dft/dft.h -1 +1 Chris@10: M ./dft/dftw-direct.c -1 +1 Chris@10: M ./dft/dftw-directsq.c -1 +1 Chris@10: M ./dft/dftw-generic.c -1 +1 Chris@10: M ./dft/dftw-genericbuf.c -1 +1 Chris@10: M ./dft/direct.c -1 +1 Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./dft/indirect-transpose.c -1 +1 Chris@10: M ./dft/indirect.c -1 +1 Chris@10: M ./dft/kdft-dif.c -1 +1 Chris@10: M ./dft/kdft-difsq.c -1 +1 Chris@10: M ./dft/kdft-dit.c -1 +1 Chris@10: M ./dft/kdft.c -1 +1 Chris@10: M ./dft/nop.c -1 +1 Chris@10: M ./dft/plan.c -1 +1 Chris@10: M ./dft/problem.c -1 +1 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./dft/rank-geq2.c -1 +1 Chris@10: M ./dft/scalar/n.c -1 +1 Chris@10: M ./dft/scalar/n.h -1 +1 Chris@10: M ./dft/scalar/t.c -1 +1 Chris@10: M ./dft/scalar/t.h -1 +1 Chris@10: M ./dft/simd/common/genus.c -1 +1 Chris@10: M ./dft/simd/n1b.h -1 +1 Chris@10: M ./dft/simd/n1f.h -1 +1 Chris@10: M ./dft/simd/n2b.h -1 +1 Chris@10: M ./dft/simd/n2f.h -1 +1 Chris@10: M ./dft/simd/n2s.h -1 +1 Chris@10: M ./dft/simd/q1b.h -1 +1 Chris@10: M ./dft/simd/q1f.h -1 +1 Chris@10: M ./dft/simd/t1b.h -1 +1 Chris@10: M ./dft/simd/t1bu.h -1 +1 Chris@10: M ./dft/simd/t1f.h -1 +1 Chris@10: M ./dft/simd/t1fu.h -1 +1 Chris@10: M ./dft/simd/t2b.h -1 +1 Chris@10: M ./dft/simd/t2f.h -1 +1 Chris@10: M ./dft/simd/t3b.h -1 +1 Chris@10: M ./dft/simd/t3f.h -1 +1 Chris@10: M ./dft/simd/ts.h -1 +1 Chris@10: M ./dft/solve.c -1 +1 Chris@10: M ./dft/vrank-geq1.c -1 +1 Chris@10: M ./dft/zero.c -1 +1 Chris@10: M ./doc/f77_wisdom.f -1 +1 Chris@10: M ./doc/license.texi -3 +3 Chris@10: M ./genfft/algsimp.ml -1 +1 Chris@10: M ./genfft/algsimp.mli -1 +1 Chris@10: M ./genfft/annotate.ml -1 +1 Chris@10: M ./genfft/annotate.mli -1 +1 Chris@10: M ./genfft/assoctable.ml -1 +1 Chris@10: M ./genfft/assoctable.mli -1 +1 Chris@10: M ./genfft/c.ml -1 +1 Chris@10: M ./genfft/c.mli -1 +1 Chris@10: M ./genfft/complex.ml -1 +1 Chris@10: M ./genfft/complex.mli -1 +1 Chris@10: M ./genfft/conv.ml -1 +1 Chris@10: M ./genfft/conv.mli -1 +1 Chris@10: M ./genfft/dag.ml -1 +1 Chris@10: M ./genfft/dag.mli -1 +1 Chris@10: M ./genfft/expr.ml -1 +1 Chris@10: M ./genfft/expr.mli -1 +1 Chris@10: M ./genfft/fft.ml -1 +1 Chris@10: M ./genfft/fft.mli -1 +1 Chris@10: M ./genfft/gen_hc2c.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft_c.ml -1 +1 Chris@10: M ./genfft/gen_hc2hc.ml -1 +1 Chris@10: M ./genfft/gen_mdct.ml -1 +1 Chris@10: M ./genfft/gen_notw.ml -1 +1 Chris@10: M ./genfft/gen_notw_c.ml -1 +1 Chris@10: M ./genfft/gen_r2cb.ml -1 +1 Chris@10: M ./genfft/gen_r2cf.ml -1 +1 Chris@10: M ./genfft/gen_r2r.ml -1 +1 Chris@10: M ./genfft/gen_twiddle.ml -1 +1 Chris@10: M ./genfft/gen_twiddle_c.ml -1 +1 Chris@10: M ./genfft/gen_twidsq.ml -1 +1 Chris@10: M ./genfft/gen_twidsq_c.ml -1 +1 Chris@10: M ./genfft/genutil.ml -1 +1 Chris@10: M ./genfft/littlesimp.ml -1 +1 Chris@10: M ./genfft/littlesimp.mli -1 +1 Chris@10: M ./genfft/magic.ml -1 +1 Chris@10: M ./genfft/monads.ml -1 +1 Chris@10: M ./genfft/number.ml -1 +1 Chris@10: M ./genfft/number.mli -1 +1 Chris@10: M ./genfft/oracle.ml -1 +1 Chris@10: M ./genfft/oracle.mli -1 +1 Chris@10: M ./genfft/schedule.ml -1 +1 Chris@10: M ./genfft/schedule.mli -1 +1 Chris@10: M ./genfft/simd.ml -1 +1 Chris@10: M ./genfft/simd.mli -1 +1 Chris@10: M ./genfft/simdmagic.ml -1 +1 Chris@10: M ./genfft/to_alist.ml -1 +1 Chris@10: M ./genfft/to_alist.mli -1 +1 Chris@10: M ./genfft/trig.ml -1 +1 Chris@10: M ./genfft/trig.mli -1 +1 Chris@10: M ./genfft/twiddle.ml -1 +1 Chris@10: M ./genfft/twiddle.mli -1 +1 Chris@10: M ./genfft/unique.ml -1 +1 Chris@10: M ./genfft/unique.mli -1 +1 Chris@10: M ./genfft/util.ml -1 +1 Chris@10: M ./genfft/util.mli -1 +1 Chris@10: M ./genfft/variable.ml -1 +1 Chris@10: M ./genfft/variable.mli -1 +1 Chris@10: M ./kernel/align.c -1 +1 Chris@10: M ./kernel/alloc.c -1 +1 Chris@10: M ./kernel/assert.c -1 +1 Chris@10: M ./kernel/awake.c -1 +1 Chris@10: M ./kernel/buffered.c -1 +1 Chris@10: M ./kernel/cpy1d.c -1 +1 Chris@10: M ./kernel/cpy2d-pair.c -1 +1 Chris@10: M ./kernel/cpy2d.c -1 +1 Chris@10: M ./kernel/ct.c -1 +1 Chris@10: M ./kernel/debug.c -1 +1 Chris@10: M ./kernel/extract-reim.c -1 +1 Chris@10: M ./kernel/hash.c -1 +1 Chris@10: M ./kernel/iabs.c -1 +1 Chris@10: M ./kernel/ifftw.h -1 +1 Chris@10: M ./kernel/kalloc.c -1 +1 Chris@10: M ./kernel/md5-1.c -1 +1 Chris@10: M ./kernel/md5.c -1 +1 Chris@10: M ./kernel/minmax.c -1 +1 Chris@10: M ./kernel/ops.c -1 +1 Chris@10: M ./kernel/pickdim.c -1 +1 Chris@10: M ./kernel/plan.c -1 +1 Chris@10: M ./kernel/planner.c -1 +1 Chris@10: M ./kernel/primes.c -1 +1 Chris@10: M ./kernel/print.c -1 +1 Chris@10: M ./kernel/problem.c -1 +1 Chris@10: M ./kernel/rader.c -1 +1 Chris@10: M ./kernel/scan.c -1 +1 Chris@10: M ./kernel/solver.c -1 +1 Chris@10: M ./kernel/solvtab.c -1 +1 Chris@10: M ./kernel/stride.c -1 +1 Chris@10: M ./kernel/tensor.c -1 +1 Chris@10: M ./kernel/tensor1.c -1 +1 Chris@10: M ./kernel/tensor2.c -1 +1 Chris@10: M ./kernel/tensor3.c -1 +1 Chris@10: M ./kernel/tensor4.c -1 +1 Chris@10: M ./kernel/tensor5.c -1 +1 Chris@10: M ./kernel/tensor7.c -1 +1 Chris@10: M ./kernel/tensor8.c -1 +1 Chris@10: M ./kernel/tensor9.c -1 +1 Chris@10: M ./kernel/tile2d.c -1 +1 Chris@10: M ./kernel/timer.c -1 +1 Chris@10: M ./kernel/transpose.c -1 +1 Chris@10: M ./kernel/trig.c -1 +1 Chris@10: M ./kernel/twiddle.c -1 +1 Chris@10: M ./libbench2/aligned-main.c -1 +1 Chris@10: M ./libbench2/bench-main.c -1 +1 Chris@10: M ./libbench2/bench-user.h -1 +1 Chris@10: M ./libbench2/bench.h -1 +1 Chris@10: M ./libbench2/can-do.c -1 +1 Chris@10: M ./libbench2/dotens2.c -1 +1 Chris@10: M ./libbench2/info.c -1 +1 Chris@10: M ./libbench2/main.c -1 +1 Chris@10: M ./libbench2/my-getopt.c -1 +1 Chris@10: M ./libbench2/my-getopt.h -1 +1 Chris@10: M ./libbench2/problem.c -1 +1 Chris@10: M ./libbench2/report.c -1 +1 Chris@10: M ./libbench2/speed.c -1 +1 Chris@10: M ./libbench2/tensor.c -1 +1 Chris@10: M ./libbench2/timer.c -1 +1 Chris@10: M ./libbench2/useropt.c -1 +1 Chris@10: M ./libbench2/util.c -1 +1 Chris@10: M ./libbench2/verify-dft.c -1 +1 Chris@10: M ./libbench2/verify-lib.c -1 +1 Chris@10: M ./libbench2/verify-r2r.c -1 +1 Chris@10: M ./libbench2/verify-rdft2.c -1 +1 Chris@10: M ./libbench2/verify.c -1 +1 Chris@10: M ./libbench2/verify.h -1 +1 Chris@10: M ./libbench2/zero.c -1 +1 Chris@10: M ./mpi/any-true.c -1 +1 Chris@10: M ./mpi/api.c -1 +1 Chris@10: M ./mpi/block.c -1 +1 Chris@10: M ./mpi/choose-radix.c -1 +1 Chris@10: M ./mpi/conf.c -1 +1 Chris@10: M ./mpi/dft-problem.c -1 +1 Chris@10: M ./mpi/dft-rank-geq2-transposed.c -1 +1 Chris@10: M ./mpi/dft-rank-geq2.c -1 +1 Chris@10: M ./mpi/dft-rank1-bigvec.c -1 +1 Chris@10: M ./mpi/dft-rank1.c -1 +1 Chris@10: M ./mpi/dft-serial.c -1 +1 Chris@10: M ./mpi/dft-solve.c -1 +1 Chris@10: M ./mpi/dtensor.c -1 +1 Chris@10: M ./mpi/ifftw-mpi.h -1 +1 Chris@10: M ./mpi/mpi-dft.h -1 +1 Chris@10: M ./mpi/mpi-rdft.h -1 +1 Chris@10: M ./mpi/mpi-rdft2.h -1 +1 Chris@10: M ./mpi/mpi-transpose.h -1 +1 Chris@10: M ./mpi/rdft-problem.c -1 +1 Chris@10: M ./mpi/rdft-rank-geq2-transposed.c -1 +1 Chris@10: M ./mpi/rdft-rank-geq2.c -1 +1 Chris@10: M ./mpi/rdft-rank1-bigvec.c -1 +1 Chris@10: M ./mpi/rdft-serial.c -1 +1 Chris@10: M ./mpi/rdft-solve.c -1 +1 Chris@10: M ./mpi/rdft2-problem.c -1 +1 Chris@10: M ./mpi/rdft2-rank-geq2-transposed.c -1 +1 Chris@10: M ./mpi/rdft2-rank-geq2.c -1 +1 Chris@10: M ./mpi/rdft2-serial.c -1 +1 Chris@10: M ./mpi/rdft2-solve.c -1 +1 Chris@10: M ./mpi/rearrange.c -1 +1 Chris@10: M ./mpi/testsched.c -1 +1 Chris@10: M ./mpi/transpose-alltoall.c -1 +1 Chris@10: M ./mpi/transpose-pairwise.c -1 +1 Chris@10: M ./mpi/transpose-problem.c -1 +1 Chris@10: M ./mpi/transpose-recurse.c -1 +1 Chris@10: M ./mpi/transpose-solve.c -1 +1 Chris@10: M ./mpi/wisdom-api.c -1 +1 Chris@10: M ./rdft/buffered.c -1 +1 Chris@10: M ./rdft/buffered2.c -1 +1 Chris@10: M ./rdft/codelet-rdft.h -1 +1 Chris@10: M ./rdft/conf.c -1 +1 Chris@10: M ./rdft/ct-hc2c-direct.c -1 +1 Chris@10: M ./rdft/ct-hc2c.c -1 +1 Chris@10: M ./rdft/ct-hc2c.h -1 +1 Chris@10: M ./rdft/dft-r2hc.c -1 +1 Chris@10: M ./rdft/dht-r2hc.c -1 +1 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/direct-r2c.c -1 +1 Chris@10: M ./rdft/direct-r2r.c -1 +1 Chris@10: M ./rdft/direct2.c -1 +1 Chris@10: M ./rdft/generic.c -1 +1 Chris@10: M ./rdft/hc2hc-direct.c -1 +1 Chris@10: M ./rdft/hc2hc-generic.c -1 +1 Chris@10: M ./rdft/hc2hc.c -1 +1 Chris@10: M ./rdft/hc2hc.h -1 +1 Chris@10: M ./rdft/indirect.c -1 +1 Chris@10: M ./rdft/khc2c.c -1 +1 Chris@10: M ./rdft/khc2hc.c -1 +1 Chris@10: M ./rdft/kr2c.c -1 +1 Chris@10: M ./rdft/kr2r.c -1 +1 Chris@10: M ./rdft/nop.c -1 +1 Chris@10: M ./rdft/nop2.c -1 +1 Chris@10: M ./rdft/plan.c -1 +1 Chris@10: M ./rdft/plan2.c -1 +1 Chris@10: M ./rdft/problem.c -1 +1 Chris@10: M ./rdft/problem2.c -1 +1 Chris@10: M ./rdft/rank-geq2-rdft2.c -1 +1 Chris@10: M ./rdft/rank-geq2.c -1 +1 Chris@10: M ./rdft/rank0-rdft2.c -1 +1 Chris@10: M ./rdft/rank0.c -1 +1 Chris@10: M ./rdft/rdft-dht.c -1 +1 Chris@10: M ./rdft/rdft.h -1 +1 Chris@10: M ./rdft/rdft2-inplace-strides.c -1 +1 Chris@10: M ./rdft/rdft2-rdft.c -1 +1 Chris@10: M ./rdft/rdft2-strides.c -1 +1 Chris@10: M ./rdft/rdft2-tensor-max-index.c -1 +1 Chris@10: M ./rdft/scalar/hb.h -1 +1 Chris@10: M ./rdft/scalar/hc2c.c -1 +1 Chris@10: M ./rdft/scalar/hc2cb.h -1 +1 Chris@10: M ./rdft/scalar/hc2cf.h -1 +1 Chris@10: M ./rdft/scalar/hf.h -1 +1 Chris@10: M ./rdft/scalar/hfb.c -1 +1 Chris@10: M ./rdft/scalar/r2c.c -1 +1 Chris@10: M ./rdft/scalar/r2cb.h -1 +1 Chris@10: M ./rdft/scalar/r2cbIII.h -1 +1 Chris@10: M ./rdft/scalar/r2cf.h -1 +1 Chris@10: M ./rdft/scalar/r2cfII.h -1 +1 Chris@10: M ./rdft/scalar/r2r.c -1 +1 Chris@10: M ./rdft/scalar/r2r.h -1 +1 Chris@10: M ./rdft/simd/common/genus.c -1 +1 Chris@10: M ./rdft/simd/hc2cbv.h -1 +1 Chris@10: M ./rdft/simd/hc2cfv.h -1 +1 Chris@10: M ./rdft/solve.c -1 +1 Chris@10: M ./rdft/solve2.c -1 +1 Chris@10: M ./rdft/vrank-geq1-rdft2.c -1 +1 Chris@10: M ./rdft/vrank-geq1.c -1 +1 Chris@10: M ./rdft/vrank3-transpose.c -1 +1 Chris@10: M ./reodft/conf.c -1 +1 Chris@10: M ./reodft/redft00e-r2hc-pad.c -1 +1 Chris@10: M ./reodft/redft00e-r2hc.c -1 +1 Chris@10: M ./reodft/reodft.h -1 +1 Chris@10: M ./reodft/reodft00e-splitradix.c -1 +1 Chris@10: M ./reodft/reodft010e-r2hc.c -1 +1 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -1 +1 Chris@10: M ./reodft/reodft11e-r2hc.c -1 +1 Chris@10: M ./reodft/reodft11e-radix2.c -1 +1 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -1 +1 Chris@10: M ./reodft/rodft00e-r2hc.c -1 +1 Chris@10: M ./simd-support/altivec.c -1 +1 Chris@10: M ./simd-support/amd64-cpuid.h -1 +1 Chris@10: M ./simd-support/avx.c -1 +1 Chris@10: M ./simd-support/neon.c -1 +1 Chris@10: M ./simd-support/simd-altivec.h -1 +1 Chris@10: M ./simd-support/simd-avx.h -1 +1 Chris@10: M ./simd-support/simd-common.h -1 +1 Chris@10: M ./simd-support/simd-neon.h -1 +1 Chris@10: M ./simd-support/simd-sse2.h -1 +1 Chris@10: M ./simd-support/sse2-nonportable.c -1 +1 Chris@10: M ./simd-support/sse2.c -1 +1 Chris@10: M ./simd-support/taint.c -1 +1 Chris@10: M ./simd-support/x86-cpuid.h -1 +1 Chris@10: M ./simd/altivec.c -1 +1 Chris@10: M ./simd/mips_ps.c -1 +1 Chris@10: M ./simd/mips_ps.h -1 +1 Chris@10: M ./simd/nonportable/sse.c -1 +1 Chris@10: M ./simd/nonportable/sse2.c -1 +1 Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: M ./simd/simd-mips_ps.h -1 +1 Chris@10: M ./simd/simd-sse.h -1 +1 Chris@10: M ./simd/simd-sse2.h -1 +1 Chris@10: M ./simd/simd.h -1 +1 Chris@10: M ./simd/sse.c -1 +1 Chris@10: M ./simd/sse2.c -1 +1 Chris@10: M ./simd/taint.c -1 +1 Chris@10: M ./simd/x86-cpuid.h -1 +1 Chris@10: M ./threads/api.c -1 +1 Chris@10: M ./threads/conf.c -1 +1 Chris@10: M ./threads/ct.c -1 +1 Chris@10: M ./threads/dft-vrank-geq1.c -1 +1 Chris@10: M ./threads/f77api.c -1 +1 Chris@10: M ./threads/f77funcs.h -1 +1 Chris@10: M ./threads/hc2hc.c -1 +1 Chris@10: M ./threads/openmp.c -1 +1 Chris@10: M ./threads/rdft-vrank-geq1.c -1 +1 Chris@10: M ./threads/threads.c -1 +1 Chris@10: M ./threads/threads.h -1 +1 Chris@10: M ./threads/vrank-geq1-rdft2.c -1 +1 Chris@10: M ./tools/fftw-wisdom-to-conf.1 -1 +1 Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: M ./tools/fftw-wisdom.c -1 +1 Chris@10: M ./tools/fftw_wisdom.1.in -1 +1 Chris@10: Chris@10: Thu Aug 11 14:54:38 EDT 2011 stevenj@fftw.org Chris@10: * allow specifying TRANSPOSE_{IN/OUT} transpose plans, since libbench does not canonicalize rnk=1 n=1 plans as rnk=0 Chris@10: Chris@10: M ./mpi/mpi-bench.c -2 +2 Chris@10: Chris@10: Thu Aug 11 14:17:24 EDT 2011 stevenj@fftw.org Chris@10: * check.pl should occasionally check DESTROY_INPUT problems too (especially since those enable slightly different algorithms in MPI) Chris@10: Chris@10: M ./tests/check.pl +1 Chris@10: Chris@10: Thu Aug 11 12:37:51 EDT 2011 stevenj@fftw.org Chris@10: * unify post-MPI transpose handling in pairwise and alltoall solvers; should make the former faster in the destroy-input out-of-place case, and the latter more widely applicable Chris@10: Chris@10: M ./mpi/mpi-transpose.h +5 Chris@10: M ./mpi/transpose-alltoall.c -69 +24 Chris@10: M ./mpi/transpose-pairwise.c -66 +97 Chris@10: Chris@10: Mon Aug 8 10:06:14 EDT 2011 athena@fftw.org Chris@10: * Add support for ARM NEON Chris@10: Chris@10: M ./Makefile.am -1 +6 Chris@10: M ./api/version.c +4 Chris@10: M ./configure.ac -5 +23 Chris@10: M ./dft/codelet-dft.h +1 Chris@10: M ./dft/conf.c +4 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: A ./dft/simd/neon/ Chris@10: A ./dft/simd/neon/Makefile.am Chris@10: M ./kernel/ifftw.h +2 Chris@10: M ./m4/ax_cc_maxopt.m4 +10 Chris@10: M ./rdft/codelet-rdft.h +1 Chris@10: M ./rdft/conf.c +4 Chris@10: M ./rdft/simd/Makefile.am -1 +1 Chris@10: A ./rdft/simd/neon/ Chris@10: A ./rdft/simd/neon/Makefile.am Chris@10: M ./simd-support/Makefile.am -1 +2 Chris@10: A ./simd-support/neon.c Chris@10: M ./simd-support/simd-common.h +3 Chris@10: A ./simd-support/simd-neon.h Chris@10: Chris@10: Fri Aug 5 17:25:32 EDT 2011 stevenj@fftw.org Chris@10: * more C++ paranoia Chris@10: Chris@10: M ./tests/fftw-bench.h +10 Chris@10: Chris@10: Fri Aug 5 17:02:00 EDT 2011 stevenj@fftw.org Chris@10: * tentative version bump for 3.3.1 Chris@10: Chris@10: M ./NEWS +5 Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Fri Aug 5 16:52:28 EDT 2011 stevenj@fftw.org Chris@10: * fixes so that MPI code compiles when MPICC is a C++ compiler, even if the serial code is compiled with a C compiler; thanks to Kyle Spyksma for the bug report Chris@10: Chris@10: M ./api/api.h +12 Chris@10: M ./api/plan-guru-r2r.h -2 Chris@10: M ./api/plan-many-r2r.c -2 Chris@10: M ./dft/dft.h +9 Chris@10: M ./kernel/ifftw.h +9 Chris@10: M ./mpi/api.c -3 +1 Chris@10: M ./mpi/dtensor.c -1 +1 Chris@10: M ./mpi/ifftw-mpi.h -1 +4 Chris@10: M ./mpi/mpi-bench.c -1 +1 Chris@10: M ./rdft/rdft.h +9 Chris@10: Chris@10: Fri Aug 5 16:04:06 EDT 2011 stevenj@fftw.org Chris@10: * use correct precision in f03-wrap.c, avoiding a (harmless) implicit pointer cast that prevented compilation under C++; thanks to Kyle Spyksma for the bug report Chris@10: Chris@10: M ./mpi/genf03-wrap.pl +2 Chris@10: Chris@10: Fri Aug 5 14:04:32 EDT 2011 stevenj@fftw.org Chris@10: * manual typo Chris@10: Chris@10: M ./doc/modern-fortran.texi -1 +1 Chris@10: Chris@10: Wed Jul 27 12:50:25 EDT 2011 athena@fftw.org Chris@10: tagged fftw-3.3 Chris@10: Chris@10: Chris@10: Tue Jul 26 20:55:45 EDT 2011 athena@fftw.org Chris@10: * Honor WITH_OUR_MALLOC in libbench2 Chris@10: Chris@10: M ./libbench2/util.c -6 +5 Chris@10: Chris@10: Tue Jul 26 20:27:28 EDT 2011 athena@fftw.org Chris@10: * fixed typo: incorrect name of combined threads library on Windows Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: Chris@10: Mon Jul 25 14:38:20 EDT 2011 stevenj@fftw.org Chris@10: * 3.3 version bump & NEWS Chris@10: Chris@10: M ./NEWS -5 +16 Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Mon Jul 25 14:37:48 EDT 2011 stevenj@fftw.org Chris@10: * use int(..., C_SIZE_T) rather than declaring another variable in the Fortran examples Chris@10: Chris@10: M ./doc/modern-fortran.texi -12 +9 Chris@10: Chris@10: Wed Jul 13 05:02:32 EDT 2011 stevenj@fftw.org Chris@10: * typo, thanks to Rhys Ulerich for the comment Chris@10: Chris@10: M ./doc/mpi.texi -1 +1 Chris@10: Chris@10: Mon Jul 11 14:39:52 EDT 2011 athena@fftw.org Chris@10: * Fix bug in bubblesort Chris@10: Chris@10: Bubblesort was not sorting. This was a bug in the benchmark library Chris@10: (not in FFTW per se), and it impacted the benchmark program Chris@10: with --report-time and --report-mflops causing it to output Chris@10: an incorrect value for the median. (The minimum, maximum, and Chris@10: average value were correct.) Thanks Dima Baksheev of Intel for Chris@10: reporting this bug. Chris@10: Chris@10: Chris@10: M ./libbench2/report.c -2 +2 Chris@10: Chris@10: Fri Jul 8 13:35:59 EDT 2011 stevenj@fftw.org Chris@10: * small manual typos Chris@10: Chris@10: M ./doc/modern-fortran.texi -2 +2 Chris@10: Chris@10: Wed Jul 6 10:49:40 EDT 2011 athena@fftw.org Chris@10: * Detection of altivec.h requires $ALTIVEC_CFLAGS Chris@10: Chris@10: M ./configure.ac -5 +4 Chris@10: Chris@10: Tue Jul 5 19:58:47 EDT 2011 athena@fftw.org Chris@10: * Introduce fake dependency so that my-getopt.c is recompiled Chris@10: Chris@10: my-getopt.c does not depend on anything, and so it is not rebuilt when Chris@10: reconfiguring for a different ISA (e.g., CC="gcc -m32" vs CC="gcc Chris@10: -m64"). Add a fake dependency on so that the file is Chris@10: recompiled. Chris@10: Chris@10: Chris@10: Chris@10: M ./libbench2/my-getopt.c +1 Chris@10: Chris@10: Tue Jul 5 18:53:36 EDT 2011 stevenj@fftw.org Chris@10: * support compiling/installing --enable-threads --enable-openmp at the same time, although in this case the test program only uses the threads variety. Update documentation accordingly, and in general expand the documentation of the OpenMP support Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac -21 +13 Chris@10: M ./doc/install.texi -14 +13 Chris@10: M ./doc/threads.texi -19 +43 Chris@10: M ./mpi/Makefile.am -2 +6 Chris@10: M ./tests/Makefile.am -2 +6 Chris@10: M ./tests/fftw-bench.c -1 +3 Chris@10: M ./threads/Makefile.am -6 +20 Chris@10: M ./threads/conf.c -4 Chris@10: M ./threads/openmp.c -4 Chris@10: M ./threads/threads.c -3 Chris@10: M ./tools/Makefile.am -4 +8 Chris@10: Chris@10: Tue Jul 5 16:04:03 EDT 2011 stevenj@fftw.org Chris@10: * call omp_set_num_threads in fftw-bench so that the number of OpenMP threads corresponds with the number of FFTW threads Chris@10: Chris@10: M ./tests/fftw-bench.c +5 Chris@10: Chris@10: Tue Jul 5 16:03:06 EDT 2011 stevenj@fftw.org Chris@10: * when --enable-openmp, install as fftw3_omp rather than fftw3_threads, so that both the POSIX threads and OpenMP variants of FFTW can be installed at once Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +4 Chris@10: M ./threads/Makefile.am -4 +4 Chris@10: Chris@10: Sat Jul 2 02:21:22 EDT 2011 stevenj@fftw.org Chris@10: * don't even declare an fftw_execute interface in Fortran, since it is unsafe and we recommend against it anyway; thanks to Arjen Markus for the suggestion Chris@10: Chris@10: M ./api/f03api.sh -1 +1 Chris@10: M ./doc/modern-fortran.texi -2 +4 Chris@10: Chris@10: Fri Jul 1 14:35:44 EDT 2011 athena@fftw.org Chris@10: * consistently use the order single, double, long double Chris@10: Chris@10: M ./support/BUILD-MINGW32.sh -1 +1 Chris@10: M ./support/BUILD-MINGW64.sh -1 +1 Chris@10: Chris@10: Wed Jun 29 17:27:06 EDT 2011 athena@fftw.org Chris@10: * MSVC AVX 64-bit detection does not work, punt for now. Chris@10: Chris@10: M ./simd-support/amd64-cpuid.h -16 +2 Chris@10: Chris@10: Wed Jun 29 15:52:27 EDT 2011 stevenj@fftw.org Chris@10: * fixed typo, added note on transposed flags for r2c/c2r; thanks to Rhys Ulerich for the suggestions Chris@10: Chris@10: M ./doc/mpi.texi -1 +14 Chris@10: Chris@10: Wed Jun 29 09:41:39 EDT 2011 athena@fftw.org Chris@10: * fixes for compiling with MSVC (untested) Chris@10: Chris@10: M ./simd-support/amd64-cpuid.h -2 +16 Chris@10: M ./simd-support/simd-avx.h -11 +16 Chris@10: M ./simd-support/x86-cpuid.h -1 +7 Chris@10: Chris@10: Tue Jun 28 16:48:36 EDT 2011 athena@fftw.org Chris@10: * comment Chris@10: Chris@10: M ./simd-support/sse2-nonportable.c +3 Chris@10: Chris@10: Mon Jun 27 21:01:56 EDT 2011 stevenj@fftw.org Chris@10: * rm extraneous line break in HTML output ... I hate texinfo Chris@10: Chris@10: M ./doc/reference.texi -6 +3 Chris@10: Chris@10: Mon Jun 27 06:43:57 EDT 2011 athena@fftw.org Chris@10: tagged fftw-3.3-beta1 Chris@10: Chris@10: Chris@10: Mon Jun 27 00:47:33 EDT 2011 stevenj@fftw.org Chris@10: * maintainer-clean should delete html directory (otherwwise we keep obsolete HTML files in the dist tarball, sigh) Chris@10: Chris@10: M ./doc/Makefile.am -2 +2 Chris@10: Chris@10: Mon Jun 27 00:45:28 EDT 2011 stevenj@fftw.org Chris@10: tagged fftw-3.3-beta1 Chris@10: Chris@10: Chris@10: Sun Jun 26 23:36:32 EDT 2011 stevenj@fftw.org Chris@10: * update copyright year in manual Chris@10: Chris@10: M ./doc/license.texi -2 +2 Chris@10: Chris@10: Sun Jun 26 22:52:54 EDT 2011 stevenj@fftw.org Chris@10: * whoops, don't dist .f03 headers, since those are built by the user's Makefile Chris@10: Chris@10: M ./api/Makefile.am -1 +2 Chris@10: M ./mpi/Makefile.am -1 +2 Chris@10: Chris@10: Sun Jun 26 22:43:49 EDT 2011 stevenj@fftw.org Chris@10: * fix embarrassing deadlock/crashing bug in my previous nowisdom_hook fix -- I forgot to handle the case where one process has wisdom and another one doesn't, requiring a nowisdom_hook in the latter case; this should only affect MPI transforms since otherwise these hook functions are NULL Chris@10: Chris@10: M ./kernel/ifftw.h +1 Chris@10: M ./kernel/planner.c -44 +48 Chris@10: M ./mpi/api.c +14 Chris@10: Chris@10: Sun Jun 26 21:02:15 EDT 2011 stevenj@fftw.org Chris@10: * subsubheadings, MPI transpose reference Chris@10: Chris@10: M ./doc/mpi.texi -7 +53 Chris@10: Chris@10: Sun Jun 26 20:48:53 EDT 2011 stevenj@fftw.org Chris@10: * add MPI plan reference Chris@10: Chris@10: M ./doc/mpi.texi -5 +197 Chris@10: Chris@10: Sun Jun 26 17:07:21 EDT 2011 stevenj@fftw.org Chris@10: * portions of MPI reference docs; tweaks to NEWS Chris@10: Chris@10: M ./NEWS -2 +3 Chris@10: M ./doc/mpi.texi -12 +264 Chris@10: Chris@10: Sun Jun 26 12:40:43 EDT 2011 stevenj@fftw.org Chris@10: * use $(CHECK_PL_OPTS) more consistently Chris@10: Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Sun Jun 26 10:04:54 EDT 2011 athena@fftw.org Chris@10: * accept \r\n as well as \n. Grrr... Chris@10: Chris@10: M ./tests/check.pl -1 +2 Chris@10: Chris@10: Sun Jun 26 09:52:11 EDT 2011 athena@fftw.org Chris@10: * new configure option --with-incoming-stack-boundary=N Chris@10: Chris@10: This option selects CFLAGS to align the stack at all externally-callable Chris@10: functions. This currently comprises api/* and threads/* Chris@10: Chris@10: Chris@10: M ./api/Makefile.am +1 Chris@10: M ./configure.ac +14 Chris@10: M ./support/BUILD-MINGW32.sh -1 +1 Chris@10: M ./threads/Makefile.am +1 Chris@10: Chris@10: Sun Jun 26 09:51:37 EDT 2011 athena@fftw.org Chris@10: * add -fomit-frame-pointer back Chris@10: Chris@10: Somehow -O3 does not imply -fomit-frame-pointer on ia32 Chris@10: Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -2 +3 Chris@10: Chris@10: Sun Jun 26 07:20:27 EDT 2011 athena@fftw.org Chris@10: * Note that removal of mips-ps is temporary. Chris@10: Chris@10: M ./NEWS -3 +5 Chris@10: Chris@10: Sat Jun 25 23:15:03 EDT 2011 stevenj@fftw.org Chris@10: * update copyright year Chris@10: Chris@10: M ./COPYRIGHT -2 +2 Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/configure.c -2 +2 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute-split-dft-c2r.c -2 +2 Chris@10: M ./api/execute-split-dft-r2c.c -2 +2 Chris@10: M ./api/execute-split-dft.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/export-wisdom-to-file.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -2 +2 Chris@10: M ./api/export-wisdom.c -2 +2 Chris@10: M ./api/f77api.c -2 +2 Chris@10: M ./api/f77funcs.h -2 +2 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/flops.c -2 +2 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: M ./api/import-wisdom-from-file.c -2 +2 Chris@10: M ./api/import-wisdom-from-string.c -2 +2 Chris@10: M ./api/import-wisdom.c -2 +2 Chris@10: M ./api/malloc.c -2 +2 Chris@10: M ./api/map-r2r-kind.c -2 +2 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./api/mkprinter-file.c -2 +2 Chris@10: M ./api/mktensor-iodims.h -2 +2 Chris@10: M ./api/mktensor-rowmajor.c -2 +2 Chris@10: M ./api/plan-dft-1d.c -2 +2 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r.c -2 +2 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -2 +2 Chris@10: M ./api/plan-dft-r2c.c -2 +2 Chris@10: M ./api/plan-dft.c -2 +2 Chris@10: M ./api/plan-guru-dft-c2r.h -2 +2 Chris@10: M ./api/plan-guru-dft-r2c.h -2 +2 Chris@10: M ./api/plan-guru-dft.h -2 +2 Chris@10: M ./api/plan-guru-r2r.h -2 +2 Chris@10: M ./api/plan-guru-split-dft-c2r.h -2 +2 Chris@10: M ./api/plan-guru-split-dft-r2c.h -2 +2 Chris@10: M ./api/plan-guru-split-dft.h -2 +2 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: M ./api/plan-many-dft.c -2 +2 Chris@10: M ./api/plan-many-r2r.c -2 +2 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -2 +2 Chris@10: M ./api/plan-r2r-3d.c -2 +2 Chris@10: M ./api/plan-r2r.c -2 +2 Chris@10: M ./api/print-plan.c -2 +2 Chris@10: M ./api/rdft2-pad.c -2 +2 Chris@10: M ./api/the-planner.c -2 +2 Chris@10: M ./api/version.c -2 +2 Chris@10: M ./api/x77.h -2 +2 Chris@10: M ./dft/bluestein.c -2 +2 Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/codelet-dft.h -2 +2 Chris@10: M ./dft/conf.c -2 +2 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/ct.h -2 +2 Chris@10: M ./dft/dft.h -2 +2 Chris@10: M ./dft/dftw-direct.c -2 +2 Chris@10: M ./dft/dftw-directsq.c -2 +2 Chris@10: M ./dft/dftw-generic.c -2 +2 Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect-transpose.c -2 +2 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/kdft-dif.c -2 +2 Chris@10: M ./dft/kdft-difsq.c -2 +2 Chris@10: M ./dft/kdft-dit.c -2 +2 Chris@10: M ./dft/kdft.c -2 +2 Chris@10: M ./dft/nop.c -2 +2 Chris@10: M ./dft/plan.c -2 +2 Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/scalar/n.c -2 +2 Chris@10: M ./dft/scalar/n.h -2 +2 Chris@10: M ./dft/scalar/t.c -2 +2 Chris@10: M ./dft/scalar/t.h -2 +2 Chris@10: M ./dft/simd/common/genus.c -2 +2 Chris@10: M ./dft/simd/n1b.h -2 +2 Chris@10: M ./dft/simd/n1f.h -2 +2 Chris@10: M ./dft/simd/n2b.h -2 +2 Chris@10: M ./dft/simd/n2f.h -2 +2 Chris@10: M ./dft/simd/n2s.h -2 +2 Chris@10: M ./dft/simd/q1b.h -2 +2 Chris@10: M ./dft/simd/q1f.h -2 +2 Chris@10: M ./dft/simd/t1b.h -2 +2 Chris@10: M ./dft/simd/t1bu.h -2 +2 Chris@10: M ./dft/simd/t1f.h -2 +2 Chris@10: M ./dft/simd/t1fu.h -2 +2 Chris@10: M ./dft/simd/t2b.h -2 +2 Chris@10: M ./dft/simd/t2f.h -2 +2 Chris@10: M ./dft/simd/t3b.h -2 +2 Chris@10: M ./dft/simd/t3f.h -2 +2 Chris@10: M ./dft/simd/ts.h -2 +2 Chris@10: M ./dft/solve.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./dft/zero.c -2 +2 Chris@10: M ./doc/f77_wisdom.f -2 +2 Chris@10: M ./genfft/algsimp.ml -2 +2 Chris@10: M ./genfft/algsimp.mli -2 +2 Chris@10: M ./genfft/annotate.ml -2 +2 Chris@10: M ./genfft/annotate.mli -2 +2 Chris@10: M ./genfft/assoctable.ml -2 +2 Chris@10: M ./genfft/assoctable.mli -2 +2 Chris@10: M ./genfft/c.ml -2 +2 Chris@10: M ./genfft/c.mli -2 +2 Chris@10: M ./genfft/complex.ml -2 +2 Chris@10: M ./genfft/complex.mli -2 +2 Chris@10: M ./genfft/conv.ml -2 +2 Chris@10: M ./genfft/conv.mli -2 +2 Chris@10: M ./genfft/dag.ml -2 +2 Chris@10: M ./genfft/dag.mli -2 +2 Chris@10: M ./genfft/expr.ml -2 +2 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/fft.ml -2 +2 Chris@10: M ./genfft/fft.mli -2 +2 Chris@10: M ./genfft/gen_hc2c.ml -2 +2 Chris@10: M ./genfft/gen_hc2cdft.ml -2 +2 Chris@10: M ./genfft/gen_hc2cdft_c.ml -2 +2 Chris@10: M ./genfft/gen_hc2hc.ml -2 +2 Chris@10: M ./genfft/gen_mdct.ml -2 +2 Chris@10: M ./genfft/gen_notw.ml -2 +2 Chris@10: M ./genfft/gen_notw_c.ml -2 +2 Chris@10: M ./genfft/gen_r2cb.ml -2 +2 Chris@10: M ./genfft/gen_r2cf.ml -2 +2 Chris@10: M ./genfft/gen_r2r.ml -2 +2 Chris@10: M ./genfft/gen_twiddle.ml -2 +2 Chris@10: M ./genfft/gen_twiddle_c.ml -2 +2 Chris@10: M ./genfft/gen_twidsq.ml -2 +2 Chris@10: M ./genfft/gen_twidsq_c.ml -2 +2 Chris@10: M ./genfft/genutil.ml -2 +2 Chris@10: M ./genfft/littlesimp.ml -2 +2 Chris@10: M ./genfft/littlesimp.mli -2 +2 Chris@10: M ./genfft/magic.ml -2 +2 Chris@10: M ./genfft/monads.ml -2 +2 Chris@10: M ./genfft/number.ml -2 +2 Chris@10: M ./genfft/number.mli -2 +2 Chris@10: M ./genfft/oracle.ml -2 +2 Chris@10: M ./genfft/oracle.mli -2 +2 Chris@10: M ./genfft/schedule.ml -2 +2 Chris@10: M ./genfft/schedule.mli -2 +2 Chris@10: M ./genfft/simd.ml -2 +2 Chris@10: M ./genfft/simd.mli -2 +2 Chris@10: M ./genfft/simdmagic.ml -2 +2 Chris@10: M ./genfft/to_alist.ml -2 +2 Chris@10: M ./genfft/to_alist.mli -2 +2 Chris@10: M ./genfft/trig.ml -2 +2 Chris@10: M ./genfft/trig.mli -2 +2 Chris@10: M ./genfft/twiddle.ml -2 +2 Chris@10: M ./genfft/twiddle.mli -2 +2 Chris@10: M ./genfft/unique.ml -2 +2 Chris@10: M ./genfft/unique.mli -2 +2 Chris@10: M ./genfft/util.ml -2 +2 Chris@10: M ./genfft/util.mli -2 +2 Chris@10: M ./genfft/variable.ml -2 +2 Chris@10: M ./genfft/variable.mli -2 +2 Chris@10: M ./kernel/align.c -2 +2 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/assert.c -2 +2 Chris@10: M ./kernel/awake.c -2 +2 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/cpy1d.c -2 +2 Chris@10: M ./kernel/cpy2d-pair.c -2 +2 Chris@10: M ./kernel/cpy2d.c -2 +2 Chris@10: M ./kernel/ct.c -2 +2 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./kernel/debug.c -2 +2 Chris@10: M ./kernel/extract-reim.c -2 +2 Chris@10: M ./kernel/hash.c -2 +2 Chris@10: M ./kernel/iabs.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/kalloc.c -2 +2 Chris@10: M ./kernel/md5-1.c -2 +2 Chris@10: M ./kernel/md5.c -2 +2 Chris@10: M ./kernel/minmax.c -2 +2 Chris@10: M ./kernel/ops.c -2 +2 Chris@10: M ./kernel/pickdim.c -2 +2 Chris@10: M ./kernel/plan.c -2 +2 Chris@10: M ./kernel/primes.c -2 +2 Chris@10: M ./kernel/print.c -2 +2 Chris@10: M ./kernel/problem.c -2 +2 Chris@10: M ./kernel/rader.c -2 +2 Chris@10: M ./kernel/scan.c -2 +2 Chris@10: M ./kernel/solver.c -2 +2 Chris@10: M ./kernel/solvtab.c -2 +2 Chris@10: M ./kernel/stride.c -2 +2 Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: M ./kernel/tensor1.c -2 +2 Chris@10: M ./kernel/tensor2.c -2 +2 Chris@10: M ./kernel/tensor3.c -2 +2 Chris@10: M ./kernel/tensor4.c -2 +2 Chris@10: M ./kernel/tensor5.c -2 +2 Chris@10: M ./kernel/tensor7.c -2 +2 Chris@10: M ./kernel/tensor8.c -2 +2 Chris@10: M ./kernel/tensor9.c -2 +2 Chris@10: M ./kernel/tile2d.c -2 +2 Chris@10: M ./kernel/timer.c -2 +2 Chris@10: M ./kernel/transpose.c -2 +2 Chris@10: M ./kernel/trig.c -2 +2 Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: M ./libbench2/dotens2.c -2 +2 Chris@10: M ./libbench2/my-getopt.c -2 +2 Chris@10: M ./libbench2/my-getopt.h -2 +2 Chris@10: M ./libbench2/verify-dft.c -2 +2 Chris@10: M ./libbench2/verify-lib.c -2 +2 Chris@10: M ./libbench2/verify-r2r.c -2 +2 Chris@10: M ./libbench2/verify-rdft2.c -2 +2 Chris@10: M ./libbench2/verify.h -2 +2 Chris@10: M ./mpi/any-true.c -2 +2 Chris@10: M ./mpi/api.c -2 +2 Chris@10: M ./mpi/block.c -2 +2 Chris@10: M ./mpi/choose-radix.c -2 +2 Chris@10: M ./mpi/conf.c -2 +2 Chris@10: M ./mpi/dft-problem.c -2 +2 Chris@10: M ./mpi/dft-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/dft-rank-geq2.c -2 +2 Chris@10: M ./mpi/dft-rank1-bigvec.c -2 +2 Chris@10: M ./mpi/dft-rank1.c -2 +2 Chris@10: M ./mpi/dft-serial.c -2 +2 Chris@10: M ./mpi/dft-solve.c -2 +2 Chris@10: M ./mpi/dtensor.c -2 +2 Chris@10: M ./mpi/fftw3-mpi.h -2 +2 Chris@10: M ./mpi/ifftw-mpi.h -2 +2 Chris@10: M ./mpi/mpi-dft.h -2 +2 Chris@10: M ./mpi/mpi-rdft.h -2 +2 Chris@10: M ./mpi/mpi-rdft2.h -2 +2 Chris@10: M ./mpi/mpi-transpose.h -2 +2 Chris@10: M ./mpi/rdft-problem.c -2 +2 Chris@10: M ./mpi/rdft-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/rdft-rank-geq2.c -2 +2 Chris@10: M ./mpi/rdft-rank1-bigvec.c -2 +2 Chris@10: M ./mpi/rdft-serial.c -2 +2 Chris@10: M ./mpi/rdft-solve.c -2 +2 Chris@10: M ./mpi/rdft2-problem.c -2 +2 Chris@10: M ./mpi/rdft2-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/rdft2-rank-geq2.c -2 +2 Chris@10: M ./mpi/rdft2-serial.c -2 +2 Chris@10: M ./mpi/rdft2-solve.c -2 +2 Chris@10: M ./mpi/rearrange.c -2 +2 Chris@10: M ./mpi/testsched.c -1 +1 Chris@10: M ./mpi/transpose-alltoall.c -2 +2 Chris@10: M ./mpi/transpose-pairwise.c -2 +2 Chris@10: M ./mpi/transpose-problem.c -2 +2 Chris@10: M ./mpi/transpose-recurse.c -2 +2 Chris@10: M ./mpi/transpose-solve.c -2 +2 Chris@10: M ./mpi/wisdom-api.c -2 +2 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/codelet-rdft.h -2 +2 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/ct-hc2c-direct.c -2 +2 Chris@10: M ./rdft/ct-hc2c.c -2 +2 Chris@10: M ./rdft/ct-hc2c.h -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: M ./rdft/direct-r2c.c -2 +2 Chris@10: M ./rdft/direct-r2r.c -2 +2 Chris@10: M ./rdft/direct2.c -2 +2 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-direct.c -2 +2 Chris@10: M ./rdft/hc2hc-generic.c -2 +2 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/hc2hc.h -2 +2 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: M ./rdft/khc2c.c -2 +2 Chris@10: M ./rdft/khc2hc.c -2 +2 Chris@10: M ./rdft/kr2c.c -2 +2 Chris@10: M ./rdft/kr2r.c -2 +2 Chris@10: M ./rdft/nop.c -2 +2 Chris@10: M ./rdft/nop2.c -2 +2 Chris@10: M ./rdft/plan.c -2 +2 Chris@10: M ./rdft/plan2.c -2 +2 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rank0-rdft2.c -2 +2 Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -2 +2 Chris@10: M ./rdft/rdft2-rdft.c -2 +2 Chris@10: M ./rdft/rdft2-strides.c -2 +2 Chris@10: M ./rdft/rdft2-tensor-max-index.c -2 +2 Chris@10: M ./rdft/scalar/hb.h -2 +2 Chris@10: M ./rdft/scalar/hc2c.c -2 +2 Chris@10: M ./rdft/scalar/hc2cb.h -2 +2 Chris@10: M ./rdft/scalar/hc2cf.h -2 +2 Chris@10: M ./rdft/scalar/hf.h -2 +2 Chris@10: M ./rdft/scalar/hfb.c -2 +2 Chris@10: M ./rdft/scalar/r2c.c -2 +2 Chris@10: M ./rdft/scalar/r2cb.h -2 +2 Chris@10: M ./rdft/scalar/r2cbIII.h -2 +2 Chris@10: M ./rdft/scalar/r2cf.h -2 +2 Chris@10: M ./rdft/scalar/r2cfII.h -2 +2 Chris@10: M ./rdft/scalar/r2r.c -2 +2 Chris@10: M ./rdft/scalar/r2r.h -2 +2 Chris@10: M ./rdft/simd/common/genus.c -2 +2 Chris@10: M ./rdft/simd/hc2cbv.h -2 +2 Chris@10: M ./rdft/simd/hc2cfv.h -2 +2 Chris@10: M ./rdft/solve.c -2 +2 Chris@10: M ./rdft/solve2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: M ./reodft/conf.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft.h -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./simd/altivec.c -2 +2 Chris@10: M ./simd/nonportable/sse.c -2 +2 Chris@10: M ./simd/nonportable/sse2.c -2 +2 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: M ./simd/simd-sse.h -2 +2 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: M ./simd/simd.h -2 +2 Chris@10: M ./simd/sse.c -2 +2 Chris@10: M ./simd/sse2.c -2 +2 Chris@10: M ./simd/taint.c -2 +2 Chris@10: M ./simd/x86-cpuid.h -2 +2 Chris@10: M ./simd-support/altivec.c -2 +2 Chris@10: M ./simd-support/amd64-cpuid.h -2 +2 Chris@10: M ./simd-support/avx.c -2 +2 Chris@10: M ./simd-support/simd-altivec.h -2 +2 Chris@10: M ./simd-support/simd-avx.h -2 +2 Chris@10: M ./simd-support/simd-common.h -2 +2 Chris@10: M ./simd-support/simd-sse2.h -2 +2 Chris@10: M ./simd-support/sse2-nonportable.c -2 +2 Chris@10: M ./simd-support/sse2.c -2 +2 Chris@10: M ./simd-support/taint.c -2 +2 Chris@10: M ./simd-support/x86-cpuid.h -2 +2 Chris@10: M ./threads/api.c -2 +2 Chris@10: M ./threads/conf.c -2 +2 Chris@10: M ./threads/ct.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/f77api.c -2 +2 Chris@10: M ./threads/f77funcs.h -2 +2 Chris@10: M ./threads/hc2hc.c -2 +2 Chris@10: M ./threads/openmp.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/threads.c -2 +2 Chris@10: M ./threads/threads.h -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./tools/fftw-wisdom-to-conf.1 -4 +4 Chris@10: M ./tools/fftw-wisdom-to-conf.in -2 +2 Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: M ./tools/fftw_wisdom.1.in -4 +4 Chris@10: Chris@10: Sat Jun 25 21:33:13 EDT 2011 stevenj@fftw.org Chris@10: * updated NEWS Chris@10: Chris@10: M ./NEWS -4 +53 Chris@10: Chris@10: Sat Jun 25 20:29:55 EDT 2011 stevenj@fftw.org Chris@10: * fixes to Fortran interface and docs Chris@10: Chris@10: M ./api/genf03.pl -12 +22 Chris@10: M ./doc/modern-fortran.texi -5 +14 Chris@10: M ./doc/mpi.texi -13 +16 Chris@10: Chris@10: Sat Jun 25 17:43:31 EDT 2011 stevenj@fftw.org Chris@10: * initial stab at MPI Fortran docs Chris@10: Chris@10: M ./doc/mpi.texi +138 Chris@10: Chris@10: Sat Jun 25 16:43:31 EDT 2011 stevenj@fftw.org Chris@10: * correct description of what MPI standard says about I/O (I can't believe this crap) Chris@10: Chris@10: M ./doc/mpi.texi -11 +21 Chris@10: Chris@10: Sat Jun 25 15:14:07 EDT 2011 stevenj@fftw.org Chris@10: * more MPI documentation; mention `fftw_alloc' functions earlier in the manual Chris@10: Chris@10: M ./doc/fftw3.texi +12 Chris@10: M ./doc/mpi.texi -30 +158 Chris@10: M ./doc/other.texi +9 Chris@10: M ./doc/tutorial.texi -11 +14 Chris@10: Chris@10: Sat Jun 25 13:40:19 EDT 2011 stevenj@fftw.org Chris@10: * clarification about --enable-sse2 Chris@10: Chris@10: M ./doc/install.texi -6 +8 Chris@10: Chris@10: Sat Jun 25 13:31:25 EDT 2011 athena@fftw.org Chris@10: * Update mingw build scripts for fftw-3.3 Chris@10: Chris@10: M ./kernel/kalloc.c -2 +2 Chris@10: M ./support/BUILD-MINGW32.sh -3 +3 Chris@10: M ./support/BUILD-MINGW64.sh -4 +4 Chris@10: Chris@10: Sat Jun 25 08:52:13 EDT 2011 athena@fftw.org Chris@10: * Fix typo: EXTRADIST => EXTRA_DIST Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: Chris@10: Fri Jun 24 23:52:19 EDT 2011 stevenj@fftw.org Chris@10: * finished draft "modern fortran" chapter Chris@10: Chris@10: M ./doc/modern-fortran.texi -9 +91 Chris@10: Chris@10: Fri Jun 24 20:47:49 EDT 2011 stevenj@fftw.org Chris@10: * include FFTW_EXTERN prototypes for wrappers, so that they are properly exported to DLLs on Windows (sigh) Chris@10: Chris@10: M ./mpi/f03-wrap.sh +4 Chris@10: Chris@10: Fri Jun 24 16:52:30 EDT 2011 athena@fftw.org Chris@10: * use malloc() instead of alloca() for large buffers Chris@10: Chris@10: The proximate cause for this patch is that OpenBSD/i386 reserves 256KB Chris@10: stack size per thread. We were allocating a buffer of size Chris@10: 128*130*sizeof(fftw_complex) that exceeds the stack. Chris@10: Chris@10: While 128*130*sizeof(fftw_complex) = 260KiB is the worst case for Chris@10: normal configurations, it is a good idea to limit stack allocation Chris@10: just in case. Also, the generic solver might in principle generate Chris@10: unbounded buffers, even though it is normally disabled for n > 137. Chris@10: Chris@10: So, as an added precaution, we now never stack-allocate buffers larger Chris@10: than 64KiB, which ought to be enough for anybody. Chris@10: Chris@10: Chris@10: M ./dft/dftw-direct.c -2 +3 Chris@10: M ./dft/direct.c -2 +3 Chris@10: M ./dft/generic.c -2 +3 Chris@10: M ./kernel/ifftw.h -7 +31 Chris@10: M ./rdft/ct-hc2c-direct.c -2 +3 Chris@10: M ./rdft/direct-r2c.c -2 +3 Chris@10: M ./rdft/generic.c -4 +6 Chris@10: M ./rdft/hc2hc-direct.c -2 +3 Chris@10: Chris@10: Fri Jun 24 16:32:30 EDT 2011 stevenj@fftw.org Chris@10: * don't imply that AVX is available on Pentium III; note that MIPS Paired Single is currently only in FFTW 3.2.x Chris@10: Chris@10: M ./doc/install.texi -3 +2 Chris@10: M ./doc/other.texi -3 +4 Chris@10: Chris@10: Fri Jun 24 16:05:27 EDT 2011 stevenj@fftw.org Chris@10: * silence annoying gfortran warnings Chris@10: Chris@10: M ./api/genf03.pl -1 +6 Chris@10: M ./configure.ac +14 Chris@10: M ./mpi/Makefile.am -3 +6 Chris@10: Chris@10: Fri Jun 24 14:59:30 EDT 2011 stevenj@fftw.org Chris@10: * a couple MPI Fortran 2003 fixes; changed MPI flags to not use 1<<31 since Fortran (not having unsigned integers) does not allow us to declare that constant in a portable way Chris@10: Chris@10: M ./mpi/api.c -1 +1 Chris@10: M ./mpi/f03api.sh -2 +2 Chris@10: M ./mpi/fftw3-mpi.h -4 +4 Chris@10: Chris@10: Fri Jun 24 15:05:05 EDT 2011 athena@fftw.org Chris@10: * Fix libtool shared version info. Chris@10: Chris@10: FFTW-3.3.x should be a direct drop-in replacement for all FFTW-3.x.y Chris@10: versions. Chris@10: Chris@10: Chris@10: M ./configure.ac -1 +4 Chris@10: Chris@10: Fri Jun 24 14:38:47 EDT 2011 stevenj@fftw.org Chris@10: * add MPI Fortran API and wrappers Chris@10: Chris@10: M ./api/Makefile.am -2 +3 Chris@10: M ./api/f03api.sh -3 +5 Chris@10: M ./api/genf03.pl -2 +14 Chris@10: M ./mpi/Makefile.am -3 +14 Chris@10: R ./mpi/README Chris@10: A ./mpi/f03-wrap.sh Chris@10: A ./mpi/f03api.sh Chris@10: M ./mpi/fftw3-mpi.h -4 +4 Chris@10: A ./mpi/genf03-wrap.pl Chris@10: Chris@10: Fri Jun 24 14:51:12 EDT 2011 athena@fftw.org Chris@10: * Do not require fig2dev on the user's machine Chris@10: Chris@10: Distribute the manual's figures in PDF/PS/PNG form instead. Chris@10: Chris@10: M ./doc/Makefile.am -10 +16 Chris@10: Chris@10: Fri Jun 24 11:52:44 EDT 2011 athena@fftw.org Chris@10: * Remove --enable-portable-binary, --with-gcc-arch from documentation. Chris@10: Chris@10: M ./NEWS +3 Chris@10: M ./doc/install.texi -23 +2 Chris@10: Chris@10: Fri Jun 24 11:48:48 EDT 2011 athena@fftw.org Chris@10: * Forget about specifying nonportable CFLAGS. Let the user do it if he wants. Chris@10: Chris@10: M ./m4/Makefile.am -4 +3 Chris@10: M ./m4/ax_cc_maxopt.m4 -24 +10 Chris@10: R ./m4/ax_gcc_archflag.m4 Chris@10: R ./m4/ax_gcc_x86_cpuid.m4 Chris@10: Chris@10: Fri Jun 24 11:48:25 EDT 2011 athena@fftw.org Chris@10: * Add "-avx" to version string when appropriate. Chris@10: Chris@10: M ./api/version.c +4 Chris@10: Chris@10: Fri Jun 24 10:26:38 EDT 2011 athena@fftw.org Chris@10: * change 3.3-alpha => 3.3-beta1 Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Fri Jun 24 09:25:49 EDT 2011 athena@fftw.org Chris@10: * Extend OUR_MALLOC16 to larger alignments Chris@10: Chris@10: Make it work for 32-byte alignment and beyond, as needed by AVX. Chris@10: Rename --with-our-malloc16 to --with-our-malloc. Keep old --with-our-malloc16 Chris@10: flag for compatibility. Chris@10: Chris@10: Chris@10: M ./configure.ac -3 +4 Chris@10: M ./kernel/kalloc.c -9 +10 Chris@10: Chris@10: Fri Jun 24 09:19:38 EDT 2011 athena@fftw.org Chris@10: * Fix typo Chris@10: Chris@10: M ./doc/install.texi -1 +1 Chris@10: Chris@10: Fri Jun 24 09:10:26 EDT 2011 athena@fftw.org Chris@10: * One pass over the manual. Chris@10: Chris@10: M ./NEWS -1 +6 Chris@10: M ./doc/Makefile.am -5 +2 Chris@10: M ./doc/install.texi -16 +10 Chris@10: M ./doc/intro.texi -1 +1 Chris@10: M ./doc/other.texi -2 +2 Chris@10: M ./doc/reference.texi -5 +5 Chris@10: M ./doc/tutorial.texi -32 +32 Chris@10: Chris@10: Fri Jun 24 08:19:03 EDT 2011 athena@fftw.org Chris@10: * eliminate the WITH_ALIGNED_STACK hack Chris@10: Chris@10: This is 2011 and I have no system with incorrect stack alignment. Chris@10: Chris@10: Chris@10: M ./TODO -2 Chris@10: M ./api/apiplan.c -9 +4 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute-split-dft-c2r.c -2 +2 Chris@10: M ./api/execute-split-dft-r2c.c -2 +2 Chris@10: M ./api/execute-split-dft.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/f77funcs.h -16 +16 Chris@10: M ./doc/Makefile.am -2 +2 Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: M ./doc/intro.texi -3 +4 Chris@10: M ./doc/legacy-fortran.texi -1 +1 Chris@10: M ./doc/other.texi -51 +4 Chris@10: M ./doc/reference.texi -1 +1 Chris@10: M ./doc/tutorial.texi -9 +5 Chris@10: M ./kernel/ifftw.h -51 Chris@10: M ./threads/ct.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/hc2hc.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: Chris@10: Fri Jun 24 07:49:47 EDT 2011 athena@fftw.org Chris@10: * enable both threaded and unthreaded wisdom in tools/fftw-wisdom Chris@10: Chris@10: M ./TODO -7 +2 Chris@10: M ./tools/fftw-wisdom.c -18 +14 Chris@10: Chris@10: Fri Jun 24 02:40:04 EDT 2011 stevenj@fftw.org Chris@10: * clarification Chris@10: Chris@10: M ./doc/modern-fortran.texi -2 +2 Chris@10: Chris@10: Fri Jun 24 02:24:01 EDT 2011 stevenj@fftw.org Chris@10: * check for error code in example Chris@10: Chris@10: M ./doc/modern-fortran.texi +2 Chris@10: Chris@10: Fri Jun 24 02:22:18 EDT 2011 stevenj@fftw.org Chris@10: * cleanup - since NATIVE_MALLOC is always malloc, delete this #define Chris@10: Chris@10: M ./api/export-wisdom-to-string.c -1 +1 Chris@10: M ./kernel/ifftw.h -2 Chris@10: Chris@10: Fri Jun 24 02:19:44 EDT 2011 stevenj@fftw.org Chris@10: * document wisdom string import/export in Fortran Chris@10: Chris@10: M ./doc/modern-fortran.texi +49 Chris@10: Chris@10: Fri Jun 24 02:11:40 EDT 2011 stevenj@fftw.org Chris@10: * bug fix - NATIVE_MALLOC should always be plain malloc, even in debug_malloc mode, because it is used in the API to return things that should be deallocated with free(); correspondingly, be sure to use free() ansd not X(free) with this Chris@10: Chris@10: M ./kernel/ifftw.h -1 +1 Chris@10: M ./mpi/wisdom-api.c -1 +1 Chris@10: Chris@10: Fri Jun 24 01:35:27 EDT 2011 stevenj@fftw.org Chris@10: * declaration style Chris@10: Chris@10: M ./doc/modern-fortran.texi -2 +4 Chris@10: Chris@10: Fri Jun 24 01:25:36 EDT 2011 stevenj@fftw.org Chris@10: * document wisdom file export/import from Fortran; add export/import_to/from_filename functions for convenience Chris@10: Chris@10: M ./api/export-wisdom-to-file.c +11 Chris@10: M ./api/fftw3.h +2 Chris@10: M ./api/import-wisdom-from-file.c +10 Chris@10: M ./doc/modern-fortran.texi -9 +79 Chris@10: M ./doc/other.texi -5 +6 Chris@10: M ./doc/reference.texi -11 +22 Chris@10: Chris@10: Thu Jun 23 19:19:43 EDT 2011 stevenj@fftw.org Chris@10: * more fortran docs Chris@10: Chris@10: M ./doc/legacy-fortran.texi -5 +6 Chris@10: M ./doc/modern-fortran.texi -10 +167 Chris@10: M ./doc/reference.texi +2 Chris@10: Chris@10: Thu Jun 23 17:50:30 EDT 2011 stevenj@fftw.org Chris@10: * enforce 132-character line-length limit that is the default in Fortran Chris@10: Chris@10: M ./api/genf03.pl -3 +26 Chris@10: Chris@10: Wed Jun 22 23:27:31 EDT 2011 stevenj@fftw.org Chris@10: * the F03 standard is ambiguous about whether types can be assigned to wider types as formal parameters with VALUE attributes, and e.g. gfortran interprets it to disallow this code Chris@10: Chris@10: M ./doc/modern-fortran.texi -2 +4 Chris@10: Chris@10: Thu Jun 23 18:12:10 EDT 2011 athena@fftw.org Chris@10: * Add md5 hash of fftw's configuration to wisdom file Chris@10: Chris@10: People were already confused by threaded vs unthreaded wisdom, and now Chris@10: things will be even worse because we enable/disable AVX codelets at Chris@10: runtime. Accept incoming wisdom only if it was produced by the same Chris@10: configuration (modulo MD5). Chris@10: Chris@10: M ./kernel/planner.c -2 +33 Chris@10: Chris@10: Thu Jun 23 09:01:27 EDT 2011 athena@fftw.org Chris@10: * distribute fftw3.f03.in Chris@10: Chris@10: M ./api/Makefile.am -1 +1 Chris@10: Chris@10: Wed Jun 22 22:02:18 EDT 2011 stevenj@fftw.org Chris@10: * more Fortran documentation Chris@10: Chris@10: M ./doc/modern-fortran.texi -1 +95 Chris@10: Chris@10: Wed Jun 22 20:10:39 EDT 2011 stevenj@fftw.org Chris@10: * correct comment Chris@10: Chris@10: M ./api/genf03.pl -1 +2 Chris@10: Chris@10: Wed Jun 22 20:26:18 EDT 2011 athena@fftw.org Chris@10: * Use "sh FOO.sh" instead of "./FOO.sh" to avoid chmod +x. Chris@10: Chris@10: M ./api/Makefile.am -1 +1 Chris@10: Chris@10: Wed Jun 22 19:19:05 EDT 2011 stevenj@fftw.org Chris@10: * document fftw_alloc_real/complex ... should we switch to using these in the tutorial examples? Chris@10: Chris@10: M ./doc/reference.texi +19 Chris@10: Chris@10: Wed Jun 22 19:07:49 EDT 2011 stevenj@fftw.org Chris@10: * whoops, added missing file Chris@10: Chris@10: A ./api/genf03.pl Chris@10: Chris@10: Wed Jun 22 18:46:01 EDT 2011 athena@fftw.org Chris@10: * Note addition of AVX. Chris@10: Chris@10: M ./NEWS +2 Chris@10: Chris@10: Wed Jun 22 18:43:43 EDT 2011 athena@fftw.org Chris@10: * In SSE2, AVX: use FMA macros when applicable. Chris@10: Chris@10: Makes it easier to play with fma4 and fma3 when it comes out. Chris@10: Chris@10: M ./simd-support/simd-avx.h -21 +21 Chris@10: M ./simd-support/simd-sse2.h -19 +19 Chris@10: Chris@10: Wed Jun 22 18:16:45 EDT 2011 stevenj@fftw.org Chris@10: * all modern Fortran compilers can call FFTW's C interfface directly -- support this, and in particular generate a Fortran 2003 interface file from fftw3.h so that Fortran code calling FFTW can be typechecked ((addressing the source of a lot of Fortran-user problems) Chris@10: Chris@10: ./doc/fortran.texi -> ./doc/legacy-fortran.texi Chris@10: M ./api/Makefile.am -3 +12 Chris@10: A ./api/f03api.sh Chris@10: M ./api/fftw3.h -3 +11 Chris@10: M ./api/malloc.c +18 Chris@10: M ./configure.ac +12 Chris@10: M ./doc/Makefile.am -1 +1 Chris@10: M ./doc/fftw3.texi -2 +4 Chris@10: M ./doc/install.texi -2 +2 Chris@10: M ./doc/intro.texi -5 +6 Chris@10: M ./doc/legacy-fortran.texi -23 +38 Chris@10: A ./doc/modern-fortran.texi Chris@10: M ./doc/mpi.texi -1 +1 Chris@10: M ./doc/other.texi -2 +4 Chris@10: M ./doc/upgrading.texi -1 +1 Chris@10: Chris@10: Wed Jun 22 13:10:02 EDT 2011 athena@fftw.org Chris@10: * some cleanup of SSE2 macros Chris@10: Chris@10: M ./simd-support/simd-sse2.h -27 +16 Chris@10: Chris@10: Wed Jun 22 07:38:18 EDT 2011 athena@fftw.org Chris@10: * don't use -xHost on ICC Chris@10: Chris@10: -xHost with ICC is problematic. On icc-12.0.0, "-mavx -xHost" Chris@10: overrides -mavx with -xHost, generating SSE2 code instead of AVX code. Chris@10: ICC does not seem to support -mtune=host or equivalent non-ABI Chris@10: changing flag. Chris@10: Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -31 +5 Chris@10: Chris@10: Tue Jun 21 20:35:36 EDT 2011 athena@fftw.org Chris@10: * Complete AVX implementation for split codelets Chris@10: Chris@10: M ./configure.ac -36 +38 Chris@10: M ./simd-support/simd-avx.h -41 +26 Chris@10: Chris@10: Tue Jun 21 19:37:14 EDT 2011 stevenj@fftw.org Chris@10: * whoops, missing altivec conf patches Chris@10: Chris@10: M ./dft/codelet-dft.h +1 Chris@10: M ./dft/conf.c +4 Chris@10: M ./rdft/codelet-rdft.h +1 Chris@10: M ./rdft/conf.c +4 Chris@10: Chris@10: Tue Jun 21 19:12:45 EDT 2011 stevenj@fftw.org Chris@10: * some BSD ar versions (e.g. on MacOS X) give an error if there are no object files, so we cannot build empty libraries Chris@10: Chris@10: M ./dft/simd/altivec/Makefile.am -4 +3 Chris@10: M ./dft/simd/avx/Makefile.am -3 +4 Chris@10: M ./dft/simd/sse2/Makefile.am -4 +3 Chris@10: M ./rdft/simd/altivec/Makefile.am -4 +3 Chris@10: M ./rdft/simd/avx/Makefile.am -3 +4 Chris@10: M ./rdft/simd/sse2/Makefile.am -3 +4 Chris@10: Chris@10: Tue Jun 21 19:12:12 EDT 2011 stevenj@fftw.org Chris@10: * re-insertion of Altivec code Chris@10: Chris@10: M ./Makefile.am -1 +6 Chris@10: M ./configure.ac -8 +13 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: A ./dft/simd/altivec/ Chris@10: A ./dft/simd/altivec/Makefile.am Chris@10: M ./kernel/ifftw.h -2 +9 Chris@10: M ./rdft/simd/Makefile.am -1 +1 Chris@10: A ./rdft/simd/altivec/ Chris@10: A ./rdft/simd/altivec/Makefile.am Chris@10: M ./simd-support/Makefile.am -1 +1 Chris@10: A ./simd-support/altivec.c Chris@10: A ./simd-support/simd-altivec.h Chris@10: M ./simd-support/simd-common.h +3 Chris@10: Chris@10: Tue Jun 21 16:26:09 EDT 2011 athena@fftw.org Chris@10: * Implement faster AVX loads/stores. Chris@10: Chris@10: M ./simd-support/simd-avx.h -21 +28 Chris@10: Chris@10: Tue Jun 21 16:03:24 EDT 2011 athena@fftw.org Chris@10: * Initial AVX256/single implementation Chris@10: Chris@10: This should be correct but slow. I need to figure out how to implement Chris@10: noncontiguous loads/stores efficiently. Chris@10: Chris@10: M ./simd-support/simd-avx.h -53 +133 Chris@10: Chris@10: Tue Jun 21 14:13:57 EDT 2011 athena@fftw.org Chris@10: * fix AVX alignment Chris@10: Chris@10: M ./simd-support/simd-common.h -6 +8 Chris@10: Chris@10: Tue Jun 21 14:07:28 EDT 2011 athena@fftw.org Chris@10: * rename avx256d -> avx Chris@10: Chris@10: AVX will work in both double and single precision, like SSE2. Chris@10: Chris@10: ./dft/simd/avx256d -> ./dft/simd/avx Chris@10: ./rdft/simd/avx256d -> ./rdft/simd/avx Chris@10: ./simd-support/avx256d.c -> ./simd-support/avx.c Chris@10: ./simd-support/simd-avx256d.h -> ./simd-support/simd-avx.h Chris@10: M ./Makefile.am -4 +4 Chris@10: M ./configure.ac -11 +6 Chris@10: M ./dft/codelet-dft.h -1 +1 Chris@10: M ./dft/conf.c -3 +3 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./dft/simd/avx/Makefile.am -4 +4 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./rdft/codelet-rdft.h -1 +1 Chris@10: M ./rdft/conf.c -3 +3 Chris@10: M ./rdft/simd/Makefile.am -1 +1 Chris@10: M ./rdft/simd/avx/Makefile.am -4 +4 Chris@10: M ./simd-support/Makefile.am -1 +1 Chris@10: M ./simd-support/avx.c -3 +3 Chris@10: M ./simd-support/simd-avx.h -2 +2 Chris@10: Chris@10: Tue Jun 21 13:52:20 EDT 2011 athena@fftw.org Chris@10: * remove CODELET_OPTIM Chris@10: Chris@10: In the old 32-bit gcc-3.x days we used to play games with gcc to force Chris@10: it to produce decent code. Now gcc has gotten smarter and it produces Chris@10: indecent code no matter what we do, so it is safe to remove these hacks. Chris@10: Chris@10: Chris@10: M ./api/version.c -5 +5 Chris@10: M ./configure.ac -40 Chris@10: M ./support/Makefile.codelets -6 Chris@10: Chris@10: Tue Jun 21 09:57:31 EDT 2011 athena@fftw.org Chris@10: * work around gcc/icc quirks Chris@10: Chris@10: M ./simd-support/simd-avx256d.h -14 +45 Chris@10: Chris@10: Tue Jun 21 09:56:07 EDT 2011 athena@fftw.org Chris@10: * Add remarks in places where we work around gcc quirks Chris@10: Chris@10: M ./simd-support/simd-sse2.h -7 +16 Chris@10: Chris@10: Mon Jun 20 21:17:59 EDT 2011 stevenj@fftw.org Chris@10: * remove the libbench directory (which we have kept lingering in the repository for years due to CVS's inability to remove directories) Chris@10: Chris@10: R ./libbench/Makefile.am Chris@10: R ./libbench/accopy-from.c Chris@10: R ./libbench/accopy-to.c Chris@10: R ./libbench/acopy.c Chris@10: R ./libbench/allocate.c Chris@10: R ./libbench/ascale.c Chris@10: R ./libbench/aset.c Chris@10: R ./libbench/bench-main.c Chris@10: R ./libbench/bench-user.h Chris@10: R ./libbench/bench.h Chris@10: R ./libbench/caadd.c Chris@10: R ./libbench/cacopy.c Chris@10: R ./libbench/can-do.c Chris@10: R ./libbench/cascale.c Chris@10: R ./libbench/caset.c Chris@10: R ./libbench/casub.c Chris@10: R ./libbench/ccopy-from.c Chris@10: R ./libbench/ccopy-to.c Chris@10: R ./libbench/copy-c2c-from.c Chris@10: R ./libbench/copy-c2c-to.c Chris@10: R ./libbench/copy-c2h-1d-fftpack.c Chris@10: R ./libbench/copy-c2h-1d-halfcomplex.c Chris@10: R ./libbench/copy-c2h-1d-packed.c Chris@10: R ./libbench/copy-c2h-1d-unpacked-ri.c Chris@10: R ./libbench/copy-c2h-unpacked.c Chris@10: R ./libbench/copy-c2h.c Chris@10: R ./libbench/copy-c2r-packed.c Chris@10: R ./libbench/copy-c2r-unpacked.c Chris@10: R ./libbench/copy-c2r.c Chris@10: R ./libbench/copy-c2ri.c Chris@10: R ./libbench/copy-h2c-1d-fftpack.c Chris@10: R ./libbench/copy-h2c-1d-halfcomplex.c Chris@10: R ./libbench/copy-h2c-1d-packed.c Chris@10: R ./libbench/copy-h2c-1d-unpacked-ri.c Chris@10: R ./libbench/copy-h2c-unpacked.c Chris@10: R ./libbench/copy-h2c.c Chris@10: R ./libbench/copy-r2c-packed.c Chris@10: R ./libbench/copy-r2c-unpacked.c Chris@10: R ./libbench/copy-r2c.c Chris@10: R ./libbench/copy-ri2c.c Chris@10: R ./libbench/deallocate.c Chris@10: R ./libbench/getopt-utils.c Chris@10: R ./libbench/getopt.c Chris@10: R ./libbench/getopt.h Chris@10: R ./libbench/getopt1.c Chris@10: R ./libbench/info.c Chris@10: R ./libbench/log2.c Chris@10: R ./libbench/main.c Chris@10: R ./libbench/mflops.c Chris@10: R ./libbench/mp.c Chris@10: R ./libbench/ovtpvt.c Chris@10: R ./libbench/pow2.c Chris@10: R ./libbench/prime.c Chris@10: R ./libbench/problem.c Chris@10: R ./libbench/report.c Chris@10: R ./libbench/speed.c Chris@10: R ./libbench/timer.c Chris@10: R ./libbench/unnormalize.c Chris@10: R ./libbench/util.c Chris@10: R ./libbench/verify.c Chris@10: R ./libbench/zero.c Chris@10: R ./libbench/ Chris@10: Chris@10: Mon Jun 20 21:17:14 EDT 2011 stevenj@fftw.org Chris@10: * update URLs Chris@10: Chris@10: M ./doc/install.texi -2 +2 Chris@10: M ./doc/intro.texi -1 +1 Chris@10: M ./doc/license.texi -1 +1 Chris@10: M ./doc/mpi.texi -1 +1 Chris@10: M ./doc/other.texi -1 +1 Chris@10: Chris@10: Mon Jun 20 20:53:31 EDT 2011 stevenj@fftw.org Chris@10: * whoops, forgot to check in alignment change Chris@10: Chris@10: M ./simd-support/simd-common.h -3 +6 Chris@10: Chris@10: Mon Jun 20 20:22:23 EDT 2011 athena@fftw.org Chris@10: * "test X = Y" requires spaces around "=" Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Mon Jun 20 19:18:52 EDT 2011 stevenj@fftw.org Chris@10: * indenting Chris@10: Chris@10: M ./simd-support/simd-sse2.h -3 +3 Chris@10: Chris@10: Mon Jun 20 18:57:10 EDT 2011 stevenj@fftw.org Chris@10: * merge back in SSE support, now combined with SSE2; --enable-sse2 now works in both single and double precision, and simd-sse2.h contains both the double- and single-precision code (which overlap a lot); in single precision it is still compiled for SSE-only (SSE2 is only required for double) Chris@10: Chris@10: M ./configure.ac -16 +9 Chris@10: M ./simd-support/simd-sse2.h -54 +181 Chris@10: M ./simd-support/sse2-nonportable.c -2 +6 Chris@10: M ./simd-support/sse2.c -4 +11 Chris@10: Chris@10: Mon Jun 20 16:02:07 EDT 2011 athena@fftw.org Chris@10: * Implement AVX autodetection (gcc-only so far) Chris@10: Chris@10: M ./simd-support/Makefile.am -1 +1 Chris@10: A ./simd-support/amd64-cpuid.h Chris@10: M ./simd-support/avx256d.c -2 +32 Chris@10: M ./simd-support/x86-cpuid.h +33 Chris@10: Chris@10: Mon Jun 20 14:25:54 EDT 2011 athena@fftw.org Chris@10: * Add VZEROUPPER at the end of AVX codelets Chris@10: Chris@10: If the Intel Optimization Manual is to be believed, we need to wave a Chris@10: dead chicken before transitioning from AVX code to SSE code. I am Chris@10: supposed to believe that there is a transition penalty for doing so, Chris@10: unless one uses a magic VZEROUPPER instruction that apparently has Chris@10: zero cost. Whatever. Chris@10: Chris@10: Chris@10: M ./genfft/c.ml +2 Chris@10: M ./genfft/c.mli +1 Chris@10: M ./genfft/gen_hc2c.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft_c.ml -1 +1 Chris@10: M ./genfft/gen_hc2hc.ml -1 +1 Chris@10: M ./genfft/gen_mdct.ml -1 +1 Chris@10: M ./genfft/gen_notw.ml -1 +1 Chris@10: M ./genfft/gen_notw_c.ml -1 +1 Chris@10: M ./genfft/gen_r2cb.ml -1 +1 Chris@10: M ./genfft/gen_r2cf.ml -1 +1 Chris@10: M ./genfft/gen_r2r.ml -1 +1 Chris@10: M ./genfft/gen_twiddle.ml -1 +1 Chris@10: M ./genfft/gen_twiddle_c.ml -1 +1 Chris@10: M ./genfft/gen_twidsq.ml -1 +1 Chris@10: M ./genfft/gen_twidsq_c.ml -1 +1 Chris@10: M ./genfft/genutil.ml -2 +2 Chris@10: M ./genfft/simd.ml +1 Chris@10: M ./simd-support/simd-avx256d.h +5 Chris@10: M ./simd-support/simd-sse2.h +2 Chris@10: Chris@10: Mon Jun 20 10:21:25 EDT 2011 athena@fftw.org Chris@10: * Move RDFT to new simd scheme Chris@10: Chris@10: A ./rdft/simd/avx256d/ Chris@10: R ./rdft/simd/codelets/Makefile.am Chris@10: R ./rdft/simd/codelets/ Chris@10: A ./rdft/simd/common/ Chris@10: A ./rdft/simd/sse2/ Chris@10: M ./Makefile.am -2 +4 Chris@10: M ./configure.ac +4 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./genfft/gen_hc2cdft_c.ml -2 +2 Chris@10: M ./kernel/ifftw.h +1 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/codelet-rdft.h -1 +2 Chris@10: M ./rdft/conf.c -2 +7 Chris@10: M ./rdft/simd/Makefile.am -6 +2 Chris@10: A ./rdft/simd/avx256d/Makefile.am Chris@10: A ./rdft/simd/codlist.mk Chris@10: A ./rdft/simd/common/Makefile.am Chris@10: A ./rdft/simd/common/genus.c Chris@10: R ./rdft/simd/hc2cbv.c Chris@10: M ./rdft/simd/hc2cbv.h -2 +2 Chris@10: R ./rdft/simd/hc2cfv.c Chris@10: M ./rdft/simd/hc2cfv.h -2 +2 Chris@10: A ./rdft/simd/simd.mk Chris@10: A ./rdft/simd/sse2/Makefile.am Chris@10: Chris@10: Mon Jun 20 09:23:38 EDT 2011 athena@fftw.org Chris@10: * New SIMD build system Chris@10: Chris@10: We now support multiple SIMD extensions in the same binary, e.g. Chris@10: --enable-sse2 --enable-avx. This patch adds the necessary Chris@10: infrastructure for SSE2/AVX and complex DFT. Later patches will add Chris@10: RDFT and SSE/ALTIVEC/etc. Chris@10: Chris@10: Chris@10: A ./dft/simd/avx256d/ Chris@10: A ./simd-support/ Chris@10: R ./dft/simd/codelets/Makefile.am Chris@10: R ./dft/simd/codelets/ Chris@10: A ./dft/simd/common/ Chris@10: A ./dft/simd/sse2/ Chris@10: M ./Makefile.am -11 +13 Chris@10: M ./configure.ac -47 +74 Chris@10: M ./dft/codelet-dft.h -4 +2 Chris@10: M ./dft/conf.c -2 +7 Chris@10: M ./dft/scalar/codelets/Makefile.am -1 +1 Chris@10: M ./dft/simd/Makefile.am -6 +3 Chris@10: A ./dft/simd/avx256d/Makefile.am Chris@10: A ./dft/simd/codlist.mk Chris@10: A ./dft/simd/common/Makefile.am Chris@10: A ./dft/simd/common/genus.c Chris@10: R ./dft/simd/n1b.c Chris@10: M ./dft/simd/n1b.h -2 +2 Chris@10: R ./dft/simd/n1f.c Chris@10: M ./dft/simd/n1f.h -2 +2 Chris@10: R ./dft/simd/n2b.c Chris@10: M ./dft/simd/n2b.h -2 +2 Chris@10: R ./dft/simd/n2f.c Chris@10: M ./dft/simd/n2f.h -2 +2 Chris@10: R ./dft/simd/n2s.c Chris@10: M ./dft/simd/n2s.h -2 +2 Chris@10: R ./dft/simd/q1b.c Chris@10: M ./dft/simd/q1b.h -2 +2 Chris@10: R ./dft/simd/q1f.c Chris@10: M ./dft/simd/q1f.h -2 +2 Chris@10: A ./dft/simd/simd.mk Chris@10: A ./dft/simd/sse2/Makefile.am Chris@10: R ./dft/simd/t.c Chris@10: M ./dft/simd/t1b.h -2 +2 Chris@10: M ./dft/simd/t1bu.h -2 +2 Chris@10: M ./dft/simd/t1f.h -2 +2 Chris@10: M ./dft/simd/t1fu.h -2 +2 Chris@10: M ./dft/simd/t2b.h -2 +2 Chris@10: M ./dft/simd/t2f.h -2 +2 Chris@10: M ./dft/simd/t3b.h -2 +2 Chris@10: M ./dft/simd/t3f.h -2 +2 Chris@10: R ./dft/simd/ts.c Chris@10: M ./dft/simd/ts.h -2 +2 Chris@10: M ./genfft/gen_twiddle.ml -2 +2 Chris@10: M ./genfft/genutil.ml -2 +5 Chris@10: M ./kernel/ifftw.h -2 +6 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/conf.c -1 +1 Chris@10: M ./rdft/scalar/r2cb/Makefile.am +1 Chris@10: M ./rdft/scalar/r2cf/Makefile.am +1 Chris@10: M ./rdft/scalar/r2r/Makefile.am +1 Chris@10: A ./simd-support/Makefile.am Chris@10: A ./simd-support/avx256d.c Chris@10: A ./simd-support/simd-avx256d.h Chris@10: A ./simd-support/simd-common.h Chris@10: A ./simd-support/simd-sse2.h Chris@10: A ./simd-support/sse2-nonportable.c Chris@10: A ./simd-support/sse2.c Chris@10: A ./simd-support/taint.c Chris@10: A ./simd-support/x86-cpuid.h Chris@10: M ./support/Makefile.codelets -21 +22 Chris@10: Chris@10: Sun Jun 19 12:29:27 EDT 2011 stevenj@fftw.org Chris@10: * italicize Latin quote Chris@10: Chris@10: M ./doc/reference.texi -2 +2 Chris@10: Chris@10: Sun Jun 19 12:26:34 EDT 2011 stevenj@fftw.org Chris@10: * work around incredibly annoying makeinfo bug -- for HTML output, in any paragraph ending with an @index command, two blank lines are needed to create a paragraph break ... our HTML output has apparently been screwed up for years Chris@10: Chris@10: M ./doc/acknowledgements.texi +2 Chris@10: M ./doc/fortran.texi +1 Chris@10: M ./doc/install.texi +2 Chris@10: M ./doc/intro.texi +1 Chris@10: M ./doc/mpi.texi +20 Chris@10: M ./doc/other.texi +2 Chris@10: M ./doc/reference.texi +21 Chris@10: M ./doc/threads.texi +2 Chris@10: M ./doc/tutorial.texi +13 Chris@10: Chris@10: Sun Jun 19 12:01:39 EDT 2011 stevenj@fftw.org Chris@10: * conjugate-pair algorithm turns out not to be due to djb, but it was pointed out to us by djb Chris@10: Chris@10: M ./doc/intro.texi -3 +4 Chris@10: Chris@10: Sun Jun 19 11:59:21 EDT 2011 stevenj@fftw.org Chris@10: * new-array execute functions are *not* the same for MPI, since the problems are different Chris@10: Chris@10: M ./doc/mpi.texi -2 +3 Chris@10: Chris@10: Sun Jun 19 11:57:52 EDT 2011 stevenj@fftw.org Chris@10: * tweaks to MPI manual Chris@10: Chris@10: M ./doc/mpi.texi -19 +44 Chris@10: M ./doc/reference.texi -1 +1 Chris@10: Chris@10: Sun Jun 19 00:47:23 EDT 2011 stevenj@fftw.org Chris@10: * only call MPI_Init_thread for MPI version >= 2 Chris@10: Chris@10: M ./mpi/mpi-bench.c +5 Chris@10: Chris@10: Sun Jun 19 00:35:44 EDT 2011 stevenj@fftw.org Chris@10: * document quad precision in Fortran Chris@10: Chris@10: M ./doc/fortran.texi -9 +12 Chris@10: Chris@10: Sun Jun 19 00:16:10 EDT 2011 stevenj@fftw.org Chris@10: * use -lquadmath for quad-precision library in pkgconfig file Chris@10: Chris@10: M ./configure.ac +2 Chris@10: M ./fftw.pc.in -1 +1 Chris@10: Chris@10: Sun Jun 19 00:10:33 EDT 2011 stevenj@fftw.org Chris@10: * document quad precision Chris@10: Chris@10: M ./doc/install.texi -1 +11 Chris@10: M ./doc/reference.texi +8 Chris@10: Chris@10: Sat Jun 18 21:26:24 EDT 2011 stevenj@fftw.org Chris@10: * there is currently no standard quad-precision type in MPI, so don't allow MPI support to be compiled with --enable-quad-precision Chris@10: Chris@10: M ./configure.ac +3 Chris@10: M ./mpi/ifftw-mpi.h +2 Chris@10: M ./mpi/mpi-bench.c +2 Chris@10: Chris@10: Sat Jun 18 21:25:51 EDT 2011 stevenj@fftw.org Chris@10: * make quad-precision library libfftwq Chris@10: Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Sat Jun 18 21:19:50 EDT 2011 stevenj@fftw.org Chris@10: * use --estimate in check script with --enable-random-estimator Chris@10: Chris@10: M ./configure.ac +2 Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./tests/Makefile.am -5 +5 Chris@10: Chris@10: Sat Jun 18 20:36:51 EDT 2011 stevenj@fftw.org Chris@10: * bug fix - correct crashing interaction between threads and debug-malloc Chris@10: Chris@10: M ./threads/threads.c +4 Chris@10: Chris@10: Sat Jun 18 18:41:48 EDT 2011 stevenj@fftw.org Chris@10: * whoops, typo Chris@10: Chris@10: M ./doc/reference.texi -1 +1 Chris@10: Chris@10: Sat Jun 18 18:35:14 EDT 2011 stevenj@fftw.org Chris@10: * clarification of when fftw_cost may return 0 Chris@10: Chris@10: M ./doc/reference.texi -2 +5 Chris@10: Chris@10: Sat Jun 18 18:31:26 EDT 2011 stevenj@fftw.org Chris@10: * corrected manual and test program for proper interaction of MPI and threads Chris@10: Chris@10: M ./doc/mpi.texi -20 +59 Chris@10: M ./mpi/mpi-bench.c +13 Chris@10: M ./tests/fftw-bench.c -4 +10 Chris@10: M ./tests/fftw-bench.h +5 Chris@10: Chris@10: Sat Jun 18 17:13:52 EDT 2011 stevenj@fftw.org Chris@10: * need --mpi restriction in mpi+threads check Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: Chris@10: Sat Jun 18 12:09:04 EDT 2011 stevenj@fftw.org Chris@10: * split fftw3.texi into multiple files for ease of editing Chris@10: Chris@10: M ./doc/Makefile.am -2 +1 Chris@10: A ./doc/acknowledgements.texi Chris@10: A ./doc/cindex.texi Chris@10: M ./doc/fftw3.texi -6158 +14 Chris@10: A ./doc/findex.texi Chris@10: A ./doc/fortran.texi Chris@10: A ./doc/install.texi Chris@10: A ./doc/intro.texi Chris@10: A ./doc/license.texi Chris@10: A ./doc/mpi.texi Chris@10: A ./doc/other.texi Chris@10: A ./doc/reference.texi Chris@10: A ./doc/threads.texi Chris@10: A ./doc/tutorial.texi Chris@10: A ./doc/upgrading.texi Chris@10: Chris@10: Fri Jun 17 17:26:50 EDT 2011 stevenj@fftw.org Chris@10: * merge recent Cell deletion with MPI branch Chris@10: Chris@10: M ./Makefile.am -3 +2 Chris@10: M ./NEWS +40 Chris@10: M ./configure.ac -1 +1 Chris@10: M ./doc/fftw3.texi -95 +909 Chris@10: Chris@10: Fri Jun 17 01:52:59 EDT 2011 stevenj@fftw.org Chris@10: * whoops, incorrect assertion Chris@10: Chris@10: M ./mpi/transpose-problem.c -2 +2 Chris@10: Chris@10: Fri Jun 17 01:52:51 EDT 2011 stevenj@fftw.org Chris@10: * comment fix Chris@10: Chris@10: M ./mpi/block.c -2 +2 Chris@10: Chris@10: Thu Jun 16 23:30:27 EDT 2011 stevenj@fftw.org Chris@10: * check if pln creation failed (e.g. for split input) bbefore calling setup_gather_scatter, to prevent crashes Chris@10: Chris@10: M ./mpi/mpi-bench.c -1 +1 Chris@10: Chris@10: Thu Jun 16 23:26:48 EDT 2011 stevenj@fftw.org Chris@10: * bug fix -- transpose-recurse is only applicable if subtransposes fit in the same space (unless I change the allocation routine, but this would seem to require looking at all possible recursive invocations of transpose-recurse) Chris@10: Chris@10: M ./mpi/transpose-recurse.c +28 Chris@10: Chris@10: Mon Apr 11 17:58:03 EDT 2011 stevenj@fftw.org Chris@10: * yikes, any_true check on subplan creation should be in comm, not comm2, so that all processes know if failure occurred Chris@10: Chris@10: M ./mpi/transpose-recurse.c -6 +4 Chris@10: Chris@10: Mon Apr 11 17:00:46 EDT 2011 stevenj@fftw.org Chris@10: * add wisdom_ok_hook to enforce wisdom synchronization on MPI problems, apparently fixing a longstanding deadlock/crash bug Chris@10: Chris@10: M ./kernel/ifftw.h +1 Chris@10: M ./kernel/planner.c +7 Chris@10: M ./mpi/api.c -13 +61 Chris@10: Chris@10: Mon Apr 11 15:10:22 EDT 2011 stevenj@fftw.org Chris@10: * add a check (in DEBUG mode only) that all processes produce the same hash of MPI problems; don't include alignment in MPI problem hash because it may differ between processes for unaligned malloc Chris@10: Chris@10: M ./mpi/any-true.c +21 Chris@10: M ./mpi/dft-problem.c -2 +6 Chris@10: M ./mpi/ifftw-mpi.h +1 Chris@10: M ./mpi/rdft-problem.c -2 +6 Chris@10: M ./mpi/rdft2-problem.c -2 +6 Chris@10: M ./mpi/transpose-problem.c -2 +6 Chris@10: Chris@10: Fri Apr 8 18:46:54 EDT 2011 stevenj@fftw.org Chris@10: * use cost_hook in random_estimate Chris@10: Chris@10: M ./kernel/planner.c -1 +1 Chris@10: Chris@10: Sun Mar 6 23:33:53 EST 2011 stevenj@fftw.org Chris@10: * added mpi new-array execute functions; thanks to Guo Luo for the bug report Chris@10: Chris@10: M ./mpi/api.c +23 Chris@10: M ./mpi/fftw3-mpi.h -1 +6 Chris@10: Chris@10: Wed Feb 9 21:29:17 EST 2011 stevenj@fftw.org Chris@10: * MPI may not support tags > 2^15-1 (e.g. Cray MPI requires tags < 2^24); thanks to Jonathan Bentz for the bug report. Chris@10: Chris@10: M ./mpi/transpose-pairwise.c -4 +4 Chris@10: Chris@10: Wed Feb 2 12:21:30 EST 2011 stevenj@fftw.org Chris@10: * fix merge conflicts Chris@10: Chris@10: M! ./configure.ac -1 +1 Chris@10: Chris@10: Sat Nov 15 22:33:20 EST 2008 stevenj@fftw.org Chris@10: tagged fftw-3.3alpha1 Chris@10: Chris@10: Chris@10: Sat Nov 15 20:33:33 EST 2008 stevenj@fftw.org Chris@10: * version bump for 3.3alpha1 Chris@10: Chris@10: M! ./NEWS -40 Chris@10: M! ./configure.ac -2 +2 Chris@10: Chris@10: Sun Oct 26 22:47:07 EDT 2008 stevenj@fftw.org Chris@10: * re-added mpi/Makefile Chris@10: Chris@10: M ./configure.ac +2 Chris@10: Chris@10: Sat Oct 25 17:14:42 EDT 2008 stevenj@fftw.org Chris@10: * re-add MPI to dist Chris@10: Chris@10: M! ./Makefile.am -2 +3 Chris@10: M! ./doc/fftw3.texi -9 +134 Chris@10: Chris@10: Sat Jun 18 08:50:13 EDT 2011 athena@fftw.org Chris@10: * remove obsolete Cell code Chris@10: Chris@10: M ./kernel/align.c -1 +1 Chris@10: M ./kernel/ifftw.h -9 +2 Chris@10: M ./rdft/rank0.c -83 Chris@10: M ./simd/simd.h -6 Chris@10: M ./simd/taint.c -1 +1 Chris@10: M ./tests/fftw-bench.c -7 Chris@10: Chris@10: Fri Jun 17 23:31:33 EDT 2011 stevenj@fftw.org Chris@10: * bug fix in accuracy test, which prevented us from consiistently determining accuracy in > double precision Chris@10: Chris@10: M ./libbench2/mp.c -2 +2 Chris@10: Chris@10: Fri Jun 17 20:05:13 EDT 2011 athena@fftw.org Chris@10: * do not check for gcc version before checking for gcc Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Fri Jun 17 18:56:37 EDT 2011 stevenj@fftw.org Chris@10: * require gcc 4.6.0 or later for --enable-quad-precision, to match fftw3.h header file; no need to mark this as EXPERIMENTAL (make check passes, and support in gcc 4.6 seems reasonably complete) Chris@10: Chris@10: M ./configure.ac -1 +2 Chris@10: Chris@10: Fri Jun 17 18:51:41 EDT 2011 stevenj@fftw.org Chris@10: * need ugly __attribute__ to use __float128 with _Complex, ugh Chris@10: Chris@10: M ./api/fftw3.h +9 Chris@10: Chris@10: Fri Jun 17 18:23:05 EDT 2011 stevenj@fftw.org Chris@10: * --verify tolerance in quad precision changed to 1e-29 Chris@10: Chris@10: M ./libbench2/bench-main.c -1 +1 Chris@10: Chris@10: Fri Jun 17 18:22:38 EDT 2011 stevenj@fftw.org Chris@10: * quad-precision F77 api should use "qfftw" prefix Chris@10: Chris@10: M ./api/x77.h +3 Chris@10: Chris@10: Fri Jun 17 18:22:27 EDT 2011 stevenj@fftw.org Chris@10: * rm extraneous space from fftw3.h Chris@10: Chris@10: M ./api/fftw3.h -1 +1 Chris@10: Chris@10: Fri Jun 17 18:05:10 EDT 2011 stevenj@fftw.org Chris@10: * use cosq etcetera with libquadmath in libbench2, so that --verify correctly gives ~33 decimal places in shift test Chris@10: Chris@10: M ./libbench2/verify.h -1 +4 Chris@10: Chris@10: Fri Jun 17 17:52:51 EDT 2011 stevenj@fftw.org Chris@10: * libquadmath ships with gcc 4.6.0, so we should require this library for sinq/cosq with --enable-quad-precision; also, include the __float128 FFTW functions in the header file for gcc >= 4.6 on i86/x86??_64/ia64 Chris@10: Chris@10: M ./Makefile.am +5 Chris@10: M ./api/fftw3.h -1 +6 Chris@10: M ./configure.ac -1 +5 Chris@10: M ./kernel/trig.c -13 +4 Chris@10: Chris@10: Fri Jun 17 16:54:01 EDT 2011 stevenj@fftw.org Chris@10: * typo in manual for fftw_cost Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Fri Jun 17 16:48:24 EDT 2011 stevenj@fftw.org Chris@10: * fix fftw_cost function: pcost needs to be saved in mkapiplan, since the plan is re-created from wisdom Chris@10: Chris@10: M ./api/apiplan.c +5 Chris@10: M ./tests/fftw-bench.c -2 +3 Chris@10: Chris@10: Fri Jun 17 16:42:25 EDT 2011 athena@fftw.org Chris@10: * removed support for the Cell Broadband Engine Chris@10: Chris@10: R ./cell/spu/Makefile.am Chris@10: R ./cell/spu/alloc.spuc Chris@10: R ./cell/spu/copy.spuc Chris@10: R ./cell/spu/dft.spuc Chris@10: R ./cell/spu/dma.spuc Chris@10: R ./cell/spu/execute.spuc Chris@10: R ./cell/spu/fftw-spu.h Chris@10: R ./cell/spu/main.spuc Chris@10: R ./cell/spu/planner.spuc Chris@10: R ./cell/spu/spu-double.h Chris@10: R ./cell/spu/spu-single.h Chris@10: R ./cell/spu/transpose.spuc Chris@10: R ./cell/spu/ Chris@10: R ./cell/Makefile.am Chris@10: R ./cell/cell.c Chris@10: R ./cell/conf.c Chris@10: R ./cell/copy.c Chris@10: R ./cell/dft-direct-cell.c Chris@10: R ./cell/fftw-cell.h Chris@10: R ./cell/plans-double.c Chris@10: R ./cell/plans-single.c Chris@10: R ./cell/spufftw-embed.S Chris@10: R ./cell/transpose.c Chris@10: R ./cell/ Chris@10: M ./AUTHORS -2 +3 Chris@10: M ./Makefile.am -11 +5 Chris@10: M ./NEWS +2 Chris@10: R ./README.Cell Chris@10: M ./api/configure.c -3 Chris@10: M ./configure.ac -18 +1 Chris@10: M ./dft/dft.h -4 Chris@10: M ./dft/rank-geq2.c -6 Chris@10: M ./doc/fftw3.texi -128 +4 Chris@10: Chris@10: Tue May 24 06:51:07 EDT 2011 athena@fftw.org Chris@10: * Undo previous change; the typo was not a typo after all. Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Mon May 23 05:08:05 EDT 2011 athena@fftw.org Chris@10: * Fix typo in manual Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Sat May 21 17:37:50 EDT 2011 athena@fftw.org Chris@10: * clarify intent about canonicalization of tensor in tensor_compress_contiguous() Chris@10: Chris@10: M ./kernel/tensor7.c -1 +9 Chris@10: Chris@10: Sat May 21 17:30:31 EDT 2011 athena@fftw.org Chris@10: * avoid useless canonicalization in tensor_compress_contiguous() Chris@10: Chris@10: M ./kernel/tensor7.c -1 +1 Chris@10: Chris@10: Sat May 21 17:24:57 EDT 2011 athena@fftw.org Chris@10: * Fix tensor_compress_contiguous Chris@10: Chris@10: tensor_compress_contiguous() was supposed to sort dimensions by Chris@10: descending istride, and then compress adjacent dimensions. This Chris@10: property was lost once we changed the canonical order of strides to be Chris@10: sorted by descending min{istride,ostride}. Chris@10: Chris@10: Change tensor_compress_contiguous() to sort by descending istride Chris@10: again, which is necessary for its correctness, and then canonicalize Chris@10: at the end. Chris@10: Chris@10: Chris@10: M ./kernel/tensor7.c -20 +41 Chris@10: Chris@10: Sun May 8 18:47:26 EDT 2011 athena@fftw.org Chris@10: * Don't distribute obsolete .depend Chris@10: Chris@10: M ./genfft/Makefile.am -2 +2 Chris@10: Chris@10: Sun May 8 18:05:36 EDT 2011 athena@fftw.org Chris@10: * Use ocamlbuild for building genfft Chris@10: Chris@10: Remove the old Makefile cruft to support ocaml, and use ocamlbuild Chris@10: instead. Chris@10: Chris@10: M ./bootstrap.sh -2 Chris@10: M ./configure.ac -10 +1 Chris@10: M ./genfft/Makefile.am -151 +8 Chris@10: M ./m4/Makefile.am -1 +1 Chris@10: R ./m4/ocaml.m4 Chris@10: M ./support/Makefile.codelets -13 +13 Chris@10: Chris@10: Sun May 8 18:03:07 EDT 2011 athena@fftw.org Chris@10: * Do not use __float128 unless BENCHFFT_QUAD is defined Chris@10: Chris@10: Otherwise, compilation fails on compilers that do not support Chris@10: __float128. Chris@10: Chris@10: Chris@10: M ./libbench2/bench-user.h +5 Chris@10: Chris@10: Fri Apr 8 13:15:54 EDT 2011 stevenj@fftw.org Chris@10: * fix configure --help string for --disable-alloca (since default is enabled) Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Fri Apr 8 13:09:56 EDT 2011 stevenj@fftw.org Chris@10: * add "random estimator" for debugging purposes; note that this is best used with ESTIMATE_PATIENT mode Chris@10: Chris@10: M ./api/apiplan.c +4 Chris@10: M ./configure.ac +5 Chris@10: M ./kernel/ifftw.h +4 Chris@10: M ./kernel/planner.c +27 Chris@10: M ./tests/fftw-bench.c +7 Chris@10: Chris@10: Tue Apr 5 14:47:56 EDT 2011 stevenj@fftw.org Chris@10: * add AC_CHECK_DECLS for srand48; thanks to Ralf Wildenhues for the bug report Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./libbench2/util.c +3 Chris@10: Chris@10: Sat Feb 5 17:00:40 EST 2011 stevenj@fftw.org Chris@10: * experimental support for gcc's __float128 quad-precision type Chris@10: Chris@10: M ./api/fftw3.h +3 Chris@10: M ./configure.ac -2 +14 Chris@10: M ./kernel/ifftw.h -2 +10 Chris@10: M ./kernel/trig.c -1 +16 Chris@10: M ./libbench2/bench-main.c +2 Chris@10: M ./libbench2/bench-user.h +4 Chris@10: M ./libbench2/info.c -1 +2 Chris@10: M ./libbench2/verify.h +6 Chris@10: M ./simd/simd-sse2.h -1 +1 Chris@10: M ./tests/fftw-bench.h +2 Chris@10: M ./tools/fftw-wisdom.c +2 Chris@10: Chris@10: Sun Oct 24 14:33:59 EDT 2010 athena@fftw.org Chris@10: * guarantee that "timelimit < 0" means "no timeout" Chris@10: "timelimit < 0" was always meant to be equivalent to Chris@10: "timelimit = HUGENUM", but this was not true in all cases, Chris@10: causing some obscure wisdom behavior. Chris@10: Chris@10: Thanks William Andrew Burnson for the bug report. Chris@10: Chris@10: Chris@10: M ./api/mapflags.c -1 +1 Chris@10: Chris@10: Sun Oct 24 14:32:20 EDT 2010 athena@fftw.org Chris@10: * compile with --enable-fma and SSE, SSE2 Chris@10: Allow compilation with --enable-fma and --enable-sse, --enable-sse2. Chris@10: This is a bad idea performance-wise, but people will try anyway. Chris@10: Chris@10: M ./simd/simd-sse.h +5 Chris@10: M ./simd/simd-sse2.h +5 Chris@10: Chris@10: Sun Jul 11 13:34:06 EDT 2010 athena@fftw.org Chris@10: * Make threads.c compiler with c++ Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Sun Jul 11 10:05:05 EDT 2010 athena@fftw.org Chris@10: * Attempt at clarifying the advanced interface doc. Chris@10: Chris@10: M ./doc/fftw3.texi -72 +151 Chris@10: Chris@10: Sun Jul 11 07:37:27 EDT 2010 athena@fftw.org Chris@10: * rename rfftwnd html picture Chris@10: Chris@10: It turns out that texinfo with pdf output reads .png Chris@10: files in preference to .pdf files (when did this change?). Chris@10: I renamed the .png figure to avoid producing an ugly pdf file. Chris@10: Chris@10: M ./doc/Makefile.am -4 +4 Chris@10: M ./doc/fftw3.texi +5 Chris@10: Chris@10: Tue Mar 30 19:43:22 EDT 2010 stevenj@fftw.org Chris@10: * added fftw_cost function; this is the second time people have asked for this, and there is a reasonable use for it in comparing e.g. oout-of-place vs. in-place plans Chris@10: Chris@10: M ./NEWS +4 Chris@10: M ./api/f77funcs.h +10 Chris@10: M ./api/fftw3.h +1 Chris@10: M ./api/flops.c +5 Chris@10: M ./doc/fftw3.texi +17 Chris@10: Chris@10: Tue Mar 2 18:55:49 EST 2010 stevenj@fftw.org Chris@10: * documented that --enable-debug-malloc causes fftw_execute to be thread-unsafe (thanks to Alexis Rohou for the problem report) Chris@10: Chris@10: M ./doc/fftw3.texi +8 Chris@10: Chris@10: Fri Jan 22 19:42:08 EST 2010 athena@fftw.org Chris@10: * Added FAQ about how to transpose matrices using FFTW. Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +33 Chris@10: Chris@10: Thu Jan 7 20:16:57 EST 2010 stevenj@fftw.org Chris@10: * catch FMS (instead of generating FMA(_,_,NEG(_)) with h -generic-arith option Chris@10: Chris@10: M ./genfft/c.ml +2 Chris@10: Chris@10: Fri Dec 11 07:01:26 EST 2009 athena@fftw.org Chris@10: * note future wisdom enhancements. Chris@10: Chris@10: M ./TODO +5 Chris@10: Chris@10: Mon Oct 19 20:21:05 EDT 2009 athena@fftw.org Chris@10: * Use SIMD flags when checking for xmmintrin.h Chris@10: Chris@10: This prevents an obnoxious warning from configure. Chris@10: Chris@10: M ./configure.ac -3 +3 Chris@10: Chris@10: Sat Aug 29 20:47:56 EDT 2009 athena@fftw.org Chris@10: * new bug Chris@10: Chris@10: M ./TODO +4 Chris@10: Chris@10: Sun Jul 26 00:40:11 EDT 2009 stevenj@fftw.org Chris@10: * typo (s/man1/many) Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Tue Jul 14 14:19:08 EDT 2009 stevenj@fftw.org Chris@10: * BUILD-MINGW32 script, updated Windows README Chris@10: Chris@10: A ./support/BUILD-MINGW32.sh Chris@10: M ./support/BUILD-MINGW64.sh -4 +4 Chris@10: Chris@10: Mon Jul 13 09:40:38 EDT 2009 fftw@fftw.org Chris@10: * cleanup BUILD-MINGW64.sh Chris@10: Chris@10: M ./support/BUILD-MINGW64.sh -19 +27 Chris@10: Chris@10: Sun Jul 12 06:34:46 EDT 2009 athena@fftw.org Chris@10: * Update NEWS, version number for 3.2.2 release. Chris@10: Chris@10: M ./NEWS +3 Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Sat Jul 11 22:28:38 EDT 2009 athena@fftw.org Chris@10: * Reintroduce the pruning heuristic in ESTIMATE mode for r2r problems. Chris@10: Chris@10: Somehow, we lost this feature between fftw-3.1.3 and fftw-3.2. Chris@10: Chris@10: M ./rdft/hc2hc-direct.c +3 Chris@10: Chris@10: Thu Jun 25 07:39:04 EDT 2009 athena@fftw.org Chris@10: * don't use pshared=1 in sem_init Chris@10: Chris@10: pshared is really not necessary, and it is not supported on Chris@10: GNU/kFreeBSD. Thanks Petr Salinger for the bug report. Chris@10: Chris@10: M ./threads/threads.c -2 +2 Chris@10: Chris@10: Thu Jun 11 19:35:40 EDT 2009 fftw@fftw.org Chris@10: * Add mingw64 build file so that we can track it. Chris@10: Chris@10: A ./support/BUILD-MINGW64.sh Chris@10: Chris@10: Wed Jun 10 12:10:58 EDT 2009 fftw@fftw.org Chris@10: * note 3.2.2 NEWS Chris@10: Chris@10: M ./NEWS +7 Chris@10: Chris@10: Wed Jun 10 12:04:54 EDT 2009 fftw@fftw.org Chris@10: * add --disable-alloca to configure Chris@10: Chris@10: It looks like alloca() is broken on mingw64, and thus Chris@10: we need to disable it explicitly. Chris@10: Chris@10: M ./configure.ac -1 +6 Chris@10: M ./kernel/ifftw.h -1 +1 Chris@10: Chris@10: Sun Apr 26 16:33:10 EDT 2009 athena@fftw.org Chris@10: * Note in FAQ that --enable-k7 has been discontinued. Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -4 +7 Chris@10: Chris@10: Tue Mar 24 09:16:18 EDT 2009 athena@fftw.org Chris@10: * clarified small confusion in fftw_cleanup documentation Chris@10: Chris@10: M ./doc/fftw3.texi -2 +3 Chris@10: Chris@10: Thu Mar 19 13:18:06 EDT 2009 stevenj@fftw.org Chris@10: * fix documentation of dfftw_init_threads to indicate thaat it takes an argument (since the C version returns a value); thanks t Hans Johnnston for the bug report Chris@10: Chris@10: M ./doc/fftw3.texi -1 +5 Chris@10: Chris@10: Thu Mar 12 13:12:13 EDT 2009 fftw@fftw.org Chris@10: * if possible, use a 128-bit type for copy Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./kernel/cpy2d.c -9 +38 Chris@10: Chris@10: Tue Mar 10 12:49:51 EDT 2009 fftw@fftw.org Chris@10: * add size-128 simd codelets Chris@10: Chris@10: It's about time Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -2 +2 Chris@10: Chris@10: Mon Mar 9 20:29:16 EDT 2009 athena@fftw.org Chris@10: * copy two floats as a double when possible Chris@10: Chris@10: Resurrect the old hack of copying two floats as a double, Chris@10: which makes some difference in these days of 64 bit boxes. Chris@10: Chris@10: Chris@10: M ./kernel/cpy2d.c -16 +34 Chris@10: Chris@10: Sun Mar 8 18:08:04 EDT 2009 athena@fftw.org Chris@10: * fixed (harmless) confusion of strides Chris@10: Chris@10: RS and VS were swapped in dftw-direct.c. This is a bug, but Chris@10: it is harmless unless one uses fixed-stride codelets, which we Chris@10: do not. Chris@10: Chris@10: M ./dft/dftw-direct.c -7 +7 Chris@10: Chris@10: Sun Mar 8 10:29:49 EDT 2009 athena@fftw.org Chris@10: * oops, I checked in debug code accidentally. Chris@10: Chris@10: M ./kernel/ct.c -1 Chris@10: Chris@10: Sat Feb 14 19:01:00 EST 2009 athena@fftw.org Chris@10: * Change TLO email address since Magdalen is no longer there. Chris@10: Chris@10: M ./commercialize.sh -1 +1 Chris@10: Chris@10: Sat Feb 14 18:18:45 EST 2009 stevenj@fftw.org Chris@10: * quote arguments to bench in test script on the off-chance that '*' would be expanded by the shell into a valid filename, and also to avoid shell confusion on Cygwin that "//" begins the name of a Windows network mountpoint Chris@10: Chris@10: M ./tests/check.pl -1 +1 Chris@10: Chris@10: Sat Feb 14 18:17:23 EST 2009 stevenj@fftw.org Chris@10: tagged fftw-3.2.1 Chris@10: Chris@10: Chris@10: Sun Mar 8 10:02:59 EDT 2009 athena@fftw.org Chris@10: * stricter conditions for Cooley-Tukey being ugly Chris@10: Chris@10: It turns out that m=2 in the leaf of Cooley-Tukey may be Chris@10: advantageous in certain cases, eg. i512v512 on AMD Shanghai: Chris@10: Chris@10: (dft-buffered-512-x128/512-6 Chris@10: (dft-ct-dit/4 Chris@10: (dftw-direct-4/24-x128 "t2fv_4") Chris@10: (dft-vrank>=1-x4/1 Chris@10: (dft-ct-dit/64 Chris@10: (dftw-direct-64/504-x128 "t2fv_64") Chris@10: (dft-vrank>=1-x64/1 Chris@10: (dft-direct-2-x128 "n2fv_2"))))) Chris@10: (dft-r2hc-1 Chris@10: (rdft-rank0-tiled/2-x128-x512)) Chris@10: (dft-nop)) Chris@10: Chris@10: Presumably this works around the 2 way associativity of the L1 cache. Chris@10: Chris@10: M ./dft/dftw-direct.c -1 +1 Chris@10: M ./kernel/ct.c -2 +3 Chris@10: M ./kernel/ifftw.h -1 +1 Chris@10: M ./rdft/ct-hc2c-direct.c -1 +1 Chris@10: M ./rdft/hc2hc-direct.c -3 +3 Chris@10: Chris@10: Mon Feb 9 19:46:00 EST 2009 stevenj@fftw.org Chris@10: * disable Windows QueryPerformanceCounter code, since it requires us to pull in windows.h in ifftw.h and causes namespace conflicts; gettimeofday seems to work well enough and has had few complaints Chris@10: Chris@10: M ./kernel/ifftw.h -4 +1 Chris@10: M ./kernel/timer.c -1 +8 Chris@10: Chris@10: Wed Feb 4 22:55:54 EST 2009 stevenj@fftw.org Chris@10: * version bump for 3.2.1, updated NEWS Chris@10: Chris@10: M ./NEWS +10 Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Wed Feb 4 22:27:28 EST 2009 stevenj@fftw.org Chris@10: * recommend that users avoid fftw_execute in Fortran, instead using dfftw_execute_dft and friends so that the compiler knows that the input/output arrays are used Chris@10: Chris@10: M ./doc/fftw3.texi -10 +80 Chris@10: Chris@10: Wed Jan 21 16:02:08 EST 2009 stevenj@fftw.org Chris@10: * prefer windows queryperformancecounter to gettimeofday on Windows, thanks to David Price for the suggestion Chris@10: Chris@10: M ./kernel/ifftw.h -1 +4 Chris@10: M ./kernel/timer.c -1 +20 Chris@10: Chris@10: Sun Feb 1 14:34:49 EST 2009 athena@fftw.org Chris@10: * compilation fixes in case snprintf() is defined as a macro. Chris@10: Chris@10: M ./libbench2/report.c -8 +7 Chris@10: Chris@10: Wed Jan 28 20:19:04 EST 2009 athena@fftw.org Chris@10: * Automake does not like continuation lines beginning with a comment. Chris@10: Chris@10: M ./rdft/scalar/r2cb/Makefile.am -1 +1 Chris@10: Chris@10: Wed Jan 28 18:24:39 EST 2009 athena@fftw.org Chris@10: * Add r2cb_2.c Chris@10: Chris@10: r2cb_2.c is needed for problem rb2, which is not equivalent to Chris@10: rf2 (unlike kb2, which is equivalent to kf2). Chris@10: Chris@10: This change would not matter much except that rb2 is generated Chris@10: when reducing backward rdft2 to dft, and the absence of the codelet was Chris@10: preventing radix 2 from being employed at all in this case. Chris@10: Chris@10: M ./rdft/scalar/r2cb/Makefile.am -6 +4 Chris@10: Chris@10: Sat Jan 10 06:47:22 EST 2009 athena@fftw.org Chris@10: * handle the case vecsz->rnk == 0 correctly. Chris@10: Chris@10: M ./dft/buffered.c -1 +3 Chris@10: M ./rdft/buffered.c -1 +3 Chris@10: M ./rdft/buffered2.c -1 +3 Chris@10: Chris@10: Fri Dec 19 15:20:36 EST 2008 stevenj@fftw.org Chris@10: * Macs are no longer ppc-based; thanks to Charles Collicutt for the FAQ update Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -3 +3 Chris@10: Chris@10: Mon Dec 8 18:08:33 EST 2008 stevenj@fftw.org Chris@10: * use new multiple-nbuf code in rdft/buffered, like for dft/buffered Chris@10: Chris@10: M ./rdft/buffered.c -11 +25 Chris@10: M ./rdft/buffered2.c -12 +26 Chris@10: Chris@10: Sat Dec 6 16:34:36 EST 2008 stevenj@fftw.org Chris@10: * make x86_cpuid macro work on x86_64 Chris@10: Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 -1 +15 Chris@10: Chris@10: Sat Dec 6 09:20:37 EST 2008 athena@fftw.org Chris@10: * Allow automatic choice of buffer size in dft/buffered.c Chris@10: Chris@10: Try a couple of different buffer sizes in buffered transforms, Chris@10: since this seems to make a difference on some Core2 models. Chris@10: Chris@10: Chris@10: M ./api/mapflags.c -2 Chris@10: M ./dft/buffered.c -10 +25 Chris@10: M ./kernel/buffered.c -3 +19 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: M ./rdft/buffered.c -2 +3 Chris@10: M ./rdft/buffered2.c -2 +3 Chris@10: M ./rdft/rdft2-rdft.c -1 +1 Chris@10: Chris@10: Tue Dec 2 19:18:30 EST 2008 athena@fftw.org Chris@10: * libbench2: do not assume that split-complex arrays are stride-1 Chris@10: Chris@10: M ./libbench2/verify-dft.c -4 +3 Chris@10: M ./libbench2/verify.c -1 +3 Chris@10: M ./tests/bench.c -3 +2 Chris@10: Chris@10: Tue Dec 2 18:39:43 EST 2008 stevenj@fftw.org Chris@10: * updated NEWS Chris@10: Chris@10: M ./NEWS +14 Chris@10: Chris@10: Tue Dec 2 18:30:00 EST 2008 stevenj@fftw.org Chris@10: * date fix Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +1 Chris@10: Chris@10: Tue Dec 2 18:29:06 EST 2008 stevenj@fftw.org Chris@10: * updated icc flags -- now prefer -xHost (-xN etc. seem t be obsolete), check for new spelling -ansi-alias, and use -malign-double like we do for gcc Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -2 +14 Chris@10: Chris@10: Tue Dec 2 18:28:03 EST 2008 stevenj@fftw.org Chris@10: * use $ax_cv_c_compiler_vendor rather than $GCC, as the former is more reliable (icc incorrectly self-identifies as gcc on MacOS where we don't use -no-gcc) Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Tue Dec 2 17:55:36 EST 2008 stevenj@fftw.org Chris@10: * don't use -no-gcc for icc on MacOS Chris@10: Chris@10: M ./configure.ac -2 +7 Chris@10: Chris@10: Tue Dec 2 17:34:04 EST 2008 stevenj@fftw.org Chris@10: * document some more bench options Chris@10: Chris@10: M ./tests/README -1 +17 Chris@10: Chris@10: Wed Nov 19 16:55:13 EST 2008 stevenj@fftw.org Chris@10: * make it clearer that --enable-openmp and --enable-threads are mutually exclusive; thanks to Long To for his comments Chris@10: Chris@10: M ./doc/fftw3.texi -6 +10 Chris@10: Chris@10: Mon Nov 17 20:16:28 EST 2008 stevenj@fftw.org Chris@10: * version bump to 3.2.1, use explicit Makefile.am for m4 subdirectory so that tarball does not include random files in there when you do 'make dist' Chris@10: Chris@10: M ./Makefile.am -3 +2 Chris@10: M ./configure.ac -2 +4 Chris@10: A ./m4/Makefile.am Chris@10: Chris@10: Sat Nov 15 21:12:58 EST 2008 stevenj@fftw.org Chris@10: * document behavior of FFTW guru arrays, and in particular the odd behavior of the plan_guru_r2r routine in Fortran (thanks to Alexander Pozdneev for the bug report) Chris@10: Chris@10: M ./doc/fftw3.texi +10 Chris@10: Chris@10: Sat Nov 15 21:03:14 EST 2008 stevenj@fftw.org Chris@10: tagged fftw-3.2 Chris@10: Chris@10: Chris@10: Mon Nov 10 20:21:32 EST 2008 stevenj@fftw.org Chris@10: * version bump to 3.2, updated copyright year Chris@10: Chris@10: M ./COPYRIGHT -2 +2 Chris@10: M ./NEWS -1 +6 Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/configure.c -2 +2 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute-split-dft-c2r.c -2 +2 Chris@10: M ./api/execute-split-dft-r2c.c -2 +2 Chris@10: M ./api/execute-split-dft.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/export-wisdom-to-file.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -2 +2 Chris@10: M ./api/export-wisdom.c -2 +2 Chris@10: M ./api/f77api.c -2 +2 Chris@10: M ./api/f77funcs.h -2 +2 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/flops.c -2 +2 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: M ./api/import-wisdom-from-file.c -2 +2 Chris@10: M ./api/import-wisdom-from-string.c -2 +2 Chris@10: M ./api/import-wisdom.c -2 +2 Chris@10: M ./api/malloc.c -2 +2 Chris@10: M ./api/map-r2r-kind.c -2 +2 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./api/mkprinter-file.c -2 +2 Chris@10: M ./api/mktensor-iodims.h -2 +2 Chris@10: M ./api/mktensor-rowmajor.c -2 +2 Chris@10: M ./api/plan-dft-1d.c -2 +2 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r.c -2 +2 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -2 +2 Chris@10: M ./api/plan-dft-r2c.c -2 +2 Chris@10: M ./api/plan-dft.c -2 +2 Chris@10: M ./api/plan-guru-dft-c2r.h -2 +2 Chris@10: M ./api/plan-guru-dft-r2c.h -2 +2 Chris@10: M ./api/plan-guru-dft.h -2 +2 Chris@10: M ./api/plan-guru-r2r.h -2 +2 Chris@10: M ./api/plan-guru-split-dft-c2r.h -2 +2 Chris@10: M ./api/plan-guru-split-dft-r2c.h -2 +2 Chris@10: M ./api/plan-guru-split-dft.h -2 +2 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: M ./api/plan-many-dft.c -2 +2 Chris@10: M ./api/plan-many-r2r.c -2 +2 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -2 +2 Chris@10: M ./api/plan-r2r-3d.c -2 +2 Chris@10: M ./api/plan-r2r.c -2 +2 Chris@10: M ./api/print-plan.c -2 +2 Chris@10: M ./api/rdft2-pad.c -2 +2 Chris@10: M ./api/the-planner.c -2 +2 Chris@10: M ./api/version.c -2 +2 Chris@10: M ./api/x77.h -2 +2 Chris@10: M ./configure.ac -2 +2 Chris@10: M ./dft/bluestein.c -2 +2 Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/codelet-dft.h -2 +2 Chris@10: M ./dft/conf.c -2 +2 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/ct.h -2 +2 Chris@10: M ./dft/dft.h -2 +2 Chris@10: M ./dft/dftw-direct.c -2 +2 Chris@10: M ./dft/dftw-directsq.c -2 +2 Chris@10: M ./dft/dftw-generic.c -2 +2 Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect-transpose.c -2 +2 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/kdft-dif.c -2 +2 Chris@10: M ./dft/kdft-difsq.c -2 +2 Chris@10: M ./dft/kdft-dit.c -2 +2 Chris@10: M ./dft/kdft.c -2 +2 Chris@10: M ./dft/nop.c -2 +2 Chris@10: M ./dft/plan.c -2 +2 Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/scalar/n.c -2 +2 Chris@10: M ./dft/scalar/n.h -2 +2 Chris@10: M ./dft/scalar/t.c -2 +2 Chris@10: M ./dft/scalar/t.h -2 +2 Chris@10: M ./dft/simd/n1b.c -2 +2 Chris@10: M ./dft/simd/n1b.h -2 +2 Chris@10: M ./dft/simd/n1f.c -2 +2 Chris@10: M ./dft/simd/n1f.h -2 +2 Chris@10: M ./dft/simd/n2b.c -2 +2 Chris@10: M ./dft/simd/n2b.h -2 +2 Chris@10: M ./dft/simd/n2f.c -2 +2 Chris@10: M ./dft/simd/n2f.h -2 +2 Chris@10: M ./dft/simd/n2s.c -2 +2 Chris@10: M ./dft/simd/n2s.h -2 +2 Chris@10: M ./dft/simd/q1b.c -2 +2 Chris@10: M ./dft/simd/q1b.h -2 +2 Chris@10: M ./dft/simd/q1f.c -2 +2 Chris@10: M ./dft/simd/q1f.h -2 +2 Chris@10: M ./dft/simd/t.c -2 +2 Chris@10: M ./dft/simd/t1b.h -2 +2 Chris@10: M ./dft/simd/t1bu.h -2 +2 Chris@10: M ./dft/simd/t1f.h -2 +2 Chris@10: M ./dft/simd/t1fu.h -2 +2 Chris@10: M ./dft/simd/t2b.h -2 +2 Chris@10: M ./dft/simd/t2f.h -2 +2 Chris@10: M ./dft/simd/t3b.h -2 +2 Chris@10: M ./dft/simd/t3f.h -2 +2 Chris@10: M ./dft/simd/ts.c -2 +2 Chris@10: M ./dft/simd/ts.h -2 +2 Chris@10: M ./dft/solve.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./dft/zero.c -2 +2 Chris@10: M ./doc/f77_wisdom.f -2 +2 Chris@10: M ./genfft/algsimp.ml -2 +2 Chris@10: M ./genfft/algsimp.mli -2 +2 Chris@10: M ./genfft/annotate.ml -2 +2 Chris@10: M ./genfft/annotate.mli -2 +2 Chris@10: M ./genfft/assoctable.ml -2 +2 Chris@10: M ./genfft/assoctable.mli -2 +2 Chris@10: M ./genfft/c.ml -2 +2 Chris@10: M ./genfft/c.mli -2 +2 Chris@10: M ./genfft/complex.ml -2 +2 Chris@10: M ./genfft/complex.mli -2 +2 Chris@10: M ./genfft/conv.ml -2 +2 Chris@10: M ./genfft/conv.mli -2 +2 Chris@10: M ./genfft/dag.ml -2 +2 Chris@10: M ./genfft/dag.mli -2 +2 Chris@10: M ./genfft/expr.ml -2 +2 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/fft.ml -2 +2 Chris@10: M ./genfft/fft.mli -2 +2 Chris@10: M ./genfft/gen_hc2c.ml -2 +2 Chris@10: M ./genfft/gen_hc2cdft.ml -2 +2 Chris@10: M ./genfft/gen_hc2cdft_c.ml -2 +2 Chris@10: M ./genfft/gen_hc2hc.ml -2 +2 Chris@10: M ./genfft/gen_mdct.ml -2 +2 Chris@10: M ./genfft/gen_notw.ml -2 +2 Chris@10: M ./genfft/gen_notw_c.ml -2 +2 Chris@10: M ./genfft/gen_r2cb.ml -2 +2 Chris@10: M ./genfft/gen_r2cf.ml -2 +2 Chris@10: M ./genfft/gen_r2r.ml -2 +2 Chris@10: M ./genfft/gen_twiddle.ml -2 +2 Chris@10: M ./genfft/gen_twiddle_c.ml -2 +2 Chris@10: M ./genfft/gen_twidsq.ml -2 +2 Chris@10: M ./genfft/gen_twidsq_c.ml -2 +2 Chris@10: M ./genfft/genutil.ml -2 +2 Chris@10: M ./genfft/littlesimp.ml -2 +2 Chris@10: M ./genfft/littlesimp.mli -2 +2 Chris@10: M ./genfft/magic.ml -2 +2 Chris@10: M ./genfft/monads.ml -2 +2 Chris@10: M ./genfft/number.ml -2 +2 Chris@10: M ./genfft/number.mli -2 +2 Chris@10: M ./genfft/oracle.ml -2 +2 Chris@10: M ./genfft/oracle.mli -2 +2 Chris@10: M ./genfft/schedule.ml -2 +2 Chris@10: M ./genfft/schedule.mli -2 +2 Chris@10: M ./genfft/simd.ml -2 +2 Chris@10: M ./genfft/simd.mli -2 +2 Chris@10: M ./genfft/simdmagic.ml -2 +2 Chris@10: M ./genfft/to_alist.ml -2 +2 Chris@10: M ./genfft/to_alist.mli -2 +2 Chris@10: M ./genfft/trig.ml -2 +2 Chris@10: M ./genfft/trig.mli -2 +2 Chris@10: M ./genfft/twiddle.ml -2 +2 Chris@10: M ./genfft/twiddle.mli -2 +2 Chris@10: M ./genfft/unique.ml -2 +2 Chris@10: M ./genfft/unique.mli -2 +2 Chris@10: M ./genfft/util.ml -2 +2 Chris@10: M ./genfft/util.mli -2 +2 Chris@10: M ./genfft/variable.ml -2 +2 Chris@10: M ./genfft/variable.mli -2 +2 Chris@10: M ./kernel/align.c -2 +2 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/assert.c -2 +2 Chris@10: M ./kernel/awake.c -2 +2 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/cpy1d.c -2 +2 Chris@10: M ./kernel/cpy2d-pair.c -2 +2 Chris@10: M ./kernel/cpy2d.c -2 +2 Chris@10: M ./kernel/ct.c -2 +2 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./kernel/debug.c -2 +2 Chris@10: M ./kernel/extract-reim.c -2 +2 Chris@10: M ./kernel/hash.c -2 +2 Chris@10: M ./kernel/iabs.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/kalloc.c -2 +2 Chris@10: M ./kernel/md5-1.c -2 +2 Chris@10: M ./kernel/md5.c -2 +2 Chris@10: M ./kernel/minmax.c -2 +2 Chris@10: M ./kernel/ops.c -2 +2 Chris@10: M ./kernel/pickdim.c -2 +2 Chris@10: M ./kernel/plan.c -2 +2 Chris@10: M ./kernel/primes.c -2 +2 Chris@10: M ./kernel/print.c -2 +2 Chris@10: M ./kernel/problem.c -2 +2 Chris@10: M ./kernel/rader.c -2 +2 Chris@10: M ./kernel/scan.c -2 +2 Chris@10: M ./kernel/solver.c -2 +2 Chris@10: M ./kernel/solvtab.c -2 +2 Chris@10: M ./kernel/stride.c -2 +2 Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: M ./kernel/tensor1.c -2 +2 Chris@10: M ./kernel/tensor2.c -2 +2 Chris@10: M ./kernel/tensor3.c -2 +2 Chris@10: M ./kernel/tensor4.c -2 +2 Chris@10: M ./kernel/tensor5.c -2 +2 Chris@10: M ./kernel/tensor7.c -2 +2 Chris@10: M ./kernel/tensor8.c -2 +2 Chris@10: M ./kernel/tensor9.c -2 +2 Chris@10: M ./kernel/tile2d.c -2 +2 Chris@10: M ./kernel/timer.c -2 +2 Chris@10: M ./kernel/transpose.c -2 +2 Chris@10: M ./kernel/trig.c -2 +2 Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: M ./libbench2/dotens2.c -2 +2 Chris@10: M ./libbench2/my-getopt.c -2 +2 Chris@10: M ./libbench2/my-getopt.h -2 +2 Chris@10: M ./libbench2/verify-dft.c -2 +2 Chris@10: M ./libbench2/verify-lib.c -2 +2 Chris@10: M ./libbench2/verify-r2r.c -2 +2 Chris@10: M ./libbench2/verify-rdft2.c -2 +2 Chris@10: M ./libbench2/verify.h -2 +2 Chris@10: M ./mpi/any-true.c -2 +2 Chris@10: M ./mpi/api.c -2 +2 Chris@10: M ./mpi/block.c -2 +2 Chris@10: M ./mpi/choose-radix.c -2 +2 Chris@10: M ./mpi/conf.c -2 +2 Chris@10: M ./mpi/dft-problem.c -2 +2 Chris@10: M ./mpi/dft-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/dft-rank-geq2.c -2 +2 Chris@10: M ./mpi/dft-rank1-bigvec.c -2 +2 Chris@10: M ./mpi/dft-rank1.c -2 +2 Chris@10: M ./mpi/dft-serial.c -2 +2 Chris@10: M ./mpi/dft-solve.c -2 +2 Chris@10: M ./mpi/dtensor.c -2 +2 Chris@10: M ./mpi/fftw3-mpi.h -2 +2 Chris@10: M ./mpi/ifftw-mpi.h -2 +2 Chris@10: M ./mpi/mpi-dft.h -2 +2 Chris@10: M ./mpi/mpi-rdft.h -2 +2 Chris@10: M ./mpi/mpi-rdft2.h -2 +2 Chris@10: M ./mpi/mpi-transpose.h -2 +2 Chris@10: M ./mpi/rdft-problem.c -2 +2 Chris@10: M ./mpi/rdft-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/rdft-rank-geq2.c -2 +2 Chris@10: M ./mpi/rdft-rank1-bigvec.c -2 +2 Chris@10: M ./mpi/rdft-serial.c -2 +2 Chris@10: M ./mpi/rdft-solve.c -2 +2 Chris@10: M ./mpi/rdft2-problem.c -2 +2 Chris@10: M ./mpi/rdft2-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/rdft2-rank-geq2.c -2 +2 Chris@10: M ./mpi/rdft2-serial.c -2 +2 Chris@10: M ./mpi/rdft2-solve.c -2 +2 Chris@10: M ./mpi/rearrange.c -2 +2 Chris@10: M ./mpi/testsched.c -2 +2 Chris@10: M ./mpi/transpose-alltoall.c -2 +2 Chris@10: M ./mpi/transpose-pairwise.c -2 +2 Chris@10: M ./mpi/transpose-problem.c -2 +2 Chris@10: M ./mpi/transpose-recurse.c -2 +2 Chris@10: M ./mpi/transpose-solve.c -2 +2 Chris@10: M ./mpi/wisdom-api.c -2 +2 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/codelet-rdft.h -2 +2 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/ct-hc2c-direct.c -2 +2 Chris@10: M ./rdft/ct-hc2c.c -2 +2 Chris@10: M ./rdft/ct-hc2c.h -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: M ./rdft/direct-r2c.c -2 +2 Chris@10: M ./rdft/direct-r2r.c -2 +2 Chris@10: M ./rdft/direct2.c -2 +2 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-direct.c -2 +2 Chris@10: M ./rdft/hc2hc-generic.c -2 +2 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/hc2hc.h -2 +2 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: M ./rdft/khc2c.c -2 +2 Chris@10: M ./rdft/khc2hc.c -2 +2 Chris@10: M ./rdft/kr2c.c -2 +2 Chris@10: M ./rdft/kr2r.c -2 +2 Chris@10: M ./rdft/nop.c -2 +2 Chris@10: M ./rdft/nop2.c -2 +2 Chris@10: M ./rdft/plan.c -2 +2 Chris@10: M ./rdft/plan2.c -2 +2 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rank0-rdft2.c -2 +2 Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -2 +2 Chris@10: M ./rdft/rdft2-rdft.c -2 +2 Chris@10: M ./rdft/rdft2-strides.c -2 +2 Chris@10: M ./rdft/rdft2-tensor-max-index.c -2 +2 Chris@10: M ./rdft/scalar/hb.h -2 +2 Chris@10: M ./rdft/scalar/hc2c.c -2 +2 Chris@10: M ./rdft/scalar/hc2cb.h -2 +2 Chris@10: M ./rdft/scalar/hc2cf.h -2 +2 Chris@10: M ./rdft/scalar/hf.h -2 +2 Chris@10: M ./rdft/scalar/hfb.c -2 +2 Chris@10: M ./rdft/scalar/r2c.c -2 +2 Chris@10: M ./rdft/scalar/r2cb.h -2 +2 Chris@10: M ./rdft/scalar/r2cbIII.h -2 +2 Chris@10: M ./rdft/scalar/r2cf.h -2 +2 Chris@10: M ./rdft/scalar/r2cfII.h -2 +2 Chris@10: M ./rdft/scalar/r2r.c -2 +2 Chris@10: M ./rdft/scalar/r2r.h -2 +2 Chris@10: M ./rdft/simd/hc2cbv.c -2 +2 Chris@10: M ./rdft/simd/hc2cbv.h -2 +2 Chris@10: M ./rdft/simd/hc2cfv.c -2 +2 Chris@10: M ./rdft/simd/hc2cfv.h -2 +2 Chris@10: M ./rdft/solve.c -2 +2 Chris@10: M ./rdft/solve2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: M ./reodft/conf.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft.h -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./simd/altivec.c -2 +2 Chris@10: M ./simd/nonportable/sse.c -2 +2 Chris@10: M ./simd/nonportable/sse2.c -2 +2 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: M ./simd/simd-sse.h -2 +2 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: M ./simd/simd.h -2 +2 Chris@10: M ./simd/sse.c -2 +2 Chris@10: M ./simd/sse2.c -2 +2 Chris@10: M ./simd/taint.c -2 +2 Chris@10: M ./simd/x86-cpuid.h -2 +2 Chris@10: M ./threads/api.c -2 +2 Chris@10: M ./threads/conf.c -2 +2 Chris@10: M ./threads/ct.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/f77api.c -2 +2 Chris@10: M ./threads/f77funcs.h -2 +2 Chris@10: M ./threads/hc2hc.c -2 +2 Chris@10: M ./threads/openmp.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/threads.c -2 +2 Chris@10: M ./threads/threads.h -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./tools/fftw-wisdom-to-conf.1 -4 +4 Chris@10: M ./tools/fftw-wisdom-to-conf.in -2 +2 Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: M ./tools/fftw_wisdom.1.in -4 +4 Chris@10: Chris@10: Wed Nov 5 16:40:31 EST 2008 athena@fftw.org Chris@10: * Store GPLv2 in darcs because automake installs GPLv3 these days. Chris@10: Chris@10: A ./COPYING Chris@10: Chris@10: Thu Oct 30 15:03:41 EDT 2008 athena@fftw.org Chris@10: * stylistic changes, comments Chris@10: Chris@10: M ./threads/threads.c -13 +17 Chris@10: Chris@10: Thu Oct 30 14:40:14 EDT 2008 athena@fftw.org Chris@10: * oops Chris@10: Chris@10: M ./threads/threads.c -27 +28 Chris@10: Chris@10: Thu Oct 30 14:30:08 EDT 2008 athena@fftw.org Chris@10: * simplification of the threading machinery Chris@10: Chris@10: M ./threads/threads.c -12 +7 Chris@10: Chris@10: Thu Oct 30 14:22:40 EDT 2008 athena@fftw.org Chris@10: * typo Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Thu Oct 30 13:42:07 EDT 2008 athena@fftw.org Chris@10: * [SECOND ATTEMPT] do not assume that a semaphore can be freed just because nobody is using it Chris@10: Chris@10: Let S be a semaphore, initially 0. Let thread A execute UP(S); Chris@10: let thread B execute DOWN(S); free(&S); It is unclear whether this Chris@10: code is correct with posix semaphores. The problem is whether UP() Chris@10: uses S after allowing DOWN() to continue; this seems to be the Chris@10: case in the glibc-2.7 implementation, and thus the pattern above Chris@10: seems to be incorrect. Avoid using such a pattern, and introduce Chris@10: a global semaphore for the unavoidable case when nothing else Chris@10: can be depended upon. Chris@10: Chris@10: Chris@10: M ./threads/threads.c -61 +80 Chris@10: Chris@10: Wed Oct 29 20:09:39 EDT 2008 stevenj@fftw.org Chris@10: * updated cpu codes from x86-1.21 Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -5 +7 Chris@10: Chris@10: Wed Oct 29 16:24:16 EDT 2008 athena@fftw.org Chris@10: * Previous change was bogus, need to find another way. Chris@10: Chris@10: M ./threads/threads.c -56 +42 Chris@10: Chris@10: Wed Oct 29 12:22:20 EDT 2008 athena@fftw.org Chris@10: * do not assume that a semaphore can be freed just because nobody is using it Chris@10: Chris@10: Let S be a semaphore, initially 0. Let thread A execute UP(S); Chris@10: let thread B execute DOWN(S); free(&S); It is unclear whether this Chris@10: code is correct with posix semaphores. The problem is whether UP() Chris@10: uses S after allowing DOWN() to continue; this seems to be the Chris@10: case in the glibc-2.7 implementation, and thus the pattern above Chris@10: seems to be incorrect. Avoid using such a pattern, and introduce Chris@10: a global semaphore for the unavoidable case when nothing else Chris@10: can be depended upon. Chris@10: Chris@10: Chris@10: M ./threads/threads.c -42 +56 Chris@10: Chris@10: Mon Oct 27 23:38:02 EDT 2008 stevenj@fftw.org Chris@10: * don't need PROG_AS any more Chris@10: Chris@10: M ./configure.ac -1 Chris@10: R ./m4/amx_prog_as.m4 Chris@10: Chris@10: Sun Oct 26 23:41:11 EDT 2008 stevenj@fftw.org Chris@10: * use AC_CONFIG_MACRO_DIR macro Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Sun Oct 26 10:08:44 EDT 2008 athena@fftw.org Chris@10: * Remove mpi/Makefile from configure.ac Chris@10: Otherwise, the tarball breaks because mpi/ is not in Chris@10: the distribution. Chris@10: Chris@10: M ./configure.ac -2 Chris@10: Chris@10: Sat Oct 25 17:13:50 EDT 2008 stevenj@fftw.org Chris@10: * remove MPI from dist until FFTW 3.3 Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./NEWS +3 Chris@10: M ./doc/fftw3.texi -948 +9 Chris@10: Chris@10: Sat Oct 25 17:12:35 EDT 2008 stevenj@fftw.org Chris@10: * use MPIRUN even for -np 1 Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: Chris@10: Fri Jul 18 17:17:08 EDT 2008 stevenj@fftw.org Chris@10: * use new gcc arch=native flag as fallback Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -5 +5 Chris@10: Chris@10: Sat Oct 25 13:36:40 EDT 2008 athena@fftw.org Chris@10: * Use sem_t to implement mutexes Chris@10: Use sem_t instead of pthread_mutex_t to implement mutexes. Chris@10: It seems like pthread mutexes hang on linux-2.6.22 after several Chris@10: days of tests; the hang does not occur on linux >= 2.6.24 Chris@10: or when we use sem_t instead of pthread_mutex_t. The Chris@10: situation is still quite mysterious but this code seems to Chris@10: work. Chris@10: Chris@10: M ./threads/threads.c -22 +34 Chris@10: Chris@10: Thu Oct 23 13:32:45 EDT 2008 athena@fftw.org Chris@10: * print informative message when pstring is NULL. Chris@10: Chris@10: M ./libbench2/verify.c -1 +2 Chris@10: Chris@10: Sun Oct 19 16:00:07 EDT 2008 athena@fftw.org Chris@10: * Fix incorrect alignment in dftw-generic. Chris@10: Chris@10: Multithreaded dftw-generic is supposed to process only a slice Chris@10: of the array, but we were planning with the alignment of the Chris@10: original array rather than the slice. This led to unaligned Chris@10: accesses in certain obscure situations. Chris@10: Chris@10: Chris@10: M ./NEWS +3 Chris@10: M ./configure.ac -1 +1 Chris@10: M ./dft/dftw-generic.c -1 +2 Chris@10: Chris@10: Mon Aug 18 17:27:26 EDT 2008 Matteo Frigo Chris@10: * Paranoia: do not create OS threads while holding locks. Chris@10: Chris@10: Glibc at least plays silly games such as keeping a global variable Chris@10: that records whether there is more than one thread in the process, and Chris@10: it does not perform atomic operations if the variable says that there Chris@10: is only one thread. Who knows how this interacts with creating Chris@10: threads while holding a lock. Some day some genius will come up with Chris@10: some ``optimization'' that breaks everything. Chris@10: Chris@10: M ./threads/threads.c -7 +13 Chris@10: Chris@10: Wed Aug 6 07:41:46 EDT 2008 athena@fftw.org Chris@10: * Welcome to the quadcore era Chris@10: Chris@10: M ./mkdist.sh -1 +1 Chris@10: Chris@10: Mon Jun 16 16:46:39 EDT 2008 stevenj@fftw.org Chris@10: * backslash is technically not allowed in "echo" arguments; thanks to Debian Bug#486046 for pointing out problem and solution (and Raphael Geissert and Vincent Zweije, in particular) Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: Chris@10: Sun May 4 12:15:24 EDT 2008 stevenj@fftw.org Chris@10: * note problem with test program in gcc 4.1.2-4.2; thanks to Raymond Rogers for reporting it Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +6 Chris@10: Chris@10: Fri May 2 19:21:30 EDT 2008 stevenj@fftw.org Chris@10: * output count of constants along with other statistics Chris@10: Chris@10: M ./genfft/c.ml -3 +8 Chris@10: Chris@10: Sat Apr 19 14:15:03 EDT 2008 athena@fftw.org Chris@10: * Lower priority of unaligned SIMD codelets. Chris@10: List t1[fb]uv_* codelets before the corresponding Chris@10: aligned codelets, since the estimator picks the Chris@10: latter ones in case of a tie and aligned codelets Chris@10: are preferable. Chris@10: Chris@10: In other words, this is a hack. Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -3 +2 Chris@10: Chris@10: Sat Apr 19 08:55:46 EDT 2008 Matteo Frigo Chris@10: * There is no point in using higher radices for unaligned codelets. Chris@10: Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -4 +2 Chris@10: Chris@10: Fri Apr 18 19:01:27 EDT 2008 stevenj@fftw.org Chris@10: * support generating loopless, strideless r2r codelets Chris@10: Chris@10: M ./genfft/gen_r2r.ml -7 +17 Chris@10: Chris@10: Fri Apr 18 19:00:25 EDT 2008 stevenj@fftw.org Chris@10: * added Magic.threemult to use 3+3 complex-multiply variant when possible Chris@10: Chris@10: M ./genfft/complex.ml -2 +13 Chris@10: M ./genfft/magic.ml +4 Chris@10: Chris@10: Thu Apr 10 19:53:31 EDT 2008 stevenj@fftw.org Chris@10: * fix documentation bug - export_wisdom_to_string returns a string that should be deallocated with free, not fftw_free (thanks to Stein Vidar Hagfors Haugan for the bug report) Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Mon Jan 21 01:11:44 EST 2008 stevenj@fftw.org Chris@10: * bsd calls x86_64 "amd64"; thanks to Fernando Herrero Carron for the bug report Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +3 Chris@10: Chris@10: Tue Jan 1 12:29:56 EST 2008 stevenj@fftw.org Chris@10: * fix typo in manual, thanks to Yinon Ehrlich Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Mon Dec 3 13:57:13 EST 2007 stevenj@fftw.org Chris@10: * note problem with gcc 3.4.4 on x86_64, thanks to Uwe Hollerbach for the report Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +4 Chris@10: Chris@10: Tue Nov 13 16:19:22 EST 2007 stevenj@fftw.org Chris@10: * bump shared-lib revision Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Tue Nov 13 16:16:49 EST 2007 stevenj@fftw.org Chris@10: * update NEWS for alpha3 Chris@10: Chris@10: M ./NEWS +23 Chris@10: Chris@10: Tue Oct 2 13:53:04 EDT 2007 stevenj@fftw.org Chris@10: * fixed URL Chris@10: Chris@10: M ./m4/ax_openmp.m4 -1 +1 Chris@10: Chris@10: Mon Sep 17 19:38:29 EDT 2007 stevenj@fftw.org Chris@10: * added missing prototype Chris@10: Chris@10: M ./tests/fftw-bench.c +4 Chris@10: Chris@10: Tue Aug 14 22:35:06 EDT 2007 stevenj@fftw.org Chris@10: * terminology tweak Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Wed Aug 1 18:44:21 EDT 2007 stevenj@fftw.org Chris@10: * check for pathscale compilers (thanks to Julian Cummings) Chris@10: Chris@10: M ./m4/ax_compiler_vendor.m4 -2 +2 Chris@10: Chris@10: Sat Sep 15 18:02:32 EDT 2007 athena@fftw.org Chris@10: * Avoid possible conflict with Windows include files. Chris@10: Chris@10: M ./threads/threads.c +5 Chris@10: Chris@10: Tue Aug 7 21:26:05 EDT 2007 athena@fftw.org Chris@10: * Distribute codlist.c for SIMD codelets in the commercial tarball. Chris@10: Chris@10: M ./commercialize.sh -1 +1 Chris@10: Chris@10: Wed Aug 1 10:33:41 EDT 2007 stevenj@fftw.org Chris@10: * some documentation clarifications, and documented FFTW_WISDOM_ONLY, at the suggestion of Mario Emmenlauer and Phil Dumont Chris@10: Chris@10: M ./api/apiplan.c -3 +3 Chris@10: M ./doc/fftw3.texi -3 +24 Chris@10: Chris@10: Tue Jul 31 16:52:56 EDT 2007 stevenj@fftw.org Chris@10: * bug fix in test program for vrank-3 transpose plans with vl=1 Chris@10: Chris@10: M ./mpi/mpi-bench.c -1 +1 Chris@10: Chris@10: Sun Jul 29 17:02:46 EDT 2007 stevenj@fftw.org Chris@10: * only run mpi checks for --enable-mpi Chris@10: Chris@10: M ./mpi/Makefile.am +4 Chris@10: Chris@10: Sun Jul 29 16:45:30 EDT 2007 stevenj@fftw.org Chris@10: * check for NULL return from spe_context_create in case SPE_MAP_PS not supported Chris@10: Chris@10: M ./cell/cell.c -4 +9 Chris@10: Chris@10: Sun Jul 29 15:56:57 EDT 2007 stevenj@fftw.org Chris@10: * use problem-state pointer to write SPE mailbox with lower latency (makes a significant performance difference for N < 32k), thanks to Jan Wagner for suggestion Chris@10: Chris@10: M ./cell/cell.c +16 Chris@10: Chris@10: Sun Jul 29 14:22:08 EDT 2007 stevenj@fftw.org Chris@10: * port cell code to SDK2.1 (libspe2), since libspe1 API is deprecated and can't be used in code that also uses libspe2 API Chris@10: Chris@10: M ./cell/cell.c -3 +49 Chris@10: M ./cell/copy.c -1 +1 Chris@10: M ./cell/dft-direct-cell.c -1 +1 Chris@10: M ./cell/fftw-cell.h -4 +5 Chris@10: M ./cell/spu/main.spuc -4 +4 Chris@10: M ./cell/transpose.c -1 +1 Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Sun Jul 29 11:46:24 EDT 2007 stevenj@fftw.org Chris@10: * bug fix: ego->W allocated with cell_aligned_malloc, so deallocate with free, not X(ifree0) Chris@10: Chris@10: M ./cell/dft-direct-cell.c -1 +1 Chris@10: Chris@10: Mon Jul 2 15:57:12 EDT 2007 stevenj@fftw.org Chris@10: * removed obsolete reference to CVS id Chris@10: Chris@10: M ./api/fftw3.h -4 Chris@10: Chris@10: Mon May 21 14:25:39 EDT 2007 athena@fftw.org Chris@10: * cycle counter for sun compiler Chris@10: Chris@10: M ./kernel/cycle.h -1 +1 Chris@10: Chris@10: Wed May 9 19:49:11 EDT 2007 stevenj@fftw.org Chris@10: * use __inline instead of inline for AIX routines (__inline is supported by gcc and xlc, whereas apparently "inline" is only supported by xlc if you specify -qlanglvl=stdc99 or similar); thanks to Jeff Haferman for the bug report Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Mon Apr 30 15:37:56 EDT 2007 stevenj@fftw.org Chris@10: * fixed incorrect type prefix (fftw_ vs. X(...)) in mpi/wisdom-api.c; thanks to Eric A. Borisch for the bug report Chris@10: Chris@10: M ./mpi/wisdom-api.c -6 +6 Chris@10: Chris@10: Wed Apr 25 21:21:39 EDT 2007 stevenj@fftw.org Chris@10: * some cleanups in MPI make check Chris@10: Chris@10: M ./mpi/Makefile.am -7 +11 Chris@10: Chris@10: Wed Apr 25 21:19:27 EDT 2007 stevenj@fftw.org Chris@10: * re-enable heuristic in the common case where we are not compiling for Cell Chris@10: Chris@10: M ./dft/rank-geq2.c -1 +3 Chris@10: Chris@10: Tue Apr 24 17:42:43 EDT 2007 athena@fftw.org Chris@10: * Removed duplicate codelet names, was breaking linker. Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -4 +4 Chris@10: Chris@10: Tue Apr 24 11:38:16 EDT 2007 stevenj@fftw.org Chris@10: * added more codelets of sizes 5/10/20/25 to improve speed for round decimal sizes (speed improvements of 10-20%, at cost of 10-30% in library size) Chris@10: Chris@10: M ./dft/scalar/codelets/Makefile.am -3 +6 Chris@10: M ./dft/simd/codelets/Makefile.am -14 +26 Chris@10: M ./rdft/scalar/r2cb/Makefile.am -8 +16 Chris@10: M ./rdft/scalar/r2cf/Makefile.am -8 +16 Chris@10: M ./rdft/simd/codelets/Makefile.am -2 +4 Chris@10: Chris@10: Sat Mar 24 18:40:47 EDT 2007 stevenj@fftw.org Chris@10: * for 1d prime sizes, punt and return serial plan Chris@10: Chris@10: M ./mpi/api.c +4 Chris@10: Chris@10: Sat Mar 24 18:24:55 EDT 2007 stevenj@fftw.org Chris@10: * output reminders of the problem during bench --verify Chris@10: Chris@10: M ./libbench2/verify.c -2 +5 Chris@10: Chris@10: Sat Mar 24 18:10:24 EDT 2007 stevenj@fftw.org Chris@10: * bug fix - missing solver->destroy initializer in rdft2-rdft Chris@10: Chris@10: M ./rdft/rdft2-rdft.c -1 +1 Chris@10: Chris@10: Fri Mar 23 11:12:19 EDT 2007 stevenj@fftw.org Chris@10: * -static, in --enable-debug, doesn't work on MacOS X (according to Daniel Oberhoff) Chris@10: Chris@10: M ./configure.ac -1 Chris@10: Chris@10: Wed Mar 21 22:23:06 EDT 2007 stevenj@fftw.org Chris@10: * fix MPI r2c/c2r to work with howmany > 1 Chris@10: Chris@10: M ./mpi/rdft2-rank-geq2-transposed.c -1 +2 Chris@10: M ./mpi/rdft2-rank-geq2.c -1 +2 Chris@10: M ./mpi/rdft2-serial.c -2 +3 Chris@10: M ./tests/check.pl -1 Chris@10: Chris@10: Wed Mar 21 18:44:41 EDT 2007 stevenj@fftw.org Chris@10: * rm MPI version from TODO Chris@10: Chris@10: M ./TODO -2 Chris@10: Chris@10: Wed Mar 21 18:34:40 EDT 2007 stevenj@fftw.org Chris@10: * added 'make bigcheck' for MPI (no paranoid-check, unfortunately), and properly get MPIRUN from configure Chris@10: Chris@10: M ./configure.ac +2 Chris@10: M ./mpi/Makefile.am -7 +10 Chris@10: Chris@10: Wed Mar 21 18:23:18 EDT 2007 stevenj@fftw.org Chris@10: * bug fix - incorrect local_size returned for 1d bigvec case Chris@10: Chris@10: M ./mpi/api.c -1 +1 Chris@10: Chris@10: Wed Mar 21 03:13:54 EDT 2007 stevenj@fftw.org Chris@10: * hack to specify MPI_TRANSPOSED_IN/OUT via "[" and "]" in libbench2 problem Chris@10: Chris@10: M ./libbench2/bench-user.h +1 Chris@10: M ./libbench2/problem.c +6 Chris@10: M ./mpi/mpi-bench.c +12 Chris@10: M ./tests/check.pl -2 +7 Chris@10: Chris@10: Wed Mar 21 02:58:11 EDT 2007 stevenj@fftw.org Chris@10: * added MPI 'make check', still needs a bit of work Chris@10: Chris@10: M ./mpi/Makefile.am +28 Chris@10: M ./tests/check.pl +31 Chris@10: Chris@10: Wed Mar 21 02:47:10 EDT 2007 stevenj@fftw.org Chris@10: * bug fix in r2r transposed-input case Chris@10: Chris@10: M ./mpi/rdft-problem.c +3 Chris@10: Chris@10: Wed Mar 21 02:46:25 EDT 2007 stevenj@fftw.org Chris@10: * don't output more than 300 erroneous outputs (unless verbose > 2) Chris@10: Chris@10: M ./libbench2/verify-lib.c -2 +3 Chris@10: M ./libbench2/verify-r2r.c -2 +3 Chris@10: Chris@10: Wed Mar 21 01:48:54 EDT 2007 stevenj@fftw.org Chris@10: * fixed bug in transposed-in c2r MPI transforms ... seems to be working, finally Chris@10: Chris@10: M ./mpi/rdft2-problem.c -5 +6 Chris@10: Chris@10: Wed Mar 21 00:41:32 EDT 2007 stevenj@fftw.org Chris@10: * some fixes to MPI r2c/c2r transforms with transposed output/input Chris@10: Chris@10: M ./mpi/rdft2-rank-geq2-transposed.c -11 +17 Chris@10: Chris@10: Wed Mar 21 00:40:25 EDT 2007 stevenj@fftw.org Chris@10: * typos Chris@10: Chris@10: M ./NEWS -4 +4 Chris@10: Chris@10: Tue Mar 20 19:53:02 EDT 2007 stevenj@fftw.org Chris@10: * bug fix for mpi-bench with r2c/c2r: allocate a little bit extra to make sure that padding is allocated Chris@10: Chris@10: M ./libbench2/allocate.c -2 +4 Chris@10: Chris@10: Tue Mar 20 19:19:13 EDT 2007 stevenj@fftw.org Chris@10: * fix typo, thanks to Ernest Turro for the bug report Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Tue Mar 20 01:39:06 EDT 2007 stevenj@fftw.org Chris@10: * spacing tweaks Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Tue Mar 20 00:53:11 EDT 2007 stevenj@fftw.org Chris@10: * Ralf Wildenhues is the one who pointed out that the self-communication could fill in the stalls in the pairwise schedule Chris@10: Chris@10: M ./mpi/testsched.c -3 +2 Chris@10: Chris@10: Tue Mar 20 00:22:25 EDT 2007 stevenj@fftw.org Chris@10: * add TRANSPOSED_OUT/IN support for r2c/c2r, respectively Chris@10: Chris@10: M ./doc/fftw3.texi -4 +1 Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/README -3 Chris@10: M ./mpi/api.c -6 +15 Chris@10: M ./mpi/conf.c -1 +1 Chris@10: M ./mpi/rdft2-problem.c -18 +9 Chris@10: A ./mpi/rdft2-rank-geq2-transposed.c Chris@10: Chris@10: Mon Mar 19 21:45:34 EDT 2007 stevenj@fftw.org Chris@10: * yikes! fixed likely deadlock bug in MPI Chris@10: Chris@10: M ./mpi/transpose-recurse.c -2 +2 Chris@10: Chris@10: Mon Mar 19 21:38:52 EDT 2007 stevenj@fftw.org Chris@10: * comment Chris@10: Chris@10: M ./mpi/transpose-pairwise.c -2 +8 Chris@10: Chris@10: Mon Mar 19 21:30:44 EDT 2007 stevenj@fftw.org Chris@10: * s/alpha1/alpha2/ Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Mon Mar 19 00:39:47 EDT 2007 stevenj@fftw.org Chris@10: * include README in dist tarball Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: Chris@10: Mon Mar 19 00:35:43 EDT 2007 stevenj@fftw.org Chris@10: * added MPI r2c/c2r transforms, some more documentation Chris@10: Chris@10: M ./NEWS -4 +9 Chris@10: M ./doc/fftw3.texi -6 +73 Chris@10: M ./libbench2/bench-user.h +2 Chris@10: M ./libbench2/problem.c -2 +4 Chris@10: M ./mpi/Makefile.am -1 +2 Chris@10: A ./mpi/README Chris@10: M ./mpi/api.c +140 Chris@10: M ./mpi/conf.c +4 Chris@10: M ./mpi/fftw3-mpi.h -88 +116 Chris@10: M ./mpi/mpi-bench.c -2 +102 Chris@10: A ./mpi/mpi-rdft2.h Chris@10: A ./mpi/rdft2-problem.c Chris@10: A ./mpi/rdft2-rank-geq2.c Chris@10: A ./mpi/rdft2-serial.c Chris@10: A ./mpi/rdft2-solve.c Chris@10: Chris@10: Sun Mar 18 23:14:29 EDT 2007 stevenj@fftw.org Chris@10: * set version to 3.2alpha2 Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sun Mar 18 19:12:18 EDT 2007 stevenj@fftw.org Chris@10: * changed --enable-mips_ps to --enable-mips-ps; added Cell section to manual (from README.Cell); many minor updates to manual Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./doc/fftw3.texi -15 +243 Chris@10: Chris@10: Sun Mar 18 15:27:06 EDT 2007 stevenj@fftw.org Chris@10: * whoops, need to sync costs in problem_mpi_rdft Chris@10: Chris@10: M ./mpi/api.c +3 Chris@10: Chris@10: Sun Mar 18 12:44:49 EDT 2007 stevenj@fftw.org Chris@10: * documented guru64 interface Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: M ./doc/fftw3.texi -11 +81 Chris@10: Chris@10: Sun Mar 18 02:57:46 EDT 2007 stevenj@fftw.org Chris@10: * typo Chris@10: Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: Chris@10: Sun Mar 18 02:45:09 EDT 2007 stevenj@fftw.org Chris@10: * bumped copyright year to 2007 Chris@10: Chris@10: M ./COPYRIGHT -2 +2 Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/configure.c -2 +2 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute-split-dft-c2r.c -2 +2 Chris@10: M ./api/execute-split-dft-r2c.c -2 +2 Chris@10: M ./api/execute-split-dft.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/export-wisdom-to-file.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -2 +2 Chris@10: M ./api/export-wisdom.c -2 +2 Chris@10: M ./api/f77api.c -2 +2 Chris@10: M ./api/f77funcs.h -2 +2 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/flops.c -2 +2 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: M ./api/import-wisdom-from-file.c -2 +2 Chris@10: M ./api/import-wisdom-from-string.c -2 +2 Chris@10: M ./api/import-wisdom.c -2 +2 Chris@10: M ./api/malloc.c -2 +2 Chris@10: M ./api/map-r2r-kind.c -2 +2 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./api/mkprinter-file.c -2 +2 Chris@10: M ./api/mktensor-iodims.h -2 +2 Chris@10: M ./api/mktensor-rowmajor.c -2 +2 Chris@10: M ./api/plan-dft-1d.c -2 +2 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r.c -2 +2 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -2 +2 Chris@10: M ./api/plan-dft-r2c.c -2 +2 Chris@10: M ./api/plan-dft.c -2 +2 Chris@10: M ./api/plan-guru-dft-c2r.h -2 +2 Chris@10: M ./api/plan-guru-dft-r2c.h -2 +2 Chris@10: M ./api/plan-guru-dft.h -2 +2 Chris@10: M ./api/plan-guru-r2r.h -2 +2 Chris@10: M ./api/plan-guru-split-dft-c2r.h -2 +2 Chris@10: M ./api/plan-guru-split-dft-r2c.h -2 +2 Chris@10: M ./api/plan-guru-split-dft.h -2 +2 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: M ./api/plan-many-dft.c -2 +2 Chris@10: M ./api/plan-many-r2r.c -2 +2 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -2 +2 Chris@10: M ./api/plan-r2r-3d.c -2 +2 Chris@10: M ./api/plan-r2r.c -2 +2 Chris@10: M ./api/print-plan.c -2 +2 Chris@10: M ./api/rdft2-pad.c -2 +2 Chris@10: M ./api/the-planner.c -2 +2 Chris@10: M ./api/version.c -2 +2 Chris@10: M ./api/x77.h -2 +2 Chris@10: M ./dft/bluestein.c -2 +2 Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/codelet-dft.h -2 +2 Chris@10: M ./dft/conf.c -2 +2 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/ct.h -2 +2 Chris@10: M ./dft/dft.h -2 +2 Chris@10: M ./dft/dftw-direct.c -2 +2 Chris@10: M ./dft/dftw-directsq.c -2 +2 Chris@10: M ./dft/dftw-generic.c -2 +2 Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect-transpose.c -2 +2 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/kdft-dif.c -2 +2 Chris@10: M ./dft/kdft-difsq.c -2 +2 Chris@10: M ./dft/kdft-dit.c -2 +2 Chris@10: M ./dft/kdft.c -2 +2 Chris@10: M ./dft/nop.c -2 +2 Chris@10: M ./dft/plan.c -2 +2 Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/scalar/n.c -2 +2 Chris@10: M ./dft/scalar/n.h -2 +2 Chris@10: M ./dft/scalar/t.c -2 +2 Chris@10: M ./dft/scalar/t.h -2 +2 Chris@10: M ./dft/simd/n1b.c -2 +2 Chris@10: M ./dft/simd/n1b.h -2 +2 Chris@10: M ./dft/simd/n1f.c -2 +2 Chris@10: M ./dft/simd/n1f.h -2 +2 Chris@10: M ./dft/simd/n2b.c -2 +2 Chris@10: M ./dft/simd/n2b.h -2 +2 Chris@10: M ./dft/simd/n2f.c -2 +2 Chris@10: M ./dft/simd/n2f.h -2 +2 Chris@10: M ./dft/simd/n2s.c -2 +2 Chris@10: M ./dft/simd/n2s.h -2 +2 Chris@10: M ./dft/simd/q1b.c -2 +2 Chris@10: M ./dft/simd/q1b.h -2 +2 Chris@10: M ./dft/simd/q1f.c -2 +2 Chris@10: M ./dft/simd/q1f.h -2 +2 Chris@10: M ./dft/simd/t.c -2 +2 Chris@10: M ./dft/simd/t1b.h -2 +2 Chris@10: M ./dft/simd/t1bu.h -2 +2 Chris@10: M ./dft/simd/t1f.h -2 +2 Chris@10: M ./dft/simd/t1fu.h -2 +2 Chris@10: M ./dft/simd/t2b.h -2 +2 Chris@10: M ./dft/simd/t2f.h -2 +2 Chris@10: M ./dft/simd/t3b.h -2 +2 Chris@10: M ./dft/simd/t3f.h -2 +2 Chris@10: M ./dft/simd/ts.c -2 +2 Chris@10: M ./dft/simd/ts.h -2 +2 Chris@10: M ./dft/solve.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./dft/zero.c -2 +2 Chris@10: M ./doc/f77_wisdom.f -2 +2 Chris@10: M ./genfft/algsimp.ml -2 +2 Chris@10: M ./genfft/algsimp.mli -2 +2 Chris@10: M ./genfft/annotate.ml -2 +2 Chris@10: M ./genfft/annotate.mli -2 +2 Chris@10: M ./genfft/assoctable.ml -2 +2 Chris@10: M ./genfft/assoctable.mli -2 +2 Chris@10: M ./genfft/c.ml -2 +2 Chris@10: M ./genfft/c.mli -2 +2 Chris@10: M ./genfft/complex.ml -2 +2 Chris@10: M ./genfft/complex.mli -2 +2 Chris@10: M ./genfft/conv.ml -2 +2 Chris@10: M ./genfft/conv.mli -2 +2 Chris@10: M ./genfft/dag.ml -2 +2 Chris@10: M ./genfft/dag.mli -2 +2 Chris@10: M ./genfft/expr.ml -2 +2 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/fft.ml -2 +2 Chris@10: M ./genfft/fft.mli -2 +2 Chris@10: M ./genfft/gen_hc2c.ml -2 +2 Chris@10: M ./genfft/gen_hc2cdft.ml -2 +2 Chris@10: M ./genfft/gen_hc2cdft_c.ml -2 +2 Chris@10: M ./genfft/gen_hc2hc.ml -2 +2 Chris@10: M ./genfft/gen_mdct.ml -2 +2 Chris@10: M ./genfft/gen_notw.ml -2 +2 Chris@10: M ./genfft/gen_notw_c.ml -2 +2 Chris@10: M ./genfft/gen_r2cb.ml -2 +2 Chris@10: M ./genfft/gen_r2cf.ml -2 +2 Chris@10: M ./genfft/gen_r2r.ml -2 +2 Chris@10: M ./genfft/gen_twiddle.ml -2 +2 Chris@10: M ./genfft/gen_twiddle_c.ml -2 +2 Chris@10: M ./genfft/gen_twidsq.ml -2 +2 Chris@10: M ./genfft/gen_twidsq_c.ml -2 +2 Chris@10: M ./genfft/genutil.ml -2 +2 Chris@10: M ./genfft/littlesimp.ml -2 +2 Chris@10: M ./genfft/littlesimp.mli -2 +2 Chris@10: M ./genfft/magic.ml -2 +2 Chris@10: M ./genfft/monads.ml -2 +2 Chris@10: M ./genfft/number.ml -2 +2 Chris@10: M ./genfft/number.mli -2 +2 Chris@10: M ./genfft/oracle.ml -2 +2 Chris@10: M ./genfft/oracle.mli -2 +2 Chris@10: M ./genfft/schedule.ml -2 +2 Chris@10: M ./genfft/schedule.mli -2 +2 Chris@10: M ./genfft/simd.ml -2 +2 Chris@10: M ./genfft/simd.mli -2 +2 Chris@10: M ./genfft/simdmagic.ml -2 +2 Chris@10: M ./genfft/to_alist.ml -2 +2 Chris@10: M ./genfft/to_alist.mli -2 +2 Chris@10: M ./genfft/trig.ml -2 +2 Chris@10: M ./genfft/trig.mli -2 +2 Chris@10: M ./genfft/twiddle.ml -2 +2 Chris@10: M ./genfft/twiddle.mli -2 +2 Chris@10: M ./genfft/unique.ml -2 +2 Chris@10: M ./genfft/unique.mli -2 +2 Chris@10: M ./genfft/util.ml -2 +2 Chris@10: M ./genfft/util.mli -2 +2 Chris@10: M ./genfft/variable.ml -2 +2 Chris@10: M ./genfft/variable.mli -2 +2 Chris@10: M ./kernel/align.c -2 +2 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/assert.c -2 +2 Chris@10: M ./kernel/awake.c -2 +2 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/cpy1d.c -2 +2 Chris@10: M ./kernel/cpy2d-pair.c -2 +2 Chris@10: M ./kernel/cpy2d.c -2 +2 Chris@10: M ./kernel/ct.c -2 +2 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./kernel/debug.c -2 +2 Chris@10: M ./kernel/extract-reim.c -2 +2 Chris@10: M ./kernel/hash.c -2 +2 Chris@10: M ./kernel/iabs.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/kalloc.c -2 +2 Chris@10: M ./kernel/md5-1.c -2 +2 Chris@10: M ./kernel/md5.c -2 +2 Chris@10: M ./kernel/minmax.c -2 +2 Chris@10: M ./kernel/ops.c -2 +2 Chris@10: M ./kernel/pickdim.c -2 +2 Chris@10: M ./kernel/plan.c -2 +2 Chris@10: M ./kernel/primes.c -2 +2 Chris@10: M ./kernel/print.c -2 +2 Chris@10: M ./kernel/problem.c -2 +2 Chris@10: M ./kernel/rader.c -2 +2 Chris@10: M ./kernel/scan.c -2 +2 Chris@10: M ./kernel/solver.c -2 +2 Chris@10: M ./kernel/solvtab.c -2 +2 Chris@10: M ./kernel/stride.c -2 +2 Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: M ./kernel/tensor1.c -2 +2 Chris@10: M ./kernel/tensor2.c -2 +2 Chris@10: M ./kernel/tensor3.c -2 +2 Chris@10: M ./kernel/tensor4.c -2 +2 Chris@10: M ./kernel/tensor5.c -2 +2 Chris@10: M ./kernel/tensor7.c -2 +2 Chris@10: M ./kernel/tensor8.c -2 +2 Chris@10: M ./kernel/tensor9.c -2 +2 Chris@10: M ./kernel/tile2d.c -2 +2 Chris@10: M ./kernel/timer.c -2 +2 Chris@10: M ./kernel/transpose.c -2 +2 Chris@10: M ./kernel/trig.c -2 +2 Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: M ./libbench2/dotens2.c -2 +2 Chris@10: M ./libbench2/my-getopt.c -2 +2 Chris@10: M ./libbench2/my-getopt.h -2 +2 Chris@10: M ./libbench2/verify-dft.c -2 +2 Chris@10: M ./libbench2/verify-lib.c -2 +2 Chris@10: M ./libbench2/verify-r2r.c -2 +2 Chris@10: M ./libbench2/verify-rdft2.c -2 +2 Chris@10: M ./libbench2/verify.h -2 +2 Chris@10: M ./mpi/any-true.c -2 +2 Chris@10: M ./mpi/api.c -2 +2 Chris@10: M ./mpi/block.c -2 +2 Chris@10: M ./mpi/choose-radix.c -2 +2 Chris@10: M ./mpi/conf.c -2 +2 Chris@10: M ./mpi/dft-problem.c -2 +2 Chris@10: M ./mpi/dft-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/dft-rank-geq2.c -2 +2 Chris@10: M ./mpi/dft-rank1-bigvec.c -2 +2 Chris@10: M ./mpi/dft-rank1.c -2 +2 Chris@10: M ./mpi/dft-serial.c -2 +2 Chris@10: M ./mpi/dft-solve.c -2 +2 Chris@10: M ./mpi/dtensor.c -2 +2 Chris@10: M ./mpi/fftw3-mpi.h -2 +2 Chris@10: M ./mpi/ifftw-mpi.h -2 +2 Chris@10: M ./mpi/mpi-dft.h -2 +2 Chris@10: M ./mpi/mpi-rdft.h -2 +2 Chris@10: M ./mpi/mpi-transpose.h -2 +2 Chris@10: M ./mpi/rdft-problem.c -2 +2 Chris@10: M ./mpi/rdft-rank-geq2-transposed.c -2 +2 Chris@10: M ./mpi/rdft-rank-geq2.c -2 +2 Chris@10: M ./mpi/rdft-rank1-bigvec.c -2 +2 Chris@10: M ./mpi/rdft-serial.c -2 +2 Chris@10: M ./mpi/rdft-solve.c -2 +2 Chris@10: M ./mpi/rearrange.c -2 +2 Chris@10: M ./mpi/testsched.c -2 +2 Chris@10: M ./mpi/transpose-alltoall.c -2 +2 Chris@10: M ./mpi/transpose-pairwise.c -2 +2 Chris@10: M ./mpi/transpose-problem.c -2 +2 Chris@10: M ./mpi/transpose-recurse.c -2 +2 Chris@10: M ./mpi/transpose-solve.c -2 +2 Chris@10: M ./mpi/wisdom-api.c -2 +2 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/codelet-rdft.h -2 +2 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/ct-hc2c-direct.c -2 +2 Chris@10: M ./rdft/ct-hc2c.c -2 +2 Chris@10: M ./rdft/ct-hc2c.h -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: M ./rdft/direct-r2c.c -2 +2 Chris@10: M ./rdft/direct-r2r.c -2 +2 Chris@10: M ./rdft/direct2.c -2 +2 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-direct.c -2 +2 Chris@10: M ./rdft/hc2hc-generic.c -2 +2 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/hc2hc.h -2 +2 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: M ./rdft/khc2c.c -2 +2 Chris@10: M ./rdft/khc2hc.c -2 +2 Chris@10: M ./rdft/kr2c.c -2 +2 Chris@10: M ./rdft/kr2r.c -2 +2 Chris@10: M ./rdft/nop.c -2 +2 Chris@10: M ./rdft/nop2.c -2 +2 Chris@10: M ./rdft/plan.c -2 +2 Chris@10: M ./rdft/plan2.c -2 +2 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rank0-rdft2.c -2 +2 Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -2 +2 Chris@10: M ./rdft/rdft2-rdft.c -2 +2 Chris@10: M ./rdft/rdft2-strides.c -2 +2 Chris@10: M ./rdft/rdft2-tensor-max-index.c -2 +2 Chris@10: M ./rdft/scalar/hb.h -2 +2 Chris@10: M ./rdft/scalar/hc2c.c -2 +2 Chris@10: M ./rdft/scalar/hc2cb.h -2 +2 Chris@10: M ./rdft/scalar/hc2cf.h -2 +2 Chris@10: M ./rdft/scalar/hf.h -2 +2 Chris@10: M ./rdft/scalar/hfb.c -2 +2 Chris@10: M ./rdft/scalar/r2c.c -2 +2 Chris@10: M ./rdft/scalar/r2cb.h -2 +2 Chris@10: M ./rdft/scalar/r2cbIII.h -2 +2 Chris@10: M ./rdft/scalar/r2cf.h -2 +2 Chris@10: M ./rdft/scalar/r2cfII.h -2 +2 Chris@10: M ./rdft/scalar/r2r.c -2 +2 Chris@10: M ./rdft/scalar/r2r.h -2 +2 Chris@10: M ./rdft/simd/hc2cbv.c -2 +2 Chris@10: M ./rdft/simd/hc2cbv.h -2 +2 Chris@10: M ./rdft/simd/hc2cfv.c -2 +2 Chris@10: M ./rdft/simd/hc2cfv.h -2 +2 Chris@10: M ./rdft/solve.c -2 +2 Chris@10: M ./rdft/solve2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: M ./reodft/conf.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft.h -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./simd/altivec.c -2 +2 Chris@10: M ./simd/nonportable/sse.c -2 +2 Chris@10: M ./simd/nonportable/sse2.c -2 +2 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: M ./simd/simd-sse.h -2 +2 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: M ./simd/simd.h -2 +2 Chris@10: M ./simd/sse.c -2 +2 Chris@10: M ./simd/sse2.c -2 +2 Chris@10: M ./simd/taint.c -2 +2 Chris@10: M ./simd/x86-cpuid.h -2 +2 Chris@10: M ./threads/api.c -2 +2 Chris@10: M ./threads/conf.c -2 +2 Chris@10: M ./threads/ct.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/f77api.c -2 +2 Chris@10: M ./threads/f77funcs.h -2 +2 Chris@10: M ./threads/hc2hc.c -2 +2 Chris@10: M ./threads/openmp.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/threads.c -2 +2 Chris@10: M ./threads/threads.h -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./tools/fftw-wisdom-to-conf.1 -4 +4 Chris@10: M ./tools/fftw-wisdom-to-conf.in -2 +2 Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: M ./tools/fftw_wisdom.1.in -4 +4 Chris@10: Chris@10: Sun Mar 18 01:41:40 EDT 2007 stevenj@fftw.org Chris@10: * noted CodeSourcery in AUTHORS Chris@10: Chris@10: M ./AUTHORS -1 +2 Chris@10: Chris@10: Sun Mar 18 01:25:00 EDT 2007 stevenj@fftw.org Chris@10: * more MPI documentation Chris@10: Chris@10: M ./doc/fftw3.texi -2 +61 Chris@10: Chris@10: Sat Mar 17 23:15:04 EDT 2007 stevenj@fftw.org Chris@10: * added MPI multi-dimensional rdft solvers & tests Chris@10: Chris@10: M ./mpi/Makefile.am -2 +4 Chris@10: M ./mpi/api.c +104 Chris@10: M ./mpi/conf.c +5 Chris@10: M ./mpi/fftw3-mpi.h -73 +90 Chris@10: M ./mpi/mpi-bench.c -1 +94 Chris@10: A ./mpi/mpi-rdft.h Chris@10: A ./mpi/rdft-problem.c Chris@10: A ./mpi/rdft-rank-geq2-transposed.c Chris@10: A ./mpi/rdft-rank-geq2.c Chris@10: A ./mpi/rdft-rank1-bigvec.c Chris@10: A ./mpi/rdft-serial.c Chris@10: A ./mpi/rdft-solve.c Chris@10: Chris@10: Sat Mar 17 22:52:00 EDT 2007 stevenj@fftw.org Chris@10: * whoops Chris@10: Chris@10: M ./mpi/dft-rank1-bigvec.c -1 +1 Chris@10: Chris@10: Sat Mar 17 22:43:54 EDT 2007 stevenj@fftw.org Chris@10: * clarification - fftw_mpi_init should be called before importing wisdom Chris@10: Chris@10: M ./doc/fftw3.texi -2 +4 Chris@10: Chris@10: Sat Mar 17 19:49:37 EDT 2007 stevenj@fftw.org Chris@10: * kindx/y/z -> kind0/1/2 for consistency Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./doc/fftw3.texi -9 +9 Chris@10: Chris@10: Sat Mar 17 19:34:02 EDT 2007 stevenj@fftw.org Chris@10: * typo Chris@10: Chris@10: M ./mpi/dft-rank-geq2-transposed.c -1 +1 Chris@10: Chris@10: Sat Mar 17 19:14:16 EDT 2007 stevenj@fftw.org Chris@10: * some refactoring in preparation for mpi-rdft Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/api.c -1 +1 Chris@10: M ./mpi/dft-problem.c -2 +2 Chris@10: M ./mpi/dft-rank1-bigvec.c -53 +5 Chris@10: M ./mpi/dtensor.c -3 +3 Chris@10: M ./mpi/ifftw-mpi.h -1 +29 Chris@10: A ./mpi/rearrange.c Chris@10: Chris@10: Sat Mar 17 18:12:45 EDT 2007 stevenj@fftw.org Chris@10: * documented more stuff for MPI Chris@10: Chris@10: M ./doc/fftw3.texi -2 +112 Chris@10: Chris@10: Sat Mar 17 15:41:23 EDT 2007 stevenj@fftw.org Chris@10: * added NEWS for 3.2alpha Chris@10: Chris@10: M ./NEWS -1 +37 Chris@10: Chris@10: Sat Mar 17 14:50:22 EDT 2007 stevenj@fftw.org Chris@10: * documented MPI transpose routines Chris@10: Chris@10: M ./doc/fftw3.texi -1 +130 Chris@10: Chris@10: Sat Mar 17 08:57:30 EDT 2007 athena@fftw.org Chris@10: * Removed unused variables Chris@10: Chris@10: M ./rdft/direct-r2c.c -3 Chris@10: Chris@10: Fri Mar 16 14:47:10 EDT 2007 athena@fftw.org Chris@10: * Preparing for interim release of Cell code. Chris@10: Chris@10: M ./AUTHORS +2 Chris@10: M ./configure.ac -1 +1 Chris@10: M ./doc/fftw3.texi +3 Chris@10: Chris@10: Thu Feb 8 12:23:43 EST 2007 athena@fftw.org Chris@10: * Added README.Cell Chris@10: Chris@10: A ./README.Cell Chris@10: M ./Makefile.am -1 +2 Chris@10: M ./README.Cell +103 Chris@10: Chris@10: Sat Mar 10 19:17:40 EST 2007 athena@fftw.org Chris@10: * Synchronized with main branch Chris@10: Chris@10: M ./simd/simd.h +10 Chris@10: Chris@10: Mon Jan 22 17:43:56 EST 2007 athena@fftw.org Chris@10: * Adapted vrecur heuristic to Cell. Chris@10: Chris@10: M ./cell/dft-direct-cell.c -1 +28 Chris@10: M ./dft/dft.h +4 Chris@10: Chris@10: Sun Jan 21 19:09:33 EST 2007 athena@fftw.org Chris@10: * synchronize with main branch. Chris@10: Chris@10: Chris@10: Sun Jan 21 14:42:00 EST 2007 athena@fftw.org Chris@10: * synchronized with main branch Chris@10: Chris@10: Chris@10: Thu Jan 18 20:29:22 EST 2007 athena@fftw.org Chris@10: * Increased MAX_N to 32K/sizeof(R). Chris@10: Chris@10: M ./cell/dft-direct-cell.c -23 +35 Chris@10: M ./cell/fftw-cell.h -3 +3 Chris@10: M ./cell/plans-double.c -9 +1033 Chris@10: M ./cell/plans-single.c -13 +525 Chris@10: M ./cell/spu/dft.spuc -2 +8 Chris@10: Chris@10: Thu Jan 18 13:43:51 EST 2007 Matteo Frigo Chris@10: * Added pointer to solver->destroy which is used in the Cell branch. Chris@10: Chris@10: M ./mpi/dft-rank-geq2-transposed.c -1 +1 Chris@10: M ./mpi/dft-rank1-bigvec.c -1 +1 Chris@10: M ./mpi/dft-rank1.c -1 +1 Chris@10: M ./mpi/transpose-recurse.c -1 +1 Chris@10: Chris@10: Thu Jan 18 12:09:26 EST 2007 athena@fftw.org Chris@10: * Updated copyright notices Chris@10: Chris@10: R ./mpi/transpose-radix2.c Chris@10: M ./cell/cell.c -2 +1 Chris@10: M ./cell/conf.c -2 +1 Chris@10: M ./cell/copy.c -2 +1 Chris@10: M ./cell/dft-direct-cell.c -2 +1 Chris@10: M ./cell/fftw-cell.h +19 Chris@10: M ./cell/spu/alloc.spuc +19 Chris@10: M ./cell/spu/copy.spuc +19 Chris@10: M ./cell/spu/dft.spuc +18 Chris@10: M ./cell/spu/dma.spuc +19 Chris@10: M ./cell/spu/execute.spuc +19 Chris@10: M ./cell/spu/fftw-spu.h +19 Chris@10: M ./cell/spu/main.spuc +19 Chris@10: M ./cell/spu/planner.spuc +19 Chris@10: M ./cell/spu/spu-double.h +19 Chris@10: M ./cell/spu/spu-single.h +19 Chris@10: M ./cell/spu/transpose.spuc +19 Chris@10: M ./cell/spufftw-embed.S +19 Chris@10: M ./cell/transpose.c -2 +1 Chris@10: Chris@10: Fri Jan 12 12:54:43 EST 2007 athena@fftw.org Chris@10: * Use mfc_read_tag_status_all() instead of spu_mfcstat(2), since the former seems to be standardized. Chris@10: Chris@10: M ./cell/spu/dma.spuc -1 +1 Chris@10: Chris@10: Thu Jan 11 14:55:08 EST 2007 athena@fftw.org Chris@10: * Silence some int/INT warnings. Chris@10: Chris@10: M ./cell/dft-direct-cell.c -21 +24 Chris@10: Chris@10: Wed Jan 10 18:19:53 EST 2007 athena@fftw.org Chris@10: * Note incompatibility of --enable-cell with --enable-threads Chris@10: Chris@10: M ./configure.ac +3 Chris@10: Chris@10: Wed Jan 10 17:57:10 EST 2007 athena@fftw.org Chris@10: * forgot to add file Chris@10: Chris@10: A ./cell/spufftw-embed.S Chris@10: Chris@10: Wed Jan 10 17:45:16 EST 2007 athena@fftw.org Chris@10: * 64-bit cleanup Chris@10: Chris@10: M ./cell/Makefile.am -4 +3 Chris@10: M ./cell/cell.c -7 +21 Chris@10: M ./cell/copy.c -2 +2 Chris@10: M ./cell/dft-direct-cell.c -11 +28 Chris@10: M ./cell/fftw-cell.h -8 +10 Chris@10: M ./cell/spu/dft.spuc -3 +3 Chris@10: M ./cell/spu/dma.spuc -7 +9 Chris@10: M ./cell/spu/fftw-spu.h -2 +2 Chris@10: M ./cell/transpose.c -1 +1 Chris@10: M ./configure.ac +3 Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +1 Chris@10: Chris@10: Wed Jan 10 13:47:20 EST 2007 athena@fftw.org Chris@10: * Use -mcpu=cell where appropriate. Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 +1 Chris@10: Chris@10: Tue Dec 26 21:35:59 EST 2006 athena@fftw.org Chris@10: * synchronized with main Chris@10: Chris@10: M ./dft/buffered.c -14 +1 Chris@10: Chris@10: Sun Dec 24 20:58:25 EST 2006 athena@fftw.org Chris@10: * synchronized with main branch Chris@10: Chris@10: M ./Makefile.am -12 +14 Chris@10: Chris@10: Sun Dec 24 13:47:37 EST 2006 athena@fftw.org Chris@10: * synchronized with main branch, updated to new sdk. Chris@10: Chris@10: M! ./cell/spu/Makefile.am -4 +3 Chris@10: M! ./cell/spu/dma.spuc -1 Chris@10: M! ./cell/spu/execute.spuc -2 +2 Chris@10: M! ./cell/spu/fftw-spu.h -29 +29 Chris@10: Chris@10: Thu Dec 21 17:17:41 EST 2006 athena@fftw.org Chris@10: * removed obsolete file Chris@10: Chris@10: R ./rdft/rdft2-radix2.c Chris@10: Chris@10: Tue Dec 19 15:17:20 EST 2006 athena@fftw.org Chris@10: * synchronized with main branch Chris@10: Chris@10: R ./rdft/direct.c Chris@10: M ./rdft/direct-r2c.c -1 +1 Chris@10: M ./rdft/direct-r2r.c -1 +1 Chris@10: Chris@10: Tue Dec 19 11:27:38 EST 2006 athena@fftw.org Chris@10: * Synchronized with main branch Chris@10: Chris@10: M! ./rdft/ct-hc2c.c -1 +1 Chris@10: M! ./rdft/direct2.c -13 +1 Chris@10: Chris@10: Fri Dec 15 16:04:31 EST 2006 athena@fftw.org Chris@10: * resolved conflict with main branch Chris@10: Chris@10: M ./tests/fftw-bench.c -3 +2 Chris@10: Chris@10: Fri Dec 8 14:43:50 EST 2006 athena@fftw.org Chris@10: * Fixes for compilation in subdirectories Chris@10: Chris@10: M ./cell/spu/Makefile.am -5 +3 Chris@10: Chris@10: Fri Dec 8 12:46:00 EST 2006 athena@fftw.org Chris@10: * Silence warning Chris@10: Chris@10: M ./rdft/rank0.c -2 Chris@10: Chris@10: Fri Dec 8 12:24:19 EST 2006 athena@fftw.org Chris@10: * silence warning Chris@10: Chris@10: M ./rdft/rank0.c -1 +1 Chris@10: Chris@10: Thu Dec 7 15:18:17 EST 2006 athena@fftw.org Chris@10: * Commented a particularly obscure piece of code. Chris@10: Chris@10: M ./cell/spu/dma.spuc -1 +14 Chris@10: Chris@10: Thu Dec 7 11:53:29 EST 2006 athena@fftw.org Chris@10: * Reorganized, clarified conditions for applicability of the DFT solver. Chris@10: Chris@10: M ./cell/dft-direct-cell.c -193 +178 Chris@10: Chris@10: Mon Dec 4 21:33:49 EST 2006 athena@fftw.org Chris@10: * Minor changes Chris@10: Chris@10: M ./cell/dft-direct-cell.c -1 +6 Chris@10: M ./cell/spu/dft.spuc -1 +7 Chris@10: Chris@10: Mon Dec 4 17:43:28 EST 2006 athena@fftw.org Chris@10: * Clarified comment Chris@10: Chris@10: M ./cell/dft-direct-cell.c -2 +2 Chris@10: Chris@10: Mon Dec 4 16:49:06 EST 2006 athena@fftw.org Chris@10: * Less incorrect conditions for fitting into local store. Chris@10: Chris@10: M ./cell/dft-direct-cell.c -5 +12 Chris@10: Chris@10: Mon Dec 4 16:08:24 EST 2006 athena@fftw.org Chris@10: * Implemented DECDIF+TRANSPOSE on Cell Chris@10: Chris@10: R ./dft/ctsq.c Chris@10: M ./cell/dft-direct-cell.c -93 +142 Chris@10: M ./cell/fftw-cell.h -8 +3 Chris@10: M ./cell/spu/dft.spuc -12 +21 Chris@10: M ./dft/indirect-transpose.c -1 +2 Chris@10: Chris@10: Fri Dec 1 17:42:55 EST 2006 athena@fftw.org Chris@10: * relaxed conditions of applicability of SPE Chris@10: Chris@10: M ./cell/dft-direct-cell.c -31 +42 Chris@10: Chris@10: Fri Dec 1 16:28:10 EST 2006 athena@fftw.org Chris@10: * tweaks Chris@10: Chris@10: M ./cell/dft-direct-cell.c -3 +2 Chris@10: Chris@10: Fri Dec 1 14:35:17 EST 2006 athena@fftw.org Chris@10: * Implemented Cell opcounts Chris@10: Chris@10: M ./cell/dft-direct-cell.c -4 +55 Chris@10: M ./cell/spu/execute.spuc +1 Chris@10: Chris@10: Fri Dec 1 13:38:44 EST 2006 athena@fftw.org Chris@10: * minor cleanup Chris@10: Chris@10: M ./cell/dft-direct-cell.c -9 +4 Chris@10: Chris@10: Fri Dec 1 11:16:52 EST 2006 athena@fftw.org Chris@10: * use [c0 s0 c1 s1] format for Cell twiddle factors, rather than [c0 c1 s0 s1]. This makes life easier and there is no speed penalty on Cell (unlike Altivec). Chris@10: Chris@10: M ./cell/copy.c -6 +6 Chris@10: M ./cell/dft-direct-cell.c -46 +15 Chris@10: M ./cell/spu/planner.spuc -2 +3 Chris@10: M ./cell/spu/spu-single.h -4 +4 Chris@10: Chris@10: Wed Nov 29 18:02:54 EST 2006 athena@fftw.org Chris@10: * Implemented SPE-accelerated copies Chris@10: Chris@10: A ./cell/copy.c Chris@10: M ./cell/Makefile.am -2 +2 Chris@10: M ./cell/copy.c +81 Chris@10: M ./cell/fftw-cell.h +14 Chris@10: M ./cell/spu/copy.spuc -69 +31 Chris@10: M ./cell/spu/dma.spuc -2 +78 Chris@10: M ./cell/spu/fftw-spu.h -4 +6 Chris@10: M ./cell/spu/main.spuc +4 Chris@10: M ./cell/spu/transpose.spuc -14 +6 Chris@10: M ./cell/transpose.c -4 +4 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./rdft/rank0.c +52 Chris@10: Chris@10: Wed Nov 29 12:11:08 EST 2006 athena@fftw.org Chris@10: * allow SPEs to compute vrank-0 problems. Chris@10: Chris@10: M ./cell/dft-direct-cell.c -10 +8 Chris@10: M ./cell/spu/dma.spuc -4 +4 Chris@10: Chris@10: Tue Nov 28 18:03:07 EST 2006 athena@fftw.org Chris@10: * eliminated DMA lists Chris@10: Chris@10: M ./cell/spu/dft.spuc -4 +4 Chris@10: M ./cell/spu/dma.spuc -119 +62 Chris@10: M ./cell/spu/fftw-spu.h -3 Chris@10: M ./cell/spu/main.spuc -2 +2 Chris@10: M ./cell/spu/transpose.spuc -6 +6 Chris@10: Chris@10: Tue Nov 28 14:22:05 EST 2006 athena@fftw.org Chris@10: * Conservatively force all dimensions to be 0 (mod VL) in cell, since otherwise it is too hard to get all cases right. Chris@10: Chris@10: M ./cell/dft-direct-cell.c -11 +10 Chris@10: M ./cell/spu/dft.spuc -1 +2 Chris@10: Chris@10: Tue Nov 28 12:39:01 EST 2006 athena@fftw.org Chris@10: * Check alignment of strides when transposing on Cell. Chris@10: Chris@10: M ./rdft/rank0.c -1 +10 Chris@10: Chris@10: Tue Nov 28 12:19:09 EST 2006 athena@fftw.org Chris@10: * consistent usage of FFT_SIGN Chris@10: Chris@10: M ./cell/dft-direct-cell.c -3 +3 Chris@10: M ./cell/spu/dft.spuc -2 +2 Chris@10: Chris@10: Tue Nov 28 11:35:38 EST 2006 athena@fftw.org Chris@10: * clever transposition algorithm without buffering Chris@10: Chris@10: M ./cell/spu/copy.spuc -178 +59 Chris@10: M ./cell/spu/dft.spuc -52 +8 Chris@10: M ./cell/spu/dma.spuc -54 +97 Chris@10: M ./cell/spu/fftw-spu.h -13 +3 Chris@10: M ./cell/spu/transpose.spuc -13 +11 Chris@10: Chris@10: Mon Nov 27 14:08:28 EST 2006 athena@fftw.org Chris@10: * Fixed tracking of dependencies Chris@10: Chris@10: M ./cell/spu/Makefile.am -1 +3 Chris@10: Chris@10: Mon Nov 27 14:03:53 EST 2006 athena@fftw.org Chris@10: * implemented 1D transforms, various tweaks Chris@10: Chris@10: M ./cell/cell.c -4 +14 Chris@10: M ./cell/conf.c +1 Chris@10: M ./cell/dft-direct-cell.c -15 +196 Chris@10: M ./cell/fftw-cell.h -2 +13 Chris@10: M ./cell/spu/Makefile.am +1 Chris@10: M ./cell/spu/alloc.spuc +1 Chris@10: M ./cell/spu/copy.spuc -34 +151 Chris@10: M ./cell/spu/dft.spuc -2 +37 Chris@10: M ./cell/spu/dma.spuc -13 +9 Chris@10: M ./cell/spu/fftw-spu.h -5 +9 Chris@10: M ./cell/spu/main.spuc -6 +3 Chris@10: Chris@10: Wed Nov 22 15:43:36 EST 2006 athena@fftw.org Chris@10: * no need to poll mailbox on spu side Chris@10: Chris@10: M ./cell/spu/main.spuc -1 Chris@10: Chris@10: Wed Nov 22 14:08:24 EST 2006 athena@fftw.org Chris@10: * increased maximum size handled by spe Chris@10: Chris@10: M ./cell/fftw-cell.h -2 +2 Chris@10: M ./cell/plans-double.c -2 +1027 Chris@10: M ./cell/plans-single.c -2 +514 Chris@10: M ./cell/spu/alloc.spuc -1 +1 Chris@10: M ./cell/spu/planner.spuc -1 +1 Chris@10: Chris@10: Tue Nov 21 16:23:17 EST 2006 athena@fftw.org Chris@10: * allow vrank<=2 problems in SPEs to avoid the vecloop overhead (grrr...) Chris@10: Chris@10: M ./cell/dft-direct-cell.c -40 +114 Chris@10: M ./cell/fftw-cell.h -1 +15 Chris@10: M ./cell/spu/dft.spuc -27 +37 Chris@10: M ./dft/rank-geq2.c +4 Chris@10: Chris@10: Mon Nov 20 14:41:45 EST 2006 athena@fftw.org Chris@10: * added emacs mode Chris@10: Chris@10: M ./cell/spu/copy.spuc +1 Chris@10: Chris@10: Mon Nov 20 09:34:12 EST 2006 athena@fftw.org Chris@10: * revised transpose, cleanup Chris@10: Chris@10: A ./cell/spu/transpose.spuc Chris@10: A ./cell/spu/dft.spuc Chris@10: M ./cell/spu/Makefile.am -1 +1 Chris@10: M ./cell/spu/dft.spuc +165 Chris@10: M ./cell/spu/fftw-spu.h +5 Chris@10: M ./cell/spu/main.spuc -224 +2 Chris@10: M ./cell/spu/transpose.spuc +70 Chris@10: Chris@10: Sun Nov 19 20:20:23 EST 2006 athena@fftw.org Chris@10: * added file Chris@10: Chris@10: A ./cell/spu/Makefile.am Chris@10: Chris@10: Sun Nov 19 20:18:35 EST 2006 athena@fftw.org Chris@10: * removed file Chris@10: Chris@10: R ./cell/spu/generate.sh Chris@10: Chris@10: Sun Nov 19 20:15:38 EST 2006 athena@fftw.org Chris@10: * better automake integration Chris@10: Chris@10: M! ./cell/Makefile.am -9 +4 Chris@10: R! ./cell/spu/alloc.c Chris@10: A! ./cell/spu/alloc.spuc Chris@10: R! ./cell/spu/copy.c Chris@10: A! ./cell/spu/copy.spuc Chris@10: R! ./cell/spu/dma.c Chris@10: A! ./cell/spu/dma.spuc Chris@10: R! ./cell/spu/execute.c Chris@10: A! ./cell/spu/execute.spuc Chris@10: R! ./cell/spu/main.c Chris@10: A! ./cell/spu/main.spuc Chris@10: R! ./cell/spu/planner.c Chris@10: A! ./cell/spu/planner.spuc Chris@10: R! ./cell/spu/spu_n2fv_10.c Chris@10: R! ./cell/spu/spu_n2fv_11.c Chris@10: R! ./cell/spu/spu_n2fv_12.c Chris@10: R! ./cell/spu/spu_n2fv_13.c Chris@10: R! ./cell/spu/spu_n2fv_14.c Chris@10: R! ./cell/spu/spu_n2fv_15.c Chris@10: R! ./cell/spu/spu_n2fv_16.c Chris@10: R! ./cell/spu/spu_n2fv_2.c Chris@10: R! ./cell/spu/spu_n2fv_3.c Chris@10: R! ./cell/spu/spu_n2fv_32.c Chris@10: R! ./cell/spu/spu_n2fv_4.c Chris@10: R! ./cell/spu/spu_n2fv_5.c Chris@10: R! ./cell/spu/spu_n2fv_6.c Chris@10: R! ./cell/spu/spu_n2fv_7.c Chris@10: R! ./cell/spu/spu_n2fv_8.c Chris@10: R! ./cell/spu/spu_n2fv_9.c Chris@10: R! ./cell/spu/spu_t1fv_10.c Chris@10: R! ./cell/spu/spu_t1fv_12.c Chris@10: R! ./cell/spu/spu_t1fv_15.c Chris@10: R! ./cell/spu/spu_t1fv_16.c Chris@10: R! ./cell/spu/spu_t1fv_2.c Chris@10: R! ./cell/spu/spu_t1fv_3.c Chris@10: R! ./cell/spu/spu_t1fv_32.c Chris@10: R! ./cell/spu/spu_t1fv_4.c Chris@10: R! ./cell/spu/spu_t1fv_5.c Chris@10: R! ./cell/spu/spu_t1fv_6.c Chris@10: R! ./cell/spu/spu_t1fv_7.c Chris@10: R! ./cell/spu/spu_t1fv_8.c Chris@10: R! ./cell/spu/spu_t1fv_9.c Chris@10: M! ./dft/indirect-transpose.c -1 +1 Chris@10: M! ./rdft/direct2.c -1 +13 Chris@10: Chris@10: Sat Nov 18 20:14:29 EST 2006 athena@fftw.org Chris@10: * changed algorithm for computing chunk size Chris@10: Chris@10: M ./cell/spu/main.c -32 +36 Chris@10: Chris@10: Sat Nov 18 19:18:11 EST 2006 athena@fftw.org Chris@10: * implemented transpose, various fixes. Chris@10: Chris@10: M! ./cell/Makefile.am -2 +2 Chris@10: M! ./cell/cell.c -2 +10 Chris@10: M! ./cell/conf.c -1 +1 Chris@10: M! ./cell/dft-direct-cell.c -13 +29 Chris@10: M! ./cell/fftw-cell.h -3 +13 Chris@10: A! ./cell/spu/copy.c Chris@10: A! ./cell/spu/dma.c Chris@10: M! ./cell/spu/fftw-spu.h +27 Chris@10: M! ./cell/spu/main.c -169 +83 Chris@10: A! ./cell/transpose.c Chris@10: M! ./configure.ac -2 +5 Chris@10: M! ./kernel/ifftw.h +5 Chris@10: M! ./rdft/rank0.c +24 Chris@10: M! ./tests/fftw-bench.c +4 Chris@10: Chris@10: Thu Nov 16 16:33:50 EST 2006 athena@fftw.org Chris@10: * Added explicit destructor to all solvers to help with the cell port. Chris@10: Chris@10: M! ./Makefile.am -2 +6 Chris@10: M! ./cell/cell.c -4 +22 Chris@10: M! ./cell/dft-direct-cell.c -1 +8 Chris@10: M! ./cell/fftw-cell.h -1 +3 Chris@10: M! ./configure.ac +1 Chris@10: M! ./dft/bluestein.c -1 +1 Chris@10: M! ./dft/buffered.c -1 +1 Chris@10: M! ./dft/ct.c -1 +1 Chris@10: A! ./dft/ctsq.c Chris@10: M! ./dft/direct.c -1 +1 Chris@10: M! ./dft/generic.c -1 +1 Chris@10: M! ./dft/indirect-transpose.c -1 +1 Chris@10: M! ./dft/indirect.c -1 +1 Chris@10: M! ./dft/nop.c -1 +1 Chris@10: M! ./dft/rader.c -1 +1 Chris@10: M! ./dft/rank-geq2.c -1 +1 Chris@10: M! ./dft/vrank-geq1.c -1 +1 Chris@10: M! ./kernel/ifftw.h +1 Chris@10: M! ./kernel/solver.c -1 +4 Chris@10: M! ./mpi/dft-rank-geq2.c -1 +1 Chris@10: M! ./mpi/dft-serial.c -1 +1 Chris@10: M! ./mpi/transpose-alltoall.c -1 +1 Chris@10: M! ./mpi/transpose-pairwise.c -1 +1 Chris@10: A! ./mpi/transpose-radix2.c Chris@10: M! ./rdft/buffered.c -1 +1 Chris@10: M! ./rdft/buffered2.c -1 +1 Chris@10: M! ./rdft/dft-r2hc.c -1 +1 Chris@10: M! ./rdft/dht-r2hc.c -1 +1 Chris@10: M! ./rdft/dht-rader.c -1 +1 Chris@10: A! ./rdft/direct.c Chris@10: M! ./rdft/direct2.c -1 +1 Chris@10: M! ./rdft/generic.c -1 +1 Chris@10: M! ./rdft/hc2hc.c -1 +1 Chris@10: M! ./rdft/indirect.c -1 +1 Chris@10: M! ./rdft/nop.c -1 +1 Chris@10: M! ./rdft/nop2.c -1 +1 Chris@10: M! ./rdft/rank-geq2-rdft2.c -1 +1 Chris@10: M! ./rdft/rank-geq2.c -1 +1 Chris@10: M! ./rdft/rank0-rdft2.c -1 +1 Chris@10: M! ./rdft/rank0.c -1 +1 Chris@10: M! ./rdft/rdft-dht.c -1 +1 Chris@10: A! ./rdft/rdft2-radix2.c Chris@10: M! ./rdft/vrank-geq1-rdft2.c -1 +1 Chris@10: M! ./rdft/vrank-geq1.c -1 +1 Chris@10: M! ./rdft/vrank3-transpose.c -1 +1 Chris@10: M! ./reodft/redft00e-r2hc-pad.c -1 +1 Chris@10: M! ./reodft/redft00e-r2hc.c -1 +1 Chris@10: M! ./reodft/reodft00e-splitradix.c -1 +1 Chris@10: M! ./reodft/reodft010e-r2hc.c -1 +1 Chris@10: M! ./reodft/reodft11e-r2hc-odd.c -1 +1 Chris@10: M! ./reodft/reodft11e-r2hc.c -1 +1 Chris@10: M! ./reodft/reodft11e-radix2.c -1 +1 Chris@10: M! ./reodft/rodft00e-r2hc-pad.c -1 +1 Chris@10: M! ./reodft/rodft00e-r2hc.c -1 +1 Chris@10: M! ./tests/fftw-bench.c -1 Chris@10: M! ./threads/ct.c -1 +1 Chris@10: M! ./threads/dft-vrank-geq1.c -1 +1 Chris@10: M! ./threads/hc2hc.c -1 +1 Chris@10: M! ./threads/rdft-vrank-geq1.c -1 +1 Chris@10: M! ./threads/vrank-geq1-rdft2.c -1 +1 Chris@10: Chris@10: Thu Nov 16 15:22:15 EST 2006 athena@fftw.org Chris@10: * consistent use of #if vs. #ifdef Chris@10: Chris@10: M ./api/configure.c +2 Chris@10: M ./cell/cell.c -1 +1 Chris@10: M ./cell/conf.c -1 +1 Chris@10: M ./cell/dft-direct-cell.c -1 +1 Chris@10: M ./cell/plans-double.c -1 +1 Chris@10: M ./cell/plans-single.c -1 +1 Chris@10: Chris@10: Thu Nov 16 15:15:34 EST 2006 athena@fftw.org Chris@10: * Additional Cell double codelets, better automake integration Chris@10: Chris@10: A! ./cell/spu/spu_n2fv_11.c Chris@10: A! ./cell/spu/spu_n2fv_13.c Chris@10: A! ./cell/spu/spu_n2fv_15.c Chris@10: A! ./cell/spu/spu_n2fv_3.c Chris@10: A! ./cell/spu/spu_n2fv_5.c Chris@10: A! ./cell/spu/spu_n2fv_7.c Chris@10: A! ./cell/spu/spu_n2fv_9.c Chris@10: M! ./Makefile.am -7 +5 Chris@10: M! ./cell/Makefile.am -2 +7 Chris@10: M! ./cell/cell.c -11 +7 Chris@10: M! ./cell/conf.c +26 Chris@10: M! ./cell/dft-direct-cell.c -9 +14 Chris@10: M! ./cell/fftw-cell.h -2 Chris@10: M! ./cell/plans-double.c -196 +201 Chris@10: M! ./cell/plans-single.c -3 +9 Chris@10: M! ./cell/spu/execute.c +10 Chris@10: M! ./cell/spu/fftw-spu.h -8 +22 Chris@10: M! ./cell/spu/generate.sh -1 +6 Chris@10: M! ./cell/spu/main.c -2 +1 Chris@10: M! ./cell/spu/planner.c +3 Chris@10: M! ./cell/spu/spu_n2fv_11.c +118 Chris@10: M! ./cell/spu/spu_n2fv_13.c +155 Chris@10: M! ./cell/spu/spu_n2fv_15.c +132 Chris@10: M! ./cell/spu/spu_n2fv_3.c +37 Chris@10: M! ./cell/spu/spu_n2fv_5.c +52 Chris@10: M! ./cell/spu/spu_n2fv_7.c +68 Chris@10: M! ./cell/spu/spu_n2fv_9.c +103 Chris@10: M! ./configure.ac +6 Chris@10: M! ./dft/buffered.c -1 +14 Chris@10: M! ./kernel/align.c -1 +1 Chris@10: M! ./kernel/ifftw.h -2 +2 Chris@10: M! ./simd/simd.h -4 Chris@10: M! ./simd/taint.c -1 +1 Chris@10: Chris@10: Thu Nov 16 12:43:34 EST 2006 athena@fftw.org Chris@10: * Use dma lists. Chris@10: Chris@10: M ./cell/spu/main.c -21 +86 Chris@10: Chris@10: Thu Nov 16 11:03:46 EST 2006 athena@fftw.org Chris@10: * converted to automake Chris@10: Chris@10: R ./cell/spu/Makefile.in Chris@10: M ./cell/Makefile.am +1 Chris@10: Chris@10: Wed Nov 15 18:00:12 EST 2006 athena@fftw.org Chris@10: * Initial port to Cell Broadband Engine. Chris@10: Chris@10: A! ./cell/ Chris@10: A! ./cell/spu/ Chris@10: A! ./cell/Makefile.am Chris@10: A! ./cell/fftw-cell.h Chris@10: A! ./cell/plans-single.c Chris@10: A! ./cell/spu/Makefile.in Chris@10: A! ./cell/spu/execute.c Chris@10: A! ./cell/spu/fftw-spu.h Chris@10: A! ./cell/spu/generate.sh Chris@10: A! ./cell/spu/main.c Chris@10: A! ./cell/spu/planner.c Chris@10: A! ./cell/spu/spu-single.h Chris@10: A! ./cell/spu/spu_n2fv_10.c Chris@10: A! ./cell/spu/spu_n2fv_12.c Chris@10: A! ./cell/spu/spu_n2fv_14.c Chris@10: A! ./cell/spu/spu_n2fv_16.c Chris@10: A! ./cell/spu/spu_n2fv_2.c Chris@10: A! ./cell/spu/spu_n2fv_32.c Chris@10: A! ./cell/spu/spu_n2fv_4.c Chris@10: A! ./cell/spu/spu_n2fv_6.c Chris@10: A! ./cell/spu/spu_n2fv_8.c Chris@10: A! ./cell/spu/spu_t1fv_10.c Chris@10: A! ./cell/spu/spu_t1fv_12.c Chris@10: A! ./cell/spu/spu_t1fv_15.c Chris@10: A! ./cell/spu/spu_t1fv_16.c Chris@10: A! ./cell/spu/spu_t1fv_2.c Chris@10: A! ./cell/spu/spu_t1fv_3.c Chris@10: A! ./cell/spu/spu_t1fv_32.c Chris@10: A! ./cell/spu/spu_t1fv_4.c Chris@10: A! ./cell/spu/spu_t1fv_5.c Chris@10: A! ./cell/spu/spu_t1fv_6.c Chris@10: A! ./cell/spu/spu_t1fv_7.c Chris@10: A! ./cell/spu/spu_t1fv_8.c Chris@10: A! ./cell/spu/spu_t1fv_9.c Chris@10: A! ./cell/cell.c Chris@10: A! ./cell/conf.c Chris@10: A! ./cell/dft-direct-cell.c Chris@10: A! ./cell/spu/spu-double.h Chris@10: A! ./cell/plans-double.c Chris@10: A! ./cell/spu/alloc.c Chris@10: M! ./Makefile.am -14 +14 Chris@10: M! ./api/configure.c +1 Chris@10: M! ./cell/Makefile.am +22 Chris@10: M! ./cell/cell.c +90 Chris@10: M! ./cell/conf.c +12 Chris@10: M! ./cell/dft-direct-cell.c +302 Chris@10: M! ./cell/fftw-cell.h +64 Chris@10: M! ./cell/plans-double.c +2056 Chris@10: M! ./cell/plans-single.c +1032 Chris@10: M! ./cell/spu/Makefile.in +35 Chris@10: M! ./cell/spu/alloc.c +24 Chris@10: M! ./cell/spu/execute.c +68 Chris@10: M! ./cell/spu/fftw-spu.h +86 Chris@10: M! ./cell/spu/generate.sh +16 Chris@10: M! ./cell/spu/main.c +281 Chris@10: M! ./cell/spu/planner.c +187 Chris@10: M! ./cell/spu/spu-double.h +105 Chris@10: M! ./cell/spu/spu-single.h +120 Chris@10: M! ./cell/spu/spu_n2fv_10.c +100 Chris@10: M! ./cell/spu/spu_n2fv_12.c +109 Chris@10: M! ./cell/spu/spu_n2fv_14.c +142 Chris@10: M! ./cell/spu/spu_n2fv_16.c +146 Chris@10: M! ./cell/spu/spu_n2fv_2.c +33 Chris@10: M! ./cell/spu/spu_n2fv_32.c +313 Chris@10: M! ./cell/spu/spu_n2fv_4.c +44 Chris@10: M! ./cell/spu/spu_n2fv_6.c +62 Chris@10: M! ./cell/spu/spu_n2fv_8.c +75 Chris@10: M! ./cell/spu/spu_t1fv_10.c +104 Chris@10: M! ./cell/spu/spu_t1fv_12.c +114 Chris@10: M! ./cell/spu/spu_t1fv_15.c +162 Chris@10: M! ./cell/spu/spu_t1fv_16.c +153 Chris@10: M! ./cell/spu/spu_t1fv_2.c +32 Chris@10: M! ./cell/spu/spu_t1fv_3.c +41 Chris@10: M! ./cell/spu/spu_t1fv_32.c +329 Chris@10: M! ./cell/spu/spu_t1fv_4.c +44 Chris@10: M! ./cell/spu/spu_t1fv_5.c +60 Chris@10: M! ./cell/spu/spu_t1fv_6.c +63 Chris@10: M! ./cell/spu/spu_t1fv_7.c +81 Chris@10: M! ./cell/spu/spu_t1fv_8.c +77 Chris@10: M! ./cell/spu/spu_t1fv_9.c +120 Chris@10: M! ./configure.ac +2 Chris@10: M! ./kernel/ifftw.h -17 +21 Chris@10: M! ./libbench2/util.c -1 +1 Chris@10: M! ./tests/fftw-bench.c +1 Chris@10: Chris@10: Wed Mar 14 10:19:53 EDT 2007 athena@fftw.org Chris@10: * Remove Codesourcery contributions from commercial tarball. Chris@10: Chris@10: M ./commercialize.sh +7 Chris@10: Chris@10: Wed Mar 14 08:59:18 EDT 2007 athena@fftw.org Chris@10: * Added FFTW_WISDOM_ONLY, at the request of Phil Dumont. Chris@10: Chris@10: M ./api/apiplan.c -19 +29 Chris@10: M ./api/fftw3.h +1 Chris@10: M ./tests/fftw-bench.c +1 Chris@10: Chris@10: Tue Mar 13 00:32:05 EDT 2007 stevenj@fftw.org Chris@10: * fixed potential MPI deadlock if timer misbehaves Chris@10: Chris@10: M ./kernel/timer.c -3 +3 Chris@10: Chris@10: Mon Mar 12 23:31:52 EDT 2007 stevenj@fftw.org Chris@10: * more work on MPI documentation Chris@10: Chris@10: M ./doc/fftw3.texi -31 +309 Chris@10: M ./mpi/api.c -3 +4 Chris@10: M ./mpi/fftw3-mpi.h -19 +19 Chris@10: Chris@10: Tue Feb 27 13:48:43 EST 2007 stevenj@fftw.org Chris@10: * index Chris@10: Chris@10: M ./doc/fftw3.texi +1 Chris@10: Chris@10: Tue Feb 27 13:46:45 EST 2007 stevenj@fftw.org Chris@10: * rename "new-data execute" to "new-array execute", since of course you do not need a new array to have new data Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +2 Chris@10: M ./doc/fftw3.texi -12 +12 Chris@10: Chris@10: Tue Feb 27 13:43:55 EST 2007 stevenj@fftw.org Chris@10: * consistency with manual (guru execute -> new-data execute) Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Tue Feb 27 13:42:24 EST 2007 stevenj@fftw.org Chris@10: * texinfo fixes; renamed "guru execute" section to "new-data execute", since previously it seemed to lead to endless confusion with the guru planner API Chris@10: Chris@10: M ./doc/fftw3.texi -60 +83 Chris@10: Chris@10: Mon Feb 26 18:57:11 EST 2007 stevenj@fftw.org Chris@10: * consistently use n0/n1/.. everywhere instead of nx/ny/... (for consistency with d-dimensional case n[0], n[1], ...) ... first start at MPI documentation Chris@10: Chris@10: M ./api/fftw3.h -10 +10 Chris@10: M ./doc/fftw3.texi -138 +374 Chris@10: M ./mpi/fftw3-mpi.h -27 +27 Chris@10: Chris@10: Sat Mar 10 18:48:05 EST 2007 athena@fftw.org Chris@10: * Changed C++-style comment into K&R Chris@10: Chris@10: M ./kernel/cycle.h -1 +1 Chris@10: Chris@10: Sat Mar 10 18:47:12 EST 2007 athena@fftw.org Chris@10: * Forgot to add file Chris@10: Chris@10: A ./simd/simd-mips_ps.h Chris@10: Chris@10: Sat Mar 10 18:44:39 EST 2007 athena@fftw.org Chris@10: * Note removal of K7 support. Chris@10: Chris@10: M ./AUTHORS -1 +4 Chris@10: Chris@10: Sat Mar 10 18:41:52 EST 2007 athena@fftw.org Chris@10: * Updated manual for MIPS PS Chris@10: Chris@10: M ./doc/fftw3.texi -9 +12 Chris@10: Chris@10: Sat Mar 10 18:37:07 EST 2007 athena@fftw.org Chris@10: * Adopted MIPS_PS patches from Codesourcery. Chris@10: Chris@10: A ./simd/mips_ps.c Chris@10: A ./simd/mips_ps.h Chris@10: M ./configure.ac +23 Chris@10: M ./kernel/cycle.h +43 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/timer.c -1 +3 Chris@10: M ./simd/Makefile.am -1 +1 Chris@10: M ./simd/mips_ps.c +45 Chris@10: M ./simd/mips_ps.h +273 Chris@10: M ./simd/simd.h +4 Chris@10: Chris@10: Sun Feb 25 11:34:51 EST 2007 athena@fftw.org Chris@10: * Incorrect initialization of win32 semaphores Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Tue Jan 30 11:43:09 EST 2007 stevenj@fftw.org Chris@10: * win32 fixes (I think, still untested) Chris@10: Chris@10: M ./threads/threads.c -9 +10 Chris@10: Chris@10: Fri Jan 19 17:31:47 EST 2007 stevenj@fftw.org Chris@10: * message-size heuristic in tranpose-recurse Chris@10: Chris@10: M ./mpi/transpose-recurse.c -3 +15 Chris@10: Chris@10: Tue Jan 30 08:53:55 EST 2007 athena@fftw.org Chris@10: * Threading layer for Win32, completely untested. Chris@10: Chris@10: M ./threads/threads.c +69 Chris@10: Chris@10: Mon Jan 29 14:26:30 EST 2007 athena@fftw.org Chris@10: * Check for EINTR after sem_wait(), as suggested by Chip Salzenberg. Chris@10: Chris@10: M ./threads/threads.c -1 +11 Chris@10: Chris@10: Mon Jan 22 13:58:23 EST 2007 athena@fftw.org Chris@10: * Force vector recursion by means of a separate function pointer. I need this for Cell. Chris@10: Chris@10: M ./dft/ct.c -3 +7 Chris@10: M ./dft/ct.h -2 +8 Chris@10: M ./dft/dft.h -15 Chris@10: M ./dft/dftw-direct.c -2 +2 Chris@10: M ./dft/dftw-directsq.c -2 +3 Chris@10: M ./dft/dftw-generic.c -2 +2 Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -4 Chris@10: M ./threads/ct.c -1 +3 Chris@10: M ./threads/threads.h -1 +3 Chris@10: Chris@10: Mon Jan 22 09:28:35 EST 2007 athena@fftw.org Chris@10: * Merge multiplications by twiddle with multiplications by i for faster r2c transforms. Chris@10: Chris@10: M ./genfft/algsimp.ml +4 Chris@10: M ./genfft/simd.ml +4 Chris@10: M ./simd/simd-altivec.h +24 Chris@10: M ./simd/simd-sse.h +18 Chris@10: M ./simd/simd-sse2.h +18 Chris@10: Chris@10: Sun Jan 21 19:02:44 EST 2007 athena@fftw.org Chris@10: * Disabled vector recursion, too messy. Chris@10: Chris@10: M ./dft/dft.h -2 +13 Chris@10: Chris@10: Sun Jan 21 14:23:35 EST 2007 athena@fftw.org Chris@10: * Changed heuristics for vector recursion. Chris@10: Chris@10: As in fftw-3.1, NO_VRECURSE disables vector recursion. As an Chris@10: exception, however, vector recursion is allowed when the predicate Chris@10: VRECURSE_ANYWAYP is true. We need some form of vector recursion to Chris@10: obtain decent plans on Cell, and this solution captures the common Chris@10: cases without increasing planning time too much. Chris@10: Chris@10: Chris@10: M ./api/fftw3.h -1 +1 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./dft/ct.c -10 +6 Chris@10: M ./dft/dft.h +4 Chris@10: M ./dft/vrank-geq1.c -6 +3 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner.c -1 +1 Chris@10: M ./rdft/ct-hc2c.c -9 +4 Chris@10: M ./rdft/hc2hc.c -9 +4 Chris@10: M ./rdft/vrank-geq1.c -7 Chris@10: Chris@10: Sun Dec 17 22:31:17 EST 2006 athena@fftw.org Chris@10: * fixed hc2c for vector-recursion branch Chris@10: Chris@10: M ./rdft/ct-hc2c.c -2 +7 Chris@10: Chris@10: Tue Dec 5 12:52:36 EST 2006 athena@fftw.org Chris@10: * switch to default vector recursion Chris@10: Chris@10: M ./api/fftw3.h -1 +1 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./dft/ct.c -4 +6 Chris@10: M ./dft/vrank-geq1.c +7 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner.c -1 +1 Chris@10: M ./rdft/hc2hc.c -2 +7 Chris@10: M ./rdft/vrank-geq1.c -2 +9 Chris@10: Chris@10: Sat Jan 20 23:37:33 EST 2007 athena@fftw.org Chris@10: * Smarter algorithm for selection of nbuf. Chris@10: Chris@10: M ./kernel/buffered.c -8 +6 Chris@10: Chris@10: Sat Jan 20 22:15:33 EST 2007 athena@fftw.org Chris@10: * Increased buffer sizes according to Moore's law. Chris@10: Chris@10: M ./kernel/buffered.c -3 +5 Chris@10: Chris@10: Fri Jan 19 16:02:00 EST 2007 stevenj@fftw.org Chris@10: * fix another MPI synchronization bug -- several more places where cost_hook must be called to synchronized process timings (sigh) Chris@10: Chris@10: M ./api/flops.c -3 +3 Chris@10: M ./kernel/ifftw.h -4 +5 Chris@10: M ./kernel/planner.c -10 +6 Chris@10: M ./kernel/timer.c -4 +15 Chris@10: M ./mpi/api.c -3 +1 Chris@10: Chris@10: Fri Jan 19 12:08:07 EST 2007 athena@fftw.org Chris@10: * Set havewisdom=0 when calling forget_wisdom() in the test program. Chris@10: Chris@10: M ./tests/fftw-bench.c -1 +3 Chris@10: Chris@10: Fri Jan 19 10:29:56 EST 2007 stevenj@fftw.org Chris@10: * remove redundant check Chris@10: Chris@10: M ./kernel/planner.c -4 +1 Chris@10: Chris@10: Thu Jan 18 22:37:59 EST 2007 stevenj@fftw.org Chris@10: * fixed potential (unlikely) bug in wisdom import (triggered when importing impatient wisdom after creating more patient plans, but apparently only for nonstandard configure.c configurations) Chris@10: Chris@10: M ./kernel/planner.c -1 +5 Chris@10: Chris@10: Thu Jan 18 21:50:14 EST 2007 stevenj@fftw.org Chris@10: * added functions to gather/broadcast wisdom for MPI Chris@10: Chris@10: M ./configure.ac -3 +8 Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/api.c -3 +3 Chris@10: M ./mpi/fftw3-mpi.h -1 +5 Chris@10: M ./mpi/mpi-bench.c -19 +40 Chris@10: A ./mpi/wisdom-api.c Chris@10: M ./tests/bench.c +10 Chris@10: M ./tests/fftw-bench.c -3 +3 Chris@10: M ./tests/fftw-bench.h +2 Chris@10: Chris@10: Thu Jan 11 18:33:17 EST 2007 stevenj@fftw.org Chris@10: * whoops, another int/INT bug Chris@10: Chris@10: M ./mpi/transpose-pairwise.c -1 +1 Chris@10: Chris@10: Thu Jan 11 17:42:24 EST 2007 stevenj@fftw.org Chris@10: * whoops, fixed bug in transpose-recurse for r != m Chris@10: Chris@10: M ./mpi/transpose-recurse.c -10 +10 Chris@10: Chris@10: Thu Jan 11 17:25:36 EST 2007 stevenj@fftw.org Chris@10: * canonicalize mpi-transposed flags by setting TRANSPOSED_IN/OUT where possible Chris@10: Chris@10: M ./mpi/transpose-problem.c -1 +8 Chris@10: Chris@10: Thu Jan 11 17:16:24 EST 2007 stevenj@fftw.org Chris@10: * replace transpose-radix2 with much more general transpose-recurse solver Chris@10: Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h +5 Chris@10: M ./kernel/tensor2.c -21 Chris@10: A ./kernel/tensor3.c Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/conf.c -1 +1 Chris@10: M ./mpi/dft-rank-geq2-transposed.c -1 Chris@10: M ./mpi/mpi-bench.c +1 Chris@10: M ./mpi/mpi-transpose.h -1 +1 Chris@10: R ./mpi/transpose-radix2.c Chris@10: A ./mpi/transpose-recurse.c Chris@10: Chris@10: Wed Jan 10 20:23:48 EST 2007 stevenj@fftw.org Chris@10: * rename transpose-inplace to transpose-pairwise, as the algorithm is not restricted to inplace operation Chris@10: Chris@10: ./mpi/transpose-inplace.c -> ./mpi/transpose-pairwise.c Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/conf.c -1 +1 Chris@10: M ./mpi/mpi-transpose.h -1 +1 Chris@10: M ./mpi/testsched.c -1 +1 Chris@10: M ./mpi/transpose-pairwise.c -9 +9 Chris@10: Chris@10: Wed Jan 10 14:39:08 EST 2007 stevenj@fftw.org Chris@10: * whoops, some int/INT bugs Chris@10: Chris@10: M ./mpi/api.c -2 +2 Chris@10: M ./mpi/block.c -3 +3 Chris@10: M ./mpi/ifftw-mpi.h -5 +5 Chris@10: Chris@10: Tue Jan 9 18:50:07 EST 2007 stevenj@fftw.org Chris@10: * fix FAQ Makefile for vpath builds Chris@10: Chris@10: M ./doc/FAQ/Makefile.am -4 +7 Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Tue Jan 9 20:22:11 EST 2007 athena@fftw.org Chris@10: * Missing ``static'' keyword. Chris@10: Chris@10: M ./rdft/direct-r2c.c -2 +2 Chris@10: Chris@10: Tue Jan 9 20:13:18 EST 2007 athena@fftw.org Chris@10: * Minor cleanup. Chris@10: Chris@10: M ./dft/direct.c -27 +16 Chris@10: Chris@10: Tue Jan 9 00:04:03 EST 2007 stevenj@fftw.org Chris@10: * interleave twiddle mults with DFTs (should we use dftw?) Chris@10: Chris@10: M ./mpi/dft-rank1.c -41 +44 Chris@10: Chris@10: Mon Jan 8 18:35:41 EST 2007 stevenj@fftw.org Chris@10: * simplified (and somewhat sped up) dft-rank1 by exploiting dft-rank1-bigvec Chris@10: Chris@10: M ./mpi/dft-rank-geq2-transposed.c -1 +1 Chris@10: M ./mpi/dft-rank-geq2.c -1 +1 Chris@10: M ./mpi/dft-rank1-bigvec.c -2 +4 Chris@10: M ./mpi/dft-rank1.c -175 +181 Chris@10: M ./mpi/transpose-alltoall.c -6 +3 Chris@10: M ./mpi/transpose-inplace.c -6 +3 Chris@10: M ./mpi/transpose-radix2.c -7 +2 Chris@10: Chris@10: Sun Jan 7 00:31:31 EST 2007 stevenj@fftw.org Chris@10: * rearranged TRANSPOSED format, numerous speedups Chris@10: Chris@10: Split the TRANSPOSED and non-TRANSPOSED rank-geq2 solvers, and changed Chris@10: the DFT TRANSPOSED format to be more like fftw2 (both globally and Chris@10: locally transposed). In general, more emphasis on arranging the data Chris@10: contiguously for the DFTs, and more flexibility in intermediate Chris@10: transposed formats. Also disable NO_SLOW when planning transposes, Chris@10: since otherwise non-square in-place transposes gratuitously put the Chris@10: planner in SLOW mode. Chris@10: Chris@10: Currently, dft-rank1-bigvec has 5 variants (or 10, if DESTROY_INPUT). Chris@10: It looks like only 2 of these are commonly used, so I should probably Chris@10: add some UGLY tags once I do more benchmarking. Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/api.c -14 +34 Chris@10: M ./mpi/conf.c +1 Chris@10: M ./mpi/dft-problem.c -2 +12 Chris@10: A ./mpi/dft-rank-geq2-transposed.c Chris@10: M ./mpi/dft-rank-geq2.c -131 +60 Chris@10: M ./mpi/dft-rank1-bigvec.c -77 +123 Chris@10: M ./mpi/dft-rank1.c -2 +2 Chris@10: M ./mpi/ifftw-mpi.h -2 +3 Chris@10: M ./mpi/mpi-bench.c +13 Chris@10: M ./mpi/mpi-dft.h -1 +2 Chris@10: M ./mpi/transpose-alltoall.c -32 +39 Chris@10: M ./mpi/transpose-inplace.c -53 +69 Chris@10: M ./mpi/transpose-radix2.c -37 +43 Chris@10: Chris@10: Thu Jan 4 19:13:17 EST 2007 stevenj@fftw.org Chris@10: * add bench_cost_postprocess to prevent deadlocks in mpi-bench Chris@10: Chris@10: M ./libbench2/Makefile.am -5 +6 Chris@10: A ./libbench2/bench-cost-postprocess.c Chris@10: M ./libbench2/bench-user.h +1 Chris@10: M ./libbench2/speed.c -2 +2 Chris@10: M ./mpi/mpi-bench.c +7 Chris@10: Chris@10: Thu Jan 4 16:46:29 EST 2007 stevenj@fftw.org Chris@10: * whoops Chris@10: Chris@10: M ./m4/acx_pthread.m4 -1 +1 Chris@10: Chris@10: Wed Jan 3 14:23:42 EST 2007 stevenj@fftw.org Chris@10: * pass proper pointer types as arguments, so that ACX_PTHREAD still works with C++ and -Werror (thanks to Ewald Arnold for the suggestion) Chris@10: Chris@10: M ./m4/acx_pthread.m4 -3 +6 Chris@10: Chris@10: Mon Jan 1 19:30:43 EST 2007 athena@fftw.org Chris@10: * Renamed [io]vs => [io]vs_by_nbuf, which is more appropriate and would have saved me 30mins debugging. Chris@10: Chris@10: M ./dft/buffered.c -6 +6 Chris@10: Chris@10: Mon Jan 1 18:52:38 EST 2007 stevenj@fftw.org Chris@10: * add --with-g77-wrappers option & always include g77 wrappers on GNU systems and/or with gfortran Chris@10: Chris@10: Upcoming GNU/Linux distros will most likely switch to configuring FFTW Chris@10: with gfortran by default, since g77 isn't even included with recent gcc Chris@10: versions. However, we still want to include g77-compatible wrappers in Chris@10: this case (two underscores) in addition to gfortran wrappers (one Chris@10: underscore) lest we silently break binary compatibility and provoke Chris@10: lots of annoying emails. Chris@10: Chris@10: M ./api/x77.h +6 Chris@10: M ./configure.ac +11 Chris@10: Chris@10: Mon Jan 1 16:48:36 EST 2007 stevenj@fftw.org Chris@10: * use AC_HELP_STRING for --disable-fortran Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Mon Jan 1 15:56:12 EST 2007 stevenj@fftw.org Chris@10: * terminology Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Sat Dec 30 16:18:35 EST 2006 athena@fftw.org Chris@10: * Free buffers before calling cldrest. Chris@10: Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./rdft/buffered.c -4 +4 Chris@10: Chris@10: Fri Dec 29 10:52:15 EST 2006 athena@fftw.org Chris@10: * Removed obsolete code. Chris@10: Chris@10: M ./simd/simd-sse.h -22 +9 Chris@10: Chris@10: Thu Dec 28 21:37:48 EST 2006 athena@fftw.org Chris@10: * Attempt to work-around old gcc bugs in a more efficient fashion that does not lose performance on newer gcc's. Chris@10: Chris@10: M ./simd/simd-sse.h -13 +24 Chris@10: Chris@10: Thu Dec 28 16:10:33 EST 2006 athena@fftw.org Chris@10: * Make sure that the speed() input is zero even in paranoid mode. Chris@10: Chris@10: M ./libbench2/speed.c +6 Chris@10: Chris@10: Thu Dec 28 11:41:46 EST 2006 athena@fftw.org Chris@10: * cld0 and cldm problems must be tainted because they are used in a v-loop. Chris@10: Chris@10: M ./rdft/ct-hc2c-direct.c -2 +5 Chris@10: M ./rdft/hc2hc-direct.c -2 +3 Chris@10: Chris@10: Wed Dec 27 17:17:45 EST 2006 athena@fftw.org Chris@10: * Run paranoid-check in patient mode. Chris@10: Chris@10: M ./tests/Makefile.am -5 +5 Chris@10: Chris@10: Wed Dec 27 10:51:42 EST 2006 athena@fftw.org Chris@10: * Fixed incorrect initialization to zero. Chris@10: Chris@10: M ./rdft/problem2.c -1 +1 Chris@10: Chris@10: Wed Dec 27 09:33:02 EST 2006 athena@fftw.org Chris@10: * Fixed wrong TAINT() Chris@10: Chris@10: M ./rdft/buffered2.c -1 +1 Chris@10: Chris@10: Tue Dec 26 22:50:28 EST 2006 athena@fftw.org Chris@10: * Grrrrr... Chris@10: Chris@10: M ./libbench2/verify-rdft2.c -1 +1 Chris@10: Chris@10: Tue Dec 26 22:48:44 EST 2006 athena@fftw.org Chris@10: * Give up trying to verify rdft2 when vrank=-infinity. Chris@10: Chris@10: M ./libbench2/verify-rdft2.c +3 Chris@10: Chris@10: Tue Dec 26 22:31:38 EST 2006 athena@fftw.org Chris@10: * typo Chris@10: Chris@10: M ./dft/direct.c -1 +2 Chris@10: Chris@10: Tue Dec 26 21:54:53 EST 2006 athena@fftw.org Chris@10: * Correctly verify rdft2 when vrank = -infinity. Chris@10: Chris@10: M ./libbench2/verify-rdft2.c -1 +1 Chris@10: Chris@10: Tue Dec 26 21:25:02 EST 2006 athena@fftw.org Chris@10: * rdft/buffered2.c now generates rdft2 subproblems, not rdft. Chris@10: Chris@10: The old rdft2->rdft reduction is now in rdft/rdft2-rdft.c Chris@10: and still does way too much. Chris@10: Chris@10: M ./TODO -3 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/buffered.c -1 +1 Chris@10: M ./rdft/buffered2.c -190 +200 Chris@10: M ./rdft/conf.c +1 Chris@10: M ./rdft/problem2.c -5 +6 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -5 +5 Chris@10: A ./rdft/rdft2-rdft.c Chris@10: M ./rdft/rdft2-strides.c -5 +5 Chris@10: M ./rdft/vrank-geq1-rdft2.c -11 +10 Chris@10: Chris@10: Tue Dec 26 14:03:27 EST 2006 athena@fftw.org Chris@10: * Buffer the input in hc2r problems, as opposed to the output. Chris@10: Chris@10: M ./rdft/buffered.c -34 +104 Chris@10: M ./rdft/buffered2.c -1 Chris@10: Chris@10: Tue Dec 26 10:02:59 EST 2006 athena@fftw.org Chris@10: * streamlined buffered solvers Chris@10: Chris@10: M ./dft/buffered.c -66 +14 Chris@10: M ./kernel/buffered.c -4 +25 Chris@10: M ./kernel/ifftw.h -1 +4 Chris@10: M ./kernel/primes.c +11 Chris@10: M ./rdft/buffered.c -63 +13 Chris@10: M ./rdft/buffered2.c -54 +7 Chris@10: Chris@10: Mon Dec 25 16:08:22 EST 2006 athena@fftw.org Chris@10: * c++ compatibility Chris@10: Chris@10: M ./libbench2/verify-r2r.c -2 +2 Chris@10: Chris@10: Sun Dec 24 20:27:23 EST 2006 athena@fftw.org Chris@10: * Gratuitous renaming of directories and files since the old naming was becoming too inconsistent for my taste. Chris@10: Chris@10: ./dft/codelets -> ./dft/scalar Chris@10: ./dft/scalar/standard -> ./dft/scalar/codelets Chris@10: ./rdft/codelets -> ./rdft/scalar Chris@10: M ./Makefile.am -14 +15 Chris@10: M ./configure.ac -6 +6 Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/scalar/Makefile.am -3 +3 Chris@10: M ./dft/scalar/codelets/Makefile.am -3 +3 Chris@10: M ./dft/simd/codelets/Makefile.am -2 +2 Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/scalar/Makefile.am -2 +2 Chris@10: M ./rdft/scalar/r2cb/Makefile.am -3 +3 Chris@10: M ./rdft/scalar/r2cf/Makefile.am -3 +3 Chris@10: M ./rdft/scalar/r2r/Makefile.am -3 +3 Chris@10: Chris@10: Sun Dec 24 20:11:50 EST 2006 athena@fftw.org Chris@10: * Fixed another dftw bug (sigh) Chris@10: Chris@10: M ./threads/ct.c -7 +3 Chris@10: Chris@10: Sun Dec 24 11:48:01 EST 2006 athena@fftw.org Chris@10: * Removed debugging leftovers. Chris@10: Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: Chris@10: Sun Dec 24 11:34:17 EST 2006 athena@fftw.org Chris@10: * Moved dftw-generic* to new dftw protocol. Chris@10: Chris@10: M ./dft/dftw-generic.c -30 +24 Chris@10: M ./dft/dftw-genericbuf.c -13 +13 Chris@10: Chris@10: Sun Dec 24 09:37:19 EST 2006 athena@fftw.org Chris@10: * Oops Chris@10: Chris@10: M ./tests/hook.c -1 +1 Chris@10: Chris@10: Sun Dec 24 09:31:46 EST 2006 athena@fftw.org Chris@10: * Fixed wrong verification of rank-1 rdft2 Chris@10: Chris@10: M ./tests/hook.c -4 +6 Chris@10: Chris@10: Sat Dec 23 20:11:29 EST 2006 athena@fftw.org Chris@10: * minor tweaks Chris@10: Chris@10: M ./dft/dftw-direct.c -2 +4 Chris@10: Chris@10: Sat Dec 23 19:50:03 EST 2006 athena@fftw.org Chris@10: * Removed obsolete items. Chris@10: Chris@10: M ./TODO -13 Chris@10: Chris@10: Sat Dec 23 17:56:37 EST 2006 athena@fftw.org Chris@10: * Modified the problem_dftw invocation protocol. Chris@10: Chris@10: apply() now requires pointers to the beginning of the full array. Chris@10: Each thread processes a slice mb <= m < me. This protocol is Chris@10: consistent with the one used in hc2hc, where there is no other choice. Chris@10: Chris@10: Chris@10: Chris@10: M ./dft/dftw-direct.c -27 +31 Chris@10: M ./dft/dftw-directsq.c -1 +3 Chris@10: M ./threads/ct.c -7 +2 Chris@10: Chris@10: Sat Dec 23 16:18:25 EST 2006 athena@fftw.org Chris@10: * typo Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Sat Dec 23 16:06:56 EST 2006 athena@fftw.org Chris@10: * changed hc2hc twiddle storage to be the same as hc2c Chris@10: Chris@10: M ./genfft/gen_hc2hc.ml -2 +3 Chris@10: M ./rdft/hc2hc-direct.c -14 +11 Chris@10: Chris@10: Sat Dec 23 15:16:36 EST 2006 athena@fftw.org Chris@10: * Allowed extra_iter in dftw-direct. Rationalized twiddle factors in hc2c. Chris@10: Chris@10: M ./dft/dftw-direct.c -10 +38 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +12 Chris@10: M ./dft/simd/t.c -2 +45 Chris@10: A ./dft/simd/t1bu.h Chris@10: A ./dft/simd/t1fu.h Chris@10: M ./genfft/gen_hc2c.ml -2 +3 Chris@10: M ./genfft/gen_hc2cdft.ml -2 +3 Chris@10: M ./genfft/gen_hc2cdft_c.ml -1 +1 Chris@10: M ./genfft/gen_hc2hc.ml -1 +1 Chris@10: M ./genfft/gen_twiddle.ml -1 +1 Chris@10: M ./genfft/gen_twiddle_c.ml -1 +1 Chris@10: M ./genfft/gen_twidsq.ml -1 +1 Chris@10: M ./genfft/gen_twidsq_c.ml -1 +1 Chris@10: M ./genfft/twiddle.ml -12 +10 Chris@10: M ./genfft/twiddle.mli -1 +1 Chris@10: M ./kernel/twiddle.c -4 +8 Chris@10: M ./rdft/ct-hc2c-direct.c -4 +7 Chris@10: M ./rdft/simd/hc2cbv.h -2 +2 Chris@10: M ./rdft/simd/hc2cfv.h -2 +2 Chris@10: M ./simd/simd-altivec.h -12 +9 Chris@10: M ./simd/simd-sse.h -12 +9 Chris@10: M ./simd/simd-sse2.h -11 +6 Chris@10: Chris@10: Sat Dec 23 10:37:11 EST 2006 athena@fftw.org Chris@10: * Implemented unmentionable hack to use 4-way SIMD with an odd number of Chris@10: iterations. Chris@10: Chris@10: M ./dft/direct.c -9 +34 Chris@10: M ./rdft/ct-hc2c-direct.c -43 +88 Chris@10: M ./simd/simd-altivec.h -1 +3 Chris@10: M ./simd/simd-sse.h -1 +3 Chris@10: Chris@10: Fri Dec 22 22:13:30 EST 2006 athena@fftw.org Chris@10: * altivec support for new codelets Chris@10: Chris@10: M ./simd/simd-altivec.h -7 +28 Chris@10: Chris@10: Fri Dec 22 19:09:15 EST 2006 athena@fftw.org Chris@10: * fixed incorrect computation of W Chris@10: Chris@10: M ./genfft/gen_hc2cdft_c.ml -1 +1 Chris@10: Chris@10: Fri Dec 22 18:51:22 EST 2006 athena@fftw.org Chris@10: * Implemented 4-way simd hc2cdftv Chris@10: Chris@10: Also eliminated the twiddle_shift hack. A zillion changes dictated Chris@10: by this choice, which was in turn necessary for the hc2cdftv thing Chris@10: to work. Chris@10: Chris@10: Chris@10: M ./Makefile.am -13 +8 Chris@10: M ./configure.ac -3 Chris@10: M ./dft/codelet-dft.h -9 +9 Chris@10: M ./dft/codelets/t.c -4 +4 Chris@10: M ./dft/dftw-direct.c -35 +30 Chris@10: M ./dft/dftw-directsq.c -7 +4 Chris@10: M ./dft/simd/q1b.c -6 +6 Chris@10: M ./dft/simd/q1f.c -6 +6 Chris@10: M ./dft/simd/t.c -14 +14 Chris@10: M ./dft/simd/t3b.h +1 Chris@10: M ./dft/simd/t3f.h +1 Chris@10: M ./dft/simd/ts.c -6 +6 Chris@10: M ./dft/simd/ts.h +1 Chris@10: M ./genfft/Makefile.am -17 +10 Chris@10: R ./genfft/gen_conv.ml Chris@10: M ./genfft/gen_hc2c.ml -25 +19 Chris@10: M ./genfft/gen_hc2cdft.ml -19 +19 Chris@10: M ./genfft/gen_hc2cdft_c.ml -20 +24 Chris@10: M ./genfft/gen_hc2hc.ml -23 +17 Chris@10: M ./genfft/gen_mdct.ml -9 +9 Chris@10: M ./genfft/gen_notw.ml -8 +8 Chris@10: M ./genfft/gen_notw_c.ml -6 +6 Chris@10: M ./genfft/gen_r2cb.ml -9 +9 Chris@10: M ./genfft/gen_r2cf.ml -9 +9 Chris@10: M ./genfft/gen_r2r.ml -6 +6 Chris@10: M ./genfft/gen_twiddle.ml -37 +39 Chris@10: M ./genfft/gen_twiddle_c.ml -38 +42 Chris@10: M ./genfft/gen_twidsq.ml -42 +44 Chris@10: M ./genfft/gen_twidsq_c.ml -44 +47 Chris@10: M ./genfft/genutil.ml -6 +6 Chris@10: M ./genfft/simd.ml -6 +7 Chris@10: M ./genfft/simd.mli -2 +1 Chris@10: M ./genfft/variable.ml -7 +12 Chris@10: M ./genfft/variable.mli -1 +3 Chris@10: M ./kernel/ifftw.h -7 Chris@10: M ./kernel/twiddle.c -14 Chris@10: M ./rdft/codelet-rdft.h -5 +5 Chris@10: M ./rdft/codelets/hc2c.c -2 +2 Chris@10: M ./rdft/conf.c -1 +1 Chris@10: M ./rdft/ct-hc2c-direct.c -47 +36 Chris@10: M ./rdft/ct-hc2c.c -1 +3 Chris@10: M ./rdft/hc2hc-direct.c -39 +31 Chris@10: M ./rdft/simd/codelets/Makefile.am -2 +2 Chris@10: M ./rdft/simd/hc2cbv.c -7 +8 Chris@10: M ./rdft/simd/hc2cbv.h -4 +3 Chris@10: M ./rdft/simd/hc2cfv.c -7 +8 Chris@10: M ./rdft/simd/hc2cfv.h -4 +3 Chris@10: M ./simd/simd-sse.h +4 Chris@10: M ./simd/simd-sse2.h -3 +5 Chris@10: M ./simd/simd.h -2 Chris@10: M ./tests/Makefile.am +9 Chris@10: Chris@10: Fri Dec 22 08:45:46 EST 2006 athena@fftw.org Chris@10: * Fixed verification of rdft2 problems with new format. Chris@10: Chris@10: M ./tests/hook.c -1 +5 Chris@10: Chris@10: Fri Dec 22 00:05:59 EST 2006 athena@fftw.org Chris@10: * Added file Chris@10: Chris@10: A ./rdft/simd/Makefile.am Chris@10: Chris@10: Fri Dec 22 00:02:50 EST 2006 athena@fftw.org Chris@10: * Hmm, previous commit did not work Chris@10: Chris@10: A ./rdft/simd/ Chris@10: A ./rdft/simd/codelets/ Chris@10: A ./rdft/simd/codelets/Makefile.am Chris@10: A ./rdft/simd/hc2hcv.h Chris@10: ./rdft/simd/hc2hcv.h -> ./rdft/simd/hc2cfv.h Chris@10: A ./rdft/simd/hc2cbv.h Chris@10: A ./rdft/simd/hc2cv.c Chris@10: ./rdft/simd/hc2cv.c -> ./rdft/simd/hc2cfv.c Chris@10: M ./rdft/simd/codelets/Makefile.am +51 Chris@10: A ./rdft/simd/hc2cbv.c Chris@10: M ./rdft/simd/hc2cbv.h +29 Chris@10: M ./rdft/simd/hc2cfv.c +41 Chris@10: M ./rdft/simd/hc2cfv.h +29 Chris@10: Chris@10: Thu Dec 21 23:58:33 EST 2006 athena@fftw.org Chris@10: * Added SIMD r2cdft codelets. Chris@10: Chris@10: A ./genfft/gen_hc2cdft_c.ml Chris@10: M ./Makefile.am -14 +22 Chris@10: M ./configure.ac +5 Chris@10: M ./dft/simd/n1b.h -1 Chris@10: M ./dft/simd/n1f.h -1 Chris@10: M ./dft/simd/n2b.h -1 Chris@10: M ./dft/simd/n2f.h -1 Chris@10: M ./dft/simd/n2s.h -1 Chris@10: M ./genfft/Makefile.am -8 +14 Chris@10: M ./genfft/algsimp.ml -2 +12 Chris@10: M ./genfft/c.ml +1 Chris@10: M ./genfft/expr.ml -1 +3 Chris@10: M ./genfft/expr.mli -1 +1 Chris@10: M ./genfft/gen_hc2cdft_c.ml +217 Chris@10: M ./genfft/simd.ml -2 +9 Chris@10: M ./genfft/to_alist.ml +1 Chris@10: M ./kernel/ifftw.h +6 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/codelet-rdft.h +5 Chris@10: M ./rdft/codelets/hc2c.c -2 +12 Chris@10: M ./rdft/conf.c +4 Chris@10: M ./rdft/ct-hc2c-direct.c -5 +60 Chris@10: M ./rdft/hc2hc-direct.c -5 Chris@10: M ./rdft/hc2hc.h -2 Chris@10: M ./rdft/khc2hc.c -1 Chris@10: M ./simd/nonportable/sse.c -9 +4 Chris@10: M ./simd/nonportable/sse2.c -9 +4 Chris@10: M ./simd/simd-sse.h -4 +4 Chris@10: M ./simd/simd-sse2.h -4 +7 Chris@10: M ./simd/sse.c -2 +2 Chris@10: M ./simd/sse2.c -2 +2 Chris@10: M ./support/Makefile.codelets +1 Chris@10: Chris@10: Thu Dec 21 21:19:21 EST 2006 athena@fftw.org Chris@10: * Bug in buffering, grrr... Chris@10: Chris@10: M ./rdft/ct-hc2c-direct.c -1 +1 Chris@10: Chris@10: Thu Dec 21 20:58:14 EST 2006 athena@fftw.org Chris@10: * Oops, memory leak. Chris@10: Chris@10: M ./rdft/hc2hc-direct.c +1 Chris@10: Chris@10: Thu Dec 21 17:12:31 EST 2006 athena@fftw.org Chris@10: * minor changes, cleanup. Chris@10: Chris@10: M ./rdft/ct-hc2c.c -2 Chris@10: M ./rdft/hc2hc.c -16 +10 Chris@10: M ./threads/hc2hc.c -17 +12 Chris@10: Chris@10: Wed Dec 20 22:09:28 EST 2006 athena@fftw.org Chris@10: * Unified hc2hc-direct, hc2hc-directbuf. Cleanup. Chris@10: Chris@10: M ./rdft/Makefile.am -4 +3 Chris@10: M ./rdft/ct-hc2c-direct.c -8 +8 Chris@10: R ./rdft/hc2hc-common.c Chris@10: M ./rdft/hc2hc-direct.c -44 +153 Chris@10: R ./rdft/hc2hc-directbuf.c Chris@10: M ./rdft/hc2hc.h -4 Chris@10: Chris@10: Wed Dec 20 17:55:56 EST 2006 athena@fftw.org Chris@10: * removed obsolete rdft2-radix2 Chris@10: Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/conf.c -1 Chris@10: M ./rdft/rdft.h -1 Chris@10: R ./rdft/rdft2-radix2.c Chris@10: Chris@10: Wed Dec 20 17:51:20 EST 2006 athena@fftw.org Chris@10: * implemented reduction rdft2->dft Chris@10: Chris@10: A ./genfft/gen_hc2cdft.ml Chris@10: M ./genfft/Makefile.am -9 +16 Chris@10: M ./genfft/gen_hc2c.ml -1 +1 Chris@10: M ./genfft/gen_hc2cdft.ml +207 Chris@10: M ./rdft/codelet-rdft.h -2 +7 Chris@10: M ./rdft/codelets/r2cb/Makefile.am -1 +12 Chris@10: M ./rdft/codelets/r2cf/Makefile.am -1 +12 Chris@10: M ./rdft/ct-hc2c-direct.c -5 +8 Chris@10: M ./rdft/ct-hc2c.c -24 +95 Chris@10: M ./rdft/ct-hc2c.h -2 +6 Chris@10: M ./rdft/khc2c.c -2 +3 Chris@10: M ./support/Makefile.codelets +1 Chris@10: Chris@10: Wed Dec 20 09:29:39 EST 2006 athena@fftw.org Chris@10: * Implemented buffered direct-r2c, direct-hc2c. Chris@10: Also, removed some old cruft: Chris@10: Chris@10: * okp() functions were never used and a pain to maintain---now they Chris@10: are gone. Chris@10: Chris@10: * ``m'' in hc2hc and hc2c codelets is now the number of iterations, Chris@10: not the ``logical'' m. Chris@10: Chris@10: M ./genfft/gen_hc2c.ml -5 +3 Chris@10: M ./genfft/gen_hc2hc.ml -9 +9 Chris@10: M ./genfft/gen_r2cb.ml -4 +1 Chris@10: M ./genfft/gen_r2cf.ml -4 +1 Chris@10: M ./genfft/gen_r2r.ml -5 +1 Chris@10: M ./rdft/codelet-rdft.h -30 +4 Chris@10: M ./rdft/codelets/hc2c.c -14 +2 Chris@10: M ./rdft/codelets/hfb.c -14 +2 Chris@10: M ./rdft/codelets/r2c.c -18 +4 Chris@10: M ./rdft/codelets/r2r.c -15 +1 Chris@10: M ./rdft/ct-hc2c-direct.c -37 +127 Chris@10: M ./rdft/direct-r2c.c -41 +189 Chris@10: M ./rdft/direct-r2r.c -15 +6 Chris@10: M ./rdft/direct2.c -20 +11 Chris@10: M ./rdft/hc2hc-direct.c -23 +9 Chris@10: M ./rdft/hc2hc-directbuf.c -79 +42 Chris@10: M ./rdft/kr2c.c +1 Chris@10: M ./rdft/rdft.h +1 Chris@10: Chris@10: Tue Dec 19 17:07:04 EST 2006 stevenj@fftw.org Chris@10: * added memcpy-loop rank0 solver (it makes a 5-20% difference for transposes of large tuples) Chris@10: Chris@10: M ./rdft/rank0.c +33 Chris@10: Chris@10: Tue Dec 19 16:15:54 EST 2006 stevenj@fftw.org Chris@10: * new variable to disable libbench2's problem allocation during speed benchmarking (to benchmark MPI transforms where the array does not fit into the memory of a single process) Chris@10: Chris@10: M ./libbench2/bench-user.h +2 Chris@10: M ./libbench2/speed.c -3 +8 Chris@10: M ./mpi/mpi-bench.c -5 +15 Chris@10: Chris@10: Tue Dec 19 14:55:08 EST 2006 stevenj@fftw.org Chris@10: * allow transpose-inplace to use input as scratch for DESTROY_INPUT plans (to avoid non-square in-place transpositions) ... on supersgj, the planner often prefers transpose-inplace to transpose-alltoall in this case (apparently MPI_Alltoall in LAM MPI isn't that great) Chris@10: Chris@10: M ./mpi/transpose-inplace.c -52 +99 Chris@10: Chris@10: Tue Dec 19 17:07:14 EST 2006 athena@fftw.org Chris@10: * For some reason HB2 codelets were not generated. Chris@10: Chris@10: M ./rdft/codelets/r2cb/Makefile.am -2 +2 Chris@10: M ./rdft/codelets/r2cf/Makefile.am -1 +1 Chris@10: Chris@10: Tue Dec 19 15:12:39 EST 2006 athena@fftw.org Chris@10: * split rdft/direct.c into direct-r2r and direct-r2c, since the file was getting out of control. Chris@10: Chris@10: A ./rdft/direct-r2c.c Chris@10: A ./rdft/direct-r2r.c Chris@10: R ./rdft/direct.c Chris@10: M ./rdft/Makefile.am -4 +5 Chris@10: M ./rdft/direct-r2c.c +195 Chris@10: M ./rdft/direct-r2r.c +154 Chris@10: Chris@10: Tue Dec 19 02:59:35 EST 2006 stevenj@fftw.org Chris@10: * added dft-rank1 solver - MPI now supports 1d complex DFTs! Chris@10: Chris@10: M ./mpi/Makefile.am -2 +2 Chris@10: M ./mpi/api.c -9 +54 Chris@10: A ./mpi/choose-radix.c Chris@10: M ./mpi/conf.c +1 Chris@10: M ./mpi/dft-problem.c -2 +2 Chris@10: A ./mpi/dft-rank1.c Chris@10: M ./mpi/fftw3-mpi.h -2 +8 Chris@10: M ./mpi/ifftw-mpi.h +4 Chris@10: M ./mpi/mpi-bench.c -25 +24 Chris@10: M ./mpi/mpi-dft.h -4 +5 Chris@10: M ./mpi/mpi-transpose.h -3 +3 Chris@10: M ./mpi/transpose-problem.c -1 +1 Chris@10: M ./tests/bench.c -8 +8 Chris@10: M ./tests/fftw-bench.h -1 +1 Chris@10: Chris@10: Tue Dec 19 01:27:20 EST 2006 stevenj@fftw.org Chris@10: * fftw_flops must call cost_hook directly; iestimate_cost always uses COST_MAX Chris@10: Chris@10: M ./api/flops.c -1 +7 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner.c -4 +3 Chris@10: Chris@10: Tue Dec 19 01:16:54 EST 2006 stevenj@fftw.org Chris@10: * fftw_flops and fftw_estimate_cost must now be called from every process, to prevent deadlocks in the MPI code (since they sum/max the cost over all processes) Chris@10: Chris@10: M ./tests/fftw-bench.c -7 +11 Chris@10: Chris@10: Tue Dec 19 00:55:34 EST 2006 stevenj@fftw.org Chris@10: * whoops, typo in assert Chris@10: Chris@10: M ./mpi/block.c -1 +1 Chris@10: Chris@10: Tue Dec 19 00:51:07 EST 2006 stevenj@fftw.org Chris@10: * remove multiplication by FFT_SIGN Chris@10: Chris@10: M ./kernel/trig.c -2 +7 Chris@10: Chris@10: Mon Dec 18 22:51:45 EST 2006 stevenj@fftw.org Chris@10: * need to synchronize ESTIMATE costs in MPI planner, and sum ESTIMATE costs for flop reporting: generalize measure_hook to cost_hook(..., {COST_SUM, COST_MAX}) Chris@10: Chris@10: M ./api/flops.c -1 +1 Chris@10: M ./kernel/ifftw.h -3 +6 Chris@10: M ./kernel/planner.c -8 +12 Chris@10: M ./mpi/api.c -6 +7 Chris@10: Chris@10: Mon Dec 18 15:36:15 EST 2006 stevenj@fftw.org Chris@10: * previous patch slowed down transpose-alltoall when TRANSPOSED_IN and DESTROY_INPUT; now allow planner to choose old behavior in this case Chris@10: Chris@10: M ./mpi/transpose-alltoall.c -11 +31 Chris@10: Chris@10: Mon Dec 18 15:15:39 EST 2006 stevenj@fftw.org Chris@10: * transpose-alltoall doesn't require input to be destroyed if TRANSPOSED_IN is set Chris@10: Chris@10: M ./mpi/transpose-alltoall.c -26 +41 Chris@10: Chris@10: Mon Dec 18 17:41:25 EST 2006 athena@fftw.org Chris@10: * Added t2-style hc2c codelets, fixed typos. Chris@10: Chris@10: M ./api/plan-guru-dft-c2r.h -1 +1 Chris@10: M ./api/plan-guru-dft-r2c.h -1 +1 Chris@10: M ./api/plan-guru-dft.h -2 +2 Chris@10: M ./rdft/codelets/r2cb/Makefile.am -1 +9 Chris@10: M ./rdft/codelets/r2cf/Makefile.am -1 +9 Chris@10: Chris@10: Mon Dec 18 16:02:42 EST 2006 athena@fftw.org Chris@10: * Renamed certain variables to avoid calling an output stride `is'. Chris@10: Chris@10: M ./rdft/codelet-rdft.h -1 +1 Chris@10: M ./rdft/direct.c -17 +17 Chris@10: M ./rdft/direct2.c -7 +7 Chris@10: Chris@10: Mon Dec 18 15:54:43 EST 2006 athena@fftw.org Chris@10: * Oops---wrong stride Chris@10: Chris@10: M ./rdft/ct-hc2c.c -1 +1 Chris@10: Chris@10: Mon Dec 18 14:59:16 EST 2006 athena@fftw.org Chris@10: * Oops, forgot to add file Chris@10: Chris@10: A ./rdft/kr2c.c Chris@10: Chris@10: Mon Dec 18 14:17:02 EST 2006 athena@fftw.org Chris@10: * Renamed r2hc/hc2r codelets to r2c Chris@10: After the recent changes, r2hc/hc2r codelets became rdft2 Chris@10: problems, so I renamed them accordingly to r2cf/r2cb. Chris@10: Codelet parameters are now a real array and a complex array, instead Chris@10: of an input array and an output array, and forward and backward Chris@10: codelets have the same type, which removes some clutter from the rdft Chris@10: code. Chris@10: Chris@10: ./genfft/gen_hc2r.ml -> ./genfft/gen_r2cb.ml Chris@10: ./genfft/gen_r2hc.ml -> ./genfft/gen_r2cf.ml Chris@10: ./rdft/codelets/hc2r -> ./rdft/codelets/r2cb Chris@10: ./rdft/codelets/r2hc -> ./rdft/codelets/r2cf Chris@10: A ./rdft/codelets/r2c.c Chris@10: A ./rdft/codelets/r2cb.h Chris@10: A ./rdft/codelets/r2cbIII.h Chris@10: A ./rdft/codelets/r2cf.h Chris@10: A ./rdft/codelets/r2cfII.h Chris@10: M ./Makefile.am -2 +2 Chris@10: M ./configure.ac -2 +2 Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: M ./genfft/Makefile.am -9 +9 Chris@10: M ./genfft/gen_r2cb.ml -58 +51 Chris@10: M ./genfft/gen_r2cf.ml -58 +51 Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/codelet-rdft.h -59 +16 Chris@10: M ./rdft/codelets/Makefile.am -3 +3 Chris@10: R ./rdft/codelets/hc2r.c Chris@10: R ./rdft/codelets/hc2r.h Chris@10: R ./rdft/codelets/hc2rIII.h Chris@10: M ./rdft/codelets/r2c.c +51 Chris@10: M ./rdft/codelets/r2cb/Makefile.am -21 +21 Chris@10: M ./rdft/codelets/r2cb.h +23 Chris@10: M ./rdft/codelets/r2cbIII.h +23 Chris@10: M ./rdft/codelets/r2cf/Makefile.am -20 +20 Chris@10: M ./rdft/codelets/r2cf.h +23 Chris@10: M ./rdft/codelets/r2cfII.h +23 Chris@10: R ./rdft/codelets/r2hc.c Chris@10: R ./rdft/codelets/r2hc.h Chris@10: R ./rdft/codelets/r2hcII.h Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/ct-hc2c.c -2 +2 Chris@10: M ./rdft/direct.c -46 +28 Chris@10: M ./rdft/direct2.c -72 +23 Chris@10: R ./rdft/khc2r.c Chris@10: R ./rdft/kr2hc.c Chris@10: M ./rdft/rdft.h -4 +2 Chris@10: M ./support/Makefile.codelets -2 +2 Chris@10: Chris@10: Mon Dec 18 10:48:07 EST 2006 athena@fftw.org Chris@10: * Implemented backward radix-2k rdft2. Chris@10: Chris@10: A ./rdft/codelets/hc2cf.h Chris@10: A ./rdft/codelets/hc2cb.h Chris@10: M ./genfft/gen_hc2c.ml -26 +30 Chris@10: M ./rdft/codelets/Makefile.am -1 +1 Chris@10: M ./rdft/codelets/hc2c.c -1 +5 Chris@10: R ./rdft/codelets/hc2c.h Chris@10: M ./rdft/codelets/hc2cb.h +23 Chris@10: M ./rdft/codelets/hc2cf.h +23 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -1 +11 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -7 +7 Chris@10: M ./rdft/ct-hc2c.c -3 +25 Chris@10: M ./rdft/rdft2-radix2.c -72 +1 Chris@10: Chris@10: Mon Dec 18 13:56:09 EST 2006 stevenj@fftw.org Chris@10: * move extract_reim into kernel, since it is used by internal MPI stuff and not just in the API code any more Chris@10: Chris@10: ./api/extract-reim.c -> ./kernel/extract-reim.c Chris@10: M ./api/Makefile.am -1 +1 Chris@10: M ./api/api.h -1 +1 Chris@10: M ./api/plan-many-dft-c2r.c -1 +1 Chris@10: M ./api/plan-many-dft-r2c.c -1 +1 Chris@10: M ./api/plan-many-dft.c -2 +2 Chris@10: M ./kernel/Makefile.am -5 +6 Chris@10: M ./kernel/extract-reim.c -6 +6 Chris@10: M ./kernel/ifftw.h +1 Chris@10: M ./mpi/dft-rank-geq2.c -3 Chris@10: M ./mpi/dft-rank1-bigvec.c -3 Chris@10: M ./mpi/dft-serial.c -3 Chris@10: Chris@10: Mon Dec 18 08:40:14 EST 2006 athena@fftw.org Chris@10: * Do not check r1==cr unless rnk>0 Chris@10: Chris@10: M ./rdft/problem2.c -1 +5 Chris@10: Chris@10: Sun Dec 17 21:03:50 EST 2006 athena@fftw.org Chris@10: * Implemented radix-2k RDFT2, forward only for now Chris@10: Chris@10: M ./api/execute-dft-c2r.c -1 +2 Chris@10: M ./api/execute-dft-r2c.c -1 +2 Chris@10: M ./api/execute-split-dft-c2r.c -1 +2 Chris@10: M ./api/execute-split-dft-r2c.c -1 +2 Chris@10: M ./api/f77funcs.h -4 +8 Chris@10: M ./api/plan-guru-dft-c2r.h -8 +8 Chris@10: M ./api/plan-guru-dft-r2c.h -9 +9 Chris@10: M ./api/plan-guru-split-dft-c2r.h -6 +6 Chris@10: M ./api/plan-guru-split-dft-r2c.h -9 +9 Chris@10: M ./api/plan-many-dft-c2r.c -3 +3 Chris@10: M ./api/plan-many-dft-r2c.c -1 +1 Chris@10: M ./dft/problem.c -3 +5 Chris@10: M ./genfft/Makefile.am -5 +11 Chris@10: A ./genfft/gen_hc2c.ml Chris@10: M ./genfft/gen_hc2hc.ml -30 +30 Chris@10: M ./genfft/gen_hc2r.ml -7 +16 Chris@10: M ./genfft/gen_r2hc.ml -8 +18 Chris@10: M ./rdft/Makefile.am -1 +2 Chris@10: M ./rdft/buffered2.c -31 +45 Chris@10: M ./rdft/codelet-rdft.h -11 +41 Chris@10: M ./rdft/codelets/Makefile.am -1 +1 Chris@10: A ./rdft/codelets/hc2c.c Chris@10: A ./rdft/codelets/hc2c.h Chris@10: M ./rdft/codelets/hc2r.c -4 +5 Chris@10: M ./rdft/codelets/hfb.c -4 +4 Chris@10: M ./rdft/codelets/r2hc.c -4 +5 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +11 Chris@10: A ./rdft/ct-hc2c-direct.c Chris@10: A ./rdft/ct-hc2c.c Chris@10: A ./rdft/ct-hc2c.h Chris@10: M ./rdft/direct.c -16 +21 Chris@10: M ./rdft/direct2.c -15 +32 Chris@10: M ./rdft/hc2hc-common.c -2 +2 Chris@10: M ./rdft/hc2hc-direct.c -4 +4 Chris@10: M ./rdft/hc2hc-directbuf.c -8 +8 Chris@10: M ./rdft/hc2hc-generic.c -4 +3 Chris@10: M ./rdft/hc2hc.c -4 +4 Chris@10: A ./rdft/khc2c.c Chris@10: M ./rdft/nop2.c -7 +10 Chris@10: M ./rdft/problem.c -2 +1 Chris@10: M ./rdft/problem2.c -38 +117 Chris@10: M ./rdft/rank-geq2-rdft2.c -13 +17 Chris@10: M ./rdft/rank0-rdft2.c -35 +44 Chris@10: M ./rdft/rdft.h -8 +27 Chris@10: M ./rdft/rdft2-inplace-strides.c -2 +6 Chris@10: M ./rdft/rdft2-radix2.c -131 +37 Chris@10: M ./rdft/solve2.c -1 +3 Chris@10: M ./rdft/vrank-geq1-rdft2.c -7 +7 Chris@10: M ./support/Makefile.codelets +1 Chris@10: M ./tests/hook.c -8 +23 Chris@10: M ./threads/vrank-geq1-rdft2.c -10 +10 Chris@10: Chris@10: Mon Dec 18 01:23:45 EST 2006 stevenj@fftw.org Chris@10: * separate TRANSPOSED/SCRAMBLED flags internally (this is required so that dft-rank1-bigvec and the future dft-rank1 won't have incompatible SCRAMBLED formats) Chris@10: Chris@10: M ./mpi/api.c -16 +2 Chris@10: M ./mpi/dft-rank-geq2.c -7 +7 Chris@10: M ./mpi/dft-rank1-bigvec.c -8 +9 Chris@10: M ./mpi/dft-serial.c -1 +1 Chris@10: M ./mpi/ifftw-mpi.h -1 +8 Chris@10: M ./mpi/mpi-dft.h -1 +2 Chris@10: M ./mpi/mpi-transpose.h -2 +2 Chris@10: M ./mpi/transpose-alltoall.c -3 +4 Chris@10: M ./mpi/transpose-inplace.c -6 +7 Chris@10: M ./mpi/transpose-radix2.c -5 +6 Chris@10: Chris@10: Mon Dec 18 01:02:27 EST 2006 stevenj@fftw.org Chris@10: * ops_add -> ops_add2 where possible, to shrink code Chris@10: Chris@10: M ./mpi/dft-rank-geq2.c -8 +4 Chris@10: M ./mpi/dft-rank1-bigvec.c -4 +2 Chris@10: M ./mpi/transpose-alltoall.c -6 +3 Chris@10: M ./mpi/transpose-inplace.c -8 +4 Chris@10: M ./mpi/transpose-radix2.c -10 +5 Chris@10: M ./rdft/vrank3-transpose.c -9 +5 Chris@10: Chris@10: Mon Dec 18 00:43:02 EST 2006 stevenj@fftw.org Chris@10: * added dft-rank1-bigvec solver (easy case for 1d parallel transforms) Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/api.c -2 +14 Chris@10: M ./mpi/conf.c +1 Chris@10: A ./mpi/dft-rank1-bigvec.c Chris@10: M ./mpi/mpi-dft.h +1 Chris@10: Chris@10: Sun Dec 17 20:42:21 EST 2006 stevenj@fftw.org Chris@10: * rewrote MPI stuff to use dtensor data structure Chris@10: Chris@10: A dtensor is an ordered tuple of triplets (n, ib, ob) giving the size of Chris@10: a dimension (n) and its input and output block sizes of a distributed Chris@10: row-major multi-dimensional array. An MPI DFT (etc.) is now specified Chris@10: in terms of dtensors, which provide a much more flexible data layout. Chris@10: Chris@10: For example, we can now describe multidimensional block distributions, Chris@10: which are important if the number of processors is greater than the Chris@10: size of any given dimension. Currently, we only have solvers for Chris@10: 1d slab distributions, and this is all that is supported in the basic Chris@10: and advanced APIs. The guru API allows one to specify more general Chris@10: distributions, however, which will be useful when/if we have solvers Chris@10: for this case. Chris@10: Chris@10: We now also don't need a TRANSPOSED flag, at least internally, since Chris@10: TRANSPOSED multi-dimensional DFT plans just correspond to dtensors Chris@10: where the input and output block distributions are different. Chris@10: Chris@10: Other changes include the use of the XM(foo) macro for X(mpi_foo). Chris@10: Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/any-true.c -1 +1 Chris@10: M ./mpi/api.c -129 +335 Chris@10: M ./mpi/block.c -32 +95 Chris@10: M ./mpi/conf.c -6 +6 Chris@10: M ./mpi/dft-problem.c -46 +33 Chris@10: M ./mpi/dft-rank-geq2.c -60 +72 Chris@10: M ./mpi/dft-serial.c -26 +20 Chris@10: M ./mpi/dft-solve.c -1 +1 Chris@10: A ./mpi/dtensor.c Chris@10: M ./mpi/fftw3-mpi.h -24 +38 Chris@10: M ./mpi/ifftw-mpi.h -22 +60 Chris@10: M ./mpi/mpi-bench.c -115 +289 Chris@10: M ./mpi/mpi-dft.h -12 +13 Chris@10: M ./mpi/mpi-transpose.h -10 +11 Chris@10: M ./mpi/transpose-alltoall.c -17 +16 Chris@10: M ./mpi/transpose-inplace.c -24 +21 Chris@10: M ./mpi/transpose-problem.c -19 +18 Chris@10: M ./mpi/transpose-radix2.c -13 +11 Chris@10: M ./mpi/transpose-solve.c -1 +1 Chris@10: M ./tests/bench.c +4 Chris@10: M ./tests/fftw-bench.c +2 Chris@10: M ./tests/fftw-bench.h +1 Chris@10: Chris@10: Fri Dec 15 16:01:23 EST 2006 athena@fftw.org Chris@10: * Distinguished mutexes from semaphores. Chris@10: The distinction is useful because the linux implementation of Chris@10: sem_post() in unnecessarily slow when semaphores are used for mutual Chris@10: exclusion. This change made spinlocks messier to implement, so I Chris@10: excised them. Chris@10: Chris@10: M ./tests/fftw-bench.c -8 Chris@10: M ./threads/threads.c -105 +40 Chris@10: Chris@10: Fri Dec 15 12:46:11 EST 2006 athena@fftw.org Chris@10: * Use posix semaphores where available. Chris@10: Paranoid declaration of all shared variables as ``volatile''. Paranoid Chris@10: initialization of all shared variables within locks. Chris@10: Chris@10: M ./threads/threads.c -36 +68 Chris@10: Chris@10: Wed Dec 13 20:12:13 EST 2006 Matteo Frigo Chris@10: * paranoia Chris@10: Chris@10: M ./threads/threads.c -2 +2 Chris@10: Chris@10: Tue Dec 12 17:28:13 EST 2006 stevenj@fftw.org Chris@10: * punt on detecting unsolvable rdft2 problems; make r==iio rdft2 problems unsolvable, since it doesn't look like we've consistently checked for this case and it's not clear why we would want to support it (it was also not documented in the manual) Chris@10: Chris@10: M ./doc/fftw3.texi -1 +2 Chris@10: M ./rdft/problem2.c -30 +2 Chris@10: Chris@10: Sat Dec 9 12:14:13 EST 2006 athena@fftw.org Chris@10: * Obey stupid const rules Chris@10: Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -1 +1 Chris@10: M ./dft/dft.h -3 +3 Chris@10: M ./dft/indirect.c -4 +4 Chris@10: M ./dft/problem.c -9 +8 Chris@10: M ./kernel/ifftw.h -5 +5 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./kernel/problem.c -4 +4 Chris@10: M ./rdft/buffered2.c -1 +1 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/indirect.c -4 +4 Chris@10: M ./rdft/problem.c -14 +13 Chris@10: M ./rdft/problem2.c -9 +8 Chris@10: M ./rdft/rank-geq2-rdft2.c -1 +1 Chris@10: M ./rdft/rdft-dht.c -1 +1 Chris@10: M ./rdft/rdft.h -13 +13 Chris@10: M ./rdft/rdft2-radix2.c -5 +5 Chris@10: Chris@10: Fri Dec 8 18:21:50 EST 2006 stevenj@fftw.org Chris@10: * added unsolvable check for rdft2 problem Chris@10: Chris@10: An in-place rdft2 problem is ill-formed if the real data, including the Chris@10: extra "padding" elements, do not coincide with the complex data. Chris@10: Chris@10: CHANGE: the new code considers all in-place split r2c and c2r problems Chris@10: to be ill-formed. Previously, these could be done, but only if the Chris@10: entire multi-dimensional array fit into the buffer, which is kind of Chris@10: stupid. I'm not sure it's worth it to even try to support the Chris@10: split in-place r2c case. Chris@10: Chris@10: M ./rdft/problem2.c -1 +33 Chris@10: Chris@10: Fri Dec 8 13:47:53 EST 2006 stevenj@fftw.org Chris@10: * check in-placeness after joining taints Chris@10: Chris@10: M ./dft/problem.c -9 +9 Chris@10: Chris@10: Fri Dec 8 13:43:44 EST 2006 athena@fftw.org Chris@10: * Grrr... paranoid-check was not testing in exhaustive mode Chris@10: Chris@10: M ./tests/Makefile.am -5 +5 Chris@10: Chris@10: Fri Dec 8 10:00:30 EST 2006 Matteo Frigo Chris@10: * Implemented PROBLEM_UNSOLVABLE. Chris@10: In-place DFT and RDFT problems with inconsistent I/O strides are Chris@10: now unsolvable, and we don't check for them any longer in solvers. Chris@10: Chris@10: While I was at it, declared all problem pointers to be ``const'' Chris@10: for extra safety. Chris@10: Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -3 +5 Chris@10: M ./dft/buffered.c -4 +1 Chris@10: M ./dft/dft.h -4 +4 Chris@10: M ./dft/direct.c -8 +6 Chris@10: M ./dft/indirect.c -4 +4 Chris@10: M ./dft/problem.c -11 +15 Chris@10: M ./dft/vrank-geq1.c -3 Chris@10: M ./kernel/ifftw.h -6 +10 Chris@10: M ./kernel/planner.c -6 +6 Chris@10: M ./kernel/problem.c -1 +40 Chris@10: M ./rdft/buffered.c -4 +1 Chris@10: M ./rdft/buffered2.c -1 +1 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/direct.c -4 +3 Chris@10: M ./rdft/indirect.c -4 +4 Chris@10: M ./rdft/problem.c -14 +17 Chris@10: M ./rdft/problem2.c -10 +10 Chris@10: M ./rdft/rank-geq2-rdft2.c -1 +1 Chris@10: M ./rdft/rdft-dht.c -1 +1 Chris@10: M ./rdft/rdft.h -13 +13 Chris@10: M ./rdft/rdft2-radix2.c -5 +5 Chris@10: M ./rdft/vrank-geq1.c -2 Chris@10: Chris@10: Thu Dec 7 20:13:46 EST 2006 Matteo Frigo Chris@10: * Avoid qsort'ing one element. Chris@10: Chris@10: M ./kernel/tensor7.c -6 +2 Chris@10: Chris@10: Thu Dec 7 18:25:47 EST 2006 Matteo Frigo Chris@10: * In-place vrank>=1 is now applicable only if the problem is really in-place. Chris@10: Chris@10: M ./dft/buffered.c -7 +12 Chris@10: M ./dft/direct.c -11 +11 Chris@10: M ./dft/vrank-geq1.c +3 Chris@10: M ./kernel/ifftw.h +2 Chris@10: M ./kernel/tensor7.c +41 Chris@10: M ./rdft/buffered.c -7 +12 Chris@10: M ./rdft/direct.c -7 +5 Chris@10: M ./rdft/vrank-geq1.c +2 Chris@10: Chris@10: Tue Dec 5 12:21:38 EST 2006 athena@fftw.org Chris@10: * unused variable Chris@10: Chris@10: M ./kernel/align.c +1 Chris@10: Chris@10: Sun Dec 3 19:16:33 EST 2006 Matteo Frigo Chris@10: * Removed CVS $Id$ everywhere, since darcs does not update them. Chris@10: Chris@10: M ./api/fftw3.h -1 Chris@10: M ./api/version.c -1 Chris@10: M ./commercialize.sh +1 Chris@10: M ./dft/buffered.c -1 Chris@10: M ./dft/codelet-dft.h -1 Chris@10: M ./dft/codelets/standard/Makefile.am +1 Chris@10: M ./dft/conf.c -1 Chris@10: M ./dft/ct.c -1 Chris@10: M ./dft/dft.h -1 Chris@10: M ./dft/dftw-direct.c -1 Chris@10: M ./dft/dftw-directsq.c -1 Chris@10: M ./dft/direct.c -1 Chris@10: M ./dft/indirect.c -1 Chris@10: M ./dft/kdft-dif.c -1 Chris@10: M ./dft/kdft-difsq.c -1 Chris@10: M ./dft/kdft-dit.c -1 Chris@10: M ./dft/kdft.c -1 Chris@10: M ./dft/nop.c -1 Chris@10: M ./dft/plan.c -1 Chris@10: M ./dft/problem.c -1 Chris@10: M ./dft/rank-geq2.c -1 Chris@10: M ./dft/solve.c -1 Chris@10: M ./dft/vrank-geq1.c -1 Chris@10: M ./dft/zero.c -1 Chris@10: M ./doc/fftw3.texi -1 Chris@10: M ./genfft/algsimp.ml -2 Chris@10: M ./genfft/algsimp.mli -2 Chris@10: M ./genfft/annotate.ml -2 Chris@10: M ./genfft/annotate.mli -1 Chris@10: M ./genfft/assoctable.ml -1 Chris@10: M ./genfft/assoctable.mli -1 Chris@10: M ./genfft/c.ml -1 Chris@10: M ./genfft/c.mli -1 Chris@10: M ./genfft/complex.ml -1 Chris@10: M ./genfft/complex.mli -1 Chris@10: M ./genfft/conv.mli -1 Chris@10: M ./genfft/dag.ml -1 Chris@10: M ./genfft/dag.mli -1 Chris@10: M ./genfft/expr.ml -1 Chris@10: M ./genfft/expr.mli -1 Chris@10: M ./genfft/fft.ml -2 Chris@10: M ./genfft/fft.mli -2 Chris@10: M ./genfft/gen_conv.ml -3 +1 Chris@10: M ./genfft/gen_hc2hc.ml -3 +1 Chris@10: M ./genfft/gen_hc2r.ml -3 +1 Chris@10: M ./genfft/gen_mdct.ml -3 +1 Chris@10: M ./genfft/gen_notw.ml -3 +1 Chris@10: M ./genfft/gen_notw_c.ml -3 +1 Chris@10: M ./genfft/gen_r2hc.ml -3 +1 Chris@10: M ./genfft/gen_r2r.ml -3 +1 Chris@10: M ./genfft/gen_twiddle.ml -3 +1 Chris@10: M ./genfft/gen_twiddle_c.ml -3 +1 Chris@10: M ./genfft/gen_twidsq.ml -3 +1 Chris@10: M ./genfft/gen_twidsq_c.ml -3 +1 Chris@10: M ./genfft/genutil.ml -8 +1 Chris@10: M ./genfft/littlesimp.ml -1 Chris@10: M ./genfft/littlesimp.mli -1 Chris@10: M ./genfft/magic.ml -1 Chris@10: M ./genfft/monads.ml -1 Chris@10: M ./genfft/number.ml -1 Chris@10: M ./genfft/number.mli -1 Chris@10: M ./genfft/oracle.ml -1 Chris@10: M ./genfft/oracle.mli -1 Chris@10: M ./genfft/schedule.ml -1 Chris@10: M ./genfft/schedule.mli -1 Chris@10: M ./genfft/simd.ml -1 Chris@10: M ./genfft/simd.mli -1 Chris@10: M ./genfft/simdmagic.ml -1 Chris@10: M ./genfft/to_alist.ml -1 Chris@10: M ./genfft/to_alist.mli -1 Chris@10: M ./genfft/trig.ml -1 Chris@10: M ./genfft/trig.mli -1 Chris@10: M ./genfft/twiddle.ml -1 Chris@10: M ./genfft/twiddle.mli -1 Chris@10: M ./genfft/unique.ml -1 Chris@10: M ./genfft/unique.mli -1 Chris@10: M ./genfft/util.ml -1 Chris@10: M ./genfft/util.mli -1 Chris@10: M ./genfft/variable.ml -1 Chris@10: M ./genfft/variable.mli -1 Chris@10: M ./kernel/align.c -1 Chris@10: M ./kernel/alloc.c -1 Chris@10: M ./kernel/assert.c -1 Chris@10: M ./kernel/awake.c -1 Chris@10: M ./kernel/cycle.h -1 Chris@10: M ./kernel/debug.c -1 Chris@10: M ./kernel/iabs.c -1 Chris@10: M ./kernel/ifftw.h -1 Chris@10: M ./kernel/kalloc.c -1 Chris@10: M ./kernel/minmax.c -1 Chris@10: M ./kernel/ops.c -1 Chris@10: M ./kernel/pickdim.c -1 Chris@10: M ./kernel/plan.c -1 Chris@10: M ./kernel/planner.c -1 Chris@10: M ./kernel/primes.c -1 Chris@10: M ./kernel/print.c -1 Chris@10: M ./kernel/problem.c -1 Chris@10: M ./kernel/scan.c -1 Chris@10: M ./kernel/solver.c -1 Chris@10: M ./kernel/solvtab.c -1 Chris@10: M ./kernel/stride.c -1 Chris@10: M ./kernel/tensor.c -1 Chris@10: M ./kernel/tensor1.c -1 Chris@10: M ./kernel/tensor2.c -1 Chris@10: M ./kernel/tensor4.c -1 Chris@10: M ./kernel/tensor5.c -1 Chris@10: M ./kernel/tensor7.c -1 Chris@10: M ./kernel/tensor8.c -1 Chris@10: M ./kernel/tensor9.c -1 Chris@10: M ./kernel/timer.c -1 Chris@10: M ./kernel/trig.c -1 Chris@10: M ./kernel/twiddle.c -1 Chris@10: M ./libbench/accopy-from.c -1 Chris@10: M ./libbench/accopy-to.c -1 Chris@10: M ./libbench/allocate.c -1 Chris@10: M ./libbench/bench-main.c -1 Chris@10: M ./libbench/bench-user.h -1 Chris@10: M ./libbench/bench.h -1 Chris@10: M ./libbench/can-do.c -1 Chris@10: M ./libbench/ccopy-from.c -1 Chris@10: M ./libbench/ccopy-to.c -1 Chris@10: M ./libbench/deallocate.c -1 Chris@10: M ./libbench/getopt-utils.c -1 Chris@10: M ./libbench/info.c -1 Chris@10: M ./libbench/main.c -1 Chris@10: M ./libbench/prime.c -1 Chris@10: M ./libbench/problem.c -1 Chris@10: M ./libbench/report.c -1 Chris@10: M ./libbench/speed.c -1 Chris@10: M ./libbench/timer.c -1 Chris@10: M ./libbench/verify.c -1 Chris@10: M ./libbench/zero.c -1 Chris@10: M ./libbench2/aligned-main.c -1 Chris@10: M ./libbench2/allocate.c -1 Chris@10: M ./libbench2/bench-main.c -1 Chris@10: M ./libbench2/bench-user.h -1 Chris@10: M ./libbench2/bench.h -1 Chris@10: M ./libbench2/can-do.c -1 Chris@10: M ./libbench2/dotens2.c -1 Chris@10: M ./libbench2/info.c -1 Chris@10: M ./libbench2/main.c -1 Chris@10: M ./libbench2/problem.c -1 Chris@10: M ./libbench2/report.c -1 Chris@10: M ./libbench2/speed.c -1 Chris@10: M ./libbench2/tensor.c -1 Chris@10: M ./libbench2/timer.c -1 Chris@10: M ./libbench2/useropt.c -1 Chris@10: M ./libbench2/verify-dft.c -1 Chris@10: M ./libbench2/verify-lib.c -1 Chris@10: M ./libbench2/verify-rdft2.c -1 Chris@10: M ./libbench2/verify.c -1 Chris@10: M ./libbench2/zero.c -1 Chris@10: M ./mpi/conf.c -1 Chris@10: M ./rdft/buffered.c -1 Chris@10: M ./rdft/buffered2.c -1 Chris@10: M ./rdft/codelets/hc2r/Makefile.am +1 Chris@10: M ./rdft/codelets/r2hc/Makefile.am +1 Chris@10: M ./rdft/conf.c -1 Chris@10: M ./rdft/dft-r2hc.c -1 Chris@10: M ./rdft/dht-r2hc.c -1 Chris@10: M ./rdft/direct.c -1 Chris@10: M ./rdft/direct2.c -1 Chris@10: M ./rdft/hc2hc-direct.c -1 Chris@10: M ./rdft/hc2hc-directbuf.c -1 Chris@10: M ./rdft/indirect.c -1 Chris@10: M ./rdft/khc2hc.c -1 Chris@10: M ./rdft/khc2r.c -1 Chris@10: M ./rdft/kr2hc.c -1 Chris@10: M ./rdft/kr2r.c -1 Chris@10: M ./rdft/nop.c -1 Chris@10: M ./rdft/nop2.c -1 Chris@10: M ./rdft/plan.c -1 Chris@10: M ./rdft/plan2.c -1 Chris@10: M ./rdft/problem.c -1 Chris@10: M ./rdft/problem2.c -1 Chris@10: M ./rdft/rank-geq2-rdft2.c -1 Chris@10: M ./rdft/rank-geq2.c -1 Chris@10: M ./rdft/rank0-rdft2.c -1 Chris@10: M ./rdft/rank0.c -1 Chris@10: M ./rdft/rdft-dht.c -1 Chris@10: M ./rdft/rdft2-inplace-strides.c -1 Chris@10: M ./rdft/rdft2-radix2.c -1 Chris@10: M ./rdft/rdft2-tensor-max-index.c -1 Chris@10: M ./rdft/solve.c -1 Chris@10: M ./rdft/solve2.c -1 Chris@10: M ./rdft/vrank-geq1-rdft2.c -1 Chris@10: M ./rdft/vrank-geq1.c -1 Chris@10: M ./rdft/vrank3-transpose.c -1 Chris@10: M ./reodft/conf.c -1 Chris@10: M ./reodft/redft00e-r2hc-pad.c -1 Chris@10: M ./reodft/redft00e-r2hc.c -1 Chris@10: M ./reodft/reodft00e-splitradix.c -1 Chris@10: M ./reodft/reodft010e-r2hc.c -1 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -1 Chris@10: M ./reodft/reodft11e-r2hc.c -1 Chris@10: M ./reodft/reodft11e-radix2.c -1 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -1 Chris@10: M ./reodft/rodft00e-r2hc.c -1 Chris@10: M ./simd/altivec.c -1 Chris@10: M ./simd/nonportable/sse.c -1 Chris@10: M ./simd/nonportable/sse2.c -1 Chris@10: M ./simd/sse.c -1 Chris@10: M ./simd/sse2.c -1 Chris@10: M ./simd/taint.c -1 Chris@10: M ./simd/x86-cpuid.h -1 Chris@10: M ./threads/conf.c -1 Chris@10: M ./threads/ct.c -1 Chris@10: M ./threads/dft-vrank-geq1.c -1 Chris@10: M ./threads/rdft-vrank-geq1.c -1 Chris@10: M ./threads/vrank-geq1-rdft2.c -1 Chris@10: Chris@10: Sun Dec 3 16:11:17 EST 2006 Matteo Frigo Chris@10: * generalized dftw to encompass q codelets. As a side effect, q codelets are now threaded. Chris@10: Chris@10: R ./dft/ctsq.c Chris@10: A ./dft/dftw-directsq.c Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/ct.c -25 +55 Chris@10: M ./dft/ct.h -6 +9 Chris@10: M ./dft/dftw-direct.c -59 +72 Chris@10: M ./dft/dftw-directsq.c +163 Chris@10: M ./dft/dftw-generic.c -24 +32 Chris@10: M ./dft/dftw-genericbuf.c -30 +37 Chris@10: M ./dft/indirect-transpose.c -1 +1 Chris@10: M ./dft/kdft-difsq.c -1 +1 Chris@10: M ./libbench2/my-getopt.c -3 +3 Chris@10: M ./threads/ct.c -28 +53 Chris@10: Chris@10: Sat Nov 25 16:34:38 EST 2006 stevenj@fftw.org Chris@10: * add missing __declspec attribute to threads API functions when compiling for Windows (thanks to Robert O. Morris for the bug report) Chris@10: Chris@10: M ./api/f77api.c -29 +1 Chris@10: M ./api/x77.h +28 Chris@10: M ./threads/f77api.c +13 Chris@10: M ./threads/f77funcs.h -3 +3 Chris@10: Chris@10: Mon Nov 20 17:39:20 EST 2006 stevenj@fftw.org Chris@10: * add AC_SUBST to AX_OPENMP, thanks to Sebastien Maret for the suggestion Chris@10: Chris@10: M ./m4/ax_openmp.m4 -1 +2 Chris@10: Chris@10: Mon Oct 16 23:02:29 EDT 2006 stevenj@fftw.org Chris@10: * not gcc bug for MIPS (thanks to Jonathan Day) Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +4 Chris@10: Chris@10: Sat Sep 23 17:52:36 EDT 2006 stevenj@fftw.org Chris@10: * in maintainer/debug mode, don't modify CFLAGS if they were explicitly set (-pedantic seems to cause problems with LAM's mpicc, so I need a way to override) Chris@10: Chris@10: M ./configure.ac +2 Chris@10: Chris@10: Wed Nov 22 18:14:47 EST 2006 athena@fftw.org Chris@10: * Removed obsolete comment. Chris@10: Chris@10: M ./dft/dftw-generic.c -1 Chris@10: Chris@10: Sun Nov 19 11:21:44 EST 2006 athena@fftw.org Chris@10: * Use p->v when comparing TW_FULL fields. Chris@10: Chris@10: M ./kernel/twiddle.c -2 +1 Chris@10: Chris@10: Thu Nov 16 14:49:05 EST 2006 athena@fftw.org Chris@10: * removed useless definition Chris@10: Chris@10: M ./simd/simd.h -1 Chris@10: Chris@10: Mon Nov 13 09:18:32 EST 2006 athena@fftw.org Chris@10: * paranoid avoidance of integer overflows Chris@10: Chris@10: M ./libbench2/timer.c -3 +3 Chris@10: Chris@10: Mon Nov 13 09:00:11 EST 2006 athena@fftw.org Chris@10: * avoid potential overflows in cycle counters Chris@10: At the suggestion of Alex Cichowski, convert all ticks Chris@10: to double before operating on them, to avoid potential Chris@10: signed/unsigned confusion and integer overflow. Chris@10: Chris@10: M ./kernel/cycle.h -5 +8 Chris@10: Chris@10: Sun Nov 5 09:00:52 EST 2006 Matteo Frigo Chris@10: * Removed unused struct field Chris@10: Chris@10: M ./threads/threads.c -1 Chris@10: Chris@10: Sat Nov 4 09:43:13 EST 2006 Matteo Frigo Chris@10: * use pthread condition variables instead of semaphores Chris@10: Condition variables are more likely to be portable everywhere, and Chris@10: somehow they appear to introduce less overhead at least on my Chris@10: linux box. Chris@10: Chris@10: M ./threads/threads.c -6 +19 Chris@10: Chris@10: Tue Oct 31 20:45:24 EST 2006 Matteo Frigo Chris@10: * Imprecise help message. Chris@10: Chris@10: M ./libbench2/my-getopt.c -1 +1 Chris@10: Chris@10: Mon Oct 30 20:13:35 EST 2006 athena@fftw.org Chris@10: * Experimental implementation of spinlocks. Chris@10: This patch implements spinlocks via a semi-portable hack, and adds Chris@10: the -ospinlocks option to the bench program so that we can play with them. Chris@10: Chris@10: M ./tests/fftw-bench.c +8 Chris@10: M ./threads/threads.c -28 +104 Chris@10: Chris@10: Thu Oct 26 22:29:18 EDT 2006 athena@fftw.org Chris@10: * Updated manual for new openmp configure options. Chris@10: Chris@10: M ./doc/fftw3.texi -20 +17 Chris@10: Chris@10: Thu Oct 26 21:52:39 EDT 2006 athena@fftw.org Chris@10: * Added back openmp. Chris@10: Rationalized threads naming conventions: Chris@10: Chris@10: * threads explicitly managed by us are enabled by --enable-threads, Chris@10: predicated on HAVE_THREADS, etc. Chris@10: * openmp is enabled by --enable-openmp, predicated on HAVE_OPENMP, etc. Chris@10: * SMP denotes either THREADS or OPENMP. Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac -10 +27 Chris@10: M ./kernel/alloc.c -1 +1 Chris@10: M ./kernel/ifftw.h -1 +1 Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./tests/Makefile.am -5 +5 Chris@10: M ./tests/fftw-bench.c -3 +7 Chris@10: M ./threads/Makefile.am -2 +2 Chris@10: M ./threads/api.c -1 +16 Chris@10: M ./threads/conf.c -2 +2 Chris@10: A ./threads/openmp.c Chris@10: M ./threads/threads.c -13 +19 Chris@10: M ./threads/threads.h -2 +2 Chris@10: M ./tools/Makefile.am -1 +1 Chris@10: M ./tools/fftw-wisdom.c -5 +9 Chris@10: Chris@10: Mon Oct 23 20:14:31 EDT 2006 athena@fftw.org Chris@10: * different thread protocols Chris@10: Chris@10: M ./threads/threads.c -80 +71 Chris@10: Chris@10: Sun Oct 22 14:49:32 EDT 2006 athena@fftw.org Chris@10: * fix memory leak Chris@10: Added pthread_attr_destroy to avoid memory leak. Chris@10: Chris@10: M ./threads/threads.c +1 Chris@10: Chris@10: Sun Oct 22 14:23:30 EDT 2006 athena@fftw.org Chris@10: * Experimental new pthread implementation that recycles threads. Chris@10: Chris@10: M ./threads/threads.c -528 +155 Chris@10: Chris@10: Tue Oct 24 23:28:10 EDT 2006 athena@fftw.org Chris@10: * switched buddies Chris@10: Switched order of buddies in rdft2 rank-geq2 for consistency Chris@10: with analogous dft and rdft solvers. Furthermore, this change reduces Chris@10: the MEASURE planning time for rank == 3. Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -4 +1 Chris@10: Chris@10: Sat Oct 21 10:10:00 EDT 2006 Matteo Frigo Chris@10: * Typo. Chris@10: Chris@10: M ./tools/fftw_wisdom.1.in -1 +1 Chris@10: Chris@10: Sat Oct 21 09:56:50 EDT 2006 Matteo Frigo Chris@10: * Out of place is the default. Thanks to Kirk Kern for pointing this out. Chris@10: Chris@10: M ./tests/README -2 +2 Chris@10: Chris@10: Fri Sep 29 01:36:11 EDT 2006 stevenj@fftw.org Chris@10: * rename "test" to "tst", since a user (Igor Levicki) reports that "test" is a reserved words in some x86 assemblers Chris@10: Chris@10: M ./simd/x86-cpuid.h -10 +10 Chris@10: Chris@10: Tue Sep 26 09:01:08 EDT 2006 Matteo Frigo Chris@10: * Stylistic change. Chris@10: Chris@10: M ./kernel/timer.c -2 +1 Chris@10: Chris@10: Tue Sep 26 08:45:37 EDT 2006 athena@fftw.org Chris@10: * Do not set tmin=1e10, since a large FFT may take longer than that. Chris@10: Chris@10: M ./kernel/timer.c -5 +4 Chris@10: Chris@10: Sat Sep 23 22:07:10 EDT 2006 athena@fftw.org Chris@10: * Disable certain gcc optimizations. Chris@10: When PRECOMPUTE_ARRAY_INDICES is #define'd, array indices have the Chris@10: form array[stride[k]] for compile-time constant k. Apparently new Chris@10: gcc's copy stride[k] onto the stack before the codelet loop, which is Chris@10: an idiotic optimization if ever there was one. This patch confuses Chris@10: gcc enough to prevent this optimization. Chris@10: Chris@10: M ./kernel/ifftw.h -3 +6 Chris@10: M ./kernel/stride.c -1 +1 Chris@10: Chris@10: Sat Sep 23 13:02:58 EDT 2006 stevenj@fftw.org Chris@10: * re-enable TOMS algorithm - it is the best for large vector lengths, since for such sizes the cache line is not an issue and the bookkeeping overhead is negligible Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -18 +15 Chris@10: Chris@10: Thu Sep 21 15:40:15 EDT 2006 stevenj@fftw.org Chris@10: * add measure_hook so that MPI can synchronize timing measurements (otherwise different processors might end up with different MPI plans, yikes!) Chris@10: Chris@10: M ./kernel/ifftw.h +2 Chris@10: M ./kernel/planner.c +5 Chris@10: M ./mpi/api.c -2 +26 Chris@10: Chris@10: Tue Sep 19 21:26:19 EDT 2006 stevenj@fftw.org Chris@10: * added O(p log p) transpose algorithm (radix 2) Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/conf.c +1 Chris@10: M ./mpi/mpi-transpose.h +1 Chris@10: A ./mpi/transpose-radix2.c Chris@10: Chris@10: Tue Sep 19 21:05:09 EDT 2006 stevenj@fftw.org Chris@10: * comments Chris@10: Chris@10: M ./mpi/testsched.c -8 +10 Chris@10: Chris@10: Tue Sep 19 19:54:58 EDT 2006 stevenj@fftw.org Chris@10: * whoops Chris@10: Chris@10: ./mpi/any_true.c -> ./mpi/any-true.c Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: Chris@10: Tue Sep 19 19:49:01 EDT 2006 stevenj@fftw.org Chris@10: * synchronize planning so that if one process fails to create a plan then all of them do. Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: A ./mpi/any_true.c Chris@10: M ./mpi/dft-rank-geq2.c -4 +4 Chris@10: M ./mpi/dft-serial.c -1 +1 Chris@10: M ./mpi/ifftw-mpi.h +3 Chris@10: M ./mpi/transpose-alltoall.c -5 +5 Chris@10: M ./mpi/transpose-inplace.c -5 +5 Chris@10: Chris@10: Tue Sep 19 18:17:38 EDT 2006 stevenj@fftw.org Chris@10: * call MPI_Alltoall instead of MPI_Alltoallv for equal-blocks case, in case MPI implementation has special optimizations for the common case of equal sizes Chris@10: Chris@10: M ./mpi/transpose-alltoall.c -6 +19 Chris@10: Chris@10: Tue Sep 19 12:07:35 EDT 2006 stevenj@fftw.org Chris@10: * whoops Chris@10: Chris@10: M ./tools/Makefile.am -1 +1 Chris@10: Chris@10: Tue Sep 19 02:20:06 EDT 2006 stevenj@fftw.org Chris@10: * typo in comment Chris@10: Chris@10: M ./mpi/testsched.c -3 +3 Chris@10: Chris@10: Tue Sep 19 01:58:55 EDT 2006 stevenj@fftw.org Chris@10: * more filename simplifications Chris@10: Chris@10: ./mpi/mpi-dft-rank-geq2.c -> ./mpi/dft-rank-geq2.c Chris@10: ./mpi/mpi-dft-serial.c -> ./mpi/dft-serial.c Chris@10: ./mpi/mpi-transpose-alltoall.c -> ./mpi/transpose-alltoall.c Chris@10: ./mpi/mpi-transpose-inplace.c -> ./mpi/transpose-inplace.c Chris@10: ./mpi/problem.c -> ./mpi/dft-problem.c Chris@10: ./mpi/solve.c -> ./mpi/dft-solve.c Chris@10: ./mpi/tproblem.c -> ./mpi/transpose-problem.c Chris@10: ./mpi/tsolve.c -> ./mpi/transpose-solve.c Chris@10: M ./mpi/Makefile.am -3 +3 Chris@10: Chris@10: Tue Sep 19 01:49:52 EDT 2006 stevenj@fftw.org Chris@10: * canonicalize file names (hyphens, not underscores) Chris@10: Chris@10: ./mpi/fftw3_mpi.h -> ./mpi/fftw3-mpi.h Chris@10: ./mpi/ifftw_mpi.h -> ./mpi/ifftw-mpi.h Chris@10: ./mpi/mpi_bench.c -> ./mpi/mpi-bench.c Chris@10: ./mpi/mpi_dft.h -> ./mpi/mpi-dft.h Chris@10: ./mpi/mpi_transpose.h -> ./mpi/mpi-transpose.h Chris@10: ./mpi/mpi_transpose_alltoall.c -> ./mpi/mpi-transpose-alltoall.c Chris@10: ./mpi/mpi_transpose_inplace.c -> ./mpi/mpi-transpose-inplace.c Chris@10: ./tests/fftw_bench.h -> ./tests/fftw-bench.h Chris@10: ./tests/fftw_bench_common.c -> ./tests/fftw-bench.c Chris@10: M ./mpi/Makefile.am -5 +5 Chris@10: M ./mpi/api.c -4 +4 Chris@10: M ./mpi/block.c -1 +1 Chris@10: M ./mpi/conf.c -2 +2 Chris@10: M ./mpi/mpi-bench.c -2 +2 Chris@10: M ./mpi/mpi-dft-rank-geq2.c -2 +2 Chris@10: M ./mpi/mpi-dft-serial.c -1 +1 Chris@10: M ./mpi/mpi-dft.h -1 +1 Chris@10: M ./mpi/mpi-transpose-alltoall.c -1 +1 Chris@10: M ./mpi/mpi-transpose-inplace.c -1 +1 Chris@10: M ./mpi/mpi-transpose.h -1 +1 Chris@10: M ./mpi/problem.c -1 +1 Chris@10: M ./mpi/solve.c -1 +1 Chris@10: M ./mpi/tproblem.c -1 +1 Chris@10: M ./mpi/tsolve.c -1 +1 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/bench.c -2 +2 Chris@10: M ./tests/fftw-bench.c -1 +1 Chris@10: Chris@10: Tue Sep 19 01:34:07 EDT 2006 stevenj@fftw.org Chris@10: * add mpi-dft-serial Chris@10: Chris@10: M ./mpi/Makefile.am -1 +1 Chris@10: M ./mpi/conf.c +1 Chris@10: M ./mpi/mpi-dft-rank-geq2.c -3 +9 Chris@10: A ./mpi/mpi-dft-serial.c Chris@10: M ./mpi/mpi_dft.h +1 Chris@10: M ./mpi/mpi_transpose_alltoall.c -1 +1 Chris@10: M ./mpi/mpi_transpose_inplace.c -1 +1 Chris@10: Chris@10: Tue Sep 19 01:21:47 EDT 2006 stevenj@fftw.org Chris@10: * silence warnings Chris@10: Chris@10: M ./dft/indirect-transpose.c -1 Chris@10: M ./libbench2/problem.c -2 Chris@10: Chris@10: Tue Sep 19 00:31:59 EDT 2006 stevenj@fftw.org Chris@10: * make "t" problem semantics match FFTW_MPI_TRANSPOSED Chris@10: Chris@10: M ./libbench2/problem.c -4 +3 Chris@10: M ./mpi/mpi_bench.c -8 +18 Chris@10: Chris@10: Mon Sep 18 23:50:43 EDT 2006 stevenj@fftw.org Chris@10: * whoops, fixed backwards mpi_dft Chris@10: Chris@10: M ./mpi/mpi-dft-rank-geq2.c -7 +7 Chris@10: Chris@10: Mon Sep 18 22:26:31 EDT 2006 stevenj@fftw.org Chris@10: * initial stab at rank-geq2 mpi-dft; seems to be mostly working Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +1 Chris@10: M ./mpi/Makefile.am -1 +4 Chris@10: M ./mpi/api.c -26 +180 Chris@10: M ./mpi/conf.c +2 Chris@10: M ./mpi/fftw3_mpi.h -11 +51 Chris@10: M ./mpi/ifftw_mpi.h -4 +1 Chris@10: A ./mpi/mpi-dft-rank-geq2.c Chris@10: M ./mpi/mpi_bench.c -55 +166 Chris@10: M ./mpi/mpi_dft.h -2 +2 Chris@10: M ./mpi/mpi_transpose.h -6 +1 Chris@10: M ./mpi/mpi_transpose_alltoall.c -7 +10 Chris@10: M ./mpi/mpi_transpose_inplace.c -7 +8 Chris@10: M ./mpi/problem.c -2 +7 Chris@10: A ./mpi/solve.c Chris@10: M ./mpi/tproblem.c -2 +2 Chris@10: Chris@10: Sun Sep 17 13:41:32 EDT 2006 stevenj@fftw.org Chris@10: * support SCRAMBLED_OUT in alltoall transpose Chris@10: Chris@10: M ./kernel/ifftw.h +4 Chris@10: M ./kernel/tensor2.c +20 Chris@10: M ./mpi/ifftw_mpi.h -1 +16 Chris@10: M ./mpi/mpi_dft.h -1 +1 Chris@10: M ./mpi/mpi_transpose_alltoall.c -23 +48 Chris@10: M ./mpi/mpi_transpose_inplace.c -23 +2 Chris@10: M ./mpi/problem.c -2 +3 Chris@10: Chris@10: Sun Sep 17 12:34:30 EDT 2006 stevenj@fftw.org Chris@10: * skeleton of future support for block-cyclic Chris@10: Chris@10: M ./mpi/block.c -7 +27 Chris@10: M ./mpi/ifftw_mpi.h +1 Chris@10: M ./mpi/mpi_transpose_alltoall.c +2 Chris@10: M ./mpi/mpi_transpose_inplace.c -2 +5 Chris@10: Chris@10: Sun Sep 17 12:11:19 EDT 2006 stevenj@fftw.org Chris@10: * test program now checks scrambled in/out via -obflag=28/29 Chris@10: Chris@10: M ./mpi/api.c -1 +1 Chris@10: M ./mpi/fftw3_mpi.h -1 +1 Chris@10: M ./mpi/mpi_bench.c -13 +59 Chris@10: M ./mpi/problem.c -1 Chris@10: Chris@10: Sun Sep 17 11:58:36 EDT 2006 stevenj@fftw.org Chris@10: * added -obflag to make it easier to set high-order bits Chris@10: Chris@10: M ./tests/fftw_bench_common.c +1 Chris@10: Chris@10: Sun Sep 17 01:30:51 EDT 2006 stevenj@fftw.org Chris@10: * use proper child plans for 2nd transpose in transpose_alltoall; implement opcount in transpose_inplace Chris@10: Chris@10: M ./mpi/mpi_transpose_alltoall.c -47 +47 Chris@10: M ./mpi/mpi_transpose_inplace.c -1 +10 Chris@10: Chris@10: Sun Sep 17 01:08:01 EDT 2006 stevenj@fftw.org Chris@10: * fix in test program for transposes of vectors -- transpose routines seem to completely work now (except for scrambled in/out, which is untested) Chris@10: Chris@10: M ./mpi/mpi_bench.c -1 +1 Chris@10: Chris@10: Sun Sep 17 01:01:16 EDT 2006 stevenj@fftw.org Chris@10: * fixed bug in transpose_alltoall for unequal blocks Chris@10: Chris@10: M ./mpi/mpi_transpose_alltoall.c -2 +2 Chris@10: M ./mpi/mpi_transpose_inplace.c -2 +2 Chris@10: Chris@10: Sat Sep 16 15:29:46 EDT 2006 stevenj@fftw.org Chris@10: * correctly handle cld2rest Chris@10: Chris@10: M ./mpi/mpi_transpose_inplace.c -24 +28 Chris@10: Chris@10: Sat Sep 16 15:29:31 EDT 2006 stevenj@fftw.org Chris@10: * some debugging code and other fixes Chris@10: Chris@10: M ./mpi/mpi_bench.c -13 +34 Chris@10: Chris@10: Sat Sep 16 14:54:30 EDT 2006 stevenj@fftw.org Chris@10: * whoops, forgot to check in mpi_bench.c file Chris@10: Chris@10: A ./mpi/mpi_bench.c Chris@10: Chris@10: Sat Sep 16 14:54:02 EDT 2006 stevenj@fftw.org Chris@10: * added bench_exit routine so that it can be overridden (by MPI_Abort) if needed Chris@10: Chris@10: M ./libbench2/Makefile.am -5 +5 Chris@10: A ./libbench2/bench-exit.c Chris@10: M ./libbench2/bench-user.h +1 Chris@10: M ./libbench2/util.c -1 +1 Chris@10: M ./libbench2/verify-lib.c -1 +1 Chris@10: M ./libbench2/verify-r2r.c -1 +1 Chris@10: Chris@10: Sat Sep 16 14:52:56 EDT 2006 stevenj@fftw.org Chris@10: * bug fix in mpi_transpose_inplace for case where some processors are idle Chris@10: Chris@10: M ./mpi/mpi_transpose_inplace.c -3 +4 Chris@10: Chris@10: Fri Sep 15 18:47:13 EDT 2006 stevenj@fftw.org Chris@10: * allow vecloop for sz->rnk==0 in exceptional (SLOW) cases, e.g. it is necessary for loops of non-square transposes (otherwise e.g. ik1v5:200:200x10:20:1x20:1:10 planning fails) Chris@10: Chris@10: M ./rdft/vrank-geq1.c -2 +6 Chris@10: Chris@10: Thu Sep 14 23:36:48 EDT 2006 stevenj@fftw.org Chris@10: * first pass at working mpi_bench test program; transpose seems to work iff dimensions are divisible by #processors Chris@10: Chris@10: M ./libbench2/Makefile.am -1 +3 Chris@10: A ./libbench2/after-ccopy-from.c Chris@10: A ./libbench2/after-ccopy-to.c Chris@10: A ./libbench2/after-hccopy-from.c Chris@10: A ./libbench2/after-hccopy-to.c Chris@10: A ./libbench2/after-rcopy-from.c Chris@10: A ./libbench2/after-rcopy-to.c Chris@10: M ./libbench2/bench-main.c -4 +6 Chris@10: M ./libbench2/bench-user.h +10 Chris@10: M ./libbench2/verify-dft.c +2 Chris@10: M ./libbench2/verify-r2r.c +4 Chris@10: M ./libbench2/verify-rdft2.c +4 Chris@10: M ./mpi/api.c -2 +6 Chris@10: Chris@10: Wed Sep 13 17:28:07 EDT 2006 stevenj@fftw.org Chris@10: * whoops Chris@10: Chris@10: M ./mpi/mpi_transpose_inplace.c -2 +2 Chris@10: Chris@10: Tue Sep 12 22:27:03 EDT 2006 stevenj@fftw.org Chris@10: * initial stub for mpi_bench Chris@10: Chris@10: M ./Makefile.am -2 +2 Chris@10: M ./libbench2/bench-main.c +3 Chris@10: M ./libbench2/bench-user.h +1 Chris@10: M ./libbench2/verify-lib.c -1 +1 Chris@10: M ./mpi/Makefile.am -1 +14 Chris@10: M ./tests/bench.c +9 Chris@10: M ./tests/fftw_bench.h +1 Chris@10: M ./tests/fftw_bench_common.c +2 Chris@10: Chris@10: Tue Sep 12 21:54:31 EDT 2006 stevenj@fftw.org Chris@10: * do no output at all if verbose < 0 (for use with MPI, where we only want output from process 0) Chris@10: Chris@10: M ./libbench2/bench-main.c -1 +1 Chris@10: M ./libbench2/bench.h +1 Chris@10: M ./libbench2/ovtpvt.c -1 +15 Chris@10: M ./libbench2/useropt.c -1 +1 Chris@10: M ./libbench2/util.c -2 +1 Chris@10: M ./libbench2/verify-lib.c -5 +4 Chris@10: M ./libbench2/verify-r2r.c -4 +4 Chris@10: Chris@10: Tue Sep 12 21:39:15 EDT 2006 stevenj@fftw.org Chris@10: * whoops Chris@10: Chris@10: M ./tests/fftw_bench_common.c +3 Chris@10: Chris@10: Tue Sep 12 21:31:40 EDT 2006 stevenj@fftw.org Chris@10: * split bench.c into bench.c and fftw_bench_common.c so that we can re-use some of the code in the MPI test program Chris@10: Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/bench.c -210 +6 Chris@10: A ./tests/fftw_bench.h Chris@10: A ./tests/fftw_bench_common.c Chris@10: M ./tools/Makefile.am -1 +2 Chris@10: Chris@10: Tue Sep 12 21:00:36 EDT 2006 stevenj@fftw.org Chris@10: * MPI stuff at least compiles now Chris@10: Chris@10: M ./configure.ac -9 +11 Chris@10: M ./kernel/ifftw.h +7 Chris@10: M ./mpi/api.c -3 +4 Chris@10: M ./mpi/fftw3_mpi.h +1 Chris@10: M ./mpi/mpi_transpose_alltoall.c -10 +12 Chris@10: M ./mpi/mpi_transpose_inplace.c -11 +14 Chris@10: M ./mpi/problem.c -5 +4 Chris@10: M ./mpi/tproblem.c -4 +3 Chris@10: Chris@10: Mon Sep 11 22:26:36 EDT 2006 stevenj@fftw.org Chris@10: * initial (nonfunctional) start at MPI support (similar to FFTW 2.x in spirit, but mostly rewritten) Chris@10: Chris@10: A ./mpi/ Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +11 Chris@10: A ./m4/acx_mpi.m4 Chris@10: A ./mpi/Makefile.am Chris@10: A ./mpi/api.c Chris@10: A ./mpi/block.c Chris@10: A ./mpi/conf.c Chris@10: A ./mpi/fftw3_mpi.h Chris@10: A ./mpi/ifftw_mpi.h Chris@10: A ./mpi/mpi_dft.h Chris@10: A ./mpi/mpi_transpose.h Chris@10: A ./mpi/mpi_transpose_alltoall.c Chris@10: A ./mpi/mpi_transpose_inplace.c Chris@10: A ./mpi/problem.c Chris@10: A ./mpi/testsched.c Chris@10: A ./mpi/tproblem.c Chris@10: A ./mpi/tsolve.c Chris@10: Chris@10: Mon Sep 11 22:25:38 EDT 2006 stevenj@fftw.org Chris@10: * make X(plan_awake) work for NULL argument to reduce code size Chris@10: Chris@10: M ./kernel/plan.c -3 +5 Chris@10: M ./rdft/vrank3-transpose.c -3 +3 Chris@10: Chris@10: Fri Sep 15 23:47:08 EDT 2006 stevenj@fftw.org Chris@10: * -mt should go before -mthreads to avoid spurious warnings on HPUX (thanks to Peter O'Gorman for the bug report) Chris@10: Chris@10: M ./m4/acx_pthread.m4 -2 +3 Chris@10: Chris@10: Mon Sep 11 13:53:44 EDT 2006 stevenj@fftw.org Chris@10: * Fortran init_threads wrapper didn't return result; thanks to Markus Wetzstein for the bug report Chris@10: Chris@10: M ./threads/f77funcs.h -2 +2 Chris@10: Chris@10: Thu Sep 7 18:43:55 EDT 2006 stevenj@fftw.org Chris@10: * make sure wrappers are included even if Fortran compiler was not detected (unless --disable-fortran was specified explicitly) ... this was supposed to be done before, but the definition was in the wrong place, grr Chris@10: Chris@10: M ./api/f77api.c -21 +21 Chris@10: Chris@10: Thu Aug 31 19:33:29 EDT 2006 stevenj@fftw.org Chris@10: * I'm sick of answering this question about non-deterministic results Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +13 Chris@10: Chris@10: Tue Aug 22 21:27:29 EDT 2006 Matteo Frigo Chris@10: * Add --tag=CC flag to libtool. Chris@10: Chris@10: This change is consistent with the libtool invocation in the latest Chris@10: automake, and is required to compile with (some version of) xlc. Chris@10: Chris@10: Chris@10: Chris@10: M ./support/Makefile.codelets -2 +3 Chris@10: Chris@10: Mon Aug 21 21:40:36 EDT 2006 athena@fftw.org Chris@10: * avoid ``fma'' because it is defined in c99. Chris@10: Chris@10: M ./tests/bench.c -3 +3 Chris@10: Chris@10: Sun Aug 20 11:40:53 EDT 2006 Matteo Frigo Chris@10: * Obey -standalone flag. Chris@10: Chris@10: M ./genfft/gen_twiddle.ml -2 +4 Chris@10: Chris@10: Sat Aug 19 13:34:27 EDT 2006 Matteo Frigo Chris@10: * obey -standalone when generating simd codelets Chris@10: Chris@10: M ./genfft/gen_notw_c.ml -2 +1 Chris@10: M ./genfft/gen_twiddle_c.ml -2 +3 Chris@10: Chris@10: Sat Aug 19 13:33:43 EDT 2006 Matteo Frigo Chris@10: * removed obsolete athfft Chris@10: Chris@10: M ./genfft/Makefile.am -10 +9 Chris@10: R ./genfft/gen_athnotw.ml Chris@10: R ./genfft/gen_athtw.ml Chris@10: Chris@10: Thu Aug 17 21:50:50 EDT 2006 stevenj@fftw.org Chris@10: * updated citation to Proc. IEEE paper Chris@10: Chris@10: M ./doc/fftw3.texi -8 +5 Chris@10: Chris@10: Thu Aug 17 21:47:05 EDT 2006 stevenj@fftw.org Chris@10: * use darcs changes --summary to make nice changelog; emacs fill-region hack is obsolete Chris@10: Chris@10: M ./mkdist.sh -3 +1 Chris@10: Chris@10: Mon Aug 14 17:53:19 EDT 2006 athena@fftw.org Chris@10: * removed timer calibration Chris@10: Timer calibration seems not to work any longer on recent processors--- Chris@10: too much noise. I have remove it completely. Chris@10: Chris@10: M ./libbench2/Makefile.am -1 +1 Chris@10: M ./libbench2/bench.h -2 Chris@10: M ./libbench2/timer.c -70 +3 Chris@10: R ./libbench2/timer2.c Chris@10: Chris@10: Mon Aug 14 10:47:15 EDT 2006 Matteo Frigo Chris@10: * removed k7 Chris@10: Removed obsolete k7 support. Chris@10: Chris@10: M ./Makefile.am -10 +2 Chris@10: M ./NEWS +3 Chris@10: M ./TODO -5 Chris@10: M ./api/version.c -3 Chris@10: M ./bootstrap.sh -2 Chris@10: M ./commercialize.sh -8 +1 Chris@10: M ./configure.ac -13 Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/codelet-dft.h -4 Chris@10: M ./dft/conf.c -3 Chris@10: R ./dft/k7/Makefile.am Chris@10: R ./dft/k7/codelets/Makefile.am Chris@10: R ./dft/k7/codelets/ Chris@10: R ./dft/k7/k7.c Chris@10: R ./dft/k7/ Chris@10: M ./doc/fftw3.texi -31 +22 Chris@10: R ./genfft-k7/Makefile.am Chris@10: R ./genfft-k7/algsimp.ml Chris@10: R ./genfft-k7/algsimp.mli Chris@10: R ./genfft-k7/assignmentsToVfpinstrs.ml Chris@10: R ./genfft-k7/assignmentsToVfpinstrs.mli Chris@10: R ./genfft-k7/assoctable.ml Chris@10: R ./genfft-k7/assoctable.mli Chris@10: R ./genfft-k7/balanceVfpinstrs.ml Chris@10: R ./genfft-k7/balanceVfpinstrs.mli Chris@10: R ./genfft-k7/complex.ml Chris@10: R ./genfft-k7/complex.mli Chris@10: R ./genfft-k7/expr.ml Chris@10: R ./genfft-k7/expr.mli Chris@10: R ./genfft-k7/fft.ml Chris@10: R ./genfft-k7/fft.mli Chris@10: R ./genfft-k7/genUtil.ml Chris@10: R ./genfft-k7/gen_notw.ml Chris@10: R ./genfft-k7/gen_twiddle.ml Chris@10: R ./genfft-k7/id.ml Chris@10: R ./genfft-k7/id.mli Chris@10: R ./genfft-k7/k7Basics.ml Chris@10: R ./genfft-k7/k7Basics.mli Chris@10: R ./genfft-k7/k7ExecutionModel.ml Chris@10: R ./genfft-k7/k7ExecutionModel.mli Chris@10: R ./genfft-k7/k7FlatInstructionScheduling.ml Chris@10: R ./genfft-k7/k7FlatInstructionScheduling.mli Chris@10: R ./genfft-k7/k7InstructionSchedulingBasics.ml Chris@10: R ./genfft-k7/k7InstructionSchedulingBasics.mli Chris@10: R ./genfft-k7/k7RegisterAllocationBasics.ml Chris@10: R ./genfft-k7/k7RegisterAllocationBasics.mli Chris@10: R ./genfft-k7/k7RegisterAllocator.ml Chris@10: R ./genfft-k7/k7RegisterAllocator.mli Chris@10: R ./genfft-k7/k7RegisterAllocatorEATranslation.ml Chris@10: R ./genfft-k7/k7RegisterAllocatorEATranslation.mli Chris@10: R ./genfft-k7/k7RegisterAllocatorInit.ml Chris@10: R ./genfft-k7/k7RegisterAllocatorInit.mli Chris@10: R ./genfft-k7/k7RegisterReallocation.ml Chris@10: R ./genfft-k7/k7RegisterReallocation.mli Chris@10: R ./genfft-k7/k7Translate.ml Chris@10: R ./genfft-k7/k7Translate.mli Chris@10: R ./genfft-k7/k7Unparsing.ml Chris@10: R ./genfft-k7/k7Unparsing.mli Chris@10: R ./genfft-k7/k7Vectorization.ml Chris@10: R ./genfft-k7/k7Vectorization.mli Chris@10: R ./genfft-k7/littlesimp.ml Chris@10: R ./genfft-k7/littlesimp.mli Chris@10: R ./genfft-k7/magic.ml Chris@10: R ./genfft-k7/memoMonad.ml Chris@10: R ./genfft-k7/memoMonad.mli Chris@10: R ./genfft-k7/monads.ml Chris@10: R ./genfft-k7/nonDetMonad.ml Chris@10: R ./genfft-k7/nonDetMonad.mli Chris@10: R ./genfft-k7/nullVectorization.ml Chris@10: R ./genfft-k7/nullVectorization.mli Chris@10: R ./genfft-k7/number.ml Chris@10: R ./genfft-k7/number.mli Chris@10: R ./genfft-k7/oracle.ml Chris@10: R ./genfft-k7/oracle.mli Chris@10: R ./genfft-k7/stateMonad.ml Chris@10: R ./genfft-k7/stateMonad.mli Chris@10: R ./genfft-k7/to_alist.ml Chris@10: R ./genfft-k7/to_alist.mli Chris@10: R ./genfft-k7/twiddle.ml Chris@10: R ./genfft-k7/twiddle.mli Chris@10: R ./genfft-k7/util.ml Chris@10: R ./genfft-k7/util.mli Chris@10: R ./genfft-k7/vAnnotatedScheduler.ml Chris@10: R ./genfft-k7/vAnnotatedScheduler.mli Chris@10: R ./genfft-k7/vDag.ml Chris@10: R ./genfft-k7/vDag.mli Chris@10: R ./genfft-k7/vFpBasics.ml Chris@10: R ./genfft-k7/vFpBasics.mli Chris@10: R ./genfft-k7/vFpUnparsing.ml Chris@10: R ./genfft-k7/vFpUnparsing.mli Chris@10: R ./genfft-k7/vImproveSchedule.ml Chris@10: R ./genfft-k7/vImproveSchedule.mli Chris@10: R ./genfft-k7/vK7Optimization.ml Chris@10: R ./genfft-k7/vK7Optimization.mli Chris@10: R ./genfft-k7/vScheduler.ml Chris@10: R ./genfft-k7/vScheduler.mli Chris@10: R ./genfft-k7/vSimdBasics.ml Chris@10: R ./genfft-k7/vSimdBasics.mli Chris@10: R ./genfft-k7/vSimdIndexing.ml Chris@10: R ./genfft-k7/vSimdIndexing.mli Chris@10: R ./genfft-k7/vSimdUnparsing.ml Chris@10: R ./genfft-k7/vSimdUnparsing.mli Chris@10: R ./genfft-k7/variable.ml Chris@10: R ./genfft-k7/variable.mli Chris@10: R ./genfft-k7/ Chris@10: M ./kernel/align.c -2 Chris@10: M ./kernel/ifftw.h -5 +1 Chris@10: M ./mkdist.sh -1 +1 Chris@10: M ./support/Makefile.am -1 +1 Chris@10: M ./support/Makefile.codelets -6 +1 Chris@10: R ./support/codelet_asmprelude Chris@10: Chris@10: Sun Aug 13 11:02:11 EDT 2006 athena@fftw.org Chris@10: * Use darcs instead of cvs. Chris@10: Chris@10: M ./mkdist.sh -5 +2 Chris@10: Chris@10: Wed Jul 19 08:52:15 EDT 2006 athena Chris@10: * [project @ 2006-07-19 12:52:15 by athena] Chris@10: Treat a the string "-" as a nonoption. Chris@10: Chris@10: M ./libbench2/my-getopt.c -1 +4 Chris@10: Chris@10: Tue Jul 4 17:10:47 EDT 2006 stevenj Chris@10: * [project @ 2006-07-04 21:10:47 by stevenj] Chris@10: comment out pkginclude dir for now Chris@10: Chris@10: M ./Makefile.am -2 +2 Chris@10: M ./api/Makefile.am -2 +2 Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./reodft/Makefile.am -2 +2 Chris@10: M ./threads/Makefile.am -2 +2 Chris@10: Chris@10: Mon Jul 3 20:51:08 EDT 2006 stevenj Chris@10: * [project @ 2006-07-04 00:51:08 by stevenj] Chris@10: make sure CCAS = CC to avoid libtool confusion Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Fri Jun 23 04:07:31 EDT 2006 stevenj Chris@10: * [project @ 2006-06-23 08:07:31 by stevenj] Chris@10: install x77.h guru.h guru64.h in pkgincludedir Chris@10: Chris@10: M ./api/Makefile.am -1 +1 Chris@10: Chris@10: Fri Jun 23 04:03:42 EDT 2006 stevenj Chris@10: * [project @ 2006-06-23 08:03:42 by stevenj] Chris@10: whitespace Chris@10: Chris@10: M ./configure.ac -3 Chris@10: Chris@10: Fri Jun 23 02:33:45 EDT 2006 stevenj Chris@10: * [project @ 2006-06-23 06:33:45 by stevenj] Chris@10: support cycle counter with xlc on Linux/ppc Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Tue Jun 20 08:16:08 EDT 2006 athena Chris@10: * [project @ 2006-06-20 12:16:08 by athena] Chris@10: Stylistic change. Chris@10: Chris@10: M ./tools/fftw-wisdom.c -1 +1 Chris@10: Chris@10: Tue Jun 20 02:20:34 EDT 2006 stevenj Chris@10: * [project @ 2006-06-20 06:20:34 by stevenj] Chris@10: bump date Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -1 +1 Chris@10: Chris@10: Tue Jun 20 02:20:06 EDT 2006 stevenj Chris@10: * [project @ 2006-06-20 06:20:06 by stevenj] Chris@10: correct bug reported by Andrew Salamon ... --enable-portable-binary was Chris@10: ignored (or rather, treated unpredictably) due to typo, grrr Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -1 +1 Chris@10: Chris@10: Thu Jun 1 20:30:06 EDT 2006 stevenj Chris@10: * [project @ 2006-06-02 00:30:06 by stevenj] Chris@10: install 'internal' header files into includedir/fftw3/, includedir/fftw3f/, etcetera....this will make it easier to write external libraries that plug into FFTW internals, e.g. to add new solvers Chris@10: Chris@10: M ./Makefile.am +3 Chris@10: M ./api/Makefile.am +3 Chris@10: M ./dft/Makefile.am +3 Chris@10: M ./kernel/Makefile.am +3 Chris@10: M ./rdft/Makefile.am +3 Chris@10: M ./reodft/Makefile.am +3 Chris@10: M ./threads/Makefile.am +3 Chris@10: Chris@10: Mon May 29 23:59:19 EDT 2006 stevenj Chris@10: * [project @ 2006-05-30 03:59:19 by stevenj] Chris@10: bug fix, thanks to James Donald for the bug report (only affects experimental semaphore stuff) Chris@10: Chris@10: M ./threads/threads.c +1 Chris@10: Chris@10: Mon May 29 23:58:16 EDT 2006 stevenj Chris@10: * [project @ 2006-05-30 03:58:16 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./NEWS -3 +4 Chris@10: Chris@10: Mon May 29 21:02:50 EDT 2006 stevenj Chris@10: * [project @ 2006-05-30 01:02:50 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./m4/acx_pthread.m4 -1 +3 Chris@10: Chris@10: Sat May 27 19:36:15 EDT 2006 stevenj Chris@10: * [project @ 2006-05-27 23:36:15 by stevenj] Chris@10: version bump Chris@10: Chris@10: M ./m4/acx_pthread.m4 -1 +1 Chris@10: Chris@10: Sat May 27 14:54:47 EDT 2006 stevenj Chris@10: * [project @ 2006-05-27 18:54:47 by stevenj] Chris@10: only check for xlc_r/cc_r if we are not using gcc Chris@10: Chris@10: M ./m4/acx_pthread.m4 -2 +3 Chris@10: Chris@10: Fri May 26 15:00:38 EDT 2006 stevenj Chris@10: * [project @ 2006-05-26 19:00:38 by stevenj] Chris@10: use ptrdiff_t (it's C89 and standard C++, hooray) Chris@10: Chris@10: M ./api/fftw3.h -10 +5 Chris@10: Chris@10: Fri May 26 12:59:33 EDT 2006 stevenj Chris@10: * [project @ 2006-05-26 16:59:33 by stevenj] Chris@10: version bump Chris@10: Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Fri May 26 12:57:32 EDT 2006 stevenj Chris@10: * [project @ 2006-05-26 16:57:32 by stevenj] Chris@10: noted 64-bit guru API Chris@10: Chris@10: M ./NEWS +7 Chris@10: Chris@10: Fri May 26 12:53:09 EDT 2006 stevenj Chris@10: * [project @ 2006-05-26 16:53:09 by stevenj] Chris@10: note that newer versions of VC++ support long long Chris@10: Chris@10: M ./api/fftw3.h -2 +3 Chris@10: Chris@10: Fri May 26 12:46:09 EDT 2006 stevenj Chris@10: * [project @ 2006-05-26 16:46:09 by stevenj] Chris@10: try harder to get a portable 64-bit type Chris@10: Chris@10: M ./api/fftw3.h -2 +6 Chris@10: Chris@10: Thu May 25 22:04:18 EDT 2006 stevenj Chris@10: * [project @ 2006-05-26 02:04:18 by stevenj] Chris@10: added draft guru64 API Chris@10: Chris@10: A ./api/guru.h Chris@10: A ./api/guru64.h Chris@10: A ./api/mktensor-iodims.h Chris@10: A ./api/mktensor-iodims64.c Chris@10: A ./api/plan-guru-dft-c2r.h Chris@10: A ./api/plan-guru-dft-r2c.h Chris@10: A ./api/plan-guru-dft.h Chris@10: A ./api/plan-guru-r2r.h Chris@10: A ./api/plan-guru-split-dft-c2r.h Chris@10: A ./api/plan-guru-split-dft-r2c.h Chris@10: A ./api/plan-guru-split-dft.h Chris@10: A ./api/plan-guru64-dft-c2r.c Chris@10: A ./api/plan-guru64-dft-r2c.c Chris@10: A ./api/plan-guru64-dft.c Chris@10: A ./api/plan-guru64-r2r.c Chris@10: A ./api/plan-guru64-split-dft-c2r.c Chris@10: A ./api/plan-guru64-split-dft-r2c.c Chris@10: A ./api/plan-guru64-split-dft.c Chris@10: M ./api/Makefile.am -1 +7 Chris@10: M ./api/api.h +3 Chris@10: M ./api/fftw3.h -2 +56 Chris@10: M ./api/guru.h +4 Chris@10: M ./api/guru64.h +4 Chris@10: M ./api/mktensor-iodims.c -62 +2 Chris@10: M ./api/mktensor-iodims.h +62 Chris@10: M ./api/mktensor-iodims64.c +2 Chris@10: M ./api/plan-guru-dft-c2r.c -44 +2 Chris@10: M ./api/plan-guru-dft-c2r.h +44 Chris@10: M ./api/plan-guru-dft-r2c.c -43 +2 Chris@10: M ./api/plan-guru-dft-r2c.h +43 Chris@10: M ./api/plan-guru-dft.c -44 +2 Chris@10: M ./api/plan-guru-dft.h +44 Chris@10: M ./api/plan-guru-r2r.c -47 +2 Chris@10: M ./api/plan-guru-r2r.h +47 Chris@10: M ./api/plan-guru-split-dft-c2r.c -40 +2 Chris@10: M ./api/plan-guru-split-dft-c2r.h +40 Chris@10: M ./api/plan-guru-split-dft-r2c.c -39 +2 Chris@10: M ./api/plan-guru-split-dft-r2c.h +39 Chris@10: M ./api/plan-guru-split-dft.c -39 +2 Chris@10: M ./api/plan-guru-split-dft.h +39 Chris@10: M ./api/plan-guru64-dft-c2r.c +2 Chris@10: M ./api/plan-guru64-dft-r2c.c +2 Chris@10: M ./api/plan-guru64-dft.c +2 Chris@10: M ./api/plan-guru64-r2r.c +2 Chris@10: M ./api/plan-guru64-split-dft-c2r.c +2 Chris@10: M ./api/plan-guru64-split-dft-r2c.c +2 Chris@10: M ./api/plan-guru64-split-dft.c +2 Chris@10: Chris@10: Mon May 22 16:41:44 EDT 2006 stevenj Chris@10: * [project @ 2006-05-22 20:41:44 by stevenj] Chris@10: added FIXME note Chris@10: Chris@10: M ./m4/acx_pthread.m4 +1 Chris@10: Chris@10: Mon May 22 16:40:30 EDT 2006 stevenj Chris@10: * [project @ 2006-05-22 20:40:30 by stevenj] Chris@10: check for xlc_r in addition to cc_r; thanks to Guy Moebs for the bug report Chris@10: Chris@10: M ./m4/acx_pthread.m4 -3 +3 Chris@10: Chris@10: Fri Apr 21 12:35:25 EDT 2006 stevenj Chris@10: * [project @ 2006-04-21 16:35:25 by stevenj] Chris@10: added note about gcc 4.0.1 on MacOS/Intel Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +4 Chris@10: Chris@10: Thu Apr 20 23:08:42 EDT 2006 stevenj Chris@10: * [project @ 2006-04-21 03:08:42 by stevenj] Chris@10: added code for Core Duo; thanks to Eric Branlund Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +3 Chris@10: Chris@10: Thu Apr 20 20:21:03 EDT 2006 stevenj Chris@10: * [project @ 2006-04-21 00:21:03 by stevenj] Chris@10: fixed failure for -fPIC or for gcc-4 on Apple Intel machines; thanks to Chris@10: Eric Branlund for the bug report Chris@10: Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 -4 +12 Chris@10: Chris@10: Tue Apr 11 20:00:31 EDT 2006 athena Chris@10: * [project @ 2006-04-12 00:00:31 by athena] Chris@10: Use -maltivec when checking for altivec.h. Chris@10: Chris@10: M ./configure.ac -1 +15 Chris@10: Chris@10: Mon Apr 3 15:52:44 EDT 2006 stevenj Chris@10: * [project @ 2006-04-03 19:52:44 by stevenj] Chris@10: note planner overwriting input in planner-flags reference Chris@10: Chris@10: M ./doc/fftw3.texi -1 +6 Chris@10: Chris@10: Tue Mar 28 09:05:26 EST 2006 athena Chris@10: * [project @ 2006-03-28 14:05:26 by athena] Chris@10: FAQ entry about --enable-k7 in 64-bit mode. Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +11 Chris@10: Chris@10: Mon Mar 27 23:41:05 EST 2006 stevenj Chris@10: * [project @ 2006-03-28 04:41:05 by stevenj] Chris@10: sprintf -> snprintf, to avoid (harmless) complaints by users/compilers Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./libbench2/report.c -12 +19 Chris@10: M ./tools/fftw-wisdom.c +4 Chris@10: Chris@10: Mon Mar 27 23:30:22 EST 2006 stevenj Chris@10: * [project @ 2006-03-28 04:30:22 by stevenj] Chris@10: silence compiler warning Chris@10: Chris@10: M ./kernel/align.c -5 +6 Chris@10: Chris@10: Fri Mar 17 09:20:10 EST 2006 athena Chris@10: * [project @ 2006-03-17 14:20:10 by athena] Chris@10: Remove dft/codelets/inplace, add simd/nonportable to list of Chris@10: directories to be compiled on non-unix systems. Chris@10: Chris@10: M ./doc/fftw3.texi -13 +14 Chris@10: Chris@10: Sat Aug 12 23:52:49 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-1-1 Chris@10: Chris@10: Chris@10: Sat Mar 4 16:17:56 EST 2006 stevenj Chris@10: * [project @ 2006-03-04 21:17:56 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 4 16:13:08 EST 2006 stevenj Chris@10: * [project @ 2006-03-04 21:13:08 by stevenj] Chris@10: note that we align the stack ourselves if necessary, with gcc and icc Chris@10: Chris@10: M ./doc/fftw3.texi -8 +10 Chris@10: Chris@10: Sat Mar 4 16:08:16 EST 2006 stevenj Chris@10: * [project @ 2006-03-04 21:08:16 by stevenj] Chris@10: clearer distinction between static and automatic storage in C Chris@10: Chris@10: M ./doc/fftw3.texi -29 +30 Chris@10: Chris@10: Sat Feb 25 20:27:01 EST 2006 stevenj Chris@10: * [project @ 2006-02-26 01:27:01 by stevenj] Chris@10: rm unused var Chris@10: Chris@10: M ./libbench2/verify-lib.c -2 +1 Chris@10: Chris@10: Sat Feb 25 17:30:28 EST 2006 athena Chris@10: * [project @ 2006-02-25 22:30:28 by athena] Chris@10: Improved usage of goto (Dijkstra miserere nostri) Chris@10: Chris@10: M ./libbench2/my-getopt.c -18 +17 Chris@10: Chris@10: Sat Feb 25 14:19:15 EST 2006 stevenj Chris@10: * [project @ 2006-02-25 19:19:15 by stevenj] Chris@10: boilerplate Chris@10: Chris@10: M ./libbench2/my-getopt.h +13 Chris@10: Chris@10: Sat Feb 25 14:14:40 EST 2006 stevenj Chris@10: * [project @ 2006-02-25 19:14:40 by stevenj] Chris@10: update for upcoming 3.1.1 Chris@10: Chris@10: M ./NEWS +17 Chris@10: Chris@10: Sat Feb 25 13:57:34 EST 2006 stevenj Chris@10: * [project @ 2006-02-25 18:57:34 by stevenj] Chris@10: replace obsolete IMPATIENT with MEASURE Chris@10: Chris@10: M ./tools/fftw-wisdom.c -1 +3 Chris@10: M ./tools/fftw_wisdom.1.in -2 +2 Chris@10: Chris@10: Sat Feb 25 13:52:25 EST 2006 stevenj Chris@10: * [project @ 2006-02-25 18:52:25 by stevenj] Chris@10: corrected comment Chris@10: Chris@10: M ./tools/fftw-wisdom.c -1 +1 Chris@10: Chris@10: Sat Feb 25 10:19:26 EST 2006 athena Chris@10: * [project @ 2006-02-25 15:19:26 by athena] Chris@10: -v does not take an argument. Chris@10: Chris@10: M ./tools/fftw-wisdom.c -6 +3 Chris@10: Chris@10: Sat Feb 25 10:17:18 EST 2006 athena Chris@10: * [project @ 2006-02-25 15:17:18 by athena] Chris@10: Obey the unix convention that -ab = -a -b Chris@10: Chris@10: M ./libbench2/my-getopt.c -15 +24 Chris@10: Chris@10: Fri Feb 24 23:13:49 EST 2006 stevenj Chris@10: * [project @ 2006-02-25 04:13:49 by stevenj] Chris@10: minor fixes (return error on unrecognized option) Chris@10: Chris@10: M ./libbench2/bench-main.c -2 +3 Chris@10: M ./libbench2/my-getopt.c +5 Chris@10: M ./tools/fftw-wisdom.c -3 +4 Chris@10: Chris@10: Fri Feb 24 22:46:12 EST 2006 stevenj Chris@10: * [project @ 2006-02-25 03:46:12 by stevenj] Chris@10: ugh Chris@10: Chris@10: M ./tools/fftw-wisdom.c -34 +30 Chris@10: Chris@10: Fri Feb 24 21:42:56 EST 2006 athena Chris@10: * [project @ 2006-02-25 02:42:56 by athena] Chris@10: require exact match for long options. Chris@10: Chris@10: M ./libbench2/my-getopt.c -1 +2 Chris@10: Chris@10: Fri Feb 24 21:38:02 EST 2006 athena Chris@10: * [project @ 2006-02-25 02:38:02 by athena] Chris@10: better fix Chris@10: Chris@10: M ./libbench2/my-getopt.c -1 +2 Chris@10: Chris@10: Fri Feb 24 21:37:06 EST 2006 athena Chris@10: * [project @ 2006-02-25 02:37:06 by athena] Chris@10: Fix Chris@10: Chris@10: M ./libbench2/my-getopt.c -1 +1 Chris@10: Chris@10: Fri Feb 24 21:25:48 EST 2006 athena Chris@10: * [project @ 2006-02-25 02:25:48 by athena] Chris@10: nothing Chris@10: Chris@10: A ./libbench2/my-getopt.c Chris@10: A ./libbench2/my-getopt.h Chris@10: M ./libbench2/Makefile.am -5 +4 Chris@10: M ./libbench2/bench-main.c -53 +48 Chris@10: M ./libbench2/bench.h -5 +1 Chris@10: R ./libbench2/getopt-utils.c Chris@10: R ./libbench2/getopt.c Chris@10: R ./libbench2/getopt.h Chris@10: R ./libbench2/getopt1.c Chris@10: M ./libbench2/my-getopt.c +153 Chris@10: M ./libbench2/my-getopt.h +33 Chris@10: Chris@10: Mon Feb 20 17:37:21 EST 2006 stevenj Chris@10: * [project @ 2006-02-20 22:37:21 by stevenj] Chris@10: rm transpose-indirect-inplace solver, which was buggy Chris@10: Chris@10: M ./dft/indirect-transpose.c -54 +11 Chris@10: Chris@10: Wed Feb 15 08:43:05 EST 2006 athena Chris@10: * [project @ 2006-02-15 13:43:05 by athena] Chris@10: Comment fix. Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Wed Feb 15 08:18:41 EST 2006 athena Chris@10: * [project @ 2006-02-15 13:18:41 by athena] Chris@10: Cycle counter for Visual C++ x86-64, courtesy of Dirk Michaelis Chris@10: Chris@10: M ./kernel/cycle.h -4 +5 Chris@10: Chris@10: Tue Feb 14 19:17:30 EST 2006 stevenj Chris@10: * [project @ 2006-02-15 00:17:30 by stevenj] Chris@10: rfftwnd.png is in builddir Chris@10: Chris@10: M ./doc/Makefile.am -1 +1 Chris@10: Chris@10: Tue Feb 14 19:03:27 EST 2006 stevenj Chris@10: * [project @ 2006-02-15 00:03:27 by stevenj] Chris@10: fixed typo: --enable-portable-binary, not --with Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Mon Feb 13 07:59:06 EST 2006 athena Chris@10: * [project @ 2006-02-13 12:59:06 by athena] Chris@10: estimator tweaks. Chris@10: Chris@10: M ./dft/dftw-direct.c -2 +3 Chris@10: M ./rdft/hc2hc-direct.c -2 +2 Chris@10: Chris@10: Sun Feb 12 20:43:39 EST 2006 athena Chris@10: * [project @ 2006-02-13 01:43:39 by athena] Chris@10: sse/sse2 support for t3?v codelets Chris@10: Chris@10: M ./simd/simd-sse.h +22 Chris@10: M ./simd/simd-sse2.h -8 +20 Chris@10: Chris@10: Sun Feb 12 20:39:22 EST 2006 athena Chris@10: * [project @ 2006-02-13 01:39:22 by athena] Chris@10: Use CEXP instead of SIN/COS. Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Sun Feb 12 20:12:10 EST 2006 athena Chris@10: * [project @ 2006-02-13 01:12:10 by athena] Chris@10: bug in randomized cse eliminator. Chris@10: Chris@10: M ./genfft/oracle.ml -3 +5 Chris@10: Chris@10: Sun Feb 12 18:34:12 EST 2006 athena Chris@10: * [project @ 2006-02-12 23:34:12 by athena] Chris@10: Added support for t2-style simd codelets. This is altivec only for Chris@10: now; sse/sse2 don't even compile yet. Chris@10: Chris@10: A ./dft/simd/t3b.h Chris@10: A ./dft/simd/t3f.h Chris@10: M ./dft/simd/Makefile.am -2 +2 Chris@10: M ./dft/simd/codelets/Makefile.am -2 +11 Chris@10: M ./dft/simd/t3b.h +34 Chris@10: M ./dft/simd/t3f.h +34 Chris@10: M ./genfft/algsimp.ml -4 +26 Chris@10: M ./genfft/annotate.ml -5 +3 Chris@10: M ./genfft/c.ml -10 +9 Chris@10: M ./genfft/c.mli -2 +2 Chris@10: M ./genfft/complex.ml -93 +6 Chris@10: M ./genfft/complex.mli -5 +3 Chris@10: M ./genfft/expr.ml -6 +15 Chris@10: M ./genfft/expr.mli -2 +4 Chris@10: M ./genfft/gen_athtw.ml -3 +3 Chris@10: M ./genfft/gen_conv.ml -5 +3 Chris@10: M ./genfft/gen_hc2hc.ml -4 +3 Chris@10: M ./genfft/gen_hc2r.ml -5 +3 Chris@10: M ./genfft/gen_mdct.ml -7 +3 Chris@10: M ./genfft/gen_notw.ml -5 +2 Chris@10: M ./genfft/gen_notw_c.ml -5 +2 Chris@10: M ./genfft/gen_r2hc.ml -4 +3 Chris@10: M ./genfft/gen_r2r.ml -4 +3 Chris@10: M ./genfft/gen_twiddle.ml -3 +3 Chris@10: M ./genfft/gen_twiddle_c.ml -10 +4 Chris@10: M ./genfft/gen_twidsq.ml -7 +3 Chris@10: M ./genfft/gen_twidsq_c.ml -15 +5 Chris@10: M ./genfft/magic.ml -4 +4 Chris@10: M ./genfft/oracle.ml -1 +5 Chris@10: M ./genfft/schedule.ml -3 +3 Chris@10: M ./genfft/simd.ml -5 +11 Chris@10: M ./genfft/to_alist.ml -4 +16 Chris@10: M ./genfft/trig.ml -3 +2 Chris@10: M ./genfft/twiddle.ml -265 +35 Chris@10: M ./genfft/twiddle.mli -4 +4 Chris@10: M ./simd/simd-altivec.h +28 Chris@10: Chris@10: Sun Feb 12 15:30:27 EST 2006 athena Chris@10: * [project @ 2006-02-12 20:30:27 by athena] Chris@10: Added support for t2-style simd split-complex codelets. Chris@10: Chris@10: A ./dft/simd/ts.c Chris@10: A ./dft/simd/ts.h Chris@10: M ./dft/simd/Makefile.am -2 +2 Chris@10: M ./dft/simd/codelets/Makefile.am -2 +7 Chris@10: R ./dft/simd/t1s.c Chris@10: R ./dft/simd/t1s.h Chris@10: M ./dft/simd/ts.c +50 Chris@10: M ./dft/simd/ts.h +33 Chris@10: M ./genfft/twiddle.ml -5 +3 Chris@10: Chris@10: Fri Feb 10 18:21:28 EST 2006 stevenj Chris@10: * [project @ 2006-02-10 23:21:28 by stevenj] Chris@10: Chris@10: M ./m4/ax_openmp.m4 -1 +1 Chris@10: Chris@10: Fri Feb 10 18:19:46 EST 2006 stevenj Chris@10: * [project @ 2006-02-10 23:19:46 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./m4/ax_openmp.m4 -2 +2 Chris@10: Chris@10: Fri Feb 10 18:00:35 EST 2006 stevenj Chris@10: * [project @ 2006-02-10 23:00:35 by stevenj] Chris@10: windows DLL stuff for Fortran interface Chris@10: Chris@10: M ./api/f77api.c +7 Chris@10: M ./api/f77funcs.h -43 +43 Chris@10: Chris@10: Fri Feb 10 09:48:52 EST 2006 athena Chris@10: * [project @ 2006-02-10 14:48:52 by athena] Chris@10: Bumped version to 3.1.1 Chris@10: Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Fri Feb 10 09:18:39 EST 2006 athena Chris@10: * [project @ 2006-02-10 14:18:39 by athena] Chris@10: Precompute array indices on x86-64. Speeds up Pentium IV and makes no Chris@10: appreciable difference on AMD. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Tue Feb 7 22:01:36 EST 2006 athena Chris@10: * [project @ 2006-02-08 03:01:36 by athena] Chris@10: Check whether the processor supports CPUID before issuing the Chris@10: instruction. (Grrr...) Code contributed by Eric J. Korpela. Chris@10: Chris@10: A ./simd/x86-cpuid.h Chris@10: M ./simd/Makefile.am -1 +1 Chris@10: M ./simd/sse.c -24 +6 Chris@10: M ./simd/sse2.c -26 +8 Chris@10: M ./simd/x86-cpuid.h +127 Chris@10: Chris@10: Tue Feb 7 21:36:47 EST 2006 athena Chris@10: * [project @ 2006-02-08 02:36:47 by athena] Chris@10: icc supports x86_64 these days. Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Sun Feb 5 18:19:55 EST 2006 athena Chris@10: * [project @ 2006-02-05 23:19:55 by athena] Chris@10: Paranoia. Chris@10: Chris@10: M ./kernel/primes.c -2 +2 Chris@10: Chris@10: Mon Jan 30 15:27:53 EST 2006 stevenj Chris@10: * [project @ 2006-01-30 20:27:53 by stevenj] Chris@10: whoops, fixed assert (y <= x) Chris@10: Chris@10: M ./kernel/primes.c -2 +2 Chris@10: Chris@10: Mon Jan 30 15:26:22 EST 2006 stevenj Chris@10: * [project @ 2006-01-30 20:26:22 by stevenj] Chris@10: note that safe_mulmod requires {x,y} < p (or at least < 2p), and added Chris@10: assert Chris@10: Chris@10: M ./kernel/primes.c -2 +4 Chris@10: Chris@10: Mon Jan 30 11:09:32 EST 2006 athena Chris@10: * [project @ 2006-01-30 16:09:32 by athena] Chris@10: fixed aix/xlc lossage Chris@10: Chris@10: M ./libbench2/bench-user.h -2 +2 Chris@10: M ./libbench2/timer.c -4 +4 Chris@10: Chris@10: Sun Jan 29 20:42:51 EST 2006 athena Chris@10: * [project @ 2006-01-30 01:42:51 by athena] Chris@10: In the impuse test, normalize the impulse so that the impulse and the Chris@10: random vectors have roughly the same L2 norm. This change reduces the Chris@10: number of bits that we lose because of floating-point cancellation, so Chris@10: that we can focus on the bits that we lose because of bugs. Chris@10: Chris@10: M ./libbench2/verify-lib.c -3 +3 Chris@10: Chris@10: Sun Jan 29 20:37:47 EST 2006 athena Chris@10: * [project @ 2006-01-30 01:37:47 by athena] Chris@10: Compute omega in trigreal precision, as opposed to R. Chris@10: Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: Chris@10: Sat Aug 12 23:51:14 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-1 Chris@10: Chris@10: Chris@10: Fri Jan 27 19:16:22 EST 2006 stevenj Chris@10: * [project @ 2006-01-28 00:16:22 by stevenj] Chris@10: add --with-combined-threads option as workaround to Windows inability to build shared libs with dependencies Chris@10: Chris@10: M ./Makefile.am -1 +7 Chris@10: M ./configure.ac -1 +3 Chris@10: M ./tests/Makefile.am +2 Chris@10: M ./threads/Makefile.am +4 Chris@10: M ./tools/Makefile.am +2 Chris@10: Chris@10: Fri Jan 27 17:20:45 EST 2006 stevenj Chris@10: * [project @ 2006-01-27 22:20:45 by stevenj] Chris@10: libfftw3_threads should *not* used -no-undefined because, in fact, it is not true -- this library depends on -lfftw3, and is not self-contained Chris@10: Chris@10: M ./threads/Makefile.am -1 +1 Chris@10: Chris@10: Thu Jan 26 22:04:34 EST 2006 stevenj Chris@10: * [project @ 2006-01-27 03:04:34 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./NEWS -3 +10 Chris@10: Chris@10: Thu Jan 26 21:10:50 EST 2006 athena Chris@10: * [project @ 2006-01-27 02:10:50 by athena] Chris@10: Added paranoid stack alignment when awaking plans. While I was at it, Chris@10: removed obsolete, redundant AWAKE macro. Chris@10: Chris@10: M ./api/apiplan.c -3 +8 Chris@10: M ./dft/bluestein.c -1 +1 Chris@10: M ./dft/buffered.c -4 +4 Chris@10: M ./dft/ct.c -3 +3 Chris@10: M ./dft/ctsq.c -2 +2 Chris@10: M ./dft/dftw-generic.c -1 +1 Chris@10: M ./dft/dftw-genericbuf.c -1 +1 Chris@10: M ./dft/indirect-transpose.c -3 +3 Chris@10: M ./dft/indirect.c -3 +3 Chris@10: M ./dft/rader.c -3 +3 Chris@10: M ./dft/rank-geq2.c -3 +3 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: M ./kernel/timer.c -3 +3 Chris@10: M ./rdft/buffered.c -4 +4 Chris@10: M ./rdft/buffered2.c -3 +3 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -3 +3 Chris@10: M ./rdft/hc2hc-direct.c -3 +3 Chris@10: M ./rdft/hc2hc-directbuf.c -3 +3 Chris@10: M ./rdft/hc2hc-generic.c -2 +2 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/indirect.c -3 +3 Chris@10: M ./rdft/rank-geq2-rdft2.c -3 +3 Chris@10: M ./rdft/rank-geq2.c -3 +3 Chris@10: M ./rdft/rank0-rdft2.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -4 +4 Chris@10: M ./reodft/redft00e-r2hc-pad.c -3 +3 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft00e-splitradix.c -3 +3 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -3 +3 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./tests/hook.c -2 +2 Chris@10: M ./threads/ct.c -3 +3 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/hc2hc.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: Chris@10: Thu Jan 26 20:54:39 EST 2006 athena Chris@10: * [project @ 2006-01-27 01:54:39 by athena] Chris@10: Updated for 3.1. Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Thu Jan 26 19:15:12 EST 2006 athena Chris@10: * [project @ 2006-01-27 00:15:12 by athena] Chris@10: ditched one alignment check and noted that we should eliminate the rest as well Chris@10: Chris@10: M ./TODO +3 Chris@10: M ./libbench2/bench-main.c -13 +1 Chris@10: Chris@10: Wed Jan 25 23:02:19 EST 2006 athena Chris@10: * [project @ 2006-01-26 04:02:19 by athena] Chris@10: alignment hack Chris@10: Chris@10: M ./libbench2/bench-main.c -2 +4 Chris@10: Chris@10: Wed Jan 25 22:05:11 EST 2006 athena Chris@10: * [project @ 2006-01-26 03:05:11 by athena] Chris@10: detect pentium M Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +2 Chris@10: Chris@10: Wed Jan 25 18:42:58 EST 2006 stevenj Chris@10: * [project @ 2006-01-25 23:42:58 by stevenj] Chris@10: don't trust host_cpu if it claims we are on i386/i486, and call cpuid anyway (if it fails we use no arch flag). This is needed on FreeBSD Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +1 Chris@10: Chris@10: Wed Jan 25 18:00:04 EST 2006 stevenj Chris@10: * [project @ 2006-01-25 23:00:04 by stevenj] Chris@10: suggest --with-our-malloc16 in error message Chris@10: Chris@10: M ./kernel/kalloc.c -2 +2 Chris@10: Chris@10: Tue Jan 24 19:53:34 EST 2006 stevenj Chris@10: * [project @ 2006-01-25 00:53:34 by stevenj] Chris@10: ditto for -no-gcc Chris@10: Chris@10: M ./configure.ac -2 +4 Chris@10: Chris@10: Tue Jan 24 19:51:08 EST 2006 stevenj Chris@10: * [project @ 2006-01-25 00:51:08 by stevenj] Chris@10: flags required for successfull compilation should be added even if the Chris@10: user overrides CFLAGS Chris@10: Chris@10: M ./configure.ac -9 +12 Chris@10: Chris@10: Tue Jan 24 18:43:59 EST 2006 stevenj Chris@10: * [project @ 2006-01-24 23:43:59 by stevenj] Chris@10: upcoming gcc OpenMP support uses -fopenmp Chris@10: Chris@10: M ./m4/ax_openmp.m4 -4 +4 Chris@10: Chris@10: Tue Jan 24 18:26:59 EST 2006 stevenj Chris@10: * [project @ 2006-01-24 23:26:59 by stevenj] Chris@10: note that PGI uses -mp as well Chris@10: Chris@10: M ./m4/ax_openmp.m4 -1 +1 Chris@10: Chris@10: Mon Jan 23 15:31:24 EST 2006 athena Chris@10: * [project @ 2006-01-23 20:31:24 by athena] Chris@10: my best guess at how to fix the microsoft crap du jour Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./simd/sse.c -65 +70 Chris@10: M ./simd/sse2.c -62 +67 Chris@10: Chris@10: Mon Jan 23 14:05:14 EST 2006 stevenj Chris@10: * [project @ 2006-01-23 19:05:14 by stevenj] Chris@10: use -Masmkeyword for PGI cycle counter, grr Chris@10: Chris@10: M ./configure.ac +5 Chris@10: M ./kernel/cycle.h -2 +4 Chris@10: Chris@10: Sun Jan 22 18:09:06 EST 2006 athena Chris@10: * [project @ 2006-01-22 23:09:06 by athena] Chris@10: Bumped version number to 3.1. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Jan 21 10:03:59 EST 2006 athena Chris@10: * [project @ 2006-01-21 15:03:59 by athena] Chris@10: Report that --enable-k7 is incompatible with --enable-shared. Chris@10: Chris@10: M ./configure.ac +3 Chris@10: Chris@10: Sat Jan 21 09:17:54 EST 2006 athena Chris@10: * [project @ 2006-01-21 14:17:54 by athena] Chris@10: Do not use empty libraries in LIBADD, since otherwise the linker fails Chris@10: on Solaris. Chris@10: Chris@10: M ./Makefile.am -7 +20 Chris@10: Chris@10: Wed Jan 18 10:47:59 EST 2006 stevenj Chris@10: * [project @ 2006-01-18 15:47:59 by stevenj] Chris@10: warn end-users away from this file Chris@10: Chris@10: M ./bootstrap.sh -1 +10 Chris@10: Chris@10: Tue Jan 17 16:16:42 EST 2006 athena Chris@10: * [project @ 2006-01-17 21:16:42 by athena] Chris@10: Gcc sucks. Chris@10: Chris@10: M ./simd/simd-sse.h -1 +2 Chris@10: Chris@10: Tue Jan 17 11:48:55 EST 2006 athena Chris@10: * [project @ 2006-01-17 16:48:55 by athena] Chris@10: Disabled checks that may turn out to be too paranoid. Chris@10: Chris@10: M ./tests/hook.c -1 +2 Chris@10: Chris@10: Tue Jan 17 10:35:03 EST 2006 athena Chris@10: * [project @ 2006-01-17 15:35:03 by athena] Chris@10: Some paranoid checks. Chris@10: Chris@10: M ./tests/hook.c +16 Chris@10: Chris@10: Tue Jan 17 09:31:08 EST 2006 athena Chris@10: * [project @ 2006-01-17 14:31:08 by athena] Chris@10: Flush stdout after printing. Chris@10: Chris@10: M ./libbench2/ovtpvt.c +1 Chris@10: Chris@10: Tue Jan 17 08:28:18 EST 2006 athena Chris@10: * [project @ 2006-01-17 13:28:18 by athena] Chris@10: Run the leak detector in all cases, not just when verbose > 2. Chris@10: Chris@10: M ./kernel/alloc.c -2 +7 Chris@10: M ./tests/bench.c -2 +1 Chris@10: Chris@10: Tue Jan 17 08:11:41 EST 2006 athena Chris@10: * [project @ 2006-01-17 13:11:41 by athena] Chris@10: Eliminate calls to pow(), rint(). Chris@10: Chris@10: M ./api/mapflags.c -8 +3 Chris@10: Chris@10: Tue Jan 17 00:45:06 EST 2006 stevenj Chris@10: * [project @ 2006-01-17 05:45:06 by stevenj] Chris@10: put # in first column, for stylistic consistency Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: Chris@10: Tue Jan 17 00:17:27 EST 2006 athena Chris@10: * [project @ 2006-01-17 05:17:27 by athena] Chris@10: Made timeout part of impatience flags, in order to improve the Chris@10: usability of wisdom. Also, fixed bogus error recovery logic in Chris@10: planner.c:imprt(). Chris@10: Chris@10: M ./api/mapflags.c -1 +36 Chris@10: M ./kernel/ifftw.h -4 +7 Chris@10: M ./kernel/planner.c -23 +70 Chris@10: Chris@10: Mon Jan 16 23:03:34 EST 2006 stevenj Chris@10: * [project @ 2006-01-17 04:03:33 by stevenj] Chris@10: make timelimit < 0 .eq. FFTW_NO_TIMELIMIT Chris@10: Chris@10: M ./api/apiplan.c -1 +1 Chris@10: M ./api/fftw3.h -1 +3 Chris@10: M ./doc/fftw3.texi -5 +9 Chris@10: M ./kernel/planner.c -3 +3 Chris@10: Chris@10: Mon Jan 16 21:52:01 EST 2006 athena Chris@10: * [project @ 2006-01-17 02:52:01 by athena] Chris@10: Eliminated the FFTW_TIMELIMIT flag in favor of this simpler logic: Chris@10: fftw_set_timelimit(0) disables time limit. Chris@10: fftw_set_timelimit(X), X>0 sets the time limit to X. Chris@10: Chris@10: M ./api/apiplan.c -6 +1 Chris@10: M ./api/fftw3.h -2 +1 Chris@10: M ./api/the-planner.c +7 Chris@10: M ./doc/fftw3.texi -30 +27 Chris@10: M ./kernel/planner.c -2 +4 Chris@10: M ./tests/bench.c -1 Chris@10: Chris@10: Mon Jan 16 08:38:04 EST 2006 athena Chris@10: * [project @ 2006-01-16 13:38:04 by athena] Chris@10: Force the use of the estimator when wisdom fails because of md5 Chris@10: collisions, otherwise the planner takes forever. Chris@10: Chris@10: M ./api/apiplan.c -2 +10 Chris@10: Chris@10: Sun Jan 15 21:30:31 EST 2006 athena Chris@10: * [project @ 2006-01-16 02:30:31 by athena] Chris@10: Ranted about how broken gcc-4 is. Chris@10: Chris@10: M ./kernel/ifftw.h -5 +40 Chris@10: Chris@10: Sun Jan 15 19:59:38 EST 2006 stevenj Chris@10: * [project @ 2006-01-16 00:59:38 by stevenj] Chris@10: change fftw_timelimit global var to fftw_set_timelimit(double) function, for simpler usage with shared libraries and for consistency with e.g. set_numthreads Chris@10: Chris@10: M ./api/apiplan.c -2 +4 Chris@10: M ./api/fftw3.h -2 +3 Chris@10: M ./doc/fftw3.texi -11 +11 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Jan 15 19:32:27 EST 2006 athena Chris@10: * [project @ 2006-01-16 00:32:27 by athena] Chris@10: Minor tweaks. Chris@10: Chris@10: M ./doc/fftw3.texi -8 +11 Chris@10: Chris@10: Sun Jan 15 16:32:54 EST 2006 athena Chris@10: * [project @ 2006-01-15 21:32:54 by athena] Chris@10: tweaks to make sure that time_n() is always called from the same stack position. Chris@10: Chris@10: M ./libbench2/timer.c -12 +12 Chris@10: Chris@10: Sun Jan 15 16:09:53 EST 2006 athena Chris@10: * [project @ 2006-01-15 21:09:53 by athena] Chris@10: Major simplification of the timer calibration logic. Also, use an FFT Chris@10: as a unit of work instead of the old pointer chasing, because God Chris@10: knows how pointer chasing interacts with the idiotic cache-hit Chris@10: speculation on the Pentium IV. Chris@10: Chris@10: M ./libbench2/bench.h -2 +3 Chris@10: M ./libbench2/speed.c -5 +8 Chris@10: M ./libbench2/timer.c -96 +35 Chris@10: M ./libbench2/timer2.c -7 +99 Chris@10: Chris@10: Sun Jan 15 15:12:08 EST 2006 athena Chris@10: * [project @ 2006-01-15 20:12:08 by athena] Chris@10: Fixed broken aligment checks when sizeof(R)==12. Chris@10: Chris@10: M ./kernel/align.c -3 +8 Chris@10: Chris@10: Sun Jan 15 10:36:40 EST 2006 athena Chris@10: * [project @ 2006-01-15 15:36:40 by athena] Chris@10: Manual unrolling of loop. Chris@10: Chris@10: M ./libbench2/timer2.c -5 +2 Chris@10: Chris@10: Sun Jan 15 10:12:55 EST 2006 athena Chris@10: * [project @ 2006-01-15 15:12:55 by athena] Chris@10: Various improvements to timer calibration routines. Chris@10: Chris@10: A ./libbench2/timer2.c Chris@10: M ./libbench2/Makefile.am -1 +1 Chris@10: M ./libbench2/bench.h -1 +2 Chris@10: M ./libbench2/timer.c -18 +32 Chris@10: M ./libbench2/timer2.c +36 Chris@10: Chris@10: Sat Jan 14 22:16:09 EST 2006 athena Chris@10: * [project @ 2006-01-15 03:16:09 by athena] Chris@10: cygwin defines __CYGWIN__, not __WIN32__ etc. Chris@10: Chris@10: M ./libbench2/timer.c -16 +17 Chris@10: Chris@10: Sat Jan 14 20:40:12 EST 2006 athena Chris@10: * [project @ 2006-01-15 01:40:12 by athena] Chris@10: fixed confusion between libbench and user timers Chris@10: Chris@10: M ./libbench2/bench-user.h -3 +7 Chris@10: M ./libbench2/speed.c -5 +5 Chris@10: M ./libbench2/timer.c -6 +8 Chris@10: M ./tests/bench.c -8 +8 Chris@10: Chris@10: Sat Jan 14 12:32:44 EST 2006 stevenj Chris@10: * [project @ 2006-01-14 17:32:44 by stevenj] Chris@10: update Chris@10: Chris@10: M ./NEWS -2 +4 Chris@10: Chris@10: Sat Jan 14 10:24:11 EST 2006 athena Chris@10: * [project @ 2006-01-14 15:24:11 by athena] Chris@10: Comment. Chris@10: Chris@10: M ./simd/simd-sse.h +5 Chris@10: Chris@10: Sat Jan 14 10:19:28 EST 2006 athena Chris@10: * [project @ 2006-01-14 15:19:28 by athena] Chris@10: Workaround gcc bug. Chris@10: Chris@10: M ./simd/simd-sse.h -2 +3 Chris@10: Chris@10: Fri Jan 13 19:13:18 EST 2006 athena Chris@10: * [project @ 2006-01-14 00:13:18 by athena] Chris@10: Switched to -beta2. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Thu Jan 12 22:21:57 EST 2006 athena Chris@10: * [project @ 2006-01-13 03:21:57 by athena] Chris@10: Fixed technically correct but highly obfuscated use of the enum tag Chris@10: R2HC as a null pointer. Chris@10: Chris@10: M ./rdft/buffered.c -5 +3 Chris@10: M ./rdft/indirect.c -4 +3 Chris@10: M ./rdft/problem.c -1 +8 Chris@10: M ./rdft/rank0-rdft2.c -4 +3 Chris@10: M ./rdft/rdft.h +1 Chris@10: M ./rdft/vrank3-transpose.c -25 +13 Chris@10: Chris@10: Thu Jan 12 19:25:20 EST 2006 stevenj Chris@10: * [project @ 2006-01-13 00:25:20 by stevenj] Chris@10: --enable-unsafe-mulmod is obsolete Chris@10: Chris@10: M ./configure.ac -4 Chris@10: Chris@10: Thu Jan 12 19:23:18 EST 2006 athena Chris@10: * [project @ 2006-01-13 00:23:18 by athena] Chris@10: More thoughts. Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Thu Jan 12 19:17:57 EST 2006 athena Chris@10: * [project @ 2006-01-13 00:17:57 by athena] Chris@10: Removed loop unrolling because it slows things down on at least one Chris@10: powerpc and it generates clumsy x86 code. Chris@10: Chris@10: M ./rdft/buffered2.c -50 +12 Chris@10: Chris@10: Thu Jan 12 19:17:35 EST 2006 stevenj Chris@10: * [project @ 2006-01-13 00:17:35 by stevenj] Chris@10: tweaks Chris@10: Chris@10: M ./kernel/kalloc.c -6 +6 Chris@10: Chris@10: Thu Jan 12 15:55:52 EST 2006 stevenj Chris@10: * [project @ 2006-01-12 20:55:52 by stevenj] Chris@10: MacOSX x86 ABI specifies that the stack is kept 16-byte aligned Chris@10: Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: Chris@10: Thu Jan 12 12:46:49 EST 2006 athena Chris@10: * [project @ 2006-01-12 17:46:49 by athena] Chris@10: ``ret'' is a reserved word in the evil empire. Chris@10: Chris@10: M ./kernel/cycle.h -5 +5 Chris@10: Chris@10: Thu Jan 12 08:31:43 EST 2006 athena Chris@10: * [project @ 2006-01-12 13:31:09 by athena] Chris@10: Changed ret => result because ret ``is a reserved word'' in the evil Chris@10: empire. Chris@10: Chris@10: M ./simd/sse.c -4 +4 Chris@10: M ./simd/sse2.c -4 +4 Chris@10: Chris@10: Wed Jan 11 19:30:42 EST 2006 athena Chris@10: * [project @ 2006-01-12 00:30:42 by athena] Chris@10: Workaround Visual c++ lossage. Chris@10: Chris@10: M ./simd/simd-sse2.h -5 +4 Chris@10: Chris@10: Wed Jan 11 19:26:16 EST 2006 athena Chris@10: * [project @ 2006-01-12 00:26:16 by athena] Chris@10: Workaround visual c++ lossage. Chris@10: Chris@10: M ./simd/simd-sse.h +38 Chris@10: Chris@10: Wed Jan 11 19:10:52 EST 2006 athena Chris@10: * [project @ 2006-01-12 00:10:52 by athena] Chris@10: isprint() is guaranteed to work for unsigned char + EOF only. Chris@10: Chris@10: M ./libbench2/getopt-utils.c -2 +3 Chris@10: Chris@10: Wed Jan 11 13:47:49 EST 2006 stevenj Chris@10: * [project @ 2006-01-11 18:47:49 by stevenj] Chris@10: rm obsolete fixme Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Wed Jan 11 13:38:46 EST 2006 stevenj Chris@10: * [project @ 2006-01-11 18:38:46 by stevenj] Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Wed Jan 11 13:32:26 EST 2006 stevenj Chris@10: * [project @ 2006-01-11 18:32:26 by stevenj] Chris@10: fix comment Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -2 +3 Chris@10: Chris@10: Wed Jan 11 12:27:05 EST 2006 athena Chris@10: * [project @ 2006-01-11 17:27:05 by athena] Chris@10: Paranoid use of K(x) for all constants x, to avoid runtime double->float conversions on sufficiently stupid compilers. Chris@10: Chris@10: M ./dft/bluestein.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/rank0-rdft2.c -11 +11 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -3 +3 Chris@10: Chris@10: Tue Jan 10 20:10:38 EST 2006 athena Chris@10: * [project @ 2006-01-11 01:10:38 by athena] Chris@10: Workaround to gcc nonsense. Chris@10: Chris@10: M ./simd/simd-sse.h -1 +15 Chris@10: Chris@10: Tue Jan 10 18:44:28 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 23:44:28 by stevenj] Chris@10: bug fix: infinite loop in transpose-cut planning Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -6 +14 Chris@10: Chris@10: Tue Jan 10 18:12:14 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 23:12:14 by stevenj] Chris@10: clarified comment Chris@10: Chris@10: M ./api/fftw3.h -6 +6 Chris@10: Chris@10: Tue Jan 10 18:10:32 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 23:10:32 by stevenj] Chris@10: more Windows decorations Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Tue Jan 10 17:57:45 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 22:57:45 by stevenj] Chris@10: added FIXME comment Chris@10: Chris@10: M ./support/Makefile.codelets +1 Chris@10: Chris@10: Tue Jan 10 17:52:07 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 22:52:07 by stevenj] Chris@10: 'make clean' should not delete codlist.c since it is included in the dist tarball Chris@10: Chris@10: M ./support/Makefile.codelets -2 +3 Chris@10: Chris@10: Tue Jan 10 17:50:12 EST 2006 athena Chris@10: * [project @ 2006-01-10 22:50:12 by athena] Chris@10: Change threshold for ``large'' Cooley-Tukey to 256K from 64K, since it Chris@10: seems to benefit the Pentium IV with sse and the planning cost is not Chris@10: too horrible. Chris@10: Chris@10: M ./dft/dftw-direct.c -2 +2 Chris@10: Chris@10: Tue Jan 10 17:45:11 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 22:45:11 by stevenj] Chris@10: more missing Windows DLL decorations Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: Chris@10: Tue Jan 10 17:41:28 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 22:41:28 by stevenj] Chris@10: remove unused var Chris@10: Chris@10: M ./rdft/dht-rader.c -2 Chris@10: Chris@10: Tue Jan 10 14:00:50 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 19:00:50 by stevenj] Chris@10: allow compiler threads, if enabled, to take precedence over explicit threads Chris@10: Chris@10: M ./threads/threads.c -17 +17 Chris@10: Chris@10: Tue Jan 10 12:30:09 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 17:30:09 by stevenj] Chris@10: Chris@10: M ./api/api.h -2 +2 Chris@10: Chris@10: Tue Jan 10 12:21:56 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 17:21:56 by stevenj] Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Tue Jan 10 09:13:20 EST 2006 athena Chris@10: * [project @ 2006-01-10 14:13:20 by athena] Chris@10: Fixed comment typo. Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Tue Jan 10 08:59:22 EST 2006 athena Chris@10: * [project @ 2006-01-10 13:59:22 by athena] Chris@10: Rearranged timeout checks so as to eliminate one of them. Chris@10: Chris@10: M ./kernel/planner.c -9 +4 Chris@10: Chris@10: Tue Jan 10 08:56:55 EST 2006 athena Chris@10: * [project @ 2006-01-10 13:56:55 by athena] Chris@10: Converted residual CK() -> A(). Chris@10: Chris@10: M ./kernel/plan.c -2 +2 Chris@10: Chris@10: Tue Jan 10 08:36:13 EST 2006 athena Chris@10: * [project @ 2006-01-10 13:36:13 by athena] Chris@10: Maintain the invariant TIMED_OUT ==> NEED_TIMEOUT_CHECK. Chris@10: Chris@10: M ./kernel/planner.c -5 +10 Chris@10: Chris@10: Tue Jan 10 08:24:41 EST 2006 athena Chris@10: * [project @ 2006-01-10 13:24:41 by athena] Chris@10: silence some 64-bit warnings Chris@10: Chris@10: M ./api/mapflags.c -1 +1 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/buffered.c -1 +1 Chris@10: M ./kernel/md5.c -1 +1 Chris@10: M ./kernel/scan.c -2 +3 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: Chris@10: Tue Jan 10 07:58:48 EST 2006 athena Chris@10: * [project @ 2006-01-10 12:58:48 by athena] Chris@10: Assertions. Chris@10: Chris@10: M ./tests/hook.c +3 Chris@10: Chris@10: Tue Jan 10 00:14:00 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 05:14:00 by stevenj] Chris@10: some condensing Chris@10: Chris@10: M ./kernel/timer.c -10 +5 Chris@10: Chris@10: Tue Jan 10 00:03:32 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 05:03:32 by stevenj] Chris@10: eliminate X(seconds) in favor of X(elapsed_since), in paranoia of clock wrap Chris@10: Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./kernel/ifftw.h -3 +33 Chris@10: M ./kernel/planner.c -4 +4 Chris@10: M ./kernel/timer.c -58 +25 Chris@10: Chris@10: Mon Jan 9 23:21:21 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 04:21:21 by stevenj] Chris@10: Chris@10: M ./kernel/timer.c -2 +2 Chris@10: Chris@10: Mon Jan 9 23:21:06 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 04:21:06 by stevenj] Chris@10: hmm, a bit more pessimistic about clock wrapping Chris@10: Chris@10: M ./kernel/timer.c -4 +4 Chris@10: Chris@10: Mon Jan 9 23:20:26 EST 2006 athena Chris@10: * [project @ 2006-01-10 04:20:26 by athena] Chris@10: Revert to md5uint = unsigned int whenever possible, so as to Chris@10: avoid wasting space for unsigned long on 64-bit machines. Chris@10: Chris@10: M ./configure.ac +2 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: Chris@10: Mon Jan 9 23:12:27 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 04:12:27 by stevenj] Chris@10: note why clock() wrap should not be a concern Chris@10: Chris@10: M ./kernel/timer.c -1 +6 Chris@10: Chris@10: Mon Jan 9 22:57:16 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 03:57:16 by stevenj] Chris@10: bugfix in recent timeout changes - check for case where last solver times out Chris@10: Chris@10: M ./kernel/planner.c -2 +8 Chris@10: Chris@10: Mon Jan 9 22:40:26 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 03:40:26 by stevenj] Chris@10: started changes list from beta Chris@10: Chris@10: M ./NEWS -1 +11 Chris@10: Chris@10: Mon Jan 9 22:34:13 EST 2006 athena Chris@10: * [project @ 2006-01-10 03:34:13 by athena] Chris@10: Paranoia. Chris@10: Chris@10: M ./api/mapflags.c +4 Chris@10: Chris@10: Mon Jan 9 22:27:37 EST 2006 athena Chris@10: * [project @ 2006-01-10 03:27:37 by athena] Chris@10: Paranoid assertions. Chris@10: Chris@10: M ./kernel/planner.c -3 +7 Chris@10: Chris@10: Mon Jan 9 22:13:32 EST 2006 athena Chris@10: * [project @ 2006-01-10 03:13:32 by athena] Chris@10: Added FIXME comment stating the 64-bit uncleaniness of Chris@10: fftw_tensor_to_bench_tensor(). Chris@10: Chris@10: M ./tests/hook.c +1 Chris@10: Chris@10: Mon Jan 9 22:06:05 EST 2006 athena Chris@10: * [project @ 2006-01-10 03:06:05 by athena] Chris@10: Another 64-bit bug. Chris@10: Chris@10: M ./dft/simd/t.c -1 +1 Chris@10: Chris@10: Mon Jan 9 21:54:07 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 02:54:07 by stevenj] Chris@10: more Windows DLL nonsense Chris@10: Chris@10: M ./api/api.h -5 +15 Chris@10: M ./kernel/ifftw.h -3 +5 Chris@10: M ./tests/hook.c +2 Chris@10: Chris@10: Mon Jan 9 21:18:25 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 02:18:25 by stevenj] Chris@10: some additional dllexport tags required to build the test program, due to internal stuff called by hook.c Chris@10: Chris@10: M ./api/api.h -2 +3 Chris@10: M ./kernel/ifftw.h -6 +17 Chris@10: Chris@10: Mon Jan 9 20:31:15 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:31:15 by stevenj] Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Mon Jan 9 20:30:19 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:30:19 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./api/fftw3.h -4 +6 Chris@10: Chris@10: Mon Jan 9 20:20:28 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:17:11 by stevenj] Chris@10: Chris@10: M ./api/api.h -7 +7 Chris@10: M ./api/fftw3.h -4 +3 Chris@10: Chris@10: Mon Jan 9 20:16:50 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:16:50 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: Chris@10: Mon Jan 9 20:12:23 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:12:23 by stevenj] Chris@10: define FFTW_DLL if DLL_EXPORT (defined by libtool) is supplied Chris@10: Chris@10: M ./api/api.h +10 Chris@10: Chris@10: Mon Jan 9 20:05:11 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:05:11 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Mon Jan 9 20:00:47 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 01:00:47 by stevenj] Chris@10: another stab at Windows DLL mess Chris@10: Chris@10: M ./api/fftw3.h -172 +185 Chris@10: Chris@10: Mon Jan 9 19:23:42 EST 2006 athena Chris@10: * [project @ 2006-01-10 00:23:42 by athena] Chris@10: 64-bit clean SIMD header file. I missed those because sparse Chris@10: does not know vector types. Grrr... Chris@10: Chris@10: M ./simd/simd-altivec.h -11 +11 Chris@10: M ./simd/simd-sse.h -6 +6 Chris@10: M ./simd/simd-sse2.h -4 +4 Chris@10: Chris@10: Mon Jan 9 19:08:36 EST 2006 stevenj Chris@10: * [project @ 2006-01-10 00:08:36 by stevenj] Chris@10: this option is called AC_DISABLE_SHARED in the documentation Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Mon Jan 9 17:34:13 EST 2006 stevenj Chris@10: * [project @ 2006-01-09 22:34:13 by stevenj] Chris@10: fixed --with-gcc-arch to work when cross-compiling Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +4 Chris@10: Chris@10: Mon Jan 9 12:04:04 EST 2006 athena Chris@10: * [project @ 2006-01-09 17:04:04 by athena] Chris@10: Moved the timeout check back into the search loop, sicut erat in Chris@10: principio. This gives us a precise control over the timeout. To Chris@10: avoid the overhead of X(seconds)(), only call X(seconds)() if some Chris@10: time measurement was taken since the last call to X(seconds)(). Chris@10: Chris@10: M ./api/apiplan.c -11 +1 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -5 +36 Chris@10: Chris@10: Mon Jan 9 00:07:40 EST 2006 stevenj Chris@10: * [project @ 2006-01-09 05:07:40 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -5 +5 Chris@10: Chris@10: Sun Jan 8 23:58:23 EST 2006 stevenj Chris@10: * [project @ 2006-01-09 04:58:23 by stevenj] Chris@10: generalized transpose-cut routine to be able to call transpose-gcd recursivly; TOMS follow-the-cycles algorithm now seems to be completely superseded Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -111 +142 Chris@10: Chris@10: Sun Jan 8 20:53:18 EST 2006 stevenj Chris@10: * [project @ 2006-01-09 01:53:18 by stevenj] Chris@10: Chris@10: M ./threads/threads.c -3 +2 Chris@10: Chris@10: Sun Jan 8 20:52:16 EST 2006 stevenj Chris@10: * [project @ 2006-01-09 01:52:16 by stevenj] Chris@10: ignore errors from setscope -- POSIX standard does not require PTHREAD_SCOPE_SYSTEM to be supported, and PTHREAD_SCOPE_PROCESS is usually okay in that case Chris@10: Chris@10: M ./threads/threads.c -7 +9 Chris@10: Chris@10: Sun Jan 8 15:58:40 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 20:58:40 by stevenj] Chris@10: added TODO comment Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -1 +7 Chris@10: Chris@10: Sun Jan 8 15:39:28 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 20:39:28 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -8 +18 Chris@10: Chris@10: Sun Jan 8 14:44:23 EST 2006 athena Chris@10: * [project @ 2006-01-08 19:44:23 by athena] Chris@10: Boasted ``much faster altivec performance''. Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Sun Jan 8 11:44:52 EST 2006 athena Chris@10: * [project @ 2006-01-08 16:44:52 by athena] Chris@10: Added a new pass to the generator to schedule for the pipeline Chris@10: latency. (This schedule modifies the ``optimal'' cache-oblivious Chris@10: schedule and hence it uses more registers.) Chris@10: Chris@10: This pass is currently: Chris@10: Chris@10: * disabled for non-fma code, under the assumption that this will Chris@10: run on a register-starved fma. Chris@10: Chris@10: * enabled for non-simd fma code, under the assumption that this will Chris@10: run on a processor with 32 or more FP registers. The latency of 4 Chris@10: is conservative and does not introduce too much register pressure. Chris@10: Chris@10: * enabled for simd fma code, under the assumption that this will run Chris@10: on altivec. The latency of 8 seems to produce the best results. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: M ./genfft/annotate.ml -3 +58 Chris@10: M ./genfft/magic.ml -3 +5 Chris@10: M ./genfft/schedule.ml -33 +5 Chris@10: M ./support/Makefile.codelets -2 +2 Chris@10: M ./support/twovers.sh -1 +1 Chris@10: Chris@10: Sun Jan 8 03:13:53 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 08:13:53 by stevenj] Chris@10: fixed estimator for vrank3-transpose Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -6 +22 Chris@10: Chris@10: Sun Jan 8 02:02:11 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 07:02:11 by stevenj] Chris@10: more detail on VC++ workaround Chris@10: Chris@10: M ./NEWS -1 +2 Chris@10: Chris@10: Sun Jan 8 00:19:19 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 05:19:19 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Sun Jan 8 00:16:20 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 05:16:20 by stevenj] Chris@10: screw it, just use planner for all sub-transposes in vrank3-transpose (still just use memcpy for contiguous copies, though) Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -95 +185 Chris@10: Chris@10: Sat Jan 7 23:13:45 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 04:13:45 by stevenj] Chris@10: add an assert Chris@10: Chris@10: M ./kernel/tile2d.c +2 Chris@10: Chris@10: Sat Jan 7 21:57:34 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 02:57:34 by stevenj] Chris@10: vrank3-transpose now uses planner to decide whether to use cpy2d, cpy2d_tiled, etc. Chris@10: Chris@10: M ./kernel/ifftw.h -1 +7 Chris@10: M ./rdft/rank0.c -7 +5 Chris@10: M ./rdft/vrank3-transpose.c -217 +231 Chris@10: Chris@10: Sat Jan 7 20:57:16 EST 2006 stevenj Chris@10: * [project @ 2006-01-08 01:57:16 by stevenj] Chris@10: too annoying to have isqrt unexpectedly fail for n==0 Chris@10: Chris@10: M ./kernel/primes.c -2 +4 Chris@10: Chris@10: Sat Jan 7 17:49:37 EST 2006 stevenj Chris@10: * [project @ 2006-01-07 22:49:37 by stevenj] Chris@10: clarifications Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: M ./doc/fftw3.texi -2 +5 Chris@10: Chris@10: Sat Jan 7 16:39:20 EST 2006 stevenj Chris@10: * [project @ 2006-01-07 21:39:20 by stevenj] Chris@10: comment fix Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -4 +4 Chris@10: Chris@10: Sat Jan 7 15:16:22 EST 2006 stevenj Chris@10: * [project @ 2006-01-07 20:16:22 by stevenj] Chris@10: more faq updates Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -8 +9 Chris@10: Chris@10: Sat Jan 7 15:12:16 EST 2006 stevenj Chris@10: * [project @ 2006-01-07 20:12:16 by stevenj] Chris@10: enable fma on hppa, update FAQ entry Chris@10: Chris@10: M ./configure.ac +1 Chris@10: M ./doc/FAQ/fftw-faq.bfnn -3 +11 Chris@10: Chris@10: Sat Aug 12 23:47:06 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-1-beta1 Chris@10: Chris@10: Chris@10: Sat Jan 7 14:06:31 EST 2006 athena Chris@10: * [project @ 2006-01-07 19:06:31 by athena] Chris@10: Accomodate different semantics of 'const' in C and C++ Chris@10: Chris@10: M ./dft/simd/t.c +4 Chris@10: Chris@10: Fri Jan 6 23:40:53 EST 2006 athena Chris@10: * [project @ 2006-01-07 04:40:53 by athena] Chris@10: Altivec is called VMX in IBM land. Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Fri Jan 6 23:40:16 EST 2006 athena Chris@10: * [project @ 2006-01-07 04:40:16 by athena] Chris@10: Noted faster altivec support. Chris@10: Chris@10: M ./NEWS +2 Chris@10: Chris@10: Fri Jan 6 21:49:10 EST 2006 stevenj Chris@10: * [project @ 2006-01-07 02:49:10 by stevenj] Chris@10: updated icc flag detection Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -1 +1 Chris@10: Chris@10: Fri Jan 6 10:01:50 EST 2006 athena Chris@10: * [project @ 2006-01-06 15:01:50 by athena] Chris@10: Note ``memoize triggen''. Chris@10: Chris@10: M ./TODO -2 +1 Chris@10: Chris@10: Fri Jan 6 09:36:51 EST 2006 athena Chris@10: * [project @ 2006-01-06 14:36:51 by athena] Chris@10: Use --enable-threads to generate dependencies in the threads/ directory. Chris@10: Chris@10: M ./mkdist.sh -1 +1 Chris@10: Chris@10: Fri Jan 6 09:26:29 EST 2006 athena Chris@10: * [project @ 2006-01-06 14:26:29 by athena] Chris@10: Workaround to icc #defining __GNUC__. Chris@10: Chris@10: M ./kernel/ifftw.h -14 +14 Chris@10: Chris@10: Fri Jan 6 09:21:19 EST 2006 athena Chris@10: * [project @ 2006-01-06 14:21:19 by athena] Chris@10: Switched name to 3.1-beta1. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Thu Jan 5 23:08:44 EST 2006 athena Chris@10: * [project @ 2006-01-06 04:08:44 by athena] Chris@10: More thoughts. Chris@10: Chris@10: M ./TODO +12 Chris@10: Chris@10: Thu Jan 5 22:30:51 EST 2006 athena Chris@10: * [project @ 2006-01-06 03:30:51 by athena] Chris@10: Note wish that (block_size % 4) == 0. Chris@10: Chris@10: M ./TODO +3 Chris@10: Chris@10: Thu Jan 5 22:19:09 EST 2006 athena Chris@10: * [project @ 2006-01-06 03:19:09 by athena] Chris@10: Check alignment of mstart, mcount in SIMD codelets. Chris@10: Chris@10: M ./dft/codelet-dft.h -2 +3 Chris@10: M ./dft/codelets/t.c -2 +3 Chris@10: M ./dft/ctsq.c -2 +3 Chris@10: M ./dft/dftw-direct.c -11 +17 Chris@10: M ./dft/k7/k7.c -3 +3 Chris@10: M ./dft/simd/q1b.c -1 +3 Chris@10: M ./dft/simd/q1f.c -1 +3 Chris@10: M ./dft/simd/t.c -9 +11 Chris@10: M ./dft/simd/t1s.c -3 +5 Chris@10: M ./threads/ct.c -3 +1 Chris@10: M ./threads/hc2hc.c -2 Chris@10: Chris@10: Thu Jan 5 21:56:19 EST 2006 athena Chris@10: * [project @ 2006-01-06 02:56:19 by athena] Chris@10: Enable threads at bootstrap time, so I get the compiler warnings that Chris@10: I would otherwise ignore. Chris@10: Chris@10: M ./bootstrap.sh -1 +1 Chris@10: Chris@10: Thu Jan 5 18:23:15 EST 2006 athena Chris@10: * [project @ 2006-01-05 23:23:15 by athena] Chris@10: made compilable by c++ Chris@10: Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: Chris@10: Thu Jan 5 17:39:02 EST 2006 athena Chris@10: * [project @ 2006-01-05 22:39:02 by athena] Chris@10: FIXED: incorrect twiddle_shift() Chris@10: Chris@10: M ./kernel/twiddle.c -13 +19 Chris@10: Chris@10: Thu Jan 5 16:01:51 EST 2006 athena Chris@10: * [project @ 2006-01-05 21:01:51 by athena] Chris@10: Replaced remnants of awake flag with the new enum wakefulness type. Chris@10: Chris@10: M ./reodft/redft00e-r2hc.c -4 +5 Chris@10: M ./reodft/reodft11e-r2hc.c -5 +7 Chris@10: M ./reodft/rodft00e-r2hc.c -4 +5 Chris@10: M ./threads/ct.c -4 +4 Chris@10: M ./threads/dft-vrank-geq1.c -3 +3 Chris@10: M ./threads/hc2hc.c -3 +3 Chris@10: M ./threads/rdft-vrank-geq1.c -3 +3 Chris@10: M ./threads/vrank-geq1-rdft2.c -3 +3 Chris@10: Chris@10: Thu Jan 5 11:20:59 EST 2006 athena Chris@10: * [project @ 2006-01-05 16:20:59 by athena] Chris@10: Oops---there is no need to find a free slot. Chris@10: Chris@10: M ./kernel/planner.c -10 +1 Chris@10: Chris@10: Thu Jan 5 09:41:58 EST 2006 athena Chris@10: * [project @ 2006-01-05 14:41:58 by athena] Chris@10: Assertions. Chris@10: Chris@10: M ./kernel/planner.c -1 +3 Chris@10: Chris@10: Thu Jan 5 09:29:55 EST 2006 athena Chris@10: * [project @ 2006-01-05 14:29:55 by athena] Chris@10: Commented the hash table lookup algorithm. Chris@10: Chris@10: M ./kernel/planner.c -10 +14 Chris@10: Chris@10: Thu Jan 5 09:12:00 EST 2006 athena Chris@10: * [project @ 2006-01-05 14:12:00 by athena] Chris@10: Fixed infinite loop in hashtable lookup/insert. Grrr... Chris@10: Chris@10: M ./kernel/planner.c -12 +31 Chris@10: Chris@10: Wed Jan 4 22:04:28 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 03:04:26 by stevenj] Chris@10: updated copyright years to 2006 Chris@10: Chris@10: M ./COPYRIGHT -2 +2 Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/configure.c -2 +2 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute-split-dft-c2r.c -2 +2 Chris@10: M ./api/execute-split-dft-r2c.c -2 +2 Chris@10: M ./api/execute-split-dft.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/export-wisdom-to-file.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -2 +2 Chris@10: M ./api/export-wisdom.c -2 +2 Chris@10: M ./api/extract-reim.c -2 +2 Chris@10: M ./api/f77api.c -2 +2 Chris@10: M ./api/f77funcs.h -2 +2 Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./api/flops.c -2 +2 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: M ./api/import-wisdom-from-file.c -2 +2 Chris@10: M ./api/import-wisdom-from-string.c -2 +2 Chris@10: M ./api/import-wisdom.c -2 +2 Chris@10: M ./api/malloc.c -2 +2 Chris@10: M ./api/map-r2r-kind.c -2 +2 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./api/mkprinter-file.c -2 +2 Chris@10: M ./api/mktensor-iodims.c -2 +2 Chris@10: M ./api/mktensor-rowmajor.c -2 +2 Chris@10: M ./api/plan-dft-1d.c -2 +2 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r.c -2 +2 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -2 +2 Chris@10: M ./api/plan-dft-r2c.c -2 +2 Chris@10: M ./api/plan-dft.c -2 +2 Chris@10: M ./api/plan-guru-dft-c2r.c -2 +2 Chris@10: M ./api/plan-guru-dft-r2c.c -2 +2 Chris@10: M ./api/plan-guru-dft.c -2 +2 Chris@10: M ./api/plan-guru-r2r.c -2 +2 Chris@10: M ./api/plan-guru-split-dft-c2r.c -2 +2 Chris@10: M ./api/plan-guru-split-dft-r2c.c -2 +2 Chris@10: M ./api/plan-guru-split-dft.c -2 +2 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: M ./api/plan-many-dft.c -2 +2 Chris@10: M ./api/plan-many-r2r.c -2 +2 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -2 +2 Chris@10: M ./api/plan-r2r-3d.c -2 +2 Chris@10: M ./api/plan-r2r.c -2 +2 Chris@10: M ./api/print-plan.c -2 +2 Chris@10: M ./api/rdft2-pad.c -2 +2 Chris@10: M ./api/the-planner.c -2 +2 Chris@10: M ./api/version.c -3 +3 Chris@10: M ./api/x77.h -2 +2 Chris@10: M ./dft/bluestein.c -2 +2 Chris@10: M ./dft/buffered.c -3 +3 Chris@10: M ./dft/codelet-dft.h -3 +3 Chris@10: M ./dft/codelets/n.c -2 +2 Chris@10: M ./dft/codelets/n.h -2 +2 Chris@10: M ./dft/codelets/t.c -2 +2 Chris@10: M ./dft/codelets/t.h -2 +2 Chris@10: M ./dft/conf.c -3 +3 Chris@10: M ./dft/ct.c -3 +3 Chris@10: M ./dft/ct.h -2 +2 Chris@10: M ./dft/ctsq.c -3 +3 Chris@10: M ./dft/dft.h -3 +3 Chris@10: M ./dft/dftw-direct.c -3 +3 Chris@10: M ./dft/dftw-generic.c -2 +2 Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect-transpose.c -2 +2 Chris@10: M ./dft/indirect.c -3 +3 Chris@10: M ./dft/k7/k7.c -3 +3 Chris@10: M ./dft/kdft-dif.c -3 +3 Chris@10: M ./dft/kdft-difsq.c -3 +3 Chris@10: M ./dft/kdft-dit.c -3 +3 Chris@10: M ./dft/kdft.c -3 +3 Chris@10: M ./dft/nop.c -3 +3 Chris@10: M ./dft/plan.c -3 +3 Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./dft/rank-geq2.c -3 +3 Chris@10: M ./dft/simd/n1b.c -2 +2 Chris@10: M ./dft/simd/n1b.h -2 +2 Chris@10: M ./dft/simd/n1f.c -2 +2 Chris@10: M ./dft/simd/n1f.h -2 +2 Chris@10: M ./dft/simd/n2b.c -2 +2 Chris@10: M ./dft/simd/n2b.h -2 +2 Chris@10: M ./dft/simd/n2f.c -2 +2 Chris@10: M ./dft/simd/n2f.h -2 +2 Chris@10: M ./dft/simd/n2s.c -2 +2 Chris@10: M ./dft/simd/n2s.h -2 +2 Chris@10: M ./dft/simd/q1b.c -2 +2 Chris@10: M ./dft/simd/q1b.h -2 +2 Chris@10: M ./dft/simd/q1f.c -2 +2 Chris@10: M ./dft/simd/q1f.h -2 +2 Chris@10: M ./dft/simd/t.c -2 +2 Chris@10: M ./dft/simd/t1b.h -2 +2 Chris@10: M ./dft/simd/t1f.h -2 +2 Chris@10: M ./dft/simd/t1s.c -2 +2 Chris@10: M ./dft/simd/t1s.h -2 +2 Chris@10: M ./dft/simd/t2b.h -2 +2 Chris@10: M ./dft/simd/t2f.h -2 +2 Chris@10: M ./dft/solve.c -3 +3 Chris@10: M ./dft/vrank-geq1.c -3 +3 Chris@10: M ./dft/zero.c -3 +3 Chris@10: M ./doc/f77_wisdom.f -2 +2 Chris@10: M ./doc/fftw3.texi -1 +1 Chris@10: M ./genfft/algsimp.ml -4 +4 Chris@10: M ./genfft/algsimp.mli -3 +3 Chris@10: M ./genfft/annotate.ml -4 +4 Chris@10: M ./genfft/annotate.mli -3 +3 Chris@10: M ./genfft/assoctable.ml -3 +3 Chris@10: M ./genfft/assoctable.mli -3 +3 Chris@10: M ./genfft/c.ml -3 +3 Chris@10: M ./genfft/c.mli -3 +3 Chris@10: M ./genfft/complex.ml -3 +3 Chris@10: M ./genfft/complex.mli -3 +3 Chris@10: M ./genfft/conv.ml -2 +2 Chris@10: M ./genfft/conv.mli -3 +3 Chris@10: M ./genfft/dag.ml -3 +3 Chris@10: M ./genfft/dag.mli -3 +3 Chris@10: M ./genfft/expr.ml -3 +3 Chris@10: M ./genfft/expr.mli -3 +3 Chris@10: M ./genfft/fft.ml -4 +4 Chris@10: M ./genfft/fft.mli -3 +3 Chris@10: M ./genfft/gen_athnotw.ml -4 +4 Chris@10: M ./genfft/gen_athtw.ml -4 +4 Chris@10: M ./genfft/gen_conv.ml -4 +4 Chris@10: M ./genfft/gen_hc2hc.ml -4 +4 Chris@10: M ./genfft/gen_hc2r.ml -4 +4 Chris@10: M ./genfft/gen_mdct.ml -4 +4 Chris@10: M ./genfft/gen_notw.ml -4 +4 Chris@10: M ./genfft/gen_notw_c.ml -4 +4 Chris@10: M ./genfft/gen_r2hc.ml -4 +4 Chris@10: M ./genfft/gen_r2r.ml -4 +4 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twiddle_c.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./genfft/gen_twidsq_c.ml -4 +4 Chris@10: M ./genfft/genutil.ml -3 +3 Chris@10: M ./genfft/littlesimp.ml -3 +3 Chris@10: M ./genfft/littlesimp.mli -3 +3 Chris@10: M ./genfft/magic.ml -3 +3 Chris@10: M ./genfft/monads.ml -3 +3 Chris@10: M ./genfft/number.ml -3 +3 Chris@10: M ./genfft/number.mli -3 +3 Chris@10: M ./genfft/oracle.ml -3 +3 Chris@10: M ./genfft/oracle.mli -3 +3 Chris@10: M ./genfft/schedule.ml -3 +3 Chris@10: M ./genfft/schedule.mli -3 +3 Chris@10: M ./genfft/simd.ml -3 +3 Chris@10: M ./genfft/simd.mli -3 +3 Chris@10: M ./genfft/simdmagic.ml -3 +3 Chris@10: M ./genfft/to_alist.ml -3 +3 Chris@10: M ./genfft/to_alist.mli -3 +3 Chris@10: M ./genfft/trig.ml -3 +3 Chris@10: M ./genfft/trig.mli -3 +3 Chris@10: M ./genfft/twiddle.ml -3 +3 Chris@10: M ./genfft/twiddle.mli -3 +3 Chris@10: M ./genfft/unique.ml -3 +3 Chris@10: M ./genfft/unique.mli -3 +3 Chris@10: M ./genfft/util.ml -3 +3 Chris@10: M ./genfft/util.mli -3 +3 Chris@10: M ./genfft/variable.ml -3 +3 Chris@10: M ./genfft/variable.mli -3 +3 Chris@10: M ./genfft-k7/algsimp.ml -4 +4 Chris@10: M ./genfft-k7/algsimp.mli -3 +3 Chris@10: M ./genfft-k7/assoctable.ml -3 +3 Chris@10: M ./genfft-k7/assoctable.mli -3 +3 Chris@10: M ./genfft-k7/complex.ml -1 +1 Chris@10: M ./genfft-k7/complex.mli -1 +1 Chris@10: M ./genfft-k7/expr.ml -3 +3 Chris@10: M ./genfft-k7/expr.mli -3 +3 Chris@10: M ./genfft-k7/fft.ml -4 +4 Chris@10: M ./genfft-k7/gen_notw.ml -1 +1 Chris@10: M ./genfft-k7/littlesimp.ml -3 +3 Chris@10: M ./genfft-k7/littlesimp.mli -3 +3 Chris@10: M ./genfft-k7/monads.ml -3 +3 Chris@10: M ./genfft-k7/number.ml -3 +3 Chris@10: M ./genfft-k7/number.mli -3 +3 Chris@10: M ./genfft-k7/oracle.ml -3 +3 Chris@10: M ./genfft-k7/oracle.mli -3 +3 Chris@10: M ./genfft-k7/to_alist.ml -3 +3 Chris@10: M ./genfft-k7/to_alist.mli -3 +3 Chris@10: M ./genfft-k7/twiddle.ml -3 +3 Chris@10: M ./genfft-k7/twiddle.mli -3 +3 Chris@10: M ./genfft-k7/vScheduler.mli -1 +1 Chris@10: M ./kernel/align.c -3 +3 Chris@10: M ./kernel/alloc.c -3 +3 Chris@10: M ./kernel/assert.c -3 +3 Chris@10: M ./kernel/awake.c -3 +3 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/cpy1d.c -2 +2 Chris@10: M ./kernel/cpy2d-pair.c -2 +2 Chris@10: M ./kernel/cpy2d.c -2 +2 Chris@10: M ./kernel/ct.c -2 +2 Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: M ./kernel/debug.c -3 +3 Chris@10: M ./kernel/hash.c -2 +2 Chris@10: M ./kernel/iabs.c -3 +3 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./kernel/kalloc.c -3 +3 Chris@10: M ./kernel/md5-1.c -2 +2 Chris@10: M ./kernel/md5.c -2 +2 Chris@10: M ./kernel/minmax.c -3 +3 Chris@10: M ./kernel/ops.c -3 +3 Chris@10: M ./kernel/pickdim.c -3 +3 Chris@10: M ./kernel/plan.c -3 +3 Chris@10: M ./kernel/primes.c -3 +3 Chris@10: M ./kernel/print.c -3 +3 Chris@10: M ./kernel/problem.c -3 +3 Chris@10: M ./kernel/rader.c -2 +2 Chris@10: M ./kernel/scan.c -3 +3 Chris@10: M ./kernel/solver.c -3 +3 Chris@10: M ./kernel/solvtab.c -3 +3 Chris@10: M ./kernel/stride.c -3 +3 Chris@10: M ./kernel/tensor.c -3 +3 Chris@10: M ./kernel/tensor1.c -3 +3 Chris@10: M ./kernel/tensor2.c -3 +3 Chris@10: M ./kernel/tensor4.c -3 +3 Chris@10: M ./kernel/tensor5.c -3 +3 Chris@10: M ./kernel/tensor7.c -3 +3 Chris@10: M ./kernel/tensor8.c -3 +3 Chris@10: M ./kernel/tensor9.c -3 +3 Chris@10: M ./kernel/tile2d.c -2 +2 Chris@10: M ./kernel/timer.c -3 +3 Chris@10: M ./kernel/transpose.c -2 +2 Chris@10: M ./kernel/trig.c -3 +3 Chris@10: M ./kernel/twiddle.c -3 +3 Chris@10: M ./libbench/accopy-from.c -1 +1 Chris@10: M ./libbench/accopy-to.c -1 +1 Chris@10: M ./libbench/allocate.c -1 +1 Chris@10: M ./libbench/bench-main.c -1 +1 Chris@10: M ./libbench/bench-user.h -1 +1 Chris@10: M ./libbench/bench.h -1 +1 Chris@10: M ./libbench/can-do.c -1 +1 Chris@10: M ./libbench/ccopy-from.c -1 +1 Chris@10: M ./libbench/ccopy-to.c -1 +1 Chris@10: M ./libbench/deallocate.c -1 +1 Chris@10: M ./libbench/getopt-utils.c -1 +1 Chris@10: M ./libbench/info.c -1 +1 Chris@10: M ./libbench/main.c -1 +1 Chris@10: M ./libbench/prime.c -1 +1 Chris@10: M ./libbench/problem.c -1 +1 Chris@10: M ./libbench/report.c -1 +1 Chris@10: M ./libbench/speed.c -1 +1 Chris@10: M ./libbench/timer.c -1 +1 Chris@10: M ./libbench/verify.c -1 +1 Chris@10: M ./libbench/zero.c -1 +1 Chris@10: M ./libbench2/aligned-main.c -1 +1 Chris@10: M ./libbench2/allocate.c -1 +1 Chris@10: M ./libbench2/can-do.c -1 +1 Chris@10: M ./libbench2/dotens2.c -3 +3 Chris@10: M ./libbench2/getopt-utils.c -1 +1 Chris@10: M ./libbench2/info.c -1 +1 Chris@10: M ./libbench2/main.c -1 +1 Chris@10: M ./libbench2/report.c -1 +1 Chris@10: M ./libbench2/tensor.c -1 +1 Chris@10: M ./libbench2/useropt.c -1 +1 Chris@10: M ./libbench2/verify-dft.c -3 +3 Chris@10: M ./libbench2/verify-lib.c -3 +3 Chris@10: M ./libbench2/verify-r2r.c -2 +2 Chris@10: M ./libbench2/verify-rdft2.c -3 +3 Chris@10: M ./libbench2/verify.c -1 +1 Chris@10: M ./libbench2/verify.h -2 +2 Chris@10: M ./libbench2/zero.c -1 +1 Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +1 Chris@10: M ./rdft/buffered.c -3 +3 Chris@10: M ./rdft/buffered2.c -3 +3 Chris@10: M ./rdft/codelet-rdft.h -2 +2 Chris@10: M ./rdft/codelets/hb.h -2 +2 Chris@10: M ./rdft/codelets/hc2r.c -2 +2 Chris@10: M ./rdft/codelets/hc2r.h -2 +2 Chris@10: M ./rdft/codelets/hc2rIII.h -2 +2 Chris@10: M ./rdft/codelets/hf.h -2 +2 Chris@10: M ./rdft/codelets/hfb.c -2 +2 Chris@10: M ./rdft/codelets/r2hc.c -2 +2 Chris@10: M ./rdft/codelets/r2hc.h -2 +2 Chris@10: M ./rdft/codelets/r2hcII.h -2 +2 Chris@10: M ./rdft/codelets/r2r.c -2 +2 Chris@10: M ./rdft/codelets/r2r.h -2 +2 Chris@10: M ./rdft/conf.c -3 +3 Chris@10: M ./rdft/dft-r2hc.c -3 +3 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: M ./rdft/direct.c -3 +3 Chris@10: M ./rdft/direct2.c -3 +3 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-common.c -2 +2 Chris@10: M ./rdft/hc2hc-direct.c -3 +3 Chris@10: M ./rdft/hc2hc-directbuf.c -3 +3 Chris@10: M ./rdft/hc2hc-generic.c -2 +2 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/hc2hc.h -2 +2 Chris@10: M ./rdft/indirect.c -3 +3 Chris@10: M ./rdft/khc2hc.c -3 +3 Chris@10: M ./rdft/khc2r.c -3 +3 Chris@10: M ./rdft/kr2hc.c -3 +3 Chris@10: M ./rdft/kr2r.c -3 +3 Chris@10: M ./rdft/nop.c -3 +3 Chris@10: M ./rdft/nop2.c -3 +3 Chris@10: M ./rdft/plan.c -3 +3 Chris@10: M ./rdft/plan2.c -3 +3 Chris@10: M ./rdft/problem.c -3 +3 Chris@10: M ./rdft/problem2.c -3 +3 Chris@10: M ./rdft/rank-geq2-rdft2.c -3 +3 Chris@10: M ./rdft/rank-geq2.c -3 +3 Chris@10: M ./rdft/rank0-rdft2.c -3 +3 Chris@10: M ./rdft/rank0.c -3 +3 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -3 +3 Chris@10: M ./rdft/rdft2-strides.c -2 +2 Chris@10: M ./rdft/rdft2-tensor-max-index.c -3 +3 Chris@10: M ./rdft/solve.c -3 +3 Chris@10: M ./rdft/solve2.c -3 +3 Chris@10: M ./rdft/vrank-geq1-rdft2.c -3 +3 Chris@10: M ./rdft/vrank-geq1.c -3 +3 Chris@10: M ./rdft/vrank3-transpose.c -3 +3 Chris@10: M ./reodft/conf.c -3 +3 Chris@10: M ./reodft/redft00e-r2hc-pad.c -3 +3 Chris@10: M ./reodft/redft00e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft.h -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -3 +3 Chris@10: M ./reodft/reodft11e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft11e-radix2.c -3 +3 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -3 +3 Chris@10: M ./reodft/rodft00e-r2hc.c -3 +3 Chris@10: M ./simd/altivec.c -3 +3 Chris@10: M ./simd/nonportable/sse.c -3 +3 Chris@10: M ./simd/nonportable/sse2.c -3 +3 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: M ./simd/simd-sse.h -2 +2 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: M ./simd/simd.h -2 +2 Chris@10: M ./simd/sse.c -3 +3 Chris@10: M ./simd/sse2.c -3 +3 Chris@10: M ./simd/taint.c -3 +3 Chris@10: M ./threads/api.c -2 +2 Chris@10: M ./threads/conf.c -3 +3 Chris@10: M ./threads/ct.c -3 +3 Chris@10: M ./threads/dft-vrank-geq1.c -3 +3 Chris@10: M ./threads/f77api.c -2 +2 Chris@10: M ./threads/f77funcs.h -2 +2 Chris@10: M ./threads/hc2hc.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -3 +3 Chris@10: M ./threads/threads.c -2 +2 Chris@10: M ./threads/threads.h -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -3 +3 Chris@10: M ./tools/fftw-wisdom-to-conf.1 -4 +4 Chris@10: M ./tools/fftw-wisdom-to-conf.in -2 +2 Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: M ./tools/fftw_wisdom.1.in -4 +4 Chris@10: Chris@10: Wed Jan 4 21:57:23 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 02:57:23 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -7 +4 Chris@10: Chris@10: Wed Jan 4 21:52:18 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 02:52:18 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +1 Chris@10: Chris@10: Wed Jan 4 21:51:40 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 02:51:40 by stevenj] Chris@10: more updates for recent pentia/amd Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -5 +12 Chris@10: Chris@10: Wed Jan 4 20:57:47 EST 2006 athena Chris@10: * [project @ 2006-01-05 01:57:47 by athena] Chris@10: Pruned TODO. Chris@10: Chris@10: M ./TODO -6 +2 Chris@10: Chris@10: Wed Jan 4 20:43:41 EST 2006 athena Chris@10: * [project @ 2006-01-05 01:43:41 by athena] Chris@10: Prototype of problem_destroy() Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +2 Chris@10: M ./libbench2/bench.h -2 +1 Chris@10: Chris@10: Wed Jan 4 20:43:13 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 01:43:13 by stevenj] Chris@10: rm obsoleted TODOs Chris@10: Chris@10: M ./TODO -5 Chris@10: Chris@10: Wed Jan 4 20:37:24 EST 2006 athena Chris@10: * [project @ 2006-01-05 01:37:24 by athena] Chris@10: Fallback to 970 if neither -mcpu=power5 nor -mcpu=power4 are supported. Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +3 Chris@10: Chris@10: Wed Jan 4 20:29:07 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 01:29:07 by stevenj] Chris@10: NEWS updates, clarifications, and reorganization Chris@10: Chris@10: M ./NEWS -6 +10 Chris@10: Chris@10: Wed Jan 4 19:54:41 EST 2006 stevenj Chris@10: * [project @ 2006-01-05 00:54:40 by stevenj] Chris@10: remove some compiler warnings, add an assert check, make estimator work properly for nop plans Chris@10: Chris@10: M ./dft/dftw-genericbuf.c +3 Chris@10: M ./kernel/planner.c -1 +2 Chris@10: M ./kernel/trig.c -1 +3 Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 +2 Chris@10: M ./rdft/dft-r2hc.c -1 +2 Chris@10: Chris@10: Tue Jan 3 19:34:04 EST 2006 athena Chris@10: * [project @ 2006-01-04 00:34:03 by athena] Chris@10: Two big changes: Chris@10: Chris@10: 1) revised the twiddle generation machinery, to avoid generating Chris@10: twiddles when measuring, and to use a faster O(sqrt(N)) table Chris@10: when this entails no loss of precision. Chris@10: Chris@10: 2) implemented new ALLOW_PRUNING estimator hack. Chris@10: Chris@10: M ./api/apiplan.c -2 +10 Chris@10: M ./api/fftw3.h -1 +2 Chris@10: M ./api/mapflags.c -1 +4 Chris@10: M ./configure.ac -8 +9 Chris@10: M ./dft/bluestein.c -12 +18 Chris@10: M ./dft/buffered.c -5 +5 Chris@10: M ./dft/ct.c -4 +7 Chris@10: M ./dft/ctsq.c -4 +4 Chris@10: M ./dft/dftw-direct.c -3 +4 Chris@10: M ./dft/dftw-generic.c -131 +19 Chris@10: M ./dft/dftw-genericbuf.c -76 +36 Chris@10: M ./dft/direct.c -1 +2 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect-transpose.c -4 +4 Chris@10: M ./dft/indirect.c -4 +4 Chris@10: M ./dft/rader.c -17 +22 Chris@10: M ./dft/rank-geq2.c -4 +4 Chris@10: M ./dft/vrank-geq1.c -3 +3 Chris@10: M ./genfft/twiddle.ml -22 +12 Chris@10: M ./kernel/awake.c -3 +3 Chris@10: M ./kernel/ifftw.h -23 +48 Chris@10: M ./kernel/plan.c -14 +9 Chris@10: M ./kernel/planner.c -1 +8 Chris@10: M ./kernel/timer.c -3 +3 Chris@10: M ./kernel/trig.c -41 +166 Chris@10: M ./kernel/twiddle.c -64 +103 Chris@10: M ./libbench2/bench-main.c -3 +8 Chris@10: M ./libbench2/bench.h -2 +2 Chris@10: M ./libbench2/problem.c -4 +3 Chris@10: M ./libbench2/speed.c -2 +8 Chris@10: M ./rdft/buffered.c -5 +5 Chris@10: M ./rdft/buffered2.c -4 +4 Chris@10: M ./rdft/dft-r2hc.c -3 +3 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./rdft/dht-rader.c -16 +23 Chris@10: M ./rdft/direct.c -1 +3 Chris@10: M ./rdft/direct2.c -1 +2 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-direct.c -5 +7 Chris@10: M ./rdft/hc2hc-directbuf.c -5 +5 Chris@10: M ./rdft/hc2hc-generic.c -6 +7 Chris@10: M ./rdft/hc2hc.c -3 +7 Chris@10: M ./rdft/indirect.c -4 +4 Chris@10: M ./rdft/rank-geq2-rdft2.c -4 +4 Chris@10: M ./rdft/rank-geq2.c -4 +4 Chris@10: M ./rdft/rank0-rdft2.c -3 +3 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -4 +5 Chris@10: M ./rdft/vrank-geq1-rdft2.c -3 +3 Chris@10: M ./rdft/vrank-geq1.c -3 +3 Chris@10: M ./reodft/redft00e-r2hc-pad.c -4 +4 Chris@10: M ./reodft/reodft00e-splitradix.c -5 +6 Chris@10: M ./reodft/reodft010e-r2hc.c -4 +5 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -3 +3 Chris@10: M ./reodft/reodft11e-radix2.c -5 +7 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -4 +4 Chris@10: M ./tests/hook.c -3 +6 Chris@10: Chris@10: Sat Dec 24 22:08:29 EST 2005 athena Chris@10: * [project @ 2005-12-25 03:08:29 by athena] Chris@10: Estimator tweaks, mostly to favor generic over rader for small n. Chris@10: Chris@10: M ./dft/generic.c +2 Chris@10: M ./rdft/generic.c +2 Chris@10: Chris@10: Sat Dec 24 17:55:47 EST 2005 athena Chris@10: * [project @ 2005-12-24 22:55:47 by athena] Chris@10: Grrr... missing break statement in switch. Chris@10: Chris@10: M ./tests/hook.c +1 Chris@10: Chris@10: Sat Dec 24 16:08:50 EST 2005 athena Chris@10: * [project @ 2005-12-24 21:08:49 by athena] Chris@10: Swapped fields TW and OPS in struct ct_desc_s, to make k7 asm Chris@10: code insensitive to -malign-double. For consistency, changed Chris@10: struct hc2hc_desc_s in the same way. Chris@10: Chris@10: M ./dft/codelet-dft.h -2 +2 Chris@10: M ./genfft/gen_hc2hc.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -3 +3 Chris@10: M ./genfft/gen_twiddle_c.ml -3 +3 Chris@10: M ./genfft/gen_twidsq.ml -3 +3 Chris@10: M ./genfft/gen_twidsq_c.ml -3 +3 Chris@10: M ./genfft-k7/gen_twiddle.ml -3 +2 Chris@10: M ./rdft/codelet-rdft.h -1 +1 Chris@10: Chris@10: Sat Dec 24 16:00:42 EST 2005 athena Chris@10: * [project @ 2005-12-24 21:00:42 by athena] Chris@10: Wrong check for infeasible slvndx in imprt(). Chris@10: Chris@10: M ./kernel/planner.c -4 +4 Chris@10: Chris@10: Sat Dec 24 15:56:59 EST 2005 athena Chris@10: * [project @ 2005-12-24 20:56:59 by athena] Chris@10: Removed obsolete function invoke_solver_if_correct_kind(). Chris@10: Chris@10: M ./kernel/planner.c -10 +1 Chris@10: Chris@10: Sat Dec 24 14:22:12 EST 2005 athena Chris@10: * [project @ 2005-12-24 19:22:12 by athena] Chris@10: Faster implementation of safe_mulmod(), avoiding divisions altogether. Chris@10: Works for 0 <= p <= INT_MAX. Chris@10: Chris@10: M ./kernel/primes.c -13 +22 Chris@10: Chris@10: Sat Dec 24 12:05:54 EST 2005 athena Chris@10: * [project @ 2005-12-24 17:05:54 by athena] Chris@10: FFTW_ALLOW_LARGE_GENERIC must belong to flags->l, it cannot be Chris@10: overridden by fftw. Chris@10: Chris@10: M ./api/mapflags.c -2 +2 Chris@10: Chris@10: Fri Dec 23 20:46:24 EST 2005 stevenj Chris@10: * [project @ 2005-12-24 01:46:24 by stevenj] Chris@10: no more need for limits.h, add some explanatory comments Chris@10: Chris@10: M ./kernel/primes.c -4 +9 Chris@10: Chris@10: Fri Dec 23 17:50:25 EST 2005 athena Chris@10: * [project @ 2005-12-23 22:50:25 by athena] Chris@10: Paranoia. Chris@10: Chris@10: M ./dft/k7/k7.c -1 +8 Chris@10: Chris@10: Fri Dec 23 17:40:41 EST 2005 athena Chris@10: * [project @ 2005-12-23 22:40:41 by athena] Chris@10: Fixed subtle bug involving overflow of the slvndx field in flags_t. Chris@10: Chris@10: M ./kernel/ifftw.h -5 +6 Chris@10: M ./kernel/planner.c -18 +26 Chris@10: Chris@10: Fri Dec 23 16:33:56 EST 2005 athena Chris@10: * [project @ 2005-12-23 21:33:56 by athena] Chris@10: Note 64-bit clean. Chris@10: Chris@10: M ./NEWS +2 Chris@10: Chris@10: Fri Dec 23 15:34:32 EST 2005 athena Chris@10: * [project @ 2005-12-23 20:34:32 by athena] Chris@10: Threads are now 64-bit clean Chris@10: Chris@10: M ./threads/ct.c -12 +13 Chris@10: M ./threads/dft-vrank-geq1.c -20 +17 Chris@10: M ./threads/hc2hc.c -7 +8 Chris@10: M ./threads/rdft-vrank-geq1.c -18 +15 Chris@10: M ./threads/threads.h -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -20 +20 Chris@10: Chris@10: Fri Dec 23 13:00:31 EST 2005 athena Chris@10: * [project @ 2005-12-23 18:00:31 by athena] Chris@10: Restored the old numbering TW_NEXT=3 etc, because the k7 code depends Chris@10: on it. Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: Chris@10: Fri Dec 23 11:58:00 EST 2005 athena Chris@10: * [project @ 2005-12-23 16:58:00 by athena] Chris@10: Portable implementation of MULMOD() and safe_mulmod(). Chris@10: Removed all unnecessary AC_CHECK_SIZEOF() from configure.ac. Chris@10: Chris@10: M ./configure.ac -5 Chris@10: M ./kernel/ifftw.h -27 +3 Chris@10: M ./kernel/primes.c -8 +3 Chris@10: Chris@10: Thu Dec 22 11:12:29 EST 2005 athena Chris@10: * [project @ 2005-12-22 16:12:29 by athena] Chris@10: Inline the loop body in r2r codelets like we do everywhere else. Chris@10: Chris@10: M ./genfft/gen_r2r.ml -44 +34 Chris@10: Chris@10: Thu Dec 22 10:48:53 EST 2005 athena Chris@10: * [project @ 2005-12-22 15:48:53 by athena] Chris@10: Oops. Chris@10: Chris@10: M ./dft/conf.c -3 +3 Chris@10: Chris@10: Thu Dec 22 10:25:15 EST 2005 athena Chris@10: * [project @ 2005-12-22 15:25:15 by athena] Chris@10: Renamed X(sin_and_cos)() to X(cexp)(). Chris@10: Chris@10: M ./dft/bluestein.c -1 +1 Chris@10: M ./dft/dftw-generic.c -2 +2 Chris@10: M ./dft/dftw-genericbuf.c -2 +2 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/trig.c -2 +2 Chris@10: M ./kernel/twiddle.c -4 +3 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Wed Dec 21 22:49:58 EST 2005 athena Chris@10: * [project @ 2005-12-22 03:49:58 by athena] Chris@10: Somewhat faster generation of twiddle factors. Chris@10: Chris@10: M ./dft/bluestein.c -2 +1 Chris@10: M ./dft/conf.c -3 +3 Chris@10: M ./dft/dftw-generic.c -8 +5 Chris@10: M ./dft/dftw-genericbuf.c -8 +6 Chris@10: M ./dft/rader.c -3 +4 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -4 +3 Chris@10: M ./kernel/trig.c -5 +67 Chris@10: R ./kernel/trig1.c Chris@10: M ./kernel/twiddle.c -11 +15 Chris@10: M ./rdft/dht-rader.c -2 +4 Chris@10: Chris@10: Tue Dec 20 23:50:01 EST 2005 athena Chris@10: * [project @ 2005-12-21 04:50:01 by athena] Chris@10: tweaks Chris@10: Chris@10: M ./kernel/md5.c -2 +1 Chris@10: Chris@10: Tue Dec 20 22:29:19 EST 2005 athena Chris@10: * [project @ 2005-12-21 03:29:19 by athena] Chris@10: Sped up planner, esp. in estimate mode. The planner now classifies Chris@10: all solvers into DFT, RDFT, and RDFT2, and it only invokes solvers Chris@10: appropriate for the problem being planned. Because we have several Chris@10: hundred solvers, the overhead of calling irrelevant solvers is Chris@10: significant, and this modification mitigates the issue somewhat. Chris@10: Chris@10: M ./dft/bluestein.c -14 +10 Chris@10: M ./dft/buffered.c -31 +30 Chris@10: M ./dft/ct.c -16 +13 Chris@10: M ./dft/ctsq.c -23 +20 Chris@10: M ./dft/dft.h -4 +1 Chris@10: M ./dft/direct.c -69 +62 Chris@10: M ./dft/generic.c -12 +8 Chris@10: M ./dft/indirect-transpose.c -23 +19 Chris@10: M ./dft/indirect.c -37 +33 Chris@10: M ./dft/nop.c -16 +15 Chris@10: M ./dft/problem.c -6 +2 Chris@10: M ./dft/rader.c -11 +7 Chris@10: M ./dft/rank-geq2.c -13 +9 Chris@10: M ./dft/vrank-geq1.c -16 +12 Chris@10: M ./kernel/ifftw.h -1 +25 Chris@10: M ./kernel/planner.c -5 +29 Chris@10: M ./rdft/buffered.c -31 +30 Chris@10: M ./rdft/buffered2.c -8 +5 Chris@10: M ./rdft/dft-r2hc.c -10 +6 Chris@10: M ./rdft/dht-r2hc.c -12 +9 Chris@10: M ./rdft/dht-rader.c -13 +9 Chris@10: M ./rdft/direct.c -49 +45 Chris@10: M ./rdft/direct2.c -45 +41 Chris@10: M ./rdft/generic.c -13 +9 Chris@10: M ./rdft/hc2hc.c -19 +15 Chris@10: M ./rdft/indirect.c -30 +26 Chris@10: M ./rdft/nop.c -16 +13 Chris@10: M ./rdft/nop2.c -17 +15 Chris@10: M ./rdft/problem.c -6 +2 Chris@10: M ./rdft/problem2.c -6 +2 Chris@10: M ./rdft/rank-geq2-rdft2.c -20 +16 Chris@10: M ./rdft/rank-geq2.c -13 +9 Chris@10: M ./rdft/rank0-rdft2.c -13 +10 Chris@10: M ./rdft/rank0.c -14 +10 Chris@10: M ./rdft/rdft-dht.c -15 +13 Chris@10: M ./rdft/rdft.h -6 Chris@10: M ./rdft/rdft2-radix2.c -25 +17 Chris@10: M ./rdft/vrank-geq1-rdft2.c -13 +12 Chris@10: M ./rdft/vrank-geq1.c -15 +11 Chris@10: M ./rdft/vrank3-transpose.c -27 +24 Chris@10: M ./reodft/redft00e-r2hc-pad.c -12 +9 Chris@10: M ./reodft/redft00e-r2hc.c -12 +9 Chris@10: M ./reodft/reodft00e-splitradix.c -17 +14 Chris@10: M ./reodft/reodft010e-r2hc.c -12 +9 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -12 +9 Chris@10: M ./reodft/reodft11e-r2hc.c -11 +9 Chris@10: M ./reodft/reodft11e-radix2.c -12 +9 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -12 +8 Chris@10: M ./reodft/rodft00e-r2hc.c -11 +8 Chris@10: M ./tests/hook.c -92 +98 Chris@10: Chris@10: Mon Dec 19 22:04:00 EST 2005 athena Chris@10: * [project @ 2005-12-20 03:04:00 by athena] Chris@10: Eliminated all calls to sprintf() in favor of own routines, so as not Chris@10: to force users to link stdio and the associated locale/pthreads crap. Chris@10: Chris@10: M ./kernel/print.c -35 +50 Chris@10: Chris@10: Mon Dec 19 21:27:25 EST 2005 athena Chris@10: * [project @ 2005-12-20 02:27:25 by athena] Chris@10: Implemented routine to print INT, removing the need for c99's Chris@10: %td format. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: M ./kernel/print.c -7 +27 Chris@10: Chris@10: Mon Dec 19 12:06:33 EST 2005 athena Chris@10: * [project @ 2005-12-19 17:06:33 by athena] Chris@10: info->n is size_t Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Sun Dec 18 18:15:04 EST 2005 athena Chris@10: * [project @ 2005-12-18 23:15:04 by athena] Chris@10: Explicit casts in front of pointer difference in printf() context, Chris@10: just in case INT != ptrdiff_t. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -3 +3 Chris@10: Chris@10: Sun Dec 18 16:52:38 EST 2005 athena Chris@10: * [project @ 2005-12-18 21:52:38 by athena] Chris@10: Forgot to add %D to print.c Chris@10: Chris@10: M ./kernel/print.c -1 +6 Chris@10: Chris@10: Sun Dec 18 16:43:26 EST 2005 athena Chris@10: * [project @ 2005-12-18 21:43:26 by athena] Chris@10: Use %D as format character for type INT. Chris@10: Chris@10: M ./dft/bluestein.c -1 +1 Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/ctsq.c -2 +2 Chris@10: M ./dft/dftw-direct.c -3 +3 Chris@10: M ./dft/dftw-generic.c -1 +1 Chris@10: M ./dft/dftw-genericbuf.c -1 +1 Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/print.c -9 +1 Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/direct.c -2 +2 Chris@10: M ./rdft/direct2.c -2 +2 Chris@10: M ./rdft/generic.c -1 +1 Chris@10: M ./rdft/hc2hc-direct.c -2 +2 Chris@10: M ./rdft/hc2hc-directbuf.c -2 +2 Chris@10: M ./rdft/hc2hc-generic.c -1 +1 Chris@10: M ./rdft/hc2hc.c -1 +1 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft00e-splitradix.c -3 +3 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: Chris@10: Sun Dec 18 15:14:03 EST 2005 athena Chris@10: * [project @ 2005-12-18 20:14:03 by athena] Chris@10: Changed type of an_int_guaranteed_to_be_zero. Changed name as well. Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./kernel/stride.c -2 +2 Chris@10: Chris@10: Sun Dec 18 14:41:31 EST 2005 athena Chris@10: * [project @ 2005-12-18 19:41:31 by athena] Chris@10: converted %o -> INT Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./kernel/print.c -8 +6 Chris@10: Chris@10: Sat Dec 17 20:28:50 EST 2005 athena Chris@10: * [project @ 2005-12-18 01:28:50 by athena] Chris@10: Major 64-bit cleanup. Chris@10: Chris@10: M ./dft/bluestein.c -14 +15 Chris@10: M ./dft/buffered.c -20 +20 Chris@10: M ./dft/codelet-dft.h -17 +17 Chris@10: M ./dft/codelets/n.c -1 +1 Chris@10: M ./dft/codelets/t.c -1 +1 Chris@10: M ./dft/ct.c -6 +6 Chris@10: M ./dft/ct.h -5 +5 Chris@10: M ./dft/ctsq.c -6 +6 Chris@10: M ./dft/dftw-direct.c -17 +17 Chris@10: M ./dft/dftw-generic.c -28 +28 Chris@10: M ./dft/dftw-genericbuf.c -25 +25 Chris@10: M ./dft/direct.c -16 +14 Chris@10: M ./dft/generic.c -8 +8 Chris@10: M ./dft/indirect-transpose.c -5 +5 Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./dft/rader.c -10 +10 Chris@10: M ./dft/simd/n1b.c -1 +1 Chris@10: M ./dft/simd/n1f.c -1 +1 Chris@10: M ./dft/simd/n2b.c -1 +1 Chris@10: M ./dft/simd/n2f.c -1 +1 Chris@10: M ./dft/simd/n2s.c -1 +1 Chris@10: M ./dft/simd/q1b.c -1 +1 Chris@10: M ./dft/simd/q1f.c -1 +1 Chris@10: M ./dft/simd/t.c -5 +5 Chris@10: M ./dft/simd/t1s.c -1 +1 Chris@10: M ./dft/vrank-geq1.c -5 +5 Chris@10: M ./dft/zero.c -3 +3 Chris@10: M ./genfft/gen_hc2hc.ml -5 +5 Chris@10: M ./genfft/gen_hc2r.ml -6 +6 Chris@10: M ./genfft/gen_notw.ml -6 +6 Chris@10: M ./genfft/gen_notw_c.ml -6 +6 Chris@10: M ./genfft/gen_r2hc.ml -6 +6 Chris@10: M ./genfft/gen_r2r.ml -6 +6 Chris@10: M ./genfft/gen_twiddle.ml -5 +5 Chris@10: M ./genfft/gen_twiddle_c.ml -5 +5 Chris@10: M ./genfft/gen_twidsq.ml -5 +5 Chris@10: M ./genfft/gen_twidsq_c.ml -5 +5 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/cpy1d.c -2 +2 Chris@10: M ./kernel/cpy2d-pair.c -9 +7 Chris@10: M ./kernel/cpy2d.c -20 +19 Chris@10: M ./kernel/ct.c -1 +1 Chris@10: M ./kernel/iabs.c -3 +3 Chris@10: M ./kernel/ifftw.h -81 +99 Chris@10: M ./kernel/md5-1.c -5 +5 Chris@10: M ./kernel/minmax.c -3 +3 Chris@10: M ./kernel/ops.c -4 +4 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./kernel/primes.c -27 +29 Chris@10: M ./kernel/rader.c -3 +3 Chris@10: M ./kernel/solvtab.c -1 +2 Chris@10: M ./kernel/stride.c -3 +3 Chris@10: M ./kernel/tensor.c -7 +8 Chris@10: M ./kernel/tensor1.c -2 +2 Chris@10: M ./kernel/tensor2.c -3 +21 Chris@10: M ./kernel/tensor4.c -8 +8 Chris@10: M ./kernel/tensor7.c -10 +16 Chris@10: M ./kernel/tile2d.c -31 +8 Chris@10: M ./kernel/transpose.c -12 +12 Chris@10: M ./kernel/trig.c -13 +3 Chris@10: M ./kernel/twiddle.c -23 +23 Chris@10: M ./rdft/buffered.c -15 +15 Chris@10: M ./rdft/buffered2.c -30 +30 Chris@10: M ./rdft/codelet-rdft.h -29 +29 Chris@10: M ./rdft/codelets/hc2r.c -1 +1 Chris@10: M ./rdft/codelets/hfb.c -1 +1 Chris@10: M ./rdft/codelets/r2hc.c -1 +1 Chris@10: M ./rdft/codelets/r2r.c -1 +1 Chris@10: M ./rdft/dft-r2hc.c -12 +12 Chris@10: M ./rdft/dht-r2hc.c -5 +5 Chris@10: M ./rdft/dht-rader.c -12 +12 Chris@10: M ./rdft/direct.c -11 +11 Chris@10: M ./rdft/direct2.c -8 +8 Chris@10: M ./rdft/generic.c -14 +14 Chris@10: M ./rdft/hc2hc-common.c -3 +3 Chris@10: M ./rdft/hc2hc-direct.c -13 +13 Chris@10: M ./rdft/hc2hc-directbuf.c -24 +24 Chris@10: M ./rdft/hc2hc-generic.c -48 +31 Chris@10: M ./rdft/hc2hc.c -5 +5 Chris@10: M ./rdft/hc2hc.h -7 +7 Chris@10: M ./rdft/problem.c -3 +3 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: M ./rdft/rank0-rdft2.c -7 +7 Chris@10: M ./rdft/rank0.c -11 +11 Chris@10: M ./rdft/rdft-dht.c -9 +9 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -23 +23 Chris@10: M ./rdft/rdft2-strides.c -1 +1 Chris@10: M ./rdft/rdft2-tensor-max-index.c -4 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -6 +6 Chris@10: M ./rdft/vrank-geq1.c -5 +5 Chris@10: M ./rdft/vrank3-transpose.c -53 +54 Chris@10: M ./reodft/redft00e-r2hc-pad.c -11 +11 Chris@10: M ./reodft/redft00e-r2hc.c -11 +11 Chris@10: M ./reodft/reodft00e-splitradix.c -16 +17 Chris@10: M ./reodft/reodft010e-r2hc.c -26 +26 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -19 +19 Chris@10: M ./reodft/reodft11e-r2hc.c -16 +16 Chris@10: M ./reodft/reodft11e-radix2.c -25 +25 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -11 +11 Chris@10: M ./reodft/rodft00e-r2hc.c -11 +11 Chris@10: M ./simd/sse2.c -2 +3 Chris@10: M ./simd/taint.c -2 +2 Chris@10: Chris@10: Sat Aug 12 23:34:43 EDT 2006 Unknown tagger Chris@10: tagged before-64bit-rewrite Chris@10: Chris@10: Chris@10: Wed Dec 7 22:39:01 EST 2005 stevenj Chris@10: * [project @ 2005-12-08 03:39:01 by stevenj] Chris@10: PGI x86-64 cycle counter, courtesy Cristiano Calonaci Chris@10: Chris@10: M ./kernel/cycle.h -1 +12 Chris@10: Chris@10: Mon Dec 5 21:25:57 EST 2005 athena Chris@10: * [project @ 2005-12-06 02:25:57 by athena] Chris@10: Must insert into hash table when wisdom_state == WISDOM_ONLY, Chris@10: otherwise wisdom does not work. Chris@10: Chris@10: M ./kernel/planner.c -2 +3 Chris@10: Chris@10: Sat Oct 8 18:08:44 EDT 2005 stevenj Chris@10: * [project @ 2005-10-08 22:07:37 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./m4/acx_pthread.m4 -1 +2 Chris@10: Chris@10: Sun Oct 2 11:49:13 EDT 2005 athena Chris@10: * [project @ 2005-10-02 15:49:13 by athena] Chris@10: Paranoia: made planner robust against MD5 collisions. Chris@10: Chris@10: M ./api/apiplan.c -6 +37 Chris@10: M ./kernel/ifftw.h -1 +19 Chris@10: M ./kernel/planner.c -25 +55 Chris@10: Chris@10: Tue Sep 27 22:33:18 EDT 2005 athena Chris@10: * [project @ 2005-09-28 02:33:18 by athena] Chris@10: Note that --enable-3dnow is unsupported. Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +2 Chris@10: Chris@10: Tue Sep 27 22:31:04 EDT 2005 athena Chris@10: * [project @ 2005-09-28 02:31:04 by athena] Chris@10: * Removed --enable-3dnow support. Chris@10: Chris@10: * SIMD support for split complex arrays. Chris@10: Chris@10: M ./NEWS +4 Chris@10: Chris@10: Tue Sep 27 22:28:41 EDT 2005 athena Chris@10: * [project @ 2005-09-28 02:28:40 by athena] Chris@10: Removed --enabled-3dnow, since it is becoming useless as the world Chris@10: moves to x86-64, and it is a pain to maintain. (We should probably Chris@10: remove the k7 stuff as well.) Chris@10: Chris@10: M ./api/version.c -5 +1 Chris@10: M ./configure.ac -9 Chris@10: M ./genfft/gen_notw.ml -6 +4 Chris@10: M ./genfft/gen_notw_c.ml -4 +2 Chris@10: M ./genfft/gen_twiddle.ml -5 +3 Chris@10: M ./genfft/gen_twiddle_c.ml -4 +2 Chris@10: M ./genfft/gen_twidsq_c.ml -4 +2 Chris@10: M ./kernel/align.c -4 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: R ./simd/3dnow.c Chris@10: M ./simd/Makefile.am -2 +2 Chris@10: R ./simd/simd-3dnow.h Chris@10: M ./simd/simd-altivec.h -3 Chris@10: M ./simd/simd-sse.h -3 Chris@10: M ./simd/simd-sse2.h -3 Chris@10: M ./simd/simd.h -4 Chris@10: Chris@10: Tue Sep 27 21:59:16 EDT 2005 athena Chris@10: * [project @ 2005-09-28 01:59:16 by athena] Chris@10: Missing BEGIN_SIMD(), END_SIMD() statements. Chris@10: Chris@10: M ./genfft/gen_notw.ml -4 +6 Chris@10: M ./genfft/gen_twiddle.ml -3 +5 Chris@10: Chris@10: Tue Sep 27 12:16:08 EDT 2005 athena Chris@10: * [project @ 2005-09-27 16:16:08 by athena] Chris@10: Tweaks Chris@10: Chris@10: M ./simd/simd-sse.h -12 +9 Chris@10: Chris@10: Tue Sep 27 10:04:32 EDT 2005 athena Chris@10: * [project @ 2005-09-27 14:04:32 by athena] Chris@10: Fixed wrong opcount for simd codelets. Chris@10: Chris@10: M ./genfft/to_alist.ml -9 +9 Chris@10: Chris@10: Tue Sep 27 09:25:50 EDT 2005 athena Chris@10: * [project @ 2005-09-27 13:25:50 by athena] Chris@10: Fixed wrong opcount for simd codelets. Chris@10: Chris@10: M ./dft/dftw-direct.c -2 +2 Chris@10: Chris@10: Mon Sep 26 22:58:19 EDT 2005 athena Chris@10: * [project @ 2005-09-27 02:58:19 by athena] Chris@10: fixed flop counts Chris@10: Chris@10: M ./genfft/c.ml -23 +17 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: M ./simd/simd-sse2.h -9 +9 Chris@10: Chris@10: Mon Sep 26 22:34:40 EDT 2005 athena Chris@10: * [project @ 2005-09-27 02:34:40 by athena] Chris@10: Silence warnings Chris@10: Chris@10: M ./simd/simd-sse2.h +6 Chris@10: Chris@10: Mon Sep 26 20:52:36 EDT 2005 athena Chris@10: * [project @ 2005-09-27 00:52:36 by athena] Chris@10: Implemented split-complex SIMD codelets Chris@10: Chris@10: A ./dft/simd/n2s.c Chris@10: A ./dft/simd/n2s.h Chris@10: A ./dft/simd/t1s.c Chris@10: A ./dft/simd/t1s.h Chris@10: M ./dft/simd/Makefile.am -2 +3 Chris@10: M ./dft/simd/codelets/Makefile.am -6 +18 Chris@10: M ./dft/simd/n2s.c +50 Chris@10: M ./dft/simd/n2s.h +28 Chris@10: M ./dft/simd/t1s.c +48 Chris@10: M ./dft/simd/t1s.h +33 Chris@10: M ./genfft/annotate.ml -3 +3 Chris@10: M ./genfft/c.ml -1 +3 Chris@10: M ./genfft/gen_hc2hc.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -5 +26 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twiddle_c.ml -3 +3 Chris@10: M ./genfft/gen_twidsq_c.ml -3 +3 Chris@10: M ./genfft/genutil.ml -3 +3 Chris@10: M ./genfft/simd.ml -14 +8 Chris@10: M ./genfft/twiddle.ml -3 +3 Chris@10: M ./genfft/twiddle.mli -2 +2 Chris@10: M ./simd/simd-altivec.h -7 +33 Chris@10: M ./simd/simd-sse.h +33 Chris@10: M ./simd/simd-sse2.h -2 +38 Chris@10: M ./simd/simd.h +2 Chris@10: Chris@10: Sun Sep 25 22:25:35 EDT 2005 athena Chris@10: * [project @ 2005-09-26 02:25:35 by athena] Chris@10: Generalized the ``store pairs'' trick (now called ``store multiple''). Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -2 +2 Chris@10: M ./genfft/annotate.ml -37 +25 Chris@10: M ./genfft/annotate.mli -4 +2 Chris@10: M ./genfft/expr.ml -6 +6 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/gen_notw_c.ml -11 +13 Chris@10: M ./genfft/simd.ml -6 +10 Chris@10: M ./genfft/simdmagic.ml -3 +3 Chris@10: M ./simd/simd-3dnow.h -2 +2 Chris@10: M ./simd/simd-altivec.h -2 +4 Chris@10: M ./simd/simd-sse.h -3 +3 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: Chris@10: Sun Sep 25 18:58:20 EDT 2005 athena Chris@10: * [project @ 2005-09-25 22:58:20 by athena] Chris@10: Silence some warnings. Chris@10: Chris@10: M ./simd/simd-altivec.h -4 +4 Chris@10: Chris@10: Sat Sep 24 12:37:16 EDT 2005 athena Chris@10: * [project @ 2005-09-24 16:37:16 by athena] Chris@10: Removed obsolete cruft Chris@10: Chris@10: M ./simd/simd-altivec.h -6 +6 Chris@10: Chris@10: Mon Sep 19 22:55:19 EDT 2005 athena Chris@10: * [project @ 2005-09-20 02:55:19 by athena] Chris@10: Re-enabled check for because OSX requires it. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./simd/simd-altivec.h +2 Chris@10: Chris@10: Sun Sep 11 11:03:03 EDT 2005 athena Chris@10: * [project @ 2005-09-11 15:03:03 by athena] Chris@10: Check for sizeof(unsigned int) unconditionally, because the Chris@10: result is used by ifftw.h. Chris@10: Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Sun Sep 11 10:59:40 EDT 2005 athena Chris@10: * [project @ 2005-09-11 14:59:40 by athena] Chris@10: Higher size limit for t2 codelets. Chris@10: Chris@10: M ./dft/simd/t.c -1 +1 Chris@10: Chris@10: Sun Sep 11 10:50:37 EDT 2005 athena Chris@10: * [project @ 2005-09-11 14:50:37 by athena] Chris@10: Heuristic: do not use t2 simd codelets for N>1024. Chris@10: Chris@10: A ./dft/simd/t.c Chris@10: M ./dft/simd/Makefile.am -2 +2 Chris@10: M ./dft/simd/t.c +98 Chris@10: R ./dft/simd/t1b.c Chris@10: R ./dft/simd/t1f.c Chris@10: M ./dft/simd/t2b.h -2 +1 Chris@10: M ./dft/simd/t2f.h -2 +1 Chris@10: Chris@10: Mon Sep 5 22:22:50 EDT 2005 athena Chris@10: * [project @ 2005-09-06 02:22:50 by athena] Chris@10: Larger tolerance in timer calibration routine. Chris@10: Chris@10: M ./libbench2/timer.c -2 +2 Chris@10: Chris@10: Mon Sep 5 16:03:33 EDT 2005 athena Chris@10: * [project @ 2005-09-05 20:03:33 by athena] Chris@10: #include unconditionally. (There is no point in checking.) Chris@10: Chris@10: M ./configure.ac -2 +1 Chris@10: M ./simd/simd-altivec.h -2 Chris@10: Chris@10: Mon Sep 5 15:23:27 EDT 2005 athena Chris@10: * [project @ 2005-09-05 19:23:27 by athena] Chris@10: Removed SSE and SSE2 asm because it was bitrotting. Use the Intel Chris@10: API instead, which seems to be supported by gcc >= 3.3. Chris@10: Moved files that require -msse, -msse2 to new directory. Chris@10: Chris@10: A ./simd/nonportable/ Chris@10: A ./simd/nonportable/Makefile.am Chris@10: A ./simd/nonportable/sse.c Chris@10: A ./simd/nonportable/sse2.c Chris@10: M ./Makefile.am +1 Chris@10: M ./configure.ac +1 Chris@10: M ./simd/Makefile.am -4 +3 Chris@10: M ./simd/nonportable/Makefile.am +8 Chris@10: M ./simd/nonportable/sse.c +43 Chris@10: M ./simd/nonportable/sse2.c +43 Chris@10: M ./simd/simd-sse.h -101 +19 Chris@10: M ./simd/simd-sse2.h -75 +7 Chris@10: R ./simd/sse-aux.c Chris@10: M ./simd/sse.c -5 +7 Chris@10: R ./simd/sse2-aux.c Chris@10: M ./simd/sse2.c -5 +7 Chris@10: Chris@10: Mon Sep 5 12:56:28 EDT 2005 athena Chris@10: * [project @ 2005-09-05 16:56:28 by athena] Chris@10: Parse cputypes of the form 7447A,altivecsupported Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +1 Chris@10: Chris@10: Mon Sep 5 12:52:30 EDT 2005 athena Chris@10: * [project @ 2005-09-05 16:52:30 by athena] Chris@10: Distinguish powerpc 7400 from the 7450, which has a different Chris@10: pipeline. Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 +2 Chris@10: Chris@10: Mon Sep 5 12:46:00 EDT 2005 athena Chris@10: * [project @ 2005-09-05 16:46:00 by athena] Chris@10: Paranoia: define RIGHT_CPU unconditionally. Chris@10: Chris@10: M ./simd/simd-altivec.h -3 +3 Chris@10: Chris@10: Thu Aug 11 20:56:41 EDT 2005 athena Chris@10: * [project @ 2005-08-12 00:56:41 by athena] Chris@10: Removed obsolete name fftw-wisdom2c. Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: Chris@10: Thu Aug 11 20:55:59 EDT 2005 athena Chris@10: * [project @ 2005-08-12 00:55:59 by athena] Chris@10: Avoid creation of temporary files---use cpp magic instead. Chris@10: This fix solves a security bug and avoids nonportable tempfile Chris@10: creation hacks. Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -13 +20 Chris@10: Chris@10: Fri Aug 5 10:03:02 EDT 2005 athena Chris@10: * [project @ 2005-08-05 14:03:02 by athena] Chris@10: Workaround for with gcc-3.3 altivec bug. Chris@10: Chris@10: M ./configure.ac -1 +3 Chris@10: M ./simd/altivec.c -3 +2 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: Chris@10: Wed Jun 15 21:36:46 EDT 2005 stevenj Chris@10: * [project @ 2005-06-16 01:36:46 by stevenj] Chris@10: solaris fix: check -pthreads first since gcc does not like -pthread but chokes due to stubbed libc (grr) Chris@10: Chris@10: M ./m4/acx_pthread.m4 -2 +2 Chris@10: Chris@10: Fri Jun 3 17:19:56 EDT 2005 stevenj Chris@10: * [project @ 2005-06-03 21:19:56 by stevenj] Chris@10: note that VC++ bug was fixed in 2005 Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -4 +4 Chris@10: Chris@10: Mon May 30 16:30:45 EDT 2005 stevenj Chris@10: * [project @ 2005-05-30 20:30:32 by stevenj] Chris@10: generalized ax_cc_vendor to ax_compiler_vendor Chris@10: Chris@10: A ./m4/ax_compiler_vendor.m4 Chris@10: M ./configure.ac -2 +2 Chris@10: M ./m4/ax_cc_maxopt.m4 -5 +5 Chris@10: R ./m4/ax_cc_vendor.m4 Chris@10: M ./m4/ax_compiler_vendor.m4 +30 Chris@10: Chris@10: Mon May 30 15:55:07 EDT 2005 stevenj Chris@10: * [project @ 2005-05-30 19:55:07 by stevenj] Chris@10: updated message Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -2 +2 Chris@10: Chris@10: Mon May 30 15:45:14 EDT 2005 stevenj Chris@10: * [project @ 2005-05-30 19:45:14 by stevenj] Chris@10: update for new AC archive format Chris@10: Chris@10: M ./m4/acx_pthread.m4 -1 +4 Chris@10: M ./m4/ax_cc_maxopt.m4 -4 +9 Chris@10: M ./m4/ax_cc_vendor.m4 +6 Chris@10: M ./m4/ax_check_compiler_flags.m4 -1 +4 Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -1 +4 Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +5 Chris@10: M ./m4/ax_gcc_version.m4 -1 +4 Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 -1 +4 Chris@10: M ./m4/ax_openmp.m4 -1 +4 Chris@10: Chris@10: Mon May 23 23:12:22 EDT 2005 stevenj Chris@10: * [project @ 2005-05-24 03:12:22 by stevenj] Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Mon May 23 18:17:38 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 22:17:38 by stevenj] Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Mon May 23 18:13:08 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 22:13:08 by stevenj] Chris@10: more notes Chris@10: Chris@10: M ./NEWS -1 +9 Chris@10: Chris@10: Sun May 22 23:37:08 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 03:37:08 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -8 +8 Chris@10: Chris@10: Sun May 22 22:37:50 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 02:37:50 by stevenj] Chris@10: note icc 8.x annoyance Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +3 Chris@10: Chris@10: Sun May 22 22:36:04 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 02:36:04 by stevenj] Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Sun May 22 22:35:34 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 02:35:34 by stevenj] Chris@10: note gcc 3.4.[0123] bug, which is fixed in gcc 3.4.4 Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +5 Chris@10: Chris@10: Sun May 22 22:21:26 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 02:21:26 by stevenj] Chris@10: added automatic detection of icc architecture flag Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -1 +34 Chris@10: Chris@10: Sun May 22 21:47:19 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 01:47:19 by stevenj] Chris@10: add -no-gcc to icc flags...even if it is Intel's fault, I'm sick of dealing with bug reports about this Chris@10: Chris@10: M ./configure.ac +2 Chris@10: Chris@10: Sun May 22 21:40:59 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 01:40:59 by stevenj] Chris@10: added @cindex portability Chris@10: Chris@10: M ./doc/fftw3.texi -2 +11 Chris@10: Chris@10: Sun May 22 21:34:10 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 01:34:10 by stevenj] Chris@10: note --without-gcc-arch Chris@10: Chris@10: M ./doc/fftw3.texi -2 +4 Chris@10: Chris@10: Sun May 22 20:54:54 EDT 2005 stevenj Chris@10: * [project @ 2005-05-23 00:54:54 by stevenj] Chris@10: bsd ppc detection; some odd 603 types Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +3 Chris@10: Chris@10: Sun May 22 11:53:20 EDT 2005 stevenj Chris@10: * [project @ 2005-05-22 15:53:20 by stevenj] Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Sat May 21 20:34:52 EDT 2005 stevenj Chris@10: * [project @ 2005-05-22 00:34:52 by stevenj] Chris@10: ensure no spaces in cputype Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Sat May 21 20:31:41 EDT 2005 stevenj Chris@10: * [project @ 2005-05-22 00:31:41 by stevenj] Chris@10: nevermind Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Sat May 21 20:30:08 EDT 2005 stevenj Chris@10: * [project @ 2005-05-22 00:30:08 by stevenj] Chris@10: more bsd stuff Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Sat May 21 20:28:40 EDT 2005 stevenj Chris@10: * [project @ 2005-05-22 00:28:40 by stevenj] Chris@10: added BSD cpu detection for SPARC and better super/hypersparc detection Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -5 +5 Chris@10: Chris@10: Sat May 21 20:22:11 EDT 2005 stevenj Chris@10: * [project @ 2005-05-22 00:22:11 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +4 Chris@10: Chris@10: Fri May 20 19:40:09 EDT 2005 stevenj Chris@10: * [project @ 2005-05-20 23:40:09 by stevenj] Chris@10: "alternate" == "alternative" is US-centric Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Fri May 20 19:36:26 EDT 2005 stevenj Chris@10: * [project @ 2005-05-20 23:36:26 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri May 20 01:28:34 EDT 2005 stevenj Chris@10: * [project @ 2005-05-20 05:28:34 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +2 Chris@10: Chris@10: Tue May 17 18:56:46 EDT 2005 stevenj Chris@10: * [project @ 2005-05-17 22:56:46 by stevenj] Chris@10: print out estimate-planner time from can_do in verbose>2 mode Chris@10: Chris@10: M ./tests/bench.c -1 +7 Chris@10: Chris@10: Mon May 9 00:47:19 EDT 2005 stevenj Chris@10: * [project @ 2005-05-09 04:47:19 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./m4/ax_cc_vendor.m4 +1 Chris@10: Chris@10: Thu May 5 23:47:55 EDT 2005 stevenj Chris@10: * [project @ 2005-05-06 03:47:55 by stevenj] Chris@10: fixes for building Windows DLLs with Cygwin; thanks in part to Stephane Fillod Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./api/api.h +2 Chris@10: M ./api/fftw3.h -2 +12 Chris@10: M ./configure.ac -2 +4 Chris@10: M ./threads/Makefile.am -1 +1 Chris@10: Chris@10: Fri Apr 22 19:47:43 EDT 2005 stevenj Chris@10: * [project @ 2005-04-22 23:47:43 by stevenj] Chris@10: -ffast-math seems to produce code that is either about the same speed or slightly faster (gcc 3.3 and 4.0, x86) Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -1 +4 Chris@10: Chris@10: Fri Apr 22 19:18:23 EDT 2005 stevenj Chris@10: * [project @ 2005-04-22 23:18:23 by stevenj] Chris@10: power5 fallback to power4 sched for older gcc's Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Fri Apr 22 19:14:53 EDT 2005 stevenj Chris@10: * [project @ 2005-04-22 23:14:53 by stevenj] Chris@10: check for power5 Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -1 +2 Chris@10: Chris@10: Tue Apr 19 21:55:13 EDT 2005 athena Chris@10: * [project @ 2005-04-20 01:55:13 by athena] Chris@10: Removed clause #3 Chris@10: Chris@10: M ./api/fftw3.h -5 +1 Chris@10: Chris@10: Tue Apr 19 21:44:57 EDT 2005 stevenj Chris@10: * [project @ 2005-04-20 01:44:57 by stevenj] Chris@10: license clarification Chris@10: Chris@10: M ./api/fftw3.h -1 +4 Chris@10: Chris@10: Tue Apr 19 21:42:51 EDT 2005 athena Chris@10: * [project @ 2005-04-20 01:42:51 by athena] Chris@10: Changed license of fftw3.h to X11. Chris@10: Chris@10: M ./api/fftw3.h -12 +24 Chris@10: Chris@10: Mon Apr 11 13:15:12 EDT 2005 stevenj Chris@10: * [project @ 2005-04-11 17:15:12 by stevenj] Chris@10: delete fixed-input code Chris@10: Chris@10: M ./genfft/gen_conv.ml -3 +2 Chris@10: Chris@10: Sun Apr 10 16:33:24 EDT 2005 athena Chris@10: * [project @ 2005-04-10 20:33:24 by athena] Chris@10: joned L-U-planner branch Chris@10: Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/fftw3.h -4 +5 Chris@10: M ./api/mapflags.c -19 +35 Chris@10: M ./dft/bluestein.c -6 +7 Chris@10: M ./dft/buffered.c -18 +11 Chris@10: M ./dft/ct.c -7 +5 Chris@10: M ./dft/dftw-direct.c -1 +4 Chris@10: M ./dft/dftw-generic.c -2 +4 Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./dft/indirect-transpose.c -1 +1 Chris@10: M ./dft/indirect.c -8 +8 Chris@10: M ./dft/rader.c -14 +16 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/ifftw.h -49 +71 Chris@10: M ./kernel/planner.c -97 +146 Chris@10: M ./rdft/buffered.c -15 +9 Chris@10: M ./rdft/buffered2.c -9 +6 Chris@10: M ./rdft/dft-r2hc.c -7 +4 Chris@10: M ./rdft/dht-r2hc.c -7 +7 Chris@10: M ./rdft/dht-rader.c -13 +15 Chris@10: M ./rdft/generic.c -1 +1 Chris@10: M ./rdft/hc2hc-generic.c -1 +1 Chris@10: M ./rdft/hc2hc.c -3 +2 Chris@10: M ./rdft/indirect.c -6 +4 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -5 +5 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -6 +6 Chris@10: M ./reodft/redft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft00e-splitradix.c -34 +17 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./tests/bench.c -1 +1 Chris@10: M ./tests/hook.c -1 +1 Chris@10: Chris@10: Thu Apr 7 23:15:02 EDT 2005 stevenj Chris@10: * [project @ 2005-04-08 03:15:02 by stevenj] Chris@10: ref Chris@10: Chris@10: M ./reodft/reodft00e-splitradix.c -16 +33 Chris@10: Chris@10: Thu Apr 7 00:11:13 EDT 2005 stevenj Chris@10: * [project @ 2005-04-07 04:10:26 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./genfft/gen_r2r.ml -3 +3 Chris@10: Chris@10: Wed Apr 6 22:06:21 EDT 2005 stevenj Chris@10: * [project @ 2005-04-07 02:06:21 by stevenj] Chris@10: added (optional) new split-radix algorithm, enabled with -newsplit; also new -standalone option to omit desc; also -unitary, -normalization, and -normsqr options to generate r2r codelets with various normalization (to match lit. in DCT-II, use: -unitary -normsqr 2) Chris@10: Chris@10: M ./genfft/complex.ml -1 +22 Chris@10: M ./genfft/complex.mli -1 +7 Chris@10: M ./genfft/fft.ml -4 +86 Chris@10: M ./genfft/gen_athtw.ml -3 +3 Chris@10: M ./genfft/gen_hc2r.ml -4 +4 Chris@10: M ./genfft/gen_notw.ml -5 +4 Chris@10: M ./genfft/gen_notw_c.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -4 +4 Chris@10: M ./genfft/gen_r2r.ml -19 +43 Chris@10: M ./genfft/magic.ml -1 +6 Chris@10: M ./genfft/number.ml -4 +15 Chris@10: M ./genfft/number.mli -1 +2 Chris@10: Chris@10: Fri Mar 25 08:59:43 EST 2005 athena Chris@10: * [project @ 2005-03-25 13:59:43 by athena] Chris@10: Moved timeout check outside the search loop, because X(seconds) is Chris@10: expensive. Chris@10: Chris@10: M ./kernel/planner.c -9 +7 Chris@10: Chris@10: Sun Mar 20 18:35:53 EST 2005 athena Chris@10: * [project @ 2005-03-20 23:35:53 by athena] Chris@10: Enable vector recursion for in-place problems, otherwise Chris@10: dftw-genericbuf works only in PATIENT mode. Chris@10: Chris@10: M ./dft/ct.c -2 +4 Chris@10: Chris@10: Sun Mar 20 17:53:58 EST 2005 athena Chris@10: * [project @ 2005-03-20 22:53:58 by athena] Chris@10: oops Chris@10: Chris@10: M ./dft/dftw-genericbuf.c -1 +1 Chris@10: Chris@10: Sun Mar 20 17:49:13 EST 2005 athena Chris@10: * [project @ 2005-03-20 22:49:13 by athena] Chris@10: make solver UGLY for small N Chris@10: Chris@10: M ./dft/dftw-genericbuf.c -3 +13 Chris@10: Chris@10: Sun Mar 20 17:16:37 EST 2005 athena Chris@10: * [project @ 2005-03-20 22:16:37 by athena] Chris@10: new dftw-genericbuf solver Chris@10: Chris@10: A ./dft/dftw-genericbuf.c Chris@10: Chris@10: Sun Mar 20 16:12:44 EST 2005 athena Chris@10: * [project @ 2005-03-20 21:12:44 by athena] Chris@10: new dftw-genericbuf solver Chris@10: Chris@10: M ./dft/Makefile.am -4 +4 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: Chris@10: Thu Mar 17 21:48:19 EST 2005 athena Chris@10: * [project @ 2005-03-18 02:48:19 by athena] Chris@10: Hmm... what was I thinking? Chris@10: Chris@10: M ./simd/sse2-aux.c -2 +2 Chris@10: Chris@10: Thu Mar 17 19:20:54 EST 2005 athena Chris@10: * [project @ 2005-03-18 00:20:54 by athena] Chris@10: Workaround for a MSVC bug. Chris@10: Chris@10: M ./simd/simd-sse2.h -1 +6 Chris@10: M ./simd/sse2-aux.c -1 +8 Chris@10: Chris@10: Thu Mar 17 08:18:39 EST 2005 athena Chris@10: * [project @ 2005-03-17 13:18:39 by athena] Chris@10: Workaround for a MSVC bug that was reported by Eddie Yee. Chris@10: Chris@10: M ./simd/simd-sse.h -1 +6 Chris@10: M ./simd/sse-aux.c -1 +8 Chris@10: Chris@10: Tue Mar 15 13:25:53 EST 2005 athena Chris@10: * [project @ 2005-03-15 18:25:53 by athena] Chris@10: try both contiguous input and contiguous output when in doubt Chris@10: Chris@10: M ./rdft/rank0.c -3 +26 Chris@10: Chris@10: Tue Mar 15 08:44:41 EST 2005 athena Chris@10: * [project @ 2005-03-15 13:44:41 by athena] Chris@10: Added genfft flag -precompute-twiddles which moves the computation of Chris@10: the twiddle factors before the main schedule. This flag produces Chris@10: smaller code everywhere, and slightly faster code on powerpc. Chris@10: I observe no speed difference on x86. Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -3 +3 Chris@10: M ./genfft/genutil.ml -3 +5 Chris@10: M ./genfft/magic.ml -1 +5 Chris@10: M ./genfft/schedule.ml -1 +53 Chris@10: M ./genfft/schedule.mli -1 +2 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -1 +1 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +1 Chris@10: Chris@10: Mon Mar 14 21:43:53 EST 2005 stevenj Chris@10: * [project @ 2005-03-15 02:43:53 by stevenj] Chris@10: sp Chris@10: Chris@10: M ./kernel/kalloc.c -2 +2 Chris@10: Chris@10: Mon Mar 14 21:43:05 EST 2005 stevenj Chris@10: * [project @ 2005-03-15 02:43:05 by stevenj] Chris@10: whoops, spelling error (thanks to Steve Eddins for bug report) Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Sat Mar 12 15:03:47 EST 2005 athena Chris@10: * [project @ 2005-03-12 20:03:45 by athena] Chris@10: Do not approximate pcost = vl * child->pcost unless child is guaranteed Chris@10: not to be a simple codelet. Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +4 Chris@10: M ./rdft/vrank-geq1.c -2 +4 Chris@10: Chris@10: Wed Mar 9 20:00:02 EST 2005 athena Chris@10: * [project @ 2005-03-10 01:00:02 by athena] Chris@10: Relaxed applicability conditions. Chris@10: Chris@10: M ./dft/direct.c -6 +3 Chris@10: Chris@10: Wed Mar 9 00:05:47 EST 2005 athena Chris@10: * [project @ 2005-03-09 05:05:47 by athena] Chris@10: Minor optimization Chris@10: Chris@10: M ./dft/dftw-generic.c -8 +8 Chris@10: Chris@10: Tue Mar 8 22:14:02 EST 2005 athena Chris@10: * [project @ 2005-03-09 03:14:02 by athena] Chris@10: Interpret K to mean *1024. Similarly for M. Chris@10: Chris@10: M ./libbench2/problem.c -1 +12 Chris@10: Chris@10: Tue Mar 8 20:44:25 EST 2005 athena Chris@10: * [project @ 2005-03-09 01:44:25 by athena] Chris@10: Hmm... somehow some previous commit got lost. Chris@10: Chris@10: M ./kernel/primes.c -2 +8 Chris@10: Chris@10: Tue Mar 8 20:30:42 EST 2005 athena Chris@10: * [project @ 2005-03-09 01:30:42 by athena] Chris@10: Paranoia Chris@10: Chris@10: M ./dft/ct.c -2 +2 Chris@10: Chris@10: Mon Mar 7 14:30:01 EST 2005 stevenj Chris@10: * [project @ 2005-03-07 19:30:01 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Mon Mar 7 14:29:43 EST 2005 stevenj Chris@10: * [project @ 2005-03-07 19:29:43 by stevenj] Chris@10: move fftw-specific HP/UX tweak into configure.ac Chris@10: Chris@10: M ./configure.ac +3 Chris@10: M ./m4/ax_cc_maxopt.m4 -2 +2 Chris@10: Chris@10: Mon Mar 7 14:19:24 EST 2005 stevenj Chris@10: * [project @ 2005-03-07 19:19:24 by stevenj] Chris@10: ax_cc_family -> ax_cc_vendor (vendor names are easier to remember), add checks for many new compilers, use in ax_cc_maxopt Chris@10: Chris@10: A ./m4/ax_cc_vendor.m4 Chris@10: M ./configure.ac -3 +3 Chris@10: R ./m4/ax_cc_family.m4 Chris@10: M ./m4/ax_cc_maxopt.m4 -65 +42 Chris@10: M ./m4/ax_cc_vendor.m4 +23 Chris@10: Chris@10: Sun Mar 6 21:36:05 EST 2005 athena Chris@10: * [project @ 2005-03-07 02:36:05 by athena] Chris@10: Count FMA as one flop in estimator when HAVE_FMA Chris@10: Chris@10: M ./kernel/planner.c -1 +7 Chris@10: Chris@10: Sun Mar 6 19:16:06 EST 2005 athena Chris@10: * [project @ 2005-03-07 00:16:06 by athena] Chris@10: Do not try radix-2 generic. Chris@10: Chris@10: M ./dft/dftw-generic.c -1 +1 Chris@10: Chris@10: Sun Mar 6 13:04:23 EST 2005 athena Chris@10: * [project @ 2005-03-06 18:04:23 by athena] Chris@10: Use -O3 for xlc now that we use -O for CODELET_OPTIM Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -4 +4 Chris@10: Chris@10: Sun Mar 6 13:02:41 EST 2005 athena Chris@10: * [project @ 2005-03-06 18:02:41 by athena] Chris@10: New AX_CC_FAMILY macro, that detects the compiler based on symbols Chris@10: that it defines (as opposed to the name of the compiler). Chris@10: We need to start use this strategy everywhere else. Chris@10: Chris@10: A ./m4/ax_cc_family.m4 Chris@10: M ./configure.ac -25 +29 Chris@10: M ./m4/ax_cc_family.m4 +18 Chris@10: Chris@10: Sun Mar 6 11:33:15 EST 2005 athena Chris@10: * [project @ 2005-03-06 16:33:15 by athena] Chris@10: Runtime checks to guarantee small strides. Chris@10: Chris@10: M ./dft/direct.c -2 +18 Chris@10: Chris@10: Sat Mar 5 20:09:25 EST 2005 athena Chris@10: * [project @ 2005-03-06 01:09:25 by athena] Chris@10: Reduced the search space for rank-0 transforms Chris@10: Chris@10: M ./dft/vrank-geq1.c -6 +6 Chris@10: M ./kernel/tensor7.c -8 +20 Chris@10: M ./rdft/rank0.c -50 +48 Chris@10: M ./rdft/vrank-geq1.c -6 +5 Chris@10: Chris@10: Fri Mar 4 17:50:29 EST 2005 stevenj Chris@10: * [project @ 2005-03-04 22:50:29 by stevenj] Chris@10: little assert Chris@10: Chris@10: M ./kernel/primes.c -8 +3 Chris@10: Chris@10: Tue Mar 1 09:19:16 EST 2005 athena Chris@10: * [project @ 2005-03-01 14:19:16 by athena] Chris@10: Implemented directbuf, enabled for now. Chris@10: Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/dftw-direct.c -10 +7 Chris@10: M ./dft/direct.c -11 +135 Chris@10: M ./dft/kdft.c -3 +3 Chris@10: Chris@10: Mon Feb 28 22:21:14 EST 2005 athena Chris@10: * [project @ 2005-03-01 03:21:14 by athena] Chris@10: Unified dftw-direct, dftw-directbuf in an attempt to tame code Chris@10: growth Chris@10: Chris@10: M ./dft/Makefile.am -4 +4 Chris@10: M ./dft/dftw-direct.c -15 +130 Chris@10: R ./dft/dftw-directbuf.c Chris@10: M ./dft/kdft-dif.c -2 +1 Chris@10: M ./dft/kdft-dit.c -2 +1 Chris@10: Chris@10: Sun Feb 27 13:51:24 EST 2005 stevenj Chris@10: * [project @ 2005-02-27 18:51:24 by stevenj] Chris@10: fixed copyright Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Feb 26 22:21:03 EST 2005 athena Chris@10: * [project @ 2005-02-27 03:21:03 by athena] Chris@10: silence warnings Chris@10: Chris@10: M ./rdft/rank0.c -4 +5 Chris@10: Chris@10: Sat Feb 26 22:19:16 EST 2005 athena Chris@10: * [project @ 2005-02-27 03:19:16 by athena] Chris@10: oops Chris@10: Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: Chris@10: Sat Feb 26 21:28:39 EST 2005 athena Chris@10: * [project @ 2005-02-27 02:28:39 by athena] Chris@10: Tweaking while thinking about a higher-rank transposer (bitreverser) Chris@10: Chris@10: M ./rdft/rank0.c -44 +63 Chris@10: Chris@10: Sat Feb 26 20:06:49 EST 2005 athena Chris@10: * [project @ 2005-02-27 01:06:49 by athena] Chris@10: Transposed the buffer, and skewed it. This allows for contiguous Chris@10: copy operations, and the codelet should not incur associativity Chris@10: conflicts if the buffer is large. Chris@10: Chris@10: M ./dft/dftw-directbuf.c -20 +26 Chris@10: Chris@10: Sat Feb 26 18:14:11 EST 2005 stevenj Chris@10: * [project @ 2005-02-26 23:14:11 by stevenj] Chris@10: make tensor_max_index more reasonable (take maximum of input and output Chris@10: max indices, computed separately) Chris@10: Chris@10: M ./kernel/tensor4.c -4 +5 Chris@10: Chris@10: Sat Feb 26 10:04:30 EST 2005 athena Chris@10: * [project @ 2005-02-26 15:04:30 by athena] Chris@10: Use cpy2d instead of cpy2d_tiled, because vl may be too large. Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -5 +5 Chris@10: Chris@10: Sat Feb 26 00:31:52 EST 2005 athena Chris@10: * [project @ 2005-02-26 05:31:52 by athena] Chris@10: Fixed old bug that was introduced with yesterday's changes. Chris@10: Chris@10: M ./genfft/annotate.ml -4 +6 Chris@10: Chris@10: Fri Feb 25 21:54:23 EST 2005 athena Chris@10: * [project @ 2005-02-26 02:54:23 by athena] Chris@10: ``Interesting'' switch statement. Chris@10: Chris@10: M ./kernel/cpy1d.c -3 +23 Chris@10: Chris@10: Fri Feb 25 12:29:54 EST 2005 athena Chris@10: * [project @ 2005-02-25 17:29:54 by athena] Chris@10: Disabled -reorder-loads -reorder-stores, since they seem to do Chris@10: nothing. Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Fri Feb 25 12:19:10 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 17:19:03 by stevenj] Chris@10: Because of the recent changes to kernel/pickdim.c, splitrnk=0 is no Chris@10: longer equivalent to splitrnk=1 for rnk < 4, where the latter is the Chris@10: FFTW2 behavior. For small rnk, however, I observe the planner to pretty Chris@10: consistently choose the FFTW2 behavior (splitrnk=1), despite its not Chris@10: being asymptotically optimal in the cache oblivious sense. So, make Chris@10: splitrnk=1 instead of splitrnk=0 the default in FFTW_MEASURE and Chris@10: FFTW_ESTIMATE modes (rnk > 3 is pretty rare in practice anyway). Chris@10: Chris@10: M ./dft/rank-geq2.c -5 +2 Chris@10: M ./rdft/rank-geq2.c -5 +2 Chris@10: Chris@10: Fri Feb 25 00:33:27 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 05:33:27 by stevenj] Chris@10: tweak Chris@10: Chris@10: M ./dft/indirect-transpose.c -3 +4 Chris@10: Chris@10: Fri Feb 25 00:29:09 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 05:29:09 by stevenj] Chris@10: slight relaxation Chris@10: Chris@10: M ./dft/indirect-transpose.c -2 +4 Chris@10: Chris@10: Fri Feb 25 00:21:00 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 05:21:00 by stevenj] Chris@10: cruft Chris@10: Chris@10: M ./dft/indirect-transpose.c -2 Chris@10: Chris@10: Fri Feb 25 00:03:14 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 05:03:13 by stevenj] Chris@10: added experimental indirect-transpose solver: when transforming the columns of the matrix, allow us to do a transpose to make the DFTs contiguous Chris@10: Chris@10: A ./dft/indirect-transpose.c Chris@10: M ./dft/Makefile.am -3 +3 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/indirect-transpose.c +280 Chris@10: M ./dft/indirect.c -3 +10 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/tensor4.c -1 +32 Chris@10: Chris@10: Thu Feb 24 23:04:58 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 04:04:58 by stevenj] Chris@10: check for abort() Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Thu Feb 24 23:04:43 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 04:04:43 by stevenj] Chris@10: call abort() on failed assertion Chris@10: Chris@10: M ./kernel/assert.c -1 +5 Chris@10: Chris@10: Thu Feb 24 21:17:23 EST 2005 athena Chris@10: * [project @ 2005-02-25 02:17:23 by athena] Chris@10: Forgot to change X(isqrt) -> isqrt_maybe Chris@10: Chris@10: M ./kernel/primes.c -2 +2 Chris@10: Chris@10: Thu Feb 24 20:18:59 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 01:17:59 by stevenj] Chris@10: require finite_rnk Chris@10: Chris@10: M ./dft/rank-geq2.c -1 +2 Chris@10: M ./rdft/rank-geq2-rdft2.c -1 +2 Chris@10: M ./rdft/rank-geq2.c -1 +2 Chris@10: Chris@10: Thu Feb 24 20:07:38 EST 2005 stevenj Chris@10: * [project @ 2005-02-25 01:07:38 by stevenj] Chris@10: #ifdef HAVE_STRING_H must come after rdft.h so that we get config.h Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -3 +3 Chris@10: Chris@10: Thu Feb 24 18:59:40 EST 2005 athena Chris@10: * [project @ 2005-02-24 23:59:38 by athena] Chris@10: Implemented reordering of loads and stores so that the real and Chris@10: imaginary part are loaded/stored together. This should improve Chris@10: out-of-cache performance in the presence of associativity conflicts, Chris@10: and maybe worsen in-cache performance because of worse scheduling. Chris@10: Enabled for now, for experimental purposes. Chris@10: Chris@10: M ./genfft/annotate.ml -30 +76 Chris@10: M ./genfft/magic.ml -4 +10 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: M ./support/twovers.sh -1 +1 Chris@10: Chris@10: Thu Feb 24 18:10:49 EST 2005 stevenj Chris@10: * [project @ 2005-02-24 23:10:49 by stevenj] Chris@10: fix comment Chris@10: Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Chris@10: Chris@10: Thu Feb 24 18:10:23 EST 2005 stevenj Chris@10: * [project @ 2005-02-24 23:10:23 by stevenj] Chris@10: better message Chris@10: Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Chris@10: Chris@10: Thu Feb 24 18:08:36 EST 2005 stevenj Chris@10: * [project @ 2005-02-24 23:08:36 by stevenj] Chris@10: use gcc version > 3.0 as fallback in check for alignment bug Chris@10: Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Chris@10: Chris@10: Thu Feb 24 18:02:31 EST 2005 stevenj Chris@10: * [project @ 2005-02-24 23:02:31 by stevenj] Chris@10: don't use -malign-double unconditionally (it is only available on x86) Chris@10: Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -2 +3 Chris@10: Chris@10: Thu Feb 24 12:03:30 EST 2005 athena Chris@10: * [project @ 2005-02-24 17:03:30 by athena] Chris@10: Subtler selection of tilesz. Chris@10: Chris@10: M ./kernel/transpose.c -3 +8 Chris@10: Chris@10: Thu Feb 24 11:52:25 EST 2005 athena Chris@10: * [project @ 2005-02-24 16:52:25 by athena] Chris@10: Call cpy2d_tiledbuf, not cpy2d_tiled. Chris@10: Chris@10: M ./rdft/rank0.c -5 +5 Chris@10: Chris@10: Thu Feb 24 11:29:28 EST 2005 athena Chris@10: * [project @ 2005-02-24 16:29:28 by athena] Chris@10: buffer sizes were wrong :-( Chris@10: Chris@10: M ./kernel/cpy2d.c -1 +1 Chris@10: M ./kernel/transpose.c -2 +2 Chris@10: Chris@10: Thu Feb 24 11:19:01 EST 2005 athena Chris@10: * [project @ 2005-02-24 16:19:01 by athena] Chris@10: Single function for computing tile size. Eliminate spurious assertions. Chris@10: Chris@10: M ./kernel/cpy2d.c -3 +8 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/tile2d.c +5 Chris@10: M ./kernel/transpose.c -4 +6 Chris@10: M ./rdft/rank0.c -3 +1 Chris@10: Chris@10: Thu Feb 24 10:00:02 EST 2005 athena Chris@10: * [project @ 2005-02-24 15:00:02 by athena] Chris@10: Do tiling recursively. Chris@10: Chris@10: M ./kernel/tile2d.c +28 Chris@10: Chris@10: Thu Feb 24 09:40:30 EST 2005 athena Chris@10: * [project @ 2005-02-24 14:40:30 by athena] Chris@10: Reworked tiled transposes; provide tiling with and without buffering. Chris@10: I can't believe that one has to waste his life with this @#$%. Chris@10: Chris@10: A ./kernel/tile2d.c Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/cpy2d.c -36 +60 Chris@10: M ./kernel/ifftw.h -2 +10 Chris@10: M ./kernel/tile2d.c +41 Chris@10: M ./kernel/transpose.c -64 +122 Chris@10: M ./rdft/rank0.c -6 +41 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Wed Feb 23 22:21:19 EST 2005 athena Chris@10: * [project @ 2005-02-24 03:21:19 by athena] Chris@10: Clarified logic. I am not sure why the code was so confusing to begin Chris@10: with. The computation of *dp in the which_dim == 0 case was also Chris@10: wrong, returning e.g. *dp == -1 if sz->rnk == 1. Chris@10: Chris@10: M ./kernel/pickdim.c -6 +6 Chris@10: Chris@10: Wed Feb 23 22:00:15 EST 2005 athena Chris@10: * [project @ 2005-02-24 03:00:15 by athena] Chris@10: Enable aggressive inlining in codelets only, to avoid code bloat. Chris@10: Chris@10: M ./configure.ac -9 +10 Chris@10: Chris@10: Wed Feb 23 21:51:50 EST 2005 athena Chris@10: * [project @ 2005-02-24 02:51:50 by athena] Chris@10: Removed cache-oblivious copy/transpose algorithms in favor of Chris@10: explicitly blocked algorithms. The cache-oblivious algorithms fail if Chris@10: there are associativity conflicts, in which case buffering is Chris@10: necessary, as per Carter and Gatlin. Once you set the buffer size, Chris@10: there is no point whatsoever to do the algorithm recursively, and you Chris@10: may as well use blocking. Chris@10: Chris@10: M ./kernel/Makefile.am -2 +1 Chris@10: M ./kernel/cpy2d.c -62 +40 Chris@10: M ./kernel/ifftw.h -11 +11 Chris@10: M ./kernel/primes.c -4 +10 Chris@10: R ./kernel/transpose-rec.c Chris@10: M ./kernel/transpose.c +54 Chris@10: M ./rdft/rank0.c -49 +28 Chris@10: M ./rdft/vrank3-transpose.c -6 +6 Chris@10: Chris@10: Wed Feb 23 18:46:12 EST 2005 stevenj Chris@10: * [project @ 2005-02-23 23:46:12 by stevenj] Chris@10: --disable-fortran now differs from --enable-fortran that fails Chris@10: Chris@10: M ./configure.ac -2 +4 Chris@10: Chris@10: Wed Feb 23 18:42:21 EST 2005 stevenj Chris@10: * [project @ 2005-02-23 23:42:21 by stevenj] Chris@10: comment tweak Chris@10: Chris@10: M ./api/f77api.c -1 +1 Chris@10: Chris@10: Wed Feb 23 18:41:14 EST 2005 stevenj Chris@10: * [project @ 2005-02-23 23:41:14 by stevenj] Chris@10: If a Fortran compiler was not detected, just make our best guess at Chris@10: what wrappers to use...I'm sick of dealing with user complaints from Chris@10: cases where wrapper detection fails for whatever reason. Chris@10: Chris@10: M ./api/f77api.c +21 Chris@10: Chris@10: Wed Feb 23 18:10:40 EST 2005 stevenj Chris@10: * [project @ 2005-02-23 23:10:40 by stevenj] Chris@10: fflush(stdout) after print_plan, in case F77 doesn't Chris@10: Chris@10: M ./api/f77funcs.h +1 Chris@10: Chris@10: Tue Feb 22 22:54:42 EST 2005 athena Chris@10: * [project @ 2005-02-23 03:54:42 by athena] Chris@10: --enable-sse is necessary after all, to generate all dependencies Chris@10: correctly. Chris@10: Chris@10: M ./mkdist.sh -1 +1 Chris@10: Chris@10: Tue Feb 22 22:32:06 EST 2005 athena Chris@10: * [project @ 2005-02-23 03:32:06 by athena] Chris@10: Put cpy2d_pair into its own file, so that I can experiment with Chris@10: buffering of nontwiddle codelets. Chris@10: Chris@10: A ./kernel/cpy2d-pair.c Chris@10: M ./dft/dftw-directbuf.c -28 +7 Chris@10: M ./kernel/Makefile.am -5 +6 Chris@10: M ./kernel/cpy2d-pair.c +61 Chris@10: M ./kernel/ifftw.h -1 +11 Chris@10: Chris@10: Tue Feb 22 20:07:11 EST 2005 athena Chris@10: * [project @ 2005-02-23 01:07:11 by athena] Chris@10: Copy rfftwnd.png from ${srcdir}, not $PWD Chris@10: Chris@10: M ./doc/Makefile.am -1 +1 Chris@10: Chris@10: Tue Feb 22 17:08:48 EST 2005 athena Chris@10: * [project @ 2005-02-22 22:08:48 by athena] Chris@10: Do not bother memcpy-ing complex numbers. Chris@10: Chris@10: M ./rdft/rank0.c -2 +6 Chris@10: Chris@10: Tue Feb 22 16:20:46 EST 2005 athena Chris@10: * [project @ 2005-02-22 21:20:46 by athena] Chris@10: Tighther layout of buffers. I am not sure it matters, but just in case... Chris@10: Chris@10: M ./kernel/cpy2d.c -2 +2 Chris@10: M ./kernel/transpose-rec.c -4 +4 Chris@10: Chris@10: Tue Feb 22 10:13:02 EST 2005 athena Chris@10: * [project @ 2005-02-22 15:13:02 by athena] Chris@10: Usec cpy1d for rank-0 copies Chris@10: Chris@10: M ./rdft/rank0.c -3 +2 Chris@10: Chris@10: Tue Feb 22 10:06:13 EST 2005 athena Chris@10: * [project @ 2005-02-22 15:06:13 by athena] Chris@10: Implemented in-place transposes with buffering. Moved Chris@10: copy/transposition routines into own files, so that we can reuse them Chris@10: from multiple places. TODO: merge vrank3-transpose.c with rank0.c, or Chris@10: rename vrank3-transpose.c to rank0-fancy.c or something like that; Chris@10: decide whether square in-place transposes should be in rank0.c or Chris@10: vrank3-transpose.c; apply FIXME's in vrank3-transpose.c. Chris@10: Chris@10: A ./kernel/cpy1d.c Chris@10: A ./kernel/cpy2d.c Chris@10: A ./kernel/transpose-rec.c Chris@10: A ./kernel/transpose.c Chris@10: M ./kernel/Makefile.am -5 +5 Chris@10: M ./kernel/cpy1d.c +50 Chris@10: M ./kernel/cpy2d.c +154 Chris@10: M ./kernel/ifftw.h -1 +28 Chris@10: M ./kernel/transpose-rec.c +143 Chris@10: M ./kernel/transpose.c +72 Chris@10: M ./rdft/rank0.c -160 +85 Chris@10: M ./rdft/vrank3-transpose.c -231 +24 Chris@10: Chris@10: Mon Feb 21 23:29:52 EST 2005 athena Chris@10: * [project @ 2005-02-22 04:29:52 by athena] Chris@10: Indentation should be printed after newline, not at the beginning Chris@10: of print() Chris@10: Chris@10: M ./kernel/print.c -6 +11 Chris@10: Chris@10: Mon Feb 21 10:07:24 EST 2005 athena Chris@10: * [project @ 2005-02-21 15:07:24 by athena] Chris@10: generalized in anticipation of more complicated solvers. Chris@10: Chris@10: M ./rdft/rank0.c -24 +45 Chris@10: Chris@10: Sun Feb 20 22:18:59 EST 2005 athena Chris@10: * [project @ 2005-02-21 03:18:59 by athena] Chris@10: Implemented buffered recursive transpose Chris@10: Chris@10: M ./rdft/rank0.c -7 +85 Chris@10: Chris@10: Sun Feb 20 18:27:29 EST 2005 athena Chris@10: * [project @ 2005-02-20 23:27:29 by athena] Chris@10: Fixed comment Chris@10: Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: Chris@10: Sun Feb 20 18:22:15 EST 2005 athena Chris@10: * [project @ 2005-02-20 23:22:15 by athena] Chris@10: grand unification of rank0 solvers Chris@10: Chris@10: M ./rdft/Makefile.am -3 +2 Chris@10: M ./rdft/conf.c -2 +1 Chris@10: R ./rdft/rank0-vrank2.c Chris@10: M ./rdft/rank0.c -93 +173 Chris@10: M ./rdft/rdft.h -1 Chris@10: Chris@10: Sun Feb 20 15:35:24 EST 2005 athena Chris@10: * [project @ 2005-02-20 20:35:24 by athena] Chris@10: manual tail-recursion optimization Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -7 +11 Chris@10: Chris@10: Sat Feb 19 17:57:44 EST 2005 athena Chris@10: * [project @ 2005-02-19 22:57:44 by athena] Chris@10: implemented check for transpositions Chris@10: Chris@10: M ./libbench2/verify-lib.c -11 +11 Chris@10: M ./libbench2/verify-r2r.c -1 +1 Chris@10: M ./tests/check.pl -1 +19 Chris@10: Chris@10: Sat Feb 19 17:28:43 EST 2005 athena Chris@10: * [project @ 2005-02-19 22:28:43 by athena] Chris@10: Previous fix was wrong for rdft2 problems. Chris@10: Chris@10: M ./libbench2/verify-lib.c -3 +3 Chris@10: Chris@10: Sat Feb 19 17:23:36 EST 2005 athena Chris@10: * [project @ 2005-02-19 22:23:36 by athena] Chris@10: vecsz->rnk must be finite for this solver to apply. Chris@10: Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: Chris@10: Sat Feb 19 17:15:19 EST 2005 athena Chris@10: * [project @ 2005-02-19 22:15:19 by athena] Chris@10: unified the various simple'' transposers Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -101 +51 Chris@10: Chris@10: Sat Feb 19 16:55:29 EST 2005 athena Chris@10: * [project @ 2005-02-19 21:55:29 by athena] Chris@10: Fixed stupid bug in rec_transpose_swap. Fixed stupid verifier that did not catch the bug. Chris@10: Chris@10: M ./libbench2/verify-lib.c -17 +26 Chris@10: M ./libbench2/verify-r2r.c -2 +3 Chris@10: M ./rdft/vrank3-transpose.c -5 +5 Chris@10: Chris@10: Sat Feb 19 15:24:03 EST 2005 athena Chris@10: * [project @ 2005-02-19 20:24:03 by athena] Chris@10: Minor cleanup of transposition routines. Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -188 +83 Chris@10: Chris@10: Sat Feb 19 09:31:14 EST 2005 athena Chris@10: * [project @ 2005-02-19 14:31:14 by athena] Chris@10: Make the batch size B=Theta(r) instead of B=Theta(1) in buffered Chris@10: twiddle solvers. Theory: for cache line size L, we want B = Omega(L) Chris@10: to utilize the cache line fully. We also want B*r =O(Z), where Z is Chris@10: the size of the cache. It is safe to assume that Z = Theta(L^2): Chris@10: cache designers will tend to make L as large as they can get away Chris@10: with, because they don't have to program the machines that they build, Chris@10: and Z < Theta(L^2) will screw up the little matrix transposition Chris@10: benchmarks that they use to design the cache. Hence, B=Theta(r) is Chris@10: the right number. Chris@10: Chris@10: M ./dft/dftw-directbuf.c -12 +17 Chris@10: M ./rdft/hc2hc-directbuf.c -16 +20 Chris@10: Chris@10: Fri Feb 18 23:47:22 EST 2005 stevenj Chris@10: * [project @ 2005-02-19 04:47:22 by stevenj] Chris@10: for --enable-portable-binary, only try -mcpu=$arch and -m$arch on x86, Chris@10: since these generate non-portable code on every other target (and Chris@10: some other targets, like Alpha, don't support -mtune=$arch). Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +5 Chris@10: Chris@10: Thu Feb 17 21:15:42 EST 2005 athena Chris@10: * [project @ 2005-02-18 02:15:42 by athena] Chris@10: gcc/aix defines _POWER, not __powerpc__ like the rest of the world Chris@10: does. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Wed Feb 16 22:30:27 EST 2005 athena Chris@10: * [project @ 2005-02-17 03:30:27 by athena] Chris@10: enable fma for ia64, since it seems to help with the hpux compiler. Chris@10: Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Wed Feb 16 21:47:48 EST 2005 athena Chris@10: * [project @ 2005-02-17 02:47:48 by athena] Chris@10: Chris@10: M ./TODO -1 +2 Chris@10: Chris@10: Wed Feb 16 15:27:18 EST 2005 athena Chris@10: * [project @ 2005-02-16 20:27:18 by athena] Chris@10: Fixes for darwin Chris@10: Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: Chris@10: Wed Feb 16 14:27:42 EST 2005 athena Chris@10: * [project @ 2005-02-16 19:27:42 by athena] Chris@10: Made the correctness of the code more obvious. Chris@10: Chris@10: M ./api/apiplan.c -14 +16 Chris@10: Chris@10: Wed Feb 16 12:30:29 EST 2005 stevenj Chris@10: * [project @ 2005-02-16 17:30:29 by stevenj] Chris@10: s/with-portable-binary/enable-portable-binary/ to be GNUlly correct; I'm sticking with --with-gcc-arch=arch, however, as --enable-gcc-arch=arch has the wrong connotations for me Chris@10: Chris@10: M ./NEWS -2 +2 Chris@10: M ./m4/ax_cc_maxopt.m4 -3 +3 Chris@10: Chris@10: Wed Feb 16 11:44:48 EST 2005 stevenj Chris@10: * [project @ 2005-02-16 16:44:48 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/apiplan.c -2 +2 Chris@10: Chris@10: Wed Feb 16 11:23:38 EST 2005 stevenj Chris@10: * [project @ 2005-02-16 16:23:38 by stevenj] Chris@10: bless wisdom with patience used to create it Chris@10: Chris@10: M ./api/apiplan.c -1 +1 Chris@10: Chris@10: Wed Feb 16 11:18:56 EST 2005 stevenj Chris@10: * [project @ 2005-02-16 16:18:56 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/apiplan.c -7 +10 Chris@10: Chris@10: Wed Feb 16 10:50:28 EST 2005 stevenj Chris@10: * [project @ 2005-02-16 15:50:28 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/apiplan.c -1 +2 Chris@10: Chris@10: Tue Feb 15 23:53:53 EST 2005 stevenj Chris@10: * [project @ 2005-02-16 04:53:53 by stevenj] Chris@10: added 'timed' planner option Chris@10: Chris@10: M ./NEWS +3 Chris@10: M ./TODO -6 Chris@10: M ./api/apiplan.c -16 +56 Chris@10: M ./api/fftw3.h -1 +3 Chris@10: M ./doc/fftw3.texi -1 +27 Chris@10: M ./kernel/ifftw.h -1 +6 Chris@10: M ./kernel/planner.c -2 +14 Chris@10: M ./kernel/timer.c -1 +13 Chris@10: M ./tests/bench.c +5 Chris@10: Chris@10: Tue Feb 15 23:08:29 EST 2005 athena Chris@10: * [project @ 2005-02-16 04:08:27 by athena] Chris@10: Do not use SIMD_CFLAGS. The theory is that if taint.c is unsafe Chris@10: with SIMD_CFLAGS, then all files in this directory are as well. Chris@10: Conversely, if these files require SIMD_CFLAGS because they include Chris@10: "simd.h", then taint.c requires SIMD_CFLAGS as well, and thus we need Chris@10: some other hack. Chris@10: Chris@10: M ./dft/simd/Makefile.am -1 Chris@10: M ./simd/Makefile.am -8 +3 Chris@10: Chris@10: Tue Feb 15 22:49:05 EST 2005 athena Chris@10: * [project @ 2005-02-16 03:49:05 by athena] Chris@10: Do not override CFLAGS in Makefile.am. Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -1 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -1 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 Chris@10: M ./rdft/codelets/r2r/Makefile.am -1 Chris@10: M ./support/Makefile.codelets -1 +6 Chris@10: Chris@10: Tue Feb 15 10:30:12 EST 2005 athena Chris@10: * [project @ 2005-02-15 15:30:12 by athena] Chris@10: Allow users to build long double version even if sizeof(long double) Chris@10: == sizeof(double) Chris@10: Chris@10: M ./configure.ac -3 Chris@10: Chris@10: Mon Feb 14 19:55:38 EST 2005 athena Chris@10: * [project @ 2005-02-15 00:55:38 by athena] Chris@10: Updated for 3.1 Chris@10: Chris@10: M ./commercialize.sh -12 +20 Chris@10: Chris@10: Mon Feb 14 19:07:14 EST 2005 athena Chris@10: * [project @ 2005-02-15 00:07:14 by athena] Chris@10: Oops, version.h is no longer used Chris@10: Chris@10: M ./api/version.c -2 +1 Chris@10: Chris@10: Mon Feb 14 18:51:05 EST 2005 athena Chris@10: * [project @ 2005-02-14 23:51:05 by athena] Chris@10: unified fma and non-fma versions Chris@10: Chris@10: A ./support/twovers.sh Chris@10: M ./api/Makefile.am -5 +2 Chris@10: M ./api/version.c -2 +6 Chris@10: M ./configure.ac -5 +10 Chris@10: M ./dft/codelets/standard/Makefile.am -7 +7 Chris@10: M ./dft/simd/codelets/Makefile.am -10 +10 Chris@10: M ./m4/ocaml.m4 -9 +3 Chris@10: M ./mkdist.sh -5 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -4 +4 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -4 +4 Chris@10: M ./rdft/codelets/r2r/Makefile.am -9 +9 Chris@10: M ./support/Makefile.am -2 +2 Chris@10: M ./support/Makefile.codelets -7 +2 Chris@10: M ./support/twovers.sh +17 Chris@10: Chris@10: Mon Feb 14 14:12:09 EST 2005 athena Chris@10: * [project @ 2005-02-14 19:12:09 by athena] Chris@10: forgot to remove inplace/Makefile from configure.ac Chris@10: Chris@10: M ./configure.ac -1 Chris@10: Chris@10: Mon Feb 14 12:08:52 EST 2005 athena Chris@10: * [project @ 2005-02-14 17:08:49 by athena] Chris@10: Merged dft/codelets/inplace with the main dft/codelets/standard Chris@10: directory. This step makes dft codelets consistent with the rest Chris@10: of the naming conventions, and will simplify the eventual merge Chris@10: of fma and non-fma codelets. Chris@10: Chris@10: M ./Makefile.am -1 Chris@10: M ./dft/codelet-dft.h -2 +1 Chris@10: M ./dft/codelets/Makefile.am -1 +1 Chris@10: R ./dft/codelets/inplace/Makefile.am Chris@10: R ./dft/codelets/inplace/ Chris@10: M ./dft/codelets/standard/Makefile.am -9 +31 Chris@10: M ./dft/conf.c -2 +1 Chris@10: Chris@10: Mon Feb 14 11:16:15 EST 2005 athena Chris@10: * [project @ 2005-02-14 16:16:15 by athena] Chris@10: inline altivec constants, since gcc seems to generate better code this way. Chris@10: Chris@10: M ./simd/altivec.c -21 +1 Chris@10: M ./simd/simd-altivec.h -16 +18 Chris@10: Chris@10: Sun Feb 13 18:17:32 EST 2005 athena Chris@10: * [project @ 2005-02-13 23:17:32 by athena] Chris@10: group altivec constants into a single array, for faster access Chris@10: Chris@10: M ./simd/altivec.c -11 +17 Chris@10: M ./simd/simd-altivec.h -20 +11 Chris@10: Chris@10: Sun Feb 13 18:15:37 EST 2005 athena Chris@10: * [project @ 2005-02-13 23:15:37 by athena] Chris@10: code cleanup Chris@10: Chris@10: M ./genfft/c.ml -23 +9 Chris@10: M ./genfft/c.mli -2 +1 Chris@10: M ./genfft/simd.ml -6 +6 Chris@10: Chris@10: Sun Feb 13 10:29:32 EST 2005 athena Chris@10: * [project @ 2005-02-13 15:29:32 by athena] Chris@10: removed some unused stuff Chris@10: Chris@10: M ./genfft/c.ml -3 +1 Chris@10: M ./genfft/c.mli -2 +1 Chris@10: Chris@10: Sat Feb 12 22:04:40 EST 2005 athena Chris@10: * [project @ 2005-02-13 03:04:40 by athena] Chris@10: New twiddle scheme for altivec, 3dnow Chris@10: Chris@10: M ./simd/simd-3dnow.h -4 +34 Chris@10: M ./simd/simd-altivec.h -4 +27 Chris@10: Chris@10: Sat Feb 12 20:17:35 EST 2005 athena Chris@10: * [project @ 2005-02-13 01:17:35 by athena] Chris@10: Implemented new twiddle scheme for sse2 Chris@10: Chris@10: M ./simd/simd-sse2.h -4 +26 Chris@10: Chris@10: Sat Feb 12 19:57:46 EST 2005 athena Chris@10: * [project @ 2005-02-13 00:57:40 by athena] Chris@10: Implemented experimental t2* codelets, which store twiddle factors Chris@10: in a more convenient format, at the expense of twice the storage. Chris@10: Currently only SSE works; I have to port SSE2, altivec, etc. to the Chris@10: new scheme. After this, we will decide whether these codelets Chris@10: are worth the price. Chris@10: Chris@10: A ./dft/simd/t2b.h Chris@10: A ./dft/simd/t2f.h Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./dft/simd/codelets/Makefile.am -2 +14 Chris@10: M ./dft/simd/q1b.h +5 Chris@10: M ./dft/simd/q1f.h +5 Chris@10: M ./dft/simd/t1b.h +5 Chris@10: M ./dft/simd/t1f.h +5 Chris@10: M ./dft/simd/t2b.h +36 Chris@10: M ./dft/simd/t2f.h +36 Chris@10: M ./simd/simd-sse.h -36 +26 Chris@10: Chris@10: Fri Feb 11 08:07:12 EST 2005 athena Chris@10: * [project @ 2005-02-11 13:07:12 by athena] Chris@10: Forgot to define SIMD_STRIDE_OKPAIR Chris@10: Chris@10: M ./simd/simd-altivec.h +1 Chris@10: Chris@10: Thu Feb 10 22:20:00 EST 2005 athena Chris@10: * [project @ 2005-02-11 03:20:00 by athena] Chris@10: fixed sse2, 3dnow, and altivec, as promised Chris@10: Chris@10: M ./simd/simd-3dnow.h -5 +3 Chris@10: M ./simd/simd-altivec.h -1 +4 Chris@10: M ./simd/simd-sse.h +1 Chris@10: M ./simd/simd-sse2.h -5 +3 Chris@10: Chris@10: Thu Feb 10 21:47:40 EST 2005 athena Chris@10: * [project @ 2005-02-11 02:47:33 by athena] Chris@10: Generate n2?v_* codelets in such a way that we may or may not Chris@10: pair stores, depending on which mode happens to work best on Chris@10: a particular SIMD implementation. sse2, 3dnow, and altivec Chris@10: are currently broken---will fix soon. Chris@10: Chris@10: M ./dft/simd/n2b.c -1 +1 Chris@10: M ./dft/simd/n2f.c -1 +1 Chris@10: M ./genfft/annotate.ml -7 +9 Chris@10: M ./genfft/expr.ml -4 +6 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/simd.ml -3 +7 Chris@10: M ./simd/simd-sse.h -1 +6 Chris@10: Chris@10: Thu Feb 10 08:53:22 EST 2005 athena Chris@10: * [project @ 2005-02-10 13:53:22 by athena] Chris@10: instantiate altivec constants only once Chris@10: Chris@10: M ./simd/altivec.c -1 +6 Chris@10: M ./simd/simd-altivec.h -4 +10 Chris@10: Chris@10: Thu Feb 10 06:37:56 EST 2005 athena Chris@10: * [project @ 2005-02-10 11:37:56 by athena] Chris@10: Fixed alignment checks for new SIMD scheme Chris@10: Chris@10: M ./dft/simd/n2b.c -3 +3 Chris@10: M ./dft/simd/n2f.c -3 +3 Chris@10: Chris@10: Wed Feb 9 21:35:01 EST 2005 athena Chris@10: * [project @ 2005-02-10 02:35:01 by athena] Chris@10: Change n2?v_* codelets to store pairs of vectors, with implicit Chris@10: 2x2 transposition. Works for 2-way SIMD as well. Tested with sse Chris@10: and sse2. I haven't tried altivec yet, but I observed a huge Chris@10: speedup when I transformed one codelet by hand. Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -8 +6 Chris@10: M ./genfft/annotate.ml -3 +48 Chris@10: M ./genfft/annotate.mli -2 +5 Chris@10: M ./genfft/expr.ml -2 +5 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/gen_notw_c.ml -3 +17 Chris@10: M ./genfft/genutil.ml -3 +2 Chris@10: M ./genfft/simd.ml -5 +9 Chris@10: M ./genfft/simdmagic.ml -1 +3 Chris@10: M ./simd/simd-3dnow.h +6 Chris@10: M ./simd/simd-altivec.h +10 Chris@10: M ./simd/simd-sse.h +6 Chris@10: M ./simd/simd-sse2.h +6 Chris@10: Chris@10: Tue Feb 8 21:28:38 EST 2005 athena Chris@10: * [project @ 2005-02-09 02:28:38 by athena] Chris@10: Resurrected old DIF codelets for experimental purposes. They Chris@10: are disabled for now, but I am keeping the setup around for Chris@10: future reference. Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -1 +14 Chris@10: Chris@10: Tue Feb 8 20:10:19 EST 2005 stevenj Chris@10: * [project @ 2005-02-09 01:09:12 by stevenj] Chris@10: Chris@10: M ./doc/fftw3.texi -7 +7 Chris@10: Chris@10: Tue Feb 8 19:37:09 EST 2005 stevenj Chris@10: * [project @ 2005-02-09 00:37:09 by stevenj] Chris@10: clarifications, document --with-portable-binary and --with-gcc-arch Chris@10: Chris@10: M ./doc/fftw3.texi -17 +30 Chris@10: Chris@10: Tue Feb 8 19:23:41 EST 2005 stevenj Chris@10: * [project @ 2005-02-09 00:20:56 by stevenj] Chris@10: Chris@10: M ./NEWS -7 +7 Chris@10: Chris@10: Tue Feb 8 01:36:22 EST 2005 stevenj Chris@10: * [project @ 2005-02-08 06:36:22 by stevenj] Chris@10: more change comments Chris@10: Chris@10: M ./NEWS +61 Chris@10: Chris@10: Tue Feb 8 00:41:38 EST 2005 stevenj Chris@10: * [project @ 2005-02-08 05:41:38 by stevenj] Chris@10: fma is definitely beneficial on Itanium with the HP/UX compiler Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Mon Feb 7 22:58:47 EST 2005 athena Chris@10: * [project @ 2005-02-08 03:58:47 by athena] Chris@10: Silence warnings. Chris@10: Chris@10: M ./libbench2/bench-main.c -3 +4 Chris@10: Chris@10: Mon Feb 7 22:55:49 EST 2005 stevenj Chris@10: * [project @ 2005-02-08 03:55:49 by stevenj] Chris@10: when we compile our own getopt, change symbol names to avoid conflicts (e.g. avoid build failure on MacOS X with --enable-shared) Chris@10: Chris@10: M ./libbench2/getopt.h +26 Chris@10: Chris@10: Mon Feb 7 22:36:42 EST 2005 stevenj Chris@10: * [project @ 2005-02-08 03:36:42 by stevenj] Chris@10: grr, more bugfixes for in-place case Chris@10: Chris@10: M ./reodft/reodft00e-splitradix.c -9 +13 Chris@10: Chris@10: Mon Feb 7 22:29:35 EST 2005 athena Chris@10: * [project @ 2005-02-08 03:29:35 by athena] Chris@10: removed relics of FRANZ mode Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -19 Chris@10: Chris@10: Mon Feb 7 18:48:36 EST 2005 athena Chris@10: * [project @ 2005-02-07 23:48:36 by athena] Chris@10: Somehow xlc does not like ``vector int dummy;'' Chris@10: Chris@10: M ./simd/altivec.c -2 +2 Chris@10: Chris@10: Mon Feb 7 13:59:47 EST 2005 athena Chris@10: * [project @ 2005-02-07 18:59:47 by athena] Chris@10: There is no need to enable sse to make the distribution. This might Chris@10: have been true in the past but not anymore. Chris@10: Chris@10: M ./mkdist.sh -6 +8 Chris@10: Chris@10: Mon Feb 7 13:55:17 EST 2005 athena Chris@10: * [project @ 2005-02-07 18:55:17 by athena] Chris@10: Oops---included fortran file in C sources Chris@10: Chris@10: M ./api/Makefile.am -1 +1 Chris@10: Chris@10: Mon Feb 7 13:42:45 EST 2005 athena Chris@10: * [project @ 2005-02-07 18:42:45 by athena] Chris@10: Set version string at ``make dist'' time, not at ``configure'' time, Chris@10: so we know whether a user is using the fma version or not. Chris@10: Chris@10: M ./api/Makefile.am -2 +9 Chris@10: M ./api/version.c -2 +3 Chris@10: Chris@10: Sun Feb 6 17:00:33 EST 2005 athena Chris@10: * [project @ 2005-02-06 22:00:33 by athena] Chris@10: Removed useless files Chris@10: Chris@10: R ./genfft/gen_hc2r_noinline.ml Chris@10: R ./genfft/gen_notw_noinline.ml Chris@10: R ./genfft/gen_notw_noinline_c.ml Chris@10: R ./genfft/gen_r2hc_noinline.ml Chris@10: Chris@10: Sun Feb 6 16:59:39 EST 2005 athena Chris@10: * [project @ 2005-02-06 21:59:39 by athena] Chris@10: Different (simpler?) way to prevent the compiler from optimizing loop Chris@10: inductive variables. We now explicitly corrupt stride variables by Chris@10: xor-ing them with another variable that happens to be zero (but the Chris@10: compiler does not know it). In this way, the compiler does not Chris@10: attempt to extract a zillion loop indices from codelets, which would Chris@10: overflow the register set. Set the -fno-loop-optimize flag to further Chris@10: help the process. Chris@10: Chris@10: Consequences: removed m* codelets. Smaller library size. Slightly Chris@10: faster code with gcc/powerpc (including altivec). Much faster code Chris@10: with xlc/powerpc. No changes for gcc/pentium. Maybe slightly faster Chris@10: with icc/pentium. Chris@10: Chris@10: M ./configure.ac -11 +9 Chris@10: M ./dft/codelets/standard/Makefile.am -52 +3 Chris@10: M ./dft/simd/codelets/Makefile.am -30 +6 Chris@10: M ./genfft/Makefile.am -36 +10 Chris@10: M ./genfft/c.ml -1 +4 Chris@10: M ./genfft/c.mli -1 +2 Chris@10: M ./genfft/gen_hc2hc.ml -3 +5 Chris@10: M ./genfft/gen_hc2r.ml -3 +7 Chris@10: M ./genfft/gen_notw.ml -3 +6 Chris@10: M ./genfft/gen_notw_c.ml -3 +6 Chris@10: M ./genfft/gen_r2hc.ml -3 +7 Chris@10: M ./genfft/gen_twiddle.ml -3 +5 Chris@10: M ./genfft/gen_twiddle_c.ml -3 +5 Chris@10: M ./genfft/gen_twidsq.ml -3 +6 Chris@10: M ./genfft/gen_twidsq_c.ml -3 +6 Chris@10: M ./genfft/genutil.ml -1 +2 Chris@10: M ./genfft/simd.ml -1 +2 Chris@10: M ./kernel/ifftw.h -3 +7 Chris@10: M ./kernel/stride.c -1 +3 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -13 +3 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -13 +3 Chris@10: M ./support/Makefile.codelets -4 Chris@10: Chris@10: Sat Feb 5 18:51:08 EST 2005 stevenj Chris@10: * [project @ 2005-02-05 23:51:08 by stevenj] Chris@10: paranoia about in-place rodft00 plans Chris@10: Chris@10: M ./reodft/reodft00e-splitradix.c -3 +17 Chris@10: Chris@10: Sat Feb 5 18:39:55 EST 2005 stevenj Chris@10: * [project @ 2005-02-05 23:39:55 by stevenj] Chris@10: don't believe pcost when using the estimator...there is no point, and Chris@10: it screws up estimator hacks to prefer in-codelet loops to vecloops Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Sat Feb 5 18:34:25 EST 2005 athena Chris@10: * [project @ 2005-02-05 23:34:25 by athena] Chris@10: Reduced optimization level from -O3 to -O for xlc, since -O generates Chris@10: faster code. Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -4 +4 Chris@10: Chris@10: Sat Feb 5 16:26:58 EST 2005 stevenj Chris@10: * [project @ 2005-02-05 21:26:58 by stevenj] Chris@10: whoops, only applicable to redft00/rodft00 plans Chris@10: Chris@10: M ./reodft/reodft00e-splitradix.c -1 +2 Chris@10: Chris@10: Sat Feb 5 16:22:39 EST 2005 stevenj Chris@10: * [project @ 2005-02-05 21:22:39 by stevenj] Chris@10: fixed in-place operation, and don't create size-0 sub-plans Chris@10: Chris@10: M ./reodft/reodft00e-splitradix.c -17 +16 Chris@10: Chris@10: Fri Feb 4 11:30:30 EST 2005 athena Chris@10: * [project @ 2005-02-04 16:30:30 by athena] Chris@10: Autodetect altivec on linux. This code works with gcc-3.4 and Chris@10: -maltivec, with or without -mabi=altivec. The code *should* work with Chris@10: gcc-3.3 without -mabi=altivec. However, disabling -mabi=altivec on Chris@10: gcc-3.4 produces much worse code (I don't know why). Chris@10: Chris@10: M ./simd/altivec.c -5 +31 Chris@10: Chris@10: Fri Jan 28 00:04:58 EST 2005 stevenj Chris@10: * [project @ 2005-01-28 05:04:58 by stevenj] Chris@10: update reference Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Thu Jan 27 15:48:28 EST 2005 stevenj Chris@10: * [project @ 2005-01-27 20:48:28 by stevenj] Chris@10: note that DCT-II/III are often called the'' DCT/DCT Chris@10: Chris@10: M ./doc/fftw3.texi -11 +17 Chris@10: Chris@10: Fri Jan 21 14:42:04 EST 2005 stevenj Chris@10: * [project @ 2005-01-21 19:42:04 by stevenj] Chris@10: added MSVC++ for ia64 (based on information at http://www.intel.com/cd/ids/developer/asmo-na/eng/19949.htm?prn=Y) Chris@10: Chris@10: M ./kernel/cycle.h -1 +21 Chris@10: Chris@10: Fri Jan 21 14:22:50 EST 2005 stevenj Chris@10: * [project @ 2005-01-21 19:22:50 by stevenj] Chris@10: vc++ defines _M_AMD64 on x86-64, apparently Chris@10: Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: Chris@10: Tue Jan 18 22:30:27 EST 2005 stevenj Chris@10: * [project @ 2005-01-19 03:30:27 by stevenj] Chris@10: avoid gratuitous breakage with -Werror, requested by Simon Perreault Chris@10: Chris@10: M ./m4/acx_pthread.m4 -2 +2 Chris@10: Chris@10: Mon Jan 17 18:54:55 EST 2005 stevenj Chris@10: * [project @ 2005-01-17 23:54:55 by stevenj] Chris@10: comment typo Chris@10: Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Chris@10: Chris@10: Sat Jan 15 16:56:23 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 21:56:23 by stevenj] Chris@10: bumped shared-lib revision# Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Jan 15 16:35:42 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 21:35:42 by stevenj] Chris@10: add X(estimate_cost) to get estimator cost, and print from bench, to aid in tweaking estimator Chris@10: Chris@10: M ./api/fftw3.h -1 +2 Chris@10: M ./api/flops.c +5 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -6 +11 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sat Jan 15 14:57:56 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 19:57:56 by stevenj] Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Sat Jan 15 12:57:07 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 17:57:07 by stevenj] Chris@10: formatting fix Chris@10: Chris@10: M ./doc/fftw3.texi -3 +9 Chris@10: Chris@10: Sat Jan 15 12:31:28 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 17:31:28 by stevenj] Chris@10: tweaks Chris@10: Chris@10: M ./doc/fftw3.texi -17 +17 Chris@10: M ./reodft/Makefile.am +1 Chris@10: M ./reodft/conf.c -4 +7 Chris@10: Chris@10: Sat Jan 15 12:03:24 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 17:03:24 by stevenj] Chris@10: use less buffer space Chris@10: Chris@10: M ./reodft/reodft00e-splitradix.c -22 +23 Chris@10: Chris@10: Sat Jan 15 01:41:58 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 06:41:58 by stevenj] Chris@10: added split-radix-based dct/dst I for odd n Chris@10: Chris@10: A ./reodft/reodft00e-splitradix.c Chris@10: M ./doc/fftw3.texi -16 +21 Chris@10: M ./reodft/Makefile.am -1 +1 Chris@10: M ./reodft/conf.c -1 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +4 Chris@10: M ./reodft/reodft.h +1 Chris@10: M ./reodft/reodft00e-splitradix.c +337 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +4 Chris@10: Chris@10: Fri Jan 14 21:50:08 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 02:50:08 by stevenj] Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: Chris@10: Fri Jan 14 21:49:55 EST 2005 stevenj Chris@10: * [project @ 2005-01-15 02:49:55 by stevenj] Chris@10: warn silly users who confuse CVS id with FFTW version Chris@10: Chris@10: M ./api/fftw3.h -1 +3 Chris@10: Chris@10: Fri Jan 14 16:57:36 EST 2005 stevenj Chris@10: * [project @ 2005-01-14 21:57:36 by stevenj] Chris@10: get sparc cpu type on solaris as well as with linux Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -13 +11 Chris@10: Chris@10: Thu Jan 13 19:21:58 EST 2005 stevenj Chris@10: * [project @ 2005-01-14 00:21:58 by stevenj] Chris@10: detect prescott mobile (f37) Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Thu Jan 13 18:09:52 EST 2005 stevenj Chris@10: * [project @ 2005-01-13 23:09:52 by stevenj] Chris@10: use cpuid for x86_64 as well as i[56]86 Chris@10: Chris@10: M ./bootstrap.sh +1 Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Thu Jan 13 17:59:55 EST 2005 stevenj Chris@10: * [project @ 2005-01-13 22:59:55 by stevenj] Chris@10: update with x86info 1.7 and other sources (identify k8, nocona, etc), handle nonzero leading bytes in eax Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -23 +27 Chris@10: Chris@10: Thu Jan 13 16:30:33 EST 2005 stevenj Chris@10: * [project @ 2005-01-13 21:30:33 by stevenj] Chris@10: compactified check for JOINABLE; use AC_DEFINE_UNQUOTED instead of AC_DEFINE for PTHREAD_CREATE_JOINABLE (thanks to Oliver Niekrenz for the bug report) Chris@10: Chris@10: M ./m4/acx_pthread.m4 -23 +16 Chris@10: Chris@10: Wed Jan 12 12:22:13 EST 2005 athena Chris@10: * [project @ 2005-01-12 17:22:13 by athena] Chris@10: The scheduler hack was incorrect because it swapped instructions Chris@10: of the form A = *B and *B = C. Fixed. Chris@10: Chris@10: M ./genfft/annotate.ml -7 +13 Chris@10: Chris@10: Tue Jan 11 22:13:24 EST 2005 athena Chris@10: * [project @ 2005-01-12 03:13:24 by athena] Chris@10: Quote expressions such as ``if test $FOO = yes'' when $FOO may be Chris@10: empty. Also, $GCC is set to either ``yes'' or empty, never to ``no''. Chris@10: Chris@10: M ./m4/ax_cc_maxopt.m4 -2 +2 Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Tue Jan 11 19:30:47 EST 2005 athena Chris@10: * [project @ 2005-01-12 00:30:47 by athena] Chris@10: Hmm---somehow the previous commit did not work. Chris@10: Chris@10: M ./TODO -6 Chris@10: M ./configure.ac +9 Chris@10: M ./simd/altivec.c -3 +3 Chris@10: Chris@10: Tue Jan 11 16:54:45 EST 2005 athena Chris@10: * [project @ 2005-01-11 21:54:45 by athena] Chris@10: Fixed various gcc-related problems on powerpc: Chris@10: - gcc-3.4 becomes totally confused by expressions like Chris@10: vec_add(a, vec_add(b, vec_add(c, ...))) Chris@10: The compiler uses gigabytes of memory and then crashes, presumably Chris@10: because of the exponential-time search problem involved in typing the Chris@10: above expression (since vec_add can take either ints or floats). Chris@10: I changed VADD and similar macros to be inline functions, thus Chris@10: constraining the type system. Chris@10: Chris@10: - New flags Chris@10: --param inline-unit-growth=1000 --param large-function-growth=1000 Chris@10: to work around limitations of the gcc-3.4 inliner. Chris@10: Chris@10: M ./simd/simd-altivec.h -14 +12 Chris@10: Chris@10: Mon Jan 10 21:27:24 EST 2005 athena Chris@10: * [project @ 2005-01-11 02:27:24 by athena] Chris@10: Check for HAVE_ALTIVEC_H Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +2 Chris@10: Chris@10: Mon Jan 10 21:09:30 EST 2005 athena Chris@10: * [project @ 2005-01-11 02:09:30 by athena] Chris@10: Remove support for altivec using gcc builtins, since these keep Chris@10: changing across gcc versions. These changes work on gcc-3.4/linux; I Chris@10: haven't tried MacOS X yet. (The altivec ``spec'' differs between Chris@10: Motorola/Apple and gcc, grrr...) Chris@10: Chris@10: M ./configure.ac -1 +2 Chris@10: M ./simd/altivec.c -8 +5 Chris@10: M ./simd/simd-altivec.h -69 +5 Chris@10: Chris@10: Mon Jan 10 18:57:30 EST 2005 athena Chris@10: * [project @ 2005-01-10 23:57:30 by athena] Chris@10: Stylistic changes Chris@10: Chris@10: M ./rdft/rank0-vrank2.c -21 +21 Chris@10: Chris@10: Mon Jan 10 17:34:41 EST 2005 athena Chris@10: * [project @ 2005-01-10 22:34:41 by athena] Chris@10: Changed incorrect ugliness condition. Chris@10: Chris@10: M ./rdft/dft-r2hc.c -4 +2 Chris@10: Chris@10: Mon Jan 10 16:09:43 EST 2005 stevenj Chris@10: * [project @ 2005-01-10 21:09:43 by stevenj] Chris@10: note x86info version number that was used, to make it easier to update Chris@10: the cpuid for changes in later versions Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Mon Jan 10 15:00:51 EST 2005 athena Chris@10: * [project @ 2005-01-10 20:00:51 by athena] Chris@10: Make dft-r2hc non-UGLY for rank-0 problems Chris@10: Chris@10: M ./rdft/dft-r2hc.c -1 +6 Chris@10: Chris@10: Mon Jan 10 14:50:23 EST 2005 athena Chris@10: * [project @ 2005-01-10 19:50:23 by athena] Chris@10: Do not use -mcpu=970 on power4 processors, because power4 does Chris@10: not have altivec. Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Mon Jan 10 14:48:47 EST 2005 athena Chris@10: * [project @ 2005-01-10 19:48:47 by athena] Chris@10: Note gcc-3.4 problem with inlining. Chris@10: Chris@10: M ./TODO +5 Chris@10: Chris@10: Mon Jan 10 13:51:08 EST 2005 athena Chris@10: * [project @ 2005-01-10 18:51:08 by athena] Chris@10: Oops, forgot to remove ``static'' from the declaration of noninlinable Chris@10: functions. Chris@10: Chris@10: M ./genfft/gen_hc2r_noinline.ml -3 +3 Chris@10: M ./genfft/gen_notw_noinline_c.ml -3 +3 Chris@10: Chris@10: Mon Jan 10 12:31:26 EST 2005 athena Chris@10: * [project @ 2005-01-10 17:31:26 by athena] Chris@10: Recognize power4. Use ``head -n COUNT'' instead of obsolete ``head Chris@10: -COUNT'' (which fails on gentoo). Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +4 Chris@10: Chris@10: Sun Jan 9 22:12:16 EST 2005 athena Chris@10: * [project @ 2005-01-10 03:12:16 by athena] Chris@10: Remind to add FAQ entry concerning gcc-3.4.[1-3] crashes. Chris@10: Chris@10: M ./TODO +4 Chris@10: Chris@10: Sun Jan 9 21:53:08 EST 2005 stevenj Chris@10: * [project @ 2005-01-10 02:53:08 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./m4/ax_gcc_version.m4 -3 +3 Chris@10: Chris@10: Sun Jan 9 21:48:02 EST 2005 stevenj Chris@10: * [project @ 2005-01-10 02:48:02 by stevenj] Chris@10: support checking for major.minor.patchlevel Chris@10: Chris@10: M ./m4/ax_gcc_version.m4 -10 +11 Chris@10: Chris@10: Sun Jan 9 21:40:18 EST 2005 athena Chris@10: * [project @ 2005-01-10 02:40:18 by athena] Chris@10: Revert CODELET_OPTIM to -O on IA32, which is faster than -O2. Chris@10: Chris@10: M ./configure.ac -3 +4 Chris@10: Chris@10: Sun Jan 9 20:30:12 EST 2005 athena Chris@10: * [project @ 2005-01-10 01:30:12 by athena] Chris@10: /bin/sh allows no spaces in assignments. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sun Jan 9 20:05:55 EST 2005 athena Chris@10: * [project @ 2005-01-10 01:05:55 by athena] Chris@10: Make non-inlinable functions external, so that gcc becomes confused Chris@10: and does not try to inline them. Chris@10: Chris@10: M ./genfft/gen_hc2r_noinline.ml -5 +5 Chris@10: M ./genfft/gen_notw_noinline.ml -5 +5 Chris@10: M ./genfft/gen_notw_noinline_c.ml -5 +5 Chris@10: Chris@10: Sun Jan 9 13:44:25 EST 2005 athena Chris@10: * [project @ 2005-01-09 18:44:25 by athena] Chris@10: Add -fno-web to CFLAGS, because -fweb destroys FMAs. Chris@10: Chris@10: M ./configure.ac -13 +18 Chris@10: Chris@10: Sun Jan 9 10:31:47 EST 2005 athena Chris@10: * [project @ 2005-01-09 15:31:47 by athena] Chris@10: Allow -mcpu=970 besides -mcpu=G5 Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Sun Jan 9 10:26:20 EST 2005 athena Chris@10: * [project @ 2005-01-09 15:26:20 by athena] Chris@10: configure was not using -fno-schedule-insns :-( Chris@10: Chris@10: M ./configure.ac -2 +1 Chris@10: Chris@10: Sun Jan 9 08:52:40 EST 2005 athena Chris@10: * [project @ 2005-01-09 13:52:40 by athena] Chris@10: In mkplan() and elsewhere, use solver index instead of solver Chris@10: *pointer*, which looks marginally clearer. Chris@10: Chris@10: M ./kernel/planner.c -18 +22 Chris@10: Chris@10: Sun Jan 9 08:15:36 EST 2005 athena Chris@10: * [project @ 2005-01-09 13:15:36 by athena] Chris@10: Split planner hash table into two tables, for blessed and unblessed Chris@10: solutions respectively. Now an unblessed solution never overwrites a Chris@10: blessed solution, thus avoiding wisdom leakage by construction. Chris@10: Further, forget() is now a O(1) operation, which speeds up the Chris@10: estimator when the wisdom table is large. Chris@10: Chris@10: M ./TODO -6 Chris@10: M ./kernel/ifftw.h -14 +17 Chris@10: M ./kernel/planner.c -157 +126 Chris@10: Chris@10: Sat Jan 8 21:19:45 EST 2005 athena Chris@10: * [project @ 2005-01-09 02:19:45 by athena] Chris@10: New TODO idea. Chris@10: Chris@10: M ./TODO +7 Chris@10: Chris@10: Thu Jan 6 11:02:29 EST 2005 athena Chris@10: * [project @ 2005-01-06 16:02:29 by athena] Chris@10: Split search() into two routines to make the UGLY/NO_UGLY logic Chris@10: obvious. Chris@10: Chris@10: M ./kernel/planner.c -41 +48 Chris@10: Chris@10: Fri Dec 17 16:08:54 EST 2004 stevenj Chris@10: * [project @ 2004-12-17 21:08:54 by stevenj] Chris@10: push/pop 64-bit registers on ia64; thanks to Orion Poplawski for the fix Chris@10: Chris@10: M ./simd/3dnow.c -1 +13 Chris@10: M ./simd/sse.c -1 +7 Chris@10: M ./simd/sse2.c -1 +7 Chris@10: Chris@10: Thu Dec 9 21:41:09 EST 2004 stevenj Chris@10: * [project @ 2004-12-10 02:41:09 by stevenj] Chris@10: patch from FreeBSD ports - FreeBSD does not have memalign, but its Chris@10: malloc is 16-byte aligned Chris@10: Chris@10: M ./kernel/kalloc.c -1 +5 Chris@10: Chris@10: Tue Nov 23 17:06:47 EST 2004 stevenj Chris@10: * [project @ 2004-11-23 22:06:47 by stevenj] Chris@10: don't compile taint.c with SIMD_CFLAGS (fixed Debian bug #259612) Chris@10: Chris@10: M ./simd/Makefile.am -3 +10 Chris@10: Chris@10: Thu Nov 18 11:37:32 EST 2004 stevenj Chris@10: * [project @ 2004-11-18 16:37:32 by stevenj] Chris@10: revert incorrect change -- codlist.c should be rebuilt, but it is built in the build directory and not in the source directory Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Wed Nov 17 22:53:53 EST 2004 stevenj Chris@10: * [project @ 2004-11-18 03:53:53 by stevenj] Chris@10: $(CODLIST) should be rebuilt only if Makefile.am changes, or Chris@10: alternatively only in maintainer mode, to prevent stomping in the Chris@10: source directory during user builds. (Thanks to Grant Cook for the Chris@10: bug report.) Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Sat Nov 13 13:43:01 EST 2004 stevenj Chris@10: * [project @ 2004-11-13 18:43:01 by stevenj] Chris@10: corrected #ifdef for icc/ia64, thanks to Matt Boman Chris@10: Chris@10: M ./kernel/cycle.h -16 +18 Chris@10: Chris@10: Sat Nov 13 13:34:55 EST 2004 stevenj Chris@10: * [project @ 2004-11-13 18:34:55 by stevenj] Chris@10: spelling correction (Larsen, not Larson) Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Mon Nov 8 22:12:39 EST 2004 stevenj Chris@10: * [project @ 2004-11-09 03:12:39 by stevenj] Chris@10: use standard withval Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +3 Chris@10: Chris@10: Mon Nov 8 22:09:16 EST 2004 stevenj Chris@10: * [project @ 2004-11-09 03:09:16 by stevenj] Chris@10: match doc Chris@10: Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 -2 +2 Chris@10: Chris@10: Mon Nov 8 22:00:34 EST 2004 stevenj Chris@10: * [project @ 2004-11-09 03:00:34 by stevenj] Chris@10: formatting Chris@10: Chris@10: M ./m4/ax_openmp.m4 -3 +3 Chris@10: Chris@10: Mon Nov 8 21:59:33 EST 2004 stevenj Chris@10: * [project @ 2004-11-09 02:59:33 by stevenj] Chris@10: make sure OPENMP_CFLAGS environment variable is used correctly Chris@10: Chris@10: M ./m4/ax_openmp.m4 -2 +6 Chris@10: Chris@10: Mon Nov 8 21:46:50 EST 2004 stevenj Chris@10: * [project @ 2004-11-09 02:46:50 by stevenj] Chris@10: replace ax_check_cc_flags with more generic ax_check_compiler_flags Chris@10: Chris@10: A ./m4/ax_check_compiler_flags.m4 Chris@10: M ./configure.ac -13 +8 Chris@10: M ./m4/ax_cc_maxopt.m4 -7 +6 Chris@10: R ./m4/ax_check_cc_flags.m4 Chris@10: M ./m4/ax_check_compiler_flags.m4 +37 Chris@10: M ./m4/ax_gcc_aligns_stack.m4 -5 +4 Chris@10: M ./m4/ax_gcc_archflag.m4 -3 +3 Chris@10: Chris@10: Mon Nov 8 17:49:42 EST 2004 stevenj Chris@10: * [project @ 2004-11-08 22:49:42 by stevenj] Chris@10: separate macro for OpenMP test Chris@10: Chris@10: A ./m4/ax_openmp.m4 Chris@10: M ./configure.ac -40 +6 Chris@10: M ./m4/ax_cc_maxopt.m4 -3 +5 Chris@10: M ./m4/ax_openmp.m4 +58 Chris@10: Chris@10: Fri Nov 5 16:24:22 EST 2004 stevenj Chris@10: * [project @ 2004-11-05 21:24:22 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Oct 29 00:48:13 EDT 2004 stevenj Chris@10: * [project @ 2004-10-29 04:48:13 by stevenj] Chris@10: Chris@10: M ./configure.ac -12 +19 Chris@10: Chris@10: Thu Oct 28 00:09:38 EDT 2004 stevenj Chris@10: * [project @ 2004-10-28 04:09:38 by stevenj] Chris@10: better guessing of sparc type on Linux Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +15 Chris@10: Chris@10: Wed Oct 27 13:44:08 EDT 2004 stevenj Chris@10: * [project @ 2004-10-27 17:44:08 by stevenj] Chris@10: note default Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Wed Oct 27 13:41:57 EDT 2004 stevenj Chris@10: * [project @ 2004-10-27 17:41:57 by stevenj] Chris@10: tweak Chris@10: Chris@10: M ./m4/ax_gcc_archflag.m4 -2 +2 Chris@10: Chris@10: Wed Oct 27 13:34:25 EDT 2004 stevenj Chris@10: * [project @ 2004-10-27 17:34:25 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 -2 +3 Chris@10: Chris@10: Wed Oct 27 13:31:10 EDT 2004 stevenj Chris@10: * [project @ 2004-10-27 17:31:10 by stevenj] Chris@10: whoops, m4 is EXTRA_DIST, not SUBDIR, since it doesn't have a Makefile Chris@10: Chris@10: M ./Makefile.am -2 +2 Chris@10: Chris@10: Wed Oct 27 13:16:57 EDT 2004 stevenj Chris@10: * [project @ 2004-10-27 17:16:57 by stevenj] Chris@10: silence warnings Chris@10: Chris@10: M ./m4/ocaml.m4 -2 +2 Chris@10: Chris@10: Wed Oct 27 13:14:22 EDT 2004 stevenj Chris@10: * [project @ 2004-10-27 17:14:22 by stevenj] Chris@10: clean up m4 macros; try to detect correct gcc -march flag on x86; new --with-portable-binary, --with-gcc-arch= flags; use -O2 for codelets with gcc 3.4 to work around bug Chris@10: Chris@10: A ./m4/ Chris@10: A ./m4/acx_pthread.m4 Chris@10: A ./m4/amx_prog_as.m4 Chris@10: A ./m4/ax_cc_maxopt.m4 Chris@10: A ./m4/ax_check_cc_flags.m4 Chris@10: A ./m4/ax_gcc_aligns_stack.m4 Chris@10: A ./m4/ax_gcc_archflag.m4 Chris@10: A ./m4/ax_gcc_version.m4 Chris@10: A ./m4/ax_gcc_x86_cpuid.m4 Chris@10: A ./m4/ocaml.m4 Chris@10: M ./Makefile.am -2 +4 Chris@10: R ./acinclude.m4 Chris@10: R ./acx_pthread.m4 Chris@10: M ./bootstrap.sh -12 +10 Chris@10: M ./configure.ac -11 +26 Chris@10: M ./m4/acx_pthread.m4 +240 Chris@10: M ./m4/amx_prog_as.m4 +11 Chris@10: M ./m4/ax_cc_maxopt.m4 +126 Chris@10: M ./m4/ax_check_cc_flags.m4 +37 Chris@10: M ./m4/ax_gcc_aligns_stack.m4 +47 Chris@10: M ./m4/ax_gcc_archflag.m4 +147 Chris@10: M ./m4/ax_gcc_version.m4 +34 Chris@10: M ./m4/ax_gcc_x86_cpuid.m4 +37 Chris@10: M ./m4/ocaml.m4 +84 Chris@10: Chris@10: Tue Oct 26 16:46:14 EDT 2004 stevenj Chris@10: * [project @ 2004-10-26 20:46:14 by stevenj] Chris@10: rename cexp -> mcexp to avoid conflict with C99 builtin Chris@10: Chris@10: M ./libbench2/mp.c -3 +3 Chris@10: Chris@10: Mon Oct 25 16:58:23 EDT 2004 stevenj Chris@10: * [project @ 2004-10-25 20:58:23 by stevenj] Chris@10: use basename , w/o args, for compiler-name comparisons; also detect Compaq ccc on alpha-linus Chris@10: Chris@10: M ./acinclude.m4 -4 +10 Chris@10: Chris@10: Sun Oct 24 22:05:10 EDT 2004 stevenj Chris@10: * [project @ 2004-10-25 02:05:10 by stevenj] Chris@10: note recent icc problems Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -6 +9 Chris@10: Chris@10: Sun Oct 24 02:10:12 EDT 2004 stevenj Chris@10: * [project @ 2004-10-24 06:10:12 by stevenj] Chris@10: whoops, disable semaphores again (for now) Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Sun Oct 24 02:04:58 EDT 2004 stevenj Chris@10: * [project @ 2004-10-24 06:04:58 by stevenj] Chris@10: POSIX semaphores are *not* the same as SYSV semaphores Chris@10: Chris@10: M ./threads/threads.c -2 +2 Chris@10: Chris@10: Sun Oct 24 01:18:14 EDT 2004 stevenj Chris@10: * [project @ 2004-10-24 05:18:14 by stevenj] Chris@10: re-implement threaded stuff; dftw now takes parameters to indicate a portion of m loop Chris@10: Chris@10: A ./rdft/hc2hc.c Chris@10: A ./rdft/hc2hc.h Chris@10: A ./threads/ct.c Chris@10: A ./threads/hc2hc.c Chris@10: M ./dft/conf.c -2 +2 Chris@10: M ./dft/ct.c -8 +11 Chris@10: M ./dft/ct.h -8 +11 Chris@10: M ./dft/ctsq.c -2 +2 Chris@10: M ./dft/dft.h -5 +2 Chris@10: M ./dft/dftw-direct.c -15 +29 Chris@10: M ./dft/dftw-directbuf.c -15 +30 Chris@10: M ./dft/dftw-generic.c -14 +31 Chris@10: M ./dft/kdft-dif.c -7 +3 Chris@10: M ./dft/kdft-difsq.c -2 +2 Chris@10: M ./dft/kdft-dit.c -7 +3 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/twiddle.c -1 +10 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: R ./rdft/ct.c Chris@10: R ./rdft/ct.h Chris@10: M ./rdft/hc2hc-common.c -4 +33 Chris@10: M ./rdft/hc2hc-direct.c -26 +42 Chris@10: M ./rdft/hc2hc-directbuf.c -25 +44 Chris@10: M ./rdft/hc2hc-generic.c -37 +104 Chris@10: M ./rdft/hc2hc.c +221 Chris@10: M ./rdft/hc2hc.h +60 Chris@10: M ./rdft/khc2hc.c -8 +4 Chris@10: M ./rdft/rdft.h -3 +1 Chris@10: M ./threads/Makefile.am -2 +2 Chris@10: R ./threads/ct-dit.c Chris@10: M ./threads/ct.c +253 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: R ./threads/hc2hc-dif.c Chris@10: R ./threads/hc2hc-dit.c Chris@10: M ./threads/hc2hc.c +238 Chris@10: M ./threads/threads.c -27 +8 Chris@10: M ./threads/threads.h -5 +4 Chris@10: Chris@10: Thu Oct 21 20:44:51 EDT 2004 stevenj Chris@10: * [project @ 2004-10-22 00:44:51 by stevenj] Chris@10: more C++ notes Chris@10: Chris@10: M ./doc/fftw3.texi -4 +11 Chris@10: Chris@10: Thu Oct 14 09:50:38 EDT 2004 stevenj Chris@10: * [project @ 2004-10-14 13:50:38 by stevenj] Chris@10: note bug report for VC++ 6.0 from Dale Dickerhoof Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +4 Chris@10: Chris@10: Fri Oct 1 16:06:59 EDT 2004 stevenj Chris@10: * [project @ 2004-10-01 20:06:59 by stevenj] Chris@10: fmt Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: Chris@10: Fri Oct 1 15:59:17 EDT 2004 stevenj Chris@10: * [project @ 2004-10-01 19:58:50 by stevenj] Chris@10: comment typo Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -3 +3 Chris@10: Chris@10: Fri Oct 1 15:48:09 EDT 2004 stevenj Chris@10: * [project @ 2004-10-01 19:48:09 by stevenj] Chris@10: bug fix -- ishift/oshift only apply to execution of child plan Chris@10: Chris@10: M ./rdft/dft-r2hc.c -9 +4 Chris@10: Chris@10: Thu Sep 30 21:12:47 EDT 2004 athena Chris@10: * [project @ 2004-10-01 01:12:47 by athena] Chris@10: New planner that tries never to lose wisdom. Chris@10: Chris@10: M ./api/fftw3.h -2 +1 Chris@10: M ./api/mapflags.c -1 Chris@10: M ./kernel/ifftw.h -6 +4 Chris@10: M ./kernel/planner.c -137 +183 Chris@10: Chris@10: Thu Sep 30 13:36:43 EDT 2004 athena Chris@10: * [project @ 2004-09-30 17:36:43 by athena] Chris@10: Nested comment was triggering a warning. Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Fri Sep 10 15:20:07 EDT 2004 stevenj Chris@10: * [project @ 2004-09-10 19:20:07 by stevenj] Chris@10: system "root" under dgjpp is /dev/env/DJDIR, not /dev/env/DJGPP, Chris@10: according to djgpp's libc.info; patch confirmed with J. M. Guerrero Chris@10: Chris@10: M ./api/import-system-wisdom.c -1 +1 Chris@10: Chris@10: Wed Sep 8 18:50:03 EDT 2004 stevenj Chris@10: * [project @ 2004-09-08 22:50:03 by stevenj] Chris@10: some minor portability fixes for djgpp; thanks to Juan Manuel Guerrero for the patch Chris@10: Chris@10: M ./api/import-system-wisdom.c -1 +8 Chris@10: M ./tests/Makefile.am -4 +4 Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: Chris@10: Thu Aug 19 12:41:23 EDT 2004 stevenj Chris@10: * [project @ 2004-08-19 16:41:23 by stevenj] Chris@10: pointer to tutorial for quick start Chris@10: Chris@10: M ./README +3 Chris@10: Chris@10: Thu Aug 19 12:39:50 EDT 2004 stevenj Chris@10: * [project @ 2004-08-19 16:39:50 by stevenj] Chris@10: point users to manual Chris@10: Chris@10: M ./api/fftw3.h -1 +16 Chris@10: Chris@10: Sat Aug 7 13:42:22 EDT 2004 stevenj Chris@10: * [project @ 2004-08-07 17:42:22 by stevenj] Chris@10: minor typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sun Jul 18 18:54:18 EDT 2004 stevenj Chris@10: * [project @ 2004-07-18 22:54:18 by stevenj] Chris@10: use __DECCXX for Compaq cxx, not Linux-specific symbol Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Fri Jul 16 13:55:25 EDT 2004 stevenj Chris@10: * [project @ 2004-07-16 17:55:25 by stevenj] Chris@10: patch by John Bowman to make cycle counter work with DEC cxx under Linux Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Wed Jun 30 00:45:10 EDT 2004 stevenj Chris@10: * [project @ 2004-06-30 04:45:10 by stevenj] Chris@10: updated pruned FFT discussion, with link to further details on www.fftw.org/pruned.html Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -8 +16 Chris@10: M ./doc/FAQ/html.refs +1 Chris@10: Chris@10: Mon Jun 14 20:08:27 EDT 2004 stevenj Chris@10: * [project @ 2004-06-15 00:08:27 by stevenj] Chris@10: darwin is based on freebsd Chris@10: Chris@10: M ./acx_pthread.m4 -2 +2 Chris@10: Chris@10: Thu Jun 3 14:23:41 EDT 2004 stevenj Chris@10: * [project @ 2004-06-03 18:23:41 by stevenj] Chris@10: in --with-windows-f77-mangling, add lowercase + single underscore for Intel compilers, etc. (thanks to David Gomez for the bug report) Chris@10: Chris@10: M ./api/f77api.c +5 Chris@10: Chris@10: Wed Apr 7 00:46:07 EDT 2004 stevenj Chris@10: * [project @ 2004-04-07 04:46:07 by stevenj] Chris@10: whoops, extra alignment check Chris@10: Chris@10: M ./rdft/rank0-vrank2.c -1 +5 Chris@10: Chris@10: Wed Apr 7 00:16:49 EDT 2004 stevenj Chris@10: * [project @ 2004-04-07 04:16:49 by stevenj] Chris@10: disable most 2-float-as-double copying, add alignment check in one remaining place Chris@10: Chris@10: M ./kernel/ifftw.h -1 +11 Chris@10: M ./rdft/rank0-vrank2.c -22 +25 Chris@10: M ./rdft/vrank3-transpose.c -6 +6 Chris@10: Chris@10: Tue Apr 6 13:49:13 EDT 2004 stevenj Chris@10: * [project @ 2004-04-06 17:49:13 by stevenj] Chris@10: make sure it is clear that real-even/odd refers to symmetry, not size Chris@10: Chris@10: M ./doc/fftw3.texi -7 +7 Chris@10: Chris@10: Mon Apr 5 20:18:29 EDT 2004 stevenj Chris@10: * [project @ 2004-04-06 00:18:29 by stevenj] Chris@10: optimization Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -11 +48 Chris@10: Chris@10: Fri Apr 2 21:31:00 EST 2004 stevenj Chris@10: * [project @ 2004-04-03 02:31:00 by stevenj] Chris@10: separate cutoff for ugliness...these cutoffs are still not ideal Chris@10: Chris@10: M ./rdft/vrank3-transpose.c -3 +4 Chris@10: Chris@10: Fri Apr 2 21:30:17 EST 2004 stevenj Chris@10: * [project @ 2004-04-03 02:30:17 by stevenj] Chris@10: transpose.c is gone Chris@10: Chris@10: M ./kernel/ifftw.h -15 +1 Chris@10: Chris@10: Fri Apr 2 21:18:27 EST 2004 stevenj Chris@10: * [project @ 2004-04-03 02:18:27 by stevenj] Chris@10: move all rank0 transforms to rdft Chris@10: Chris@10: A ./rdft/rank0-vrank2.c Chris@10: A ./rdft/vrank3-transpose.c Chris@10: M ./configure.ac +1 Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/conf.c -4 +1 Chris@10: R ./dft/rank0.c Chris@10: R ./dft/vrank2-transpose.c Chris@10: R ./dft/vrank3-transpose.c Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: R ./kernel/transpose.c Chris@10: M ./rdft/Makefile.am -2 +3 Chris@10: M ./rdft/conf.c -1 +3 Chris@10: M ./rdft/dft-r2hc.c -20 +36 Chris@10: M ./rdft/rank0-vrank2.c +289 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/vrank3-transpose.c +943 Chris@10: Chris@10: Fri Apr 2 20:35:35 EST 2004 stevenj Chris@10: * [project @ 2004-04-03 01:35:35 by stevenj] Chris@10: enable fp-moves/us comparison of rank-0 transforms Chris@10: Chris@10: M ./libbench2/mflops.c -2 +8 Chris@10: M ./libbench2/report.c -3 +7 Chris@10: Chris@10: Thu Apr 1 16:13:22 EST 2004 stevenj Chris@10: * [project @ 2004-04-01 21:13:22 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/transpose.c -1 +1 Chris@10: Chris@10: Thu Apr 1 15:25:30 EST 2004 stevenj Chris@10: * [project @ 2004-04-01 20:25:30 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/tensor7.c -3 +3 Chris@10: Chris@10: Wed Mar 31 18:11:02 EST 2004 stevenj Chris@10: * [project @ 2004-03-31 23:11:02 by stevenj] Chris@10: sort tensor dims by stride absolute values, not strides Chris@10: Chris@10: M ./kernel/tensor7.c -5 +7 Chris@10: Chris@10: Tue Mar 30 20:22:50 EST 2004 stevenj Chris@10: * [project @ 2004-03-31 01:22:50 by stevenj] Chris@10: Chris@10: M ./kernel/transpose.c -2 +1 Chris@10: Chris@10: Tue Mar 30 19:44:54 EST 2004 stevenj Chris@10: * [project @ 2004-03-31 00:44:54 by stevenj] Chris@10: added improved transpose algorithm for N x M where |N-M| is small Chris@10: Chris@10: M ./dft/dftw-generic.c +1 Chris@10: M ./dft/vrank2-transpose.c -7 +10 Chris@10: M ./dft/vrank3-transpose.c -6 +7 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/transpose.c -15 +73 Chris@10: M ./libbench2/problem.c -2 +4 Chris@10: Chris@10: Tue Mar 30 19:41:14 EST 2004 stevenj Chris@10: * [project @ 2004-03-31 00:41:14 by stevenj] Chris@10: check to make sure SIMD matches precision, and make sure user doesn't select both SSE and SSE2 Chris@10: Chris@10: M ./configure.ac -2 +7 Chris@10: Chris@10: Sun Mar 28 09:26:38 EST 2004 athena Chris@10: * [project @ 2004-03-28 14:26:38 by athena] Chris@10: Implemented hc2hc-generic hc2r. Chris@10: Chris@10: M ./rdft/hc2hc-generic.c -29 +79 Chris@10: Chris@10: Thu Mar 25 11:19:25 EST 2004 athena Chris@10: * [project @ 2004-03-25 16:19:25 by athena] Chris@10: Inverted loop for stride-1 access. Chris@10: Chris@10: M ./rdft/hc2hc-generic.c -12 +18 Chris@10: Chris@10: Thu Mar 25 11:18:49 EST 2004 athena Chris@10: * [project @ 2004-03-25 16:18:49 by athena] Chris@10: Swapped j <-> k for consistency Chris@10: Chris@10: M ./dft/dftw-generic.c -3 +3 Chris@10: Chris@10: Tue Mar 23 12:08:07 EST 2004 athena Chris@10: * [project @ 2004-03-23 17:08:07 by athena] Chris@10: Require that R be odd Chris@10: Chris@10: M ./rdft/hc2hc-generic.c -1 +1 Chris@10: Chris@10: Tue Mar 23 11:49:01 EST 2004 athena Chris@10: * [project @ 2004-03-23 16:49:01 by athena] Chris@10: Implemented hc2hc-generic (DIT only for now). Chris@10: Chris@10: A ./rdft/hc2hc-generic.c Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/conf.c -5 +4 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/hc2hc-generic.c +216 Chris@10: M ./rdft/rdft.h +1 Chris@10: Chris@10: Mon Mar 22 14:43:16 EST 2004 athena Chris@10: * [project @ 2004-03-22 19:43:16 by athena] Chris@10: Relax equality of twiddle description, since the `i' field Chris@10: is not used by TW_FULL or TW_HALF. Chris@10: Chris@10: M ./kernel/twiddle.c -5 +17 Chris@10: Chris@10: Mon Mar 22 13:22:44 EST 2004 athena Chris@10: * [project @ 2004-03-22 18:22:41 by athena] Chris@10: Do not allocate tw_instr's on the stack. Thus, the ``consistency check'' Chris@10: in twiddle.c becomes wrong. Chris@10: Chris@10: M ./dft/dftw-generic.c -8 +1 Chris@10: M ./kernel/twiddle.c -2 +1 Chris@10: Chris@10: Mon Mar 22 13:21:28 EST 2004 athena Chris@10: * [project @ 2004-03-22 18:21:28 by athena] Chris@10: Fixed incorrect malloc()/free() logic. Chris@10: Chris@10: M ./libbench2/mp.c -3 +6 Chris@10: Chris@10: Mon Mar 22 09:04:37 EST 2004 athena Chris@10: * [project @ 2004-03-22 14:04:37 by athena] Chris@10: Silence warnings Chris@10: Chris@10: M ./rdft/hc2hc-directbuf.c -2 +2 Chris@10: Chris@10: Mon Mar 22 09:02:55 EST 2004 athena Chris@10: * [project @ 2004-03-22 14:02:55 by athena] Chris@10: Separate file for hc2hc common routines Chris@10: Chris@10: A ./rdft/hc2hc-common.c Chris@10: M ./rdft/Makefile.am -4 +4 Chris@10: M ./rdft/ct.c -35 Chris@10: M ./rdft/hc2hc-common.c +58 Chris@10: Chris@10: Mon Mar 22 08:23:56 EST 2004 athena Chris@10: * [project @ 2004-03-22 13:23:56 by athena] Chris@10: (re)Implemented buffered hc2hc. Slight simplification of Chris@10: twiddle-factors management. Chris@10: Chris@10: A ./rdft/hc2hc-directbuf.c Chris@10: M ./dft/dftw-directbuf.c -2 +2 Chris@10: M ./rdft/Makefile.am -2 +3 Chris@10: M ./rdft/ct.h +2 Chris@10: M ./rdft/hc2hc-direct.c -16 +8 Chris@10: M ./rdft/hc2hc-directbuf.c +238 Chris@10: M ./rdft/khc2hc.c -1 +2 Chris@10: Chris@10: Sun Mar 21 19:53:05 EST 2004 athena Chris@10: * [project @ 2004-03-22 00:53:05 by athena] Chris@10: Incremented libtool revision number before we forget. Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sun Mar 21 19:25:56 EST 2004 athena Chris@10: * [project @ 2004-03-22 00:25:56 by athena] Chris@10: Fixed opcnt Chris@10: Chris@10: M ./rdft/hc2hc-direct.c -2 +5 Chris@10: Chris@10: Sun Mar 21 17:56:15 EST 2004 athena Chris@10: * [project @ 2004-03-21 22:56:15 by athena] Chris@10: Renamed files. These solvers are not really cooley-tukey. Chris@10: Chris@10: A ./dft/dftw-direct.c Chris@10: A ./dft/dftw-directbuf.c Chris@10: A ./dft/dftw-generic.c Chris@10: M ./dft/Makefile.am -5 +5 Chris@10: R ./dft/ct-directw.c Chris@10: R ./dft/ct-directwbuf.c Chris@10: R ./dft/ct-generic.c Chris@10: M ./dft/dftw-direct.c +153 Chris@10: M ./dft/dftw-directbuf.c +213 Chris@10: M ./dft/dftw-generic.c +304 Chris@10: Chris@10: Sun Mar 21 12:38:45 EST 2004 athena Chris@10: * [project @ 2004-03-21 17:38:45 by athena] Chris@10: Started moving rdft/ to the new cooley-tukey ontology Chris@10: Chris@10: A ./rdft/ct.c Chris@10: A ./rdft/ct.h Chris@10: A ./rdft/hc2hc-direct.c Chris@10: A ./rdft/khc2hc.c Chris@10: M ./dft/ct.h -1 +1 Chris@10: M ./genfft/gen_hc2hc.ml -6 +3 Chris@10: M ./rdft/Makefile.am -5 +4 Chris@10: M ./rdft/codelet-rdft.h -3 +1 Chris@10: M ./rdft/ct.c +253 Chris@10: M ./rdft/ct.h +54 Chris@10: R ./rdft/hc2hc-buf.c Chris@10: R ./rdft/hc2hc-dif.c Chris@10: M ./rdft/hc2hc-direct.c +182 Chris@10: R ./rdft/hc2hc-dit.c Chris@10: R ./rdft/hc2hc.c Chris@10: R ./rdft/hc2hc.h Chris@10: R ./rdft/khc2hc-dif.c Chris@10: R ./rdft/khc2hc-dit.c Chris@10: M ./rdft/khc2hc.c +32 Chris@10: M ./rdft/rdft.h -6 +1 Chris@10: Chris@10: Sun Mar 21 10:59:42 EST 2004 athena Chris@10: * [project @ 2004-03-21 15:59:42 by athena] Chris@10: Plans in ct-*.c are subtypes of plan_dftw, not plan_dft Chris@10: Chris@10: M ./dft/ct-directw.c -2 +2 Chris@10: M ./dft/ct-directwbuf.c -2 +2 Chris@10: M ./dft/ct-generic.c -1 +1 Chris@10: Chris@10: Sun Mar 21 10:38:18 EST 2004 athena Chris@10: * [project @ 2004-03-21 15:38:18 by athena] Chris@10: Slight simplification Chris@10: Chris@10: M ./dft/ct-directw.c -2 +2 Chris@10: Chris@10: Sun Mar 21 10:20:06 EST 2004 athena Chris@10: * [project @ 2004-03-21 15:20:06 by athena] Chris@10: Minor simplification Chris@10: Chris@10: M ./dft/ct.c -4 +2 Chris@10: Chris@10: Sat Mar 20 08:43:57 EST 2004 athena Chris@10: * [project @ 2004-03-20 13:43:57 by athena] Chris@10: Workarounds for icc-8.0 nonsense. Chris@10: Chris@10: M ./simd/simd-sse.h +6 Chris@10: M ./simd/simd-sse2.h +6 Chris@10: Chris@10: Sun Mar 7 07:56:08 EST 2004 athena Chris@10: * [project @ 2004-03-07 12:56:08 by athena] Chris@10: FFTW_FORWARD is not technically an ``option''. Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Tue Feb 24 12:17:06 EST 2004 stevenj Chris@10: * [project @ 2004-02-24 17:17:06 by stevenj] Chris@10: Alejandro requested that his name be removed from @author Chris@10: Chris@10: M ./acx_pthread.m4 -4 +5 Chris@10: Chris@10: Mon Feb 23 17:42:56 EST 2004 stevenj Chris@10: * [project @ 2004-02-23 22:42:56 by stevenj] Chris@10: GNU Pth emulation library check Chris@10: Chris@10: M ./acx_pthread.m4 -3 +12 Chris@10: Chris@10: Sat Feb 21 17:51:13 EST 2004 stevenj Chris@10: * [project @ 2004-02-21 22:51:13 by stevenj] Chris@10: calling can-do calls the estimating-planner, which creates wisdom that we don't want ...we should be able to do all of the documented problems, anyway Chris@10: Chris@10: M ./tools/fftw-wisdom.c -1 +1 Chris@10: Chris@10: Sat Feb 21 17:46:06 EST 2004 stevenj Chris@10: * [project @ 2004-02-21 22:46:06 by stevenj] Chris@10: don't forget_wisdom because of side effects Chris@10: Chris@10: M ./tests/bench.c -1 Chris@10: Chris@10: Sat Feb 21 17:42:47 EST 2004 stevenj Chris@10: * [project @ 2004-02-21 22:42:47 by stevenj] Chris@10: forget wisdom from can_do Chris@10: Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Thu Feb 19 14:11:14 EST 2004 stevenj Chris@10: * [project @ 2004-02-19 19:11:14 by stevenj] Chris@10: parenthesization Chris@10: Chris@10: M ./api/malloc.c -1 +1 Chris@10: Chris@10: Fri Feb 13 07:20:31 EST 2004 athena Chris@10: * [project @ 2004-02-13 12:20:31 by athena] Chris@10: Split malloc into kernel_malloc and API malloc Chris@10: Chris@10: A ./api/malloc.c Chris@10: A ./kernel/kalloc.c Chris@10: M ./api/Makefile.am -14 +14 Chris@10: M ./api/malloc.c +32 Chris@10: M ./kernel/Makefile.am -5 +5 Chris@10: M ./kernel/alloc.c -127 +5 Chris@10: M ./kernel/ifftw.h -1 +6 Chris@10: M ./kernel/kalloc.c +140 Chris@10: M ./tests/bench.c +4 Chris@10: Chris@10: Thu Feb 12 15:42:20 EST 2004 stevenj Chris@10: * [project @ 2004-02-12 20:42:20 by stevenj] Chris@10: X(malloc) must be extern "C" Chris@10: Chris@10: M ./kernel/alloc.c -1 +4 Chris@10: Chris@10: Thu Feb 12 15:41:44 EST 2004 stevenj Chris@10: * [project @ 2004-02-12 20:41:44 by stevenj] Chris@10: satsify C++ compiler Chris@10: Chris@10: M ./dft/bluestein.c -2 +2 Chris@10: Chris@10: Thu Feb 5 20:39:14 EST 2004 stevenj Chris@10: * [project @ 2004-02-06 01:39:14 by stevenj] Chris@10: with the new flags, fma is definitely beneficial on PA-RISC with HP/UX cc Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -6 +7 Chris@10: Chris@10: Thu Feb 5 19:52:17 EST 2004 stevenj Chris@10: * [project @ 2004-02-06 00:52:17 by stevenj] Chris@10: grr, Ofaster etcetera are not supported under older versions of the compiler. Note that +Ofltacc *disables* fp-reordering optimizations (which are enabled by +Oall). +Optrs_ansi is the older version of the aliasing stuff Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Thu Feb 5 19:26:01 EST 2004 stevenj Chris@10: * [project @ 2004-02-06 00:26:01 by stevenj] Chris@10: +Otype_safety=ansi on hpux Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Thu Feb 5 19:22:34 EST 2004 stevenj Chris@10: * [project @ 2004-02-06 00:22:34 by stevenj] Chris@10: just use +Ofaster on hpux (+O3 +Onolimit +Olibcalls +Ofltacc=relaxed -Wl,+mergeseg) Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Fri Jan 30 14:17:15 EST 2004 stevenj Chris@10: * [project @ 2004-01-30 19:17:15 by stevenj] Chris@10: check for win32 threads for mingw32; thanks to Alessio Massaro Chris@10: Chris@10: M ./configure.ac +8 Chris@10: Chris@10: Thu Jan 29 15:23:33 EST 2004 stevenj Chris@10: * [project @ 2004-01-29 20:23:33 by stevenj] Chris@10: added missing 'static', thanks to Alessio Massaro Chris@10: Chris@10: M ./threads/threads.c -3 +3 Chris@10: Chris@10: Fri Jan 9 16:36:48 EST 2004 stevenj Chris@10: * [project @ 2004-01-09 21:36:48 by stevenj] Chris@10: print more like bluestein Chris@10: Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Fri Jan 9 15:45:22 EST 2004 stevenj Chris@10: * [project @ 2004-01-09 20:45:22 by stevenj] Chris@10: fixed op count for R2HC_ONLY_CONV Chris@10: Chris@10: M ./rdft/dht-rader.c -3 +2 Chris@10: Chris@10: Fri Jan 9 15:41:50 EST 2004 stevenj Chris@10: * [project @ 2004-01-09 20:41:50 by stevenj] Chris@10: include DESTROY_INPUT in buffered flags for in-place...otherwise in-place hc2r uses rdft-dhtcvs diff Chris@10: Chris@10: M ./dft/buffered.c -1 +10 Chris@10: M ./rdft/buffered.c -1 +10 Chris@10: M ./rdft/buffered2.c -1 +9 Chris@10: Chris@10: Fri Jan 9 15:41:09 EST 2004 stevenj Chris@10: * [project @ 2004-01-09 20:41:09 by stevenj] Chris@10: resurrected R2HC_ONLY_CONV option to share plans and save on planning time Chris@10: Chris@10: M ./rdft/dht-rader.c -3 +46 Chris@10: Chris@10: Fri Jan 9 14:47:00 EST 2004 stevenj Chris@10: * [project @ 2004-01-09 19:47:00 by stevenj] Chris@10: precompute folding for cyclic convolution Chris@10: Chris@10: M ./rdft/dht-rader.c -13 +9 Chris@10: Chris@10: Wed Jan 7 16:48:39 EST 2004 stevenj Chris@10: * [project @ 2004-01-07 21:48:39 by stevenj] Chris@10: minor Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Wed Jan 7 16:48:25 EST 2004 stevenj Chris@10: * [project @ 2004-01-07 21:48:25 by stevenj] Chris@10: note reports of successful compilation on Windows Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -5 +5 Chris@10: Chris@10: Wed Jan 7 14:16:16 EST 2004 stevenj Chris@10: * [project @ 2004-01-07 19:16:16 by stevenj] Chris@10: citation year Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: Chris@10: Tue Jan 6 01:07:36 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 06:07:36 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Tue Jan 6 01:07:08 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 06:07:08 by stevenj] Chris@10: comment fix Chris@10: Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Tue Jan 6 01:06:57 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 06:06:57 by stevenj] Chris@10: fixed naming cruft Chris@10: Chris@10: M ./rdft/dht-rader.c -18 +18 Chris@10: Chris@10: Tue Jan 6 00:56:16 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 05:56:16 by stevenj] Chris@10: space Chris@10: Chris@10: M ./rdft/dht-rader.c -1 Chris@10: Chris@10: Tue Jan 6 00:55:53 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 05:55:45 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Tue Jan 6 00:54:07 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 05:54:07 by stevenj] Chris@10: moved assert Chris@10: Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: Chris@10: Tue Jan 6 00:49:16 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 05:49:16 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./rdft/dht-rader.c +5 Chris@10: Chris@10: Tue Jan 6 00:41:06 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 05:41:06 by stevenj] Chris@10: delete old R2HC_ONLY_CONV hack, now defunct Chris@10: Chris@10: M ./rdft/dht-rader.c -46 Chris@10: Chris@10: Tue Jan 6 00:32:58 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 05:32:58 by stevenj] Chris@10: added padded real rader Chris@10: Chris@10: M ./rdft/dht-rader.c -56 +95 Chris@10: Chris@10: Mon Jan 5 22:56:58 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 03:56:58 by stevenj] Chris@10: removed unused var Chris@10: Chris@10: M ./rdft/generic.c -1 +1 Chris@10: Chris@10: Mon Jan 5 21:20:29 EST 2004 stevenj Chris@10: * [project @ 2004-01-06 02:20:29 by stevenj] Chris@10: handle both FFT_SIGN values Chris@10: Chris@10: M ./rdft/generic.c -2 +14 Chris@10: Chris@10: Fri Jan 2 06:07:51 EST 2004 athena Chris@10: * [project @ 2004-01-02 11:07:51 by athena] Chris@10: Oops: d->ros ==> d->ios Chris@10: Chris@10: M ./rdft/codelets/r2hc.c -1 +1 Chris@10: Chris@10: Fri Jan 2 06:05:10 EST 2004 athena Chris@10: * [project @ 2004-01-02 11:05:10 by athena] Chris@10: Oops: d->ris should have been d->iis Chris@10: Chris@10: M ./rdft/codelets/hc2r.c -1 +1 Chris@10: Chris@10: Thu Jan 1 16:00:07 EST 2004 athena Chris@10: * [project @ 2004-01-01 21:00:07 by athena] Chris@10: Removed rdft rader cooley-tukey, to be superseded by a generic Chris@10: reduction of rdft twiddle problems to dft + pre/post processing Chris@10: Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/dft.h -5 +1 Chris@10: R ./dft/rader-omega.c Chris@10: M ./dft/rader.c -2 +38 Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/conf.c -2 +1 Chris@10: R ./rdft/rader-hc2hc.c Chris@10: Chris@10: Thu Jan 1 15:44:09 EST 2004 athena Chris@10: * [project @ 2004-01-01 20:44:09 by athena] Chris@10: In anticipation of the upcoming revision of rdft, removed rdft generic Chris@10: dit/dif cooley-tukey, in favor of generic rh2c and hc2r solvers. Chris@10: Cleaned up stuff that became unused after this change, such as Chris@10: TW_GENERIC. Chris@10: Chris@10: M ./dft/ct.c -33 +3 Chris@10: M ./dft/generic.c -1 +2 Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./kernel/primes.c -1 +32 Chris@10: M ./kernel/twiddle.c -18 +1 Chris@10: M ./rdft/generic.c -258 +103 Chris@10: Chris@10: Thu Jan 1 12:59:30 EST 2004 athena Chris@10: * [project @ 2004-01-01 17:59:30 by athena] Chris@10: Removed useless file Chris@10: Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: R ./kernel/square.c Chris@10: Chris@10: Fri Dec 26 13:54:00 EST 2003 stevenj Chris@10: * [project @ 2003-12-26 18:54:00 by stevenj] Chris@10: whoops, don't call AC_F77_DUMMY_MAIN if no Fortran compiler is found; thanks to Charles Radley for the bug report. Chris@10: Chris@10: M ./configure.ac -2 +3 Chris@10: Chris@10: Fri Dec 19 13:58:05 EST 2003 stevenj Chris@10: * [project @ 2003-12-19 18:58:05 by stevenj] Chris@10: guess good flags for Solaris/intel, suggested by J. Gregory Wright Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Fri Dec 5 19:55:13 EST 2003 stevenj Chris@10: * [project @ 2003-12-06 00:55:13 by stevenj] Chris@10: blah Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: M ./doc/FAQ/html.refs -1 +1 Chris@10: Chris@10: Sun Nov 30 06:59:41 EST 2003 athena Chris@10: * [project @ 2003-11-30 11:59:41 by athena] Chris@10: DIF generic solver was destroying the input. Chris@10: Chris@10: M ./rdft/generic.c -2 +16 Chris@10: Chris@10: Sat Nov 29 19:28:39 EST 2003 athena Chris@10: * [project @ 2003-11-30 00:28:39 by athena] Chris@10: Fixed bug that caused HC2R transforms to destroy the input in Chris@10: certain cases, even if the user specified FFTW_PRESERVE_INPUT. Chris@10: Chris@10: M ./NEWS +5 Chris@10: M ./rdft/rader-hc2hc.c -2 +14 Chris@10: Chris@10: Sat Nov 29 16:49:01 EST 2003 athena Chris@10: * [project @ 2003-11-29 21:49:01 by athena] Chris@10: Implemented swap_io hack for r2r verifier. Chris@10: Chris@10: M ./libbench2/verify-r2r.c -1 +8 Chris@10: Chris@10: Thu Nov 20 22:00:53 EST 2003 stevenj Chris@10: * [project @ 2003-11-21 03:00:53 by stevenj] Chris@10: citation Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -2 +6 Chris@10: Chris@10: Fri Nov 14 20:57:55 EST 2003 athena Chris@10: * [project @ 2003-11-15 01:57:55 by athena] Chris@10: Trying to get ``make paranoid-check'' to work. (Still broken.) Chris@10: Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./tests/hook.c -9 +8 Chris@10: Chris@10: Fri Nov 14 20:05:54 EST 2003 stevenj Chris@10: * [project @ 2003-11-15 01:05:54 by stevenj] Chris@10: fixes for input-preservation tests Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +2 Chris@10: M ./libbench2/tensor.c -1 +15 Chris@10: M ./libbench2/verify-dft.c -3 +12 Chris@10: M ./libbench2/verify-lib.c -1 +4 Chris@10: M ./libbench2/verify-r2r.c -13 +33 Chris@10: M ./libbench2/verify-rdft2.c -6 +22 Chris@10: M ./libbench2/verify.h -1 +1 Chris@10: Chris@10: Fri Nov 14 19:19:31 EST 2003 athena Chris@10: * [project @ 2003-11-15 00:19:31 by athena] Chris@10: Assume FFTW_PRESERVE_INPUT unless either the `d' flag is given in the Chris@10: problem, or the problem is multidimensional c2r (which fftw3 cannot Chris@10: without destroying the input). With this change, we can at least test Chris@10: that FFTW_PRESERVE_INPUT works in the c2r 1d case. Chris@10: Chris@10: M ./tests/bench.c -12 +21 Chris@10: Chris@10: Fri Nov 14 19:14:40 EST 2003 stevenj Chris@10: * [project @ 2003-11-15 00:14:40 by stevenj] Chris@10: apply should copy back input for input-preservation check Chris@10: Chris@10: M ./libbench2/verify-dft.c -1 +3 Chris@10: M ./libbench2/verify-r2r.c +13 Chris@10: M ./libbench2/verify-rdft2.c -1 +4 Chris@10: Chris@10: Fri Nov 14 19:01:36 EST 2003 athena Chris@10: * [project @ 2003-11-15 00:01:14 by athena] Chris@10: Undone previous bogus changes Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -1 +10 Chris@10: M ./tests/bench.c -14 +10 Chris@10: M ./tests/check.pl -1 Chris@10: Chris@10: Fri Nov 14 18:27:12 EST 2003 athena Chris@10: * [project @ 2003-11-14 23:27:12 by athena] Chris@10: Check dr[fb] in addition to r[fb] Chris@10: Chris@10: M ./tests/check.pl +1 Chris@10: Chris@10: Fri Nov 14 17:33:44 EST 2003 athena Chris@10: * [project @ 2003-11-14 22:33:42 by athena] Chris@10: Fixed conditions under which the rank-geq2-rdft2 solver is applicable. Chris@10: Chris@10: The old solver was not applicable for out-of-place problems Chris@10: unless DESTROY_INPUT. This is bogus. As long as the subsolvers Chris@10: honor !DESTROY_INPUT, the solver is always applicable. Chris@10: Chris@10: Changed semantics of test program, so that PRESERVE_INPUT is always Chris@10: true unless the problem specifies destroy_input explicitly. Without Chris@10: this change, there is no way to test the new solver. Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -10 +1 Chris@10: M ./tests/bench.c -10 +14 Chris@10: Chris@10: Thu Oct 30 15:10:42 EST 2003 stevenj Chris@10: * [project @ 2003-10-30 20:10:42 by stevenj] Chris@10: added AIX OpenMP (-qsmp=omp) support; thanks to Greg Bauer Chris@10: Chris@10: M ./configure.ac +5 Chris@10: Chris@10: Thu Oct 30 10:11:39 EST 2003 athena Chris@10: * [project @ 2003-10-30 15:11:39 by athena] Chris@10: G5 CFLAGS Chris@10: Chris@10: M ./acinclude.m4 +6 Chris@10: Chris@10: Fri Oct 24 04:17:39 EDT 2003 stevenj Chris@10: * [project @ 2003-10-24 08:17:39 by stevenj] Chris@10: western FAQ Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +5 Chris@10: Chris@10: Thu Oct 23 11:34:11 EDT 2003 athena Chris@10: * [project @ 2003-10-23 15:34:11 by athena] Chris@10: Oops. Chris@10: Chris@10: M ./simd/altivec.c -1 +2 Chris@10: Chris@10: Thu Oct 23 11:28:28 EDT 2003 athena Chris@10: * [project @ 2003-10-23 15:28:27 by athena] Chris@10: Autodetect altivec Chris@10: Chris@10: M ./configure.ac -2 +3 Chris@10: M ./simd/altivec.c -1 +28 Chris@10: M ./simd/simd-altivec.h -1 +2 Chris@10: Chris@10: Wed Oct 22 01:14:10 EDT 2003 stevenj Chris@10: * [project @ 2003-10-22 05:14:10 by stevenj] Chris@10: MinGW gets confused by a single / Chris@10: Chris@10: M ./tests/check.pl -4 +4 Chris@10: Chris@10: Fri Oct 17 10:46:41 EDT 2003 athena Chris@10: * [project @ 2003-10-17 14:46:41 by athena] Chris@10: Paranoid portability fix Chris@10: Chris@10: M ./libbench2/mp.c -2 +2 Chris@10: Chris@10: Thu Oct 16 11:07:46 EDT 2003 athena Chris@10: * [project @ 2003-10-16 15:07:46 by athena] Chris@10: size -> length, which should make clear that we are not talking Chris@10: about arbitrary precision. Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Wed Oct 15 15:01:40 EDT 2003 stevenj Chris@10: * [project @ 2003-10-15 19:01:40 by stevenj] Chris@10: pruned transforms are a FAQ Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +11 Chris@10: Chris@10: Wed Oct 8 23:54:17 EDT 2003 stevenj Chris@10: * [project @ 2003-10-09 03:54:17 by stevenj] Chris@10: NO_SEARCH has already been mapped to FFTW_WISDOM_ONLY Chris@10: Chris@10: M ./TODO -1 +1 Chris@10: Chris@10: Wed Oct 8 23:53:19 EDT 2003 stevenj Chris@10: * [project @ 2003-10-09 03:53:19 by stevenj] Chris@10: newline Chris@10: Chris@10: M ./TODO +1 Chris@10: Chris@10: Sat Sep 27 20:27:32 EDT 2003 stevenj Chris@10: * [project @ 2003-09-28 00:27:32 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Sep 27 20:24:39 EDT 2003 stevenj Chris@10: * [project @ 2003-09-28 00:24:39 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -2 +4 Chris@10: Chris@10: Sat Sep 27 17:43:57 EDT 2003 stevenj Chris@10: * [project @ 2003-09-27 21:43:57 by stevenj] Chris@10: minor fix Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Sep 27 17:42:30 EDT 2003 stevenj Chris@10: * [project @ 2003-09-27 21:42:30 by stevenj] Chris@10: grammar Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Sep 27 17:29:04 EDT 2003 stevenj Chris@10: * [project @ 2003-09-27 21:29:04 by stevenj] Chris@10: html output fix Chris@10: Chris@10: M ./doc/fftw3.texi -5 +5 Chris@10: Chris@10: Sat Sep 27 17:22:48 EDT 2003 stevenj Chris@10: * [project @ 2003-09-27 21:22:48 by stevenj] Chris@10: mentioned sqrt(2) factors for DCT/DST Chris@10: Chris@10: M ./doc/fftw3.texi -5 +48 Chris@10: Chris@10: Sat Sep 27 17:07:18 EDT 2003 stevenj Chris@10: * [project @ 2003-09-27 21:07:18 by stevenj] Chris@10: FFTW_WISDOM_ONLY flag (undocumented for now), suggested by Phil Dumont Chris@10: Chris@10: M ./api/fftw3.h -1 +2 Chris@10: M ./api/mapflags.c +1 Chris@10: Chris@10: Tue Sep 23 23:36:19 EDT 2003 stevenj Chris@10: * [project @ 2003-09-24 03:36:19 by stevenj] Chris@10: removed UpTime code Chris@10: Chris@10: M ./kernel/cycle.h -20 +3 Chris@10: Chris@10: Tue Sep 23 23:27:29 EDT 2003 stevenj Chris@10: * [project @ 2003-09-24 03:27:29 by stevenj] Chris@10: updated documentation for mach_absolute_time Chris@10: Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: Chris@10: Tue Sep 23 23:25:52 EDT 2003 stevenj Chris@10: * [project @ 2003-09-24 03:25:52 by stevenj] Chris@10: use mach_absolute_time on MacOS/Darwin, as a fallback; don't bother checking for UpTime since it requires extra libs Chris@10: Chris@10: M ./configure.ac -3 +2 Chris@10: M ./kernel/cycle.h -1 +11 Chris@10: Chris@10: Tue Sep 23 22:59:29 EDT 2003 stevenj Chris@10: * [project @ 2003-09-24 02:59:29 by stevenj] Chris@10: support Apple UpTime function for asm-less xlc, grrr... Chris@10: Chris@10: M ./configure.ac -2 +3 Chris@10: M ./kernel/cycle.h -1 +19 Chris@10: Chris@10: Tue Sep 23 15:42:29 EDT 2003 stevenj Chris@10: * [project @ 2003-09-23 19:42:29 by stevenj] Chris@10: additional paranoia for xlc etc. Chris@10: Chris@10: M ./api/api.h -1 +1 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Mon Sep 22 15:28:56 EDT 2003 stevenj Chris@10: * [project @ 2003-09-22 19:28:56 by stevenj] Chris@10: work around _Complex_I weirdness in xlc, reported by Greg Allen Chris@10: Chris@10: M ./api/api.h -2 +4 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Fri Sep 5 18:03:11 EDT 2003 stevenj Chris@10: * [project @ 2003-09-05 22:03:11 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +2 Chris@10: Chris@10: Fri Sep 5 13:11:40 EDT 2003 athena Chris@10: * [project @ 2003-09-05 17:11:40 by athena] Chris@10: New script that produces commercial version. Chris@10: Chris@10: A ./commercialize.sh Chris@10: Chris@10: Fri Sep 5 07:27:06 EDT 2003 athena Chris@10: * [project @ 2003-09-05 11:27:06 by athena] Chris@10: Noted that VC++ is buggy. Noted that we know nothing about Windows. Chris@10: Noted that the sky is blue as well. Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -10 +22 Chris@10: Chris@10: Tue Sep 2 09:04:19 EDT 2003 athena Chris@10: * [project @ 2003-09-02 13:04:19 by athena] Chris@10: Noted that certain arrays are no longer used after the planner has Chris@10: completed. Chris@10: Chris@10: M ./doc/fftw3.texi -2 +12 Chris@10: Chris@10: Tue Aug 26 08:22:38 EDT 2003 athena Chris@10: * [project @ 2003-08-26 12:22:38 by athena] Chris@10: Typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Aug 25 21:27:43 EDT 2003 athena Chris@10: * [project @ 2003-08-26 01:27:43 by athena] Chris@10: New item Chris@10: Chris@10: M ./TODO -2 +1 Chris@10: Chris@10: Thu Aug 21 17:36:08 EDT 2003 stevenj Chris@10: * [project @ 2003-08-21 21:36:08 by stevenj] Chris@10: try creating output file before planning (thanks to Phil Dumont for the suggestion) Chris@10: Chris@10: M ./tools/fftw-wisdom.c -10 +10 Chris@10: Chris@10: Tue Aug 19 10:08:07 EDT 2003 athena Chris@10: * [project @ 2003-08-19 14:08:07 by athena] Chris@10: Clarified fftw_cleanup() Chris@10: Chris@10: M ./doc/fftw3.texi -4 +7 Chris@10: Chris@10: Sat Aug 16 03:13:41 EDT 2003 stevenj Chris@10: * [project @ 2003-08-16 07:13:41 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Jul 28 18:01:13 EDT 2003 stevenj Chris@10: * [project @ 2003-07-28 22:01:13 by stevenj] Chris@10: use time() instead of clock() (FIXME: what to do for non-POSIX systems?) ...thanks to JP Sugarbroad and James A. Treacy for the bug report Chris@10: Chris@10: M ./tools/fftw-wisdom.c -4 +4 Chris@10: Chris@10: Thu Jul 24 18:58:10 EDT 2003 athena Chris@10: * [project @ 2003-07-24 22:58:10 by athena] Chris@10: Need __volatile__ in sparc cycle counter. This is why the debian Chris@10: port hangs. Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Sun Jul 20 16:02:43 EDT 2003 stevenj Chris@10: * [project @ 2003-07-20 20:02:43 by stevenj] Chris@10: merged 3.0.1 notes Chris@10: Chris@10: M ./NEWS +30 Chris@10: Chris@10: Sun Jul 13 20:57:34 EDT 2003 stevenj Chris@10: * [project @ 2003-07-14 00:57:34 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./libbench2/bench-main.c -2 +2 Chris@10: Chris@10: Thu Jul 10 11:48:50 EDT 2003 athena Chris@10: * [project @ 2003-07-10 15:48:50 by athena] Chris@10: Dealing with constants in a way that seems to confuse gcc less. Chris@10: Chris@10: M ./simd/simd-sse.h -5 +2 Chris@10: M ./simd/simd-sse2.h -6 +3 Chris@10: Chris@10: Wed Jul 9 17:39:23 EDT 2003 athena Chris@10: * [project @ 2003-07-09 21:39:16 by athena] Chris@10: Enabled scheduler hack for FMA, where it seems to help. Chris@10: Chris@10: M ./genfft/annotate.ml -7 +3 Chris@10: M ./genfft/magic.ml -1 +5 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Wed Jul 9 12:57:38 EDT 2003 athena Chris@10: * [project @ 2003-07-09 16:57:38 by athena] Chris@10: Hmm---the new scheduler seems make things worse for gcc/x86, better Chris@10: for gcc/ppc, and about the same for icc/x86. Disabled for now. Chris@10: Chris@10: M ./genfft/annotate.ml -3 +6 Chris@10: Chris@10: Wed Jul 9 08:09:53 EDT 2003 athena Chris@10: * [project @ 2003-07-09 12:09:53 by athena] Chris@10: New scheduling pass that keeps ``x = a + b'' and ``y = a - b'' close Chris@10: together. This property was no longer automatic for the dags Chris@10: generated in SIMD mode. Chris@10: Chris@10: I cannot measure any speed difference due to this change. However, Chris@10: the change is justified by a minimal-screwup argument. Moreover, the Chris@10: sse2 fftw library is now 1% smaller than it was before. Chris@10: Chris@10: M ./genfft/annotate.ml -32 +70 Chris@10: Chris@10: Tue Jul 8 20:42:22 EDT 2003 athena Chris@10: * [project @ 2003-07-09 00:42:22 by athena] Chris@10: -(FNMS()) => FMS() Chris@10: Chris@10: M ./genfft/c.ml -1 +2 Chris@10: Chris@10: Sun Jul 6 13:53:23 EDT 2003 stevenj Chris@10: * [project @ 2003-07-06 17:53:23 by stevenj] Chris@10: added more convenient target name Chris@10: Chris@10: M ./doc/FAQ/Makefile.am +2 Chris@10: Chris@10: Sat Jul 5 13:30:10 EDT 2003 stevenj Chris@10: * [project @ 2003-07-05 17:30:10 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Sat Jul 5 13:19:36 EDT 2003 athena Chris@10: * [project @ 2003-07-05 17:19:36 by athena] Chris@10: Consistent naming Chris@10: Chris@10: M ./dft/ct-generic.c -2 +2 Chris@10: Chris@10: Sat Jul 5 13:05:51 EDT 2003 athena Chris@10: * [project @ 2003-07-05 17:05:51 by athena] Chris@10: Got rid of problemw. Chris@10: Chris@10: A ./dft/ct-directw.c Chris@10: A ./dft/ct-directwbuf.c Chris@10: A ./dft/ct-generic.c Chris@10: A ./dft/ct.h Chris@10: M ./dft/Makefile.am -5 +4 Chris@10: M ./dft/conf.c -3 +2 Chris@10: M ./dft/ct-directw.c +153 Chris@10: M ./dft/ct-directwbuf.c +213 Chris@10: M ./dft/ct-generic.c +304 Chris@10: M ./dft/ct.c -62 +24 Chris@10: M ./dft/ct.h +56 Chris@10: M ./dft/ctsq.c -3 +3 Chris@10: M ./dft/dft.h -39 +2 Chris@10: R ./dft/dftw-dft.c Chris@10: M ./dft/direct.c -2 +1 Chris@10: R ./dft/directw.c Chris@10: R ./dft/directwbuf.c Chris@10: M ./dft/generic.c -2 Chris@10: M ./dft/kdft-dif.c -4 +4 Chris@10: M ./dft/kdft-difsq.c -3 +3 Chris@10: M ./dft/kdft-dit.c -4 +4 Chris@10: M ./dft/plan.c -11 +1 Chris@10: R ./dft/problemw.c Chris@10: M ./dft/rader.c -2 Chris@10: M ./dft/solve.c -8 +1 Chris@10: Chris@10: Fri Jul 4 06:56:26 EDT 2003 athena Chris@10: * [project @ 2003-07-04 10:56:26 by athena] Chris@10: Increase TIME_MIN on intel only Chris@10: Chris@10: M ./kernel/cycle.h -1 +3 Chris@10: M ./kernel/timer.c -2 +2 Chris@10: Chris@10: Fri Jul 4 06:36:02 EDT 2003 athena Chris@10: * [project @ 2003-07-04 10:36:02 by athena] Chris@10: A little hack to get more consistent scheduling. Chris@10: Chris@10: M ./genfft/schedule.ml -7 +8 Chris@10: Chris@10: Thu Jul 3 16:47:42 EDT 2003 athena Chris@10: * [project @ 2003-07-03 20:47:42 by athena] Chris@10: New experimental scheduler (currently disabled). Chris@10: Chris@10: The old scheduler is ``optimal'' in the sense that it minimizes Chris@10: register pressure. The only way to reduce register pressure is to Chris@10: schedule dependent instructions as closely as possible, so as to Chris@10: minimize the life time of registers. This strategy maximizes the Chris@10: number of pipeline stalls, however. With enough registers and short Chris@10: enough pipelines, this tradeoff is fine. This is no longer the case Chris@10: for the devilish pipeline of the Pentium IV or (probably) the PowerPC Chris@10: 970. Chris@10: Chris@10: The new scheduler switches to a ``list scheduler'' for dags smaller Chris@10: than a specified size. The list scheduler executes a butterfly left Chris@10: to right one column at the time. This amounts to the best possible Chris@10: pipeline utilization, and the worst possible register pressure. Chris@10: Chris@10: The ``specified size'' defaults to 0, i.e., no change from fftw2 and Chris@10: fftw-3.0. It seems like a value of 7--10 produces the best results Chris@10: for Pentium IV (probably screwing the G3/G4 powerpcs and sparc, but I Chris@10: haven't tried.) As time goes by, we may want to increase this number Chris@10: to favor newer processors over older processors. Chris@10: Chris@10: M ./genfft/magic.ml -1 +3 Chris@10: M ./genfft/schedule.ml -9 +35 Chris@10: Chris@10: Wed Jun 25 17:43:59 EDT 2003 stevenj Chris@10: * [project @ 2003-06-25 21:43:59 by stevenj] Chris@10: remove non-portable use of tempfile; thanks to Nicolas Decoster for the patch Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: Chris@10: Wed Jun 25 17:14:03 EDT 2003 stevenj Chris@10: * [project @ 2003-06-25 21:14:03 by stevenj] Chris@10: increase stupid HP preprocessor limits Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Thu Jun 19 15:21:52 EDT 2003 athena Chris@10: * [project @ 2003-06-19 19:21:52 by athena] Chris@10: Distribute gen_mdct.ml Chris@10: Chris@10: M ./genfft/Makefile.am -8 +8 Chris@10: Chris@10: Wed Jun 11 06:55:21 EDT 2003 athena Chris@10: * [project @ 2003-06-11 10:55:21 by athena] Chris@10: Cleared int/ptrdiff_t confusions Chris@10: Chris@10: M ./rdft/buffered2.c -2 +3 Chris@10: M ./rdft/rdft2-radix2.c -5 +5 Chris@10: Chris@10: Tue Jun 10 22:15:42 EDT 2003 athena Chris@10: * [project @ 2003-06-11 02:15:41 by athena] Chris@10: Cleared int/ptrdiff_t confusion Chris@10: Chris@10: M ./dft/dftw-dft.c -1 +1 Chris@10: M ./dft/directwbuf.c -2 +2 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: M ./dft/vrank2-transpose.c -2 +2 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Sun Jun 8 09:52:57 EDT 2003 athena Chris@10: * [project @ 2003-06-08 13:52:57 by athena] Chris@10: Increased TIME_MIN. This seems to produce more reliable plans Chris@10: on Pentium IV. Chris@10: Chris@10: M ./kernel/timer.c -2 +2 Chris@10: Chris@10: Sat Jun 7 21:43:00 EDT 2003 athena Chris@10: * [project @ 2003-06-08 01:43:00 by athena] Chris@10: Removed relic -trivial-stores, which dates back to Franz's early Chris@10: experiments. Speed improved on SSE2, both with gcc and icc. Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: Chris@10: Thu Jun 5 22:29:52 EDT 2003 stevenj Chris@10: * [project @ 2003-06-06 02:29:52 by stevenj] Chris@10: fix direntry Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Thu Jun 5 13:41:34 EDT 2003 stevenj Chris@10: * [project @ 2003-06-05 17:41:34 by stevenj] Chris@10: added imdct Chris@10: Chris@10: M ./genfft/gen_mdct.ml -13 +66 Chris@10: Chris@10: Wed Jun 4 19:54:38 EDT 2003 athena Chris@10: * [project @ 2003-06-04 23:54:38 by athena] Chris@10: Collect pattern (a * b) +- (c * d) in generic-arith, because this Chris@10: operation can usually be computed with one rounding in fixed-point Chris@10: (and it possibly exposes a FMA instruction) Chris@10: Chris@10: M ./genfft/c.ml -5 +15 Chris@10: Chris@10: Wed Jun 4 15:11:29 EDT 2003 athena Chris@10: * [project @ 2003-06-04 19:11:29 by athena] Chris@10: Generic-arithmetic unparser Chris@10: Chris@10: M ./genfft/c.ml -8 +31 Chris@10: M ./genfft/magic.ml -1 +5 Chris@10: Chris@10: Sun Jun 1 09:05:30 EDT 2003 athena Chris@10: * [project @ 2003-06-01 13:05:30 by athena] Chris@10: Oops---randomized CSE was using the same random numbers Chris@10: over and over Chris@10: Chris@10: M ./genfft/oracle.ml -8 +6 Chris@10: M ./genfft-k7/oracle.ml -11 +11 Chris@10: Chris@10: Sun Jun 1 07:01:17 EDT 2003 athena Chris@10: * [project @ 2003-06-01 11:01:17 by athena] Chris@10: Paranoia. Chris@10: Chris@10: M ./genfft/c.ml -3 +3 Chris@10: Chris@10: Sun Jun 1 07:00:54 EDT 2003 athena Chris@10: * [project @ 2003-06-01 11:00:54 by athena] Chris@10: Use relative error instead of absolute error, to avoid problems Chris@10: when normalization factors are used. Chris@10: Chris@10: M ./genfft/oracle.ml -3 +5 Chris@10: Chris@10: Sat May 31 22:11:28 EDT 2003 stevenj Chris@10: * [project @ 2003-06-01 02:11:28 by stevenj] Chris@10: slight opt Chris@10: Chris@10: M ./reodft/reodft11e-radix2.c -3 +3 Chris@10: Chris@10: Sat May 31 22:10:45 EDT 2003 stevenj Chris@10: * [project @ 2003-06-01 02:10:45 by stevenj] Chris@10: slight optimization Chris@10: Chris@10: M ./reodft/reodft11e-radix2.c -3 +3 Chris@10: Chris@10: Sat May 31 20:43:31 EDT 2003 stevenj Chris@10: * [project @ 2003-06-01 00:43:31 by stevenj] Chris@10: *W is const Chris@10: Chris@10: M ./genfft/gen_mdct.ml -3 +3 Chris@10: Chris@10: Sat May 31 20:41:15 EDT 2003 stevenj Chris@10: * [project @ 2003-06-01 00:41:15 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./genfft/gen_mdct.ml -2 +4 Chris@10: Chris@10: Thu May 29 21:31:31 EDT 2003 stevenj Chris@10: * [project @ 2003-05-30 01:31:31 by stevenj] Chris@10: added experimental MDCT Chris@10: Chris@10: A ./genfft/gen_mdct.ml Chris@10: M ./genfft/Makefile.am -1 +7 Chris@10: M ./genfft/gen_mdct.ml +208 Chris@10: Chris@10: Wed May 28 22:01:37 EDT 2003 stevenj Chris@10: * [project @ 2003-05-29 02:01:37 by stevenj] Chris@10: altivec (fma) needs simd codlist.c too Chris@10: Chris@10: M ./mkdist.sh -1 +1 Chris@10: Chris@10: Wed May 28 22:00:49 EDT 2003 stevenj Chris@10: * [project @ 2003-05-29 02:00:49 by stevenj] Chris@10: make sure we include SIMD codlist.c for non-Unix folks Chris@10: Chris@10: M ./mkdist.sh -1 +1 Chris@10: Chris@10: Tue May 27 20:31:25 EDT 2003 stevenj Chris@10: * [project @ 2003-05-28 00:31:25 by stevenj] Chris@10: noted howmany_rank == 0 is a single transform Chris@10: Chris@10: M ./doc/fftw3.texi -1 +3 Chris@10: Chris@10: Tue May 27 20:02:31 EDT 2003 stevenj Chris@10: * [project @ 2003-05-28 00:02:31 by stevenj] Chris@10: further stride clarification Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Mon May 26 10:21:22 EDT 2003 athena Chris@10: * [project @ 2003-05-26 14:21:22 by athena] Chris@10: Removed transposed dftw problems. Chris@10: Chris@10: I now consider transposed dftw a Bad Idea, since it does not Chris@10: apply to the case that it was originally meant for (speed up four-step) Chris@10: and it complicates the implementation of the other thing I want to try Chris@10: (dftw m-slices). Chris@10: Chris@10: M ./dft/Makefile.am -3 +3 Chris@10: M ./dft/conf.c -2 +1 Chris@10: M ./dft/ct.c -5 +3 Chris@10: M ./dft/ctsq.c -74 +57 Chris@10: M ./dft/dft.h -9 +5 Chris@10: M ./dft/dftw-dft.c -83 Chris@10: M ./dft/directw.c -5 +1 Chris@10: M ./dft/directwbuf.c -5 +1 Chris@10: R ./dft/directwsq.c Chris@10: M ./dft/kdft-difsq.c -2 +2 Chris@10: M ./dft/problemw.c -13 +9 Chris@10: Chris@10: Mon May 26 07:22:59 EDT 2003 athena Chris@10: * [project @ 2003-05-26 11:22:59 by athena] Chris@10: Obsolete comment Chris@10: Chris@10: M ./dft/buffered.c -6 +1 Chris@10: Chris@10: Sat May 24 15:00:53 EDT 2003 athena Chris@10: * [project @ 2003-05-24 19:00:53 by athena] Chris@10: comment Chris@10: Chris@10: M ./dft/ct.c -2 +2 Chris@10: Chris@10: Sat May 24 07:20:35 EDT 2003 athena Chris@10: * [project @ 2003-05-24 11:20:35 by athena] Chris@10: Oops---wrong test NO_UGLYP instead of !NO_UGLYP Chris@10: Chris@10: M ./dft/dftw-dft.c -3 +3 Chris@10: Chris@10: Sat May 24 07:05:34 EDT 2003 athena Chris@10: * [project @ 2003-05-24 11:05:34 by athena] Chris@10: Implemented radix r, where n=r^2 * p Chris@10: Chris@10: M ./dft/ct.c -19 +53 Chris@10: Chris@10: Wed May 21 01:54:32 EDT 2003 stevenj Chris@10: * [project @ 2003-05-21 05:54:32 by stevenj] Chris@10: xlc seems to properly use fma as well Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +2 Chris@10: Chris@10: Tue May 20 23:07:43 EDT 2003 stevenj Chris@10: * [project @ 2003-05-21 03:07:43 by stevenj] Chris@10: print warning if there is no cycle counter Chris@10: Chris@10: M ./configure.ac +18 Chris@10: M ./doc/fftw3.texi -2 +3 Chris@10: Chris@10: Tue May 20 17:32:04 EDT 2003 stevenj Chris@10: * [project @ 2003-05-20 21:32:04 by stevenj] Chris@10: updated Funda reference Chris@10: Chris@10: M ./libbench2/verify-lib.c -1 +4 Chris@10: M ./libbench2/verify-r2r.c +3 Chris@10: Chris@10: Mon May 19 20:12:36 EDT 2003 athena Chris@10: * [project @ 2003-05-20 00:12:36 by athena] Chris@10: const Chris@10: Chris@10: M ./dft/ct.c -2 +2 Chris@10: Chris@10: Mon May 19 15:41:09 EDT 2003 athena Chris@10: * [project @ 2003-05-19 19:41:09 by athena] Chris@10: Implemented generic dif square transposed (q-style) solver. Chris@10: Chris@10: M ./dft/dftw-dft.c -8 +88 Chris@10: M ./dft/directwsq.c -2 +2 Chris@10: Chris@10: Mon May 19 07:00:36 EDT 2003 athena Chris@10: * [project @ 2003-05-19 11:00:36 by athena] Chris@10: applicable() is now a property of the solver (in anticipation of Chris@10: transposed solvers) Chris@10: Chris@10: M ./dft/dftw-dft.c -35 +35 Chris@10: Chris@10: Mon May 19 06:33:40 EDT 2003 athena Chris@10: * [project @ 2003-05-19 10:33:40 by athena] Chris@10: Slight cleanup Chris@10: Chris@10: M ./dft/dftw-dft.c -7 +7 Chris@10: Chris@10: Sun May 18 13:05:51 EDT 2003 athena Chris@10: * [project @ 2003-05-18 17:05:51 by athena] Chris@10: Nothing, really Chris@10: Chris@10: M ./dft/bluestein.c -9 +3 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/primes.c -1 +9 Chris@10: Chris@10: Sun May 18 09:05:20 EDT 2003 athena Chris@10: * [project @ 2003-05-18 13:05:20 by athena] Chris@10: Moved vector loop inside bytwiddle(), in anticipation of Chris@10: a q-style dftw-dit transposed solver. Chris@10: Chris@10: M ./dft/dftw-dft.c -38 +41 Chris@10: Chris@10: Sun May 18 08:52:02 EDT 2003 athena Chris@10: * [project @ 2003-05-18 12:52:02 by athena] Chris@10: Fixed flops count Chris@10: Chris@10: M ./dft/dftw-dft.c -1 +1 Chris@10: Chris@10: Sun May 18 08:47:20 EDT 2003 athena Chris@10: * [project @ 2003-05-18 12:47:20 by athena] Chris@10: style Chris@10: Chris@10: M ./dft/dftw-dft.c -5 +3 Chris@10: Chris@10: Sun May 18 07:16:34 EDT 2003 athena Chris@10: * [project @ 2003-05-18 11:16:34 by athena] Chris@10: Faster inner loop. Chris@10: Chris@10: M ./dft/dftw-dft.c -13 +25 Chris@10: Chris@10: Sat May 17 08:02:38 EDT 2003 athena Chris@10: * [project @ 2003-05-17 12:02:38 by athena] Chris@10: Print vector length Chris@10: Chris@10: M ./dft/dftw-dft.c -2 +2 Chris@10: Chris@10: Sat May 17 07:55:33 EDT 2003 athena Chris@10: * [project @ 2003-05-17 11:55:33 by athena] Chris@10: Oops Chris@10: Chris@10: M ./dft/dftw-dft.c -7 +7 Chris@10: Chris@10: Sat May 17 07:50:35 EDT 2003 athena Chris@10: * [project @ 2003-05-17 11:50:35 by athena] Chris@10: Allow vl > 1 Chris@10: Chris@10: M ./dft/dftw-dft.c -7 +16 Chris@10: Chris@10: Sat May 17 07:01:42 EDT 2003 athena Chris@10: * [project @ 2003-05-17 11:01:42 by athena] Chris@10: Radix can be derived from problem---no need to pre-specify it. Chris@10: Chris@10: M ./dft/ctsq.c -21 +12 Chris@10: Chris@10: Fri May 16 22:50:50 EDT 2003 stevenj Chris@10: * [project @ 2003-05-17 02:50:50 by stevenj] Chris@10: fixed comment Chris@10: Chris@10: M ./kernel/transpose.c -1 +1 Chris@10: Chris@10: Fri May 16 22:48:20 EDT 2003 stevenj Chris@10: * [project @ 2003-05-17 02:48:20 by stevenj] Chris@10: whoops, gcd should be static Chris@10: Chris@10: M ./kernel/transpose.c -1 +1 Chris@10: Chris@10: Fri May 16 22:40:32 EDT 2003 stevenj Chris@10: * [project @ 2003-05-17 02:40:32 by stevenj] Chris@10: more unrolling Chris@10: Chris@10: M ./kernel/transpose.c -7 +31 Chris@10: Chris@10: Fri May 16 20:03:48 EDT 2003 athena Chris@10: * [project @ 2003-05-17 00:03:48 by athena] Chris@10: Hack to avoid infinite recursion. Chris@10: Chris@10: M ./dft/bluestein.c +3 Chris@10: Chris@10: Fri May 16 19:52:43 EDT 2003 stevenj Chris@10: * [project @ 2003-05-16 23:52:43 by stevenj] Chris@10: consistency Chris@10: Chris@10: M ./dft/codelet-dft.h -4 +4 Chris@10: Chris@10: Fri May 16 19:45:15 EDT 2003 athena Chris@10: * [project @ 2003-05-16 23:45:15 by athena] Chris@10: Wrong comment. Chris@10: Chris@10: M ./dft/bluestein.c -1 +1 Chris@10: Chris@10: Fri May 16 19:45:03 EDT 2003 athena Chris@10: * [project @ 2003-05-16 23:45:03 by athena] Chris@10: Style. Chris@10: Chris@10: M ./dft/bluestein.c -3 +3 Chris@10: Chris@10: Fri May 16 18:35:27 EDT 2003 stevenj Chris@10: * [project @ 2003-05-16 22:35:27 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +2 Chris@10: Chris@10: Fri May 16 18:33:45 EDT 2003 stevenj Chris@10: * [project @ 2003-05-16 22:33:45 by stevenj] Chris@10: added allzero FAQ Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +7 Chris@10: Chris@10: Fri May 16 18:22:45 EDT 2003 stevenj Chris@10: * [project @ 2003-05-16 22:22:45 by stevenj] Chris@10: simplification: instead of cldb, just use cldf with inputs/output values swapped Chris@10: Chris@10: M ./dft/bluestein.c -22 +11 Chris@10: Chris@10: Fri May 16 15:47:17 EDT 2003 athena Chris@10: * [project @ 2003-05-16 19:47:17 by athena] Chris@10: Allow more general transform sizes. Chris@10: Chris@10: M ./dft/bluestein.c -15 +21 Chris@10: Chris@10: Fri May 16 14:22:37 EDT 2003 stevenj Chris@10: * [project @ 2003-05-16 18:22:37 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: Chris@10: Fri May 16 14:22:05 EDT 2003 stevenj Chris@10: * [project @ 2003-05-16 18:22:05 by stevenj] Chris@10: MS has __int64 type, not long long (grr) Chris@10: Chris@10: M ./kernel/ifftw.h -1 +4 Chris@10: Chris@10: Fri May 16 13:34:16 EDT 2003 athena Chris@10: * [project @ 2003-05-16 17:34:16 by athena] Chris@10: Fixed printout Chris@10: Chris@10: M ./dft/ct.c -2 +4 Chris@10: Chris@10: Fri May 16 13:23:00 EDT 2003 athena Chris@10: * [project @ 2003-05-16 17:23:00 by athena] Chris@10: Fixed flop count Chris@10: Chris@10: M ./dft/bluestein.c +5 Chris@10: Chris@10: Fri May 16 13:02:06 EDT 2003 athena Chris@10: * [project @ 2003-05-16 17:02:06 by athena] Chris@10: New bluestein solver Chris@10: Chris@10: A ./dft/bluestein.c Chris@10: M ./dft/Makefile.am -5 +6 Chris@10: M ./dft/bluestein.c +255 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: Chris@10: Fri May 16 09:51:05 EDT 2003 athena Chris@10: * [project @ 2003-05-16 13:51:05 by athena] Chris@10: Implemented generic radix. Chris@10: Chris@10: M ./dft/ct.c -14 +35 Chris@10: Chris@10: Fri May 16 08:19:38 EDT 2003 athena Chris@10: * [project @ 2003-05-16 12:19:38 by athena] Chris@10: Removed conditional branch from inner loop in generic.c Chris@10: Chris@10: M ./dft/generic.c -18 +20 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/twiddle.c -1 +18 Chris@10: Chris@10: Fri May 16 07:48:28 EDT 2003 athena Chris@10: * [project @ 2003-05-16 11:48:28 by athena] Chris@10: Simplified indexing Chris@10: Chris@10: M ./dft/generic.c -29 +20 Chris@10: Chris@10: Fri May 16 06:53:56 EDT 2003 athena Chris@10: * [project @ 2003-05-16 10:53:56 by athena] Chris@10: Better still. Chris@10: Chris@10: M ./dft/generic.c -26 +37 Chris@10: Chris@10: Fri May 16 06:24:31 EDT 2003 athena Chris@10: * [project @ 2003-05-16 10:24:31 by athena] Chris@10: Further improvement of generic solver Chris@10: Chris@10: M ./dft/generic.c -8 +8 Chris@10: Chris@10: Fri May 16 05:57:07 EDT 2003 athena Chris@10: * [project @ 2003-05-16 09:57:07 by athena] Chris@10: Cleanup Chris@10: Chris@10: M ./dft/rader.c -54 +51 Chris@10: Chris@10: Fri May 16 05:42:57 EDT 2003 athena Chris@10: * [project @ 2003-05-16 09:42:57 by athena] Chris@10: Cleanup Chris@10: Chris@10: M ./dft/dftw-dft.c -30 +25 Chris@10: Chris@10: Fri May 16 05:31:40 EDT 2003 athena Chris@10: * [project @ 2003-05-16 09:31:40 by athena] Chris@10: Generic now only works for odd sized. Added check. Chris@10: Chris@10: M ./dft/generic.c -1 +2 Chris@10: Chris@10: Thu May 15 21:53:25 EDT 2003 athena Chris@10: * [project @ 2003-05-16 01:53:25 by athena] Chris@10: Increased GENERIC_MIN_BAD because of new algorithm. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Thu May 15 21:40:27 EDT 2003 athena Chris@10: * [project @ 2003-05-16 01:40:27 by athena] Chris@10: Much, much better. Chris@10: Chris@10: M ./dft/generic.c -9 +30 Chris@10: Chris@10: Thu May 15 21:25:00 EDT 2003 athena Chris@10: * [project @ 2003-05-16 01:25:00 by athena] Chris@10: Still trying to understand why rdft-generic-dit is faster Chris@10: then dft-generic... Chris@10: Chris@10: M ./dft/generic.c -9 +8 Chris@10: Chris@10: Thu May 15 21:04:33 EDT 2003 athena Chris@10: * [project @ 2003-05-16 01:04:33 by athena] Chris@10: Nothing, really Chris@10: Chris@10: M ./dft/generic.c -2 +2 Chris@10: Chris@10: Thu May 15 20:59:45 EDT 2003 athena Chris@10: * [project @ 2003-05-16 00:59:45 by athena] Chris@10: Never be clever for the sake of being clever. Chris@10: Chris@10: M ./dft/generic.c -1 +1 Chris@10: Chris@10: Thu May 15 20:58:06 EDT 2003 athena Chris@10: * [project @ 2003-05-16 00:58:06 by athena] Chris@10: Simplified. generic-dit is gone. The solver is now out-of-place Chris@10: only---buffering is done by the buffered solver. Chris@10: Chris@10: M ./dft/generic.c -94 +45 Chris@10: Chris@10: Thu May 15 19:18:18 EDT 2003 athena Chris@10: * [project @ 2003-05-15 23:18:18 by athena] Chris@10: rader-dit is gone. Chris@10: Chris@10: M ./dft/rader.c -200 Chris@10: Chris@10: Thu May 15 19:13:03 EDT 2003 athena Chris@10: * [project @ 2003-05-15 23:13:03 by athena] Chris@10: Cast Chris@10: Chris@10: M ./dft/plan.c -2 +2 Chris@10: Chris@10: Thu May 15 19:09:07 EDT 2003 athena Chris@10: * [project @ 2003-05-15 23:09:07 by athena] Chris@10: Introduced twiddle problem ``dftw''. Changed most other things Chris@10: to deal with this change. Chris@10: Chris@10: A ./dft/ctsq.c Chris@10: A ./dft/dftw-dft.c Chris@10: A ./dft/directw.c Chris@10: A ./dft/directwbuf.c Chris@10: A ./dft/directwsq.c Chris@10: A ./dft/problemw.c Chris@10: M ./configure.ac -1 +1 Chris@10: M ./dft/Makefile.am -5 +5 Chris@10: M ./dft/buffered.c -2 +1 Chris@10: M ./dft/codelet-dft.h -13 +9 Chris@10: M ./dft/conf.c -1 +4 Chris@10: R ./dft/ct-dif.c Chris@10: R ./dft/ct-dit.c Chris@10: R ./dft/ct-ditbuf.c Chris@10: R ./dft/ct-ditf.c Chris@10: M ./dft/ct.c -91 +157 Chris@10: R ./dft/ct.h Chris@10: M ./dft/ctsq.c +213 Chris@10: M ./dft/dft.h -7 +45 Chris@10: M ./dft/dftw-dft.c +297 Chris@10: M ./dft/directw.c +172 Chris@10: M ./dft/directwbuf.c +230 Chris@10: M ./dft/directwsq.c +148 Chris@10: M ./dft/kdft-dif.c -4 +5 Chris@10: M ./dft/kdft-difsq.c -3 +3 Chris@10: M ./dft/kdft-dit.c -5 +5 Chris@10: M ./dft/plan.c -1 +11 Chris@10: M ./dft/problemw.c +103 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./dft/solve.c -1 +8 Chris@10: Chris@10: Sat Aug 12 23:17:14 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-0-1-branch Chris@10: Chris@10: Chris@10: Thu May 15 18:47:18 EDT 2003 stevenj Chris@10: * [project @ 2003-05-15 22:47:18 by stevenj] Chris@10: whoops, X(safe_mulmod) not fftw_safe_mulmod Chris@10: Chris@10: M ./kernel/primes.c -3 +3 Chris@10: Chris@10: Thu May 15 16:53:16 EDT 2003 stevenj Chris@10: * [project @ 2003-05-15 20:53:16 by stevenj] Chris@10: add VC++ versions of asm Chris@10: Chris@10: M ./simd/sse.c -1 +17 Chris@10: M ./simd/sse2.c -1 +17 Chris@10: Chris@10: Thu May 15 15:03:06 EDT 2003 stevenj Chris@10: * [project @ 2003-05-15 19:03:05 by stevenj] Chris@10: VC++ reportedly supports the intel intrinsics, but requires __inline instead of __inline__ Chris@10: Chris@10: M ./simd/simd-sse.h -1 +4 Chris@10: M ./simd/simd-sse2.h -1 +5 Chris@10: Chris@10: Thu May 15 14:32:06 EDT 2003 stevenj Chris@10: * [project @ 2003-05-15 18:32:06 by stevenj] Chris@10: precompute array indices with VC++ Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Wed May 14 21:57:39 EDT 2003 stevenj Chris@10: * [project @ 2003-05-15 01:57:39 by stevenj] Chris@10: added doc note Chris@10: Chris@10: M ./acx_pthread.m4 -1 +5 Chris@10: Chris@10: Wed May 14 19:45:54 EDT 2003 stevenj Chris@10: * [project @ 2003-05-14 23:45:54 by stevenj] Chris@10: autodetect windows Chris@10: Chris@10: M ./threads/threads.c -3 +4 Chris@10: Chris@10: Wed May 14 15:08:49 EDT 2003 stevenj Chris@10: * [project @ 2003-05-14 19:08:49 by stevenj] Chris@10: don't bother with #ifdef HAVE_CONFIG_H, since non-Unix users always forget to define it Chris@10: Chris@10: M ./libbench2/getopt.c -2 Chris@10: Chris@10: Tue May 13 16:58:07 EDT 2003 stevenj Chris@10: * [project @ 2003-05-13 20:58:07 by stevenj] Chris@10: VC++ uses __inline Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Tue May 13 14:51:26 EDT 2003 stevenj Chris@10: * [project @ 2003-05-13 18:51:26 by stevenj] Chris@10: added leak question Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +8 Chris@10: Chris@10: Mon May 12 18:26:51 EDT 2003 stevenj Chris@10: * [project @ 2003-05-12 22:26:51 by stevenj] Chris@10: LARGE_INTEGER needs windows.h (supposedly, there is some problem converting _itnt64 to double...damn MS and their nonstandard types) Chris@10: Chris@10: M ./kernel/cycle.h -1 +2 Chris@10: Chris@10: Mon May 12 18:22:16 EDT 2003 stevenj Chris@10: * [project @ 2003-05-12 22:22:16 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./libbench2/timer.c -2 +2 Chris@10: Chris@10: Mon May 12 17:16:19 EDT 2003 stevenj Chris@10: * [project @ 2003-05-12 21:16:19 by stevenj] Chris@10: added 256x256 to canonical list Chris@10: Chris@10: M ./tools/fftw-wisdom.c -1 +1 Chris@10: Chris@10: Mon May 12 07:02:06 EDT 2003 athena Chris@10: * [project @ 2003-05-12 11:02:06 by athena] Chris@10: Oops... Chris@10: Chris@10: M ./kernel/transpose.c +1 Chris@10: Chris@10: Sun May 11 11:04:46 EDT 2003 athena Chris@10: * [project @ 2003-05-11 15:04:46 by athena] Chris@10: Unrolled loops, changed cutoff Chris@10: Chris@10: M ./kernel/transpose.c -23 +80 Chris@10: Chris@10: Sun May 11 10:20:04 EDT 2003 athena Chris@10: * [project @ 2003-05-11 14:20:04 by athena] Chris@10: Do not multiply strides by 2 twice. Chris@10: Chris@10: M ./tests/bench.c -16 +15 Chris@10: Chris@10: Wed May 7 21:09:43 EDT 2003 stevenj Chris@10: * [project @ 2003-05-08 01:09:43 by stevenj] Chris@10: added 'make smallcheck' Chris@10: Chris@10: M ./tests/Makefile.am +13 Chris@10: Chris@10: Wed May 7 20:46:10 EDT 2003 stevenj Chris@10: * [project @ 2003-05-08 00:46:10 by stevenj] Chris@10: --without-cycle-counter becomes --with-slow-timer, updated docs Chris@10: Chris@10: M ./configure.ac -3 +3 Chris@10: M ./doc/fftw3.texi -15 +15 Chris@10: M ./kernel/timer.c -3 +3 Chris@10: Chris@10: Wed May 7 18:05:29 EDT 2003 stevenj Chris@10: * [project @ 2003-05-07 22:05:29 by stevenj] Chris@10: remove duplicate -openmp check; Sun requires -xopenmp Chris@10: Chris@10: M ./configure.ac -5 +5 Chris@10: Chris@10: Wed May 7 17:59:23 EDT 2003 stevenj Chris@10: * [project @ 2003-05-07 21:59:23 by stevenj] Chris@10: fixed compilation under Sun C++ Chris@10: Chris@10: M ./dft/ct-ditbuf.c -3 +3 Chris@10: M ./rdft/hc2hc-buf.c -3 +3 Chris@10: Chris@10: Wed May 7 14:24:46 EDT 2003 athena Chris@10: * [project @ 2003-05-07 18:24:46 by athena] Chris@10: Use estimator if cycle counter is unavailable, regardless Chris@10: of the FFTW_MEASURE/ESTIMATE setting. Chris@10: Chris@10: M ./kernel/planner.c -3 +13 Chris@10: M ./kernel/timer.c -57 +67 Chris@10: Chris@10: Tue May 6 23:15:34 EDT 2003 stevenj Chris@10: * [project @ 2003-05-07 03:15:34 by stevenj] Chris@10: _WIN32 (not __WIN32__) is always defined Chris@10: Chris@10: M ./kernel/cycle.h -3 +2 Chris@10: Chris@10: Tue May 6 23:11:52 EDT 2003 stevenj Chris@10: * [project @ 2003-05-07 03:11:52 by stevenj] Chris@10: minor cleanup Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Tue May 6 22:50:07 EDT 2003 stevenj Chris@10: * [project @ 2003-05-07 02:50:07 by stevenj] Chris@10: tentative VC++ stuff, some consolidation Chris@10: Chris@10: M ./kernel/cycle.h -51 +56 Chris@10: Chris@10: Tue May 6 12:17:56 EDT 2003 stevenj Chris@10: * [project @ 2003-05-06 16:17:56 by stevenj] Chris@10: made cycle.h more self-contained Chris@10: Chris@10: M ./kernel/cycle.h -3 +13 Chris@10: M ./kernel/timer.c -15 +18 Chris@10: Chris@10: Tue May 6 08:30:39 EDT 2003 athena Chris@10: * [project @ 2003-05-06 12:30:39 by athena] Chris@10: Use ``%'' flag to denote commutative operations. Chris@10: Chris@10: M ./simd/simd-3dnow.h -3 +3 Chris@10: M ./simd/simd-sse.h -3 +3 Chris@10: M ./simd/simd-sse2.h -3 +3 Chris@10: Chris@10: Mon May 5 20:42:30 EDT 2003 stevenj Chris@10: * [project @ 2003-05-06 00:42:30 by stevenj] Chris@10: MIT license, brief documentation Chris@10: Chris@10: M ./kernel/cycle.h -12 +41 Chris@10: Chris@10: Mon May 5 20:31:16 EDT 2003 stevenj Chris@10: * [project @ 2003-05-06 00:31:16 by stevenj] Chris@10: whoops, forgot f77_wisdom.f Chris@10: Chris@10: M ./doc/Makefile.am -1 +1 Chris@10: Chris@10: Sun May 4 19:37:09 EDT 2003 athena Chris@10: * [project @ 2003-05-04 23:37:09 by athena] Chris@10: Improved speed of accuracy test. Chris@10: Chris@10: M ./dft/problem.c -6 +3 Chris@10: M ./libbench/mp.c -16 +26 Chris@10: M ./libbench2/bench.h -1 +2 Chris@10: M ./libbench2/mp.c -46 +92 Chris@10: M ./libbench2/verify-lib.c -1 +3 Chris@10: M ./rdft/problem2.c -3 +2 Chris@10: Chris@10: Tue Apr 29 11:45:34 EDT 2003 athena Chris@10: * [project @ 2003-04-29 15:45:34 by athena] Chris@10: s390 cycle counter Chris@10: Chris@10: M ./kernel/cycle.h -1 +21 Chris@10: Chris@10: Sat Apr 26 12:26:15 EDT 2003 stevenj Chris@10: * [project @ 2003-04-26 16:26:15 by stevenj] Chris@10: forgot r2r directory Chris@10: Chris@10: M ./doc/fftw3.texi -10 +10 Chris@10: Chris@10: Fri Apr 25 20:52:23 EDT 2003 stevenj Chris@10: * [project @ 2003-04-26 00:52:23 by stevenj] Chris@10: delete unused files, since they don't compile any more Chris@10: Chris@10: M ./rdft/Makefile.am -2 Chris@10: R ./rdft/vrank2-transpose.c Chris@10: R ./rdft/vrank3-transpose.c Chris@10: Chris@10: Thu Apr 24 06:37:41 EDT 2003 athena Chris@10: * [project @ 2003-04-24 10:37:41 by athena] Chris@10: Better gcc code generation Chris@10: Chris@10: M ./simd/simd-sse2.h -3 +5 Chris@10: Chris@10: Wed Apr 23 15:30:50 EDT 2003 stevenj Chris@10: * [project @ 2003-04-23 19:30:50 by stevenj] Chris@10: ccc is the Compaq C compiler on Linux/alpha Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Wed Apr 23 00:06:03 EDT 2003 stevenj Chris@10: * [project @ 2003-04-23 04:06:03 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Sat Aug 12 23:16:25 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-0 Chris@10: Chris@10: Chris@10: Sat Apr 19 09:18:25 EDT 2003 athena Chris@10: * [project @ 2003-04-19 13:18:25 by athena] Chris@10: ia64 cycle counter with intel compiler. Chris@10: Chris@10: M ./kernel/cycle.h -1 +19 Chris@10: Chris@10: Fri Apr 18 18:27:30 EDT 2003 athena Chris@10: * [project @ 2003-04-18 22:27:30 by athena] Chris@10: More gcc bugs. Sigh. Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -4 +12 Chris@10: Chris@10: Fri Apr 18 18:01:49 EDT 2003 athena Chris@10: * [project @ 2003-04-18 22:01:49 by athena] Chris@10: touch ChangeLog to observe GNU standards Chris@10: Chris@10: M ./bootstrap.sh +1 Chris@10: Chris@10: Fri Apr 18 18:01:12 EDT 2003 athena Chris@10: * [project @ 2003-04-18 22:01:12 by athena] Chris@10: We now build ChangeLog automatically at distribution time Chris@10: Chris@10: R ./ChangeLog Chris@10: Chris@10: Fri Apr 18 18:00:17 EDT 2003 athena Chris@10: * [project @ 2003-04-18 22:00:17 by athena] Chris@10: Automatic ChangeLog hackery Chris@10: Chris@10: M ./mkdist.sh +7 Chris@10: Chris@10: Fri Apr 18 13:25:26 EDT 2003 stevenj Chris@10: * [project @ 2003-04-18 17:25:26 by stevenj] Chris@10: plural Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Fri Apr 18 13:25:01 EDT 2003 stevenj Chris@10: * [project @ 2003-04-18 17:25:01 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./NEWS +12 Chris@10: Chris@10: Fri Apr 18 12:59:41 EDT 2003 athena Chris@10: * [project @ 2003-04-18 16:59:41 by athena] Chris@10: Updated Chris@10: Chris@10: M ./ChangeLog -6789 +3415 Chris@10: Chris@10: Fri Apr 18 11:48:39 EDT 2003 stevenj Chris@10: * [project @ 2003-04-18 15:48:39 by stevenj] Chris@10: a -> an Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Fri Apr 18 11:47:56 EDT 2003 stevenj Chris@10: * [project @ 2003-04-18 15:47:56 by stevenj] Chris@10: hyphen Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Fri Apr 18 11:47:28 EDT 2003 stevenj Chris@10: * [project @ 2003-04-18 15:47:28 by stevenj] Chris@10: comma Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Fri Apr 18 11:46:59 EDT 2003 stevenj Chris@10: * [project @ 2003-04-18 15:46:59 by stevenj] Chris@10: minor Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -4 +4 Chris@10: Chris@10: Fri Apr 18 10:37:31 EDT 2003 athena Chris@10: * [project @ 2003-04-18 14:37:31 by athena] Chris@10: Updated Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -9 +36 Chris@10: Chris@10: Fri Apr 18 10:14:59 EDT 2003 athena Chris@10: * [project @ 2003-04-18 14:14:59 by athena] Chris@10: New script that builds the distributions Chris@10: Chris@10: A ./mkdist.sh Chris@10: Chris@10: Fri Apr 18 08:51:07 EDT 2003 athena Chris@10: * [project @ 2003-04-18 12:51:07 by athena] Chris@10: Oops again Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -4 +4 Chris@10: Chris@10: Fri Apr 18 08:39:05 EDT 2003 athena Chris@10: * [project @ 2003-04-18 12:39:05 by athena] Chris@10: Oops, forgot -sign 1 Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -4 +4 Chris@10: Chris@10: Fri Apr 18 08:28:25 EDT 2003 athena Chris@10: * [project @ 2003-04-18 12:28:25 by athena] Chris@10: Reorganization of simd codelets Chris@10: Chris@10: M ./configure.ac -4 Chris@10: M ./dft/simd/codelets/Makefile.am -136 +22 Chris@10: M ./dft/simd/n1b.c -3 +3 Chris@10: M ./dft/simd/n1b.h -3 Chris@10: M ./dft/simd/n1f.c -3 +3 Chris@10: M ./dft/simd/n1f.h -3 Chris@10: M ./dft/simd/n2b.c -3 +3 Chris@10: M ./dft/simd/n2b.h +3 Chris@10: M ./dft/simd/n2f.c -3 +3 Chris@10: M ./dft/simd/n2f.h +3 Chris@10: Chris@10: Thu Apr 17 21:21:45 EDT 2003 athena Chris@10: * [project @ 2003-04-18 01:21:45 by athena] Chris@10: k7 assembly was not updated after conversion of opcnt from Chris@10: int to double Chris@10: Chris@10: M ./genfft-k7/gen_notw.ml -5 +5 Chris@10: M ./genfft-k7/gen_twiddle.ml -5 +6 Chris@10: Chris@10: Thu Apr 17 19:15:53 EDT 2003 athena Chris@10: * [project @ 2003-04-17 23:15:53 by athena] Chris@10: Capital `X' looks bad in all-lowercase plans Chris@10: Chris@10: M ./dft/vrank2-transpose.c -2 +2 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Thu Apr 17 18:53:29 EDT 2003 athena Chris@10: * [project @ 2003-04-17 22:53:29 by athena] Chris@10: Removed redundant inline/noinline codelets Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -3 +2 Chris@10: M ./dft/simd/codelets/Makefile.am -27 +12 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -8 +4 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -8 +4 Chris@10: Chris@10: Thu Apr 17 15:25:50 EDT 2003 athena Chris@10: * [project @ 2003-04-17 19:25:50 by athena] Chris@10: New noinline Chris@10: Noinline real codelets Chris@10: Chris@10: A ./genfft/gen_hc2r_noinline.ml Chris@10: A ./genfft/gen_r2hc_noinline.ml Chris@10: M ./genfft/Makefile.am -10 +24 Chris@10: M ./genfft/gen_hc2hc.ml -5 +5 Chris@10: M ./genfft/gen_hc2r.ml -48 +31 Chris@10: M ./genfft/gen_hc2r_noinline.ml +185 Chris@10: M ./genfft/gen_r2hc.ml -49 +31 Chris@10: M ./genfft/gen_r2hc_noinline.ml +181 Chris@10: M ./genfft/gen_r2r.ml -3 +3 Chris@10: M ./kernel/ifftw.h -4 +1 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -1 +15 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +15 Chris@10: M ./support/Makefile.codelets -1 +3 Chris@10: Chris@10: Thu Apr 17 15:23:03 EDT 2003 stevenj Chris@10: * [project @ 2003-04-17 19:23:03 by stevenj] Chris@10: more ideas Chris@10: Chris@10: M ./TODO +6 Chris@10: Chris@10: Thu Apr 17 13:18:45 EDT 2003 athena Chris@10: * [project @ 2003-04-17 17:18:45 by athena] Chris@10: Removed duplicate rules. Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -6 Chris@10: Chris@10: Thu Apr 17 10:51:09 EDT 2003 athena Chris@10: * [project @ 2003-04-17 14:51:09 by athena] Chris@10: acx_pthread.m4 was not distributed Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: Chris@10: Thu Apr 17 07:21:17 EDT 2003 athena Chris@10: * [project @ 2003-04-17 11:21:17 by athena] Chris@10: Oops Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Thu Apr 17 07:07:19 EDT 2003 athena Chris@10: * [project @ 2003-04-17 11:07:19 by athena] Chris@10: Both inlined and non-inlined notw codelets. Chris@10: Chris@10: A ./genfft/gen_notw_noinline.ml Chris@10: M ./dft/codelets/standard/Makefile.am -1 +10 Chris@10: M ./genfft/Makefile.am -11 +17 Chris@10: M ./genfft/gen_notw.ml -51 +37 Chris@10: M ./genfft/gen_notw_noinline.ml +164 Chris@10: M ./support/Makefile.codelets +1 Chris@10: Chris@10: Thu Apr 17 06:44:21 EDT 2003 athena Chris@10: * [project @ 2003-04-17 10:44:21 by athena] Chris@10: Initial experiment with both inlined and non-inlined simd codelets. Chris@10: Both are included for now. Chris@10: Chris@10: A ./genfft/gen_notw_noinline_c.ml Chris@10: M ./dft/simd/codelets/Makefile.am -3 +68 Chris@10: M ./genfft/Makefile.am -9 +17 Chris@10: M ./genfft/gen_notw_noinline_c.ml +160 Chris@10: M ./support/Makefile.codelets -1 +2 Chris@10: Chris@10: Thu Apr 17 05:57:36 EDT 2003 athena Chris@10: * [project @ 2003-04-17 09:57:36 by athena] Chris@10: --enable-fma to build FMA distribution Chris@10: Chris@10: M ./configure.ac +7 Chris@10: M ./support/Makefile.codelets -1 +7 Chris@10: Chris@10: Wed Apr 16 17:21:53 EDT 2003 athena Chris@10: * [project @ 2003-04-16 21:21:53 by athena] Chris@10: Inline SIMD nontwiddle codelets Chris@10: Chris@10: M ./genfft/gen_notw_c.ml -47 +41 Chris@10: Chris@10: Wed Apr 16 16:18:29 EDT 2003 athena Chris@10: * [project @ 2003-04-16 20:18:29 by athena] Chris@10: Pathetic attempt at saving a couple of registers... Chris@10: Chris@10: M ./simd/simd-sse.h -6 +9 Chris@10: M ./simd/simd-sse2.h -6 +9 Chris@10: Chris@10: Wed Apr 16 15:51:27 EDT 2003 athena Chris@10: * [project @ 2003-04-16 19:51:27 by athena] Chris@10: for (i = 0; i < m; ++i) ==> for (i = m; i > 0; --i) Chris@10: No proof of evidence that this is any faster, but just in case... Chris@10: Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_notw_c.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -3 +3 Chris@10: M ./genfft/gen_r2r.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -5 +5 Chris@10: M ./genfft/gen_twiddle_c.ml -5 +5 Chris@10: Chris@10: Tue Apr 15 15:03:20 EDT 2003 stevenj Chris@10: * [project @ 2003-04-15 19:03:20 by stevenj] Chris@10: added hack to make sure that codelet loops are preferred to vecloop solvers in the estimator Chris@10: Chris@10: M ./dft/vrank-geq1.c -1 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -1 +2 Chris@10: M ./rdft/vrank-geq1.c -1 +2 Chris@10: Chris@10: Tue Apr 15 14:53:44 EDT 2003 stevenj Chris@10: * [project @ 2003-04-15 18:53:44 by stevenj] Chris@10: use double for flops Chris@10: Chris@10: M ./api/f77funcs.h -1 +1 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/flops.c -1 +1 Chris@10: M ./doc/fftw3.texi -3 +6 Chris@10: M ./kernel/ifftw.h -5 +5 Chris@10: M ./tests/bench.c -2 +2 Chris@10: Chris@10: Tue Apr 15 14:51:50 EDT 2003 stevenj Chris@10: * [project @ 2003-04-15 18:51:50 by stevenj] Chris@10: metrowerks reportedly supports gcc assembly extensions on ppc Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Mon Apr 14 15:00:50 EDT 2003 athena Chris@10: * [project @ 2003-04-14 19:00:48 by athena] Chris@10: foo_CFLAGS generates some automake junk that breaks the build Chris@10: on Redhat 7.3. Screw it. Chris@10: Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./simd/Makefile.am -1 +1 Chris@10: Chris@10: Mon Apr 14 12:22:59 EDT 2003 athena Chris@10: * [project @ 2003-04-14 16:22:59 by athena] Chris@10: Carefully check return status Chris@10: Chris@10: M ./tests/check.pl -1 +21 Chris@10: Chris@10: Sun Apr 13 16:46:12 EDT 2003 athena Chris@10: * [project @ 2003-04-13 20:46:12 by athena] Chris@10: Removed annoying -FMA() expressions. Chris@10: Chris@10: M ./genfft/c.ml -2 +6 Chris@10: M ./genfft/simd.ml -2 +2 Chris@10: M ./kernel/ifftw.h -1 +9 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Sat Apr 12 14:32:22 EDT 2003 athena Chris@10: * [project @ 2003-04-12 18:32:22 by athena] Chris@10: Major fma hackery Chris@10: Chris@10: M ./kernel/ifftw.h -5 +17 Chris@10: Chris@10: Sat Apr 12 14:25:43 EDT 2003 athena Chris@10: * [project @ 2003-04-12 18:25:43 by athena] Chris@10: Slight cleanup Chris@10: Chris@10: M ./api/apiplan.c -8 +5 Chris@10: Chris@10: Sat Apr 12 10:04:51 EDT 2003 athena Chris@10: * [project @ 2003-04-12 14:04:51 by athena] Chris@10: Updated version number Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Apr 12 08:03:07 EDT 2003 athena Chris@10: * [project @ 2003-04-12 12:03:07 by athena] Chris@10: Damn autoconf Chris@10: Chris@10: M ./acinclude.m4 -2 +2 Chris@10: Chris@10: Sat Apr 12 07:54:20 EDT 2003 athena Chris@10: * [project @ 2003-04-12 11:54:20 by athena] Chris@10: Recognize all 74xx processors Chris@10: Chris@10: M ./acinclude.m4 -4 +5 Chris@10: Chris@10: Sat Apr 12 07:35:17 EDT 2003 athena Chris@10: * [project @ 2003-04-12 11:35:17 by athena] Chris@10: Detect 7400 processor. Chris@10: Chris@10: M ./acinclude.m4 -3 +7 Chris@10: Chris@10: Fri Apr 11 20:42:11 EDT 2003 athena Chris@10: * [project @ 2003-04-12 00:42:11 by athena] Chris@10: No need to check for gcc-2.95 Chris@10: Chris@10: M ./acinclude.m4 -4 +2 Chris@10: Chris@10: Fri Apr 11 16:14:39 EDT 2003 stevenj Chris@10: * [project @ 2003-04-11 20:14:39 by stevenj] Chris@10: removed duplicate Chris@10: Chris@10: M ./NEWS -3 Chris@10: Chris@10: Fri Apr 11 08:45:37 EDT 2003 athena Chris@10: * [project @ 2003-04-11 12:45:37 by athena] Chris@10: mflops ==> ``mflops'' Chris@10: Chris@10: M ./libbench2/report.c -2 +2 Chris@10: Chris@10: Fri Apr 11 07:00:53 EDT 2003 athena Chris@10: * [project @ 2003-04-11 11:00:53 by athena] Chris@10: Print setup time as well Chris@10: Chris@10: M ./libbench2/report.c -3 +5 Chris@10: Chris@10: Sat Aug 12 23:14:42 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-0-beta3 Chris@10: Chris@10: Chris@10: Thu Apr 10 15:36:18 EDT 2003 athena Chris@10: * [project @ 2003-04-10 19:36:15 by athena] Chris@10: Enforce pointer equality for in-place problems. Chris@10: Chris@10: M ./dft/problem.c -4 +8 Chris@10: M ./kernel/ifftw.h -1 +4 Chris@10: M ./rdft/problem.c -4 +3 Chris@10: M ./rdft/problem2.c -4 +7 Chris@10: M ./simd/taint.c -1 +8 Chris@10: Chris@10: Wed Apr 9 17:47:54 EDT 2003 stevenj Chris@10: * [project @ 2003-04-09 21:47:26 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +406 Chris@10: M ./NEWS +28 Chris@10: Chris@10: Wed Apr 9 14:53:38 EDT 2003 stevenj Chris@10: * [project @ 2003-04-09 18:53:38 by stevenj] Chris@10: cross-ref fftw-wisdom man page Chris@10: Chris@10: M ./tests/README -1 +1 Chris@10: Chris@10: Wed Apr 9 10:13:00 EDT 2003 athena Chris@10: * [project @ 2003-04-09 14:13:00 by athena] Chris@10: Undone previous change, committed by mistake. Chris@10: Chris@10: M ./kernel/planner.c -2 +1 Chris@10: Chris@10: Wed Apr 9 10:12:24 EDT 2003 athena Chris@10: * [project @ 2003-04-09 14:12:18 by athena] Chris@10: Quick and dirty README for bench Chris@10: Chris@10: A ./tests/README Chris@10: M ./kernel/planner.c -1 +2 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/README +57 Chris@10: Chris@10: Wed Apr 9 08:50:25 EDT 2003 athena Chris@10: * [project @ 2003-04-09 12:50:25 by athena] Chris@10: Consider additional command-line arguments as problems to be Chris@10: benchmarked. Chris@10: Chris@10: M ./libbench2/bench-main.c -7 +6 Chris@10: M ./libbench2/timer.c -1 +7 Chris@10: Chris@10: Wed Apr 9 08:44:13 EDT 2003 athena Chris@10: * [project @ 2003-04-09 12:44:13 by athena] Chris@10: Default report format is now human-readable. Removed Chris@10: unnecessary complexity in benchmark reporting. Chris@10: Chris@10: M ./libbench2/bench-main.c -18 +7 Chris@10: M ./libbench2/bench.h -5 +2 Chris@10: M ./libbench2/report.c -71 +39 Chris@10: Chris@10: Wed Apr 9 06:10:40 EDT 2003 athena Chris@10: * [project @ 2003-04-09 10:10:40 by athena] Chris@10: Updated for new interleaved/split api. Chris@10: Chris@10: M ./doc/fftw3.texi -100 +180 Chris@10: Chris@10: Wed Apr 9 03:01:03 EDT 2003 stevenj Chris@10: * [project @ 2003-04-09 07:01:03 by stevenj] Chris@10: updated citation Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Tue Apr 8 19:35:59 EDT 2003 athena Chris@10: * [project @ 2003-04-08 23:35:59 by athena] Chris@10: Time for beta3 Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Tue Apr 8 17:40:59 EDT 2003 stevenj Chris@10: * [project @ 2003-04-08 21:40:59 by stevenj] Chris@10: whoops, added Chris@10: Chris@10: A ./reodft/redft00e-r2hc-pad.c Chris@10: Chris@10: Tue Apr 8 17:33:47 EDT 2003 stevenj Chris@10: * [project @ 2003-04-08 21:33:47 by stevenj] Chris@10: more comparison of different R*DFT types Chris@10: Chris@10: M ./doc/fftw3.texi -1 +32 Chris@10: Chris@10: Tue Apr 8 16:48:08 EDT 2003 stevenj Chris@10: * [project @ 2003-04-08 20:48:08 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./reodft/redft00e-r2hc.c -5 +18 Chris@10: M ./reodft/rodft00e-r2hc.c -5 +18 Chris@10: Chris@10: Tue Apr 8 16:19:39 EDT 2003 stevenj Chris@10: * [project @ 2003-04-08 20:19:39 by stevenj] Chris@10: more accurate DCT-I and DST-I, at the expense of up to a factor of 2 in speed and memory Chris@10: Chris@10: A ./reodft/rodft00e-r2hc-pad.c Chris@10: M ./reodft/Makefile.am -4 +5 Chris@10: M ./reodft/conf.c -3 +5 Chris@10: M ./reodft/reodft.h +2 Chris@10: M ./reodft/rodft00e-r2hc-pad.c +200 Chris@10: Chris@10: Tue Apr 8 05:38:09 EDT 2003 athena Chris@10: * [project @ 2003-04-08 09:38:09 by athena] Chris@10: Workaround gcc/sparc bug Chris@10: Chris@10: M ./kernel/planner.c -2 +3 Chris@10: Chris@10: Tue Apr 8 01:34:12 EDT 2003 stevenj Chris@10: * [project @ 2003-04-08 05:34:12 by stevenj] Chris@10: rumors Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Apr 7 18:54:11 EDT 2003 stevenj Chris@10: * [project @ 2003-04-07 22:54:11 by stevenj] Chris@10: added rdft2 paranoid mode Chris@10: Chris@10: M ./tests/hook.c +33 Chris@10: Chris@10: Mon Apr 7 18:47:37 EDT 2003 stevenj Chris@10: * [project @ 2003-04-07 22:47:37 by stevenj] Chris@10: added paranoid mode for r2r Chris@10: Chris@10: M ./tests/hook.c +51 Chris@10: Chris@10: Mon Apr 7 15:10:08 EDT 2003 stevenj Chris@10: * [project @ 2003-04-07 19:10:08 by stevenj] Chris@10: whoops, sincos is predefined on some systems Chris@10: Chris@10: M ./libbench2/verify-r2r.c -3 +3 Chris@10: Chris@10: Sat Apr 5 16:50:57 EST 2003 athena Chris@10: * [project @ 2003-04-05 21:50:57 by athena] Chris@10: bp->destroy_input was not initialized Chris@10: Chris@10: M ./tests/hook.c +1 Chris@10: Chris@10: Sat Apr 5 09:29:11 EST 2003 athena Chris@10: * [project @ 2003-04-05 14:29:11 by athena] Chris@10: Asserted correctness conditions for tainted pointers. Chris@10: Chris@10: (For now, use CK() while we test. They should be changed into Chris@10: A() at some point.) Chris@10: Chris@10: M ./dft/problem.c -7 +16 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./rdft/problem.c -3 +7 Chris@10: M ./rdft/problem2.c -6 +12 Chris@10: Chris@10: Sat Apr 5 08:18:23 EST 2003 athena Chris@10: * [project @ 2003-04-05 13:18:23 by athena] Chris@10: Untaint pointers before zero'ing arrays and before hashing Chris@10: Chris@10: M ./dft/problem.c -8 +10 Chris@10: M ./rdft/problem.c -4 +4 Chris@10: M ./rdft/problem2.c -9 +9 Chris@10: Chris@10: Sat Apr 5 07:11:56 EST 2003 athena Chris@10: * [project @ 2003-04-05 12:11:56 by athena] Chris@10: Alignment check did not work with icc, which seems to be Chris@10: confused by the fact that the variable is not used. Chris@10: Chris@10: M ./libbench2/bench-main.c -5 +4 Chris@10: Chris@10: Sat Apr 5 06:41:20 EST 2003 athena Chris@10: * [project @ 2003-04-05 11:41:20 by athena] Chris@10: More paranoid paranoid-check Chris@10: Chris@10: M ./tests/Makefile.am -1 +7 Chris@10: Chris@10: Sat Apr 5 06:19:25 EST 2003 athena Chris@10: * [project @ 2003-04-05 11:19:25 by athena] Chris@10: 0 == x & 7 parses as (0 == x) & 7, which is wrong Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Fri Apr 4 21:35:49 EST 2003 stevenj Chris@10: * [project @ 2003-04-05 02:35:49 by stevenj] Chris@10: alignment checks Chris@10: Chris@10: M ./dft/direct.c -1 +2 Chris@10: M ./kernel/ifftw.h -1 +12 Chris@10: M ./kernel/planner.c -1 +2 Chris@10: M ./libbench2/bench-main.c -1 +3 Chris@10: M ./rdft/direct.c -1 +4 Chris@10: M ./rdft/direct2.c -1 +3 Chris@10: Chris@10: Fri Apr 4 21:04:14 EST 2003 stevenj Chris@10: * [project @ 2003-04-05 02:04:14 by stevenj] Chris@10: prevent infinite loops in exhaustive planning Chris@10: Chris@10: M ./rdft/rdft-dht.c -1 +6 Chris@10: Chris@10: Fri Apr 4 20:58:20 EST 2003 stevenj Chris@10: * [project @ 2003-04-05 01:58:20 by stevenj] Chris@10: split/unsplit guru interface Chris@10: Chris@10: A ./api/execute-split-dft-c2r.c Chris@10: A ./api/execute-split-dft-r2c.c Chris@10: A ./api/execute-split-dft.c Chris@10: A ./api/plan-guru-split-dft-c2r.c Chris@10: A ./api/plan-guru-split-dft-r2c.c Chris@10: A ./api/plan-guru-split-dft.c Chris@10: M ./api/Makefile.am -1 +3 Chris@10: M ./api/api.h -2 +3 Chris@10: M ./api/apiplan.c -3 +4 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -3 +6 Chris@10: M ./api/execute-split-dft-c2r.c +29 Chris@10: M ./api/execute-split-dft-r2c.c +29 Chris@10: M ./api/execute-split-dft.c +29 Chris@10: M ./api/f77funcs.h -8 +73 Chris@10: M ./api/fftw3.h -5 +25 Chris@10: M ./api/mktensor-iodims.c -3 +3 Chris@10: M ./api/plan-guru-dft-c2r.c -4 +9 Chris@10: M ./api/plan-guru-dft-r2c.c -4 +9 Chris@10: M ./api/plan-guru-dft.c -4 +10 Chris@10: M ./api/plan-guru-r2r.c -3 +4 Chris@10: M ./api/plan-guru-split-dft-c2r.c +40 Chris@10: M ./api/plan-guru-split-dft-r2c.c +39 Chris@10: M ./api/plan-guru-split-dft.c +39 Chris@10: M ./api/plan-many-dft-c2r.c -1 +1 Chris@10: M ./api/plan-many-dft-r2c.c -1 +1 Chris@10: M ./api/plan-many-dft.c -1 +1 Chris@10: M ./api/plan-many-r2r.c -1 +1 Chris@10: M ./tests/bench.c -29 +15 Chris@10: Chris@10: Fri Apr 4 20:39:55 EST 2003 athena Chris@10: * [project @ 2003-04-05 01:39:55 by athena] Chris@10: Need UNTAINT in verifier too. Chris@10: Chris@10: M ./tests/hook.c -4 +4 Chris@10: Chris@10: Fri Apr 4 19:36:46 EST 2003 athena Chris@10: * [project @ 2003-04-05 00:36:46 by athena] Chris@10: Forgot #if HAVE_SIMD Chris@10: Chris@10: M ./simd/taint.c -1 +5 Chris@10: Chris@10: Fri Apr 4 19:30:37 EST 2003 athena Chris@10: * [project @ 2003-04-05 00:30:37 by athena] Chris@10: Keep track of two separate taint bits Chris@10: Chris@10: A ./simd/taint.c Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./kernel/align.c -15 +1 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./simd/Makefile.am -1 +1 Chris@10: M ./simd/simd.h -2 +16 Chris@10: M ./simd/taint.c +33 Chris@10: Chris@10: Fri Apr 4 19:16:32 EST 2003 stevenj Chris@10: * [project @ 2003-04-05 00:16:32 by stevenj] Chris@10: added NO_SIMD problem flag, made UNALIGNED an API issue (taints input pointers) Chris@10: Chris@10: M ./api/api.h +2 Chris@10: M ./api/fftw3.h -1 +3 Chris@10: M ./api/mapflags.c -1 +1 Chris@10: M ./api/plan-guru-dft-c2r.c -1 +3 Chris@10: M ./api/plan-guru-dft-r2c.c -1 +3 Chris@10: M ./api/plan-guru-dft.c -1 +4 Chris@10: M ./api/plan-guru-r2r.c -1 +2 Chris@10: M ./api/plan-many-dft-c2r.c -1 +3 Chris@10: M ./api/plan-many-dft-r2c.c -1 +3 Chris@10: M ./api/plan-many-dft.c -1 +4 Chris@10: M ./api/plan-many-r2r.c -1 +2 Chris@10: M ./dft/k7/k7.c -5 +13 Chris@10: M ./dft/simd/n1b.c -1 +1 Chris@10: M ./dft/simd/n1f.c -1 +1 Chris@10: M ./dft/simd/n2b.c +1 Chris@10: M ./dft/simd/n2f.c +1 Chris@10: M ./dft/simd/q1b.c -1 +1 Chris@10: M ./dft/simd/q1f.c -1 +1 Chris@10: M ./dft/simd/t1b.c -1 +1 Chris@10: M ./dft/simd/t1f.c -1 +1 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Fri Apr 4 18:14:14 EST 2003 stevenj Chris@10: * [project @ 2003-04-04 23:14:14 by stevenj] Chris@10: bugfix in buffered: wrong pointers passed for cldrest; also use TAINT instead of UNALIGNED in buffered2 Chris@10: Chris@10: M ./dft/buffered.c -6 +10 Chris@10: M ./rdft/buffered.c -6 +10 Chris@10: M ./rdft/buffered2.c -7 +5 Chris@10: Chris@10: Fri Apr 4 17:19:51 EST 2003 athena Chris@10: * [project @ 2003-04-04 22:19:51 by athena] Chris@10: Reverted previous change, committed accidentally Chris@10: Chris@10: M ./dft/vrank-geq1.c -3 +3 Chris@10: Chris@10: Fri Apr 4 17:18:39 EST 2003 athena Chris@10: * [project @ 2003-04-04 22:18:39 by athena] Chris@10: What was I thinking? Chris@10: Chris@10: M ./kernel/align.c -1 +2 Chris@10: Chris@10: Fri Apr 4 17:18:21 EST 2003 athena Chris@10: * [project @ 2003-04-04 22:18:21 by athena] Chris@10: Chris@10: M ./dft/vrank-geq1.c -3 +3 Chris@10: Chris@10: Fri Apr 4 16:48:32 EST 2003 stevenj Chris@10: * [project @ 2003-04-04 21:48:32 by stevenj] Chris@10: added --enable-debug-alignment Chris@10: Chris@10: M ./configure.ac +5 Chris@10: M ./libbench2/aligned-main.c -1 +7 Chris@10: Chris@10: Fri Apr 4 16:29:43 EST 2003 stevenj Chris@10: * [project @ 2003-04-04 21:29:43 by stevenj] Chris@10: X(taint) prototype, define corresponding function only if HAVE_SIMD Chris@10: Chris@10: M ./kernel/align.c -1 +3 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: Chris@10: Fri Apr 4 16:15:53 EST 2003 athena Chris@10: * [project @ 2003-04-04 21:15:53 by athena] Chris@10: Initial checkin of tained pointers Chris@10: Chris@10: M ./dft/buffered.c -4 +9 Chris@10: M ./dft/solve.c -2 +4 Chris@10: M ./dft/vrank-geq1.c -3 +3 Chris@10: M ./kernel/align.c -6 +4 Chris@10: M ./kernel/ifftw.h -2 +9 Chris@10: M ./rdft/buffered.c -4 +4 Chris@10: M ./rdft/buffered2.c -3 +2 Chris@10: M ./rdft/solve.c -2 +2 Chris@10: M ./rdft/solve2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -3 +4 Chris@10: M ./rdft/vrank-geq1.c -3 +3 Chris@10: Chris@10: Fri Apr 4 13:12:58 EST 2003 athena Chris@10: * [project @ 2003-04-04 18:12:52 by athena] Chris@10: More conservative preservation of alignment Chris@10: Chris@10: M ./dft/buffered.c -1 +2 Chris@10: M ./dft/rader.c -4 +3 Chris@10: M ./dft/simd/n2b.c -1 Chris@10: M ./dft/simd/n2f.c -1 Chris@10: M ./dft/vrank-geq1.c -6 +4 Chris@10: M ./kernel/align.c -35 +8 Chris@10: M ./kernel/ifftw.h -3 +2 Chris@10: M ./rdft/buffered.c -1 +2 Chris@10: M ./rdft/buffered2.c -1 +4 Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -5 +3 Chris@10: M ./rdft/vrank-geq1.c -5 +5 Chris@10: Chris@10: Thu Apr 3 23:16:27 EST 2003 stevenj Chris@10: * [project @ 2003-04-04 04:16:27 by stevenj] Chris@10: plan/execute with aligned stack Chris@10: Chris@10: M ./api/apiplan.c -19 +24 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/f77funcs.h -10 +10 Chris@10: Chris@10: Thu Apr 3 15:40:01 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 20:40:01 by stevenj] Chris@10: whoops, missed FFTW_MEASURE in fftw3.f Chris@10: Chris@10: M ./api/Makefile.am -1 +1 Chris@10: Chris@10: Thu Apr 3 13:44:46 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 18:44:46 by stevenj] Chris@10: use WITH_ALIGNED_STACK for experimental semaphore stuff, too Chris@10: Chris@10: M ./threads/threads.c -8 +3 Chris@10: Chris@10: Thu Apr 3 09:04:23 EST 2003 athena Chris@10: * [project @ 2003-04-03 14:04:23 by athena] Chris@10: Removed old file Chris@10: Chris@10: R ./kernel/stack.c Chris@10: Chris@10: Thu Apr 3 07:50:43 EST 2003 athena Chris@10: * [project @ 2003-04-03 12:50:43 by athena] Chris@10: Improved stack-alignment hack Chris@10: Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -4 +52 Chris@10: M ./threads/ct-dit.c -7 +6 Chris@10: M ./threads/dft-vrank-geq1.c -7 +6 Chris@10: M ./threads/hc2hc-dif.c -7 +6 Chris@10: M ./threads/hc2hc-dit.c -7 +6 Chris@10: M ./threads/rdft-vrank-geq1.c -6 +4 Chris@10: M ./threads/threads.h -5 Chris@10: M ./threads/vrank-geq1-rdft2.c -7 +6 Chris@10: Chris@10: Thu Apr 3 02:37:57 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 07:37:57 by stevenj] Chris@10: use aligned stack for experimental semaphores, too Chris@10: Chris@10: M ./threads/threads.c -2 +8 Chris@10: Chris@10: Thu Apr 3 02:17:58 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 07:17:58 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/stack.c -2 +2 Chris@10: M ./threads/ct-dit.c -2 +4 Chris@10: M ./threads/dft-vrank-geq1.c -2 +4 Chris@10: M ./threads/hc2hc-dif.c -2 +4 Chris@10: M ./threads/hc2hc-dit.c -2 +4 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +4 Chris@10: M ./threads/threads.c -24 +3 Chris@10: M ./threads/threads.h +5 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +4 Chris@10: Chris@10: Thu Apr 3 01:58:32 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 06:58:32 by stevenj] Chris@10: fix(?) for SIMD thread problems Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/stack.c -3 +3 Chris@10: M ./threads/ct-dit.c -7 +10 Chris@10: M ./threads/hc2hc-dif.c -2 +2 Chris@10: M ./threads/hc2hc-dit.c -2 +2 Chris@10: M ./threads/threads.c -6 +27 Chris@10: Chris@10: Wed Apr 2 20:33:12 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 01:33:12 by stevenj] Chris@10: noted n=1 REDFT01 case Chris@10: Chris@10: M ./doc/fftw3.texi -1 +11 Chris@10: Chris@10: Wed Apr 2 20:32:07 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 01:32:07 by stevenj] Chris@10: note about n=2 REDFT00 formula Chris@10: Chris@10: M ./doc/fftw3.texi -2 +3 Chris@10: Chris@10: Wed Apr 2 20:30:10 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 01:30:10 by stevenj] Chris@10: note about undefined REDFT00 Chris@10: Chris@10: M ./doc/fftw3.texi -1 +2 Chris@10: Chris@10: Wed Apr 2 20:18:03 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 01:18:03 by stevenj] Chris@10: noted n=1 RODFT01 case Chris@10: Chris@10: M ./doc/fftw3.texi -2 +12 Chris@10: Chris@10: Wed Apr 2 20:14:07 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 01:14:07 by stevenj] Chris@10: corrected definitions Chris@10: Chris@10: M ./doc/equation-redft11.png Chris@10: M ./doc/equation-rodft01.png Chris@10: M ./doc/equation-rodft11.png Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Wed Apr 2 19:43:59 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 00:43:59 by stevenj] Chris@10: added REODFT_KINDP, fixed nontrivial test for R2HC11 and HC2R11 (not that we support these yet anyway) Chris@10: Chris@10: M ./rdft/codelet-rdft.h +1 Chris@10: M ./rdft/problem.c -3 +3 Chris@10: M ./rdft/vrank-geq1.c -3 +3 Chris@10: Chris@10: Wed Apr 2 19:16:54 EST 2003 stevenj Chris@10: * [project @ 2003-04-03 00:16:53 by stevenj] Chris@10: size 2 hc2r and dht are equivalent to r2hc Chris@10: Chris@10: M ./rdft/codelets/hc2r/Makefile.am -3 +5 Chris@10: M ./rdft/problem.c -3 +5 Chris@10: Chris@10: Wed Apr 2 15:09:08 EST 2003 stevenj Chris@10: * [project @ 2003-04-02 20:09:08 by stevenj] Chris@10: noted overwriting in upgrading section Chris@10: Chris@10: M ./doc/fftw3.texi -1 +4 Chris@10: Chris@10: Wed Apr 2 05:25:56 EST 2003 athena Chris@10: * [project @ 2003-04-02 10:25:56 by athena] Chris@10: Moved with_aligned_stack to its own file Chris@10: Chris@10: A ./kernel/stack.c Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/align.c -46 +1 Chris@10: M ./kernel/stack.c +67 Chris@10: Chris@10: Tue Apr 1 21:11:31 EST 2003 athena Chris@10: * [project @ 2003-04-02 02:11:31 by athena] Chris@10: Fixed comments Chris@10: Chris@10: M ./kernel/align.c -6 +2 Chris@10: M ./libbench2/aligned-main.c -6 +2 Chris@10: Chris@10: Tue Apr 1 20:57:39 EST 2003 athena Chris@10: * [project @ 2003-04-02 01:57:39 by athena] Chris@10: Alignment hacks Chris@10: Chris@10: M ./kernel/align.c -1 +50 Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./libbench2/aligned-main.c -26 +14 Chris@10: M ./libbench2/bench-main.c -1 +9 Chris@10: Chris@10: Tue Apr 1 14:26:48 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 19:26:48 by stevenj] Chris@10: phew, no, previous version was okay Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Tue Apr 1 14:26:15 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 19:26:15 by stevenj] Chris@10: whoops, crap Chris@10: Chris@10: M ./threads/threads.c -2 +2 Chris@10: Chris@10: Tue Apr 1 08:01:06 EST 2003 athena Chris@10: * [project @ 2003-04-01 13:01:06 by athena] Chris@10: support sse2 in forthcoming gcc-3.3 Chris@10: Chris@10: M ./simd/simd-sse2.h -3 +11 Chris@10: Chris@10: Tue Apr 1 01:17:15 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 06:17:15 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Tue Apr 1 01:16:46 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 06:16:46 by stevenj] Chris@10: noted ac_check_headers Chris@10: Chris@10: M ./kernel/cycle.h -1 +4 Chris@10: Chris@10: Tue Apr 1 01:11:31 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 06:11:31 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./kernel/cycle.h -1 +2 Chris@10: Chris@10: Tue Apr 1 01:06:53 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 06:06:53 by stevenj] Chris@10: documented autoconf tests, so that cycle.h can be distributed separately Chris@10: Chris@10: M ./kernel/cycle.h -1 +27 Chris@10: Chris@10: Sat Aug 12 23:11:17 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-0-beta2 Chris@10: Chris@10: Chris@10: Mon Mar 31 22:12:02 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 03:12:02 by stevenj] Chris@10: IRIX is all-caps Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Mon Mar 31 22:11:42 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 03:11:42 by stevenj] Chris@10: noted Irix fix Chris@10: Chris@10: M ./NEWS +3 Chris@10: Chris@10: Mon Mar 31 22:10:33 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 03:10:33 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./threads/api.c -1 +1 Chris@10: M ./threads/threads.h -1 +1 Chris@10: Chris@10: Mon Mar 31 22:04:35 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 03:04:35 by stevenj] Chris@10: use ithreads_init so as not to confuse fftw 2 users Chris@10: Chris@10: M ./threads/threads.c -6 +8 Chris@10: Chris@10: Mon Mar 31 22:00:42 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 03:00:42 by stevenj] Chris@10: IRIX lossage Chris@10: Chris@10: M ./threads/threads.c -1 +5 Chris@10: Chris@10: Mon Mar 31 21:19:20 EST 2003 stevenj Chris@10: * [project @ 2003-04-01 02:19:20 by stevenj] Chris@10: check for -openmp (icc) among the OpenMP flags (TODO: make this a Chris@10: separate macro, with a loop instead of repeated checks) Chris@10: Chris@10: M ./configure.ac +5 Chris@10: Chris@10: Mon Mar 31 17:12:19 EST 2003 stevenj Chris@10: * [project @ 2003-03-31 22:12:19 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -1 +4 Chris@10: Chris@10: Mon Mar 31 17:01:16 EST 2003 athena Chris@10: * [project @ 2003-03-31 22:01:16 by athena] Chris@10: More liberal test for solaris CC Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Mon Mar 31 15:13:33 EST 2003 athena Chris@10: * [project @ 2003-03-31 20:13:33 by athena] Chris@10: Allow x86-64 simd Chris@10: Chris@10: M ./simd/simd-sse.h -1 +1 Chris@10: M ./simd/simd-sse2.h -1 +1 Chris@10: Chris@10: Mon Mar 31 15:13:21 EST 2003 athena Chris@10: * [project @ 2003-03-31 20:13:21 by athena] Chris@10: Added x86-64 timer code Chris@10: Chris@10: M ./kernel/cycle.h -1 +22 Chris@10: Chris@10: Mon Mar 31 13:10:54 EST 2003 stevenj Chris@10: * [project @ 2003-03-31 18:10:54 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./NEWS +4 Chris@10: Chris@10: Mon Mar 31 13:07:19 EST 2003 stevenj Chris@10: * [project @ 2003-03-31 18:07:19 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog -1378 +3497 Chris@10: Chris@10: Mon Mar 31 13:05:27 EST 2003 stevenj Chris@10: * [project @ 2003-03-31 18:05:27 by stevenj] Chris@10: colon Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Mon Mar 31 07:20:20 EST 2003 athena Chris@10: * [project @ 2003-03-31 12:20:20 by athena] Chris@10: Reorganized compiler bugs section (which is growing out of control) Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -11 +18 Chris@10: Chris@10: Mon Mar 31 07:15:20 EST 2003 athena Chris@10: * [project @ 2003-03-31 12:15:20 by athena] Chris@10: solaris gcc bug appears to be also in 2.95.2 Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Mon Mar 31 07:13:45 EST 2003 athena Chris@10: * [project @ 2003-03-31 12:13:45 by athena] Chris@10: Workaround works---there is another gcc/sparc bug elsehwere Chris@10: Chris@10: M ./kernel/planner.c -3 +3 Chris@10: Chris@10: Mon Mar 31 07:08:56 EST 2003 athena Chris@10: * [project @ 2003-03-31 12:08:56 by athena] Chris@10: Grrr, workaround does not work. Chris@10: Chris@10: M ./kernel/planner.c -3 +3 Chris@10: Chris@10: Mon Mar 31 07:02:23 EST 2003 athena Chris@10: * [project @ 2003-03-31 12:02:23 by athena] Chris@10: ADDMOD is now function, which seems to avoid gcc bugs. Chris@10: Chris@10: M ./kernel/planner.c -10 +16 Chris@10: Chris@10: Sun Mar 30 16:40:26 EST 2003 athena Chris@10: * [project @ 2003-03-30 21:40:26 by athena] Chris@10: Workaround sparc gcc bug Chris@10: Chris@10: M ./kernel/planner.c -3 +5 Chris@10: Chris@10: Sun Mar 30 15:51:59 EST 2003 stevenj Chris@10: * [project @ 2003-03-30 20:50:59 by stevenj] Chris@10: note Chris@10: Chris@10: M ./doc/fftw3.texi -4 +5 Chris@10: Chris@10: Sun Mar 30 15:34:57 EST 2003 stevenj Chris@10: * [project @ 2003-03-30 20:34:57 by stevenj] Chris@10: make non-square UGLY, for now Chris@10: Chris@10: M ./dft/vrank2-transpose.c -4 +4 Chris@10: M ./dft/vrank3-transpose.c -1 +4 Chris@10: Chris@10: Sun Mar 30 15:33:57 EST 2003 stevenj Chris@10: * [project @ 2003-03-30 20:33:57 by stevenj] Chris@10: added -o amnesia to forget_wisdom before each plan Chris@10: Chris@10: M ./tests/bench.c +5 Chris@10: Chris@10: Sun Mar 30 09:41:27 EST 2003 athena Chris@10: * [project @ 2003-03-30 14:41:27 by athena] Chris@10: Report setup time in benchmark Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +4 Chris@10: M ./libbench2/report.c -2 +2 Chris@10: M ./libbench2/speed.c -1 +4 Chris@10: Chris@10: Sat Mar 29 20:21:15 EST 2003 stevenj Chris@10: * [project @ 2003-03-30 01:20:52 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./kernel/transpose.c -1 +2 Chris@10: Chris@10: Sat Mar 29 19:11:10 EST 2003 stevenj Chris@10: * [project @ 2003-03-30 00:11:10 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Sat Mar 29 18:46:16 EST 2003 athena Chris@10: * [project @ 2003-03-29 23:46:16 by athena] Chris@10: More relaxed definition of UGLYness Chris@10: Chris@10: M ./kernel/ct.c -1 +1 Chris@10: Chris@10: Sat Mar 29 15:28:01 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 20:28:01 by stevenj] Chris@10: no more cvs id strings in header files...I'm tired of having to rebuild everything after a commit Chris@10: Chris@10: M ./rdft/codelet-rdft.h -2 Chris@10: M ./rdft/hc2hc.h -2 Chris@10: M ./rdft/rdft.h -2 Chris@10: M ./reodft/reodft.h -2 Chris@10: M ./threads/threads.h -2 Chris@10: Chris@10: Sat Mar 29 15:22:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 20:22:28 by stevenj] Chris@10: rdft2 stride unification Chris@10: Chris@10: A ./rdft/rdft2-strides.c Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/buffered2.c -20 +11 Chris@10: M ./rdft/direct2.c -5 +5 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: M ./rdft/rdft2-inplace-strides.c -9 +2 Chris@10: M ./rdft/rdft2-strides.c +38 Chris@10: M ./rdft/rdft2-tensor-max-index.c -7 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -26 +9 Chris@10: M ./threads/vrank-geq1-rdft2.c -9 +3 Chris@10: Chris@10: Sat Mar 29 14:38:23 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 19:38:23 by stevenj] Chris@10: preserve in-place-ness Chris@10: Chris@10: M ./rdft/vrank-geq1-rdft2.c -3 +3 Chris@10: Chris@10: Sat Mar 29 14:23:31 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 19:23:31 by stevenj] Chris@10: make nowisdom the default Chris@10: Chris@10: M ./tests/Makefile.am -7 +7 Chris@10: M ./tests/bench.c -2 +2 Chris@10: M ./tests/check.pl -3 +3 Chris@10: Chris@10: Sat Mar 29 14:13:18 EST 2003 athena Chris@10: * [project @ 2003-03-29 19:13:18 by athena] Chris@10: --verbose in paranoid-check produces too much output. Make it quiet. Chris@10: Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Sat Mar 29 13:45:13 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 18:45:13 by stevenj] Chris@10: fixed transpose bugs...need to check ri-ii before deciding whether Ntuple fits Chris@10: Chris@10: M ./dft/vrank2-transpose.c -5 +9 Chris@10: M ./dft/vrank3-transpose.c -7 +13 Chris@10: M ./kernel/ifftw.h -3 +5 Chris@10: M ./kernel/transpose.c -11 +21 Chris@10: Chris@10: Sat Mar 29 08:10:40 EST 2003 athena Chris@10: * [project @ 2003-03-29 13:10:40 by athena] Chris@10: try more 2^k Chris@10: Chris@10: M ./tests/check.pl +2 Chris@10: Chris@10: Sat Mar 29 08:05:41 EST 2003 athena Chris@10: * [project @ 2003-03-29 13:05:41 by athena] Chris@10: MIN_ALIGNMENT was defined after being used, causing crash in sse2. Chris@10: Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: Chris@10: Sat Mar 29 03:07:34 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 08:07:34 by stevenj] Chris@10: real transposes are currently unused, and are not needed for MPI code either Chris@10: Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: R ./kernel/tensor10.c Chris@10: M ./kernel/transpose.c -180 +18 Chris@10: M ./rdft/Makefile.am -2 +3 Chris@10: M ./rdft/conf.c -3 +5 Chris@10: Chris@10: Sat Mar 29 02:58:39 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 07:58:39 by stevenj] Chris@10: added general transpose Chris@10: Chris@10: A ./kernel/transpose.c Chris@10: M ./dft/vrank2-transpose.c -9 +43 Chris@10: M ./dft/vrank3-transpose.c -25 +57 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -1 +13 Chris@10: M ./kernel/transpose.c +523 Chris@10: Chris@10: Fri Mar 28 22:49:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 03:49:04 by stevenj] Chris@10: added transposition option Chris@10: Chris@10: M ./libbench2/problem.c -1 +18 Chris@10: Chris@10: Fri Mar 28 22:09:22 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 03:09:22 by stevenj] Chris@10: yikes, fixed incorrect applicability of transpose plans Chris@10: Chris@10: A ./kernel/tensor10.c Chris@10: M ./dft/vrank2-transpose.c -3 +3 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/tensor10.c +31 Chris@10: M ./rdft/vrank2-transpose.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Fri Mar 28 22:06:14 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 03:06:14 by stevenj] Chris@10: in the future, we might want to allow sz->rnk == 0, vecsz->rnk arbitrary to be converted to r2hc (the apply function already should work for this case)...disabled for now, though Chris@10: Chris@10: M ./rdft/dft-r2hc.c -7 +21 Chris@10: Chris@10: Fri Mar 28 19:12:08 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 00:12:08 by stevenj] Chris@10: use most_unaligned in rdft2 Chris@10: Chris@10: M ./kernel/align.c -6 +1 Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: M ./rdft/vrank-geq1-rdft2.c -6 +11 Chris@10: Chris@10: Fri Mar 28 19:11:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 00:11:47 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./tests/Makefile.am -2 +2 Chris@10: Chris@10: Fri Mar 28 19:00:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-29 00:00:21 by stevenj] Chris@10: output message when checks pass Chris@10: Chris@10: M ./tests/Makefile.am +12 Chris@10: Chris@10: Fri Mar 28 17:21:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 22:21:47 by stevenj] Chris@10: added ifndef alloca around alloca stuff Chris@10: Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: Chris@10: Fri Mar 28 13:45:50 EST 2003 athena Chris@10: * [project @ 2003-03-28 18:45:43 by athena] Chris@10: Proper alignment in rader Chris@10: Chris@10: M ./dft/rader.c -2 +4 Chris@10: M ./dft/vrank-geq1.c -4 +3 Chris@10: M ./kernel/align.c -3 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./rdft/dht-rader.c -2 +5 Chris@10: Chris@10: Fri Mar 28 12:43:23 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 17:43:23 by stevenj] Chris@10: whitespace Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: Chris@10: Fri Mar 28 12:41:39 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 17:41:39 by stevenj] Chris@10: whoops, alloca stuff inside HAVE_ALLOCA Chris@10: Chris@10: M ./kernel/ifftw.h -3 +4 Chris@10: Chris@10: Fri Mar 28 12:35:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 17:35:21 by stevenj] Chris@10: make check can afford to be a little bigger Chris@10: Chris@10: M ./tests/Makefile.am -2 +2 Chris@10: Chris@10: Fri Mar 28 12:31:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 17:31:32 by stevenj] Chris@10: use same alloca macrology as configure script Chris@10: Chris@10: M ./kernel/ifftw.h -28 +45 Chris@10: Chris@10: Fri Mar 28 03:05:15 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 08:05:15 by stevenj] Chris@10: fallback is no longer needed for mingw Chris@10: Chris@10: M ./kernel/ifftw.h -6 +1 Chris@10: Chris@10: Fri Mar 28 02:58:45 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 07:58:45 by stevenj] Chris@10: alloca fallback for gcc Chris@10: Chris@10: M ./kernel/ifftw.h -3 +4 Chris@10: Chris@10: Fri Mar 28 02:49:59 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 07:49:59 by stevenj] Chris@10: _alloca was added for MinGW, but it causes problems there Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Thu Mar 27 22:06:07 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 03:06:07 by stevenj] Chris@10: fixed most_unaligned for split format Chris@10: Chris@10: M ./kernel/align.c -5 +8 Chris@10: Chris@10: Thu Mar 27 19:01:58 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 00:01:58 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: Chris@10: Thu Mar 27 19:00:20 EST 2003 stevenj Chris@10: * [project @ 2003-03-28 00:00:20 by stevenj] Chris@10: added pkg-config Chris@10: Chris@10: A ./fftw.pc.in Chris@10: M ./Makefile.am -1 +6 Chris@10: M ./configure.ac +1 Chris@10: M ./fftw.pc.in +10 Chris@10: Chris@10: Thu Mar 27 15:59:01 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 20:59:01 by stevenj] Chris@10: fixed asserts Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +3 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Thu Mar 27 15:49:53 EST 2003 athena Chris@10: * [project @ 2003-03-27 20:49:53 by athena] Chris@10: Do not adjust r/i pointers separately. Chris@10: Chris@10: M ./kernel/align.c -4 +5 Chris@10: Chris@10: Thu Mar 27 15:17:40 EST 2003 athena Chris@10: * [project @ 2003-03-27 20:17:40 by athena] Chris@10: iForgot to add files Chris@10: Chris@10: A ./dft/simd/n2b.h Chris@10: A ./dft/simd/n2f.h Chris@10: M ./dft/simd/n2b.h +25 Chris@10: M ./dft/simd/n2f.h +25 Chris@10: Chris@10: Thu Mar 27 15:10:41 EST 2003 athena Chris@10: * [project @ 2003-03-27 20:10:36 by athena] Chris@10: Specialized n simd codelets for unit vector stride. Chris@10: Chris@10: A ./dft/simd/n2b.c Chris@10: A ./dft/simd/n2f.c Chris@10: M ./configure.ac +4 Chris@10: M ./dft/simd/Makefile.am -2 +2 Chris@10: M ./dft/simd/codelets/Makefile.am -8 +54 Chris@10: M ./dft/simd/n1b.c -3 +3 Chris@10: M ./dft/simd/n1b.h +4 Chris@10: M ./dft/simd/n1f.c -3 +3 Chris@10: M ./dft/simd/n1f.h +4 Chris@10: M ./dft/simd/n2b.c +49 Chris@10: M ./dft/simd/n2f.c +49 Chris@10: Chris@10: Thu Mar 27 08:22:03 EST 2003 athena Chris@10: * [project @ 2003-03-27 13:22:03 by athena] Chris@10: Changed version number to beta2 Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Thu Mar 27 06:37:07 EST 2003 athena Chris@10: * [project @ 2003-03-27 11:37:07 by athena] Chris@10: Changed alignment requirements for n1 simd codelets. Changed Chris@10: mechanism for detecting lack of alignment. Chris@10: Chris@10: M ./api/mapflags.c -1 +1 Chris@10: M ./dft/simd/n1b.c -4 +4 Chris@10: M ./dft/simd/n1b.h -3 Chris@10: M ./dft/simd/n1f.c -4 +4 Chris@10: M ./dft/simd/n1f.h -3 Chris@10: M ./dft/simd/q1b.c -1 +1 Chris@10: M ./dft/simd/q1f.c -1 +1 Chris@10: M ./dft/simd/t1b.c -1 +1 Chris@10: M ./dft/simd/t1f.c -1 +1 Chris@10: M ./dft/vrank-geq1.c -7 +7 Chris@10: M ./kernel/align.c -5 +42 Chris@10: M ./kernel/ifftw.h -3 +5 Chris@10: M ./rdft/vrank-geq1-rdft2.c -4 +7 Chris@10: M ./rdft/vrank-geq1.c -5 +5 Chris@10: Chris@10: Thu Mar 27 04:25:06 EST 2003 athena Chris@10: * [project @ 2003-03-27 09:25:06 by athena] Chris@10: Oops, wrong place for hook Chris@10: Chris@10: M ./tests/bench.c -2 +1 Chris@10: Chris@10: Thu Mar 27 02:37:52 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 07:37:52 by stevenj] Chris@10: added comments to codelet makefiles, to aid people wanting to generate their own code Chris@10: Chris@10: M ./dft/codelets/inplace/Makefile.am -1 +23 Chris@10: M ./dft/codelets/standard/Makefile.am +24 Chris@10: M ./dft/k7/codelets/Makefile.am +25 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +32 Chris@10: M ./rdft/codelets/hc2r/Makefile.am +25 Chris@10: M ./rdft/codelets/r2hc/Makefile.am +25 Chris@10: M ./rdft/codelets/r2r/Makefile.am +18 Chris@10: Chris@10: Thu Mar 27 01:42:27 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:42:27 by stevenj] Chris@10: Matteo is also a copyright holder Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Thu Mar 27 01:41:08 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:41:08 by stevenj] Chris@10: FORTRAN is officially Fortran, these days Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +2 Chris@10: Chris@10: Thu Mar 27 01:40:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:40:32 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Thu Mar 27 01:40:14 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:40:14 by stevenj] Chris@10: don't use "wrapper" Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Thu Mar 27 01:37:53 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:37:53 by stevenj] Chris@10: plural Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Thu Mar 27 01:35:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:35:32 by stevenj] Chris@10: grammar Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Thu Mar 27 01:33:35 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 06:33:35 by stevenj] Chris@10: better phrasing Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Wed Mar 26 22:47:58 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 03:47:58 by stevenj] Chris@10: stddef.h should not be needed anymore for this file Chris@10: Chris@10: M ./kernel/align.c -2 +1 Chris@10: Chris@10: Wed Mar 26 22:13:48 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 03:13:48 by stevenj] Chris@10: added comments for Franz mode Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -5 +5 Chris@10: Chris@10: Wed Mar 26 22:11:58 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 03:10:11 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -3 +4 Chris@10: Chris@10: Wed Mar 26 22:08:51 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 03:08:51 by stevenj] Chris@10: commented on FRANZ codelets Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -5 +5 Chris@10: Chris@10: Wed Mar 26 22:06:45 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 03:06:45 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./NEWS +23 Chris@10: Chris@10: Wed Mar 26 21:51:15 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 02:51:15 by stevenj] Chris@10: disable DIF codelets, since they are never used (apparently) except Chris@10: for some non-power-of-two sizes...improve support for the latter by Chris@10: adding size 3, 5, and 6 q^2 codelets. Chris@10: Chris@10: M ./dft/codelets/inplace/Makefile.am -4 +8 Chris@10: Chris@10: Wed Mar 26 20:07:11 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 01:07:11 by stevenj] Chris@10: DHT has no forward/backward Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Wed Mar 26 19:46:12 EST 2003 fftw Chris@10: * [project @ 2003-03-27 00:46:12 by fftw] Chris@10: added hacky way to use an arbitrary flag Chris@10: Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Wed Mar 26 19:44:31 EST 2003 athena Chris@10: * [project @ 2003-03-27 00:44:31 by athena] Chris@10: Better place to install hook Chris@10: Chris@10: M ./tests/bench.c -2 +2 Chris@10: Chris@10: Wed Mar 26 19:40:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-27 00:40:28 by stevenj] Chris@10: noted that the user should run make check if they think FFTW has a bug Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -6 +8 Chris@10: Chris@10: Wed Mar 26 17:31:16 EST 2003 athena Chris@10: * [project @ 2003-03-26 22:31:16 by athena] Chris@10: Oops, what am I thinking Chris@10: Chris@10: M ./kernel/planner.c -5 +8 Chris@10: Chris@10: Wed Mar 26 17:23:56 EST 2003 athena Chris@10: * [project @ 2003-03-26 22:23:56 by athena] Chris@10: Grrr.... fixed bug in estimator Chris@10: Chris@10: M ./kernel/planner.c -8 +5 Chris@10: Chris@10: Wed Mar 26 17:16:19 EST 2003 athena Chris@10: * [project @ 2003-03-26 22:16:19 by athena] Chris@10: Oops---the flop count was right. The estimator is broken elsewhere. Chris@10: Chris@10: M ./genfft/c.ml -5 +2 Chris@10: Chris@10: Wed Mar 26 14:28:41 EST 2003 athena Chris@10: * [project @ 2003-03-26 19:28:41 by athena] Chris@10: Fixed SIMD estimator Chris@10: Chris@10: M ./genfft/c.ml -2 +5 Chris@10: Chris@10: Wed Mar 26 07:45:03 EST 2003 athena Chris@10: * [project @ 2003-03-26 12:45:03 by athena] Chris@10: Added twidsq simd codelets Chris@10: Chris@10: A ./dft/simd/q1b.c Chris@10: A ./dft/simd/q1b.h Chris@10: A ./dft/simd/q1f.c Chris@10: A ./dft/simd/q1f.h Chris@10: A ./genfft/gen_twidsq_c.ml Chris@10: M ./dft/simd/Makefile.am -1 +2 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +12 Chris@10: M ./dft/simd/q1b.c +44 Chris@10: M ./dft/simd/q1b.h +25 Chris@10: M ./dft/simd/q1f.c +44 Chris@10: M ./dft/simd/q1f.h +25 Chris@10: M ./dft/simd/t1b.c +1 Chris@10: M ./dft/simd/t1f.c +1 Chris@10: M ./genfft/Makefile.am -8 +14 Chris@10: M ./genfft/gen_twiddle_c.ml -3 +3 Chris@10: M ./genfft/gen_twidsq_c.ml +195 Chris@10: M ./support/Makefile.codelets -1 +2 Chris@10: Chris@10: Tue Mar 25 23:33:03 EST 2003 stevenj Chris@10: * [project @ 2003-03-26 04:33:03 by stevenj] Chris@10: gensrc -> genfft Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Tue Mar 25 23:32:16 EST 2003 stevenj Chris@10: * [project @ 2003-03-26 04:32:16 by stevenj] Chris@10: newline Chris@10: Chris@10: M ./TODO +1 Chris@10: Chris@10: Tue Mar 25 19:17:08 EST 2003 athena Chris@10: * [project @ 2003-03-26 00:17:08 by athena] Chris@10: Noted need to add dif simd codelets Chris@10: Chris@10: M ./TODO +1 Chris@10: Chris@10: Tue Mar 25 13:03:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 18:03:47 by stevenj] Chris@10: noted shift Chris@10: Chris@10: M ./doc/fftw3.texi -3 +2 Chris@10: Chris@10: Tue Mar 25 13:02:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 18:02:47 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -4 +14 Chris@10: Chris@10: Tue Mar 25 12:46:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 17:46:44 by stevenj] Chris@10: need make after bootstrap Chris@10: Chris@10: M ./doc/fftw3.texi -2 +3 Chris@10: Chris@10: Tue Mar 25 12:31:49 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 17:31:49 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Tue Mar 25 12:30:56 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 17:30:56 by stevenj] Chris@10: libtool is also needed Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Tue Mar 25 12:29:52 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 17:29:52 by stevenj] Chris@10: added code generator introduction Chris@10: Chris@10: M ./doc/fftw3.texi -13 +63 Chris@10: Chris@10: Tue Mar 25 11:51:49 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 16:51:49 by stevenj] Chris@10: added support for REDFT/RODFT/DHT direct codelets Chris@10: Chris@10: A ./genfft/gen_r2r.ml Chris@10: A ./rdft/kr2r.c Chris@10: A ./rdft/codelets/r2r/ Chris@10: A ./rdft/codelets/r2r.c Chris@10: A ./rdft/codelets/r2r.h Chris@10: A ./rdft/codelets/r2r/Makefile.am Chris@10: M ./Makefile.am +1 Chris@10: M ./configure.ac +1 Chris@10: M ./genfft/Makefile.am -6 +6 Chris@10: M ./genfft/complex.ml -1 +8 Chris@10: M ./genfft/complex.mli -1 +2 Chris@10: M ./genfft/gen_r2r.ml +240 Chris@10: R ./genfft/gen_trig.ml Chris@10: M ./genfft/trig.ml -60 +47 Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/codelet-rdft.h -2 +2 Chris@10: M ./rdft/codelets/Makefile.am -2 +2 Chris@10: M ./rdft/codelets/r2r/Makefile.am +72 Chris@10: M ./rdft/codelets/r2r.c +38 Chris@10: M ./rdft/codelets/r2r.h +23 Chris@10: M ./rdft/conf.c -1 +2 Chris@10: M ./rdft/direct.c -16 +58 Chris@10: M ./rdft/kr2r.c +28 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: M ./support/Makefile.codelets -1 +2 Chris@10: Chris@10: Tue Mar 25 11:29:29 EST 2003 stevenj Chris@10: * [project @ 2003-03-25 16:29:29 by stevenj] Chris@10: noted ARM bug; thanks to Jay Treacy Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +2 Chris@10: Chris@10: Tue Mar 25 07:55:54 EST 2003 athena Chris@10: * [project @ 2003-03-25 12:55:54 by athena] Chris@10: bugfix from Stefan Chris@10: Chris@10: M ./genfft-k7/vK7Optimization.ml -1 +2 Chris@10: Chris@10: Mon Mar 24 15:59:08 EST 2003 stevenj Chris@10: * [project @ 2003-03-24 20:59:08 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Mar 24 15:58:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-24 20:58:44 by stevenj] Chris@10: caveat Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Mar 24 15:58:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-24 20:58:04 by stevenj] Chris@10: warning about DHT Chris@10: Chris@10: M ./doc/fftw3.texi -1 +8 Chris@10: Chris@10: Mon Mar 24 08:34:14 EST 2003 athena Chris@10: * [project @ 2003-03-24 13:34:14 by athena] Chris@10: Oops Chris@10: Chris@10: M ./dft/k7/codelets/Makefile.am -2 +2 Chris@10: Chris@10: Mon Mar 24 08:13:15 EST 2003 athena Chris@10: * [project @ 2003-03-24 13:13:15 by athena] Chris@10: Regression test for p4fftwgel Chris@10: Chris@10: M ./dft/k7/codelets/Makefile.am -2 +2 Chris@10: M ./tests/Makefile.am +3 Chris@10: M ./tests/check.pl +4 Chris@10: Chris@10: Mon Mar 24 03:09:06 EST 2003 stevenj Chris@10: * [project @ 2003-03-24 08:09:06 by stevenj] Chris@10: make check is faster, old tests are in make bigcheck Chris@10: Chris@10: M ./tests/Makefile.am -1 +6 Chris@10: Chris@10: Sat Mar 22 00:41:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-22 05:41:21 by stevenj] Chris@10: note Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Sat Mar 22 00:40:05 EST 2003 stevenj Chris@10: * [project @ 2003-03-22 05:40:05 by stevenj] Chris@10: whoops, line wrapping Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -2 +1 Chris@10: Chris@10: Fri Mar 21 15:10:00 EST 2003 athena Chris@10: * [project @ 2003-03-21 20:10:00 by athena] Chris@10: Franz-mode codelets even without SIMD. (disabled) Chris@10: Chris@10: M ./dft/codelets/standard/Makefile.am -2 +59 Chris@10: M ./genfft/gen_notw.ml -5 +6 Chris@10: M ./genfft/gen_twiddle.ml -4 +5 Chris@10: Chris@10: Fri Mar 21 09:09:30 EST 2003 athena Chris@10: * [project @ 2003-03-21 14:09:30 by athena] Chris@10: Bug is in netbsd-1.6, not 1.5 Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Fri Mar 21 07:45:48 EST 2003 athena Chris@10: * [project @ 2003-03-21 12:45:48 by athena] Chris@10: const cast, should placate c++ compilers. Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Thu Mar 20 18:49:49 EST 2003 stevenj Chris@10: * [project @ 2003-03-20 23:49:49 by stevenj] Chris@10: added FAQ on why plans are array-specific Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +26 Chris@10: Chris@10: Thu Mar 20 16:12:56 EST 2003 stevenj Chris@10: * [project @ 2003-03-20 21:12:56 by stevenj] Chris@10: comment fix Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -3 +3 Chris@10: Chris@10: Thu Mar 20 16:12:15 EST 2003 stevenj Chris@10: * [project @ 2003-03-20 21:12:15 by stevenj] Chris@10: noted comparison to NR Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -1 +5 Chris@10: Chris@10: Wed Mar 19 20:13:16 EST 2003 stevenj Chris@10: * [project @ 2003-03-20 01:13:16 by stevenj] Chris@10: whoops, C99 complex didn't work if complex is a macro (as it is with glibc); thanks to Keh-Cheng Chu for the bug report Chris@10: Chris@10: M ./api/fftw3.h -5 +4 Chris@10: Chris@10: Wed Mar 19 16:52:54 EST 2003 stevenj Chris@10: * [project @ 2003-03-19 21:52:54 by stevenj] Chris@10: noted in help that --enable-k7 enables 3dnow, and that --enable-3dnow is only a fallback Chris@10: Chris@10: M ./configure.ac -2 +2 Chris@10: Chris@10: Wed Mar 19 15:09:52 EST 2003 athena Chris@10: * [project @ 2003-03-19 20:09:52 by athena] Chris@10: New gcc bug. html.refs was not in repository/distribution. Chris@10: Chris@10: A ./doc/FAQ/html.refs Chris@10: M ./doc/FAQ/Makefile.am -1 +1 Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +13 Chris@10: M ./doc/FAQ/html.refs +6 Chris@10: Chris@10: Wed Mar 19 10:09:16 EST 2003 athena Chris@10: * [project @ 2003-03-19 15:09:16 by athena] Chris@10: Don't write wisdom if you don't have it. Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Tue Mar 18 15:44:41 EST 2003 athena Chris@10: * [project @ 2003-03-18 20:44:41 by athena] Chris@10: Added index entries for DHT. Similarly for DCT, DST Chris@10: Chris@10: M ./doc/fftw3.texi -1 +6 Chris@10: Chris@10: Tue Mar 18 14:50:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-18 19:50:04 by stevenj] Chris@10: execute should not go through C api, for efficiency Chris@10: Chris@10: M ./api/f77api.c +2 Chris@10: M ./api/f77funcs.h -5 +10 Chris@10: Chris@10: Tue Mar 18 06:14:51 EST 2003 athena Chris@10: * [project @ 2003-03-18 11:14:51 by athena] Chris@10: Renamed FFTW_IODIM, FFTW_R2R_KIND Chris@10: Chris@10: M ./api/fftw3.h -6 +6 Chris@10: Chris@10: Tue Mar 18 00:30:17 EST 2003 stevenj Chris@10: * [project @ 2003-03-18 05:30:17 by stevenj] Chris@10: added rfftwnd.eps to dist, so that transfig is not required for people trying to build other formats (e.g. ps); thanks to Brian Gough for the bug report Chris@10: Chris@10: M ./doc/Makefile.am -1 +1 Chris@10: Chris@10: Mon Mar 17 15:17:59 EST 2003 stevenj Chris@10: * [project @ 2003-03-17 20:17:59 by stevenj] Chris@10: pointer to upgrading section from tutorial Chris@10: Chris@10: M ./doc/fftw3.texi -1 +4 Chris@10: Chris@10: Mon Mar 17 14:44:40 EST 2003 stevenj Chris@10: * [project @ 2003-03-17 19:44:40 by stevenj] Chris@10: make print_plan and fprint_plan, so that the former can be more easily called from other languages Chris@10: Chris@10: M ./api/f77funcs.h -2 +2 Chris@10: M ./api/fftw3.h -2 +3 Chris@10: M ./api/print-plan.c -1 +6 Chris@10: M ./doc/fftw3.texi -3 +5 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Mon Mar 17 14:19:10 EST 2003 stevenj Chris@10: * [project @ 2003-03-17 19:19:10 by stevenj] Chris@10: whoops, forgot to change equation image links to .png Chris@10: Chris@10: M ./doc/fftw3.texi -14 +14 Chris@10: Chris@10: Mon Mar 17 04:15:50 EST 2003 athena Chris@10: * [project @ 2003-03-17 09:15:50 by athena] Chris@10: fixed c++ linkage problems Chris@10: Chris@10: M ./api/fftw3.h -4 +4 Chris@10: M ./api/version.c -8 +5 Chris@10: M ./support/Makefile.codelets -1 +2 Chris@10: Chris@10: Mon Mar 17 03:25:17 EST 2003 athena Chris@10: * [project @ 2003-03-17 08:25:17 by athena] Chris@10: Removed ``const'', otherwise c++ link fails Chris@10: Chris@10: M ./api/fftw3.h -4 +4 Chris@10: M ./api/version.c -4 +4 Chris@10: Chris@10: Sun Mar 16 20:24:31 EST 2003 stevenj Chris@10: * [project @ 2003-03-17 01:24:31 by stevenj] Chris@10: fixed C++ annoyances: void* casts, and global variables are static by default(?!?) Chris@10: Chris@10: M ./api/f77api.c -4 +5 Chris@10: M ./api/f77funcs.h -3 +5 Chris@10: M ./api/version.c -4 +4 Chris@10: M ./libbench2/allocate.c -9 +9 Chris@10: M ./libbench2/getopt-utils.c -2 +2 Chris@10: M ./libbench2/problem.c -3 +3 Chris@10: M ./libbench2/speed.c -2 +2 Chris@10: M ./libbench2/timer.c -2 +2 Chris@10: M ./libbench2/verify-r2r.c -3 +3 Chris@10: M ./libbench2/zero.c -9 +9 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: M ./tests/bench.c -37 +66 Chris@10: M ./tests/hook.c -2 +2 Chris@10: M ./tools/fftw-wisdom.c -1 +1 Chris@10: Chris@10: Sat Aug 12 23:06:56 EDT 2006 Unknown tagger Chris@10: tagged fftw-3-0-beta1 Chris@10: Chris@10: Chris@10: Sun Mar 16 15:29:11 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 20:29:11 by stevenj] Chris@10: ranlib bug is in binutils Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -4 +5 Chris@10: Chris@10: Sun Mar 16 15:26:42 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 20:26:42 by stevenj] Chris@10: ranlib Irix bug Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn +5 Chris@10: Chris@10: Sun Mar 16 15:13:35 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 20:13:35 by stevenj] Chris@10: start with random tests Chris@10: Chris@10: M ./tests/check.pl -1 +1 Chris@10: Chris@10: Sun Mar 16 15:00:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 20:00:03 by stevenj] Chris@10: silenced some compiler warnings, eliminated unused variables, and fixed Makefile.am for f77funcs.h Chris@10: Chris@10: M ./api/Makefile.am -5 +1 Chris@10: M ./dft/direct.c -3 +2 Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: M ./libbench2/verify-r2r.c -3 Chris@10: M ./rdft/direct.c -3 +2 Chris@10: M ./rdft/direct2.c -3 +2 Chris@10: M ./threads/Makefile.am -3 +1 Chris@10: Chris@10: Sun Mar 16 14:55:13 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 19:55:13 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Sun Mar 16 14:28:22 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 19:28:22 by stevenj] Chris@10: 3dnow is float Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sun Mar 16 14:27:45 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 19:27:45 by stevenj] Chris@10: fixed k7 docs Chris@10: Chris@10: M ./doc/fftw3.texi -17 +10 Chris@10: Chris@10: Sun Mar 16 14:19:10 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 19:19:10 by stevenj] Chris@10: SGI compilers now support inline Chris@10: Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: Chris@10: Sun Mar 16 14:18:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 19:18:32 by stevenj] Chris@10: cruft Chris@10: Chris@10: M ./kernel/cycle.h -2 +1 Chris@10: Chris@10: Sun Mar 16 14:15:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 19:15:47 by stevenj] Chris@10: texinfo doesn't like commas in nodes Chris@10: Chris@10: M ./doc/fftw3.texi -16 +15 Chris@10: Chris@10: Sun Mar 16 13:52:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:48:30 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog -47 +1389 Chris@10: M ./README +28 Chris@10: Chris@10: Sun Mar 16 13:47:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:47:44 by stevenj] Chris@10: f77funcs.c -> f77funcs.h so that people don't try to compile it Chris@10: Chris@10: A ./api/f77funcs.h Chris@10: A ./threads/f77funcs.h Chris@10: M ./api/f77api.c -4 +4 Chris@10: R ./api/f77funcs.c Chris@10: M ./api/f77funcs.h +366 Chris@10: M ./threads/f77api.c -4 +4 Chris@10: R ./threads/f77funcs.c Chris@10: M ./threads/f77funcs.h +39 Chris@10: Chris@10: Sun Mar 16 13:46:11 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:46:11 by stevenj] Chris@10: minor changes Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -7 +4 Chris@10: Chris@10: Sun Mar 16 13:39:24 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:39:24 by stevenj] Chris@10: updated compiler bug list Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -40 +20 Chris@10: Chris@10: Sun Mar 16 13:39:07 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:39:07 by stevenj] Chris@10: noted how to set CC Chris@10: Chris@10: M ./doc/fftw3.texi -4 +10 Chris@10: Chris@10: Sun Mar 16 13:01:01 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:01:01 by stevenj] Chris@10: TODONE Chris@10: Chris@10: M ./TODO -10 Chris@10: Chris@10: Sun Mar 16 13:00:42 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 18:00:42 by stevenj] Chris@10: yikes, bugfix Chris@10: Chris@10: M ./threads/vrank-geq1-rdft2.c -19 +11 Chris@10: Chris@10: Sun Mar 16 10:26:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-16 15:26:28 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Sun Mar 16 09:24:19 EST 2003 athena Chris@10: * [project @ 2003-03-16 14:24:19 by athena] Chris@10: Report SIMD extensions in version string Chris@10: Chris@10: M ./api/version.c -2 +26 Chris@10: Chris@10: Sat Mar 15 18:56:11 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 23:56:11 by stevenj] Chris@10: more verbose output Chris@10: Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Sat Mar 15 17:41:25 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 22:41:25 by stevenj] Chris@10: a couple of additional non-Unix instructions Chris@10: Chris@10: M ./doc/fftw3.texi -1 +9 Chris@10: Chris@10: Sat Mar 15 17:15:26 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 22:15:26 by stevenj] Chris@10: hyphen Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -1 +1 Chris@10: Chris@10: Sat Mar 15 17:12:29 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 22:12:29 by stevenj] Chris@10: softened Chris@10: Chris@10: M ./doc/FAQ/fftw-faq.bfnn -3 +3 Chris@10: Chris@10: Sat Mar 15 17:09:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 22:09:44 by stevenj] Chris@10: added FAQ, used PNGs Chris@10: Chris@10: A ./doc/equation-dft.png Chris@10: A ./doc/equation-dht.png Chris@10: A ./doc/equation-idft.png Chris@10: A ./doc/equation-redft00.png Chris@10: A ./doc/equation-redft01.png Chris@10: A ./doc/equation-redft10.png Chris@10: A ./doc/equation-redft11.png Chris@10: A ./doc/equation-rodft00.png Chris@10: A ./doc/equation-rodft01.png Chris@10: A ./doc/equation-rodft10.png Chris@10: A ./doc/FAQ/ Chris@10: A ./doc/FAQ/Makefile.am Chris@10: A ./doc/FAQ/bfnnconv.pl Chris@10: A ./doc/FAQ/fftw-faq.bfnn Chris@10: A ./doc/FAQ/m-ascii.pl Chris@10: A ./doc/FAQ/m-html.pl Chris@10: A ./doc/FAQ/m-info.pl Chris@10: A ./doc/FAQ/m-lout.pl Chris@10: A ./doc/FAQ/m-post.pl Chris@10: A ./doc/equation-rodft11.png Chris@10: M ./configure.ac +1 Chris@10: M ./doc/FAQ/Makefile.am +14 Chris@10: M ./doc/FAQ/bfnnconv.pl +298 Chris@10: M ./doc/FAQ/fftw-faq.bfnn +492 Chris@10: M ./doc/FAQ/m-ascii.pl +189 Chris@10: M ./doc/FAQ/m-html.pl +337 Chris@10: M ./doc/FAQ/m-info.pl +226 Chris@10: M ./doc/FAQ/m-lout.pl +242 Chris@10: M ./doc/FAQ/m-post.pl +189 Chris@10: M ./doc/Makefile.am -4 +6 Chris@10: R ./doc/equation-dft.gif Chris@10: M ./doc/equation-dft.png Chris@10: R ./doc/equation-dht.gif Chris@10: M ./doc/equation-dht.png Chris@10: R ./doc/equation-idft.gif Chris@10: M ./doc/equation-idft.png Chris@10: R ./doc/equation-redft00.gif Chris@10: M ./doc/equation-redft00.png Chris@10: R ./doc/equation-redft01.gif Chris@10: M ./doc/equation-redft01.png Chris@10: R ./doc/equation-redft10.gif Chris@10: M ./doc/equation-redft10.png Chris@10: R ./doc/equation-redft11.gif Chris@10: M ./doc/equation-redft11.png Chris@10: R ./doc/equation-rodft00.gif Chris@10: M ./doc/equation-rodft00.png Chris@10: R ./doc/equation-rodft01.gif Chris@10: M ./doc/equation-rodft01.png Chris@10: R ./doc/equation-rodft10.gif Chris@10: M ./doc/equation-rodft10.png Chris@10: R ./doc/equation-rodft11.gif Chris@10: M ./doc/equation-rodft11.png Chris@10: Chris@10: Sat Mar 15 15:29:43 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 20:29:42 by stevenj] Chris@10: great copyright update Chris@10: Chris@10: M ./COPYRIGHT -2 +2 Chris@10: M ./TODO -6 Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/configure.c -2 +2 Chris@10: M ./api/execute-dft-c2r.c -2 +2 Chris@10: M ./api/execute-dft-r2c.c -2 +2 Chris@10: M ./api/execute-dft.c -2 +2 Chris@10: M ./api/execute-r2r.c -2 +2 Chris@10: M ./api/execute.c -2 +2 Chris@10: M ./api/export-wisdom-to-file.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -2 +2 Chris@10: M ./api/export-wisdom.c -2 +2 Chris@10: M ./api/extract-reim.c -2 +2 Chris@10: M ./api/f77api.c -2 +2 Chris@10: M ./api/f77funcs.c -2 +2 Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./api/flops.c -2 +2 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: M ./api/import-wisdom-from-file.c -2 +2 Chris@10: M ./api/import-wisdom-from-string.c -2 +2 Chris@10: M ./api/import-wisdom.c -2 +2 Chris@10: M ./api/map-r2r-kind.c -2 +2 Chris@10: M ./api/mapflags.c -2 +2 Chris@10: M ./api/mkprinter-file.c -2 +2 Chris@10: M ./api/mktensor-iodims.c -2 +2 Chris@10: M ./api/mktensor-rowmajor.c -2 +2 Chris@10: M ./api/plan-dft-1d.c -2 +2 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r.c -2 +2 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -2 +2 Chris@10: M ./api/plan-dft-r2c.c -2 +2 Chris@10: M ./api/plan-dft.c -2 +2 Chris@10: M ./api/plan-guru-dft-c2r.c -2 +2 Chris@10: M ./api/plan-guru-dft-r2c.c -2 +2 Chris@10: M ./api/plan-guru-dft.c -2 +2 Chris@10: M ./api/plan-guru-r2r.c -2 +2 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: M ./api/plan-many-dft.c -2 +2 Chris@10: M ./api/plan-many-r2r.c -2 +2 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -2 +2 Chris@10: M ./api/plan-r2r-3d.c -2 +2 Chris@10: M ./api/plan-r2r.c -2 +2 Chris@10: M ./api/print-plan.c -2 +2 Chris@10: M ./api/rdft2-pad.c -2 +2 Chris@10: M ./api/the-planner.c -2 +2 Chris@10: M ./api/version.c -3 +3 Chris@10: M ./api/x77.h -2 +2 Chris@10: M ./dft/buffered.c -3 +3 Chris@10: M ./dft/codelet-dft.h -3 +3 Chris@10: M ./dft/codelets/n.c -2 +2 Chris@10: M ./dft/codelets/n.h -2 +2 Chris@10: M ./dft/codelets/t.c -2 +2 Chris@10: M ./dft/codelets/t.h -2 +2 Chris@10: M ./dft/conf.c -3 +3 Chris@10: M ./dft/ct-dif.c -3 +3 Chris@10: M ./dft/ct-dit.c -3 +3 Chris@10: M ./dft/ct-ditbuf.c -3 +3 Chris@10: M ./dft/ct-ditf.c -3 +3 Chris@10: M ./dft/ct.c -3 +3 Chris@10: M ./dft/ct.h -3 +3 Chris@10: M ./dft/dft.h -3 +3 Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect.c -3 +3 Chris@10: M ./dft/k7/k7.c -3 +3 Chris@10: M ./dft/kdft-dif.c -3 +3 Chris@10: M ./dft/kdft-difsq.c -3 +3 Chris@10: M ./dft/kdft-dit.c -3 +3 Chris@10: M ./dft/kdft.c -3 +3 Chris@10: M ./dft/nop.c -3 +3 Chris@10: M ./dft/plan.c -3 +3 Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./dft/rader-omega.c -2 +2 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./dft/rank-geq2.c -3 +3 Chris@10: M ./dft/rank0.c -3 +3 Chris@10: M ./dft/simd/n1b.c -2 +2 Chris@10: M ./dft/simd/n1b.h -2 +2 Chris@10: M ./dft/simd/n1f.c -2 +2 Chris@10: M ./dft/simd/n1f.h -2 +2 Chris@10: M ./dft/simd/t1b.c -2 +2 Chris@10: M ./dft/simd/t1b.h -2 +2 Chris@10: M ./dft/simd/t1f.c -2 +2 Chris@10: M ./dft/simd/t1f.h -2 +2 Chris@10: M ./dft/solve.c -3 +3 Chris@10: M ./dft/vrank-geq1.c -3 +3 Chris@10: M ./dft/vrank2-transpose.c -3 +3 Chris@10: M ./dft/vrank3-transpose.c -3 +3 Chris@10: M ./dft/zero.c -3 +3 Chris@10: M ./doc/f77_wisdom.f -2 +2 Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: M ./genfft/algsimp.ml -4 +4 Chris@10: M ./genfft/algsimp.mli -3 +3 Chris@10: M ./genfft/annotate.ml -4 +4 Chris@10: M ./genfft/annotate.mli -3 +3 Chris@10: M ./genfft/assoctable.ml -3 +3 Chris@10: M ./genfft/assoctable.mli -3 +3 Chris@10: M ./genfft/c.ml -3 +3 Chris@10: M ./genfft/c.mli -3 +3 Chris@10: M ./genfft/complex.ml -3 +3 Chris@10: M ./genfft/complex.mli -3 +3 Chris@10: M ./genfft/conv.ml -2 +2 Chris@10: M ./genfft/conv.mli -3 +3 Chris@10: M ./genfft/dag.ml -3 +3 Chris@10: M ./genfft/dag.mli -3 +3 Chris@10: M ./genfft/expr.ml -3 +3 Chris@10: M ./genfft/expr.mli -3 +3 Chris@10: M ./genfft/fft.ml -4 +4 Chris@10: M ./genfft/fft.mli -3 +3 Chris@10: M ./genfft/gen_athnotw.ml -4 +4 Chris@10: M ./genfft/gen_athtw.ml -4 +4 Chris@10: M ./genfft/gen_conv.ml -4 +4 Chris@10: M ./genfft/gen_hc2hc.ml -4 +4 Chris@10: M ./genfft/gen_hc2r.ml -4 +4 Chris@10: M ./genfft/gen_notw.ml -4 +4 Chris@10: M ./genfft/gen_notw_c.ml -4 +4 Chris@10: M ./genfft/gen_r2hc.ml -4 +4 Chris@10: M ./genfft/gen_trig.ml -4 +4 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twiddle_c.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./genfft/genutil.ml -3 +3 Chris@10: M ./genfft/littlesimp.ml -3 +3 Chris@10: M ./genfft/littlesimp.mli -3 +3 Chris@10: M ./genfft/magic.ml -3 +3 Chris@10: M ./genfft/monads.ml -3 +3 Chris@10: M ./genfft/number.ml -3 +3 Chris@10: M ./genfft/number.mli -3 +3 Chris@10: M ./genfft/oracle.ml -3 +3 Chris@10: M ./genfft/oracle.mli -3 +3 Chris@10: M ./genfft/schedule.ml -3 +3 Chris@10: M ./genfft/schedule.mli -3 +3 Chris@10: M ./genfft/simd.ml -3 +3 Chris@10: M ./genfft/simd.mli -3 +3 Chris@10: M ./genfft/simdmagic.ml -3 +3 Chris@10: M ./genfft/to_alist.ml -3 +3 Chris@10: M ./genfft/to_alist.mli -3 +3 Chris@10: M ./genfft/trig.ml -3 +3 Chris@10: M ./genfft/trig.mli -3 +3 Chris@10: M ./genfft/twiddle.ml -3 +3 Chris@10: M ./genfft/twiddle.mli -3 +3 Chris@10: M ./genfft/unique.ml -3 +3 Chris@10: M ./genfft/unique.mli -3 +3 Chris@10: M ./genfft/util.ml -3 +3 Chris@10: M ./genfft/util.mli -3 +3 Chris@10: M ./genfft/variable.ml -3 +3 Chris@10: M ./genfft/variable.mli -3 +3 Chris@10: M ./genfft-k7/algsimp.ml -4 +4 Chris@10: M ./genfft-k7/algsimp.mli -3 +3 Chris@10: M ./genfft-k7/assoctable.ml -3 +3 Chris@10: M ./genfft-k7/assoctable.mli -3 +3 Chris@10: M ./genfft-k7/expr.ml -3 +3 Chris@10: M ./genfft-k7/expr.mli -3 +3 Chris@10: M ./genfft-k7/fft.ml -4 +4 Chris@10: M ./genfft-k7/littlesimp.ml -3 +3 Chris@10: M ./genfft-k7/littlesimp.mli -3 +3 Chris@10: M ./genfft-k7/monads.ml -3 +3 Chris@10: M ./genfft-k7/number.ml -3 +3 Chris@10: M ./genfft-k7/number.mli -3 +3 Chris@10: M ./genfft-k7/oracle.ml -3 +3 Chris@10: M ./genfft-k7/oracle.mli -3 +3 Chris@10: M ./genfft-k7/to_alist.ml -3 +3 Chris@10: M ./genfft-k7/to_alist.mli -3 +3 Chris@10: M ./genfft-k7/twiddle.ml -3 +3 Chris@10: M ./genfft-k7/twiddle.mli -3 +3 Chris@10: M ./kernel/align.c -3 +3 Chris@10: M ./kernel/alloc.c -3 +3 Chris@10: M ./kernel/assert.c -3 +3 Chris@10: M ./kernel/awake.c -3 +3 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/ct.c -2 +2 Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: M ./kernel/debug.c -3 +3 Chris@10: M ./kernel/hash.c -2 +2 Chris@10: M ./kernel/iabs.c -3 +3 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./kernel/md5-1.c -2 +2 Chris@10: M ./kernel/md5.c -2 +2 Chris@10: M ./kernel/minmax.c -3 +3 Chris@10: M ./kernel/ops.c -3 +3 Chris@10: M ./kernel/pickdim.c -3 +3 Chris@10: M ./kernel/plan.c -3 +3 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./kernel/primes.c -3 +3 Chris@10: M ./kernel/print.c -3 +3 Chris@10: M ./kernel/problem.c -3 +3 Chris@10: M ./kernel/rader.c -2 +2 Chris@10: M ./kernel/scan.c -3 +3 Chris@10: M ./kernel/solver.c -3 +3 Chris@10: M ./kernel/solvtab.c -3 +3 Chris@10: M ./kernel/square.c -3 +3 Chris@10: M ./kernel/stride.c -3 +3 Chris@10: M ./kernel/tensor.c -3 +3 Chris@10: M ./kernel/tensor1.c -3 +3 Chris@10: M ./kernel/tensor2.c -3 +3 Chris@10: M ./kernel/tensor4.c -3 +3 Chris@10: M ./kernel/tensor5.c -3 +3 Chris@10: M ./kernel/tensor7.c -3 +3 Chris@10: M ./kernel/tensor8.c -3 +3 Chris@10: M ./kernel/tensor9.c -3 +3 Chris@10: M ./kernel/timer.c -3 +3 Chris@10: M ./kernel/trig.c -3 +3 Chris@10: M ./kernel/trig1.c -3 +3 Chris@10: M ./kernel/twiddle.c -3 +3 Chris@10: M ./libbench/bench-main.c -2 +2 Chris@10: M ./libbench/bench-user.h -2 +2 Chris@10: M ./libbench/bench.h -2 +2 Chris@10: M ./libbench/can-do.c -2 +2 Chris@10: M ./libbench/getopt-utils.c -2 +2 Chris@10: M ./libbench/info.c -2 +2 Chris@10: M ./libbench/main.c -2 +2 Chris@10: M ./libbench/prime.c -2 +2 Chris@10: M ./libbench/problem.c -2 +2 Chris@10: M ./libbench/report.c -2 +2 Chris@10: M ./libbench/speed.c -2 +2 Chris@10: M ./libbench/timer.c -2 +2 Chris@10: M ./libbench/util.c -1 +1 Chris@10: M ./libbench/verify.c -2 +2 Chris@10: M ./libbench/zero.c -2 +2 Chris@10: M ./libbench2/aligned-main.c -2 +2 Chris@10: M ./libbench2/bench-main.c -2 +2 Chris@10: M ./libbench2/bench-user.h -2 +2 Chris@10: M ./libbench2/bench.h -2 +2 Chris@10: M ./libbench2/can-do.c -2 +2 Chris@10: M ./libbench2/dotens2.c -3 +3 Chris@10: M ./libbench2/getopt-utils.c -2 +2 Chris@10: M ./libbench2/info.c -2 +2 Chris@10: M ./libbench2/main.c -2 +2 Chris@10: M ./libbench2/problem.c -2 +2 Chris@10: M ./libbench2/report.c -2 +2 Chris@10: M ./libbench2/speed.c -2 +2 Chris@10: M ./libbench2/tensor.c -2 +2 Chris@10: M ./libbench2/timer.c -2 +2 Chris@10: M ./libbench2/useropt.c -2 +2 Chris@10: M ./libbench2/util.c -1 +1 Chris@10: M ./libbench2/verify-dft.c -3 +3 Chris@10: M ./libbench2/verify-lib.c -3 +3 Chris@10: M ./libbench2/verify-r2r.c -2 +2 Chris@10: M ./libbench2/verify-rdft2.c -3 +3 Chris@10: M ./libbench2/verify.c -2 +2 Chris@10: M ./libbench2/verify.h -2 +2 Chris@10: M ./libbench2/zero.c -2 +2 Chris@10: M ./rdft/buffered.c -3 +3 Chris@10: M ./rdft/buffered2.c -3 +3 Chris@10: M ./rdft/codelet-rdft.h -3 +3 Chris@10: M ./rdft/codelets/hb.h -2 +2 Chris@10: M ./rdft/codelets/hc2r.c -2 +2 Chris@10: M ./rdft/codelets/hc2r.h -2 +2 Chris@10: M ./rdft/codelets/hc2rIII.h -2 +2 Chris@10: M ./rdft/codelets/hf.h -2 +2 Chris@10: M ./rdft/codelets/hfb.c -2 +2 Chris@10: M ./rdft/codelets/r2hc.c -2 +2 Chris@10: M ./rdft/codelets/r2hc.h -2 +2 Chris@10: M ./rdft/codelets/r2hcII.h -2 +2 Chris@10: M ./rdft/conf.c -3 +3 Chris@10: M ./rdft/dft-r2hc.c -3 +3 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./rdft/dht-rader.c -2 +2 Chris@10: M ./rdft/direct.c -3 +3 Chris@10: M ./rdft/direct2.c -3 +3 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-buf.c -3 +3 Chris@10: M ./rdft/hc2hc-dif.c -3 +3 Chris@10: M ./rdft/hc2hc-dit.c -3 +3 Chris@10: M ./rdft/hc2hc.c -3 +3 Chris@10: M ./rdft/hc2hc.h -3 +3 Chris@10: M ./rdft/indirect.c -3 +3 Chris@10: M ./rdft/khc2hc-dif.c -3 +3 Chris@10: M ./rdft/khc2hc-dit.c -3 +3 Chris@10: M ./rdft/khc2r.c -3 +3 Chris@10: M ./rdft/kr2hc.c -3 +3 Chris@10: M ./rdft/nop.c -3 +3 Chris@10: M ./rdft/nop2.c -3 +3 Chris@10: M ./rdft/plan.c -3 +3 Chris@10: M ./rdft/plan2.c -3 +3 Chris@10: M ./rdft/problem.c -3 +3 Chris@10: M ./rdft/problem2.c -3 +3 Chris@10: M ./rdft/rader-hc2hc.c -2 +2 Chris@10: M ./rdft/rank-geq2-rdft2.c -3 +3 Chris@10: M ./rdft/rank-geq2.c -3 +3 Chris@10: M ./rdft/rank0-rdft2.c -3 +3 Chris@10: M ./rdft/rank0.c -3 +3 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft.h -3 +3 Chris@10: M ./rdft/rdft2-inplace-strides.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -3 +3 Chris@10: M ./rdft/rdft2-tensor-max-index.c -3 +3 Chris@10: M ./rdft/solve.c -3 +3 Chris@10: M ./rdft/solve2.c -3 +3 Chris@10: M ./rdft/vrank-geq1-rdft2.c -3 +3 Chris@10: M ./rdft/vrank-geq1.c -3 +3 Chris@10: M ./rdft/vrank2-transpose.c -3 +3 Chris@10: M ./rdft/vrank3-transpose.c -3 +3 Chris@10: M ./reodft/conf.c -3 +3 Chris@10: M ./reodft/redft00e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft.h -3 +3 Chris@10: M ./reodft/reodft010e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -3 +3 Chris@10: M ./reodft/reodft11e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft11e-radix2.c -3 +3 Chris@10: M ./reodft/rodft00e-r2hc.c -3 +3 Chris@10: M ./simd/3dnow.c -3 +3 Chris@10: M ./simd/altivec.c -3 +3 Chris@10: M ./simd/simd-3dnow.h -2 +2 Chris@10: M ./simd/simd-altivec.h -2 +2 Chris@10: M ./simd/simd-sse.h -2 +2 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: M ./simd/simd.h -2 +2 Chris@10: M ./simd/sse-aux.c -3 +3 Chris@10: M ./simd/sse.c -3 +3 Chris@10: M ./simd/sse2-aux.c -3 +3 Chris@10: M ./simd/sse2.c -3 +3 Chris@10: M ./threads/api.c -2 +2 Chris@10: M ./threads/conf.c -3 +3 Chris@10: M ./threads/ct-dit.c -3 +3 Chris@10: M ./threads/dft-vrank-geq1.c -3 +3 Chris@10: M ./threads/f77api.c -2 +2 Chris@10: M ./threads/f77funcs.c -2 +2 Chris@10: M ./threads/hc2hc-dif.c -3 +3 Chris@10: M ./threads/hc2hc-dit.c -3 +3 Chris@10: M ./threads/rdft-vrank-geq1.c -3 +3 Chris@10: M ./threads/threads.c -2 +2 Chris@10: M ./threads/threads.h -3 +3 Chris@10: M ./threads/vrank-geq1-rdft2.c -3 +3 Chris@10: M ./tools/fftw-wisdom-to-conf.1 -4 +4 Chris@10: M ./tools/fftw-wisdom-to-conf.in -2 +2 Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: M ./tools/fftw_wisdom.1.in -4 +4 Chris@10: Chris@10: Sat Mar 15 15:14:02 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 20:14:02 by stevenj] Chris@10: threads in make check Chris@10: Chris@10: M ./TODO -4 Chris@10: M ./tests/Makefile.am -1 +5 Chris@10: M ./tests/check.pl +3 Chris@10: Chris@10: Sat Mar 15 15:11:24 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 20:11:24 by stevenj] Chris@10: fixed const warnings Chris@10: Chris@10: M ./threads/ct-dit.c -2 +2 Chris@10: M ./threads/hc2hc-dif.c -2 +2 Chris@10: M ./threads/hc2hc-dit.c -2 +2 Chris@10: Chris@10: Sat Mar 15 15:08:25 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 20:08:25 by stevenj] Chris@10: make sure spawn_loop size > 1 (it has to be at least > 0 lest we crash, but > 1 is an optimization) Chris@10: Chris@10: M ./threads/ct-dit.c -1 +2 Chris@10: M ./threads/hc2hc-dif.c -1 +2 Chris@10: M ./threads/hc2hc-dit.c -1 +2 Chris@10: Chris@10: Sat Mar 15 14:00:17 EST 2003 athena Chris@10: * [project @ 2003-03-15 19:00:17 by athena] Chris@10: hpux seems to want machine/sys/inline.h as opposed to Chris@10: machine/inline.h. Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Sat Mar 15 13:36:56 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 18:36:56 by stevenj] Chris@10: Sourceforge is really SourceForge.net, and is run by VA Chris@10: Chris@10: M ./doc/fftw3.texi -3 +4 Chris@10: Chris@10: Sat Mar 15 13:34:05 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 18:33:07 by stevenj] Chris@10: comma Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 15 13:31:42 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 18:31:42 by stevenj] Chris@10: fixed AMD company name Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 15 13:29:41 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 18:29:41 by stevenj] Chris@10: minor changes Chris@10: Chris@10: M ./doc/fftw3.texi -19 +19 Chris@10: Chris@10: Sat Mar 15 13:13:55 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 18:13:55 by stevenj] Chris@10: more emitter->read_char renaming Chris@10: Chris@10: M ./api/f77api.c -5 +5 Chris@10: M ./api/f77funcs.c -4 +4 Chris@10: Chris@10: Sat Mar 15 13:08:45 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 18:08:45 by stevenj] Chris@10: more wisdom docs, noted wisdom utilities Chris@10: Chris@10: M ./doc/fftw3.texi -35 +95 Chris@10: Chris@10: Sat Mar 15 11:41:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 16:41:32 by stevenj] Chris@10: compound adjectives are hyphenated Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 15 11:40:30 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 16:40:30 by stevenj] Chris@10: fftw does support another type of packed array via r2r Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 15 11:29:12 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 16:29:12 by stevenj] Chris@10: write_char/read_char for export/import functions Chris@10: Chris@10: M ./api/export-wisdom.c -4 +4 Chris@10: M ./api/f77api.c -5 +5 Chris@10: M ./api/f77funcs.c -4 +4 Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./api/import-wisdom.c -4 +4 Chris@10: M ./doc/f77_wisdom.f -6 +6 Chris@10: M ./doc/fftw3.texi -5 +5 Chris@10: Chris@10: Sat Mar 15 11:19:19 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 16:19:19 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./threads/threads.c -4 +8 Chris@10: Chris@10: Sat Mar 15 10:08:26 EST 2003 athena Chris@10: * [project @ 2003-03-15 15:08:26 by athena] Chris@10: Enabled randomized-cse Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Sat Mar 15 09:47:49 EST 2003 athena Chris@10: * [project @ 2003-03-15 14:47:49 by athena] Chris@10: Changed to 3.0-beta1 Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Mar 15 09:07:31 EST 2003 athena Chris@10: * [project @ 2003-03-15 14:07:31 by athena] Chris@10: First complete draft Chris@10: Chris@10: M ./doc/fftw3.texi -18 +131 Chris@10: Chris@10: Sat Mar 15 08:37:52 EST 2003 athena Chris@10: * [project @ 2003-03-15 13:37:52 by athena] Chris@10: EMITTER is a misnomer Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/import-wisdom.c -8 +6 Chris@10: Chris@10: Sat Mar 15 05:50:50 EST 2003 athena Chris@10: * [project @ 2003-03-15 10:50:50 by athena] Chris@10: Revision, wisdom tutorial, acks. Chris@10: Chris@10: M ./doc/fftw3.texi -44 +219 Chris@10: Chris@10: Fri Mar 14 22:59:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 03:59:04 by stevenj] Chris@10: noted OpenMP Chris@10: Chris@10: M ./NEWS +2 Chris@10: Chris@10: Fri Mar 14 22:38:49 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 03:38:49 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Fri Mar 14 22:38:30 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 03:38:30 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./threads/threads.c -2 +2 Chris@10: Chris@10: Fri Mar 14 22:38:05 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 03:38:05 by stevenj] Chris@10: reformatting Chris@10: Chris@10: M ./threads/threads.c -3 +1 Chris@10: Chris@10: Fri Mar 14 22:26:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 03:26:28 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./threads/threads.c -2 +2 Chris@10: Chris@10: Fri Mar 14 22:11:23 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 03:11:23 by stevenj] Chris@10: some threads fixes, and added experimental semaphore (pre-thread-spawning) and Linux spinlock support Chris@10: Chris@10: M ./tests/bench.c -1 +10 Chris@10: M ./threads/api.c +1 Chris@10: M ./threads/threads.c -8 +159 Chris@10: M ./threads/threads.h -1 +2 Chris@10: Chris@10: Fri Mar 14 20:50:46 EST 2003 stevenj Chris@10: * [project @ 2003-03-15 01:50:46 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./threads/f77funcs.c -2 +2 Chris@10: Chris@10: Fri Mar 14 18:23:03 EST 2003 stevenj Chris@10: * [project @ 2003-03-14 23:23:03 by stevenj] Chris@10: added note that FFTW_PATIENT will disable threads if they are not beneficial Chris@10: Chris@10: M ./doc/fftw3.texi -2 +5 Chris@10: Chris@10: Fri Mar 14 18:20:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-14 23:20:44 by stevenj] Chris@10: made fftw_cleanup* more restrictive, in that we don't want to Chris@10: guarantee that previously created plans will still work (they won't, Chris@10: in the case of threaded plans and fftw_cleanup_threads), and there is Chris@10: no reason to provide such a guarantee anyway. Chris@10: Chris@10: M ./doc/fftw3.texi -6 +14 Chris@10: Chris@10: Fri Mar 14 17:23:13 EST 2003 athena Chris@10: * [project @ 2003-03-14 22:23:13 by athena] Chris@10: Moved version.c from kernel/ into api/ Chris@10: Chris@10: A ./api/version.c Chris@10: M ./api/Makefile.am -1 +2 Chris@10: M ./api/version.c +28 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -4 +1 Chris@10: R ./kernel/version.c Chris@10: Chris@10: Fri Mar 14 17:19:50 EST 2003 athena Chris@10: * [project @ 2003-03-14 22:19:50 by athena] Chris@10: icc-7.0 requires -openmp Chris@10: Chris@10: M ./configure.ac +5 Chris@10: Chris@10: Fri Mar 14 14:47:52 EST 2003 athena Chris@10: * [project @ 2003-03-14 19:47:52 by athena] Chris@10: Ensure that one can do make dist given the distribution Chris@10: Chris@10: M ./doc/Makefile.am -5 +5 Chris@10: Chris@10: Fri Mar 14 14:38:11 EST 2003 athena Chris@10: * [project @ 2003-03-14 19:38:11 by athena] Chris@10: Dist fftw3.pdf, not fftw.pdf Chris@10: Chris@10: M ./doc/Makefile.am -1 +2 Chris@10: Chris@10: Fri Mar 14 14:36:25 EST 2003 athena Chris@10: * [project @ 2003-03-14 19:36:25 by athena] Chris@10: Support -onthreads=%d Chris@10: Chris@10: M ./tests/bench.c -4 +8 Chris@10: Chris@10: Fri Mar 14 14:34:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-14 19:34:21 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Fri Mar 14 14:33:27 EST 2003 stevenj Chris@10: * [project @ 2003-03-14 19:33:27 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./threads/Makefile.am -1 +3 Chris@10: Chris@10: Fri Mar 14 12:32:18 EST 2003 stevenj Chris@10: * [project @ 2003-03-14 17:32:18 by stevenj] Chris@10: fftw_real is gone Chris@10: Chris@10: M ./doc/rfftwnd.fig -1 +1 Chris@10: Chris@10: Fri Mar 14 12:26:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-14 17:26:04 by stevenj] Chris@10: typos Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Fri Mar 14 06:21:43 EST 2003 athena Chris@10: * [project @ 2003-03-14 11:21:43 by athena] Chris@10: More BENCH_DOC strings Chris@10: Chris@10: M ./api/fftw3.h -3 +8 Chris@10: M ./tests/bench.c -4 +11 Chris@10: Chris@10: Fri Mar 14 05:58:53 EST 2003 athena Chris@10: * [project @ 2003-03-14 10:58:53 by athena] Chris@10: Fixed xref's Chris@10: Chris@10: M ./doc/fftw3.texi -15 +15 Chris@10: Chris@10: Fri Mar 14 05:38:26 EST 2003 athena Chris@10: * [project @ 2003-03-14 10:38:26 by athena] Chris@10: Revised manual (esp. intro and tutorial), fixed texinfo hackery Chris@10: for figures. Chris@10: Chris@10: M ./doc/Makefile.am -8 +13 Chris@10: M ./doc/fftw3.texi -165 +280 Chris@10: R ./doc/rfftwnd.gif Chris@10: Chris@10: Wed Mar 12 02:42:33 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 07:42:33 by stevenj] Chris@10: redirect users from guru execute to advanced interface, if possible Chris@10: Chris@10: M ./doc/fftw3.texi -1 +5 Chris@10: Chris@10: Wed Mar 12 02:35:22 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 07:35:22 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Wed Mar 12 02:28:51 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 07:28:51 by stevenj] Chris@10: use correct heading level Chris@10: Chris@10: M ./doc/fftw3.texi -7 +7 Chris@10: Chris@10: Wed Mar 12 02:24:37 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 07:24:37 by stevenj] Chris@10: html generation Chris@10: Chris@10: M ./doc/Makefile.am +15 Chris@10: M ./doc/fftw3.texi -91 +90 Chris@10: Chris@10: Wed Mar 12 01:44:00 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 06:44:00 by stevenj] Chris@10: added equation GIFs Chris@10: Chris@10: A ./doc/equation-dft.gif Chris@10: A ./doc/equation-dht.gif Chris@10: A ./doc/equation-idft.gif Chris@10: A ./doc/equation-redft00.gif Chris@10: A ./doc/equation-redft01.gif Chris@10: A ./doc/equation-redft10.gif Chris@10: A ./doc/equation-redft11.gif Chris@10: A ./doc/equation-rodft00.gif Chris@10: A ./doc/equation-rodft01.gif Chris@10: A ./doc/equation-rodft10.gif Chris@10: A ./doc/equation-rodft11.gif Chris@10: M ./doc/equation-dft.gif Chris@10: M ./doc/equation-dht.gif Chris@10: M ./doc/equation-idft.gif Chris@10: M ./doc/equation-redft00.gif Chris@10: M ./doc/equation-redft01.gif Chris@10: M ./doc/equation-redft10.gif Chris@10: M ./doc/equation-redft11.gif Chris@10: M ./doc/equation-rodft00.gif Chris@10: M ./doc/equation-rodft01.gif Chris@10: M ./doc/equation-rodft10.gif Chris@10: M ./doc/equation-rodft11.gif Chris@10: Chris@10: Wed Mar 12 01:43:27 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 06:43:27 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./doc/fftw3.texi -18 +18 Chris@10: Chris@10: Wed Mar 12 01:26:46 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 06:26:46 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./doc/fftw3.texi -14 +14 Chris@10: Chris@10: Wed Mar 12 01:25:12 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 06:25:12 by stevenj] Chris@10: added multi-dimensional transform definitions Chris@10: Chris@10: M ./doc/fftw3.texi -3 +124 Chris@10: Chris@10: Wed Mar 12 00:14:03 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 05:14:03 by stevenj] Chris@10: slight changes Chris@10: Chris@10: M ./doc/fftw3.texi -3 +5 Chris@10: Chris@10: Wed Mar 12 00:06:34 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 05:06:34 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Tue Mar 11 23:50:43 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 04:50:43 by stevenj] Chris@10: added 1d version of What FFTW Really Computes Chris@10: Chris@10: M ./doc/fftw3.texi -1 +432 Chris@10: Chris@10: Tue Mar 11 21:17:54 EST 2003 stevenj Chris@10: * [project @ 2003-03-12 02:17:54 by stevenj] Chris@10: note in upgrading section about FFTW_PATIENT Chris@10: Chris@10: M ./doc/fftw3.texi -1 +7 Chris@10: Chris@10: Tue Mar 11 15:18:39 EST 2003 stevenj Chris@10: * [project @ 2003-03-11 20:18:39 by stevenj] Chris@10: added cycle-counter section Chris@10: Chris@10: M ./doc/fftw3.texi -1 +32 Chris@10: Chris@10: Tue Mar 11 14:53:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-11 19:53:44 by stevenj] Chris@10: more ideas Chris@10: Chris@10: M ./TODO +7 Chris@10: Chris@10: Mon Mar 10 17:41:35 EST 2003 stevenj Chris@10: * [project @ 2003-03-10 22:41:35 by stevenj] Chris@10: noted that indirect should probably be merged with rank-geq2, to make a rank-split solver Chris@10: Chris@10: M ./dft/indirect.c -1 +4 Chris@10: M ./rdft/indirect.c -1 +4 Chris@10: Chris@10: Fri Mar 7 03:01:52 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 08:01:52 by stevenj] Chris@10: added non-Unix installation instructions Chris@10: Chris@10: M ./doc/fftw3.texi -1 +38 Chris@10: Chris@10: Fri Mar 7 02:30:59 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:30:59 by stevenj] Chris@10: also talk about stack alignment with SSE/SSE2 Chris@10: Chris@10: M ./doc/fftw3.texi -1 +7 Chris@10: Chris@10: Fri Mar 7 02:24:07 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:24:07 by stevenj] Chris@10: made warning more dire Chris@10: Chris@10: M ./doc/fftw3.texi -3 +4 Chris@10: Chris@10: Fri Mar 7 02:13:25 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:13:25 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Mar 7 02:09:55 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:09:55 by stevenj] Chris@10: number Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Mar 7 02:09:08 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:09:08 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Fri Mar 7 02:08:01 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:08:01 by stevenj] Chris@10: minor Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Mar 7 02:04:45 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 07:04:45 by stevenj] Chris@10: minor fix Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Fri Mar 7 01:58:15 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 06:58:15 by stevenj] Chris@10: cross-ref Chris@10: Chris@10: M ./doc/fftw3.texi -2 +3 Chris@10: Chris@10: Fri Mar 7 01:57:31 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 06:57:31 by stevenj] Chris@10: minor Chris@10: Chris@10: M ./doc/fftw3.texi -7 +8 Chris@10: Chris@10: Fri Mar 7 01:53:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 06:53:28 by stevenj] Chris@10: more installation manual Chris@10: Chris@10: M ./doc/fftw3.texi -7 +151 Chris@10: Chris@10: Fri Mar 7 00:43:40 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 05:43:40 by stevenj] Chris@10: GNU-lly correct Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Mar 7 00:38:48 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 05:38:48 by stevenj] Chris@10: started installation section Chris@10: Chris@10: M ./doc/fftw3.texi -3 +33 Chris@10: Chris@10: Fri Mar 7 00:25:02 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 05:25:02 by stevenj] Chris@10: added --without-cycle-counter option as a last resort Chris@10: Chris@10: M ./configure.ac +5 Chris@10: M ./kernel/timer.c -11 +24 Chris@10: Chris@10: Fri Mar 7 00:07:12 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 05:07:12 by stevenj] Chris@10: macros with () arguments were only standardized in C99, and we don't need them anyway Chris@10: Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: Chris@10: Thu Mar 6 23:10:41 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 04:10:41 by stevenj] Chris@10: wording Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Thu Mar 6 23:03:03 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 04:03:03 by stevenj] Chris@10: parallelism Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Thu Mar 6 23:01:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 04:01:47 by stevenj] Chris@10: additions to upgrading chapter Chris@10: Chris@10: M ./doc/fftw3.texi -1 +31 Chris@10: Chris@10: Thu Mar 6 22:39:36 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 03:39:36 by stevenj] Chris@10: noted additional humility of FFTW 3 wisdom Chris@10: Chris@10: M ./doc/fftw3.texi -1 +9 Chris@10: Chris@10: Thu Mar 6 22:32:44 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 03:32:44 by stevenj] Chris@10: renaming Chris@10: Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Thu Mar 6 22:31:00 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 03:31:00 by stevenj] Chris@10: added placeholder for wisdom reference Chris@10: Chris@10: M ./doc/fftw3.texi -3 +10 Chris@10: Chris@10: Thu Mar 6 22:29:38 EST 2003 stevenj Chris@10: * [project @ 2003-03-07 03:29:38 by stevenj] Chris@10: wrote upgrading chapter Chris@10: Chris@10: M ./doc/fftw3.texi -1 +139 Chris@10: Chris@10: Thu Mar 6 18:01:10 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 23:01:10 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Thu Mar 6 18:00:43 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 23:00:43 by stevenj] Chris@10: placeholder for upgrade chapter Chris@10: Chris@10: M ./doc/fftw3.texi -3 +18 Chris@10: Chris@10: Thu Mar 6 13:47:49 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 18:47:49 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: Chris@10: Thu Mar 6 13:36:38 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 18:36:38 by stevenj] Chris@10: strengthed warning about time Chris@10: Chris@10: M ./tools/fftw_wisdom.1.in -1 +1 Chris@10: Chris@10: Thu Mar 6 13:35:42 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 18:35:42 by stevenj] Chris@10: noted -t in example Chris@10: Chris@10: M ./tools/fftw_wisdom.1.in -1 +3 Chris@10: Chris@10: Thu Mar 6 13:21:03 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 18:21:03 by stevenj] Chris@10: pay attention to WINDOWS_F77_MANGLING Chris@10: Chris@10: M ./threads/f77api.c -3 +26 Chris@10: Chris@10: Thu Mar 6 02:52:30 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:52:30 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Thu Mar 6 02:51:02 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:51:02 by stevenj] Chris@10: index Chris@10: Chris@10: M ./doc/fftw3.texi -1 +3 Chris@10: Chris@10: Thu Mar 6 02:50:38 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:50:38 by stevenj] Chris@10: documented C++ usage Chris@10: Chris@10: M ./doc/fftw3.texi -3 +18 Chris@10: Chris@10: Thu Mar 6 02:25:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:25:32 by stevenj] Chris@10: got rid of overfull hbox TeX warnings Chris@10: Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Thu Mar 6 02:20:38 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:20:38 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Thu Mar 6 02:20:13 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:20:13 by stevenj] Chris@10: noted fftw_iodim split for Fortran guru interface Chris@10: Chris@10: M ./doc/fftw3.texi -5 +16 Chris@10: Chris@10: Thu Mar 6 02:14:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 07:14:21 by stevenj] Chris@10: added guru reference Chris@10: Chris@10: M ./doc/fftw3.texi -3 +187 Chris@10: Chris@10: Wed Mar 5 22:56:05 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 03:56:05 by stevenj] Chris@10: minor Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Wed Mar 5 22:45:31 EST 2003 stevenj Chris@10: * [project @ 2003-03-06 03:45:31 by stevenj] Chris@10: use @r{...} for comment text in code examples Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Wed Mar 5 13:14:04 EST 2003 stevenj Chris@10: * [project @ 2003-03-05 18:14:04 by stevenj] Chris@10: eliminate warning Chris@10: Chris@10: M ./simd/sse.c -1 +2 Chris@10: Chris@10: Wed Mar 5 13:12:56 EST 2003 stevenj Chris@10: * [project @ 2003-03-05 18:12:56 by stevenj] Chris@10: SIMD_CFLAGS only for simd code Chris@10: Chris@10: M ./configure.ac -5 +8 Chris@10: M ./dft/simd/Makefile.am -1 +1 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: M ./kernel/align.c -12 +9 Chris@10: M ./simd/Makefile.am +1 Chris@10: Chris@10: Wed Mar 5 11:06:41 EST 2003 athena Chris@10: * [project @ 2003-03-05 16:06:41 by athena] Chris@10: Minor changes. Chris@10: Chris@10: M ./doc/fftw3.texi -89 +96 Chris@10: Chris@10: Wed Mar 5 02:13:34 EST 2003 stevenj Chris@10: * [project @ 2003-03-05 07:13:34 by stevenj] Chris@10: cross-compiling with MinGW can't detect f77 mangling, so add an option to use what seems to be the most common styles Chris@10: Chris@10: M ./api/f77api.c -3 +34 Chris@10: M ./configure.ac +5 Chris@10: Chris@10: Tue Mar 4 20:00:31 EST 2003 stevenj Chris@10: * [project @ 2003-03-05 01:00:31 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./libbench2/util.c -2 +3 Chris@10: Chris@10: Tue Mar 4 20:00:13 EST 2003 stevenj Chris@10: * [project @ 2003-03-05 01:00:13 by stevenj] Chris@10: we only use our-malloc-16 on machines where size_t == uintptr_t, so don't bother doing the right thing with the benchmark Chris@10: Chris@10: M ./libbench2/util.c -1 +1 Chris@10: Chris@10: Tue Mar 4 19:46:09 EST 2003 stevenj Chris@10: * [project @ 2003-03-05 00:46:09 by stevenj] Chris@10: support WITH_OUR_MALLOC16 Chris@10: Chris@10: M ./libbench2/util.c -2 +19 Chris@10: Chris@10: Tue Mar 4 18:50:53 EST 2003 fftw Chris@10: * [project @ 2003-03-04 23:50:53 by fftw] Chris@10: automatically add -msse etcetera for --enable-sse etcetera Chris@10: Chris@10: M ./configure.ac +22 Chris@10: Chris@10: Tue Mar 4 18:24:26 EST 2003 fftw Chris@10: * [project @ 2003-03-04 23:24:26 by fftw] Chris@10: got rid of const warning Chris@10: Chris@10: M ./tools/fftw-wisdom.c -2 +2 Chris@10: Chris@10: Tue Mar 4 18:22:48 EST 2003 fftw Chris@10: * [project @ 2003-03-04 23:22:48 by fftw] Chris@10: missing header Chris@10: Chris@10: M ./libbench2/problem.c -1 +2 Chris@10: Chris@10: Tue Mar 4 15:55:47 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 20:55:47 by stevenj] Chris@10: fixes Chris@10: Chris@10: M ./doc/fftw3.texi -15 +49 Chris@10: Chris@10: Tue Mar 4 15:53:26 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 20:53:26 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/import-system-wisdom.c -1 +1 Chris@10: Chris@10: Tue Mar 4 02:22:14 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 07:22:14 by stevenj] Chris@10: started guru reference Chris@10: Chris@10: M ./doc/fftw3.texi -1 +110 Chris@10: Chris@10: Tue Mar 4 01:44:09 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 06:44:09 by stevenj] Chris@10: use same FFTW_IODIM between precisions Chris@10: Chris@10: M ./api/fftw3.h -6 +8 Chris@10: Chris@10: Tue Mar 4 00:25:57 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 05:25:57 by stevenj] Chris@10: renamed section Chris@10: Chris@10: M ./doc/fftw3.texi -5 +5 Chris@10: Chris@10: Tue Mar 4 00:21:49 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 05:21:49 by stevenj] Chris@10: no need for "advanced" in subheadings Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Tue Mar 4 00:20:05 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 05:20:05 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Tue Mar 4 00:17:23 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 05:17:23 by stevenj] Chris@10: finished advanced interface Chris@10: Chris@10: M ./doc/fftw3.texi -1 +34 Chris@10: Chris@10: Mon Mar 3 23:26:12 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 04:26:12 by stevenj] Chris@10: more advance interface docs Chris@10: Chris@10: M ./doc/fftw3.texi -1 +46 Chris@10: Chris@10: Mon Mar 3 23:12:09 EST 2003 stevenj Chris@10: * [project @ 2003-03-04 04:12:09 by stevenj] Chris@10: fail for win32 Chris@10: Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: Chris@10: Mon Mar 3 17:18:48 EST 2003 fftw Chris@10: * [project @ 2003-03-03 22:18:48 by fftw] Chris@10: shortened help string Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Mon Mar 3 17:16:17 EST 2003 fftw Chris@10: * [project @ 2003-03-03 22:16:17 by fftw] Chris@10: fixed cross-refs Chris@10: Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: Chris@10: Mon Mar 3 17:07:27 EST 2003 fftw Chris@10: * [project @ 2003-03-03 22:07:27 by fftw] Chris@10: FFTW_POSSIBLY_UNALIGNED -> simpler FFTW_UNALIGNED in API, added bench option Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/mapflags.c -1 +1 Chris@10: M ./doc/fftw3.texi -3 +3 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Mon Mar 3 16:58:07 EST 2003 fftw Chris@10: * [project @ 2003-03-03 21:58:07 by fftw] Chris@10: whoops Chris@10: Chris@10: M ./kernel/alloc.c -3 +3 Chris@10: Chris@10: Mon Mar 3 16:52:58 EST 2003 fftw Chris@10: * [project @ 2003-03-03 21:52:58 by fftw] Chris@10: noted assumption Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Mon Mar 3 16:50:33 EST 2003 fftw Chris@10: * [project @ 2003-03-03 21:50:33 by fftw] Chris@10: provide our own malloc16 routine because of Windows lossage Chris@10: Chris@10: M ./configure.ac +5 Chris@10: M ./kernel/alloc.c -3 +35 Chris@10: Chris@10: Mon Mar 3 13:28:12 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 18:28:12 by stevenj] Chris@10: capitalization Chris@10: Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Mon Mar 3 13:26:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 18:26:32 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Mar 3 12:55:57 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 17:55:57 by stevenj] Chris@10: vertical skip looks better than indenting for setting off short paragraphs Chris@10: Chris@10: M ./doc/fftw3.texi -1 +6 Chris@10: Chris@10: Mon Mar 3 06:34:09 EST 2003 athena Chris@10: * [project @ 2003-03-03 11:34:09 by athena] Chris@10: Removed franz-mode. Automake was distributing franz files Chris@10: whether franz mode was enabled or not. Chris@10: Chris@10: M ./configure.ac -3 Chris@10: M ./dft/simd/codelets/Makefile.am -7 +2 Chris@10: Chris@10: Mon Mar 3 01:44:00 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 06:44:00 by stevenj] Chris@10: made output boundary conditions more prominent; they are important, Chris@10: because they make the different transform types inequivalent in Chris@10: parity Chris@10: Chris@10: M ./doc/fftw3.texi -4 +8 Chris@10: Chris@10: Mon Mar 3 01:17:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 06:17:28 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Mar 3 01:17:07 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 06:17:07 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Mar 3 01:10:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 06:10:28 by stevenj] Chris@10: started advanced reference Chris@10: Chris@10: M ./doc/fftw3.texi -6 +51 Chris@10: Chris@10: Mon Mar 3 00:52:02 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 05:52:02 by stevenj] Chris@10: r2r reference Chris@10: Chris@10: M ./doc/fftw3.texi -2 +209 Chris@10: Chris@10: Sun Mar 2 23:51:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 04:51:21 by stevenj] Chris@10: workaround for info formatting bug Chris@10: Chris@10: M ./doc/fftw3.texi -5 +4 Chris@10: Chris@10: Sun Mar 2 23:47:19 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 04:47:19 by stevenj] Chris@10: noted lack of fftw_malloc in Fortran Chris@10: Chris@10: M ./doc/fftw3.texi -3 +9 Chris@10: Chris@10: Sun Mar 2 23:42:52 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 04:42:32 by stevenj] Chris@10: parallelism Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Sun Mar 2 23:39:54 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 04:39:05 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -7 +7 Chris@10: Chris@10: Sun Mar 2 23:33:02 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 04:33:02 by stevenj] Chris@10: r2c/c2r reference Chris@10: Chris@10: M ./doc/fftw3.texi -40 +248 Chris@10: Chris@10: Sun Mar 2 22:44:10 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 03:44:10 by stevenj] Chris@10: table of contents was being included twice Chris@10: Chris@10: M ./doc/fftw3.texi -2 +1 Chris@10: Chris@10: Sun Mar 2 22:42:29 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 03:42:29 by stevenj] Chris@10: minor changes Chris@10: Chris@10: M ./doc/fftw3.texi -15 +16 Chris@10: Chris@10: Sun Mar 2 21:54:13 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 02:54:13 by stevenj] Chris@10: started reference section Chris@10: Chris@10: M ./doc/fftw3.texi -2 +238 Chris@10: Chris@10: Sun Mar 2 19:10:02 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 00:10:02 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/Makefile.am -2 +2 Chris@10: Chris@10: Sun Mar 2 19:03:23 EST 2003 stevenj Chris@10: * [project @ 2003-03-03 00:03:23 by stevenj] Chris@10: started ref. section Chris@10: Chris@10: M ./doc/fftw3.texi -2 +146 Chris@10: Chris@10: Sun Mar 2 18:50:58 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 23:50:58 by stevenj] Chris@10: fftw_flops takes const plan Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/flops.c -1 +1 Chris@10: Chris@10: Sun Mar 2 15:54:14 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:54:14 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sun Mar 2 15:52:41 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:52:41 by stevenj] Chris@10: added "Wisdom of Fortran?" section Chris@10: Chris@10: M ./doc/fftw3.texi -1 +39 Chris@10: Chris@10: Sun Mar 2 15:50:37 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:50:37 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/f77_wisdom.f -2 +2 Chris@10: Chris@10: Sun Mar 2 15:49:57 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:49:57 by stevenj] Chris@10: wording Chris@10: Chris@10: M ./doc/f77_wisdom.f -1 +1 Chris@10: Chris@10: Sun Mar 2 15:46:13 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:46:13 by stevenj] Chris@10: added comments Chris@10: Chris@10: M ./doc/f77_wisdom.f +25 Chris@10: Chris@10: Sun Mar 2 15:44:01 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:44:01 by stevenj] Chris@10: added example file Chris@10: Chris@10: A ./doc/f77_wisdom.f Chris@10: Chris@10: Sun Mar 2 15:37:32 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:37:32 by stevenj] Chris@10: don't print out READ WISDOM unless we have Chris@10: Chris@10: M ./tests/bench.c -6 +11 Chris@10: Chris@10: Sun Mar 2 15:36:28 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 20:36:28 by stevenj] Chris@10: EOF is not a space Chris@10: Chris@10: M ./kernel/scan.c -2 +2 Chris@10: Chris@10: Sun Mar 2 09:14:37 EST 2003 athena Chris@10: * [project @ 2003-03-02 14:14:37 by athena] Chris@10: Turn on inline by default Chris@10: Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: Chris@10: Sun Mar 2 07:11:56 EST 2003 athena Chris@10: * [project @ 2003-03-02 12:11:56 by athena] Chris@10: Optionally inline loop in notw codelets Chris@10: Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_notw_c.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -3 +3 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: Chris@10: Sun Mar 2 01:37:41 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 06:37:41 by stevenj] Chris@10: updated nodes Chris@10: Chris@10: M ./doc/fftw3.texi -5 +9 Chris@10: Chris@10: Sun Mar 2 01:37:19 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 06:37:19 by stevenj] Chris@10: wrote most of Fortran chapter Chris@10: Chris@10: M ./doc/fftw3.texi -1 +208 Chris@10: Chris@10: Sun Mar 2 00:58:37 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 05:58:37 by stevenj] Chris@10: citation Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sun Mar 2 00:57:22 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 05:57:22 by stevenj] Chris@10: added parallel FFTW chapter Chris@10: Chris@10: M ./doc/fftw3.texi -3 +209 Chris@10: Chris@10: Sat Mar 1 20:42:23 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 01:42:23 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 1 20:34:38 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 01:34:38 by stevenj] Chris@10: added inlining to TODO Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Sat Mar 1 19:36:26 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 00:36:26 by stevenj] Chris@10: added K Chris@10: Chris@10: M ./CONVENTIONS -1 +2 Chris@10: Chris@10: Sat Mar 1 19:15:18 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 00:15:18 by stevenj] Chris@10: use K for constants Chris@10: Chris@10: M ./dft/zero.c -3 +3 Chris@10: M ./kernel/trig1.c -5 +5 Chris@10: M ./rdft/generic.c -8 +8 Chris@10: M ./rdft/problem.c -3 +3 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -11 +11 Chris@10: M ./reodft/redft00e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft010e-r2hc.c -11 +11 Chris@10: M ./reodft/reodft11e-r2hc.c -7 +7 Chris@10: M ./reodft/reodft11e-radix2.c -11 +11 Chris@10: M ./reodft/rodft00e-r2hc.c -3 +3 Chris@10: Chris@10: Sat Mar 1 19:14:54 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 00:14:54 by stevenj] Chris@10: fixed cross-ref Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Mar 1 19:14:16 EST 2003 stevenj Chris@10: * [project @ 2003-03-02 00:14:16 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Sat Mar 1 18:50:43 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 23:50:43 by stevenj] Chris@10: cleanup Chris@10: Chris@10: M ./doc/fftw3.texi -8 +8 Chris@10: Chris@10: Sat Mar 1 18:46:38 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 23:46:38 by stevenj] Chris@10: "words of wisdom" by itself is a little too obscure Chris@10: Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Sat Mar 1 18:43:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 23:43:21 by stevenj] Chris@10: re-added multi-dimensional array stuff Chris@10: Chris@10: M ./doc/fftw3.texi -1 +203 Chris@10: Chris@10: Sat Mar 1 18:15:22 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 23:15:22 by stevenj] Chris@10: added alignment section Chris@10: Chris@10: M ./doc/fftw3.texi -4 +93 Chris@10: Chris@10: Sat Mar 1 16:34:21 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 21:34:21 by stevenj] Chris@10: shrunk code Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -147 +57 Chris@10: Chris@10: Fri Feb 28 20:22:00 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 01:22:00 by stevenj] Chris@10: slight compression Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -146 +122 Chris@10: Chris@10: Fri Feb 28 19:01:20 EST 2003 stevenj Chris@10: * [project @ 2003-03-01 00:01:20 by stevenj] Chris@10: style Chris@10: Chris@10: M ./doc/fftw3.texi -7 +7 Chris@10: M ./reodft/reodft11e-radix2.c -2 +2 Chris@10: Chris@10: Fri Feb 28 18:46:53 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 23:46:53 by stevenj] Chris@10: noted not in API Chris@10: Chris@10: M ./CONVENTIONS -1 +1 Chris@10: Chris@10: Fri Feb 28 18:43:14 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 23:43:14 by stevenj] Chris@10: more updates Chris@10: Chris@10: M ./CONVENTIONS -3 +4 Chris@10: Chris@10: Fri Feb 28 18:38:42 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 23:38:42 by stevenj] Chris@10: slight updates Chris@10: Chris@10: M ./CONVENTIONS -3 +4 Chris@10: Chris@10: Fri Feb 28 18:28:58 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 23:28:58 by stevenj] Chris@10: great const-ification of apply/solve and print Chris@10: Chris@10: M ./api/f77funcs.c -1 +1 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/print-plan.c -1 +1 Chris@10: M ./dft/buffered.c -5 +5 Chris@10: M ./dft/ct-dif.c -3 +3 Chris@10: M ./dft/ct-dit.c -3 +3 Chris@10: M ./dft/ct-ditbuf.c -3 +3 Chris@10: M ./dft/ct-ditf.c -3 +3 Chris@10: M ./dft/ct.c -3 +3 Chris@10: M ./dft/dft.h -3 +3 Chris@10: M ./dft/direct.c -5 +5 Chris@10: M ./dft/generic.c -4 +4 Chris@10: M ./dft/indirect.c -7 +7 Chris@10: M ./dft/nop.c -3 +3 Chris@10: M ./dft/rader.c -12 +12 Chris@10: M ./dft/rank-geq2.c -5 +5 Chris@10: M ./dft/rank0.c -12 +12 Chris@10: M ./dft/solve.c -3 +3 Chris@10: M ./dft/vrank-geq1.c -5 +5 Chris@10: M ./dft/vrank2-transpose.c -5 +5 Chris@10: M ./dft/vrank3-transpose.c -5 +5 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./rdft/buffered.c -5 +5 Chris@10: M ./rdft/buffered2.c -7 +7 Chris@10: M ./rdft/dft-r2hc.c -5 +5 Chris@10: M ./rdft/dht-r2hc.c -5 +5 Chris@10: M ./rdft/dht-rader.c -4 +4 Chris@10: M ./rdft/direct.c -7 +7 Chris@10: M ./rdft/direct2.c -7 +7 Chris@10: M ./rdft/generic.c -6 +6 Chris@10: M ./rdft/hc2hc-buf.c -5 +5 Chris@10: M ./rdft/hc2hc-dif.c -3 +3 Chris@10: M ./rdft/hc2hc-dit.c -3 +3 Chris@10: M ./rdft/hc2hc.c -3 +3 Chris@10: M ./rdft/indirect.c -7 +7 Chris@10: M ./rdft/nop.c -3 +3 Chris@10: M ./rdft/nop2.c -3 +3 Chris@10: M ./rdft/rader-hc2hc.c -6 +6 Chris@10: M ./rdft/rank-geq2-rdft2.c -7 +7 Chris@10: M ./rdft/rank-geq2.c -5 +5 Chris@10: M ./rdft/rank0-rdft2.c -9 +9 Chris@10: M ./rdft/rank0.c -8 +8 Chris@10: M ./rdft/rdft-dht.c -9 +9 Chris@10: M ./rdft/rdft.h -5 +5 Chris@10: M ./rdft/rdft2-radix2.c -12 +12 Chris@10: M ./rdft/solve.c -3 +3 Chris@10: M ./rdft/solve2.c -3 +3 Chris@10: M ./rdft/vrank-geq1-rdft2.c -7 +7 Chris@10: M ./rdft/vrank-geq1.c -5 +5 Chris@10: M ./rdft/vrank2-transpose.c -5 +5 Chris@10: M ./rdft/vrank3-transpose.c -5 +5 Chris@10: M ./reodft/redft00e-r2hc.c -5 +5 Chris@10: M ./reodft/reodft010e-r2hc.c -11 +11 Chris@10: M ./reodft/reodft11e-r2hc-odd.c -7 +7 Chris@10: M ./reodft/reodft11e-r2hc.c -7 +7 Chris@10: M ./reodft/reodft11e-radix2.c -9 +9 Chris@10: M ./reodft/rodft00e-r2hc.c -5 +5 Chris@10: M ./threads/ct-dit.c -3 +3 Chris@10: M ./threads/dft-vrank-geq1.c -5 +5 Chris@10: M ./threads/hc2hc-dif.c -3 +3 Chris@10: M ./threads/hc2hc-dit.c -3 +3 Chris@10: M ./threads/rdft-vrank-geq1.c -5 +5 Chris@10: M ./threads/vrank-geq1-rdft2.c -7 +7 Chris@10: Chris@10: Fri Feb 28 17:51:15 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 22:51:15 by stevenj] Chris@10: make fftw_execute take a const plan, to remind the user that it is re-entrant (or should be)... Chris@10: Chris@10: M ./api/execute-dft-c2r.c -1 +1 Chris@10: M ./api/execute-dft-r2c.c -1 +1 Chris@10: M ./api/execute-dft.c -1 +1 Chris@10: M ./api/execute-r2r.c -1 +1 Chris@10: M ./api/execute.c -1 +1 Chris@10: M ./api/f77funcs.c -5 +6 Chris@10: M ./api/fftw3.h -6 +6 Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Feb 28 17:29:40 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 22:29:40 by stevenj] Chris@10: weakening Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Fri Feb 28 17:28:48 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 22:28:48 by stevenj] Chris@10: note Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Fri Feb 28 17:27:10 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 22:27:10 by stevenj] Chris@10: footnote about why DHT is provided Chris@10: Chris@10: M ./doc/fftw3.texi -7 +12 Chris@10: Chris@10: Fri Feb 28 15:07:03 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 20:07:03 by stevenj] Chris@10: index Chris@10: Chris@10: M ./doc/fftw3.texi -1 +2 Chris@10: Chris@10: Fri Feb 28 15:05:48 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 20:05:48 by stevenj] Chris@10: added DHT tutorial Chris@10: Chris@10: M ./doc/fftw3.texi -2 +32 Chris@10: Chris@10: Fri Feb 28 14:36:45 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 19:36:45 by stevenj] Chris@10: fixed O(n log n) Chris@10: Chris@10: M ./doc/fftw3.texi -4 +16 Chris@10: Chris@10: Fri Feb 28 14:12:15 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 19:08:03 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./doc/fftw3.texi -4 +10 Chris@10: Chris@10: Fri Feb 28 14:06:22 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 19:06:22 by stevenj] Chris@10: slight improvements Chris@10: Chris@10: M ./doc/fftw3.texi -2 +12 Chris@10: Chris@10: Fri Feb 28 00:55:50 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 05:55:50 by stevenj] Chris@10: addition Chris@10: Chris@10: M ./doc/fftw3.texi -3 +5 Chris@10: Chris@10: Fri Feb 28 00:54:09 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 05:54:09 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -7 +8 Chris@10: Chris@10: Thu Feb 27 23:49:37 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 04:49:37 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Thu Feb 27 23:43:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 04:43:56 by stevenj] Chris@10: slight changes Chris@10: Chris@10: M ./doc/fftw3.texi -15 +21 Chris@10: Chris@10: Thu Feb 27 23:27:48 EST 2003 stevenj Chris@10: * [project @ 2003-02-28 04:27:48 by stevenj] Chris@10: added R{E,O}DFTab tutorial Chris@10: Chris@10: M ./doc/fftw3.texi -9 +105 Chris@10: Chris@10: Thu Feb 27 17:24:20 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 22:24:20 by stevenj] Chris@10: fixes Chris@10: Chris@10: M ./doc/fftw3.texi -12 +11 Chris@10: Chris@10: Thu Feb 27 17:20:42 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 22:20:42 by stevenj] Chris@10: fixes Chris@10: Chris@10: M ./doc/fftw3.texi -24 +26 Chris@10: Chris@10: Thu Feb 27 17:11:54 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 22:11:22 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Thu Feb 27 17:07:45 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 22:07:45 by stevenj] Chris@10: documented r2hc/hc2r Chris@10: Chris@10: M ./doc/fftw3.texi -6 +86 Chris@10: Chris@10: Thu Feb 27 16:19:16 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 21:19:16 by stevenj] Chris@10: minor changes Chris@10: Chris@10: M ./doc/fftw3.texi -24 +26 Chris@10: Chris@10: Thu Feb 27 13:54:06 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 18:54:06 by stevenj] Chris@10: timed planner and unifying radix-2 butterfly loops are not critical for release Chris@10: Chris@10: M ./TODO -8 +8 Chris@10: Chris@10: Thu Feb 27 13:51:20 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 18:51:20 by stevenj] Chris@10: reodft/verify.c no longer exists Chris@10: Chris@10: M ./TODO -2 Chris@10: Chris@10: Thu Feb 27 13:44:19 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 18:44:19 by stevenj] Chris@10: optimization: REDFT00 of size 2 is same as R2HC Chris@10: Chris@10: M ./rdft/problem.c -1 +5 Chris@10: Chris@10: Thu Feb 27 12:35:33 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 17:35:33 by stevenj] Chris@10: R{E,O}DFT01 of size-1 is identity Chris@10: Chris@10: M ./rdft/problem.c -3 +4 Chris@10: Chris@10: Thu Feb 27 12:15:10 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 17:15:10 by stevenj] Chris@10: minor simplification Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -31 +31 Chris@10: Chris@10: Thu Feb 27 02:46:31 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 07:46:31 by stevenj] Chris@10: fixed add count Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: Chris@10: Thu Feb 27 02:25:04 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 07:25:04 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -5 +5 Chris@10: Chris@10: Thu Feb 27 02:22:03 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 07:22:03 by stevenj] Chris@10: another optimization Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -53 +53 Chris@10: Chris@10: Thu Feb 27 01:43:00 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 06:43:00 by stevenj] Chris@10: added op counts Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +4 Chris@10: M ./reodft/reodft11e-radix2.c -2 +9 Chris@10: Chris@10: Thu Feb 27 01:29:32 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 06:29:32 by stevenj] Chris@10: cleanup Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -53 +55 Chris@10: Chris@10: Thu Feb 27 01:17:23 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 06:17:23 by stevenj] Chris@10: typo in comment Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: Chris@10: Thu Feb 27 01:13:49 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 06:13:49 by stevenj] Chris@10: fixed comment Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: Chris@10: Thu Feb 27 01:12:05 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 06:12:05 by stevenj] Chris@10: use E instead of R Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -9 +9 Chris@10: Chris@10: Thu Feb 27 01:05:39 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 06:05:39 by stevenj] Chris@10: more unrolling to eliminate if statements in loops, for speedups of 25-40% Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -25 +28 Chris@10: Chris@10: Thu Feb 27 00:27:00 EST 2003 stevenj Chris@10: * [project @ 2003-02-27 05:27:00 by stevenj] Chris@10: some loop splitting to touch each element of output buf only once and eliminate some conditionals...speeds up by 30-40% Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -31 +167 Chris@10: Chris@10: Wed Feb 26 17:48:26 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 22:48:26 by stevenj] Chris@10: comma Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: Chris@10: Wed Feb 26 17:46:17 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 22:46:17 by stevenj] Chris@10: pointer to odd case Chris@10: Chris@10: M ./reodft/reodft11e-radix2.c -1 +3 Chris@10: Chris@10: Wed Feb 26 17:40:54 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 22:40:54 by stevenj] Chris@10: precision -> accuracy (c.f. Kahan) Chris@10: Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: Chris@10: Wed Feb 26 17:36:13 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 22:36:13 by stevenj] Chris@10: added time limit for wisdom generation Chris@10: Chris@10: M ./Makefile.am -1 +4 Chris@10: M ./libbench2/bench-user.h -1 +2 Chris@10: M ./libbench2/problem.c -1 +4 Chris@10: M ./tools/fftw-wisdom.c -16 +59 Chris@10: M ./tools/fftw_wisdom.1.in +6 Chris@10: Chris@10: Wed Feb 26 13:24:36 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 18:24:36 by stevenj] Chris@10: caps Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +2 Chris@10: Chris@10: Tue Feb 25 20:56:01 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 01:56:01 by stevenj] Chris@10: another note Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -2 +3 Chris@10: Chris@10: Tue Feb 25 20:54:57 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 01:54:57 by stevenj] Chris@10: note Chris@10: Chris@10: M ./reodft/reodft11e-r2hc-odd.c -4 +5 Chris@10: Chris@10: Tue Feb 25 20:42:08 EST 2003 stevenj Chris@10: * [project @ 2003-02-26 01:42:08 by stevenj] Chris@10: added new, more accurate (hopefully) reodft11 algorithms; added --disable-debug-malloc; added --impulse-accuracy-rounds=rounds flags to libbench2 for impulse-response accuracy tests Chris@10: Chris@10: A ./reodft/reodft11e-r2hc-odd.c Chris@10: A ./reodft/reodft11e-radix2.c Chris@10: M ./configure.ac +8 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./libbench2/bench-main.c -2 +8 Chris@10: M ./libbench2/bench.h -2 +2 Chris@10: M ./libbench2/verify-dft.c -3 +4 Chris@10: M ./libbench2/verify-lib.c -18 +76 Chris@10: M ./libbench2/verify-r2r.c -7 +39 Chris@10: M ./libbench2/verify-rdft2.c -3 +4 Chris@10: M ./libbench2/verify.c -7 +10 Chris@10: M ./libbench2/verify.h -4 +7 Chris@10: M ./reodft/Makefile.am -1 +4 Chris@10: M ./reodft/conf.c -2 +4 Chris@10: M ./reodft/reodft.h -1 +3 Chris@10: M ./reodft/reodft11e-r2hc-odd.c +273 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +8 Chris@10: M ./reodft/reodft11e-radix2.c +506 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Feb 23 14:07:48 EST 2003 athena Chris@10: * [project @ 2003-02-23 19:07:48 by athena] Chris@10: fftw_wisdom.1 is in $builddir, not $srcdir Chris@10: Chris@10: M ./tools/Makefile.am -1 +1 Chris@10: Chris@10: Mon Feb 17 03:42:19 EST 2003 stevenj Chris@10: * [project @ 2003-02-17 08:42:19 by stevenj] Chris@10: pde Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Mon Feb 17 03:40:19 EST 2003 stevenj Chris@10: * [project @ 2003-02-17 08:40:19 by stevenj] Chris@10: consistent number Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Mon Feb 17 03:39:02 EST 2003 stevenj Chris@10: * [project @ 2003-02-17 08:39:02 by stevenj] Chris@10: started r2r doc Chris@10: Chris@10: M ./doc/fftw3.texi -1 +65 Chris@10: Chris@10: Mon Feb 17 02:31:51 EST 2003 stevenj Chris@10: * [project @ 2003-02-17 07:31:50 by stevenj] Chris@10: rfftwnd Chris@10: Chris@10: A ./doc/rfftwnd.fig Chris@10: A ./doc/rfftwnd.gif Chris@10: M ./doc/Makefile.am +11 Chris@10: M ./doc/fftw3.texi -26 +147 Chris@10: M ./doc/rfftwnd.fig +1148 Chris@10: M ./doc/rfftwnd.gif Chris@10: Chris@10: Sat Feb 15 17:02:07 EST 2003 stevenj Chris@10: * [project @ 2003-02-15 22:02:07 by stevenj] Chris@10: continued Chris@10: Chris@10: M ./doc/fftw3.texi -4 +32 Chris@10: Chris@10: Sat Feb 15 15:16:26 EST 2003 stevenj Chris@10: * [project @ 2003-02-15 20:16:26 by stevenj] Chris@10: started r2c/c2r docs Chris@10: Chris@10: M ./doc/fftw3.texi -8 +87 Chris@10: Chris@10: Sat Feb 15 01:12:52 EST 2003 stevenj Chris@10: * [project @ 2003-02-15 06:12:52 by stevenj] Chris@10: added r{e,o}dft11 accuracy test Chris@10: Chris@10: M ./libbench2/verify-r2r.c -23 +45 Chris@10: Chris@10: Sat Feb 15 00:42:48 EST 2003 stevenj Chris@10: * [project @ 2003-02-15 05:42:48 by stevenj] Chris@10: added more r2r accuracy checks Chris@10: Chris@10: M ./libbench2/verify-dft.c -2 +2 Chris@10: M ./libbench2/verify-lib.c -10 +4 Chris@10: M ./libbench2/verify-r2r.c -1 +127 Chris@10: M ./libbench2/verify-rdft2.c -2 +2 Chris@10: M ./libbench2/verify.h -2 +3 Chris@10: Chris@10: Fri Feb 14 19:19:54 EST 2003 athena Chris@10: * [project @ 2003-02-15 00:19:54 by athena] Chris@10: $< is a GNUism Chris@10: Chris@10: M ./tools/Makefile.am -1 +1 Chris@10: Chris@10: Wed Feb 12 21:02:16 EST 2003 stevenj Chris@10: * [project @ 2003-02-13 02:02:16 by stevenj] Chris@10: r2r test cases are in Chris@10: Chris@10: M ./TODO -1 +1 Chris@10: Chris@10: Wed Feb 12 21:01:28 EST 2003 stevenj Chris@10: * [project @ 2003-02-13 02:01:28 by stevenj] Chris@10: added vector radix to TODO Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Wed Feb 12 17:21:33 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 22:21:33 by stevenj] Chris@10: fixed cross-ref Chris@10: Chris@10: M ./tools/fftw_wisdom.1.in -1 +1 Chris@10: Chris@10: Wed Feb 12 17:19:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 22:19:56 by stevenj] Chris@10: shorter synopsis Chris@10: Chris@10: M ./tools/fftw_wisdom.1.in -1 +1 Chris@10: Chris@10: Wed Feb 12 12:53:19 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 17:53:19 by stevenj] Chris@10: obsolete Chris@10: Chris@10: R ./tests/debug.h Chris@10: Chris@10: Wed Feb 12 12:52:53 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 17:52:53 by stevenj] Chris@10: removed old dotens Chris@10: Chris@10: R ./tests/dotens.c Chris@10: R ./tests/dotens2.c Chris@10: Chris@10: Wed Feb 12 12:52:16 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 17:52:16 by stevenj] Chris@10: removed old verify files Chris@10: Chris@10: R ./tests/verify-dft.c Chris@10: R ./tests/verify-lib.c Chris@10: R ./tests/verify-rdft.c Chris@10: R ./tests/verify-reodft.c Chris@10: R ./tests/verify.h Chris@10: Chris@10: Wed Feb 12 12:37:17 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 17:37:17 by stevenj] Chris@10: disable threads support by default Chris@10: Chris@10: M ./tools/fftw-wisdom.c -4 +11 Chris@10: M ./tools/fftw_wisdom.1.in -7 Chris@10: Chris@10: Wed Feb 12 11:03:28 EST 2003 athena Chris@10: * [project @ 2003-02-12 16:03:28 by athena] Chris@10: Removed old test program Chris@10: Chris@10: M ./tests/bench.c -390 Chris@10: Chris@10: Tue Feb 11 22:30:55 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 03:30:55 by stevenj] Chris@10: joke Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: Chris@10: Tue Feb 11 22:27:44 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 03:27:44 by stevenj] Chris@10: add --help and --version, to be GNU-lly correct Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.1 +7 Chris@10: M ./tools/fftw-wisdom-to-conf.in +36 Chris@10: Chris@10: Tue Feb 11 22:27:18 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 03:27:18 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./tools/fftw_wisdom.1.in -1 Chris@10: Chris@10: Tue Feb 11 22:17:35 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 03:15:03 by stevenj] Chris@10: better help Chris@10: Chris@10: M ./tools/fftw-wisdom.c -10 +43 Chris@10: Chris@10: Tue Feb 11 21:47:35 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 02:47:35 by stevenj] Chris@10: comma Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.1 -2 +1 Chris@10: Chris@10: Tue Feb 11 21:46:12 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 02:46:12 by stevenj] Chris@10: formatting Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.1 -1 +1 Chris@10: Chris@10: Tue Feb 11 21:45:23 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 02:45:23 by stevenj] Chris@10: man pages for tools Chris@10: Chris@10: A ./tools/fftw-wisdom-to-conf.1 Chris@10: A ./tools/fftw_wisdom.1.in Chris@10: M ./configure.ac +1 Chris@10: M ./tools/Makefile.am -1 +7 Chris@10: M ./tools/fftw-wisdom-to-conf.1 +85 Chris@10: M ./tools/fftw_wisdom.1.in +190 Chris@10: Chris@10: Tue Feb 11 19:07:12 EST 2003 stevenj Chris@10: * [project @ 2003-02-12 00:07:12 by stevenj] Chris@10: added -V Chris@10: Chris@10: M ./tools/fftw-wisdom.c +25 Chris@10: Chris@10: Tue Feb 11 18:42:17 EST 2003 stevenj Chris@10: * [project @ 2003-02-11 23:42:17 by stevenj] Chris@10: added install-wisdom target Chris@10: Chris@10: M ./Makefile.am +10 Chris@10: Chris@10: Tue Feb 11 18:23:02 EST 2003 stevenj Chris@10: * [project @ 2003-02-11 23:23:02 by stevenj] Chris@10: another note Chris@10: Chris@10: M ./NEWS -1 +2 Chris@10: Chris@10: Tue Feb 11 17:32:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-11 22:32:56 by stevenj] Chris@10: started r2r accuracy tests (only three kinds covered so far) Chris@10: Chris@10: M ./libbench2/verify-dft.c -2 +2 Chris@10: M ./libbench2/verify-lib.c -10 +13 Chris@10: M ./libbench2/verify-r2r.c +103 Chris@10: M ./libbench2/verify-rdft2.c -2 +2 Chris@10: M ./libbench2/verify.c -2 +2 Chris@10: M ./libbench2/verify.h -1 +4 Chris@10: Chris@10: Mon Feb 10 22:04:18 EST 2003 stevenj Chris@10: * [project @ 2003-02-11 03:04:18 by stevenj] Chris@10: silence warning Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: Chris@10: Mon Feb 10 20:55:20 EST 2003 athena Chris@10: * [project @ 2003-02-11 01:55:20 by athena] Chris@10: gcc bug is now avoided. Chris@10: Chris@10: M ./TODO -5 Chris@10: Chris@10: Mon Feb 10 20:37:54 EST 2003 athena Chris@10: * [project @ 2003-02-11 01:37:54 by athena] Chris@10: Accuracy test Chris@10: Chris@10: A ./libbench2/mp.c Chris@10: M ./libbench2/Makefile.am -5 +6 Chris@10: M ./libbench2/bench-user.h -7 +1 Chris@10: M ./libbench2/mp.c +592 Chris@10: M ./libbench2/verify-dft.c -1 +22 Chris@10: M ./libbench2/verify-lib.c -16 +39 Chris@10: M ./libbench2/verify-r2r.c -17 +2 Chris@10: M ./libbench2/verify-rdft2.c -1 +23 Chris@10: M ./libbench2/verify.c -2 +21 Chris@10: M ./libbench2/verify.h +27 Chris@10: Chris@10: Mon Feb 10 07:59:57 EST 2003 athena Chris@10: * [project @ 2003-02-10 12:59:57 by athena] Chris@10: There is no point in precomputing strides for the long-double code, as Chris@10: multiplication by sizeof(long double) cannot be folded into the Chris@10: addressing mode. This change also fixes the gcc-2.95 bug that causes Chris@10: miscompilation of certain codelets. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Mon Feb 10 02:54:35 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 07:54:35 by stevenj] Chris@10: added random r2r tests Chris@10: Chris@10: M ./tests/check.pl -3 +19 Chris@10: Chris@10: Mon Feb 10 02:44:58 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 07:44:58 by stevenj] Chris@10: whoops, bugfix: missing stride for ro10 Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: Chris@10: Mon Feb 10 02:21:50 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 07:21:50 by stevenj] Chris@10: formatting Chris@10: Chris@10: M ./api/mapflags.c -1 +2 Chris@10: Chris@10: Sun Feb 9 23:24:52 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 04:24:52 by stevenj] Chris@10: flop counts for reodft Chris@10: Chris@10: M ./reodft/redft00e-r2hc.c -5 +12 Chris@10: M ./reodft/reodft010e-r2hc.c -3 +16 Chris@10: M ./reodft/reodft11e-r2hc.c -3 +10 Chris@10: M ./reodft/rodft00e-r2hc.c -3 +12 Chris@10: Chris@10: Sun Feb 9 23:22:15 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 04:22:15 by stevenj] Chris@10: declare aligned_main Chris@10: Chris@10: M ./libbench2/bench.h -1 +2 Chris@10: Chris@10: Sun Feb 9 20:56:06 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 01:56:06 by stevenj] Chris@10: corrected rader op counts Chris@10: Chris@10: M ./rdft/dht-rader.c -6 +5 Chris@10: M ./rdft/rader-hc2hc.c -2 +2 Chris@10: Chris@10: Sun Feb 9 20:25:32 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 01:25:32 by stevenj] Chris@10: punctuation Chris@10: Chris@10: M ./TODO -1 +1 Chris@10: Chris@10: Sun Feb 9 20:25:17 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 01:25:17 by stevenj] Chris@10: noted need for better estimator Chris@10: Chris@10: M ./TODO +4 Chris@10: Chris@10: Sun Feb 9 19:58:59 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 00:58:59 by stevenj] Chris@10: noted F77 api fix for g77 mangling incompatibility Chris@10: Chris@10: M ./NEWS +3 Chris@10: Chris@10: Sun Feb 9 19:30:55 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 00:30:55 by stevenj] Chris@10: build f77 header file of constants from fftw3.h Chris@10: Chris@10: M ./api/Makefile.am -1 +10 Chris@10: Chris@10: Sun Feb 9 19:04:53 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 00:04:53 by stevenj] Chris@10: updates Chris@10: Chris@10: M ./TODO -7 +7 Chris@10: Chris@10: Sun Feb 9 19:03:34 EST 2003 stevenj Chris@10: * [project @ 2003-02-10 00:03:34 by stevenj] Chris@10: threads f77 api Chris@10: Chris@10: A ./api/x77.h Chris@10: A ./threads/f77api.c Chris@10: A ./threads/f77funcs.c Chris@10: M ./api/Makefile.am -1 +1 Chris@10: M ./api/f77api.c -12 +1 Chris@10: M ./api/x77.h +32 Chris@10: M ./threads/Makefile.am -1 +1 Chris@10: M ./threads/f77api.c +39 Chris@10: M ./threads/f77funcs.c +39 Chris@10: Chris@10: Sun Feb 9 18:54:00 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 23:54:00 by stevenj] Chris@10: finished f77 serial api Chris@10: Chris@10: M ./api/f77api.c +15 Chris@10: M ./api/f77funcs.c +220 Chris@10: Chris@10: Sun Feb 9 18:32:26 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 23:32:26 by stevenj] Chris@10: added flops, slight cleanups Chris@10: Chris@10: M ./api/f77api.c -8 +5 Chris@10: M ./api/f77funcs.c -21 +27 Chris@10: Chris@10: Sun Feb 9 18:11:48 EST 2003 athena Chris@10: * [project @ 2003-02-09 23:11:48 by athena] Chris@10: Oops, forgot #include Chris@10: Chris@10: M ./libbench2/aligned-main.c -1 +3 Chris@10: Chris@10: Sun Feb 9 18:08:26 EST 2003 athena Chris@10: * [project @ 2003-02-09 23:08:26 by athena] Chris@10: Removed duplication of stack-alignment code Chris@10: Chris@10: A ./libbench2/aligned-main.c Chris@10: M ./libbench2/Makefile.am -5 +5 Chris@10: M ./libbench2/aligned-main.c +81 Chris@10: M ./libbench2/bench-main.c -62 +2 Chris@10: M ./libbench2/main.c -2 +2 Chris@10: M ./tools/fftw-wisdom.c -61 +1 Chris@10: Chris@10: Sun Feb 9 15:48:15 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 20:48:15 by stevenj] Chris@10: allow - to read problems from stdin Chris@10: Chris@10: M ./tools/fftw-wisdom.c -2 +14 Chris@10: Chris@10: Sun Feb 9 15:22:23 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 20:22:23 by stevenj] Chris@10: added fftw-wisdom tool Chris@10: Chris@10: A ./tools/fftw-wisdom.c Chris@10: M ./tools/Makefile.am +15 Chris@10: M ./tools/fftw-wisdom.c +298 Chris@10: Chris@10: Sun Feb 9 15:06:38 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 20:06:38 by stevenj] Chris@10: elim. warning Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Feb 9 14:24:19 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 19:24:19 by stevenj] Chris@10: destroy_input should not contaminate flags of other problems Chris@10: Chris@10: M ./tests/bench.c -1 +5 Chris@10: Chris@10: Sun Feb 9 13:06:11 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 18:06:11 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog -53 +1740 Chris@10: Chris@10: Sun Feb 9 13:01:45 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 18:01:44 by stevenj] Chris@10: removed overzealous inplace check, which caused problems for rdft2 Chris@10: Chris@10: M ./dft/rank-geq2.c -12 +1 Chris@10: M ./rdft/rank-geq2-rdft2.c -3 +2 Chris@10: M ./rdft/rank-geq2.c -12 +1 Chris@10: Chris@10: Sun Feb 9 08:14:03 EST 2003 athena Chris@10: * [project @ 2003-02-09 13:14:03 by athena] Chris@10: Consistent syntax for RNK_MINFTY tensors Chris@10: Chris@10: M ./kernel/tensor.c -4 +4 Chris@10: Chris@10: Sun Feb 9 07:31:13 EST 2003 athena Chris@10: * [project @ 2003-02-09 12:31:13 by athena] Chris@10: lisply-correct tensor print. We no longer need to parse tensors. Chris@10: Chris@10: M ./kernel/tensor.c -3 +9 Chris@10: Chris@10: Sun Feb 9 03:35:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 08:35:56 by stevenj] Chris@10: removed completed items Chris@10: Chris@10: M ./TODO -4 Chris@10: Chris@10: Sun Feb 9 03:27:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 08:27:56 by stevenj] Chris@10: slight renaming Chris@10: Chris@10: M ./libbench2/verify-r2r.c -4 +4 Chris@10: Chris@10: Sun Feb 9 03:15:28 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 08:15:28 by stevenj] Chris@10: multi-dimensional r2r verifier Chris@10: Chris@10: M ./libbench2/problem.c -2 +2 Chris@10: M ./libbench2/verify-r2r.c -81 +118 Chris@10: Chris@10: Sun Feb 9 02:40:22 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 07:40:22 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./libbench2/verify-r2r.c -1 +3 Chris@10: Chris@10: Sun Feb 9 02:38:26 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 07:38:26 by stevenj] Chris@10: slight simplification Chris@10: Chris@10: M ./libbench2/verify-r2r.c -7 +2 Chris@10: Chris@10: Sun Feb 9 02:36:25 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 07:36:25 by stevenj] Chris@10: added 1d r2r verifier (triple ugh) Chris@10: Chris@10: A ./libbench2/verify-r2r.c Chris@10: M ./libbench2/Makefile.am -2 +2 Chris@10: M ./libbench2/allocate.c -1 +17 Chris@10: M ./libbench2/bench-user.h -2 +10 Chris@10: M ./libbench2/mflops.c +1 Chris@10: M ./libbench2/problem.c -7 +68 Chris@10: M ./libbench2/verify-r2r.c +616 Chris@10: M ./libbench2/verify.c -2 +2 Chris@10: M ./libbench2/zero.c -1 +4 Chris@10: M ./tests/bench.c +101 Chris@10: Chris@10: Sat Feb 8 22:23:00 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 03:23:00 by stevenj] Chris@10: added vector transforms to random tests Chris@10: Chris@10: M ./tests/check.pl -3 +11 Chris@10: Chris@10: Sat Feb 8 20:59:07 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 01:59:07 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/direct2.c -2 +1 Chris@10: Chris@10: Sat Feb 8 19:52:58 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 00:52:58 by stevenj] Chris@10: fixed interaction between dwims for sz/vecsz with rdft2 transforms Chris@10: Chris@10: M ./libbench2/problem.c -15 +17 Chris@10: Chris@10: Sat Feb 8 19:35:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 00:35:56 by stevenj] Chris@10: added destroy_input flag/check Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +2 Chris@10: M ./libbench2/problem.c -1 +3 Chris@10: M ./libbench2/verify-dft.c -1 +4 Chris@10: M ./libbench2/verify-lib.c -1 +35 Chris@10: M ./libbench2/verify-rdft2.c -1 +5 Chris@10: M ./libbench2/verify.h +2 Chris@10: M ./tests/bench.c +7 Chris@10: Chris@10: Sat Feb 8 19:11:58 EST 2003 stevenj Chris@10: * [project @ 2003-02-09 00:11:57 by stevenj] Chris@10: added rdft2 verifier Chris@10: Chris@10: A ./api/extract-reim.c Chris@10: A ./libbench2/aset.c Chris@10: A ./libbench2/verify-rdft2.c Chris@10: M ./api/Makefile.am -3 +3 Chris@10: R ./api/dfthelp.c Chris@10: M ./api/extract-reim.c +36 Chris@10: M ./api/plan-guru-dft-c2r.c -1 +2 Chris@10: M ./api/plan-many-dft-c2r.c +2 Chris@10: M ./libbench2/Makefile.am -5 +5 Chris@10: M ./libbench2/allocate.c -7 +44 Chris@10: M ./libbench2/aset.c +10 Chris@10: M ./libbench2/bench-user.h -1 +4 Chris@10: M ./libbench2/bench.h -1 +2 Chris@10: M ./libbench2/problem.c -10 +42 Chris@10: M ./libbench2/tensor.c -1 +47 Chris@10: M ./libbench2/verify-dft.c -1 +3 Chris@10: M ./libbench2/verify-lib.c -16 +13 Chris@10: M ./libbench2/verify-rdft2.c +255 Chris@10: M ./libbench2/verify.c -2 +2 Chris@10: M ./libbench2/verify.h -1 +1 Chris@10: M ./libbench2/zero.c -2 +8 Chris@10: M ./tests/bench.c -8 +189 Chris@10: M ./tests/check.pl -1 +7 Chris@10: Chris@10: Sat Feb 8 13:31:14 EST 2003 stevenj Chris@10: * [project @ 2003-02-08 18:31:14 by stevenj] Chris@10: an additional check for in-place case Chris@10: Chris@10: M ./rdft/rdft2-radix2.c -3 +7 Chris@10: Chris@10: Fri Feb 7 17:36:56 EST 2003 stevenj Chris@10: * [project @ 2003-02-07 22:36:56 by stevenj] Chris@10: slight fix: hc2r constraints are mostly determined by sub-plan Chris@10: Chris@10: M ./rdft/rank0-rdft2.c -6 +7 Chris@10: Chris@10: Fri Feb 7 16:28:55 EST 2003 stevenj Chris@10: * [project @ 2003-02-07 21:28:55 by stevenj] Chris@10: make radix2-dft inapplicable to in-place/split case (r == rio, iio >= rio + n/2+1 != r + 1) Chris@10: Chris@10: M ./rdft/rdft2-radix2.c -3 +22 Chris@10: Chris@10: Tue Feb 4 06:36:29 EST 2003 athena Chris@10: * [project @ 2003-02-04 11:36:29 by athena] Chris@10: Allow plnr->hook to be 0 Chris@10: Chris@10: M ./kernel/planner.c -12 +12 Chris@10: M ./tests/hook.c -7 +2 Chris@10: Chris@10: Tue Feb 4 03:25:36 EST 2003 stevenj Chris@10: * [project @ 2003-02-04 08:25:36 by stevenj] Chris@10: moved dft stuff into verify-dft Chris@10: Chris@10: M ./libbench2/bench-user.h -4 +2 Chris@10: M ./libbench2/verify-dft.c -9 +84 Chris@10: M ./libbench2/verify.c -84 +2 Chris@10: Chris@10: Tue Feb 4 03:25:00 EST 2003 stevenj Chris@10: * [project @ 2003-02-04 08:25:00 by stevenj] Chris@10: cruft Chris@10: Chris@10: M ./tests/hook.c -1 Chris@10: Chris@10: Tue Feb 4 03:18:28 EST 2003 stevenj Chris@10: * [project @ 2003-02-04 08:18:28 by stevenj] Chris@10: further unify libbench2 and paranoid verifiers Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +6 Chris@10: M ./libbench2/problem.c -1 +2 Chris@10: M ./libbench2/verify.c -16 +21 Chris@10: M ./tests/bench.c -3 +5 Chris@10: M ./tests/hook.c -125 +58 Chris@10: Chris@10: Sun Feb 2 01:45:37 EST 2003 stevenj Chris@10: * [project @ 2003-02-02 06:45:37 by stevenj] Chris@10: typo in comment Chris@10: Chris@10: M ./api/import-wisdom-from-file.c -1 +1 Chris@10: Chris@10: Sat Feb 1 09:30:03 EST 2003 athena Chris@10: * [project @ 2003-02-01 14:30:03 by athena] Chris@10: Fixed p==2 case Chris@10: Chris@10: M ./kernel/primes.c -1 +4 Chris@10: Chris@10: Sat Feb 1 09:23:43 EST 2003 athena Chris@10: * [project @ 2003-02-01 14:23:43 by athena] Chris@10: Incorporated new find_generator by Greg Dionne. Chris@10: Chris@10: M ./kernel/primes.c -21 +35 Chris@10: Chris@10: Fri Jan 31 20:46:24 EST 2003 athena Chris@10: * [project @ 2003-02-01 01:46:24 by athena] Chris@10: Removed nonportable call to gettext() Chris@10: Chris@10: M ./libbench2/getopt.c +7 Chris@10: Chris@10: Wed Jan 29 19:03:43 EST 2003 athena Chris@10: * [project @ 2003-01-30 00:03:43 by athena] Chris@10: uintptr_t is in in openbsd Chris@10: Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: Chris@10: Wed Jan 29 15:41:56 EST 2003 athena Chris@10: * [project @ 2003-01-29 20:41:56 by athena] Chris@10: Huge speedups in wisdom I/O. Chris@10: Chris@10: M ./api/export-wisdom-to-string.c -2 +2 Chris@10: M ./api/export-wisdom.c -1 +1 Chris@10: M ./api/import-wisdom-from-file.c -4 +27 Chris@10: M ./api/mkprinter-file.c -5 +25 Chris@10: M ./kernel/debug.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +5 Chris@10: M ./kernel/planner.c -5 +9 Chris@10: M ./kernel/print.c -2 +7 Chris@10: M ./kernel/scan.c -4 +25 Chris@10: M ./tests/bench.c -1 +5 Chris@10: Chris@10: Tue Jan 28 19:36:51 EST 2003 athena Chris@10: * [project @ 2003-01-29 00:36:51 by athena] Chris@10: Added appropriate warning against likely future bug. Chris@10: Chris@10: M ./kernel/planner.c -1 +4 Chris@10: Chris@10: Tue Jan 28 19:00:24 EST 2003 athena Chris@10: * [project @ 2003-01-29 00:00:24 by athena] Chris@10: Don't attempt to remove bogus wisdom entries. Chris@10: Chris@10: M ./kernel/planner.c -10 +3 Chris@10: Chris@10: Tue Jan 28 18:16:24 EST 2003 athena Chris@10: * [project @ 2003-01-28 23:16:24 by athena] Chris@10: Fixed a couple of very very very nasty bugs---pointers became Chris@10: invalid after the hash table was relocated. Chris@10: Chris@10: M ./kernel/planner.c -19 +29 Chris@10: Chris@10: Tue Jan 28 07:34:10 EST 2003 athena Chris@10: * [project @ 2003-01-28 12:34:10 by athena] Chris@10: Read wisdom at can_do() time, otherwise wisdom is destroyed. Chris@10: Chris@10: M ./tests/bench.c -2 +5 Chris@10: Chris@10: Tue Jan 28 06:54:38 EST 2003 athena Chris@10: * [project @ 2003-01-28 11:54:38 by athena] Chris@10: More conservative inheritance of blessings Chris@10: Chris@10: M ./kernel/planner.c -19 +24 Chris@10: Chris@10: Tue Jan 28 06:50:20 EST 2003 athena Chris@10: * [project @ 2003-01-28 11:50:20 by athena] Chris@10: Print the same info as it is hashed Chris@10: Chris@10: M ./dft/problem.c -3 +4 Chris@10: Chris@10: Tue Jan 28 06:49:48 EST 2003 athena Chris@10: * [project @ 2003-01-28 11:49:48 by athena] Chris@10: Print name of executable when FAILURE Chris@10: Chris@10: M ./tests/check.pl -2 +2 Chris@10: Chris@10: Mon Jan 27 06:59:40 EST 2003 athena Chris@10: * [project @ 2003-01-27 11:59:40 by athena] Chris@10: New NO_SEARCH planner flag, which avoids searching altogether. Chris@10: A wisdom entry must lead to a NO_SEARCH-grade plan, or else the Chris@10: wisdom entry is bogus. Chris@10: Chris@10: M ./kernel/ifftw.h -5 +10 Chris@10: M ./kernel/planner.c -13 +30 Chris@10: Chris@10: Sun Jan 26 20:45:21 EST 2003 athena Chris@10: * [project @ 2003-01-27 01:45:21 by athena] Chris@10: Use cosl()/sinl() when appropriate Chris@10: Chris@10: M ./libbench2/verify-lib.c -3 +20 Chris@10: Chris@10: Sun Jan 26 16:29:18 EST 2003 athena Chris@10: * [project @ 2003-01-26 21:29:18 by athena] Chris@10: Use null pointers when estimating. The estimator should never Chris@10: time anything. Chris@10: Chris@10: M ./kernel/planner.c -1 +6 Chris@10: M ./libbench2/problem.c -2 +1 Chris@10: M ./libbench2/speed.c -1 +2 Chris@10: M ./libbench2/verify.c -1 +3 Chris@10: Chris@10: Sun Jan 26 15:19:01 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 20:19:01 by stevenj] Chris@10: note Chris@10: Chris@10: M ./api/f77api.c -1 +1 Chris@10: Chris@10: Sun Jan 26 15:16:22 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 20:16:22 by stevenj] Chris@10: support multiple mangling schemes with g77 Chris@10: Chris@10: A ./api/f77funcs.c Chris@10: M ./api/Makefile.am +4 Chris@10: M ./api/f77api.c -130 +41 Chris@10: M ./api/f77funcs.c +139 Chris@10: M ./configure.ac +6 Chris@10: Chris@10: Sun Jan 26 12:58:57 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 17:58:57 by stevenj] Chris@10: fixed verbose, made random tests only use selected rank, use rank <= 4, fixed final flush_problems call Chris@10: Chris@10: M ./tests/check.pl -4 +9 Chris@10: Chris@10: Sun Jan 26 12:42:49 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 17:42:49 by stevenj] Chris@10: fixed typo (count instead of maxcount) Chris@10: Chris@10: M ./tests/check.pl -2 +2 Chris@10: Chris@10: Sun Jan 26 12:12:07 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 17:12:07 by stevenj] Chris@10: hypot is no longer used Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sun Jan 26 12:07:43 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 17:07:43 by stevenj] Chris@10: check for _alloca (MSVC) Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: Chris@10: Sun Jan 26 11:56:53 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 16:56:53 by stevenj] Chris@10: slight fix in assert Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Sun Jan 26 11:55:39 EST 2003 athena Chris@10: * [project @ 2003-01-26 16:55:39 by athena] Chris@10: Allocate problem in all cases--- can_do may need correct pointers. Chris@10: Chris@10: M ./libbench2/problem.c -1 +2 Chris@10: M ./libbench2/speed.c -2 +1 Chris@10: M ./libbench2/verify.c -3 +1 Chris@10: M ./tests/bench.c -6 Chris@10: Chris@10: Sun Jan 26 11:51:27 EST 2003 athena Chris@10: * [project @ 2003-01-26 16:51:27 by athena] Chris@10: Nastier checks Chris@10: Chris@10: M ./tests/bench.c -1 +8 Chris@10: M ./tests/check.pl -9 +23 Chris@10: Chris@10: Sun Jan 26 11:51:16 EST 2003 athena Chris@10: * [project @ 2003-01-26 16:51:16 by athena] Chris@10: X(use_plan) is a relic. Chris@10: Chris@10: M ./kernel/ifftw.h -3 +1 Chris@10: M ./kernel/plan.c -11 +3 Chris@10: M ./kernel/planner.c -5 +2 Chris@10: Chris@10: Sun Jan 26 09:23:16 EST 2003 athena Chris@10: * [project @ 2003-01-26 14:23:16 by athena] Chris@10: Print full pathname of the bench executable, so that I don't get Chris@10: confused when running multiple tests for different configurations. Chris@10: Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Sun Jan 26 07:35:46 EST 2003 athena Chris@10: * [project @ 2003-01-26 12:35:46 by athena] Chris@10: Split done() into done() and cleanup(), in order to test Chris@10: multiple problems with the same planner from the command line. Chris@10: Chris@10: M ./libbench2/bench-main.c -1 +2 Chris@10: M ./libbench2/bench-user.h -1 +2 Chris@10: M ./tests/bench.c -1 +8 Chris@10: Chris@10: Sat Jan 25 20:44:49 EST 2003 athena Chris@10: * [project @ 2003-01-26 01:44:49 by athena] Chris@10: Improved readability Chris@10: Chris@10: M ./kernel/alloc.c -6 +16 Chris@10: Chris@10: Sat Jan 25 19:17:26 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 00:17:26 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./kernel/alloc.c -3 +3 Chris@10: Chris@10: Sat Jan 25 19:16:53 EST 2003 stevenj Chris@10: * [project @ 2003-01-26 00:16:53 by stevenj] Chris@10: added macos9 mpallocatealigned function Chris@10: Chris@10: M ./kernel/alloc.c -1 +19 Chris@10: Chris@10: Sat Jan 25 18:59:55 EST 2003 stevenj Chris@10: * [project @ 2003-01-25 23:59:55 by stevenj] Chris@10: sometimes __APPLE__ is defined instead of __MACOSX__ Chris@10: Chris@10: M ./kernel/alloc.c -2 +3 Chris@10: Chris@10: Sat Jan 25 18:54:39 EST 2003 stevenj Chris@10: * [project @ 2003-01-25 23:54:39 by stevenj] Chris@10: macos x malloc is already 16-byte aligned Chris@10: Chris@10: M ./kernel/alloc.c -2 +3 Chris@10: Chris@10: Sat Jan 25 13:38:32 EST 2003 athena Chris@10: * [project @ 2003-01-25 18:38:32 by athena] Chris@10: Include because uintptr_t is defined there Chris@10: on solaris. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +6 Chris@10: Chris@10: Sat Jan 25 13:22:59 EST 2003 athena Chris@10: * [project @ 2003-01-25 18:22:59 by athena] Chris@10: Oops---forgot getopt_long Chris@10: Chris@10: A ./libbench2/getopt1.c Chris@10: M ./libbench2/Makefile.am -2 +2 Chris@10: M ./libbench2/getopt1.c +188 Chris@10: Chris@10: Sat Jan 25 13:17:29 EST 2003 athena Chris@10: * [project @ 2003-01-25 18:17:29 by athena] Chris@10: Include default includes when checking for uintptr_t. Chris@10: (Otherwise solaris breaks.) Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Jan 25 12:39:52 EST 2003 athena Chris@10: * [project @ 2003-01-25 17:39:52 by athena] Chris@10: distribute check.pl Chris@10: Chris@10: M ./tests/Makefile.am +1 Chris@10: Chris@10: Sat Jan 25 12:38:34 EST 2003 athena Chris@10: * [project @ 2003-01-25 17:38:34 by athena] Chris@10: Check split format, too. Chris@10: Chris@10: M ./tests/check.pl +4 Chris@10: Chris@10: Sat Jan 25 11:48:19 EST 2003 athena Chris@10: * [project @ 2003-01-25 16:48:19 by athena] Chris@10: New tests, added make check Chris@10: Chris@10: M ./tests/Makefile.am -1 +2 Chris@10: M ./tests/check.pl -3 +66 Chris@10: Chris@10: Thu Jan 23 08:34:24 EST 2003 athena Chris@10: * [project @ 2003-01-23 13:34:24 by athena] Chris@10: More tests Chris@10: Chris@10: M ./tests/check.pl -17 +83 Chris@10: Chris@10: Tue Jan 21 20:32:12 EST 2003 athena Chris@10: * [project @ 2003-01-22 01:32:09 by athena] Chris@10: Deal with rnk(sz)=-infinity Chris@10: Chris@10: M ./api/mktensor-iodims.c -4 +9 Chris@10: M ./api/mktensor-rowmajor.c -4 +6 Chris@10: M ./libbench2/problem.c -5 +2 Chris@10: Chris@10: Tue Jan 21 10:07:16 EST 2003 athena Chris@10: * [project @ 2003-01-21 15:07:16 by athena] Chris@10: Crazy idea Chris@10: Chris@10: M ./TODO +3 Chris@10: Chris@10: Tue Jan 21 07:14:22 EST 2003 athena Chris@10: * [project @ 2003-01-21 12:14:22 by athena] Chris@10: Test program, still barely worthy of the name. Chris@10: Chris@10: A ./tests/check.pl Chris@10: Chris@10: Mon Jan 20 08:29:21 EST 2003 athena Chris@10: * [project @ 2003-01-20 13:29:21 by athena] Chris@10: Stylistic changes Chris@10: Chris@10: M ./libbench2/problem.c -15 +10 Chris@10: Chris@10: Mon Jan 20 07:03:38 EST 2003 athena Chris@10: * [project @ 2003-01-20 12:03:38 by athena] Chris@10: Implemented flops api Chris@10: Chris@10: A ./api/flops.c Chris@10: M ./api/Makefile.am -10 +10 Chris@10: M ./api/fftw3.h -2 +4 Chris@10: M ./api/flops.c +27 Chris@10: M ./tests/bench.c +3 Chris@10: Chris@10: Sun Jan 19 14:27:21 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 19:27:21 by stevenj] Chris@10: cleanup Chris@10: Chris@10: M ./libbench2/problem.c -31 +24 Chris@10: Chris@10: Sun Jan 19 14:14:49 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 19:14:49 by stevenj] Chris@10: 'v' syntax now defaults to an 'internal' (stride 1) vector, which is a more interesting case and corresponds more closely to the intuitive notion of a 'vector' transform, while '*' does the old 'external' (stride n) vector Chris@10: Chris@10: M ./libbench2/problem.c -4 +24 Chris@10: Chris@10: Sun Jan 19 13:55:35 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 18:55:35 by stevenj] Chris@10: removed '/' overloading Chris@10: Chris@10: M ./libbench2/problem.c -2 +2 Chris@10: Chris@10: Sun Jan 19 13:52:09 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 18:52:09 by stevenj] Chris@10: get rid of '*' and ',' synonyms for 'x' in problem parser; there's no need to clutter the namespace with syntax we never use Chris@10: Chris@10: M ./libbench2/problem.c -2 +2 Chris@10: Chris@10: Sun Jan 19 07:28:27 EST 2003 athena Chris@10: * [project @ 2003-01-19 12:28:27 by athena] Chris@10: Signed/unsigned fixes. Chris@10: Chris@10: M ./kernel/planner.c -4 +4 Chris@10: Chris@10: Sun Jan 19 07:09:54 EST 2003 athena Chris@10: * [project @ 2003-01-19 12:09:54 by athena] Chris@10: Test split arrays. Chris@10: Chris@10: M ./libbench2/bench-user.h -3 +9 Chris@10: M ./libbench2/verify-dft.c -14 +11 Chris@10: M ./libbench2/verify.c -13 +43 Chris@10: M ./libbench2/verify.h +1 Chris@10: M ./tests/bench.c -13 +50 Chris@10: M ./tests/hook.c -1 +4 Chris@10: Chris@10: Sat Jan 18 23:46:57 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 04:46:57 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -17 +16 Chris@10: Chris@10: Sat Jan 18 21:53:18 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 02:53:18 by stevenj] Chris@10: caps Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Jan 18 21:52:51 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 02:52:51 by stevenj] Chris@10: brackets Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Jan 18 21:52:32 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 02:52:32 by stevenj] Chris@10: quote Chris@10: Chris@10: M ./doc/fftw3.texi -1 +7 Chris@10: Chris@10: Sat Jan 18 20:53:11 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:53:11 by stevenj] Chris@10: referencing Chris@10: Chris@10: M ./doc/fftw3.texi -4 +7 Chris@10: Chris@10: Sat Jan 18 20:33:28 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:33:28 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Jan 18 20:31:41 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:31:41 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./doc/fftw3.texi -8 +8 Chris@10: Chris@10: Sat Jan 18 20:31:22 EST 2003 athena Chris@10: * [project @ 2003-01-19 01:31:22 by athena] Chris@10: Print errors when --verify. Chris@10: Chris@10: M ./libbench2/verify-dft.c -7 +12 Chris@10: M ./libbench2/verify-lib.c -30 +38 Chris@10: M ./libbench2/verify.h -12 +14 Chris@10: M ./tests/bench.c -5 +5 Chris@10: Chris@10: Sat Jan 18 20:30:27 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:30:27 by stevenj] Chris@10: improved description, noted that FFTW_ESTIMATE does not destroy arrays Chris@10: Chris@10: M ./doc/fftw3.texi -7 +8 Chris@10: Chris@10: Sat Jan 18 20:23:12 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:23:12 by stevenj] Chris@10: FFTW_DEFAULTS isn't really needed Chris@10: Chris@10: M ./api/fftw3.h -2 +1 Chris@10: Chris@10: Sat Jan 18 20:21:09 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:21:09 by stevenj] Chris@10: added FFTW_MEASURE synonym for FFTW_DEFAULTS Chris@10: Chris@10: M ./api/fftw3.h -1 +2 Chris@10: M ./doc/fftw3.texi -6 +6 Chris@10: Chris@10: Sat Jan 18 20:18:29 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:18:29 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Sat Jan 18 20:16:08 EST 2003 athena Chris@10: * [project @ 2003-01-19 01:16:08 by athena] Chris@10: Clearer name Chris@10: Chris@10: M ./tests/bench.c -2 +2 Chris@10: Chris@10: Sat Jan 18 20:13:14 EST 2003 athena Chris@10: * [project @ 2003-01-19 01:13:14 by athena] Chris@10: Completed dft api test Chris@10: Chris@10: M ./api/fftw3.h -12 +12 Chris@10: M ./libbench2/tensor.c -3 +3 Chris@10: M ./tests/bench.c -8 +65 Chris@10: Chris@10: Sat Jan 18 20:07:33 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:07:33 by stevenj] Chris@10: index Chris@10: Chris@10: M ./doc/fftw3.texi -1 +2 Chris@10: Chris@10: Sat Jan 18 20:05:50 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:05:50 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Jan 18 20:04:11 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:04:11 by stevenj] Chris@10: parallel structure Chris@10: Chris@10: M ./doc/fftw3.texi -2 +2 Chris@10: Chris@10: Sat Jan 18 20:03:18 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:03:18 by stevenj] Chris@10: fix Chris@10: Chris@10: M ./doc/fftw3.texi -4 +4 Chris@10: Chris@10: Sat Jan 18 20:00:24 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 01:00:24 by stevenj] Chris@10: joke Chris@10: Chris@10: M ./doc/fftw3.texi -5 +6 Chris@10: Chris@10: Sat Jan 18 19:59:28 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 00:59:28 by stevenj] Chris@10: recommendation to read tutorial in-order Chris@10: Chris@10: M ./doc/fftw3.texi -1 +6 Chris@10: Chris@10: Sat Jan 18 19:54:55 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 00:54:55 by stevenj] Chris@10: expanded outline Chris@10: Chris@10: M ./doc/fftw3.texi -7 +42 Chris@10: Chris@10: Sat Jan 18 19:35:52 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 00:35:52 by stevenj] Chris@10: clarification Chris@10: Chris@10: M ./doc/fftw3.texi -3 +5 Chris@10: Chris@10: Sat Jan 18 19:17:27 EST 2003 stevenj Chris@10: * [project @ 2003-01-19 00:17:27 by stevenj] Chris@10: draft complex-dft tutorial Chris@10: Chris@10: M ./doc/fftw3.texi -25 +183 Chris@10: Chris@10: Sat Jan 18 17:27:15 EST 2003 athena Chris@10: * [project @ 2003-01-18 22:27:15 by athena] Chris@10: Paranoid mode is back. Fixed dwim to do what I mean. Chris@10: Chris@10: A ./tests/hook.c Chris@10: M ./libbench2/allocate.c -2 +1 Chris@10: M ./libbench2/bench-main.c -2 +1 Chris@10: M ./libbench2/bench-user.h -8 +1 Chris@10: M ./libbench2/bench.h -2 +1 Chris@10: M ./libbench2/can-do.c -2 +1 Chris@10: M ./libbench2/dotens2.c -2 +2 Chris@10: M ./libbench2/info.c -2 +1 Chris@10: M ./libbench2/problem.c -18 +21 Chris@10: M ./libbench2/report.c -2 +1 Chris@10: M ./libbench2/speed.c -2 +1 Chris@10: M ./libbench2/timer.c -2 +1 Chris@10: M ./libbench2/util.c -1 Chris@10: M ./libbench2/verify.c -4 +3 Chris@10: M ./libbench2/verify.h +7 Chris@10: M ./libbench2/zero.c -2 +1 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/bench.c +4 Chris@10: M ./tests/hook.c +191 Chris@10: Chris@10: Sat Jan 18 17:13:51 EST 2003 stevenj Chris@10: * [project @ 2003-01-18 22:13:51 by stevenj] Chris@10: started tut. Chris@10: Chris@10: M ./doc/fftw3.texi -14 +46 Chris@10: Chris@10: Sat Jan 18 16:13:15 EST 2003 athena Chris@10: * [project @ 2003-01-18 21:13:15 by athena] Chris@10: Great renaming, so that we can include both bench-user.h and Chris@10: ifftw.h to implement the paranoid-mode hook. Chris@10: Chris@10: M ./libbench2/allocate.c -5 +5 Chris@10: M ./libbench2/bench-user.h -30 +28 Chris@10: M ./libbench2/bench.h -12 +12 Chris@10: M ./libbench2/can-do.c -2 +2 Chris@10: M ./libbench2/dotens2.c -3 +3 Chris@10: M ./libbench2/mflops.c -1 +1 Chris@10: M ./libbench2/problem.c -12 +12 Chris@10: M ./libbench2/report.c -10 +10 Chris@10: M ./libbench2/speed.c -2 +2 Chris@10: M ./libbench2/tensor.c -22 +22 Chris@10: M ./libbench2/verify-dft.c -2 +2 Chris@10: M ./libbench2/verify-lib.c -6 +6 Chris@10: M ./libbench2/verify.c -9 +9 Chris@10: M ./libbench2/verify.h -2 +2 Chris@10: M ./libbench2/zero.c -2 +2 Chris@10: M ./tests/bench.c -29 +29 Chris@10: Chris@10: Sat Jan 18 15:41:18 EST 2003 athena Chris@10: * [project @ 2003-01-18 20:41:18 by athena] Chris@10: Trying to tweak the verifier so that I can use it in Chris@10: bench.c for paranoid mode Chris@10: Chris@10: M ./libbench2/bench-user.h -1 +23 Chris@10: M ./libbench2/problem.c -3 +3 Chris@10: M ./libbench2/verify-dft.c -66 +9 Chris@10: M ./libbench2/verify-lib.c -21 +18 Chris@10: M ./libbench2/verify.c -5 +62 Chris@10: M ./libbench2/verify.h -18 +3 Chris@10: Chris@10: Sat Jan 18 10:24:05 EST 2003 athena Chris@10: * [project @ 2003-01-18 15:24:05 by athena] Chris@10: Added stride_factor for complex arrays. Chris@10: Chris@10: M ./tests/bench.c -5 +5 Chris@10: Chris@10: Sat Jan 18 10:02:11 EST 2003 athena Chris@10: * [project @ 2003-01-18 15:02:11 by athena] Chris@10: can_do now calls the planner. Chris@10: Chris@10: M ./tests/bench.c -1 +6 Chris@10: Chris@10: Sat Jan 18 09:59:24 EST 2003 athena Chris@10: * [project @ 2003-01-18 14:59:24 by athena] Chris@10: Call guru api in bench.c Chris@10: Chris@10: M ./api/plan-guru-dft.c -2 +1 Chris@10: M ./tests/bench.c -3 +47 Chris@10: Chris@10: Sat Jan 18 08:17:23 EST 2003 athena Chris@10: * [project @ 2003-01-18 13:17:23 by athena] Chris@10: Fixed prototype. Chris@10: Chris@10: M ./libbench2/bench.h -1 +3 Chris@10: M ./libbench2/zero.c -2 +2 Chris@10: Chris@10: Sat Jan 18 08:14:48 EST 2003 athena Chris@10: * [project @ 2003-01-18 13:14:48 by athena] Chris@10: Attempt to make the signed/unsigned use of flags consistent. Chris@10: Chris@10: M ./api/api.h -2 +2 Chris@10: M ./api/apiplan.c -1 +1 Chris@10: M ./api/fftw3.h -25 +25 Chris@10: M ./api/mapflags.c -8 +10 Chris@10: M ./api/plan-dft-1d.c -1 +1 Chris@10: M ./api/plan-dft-2d.c -1 +1 Chris@10: M ./api/plan-dft-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-1d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-2d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r.c -1 +1 Chris@10: M ./api/plan-dft-r2c-1d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-2d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-3d.c -1 +1 Chris@10: M ./api/plan-dft-r2c.c -1 +1 Chris@10: M ./api/plan-dft.c -1 +1 Chris@10: M ./api/plan-guru-dft-c2r.c -1 +1 Chris@10: M ./api/plan-guru-dft-r2c.c -1 +1 Chris@10: M ./api/plan-guru-dft.c -1 +1 Chris@10: M ./api/plan-guru-r2r.c -1 +1 Chris@10: M ./api/plan-many-dft-c2r.c -1 +1 Chris@10: M ./api/plan-many-dft-r2c.c -1 +1 Chris@10: M ./api/plan-many-dft.c -1 +1 Chris@10: M ./api/plan-many-r2r.c -1 +1 Chris@10: M ./api/plan-r2r-1d.c -1 +1 Chris@10: M ./api/plan-r2r-2d.c -1 +1 Chris@10: M ./api/plan-r2r-3d.c -1 +1 Chris@10: M ./api/plan-r2r.c -1 +1 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Sat Jan 18 08:03:07 EST 2003 athena Chris@10: * [project @ 2003-01-18 13:03:07 by athena] Chris@10: Implemented useropt. Chris@10: Chris@10: A ./libbench2/useropt.c Chris@10: M ./libbench2/Makefile.am -2 +3 Chris@10: M ./libbench2/bench-main.c -1 +5 Chris@10: M ./libbench2/bench-user.h -1 +2 Chris@10: M ./libbench2/useropt.c +30 Chris@10: M ./tests/bench.c -7 +24 Chris@10: Chris@10: Sat Jan 18 08:02:05 EST 2003 athena Chris@10: * [project @ 2003-01-18 13:02:05 by athena] Chris@10: The first map_flags pass must be transitive, i.e., always use the Chris@10: latest flags value as opposed to the original value. (I think.) Chris@10: Chris@10: M ./api/mapflags.c -11 +12 Chris@10: Chris@10: Sat Jan 18 07:20:19 EST 2003 athena Chris@10: * [project @ 2003-01-18 12:20:18 by athena] Chris@10: Started working on verifier Chris@10: Chris@10: A ./libbench2/dotens2.c Chris@10: A ./libbench2/verify-dft.c Chris@10: A ./libbench2/verify-lib.c Chris@10: A ./libbench2/verify.h Chris@10: M ./libbench2/Makefile.am -3 +3 Chris@10: M ./libbench2/bench-user.h -1 +4 Chris@10: M ./libbench2/dotens2.c +55 Chris@10: M ./libbench2/tensor.c -1 +8 Chris@10: M ./libbench2/verify-dft.c +119 Chris@10: M ./libbench2/verify-lib.c +399 Chris@10: M ./libbench2/verify.c -2 +9 Chris@10: M ./libbench2/verify.h +65 Chris@10: M ./tests/Makefile.am -4 +1 Chris@10: Chris@10: Fri Jan 17 14:53:28 EST 2003 stevenj Chris@10: * [project @ 2003-01-17 19:53:28 by stevenj] Chris@10: added X(threads_cleanup) Chris@10: Chris@10: M ./api/fftw3.h -1 +2 Chris@10: M ./threads/api.c +9 Chris@10: M ./threads/threads.c -6 +19 Chris@10: M ./threads/threads.h -1 +2 Chris@10: Chris@10: Fri Jan 17 10:35:56 EST 2003 athena Chris@10: * [project @ 2003-01-17 15:35:56 by athena] Chris@10: Use C style for upper and lower array bounds. Free tensors properly. Chris@10: Chris@10: M ./libbench2/allocate.c -3 +5 Chris@10: M ./libbench2/tensor.c -2 +2 Chris@10: Chris@10: Fri Jan 17 08:50:42 EST 2003 athena Chris@10: * [project @ 2003-01-17 13:50:42 by athena] Chris@10: Fixed ambiguous syntax Chris@10: Chris@10: M ./libbench2/problem.c -2 +2 Chris@10: Chris@10: Fri Jan 17 08:20:57 EST 2003 athena Chris@10: * [project @ 2003-01-17 13:20:57 by athena] Chris@10: Parse minus sign, bugfixes Chris@10: Chris@10: M ./libbench2/problem.c -4 +17 Chris@10: Chris@10: Fri Jan 17 08:11:56 EST 2003 athena Chris@10: * [project @ 2003-01-17 13:11:56 by athena] Chris@10: Skeleton libbench2 implemented (probably still buggy) Chris@10: Chris@10: A ./libbench2/ Chris@10: A ./libbench2/Makefile.am Chris@10: A ./libbench2/allocate.c Chris@10: A ./libbench2/bench-main.c Chris@10: A ./libbench2/bench-user.h Chris@10: A ./libbench2/bench.h Chris@10: A ./libbench2/can-do.c Chris@10: A ./libbench2/caset.c Chris@10: A ./libbench2/getopt-utils.c Chris@10: A ./libbench2/getopt.c Chris@10: A ./libbench2/getopt.h Chris@10: A ./libbench2/info.c Chris@10: A ./libbench2/main.c Chris@10: A ./libbench2/mflops.c Chris@10: A ./libbench2/ovtpvt.c Chris@10: A ./libbench2/pow2.c Chris@10: A ./libbench2/problem.c Chris@10: A ./libbench2/report.c Chris@10: A ./libbench2/speed.c Chris@10: A ./libbench2/tensor.c Chris@10: A ./libbench2/timer.c Chris@10: A ./libbench2/util.c Chris@10: A ./libbench2/verify.c Chris@10: A ./libbench2/zero.c Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac -1 +1 Chris@10: M ./libbench2/Makefile.am +13 Chris@10: M ./libbench2/allocate.c +55 Chris@10: M ./libbench2/bench-main.c +250 Chris@10: M ./libbench2/bench-user.h +206 Chris@10: M ./libbench2/bench.h +67 Chris@10: M ./libbench2/can-do.c +33 Chris@10: M ./libbench2/caset.c +12 Chris@10: M ./libbench2/getopt-utils.c +104 Chris@10: M ./libbench2/getopt.c +1062 Chris@10: M ./libbench2/getopt.h +180 Chris@10: M ./libbench2/info.c +59 Chris@10: M ./libbench2/main.c +40 Chris@10: M ./libbench2/mflops.c +25 Chris@10: M ./libbench2/ovtpvt.c +13 Chris@10: M ./libbench2/pow2.c +6 Chris@10: M ./libbench2/problem.c +172 Chris@10: M ./libbench2/report.c +159 Chris@10: M ./libbench2/speed.c +73 Chris@10: M ./libbench2/tensor.c +173 Chris@10: M ./libbench2/timer.c +241 Chris@10: M ./libbench2/util.c +216 Chris@10: M ./libbench2/verify.c +53 Chris@10: M ./libbench2/zero.c +36 Chris@10: M ./tests/Makefile.am -2 +2 Chris@10: M ./tests/bench.c -56 +93 Chris@10: Chris@10: Fri Jan 17 04:23:37 EST 2003 athena Chris@10: * [project @ 2003-01-17 09:23:37 by athena] Chris@10: Formatting Chris@10: Chris@10: M ./kernel/tensor4.c -10 +10 Chris@10: Chris@10: Fri Jan 17 03:15:24 EST 2003 fftw Chris@10: * [project @ 2003-01-17 08:15:24 by fftw] Chris@10: slight updates Chris@10: Chris@10: M ./doc/fftw3.texi -27 +34 Chris@10: Chris@10: Fri Jan 17 01:44:44 EST 2003 stevenj Chris@10: * [project @ 2003-01-17 06:44:44 by stevenj] Chris@10: eliminated obsolete uimin/uimax Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/buffered.c -2 +2 Chris@10: M ./kernel/ifftw.h -3 +1 Chris@10: M ./kernel/minmax.c -11 +1 Chris@10: M ./kernel/tensor4.c -4 +4 Chris@10: M ./rdft/buffered2.c -4 +4 Chris@10: M ./rdft/rdft2-inplace-strides.c -2 +2 Chris@10: M ./rdft/rdft2-tensor-max-index.c -4 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Fri Jan 17 01:40:10 EST 2003 stevenj Chris@10: * [project @ 2003-01-17 06:40:10 by stevenj] Chris@10: threads needs to have its own library, lest all programs linking to libfftw3.so need -lpthread Chris@10: Chris@10: A ./threads/api.c Chris@10: M ./Makefile.am -1 Chris@10: M ./api/Makefile.am -4 +3 Chris@10: M ./api/configure.c -2 Chris@10: M ./api/fftw3.h -2 +3 Chris@10: R ./api/plan-with-nthreads.c Chris@10: M ./tests/Makefile.am -1 +8 Chris@10: M ./tests/bench.c +5 Chris@10: M ./threads/Makefile.am -9 +6 Chris@10: M ./threads/api.c +56 Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Thu Jan 16 19:53:46 EST 2003 stevenj Chris@10: * [project @ 2003-01-17 00:53:46 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/f77api.c -2 +2 Chris@10: Chris@10: Thu Jan 16 19:53:30 EST 2003 stevenj Chris@10: * [project @ 2003-01-17 00:53:30 by stevenj] Chris@10: better name Chris@10: Chris@10: M ./api/f77api.c -2 +2 Chris@10: Chris@10: Thu Jan 16 19:52:36 EST 2003 stevenj Chris@10: * [project @ 2003-01-17 00:52:36 by stevenj] Chris@10: added more functions Chris@10: Chris@10: M ./api/f77api.c +68 Chris@10: Chris@10: Thu Jan 16 16:57:06 EST 2003 stevenj Chris@10: * [project @ 2003-01-16 21:57:06 by stevenj] Chris@10: if 'long' is big enough, use it for mulmod in preference to 'long long' Chris@10: Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: Chris@10: Thu Jan 16 14:53:41 EST 2003 stevenj Chris@10: * [project @ 2003-01-16 19:53:41 by stevenj] Chris@10: use uintptr_t for pointer alignment arithmetic Chris@10: Chris@10: M ./configure.ac -2 +9 Chris@10: M ./kernel/align.c -2 +2 Chris@10: M ./kernel/ifftw.h -5 +22 Chris@10: Chris@10: Thu Jan 16 07:58:28 EST 2003 athena Chris@10: * [project @ 2003-01-16 12:58:28 by athena] Chris@10: More signed/unsigned cleanup Chris@10: Chris@10: M ./kernel/planner.c -10 +10 Chris@10: M ./kernel/print.c -3 +3 Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: Chris@10: Thu Jan 16 07:57:40 EST 2003 athena Chris@10: * [project @ 2003-01-16 12:57:40 by athena] Chris@10: null function pointers are technically nonportable Chris@10: Chris@10: M ./kernel/solvtab.c -2 +2 Chris@10: Chris@10: Thu Jan 16 07:17:45 EST 2003 athena Chris@10: * [project @ 2003-01-16 12:17:45 by athena] Chris@10: Free short_options Chris@10: Chris@10: M ./libbench/bench-main.c -2 +3 Chris@10: Chris@10: Thu Jan 16 05:48:30 EST 2003 athena Chris@10: * [project @ 2003-01-16 10:48:30 by athena] Chris@10: Oops, forgot STACK_FREE Chris@10: Chris@10: M ./kernel/alloc.c -15 +17 Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Thu Jan 16 05:40:39 EST 2003 athena Chris@10: * [project @ 2003-01-16 10:40:39 by athena] Chris@10: Do not require memalign() unless HAVE_SIMD Chris@10: Chris@10: M ./kernel/alloc.c -6 +15 Chris@10: M ./kernel/ifftw.h -13 +18 Chris@10: Chris@10: Thu Jan 16 01:03:31 EST 2003 stevenj Chris@10: * [project @ 2003-01-16 06:03:31 by stevenj] Chris@10: MS VC++ _aligned_malloc Chris@10: Chris@10: M ./kernel/alloc.c -1 +7 Chris@10: Chris@10: Thu Jan 16 00:44:45 EST 2003 stevenj Chris@10: * [project @ 2003-01-16 05:44:45 by stevenj] Chris@10: added api fftw_malloc/free Chris@10: Chris@10: M ./api/fftw3.h -2 +5 Chris@10: M ./kernel/alloc.c -2 +10 Chris@10: Chris@10: Thu Jan 16 00:43:48 EST 2003 stevenj Chris@10: * [project @ 2003-01-16 05:43:48 by stevenj] Chris@10: silence warning Chris@10: Chris@10: M ./api/map-r2r-kind.c -1 +1 Chris@10: Chris@10: Wed Jan 15 22:39:04 EST 2003 stevenj Chris@10: * [project @ 2003-01-16 03:39:04 by stevenj] Chris@10: send error output to stderr Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +1 Chris@10: Chris@10: Wed Jan 15 13:20:35 EST 2003 athena Chris@10: * [project @ 2003-01-15 18:20:35 by athena] Chris@10: Pure paranoia. Chris@10: Chris@10: M ./kernel/tensor7.c -3 +6 Chris@10: Chris@10: Wed Jan 15 06:51:34 EST 2003 athena Chris@10: * [project @ 2003-01-15 11:51:34 by athena] Chris@10: Fixed formatting that was messed up by the conversion uint->int. Chris@10: Ensure that iodims etc are kosher. Chris@10: Chris@10: A ./kernel/tensor9.c Chris@10: M ./api/api.h -13 +20 Chris@10: M ./api/apiplan.c -9 +9 Chris@10: M ./api/configure.c -5 +5 Chris@10: M ./api/dfthelp.c -1 +1 Chris@10: M ./api/execute-dft-c2r.c -1 +1 Chris@10: M ./api/execute-dft-r2c.c -1 +1 Chris@10: M ./api/execute-dft.c -1 +1 Chris@10: M ./api/execute-r2r.c -1 +1 Chris@10: M ./api/execute.c -1 +1 Chris@10: M ./api/export-wisdom-to-file.c -4 +4 Chris@10: M ./api/export-wisdom-to-string.c -6 +6 Chris@10: M ./api/export-wisdom.c -6 +6 Chris@10: M ./api/f77api.c -58 +46 Chris@10: M ./api/fftw3.h -14 +11 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -2 +2 Chris@10: M ./api/import-wisdom-from-file.c -5 +5 Chris@10: M ./api/import-wisdom-from-string.c -4 +4 Chris@10: M ./api/import-wisdom.c -6 +6 Chris@10: M ./api/map-r2r-kind.c -37 +17 Chris@10: M ./api/mapflags.c -4 +4 Chris@10: M ./api/mkprinter-file.c -2 +2 Chris@10: M ./api/mktensor-iodims.c -2 +23 Chris@10: M ./api/mktensor-rowmajor.c -4 +20 Chris@10: M ./api/plan-dft-1d.c -2 +2 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -3 +3 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -3 +3 Chris@10: M ./api/plan-dft-c2r.c -3 +3 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -3 +3 Chris@10: M ./api/plan-dft-r2c.c -3 +5 Chris@10: M ./api/plan-dft.c -4 +6 Chris@10: M ./api/plan-guru-dft-c2r.c -10 +10 Chris@10: M ./api/plan-guru-dft-r2c.c -10 +11 Chris@10: M ./api/plan-guru-dft.c -10 +11 Chris@10: M ./api/plan-guru-r2r.c -14 +16 Chris@10: M ./api/plan-many-dft-c2r.c -34 +21 Chris@10: M ./api/plan-many-dft-r2c.c -34 +21 Chris@10: M ./api/plan-many-dft.c -18 +19 Chris@10: M ./api/plan-many-r2r.c -22 +21 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -3 +3 Chris@10: M ./api/plan-r2r-3d.c -4 +4 Chris@10: M ./api/plan-r2r.c -4 +4 Chris@10: M ./api/plan-with-nthreads.c -3 +3 Chris@10: M ./api/print-plan.c -3 +3 Chris@10: M ./api/rdft2-pad.c -2 +2 Chris@10: M ./api/the-planner.c -5 +5 Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./dft/problem.c -4 +5 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./dft/vrank2-transpose.c -2 +2 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -10 +10 Chris@10: M ./kernel/print.c -3 +3 Chris@10: M ./kernel/tensor.c -2 +4 Chris@10: M ./kernel/tensor9.c +37 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/direct.c -2 +2 Chris@10: M ./rdft/direct2.c -2 +2 Chris@10: M ./rdft/generic.c -1 +1 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/problem.c -2 +6 Chris@10: M ./rdft/problem2.c -2 +5 Chris@10: M ./rdft/rader-hc2hc.c -1 +1 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./rdft/vrank2-transpose.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -2 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -2 +2 Chris@10: Chris@10: Wed Jan 15 01:32:18 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 06:32:18 by stevenj] Chris@10: added version stamp Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -1 +2 Chris@10: Chris@10: Wed Jan 15 01:28:20 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 06:28:20 by stevenj] Chris@10: added warning Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in +1 Chris@10: Chris@10: Wed Jan 15 01:23:25 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 06:23:25 by stevenj] Chris@10: add fftw-wisdom-to-conf to BUILT_SOURCES Chris@10: Chris@10: M ./tools/Makefile.am -1 +1 Chris@10: Chris@10: Wed Jan 15 01:09:29 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 06:09:29 by stevenj] Chris@10: added const Chris@10: Chris@10: M ./tools/fftw-wisdom-to-conf.in -2 +2 Chris@10: Chris@10: Wed Jan 15 01:04:10 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 06:04:10 by stevenj] Chris@10: added wisdom-to-conf Chris@10: Chris@10: A ./tools/ Chris@10: A ./tools/Makefile.am Chris@10: A ./tools/fftw-wisdom-to-conf.in Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +3 Chris@10: M ./tools/Makefile.am +3 Chris@10: M ./tools/fftw-wisdom-to-conf.in +37 Chris@10: Chris@10: Wed Jan 15 00:23:36 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 05:23:36 by stevenj] Chris@10: include type prefix in wisdom preamble Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Wed Jan 15 00:02:31 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 05:02:31 by stevenj] Chris@10: updates Chris@10: Chris@10: M ./TODO -4 +2 Chris@10: Chris@10: Tue Jan 14 23:59:26 EST 2003 stevenj Chris@10: * [project @ 2003-01-15 04:59:26 by stevenj] Chris@10: check the_plan before printing Chris@10: Chris@10: M ./tests/bench.c -2 +3 Chris@10: Chris@10: Tue Jan 14 21:10:25 EST 2003 athena Chris@10: * [project @ 2003-01-15 02:10:25 by athena] Chris@10: Eliminated those unsigned values that would break LP64 machines. Chris@10: Chris@10: M ./ChangeLog -1 +1 Chris@10: M ./api/api.h -16 +16 Chris@10: M ./api/apiplan.c -19 +19 Chris@10: M ./api/configure.c -5 +5 Chris@10: M ./api/dfthelp.c -5 +5 Chris@10: M ./api/execute-dft-c2r.c -1 +1 Chris@10: M ./api/execute-dft-r2c.c -1 +1 Chris@10: M ./api/execute-dft.c -1 +1 Chris@10: M ./api/execute-r2r.c -1 +1 Chris@10: M ./api/execute.c -1 +1 Chris@10: M ./api/export-wisdom-to-file.c -4 +4 Chris@10: M ./api/export-wisdom-to-string.c -13 +13 Chris@10: M ./api/export-wisdom.c -7 +7 Chris@10: M ./api/f77api.c -71 +69 Chris@10: M ./api/fftw3.h -134 +136 Chris@10: M ./api/forget-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c -8 +8 Chris@10: M ./api/import-wisdom-from-file.c -6 +6 Chris@10: M ./api/import-wisdom-from-string.c -6 +6 Chris@10: M ./api/import-wisdom.c -9 +11 Chris@10: M ./api/map-r2r-kind.c -17 +40 Chris@10: M ./api/mapflags.c -18 +18 Chris@10: M ./api/mkprinter-file.c -3 +3 Chris@10: M ./api/mktensor-iodims.c -9 +9 Chris@10: M ./api/mktensor-rowmajor.c -6 +5 Chris@10: M ./api/plan-dft-1d.c -3 +2 Chris@10: M ./api/plan-dft-2d.c -5 +5 Chris@10: M ./api/plan-dft-3d.c -5 +7 Chris@10: M ./api/plan-dft-c2r-1d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-2d.c -5 +5 Chris@10: M ./api/plan-dft-c2r-3d.c -5 +7 Chris@10: M ./api/plan-dft-c2r.c -6 +3 Chris@10: M ./api/plan-dft-r2c-1d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-2d.c -5 +5 Chris@10: M ./api/plan-dft-r2c-3d.c -5 +7 Chris@10: M ./api/plan-dft-r2c.c -6 +3 Chris@10: M ./api/plan-dft.c -6 +4 Chris@10: M ./api/plan-guru-dft-c2r.c -10 +10 Chris@10: M ./api/plan-guru-dft-r2c.c -10 +10 Chris@10: M ./api/plan-guru-dft.c -10 +10 Chris@10: M ./api/plan-guru-r2r.c -13 +14 Chris@10: M ./api/plan-many-dft-c2r.c -23 +36 Chris@10: M ./api/plan-many-dft-r2c.c -23 +36 Chris@10: M ./api/plan-many-dft.c -17 +18 Chris@10: M ./api/plan-many-r2r.c -18 +22 Chris@10: M ./api/plan-r2r-1d.c -3 +2 Chris@10: M ./api/plan-r2r-2d.c -7 +8 Chris@10: M ./api/plan-r2r-3d.c -8 +11 Chris@10: M ./api/plan-r2r.c -3 +4 Chris@10: M ./api/plan-with-nthreads.c -7 +7 Chris@10: M ./api/print-plan.c -3 +3 Chris@10: M ./api/rdft2-pad.c -10 +9 Chris@10: M ./api/the-planner.c -7 +7 Chris@10: M ./configure.ac -9 +3 Chris@10: M ./dft/buffered.c -11 +11 Chris@10: M ./dft/codelet-dft.h -11 +11 Chris@10: M ./dft/codelets/n.c -1 +1 Chris@10: M ./dft/codelets/t.c -1 +1 Chris@10: M ./dft/ct-dif.c -4 +4 Chris@10: M ./dft/ct-dit.c -4 +4 Chris@10: M ./dft/ct-ditbuf.c -6 +6 Chris@10: M ./dft/ct-ditf.c -2 +2 Chris@10: M ./dft/ct.c -5 +5 Chris@10: M ./dft/ct.h -2 +2 Chris@10: M ./dft/dft.h -2 +2 Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./dft/generic.c -5 +5 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/k7/k7.c -9 +9 Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./dft/rader-omega.c -2 +2 Chris@10: M ./dft/rader.c -13 +13 Chris@10: M ./dft/rank-geq2.c -9 +9 Chris@10: M ./dft/rank0.c -7 +7 Chris@10: M ./dft/simd/n1b.c -1 +1 Chris@10: M ./dft/simd/n1f.c -1 +1 Chris@10: M ./dft/simd/t1b.c -1 +1 Chris@10: M ./dft/simd/t1f.c -1 +1 Chris@10: M ./dft/vrank-geq1.c -11 +11 Chris@10: M ./dft/vrank2-transpose.c -4 +4 Chris@10: M ./dft/vrank3-transpose.c -10 +10 Chris@10: M ./dft/zero.c -3 +3 Chris@10: M ./genfft/gen_hc2hc.ml -4 +4 Chris@10: M ./genfft/gen_hc2r.ml -4 +4 Chris@10: M ./genfft/gen_notw.ml -4 +4 Chris@10: M ./genfft/gen_notw_c.ml -4 +4 Chris@10: M ./genfft/gen_r2hc.ml -4 +4 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twiddle_c.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./kernel/align.c -5 +5 Chris@10: M ./kernel/buffered.c -3 +3 Chris@10: M ./kernel/ct.c -1 +1 Chris@10: M ./kernel/hash.c -3 +3 Chris@10: M ./kernel/iabs.c -3 +3 Chris@10: M ./kernel/ifftw.h -71 +70 Chris@10: M ./kernel/md5-1.c -3 +3 Chris@10: M ./kernel/md5.c -4 +4 Chris@10: M ./kernel/minmax.c -3 +3 Chris@10: M ./kernel/ops.c -4 +4 Chris@10: M ./kernel/pickdim.c -6 +6 Chris@10: M ./kernel/planner.c -32 +32 Chris@10: M ./kernel/primes.c -14 +14 Chris@10: M ./kernel/print.c -5 +5 Chris@10: M ./kernel/rader.c -3 +3 Chris@10: M ./kernel/scan.c -3 +3 Chris@10: M ./kernel/tensor.c -9 +9 Chris@10: M ./kernel/tensor1.c -2 +2 Chris@10: M ./kernel/tensor2.c -3 +3 Chris@10: M ./kernel/tensor4.c -10 +10 Chris@10: M ./kernel/tensor5.c -6 +6 Chris@10: M ./kernel/tensor7.c -4 +4 Chris@10: M ./kernel/trig.c -4 +4 Chris@10: M ./kernel/twiddle.c -14 +14 Chris@10: M ./libbench/acopy.c -2 +2 Chris@10: M ./libbench/allocate.c -2 +2 Chris@10: M ./libbench/ascale.c -2 +2 Chris@10: M ./libbench/aset.c -2 +2 Chris@10: M ./libbench/bench-user.h -23 +23 Chris@10: M ./libbench/bench.h -2 +2 Chris@10: M ./libbench/caadd.c -2 +2 Chris@10: M ./libbench/cacopy.c -2 +2 Chris@10: M ./libbench/cascale.c -2 +2 Chris@10: M ./libbench/caset.c -2 +2 Chris@10: M ./libbench/casub.c -2 +2 Chris@10: M ./libbench/copy-c2h-1d-fftpack.c -1 +1 Chris@10: M ./libbench/copy-c2h-1d-halfcomplex.c -1 +1 Chris@10: M ./libbench/copy-c2h-1d-packed.c -2 +2 Chris@10: M ./libbench/copy-c2h-1d-unpacked-ri.c -1 +1 Chris@10: M ./libbench/copy-c2h-unpacked.c -1 +1 Chris@10: M ./libbench/copy-c2r-packed.c -2 +2 Chris@10: M ./libbench/copy-c2r-unpacked.c -1 +1 Chris@10: M ./libbench/copy-c2ri.c -2 +2 Chris@10: M ./libbench/copy-h2c-1d-fftpack.c -1 +1 Chris@10: M ./libbench/copy-h2c-1d-halfcomplex.c -1 +1 Chris@10: M ./libbench/copy-h2c-1d-packed.c -2 +2 Chris@10: M ./libbench/copy-h2c-1d-unpacked-ri.c -1 +1 Chris@10: M ./libbench/copy-h2c-unpacked.c -9 +9 Chris@10: M ./libbench/copy-r2c-packed.c -2 +2 Chris@10: M ./libbench/copy-r2c-unpacked.c -1 +1 Chris@10: M ./libbench/copy-ri2c.c -2 +2 Chris@10: M ./libbench/getopt-utils.c -2 +2 Chris@10: M ./libbench/getopt.c -3 +3 Chris@10: M ./libbench/log2.c -2 +2 Chris@10: M ./libbench/mp.c -20 +20 Chris@10: M ./libbench/pow2.c -1 +1 Chris@10: M ./libbench/prime.c -4 +4 Chris@10: M ./libbench/problem.c -2 +2 Chris@10: M ./libbench/timer.c -2 +2 Chris@10: M ./libbench/verify.c -36 +36 Chris@10: M ./rdft/buffered.c -10 +10 Chris@10: M ./rdft/buffered2.c -22 +22 Chris@10: M ./rdft/codelet-rdft.h -17 +17 Chris@10: M ./rdft/codelets/hc2r.c -1 +1 Chris@10: M ./rdft/codelets/hfb.c -1 +1 Chris@10: M ./rdft/codelets/r2hc.c -1 +1 Chris@10: M ./rdft/dft-r2hc.c -4 +4 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./rdft/dht-rader.c -6 +6 Chris@10: M ./rdft/direct.c -6 +6 Chris@10: M ./rdft/direct2.c -5 +5 Chris@10: M ./rdft/generic.c -8 +8 Chris@10: M ./rdft/hc2hc-buf.c -7 +7 Chris@10: M ./rdft/hc2hc-dif.c -4 +4 Chris@10: M ./rdft/hc2hc-dit.c -4 +4 Chris@10: M ./rdft/hc2hc.c -5 +5 Chris@10: M ./rdft/hc2hc.h -2 +2 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: M ./rdft/problem.c -11 +11 Chris@10: M ./rdft/problem2.c -3 +3 Chris@10: M ./rdft/rader-hc2hc.c -12 +12 Chris@10: M ./rdft/rank-geq2-rdft2.c -9 +9 Chris@10: M ./rdft/rank-geq2.c -9 +9 Chris@10: M ./rdft/rank0-rdft2.c -4 +4 Chris@10: M ./rdft/rank0.c -5 +5 Chris@10: M ./rdft/rdft-dht.c -5 +5 Chris@10: M ./rdft/rdft.h -4 +4 Chris@10: M ./rdft/rdft2-inplace-strides.c -4 +4 Chris@10: M ./rdft/rdft2-radix2.c -15 +15 Chris@10: M ./rdft/rdft2-tensor-max-index.c -4 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -12 +12 Chris@10: M ./rdft/vrank-geq1.c -11 +11 Chris@10: M ./rdft/vrank2-transpose.c -4 +4 Chris@10: M ./rdft/vrank3-transpose.c -10 +10 Chris@10: M ./reodft/redft00e-r2hc.c -7 +7 Chris@10: M ./reodft/reodft010e-r2hc.c -16 +16 Chris@10: M ./reodft/reodft11e-r2hc.c -10 +10 Chris@10: M ./reodft/rodft00e-r2hc.c -7 +7 Chris@10: M ./simd/3dnow.c -5 +5 Chris@10: M ./simd/sse.c -3 +3 Chris@10: M ./simd/sse2.c -3 +3 Chris@10: M ./tests/bench.c -13 +7 Chris@10: M ./tests/dotens.c -3 +3 Chris@10: M ./tests/dotens2.c -3 +3 Chris@10: M ./tests/trigtest.c -7 +7 Chris@10: M ./tests/verify-dft.c -4 +4 Chris@10: M ./tests/verify-lib.c -41 +41 Chris@10: M ./tests/verify-rdft.c -13 +13 Chris@10: M ./tests/verify-reodft.c -48 +48 Chris@10: M ./tests/verify.h -15 +15 Chris@10: M ./threads/ct-dit.c -5 +5 Chris@10: M ./threads/dft-vrank-geq1.c -15 +15 Chris@10: M ./threads/hc2hc-dif.c -6 +6 Chris@10: M ./threads/hc2hc-dit.c -6 +6 Chris@10: M ./threads/rdft-vrank-geq1.c -15 +15 Chris@10: M ./threads/threads.c -4 +4 Chris@10: M ./threads/threads.h -3 +3 Chris@10: M ./threads/vrank-geq1-rdft2.c -15 +15 Chris@10: Chris@10: Tue Jan 14 15:14:29 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 20:14:29 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./kernel/primes.c -4 +4 Chris@10: Chris@10: Tue Jan 14 08:00:08 EST 2003 athena Chris@10: * [project @ 2003-01-14 13:00:08 by athena] Chris@10: Oops Chris@10: Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./rdft/generic.c -1 +1 Chris@10: Chris@10: Tue Jan 14 07:59:14 EST 2003 athena Chris@10: * [project @ 2003-01-14 12:59:14 by athena] Chris@10: int/uint confusion Chris@10: Chris@10: M ./dft/generic.c -1 +2 Chris@10: M ./rdft/generic.c -1 +2 Chris@10: Chris@10: Tue Jan 14 02:25:33 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 07:25:33 by stevenj] Chris@10: updated introduction and some organization Chris@10: Chris@10: M ./doc/fftw3.texi -79 +117 Chris@10: Chris@10: Tue Jan 14 01:34:46 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 06:34:46 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/f77api.c -18 +18 Chris@10: Chris@10: Tue Jan 14 01:33:04 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 06:33:04 by stevenj] Chris@10: newline Chris@10: Chris@10: M ./Makefile.am +1 Chris@10: Chris@10: Tue Jan 14 00:23:04 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 05:23:04 by stevenj] Chris@10: added win32 timer Chris@10: Chris@10: M ./libbench/timer.c -1 +23 Chris@10: Chris@10: Tue Jan 14 00:12:21 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 05:12:21 by stevenj] Chris@10: sync with kernel/alloc.c Chris@10: Chris@10: M ./libbench/util.c -2 +5 Chris@10: Chris@10: Tue Jan 14 00:03:20 EST 2003 stevenj Chris@10: * [project @ 2003-01-14 05:03:20 by stevenj] Chris@10: handle missing F77_FUNC_ Chris@10: Chris@10: M ./api/f77api.c +4 Chris@10: Chris@10: Mon Jan 13 17:42:50 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 22:42:50 by stevenj] Chris@10: used fint instead of int to make Fortran integer type easier to change Chris@10: Chris@10: M ./api/f77api.c -24 +31 Chris@10: Chris@10: Mon Jan 13 17:38:56 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 22:38:56 by stevenj] Chris@10: slight abbreviation Chris@10: Chris@10: M ./api/f77api.c -4 +4 Chris@10: Chris@10: Mon Jan 13 17:35:20 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 22:35:20 by stevenj] Chris@10: the great lengthening, part I: int -> long in api; mv mktensor-rowmajor to api Chris@10: Chris@10: A ./api/mktensor-rowmajor.c Chris@10: M ./api/Makefile.am -2 +2 Chris@10: M ./api/api.h -2 +5 Chris@10: M ./api/f77api.c -7 +7 Chris@10: M ./api/fftw3.h -44 +46 Chris@10: M ./api/mktensor-rowmajor.c +44 Chris@10: M ./api/plan-dft-1d.c -1 +1 Chris@10: M ./api/plan-dft-2d.c -2 +2 Chris@10: M ./api/plan-dft-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-1d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-2d.c -2 +2 Chris@10: M ./api/plan-dft-c2r-3d.c -2 +2 Chris@10: M ./api/plan-dft-c2r.c -1 +1 Chris@10: M ./api/plan-dft-r2c-1d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-2d.c -2 +2 Chris@10: M ./api/plan-dft-r2c-3d.c -2 +2 Chris@10: M ./api/plan-dft-r2c.c -1 +1 Chris@10: M ./api/plan-dft.c -1 +1 Chris@10: M ./api/plan-many-dft-c2r.c -7 +7 Chris@10: M ./api/plan-many-dft-r2c.c -7 +7 Chris@10: M ./api/plan-many-dft.c -6 +6 Chris@10: M ./api/plan-many-r2r.c -6 +6 Chris@10: M ./api/plan-r2r-1d.c -1 +1 Chris@10: M ./api/plan-r2r-2d.c -2 +2 Chris@10: M ./api/plan-r2r-3d.c -2 +2 Chris@10: M ./api/plan-r2r.c -1 +1 Chris@10: M ./api/rdft2-pad.c -4 +4 Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -4 +1 Chris@10: R ./kernel/tensor3.c Chris@10: M ./tests/bench.c -1 +9 Chris@10: Chris@10: Mon Jan 13 15:23:22 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 20:23:22 by stevenj] Chris@10: long types Chris@10: Chris@10: M ./configure.ac +4 Chris@10: Chris@10: Mon Jan 13 04:20:37 EST 2003 athena Chris@10: * [project @ 2003-01-13 09:20:36 by athena] Chris@10: Renamed fftw_malloc -> MALLOC, X(free) -> X(ifree), X(free0) -> Chris@10: X(ifree0), non_fftw_malloc -> NATIVE_MALLOC Chris@10: Chris@10: M ./api/apiplan.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -1 +1 Chris@10: M ./api/f77api.c -8 +8 Chris@10: M ./api/map-r2r-kind.c -1 +1 Chris@10: M ./api/plan-guru-r2r.c -1 +1 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: M ./api/plan-many-r2r.c -1 +1 Chris@10: M ./api/rdft2-pad.c -1 +1 Chris@10: M ./dft/buffered.c -6 +6 Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./dft/rader-omega.c -1 +1 Chris@10: M ./dft/rader.c -10 +10 Chris@10: M ./kernel/alloc.c -6 +6 Chris@10: M ./kernel/ifftw.h -11 +11 Chris@10: M ./kernel/plan.c -3 +3 Chris@10: M ./kernel/planner.c -12 +12 Chris@10: M ./kernel/print.c -3 +3 Chris@10: M ./kernel/problem.c -2 +2 Chris@10: M ./kernel/rader.c -3 +3 Chris@10: M ./kernel/scan.c -3 +3 Chris@10: M ./kernel/solver.c -3 +3 Chris@10: M ./kernel/stride.c -3 +3 Chris@10: M ./kernel/tensor.c -9 +9 Chris@10: M ./kernel/twiddle.c -5 +5 Chris@10: M ./rdft/buffered.c -6 +6 Chris@10: M ./rdft/buffered2.c -8 +8 Chris@10: M ./rdft/dht-rader.c -6 +6 Chris@10: M ./rdft/generic.c -1 +1 Chris@10: M ./rdft/problem.c -4 +4 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: M ./rdft/rader-hc2hc.c -10 +10 Chris@10: M ./reodft/redft00e-r2hc.c -5 +5 Chris@10: M ./reodft/reodft010e-r2hc.c -11 +11 Chris@10: M ./reodft/reodft11e-r2hc.c -7 +7 Chris@10: M ./reodft/rodft00e-r2hc.c -5 +5 Chris@10: M ./tests/verify-dft.c -15 +15 Chris@10: M ./tests/verify-rdft.c -29 +29 Chris@10: M ./tests/verify-reodft.c -15 +15 Chris@10: M ./threads/dft-vrank-geq1.c -4 +4 Chris@10: M ./threads/rdft-vrank-geq1.c -4 +4 Chris@10: M ./threads/threads.c -1 +1 Chris@10: M ./threads/vrank-geq1-rdft2.c -4 +4 Chris@10: Chris@10: Mon Jan 13 02:37:22 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 07:37:22 by stevenj] Chris@10: added beginning of Fortran interface Chris@10: Chris@10: A ./api/f77api.c Chris@10: M ./api/Makefile.am -1 +1 Chris@10: M ./api/f77api.c +145 Chris@10: Chris@10: Mon Jan 13 01:05:29 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 06:05:29 by stevenj] Chris@10: add fortran mangling check Chris@10: Chris@10: M ./configure.ac +18 Chris@10: Chris@10: Mon Jan 13 00:33:28 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 05:33:28 by stevenj] Chris@10: added guru r2r interface Chris@10: Chris@10: A ./api/execute-r2r.c Chris@10: A ./api/plan-guru-r2r.c Chris@10: M ./api/Makefile.am -3 +4 Chris@10: M ./api/execute-r2r.c +29 Chris@10: M ./api/fftw3.h -1 +8 Chris@10: M ./api/plan-guru-r2r.c +42 Chris@10: Chris@10: Mon Jan 13 00:23:26 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 05:23:26 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/fftw3.h -5 +7 Chris@10: M ./api/plan-r2r-1d.c -2 +2 Chris@10: M ./api/plan-r2r-2d.c -1 +4 Chris@10: M ./api/plan-r2r-3d.c -2 +5 Chris@10: Chris@10: Mon Jan 13 00:16:20 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 05:16:20 by stevenj] Chris@10: added r2r planner Chris@10: Chris@10: A ./api/map-r2r-kind.c Chris@10: A ./api/plan-many-r2r.c Chris@10: A ./api/plan-r2r-1d.c Chris@10: A ./api/plan-r2r-2d.c Chris@10: A ./api/plan-r2r-3d.c Chris@10: A ./api/plan-r2r.c Chris@10: M ./api/Makefile.am -2 +3 Chris@10: M ./api/fftw3.h -1 +29 Chris@10: M ./api/map-r2r-kind.c +47 Chris@10: M ./api/plan-many-r2r.c +48 Chris@10: M ./api/plan-r2r-1d.c +27 Chris@10: M ./api/plan-r2r-2d.c +29 Chris@10: M ./api/plan-r2r-3d.c +30 Chris@10: M ./api/plan-r2r.c +27 Chris@10: Chris@10: Sun Jan 12 22:58:18 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 03:58:18 by stevenj] Chris@10: more long-double checks Chris@10: Chris@10: M ./configure.ac -2 +13 Chris@10: Chris@10: Sun Jan 12 20:01:51 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 01:01:51 by stevenj] Chris@10: slight regrouping Chris@10: Chris@10: M ./kernel/planner.c -5 +7 Chris@10: Chris@10: Sun Jan 12 19:58:46 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 00:58:46 by stevenj] Chris@10: added joke Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Sun Jan 12 19:53:58 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 00:53:58 by stevenj] Chris@10: simplified rdft2 padding Chris@10: Chris@10: A ./api/rdft2-pad.c Chris@10: M ./api/Makefile.am -7 +7 Chris@10: M ./api/api.h -4 +2 Chris@10: R ./api/mktensor-rowmajor-pad.c Chris@10: M ./api/plan-many-dft-c2r.c -19 +16 Chris@10: M ./api/plan-many-dft-r2c.c -6 +18 Chris@10: M ./api/rdft2-pad.c +40 Chris@10: Chris@10: Sun Jan 12 19:02:09 EST 2003 stevenj Chris@10: * [project @ 2003-01-13 00:02:09 by stevenj] Chris@10: added comment Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: Chris@10: Sun Jan 12 18:54:49 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 23:54:49 by stevenj] Chris@10: use latest api Chris@10: Chris@10: M ./tests/bench.c -2 +1 Chris@10: Chris@10: Sun Jan 12 18:49:58 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 23:49:58 by stevenj] Chris@10: nembed should only be in advanced (many) interface, not basic interface...only a handful of people over the years have ever requested that functionality. Chris@10: Chris@10: M ./api/fftw3.h -10 +4 Chris@10: M ./api/plan-dft-1d.c -1 +1 Chris@10: M ./api/plan-dft-2d.c -1 +1 Chris@10: M ./api/plan-dft-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-1d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-2d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r.c -5 +3 Chris@10: M ./api/plan-dft-r2c-1d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-2d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-3d.c -1 +1 Chris@10: M ./api/plan-dft-r2c.c -5 +3 Chris@10: M ./api/plan-dft.c -5 +3 Chris@10: Chris@10: Sun Jan 12 18:41:57 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 23:41:57 by stevenj] Chris@10: impatient is default; generalize mapping functions using xor trick Chris@10: Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./api/mapflags.c -55 +61 Chris@10: Chris@10: Sun Jan 12 14:39:42 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 19:39:41 by stevenj] Chris@10: use NULL nembed to signal padding Chris@10: Chris@10: M ./api/mktensor-rowmajor-pad.c -5 +9 Chris@10: M ./api/plan-dft-c2r-1d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-2d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-3d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-1d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-2d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-3d.c -1 +1 Chris@10: Chris@10: Sun Jan 12 14:23:00 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 19:23:00 by stevenj] Chris@10: accept NULL nembed Chris@10: Chris@10: M ./api/plan-many-dft.c -1 +4 Chris@10: Chris@10: Sun Jan 12 13:57:13 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 18:57:13 by stevenj] Chris@10: added execute-dft-r2c/c2r Chris@10: Chris@10: A ./api/execute-dft-c2r.c Chris@10: A ./api/execute-dft-r2c.c Chris@10: M ./api/Makefile.am -5 +6 Chris@10: M ./api/execute-dft-c2r.c +29 Chris@10: M ./api/execute-dft-r2c.c +29 Chris@10: M ./api/fftw3.h -1 +4 Chris@10: Chris@10: Sun Jan 12 13:43:20 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 18:43:20 by stevenj] Chris@10: don't need dft.h Chris@10: Chris@10: M ./api/plan-dft.c -1 Chris@10: Chris@10: Sun Jan 12 13:22:14 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 18:22:14 by stevenj] Chris@10: tensors are compressed in the problem, duh Chris@10: Chris@10: M ./api/plan-many-dft-c2r.c -2 +1 Chris@10: M ./api/plan-many-dft-r2c.c -2 +1 Chris@10: Chris@10: Sun Jan 12 12:45:26 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 17:45:26 by stevenj] Chris@10: noted that posix_memalign bug is now fixed, thanks to bug report by yours truly Chris@10: Chris@10: M ./kernel/alloc.c -2 +3 Chris@10: Chris@10: Sun Jan 12 12:44:43 EST 2003 athena Chris@10: * [project @ 2003-01-12 17:44:43 by athena] Chris@10: Bug: n[3] instead of n[2]. Bug was propagated by copy-and-paste. Chris@10: Grrr... Chris@10: Chris@10: M ./api/plan-dft-3d.c -1 +1 Chris@10: M ./api/plan-dft-c2r-3d.c -1 +1 Chris@10: M ./api/plan-dft-r2c-3d.c -1 +1 Chris@10: Chris@10: Sun Jan 12 12:41:43 EST 2003 athena Chris@10: * [project @ 2003-01-12 17:41:43 by athena] Chris@10: Express plan_dft() in terms of plan_many_dft() Chris@10: Chris@10: M ./api/plan-dft.c -11 +4 Chris@10: Chris@10: Sun Jan 12 12:19:53 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 17:19:38 by stevenj] Chris@10: whoops Chris@10: Chris@10: A ./api/plan-guru-dft-c2r.c Chris@10: A ./api/plan-guru-dft-r2c.c Chris@10: A ./api/plan-guru-dft.c Chris@10: M ./api/plan-guru-dft-c2r.c +36 Chris@10: M ./api/plan-guru-dft-r2c.c +35 Chris@10: M ./api/plan-guru-dft.c +35 Chris@10: Chris@10: Sun Jan 12 06:00:46 EST 2003 athena Chris@10: * [project @ 2003-01-12 11:00:46 by athena] Chris@10: Manual skeleton. Chris@10: Chris@10: A ./doc/ Chris@10: A ./doc/Makefile.am Chris@10: A ./doc/fftw3.texi Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +1 Chris@10: M ./doc/Makefile.am -1 +2 Chris@10: M ./doc/fftw3.texi +318 Chris@10: M ./genfft-k7/vK7Optimization.ml -1 +1 Chris@10: Chris@10: Sat Jan 11 23:46:34 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 04:46:34 by stevenj] Chris@10: added r2c/c2r guru api Chris@10: Chris@10: M ./api/Makefile.am -1 +2 Chris@10: M ./api/fftw3.h -1 +12 Chris@10: Chris@10: Sat Jan 11 23:42:10 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 04:42:10 by stevenj] Chris@10: FFTW_DESTROY_INPUT is default for c2r transforms Chris@10: Chris@10: M ./api/plan-many-dft-c2r.c -1 +1 Chris@10: Chris@10: Sat Jan 11 23:36:26 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 04:36:26 by stevenj] Chris@10: added more of r2c/c2r api Chris@10: Chris@10: A ./api/plan-dft-c2r-1d.c Chris@10: A ./api/plan-dft-c2r-2d.c Chris@10: A ./api/plan-dft-c2r-3d.c Chris@10: A ./api/plan-dft-c2r.c Chris@10: A ./api/plan-dft-r2c-1d.c Chris@10: A ./api/plan-dft-r2c-2d.c Chris@10: A ./api/plan-dft-r2c-3d.c Chris@10: A ./api/plan-dft-r2c.c Chris@10: M ./api/Makefile.am -1 +3 Chris@10: M ./api/fftw3.h -69 +90 Chris@10: M ./api/plan-dft-c2r-1d.c +26 Chris@10: M ./api/plan-dft-c2r-2d.c +29 Chris@10: M ./api/plan-dft-c2r-3d.c +29 Chris@10: M ./api/plan-dft-c2r.c +32 Chris@10: M ./api/plan-dft-r2c-1d.c +26 Chris@10: M ./api/plan-dft-r2c-2d.c +29 Chris@10: M ./api/plan-dft-r2c-3d.c +29 Chris@10: M ./api/plan-dft-r2c.c +32 Chris@10: Chris@10: Sat Jan 11 21:09:41 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 02:09:41 by stevenj] Chris@10: r2c doesn't have adjustible sign Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./api/plan-many-dft-c2r.c -2 +2 Chris@10: M ./api/plan-many-dft-r2c.c -2 +2 Chris@10: Chris@10: Sat Jan 11 21:07:55 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 02:07:55 by stevenj] Chris@10: note that copyright year is out of date Chris@10: Chris@10: M ./TODO -1 +1 Chris@10: Chris@10: Sat Jan 11 21:04:23 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 02:04:23 by stevenj] Chris@10: updated api for r2c Chris@10: Chris@10: M ./api/fftw3.h -1 +17 Chris@10: Chris@10: Sat Jan 11 21:00:07 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 02:00:07 by stevenj] Chris@10: removed annoying nophys == niphys case Chris@10: Chris@10: M ./api/mktensor-rowmajor-pad.c -11 Chris@10: Chris@10: Sat Jan 11 20:58:13 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 01:58:13 by stevenj] Chris@10: added basic r2c/c2r planner Chris@10: Chris@10: A ./api/mktensor-rowmajor-pad.c Chris@10: A ./api/plan-many-dft-c2r.c Chris@10: A ./api/plan-many-dft-r2c.c Chris@10: M ./api/Makefile.am -2 +3 Chris@10: M ./api/api.h +5 Chris@10: M ./api/mktensor-rowmajor-pad.c +88 Chris@10: M ./api/plan-many-dft-c2r.c +59 Chris@10: M ./api/plan-many-dft-r2c.c +44 Chris@10: Chris@10: Sat Jan 11 19:34:14 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 00:34:14 by stevenj] Chris@10: dist should be in terms of complex values Chris@10: Chris@10: M ./api/plan-many-dft.c -1 +1 Chris@10: Chris@10: Sat Jan 11 19:14:24 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 00:14:24 by stevenj] Chris@10: added plan-with-nthreads Chris@10: Chris@10: M ./api/fftw3.h -2 +5 Chris@10: Chris@10: Sat Jan 11 19:12:51 EST 2003 stevenj Chris@10: * [project @ 2003-01-12 00:12:51 by stevenj] Chris@10: added function to set nthr Chris@10: Chris@10: A ./api/plan-with-nthreads.c Chris@10: M ./api/Makefile.am -1 +1 Chris@10: M ./api/plan-with-nthreads.c +39 Chris@10: Chris@10: Sat Jan 11 18:04:57 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 23:04:57 by stevenj] Chris@10: slight cleanup Chris@10: Chris@10: M ./api/fftw3.h -4 +5 Chris@10: Chris@10: Sat Jan 11 17:57:29 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 22:57:29 by stevenj] Chris@10: whoops Chris@10: Chris@10: A ./api/mktensor-iodims.c Chris@10: Chris@10: Sat Jan 11 17:55:39 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 22:55:39 by stevenj] Chris@10: maxlen is maximum string length, not including null termination Chris@10: Chris@10: M ./kernel/scan.c -2 +2 Chris@10: Chris@10: Sat Jan 11 17:50:49 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 22:50:49 by stevenj] Chris@10: imprt reverts hashtable on failure Chris@10: Chris@10: M ./kernel/planner.c -3 +14 Chris@10: Chris@10: Sat Jan 11 16:43:54 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 21:43:54 by stevenj] Chris@10: slight move Chris@10: Chris@10: M ./api/fftw3.h -14 +14 Chris@10: Chris@10: Sat Jan 11 16:34:56 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 21:34:56 by stevenj] Chris@10: stdio.h should be inlcuded outside of extern "C" Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: Chris@10: Sat Jan 11 16:26:35 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 21:26:35 by stevenj] Chris@10: added guru planner API Chris@10: Chris@10: M ./api/Makefile.am -2 +3 Chris@10: M ./api/api.h +1 Chris@10: M ./api/fftw3.h -1 +15 Chris@10: Chris@10: Sat Jan 11 15:54:57 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 20:54:57 by stevenj] Chris@10: added FFTW_FORWARD/BACKWARD Chris@10: Chris@10: M ./api/fftw3.h -1 +4 Chris@10: Chris@10: Sat Jan 11 15:52:17 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 20:52:17 by stevenj] Chris@10: added plan_many_dft Chris@10: Chris@10: A ./api/plan-many-dft.c Chris@10: M ./api/Makefile.am -1 +1 Chris@10: M ./api/fftw3.h -1 +9 Chris@10: M ./api/plan-many-dft.c +43 Chris@10: Chris@10: Sat Jan 11 15:44:37 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 20:44:37 by stevenj] Chris@10: indenting Chris@10: Chris@10: M ./kernel/tensor3.c -3 +3 Chris@10: Chris@10: Sat Jan 11 14:49:08 EST 2003 athena Chris@10: * [project @ 2003-01-11 19:49:08 by athena] Chris@10: Final \n Chris@10: Chris@10: M ./tests/bench.c -2 +4 Chris@10: Chris@10: Sat Jan 11 14:47:31 EST 2003 athena Chris@10: * [project @ 2003-01-11 19:47:31 by athena] Chris@10: Do not compile if not defined(FFTW_DEBUG), in order to avoid Chris@10: unused code in the shared library. Chris@10: Chris@10: M ./kernel/debug.c -1 +3 Chris@10: Chris@10: Sat Jan 11 14:45:56 EST 2003 athena Chris@10: * [project @ 2003-01-11 19:45:56 by athena] Chris@10: Implemented print_plan() Chris@10: Chris@10: A ./api/mkprinter-file.c Chris@10: A ./api/print-plan.c Chris@10: M ./api/Makefile.am -4 +4 Chris@10: M ./api/api.h +1 Chris@10: M ./api/export-wisdom-to-file.c -19 +1 Chris@10: M ./api/fftw3.h -2 +3 Chris@10: M ./api/mkprinter-file.c +39 Chris@10: M ./api/print-plan.c +29 Chris@10: M ./tests/bench.c -2 +3 Chris@10: Chris@10: Sat Jan 11 13:12:01 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 18:12:01 by stevenj] Chris@10: changed the OOP-like plan_destroy to the more-grammatical destroy_plan Chris@10: Chris@10: M ./api/apiplan.c -1 +1 Chris@10: M ./api/fftw3.h -2 +2 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sat Jan 11 12:58:04 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 17:55:49 by stevenj] Chris@10: added guru execute_dft Chris@10: Chris@10: A ./api/execute-dft.c Chris@10: M ./api/Makefile.am -4 +5 Chris@10: M ./api/execute-dft.c +29 Chris@10: M ./api/fftw3.h -2 +3 Chris@10: Chris@10: Sat Jan 11 12:38:40 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 17:38:40 by stevenj] Chris@10: allow for malloc errors in wisdom string, since non-fftw-malloc Chris@10: Chris@10: M ./api/export-wisdom-to-string.c -4 +5 Chris@10: Chris@10: Sat Jan 11 12:16:05 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 17:16:05 by stevenj] Chris@10: cleanup should reset plnr to zero so that fftw can be restarted Chris@10: Chris@10: M ./api/the-planner.c -1 +3 Chris@10: Chris@10: Sat Jan 11 12:13:18 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 17:13:18 by stevenj] Chris@10: NO_UGLY is an internal planner flag Chris@10: Chris@10: M ./api/fftw3.h -4 +3 Chris@10: M ./api/mapflags.c -3 Chris@10: Chris@10: Sat Jan 11 11:23:13 EST 2003 athena Chris@10: * [project @ 2003-01-11 16:23:13 by athena] Chris@10: Written 1d api in terms of generic n-d api. The code is less compact Chris@10: but easier to test Chris@10: Chris@10: M ./api/plan-dft-1d.c -9 +1 Chris@10: M ./tests/bench.c -27 +36 Chris@10: Chris@10: Sat Jan 11 11:07:25 EST 2003 athena Chris@10: * [project @ 2003-01-11 16:07:24 by athena] Chris@10: Added wisdom to header file, made scanners/printer static. stdio.h Chris@10: no longer needed in fftw.h, removed. Probably the printer_file Chris@10: should be reintroduced in a separate file if we ever want to Chris@10: print plans... Chris@10: Chris@10: M ./api/export-wisdom-to-file.c -2 +2 Chris@10: M ./api/export-wisdom-to-string.c -4 +4 Chris@10: M ./api/fftw3.h -2 +13 Chris@10: M ./api/import-wisdom-from-file.c -2 +2 Chris@10: M ./api/import-wisdom-from-string.c -2 +2 Chris@10: M ./kernel/alloc.c -1 +3 Chris@10: M ./kernel/assert.c -2 +2 Chris@10: M ./kernel/debug.c -2 +22 Chris@10: M ./kernel/ifftw.h -15 +1 Chris@10: M ./kernel/print.c -1 +2 Chris@10: M ./kernel/scan.c -1 +2 Chris@10: M ./tests/bench.c -13 +49 Chris@10: M ./tests/verify-lib.c -1 +2 Chris@10: M ./tests/verify-reodft.c -1 +2 Chris@10: Chris@10: Sat Jan 11 09:49:30 EST 2003 athena Chris@10: * [project @ 2003-01-11 14:49:30 by athena] Chris@10: Implemented more APIs Chris@10: Chris@10: A ./api/plan-dft-2d.c Chris@10: A ./api/plan-dft-3d.c Chris@10: A ./api/plan-dft.c Chris@10: M ./api/Makefile.am -6 +5 Chris@10: M ./api/apiplan.c -1 +1 Chris@10: M ./api/fftw3.h -13 +25 Chris@10: M ./api/plan-dft-2d.c +30 Chris@10: M ./api/plan-dft-3d.c +30 Chris@10: M ./api/plan-dft.c +40 Chris@10: M ./tests/bench.c +18 Chris@10: Chris@10: Sat Jan 11 09:21:53 EST 2003 athena Chris@10: * [project @ 2003-01-11 14:21:53 by athena] Chris@10: Added cleanup() to API Chris@10: Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./api/the-planner.c +6 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Sat Jan 11 09:17:34 EST 2003 athena Chris@10: * [project @ 2003-01-11 14:17:34 by athena] Chris@10: Started new bench.c. I had to rename plan_destroy -> Chris@10: plan_destroy_internal to avoid conflicts with API Chris@10: Chris@10: M ./api/api.h -1 Chris@10: M ./api/apiplan.c -3 +3 Chris@10: M ./api/fftw3.h -1 +3 Chris@10: M ./dft/buffered.c -7 +7 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect.c -5 +5 Chris@10: M ./dft/rader.c -8 +8 Chris@10: M ./dft/rank-geq2.c -5 +5 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/plan.c -2 +2 Chris@10: M ./kernel/planner.c -3 +3 Chris@10: M ./libbench/bench-user.h -1 +4 Chris@10: M ./libbench/bench.h -3 +1 Chris@10: M ./rdft/buffered.c -7 +7 Chris@10: M ./rdft/buffered2.c -5 +5 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/dht-rader.c -6 +6 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc.c -7 +7 Chris@10: M ./rdft/indirect.c -5 +5 Chris@10: M ./rdft/rader-hc2hc.c -7 +7 Chris@10: M ./rdft/rank-geq2-rdft2.c -5 +5 Chris@10: M ./rdft/rank-geq2.c -5 +5 Chris@10: M ./rdft/rank0-rdft2.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/bench.c -2 +103 Chris@10: M ./threads/dft-vrank-geq1.c -3 +3 Chris@10: M ./threads/rdft-vrank-geq1.c -3 +3 Chris@10: M ./threads/vrank-geq1-rdft2.c -3 +3 Chris@10: Chris@10: Sat Jan 11 02:45:39 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 07:45:39 by stevenj] Chris@10: fix types Chris@10: Chris@10: A ./api/import-system-wisdom.c Chris@10: M ./api/Makefile.am -3 +3 Chris@10: M ./api/export-wisdom.c -2 +2 Chris@10: M ./api/import-system-wisdom.c +46 Chris@10: M ./api/import-wisdom-from-file.c -2 +3 Chris@10: M ./api/import-wisdom-from-string.c -3 +4 Chris@10: M ./api/import-wisdom.c -3 +5 Chris@10: Chris@10: Sat Jan 11 02:13:25 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 07:13:25 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./api/export-wisdom-to-string.c -1 +2 Chris@10: Chris@10: Sat Jan 11 02:10:50 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 07:10:50 by stevenj] Chris@10: added wisdom api Chris@10: Chris@10: A ./api/export-wisdom-to-file.c Chris@10: A ./api/export-wisdom-to-string.c Chris@10: A ./api/export-wisdom.c Chris@10: A ./api/forget-wisdom.c Chris@10: A ./api/import-wisdom-from-file.c Chris@10: A ./api/import-wisdom-from-string.c Chris@10: A ./api/import-wisdom.c Chris@10: M ./api/Makefile.am -2 +6 Chris@10: M ./api/export-wisdom-to-file.c +47 Chris@10: M ./api/export-wisdom-to-string.c +80 Chris@10: M ./api/export-wisdom.c +44 Chris@10: M ./api/forget-wisdom.c +27 Chris@10: M ./api/import-wisdom-from-file.c +47 Chris@10: M ./api/import-wisdom-from-string.c +49 Chris@10: M ./api/import-wisdom.c +44 Chris@10: M ./kernel/Makefile.am -4 +4 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: R ./kernel/printers.c Chris@10: R ./kernel/scanners.c Chris@10: Chris@10: Sat Jan 11 01:01:17 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 06:01:17 by stevenj] Chris@10: grammar Chris@10: Chris@10: M ./api/mapflags.c -1 +1 Chris@10: Chris@10: Sat Jan 11 00:54:54 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 05:54:54 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./api/mapflags.c -1 +1 Chris@10: Chris@10: Sat Jan 11 00:52:04 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 05:52:04 by stevenj] Chris@10: implemented api/mapflags Chris@10: Chris@10: M ./api/fftw3.h -1 +23 Chris@10: M ./api/mapflags.c -1 +83 Chris@10: Chris@10: Sat Jan 11 00:48:27 EST 2003 stevenj Chris@10: * [project @ 2003-01-11 05:48:27 by stevenj] Chris@10: IMPATIENT is an api issue Chris@10: Chris@10: M ./kernel/ifftw.h -11 +1 Chris@10: Chris@10: Fri Jan 10 01:57:41 EST 2003 stevenj Chris@10: * [project @ 2003-01-10 06:57:41 by stevenj] Chris@10: removed un-needed headers Chris@10: Chris@10: M ./api/the-planner.c -4 Chris@10: Chris@10: Fri Jan 10 01:56:59 EST 2003 stevenj Chris@10: * [project @ 2003-01-10 06:56:59 by stevenj] Chris@10: mkplanner initializes nthr to 1 already Chris@10: Chris@10: M ./api/the-planner.c -1 Chris@10: Chris@10: Thu Jan 9 18:53:09 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 23:53:09 by stevenj] Chris@10: boilerplate Chris@10: Chris@10: M ./api/fftw3.h -1 +14 Chris@10: Chris@10: Thu Jan 9 18:16:39 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 23:16:39 by stevenj] Chris@10: fold vecloop into r{e,o}dft apply function to share buffer, etcetera Chris@10: Chris@10: M ./rdft/vrank-geq1.c -1 +6 Chris@10: M ./reodft/redft00e-r2hc.c -42 +45 Chris@10: M ./reodft/reodft010e-r2hc.c -138 +153 Chris@10: M ./reodft/reodft11e-r2hc.c -98 +107 Chris@10: M ./reodft/rodft00e-r2hc.c -39 +42 Chris@10: Chris@10: Thu Jan 9 18:10:19 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 23:10:19 by stevenj] Chris@10: whoops, bugfix in impulse test for vecn > 1 Chris@10: Chris@10: M ./tests/verify-reodft.c -2 +2 Chris@10: Chris@10: Thu Jan 9 14:23:51 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 19:23:51 by stevenj] Chris@10: bugfix, grr Chris@10: Chris@10: M ./rdft/hc2hc-buf.c -2 +2 Chris@10: Chris@10: Thu Jan 9 14:21:16 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 19:21:16 by stevenj] Chris@10: fixed signed-ness enum problem Chris@10: Chris@10: M ./rdft/codelet-rdft.h -2 +5 Chris@10: Chris@10: Thu Jan 9 14:12:42 EST 2003 athena Chris@10: * [project @ 2003-01-09 19:12:42 by athena] Chris@10: Explicit cast Chris@10: Chris@10: M ./kernel/md5-1.c -1 +1 Chris@10: Chris@10: Thu Jan 9 13:41:51 EST 2003 athena Chris@10: * [project @ 2003-01-09 18:41:51 by athena] Chris@10: Added configure_planner(). mkplan() behaves properly when plan is null. Chris@10: Chris@10: A ./api/configure.c Chris@10: M ./api/Makefile.am -2 +2 Chris@10: M ./api/api.h +1 Chris@10: M ./api/apiplan.c -14 +21 Chris@10: M ./api/configure.c +33 Chris@10: M ./api/fftw3.h -3 +3 Chris@10: M ./api/the-planner.c -4 +1 Chris@10: Chris@10: Thu Jan 9 06:48:53 EST 2003 athena Chris@10: * [project @ 2003-01-09 11:48:53 by athena] Chris@10: More API work Chris@10: Chris@10: A ./api/apiplan.c Chris@10: A ./api/execute.c Chris@10: A ./api/mapflags.c Chris@10: M ./api/Makefile.am -4 +2 Chris@10: M ./api/api.h -2 +3 Chris@10: M ./api/apiplan.c +60 Chris@10: M ./api/execute.c +27 Chris@10: M ./api/fftw3.h -7 +3 Chris@10: M ./api/mapflags.c +26 Chris@10: M ./api/plan-dft-1d.c -11 +4 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Thu Jan 9 05:40:34 EST 2003 athena Chris@10: * [project @ 2003-01-09 10:40:34 by athena] Chris@10: First skeleton of API infrastructure Chris@10: Chris@10: A ./api/ Chris@10: A ./api/Makefile.am Chris@10: A ./api/api.h Chris@10: A ./api/dfthelp.c Chris@10: A ./api/fftw3.h Chris@10: A ./api/plan-dft-1d.c Chris@10: A ./api/the-planner.c Chris@10: M ./Makefile.am -5 +7 Chris@10: M ./api/Makefile.am +10 Chris@10: M ./api/api.h +51 Chris@10: M ./api/dfthelp.c +37 Chris@10: M ./api/fftw3.h +67 Chris@10: M ./api/plan-dft-1d.c +43 Chris@10: M ./api/the-planner.c +42 Chris@10: M ./configure.ac +2 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/trig.c -3 +1 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Thu Jan 9 03:19:35 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 08:19:35 by stevenj] Chris@10: unsigned strikes again Chris@10: Chris@10: M ./rdft/rdft2-tensor-max-index.c -3 +3 Chris@10: Chris@10: Thu Jan 9 01:51:45 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 06:51:45 by stevenj] Chris@10: put rdft2_inplace_strides and rdft2_tensor_max_index in their own files for tighter linking Chris@10: Chris@10: A ./rdft/rdft2-inplace-strides.c Chris@10: A ./rdft/rdft2-tensor-max-index.c Chris@10: M ./rdft/Makefile.am -1 +2 Chris@10: M ./rdft/problem2.c -46 +1 Chris@10: M ./rdft/rdft2-inplace-strides.c +68 Chris@10: M ./rdft/rdft2-tensor-max-index.c +47 Chris@10: M ./rdft/vrank-geq1-rdft2.c -25 +1 Chris@10: Chris@10: Thu Jan 9 01:43:13 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 06:43:13 by stevenj] Chris@10: added rdft2_tensor_max_index...incorrect use of tensor_max_index was preventing proper loop ordering for rnk > 2 rdft2 Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +3 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -4 +27 Chris@10: Chris@10: Thu Jan 9 00:44:45 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 05:44:45 by stevenj] Chris@10: arbitrary spltrnk in rdft2 rank-geq2 Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -11 +53 Chris@10: Chris@10: Thu Jan 9 00:40:17 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 05:40:17 by stevenj] Chris@10: don't mention wisdom when non-verbose Chris@10: Chris@10: M ./tests/bench.c -2 +4 Chris@10: Chris@10: Thu Jan 9 00:02:35 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 05:02:35 by stevenj] Chris@10: bug fix: printing %T should pass tensor *, not tensor ** Chris@10: Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./rdft/problem.c -3 +3 Chris@10: M ./rdft/problem2.c -3 +3 Chris@10: Chris@10: Wed Jan 8 23:40:48 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 04:40:48 by stevenj] Chris@10: correct(?) normalization for rodft00 ... all of the even/odd transforms should be normalized according to the expanded'' DFT of ~twice the length Chris@10: Chris@10: M ./reodft/rodft00e-r2hc.c -4 +4 Chris@10: M ./tests/verify-reodft.c -2 +2 Chris@10: Chris@10: Wed Jan 8 23:18:23 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 04:18:23 by stevenj] Chris@10: fixed tests for n=1 Chris@10: Chris@10: M ./tests/verify-reodft.c -3 +6 Chris@10: Chris@10: Wed Jan 8 22:10:08 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 03:10:08 by stevenj] Chris@10: fixed bug in vector tests for rdft(2) Chris@10: Chris@10: M ./tests/bench.c -3 +5 Chris@10: Chris@10: Wed Jan 8 20:12:00 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 01:12:00 by stevenj] Chris@10: fixed handling when first rnk-1 dimensions compress to nothing (ugh) Chris@10: Chris@10: M ./rdft/problem2.c -3 +6 Chris@10: Chris@10: Wed Jan 8 20:02:35 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 01:02:35 by stevenj] Chris@10: fixed incorrect/missing rdft2 rank-0 handling Chris@10: Chris@10: A ./rdft/rank0-rdft2.c Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/conf.c -1 +2 Chris@10: M ./rdft/nop2.c -3 +6 Chris@10: M ./rdft/rank0-rdft2.c +194 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: Chris@10: Wed Jan 8 19:49:05 EST 2003 stevenj Chris@10: * [project @ 2003-01-09 00:49:05 by stevenj] Chris@10: bug fix: for rnk > 1, must compress rnk-1 dims separately (ugh) Chris@10: Chris@10: M ./rdft/problem2.c -2 +11 Chris@10: Chris@10: Wed Jan 8 17:39:14 EST 2003 stevenj Chris@10: * [project @ 2003-01-08 22:39:14 by stevenj] Chris@10: added trailing newline Chris@10: Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Wed Jan 8 17:38:02 EST 2003 stevenj Chris@10: * [project @ 2003-01-08 22:38:02 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +66 Chris@10: Chris@10: Wed Jan 8 16:53:16 EST 2003 stevenj Chris@10: * [project @ 2003-01-08 21:53:16 by stevenj] Chris@10: got rid of compiler warning Chris@10: Chris@10: M ./rdft/problem.c -2 +2 Chris@10: Chris@10: Wed Jan 8 16:49:48 EST 2003 stevenj Chris@10: * [project @ 2003-01-08 21:49:48 by stevenj] Chris@10: whoops, test r2hc and not rodft00 by default Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Wed Jan 8 16:46:24 EST 2003 stevenj Chris@10: * [project @ 2003-01-08 21:46:24 by stevenj] Chris@10: got rid of real_n...use physical n everywhere in rdft; fixed rdft sz compression; fixed rodft00 verify bug Chris@10: Chris@10: M ./rdft/buffered.c -3 +2 Chris@10: M ./rdft/indirect.c -10 +6 Chris@10: M ./rdft/problem.c -44 +43 Chris@10: M ./rdft/rank-geq2.c -10 +6 Chris@10: M ./rdft/rdft.h -3 +1 Chris@10: M ./reodft/redft00e-r2hc.c -4 +5 Chris@10: M ./reodft/reodft010e-r2hc.c -4 +3 Chris@10: M ./reodft/reodft11e-r2hc.c -4 +3 Chris@10: M ./reodft/rodft00e-r2hc.c -4 +3 Chris@10: M ./tests/bench.c -2 +2 Chris@10: M ./tests/verify-reodft.c -33 +35 Chris@10: Chris@10: Wed Jan 8 07:20:47 EST 2003 athena Chris@10: * [project @ 2003-01-08 12:20:47 by athena] Chris@10: icc-6.0 bug workaround Chris@10: Chris@10: A ./simd/sse-aux.c Chris@10: A ./simd/sse2-aux.c Chris@10: M ./simd/Makefile.am -2 +2 Chris@10: M ./simd/sse-aux.c +33 Chris@10: M ./simd/sse.c -3 +1 Chris@10: M ./simd/sse2-aux.c +34 Chris@10: M ./simd/sse2.c -3 +1 Chris@10: Chris@10: Wed Jan 8 04:21:40 EST 2003 athena Chris@10: * [project @ 2003-01-08 09:21:40 by athena] Chris@10: Reclaimed the fftw_real identifier, because I need it for the API Chris@10: Chris@10: M ./kernel/ifftw.h -12 +9 Chris@10: M ./rdft/buffered2.c -5 +5 Chris@10: M ./rdft/rader-hc2hc.c -1 +1 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Wed Jan 8 04:14:55 EST 2003 athena Chris@10: * [project @ 2003-01-08 09:14:55 by athena] Chris@10: Use recommended AC_OUTPUT syntax Chris@10: Chris@10: M ./configure.ac -1 +2 Chris@10: Chris@10: Wed Jan 8 04:00:22 EST 2003 athena Chris@10: * [project @ 2003-01-08 09:00:22 by athena] Chris@10: Removed FFTW(foo) as a synonym for X(foo). This is an API issue. Chris@10: Chris@10: M ./kernel/ifftw.h -8 +7 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Tue Jan 7 17:45:52 EST 2003 stevenj Chris@10: * [project @ 2003-01-07 22:45:52 by stevenj] Chris@10: get rid of warning Chris@10: Chris@10: M ./simd/sse2.c -1 +2 Chris@10: Chris@10: Tue Jan 7 16:22:39 EST 2003 athena Chris@10: * [project @ 2003-01-07 21:22:39 by athena] Chris@10: Renamed conflicting files */codelet.h into dft/codelet-dft.h and Chris@10: rdft/codelet-rdft.h Chris@10: Chris@10: A ./dft/codelet-dft.h Chris@10: A ./rdft/codelet-rdft.h Chris@10: A ./support/codelet_prelude.dft Chris@10: A ./support/codelet_prelude.rdft Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/codelet-dft.h +113 Chris@10: R ./dft/codelet.h Chris@10: M ./dft/codelets/inplace/Makefile.am -4 +4 Chris@10: M ./dft/codelets/n.c -1 +1 Chris@10: M ./dft/codelets/standard/Makefile.am -3 +3 Chris@10: M ./dft/codelets/t.c -1 +1 Chris@10: M ./dft/dft.h -2 +2 Chris@10: M ./dft/simd/codelets/Makefile.am -6 +6 Chris@10: M ./dft/simd/n1b.c -1 +1 Chris@10: M ./dft/simd/n1f.c -1 +1 Chris@10: M ./dft/simd/t1b.c -1 +1 Chris@10: M ./dft/simd/t1f.c -1 +1 Chris@10: M ./rdft/Makefile.am -5 +5 Chris@10: M ./rdft/codelet-rdft.h +190 Chris@10: R ./rdft/codelet.h Chris@10: M ./rdft/codelets/hc2r/Makefile.am -4 +4 Chris@10: M ./rdft/codelets/hc2r.c -1 +1 Chris@10: M ./rdft/codelets/hfb.c -1 +1 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -4 +4 Chris@10: M ./rdft/codelets/r2hc.c -1 +1 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./support/Makefile.am -1 +2 Chris@10: M ./support/Makefile.codelets -2 +4 Chris@10: R ./support/codelet_prelude Chris@10: M ./support/codelet_prelude.dft +8 Chris@10: M ./support/codelet_prelude.rdft +8 Chris@10: Chris@10: Tue Jan 7 16:21:16 EST 2003 stevenj Chris@10: * [project @ 2003-01-07 21:21:16 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +737 Chris@10: Chris@10: Tue Jan 7 15:47:24 EST 2003 athena Chris@10: * [project @ 2003-01-07 20:47:24 by athena] Chris@10: Silence warnings Chris@10: Chris@10: M ./simd/simd-3dnow.h -1 +1 Chris@10: M ./simd/simd-sse.h -1 Chris@10: M ./simd/simd-sse2.h -1 +1 Chris@10: M ./simd/sse2.c -2 +1 Chris@10: Chris@10: Tue Jan 7 15:00:14 EST 2003 stevenj Chris@10: * [project @ 2003-01-07 20:00:14 by stevenj] Chris@10: fftw2 used spltrnk=1 Chris@10: Chris@10: M ./dft/rank-geq2.c -2 +4 Chris@10: M ./rdft/rank-geq2.c -2 +4 Chris@10: Chris@10: Tue Jan 7 14:32:06 EST 2003 athena Chris@10: * [project @ 2003-01-07 19:32:06 by athena] Chris@10: Silence warning Chris@10: Chris@10: M ./dft/codelet.h -2 +1 Chris@10: M ./rdft/codelet.h -2 +1 Chris@10: M ./simd/simd-sse.h -1 +2 Chris@10: M ./simd/sse.c -2 +1 Chris@10: Chris@10: Tue Jan 7 12:13:50 EST 2003 stevenj Chris@10: * [project @ 2003-01-07 17:13:50 by stevenj] Chris@10: noted deficiency Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Tue Jan 7 07:18:51 EST 2003 athena Chris@10: * [project @ 2003-01-07 12:18:51 by athena] Chris@10: Strengthened conditions for a problem to be POSSIBLY_UNALIGNED Chris@10: Chris@10: M ./rdft/vrank-geq1-rdft2.c -4 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Tue Jan 7 05:09:42 EST 2003 athena Chris@10: * [project @ 2003-01-07 10:09:42 by athena] Chris@10: Strengthened conditions for a plan to be POSSIBLY_UNALIGNED Chris@10: Chris@10: M ./dft/vrank-geq1.c -5 +2 Chris@10: M ./kernel/align.c -1 +6 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: Chris@10: Sun Jan 5 02:43:45 EST 2003 stevenj Chris@10: * [project @ 2003-01-05 07:43:45 by stevenj] Chris@10: added copyright todo Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Sun Jan 5 02:37:31 EST 2003 stevenj Chris@10: * [project @ 2003-01-05 07:37:31 by stevenj] Chris@10: modified comment Chris@10: Chris@10: M ./kernel/planner.c -2 +3 Chris@10: Chris@10: Sun Jan 5 02:34:36 EST 2003 stevenj Chris@10: * [project @ 2003-01-05 07:33:41 by stevenj] Chris@10: fixed comment Chris@10: Chris@10: M ./tests/verify-rdft.c -3 +3 Chris@10: Chris@10: Sun Jan 5 02:31:56 EST 2003 stevenj Chris@10: * [project @ 2003-01-05 07:31:56 by stevenj] Chris@10: implemented rdft2 verify Chris@10: Chris@10: M ./TODO -4 +2 Chris@10: M ./tests/verify-rdft.c -5 +15 Chris@10: Chris@10: Sat Jan 4 16:20:42 EST 2003 stevenj Chris@10: * [project @ 2003-01-04 21:20:42 by stevenj] Chris@10: fix --enable-single Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Wed Oct 23 12:59:12 EDT 2002 stevenj Chris@10: * [project @ 2002-10-23 16:59:12 by stevenj] Chris@10: slight fixes Chris@10: Chris@10: M ./threads/threads.c -20 +14 Chris@10: Chris@10: Wed Oct 23 12:42:39 EDT 2002 stevenj Chris@10: * [project @ 2002-10-23 16:42:39 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./threads/threads.c -1 +1 Chris@10: Chris@10: Tue Oct 1 09:32:56 EDT 2002 athena Chris@10: * [project @ 2002-10-01 13:32:56 by athena] Chris@10: Experimental stuff Chris@10: Chris@10: M ./genfft/annotate.ml -2 +9 Chris@10: M ./genfft/annotate.mli -3 +2 Chris@10: M ./genfft/c.ml -4 +16 Chris@10: M ./genfft/genutil.ml -1 +11 Chris@10: M ./genfft/magic.ml -1 +5 Chris@10: Chris@10: Sat Sep 28 13:03:53 EDT 2002 athena Chris@10: * [project @ 2002-09-28 17:03:53 by athena] Chris@10: Experimental Franz mode Chris@10: Chris@10: M ./configure.ac +3 Chris@10: M ./dft/simd/codelets/Makefile.am -3 +68 Chris@10: M ./genfft/gen_notw_c.ml -7 +8 Chris@10: M ./genfft/gen_twiddle_c.ml -6 +7 Chris@10: M ./genfft/genutil.ml -2 +10 Chris@10: Chris@10: Thu Sep 26 15:14:38 EDT 2002 athena Chris@10: * [project @ 2002-09-26 19:14:38 by athena] Chris@10: const-correct Chris@10: Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: Chris@10: Thu Sep 26 15:06:38 EDT 2002 athena Chris@10: * [project @ 2002-09-26 19:06:38 by athena] Chris@10: Reuse dimcmp routine for other purposes Chris@10: Chris@10: M ./dft/vrank2-transpose.c -4 +3 Chris@10: M ./dft/vrank3-transpose.c -4 +2 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/tensor7.c -5 +5 Chris@10: M ./rdft/vrank2-transpose.c -4 +2 Chris@10: M ./rdft/vrank3-transpose.c -4 +2 Chris@10: Chris@10: Wed Sep 25 07:37:38 EDT 2002 athena Chris@10: * [project @ 2002-09-25 11:37:38 by athena] Chris@10: Use tornk1 correctly. Chris@10: Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/tensor.c -2 +3 Chris@10: M ./rdft/direct.c -3 +3 Chris@10: M ./rdft/direct2.c -3 +3 Chris@10: Chris@10: Wed Sep 25 07:36:38 EDT 2002 athena Chris@10: * [project @ 2002-09-25 11:36:38 by athena] Chris@10: Hmm... I thought I had fixed this before... Chris@10: Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: Chris@10: Tue Sep 24 21:27:49 EDT 2002 athena Chris@10: * [project @ 2002-09-25 01:27:49 by athena] Chris@10: Collect more common idioms Chris@10: Chris@10: M ./dft/buffered.c -5 +3 Chris@10: M ./dft/rank0.c -16 +3 Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: M ./rdft/buffered.c -5 +2 Chris@10: M ./rdft/buffered2.c -5 +2 Chris@10: M ./rdft/rank0.c -16 +3 Chris@10: Chris@10: Tue Sep 24 21:15:57 EDT 2002 athena Chris@10: * [project @ 2002-09-25 01:15:57 by athena] Chris@10: Still collecting common idioms... Chris@10: Chris@10: M ./dft/direct.c -10 +3 Chris@10: M ./rdft/direct.c -10 +2 Chris@10: M ./rdft/direct2.c -9 +2 Chris@10: Chris@10: Tue Sep 24 21:13:00 EDT 2002 athena Chris@10: * [project @ 2002-09-25 01:13:00 by athena] Chris@10: More garbage collection. Chris@10: Chris@10: M ./dft/direct.c -6 +5 Chris@10: M ./rdft/direct.c -6 +5 Chris@10: M ./rdft/direct2.c -6 +5 Chris@10: Chris@10: Tue Sep 24 21:08:19 EDT 2002 athena Chris@10: * [project @ 2002-09-25 01:08:19 by athena] Chris@10: More compact code Chris@10: Chris@10: M ./dft/buffered.c -6 +3 Chris@10: Chris@10: Tue Sep 24 20:54:43 EDT 2002 athena Chris@10: * [project @ 2002-09-25 00:54:43 by athena] Chris@10: Collect common pattern if (foo) free(foo) ==> free0(foo) Chris@10: Chris@10: M ./dft/buffered.c -7 +6 Chris@10: M ./dft/generic.c -2 +1 Chris@10: M ./dft/rader.c -6 +3 Chris@10: M ./kernel/alloc.c -1 +7 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -8 +4 Chris@10: M ./kernel/stride.c -3 +2 Chris@10: M ./kernel/tensor.c -5 +2 Chris@10: M ./rdft/buffered.c -3 +2 Chris@10: M ./rdft/buffered2.c -3 +2 Chris@10: M ./rdft/dht-rader.c -2 +1 Chris@10: M ./rdft/generic.c -2 +1 Chris@10: M ./rdft/problem.c -3 +2 Chris@10: M ./rdft/rader-hc2hc.c -6 +3 Chris@10: Chris@10: Tue Sep 24 20:08:44 EDT 2002 athena Chris@10: * [project @ 2002-09-25 00:08:44 by athena] Chris@10: Collect some common code in */buffered*.c Chris@10: Chris@10: A ./kernel/buffered.c Chris@10: M ./dft/buffered.c -18 +2 Chris@10: M ./kernel/Makefile.am -7 +6 Chris@10: M ./kernel/buffered.c +44 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./rdft/buffered.c -18 +3 Chris@10: M ./rdft/buffered2.c -18 +2 Chris@10: Chris@10: Tue Sep 24 19:39:22 EDT 2002 stevenj Chris@10: * [project @ 2002-09-24 23:39:22 by stevenj] Chris@10: use STRUCT_HACK #define to determing rdft kind[] allocation Chris@10: Chris@10: M ./rdft/problem.c -8 +23 Chris@10: M ./rdft/rdft.h -2 +8 Chris@10: Chris@10: Tue Sep 24 17:21:09 EDT 2002 stevenj Chris@10: * [project @ 2002-09-24 21:21:09 by stevenj] Chris@10: report total pcost of measured/estimated plans...epcost is especially useful to estimate the effects of various impatience flags on planning time for large transforms Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -1 +6 Chris@10: Chris@10: Mon Sep 23 18:49:10 EDT 2002 athena Chris@10: * [project @ 2002-09-23 22:49:10 by athena] Chris@10: Prevent unwanted inlining Chris@10: Chris@10: A ./kernel/trig1.c Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/trig.c -54 +7 Chris@10: M ./kernel/trig1.c +70 Chris@10: Chris@10: Mon Sep 23 18:37:59 EDT 2002 athena Chris@10: * [project @ 2002-09-23 22:37:59 by athena] Chris@10: Space compaction Chris@10: Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: M ./kernel/trig.c -21 +27 Chris@10: Chris@10: Mon Sep 23 11:49:32 EDT 2002 athena Chris@10: * [project @ 2002-09-23 15:49:32 by athena] Chris@10: Still reducing size Chris@10: Chris@10: A ./kernel/hash.c Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/hash.c +31 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/md5-1.c +1 Chris@10: M ./kernel/planner.c -12 +3 Chris@10: M ./kernel/scan.c -22 +1 Chris@10: Chris@10: Sun Sep 22 16:03:30 EDT 2002 athena Chris@10: * [project @ 2002-09-22 20:03:30 by athena] Chris@10: Saved another 5KB by redesigning opcnt protocol. (gasp!) Chris@10: Chris@10: M ./dft/buffered.c -5 +6 Chris@10: M ./dft/ct-dif.c -4 +3 Chris@10: M ./dft/ct-dit.c -4 +3 Chris@10: M ./dft/ct-ditbuf.c -6 +6 Chris@10: M ./dft/ct-ditf.c -4 +3 Chris@10: M ./dft/direct.c -2 +3 Chris@10: M ./dft/generic.c -4 +2 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/nop.c -2 +2 Chris@10: M ./dft/rader.c -4 +3 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +3 Chris@10: M ./dft/vrank2-transpose.c -2 +2 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: M ./kernel/ifftw.h -6 +14 Chris@10: M ./kernel/ops.c -23 +28 Chris@10: M ./kernel/plan.c -2 +2 Chris@10: M ./rdft/buffered.c -5 +6 Chris@10: M ./rdft/buffered2.c -3 +3 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/direct.c -3 +8 Chris@10: M ./rdft/direct2.c -3 +8 Chris@10: M ./rdft/generic.c -4 +3 Chris@10: M ./rdft/hc2hc-buf.c -10 +9 Chris@10: M ./rdft/hc2hc-dif.c -8 +8 Chris@10: M ./rdft/hc2hc-dit.c -8 +8 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: M ./rdft/nop.c -2 +2 Chris@10: M ./rdft/nop2.c -2 +2 Chris@10: M ./rdft/rader-hc2hc.c -8 +5 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -5 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +3 Chris@10: M ./rdft/vrank-geq1.c -2 +3 Chris@10: M ./rdft/vrank2-transpose.c -2 +2 Chris@10: M ./rdft/vrank3-transpose.c -2 +2 Chris@10: M ./threads/ct-dit.c -4 +3 Chris@10: M ./threads/dft-vrank-geq1.c -4 +3 Chris@10: M ./threads/hc2hc-dif.c -8 +8 Chris@10: M ./threads/hc2hc-dit.c -8 +8 Chris@10: M ./threads/rdft-vrank-geq1.c -4 +3 Chris@10: M ./threads/vrank-geq1-rdft2.c -4 +3 Chris@10: Chris@10: Sun Sep 22 15:00:59 EDT 2002 athena Chris@10: * [project @ 2002-09-22 19:00:59 by athena] Chris@10: More code compression Chris@10: Chris@10: A ./kernel/tensor8.c Chris@10: M ./dft/buffered.c -4 +3 Chris@10: M ./dft/direct.c -3 +2 Chris@10: M ./dft/indirect.c -4 +3 Chris@10: M ./dft/problem.c -5 +3 Chris@10: M ./dft/rank-geq2.c -9 +3 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: M ./kernel/tensor1.c -1 +6 Chris@10: M ./kernel/tensor4.c -1 +6 Chris@10: M ./kernel/tensor8.c +35 Chris@10: M ./rdft/buffered.c -4 +3 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/direct.c -3 +2 Chris@10: M ./rdft/hc2hc.c -7 +6 Chris@10: M ./rdft/indirect.c -4 +3 Chris@10: M ./rdft/problem.c -7 +4 Chris@10: M ./rdft/problem2.c -5 +3 Chris@10: M ./rdft/rank-geq2-rdft2.c -9 +3 Chris@10: M ./rdft/rank-geq2.c -13 +5 Chris@10: Chris@10: Sun Sep 22 13:27:46 EDT 2002 athena Chris@10: * [project @ 2002-09-22 17:27:46 by athena] Chris@10: Smaller code size. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/solver.c -1 +6 Chris@10: Chris@10: Sun Sep 22 12:50:36 EDT 2002 athena Chris@10: * [project @ 2002-09-22 16:50:36 by athena] Chris@10: Started unification of rader Chris@10: Chris@10: A ./dft/rader-omega.c Chris@10: M ./dft/Makefile.am -3 +3 Chris@10: M ./dft/dft.h -1 +5 Chris@10: M ./dft/rader-omega.c +57 Chris@10: M ./dft/rader.c -38 +3 Chris@10: M ./rdft/rader-hc2hc.c -37 +3 Chris@10: Chris@10: Sun Sep 22 12:35:30 EDT 2002 athena Chris@10: * [project @ 2002-09-22 16:35:30 by athena] Chris@10: Typo Chris@10: Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: Chris@10: Sun Sep 22 12:25:20 EDT 2002 athena Chris@10: * [project @ 2002-09-22 16:25:20 by athena] Chris@10: Changed protocol for destroy_plan so as to save space. Chris@10: Chris@10: M ./dft/buffered.c -8 +4 Chris@10: M ./dft/ct.c -2 +1 Chris@10: M ./dft/direct.c -2 +1 Chris@10: M ./dft/generic.c -3 +1 Chris@10: M ./dft/indirect.c -6 +3 Chris@10: M ./dft/nop.c -7 +2 Chris@10: M ./dft/rader.c -7 +3 Chris@10: M ./dft/rank-geq2.c -6 +3 Chris@10: M ./dft/rank0.c -7 +2 Chris@10: M ./dft/vrank-geq1.c -2 +1 Chris@10: M ./dft/vrank2-transpose.c -7 +2 Chris@10: M ./dft/vrank3-transpose.c -7 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/plan.c -3 +11 Chris@10: M ./kernel/problem.c -2 +3 Chris@10: M ./rdft/buffered.c -8 +4 Chris@10: M ./rdft/buffered2.c -6 +3 Chris@10: M ./rdft/dft-r2hc.c -2 +1 Chris@10: M ./rdft/dht-r2hc.c -2 +1 Chris@10: M ./rdft/dht-rader.c -7 +3 Chris@10: M ./rdft/direct.c -2 +1 Chris@10: M ./rdft/direct2.c -2 +1 Chris@10: M ./rdft/generic.c -3 +1 Chris@10: M ./rdft/hc2hc.c -14 +7 Chris@10: M ./rdft/indirect.c -6 +3 Chris@10: M ./rdft/nop.c -7 +2 Chris@10: M ./rdft/nop2.c -7 +2 Chris@10: M ./rdft/rader-hc2hc.c -9 +4 Chris@10: M ./rdft/rank-geq2-rdft2.c -6 +3 Chris@10: M ./rdft/rank-geq2.c -6 +3 Chris@10: M ./rdft/rank0.c -7 +2 Chris@10: M ./rdft/rdft-dht.c -2 +1 Chris@10: M ./rdft/rdft2-radix2.c -2 +1 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +1 Chris@10: M ./rdft/vrank-geq1.c -2 +1 Chris@10: M ./rdft/vrank2-transpose.c -7 +2 Chris@10: M ./rdft/vrank3-transpose.c -7 +2 Chris@10: M ./reodft/redft00e-r2hc.c -2 +1 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +1 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +1 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +1 Chris@10: M ./threads/dft-vrank-geq1.c -4 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -4 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -4 +2 Chris@10: Chris@10: Sun Sep 22 11:08:57 EDT 2002 athena Chris@10: * [project @ 2002-09-22 15:08:57 by athena] Chris@10: Introduced convenient function X(mkplan_d) Chris@10: Chris@10: M ./dft/buffered.c -31 +18 Chris@10: M ./dft/ct.c -5 +2 Chris@10: M ./dft/generic.c -10 +5 Chris@10: M ./dft/indirect.c -14 +8 Chris@10: M ./dft/rader.c -38 +20 Chris@10: M ./dft/rank-geq2.c -16 +12 Chris@10: M ./dft/vrank-geq1.c -9 +7 Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/planner.c -1 +8 Chris@10: M ./rdft/buffered.c -30 +20 Chris@10: M ./rdft/buffered2.c -9 +3 Chris@10: M ./rdft/dft-r2hc.c -7 +5 Chris@10: M ./rdft/dht-r2hc.c -7 +4 Chris@10: M ./rdft/dht-rader.c -23 +12 Chris@10: M ./rdft/generic.c -4 +1 Chris@10: M ./rdft/hc2hc.c -16 +10 Chris@10: M ./rdft/indirect.c -14 +9 Chris@10: M ./rdft/rader-hc2hc.c -41 +21 Chris@10: M ./rdft/rank-geq2-rdft2.c -12 +8 Chris@10: M ./rdft/rank-geq2.c -16 +12 Chris@10: M ./rdft/rdft-dht.c -5 +3 Chris@10: M ./rdft/rdft2-radix2.c -7 +3 Chris@10: M ./rdft/vrank-geq1-rdft2.c -9 +7 Chris@10: M ./rdft/vrank-geq1.c -10 +7 Chris@10: M ./reodft/redft00e-r2hc.c -3 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -3 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -3 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -3 +2 Chris@10: M ./threads/ct-dit.c -4 +4 Chris@10: M ./threads/dft-vrank-geq1.c -12 +10 Chris@10: M ./threads/hc2hc-dif.c -4 +4 Chris@10: M ./threads/hc2hc-dit.c -4 +4 Chris@10: M ./threads/rdft-vrank-geq1.c -11 +9 Chris@10: M ./threads/vrank-geq1-rdft2.c -11 +9 Chris@10: Chris@10: Sun Sep 22 10:21:36 EDT 2002 athena Chris@10: * [project @ 2002-09-22 14:21:36 by athena] Chris@10: Split tensor/md5 into separate files to allow independent linking Chris@10: and/or prevent undesidred inlining Chris@10: Chris@10: A ./kernel/md5-1.c Chris@10: A ./kernel/tensor1.c Chris@10: A ./kernel/tensor2.c Chris@10: A ./kernel/tensor3.c Chris@10: A ./kernel/tensor4.c Chris@10: A ./kernel/tensor5.c Chris@10: A ./kernel/tensor7.c Chris@10: M ./kernel/Makefile.am -3 +4 Chris@10: M ./kernel/md5-1.c +53 Chris@10: M ./kernel/md5.c -31 Chris@10: M ./kernel/tensor.c -270 +2 Chris@10: M ./kernel/tensor1.c +32 Chris@10: M ./kernel/tensor2.c +37 Chris@10: M ./kernel/tensor3.c +46 Chris@10: M ./kernel/tensor4.c +68 Chris@10: M ./kernel/tensor5.c +93 Chris@10: M ./kernel/tensor7.c +127 Chris@10: Chris@10: Sun Sep 22 09:49:09 EDT 2002 athena Chris@10: * [project @ 2002-09-22 13:49:08 by athena] Chris@10: Treat all tensors as dynamically allocated objects. They were Chris@10: dynamically allocated in part anyway, so there is no point in Chris@10: complicating the object code with the clumsy calling conventions Chris@10: for by-value structs. Chris@10: Chris@10: M ./dft/buffered.c -19 +19 Chris@10: M ./dft/ct-dif.c -5 +5 Chris@10: M ./dft/ct-dit.c -5 +5 Chris@10: M ./dft/ct-ditbuf.c -4 +5 Chris@10: M ./dft/ct-ditf.c -5 +5 Chris@10: M ./dft/ct.c -14 +14 Chris@10: M ./dft/dft.h -4 +4 Chris@10: M ./dft/direct.c -13 +13 Chris@10: M ./dft/generic.c -7 +7 Chris@10: M ./dft/indirect.c -18 +16 Chris@10: M ./dft/nop.c -5 +5 Chris@10: M ./dft/problem.c -12 +12 Chris@10: M ./dft/rader.c -12 +12 Chris@10: M ./dft/rank-geq2.c -22 +22 Chris@10: M ./dft/rank0.c -14 +14 Chris@10: M ./dft/vrank-geq1.c -11 +11 Chris@10: M ./dft/vrank2-transpose.c -9 +9 Chris@10: M ./dft/vrank3-transpose.c -12 +12 Chris@10: M ./dft/zero.c -18 +20 Chris@10: M ./kernel/ifftw.h -13 +22 Chris@10: M ./kernel/tensor.c -92 +102 Chris@10: M ./rdft/buffered.c -23 +21 Chris@10: M ./rdft/buffered2.c -24 +24 Chris@10: M ./rdft/dft-r2hc.c -11 +11 Chris@10: M ./rdft/dht-r2hc.c -6 +6 Chris@10: M ./rdft/dht-rader.c -7 +7 Chris@10: M ./rdft/direct.c -18 +18 Chris@10: M ./rdft/direct2.c -16 +16 Chris@10: M ./rdft/generic.c -8 +8 Chris@10: M ./rdft/hc2hc-buf.c -4 +5 Chris@10: M ./rdft/hc2hc-dif.c -5 +5 Chris@10: M ./rdft/hc2hc-dit.c -5 +5 Chris@10: M ./rdft/hc2hc.c -23 +23 Chris@10: M ./rdft/indirect.c -20 +18 Chris@10: M ./rdft/nop.c -5 +5 Chris@10: M ./rdft/nop2.c -4 +4 Chris@10: M ./rdft/problem.c -39 +40 Chris@10: M ./rdft/problem2.c -34 +34 Chris@10: M ./rdft/rader-hc2hc.c -11 +11 Chris@10: M ./rdft/rank-geq2-rdft2.c -23 +23 Chris@10: M ./rdft/rank-geq2.c -28 +28 Chris@10: M ./rdft/rank0.c -10 +10 Chris@10: M ./rdft/rdft-dht.c -10 +10 Chris@10: M ./rdft/rdft.h -9 +9 Chris@10: M ./rdft/rdft2-radix2.c -21 +21 Chris@10: M ./rdft/vrank-geq1-rdft2.c -11 +11 Chris@10: M ./rdft/vrank-geq1.c -11 +11 Chris@10: M ./rdft/vrank2-transpose.c -9 +9 Chris@10: M ./rdft/vrank3-transpose.c -12 +12 Chris@10: M ./reodft/redft00e-r2hc.c -10 +10 Chris@10: M ./reodft/reodft010e-r2hc.c -10 +10 Chris@10: M ./reodft/reodft11e-r2hc.c -10 +10 Chris@10: M ./reodft/rodft00e-r2hc.c -10 +10 Chris@10: M ./tests/debug.h -2 +2 Chris@10: M ./tests/dotens.c -4 +4 Chris@10: M ./tests/dotens2.c -5 +5 Chris@10: M ./tests/verify-dft.c -10 +11 Chris@10: M ./tests/verify-lib.c -15 +16 Chris@10: M ./tests/verify-rdft.c -43 +47 Chris@10: M ./tests/verify-reodft.c -28 +28 Chris@10: M ./tests/verify.h -3 +3 Chris@10: Chris@10: Sat Sep 21 18:24:55 EDT 2002 stevenj Chris@10: * [project @ 2002-09-21 22:24:55 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Sat Sep 21 18:10:07 EDT 2002 athena Chris@10: * [project @ 2002-09-21 22:10:07 by athena] Chris@10: Avoid generating NaN when n = 0. Chris@10: Chris@10: M ./tests/verify-lib.c -13 +16 Chris@10: Chris@10: Sat Sep 21 18:04:05 EDT 2002 athena Chris@10: * [project @ 2002-09-21 22:04:05 by athena] Chris@10: Saved more. Chris@10: Chris@10: M ./dft/dft.h -2 +2 Chris@10: M ./dft/problem.c -5 +5 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -2 +2 Chris@10: M ./rdft/hc2hc.c -5 +7 Chris@10: M ./rdft/problem.c -11 +11 Chris@10: M ./rdft/problem2.c -5 +5 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft.h -4 +4 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./threads/dft-vrank-geq1.c -2 +2 Chris@10: Chris@10: Sat Sep 21 17:47:36 EDT 2002 athena Chris@10: * [project @ 2002-09-21 21:47:35 by athena] Chris@10: Save 1200 bytes of object code. Do not pass structs by value whenever Chris@10: practical, because the calling protocol generates clumsy code. Chris@10: Chris@10: M ./dft/buffered.c -6 +6 Chris@10: M ./dft/ct.c -5 +5 Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./dft/indirect.c -12 +12 Chris@10: M ./dft/nop.c -2 +2 Chris@10: M ./dft/problem.c -11 +11 Chris@10: M ./dft/rank-geq2.c -21 +21 Chris@10: M ./dft/vrank-geq1.c -6 +6 Chris@10: M ./dft/vrank3-transpose.c -8 +8 Chris@10: M ./kernel/ifftw.h -20 +22 Chris@10: M ./kernel/pickdim.c -9 +9 Chris@10: M ./kernel/print.c -2 +2 Chris@10: M ./kernel/tensor.c -69 +71 Chris@10: M ./rdft/buffered.c -6 +6 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/dft-r2hc.c -3 +3 Chris@10: M ./rdft/direct.c -3 +3 Chris@10: M ./rdft/hc2hc.c -9 +9 Chris@10: M ./rdft/indirect.c -14 +14 Chris@10: M ./rdft/nop.c -2 +2 Chris@10: M ./rdft/problem.c -15 +15 Chris@10: M ./rdft/problem2.c -15 +15 Chris@10: M ./rdft/rank-geq2-rdft2.c -19 +19 Chris@10: M ./rdft/rank-geq2.c -27 +27 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -7 +7 Chris@10: M ./rdft/vrank-geq1-rdft2.c -6 +7 Chris@10: M ./rdft/vrank-geq1.c -6 +6 Chris@10: M ./rdft/vrank3-transpose.c -8 +8 Chris@10: M ./reodft/redft00e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -2 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -2 +2 Chris@10: M ./tests/verify-dft.c -6 +6 Chris@10: M ./tests/verify-lib.c -2 +2 Chris@10: M ./tests/verify-rdft.c -21 +21 Chris@10: M ./tests/verify-reodft.c -10 +10 Chris@10: M ./threads/dft-vrank-geq1.c -3 +3 Chris@10: M ./threads/rdft-vrank-geq1.c -3 +3 Chris@10: M ./threads/vrank-geq1-rdft2.c -3 +4 Chris@10: Chris@10: Sat Sep 21 12:10:21 EDT 2002 athena Chris@10: * [project @ 2002-09-21 16:10:21 by athena] Chris@10: Do not allocate buffers for rader omegas. Let the planner do it Chris@10: if necessary. Chris@10: Chris@10: M ./rdft/dht-rader.c -15 +6 Chris@10: Chris@10: Sat Sep 21 12:03:46 EDT 2002 athena Chris@10: * [project @ 2002-09-21 16:03:46 by athena] Chris@10: Check rank *before* reading kind[0], which may be undefined if rnk < 1 Chris@10: Chris@10: M ./tests/verify-rdft.c -2 +2 Chris@10: M ./tests/verify-reodft.c -2 +2 Chris@10: Chris@10: Sat Sep 21 11:48:50 EDT 2002 athena Chris@10: * [project @ 2002-09-21 15:48:50 by athena] Chris@10: Second step towards rader unification. Chris@10: Chris@10: M ./dft/rader.c -17 +7 Chris@10: M ./rdft/rader-hc2hc.c -1 +1 Chris@10: Chris@10: Sat Sep 21 11:37:06 EDT 2002 athena Chris@10: * [project @ 2002-09-21 15:37:06 by athena] Chris@10: First step towards unification of Rader code Chris@10: Chris@10: A ./kernel/rader.c Chris@10: M ./dft/rader.c -57 +12 Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -1 +9 Chris@10: M ./kernel/rader.c +68 Chris@10: M ./rdft/dht-rader.c -50 +6 Chris@10: M ./rdft/rader-hc2hc.c -57 +11 Chris@10: Chris@10: Sat Sep 21 07:58:11 EDT 2002 athena Chris@10: * [project @ 2002-09-21 11:58:11 by athena] Chris@10: Fix ugliness condition for cooley-tukey. Chris@10: Chris@10: A ./kernel/ct.c Chris@10: M ./dft/ct-dif.c -5 +3 Chris@10: M ./dft/ct-dit.c -4 +2 Chris@10: M ./dft/ct-ditbuf.c -5 +3 Chris@10: M ./kernel/Makefile.am -4 +5 Chris@10: M ./kernel/ct.c +31 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -3 +2 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./rdft/dht-rader.c -3 +3 Chris@10: M ./rdft/hc2hc-buf.c -9 +3 Chris@10: M ./rdft/hc2hc-dif.c -4 +2 Chris@10: M ./rdft/hc2hc-dit.c -4 +2 Chris@10: M ./rdft/rdft-dht.c -4 +2 Chris@10: M ./reodft/redft00e-r2hc.c -4 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -4 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -4 +2 Chris@10: M ./reodft/rodft00e-r2hc.c -4 +2 Chris@10: M ./threads/ct-dit.c -5 +3 Chris@10: M ./threads/hc2hc-dif.c -5 +3 Chris@10: M ./threads/hc2hc-dit.c -5 +3 Chris@10: Chris@10: Fri Sep 20 16:53:45 EDT 2002 athena Chris@10: * [project @ 2002-09-20 20:53:45 by athena] Chris@10: Removed RADER_MIN_GOOD and associated machinery Chris@10: Chris@10: M ./dft/rader.c -29 +6 Chris@10: M ./kernel/ifftw.h -2 +1 Chris@10: M ./rdft/dht-rader.c -14 +3 Chris@10: M ./rdft/rader-hc2hc.c -17 +5 Chris@10: Chris@10: Fri Sep 20 14:49:12 EDT 2002 athena Chris@10: * [project @ 2002-09-20 18:49:12 by athena] Chris@10: Proper cast Chris@10: Chris@10: M ./rdft/dht-r2hc.c -4 +3 Chris@10: Chris@10: Fri Sep 20 14:45:54 EDT 2002 athena Chris@10: * [project @ 2002-09-20 18:45:54 by athena] Chris@10: Typo Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Fri Sep 20 14:38:13 EDT 2002 athena Chris@10: * [project @ 2002-09-20 18:38:13 by athena] Chris@10: Implemented NO_LARGE_GENERIC Chris@10: Chris@10: M ./dft/generic.c -1 +8 Chris@10: M ./kernel/ifftw.h -20 +23 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/generic.c -2 +8 Chris@10: M ./rdft/rdft-dht.c -10 +2 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Thu Sep 19 07:48:25 EDT 2002 athena Chris@10: * [project @ 2002-09-19 11:48:24 by athena] Chris@10: Consistent macroization of NO_DHT_R2HC Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./rdft/dht-r2hc.c -4 +4 Chris@10: Chris@10: Wed Sep 18 21:47:17 EDT 2002 athena Chris@10: * [project @ 2002-09-19 01:47:17 by athena] Chris@10: NO_DHT_R2HC is a planner flag, otherwise the EXHAUSTIVE planner loops. Chris@10: Chris@10: M ./kernel/ifftw.h -7 +6 Chris@10: M ./kernel/planner.c -10 +6 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Wed Sep 18 20:47:31 EDT 2002 athena Chris@10: * [project @ 2002-09-19 00:47:31 by athena] Chris@10: Resurrected NO_EXHAUSTIVE Chris@10: Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/planner.c -8 +24 Chris@10: Chris@10: Wed Sep 18 19:31:57 EDT 2002 stevenj Chris@10: * [project @ 2002-09-18 23:31:57 by stevenj] Chris@10: au revoir, score() Chris@10: Chris@10: M ./threads/ct-dit.c -25 +22 Chris@10: M ./threads/dft-vrank-geq1.c -14 +10 Chris@10: M ./threads/hc2hc-dif.c -25 +21 Chris@10: M ./threads/hc2hc-dit.c -25 +21 Chris@10: M ./threads/rdft-vrank-geq1.c -14 +10 Chris@10: M ./threads/vrank-geq1-rdft2.c -14 +10 Chris@10: Chris@10: Wed Sep 18 19:31:05 EDT 2002 stevenj Chris@10: * [project @ 2002-09-18 23:31:05 by stevenj] Chris@10: eliminated unused Chris@10: Chris@10: M ./tests/bench.c +2 Chris@10: M ./tests/verify-reodft.c -7 +7 Chris@10: Chris@10: Wed Sep 18 18:28:44 EDT 2002 stevenj Chris@10: * [project @ 2002-09-18 22:28:44 by stevenj] Chris@10: capitalize and parenthesize SUBSUMES Chris@10: Chris@10: M ./kernel/planner.c -9 +8 Chris@10: Chris@10: Wed Sep 18 18:26:58 EDT 2002 stevenj Chris@10: * [project @ 2002-09-18 22:26:58 by stevenj] Chris@10: comment Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Wed Sep 18 18:03:18 EDT 2002 athena Chris@10: * [project @ 2002-09-18 22:03:18 by athena] Chris@10: Use flags from wisdom if wisdom is applicable. Chris@10: Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./kernel/planner.c -70 +32 Chris@10: Chris@10: Wed Sep 18 17:16:17 EDT 2002 athena Chris@10: * [project @ 2002-09-18 21:16:16 by athena] Chris@10: Removed score() machinery Chris@10: Chris@10: M ./dft/buffered.c -16 +11 Chris@10: M ./dft/ct-dif.c -23 +21 Chris@10: M ./dft/ct-dit.c -29 +25 Chris@10: M ./dft/ct-ditbuf.c -30 +24 Chris@10: M ./dft/ct-ditf.c -10 +2 Chris@10: M ./dft/direct.c -8 +2 Chris@10: M ./dft/generic.c -9 +7 Chris@10: M ./dft/indirect.c -9 +12 Chris@10: M ./dft/nop.c -8 +2 Chris@10: M ./dft/rader.c -23 +25 Chris@10: M ./dft/rank-geq2.c -14 +13 Chris@10: M ./dft/rank0.c -8 +2 Chris@10: M ./dft/vrank-geq1.c -30 +29 Chris@10: M ./dft/vrank2-transpose.c -8 +2 Chris@10: M ./dft/vrank3-transpose.c -13 +15 Chris@10: M ./kernel/ifftw.h -12 +2 Chris@10: M ./kernel/planner.c -31 +19 Chris@10: M ./rdft/buffered.c -18 +11 Chris@10: M ./rdft/buffered2.c -18 +11 Chris@10: M ./rdft/dft-r2hc.c -12 +14 Chris@10: M ./rdft/dht-r2hc.c -8 +8 Chris@10: M ./rdft/dht-rader.c -12 +12 Chris@10: M ./rdft/direct.c -9 +3 Chris@10: M ./rdft/direct2.c -9 +3 Chris@10: M ./rdft/generic.c -8 +7 Chris@10: M ./rdft/hc2hc-buf.c -30 +29 Chris@10: M ./rdft/hc2hc-dif.c -33 +25 Chris@10: M ./rdft/hc2hc-dit.c -32 +27 Chris@10: M ./rdft/indirect.c -9 +14 Chris@10: M ./rdft/nop.c -8 +2 Chris@10: M ./rdft/nop2.c -8 +2 Chris@10: M ./rdft/rader-hc2hc.c -13 +12 Chris@10: M ./rdft/rank-geq2-rdft2.c -15 +17 Chris@10: M ./rdft/rank-geq2.c -17 +19 Chris@10: M ./rdft/rank0.c -8 +2 Chris@10: M ./rdft/rdft-dht.c -11 +13 Chris@10: M ./rdft/rdft2-radix2.c -12 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -32 +29 Chris@10: M ./rdft/vrank-geq1.c -31 +31 Chris@10: M ./rdft/vrank2-transpose.c -8 +2 Chris@10: M ./rdft/vrank3-transpose.c -18 +14 Chris@10: M ./reodft/redft00e-r2hc.c -6 +7 Chris@10: M ./reodft/reodft010e-r2hc.c -6 +7 Chris@10: M ./reodft/reodft11e-r2hc.c -6 +7 Chris@10: M ./reodft/rodft00e-r2hc.c -6 +7 Chris@10: M ./tests/bench.c -2 Chris@10: Chris@10: Wed Sep 18 14:12:21 EDT 2002 athena Chris@10: * [project @ 2002-09-18 18:12:21 by athena] Chris@10: Revised planner hack Chris@10: Chris@10: M ./kernel/planner.c -4 +7 Chris@10: Chris@10: Wed Sep 18 10:14:41 EDT 2002 athena Chris@10: * [project @ 2002-09-18 14:14:41 by athena] Chris@10: Fix warning Chris@10: Chris@10: M ./simd/simd-altivec.h +2 Chris@10: Chris@10: Tue Sep 17 17:54:07 EDT 2002 athena Chris@10: * [project @ 2002-09-17 21:54:07 by athena] Chris@10: Type qualifiers. Chris@10: Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: Chris@10: Tue Sep 17 16:17:55 EDT 2002 athena Chris@10: * [project @ 2002-09-17 20:17:55 by athena] Chris@10: ESTIMATE is no longer subsumed by everything else. Chris@10: Chris@10: M ./kernel/planner.c -2 +1 Chris@10: Chris@10: Tue Sep 17 10:55:15 EDT 2002 athena Chris@10: * [project @ 2002-09-17 14:55:15 by athena] Chris@10: NO_BUFFERING is a planner flag, not a problem flag Chris@10: Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: Chris@10: Tue Sep 17 09:36:16 EDT 2002 athena Chris@10: * [project @ 2002-09-17 13:36:16 by athena] Chris@10: Maintain flags in canonical form. Chris@10: Chris@10: M ./kernel/ifftw.h -4 +6 Chris@10: M ./kernel/planner.c -13 +15 Chris@10: Chris@10: Tue Sep 17 09:09:57 EDT 2002 athena Chris@10: * [project @ 2002-09-17 13:09:56 by athena] Chris@10: In dramatic break with tradition, SUBSUME is now a partial order. I Chris@10: swear. Chris@10: Chris@10: M ./kernel/ifftw.h -4 +1 Chris@10: M ./kernel/planner.c -15 +54 Chris@10: Chris@10: Tue Sep 17 07:29:00 EDT 2002 athena Chris@10: * [project @ 2002-09-17 11:29:00 by athena] Chris@10: Added comment Chris@10: Chris@10: M ./kernel/planner.c -1 +3 Chris@10: Chris@10: Tue Sep 17 07:27:17 EDT 2002 athena Chris@10: * [project @ 2002-09-17 11:27:17 by athena] Chris@10: Inverted ESTIMATE flag, renamed USE_SCORE for consistency with the Chris@10: convention that 0 subsumes 1. Chris@10: Chris@10: M ./kernel/ifftw.h -9 +6 Chris@10: M ./kernel/planner.c -10 +7 Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Tue Sep 17 02:50:15 EDT 2002 stevenj Chris@10: * [project @ 2002-09-17 06:50:15 by stevenj] Chris@10: NO_INDIRECT -> NO_INDIRECT_OP (out-of-place only) Chris@10: Chris@10: M ./dft/indirect.c -4 +5 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./rdft/indirect.c -4 +5 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Tue Sep 17 00:40:04 EDT 2002 stevenj Chris@10: * [project @ 2002-09-17 04:40:04 by stevenj] Chris@10: hpux needs -D_REENTRANT (thanks to Clinton Roy for the bug report) Chris@10: Chris@10: M ./acx_pthread.m4 -2 +2 Chris@10: Chris@10: Mon Sep 16 23:54:34 EDT 2002 athena Chris@10: * [project @ 2002-09-17 03:54:34 by athena] Chris@10: Oops. Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Mon Sep 16 23:44:47 EDT 2002 athena Chris@10: * [project @ 2002-09-17 03:44:47 by athena] Chris@10: Yet another attempt at getting the planner right. Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./kernel/planner.c -28 +27 Chris@10: Chris@10: Mon Sep 16 21:56:14 EDT 2002 athena Chris@10: * [project @ 2002-09-17 01:56:14 by athena] Chris@10: Better coding. Chris@10: Chris@10: M ./kernel/planner.c -21 +11 Chris@10: Chris@10: Mon Sep 16 21:51:06 EDT 2002 athena Chris@10: * [project @ 2002-09-17 01:51:06 by athena] Chris@10: NO_UGLY is no longer a flag, but a separate planner field that does not Chris@10: interfere with wisdom. Chris@10: Chris@10: M ./kernel/ifftw.h -3 +2 Chris@10: M ./kernel/planner.c -14 +15 Chris@10: Chris@10: Mon Sep 16 19:04:41 EDT 2002 athena Chris@10: * [project @ 2002-09-16 23:04:41 by athena] Chris@10: Did not compile without FFTW_DEBUG Chris@10: Chris@10: M ./tests/verify-reodft.c -3 +1 Chris@10: Chris@10: Mon Sep 16 18:37:06 EDT 2002 athena Chris@10: * [project @ 2002-09-16 22:37:06 by athena] Chris@10: Changed scoring mechanism. Chris@10: Chris@10: M ./kernel/ifftw.h -5 +4 Chris@10: M ./kernel/plan.c -5 +1 Chris@10: M ./kernel/planner.c -51 +28 Chris@10: M ./tests/bench.c -5 +2 Chris@10: Chris@10: Mon Sep 16 17:13:45 EDT 2002 athena Chris@10: * [project @ 2002-09-16 21:13:45 by athena] Chris@10: Count infeasible plans Chris@10: Chris@10: M ./kernel/planner.c -4 +12 Chris@10: Chris@10: Mon Sep 16 16:36:12 EDT 2002 athena Chris@10: * [project @ 2002-09-16 20:36:12 by athena] Chris@10: curse subsumed plans before export Chris@10: Chris@10: M ./kernel/planner.c -27 +35 Chris@10: Chris@10: Mon Sep 16 15:40:46 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 19:40:46 by stevenj] Chris@10: removed ESTIMATE_BIT vs. ESTIMATE... ESTIMATE | IMPATIENT is a UI issue Chris@10: Chris@10: M ./kernel/ifftw.h -6 +4 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Mon Sep 16 15:31:39 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 19:31:39 by stevenj] Chris@10: cleanup Chris@10: Chris@10: M ./rdft/buffered2.c -6 +3 Chris@10: Chris@10: Mon Sep 16 15:28:47 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 19:28:47 by stevenj] Chris@10: use CONSERVE_MEMORY flag to prevent buffered for large sizes Chris@10: Chris@10: M ./dft/buffered.c -4 +7 Chris@10: M ./rdft/buffered.c -4 +7 Chris@10: M ./rdft/buffered2.c -5 +9 Chris@10: Chris@10: Mon Sep 16 15:16:16 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 19:16:16 by stevenj] Chris@10: moved NO_DHT_R2HC back into planner flags: there's no reason we would want this flag to block plan reuse Chris@10: Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: Chris@10: Mon Sep 16 14:59:14 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 18:59:14 by stevenj] Chris@10: whoops, commas Chris@10: Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: Chris@10: Mon Sep 16 14:58:26 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 18:58:26 by stevenj] Chris@10: problem_flags == checked in applicable, planner_flags == checked in score Chris@10: Chris@10: M ./kernel/ifftw.h -5 +7 Chris@10: Chris@10: Mon Sep 16 14:53:16 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 18:53:16 by stevenj] Chris@10: ESTIMATE should not *include* all impatience flags, even if it subsumes them; some impatience flags, like NO_INDIRECT, might make a problem unsolvable Chris@10: Chris@10: M ./kernel/ifftw.h -6 +6 Chris@10: M ./kernel/planner.c -3 +4 Chris@10: Chris@10: Mon Sep 16 00:56:29 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 04:56:29 by stevenj] Chris@10: quotatio marks Chris@10: Chris@10: M ./kernel/planner.c -3 +3 Chris@10: Chris@10: Sun Sep 15 23:55:44 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 03:55:44 by stevenj] Chris@10: delete blank line Chris@10: Chris@10: M ./kernel/planner.c -2 +1 Chris@10: Chris@10: Sun Sep 15 23:51:14 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 03:51:14 by stevenj] Chris@10: substitution Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Sun Sep 15 23:49:50 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 03:49:50 by stevenj] Chris@10: note that we are not GNUlly correct Chris@10: Chris@10: M ./kernel/planner.c -1 +10 Chris@10: Chris@10: Sun Sep 15 23:41:01 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 03:41:01 by stevenj] Chris@10: indenting Chris@10: Chris@10: M ./kernel/planner.c -3 +3 Chris@10: Chris@10: Sun Sep 15 23:37:46 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 03:37:46 by stevenj] Chris@10: more jokes Chris@10: Chris@10: M ./kernel/planner.c -2 +5 Chris@10: Chris@10: Sun Sep 15 23:20:14 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 03:20:14 by stevenj] Chris@10: NONTHREADED_ICKYP includes nthr > 1 check Chris@10: Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./rdft/hc2hc-dif.c -2 +2 Chris@10: M ./rdft/hc2hc-dit.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Sun Sep 15 22:56:44 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 02:56:44 by stevenj] Chris@10: use md5sig Chris@10: Chris@10: M ./kernel/md5.c -1 +1 Chris@10: Chris@10: Sun Sep 15 22:55:41 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 02:55:41 by stevenj] Chris@10: md5sig typedef Chris@10: Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/planner.c -9 +9 Chris@10: Chris@10: Sun Sep 15 22:35:13 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 02:35:13 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +324 Chris@10: Chris@10: Sun Sep 15 22:30:26 EDT 2002 stevenj Chris@10: * [project @ 2002-09-16 02:30:26 by stevenj] Chris@10: partially-ordered impatience Chris@10: Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/ct-dif.c -3 +3 Chris@10: M ./dft/ct-dit.c -3 +3 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/ct.c -5 +1 Chris@10: M ./dft/indirect.c -4 +4 Chris@10: M ./dft/rank-geq2.c -3 +2 Chris@10: M ./dft/vrank-geq1.c -13 +3 Chris@10: M ./kernel/ifftw.h -18 +51 Chris@10: M ./kernel/planner.c -11 +49 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -2 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: M ./rdft/hc2hc-buf.c -4 +3 Chris@10: M ./rdft/hc2hc-dif.c -4 +4 Chris@10: M ./rdft/hc2hc-dit.c -3 +3 Chris@10: M ./rdft/hc2hc.c -5 +1 Chris@10: M ./rdft/indirect.c -4 +4 Chris@10: M ./rdft/rank-geq2-rdft2.c -3 +2 Chris@10: M ./rdft/rank-geq2.c -3 +2 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -13 +3 Chris@10: M ./rdft/vrank-geq1.c -13 +3 Chris@10: M ./tests/bench.c -3 Chris@10: M ./threads/dft-vrank-geq1.c -13 +2 Chris@10: M ./threads/hc2hc-dif.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -12 +2 Chris@10: M ./threads/vrank-geq1-rdft2.c -12 +2 Chris@10: Chris@10: Sat Sep 14 19:47:56 EDT 2002 athena Chris@10: * [project @ 2002-09-14 23:47:56 by athena] Chris@10: Removed all that planner inheritance crap. Chris@10: Chris@10: M ./kernel/Makefile.am -4 +3 Chris@10: M ./kernel/ifftw.h -15 +5 Chris@10: R ./kernel/planner-naive.c Chris@10: R ./kernel/planner-score.c Chris@10: M ./kernel/planner.c -35 +128 Chris@10: M ./tests/bench.c -2 +2 Chris@10: Chris@10: Sat Sep 14 16:35:28 EDT 2002 stevenj Chris@10: * [project @ 2002-09-14 20:35:28 by stevenj] Chris@10: string.h is used for more than strlen Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Sat Sep 14 12:19:13 EDT 2002 athena Chris@10: * [project @ 2002-09-14 16:19:13 by athena] Chris@10: Reduced hashtable size by 1/6 (on 32-bit machines) at the expense Chris@10: of messier planner. Chris@10: Chris@10: M ./kernel/ifftw.h -13 +17 Chris@10: M ./kernel/planner.c -80 +87 Chris@10: Chris@10: Sat Sep 14 08:31:29 EDT 2002 athena Chris@10: * [project @ 2002-09-14 12:31:29 by athena] Chris@10: Only print wisdom if verbose > 3 Chris@10: Chris@10: M ./tests/bench.c -2 +4 Chris@10: Chris@10: Sat Sep 14 07:56:56 EDT 2002 athena Chris@10: * [project @ 2002-09-14 11:56:56 by athena] Chris@10: Changed syntax of temporaries to avoid shadowing library functions Chris@10: (which is harmless but I hate the warning) Chris@10: Chris@10: M ./genfft/variable.ml -2 +2 Chris@10: M ./genfft-k7/variable.ml -1 +1 Chris@10: Chris@10: Fri Sep 13 23:07:39 EDT 2002 stevenj Chris@10: * [project @ 2002-09-14 03:07:39 by stevenj] Chris@10: only add warnings in debug/maintainer mode, and add a few more warning flags; eliminate more warnings; add support for posix_memalign (broken in glibc, grrr) Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: M ./configure.ac -1 +9 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./kernel/alloc.c -3 +16 Chris@10: M ./kernel/assert.c -2 +2 Chris@10: M ./kernel/ifftw.h -5 +5 Chris@10: M ./kernel/md5.c -1 +1 Chris@10: M ./kernel/planner-score.c -11 +13 Chris@10: M ./kernel/primes.c -4 +4 Chris@10: M ./kernel/scan.c -2 +2 Chris@10: M ./libbench/bench-user.h -2 +2 Chris@10: M ./libbench/bench.h -3 +4 Chris@10: M ./libbench/report.c -5 +5 Chris@10: M ./libbench/timer.c -1 +2 Chris@10: M ./libbench/util.c -14 +26 Chris@10: M ./libbench/verify.c -4 +1 Chris@10: M ./rdft/rader-hc2hc.c -6 +6 Chris@10: M ./tests/bench.c -3 Chris@10: M ./tests/verify-lib.c -1 +4 Chris@10: Chris@10: Fri Sep 13 21:57:50 EDT 2002 athena Chris@10: * [project @ 2002-09-14 01:57:50 by athena] Chris@10: Explicit cast Chris@10: Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: Chris@10: Fri Sep 13 21:54:50 EDT 2002 athena Chris@10: * [project @ 2002-09-14 01:54:50 by athena] Chris@10: Use double-hashing. This allows a slightly higher load factor Chris@10: at the expense of a messier computation of the hashtable size. Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -57 +63 Chris@10: M ./kernel/primes.c -1 +7 Chris@10: Chris@10: Fri Sep 13 17:53:13 EDT 2002 stevenj Chris@10: * [project @ 2002-09-13 21:53:13 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./genfft/magic.ml -2 +2 Chris@10: Chris@10: Fri Sep 13 15:36:07 EDT 2002 athena Chris@10: * [project @ 2002-09-13 19:36:07 by athena] Chris@10: Slight change in hash table growth functions. Chris@10: Chris@10: M ./kernel/planner.c -3 +12 Chris@10: Chris@10: Fri Sep 13 14:58:22 EDT 2002 athena Chris@10: * [project @ 2002-09-13 18:58:22 by athena] Chris@10: More statistics. Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -1 +8 Chris@10: Chris@10: Fri Sep 13 10:13:02 EDT 2002 athena Chris@10: * [project @ 2002-09-13 14:13:02 by athena] Chris@10: Clearer logic. Chris@10: Chris@10: M ./kernel/planner.c -2 +3 Chris@10: Chris@10: Fri Sep 13 10:11:10 EDT 2002 athena Chris@10: * [project @ 2002-09-13 14:11:10 by athena] Chris@10: Oops. Chris@10: Chris@10: M ./kernel/planner.c -2 +1 Chris@10: Chris@10: Fri Sep 13 09:31:46 EDT 2002 athena Chris@10: * [project @ 2002-09-13 13:31:46 by athena] Chris@10: Cleaned up Chris@10: Chris@10: M ./kernel/planner.c -5 +6 Chris@10: Chris@10: Fri Sep 13 09:16:07 EDT 2002 athena Chris@10: * [project @ 2002-09-13 13:16:07 by athena] Chris@10: Deal properly with infeasible problems. Chris@10: Chris@10: M ./kernel/planner.c -4 +7 Chris@10: Chris@10: Fri Sep 13 07:15:06 EDT 2002 athena Chris@10: * [project @ 2002-09-13 11:15:06 by athena] Chris@10: Redundantly initialize hash table to prevent valgrind warnings. Chris@10: Chris@10: M ./kernel/planner.c -3 +7 Chris@10: Chris@10: Thu Sep 12 19:00:22 EDT 2002 athena Chris@10: * [project @ 2002-09-12 23:00:22 by athena] Chris@10: Removed relics from past. Chris@10: Chris@10: M ./kernel/md5.c -5 +3 Chris@10: Chris@10: Thu Sep 12 18:53:44 EDT 2002 athena Chris@10: * [project @ 2002-09-12 22:53:44 by athena] Chris@10: md5hash a problem only once. Chris@10: Chris@10: M ./kernel/ifftw.h -6 +11 Chris@10: M ./kernel/planner.c -39 +32 Chris@10: Chris@10: Thu Sep 12 16:33:49 EDT 2002 athena Chris@10: * [project @ 2002-09-12 20:33:49 by athena] Chris@10: Renamed k7 codelets Chris@10: Chris@10: M ./genfft-k7/genUtil.ml -1 +1 Chris@10: M ./genfft-k7/gen_notw.ml -3 +3 Chris@10: M ./genfft-k7/gen_twiddle.ml -4 +4 Chris@10: Chris@10: Thu Sep 12 16:32:03 EDT 2002 stevenj Chris@10: * [project @ 2002-09-12 20:32:03 by stevenj] Chris@10: FORBID_DHT_R2HC -> DHT_R2HC_VERBOTEN for consistency Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: Chris@10: Thu Sep 12 16:28:43 EDT 2002 stevenj Chris@10: * [project @ 2002-09-12 20:28:43 by stevenj] Chris@10: removed obsolete macro Chris@10: Chris@10: M ./kernel/ifftw.h -3 +1 Chris@10: Chris@10: Thu Sep 12 16:20:39 EDT 2002 athena Chris@10: * [project @ 2002-09-12 20:20:39 by athena] Chris@10: Split flags in SIMD code. Chris@10: Chris@10: M ./dft/simd/n1b.c -1 +1 Chris@10: M ./dft/simd/n1f.c -1 +1 Chris@10: M ./dft/simd/t1b.c -1 +1 Chris@10: M ./dft/simd/t1f.c -1 +1 Chris@10: Chris@10: Thu Sep 12 16:18:51 EDT 2002 athena Chris@10: * [project @ 2002-09-12 20:18:51 by athena] Chris@10: Forgot to fix threads Chris@10: Chris@10: M ./threads/dft-vrank-geq1.c -5 +6 Chris@10: M ./threads/hc2hc-dif.c -2 +2 Chris@10: M ./threads/rdft-vrank-geq1.c -5 +6 Chris@10: M ./threads/vrank-geq1-rdft2.c -5 +6 Chris@10: Chris@10: Thu Sep 12 16:10:05 EDT 2002 athena Chris@10: * [project @ 2002-09-12 20:10:05 by athena] Chris@10: Split flags into planner_flags and problem_flags Chris@10: Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/ct-dif.c -3 +3 Chris@10: M ./dft/ct-dit.c -3 +3 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/ct.c -3 +3 Chris@10: M ./dft/indirect.c -4 +4 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./dft/rank-geq2.c -2 +3 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -7 +8 Chris@10: M ./kernel/ifftw.h -22 +21 Chris@10: M ./kernel/planner-naive.c -3 +3 Chris@10: M ./kernel/planner-score.c -3 +3 Chris@10: M ./kernel/planner.c -33 +28 Chris@10: M ./rdft/buffered.c -2 +2 Chris@10: M ./rdft/buffered2.c -3 +3 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/dht-r2hc.c -3 +5 Chris@10: M ./rdft/dht-rader.c -1 +1 Chris@10: M ./rdft/hc2hc-buf.c -3 +3 Chris@10: M ./rdft/hc2hc-dif.c -4 +4 Chris@10: M ./rdft/hc2hc-dit.c -3 +3 Chris@10: M ./rdft/hc2hc.c -3 +3 Chris@10: M ./rdft/indirect.c -4 +4 Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +3 Chris@10: M ./rdft/rank-geq2.c -2 +3 Chris@10: M ./rdft/rdft-dht.c -3 +3 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -7 +8 Chris@10: M ./rdft/vrank-geq1.c -7 +8 Chris@10: M ./tests/bench.c -8 +8 Chris@10: Chris@10: Thu Sep 12 15:46:56 EDT 2002 stevenj Chris@10: * [project @ 2002-09-12 19:46:56 by stevenj] Chris@10: tetrameter Chris@10: Chris@10: M ./kernel/planner.c -2 +3 Chris@10: Chris@10: Thu Sep 12 15:11:21 EDT 2002 athena Chris@10: * [project @ 2002-09-12 19:11:21 by athena] Chris@10: Overwrite less impatient solutions properly. Chris@10: Chris@10: M ./kernel/planner.c -23 +15 Chris@10: Chris@10: Thu Sep 12 11:29:16 EDT 2002 athena Chris@10: * [project @ 2002-09-12 15:29:16 by athena] Chris@10: Oops. Chris@10: Chris@10: M ./kernel/planner.c -3 +3 Chris@10: Chris@10: Thu Sep 12 10:58:56 EDT 2002 athena Chris@10: * [project @ 2002-09-12 14:58:56 by athena] Chris@10: Keep less impatient solution in case of conflict. Paranoid Chris@10: cast to uint in certain places. Chris@10: Chris@10: M ./kernel/planner.c -4 +5 Chris@10: Chris@10: Thu Sep 12 10:02:51 EDT 2002 athena Chris@10: * [project @ 2002-09-12 14:02:51 by athena] Chris@10: Complete reimplementation of planner hash table. Chris@10: Chris@10: M ./kernel/ifftw.h -6 +7 Chris@10: M ./kernel/planner.c -143 +146 Chris@10: M ./tests/bench.c -2 +4 Chris@10: Chris@10: Thu Sep 12 07:58:45 EDT 2002 athena Chris@10: * [project @ 2002-09-12 11:58:45 by athena] Chris@10: planner->cnt was not properly decremented. Chris@10: Chris@10: M ./kernel/planner.c -13 +14 Chris@10: Chris@10: Wed Sep 11 17:52:39 EDT 2002 stevenj Chris@10: * [project @ 2002-09-11 21:52:39 by stevenj] Chris@10: typo Chris@10: Chris@10: M ./NEWS -1 +1 Chris@10: Chris@10: Mon Sep 9 17:10:45 EDT 2002 athena Chris@10: * [project @ 2002-09-09 21:10:45 by athena] Chris@10: Simplified Chris@10: Chris@10: M ./kernel/planner.c -15 +5 Chris@10: Chris@10: Mon Sep 9 17:03:32 EDT 2002 athena Chris@10: * [project @ 2002-09-09 21:03:32 by athena] Chris@10: Always overwrite old wisdom with new, in case the old is Chris@10: corrupt/conclicting. Chris@10: Chris@10: M ./kernel/planner.c -16 +8 Chris@10: Chris@10: Mon Sep 9 16:56:03 EDT 2002 stevenj Chris@10: * [project @ 2002-09-09 20:56:03 by stevenj] Chris@10: added quote/joke Chris@10: Chris@10: M ./kernel/plan.c -1 +5 Chris@10: Chris@10: Mon Sep 9 15:04:47 EDT 2002 athena Chris@10: * [project @ 2002-09-09 19:04:47 by athena] Chris@10: Completed wisdom import Chris@10: Chris@10: M ./kernel/ifftw.h -10 +11 Chris@10: M ./kernel/md5.c -1 +1 Chris@10: M ./kernel/planner.c -37 +54 Chris@10: M ./kernel/print.c -2 +3 Chris@10: M ./kernel/scan.c -73 +25 Chris@10: M ./tests/bench.c -3 +3 Chris@10: Chris@10: Mon Sep 9 10:14:22 EDT 2002 athena Chris@10: * [project @ 2002-09-09 14:14:22 by athena] Chris@10: Slight cleanup of md5 interface. Chris@10: Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/md5.c -12 +24 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: Chris@10: Tue Sep 3 22:32:43 EDT 2002 athena Chris@10: * [project @ 2002-09-04 02:32:43 by athena] Chris@10: More consistent protocol between planner and inferior. Chris@10: Chris@10: M ./kernel/planner-naive.c -2 +5 Chris@10: M ./kernel/planner-score.c -1 +2 Chris@10: M ./kernel/planner.c -3 +1 Chris@10: Chris@10: Tue Sep 3 21:08:30 EDT 2002 athena Chris@10: * [project @ 2002-09-04 01:08:30 by athena] Chris@10: I can't think of any situation where saving infeasible problems would Chris@10: be desirable. Removed relevant code. Chris@10: Chris@10: M ./kernel/planner.c -9 +3 Chris@10: Chris@10: Tue Sep 3 20:57:03 EDT 2002 athena Chris@10: * [project @ 2002-09-04 00:57:03 by athena] Chris@10: Encoder registrar's names in wisdom. Remove export_conf, since Chris@10: a separate program can now generate it. Chris@10: Chris@10: M ./kernel/ifftw.h -5 +5 Chris@10: M ./kernel/planner.c -108 +32 Chris@10: M ./kernel/solvtab.c -1 +2 Chris@10: M ./tests/bench.c -3 +1 Chris@10: Chris@10: Tue Sep 3 15:11:06 EDT 2002 athena Chris@10: * [project @ 2002-09-03 19:11:06 by athena] Chris@10: Fixed typo Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Tue Sep 3 14:52:45 EDT 2002 athena Chris@10: * [project @ 2002-09-03 18:52:45 by athena] Chris@10: Fixed broken trochaic meter. Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Tue Sep 3 09:49:50 EDT 2002 athena Chris@10: * [project @ 2002-09-03 13:49:50 by athena] Chris@10: Initialize planner->score. It is correct to leave it uninitialized, Chris@10: but I don't want people to send reports about purify complaining. Chris@10: Chris@10: M ./kernel/planner.c -1 +2 Chris@10: Chris@10: Tue Sep 3 09:03:46 EDT 2002 athena Chris@10: * [project @ 2002-09-03 13:03:46 by athena] Chris@10: More latin silliness Chris@10: Chris@10: M ./kernel/planner.c -3 +7 Chris@10: Chris@10: Mon Sep 2 17:57:32 EDT 2002 stevenj Chris@10: * [project @ 2002-09-02 21:57:32 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +193 Chris@10: Chris@10: Mon Sep 2 17:33:49 EDT 2002 stevenj Chris@10: * [project @ 2002-09-02 21:33:49 by stevenj] Chris@10: added clock() getseconds timer Chris@10: Chris@10: M ./kernel/timer.c -1 +16 Chris@10: Chris@10: Mon Sep 2 16:16:58 EDT 2002 athena Chris@10: * [project @ 2002-09-02 20:16:58 by athena] Chris@10: Oops Chris@10: Chris@10: M ./rdft/indirect.c -2 +1 Chris@10: Chris@10: Mon Sep 2 15:58:19 EDT 2002 athena Chris@10: * [project @ 2002-09-02 19:58:19 by athena] Chris@10: Experimental INDIRECT_VERBOTEN flag (not used) Chris@10: Chris@10: M ./dft/indirect.c -1 +3 Chris@10: M ./kernel/ifftw.h -5 +6 Chris@10: M ./rdft/indirect.c -3 +4 Chris@10: Chris@10: Mon Sep 2 15:36:21 EDT 2002 athena Chris@10: * [project @ 2002-09-02 19:36:21 by athena] Chris@10: Do not allow buffering in children of indirect solvers. Chris@10: Chris@10: M ./dft/buffered.c -1 +4 Chris@10: M ./dft/indirect.c -1 +3 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./rdft/buffered.c -1 +4 Chris@10: M ./rdft/buffered2.c -1 +4 Chris@10: M ./rdft/indirect.c -1 +3 Chris@10: Chris@10: Mon Sep 2 15:02:11 EDT 2002 athena Chris@10: * [project @ 2002-09-02 19:02:11 by athena] Chris@10: Oops Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Mon Sep 2 14:32:28 EDT 2002 athena Chris@10: * [project @ 2002-09-02 18:32:28 by athena] Chris@10: Hash sizeof(R) as part of wisdom. Chris@10: Chris@10: M ./kernel/planner.c -1 +2 Chris@10: Chris@10: Mon Sep 2 13:47:57 EDT 2002 stevenj Chris@10: * [project @ 2002-09-02 17:47:57 by stevenj] Chris@10: added --enable-float synonym for --enable-single (since with have --enable-long-double) Chris@10: Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Mon Sep 2 13:46:08 EDT 2002 athena Chris@10: * [project @ 2002-09-02 17:46:08 by athena] Chris@10: zerotens is now in its own file, so it does not cause dft to be linked Chris@10: in if only rdft is used. Chris@10: Chris@10: A ./dft/zero.c Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/problem.c -25 +1 Chris@10: M ./dft/zero.c +49 Chris@10: Chris@10: Mon Sep 2 11:56:37 EDT 2002 athena Chris@10: * [project @ 2002-09-02 15:56:37 by athena] Chris@10: Removed unused var. Chris@10: Chris@10: M ./kernel/planner.c -2 +1 Chris@10: Chris@10: Mon Sep 2 11:55:33 EDT 2002 athena Chris@10: * [project @ 2002-09-02 15:55:33 by athena] Chris@10: Split insert() in preparation for wisdom import Chris@10: Chris@10: M ./kernel/planner.c -5 +11 Chris@10: Chris@10: Mon Sep 2 11:46:57 EDT 2002 athena Chris@10: * [project @ 2002-09-02 15:46:57 by athena] Chris@10: Moved debugging infrastructure to test directory so that it is not Chris@10: linked into the shared library. Chris@10: Chris@10: A ./tests/debug.h Chris@10: A ./tests/dotens.c Chris@10: A ./tests/dotens2.c Chris@10: A ./tests/verify-dft.c Chris@10: A ./tests/verify-lib.c Chris@10: A ./tests/verify-rdft.c Chris@10: A ./tests/verify-reodft.c Chris@10: A ./tests/verify.h Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/dft.h -4 +1 Chris@10: R ./dft/verify.c Chris@10: M ./kernel/Makefile.am -6 +5 Chris@10: R ./kernel/dotens.c Chris@10: R ./kernel/dotens2.c Chris@10: M ./kernel/ifftw.h -17 +1 Chris@10: R ./kernel/verify-lib.c Chris@10: R ./kernel/verify.h Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/rdft.h -4 +1 Chris@10: R ./rdft/verify.c Chris@10: M ./reodft/Makefile.am -1 +1 Chris@10: M ./reodft/reodft.h -4 +1 Chris@10: R ./reodft/verify.c Chris@10: M ./tests/Makefile.am -2 +7 Chris@10: M ./tests/bench.c +3 Chris@10: M ./tests/debug.h +18 Chris@10: M ./tests/dotens.c +48 Chris@10: M ./tests/dotens2.c +56 Chris@10: M ./tests/verify-dft.c +131 Chris@10: M ./tests/verify-lib.c +386 Chris@10: M ./tests/verify-rdft.c +420 Chris@10: M ./tests/verify-reodft.c +536 Chris@10: M ./tests/verify.h +83 Chris@10: Chris@10: Mon Sep 2 11:04:54 EDT 2002 athena Chris@10: * [project @ 2002-09-02 15:04:53 by athena] Chris@10: Reactivated wisdom export Chris@10: Chris@10: M ./kernel/planner.c -42 +9 Chris@10: M ./kernel/print.c -1 +12 Chris@10: Chris@10: Sun Sep 1 21:30:58 EDT 2002 athena Chris@10: * [project @ 2002-09-02 01:30:58 by athena] Chris@10: Dump errors to stderr, not stdout. Chris@10: Chris@10: M ./kernel/verify-lib.c -2 +3 Chris@10: Chris@10: Sun Sep 1 21:26:38 EDT 2002 athena Chris@10: * [project @ 2002-09-02 01:26:38 by athena] Chris@10: Removed traverse.c. Chris@10: traverse.c is no longer need for plan blessing. I figured out Chris@10: a way to avoid using it in planner-score.c, so the file is Chris@10: now redundant. Chris@10: Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -9 +2 Chris@10: M ./kernel/planner-score.c -30 +37 Chris@10: R ./kernel/traverse.c Chris@10: M ./tests/bench.c -20 Chris@10: Chris@10: Sun Sep 1 19:51:50 EDT 2002 athena Chris@10: * [project @ 2002-09-01 23:51:50 by athena] Chris@10: Removed code made obsolete by new MD5 scheme: problem equality Chris@10: tests, scanners, and associated list of problem kinds. Chris@10: Chris@10: M ./dft/conf.c -3 +1 Chris@10: M ./dft/dft.h -3 +1 Chris@10: M ./dft/problem.c -58 +3 Chris@10: M ./kernel/align.c -7 +1 Chris@10: M ./kernel/ifftw.h -24 +4 Chris@10: M ./kernel/planner.c -25 +10 Chris@10: M ./kernel/problem.c -11 +2 Chris@10: M ./kernel/scan.c -44 +2 Chris@10: M ./kernel/scanners.c -5 +5 Chris@10: M ./kernel/tensor.c -46 +1 Chris@10: M ./rdft/conf.c -4 +1 Chris@10: M ./rdft/problem.c -78 +3 Chris@10: M ./rdft/problem2.c -58 +3 Chris@10: M ./rdft/rdft.h -4 +1 Chris@10: Chris@10: Sun Sep 1 19:22:54 EDT 2002 athena Chris@10: * [project @ 2002-09-01 23:22:53 by athena] Chris@10: Started md5 implementation Chris@10: Chris@10: A ./kernel/md5.c Chris@10: M ./dft/problem.c -9 +9 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -3 +29 Chris@10: M ./kernel/md5.c +164 Chris@10: M ./kernel/planner-naive.c -3 +3 Chris@10: M ./kernel/planner-score.c -7 +11 Chris@10: M ./kernel/planner.c -27 +47 Chris@10: M ./kernel/tensor.c -9 +7 Chris@10: M ./rdft/problem.c -12 +11 Chris@10: M ./rdft/problem2.c -10 +11 Chris@10: Chris@10: Sat Aug 31 14:00:04 EDT 2002 athena Chris@10: * [project @ 2002-08-31 18:00:04 by athena] Chris@10: Keep track of hit rate Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -3 +9 Chris@10: Chris@10: Sat Aug 31 12:44:04 EDT 2002 athena Chris@10: * [project @ 2002-08-31 16:44:04 by athena] Chris@10: Only dump when verbose > 4 Chris@10: Chris@10: M ./kernel/planner.c -2 +2 Chris@10: Chris@10: Sat Aug 31 09:55:57 EDT 2002 athena Chris@10: * [project @ 2002-08-31 13:55:57 by athena] Chris@10: Debugging infrastructure Chris@10: Chris@10: M ./dft/indirect.c -2 +1 Chris@10: M ./kernel/ifftw.h -5 +6 Chris@10: M ./kernel/plan.c -1 +6 Chris@10: M ./kernel/planner.c -7 +12 Chris@10: M ./tests/bench.c -3 +7 Chris@10: Chris@10: Sat Aug 31 09:21:48 EDT 2002 athena Chris@10: * [project @ 2002-08-31 13:21:48 by athena] Chris@10: Use debug infrastructure to dump planner. Chris@10: Chris@10: M ./kernel/planner.c -19 +10 Chris@10: M ./kernel/print.c -2 +5 Chris@10: Chris@10: Fri Aug 30 21:29:10 EDT 2002 athena Chris@10: * [project @ 2002-08-31 01:29:10 by athena] Chris@10: Do not store plans in planner, plus general planner cleanup. Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/ifftw.h -34 +37 Chris@10: M ./kernel/plan.c -15 +1 Chris@10: M ./kernel/planner-naive.c -7 +7 Chris@10: M ./kernel/planner-score.c -8 +7 Chris@10: M ./kernel/planner.c -103 +80 Chris@10: M ./kernel/scan.c -3 +3 Chris@10: M ./kernel/scanners.c -3 +3 Chris@10: M ./tests/bench.c -3 +16 Chris@10: Chris@10: Fri Aug 30 18:07:52 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 22:07:52 by stevenj] Chris@10: renamed IN_DHT_R2HC to the more general FORBID_DHT_R2HC Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./rdft/dht-r2hc.c -3 +3 Chris@10: Chris@10: Fri Aug 30 18:07:21 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 22:07:21 by stevenj] Chris@10: eliminated unused var Chris@10: Chris@10: M ./kernel/planner.c -2 +1 Chris@10: Chris@10: Fri Aug 30 12:09:48 EDT 2002 athena Chris@10: * [project @ 2002-08-30 16:09:48 by athena] Chris@10: Score planner was not working correctly when using wisdom. Fixed. Chris@10: Chris@10: M ./kernel/planner-naive.c -2 +8 Chris@10: M ./kernel/planner-score.c -2 +13 Chris@10: M ./kernel/planner.c -13 +18 Chris@10: Chris@10: Fri Aug 30 08:20:48 EDT 2002 athena Chris@10: * [project @ 2002-08-30 12:20:48 by athena] Chris@10: Use hash table in debug malloc Chris@10: Chris@10: M ./kernel/alloc.c -12 +25 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/stride.c -2 +2 Chris@10: Chris@10: Fri Aug 30 02:45:15 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 06:45:15 by stevenj] Chris@10: listed some good stuff Chris@10: Chris@10: M ./NEWS +47 Chris@10: Chris@10: Fri Aug 30 02:17:30 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 06:17:30 by stevenj] Chris@10: timed planner Chris@10: Chris@10: M ./TODO +4 Chris@10: Chris@10: Fri Aug 30 02:07:00 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 06:07:00 by stevenj] Chris@10: fma? Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Fri Aug 30 02:05:55 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 06:05:55 by stevenj] Chris@10: update Chris@10: Chris@10: M ./TODO -3 +3 Chris@10: Chris@10: Fri Aug 30 01:31:47 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 05:31:47 by stevenj] Chris@10: rader-dht -> dht-rader Chris@10: Chris@10: A ./rdft/dht-rader.c Chris@10: M ./rdft/Makefile.am -4 +4 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/dht-rader.c +422 Chris@10: R ./rdft/rader-dht.c Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: Chris@10: Fri Aug 30 01:21:37 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 05:21:37 by stevenj] Chris@10: add DHT solver, and break up rader-dht and r2hc-hc2r Chris@10: Chris@10: A ./rdft/dht-r2hc.c Chris@10: A ./rdft/rdft-dht.c Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./rdft/Makefile.am -6 +6 Chris@10: M ./rdft/buffered2.c -2 +4 Chris@10: M ./rdft/conf.c -4 +5 Chris@10: M ./rdft/dht-r2hc.c +151 Chris@10: R ./rdft/r2hc-hc2r.c Chris@10: M ./rdft/rader-dht.c -73 +21 Chris@10: M ./rdft/rank-geq2.c -1 +12 Chris@10: M ./rdft/rdft-dht.c +229 Chris@10: M ./rdft/rdft.h -3 +4 Chris@10: Chris@10: Thu Aug 29 23:20:35 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 03:20:35 by stevenj] Chris@10: another option Chris@10: Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Thu Aug 29 22:55:29 EDT 2002 stevenj Chris@10: * [project @ 2002-08-30 02:55:29 by stevenj] Chris@10: generalized indirect solvers for fftw2-like buffering and more Chris@10: Chris@10: M ./dft/indirect.c -22 +35 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/tensor.c -13 +17 Chris@10: M ./rdft/indirect.c -18 +35 Chris@10: Chris@10: Thu Aug 29 18:08:16 EDT 2002 stevenj Chris@10: * [project @ 2002-08-29 22:08:16 by stevenj] Chris@10: tensor_max_index and tensor_min_stride are now both unsigned Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +3 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./kernel/tensor.c -11 +7 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +3 Chris@10: M ./rdft/vrank-geq1.c -2 +3 Chris@10: Chris@10: Thu Aug 29 17:58:35 EDT 2002 stevenj Chris@10: * [project @ 2002-08-29 21:58:35 by stevenj] Chris@10: added iabs.c, and tensor_min_stride returns min absolute value Chris@10: Chris@10: A ./kernel/iabs.c Chris@10: M ./kernel/Makefile.am -5 +5 Chris@10: M ./kernel/iabs.c +28 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: M ./kernel/tensor.c -12 +6 Chris@10: M ./rdft/buffered2.c -9 +4 Chris@10: M ./rdft/problem2.c -8 +3 Chris@10: Chris@10: Thu Aug 29 17:31:39 EDT 2002 stevenj Chris@10: * [project @ 2002-08-29 21:31:39 by stevenj] Chris@10: bug fix in cldrest hc2c/c2hc copy loops Chris@10: Chris@10: M ./rdft/buffered2.c -9 +14 Chris@10: Chris@10: Thu Aug 29 13:45:08 EDT 2002 athena Chris@10: * [project @ 2002-08-29 17:45:08 by athena] Chris@10: Added things to do. Chris@10: Chris@10: M ./TODO -1 +1 Chris@10: Chris@10: Thu Aug 29 13:10:04 EDT 2002 stevenj Chris@10: * [project @ 2002-08-29 17:10:04 by stevenj] Chris@10: added automake prereq Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Thu Aug 29 08:36:36 EDT 2002 athena Chris@10: * [project @ 2002-08-29 12:36:36 by athena] Chris@10: Use indexed addressing Chris@10: Chris@10: M ./rdft/rdft2-radix2.c -61 +61 Chris@10: Chris@10: Thu Aug 29 08:20:55 EDT 2002 athena Chris@10: * [project @ 2002-08-29 12:20:55 by athena] Chris@10: Ooops Chris@10: Chris@10: M ./libbench/verify.c -3 +6 Chris@10: M ./rdft/rdft2-radix2.c -4 +7 Chris@10: Chris@10: Thu Aug 29 07:45:37 EDT 2002 athena Chris@10: * [project @ 2002-08-29 11:45:37 by athena] Chris@10: Oops Chris@10: Chris@10: M ./kernel/ifftw.h -10 +10 Chris@10: Chris@10: Thu Aug 29 02:32:13 EDT 2002 stevenj Chris@10: * [project @ 2002-08-29 06:32:13 by stevenj] Chris@10: updates to win32 threads code (ick) Chris@10: Chris@10: M ./threads/threads.c -4 +22 Chris@10: Chris@10: Thu Aug 29 01:44:33 EDT 2002 stevenj Chris@10: * [project @ 2002-08-29 05:44:33 by stevenj] Chris@10: added threaded version Chris@10: Chris@10: A ./acx_pthread.m4 Chris@10: A ./threads/ Chris@10: A ./threads/Makefile.am Chris@10: A ./threads/conf.c Chris@10: A ./threads/ct-dit.c Chris@10: A ./threads/dft-vrank-geq1.c Chris@10: A ./threads/hc2hc-dif.c Chris@10: A ./threads/hc2hc-dit.c Chris@10: A ./threads/rdft-vrank-geq1.c Chris@10: A ./threads/threads.c Chris@10: A ./threads/threads.h Chris@10: A ./threads/vrank-geq1-rdft2.c Chris@10: M ./Makefile.am -2 +3 Chris@10: M ./acx_pthread.m4 +226 Chris@10: M ./configure.ac +56 Chris@10: M ./dft/ct-dif.c -2 +2 Chris@10: M ./dft/ct-dit.c -2 +5 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/ct-ditf.c -2 +2 Chris@10: M ./dft/ct.c -2 +3 Chris@10: M ./dft/ct.h -1 +2 Chris@10: M ./dft/dft.h -1 +4 Chris@10: M ./dft/kdft-dif.c -1 +5 Chris@10: M ./dft/kdft-dit.c -1 +5 Chris@10: M ./dft/vrank-geq1.c -1 +4 Chris@10: M ./kernel/alloc.c -42 +56 Chris@10: M ./kernel/ifftw.h -7 +18 Chris@10: M ./kernel/planner.c -15 +24 Chris@10: M ./rdft/hc2hc-buf.c -1 +3 Chris@10: M ./rdft/hc2hc-dif.c -1 +5 Chris@10: M ./rdft/hc2hc-dit.c -1 +5 Chris@10: M ./rdft/hc2hc.c -2 +3 Chris@10: M ./rdft/hc2hc.h -1 +2 Chris@10: M ./rdft/khc2hc-dif.c -1 +5 Chris@10: M ./rdft/khc2hc-dit.c -1 +5 Chris@10: M ./rdft/rdft.h -1 +4 Chris@10: M ./rdft/vrank-geq1-rdft2.c -1 +4 Chris@10: M ./rdft/vrank-geq1.c -1 +4 Chris@10: M ./tests/Makefile.am -2 +4 Chris@10: M ./tests/bench.c -1 +5 Chris@10: M ./threads/Makefile.am +15 Chris@10: M ./threads/conf.c +41 Chris@10: M ./threads/ct-dit.c +151 Chris@10: M ./threads/dft-vrank-geq1.c +249 Chris@10: M ./threads/hc2hc-dif.c +168 Chris@10: M ./threads/hc2hc-dit.c +166 Chris@10: M ./threads/rdft-vrank-geq1.c +246 Chris@10: M ./threads/threads.c +467 Chris@10: M ./threads/threads.h +53 Chris@10: M ./threads/vrank-geq1-rdft2.c +267 Chris@10: Chris@10: Wed Aug 28 19:47:21 EDT 2002 stevenj Chris@10: * [project @ 2002-08-28 23:47:21 by stevenj] Chris@10: fix make dist Chris@10: Chris@10: M ./kernel/Makefile.am -2 +3 Chris@10: Chris@10: Wed Aug 28 15:09:03 EDT 2002 stevenj Chris@10: * [project @ 2002-08-28 19:09:03 by stevenj] Chris@10: whoops, bugfix for inverse Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -4 +9 Chris@10: Chris@10: Wed Aug 28 14:50:34 EDT 2002 athena Chris@10: * [project @ 2002-08-28 18:50:34 by athena] Chris@10: Use C9x convention for naming (fftwf etc.). Removed installable header Chris@10: files since they will be part of the API. Chris@10: Chris@10: M ./Makefile.am -5 +5 Chris@10: M ./configure.ac -2 +9 Chris@10: M ./kernel/Makefile.am -1 Chris@10: R ./kernel/dfftw3.h Chris@10: R ./kernel/fftw3.h Chris@10: M ./kernel/ifftw.h -2 +17 Chris@10: R ./kernel/lfftw3.h Chris@10: R ./kernel/sfftw3.h Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Tue Aug 27 23:34:00 EDT 2002 stevenj Chris@10: * [project @ 2002-08-28 03:34:00 by stevenj] Chris@10: allow _1 variants to accept rnk 0 (sz 1) problems Chris@10: Chris@10: M ./rdft/problem.c -3 +3 Chris@10: Chris@10: Tue Aug 27 15:56:09 EDT 2002 stevenj Chris@10: * [project @ 2002-08-27 19:56:09 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +619 Chris@10: Chris@10: Mon Aug 26 20:14:56 EDT 2002 athena Chris@10: * [project @ 2002-08-27 00:14:56 by athena] Chris@10: Loop unroll is useless Chris@10: Chris@10: M ./dft/rank0.c -28 +6 Chris@10: Chris@10: Mon Aug 26 20:00:41 EDT 2002 athena Chris@10: * [project @ 2002-08-27 00:00:41 by athena] Chris@10: Use indexed addressing Chris@10: Chris@10: M ./dft/ct-ditbuf.c -12 +9 Chris@10: Chris@10: Mon Aug 26 19:46:46 EDT 2002 athena Chris@10: * [project @ 2002-08-26 23:46:46 by athena] Chris@10: Use indexed addressing in transpose routines. (Seems to be Chris@10: slightly better on athlon.) Chris@10: Chris@10: M ./dft/vrank2-transpose.c -16 +8 Chris@10: M ./dft/vrank3-transpose.c -15 +8 Chris@10: Chris@10: Mon Aug 26 12:59:44 EDT 2002 stevenj Chris@10: * [project @ 2002-08-26 16:59:44 by stevenj] Chris@10: added comment about stability Chris@10: Chris@10: M ./reodft/redft00e-r2hc.c -1 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -1 +3 Chris@10: M ./reodft/rodft00e-r2hc.c -1 +2 Chris@10: Chris@10: Mon Aug 26 07:43:53 EDT 2002 athena Chris@10: * [project @ 2002-08-26 11:43:53 by athena] Chris@10: Approximate opcount Chris@10: Chris@10: M ./rdft/rdft2-radix2.c -6 +10 Chris@10: Chris@10: Mon Aug 26 06:38:49 EDT 2002 athena Chris@10: * [project @ 2002-08-26 10:38:49 by athena] Chris@10: Finished rdft2 via dft/rdft Chris@10: Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rank-geq2.c -2 +2 Chris@10: M ./rdft/rdft2-radix2.c -9 +149 Chris@10: Chris@10: Mon Aug 26 00:15:59 EDT 2002 stevenj Chris@10: * [project @ 2002-08-26 04:15:59 by stevenj] Chris@10: some updates Chris@10: Chris@10: M ./TODO -9 +9 Chris@10: Chris@10: Mon Aug 26 00:05:53 EDT 2002 stevenj Chris@10: * [project @ 2002-08-26 04:05:52 by stevenj] Chris@10: rdft kind is now per-dimension, added rdft/rank-geq2 Chris@10: Chris@10: A ./rdft/rank-geq2.c Chris@10: M ./rdft/Makefile.am -2 +3 Chris@10: M ./rdft/buffered.c -3 +4 Chris@10: M ./rdft/buffered2.c -5 +5 Chris@10: M ./rdft/conf.c -3 +2 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/direct.c -6 +6 Chris@10: M ./rdft/generic.c -5 +5 Chris@10: M ./rdft/hc2hc.c -10 +10 Chris@10: M ./rdft/indirect.c -2 +2 Chris@10: M ./rdft/problem.c -21 +75 Chris@10: M ./rdft/r2hc-hc2r.c -4 +4 Chris@10: M ./rdft/rader-dht.c -4 +4 Chris@10: M ./rdft/rader-hc2hc.c -4 +4 Chris@10: M ./rdft/rank-geq2.c +231 Chris@10: M ./rdft/rdft.h -4 +9 Chris@10: M ./rdft/rdft2-radix2.c -2 +2 Chris@10: M ./rdft/verify.c -5 +5 Chris@10: M ./reodft/redft00e-r2hc.c -3 +3 Chris@10: M ./reodft/reodft010e-r2hc.c -6 +6 Chris@10: M ./reodft/reodft11e-r2hc.c -5 +5 Chris@10: M ./reodft/rodft00e-r2hc.c -3 +3 Chris@10: M ./reodft/verify.c -4 +4 Chris@10: M ./tests/bench.c -2 +4 Chris@10: Chris@10: Sun Aug 25 22:45:38 EDT 2002 stevenj Chris@10: * [project @ 2002-08-26 02:45:38 by stevenj] Chris@10: added note Chris@10: Chris@10: M ./rdft/problem.c -2 +3 Chris@10: Chris@10: Sun Aug 25 22:28:12 EDT 2002 stevenj Chris@10: * [project @ 2002-08-26 02:28:12 by stevenj] Chris@10: must zero real sz Chris@10: Chris@10: M ./rdft/problem.c -2 +4 Chris@10: Chris@10: Sun Aug 25 22:06:52 EDT 2002 stevenj Chris@10: * [project @ 2002-08-26 02:06:52 by stevenj] Chris@10: unified pickdim funcs Chris@10: Chris@10: A ./kernel/pickdim.c Chris@10: M ./dft/rank-geq2.c -32 +10 Chris@10: M ./dft/vrank-geq1.c -45 +3 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -1 +6 Chris@10: M ./kernel/pickdim.c +82 Chris@10: M ./rdft/vrank-geq1-rdft2.c -45 +3 Chris@10: M ./rdft/vrank-geq1.c -45 +3 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Aug 25 14:10:55 EDT 2002 fftw Chris@10: * [project @ 2002-08-25 18:10:55 by fftw] Chris@10: silence warnings Chris@10: Chris@10: M ./libbench/mp.c -14 Chris@10: M ./rdft/codelet.h -2 +2 Chris@10: M ./rdft/indirect.c -3 +1 Chris@10: M ./rdft/rank-geq2-rdft2.c -5 +2 Chris@10: M ./rdft/verify.c -10 +7 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -7 +2 Chris@10: M ./reodft/verify.c -2 +2 Chris@10: Chris@10: Sun Aug 25 13:16:49 EDT 2002 athena Chris@10: * [project @ 2002-08-25 17:16:49 by athena] Chris@10: I had to add another planner flag to record whether pointers could Chris@10: become unaligned because of vrank-geq1 solvers (these solvers only Chris@10: plan the first element of a vector problem, but the second element Chris@10: may have a different alignment). This addition is ugly, but I don't Chris@10: see any way around it. Chris@10: Chris@10: M ./dft/codelet.h -3 +5 Chris@10: M ./dft/codelets/n.c -2 +3 Chris@10: M ./dft/codelets/t.c -2 +2 Chris@10: M ./dft/ct-dif.c -3 +3 Chris@10: M ./dft/ct-dit.c -3 +3 Chris@10: M ./dft/ct-ditbuf.c -3 +3 Chris@10: M ./dft/ct-ditf.c -2 +2 Chris@10: M ./dft/direct.c -5 +6 Chris@10: M ./dft/simd/n1b.c -1 +3 Chris@10: M ./dft/simd/n1f.c -1 +3 Chris@10: M ./dft/simd/t1b.c -1 +3 Chris@10: M ./dft/simd/t1f.c -1 +3 Chris@10: M ./dft/vrank-geq1.c -2 +11 Chris@10: M ./kernel/ifftw.h -4 +8 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +8 Chris@10: M ./rdft/vrank-geq1.c -2 +7 Chris@10: Chris@10: Sun Aug 25 10:18:25 EDT 2002 athena Chris@10: * [project @ 2002-08-25 14:18:25 by athena] Chris@10: Added thoughts Chris@10: Chris@10: M ./TODO -1 +2 Chris@10: Chris@10: Sun Aug 25 10:08:59 EDT 2002 athena Chris@10: * [project @ 2002-08-25 14:08:59 by athena] Chris@10: Implemented rdft2 via vector rdft + radix2 step Chris@10: Chris@10: A ./rdft/rdft2-radix2.c Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: R ./rdft/rdft2-dft.c Chris@10: M ./rdft/rdft2-radix2.c +325 Chris@10: Chris@10: Sat Aug 24 17:43:54 EDT 2002 athena Chris@10: * [project @ 2002-08-24 21:43:54 by athena] Chris@10: Stylistic changes Chris@10: Chris@10: M ./rdft/rdft2-dft.c -5 +5 Chris@10: Chris@10: Sat Aug 24 11:19:30 EDT 2002 athena Chris@10: * [project @ 2002-08-24 15:19:30 by athena] Chris@10: Simplified mktwiddle interface Chris@10: Chris@10: M ./dft/ct.c -7 +3 Chris@10: M ./dft/generic.c -5 +2 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/twiddle.c -1 +11 Chris@10: M ./rdft/generic.c -6 +3 Chris@10: M ./rdft/rdft2-dft.c -11 +5 Chris@10: M ./reodft/redft00e-r2hc.c -6 +2 Chris@10: M ./reodft/reodft010e-r2hc.c -5 +2 Chris@10: M ./reodft/reodft11e-r2hc.c -8 +3 Chris@10: M ./reodft/rodft00e-r2hc.c -5 +2 Chris@10: Chris@10: Sat Aug 24 11:05:08 EDT 2002 athena Chris@10: * [project @ 2002-08-24 15:05:08 by athena] Chris@10: Unification of certain vector computations. rdft2-dft is now a Chris@10: vector transform. Chris@10: Chris@10: M ./dft/ct-dif.c -2 +2 Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/ct.c -16 +2 Chris@10: M ./dft/ct.h -4 +1 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/tensor.c -1 +17 Chris@10: M ./rdft/hc2hc-dif.c -2 +2 Chris@10: M ./rdft/hc2hc-dit.c -2 +2 Chris@10: M ./rdft/hc2hc.c -18 +2 Chris@10: M ./rdft/hc2hc.h -4 +1 Chris@10: M ./rdft/rdft2-dft.c -54 +66 Chris@10: Chris@10: Fri Aug 23 20:21:25 EDT 2002 athena Chris@10: * [project @ 2002-08-24 00:21:25 by athena] Chris@10: Intel compiler seems to be still buggy Chris@10: Chris@10: M ./configure.ac +1 Chris@10: M ./simd/sse.c -1 +3 Chris@10: M ./simd/sse2.c -1 +3 Chris@10: Chris@10: Fri Aug 23 16:07:12 EDT 2002 athena Chris@10: * [project @ 2002-08-23 20:07:12 by athena] Chris@10: Streamlined twiddle protocol Chris@10: Chris@10: M ./dft/ct-dif.c -2 +3 Chris@10: M ./dft/ct-dit.c -2 +3 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/ct-ditf.c -2 +2 Chris@10: M ./dft/ct.c -13 +6 Chris@10: M ./dft/ct.h -3 +2 Chris@10: M ./dft/generic.c -16 +6 Chris@10: M ./dft/indirect.c -3 +1 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./kernel/twiddle.c -7 +14 Chris@10: M ./rdft/generic.c -18 +8 Chris@10: M ./rdft/hc2hc.c -10 +6 Chris@10: M ./rdft/rdft2-dft.c -19 +10 Chris@10: M ./reodft/redft00e-r2hc.c -15 +6 Chris@10: M ./reodft/reodft010e-r2hc.c -19 +10 Chris@10: M ./reodft/reodft11e-r2hc.c -21 +9 Chris@10: M ./reodft/rodft00e-r2hc.c -15 +6 Chris@10: Chris@10: Fri Aug 23 13:22:17 EDT 2002 athena Chris@10: * [project @ 2002-08-23 17:22:17 by athena] Chris@10: Implemented rdft2 via dft (forward only for now) Chris@10: Chris@10: A ./rdft/rdft2-dft.c Chris@10: M ./libbench/verify.c -1 +10 Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/conf.c -1 +2 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: M ./rdft/rdft2-dft.c +213 Chris@10: Chris@10: Thu Aug 22 11:29:29 EDT 2002 athena Chris@10: * [project @ 2002-08-22 15:29:29 by athena] Chris@10: More cleanup of verify Chris@10: Chris@10: M ./kernel/verify-lib.c -21 +13 Chris@10: M ./libbench/verify.c -45 +12 Chris@10: Chris@10: Thu Aug 22 11:16:03 EDT 2002 athena Chris@10: * [project @ 2002-08-22 15:16:03 by athena] Chris@10: Changed error criterion because old one was too strict Chris@10: Chris@10: M ./kernel/verify-lib.c -40 +21 Chris@10: Chris@10: Thu Aug 22 11:15:17 EDT 2002 athena Chris@10: * [project @ 2002-08-22 15:15:17 by athena] Chris@10: Disable shared Chris@10: Chris@10: M ./bootstrap.sh -1 +1 Chris@10: Chris@10: Thu Aug 22 09:19:12 EDT 2002 athena Chris@10: * [project @ 2002-08-22 13:19:12 by athena] Chris@10: Added thoughts Chris@10: Chris@10: M ./TODO +4 Chris@10: Chris@10: Thu Aug 22 09:17:28 EDT 2002 athena Chris@10: * [project @ 2002-08-22 13:17:28 by athena] Chris@10: Oops Chris@10: Chris@10: M ./dft/generic.c -1 +2 Chris@10: Chris@10: Thu Aug 22 09:11:34 EDT 2002 athena Chris@10: * [project @ 2002-08-22 13:11:34 by athena] Chris@10: Do not use inline. Minor changes. Chris@10: Chris@10: M ./dft/generic.c -1 +1 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/planner-score.c -2 +2 Chris@10: M ./kernel/tensor.c -3 +3 Chris@10: Chris@10: Wed Aug 21 16:23:26 EDT 2002 stevenj Chris@10: * [project @ 2002-08-21 20:23:26 by stevenj] Chris@10: more commented flags Chris@10: Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Tue Aug 20 19:44:43 EDT 2002 stevenj Chris@10: * [project @ 2002-08-20 23:44:43 by stevenj] Chris@10: added DCT-IV and DST-IV Chris@10: Chris@10: A ./reodft/reodft11e-r2hc.c Chris@10: M ./reodft/Makefile.am -1 +2 Chris@10: M ./reodft/conf.c -1 +2 Chris@10: M ./reodft/reodft11e-r2hc.c +297 Chris@10: M ./reodft/verify.c -1 +15 Chris@10: M ./tests/bench.c -1 +5 Chris@10: Chris@10: Tue Aug 20 16:01:36 EDT 2002 athena Chris@10: * [project @ 2002-08-20 20:01:36 by athena] Chris@10: Slight improvement in twiddle scheme Chris@10: Chris@10: M ./genfft/twiddle.ml -2 +9 Chris@10: Chris@10: Tue Aug 20 15:31:54 EDT 2002 stevenj Chris@10: * [project @ 2002-08-20 19:31:54 by stevenj] Chris@10: name fix Chris@10: Chris@10: M ./reodft/conf.c -2 +2 Chris@10: M ./reodft/reodft.h -6 +4 Chris@10: M ./reodft/reodft010e-r2hc.c -2 +2 Chris@10: Chris@10: Tue Aug 20 15:16:48 EDT 2002 stevenj Chris@10: * [project @ 2002-08-20 19:16:48 by stevenj] Chris@10: removed extraneous variable Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -5 +1 Chris@10: Chris@10: Tue Aug 20 11:46:29 EDT 2002 athena Chris@10: * [project @ 2002-08-20 15:46:29 by athena] Chris@10: Oops Chris@10: Chris@10: M ./libbench/mp.c -33 +79 Chris@10: M ./libbench/verify.c -33 +25 Chris@10: Chris@10: Tue Aug 20 08:37:45 EDT 2002 athena Chris@10: * [project @ 2002-08-20 12:37:45 by athena] Chris@10: Still playing around Chris@10: Chris@10: M ./genfft/twiddle.ml -35 +62 Chris@10: M ./kernel/trig.c -1 +3 Chris@10: Chris@10: Mon Aug 19 19:56:29 EDT 2002 athena Chris@10: * [project @ 2002-08-19 23:56:29 by athena] Chris@10: Playing around with addition chain Chris@10: Chris@10: M ./TODO -6 +11 Chris@10: M ./genfft/algsimp.ml -3 +4 Chris@10: M ./genfft/expr.ml -1 +6 Chris@10: M ./genfft/expr.mli -1 +2 Chris@10: M ./genfft/twiddle.ml -46 +64 Chris@10: M ./support/addchain.c -8 +25 Chris@10: Chris@10: Mon Aug 19 19:48:56 EDT 2002 stevenj Chris@10: * [project @ 2002-08-19 23:48:56 by stevenj] Chris@10: comments Chris@10: Chris@10: M ./reodft/redft00e-r2hc.c -1 +4 Chris@10: M ./reodft/rodft00e-r2hc.c -1 +4 Chris@10: Chris@10: Mon Aug 19 19:45:35 EDT 2002 stevenj Chris@10: * [project @ 2002-08-19 23:45:35 by stevenj] Chris@10: comment fixes Chris@10: Chris@10: M ./reodft/reodft010e-r2hc.c -5 +2 Chris@10: Chris@10: Mon Aug 19 19:40:18 EDT 2002 stevenj Chris@10: * [project @ 2002-08-19 23:40:18 by stevenj] Chris@10: added reodft stuff Chris@10: Chris@10: A ./reodft/ Chris@10: A ./reodft/Makefile.am Chris@10: A ./reodft/conf.c Chris@10: A ./reodft/redft00e-r2hc.c Chris@10: A ./reodft/reodft.h Chris@10: A ./reodft/reodft010e-r2hc.c Chris@10: A ./reodft/rodft00e-r2hc.c Chris@10: A ./reodft/verify.c Chris@10: M ./Makefile.am -2 +3 Chris@10: M ./configure.ac +2 Chris@10: M ./dft/dft.h -1 +5 Chris@10: M ./rdft/rdft.h -1 +5 Chris@10: M ./reodft/Makefile.am +6 Chris@10: M ./reodft/conf.c +37 Chris@10: M ./reodft/redft00e-r2hc.c +204 Chris@10: M ./reodft/reodft.h +44 Chris@10: M ./reodft/reodft010e-r2hc.c +400 Chris@10: M ./reodft/rodft00e-r2hc.c +199 Chris@10: M ./reodft/verify.c +521 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/bench.c -2 +11 Chris@10: Chris@10: Sun Aug 18 19:44:14 EDT 2002 athena Chris@10: * [project @ 2002-08-18 23:44:14 by athena] Chris@10: Sync with nbenchfft Chris@10: Chris@10: M ./libbench/Makefile.am +4 Chris@10: M ./libbench/verify.c -4 +3 Chris@10: Chris@10: Sun Aug 18 16:02:37 EDT 2002 athena Chris@10: * [project @ 2002-08-18 20:02:37 by athena] Chris@10: Economy of thought Chris@10: Chris@10: M ./genfft/complex.ml -23 +12 Chris@10: M ./genfft/complex.mli -3 +3 Chris@10: M ./genfft/twiddle.ml -5 +5 Chris@10: Chris@10: Sat Aug 17 15:52:05 EDT 2002 stevenj Chris@10: * [project @ 2002-08-17 19:52:05 by stevenj] Chris@10: distribute addchain.c Chris@10: Chris@10: M ./support/Makefile.am -1 +1 Chris@10: Chris@10: Sat Aug 17 14:09:11 EDT 2002 athena Chris@10: * [project @ 2002-08-17 18:09:11 by athena] Chris@10: Nothing serious Chris@10: Chris@10: M ./support/addchain.c -4 +7 Chris@10: Chris@10: Sat Aug 17 10:47:59 EDT 2002 athena Chris@10: * [project @ 2002-08-17 14:47:59 by athena] Chris@10: New twiddle policy (disabled for now) Chris@10: Chris@10: A ./support/addchain.c Chris@10: M ./genfft/twiddle.ml -1 +98 Chris@10: M ./support/addchain.c +151 Chris@10: Chris@10: Fri Aug 16 23:44:28 EDT 2002 stevenj Chris@10: * [project @ 2002-08-17 03:44:28 by stevenj] Chris@10: bug fix for hc2r (must use inverse dft) Chris@10: Chris@10: M ./rdft/rank-geq2-rdft2.c -2 +2 Chris@10: Chris@10: Fri Aug 16 20:27:10 EDT 2002 athena Chris@10: * [project @ 2002-08-17 00:27:10 by athena] Chris@10: New log3 twiddle policy Chris@10: Chris@10: M ./dft/codelets/inplace/Makefile.am -2 +2 Chris@10: M ./dft/codelets/standard/Makefile.am -1 +1 Chris@10: M ./genfft/twiddle.ml -9 +88 Chris@10: M ./rdft/codelets/hc2r/Makefile.am +6 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +1 Chris@10: Chris@10: Fri Aug 16 18:10:33 EDT 2002 athena Chris@10: * [project @ 2002-08-16 22:10:33 by athena] Chris@10: More verify cleanup Chris@10: Chris@10: M ./dft/verify.c -20 +2 Chris@10: M ./kernel/verify-lib.c -1 +19 Chris@10: M ./kernel/verify.h +3 Chris@10: M ./rdft/verify.c -23 +5 Chris@10: Chris@10: Fri Aug 16 16:31:19 EDT 2002 athena Chris@10: * [project @ 2002-08-16 20:31:19 by athena] Chris@10: Oops Chris@10: Chris@10: M ./rdft/verify.c -3 +3 Chris@10: Chris@10: Fri Aug 16 15:22:36 EDT 2002 athena Chris@10: * [project @ 2002-08-16 19:22:36 by athena] Chris@10: Economy of thought (and code) Chris@10: Chris@10: A ./kernel/verify-lib.c Chris@10: A ./kernel/verify.h Chris@10: M ./dft/verify.c -305 +12 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/verify-lib.c +394 Chris@10: M ./kernel/verify.h +80 Chris@10: M ./rdft/verify.c -364 +23 Chris@10: Chris@10: Fri Aug 16 14:05:45 EDT 2002 athena Chris@10: * [project @ 2002-08-16 18:05:45 by athena] Chris@10: Added comment Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Fri Aug 16 12:57:43 EDT 2002 athena Chris@10: * [project @ 2002-08-16 16:57:43 by athena] Chris@10: Cleaner rounding algorithm Chris@10: Chris@10: M ./libbench/mp.c -16 +27 Chris@10: Chris@10: Fri Aug 16 11:27:43 EDT 2002 athena Chris@10: * [project @ 2002-08-16 15:27:43 by athena] Chris@10: Can get away with shorter length in bluestein (I think). Chris@10: Chris@10: M ./libbench/mp.c -1 +1 Chris@10: Chris@10: Fri Aug 16 11:08:09 EDT 2002 athena Chris@10: * [project @ 2002-08-16 15:08:09 by athena] Chris@10: Portability improvements Chris@10: Chris@10: M ./libbench/mp.c -11 +4 Chris@10: Chris@10: Fri Aug 16 08:06:31 EDT 2002 athena Chris@10: * [project @ 2002-08-16 12:06:31 by athena] Chris@10: Optionally average accuracy test over many rounds Chris@10: Chris@10: M ./libbench/bench-main.c -2 +9 Chris@10: M ./libbench/bench.h -2 +2 Chris@10: M ./libbench/verify.c -29 +43 Chris@10: Chris@10: Fri Aug 16 07:50:24 EDT 2002 athena Chris@10: * [project @ 2002-08-16 11:50:24 by athena] Chris@10: More accurate formula for trig tables Chris@10: Chris@10: M ./dft/rader.c -3 +3 Chris@10: M ./rdft/rader-dht.c -3 +3 Chris@10: M ./rdft/rader-hc2hc.c -3 +3 Chris@10: Chris@10: Fri Aug 16 06:42:02 EDT 2002 athena Chris@10: * [project @ 2002-08-16 10:42:02 by athena] Chris@10: Implemented accuracy test for all integers Chris@10: Chris@10: M ./libbench/mp.c -6 +123 Chris@10: M ./libbench/verify.c -2 +1 Chris@10: Chris@10: Thu Aug 15 18:54:44 EDT 2002 athena Chris@10: * [project @ 2002-08-15 22:54:44 by athena] Chris@10: inv, neg: make static Chris@10: Chris@10: M ./libbench/mp.c -2 +2 Chris@10: Chris@10: Thu Aug 15 17:25:37 EDT 2002 athena Chris@10: * [project @ 2002-08-15 21:25:37 by athena] Chris@10: Verify was not complete for real transforms Chris@10: Chris@10: M ./libbench/verify.c -9 +34 Chris@10: Chris@10: Thu Aug 15 16:30:03 EDT 2002 athena Chris@10: * [project @ 2002-08-15 20:30:03 by athena] Chris@10: Oops Chris@10: Chris@10: M ./libbench/verify.c -3 +1 Chris@10: Chris@10: Thu Aug 15 16:29:16 EDT 2002 athena Chris@10: * [project @ 2002-08-15 20:29:16 by athena] Chris@10: Fixed hb codelets Chris@10: Chris@10: M ./genfft/gen_hc2hc.ml -3 +5 Chris@10: M ./libbench/verify.c -1 +4 Chris@10: Chris@10: Thu Aug 15 14:10:45 EDT 2002 athena Chris@10: * [project @ 2002-08-15 18:10:45 by athena] Chris@10: Changed twiddle policy Chris@10: Chris@10: M ./dft/codelets/inplace/Makefile.am -2 +2 Chris@10: M ./dft/codelets/standard/Makefile.am -1 +1 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +1 Chris@10: Chris@10: Thu Aug 15 13:32:24 EDT 2002 stevenj Chris@10: * [project @ 2002-08-15 17:32:24 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/direct2.c -3 +3 Chris@10: Chris@10: Thu Aug 15 11:01:04 EDT 2002 athena Chris@10: * [project @ 2002-08-15 15:01:04 by athena] Chris@10: No point in libbench being a shared library Chris@10: Chris@10: M ./libbench/Makefile.am -2 +2 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Thu Aug 15 09:48:37 EDT 2002 athena Chris@10: * [project @ 2002-08-15 13:48:37 by athena] Chris@10: Moved accuracy test to libbench Chris@10: Chris@10: A ./libbench/mp.c Chris@10: M ./libbench/Makefile.am -1 +1 Chris@10: M ./libbench/bench-main.c -4 +13 Chris@10: M ./libbench/bench.h -1 +3 Chris@10: M ./libbench/mp.c +439 Chris@10: M ./libbench/util.c +8 Chris@10: M ./libbench/verify.c -3 +66 Chris@10: M ./tests/Makefile.am -5 +1 Chris@10: R ./tests/accuracy.c Chris@10: R ./tests/mp.c Chris@10: Chris@10: Wed Aug 14 19:48:23 EDT 2002 athena Chris@10: * [project @ 2002-08-14 23:48:23 by athena] Chris@10: Modified accuracy test Chris@10: Chris@10: M ./tests/accuracy.c -33 +20 Chris@10: Chris@10: Wed Aug 14 08:34:26 EDT 2002 athena Chris@10: * [project @ 2002-08-14 12:34:26 by athena] Chris@10: Fixes for long double Chris@10: Chris@10: M ./tests/accuracy.c -2 +3 Chris@10: M ./tests/mp.c -1 +1 Chris@10: Chris@10: Wed Aug 14 08:17:57 EDT 2002 athena Chris@10: * [project @ 2002-08-14 12:17:57 by athena] Chris@10: Normalize input Chris@10: Chris@10: M ./tests/accuracy.c -3 +17 Chris@10: Chris@10: Wed Aug 14 07:26:41 EDT 2002 athena Chris@10: * [project @ 2002-08-14 11:26:41 by athena] Chris@10: Oops Chris@10: Chris@10: M ./tests/accuracy.c +1 Chris@10: Chris@10: Wed Aug 14 07:25:34 EDT 2002 athena Chris@10: * [project @ 2002-08-14 11:25:34 by athena] Chris@10: Also compute relative error Chris@10: Chris@10: M ./tests/accuracy.c -2 +13 Chris@10: Chris@10: Wed Aug 14 07:08:20 EDT 2002 athena Chris@10: * [project @ 2002-08-14 11:08:20 by athena] Chris@10: Loop over N Chris@10: Chris@10: M ./tests/accuracy.c -24 +32 Chris@10: Chris@10: Wed Aug 14 06:54:50 EDT 2002 athena Chris@10: * [project @ 2002-08-14 10:54:50 by athena] Chris@10: simple-minded accuracy test Chris@10: Chris@10: A ./tests/accuracy.c Chris@10: A ./tests/mp.c Chris@10: M ./tests/Makefile.am -1 +6 Chris@10: M ./tests/accuracy.c +48 Chris@10: M ./tests/mp.c +434 Chris@10: Chris@10: Wed Aug 14 03:26:06 EDT 2002 stevenj Chris@10: * [project @ 2002-08-14 07:26:06 by stevenj] Chris@10: whoops Chris@10: Chris@10: A ./rdft/rank-geq2-rdft2.c Chris@10: Chris@10: Tue Aug 13 11:42:41 EDT 2002 athena Chris@10: * [project @ 2002-08-13 15:42:41 by athena] Chris@10: fma() stuff is too nonportable, removed Chris@10: Chris@10: M ./kernel/trig.c -45 +1 Chris@10: Chris@10: Mon Aug 12 14:07:44 EDT 2002 stevenj Chris@10: * [project @ 2002-08-12 18:07:44 by stevenj] Chris@10: slight fix Chris@10: Chris@10: M ./rdft/problem.c -2 +2 Chris@10: Chris@10: Mon Aug 12 14:07:18 EDT 2002 stevenj Chris@10: * [project @ 2002-08-12 18:07:18 by stevenj] Chris@10: use table for rdft_kind_str Chris@10: Chris@10: M ./rdft/problem.c -26 +12 Chris@10: Chris@10: Mon Aug 12 13:43:08 EDT 2002 stevenj Chris@10: * [project @ 2002-08-12 17:43:08 by stevenj] Chris@10: slight fixes Chris@10: Chris@10: M ./rdft/problem2.c -6 +11 Chris@10: Chris@10: Mon Aug 12 13:31:37 EDT 2002 stevenj Chris@10: * [project @ 2002-08-12 17:31:37 by stevenj] Chris@10: multidimensional rdft2 Chris@10: Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./kernel/planner.c -1 +2 Chris@10: M ./kernel/tensor.c -5 +6 Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/buffered2.c -13 +18 Chris@10: M ./rdft/conf.c -1 +2 Chris@10: M ./rdft/direct2.c -9 +14 Chris@10: M ./rdft/nop2.c -2 +12 Chris@10: M ./rdft/problem2.c -36 +41 Chris@10: M ./rdft/rdft.h -7 +9 Chris@10: M ./rdft/vrank-geq1-rdft2.c -6 +23 Chris@10: M ./tests/bench.c -11 +36 Chris@10: Chris@10: Sat Aug 10 19:33:23 EDT 2002 stevenj Chris@10: * [project @ 2002-08-10 23:33:23 by stevenj] Chris@10: use tensor_copy_inplace Chris@10: Chris@10: M ./rdft/indirect.c -13 +5 Chris@10: Chris@10: Sat Aug 10 19:32:03 EDT 2002 stevenj Chris@10: * [project @ 2002-08-10 23:32:03 by stevenj] Chris@10: bugfix, use tensor_copy_inplace Chris@10: Chris@10: M ./dft/rank-geq2.c -4 +10 Chris@10: Chris@10: Sat Aug 10 19:30:39 EDT 2002 stevenj Chris@10: * [project @ 2002-08-10 23:30:39 by stevenj] Chris@10: use tensor_copy_inplace Chris@10: Chris@10: M ./dft/indirect.c -13 +5 Chris@10: Chris@10: Sat Aug 10 19:28:07 EDT 2002 stevenj Chris@10: * [project @ 2002-08-10 23:28:07 by stevenj] Chris@10: added tensor_copy_inplace Chris@10: Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/tensor.c -1 +19 Chris@10: Chris@10: Sat Aug 10 19:25:50 EDT 2002 stevenj Chris@10: * [project @ 2002-08-10 23:25:50 by stevenj] Chris@10: fixed trig-function table type Chris@10: Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: Chris@10: Sat Aug 10 14:41:04 EDT 2002 athena Chris@10: * [project @ 2002-08-10 18:41:04 by athena] Chris@10: Improved trig scheme Chris@10: Chris@10: M ./kernel/trig.c -8 +53 Chris@10: M ./tests/trigtest.c -8 +74 Chris@10: Chris@10: Fri Aug 9 21:05:01 EDT 2002 athena Chris@10: * [project @ 2002-08-10 01:05:01 by athena] Chris@10: Allow for testing using long double instead of pari Chris@10: Chris@10: M ./tests/trigtest.c -15 +27 Chris@10: Chris@10: Fri Aug 9 20:49:32 EDT 2002 athena Chris@10: * [project @ 2002-08-10 00:49:32 by athena] Chris@10: Yet another trig scheme. Chris@10: Chris@10: M ./kernel/trig.c -25 +24 Chris@10: M ./tests/trigtest.c -25 +22 Chris@10: Chris@10: Fri Aug 9 20:38:07 EDT 2002 athena Chris@10: * [project @ 2002-08-10 00:38:07 by athena] Chris@10: Yet another scheme Chris@10: Chris@10: M ./kernel/trig.c -4 +12 Chris@10: M ./tests/trigtest.c +7 Chris@10: Chris@10: Fri Aug 9 20:31:16 EDT 2002 athena Chris@10: * [project @ 2002-08-10 00:31:16 by athena] Chris@10: Careful with overflow Chris@10: Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: M ./kernel/trig.c -23 +23 Chris@10: M ./tests/trigtest.c -24 +17 Chris@10: Chris@10: Fri Aug 9 20:16:23 EDT 2002 athena Chris@10: * [project @ 2002-08-10 00:16:23 by athena] Chris@10: Avoid overflow Chris@10: Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: M ./kernel/trig.c -20 +24 Chris@10: M ./tests/trigtest.c -21 +43 Chris@10: Chris@10: Fri Aug 9 19:26:57 EDT 2002 athena Chris@10: * [project @ 2002-08-09 23:26:57 by athena] Chris@10: New(er) trig routines Chris@10: Chris@10: M ./dft/rader.c -8 +6 Chris@10: M ./dft/verify.c -4 +3 Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: M ./kernel/trig.c -22 +17 Chris@10: M ./kernel/twiddle.c -9 +7 Chris@10: M ./rdft/rader-dht.c -4 +3 Chris@10: M ./rdft/rader-hc2hc.c -8 +6 Chris@10: M ./rdft/verify.c -4 +3 Chris@10: M ./tests/trigtest.c -20 +19 Chris@10: Chris@10: Fri Aug 9 19:25:44 EDT 2002 athena Chris@10: * [project @ 2002-08-09 23:25:44 by athena] Chris@10: Oops Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Fri Aug 9 18:49:04 EDT 2002 athena Chris@10: * [project @ 2002-08-09 22:49:04 by athena] Chris@10: New file Chris@10: Chris@10: A ./tests/trigtest.c Chris@10: Chris@10: Fri Aug 9 13:04:00 EDT 2002 athena Chris@10: * [project @ 2002-08-09 17:04:00 by athena] Chris@10: Commented about likely gcc bug Chris@10: Chris@10: M ./TODO +4 Chris@10: Chris@10: Fri Aug 9 13:01:49 EDT 2002 athena Chris@10: * [project @ 2002-08-09 17:01:49 by athena] Chris@10: Improved accuracy of twiddle factors Chris@10: Chris@10: A ./kernel/trig.c Chris@10: M ./dft/rader.c -8 +8 Chris@10: M ./dft/verify.c -5 +4 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -10 +7 Chris@10: M ./kernel/trig.c +79 Chris@10: M ./kernel/twiddle.c -8 +10 Chris@10: M ./rdft/rader-dht.c -4 +4 Chris@10: M ./rdft/rader-hc2hc.c -8 +8 Chris@10: M ./rdft/verify.c -5 +4 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Thu Aug 8 06:36:23 EDT 2002 athena Chris@10: * [project @ 2002-08-08 10:36:23 by athena] Chris@10: Wrong comment Chris@10: Chris@10: M ./simd/simd-3dnow.h -1 Chris@10: Chris@10: Wed Aug 7 17:14:09 EDT 2002 athena Chris@10: * [project @ 2002-08-07 21:14:09 by athena] Chris@10: Experimental 3dnow port using gcc, to compare it with Stefan's stuff. Chris@10: Chris@10: A ./simd/3dnow.c Chris@10: A ./simd/simd-3dnow.h Chris@10: M ./configure.ac +6 Chris@10: M ./genfft/gen_notw_c.ml -3 +6 Chris@10: M ./genfft/gen_twiddle_c.ml -2 +4 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./simd/3dnow.c +66 Chris@10: M ./simd/Makefile.am -2 +2 Chris@10: M ./simd/simd-3dnow.h +164 Chris@10: M ./simd/simd-altivec.h +2 Chris@10: M ./simd/simd-sse.h +3 Chris@10: M ./simd/simd-sse2.h +3 Chris@10: M ./simd/simd.h +4 Chris@10: Chris@10: Wed Aug 7 12:58:10 EDT 2002 athena Chris@10: * [project @ 2002-08-07 16:58:10 by athena] Chris@10: End of AREF experiment Chris@10: Chris@10: M ./genfft/c.ml -2 +2 Chris@10: M ./kernel/ifftw.h -3 +1 Chris@10: Chris@10: Wed Aug 7 07:47:19 EDT 2002 athena Chris@10: * [project @ 2002-08-07 11:47:19 by athena] Chris@10: Oops Chris@10: Chris@10: M ./configure.ac -4 +1 Chris@10: Chris@10: Wed Aug 7 07:46:38 EDT 2002 athena Chris@10: * [project @ 2002-08-07 11:46:38 by athena] Chris@10: Pathetic attempt to reduce size of configure script Chris@10: Chris@10: M ./configure.ac -16 +8 Chris@10: Chris@10: Tue Aug 6 20:38:11 EDT 2002 athena Chris@10: * [project @ 2002-08-07 00:38:11 by athena] Chris@10: Changed array syntax for experiments. Chris@10: Chris@10: M ./genfft/c.ml -8 +10 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: Chris@10: Tue Aug 6 19:58:20 EDT 2002 athena Chris@10: * [project @ 2002-08-06 23:58:20 by athena] Chris@10: Fix warning Chris@10: Chris@10: M ./simd/simd-sse2.h +2 Chris@10: Chris@10: Tue Aug 6 13:35:28 EDT 2002 athena Chris@10: * [project @ 2002-08-06 17:35:28 by athena] Chris@10: Move nonportable stuff in one place. Chris@10: Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./kernel/align.c -1 +8 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: Chris@10: Tue Aug 6 10:32:53 EDT 2002 athena Chris@10: * [project @ 2002-08-06 14:32:53 by athena] Chris@10: Economy of thought: I didn't like having two algorithms for removing Chris@10: solutions, both correct. At least now we have the same algorithm Chris@10: copied twice. Chris@10: Chris@10: M ./kernel/planner.c -12 +22 Chris@10: Chris@10: Tue Aug 6 09:12:21 EDT 2002 athena Chris@10: * [project @ 2002-08-06 13:12:21 by athena] Chris@10: Added things to do Chris@10: Chris@10: M ./TODO -1 +1 Chris@10: Chris@10: Mon Aug 5 19:54:31 EDT 2002 stevenj Chris@10: * [project @ 2002-08-05 23:54:31 by stevenj] Chris@10: improved interaction of planner with patience flags Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner.c -12 +34 Chris@10: Chris@10: Mon Aug 5 14:17:58 EDT 2002 stevenj Chris@10: * [project @ 2002-08-05 18:17:58 by stevenj] Chris@10: set up for real-even/odd DFTs, where n is not the size of the data Chris@10: Chris@10: M ./rdft/buffered.c -3 +3 Chris@10: M ./rdft/codelet.h -7 +25 Chris@10: M ./rdft/indirect.c -4 +8 Chris@10: M ./rdft/problem.c -13 +58 Chris@10: M ./rdft/rader-hc2hc.c -1 +1 Chris@10: M ./rdft/rdft.h -1 +3 Chris@10: M ./tests/bench.c -2 +3 Chris@10: Chris@10: Sun Aug 4 23:57:51 EDT 2002 stevenj Chris@10: * [project @ 2002-08-05 03:57:51 by stevenj] Chris@10: DESTROY_INPUT flag Chris@10: Chris@10: M ./dft/ct-dif.c -4 +5 Chris@10: M ./dft/ct-dit.c -3 +5 Chris@10: M ./dft/ct-ditbuf.c -3 +5 Chris@10: M ./dft/ct-ditf.c -3 +5 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/ct.h -2 +3 Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./rdft/hc2hc-buf.c -3 +6 Chris@10: M ./rdft/hc2hc-dif.c -3 +5 Chris@10: M ./rdft/hc2hc-dit.c -3 +5 Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: M ./rdft/hc2hc.h -2 +3 Chris@10: M ./rdft/r2hc-hc2r.c -13 +20 Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Sun Aug 4 22:50:19 EDT 2002 stevenj Chris@10: * [project @ 2002-08-05 02:50:19 by stevenj] Chris@10: CLASSIC -> IMPATIENT Chris@10: Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: M ./kernel/ifftw.h -4 +4 Chris@10: M ./kernel/planner.c -3 +3 Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: M ./rdft/vrank-geq1-rdft2.c -2 +2 Chris@10: M ./rdft/vrank-geq1.c -2 +2 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Aug 4 19:05:43 EDT 2002 athena Chris@10: * [project @ 2002-08-04 23:05:43 by athena] Chris@10: Require make maintainer-clean to remove the generator, as opposed Chris@10: to make clean. In this way we can type make clean without regenerating Chris@10: all codelets. Chris@10: Chris@10: M ./genfft/Makefile.am -11 +11 Chris@10: M ./genfft-k7/Makefile.am -3 +3 Chris@10: Chris@10: Sun Aug 4 17:34:04 EDT 2002 stevenj Chris@10: * [project @ 2002-08-04 21:34:04 by stevenj] Chris@10: ESTIMATE plans are not blessed Chris@10: Chris@10: M ./kernel/planner.c -10 +13 Chris@10: Chris@10: Sun Aug 4 17:24:37 EDT 2002 stevenj Chris@10: * [project @ 2002-08-04 21:24:37 by stevenj] Chris@10: use flags in wisdom Chris@10: Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/planner.c -9 +14 Chris@10: Chris@10: Sun Aug 4 17:03:45 EDT 2002 stevenj Chris@10: * [project @ 2002-08-04 21:03:45 by stevenj] Chris@10: score now takes plnr, not flags, as arg Chris@10: Chris@10: M ./dft/buffered.c -3 +3 Chris@10: M ./dft/ct-dif.c -4 +3 Chris@10: M ./dft/ct-dit.c -3 +3 Chris@10: M ./dft/ct-ditbuf.c -3 +3 Chris@10: M ./dft/ct-ditf.c -3 +3 Chris@10: M ./dft/direct.c -3 +3 Chris@10: M ./dft/generic.c -2 +2 Chris@10: M ./dft/indirect.c -3 +3 Chris@10: M ./dft/nop.c -3 +3 Chris@10: M ./dft/rader.c -4 +4 Chris@10: M ./dft/rank-geq2.c -3 +3 Chris@10: M ./dft/rank0.c -3 +3 Chris@10: M ./dft/vrank-geq1.c -4 +4 Chris@10: M ./dft/vrank2-transpose.c -3 +3 Chris@10: M ./dft/vrank3-transpose.c -3 +3 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner-score.c -4 +3 Chris@10: M ./kernel/tensor.c -3 +3 Chris@10: M ./rdft/buffered.c -3 +3 Chris@10: M ./rdft/buffered2.c -3 +3 Chris@10: M ./rdft/dft-r2hc.c -3 +3 Chris@10: M ./rdft/direct.c -3 +3 Chris@10: M ./rdft/direct2.c -3 +3 Chris@10: M ./rdft/generic.c -2 +2 Chris@10: M ./rdft/hc2hc-buf.c -3 +3 Chris@10: M ./rdft/hc2hc-dif.c -3 +3 Chris@10: M ./rdft/hc2hc-dit.c -3 +3 Chris@10: M ./rdft/indirect.c -3 +3 Chris@10: M ./rdft/nop.c -3 +3 Chris@10: M ./rdft/nop2.c -3 +3 Chris@10: M ./rdft/r2hc-hc2r.c -3 +3 Chris@10: M ./rdft/rader-dht.c -2 +2 Chris@10: M ./rdft/rader-hc2hc.c -2 +2 Chris@10: M ./rdft/rank0.c -3 +3 Chris@10: M ./rdft/vrank-geq1-rdft2.c -4 +4 Chris@10: M ./rdft/vrank-geq1.c -4 +4 Chris@10: M ./rdft/vrank2-transpose.c -3 +3 Chris@10: M ./rdft/vrank3-transpose.c -3 +3 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Aug 4 16:37:46 EDT 2002 stevenj Chris@10: * [project @ 2002-08-04 20:37:46 by stevenj] Chris@10: align initial stack in alignment check, which should now pass for gcc 3.1.1 Chris@10: Chris@10: M ./acinclude.m4 -2 +3 Chris@10: Chris@10: Sat Aug 3 20:04:57 EDT 2002 athena Chris@10: * [project @ 2002-08-04 00:04:57 by athena] Chris@10: Detect ultrasparc (sort of) Chris@10: Chris@10: M ./acinclude.m4 +3 Chris@10: Chris@10: Sat Aug 3 19:38:17 EDT 2002 stevenj Chris@10: * [project @ 2002-08-03 23:38:17 by stevenj] Chris@10: added solvtab_rdft_r2r placeholder Chris@10: Chris@10: M ./rdft/codelet.h -1 +3 Chris@10: Chris@10: Sat Aug 3 19:34:49 EDT 2002 athena Chris@10: * [project @ 2002-08-03 23:34:49 by athena] Chris@10: Damn solaris Chris@10: Chris@10: M ./support/Makefile.codelets -2 +2 Chris@10: Chris@10: Sat Aug 3 17:55:44 EDT 2002 stevenj Chris@10: * [project @ 2002-08-03 21:55:44 by stevenj] Chris@10: use E extended precision in solvers Chris@10: Chris@10: M ./rdft/problem.c -7 +9 Chris@10: Chris@10: Sat Aug 3 17:53:29 EDT 2002 stevenj Chris@10: * [project @ 2002-08-03 21:53:29 by stevenj] Chris@10: an alternative notation for D{C,S}T: DXTio, where i/o are {0,1} Chris@10: according to whether the input/output are shifted, respectively. Chris@10: Alternatively, io is the binary representation of the usual Chris@10: DXT-{I,II,III,IV} nomenclature, minus 1. Chris@10: Chris@10: M ./rdft/codelet.h -2 +2 Chris@10: Chris@10: Sat Aug 3 17:49:11 EDT 2002 stevenj Chris@10: * [project @ 2002-08-03 21:49:11 by stevenj] Chris@10: use E extended precision in solvers Chris@10: Chris@10: M ./dft/generic.c -5 +5 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./rdft/generic.c -21 +21 Chris@10: M ./rdft/r2hc-hc2r.c -3 +3 Chris@10: M ./rdft/rader-dht.c -4 +4 Chris@10: Chris@10: Sat Aug 3 15:39:49 EDT 2002 athena Chris@10: * [project @ 2002-08-03 19:39:49 by athena] Chris@10: More portability fixes, compiler bugs workarounds, etc. Chris@10: Chris@10: M ./configure.ac -2 +4 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +4 Chris@10: Chris@10: Sat Aug 3 15:09:56 EDT 2002 athena Chris@10: * [project @ 2002-08-03 19:09:56 by athena] Chris@10: More portability work Chris@10: Chris@10: M ./configure.ac -2 +7 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./kernel/ifftw.h -9 +2 Chris@10: Chris@10: Sat Aug 3 14:33:40 EDT 2002 athena Chris@10: * [project @ 2002-08-03 18:33:40 by athena] Chris@10: Improved portability, removed gnu make dependencies Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: M ./configure.ac -4 +3 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./support/Makefile.codelets -14 +6 Chris@10: Chris@10: Sat Aug 3 13:48:53 EDT 2002 athena Chris@10: * [project @ 2002-08-03 17:48:53 by athena] Chris@10: Remember to thank XXX Chris@10: Chris@10: M ./TODO +8 Chris@10: Chris@10: Fri Aug 2 17:38:18 EDT 2002 athena Chris@10: * [project @ 2002-08-02 21:38:18 by athena] Chris@10: Multiplication on altivec requires FMA with -0.0 to be IEEE754 compliant. Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Fri Aug 2 15:26:37 EDT 2002 athena Chris@10: * [project @ 2002-08-02 19:26:37 by athena] Chris@10: Allow for extended precision in codelets Chris@10: Chris@10: M ./genfft/c.ml -3 +4 Chris@10: M ./kernel/ifftw.h -7 +9 Chris@10: Chris@10: Fri Aug 2 08:52:04 EDT 2002 athena Chris@10: * [project @ 2002-08-02 12:52:04 by athena] Chris@10: Shortened names Chris@10: Chris@10: M ./dft/codelets/inplace/Makefile.am -5 +5 Chris@10: Chris@10: Fri Aug 2 03:49:09 EDT 2002 stevenj Chris@10: * [project @ 2002-08-02 07:49:09 by stevenj] Chris@10: added infrastructure for future r2r transforms Chris@10: Chris@10: M ./TODO -3 +1 Chris@10: M ./rdft/codelet.h -3 +33 Chris@10: M ./rdft/problem.c -1 +8 Chris@10: Chris@10: Thu Aug 1 21:29:14 EDT 2002 athena Chris@10: * [project @ 2002-08-02 01:29:05 by athena] Chris@10: Version info Chris@10: Chris@10: M ./Makefile.am +1 Chris@10: M ./configure.ac +2 Chris@10: Chris@10: Thu Aug 1 21:06:22 EDT 2002 athena Chris@10: * [project @ 2002-08-02 01:06:22 by athena] Chris@10: Listened to one customer and added radix-12. Added radix-15 for Chris@10: consistency (whatever that is) Chris@10: Chris@10: M ./dft/codelets/inplace/Makefile.am -1 +1 Chris@10: M ./dft/codelets/standard/Makefile.am -1 +1 Chris@10: M ./dft/k7/codelets/Makefile.am -3 +4 Chris@10: M ./dft/simd/codelets/Makefile.am -2 +2 Chris@10: M ./kernel/align.c -2 +8 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -3 +3 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -3 +3 Chris@10: Chris@10: Thu Aug 1 19:50:53 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 23:50:53 by stevenj] Chris@10: whoops again, fixed the wrong line Chris@10: Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: Chris@10: Thu Aug 1 19:50:16 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 23:50:16 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Thu Aug 1 16:01:15 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 20:01:15 by stevenj] Chris@10: use new AC_INIT and add VERSION to wisdom Chris@10: Chris@10: M ./configure.ac -3 +4 Chris@10: M ./kernel/planner.c -4 +6 Chris@10: Chris@10: Thu Aug 1 14:56:45 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 18:56:45 by stevenj] Chris@10: mygetR -> getR Chris@10: Chris@10: M ./kernel/scan.c -3 +3 Chris@10: Chris@10: Thu Aug 1 14:56:02 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 18:56:02 by stevenj] Chris@10: scanner cleanups: just return 0/1, simplify integer reads Chris@10: Chris@10: M ./dft/problem.c -6 +4 Chris@10: M ./kernel/planner.c -14 +12 Chris@10: M ./kernel/scan.c -78 +42 Chris@10: M ./kernel/tensor.c -9 +8 Chris@10: M ./rdft/problem.c -7 +5 Chris@10: M ./rdft/problem2.c -7 +5 Chris@10: M ./tests/bench.c +7 Chris@10: Chris@10: Thu Aug 1 08:04:01 EDT 2002 athena Chris@10: * [project @ 2002-08-01 12:04:01 by athena] Chris@10: Reverted back to casting pointer to ulong Chris@10: Chris@10: M ./kernel/align.c -2 +2 Chris@10: Chris@10: Thu Aug 1 08:03:46 EDT 2002 athena Chris@10: * [project @ 2002-08-01 12:03:46 by athena] Chris@10: Cast to unsigned long, not long Chris@10: Chris@10: M ./kernel/ifftw.h -6 +6 Chris@10: Chris@10: Thu Aug 1 03:14:50 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 07:14:50 by stevenj] Chris@10: additional comment Chris@10: Chris@10: M ./kernel/scan.c -2 +4 Chris@10: Chris@10: Thu Aug 1 03:12:37 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 07:12:37 by stevenj] Chris@10: added comment Chris@10: Chris@10: M ./kernel/scan.c -1 +6 Chris@10: Chris@10: Thu Aug 1 03:03:18 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 07:03:18 by stevenj] Chris@10: added wisdom import Chris@10: Chris@10: A ./kernel/printers.c Chris@10: A ./kernel/scan.c Chris@10: A ./kernel/scanners.c Chris@10: M ./dft/conf.c -1 +3 Chris@10: M ./dft/dft.h -1 +3 Chris@10: M ./dft/problem.c -2 +31 Chris@10: M ./dft/verify.c -2 +1 Chris@10: M ./kernel/Makefile.am -3 +4 Chris@10: M ./kernel/alloc.c -2 +1 Chris@10: M ./kernel/assert.c -2 +1 Chris@10: M ./kernel/debug.c -9 +2 Chris@10: M ./kernel/ifftw.h -6 +53 Chris@10: M ./kernel/planner.c -22 +106 Chris@10: M ./kernel/print.c -2 +1 Chris@10: M ./kernel/printers.c +80 Chris@10: M ./kernel/problem.c -1 +2 Chris@10: M ./kernel/scan.c +322 Chris@10: M ./kernel/scanners.c +61 Chris@10: M ./kernel/tensor.c -4 +30 Chris@10: M ./kernel/timer.c -2 +1 Chris@10: M ./rdft/conf.c -1 +4 Chris@10: M ./rdft/problem.c -2 +32 Chris@10: M ./rdft/problem2.c -2 +32 Chris@10: M ./rdft/rdft.h -1 +4 Chris@10: M ./rdft/verify.c -2 +1 Chris@10: M ./tests/bench.c -9 +31 Chris@10: Chris@10: Wed Jul 31 23:12:05 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 03:12:05 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/align.c -1 +2 Chris@10: Chris@10: Wed Jul 31 22:06:46 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 02:06:46 by stevenj] Chris@10: use %u for alignment_of Chris@10: Chris@10: M ./dft/problem.c -3 +3 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: Chris@10: Wed Jul 31 21:47:15 EDT 2002 stevenj Chris@10: * [project @ 2002-08-01 01:47:15 by stevenj] Chris@10: ptrdiff_t form Chris@10: Chris@10: M ./kernel/align.c -2 +2 Chris@10: Chris@10: Wed Jul 31 21:33:35 EDT 2002 athena Chris@10: * [project @ 2002-08-01 01:33:35 by athena] Chris@10: Cast to avoid warning from C++ compiler Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Wed Jul 31 18:57:04 EDT 2002 athena Chris@10: * [project @ 2002-07-31 22:57:04 by athena] Chris@10: Make problem equality depend on alignments. Chris@10: Chris@10: A ./kernel/align.c Chris@10: M ./dft/problem.c -4 +10 Chris@10: M ./kernel/Makefile.am -5 +6 Chris@10: M ./kernel/align.c +35 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./rdft/problem.c -4 +9 Chris@10: M ./rdft/problem2.c -8 +13 Chris@10: M ./simd/simd.h +2 Chris@10: Chris@10: Wed Jul 31 15:45:31 EDT 2002 athena Chris@10: * [project @ 2002-07-31 19:45:31 by athena] Chris@10: Shorter names Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -22 +18 Chris@10: Chris@10: Wed Jul 31 14:38:00 EDT 2002 athena Chris@10: * [project @ 2002-07-31 18:38:00 by athena] Chris@10: Oops Chris@10: Chris@10: M ./simd/simd-sse.h -1 +1 Chris@10: Chris@10: Wed Jul 31 14:37:19 EDT 2002 athena Chris@10: * [project @ 2002-07-31 18:37:19 by athena] Chris@10: Fix warning Chris@10: Chris@10: M ./simd/simd-sse.h +2 Chris@10: Chris@10: Wed Jul 31 07:52:53 EDT 2002 athena Chris@10: * [project @ 2002-07-31 11:52:53 by athena] Chris@10: Removed silly abstraction barrier. Also, cons() terminology was Chris@10: no longer appropriate. Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/ifftw.h -16 +19 Chris@10: M ./kernel/planner-naive.c -2 +2 Chris@10: M ./kernel/planner-score.c -2 +2 Chris@10: M ./kernel/planner.c -26 +13 Chris@10: Chris@10: Tue Jul 30 22:35:24 EDT 2002 stevenj Chris@10: * [project @ 2002-07-31 02:35:24 by stevenj] Chris@10: removed register_registrar and solvtab_exec_reverse hacks Chris@10: Chris@10: M ./kernel/ifftw.h -4 +3 Chris@10: M ./kernel/planner.c -24 +15 Chris@10: M ./kernel/solvtab.c -19 +6 Chris@10: Chris@10: Tue Jul 30 19:54:41 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 23:54:41 by stevenj] Chris@10: register_registrar doesn't search whole solver list (maybe we should change register_solver instead) Chris@10: Chris@10: M ./kernel/planner.c -1 +3 Chris@10: Chris@10: Tue Jul 30 19:36:37 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 23:36:37 by stevenj] Chris@10: credit Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Tue Jul 30 19:34:16 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 23:34:16 by stevenj] Chris@10: added HP/UX ia64 support, courtesy of Teresa L. Johnson Chris@10: Chris@10: M ./kernel/cycle.h -1 +23 Chris@10: Chris@10: Tue Jul 30 13:28:33 EDT 2002 athena Chris@10: * [project @ 2002-07-30 17:28:33 by athena] Chris@10: Fixed alignment checks Chris@10: Chris@10: M ./dft/simd/n1b.c -1 +1 Chris@10: M ./dft/simd/n1f.c -1 +1 Chris@10: M ./dft/simd/t1b.c -1 +1 Chris@10: M ./dft/simd/t1f.c -1 +1 Chris@10: M ./kernel/alloc.c -1 +2 Chris@10: Chris@10: Tue Jul 30 01:20:11 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 05:20:11 by stevenj] Chris@10: ugh, wisdom id fixes in exprt_conf Chris@10: Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/planner.c -2 +4 Chris@10: M ./kernel/solvtab.c -5 +18 Chris@10: Chris@10: Tue Jul 30 00:41:15 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 04:41:15 by stevenj] Chris@10: exprt_registrars -> exprt_conf, added missing SOLVTAB_END Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner.c -5 +6 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Tue Jul 30 00:36:26 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 04:36:26 by stevenj] Chris@10: exprt_registrars should output self-contained configuration Chris@10: Chris@10: M ./kernel/planner.c -9 +20 Chris@10: Chris@10: Mon Jul 29 23:52:07 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 03:52:07 by stevenj] Chris@10: added exprt_registrars Chris@10: Chris@10: M ./dft/conf.c -12 +12 Chris@10: M ./kernel/ifftw.h -2 +7 Chris@10: M ./kernel/planner.c -4 +91 Chris@10: M ./kernel/solvtab.c -3 +5 Chris@10: M ./rdft/conf.c -18 +18 Chris@10: M ./support/Makefile.codelets -3 +3 Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Mon Jul 29 23:42:27 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 03:42:27 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/print.c -2 +3 Chris@10: Chris@10: Mon Jul 29 21:24:51 EDT 2002 athena Chris@10: * [project @ 2002-07-30 01:24:51 by athena] Chris@10: More stringent requirements on strides for SIMD codelets Chris@10: Chris@10: M ./dft/simd/n1b.c -4 +4 Chris@10: M ./dft/simd/n1b.h +3 Chris@10: M ./dft/simd/n1f.c -4 +4 Chris@10: M ./dft/simd/n1f.h +3 Chris@10: M ./dft/simd/t1b.c -2 +2 Chris@10: M ./dft/simd/t1b.h -1 +6 Chris@10: M ./dft/simd/t1f.c -2 +2 Chris@10: M ./dft/simd/t1f.h +5 Chris@10: M ./simd/simd-altivec.h +2 Chris@10: M ./simd/simd-sse.h +2 Chris@10: M ./simd/simd-sse2.h +2 Chris@10: M ./simd/simd.h -2 +3 Chris@10: Chris@10: Mon Jul 29 21:05:49 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 01:05:49 by stevenj] Chris@10: remove warning Chris@10: Chris@10: M ./rdft/buffered2.c -2 +3 Chris@10: Chris@10: Mon Jul 29 20:51:19 EDT 2002 stevenj Chris@10: * [project @ 2002-07-30 00:51:01 by stevenj] Chris@10: use %td for ptrdiff_t and %T for tensors Chris@10: Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./kernel/print.c -3 +13 Chris@10: M ./kernel/traverse.c -2 +7 Chris@10: M ./rdft/problem.c -2 +2 Chris@10: M ./rdft/problem2.c -2 +2 Chris@10: Chris@10: Mon Jul 29 16:17:11 EDT 2002 athena Chris@10: * [project @ 2002-07-29 20:17:11 by athena] Chris@10: Fix for SIMD Chris@10: Chris@10: M ./dft/buffered.c -1 +5 Chris@10: Chris@10: Mon Jul 29 15:40:53 EDT 2002 athena Chris@10: * [project @ 2002-07-29 19:40:53 by athena] Chris@10: Missing lfftw_mkstride and lfftw_stride_destroy Chris@10: Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: Chris@10: Mon Jul 29 14:34:46 EDT 2002 athena Chris@10: * [project @ 2002-07-29 18:34:46 by athena] Chris@10: Implement LDA/STA Chris@10: Chris@10: M ./simd/simd-altivec.h +10 Chris@10: Chris@10: Mon Jul 29 14:19:21 EDT 2002 athena Chris@10: * [project @ 2002-07-29 18:19:21 by athena] Chris@10: More SIMD work Chris@10: Chris@10: M ./dft/simd/n1b.c -4 +4 Chris@10: M ./dft/simd/n1f.c -4 +4 Chris@10: M ./dft/simd/t1b.c -2 +2 Chris@10: M ./dft/simd/t1f.c -2 +2 Chris@10: M ./simd/simd-altivec.h -3 +2 Chris@10: M ./simd/simd-sse.h -3 +2 Chris@10: M ./simd/simd-sse2.h -2 +1 Chris@10: M ./simd/simd.h +3 Chris@10: Chris@10: Mon Jul 29 13:16:12 EDT 2002 athena Chris@10: * [project @ 2002-07-29 17:16:12 by athena] Chris@10: Cleanup Chris@10: Chris@10: M ./simd/simd-altivec.h -9 +9 Chris@10: M ./simd/simd-sse.h -16 +28 Chris@10: M ./simd/simd-sse2.h -13 +25 Chris@10: Chris@10: Mon Jul 29 13:02:38 EDT 2002 stevenj Chris@10: * [project @ 2002-07-29 17:02:38 by stevenj] Chris@10: update Chris@10: Chris@10: M ./ChangeLog +461 Chris@10: Chris@10: Mon Jul 29 12:45:33 EDT 2002 athena Chris@10: * [project @ 2002-07-29 16:45:33 by athena] Chris@10: Also check strides in SIMD codelets Chris@10: Chris@10: M ./dft/simd/n1b.c +4 Chris@10: M ./dft/simd/n1f.c +4 Chris@10: M ./dft/simd/t1b.c +2 Chris@10: M ./dft/simd/t1f.c +2 Chris@10: M ./simd/simd-altivec.h -1 +2 Chris@10: M ./simd/simd-sse.h +2 Chris@10: M ./simd/simd-sse2.h +2 Chris@10: Chris@10: Mon Jul 29 11:26:08 EDT 2002 athena Chris@10: * [project @ 2002-07-29 15:26:08 by athena] Chris@10: Minor changes, mostly for consistency with the big-endian processor Chris@10: Chris@10: M ./simd/simd-altivec.h -18 +18 Chris@10: Chris@10: Mon Jul 29 00:50:06 EDT 2002 stevenj Chris@10: * [project @ 2002-07-29 04:50:06 by stevenj] Chris@10: added comment Chris@10: Chris@10: M ./rdft/rader-dht.c +1 Chris@10: Chris@10: Sun Jul 28 21:19:35 EDT 2002 stevenj Chris@10: * [project @ 2002-07-29 01:19:35 by stevenj] Chris@10: added code for icc's _mm_malloc (memalign replacement) Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./kernel/alloc.c -4 +11 Chris@10: Chris@10: Sun Jul 28 17:33:07 EDT 2002 stevenj Chris@10: * [project @ 2002-07-28 21:33:07 by stevenj] Chris@10: slight fixes Chris@10: Chris@10: M ./rdft/problem2.c -6 +11 Chris@10: M ./rdft/verify.c -4 +7 Chris@10: M ./tests/bench.c -1 +2 Chris@10: Chris@10: Sun Jul 28 16:28:43 EDT 2002 stevenj Chris@10: * [project @ 2002-07-28 20:28:43 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/problem2.c -3 +3 Chris@10: Chris@10: Sun Jul 28 16:13:19 EDT 2002 athena Chris@10: * [project @ 2002-07-28 20:13:19 by athena] Chris@10: Use vec_xor to change sign Chris@10: Chris@10: M ./simd/altivec.c -2 +4 Chris@10: M ./simd/simd-altivec.h -3 +10 Chris@10: Chris@10: Sun Jul 28 16:10:59 EDT 2002 stevenj Chris@10: * [project @ 2002-07-28 20:10:59 by stevenj] Chris@10: added rdft2 Chris@10: Chris@10: A ./rdft/buffered2.c Chris@10: A ./rdft/direct2.c Chris@10: A ./rdft/nop2.c Chris@10: A ./rdft/plan2.c Chris@10: A ./rdft/problem2.c Chris@10: A ./rdft/solve2.c Chris@10: A ./rdft/vrank-geq1-rdft2.c Chris@10: M ./rdft/Makefile.am -2 +4 Chris@10: M ./rdft/buffered2.c +451 Chris@10: M ./rdft/conf.c -1 +7 Chris@10: M ./rdft/direct2.c +221 Chris@10: M ./rdft/khc2r.c -3 +3 Chris@10: M ./rdft/kr2hc.c -3 +3 Chris@10: M ./rdft/nop2.c +88 Chris@10: M ./rdft/plan2.c +33 Chris@10: M ./rdft/problem2.c +175 Chris@10: M ./rdft/rdft.h -1 +53 Chris@10: M ./rdft/solve2.c +31 Chris@10: M ./rdft/verify.c -17 +195 Chris@10: M ./rdft/vrank-geq1-rdft2.c +259 Chris@10: M ./tests/bench.c -6 +35 Chris@10: Chris@10: Sun Jul 28 15:45:54 EDT 2002 athena Chris@10: * [project @ 2002-07-28 19:45:54 by athena] Chris@10: Optimized Chris@10: Chris@10: M ./simd/simd-altivec.h -17 +10 Chris@10: Chris@10: Sun Jul 28 15:11:14 EDT 2002 athena Chris@10: * [project @ 2002-07-28 19:11:14 by athena] Chris@10: Changed ALIGNMENT Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Sun Jul 28 15:09:40 EDT 2002 athena Chris@10: * [project @ 2002-07-28 19:09:40 by athena] Chris@10: alignment := 8 Chris@10: Chris@10: M ./simd/simd-sse.h -1 +1 Chris@10: Chris@10: Sun Jul 28 14:57:22 EDT 2002 athena Chris@10: * [project @ 2002-07-28 18:57:22 by athena] Chris@10: Avoid warning Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Sun Jul 28 14:53:03 EDT 2002 athena Chris@10: * [project @ 2002-07-28 18:53:03 by athena] Chris@10: Oops Chris@10: Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: Chris@10: Sun Jul 28 14:50:09 EDT 2002 athena Chris@10: * [project @ 2002-07-28 18:50:09 by athena] Chris@10: New altivec experiment Chris@10: Chris@10: M ./genfft/annotate.ml -5 +5 Chris@10: M ./genfft/genutil.ml -7 +7 Chris@10: M ./genfft/simd.ml -3 +6 Chris@10: M ./genfft/variable.ml -38 +16 Chris@10: M ./genfft/variable.mli -8 +4 Chris@10: M ./simd/simd-altivec.h -15 +15 Chris@10: M ./simd/simd-sse.h -3 +5 Chris@10: M ./simd/simd-sse2.h -2 +2 Chris@10: Chris@10: Sun Jul 28 13:48:20 EDT 2002 athena Chris@10: * [project @ 2002-07-28 17:48:20 by athena] Chris@10: Nothing Chris@10: Chris@10: M ./simd/simd-altivec.h -6 +6 Chris@10: Chris@10: Sun Jul 28 13:47:50 EDT 2002 athena Chris@10: * [project @ 2002-07-28 17:47:50 by athena] Chris@10: Oops Chris@10: Chris@10: M ./simd/simd-altivec.h -26 +19 Chris@10: Chris@10: Sun Jul 28 13:44:28 EDT 2002 athena Chris@10: * [project @ 2002-07-28 17:44:28 by athena] Chris@10: Nothing Chris@10: Chris@10: M ./simd/simd-altivec.h -19 +26 Chris@10: Chris@10: Sun Jul 28 10:38:10 EDT 2002 athena Chris@10: * [project @ 2002-07-28 14:38:10 by athena] Chris@10: Constants are now in separate file. Chris@10: Chris@10: A ./simd/altivec.c Chris@10: M ./simd/Makefile.am -1 +1 Chris@10: M ./simd/altivec.c +38 Chris@10: M ./simd/simd-altivec.h -20 +13 Chris@10: Chris@10: Sun Jul 28 07:58:37 EDT 2002 athena Chris@10: * [project @ 2002-07-28 11:58:37 by athena] Chris@10: More precise comment Chris@10: Chris@10: M ./simd/simd-altivec.h -1 +1 Chris@10: Chris@10: Sun Jul 28 07:56:40 EDT 2002 athena Chris@10: * [project @ 2002-07-28 11:56:40 by athena] Chris@10: gcc-3.1 bug workaround Chris@10: Chris@10: M ./simd/simd-altivec.h -42 +24 Chris@10: Chris@10: Sun Jul 28 01:39:54 EDT 2002 stevenj Chris@10: * [project @ 2002-07-28 05:39:54 by stevenj] Chris@10: slight optimization, and exported zerotens functions Chris@10: Chris@10: M ./dft/buffered.c -4 +4 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/problem.c -4 +4 Chris@10: M ./rdft/buffered.c -4 +4 Chris@10: M ./rdft/problem.c -4 +4 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: Chris@10: Sun Jul 28 00:54:59 EDT 2002 stevenj Chris@10: * [project @ 2002-07-28 04:54:59 by stevenj] Chris@10: should be a plan_dft, not a plan_rdft Chris@10: Chris@10: M ./rdft/dft-r2hc.c -2 +2 Chris@10: Chris@10: Sat Jul 27 21:36:46 EDT 2002 athena Chris@10: * [project @ 2002-07-28 01:36:46 by athena] Chris@10: Optimizations. Make it work with vanilla non-Apple gcc. Chris@10: Chris@10: M ./simd/simd-altivec.h -21 +82 Chris@10: Chris@10: Sat Jul 27 19:20:09 EDT 2002 stevenj Chris@10: * [project @ 2002-07-27 23:20:09 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/generic.c -5 Chris@10: Chris@10: Sat Jul 27 18:54:01 EDT 2002 stevenj Chris@10: * [project @ 2002-07-27 22:54:01 by stevenj] Chris@10: added hc2r (dif) Chris@10: Chris@10: M ./rdft/generic.c -13 +122 Chris@10: Chris@10: Sat Jul 27 18:31:43 EDT 2002 stevenj Chris@10: * [project @ 2002-07-27 22:31:43 by stevenj] Chris@10: add hc2r (dif) case Chris@10: Chris@10: M ./rdft/rader-hc2hc.c -2 +158 Chris@10: Chris@10: Sat Jul 27 15:09:40 EDT 2002 athena Chris@10: * [project @ 2002-07-27 19:09:40 by athena] Chris@10: Altivec port Chris@10: Chris@10: M ./simd/simd-altivec.h -38 +131 Chris@10: M ./support/Makefile.codelets -3 +3 Chris@10: Chris@10: Sat Jul 27 15:06:21 EDT 2002 athena Chris@10: * [project @ 2002-07-27 19:06:21 by athena] Chris@10: Fixed signed/unsigned bug. Chris@10: Chris@10: M ./kernel/twiddle.c -2 +3 Chris@10: Chris@10: Thu Jul 25 20:11:26 EDT 2002 athena Chris@10: * [project @ 2002-07-26 00:11:26 by athena] Chris@10: Make rank0 unapplicable to in-place problems. Chris@10: Chris@10: M ./dft/rank0.c -1 +2 Chris@10: M ./rdft/rank0.c -1 +2 Chris@10: Chris@10: Thu Jul 25 17:10:52 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 21:10:52 by stevenj] Chris@10: only works for r odd Chris@10: Chris@10: M ./rdft/generic.c -1 +1 Chris@10: Chris@10: Thu Jul 25 15:30:06 EDT 2002 athena Chris@10: * [project @ 2002-07-25 19:30:06 by athena] Chris@10: Reinserted much better timing-avoidance heuristic Chris@10: Chris@10: M ./kernel/planner-score.c -2 +7 Chris@10: Chris@10: Thu Jul 25 15:21:13 EDT 2002 athena Chris@10: * [project @ 2002-07-25 19:21:13 by athena] Chris@10: Score is now a property of the plan, not of the solver. Chris@10: Revised representation of closures. Chris@10: Chris@10: M ./dft/buffered.c -5 +1 Chris@10: M ./kernel/ifftw.h -2 +7 Chris@10: M ./kernel/plan.c -4 +6 Chris@10: M ./kernel/planner-score.c -17 +30 Chris@10: M ./kernel/traverse.c -17 +16 Chris@10: M ./rdft/buffered.c -5 +1 Chris@10: M ./tests/bench.c -7 +14 Chris@10: Chris@10: Thu Jul 25 06:36:51 EDT 2002 athena Chris@10: * [project @ 2002-07-25 10:36:51 by athena] Chris@10: Cosmetic changes. Added hc2r_128.c Chris@10: Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -3 +3 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -1 +1 Chris@10: Chris@10: Thu Jul 25 01:37:53 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 05:37:53 by stevenj] Chris@10: added hc2r Chris@10: Chris@10: M ./rdft/rader-dht.c -16 +55 Chris@10: Chris@10: Thu Jul 25 00:51:45 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 04:51:45 by stevenj] Chris@10: added hc2hc-difbuf Chris@10: Chris@10: A ./rdft/hc2hc-buf.c Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/hc2hc-buf.c +248 Chris@10: R ./rdft/hc2hc-ditbuf.c Chris@10: M ./rdft/khc2hc-dif.c -3 +1 Chris@10: Chris@10: Thu Jul 25 00:25:06 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 04:25:06 by stevenj] Chris@10: added rdft-dif Chris@10: Chris@10: A ./rdft/hc2hc-dif.c Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/hc2hc-dif.c +128 Chris@10: M ./rdft/hc2hc.c -3 +5 Chris@10: M ./rdft/khc2hc-dif.c -4 +3 Chris@10: M ./rdft/rdft.h -1 +3 Chris@10: Chris@10: Thu Jul 25 00:22:36 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 04:22:36 by stevenj] Chris@10: whoops, hc2r must be conjugated to have right sign Chris@10: Chris@10: M ./rdft/verify.c -3 +3 Chris@10: Chris@10: Wed Jul 24 23:27:45 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 03:27:45 by stevenj] Chris@10: slight change Chris@10: Chris@10: M ./dft/ct-dif.c -4 +6 Chris@10: Chris@10: Wed Jul 24 23:24:24 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 03:24:24 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/verify.c -3 +2 Chris@10: Chris@10: Wed Jul 24 22:46:39 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 02:46:39 by stevenj] Chris@10: support hc2r codelets Chris@10: Chris@10: A ./rdft/direct.c Chris@10: M ./rdft/Makefile.am -5 +5 Chris@10: M ./rdft/codelet.h -1 +3 Chris@10: R ./rdft/direct-r2hc.c Chris@10: M ./rdft/direct.c +227 Chris@10: M ./rdft/khc2r.c -5 +2 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: Chris@10: Wed Jul 24 22:01:53 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 02:01:53 by stevenj] Chris@10: use vector plan for r/i instead of two separate plans Chris@10: Chris@10: M ./rdft/dft-r2hc.c -36 +20 Chris@10: Chris@10: Wed Jul 24 20:36:34 EDT 2002 stevenj Chris@10: * [project @ 2002-07-25 00:36:34 by stevenj] Chris@10: hack to allow rader/generic to work in-place for small prime sizes, instead of always using buffered Chris@10: Chris@10: M ./dft/buffered.c -1 +5 Chris@10: M ./dft/rader.c -2 +2 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./rdft/buffered.c -1 +5 Chris@10: M ./rdft/rader-dht.c -1 +1 Chris@10: M ./rdft/rader-hc2hc.c -1 +1 Chris@10: Chris@10: Wed Jul 24 18:04:41 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 22:04:41 by stevenj] Chris@10: added rdft-generic Chris@10: Chris@10: A ./rdft/generic.c Chris@10: M ./rdft/Makefile.am -4 +4 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/generic.c +281 Chris@10: Chris@10: Wed Jul 24 17:27:34 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 21:27:34 by stevenj] Chris@10: fixed add count Chris@10: Chris@10: M ./dft/generic.c -1 +1 Chris@10: Chris@10: Wed Jul 24 14:52:26 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 18:52:26 by stevenj] Chris@10: again Chris@10: Chris@10: M ./rdft/rader-hc2hc.c -2 +2 Chris@10: Chris@10: Wed Jul 24 14:51:58 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 18:51:58 by stevenj] Chris@10: slight fix Chris@10: Chris@10: M ./rdft/rader-hc2hc.c -4 +5 Chris@10: Chris@10: Wed Jul 24 14:51:07 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 18:51:07 by stevenj] Chris@10: fixed comment Chris@10: Chris@10: M ./rdft/rader-hc2hc.c -3 +3 Chris@10: Chris@10: Wed Jul 24 14:41:24 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 18:41:24 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./tests/bench.c -2 Chris@10: Chris@10: Wed Jul 24 14:38:15 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 18:38:15 by stevenj] Chris@10: added rader-hc2hc Chris@10: Chris@10: A ./rdft/rader-hc2hc.c Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: M ./rdft/rader-hc2hc.c +482 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Wed Jul 24 00:07:59 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 04:07:59 by stevenj] Chris@10: whoops, initialize W Chris@10: Chris@10: M ./dft/rader.c +1 Chris@10: Chris@10: Tue Jul 23 23:03:09 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 03:03:09 by stevenj] Chris@10: strides should not be unsigned Chris@10: Chris@10: M ./rdft/rader-dht.c -2 +4 Chris@10: Chris@10: Tue Jul 23 23:02:08 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 03:02:08 by stevenj] Chris@10: more stride sign fixes Chris@10: Chris@10: M ./dft/rader.c -3 +5 Chris@10: Chris@10: Tue Jul 23 23:01:04 EDT 2002 stevenj Chris@10: * [project @ 2002-07-24 03:01:04 by stevenj] Chris@10: strides should not be unsigned! Chris@10: Chris@10: M ./dft/rader.c -1 +2 Chris@10: Chris@10: Tue Jul 23 14:55:25 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 18:55:25 by stevenj] Chris@10: added comment Chris@10: Chris@10: M ./rdft/dft-r2hc.c -1 +4 Chris@10: Chris@10: Tue Jul 23 14:52:04 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 18:52:04 by stevenj] Chris@10: another fix to op count Chris@10: Chris@10: M ./rdft/r2hc-hc2r.c -1 +4 Chris@10: Chris@10: Tue Jul 23 14:51:01 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 18:51:01 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/r2hc-hc2r.c -2 +2 Chris@10: Chris@10: Tue Jul 23 14:49:43 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 18:49:43 by stevenj] Chris@10: slight fix to op counts Chris@10: Chris@10: M ./rdft/dft-r2hc.c -3 +3 Chris@10: M ./rdft/r2hc-hc2r.c -3 +3 Chris@10: Chris@10: Tue Jul 23 14:09:19 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 18:09:18 by stevenj] Chris@10: added dft-r2hc Chris@10: Chris@10: A ./rdft/dft-r2hc.c Chris@10: M ./rdft/Makefile.am -5 +6 Chris@10: M ./rdft/conf.c -2 +3 Chris@10: M ./rdft/dft-r2hc.c +187 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: Chris@10: Tue Jul 23 02:50:12 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 06:50:12 by stevenj] Chris@10: better comment and var. name Chris@10: Chris@10: M ./rdft/rader-dht.c -7 +7 Chris@10: Chris@10: Tue Jul 23 02:39:11 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 06:39:11 by stevenj] Chris@10: fixed tests for hc2r, and added r2hc-hc2r Chris@10: Chris@10: A ./rdft/r2hc-hc2r.c Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/conf.c -1 +2 Chris@10: M ./rdft/r2hc-hc2r.c +182 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: M ./rdft/verify.c -1 +46 Chris@10: M ./tests/bench.c -1 +6 Chris@10: Chris@10: Tue Jul 23 00:45:23 EDT 2002 stevenj Chris@10: * [project @ 2002-07-23 04:45:23 by stevenj] Chris@10: added rader-dht Chris@10: Chris@10: A ./rdft/rader-dht.c Chris@10: M ./rdft/Makefile.am -2 +3 Chris@10: M ./rdft/conf.c -1 +2 Chris@10: M ./rdft/rader-dht.c +433 Chris@10: M ./rdft/rdft.h -1 +2 Chris@10: Chris@10: Mon Jul 22 21:05:12 EDT 2002 athena Chris@10: * [project @ 2002-07-23 01:05:12 by athena] Chris@10: Added r2hc_128, what the hell. Chris@10: Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +1 Chris@10: Chris@10: Mon Jul 22 20:48:59 EDT 2002 athena Chris@10: * [project @ 2002-07-23 00:48:59 by athena] Chris@10: Added codelets that compute twiddle factors Chris@10: Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +7 Chris@10: Chris@10: Mon Jul 22 19:57:16 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 23:57:16 by stevenj] Chris@10: added rdft-buffered Chris@10: Chris@10: A ./rdft/buffered.c Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/buffered.c +330 Chris@10: M ./rdft/conf.c -2 +2 Chris@10: Chris@10: Mon Jul 22 19:43:39 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 23:43:39 by stevenj] Chris@10: added hc2hc-ditbuf Chris@10: Chris@10: A ./rdft/hc2hc-ditbuf.c Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/hc2hc-ditbuf.c +189 Chris@10: M ./rdft/khc2hc-dit.c -3 +1 Chris@10: Chris@10: Mon Jul 22 14:29:04 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 18:29:04 by stevenj] Chris@10: use STACK_MALLOC (alloca), since generic radix is always small Chris@10: Chris@10: M ./dft/generic.c -3 +3 Chris@10: Chris@10: Mon Jul 22 14:22:43 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 18:22:43 by stevenj] Chris@10: small cleanup Chris@10: Chris@10: M ./rdft/hc2hc-dit.c -2 +2 Chris@10: Chris@10: Mon Jul 22 07:42:13 EDT 2002 athena Chris@10: * [project @ 2002-07-22 11:42:13 by athena] Chris@10: What the hell was I thinking? Chris@10: Chris@10: M ./rdft/problem.c -18 +8 Chris@10: Chris@10: Mon Jul 22 07:37:12 EDT 2002 athena Chris@10: * [project @ 2002-07-22 11:37:12 by athena] Chris@10: Reduced code size by using table instead of switch statement. Chris@10: Chris@10: M ./rdft/problem.c -12 +16 Chris@10: Chris@10: Mon Jul 22 07:27:06 EDT 2002 athena Chris@10: * [project @ 2002-07-22 11:27:06 by athena] Chris@10: Changed hash function to avoid collisions with DFT. Chris@10: Chris@10: M ./rdft/problem.c -4 +4 Chris@10: Chris@10: Mon Jul 22 01:37:06 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 05:37:06 by stevenj] Chris@10: added missing file, whoops Chris@10: Chris@10: A ./rdft/hc2hc-dit.c Chris@10: Chris@10: Mon Jul 22 01:24:17 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 05:24:17 by stevenj] Chris@10: whoops, generate enough twiddles for odd m Chris@10: Chris@10: M ./rdft/hc2hc.c -2 +2 Chris@10: Chris@10: Mon Jul 22 01:10:21 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 05:10:21 by stevenj] Chris@10: don't try to verify R2HCII or HC2RIII plans Chris@10: Chris@10: M ./rdft/verify.c -5 +7 Chris@10: Chris@10: Mon Jul 22 01:05:00 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 05:05:00 by stevenj] Chris@10: recursive case now works, I think Chris@10: Chris@10: M ./rdft/hc2hc.c -9 +4 Chris@10: Chris@10: Mon Jul 22 01:04:40 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 05:04:40 by stevenj] Chris@10: add extra impulse test for debugging Chris@10: Chris@10: M ./rdft/verify.c -1 +4 Chris@10: Chris@10: Mon Jul 22 01:02:38 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 05:02:38 by stevenj] Chris@10: whoops, multiply ios offset by stride (and rename to ioffset) Chris@10: Chris@10: M ./rdft/direct-r2hc.c -4 +5 Chris@10: Chris@10: Mon Jul 22 00:22:02 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 04:22:02 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/verify.c -2 +2 Chris@10: Chris@10: Sun Jul 21 23:58:14 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 03:58:14 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./rdft/verify.c -2 +2 Chris@10: Chris@10: Sun Jul 21 23:43:03 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 03:43:03 by stevenj] Chris@10: added hc2hc-dit Chris@10: Chris@10: M ./genfft/gen_hc2hc.ml -3 +3 Chris@10: M ./rdft/Makefile.am -2 +2 Chris@10: M ./rdft/hc2hc.c -10 +21 Chris@10: M ./rdft/khc2hc-dit.c -4 +3 Chris@10: Chris@10: Sun Jul 21 23:15:12 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 03:15:12 by stevenj] Chris@10: twiddles can be shared with smaller m's Chris@10: Chris@10: M ./kernel/twiddle.c -4 +4 Chris@10: Chris@10: Sun Jul 21 22:34:28 EDT 2002 stevenj Chris@10: * [project @ 2002-07-22 02:34:28 by stevenj] Chris@10: preparing for recursive rdft... Chris@10: Chris@10: A ./rdft/hc2hc.c Chris@10: A ./rdft/hc2hc.h Chris@10: M ./rdft/Makefile.am -3 +3 Chris@10: M ./rdft/codelet.h -1 +2 Chris@10: M ./rdft/codelets/hfb.c -2 +2 Chris@10: M ./rdft/hc2hc.c +249 Chris@10: M ./rdft/hc2hc.h +69 Chris@10: Chris@10: Sun Jul 21 19:31:22 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 23:31:22 by stevenj] Chris@10: slight fix, to match libbench/verify.c Chris@10: Chris@10: M ./rdft/verify.c -3 +3 Chris@10: Chris@10: Sun Jul 21 18:43:12 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 22:43:12 by stevenj] Chris@10: r2hcII has imag parts offset by n-1, not n. We can also allocate fewer strides. Chris@10: Chris@10: M ./rdft/direct-r2hc.c -4 +7 Chris@10: Chris@10: Sun Jul 21 18:27:09 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 22:27:09 by stevenj] Chris@10: delete unused var Chris@10: Chris@10: M ./rdft/rank0.c -2 +2 Chris@10: Chris@10: Sun Jul 21 02:06:53 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 06:06:53 by stevenj] Chris@10: added some rdft solvers Chris@10: Chris@10: A ./rdft/direct-r2hc.c Chris@10: A ./rdft/indirect.c Chris@10: A ./rdft/nop.c Chris@10: A ./rdft/rank0.c Chris@10: A ./rdft/vrank-geq1.c Chris@10: A ./rdft/vrank2-transpose.c Chris@10: A ./rdft/vrank3-transpose.c Chris@10: M ./rdft/Makefile.am -2 +4 Chris@10: M ./rdft/codelet.h -3 +8 Chris@10: M ./rdft/codelets/hc2r.c -2 +2 Chris@10: M ./rdft/codelets/r2hc.c -2 +2 Chris@10: M ./rdft/conf.c -4 +4 Chris@10: M ./rdft/direct-r2hc.c +174 Chris@10: M ./rdft/indirect.c +235 Chris@10: R ./rdft/khc2rIII.c Chris@10: M ./rdft/kr2hc.c -5 +2 Chris@10: R ./rdft/kr2hcII.c Chris@10: M ./rdft/nop.c +97 Chris@10: M ./rdft/problem.c -3 +20 Chris@10: M ./rdft/rank0.c +217 Chris@10: M ./rdft/rdft.h -5 +4 Chris@10: M ./rdft/vrank-geq1.c +260 Chris@10: M ./rdft/vrank2-transpose.c +130 Chris@10: M ./rdft/vrank3-transpose.c +171 Chris@10: Chris@10: Sun Jul 21 01:52:54 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 05:52:54 by stevenj] Chris@10: pass identifier in FFTW() through another macro so that the mangled name Chris@10: can itself be a preprocessor symbol Chris@10: Chris@10: M ./kernel/fftw3.h -4 +5 Chris@10: Chris@10: Sun Jul 21 01:05:21 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 05:05:21 by stevenj] Chris@10: fix in comment Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Sun Jul 21 00:47:03 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 04:47:03 by stevenj] Chris@10: bench tests rdft plans Chris@10: Chris@10: M ./Makefile.am -2 +5 Chris@10: M ./rdft/rdft.h -2 +3 Chris@10: M ./tests/bench.c -16 +36 Chris@10: Chris@10: Sun Jul 21 00:22:14 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 04:22:14 by stevenj] Chris@10: make rdft.h and dft.h compatible Chris@10: Chris@10: M ./rdft/codelet.h -4 +4 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Jul 21 00:12:19 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 04:12:19 by stevenj] Chris@10: first-draft rdft verify Chris@10: Chris@10: A ./rdft/verify.c Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/problem.c -8 +8 Chris@10: M ./rdft/rdft.h -5 +5 Chris@10: M ./rdft/verify.c +554 Chris@10: Chris@10: Sat Jul 20 22:09:15 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 02:09:15 by stevenj] Chris@10: got rid of annoying warnings Chris@10: Chris@10: M ./rdft/khc2hc-dif.c -1 +2 Chris@10: M ./rdft/khc2hc-dit.c -1 +2 Chris@10: M ./rdft/khc2r.c -1 +2 Chris@10: M ./rdft/khc2rIII.c -1 +2 Chris@10: M ./rdft/kr2hc.c -1 +2 Chris@10: M ./rdft/kr2hcII.c -1 +2 Chris@10: Chris@10: Sat Jul 20 22:07:37 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 02:07:37 by stevenj] Chris@10: added stub codelet registration for linking purposes Chris@10: Chris@10: A ./rdft/khc2hc-dif.c Chris@10: A ./rdft/khc2hc-dit.c Chris@10: A ./rdft/khc2r.c Chris@10: A ./rdft/khc2rIII.c Chris@10: A ./rdft/kr2hc.c Chris@10: A ./rdft/kr2hcII.c Chris@10: M ./rdft/Makefile.am -1 +2 Chris@10: M ./rdft/khc2hc-dif.c +31 Chris@10: M ./rdft/khc2hc-dit.c +31 Chris@10: M ./rdft/khc2r.c +31 Chris@10: M ./rdft/khc2rIII.c +31 Chris@10: M ./rdft/kr2hc.c +31 Chris@10: M ./rdft/kr2hcII.c +31 Chris@10: M ./rdft/rdft.h -2 +2 Chris@10: Chris@10: Sat Jul 20 21:46:03 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 01:46:03 by stevenj] Chris@10: basic rdft stuff Chris@10: Chris@10: A ./rdft/conf.c Chris@10: A ./rdft/plan.c Chris@10: A ./rdft/problem.c Chris@10: A ./rdft/rdft.h Chris@10: A ./rdft/solve.c Chris@10: M ./rdft/Makefile.am -1 +1 Chris@10: M ./rdft/conf.c +47 Chris@10: M ./rdft/plan.c +33 Chris@10: M ./rdft/problem.c +145 Chris@10: M ./rdft/rdft.h +82 Chris@10: M ./rdft/solve.c +31 Chris@10: Chris@10: Sat Jul 20 21:06:50 EDT 2002 stevenj Chris@10: * [project @ 2002-07-21 01:06:50 by stevenj] Chris@10: rdft codelets now compile Chris@10: Chris@10: A ./rdft/Makefile.am Chris@10: A ./rdft/codelet.h Chris@10: A ./rdft/codelets/Makefile.am Chris@10: A ./rdft/codelets/hb.h Chris@10: A ./rdft/codelets/hc2r.c Chris@10: A ./rdft/codelets/hc2r.h Chris@10: A ./rdft/codelets/hc2rIII.h Chris@10: A ./rdft/codelets/hf.h Chris@10: A ./rdft/codelets/hfb.c Chris@10: A ./rdft/codelets/r2hc.c Chris@10: A ./rdft/codelets/r2hc.h Chris@10: A ./rdft/codelets/r2hcII.h Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +2 Chris@10: M ./dft/codelet.h -33 +1 Chris@10: M ./genfft/gen_hc2hc.ml -3 +3 Chris@10: M ./kernel/ifftw.h -1 +35 Chris@10: M ./rdft/Makefile.am +6 Chris@10: M ./rdft/codelet.h +133 Chris@10: M ./rdft/codelets/Makefile.am +7 Chris@10: M ./rdft/codelets/hb.h +23 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -3 +3 Chris@10: M ./rdft/codelets/hc2r.c +44 Chris@10: M ./rdft/codelets/hc2r.h +23 Chris@10: M ./rdft/codelets/hc2rIII.h +23 Chris@10: M ./rdft/codelets/hf.h +23 Chris@10: M ./rdft/codelets/hfb.c +41 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -2 +2 Chris@10: M ./rdft/codelets/r2hc.c +44 Chris@10: M ./rdft/codelets/r2hc.h +23 Chris@10: M ./rdft/codelets/r2hcII.h +23 Chris@10: Chris@10: Sat Jul 20 18:40:31 EDT 2002 athena Chris@10: * [project @ 2002-07-20 22:40:31 by athena] Chris@10: Oops, was generating rdfts instead of hdfts Chris@10: Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: Chris@10: Sat Jul 20 18:25:47 EDT 2002 athena Chris@10: * [project @ 2002-07-20 22:25:47 by athena] Chris@10: Added hc2r codelets Chris@10: Chris@10: A ./rdft/codelets/hc2r/ Chris@10: A ./rdft/codelets/hc2r/Makefile.am Chris@10: M ./TODO +4 Chris@10: M ./configure.ac +1 Chris@10: M ./genfft/twiddle.ml -7 +14 Chris@10: M ./genfft-k7/twiddle.ml -12 +7 Chris@10: M ./kernel/twiddle.c -26 +39 Chris@10: M ./rdft/codelets/hc2r/Makefile.am -1 +42 Chris@10: Chris@10: Sat Jul 20 17:54:39 EDT 2002 athena Chris@10: * [project @ 2002-07-20 21:54:39 by athena] Chris@10: return W in hc2hc codelets Chris@10: Chris@10: M ./genfft/gen_hc2hc.ml -3 +4 Chris@10: Chris@10: Sat Jul 20 17:51:06 EDT 2002 athena Chris@10: * [project @ 2002-07-20 21:51:06 by athena] Chris@10: Some work on rdft codelets Chris@10: Chris@10: A ./rdft/ Chris@10: A ./rdft/codelets/ Chris@10: A ./rdft/codelets/r2hc/ Chris@10: A ./rdft/codelets/r2hc/Makefile.am Chris@10: M ./configure.ac +2 Chris@10: M ./dft/codelets/inplace/Makefile.am +4 Chris@10: M ./dft/codelets/standard/Makefile.am +3 Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: M ./genfft/gen_hc2hc.ml -62 +18 Chris@10: M ./genfft/gen_hc2r.ml -5 +24 Chris@10: M ./genfft/gen_r2hc.ml -5 +21 Chris@10: M ./genfft/trig.ml -22 +22 Chris@10: M ./rdft/codelets/r2hc/Makefile.am -1 +42 Chris@10: M ./support/Makefile.codelets -9 +6 Chris@10: Chris@10: Tue Jul 16 13:55:50 EDT 2002 athena Chris@10: * [project @ 2002-07-16 17:55:50 by athena] Chris@10: fix const Chris@10: Chris@10: M ./kernel/fftw3.h -3 +4 Chris@10: Chris@10: Tue Jul 16 07:00:10 EDT 2002 athena Chris@10: * [project @ 2002-07-16 11:00:10 by athena] Chris@10: Separate CFLAGS in codelets. Fix const in certain places. Chris@10: Chris@10: M ./acinclude.m4 -2 +6 Chris@10: M ./configure.ac +1 Chris@10: M ./dft/codelets/inplace/Makefile.am +1 Chris@10: M ./dft/codelets/standard/Makefile.am +1 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: M ./dft/simd/codelets/Makefile.am +1 Chris@10: M ./kernel/version.c -3 +4 Chris@10: M ./tests/bench.c -2 +9 Chris@10: Chris@10: Mon Jul 15 21:10:42 EDT 2002 stevenj Chris@10: * [project @ 2002-07-16 01:10:42 by stevenj] Chris@10: note buffering problem Chris@10: Chris@10: M ./TODO +2 Chris@10: Chris@10: Mon Jul 15 20:27:51 EDT 2002 athena Chris@10: * [project @ 2002-07-16 00:27:51 by athena] Chris@10: Removed unpredictable branch from inner loop Chris@10: Chris@10: M ./dft/generic.c -2 +2 Chris@10: Chris@10: Mon Jul 15 19:35:04 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 23:35:04 by stevenj] Chris@10: update Chris@10: Chris@10: M ./TODO -2 Chris@10: Chris@10: Mon Jul 15 19:31:39 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 23:31:39 by stevenj] Chris@10: optimization Chris@10: Chris@10: M ./dft/generic.c -1 +2 Chris@10: Chris@10: Mon Jul 15 19:28:30 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 23:28:30 by stevenj] Chris@10: added generic dit Chris@10: Chris@10: A ./dft/generic.c Chris@10: M ./dft/Makefile.am -4 +4 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/generic.c +219 Chris@10: M ./kernel/ifftw.h -2 +3 Chris@10: M ./kernel/twiddle.c -1 +13 Chris@10: Chris@10: Mon Jul 15 17:03:53 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 21:03:53 by stevenj] Chris@10: whoops, mksolver should be static Chris@10: Chris@10: M ./dft/rader.c -2 +2 Chris@10: Chris@10: Mon Jul 15 16:46:36 EDT 2002 athena Chris@10: * [project @ 2002-07-15 20:46:35 by athena] Chris@10: First implementation of gen_hc2hc, probably still buggy. Chris@10: Chris@10: A ./genfft/gen_hc2hc.ml Chris@10: M ./genfft/Makefile.am -9 +15 Chris@10: M ./genfft/algsimp.ml -4 +5 Chris@10: M ./genfft/c.ml -3 +10 Chris@10: M ./genfft/c.mli -1 +2 Chris@10: M ./genfft/gen_hc2hc.ml +220 Chris@10: M ./genfft/gen_hc2r.ml -24 +69 Chris@10: M ./genfft/gen_r2hc.ml -22 +66 Chris@10: M ./genfft/genutil.ml -13 +17 Chris@10: Chris@10: Mon Jul 15 16:40:23 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 20:40:23 by stevenj] Chris@10: don't count loading of twiddle factors in ops.other, since it isn't Chris@10: counted for the codelets Chris@10: Chris@10: M ./dft/rader.c -1 Chris@10: Chris@10: Mon Jul 15 15:13:19 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 19:13:19 by stevenj] Chris@10: plan_destroy puts plan to sleep before deallocating it, to eliminate duplicate free calls in solvers Chris@10: Chris@10: M ./dft/ct.c -2 +1 Chris@10: M ./dft/rader.c -2 Chris@10: M ./kernel/plan.c -2 +5 Chris@10: Chris@10: Mon Jul 15 15:07:41 EDT 2002 stevenj Chris@10: * [project @ 2002-07-15 19:07:41 by stevenj] Chris@10: fftw2-like vector recursion flag Chris@10: Chris@10: M ./dft/ct-dif.c -1 +6 Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/ct.c -1 +5 Chris@10: M ./dft/vrank-geq1.c -2 +10 Chris@10: M ./kernel/ifftw.h -2 +6 Chris@10: M ./tests/bench.c -2 +1 Chris@10: Chris@10: Sun Jul 14 21:01:44 EDT 2002 athena Chris@10: * [project @ 2002-07-15 01:01:44 by athena] Chris@10: More jokes Chris@10: Chris@10: M ./kernel/planner.c -1 +2 Chris@10: Chris@10: Sun Jul 14 20:36:01 EDT 2002 athena Chris@10: * [project @ 2002-07-15 00:36:01 by athena] Chris@10: Bless plan for testing purposes Chris@10: Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Sun Jul 14 20:35:49 EDT 2002 athena Chris@10: * [project @ 2002-07-15 00:35:49 by athena] Chris@10: Canonical linked-list deletion (hope it is right) Chris@10: Chris@10: M ./kernel/planner.c -12 +13 Chris@10: Chris@10: Sun Jul 14 18:26:19 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 22:26:19 by stevenj] Chris@10: use estimating planner for cld_omega Chris@10: Chris@10: M ./dft/rader.c -1 +1 Chris@10: Chris@10: Sun Jul 14 18:10:56 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 22:10:56 by stevenj] Chris@10: better internal naming Chris@10: Chris@10: M ./dft/rader.c -3 +3 Chris@10: Chris@10: Sun Jul 14 18:10:01 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 22:10:01 by stevenj] Chris@10: printing should really be fixed now, grrr Chris@10: Chris@10: M ./dft/rader.c -10 +12 Chris@10: Chris@10: Sun Jul 14 17:57:12 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 21:57:12 by stevenj] Chris@10: print all distinct child plans Chris@10: Chris@10: M ./dft/rader.c -2 +7 Chris@10: Chris@10: Sun Jul 14 17:49:21 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 21:49:21 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Jul 14 17:45:54 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 21:45:54 by stevenj] Chris@10: whoops, destroy should delete twiddle/omega from list Chris@10: Chris@10: M ./dft/rader.c -4 +2 Chris@10: Chris@10: Sun Jul 14 17:33:02 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 21:33:02 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/planner.c -3 +6 Chris@10: Chris@10: Sun Jul 14 17:12:14 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 21:12:14 by stevenj] Chris@10: added plan_bless and FORGET_ACCURSED Chris@10: Chris@10: M ./kernel/ifftw.h -2 +6 Chris@10: M ./kernel/plan.c -1 +14 Chris@10: M ./kernel/planner.c -13 +24 Chris@10: Chris@10: Sun Jul 14 16:15:43 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 20:15:43 by stevenj] Chris@10: further cleanup Chris@10: Chris@10: M ./kernel/traverse.c -6 +4 Chris@10: Chris@10: Sun Jul 14 16:14:15 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 20:14:15 by stevenj] Chris@10: slight cleanup Chris@10: Chris@10: M ./kernel/traverse.c -8 +7 Chris@10: Chris@10: Sun Jul 14 16:09:17 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 20:09:17 by stevenj] Chris@10: added traverse_plan via print (ugh) Chris@10: Chris@10: A ./kernel/traverse.c Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: M ./kernel/traverse.c +115 Chris@10: M ./tests/bench.c +12 Chris@10: Chris@10: Sun Jul 14 15:08:29 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 19:08:29 by stevenj] Chris@10: added TW_FULL, and additional n parameter for twiddles Chris@10: Chris@10: M ./dft/ct.c -3 +4 Chris@10: M ./kernel/ifftw.h -5 +5 Chris@10: M ./kernel/twiddle.c -18 +34 Chris@10: Chris@10: Sun Jul 14 15:03:51 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 19:03:51 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./kernel/planner.c -2 +4 Chris@10: Chris@10: Sun Jul 14 13:49:20 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 17:49:20 by stevenj] Chris@10: save flags before invoking solver mkplan Chris@10: Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/planner-naive.c -2 +2 Chris@10: M ./kernel/planner-score.c -2 +2 Chris@10: M ./kernel/planner.c -3 +10 Chris@10: Chris@10: Sun Jul 14 09:28:37 EDT 2002 athena Chris@10: * [project @ 2002-07-14 13:28:37 by athena] Chris@10: Chris@10: M ./TODO -6 +1 Chris@10: Chris@10: Sat Jul 13 22:17:29 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 02:17:29 by stevenj] Chris@10: added support for UNICOS _rtc() real-time-clock intrinsic function Chris@10: Chris@10: M ./configure.ac -1 +10 Chris@10: M ./kernel/cycle.h -2 +20 Chris@10: Chris@10: Sat Jul 13 22:06:35 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 02:06:35 by stevenj] Chris@10: fixed typo: HAVE_TIME_H should include time.h, not sys/time.h Chris@10: Chris@10: M ./kernel/timer.c -2 +2 Chris@10: Chris@10: Sat Jul 13 21:46:02 EDT 2002 stevenj Chris@10: * [project @ 2002-07-14 01:46:02 by stevenj] Chris@10: support AIX read_real_time timer Chris@10: Chris@10: M ./configure.ac +1 Chris@10: M ./kernel/cycle.h -1 +24 Chris@10: Chris@10: Sat Jul 13 17:02:51 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 21:02:51 by stevenj] Chris@10: use && instead of the (sigh) unportable -a Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: Chris@10: Sat Jul 13 16:38:18 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 20:38:18 by stevenj] Chris@10: use AC_HELP_STRING Chris@10: Chris@10: M ./configure.ac -11 +11 Chris@10: Chris@10: Sat Jul 13 16:05:43 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 20:05:43 by stevenj] Chris@10: support long-double precision Chris@10: Chris@10: A ./kernel/lfftw3.h Chris@10: M ./configure.ac -2 +22 Chris@10: M ./dft/codelet.h -2 +6 Chris@10: M ./dft/verify.c -10 +29 Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/dfftw3.h -1 +2 Chris@10: M ./kernel/fftw3.h -2 +5 Chris@10: M ./kernel/ifftw.h -9 +14 Chris@10: M ./kernel/lfftw3.h +25 Chris@10: M ./kernel/sfftw3.h -1 +2 Chris@10: M ./libbench/bench-main.c -3 +5 Chris@10: M ./libbench/bench-user.h -4 +8 Chris@10: M ./libbench/info.c -2 +4 Chris@10: M ./libbench/verify.c -10 +38 Chris@10: M ./simd/simd-sse2.h -1 +1 Chris@10: Chris@10: Sat Jul 13 15:48:10 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 19:48:10 by stevenj] Chris@10: whoops whoops Chris@10: Chris@10: M ./dft/rader.c -2 +2 Chris@10: Chris@10: Sat Jul 13 15:47:39 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 19:47:39 by stevenj] Chris@10: whoops Chris@10: Chris@10: M ./dft/rader.c -1 +1 Chris@10: Chris@10: Sat Jul 13 14:13:42 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 18:13:42 by stevenj] Chris@10: buffered solver strides have been fixed Chris@10: Chris@10: M ./TODO -7 Chris@10: Chris@10: Sat Jul 13 13:48:13 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 17:48:13 by stevenj] Chris@10: convention Chris@10: Chris@10: M ./dft/rader.c -1 +1 Chris@10: Chris@10: Sat Jul 13 12:50:06 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 16:50:06 by stevenj] Chris@10: share twiddle arrays in Rader Chris@10: Chris@10: M ./TODO -2 Chris@10: M ./dft/rader.c -5 +68 Chris@10: Chris@10: Sat Jul 13 12:48:10 EDT 2002 stevenj Chris@10: * [project @ 2002-07-13 16:48:10 by stevenj] Chris@10: call done() after verify Chris@10: Chris@10: M ./libbench/verify.c -1 +2 Chris@10: Chris@10: Fri Jul 12 15:42:04 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 19:42:04 by stevenj] Chris@10: output planner time with -v Chris@10: Chris@10: M ./tests/bench.c +7 Chris@10: Chris@10: Fri Jul 12 15:40:14 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 19:40:14 by stevenj] Chris@10: support double outputs Chris@10: Chris@10: M ./kernel/print.c -2 +9 Chris@10: Chris@10: Fri Jul 12 15:09:19 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 19:09:19 by stevenj] Chris@10: removed extraneous parens Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Fri Jul 12 15:08:13 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 19:08:13 by stevenj] Chris@10: increase maxbufsz to 64k; makes a big difference for large 2d transforms Chris@10: Chris@10: M ./dft/buffered.c -2 +2 Chris@10: Chris@10: Fri Jul 12 05:59:26 EDT 2002 athena Chris@10: * [project @ 2002-07-12 09:59:26 by athena] Chris@10: Fix Chris@10: Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Fri Jul 12 01:22:38 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 05:22:38 by stevenj] Chris@10: fix comment Chris@10: Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: Chris@10: Fri Jul 12 00:59:29 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 04:59:29 by stevenj] Chris@10: fix in comment Chris@10: Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: Chris@10: Fri Jul 12 00:13:13 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 04:13:13 by stevenj] Chris@10: updated Chris@10: Chris@10: M ./ChangeLog +873 Chris@10: Chris@10: Thu Jul 11 23:39:27 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 03:39:27 by stevenj] Chris@10: buffered malloc's buffers Chris@10: Chris@10: M ./TODO -3 Chris@10: Chris@10: Thu Jul 11 23:30:26 EDT 2002 stevenj Chris@10: * [project @ 2002-07-12 03:30:26 by stevenj] Chris@10: share more code between apply and apply_dit in Rader Chris@10: Chris@10: M ./TODO -2 Chris@10: M ./dft/rader.c -69 +30 Chris@10: Chris@10: Mon Jul 8 12:30:34 EDT 2002 athena Chris@10: * [project @ 2002-07-08 16:30:34 by athena] Chris@10: Polished Chris@10: Chris@10: M ./simd/simd-sse.h -18 +28 Chris@10: M ./simd/simd-sse2.h -25 +31 Chris@10: M ./simd/sse.c -4 +2 Chris@10: M ./simd/sse2.c -3 +2 Chris@10: Chris@10: Mon Jul 8 09:47:11 EDT 2002 athena Chris@10: * [project @ 2002-07-08 13:47:11 by athena] Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Mon Jul 8 09:42:08 EDT 2002 athena Chris@10: * [project @ 2002-07-08 13:42:08 by athena] Chris@10: SIMD/FMA stuff Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: M ./genfft/c.ml -7 +2 Chris@10: M ./genfft/gen_notw_c.ml -2 +3 Chris@10: M ./genfft/simd.ml -2 +13 Chris@10: M ./genfft/to_alist.ml -1 +3 Chris@10: M ./genfft/to_alist.mli -1 +2 Chris@10: M ./simd/simd-sse.h +3 Chris@10: M ./simd/simd-sse2.h +3 Chris@10: M ./support/Makefile.codelets -1 +2 Chris@10: Chris@10: Mon Jul 8 07:43:51 EDT 2002 athena Chris@10: * [project @ 2002-07-08 11:43:51 by athena] Chris@10: Avoid code duplication Chris@10: Chris@10: M ./simd/simd-sse.h -12 +6 Chris@10: Chris@10: Sun Jul 7 20:56:15 EDT 2002 athena Chris@10: * [project @ 2002-07-08 00:56:15 by athena] Chris@10: Fixes for FMA+SIMD Chris@10: Chris@10: M ./genfft/Makefile.am -3 +3 Chris@10: M ./genfft/to_alist.ml -7 +14 Chris@10: Chris@10: Sun Jul 7 20:32:01 EDT 2002 athena Chris@10: * [project @ 2002-07-08 00:32:01 by athena] Chris@10: Major changes in SIMD fftw Chris@10: Chris@10: A ./dft/simd/n1b.c Chris@10: A ./dft/simd/n1b.h Chris@10: A ./dft/simd/n1f.c Chris@10: A ./dft/simd/n1f.h Chris@10: A ./dft/simd/t1b.c Chris@10: A ./dft/simd/t1b.h Chris@10: A ./dft/simd/t1f.c Chris@10: A ./dft/simd/t1f.h Chris@10: A ./genfft/gen_notw_c.ml Chris@10: A ./genfft/gen_twiddle_c.ml Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./dft/codelets/standard/Makefile.am -6 +5 Chris@10: M ./dft/simd/Makefile.am -3 +1 Chris@10: R ./dft/simd/NAMING Chris@10: M ./dft/simd/codelets/Makefile.am -26 +23 Chris@10: M ./dft/simd/n1b.c +43 Chris@10: M ./dft/simd/n1b.h +24 Chris@10: M ./dft/simd/n1f.c +43 Chris@10: M ./dft/simd/n1f.h +24 Chris@10: R ./dft/simd/n2f.c Chris@10: R ./dft/simd/n2f.h Chris@10: R ./dft/simd/n3f.h Chris@10: R ./dft/simd/n4.c Chris@10: R ./dft/simd/n4.h Chris@10: M ./dft/simd/t1b.c +38 Chris@10: M ./dft/simd/t1b.h +25 Chris@10: M ./dft/simd/t1f.c +38 Chris@10: M ./dft/simd/t1f.h +25 Chris@10: R ./dft/simd/t2f.c Chris@10: R ./dft/simd/t2f.h Chris@10: R ./dft/simd/t3f.h Chris@10: R ./dft/simd/t4.c Chris@10: R ./dft/simd/t4.h Chris@10: M ./genfft/Makefile.am -29 +27 Chris@10: M ./genfft/algsimp.ml -16 +27 Chris@10: M ./genfft/annotate.ml -131 +11 Chris@10: M ./genfft/annotate.mli -17 +2 Chris@10: M ./genfft/c.ml -7 +23 Chris@10: M ./genfft/complex.ml -1 +3 Chris@10: M ./genfft/complex.mli -1 +2 Chris@10: M ./genfft/expr.ml -3 +21 Chris@10: M ./genfft/expr.mli -1 +6 Chris@10: M ./genfft/gen_athnotw.ml -3 +3 Chris@10: M ./genfft/gen_athtw.ml -3 +3 Chris@10: M ./genfft/gen_conv.ml -4 +4 Chris@10: M ./genfft/gen_hc2r.ml -4 +4 Chris@10: M ./genfft/gen_notw.ml -4 +4 Chris@10: M ./genfft/gen_notw_c.ml +155 Chris@10: M ./genfft/gen_r2hc.ml -4 +4 Chris@10: M ./genfft/gen_trig.ml -4 +4 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twiddle_c.ml +165 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./genfft/genutil.ml -4 +5 Chris@10: M ./genfft/oracle.ml -1 +2 Chris@10: M ./genfft/simd.ml -177 +95 Chris@10: M ./genfft/simd.mli -2 +2 Chris@10: M ./genfft/simdmagic.ml -47 +1 Chris@10: M ./genfft/to_alist.ml -3 +9 Chris@10: M ./genfft/trig.ml -1 +10 Chris@10: M ./genfft/trig.mli -1 +2 Chris@10: M ./genfft/twiddle.ml -5 +6 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./simd/simd-sse.h -61 +88 Chris@10: M ./simd/simd-sse2.h -67 +57 Chris@10: M ./simd/sse.c -1 +4 Chris@10: M ./simd/sse2.c -1 +4 Chris@10: M ./support/Makefile.codelets -67 +2 Chris@10: Chris@10: Fri Jul 5 17:32:09 EDT 2002 athena Chris@10: * [project @ 2002-07-05 21:32:09 by athena] Chris@10: Use unpck instructions instead of shuffles Chris@10: Chris@10: M ./dft/buffered.c -2 +2 Chris@10: M ./simd/simd-altivec.h -13 +13 Chris@10: M ./simd/simd-sse.h -23 +10 Chris@10: Chris@10: Fri Jul 5 15:49:14 EDT 2002 athena Chris@10: * [project @ 2002-07-05 19:49:14 by athena] Chris@10: Minor tweaks Chris@10: Chris@10: M ./dft/codelets/n.c +1 Chris@10: M ./dft/codelets/t.c +1 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/verify.c -4 +1 Chris@10: M ./kernel/ifftw.h -10 +9 Chris@10: M ./kernel/planner.c -5 +4 Chris@10: M ./tests/bench.c -2 +2 Chris@10: Chris@10: Fri Jul 5 15:02:54 EDT 2002 athena Chris@10: * [project @ 2002-07-05 19:02:54 by athena] Chris@10: Use score planner Chris@10: Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Fri Jul 5 14:49:59 EDT 2002 athena Chris@10: * [project @ 2002-07-05 18:49:59 by athena] Chris@10: Added verifier Chris@10: Chris@10: A ./dft/verify.c Chris@10: A ./kernel/dotens.c Chris@10: A ./kernel/dotens2.c Chris@10: M ./CONVENTIONS +1 Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/dft.h -1 +4 Chris@10: M ./dft/verify.c +428 Chris@10: M ./kernel/Makefile.am -1 +2 Chris@10: M ./kernel/debug.c -2 +2 Chris@10: M ./kernel/dotens.c +47 Chris@10: M ./kernel/dotens2.c +55 Chris@10: M ./kernel/ifftw.h -8 +24 Chris@10: M ./tests/bench.c -8 +14 Chris@10: Chris@10: Wed Jul 3 20:32:28 EDT 2002 athena Chris@10: * [project @ 2002-07-04 00:32:28 by athena] Chris@10: More simd codelets Chris@10: Chris@10: A ./dft/simd/n2f.c Chris@10: A ./dft/simd/n2f.h Chris@10: A ./dft/simd/n3f.h Chris@10: A ./dft/simd/t2f.c Chris@10: A ./dft/simd/t2f.h Chris@10: A ./dft/simd/t3f.h Chris@10: M ./dft/buffered.c -3 +3 Chris@10: M ./dft/codelet.h -1 +3 Chris@10: M ./dft/codelets/n.c +1 Chris@10: M ./dft/codelets/t.c +1 Chris@10: M ./dft/ct-dif.c -1 +6 Chris@10: M ./dft/ct-dit.c -1 +6 Chris@10: M ./dft/ct-ditbuf.c -9 +19 Chris@10: M ./dft/ct.c -11 +17 Chris@10: M ./dft/ct.h -3 +5 Chris@10: M ./dft/simd/Makefile.am -1 +2 Chris@10: M ./dft/simd/NAMING -7 +20 Chris@10: M ./dft/simd/codelets/Makefile.am -4 +27 Chris@10: M ./dft/simd/n2f.c +43 Chris@10: M ./dft/simd/n2f.h +25 Chris@10: M ./dft/simd/n3f.h +1 Chris@10: M ./dft/simd/n4.c -2 +1 Chris@10: M ./dft/simd/t2f.c +40 Chris@10: M ./dft/simd/t2f.h +25 Chris@10: M ./dft/simd/t3f.h +1 Chris@10: M ./dft/simd/t4.c -1 +1 Chris@10: M ./genfft/annotate.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -9 +23 Chris@10: M ./genfft/gen_twiddle.ml -6 +14 Chris@10: M ./genfft/gen_twidsq.ml -3 +9 Chris@10: M ./genfft/simd.ml -43 +69 Chris@10: M ./genfft/simdmagic.ml -11 +36 Chris@10: M ./genfft-k7/gen_notw.ml -1 +2 Chris@10: M ./genfft-k7/gen_twiddle.ml -1 +2 Chris@10: M ./kernel/alloc.c -5 +2 Chris@10: M ./kernel/ifftw.h -3 +9 Chris@10: M ./simd/simd-sse.h -28 +106 Chris@10: M ./simd/simd-sse2.h -24 +47 Chris@10: Chris@10: Tue Jul 2 16:18:09 EDT 2002 athena Chris@10: * [project @ 2002-07-02 20:18:09 by athena] Chris@10: Oops Chris@10: Chris@10: M ./dft/rank-geq2.c -3 +4 Chris@10: Chris@10: Tue Jul 2 16:13:24 EDT 2002 athena Chris@10: * [project @ 2002-07-02 20:13:24 by athena] Chris@10: Fixed classic mode Chris@10: Chris@10: M ./dft/rank-geq2.c -6 +5 Chris@10: M ./dft/vrank-geq1.c -6 +6 Chris@10: Chris@10: Tue Jul 2 15:38:36 EDT 2002 athena Chris@10: * [project @ 2002-07-02 19:38:36 by athena] Chris@10: Use LDK for constants so that we can play games. Chris@10: Chris@10: M ./genfft/simd.ml -2 +2 Chris@10: M ./simd/simd-altivec.h +1 Chris@10: M ./simd/simd-sse.h -5 +5 Chris@10: M ./simd/simd-sse2.h -2 +5 Chris@10: Chris@10: Tue Jul 2 13:15:58 EDT 2002 athena Chris@10: * [project @ 2002-07-02 17:15:58 by athena] Chris@10: Improved support for fixed strides Chris@10: Chris@10: M ./dft/codelet.h -1 +2 Chris@10: M ./dft/codelets/n.c -1 +2 Chris@10: M ./dft/simd/n4.c +3 Chris@10: M ./dft/simd/t4.c +2 Chris@10: M ./genfft/gen_notw.ml -8 +21 Chris@10: M ./genfft/genutil.ml -1 +6 Chris@10: M ./genfft/simd.ml -7 +7 Chris@10: M ./genfft/simd.mli -3 +3 Chris@10: M ./genfft-k7/gen_notw.ml -1 +2 Chris@10: M ./simd/simd-sse.h -17 +54 Chris@10: Chris@10: Tue Jul 2 10:30:58 EDT 2002 athena Chris@10: * [project @ 2002-07-02 14:30:58 by athena] Chris@10: Changed accounting of flops Chris@10: Chris@10: M ./dft/codelet.h -12 +25 Chris@10: M ./dft/codelets/n.c -4 +5 Chris@10: M ./dft/codelets/n.h -6 +2 Chris@10: M ./dft/codelets/t.c -3 +5 Chris@10: M ./dft/codelets/t.h -3 +2 Chris@10: M ./dft/ct-dif.c -3 +5 Chris@10: M ./dft/ct-dit.c -3 +5 Chris@10: M ./dft/ct-ditbuf.c -3 +4 Chris@10: M ./dft/ct-ditf.c -3 +6 Chris@10: M ./dft/direct.c -5 +5 Chris@10: M ./dft/k7/k7.c -14 +23 Chris@10: M ./dft/simd/n4.c -3 +5 Chris@10: M ./dft/simd/n4.h -5 +2 Chris@10: M ./dft/simd/t4.c -3 +5 Chris@10: M ./dft/simd/t4.h -3 +2 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -3 +3 Chris@10: M ./genfft/gen_twidsq.ml -3 +3 Chris@10: M ./genfft-k7/gen_notw.ml -2 +2 Chris@10: M ./genfft-k7/gen_twiddle.ml -2 +2 Chris@10: Chris@10: Tue Jul 2 08:51:38 EDT 2002 athena Chris@10: * [project @ 2002-07-02 12:51:38 by athena] Chris@10: Wrong code in non-fma mode Chris@10: Chris@10: M ./genfft/algsimp.ml -11 +15 Chris@10: M ./genfft-k7/algsimp.ml -8 +24 Chris@10: M ./genfft-k7/to_alist.ml -3 +3 Chris@10: M ./simd/simd-sse2.h -4 +9 Chris@10: Chris@10: Mon Jul 1 23:17:06 EDT 2002 athena Chris@10: * [project @ 2002-07-02 03:17:06 by athena] Chris@10: sse2 stuff Chris@10: Chris@10: A ./simd/simd-sse2.h Chris@10: A ./simd/sse2.c Chris@10: M ./genfft/simdmagic.ml -2 +2 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./simd/Makefile.am -1 +2 Chris@10: M ./simd/simd-sse2.h +125 Chris@10: M ./simd/sse2.c +73 Chris@10: Chris@10: Mon Jul 1 14:05:56 EDT 2002 athena Chris@10: * [project @ 2002-07-01 18:05:56 by athena] Chris@10: Identify CPUs for special codelets Chris@10: Chris@10: A ./simd/sse.c Chris@10: M ./Makefile.am +1 Chris@10: M ./dft/ct.c -4 +4 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/k7/k7.c -5 +43 Chris@10: M ./dft/simd/n4.c -1 +1 Chris@10: M ./dft/simd/t4.c -1 +1 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./simd/Makefile.am -1 +3 Chris@10: M ./simd/simd-altivec.h +2 Chris@10: M ./simd/simd-sse.h +3 Chris@10: M ./simd/sse.c +73 Chris@10: Chris@10: Mon Jul 1 09:26:42 EDT 2002 athena Chris@10: * [project @ 2002-07-01 13:26:42 by athena] Chris@10: Change split problem syntax Chris@10: Chris@10: M ./libbench/problem.c -2 +2 Chris@10: Chris@10: Mon Jul 1 09:11:39 EDT 2002 athena Chris@10: * [project @ 2002-07-01 13:11:39 by athena] Chris@10: Removed -fma flag Chris@10: Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: Chris@10: Sun Jun 30 20:08:26 EDT 2002 athena Chris@10: * [project @ 2002-07-01 00:08:26 by athena] Chris@10: Work around gcc bug Chris@10: Chris@10: M ./simd/simd-altivec.h -3 +17 Chris@10: Chris@10: Sun Jun 30 18:34:06 EDT 2002 athena Chris@10: * [project @ 2002-06-30 22:34:06 by athena] Chris@10: New simd stuff Chris@10: Chris@10: M ./genfft/algsimp.ml -8 +20 Chris@10: M ./genfft/magic.ml -1 +3 Chris@10: M ./genfft/oracle.ml -2 +2 Chris@10: M ./genfft/simd.ml -2 +4 Chris@10: M ./genfft/to_alist.ml -3 +3 Chris@10: Chris@10: Sun Jun 30 17:00:09 EDT 2002 athena Chris@10: * [project @ 2002-06-30 21:00:09 by athena] Chris@10: Added altivec support Chris@10: Chris@10: A ./simd/simd-altivec.h Chris@10: M ./dft/simd/codelets/Makefile.am -1 +1 Chris@10: M ./simd/Makefile.am -1 +1 Chris@10: M ./simd/simd-altivec.h +66 Chris@10: M ./simd/simd-sse.h -1 Chris@10: Chris@10: Sun Jun 30 14:47:47 EDT 2002 athena Chris@10: * [project @ 2002-06-30 18:47:47 by athena] Chris@10: Forgot file Chris@10: Chris@10: A ./dft/simd/t4.c Chris@10: Chris@10: Sun Jun 30 14:37:55 EDT 2002 athena Chris@10: * [project @ 2002-06-30 18:37:55 by athena] Chris@10: Progress towards simd implementation Chris@10: Chris@10: A ./dft/codelets/f.h Chris@10: A ./dft/codelets/n.c Chris@10: A ./dft/codelets/n.h Chris@10: A ./dft/codelets/q.h Chris@10: A ./dft/codelets/t.c Chris@10: A ./dft/codelets/t.h Chris@10: A ./dft/k7/k7.c Chris@10: A ./dft/simd/ Chris@10: A ./dft/simd/Makefile.am Chris@10: A ./dft/simd/NAMING Chris@10: A ./dft/simd/codelets/ Chris@10: A ./dft/simd/codelets/Makefile.am Chris@10: A ./dft/simd/n4.c Chris@10: A ./dft/simd/n4.h Chris@10: A ./dft/simd/t4.h Chris@10: A ./simd/ Chris@10: A ./simd/Makefile.am Chris@10: A ./simd/README Chris@10: A ./simd/simd-sse.h Chris@10: A ./simd/simd.h Chris@10: M ./Makefile.am -2 +5 Chris@10: M ./configure.ac -6 +32 Chris@10: M ./dft/Makefile.am -3 +3 Chris@10: R ./dft/codelet-k7.h Chris@10: M ./dft/codelet.h -11 +23 Chris@10: M ./dft/codelets/Makefile.am +4 Chris@10: M ./dft/codelets/f.h +1 Chris@10: M ./dft/codelets/inplace/Makefile.am -13 +14 Chris@10: M ./dft/codelets/n.c +34 Chris@10: M ./dft/codelets/n.h +27 Chris@10: M ./dft/codelets/q.h +1 Chris@10: M ./dft/codelets/standard/Makefile.am -13 +14 Chris@10: M ./dft/codelets/t.c +32 Chris@10: M ./dft/codelets/t.h +24 Chris@10: M ./dft/conf.c -2 +5 Chris@10: M ./dft/ct-dif.c -3 +3 Chris@10: M ./dft/ct-dit.c -5 +3 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/ct-ditf.c -4 +3 Chris@10: M ./dft/ct.c -6 +6 Chris@10: M ./dft/ct.h -3 +1 Chris@10: M ./dft/direct.c -6 +11 Chris@10: M ./dft/k7/Makefile.am -2 +1 Chris@10: M ./dft/k7/codelets/Makefile.am -33 +29 Chris@10: R ./dft/k7/ct-dif.c Chris@10: R ./dft/k7/ct-dit.c Chris@10: R ./dft/k7/ct-ditbuf.c Chris@10: R ./dft/k7/direct.c Chris@10: M ./dft/k7/k7.c +69 Chris@10: R ./dft/k7/kdft-dif.c Chris@10: R ./dft/k7/kdft-dit.c Chris@10: R ./dft/k7/kdft.c Chris@10: M ./dft/simd/Makefile.am +6 Chris@10: M ./dft/simd/NAMING +17 Chris@10: M ./dft/simd/codelets/Makefile.am +36 Chris@10: M ./dft/simd/n4.c +39 Chris@10: M ./dft/simd/n4.h +28 Chris@10: M ./dft/simd/t4.h +26 Chris@10: M ./genfft/annotate.ml -7 +7 Chris@10: M ./genfft/c.ml -1 +6 Chris@10: M ./genfft/c.mli -1 +2 Chris@10: M ./genfft/gen_notw.ml -12 +20 Chris@10: M ./genfft/gen_twiddle.ml -14 +18 Chris@10: M ./genfft/gen_twidsq.ml -7 +7 Chris@10: M ./genfft/genutil.ml -4 +17 Chris@10: M ./genfft/magic.ml -1 +3 Chris@10: M ./genfft/simd.ml -66 +34 Chris@10: M ./genfft/simd.mli -1 +7 Chris@10: M ./genfft/simdmagic.ml -2 +12 Chris@10: M ./genfft/twiddle.ml -1 +12 Chris@10: M ./genfft/twiddle.mli -1 +2 Chris@10: M ./genfft-k7/genUtil.ml -1 +1 Chris@10: M ./genfft-k7/gen_notw.ml -11 +18 Chris@10: M ./genfft-k7/gen_twiddle.ml -12 +17 Chris@10: M ./kernel/alloc.c -9 +28 Chris@10: M ./kernel/ifftw.h -10 +10 Chris@10: M ./libbench/bench-user.h -1 +2 Chris@10: M ./libbench/problem.c -1 +3 Chris@10: M ./libbench/util.c -1 Chris@10: M ./simd/Makefile.am +1 Chris@10: M ./simd/README +2 Chris@10: M ./simd/simd-sse.h +94 Chris@10: M ./simd/simd.h +33 Chris@10: M ./support/Makefile.codelets -33 +36 Chris@10: M ./tests/Makefile.am +1 Chris@10: M ./tests/bench.c -14 +63 Chris@10: Chris@10: Tue Jun 25 20:23:29 EDT 2002 athena Chris@10: * [project @ 2002-06-26 00:23:29 by athena] Chris@10: Add 128- codelet Chris@10: Chris@10: M ./dft/k7/codelets/Makefile.am -2 +3 Chris@10: Chris@10: Sat Jun 22 20:47:28 EDT 2002 athena Chris@10: * [project @ 2002-06-23 00:47:28 by athena] Chris@10: More simd changes. Ensure proper stack alignment in k7 codelets. Chris@10: Chris@10: M ./configure.ac +1 Chris@10: M ./genfft/c.ml -25 +4 Chris@10: M ./genfft/c.mli -2 +4 Chris@10: M ./genfft/expr.ml -1 +18 Chris@10: M ./genfft/expr.mli -1 +4 Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -3 +3 Chris@10: M ./genfft/gen_trig.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -3 +3 Chris@10: M ./genfft/gen_twidsq.ml -3 +3 Chris@10: M ./genfft/genutil.ml -1 +13 Chris@10: M ./genfft/simd.ml -1 +9 Chris@10: M ./genfft/simd.mli -1 +2 Chris@10: M ./genfft-k7/genUtil.ml -34 +6 Chris@10: M ./genfft-k7/gen_notw.ml -2 +2 Chris@10: M ./genfft-k7/gen_twiddle.ml -2 +2 Chris@10: M ./libbench/bench-main.c -1 +11 Chris@10: Chris@10: Sat Jun 22 13:01:33 EDT 2002 athena Chris@10: * [project @ 2002-06-22 17:01:33 by athena] Chris@10: Fixed prototypes Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/solvtab.c -2 +2 Chris@10: Chris@10: Sat Jun 22 12:53:26 EDT 2002 athena Chris@10: * [project @ 2002-06-22 16:53:26 by athena] Chris@10: Sparc cycle counter requires v9 Chris@10: Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: Chris@10: Sat Jun 22 11:45:48 EDT 2002 athena Chris@10: * [project @ 2002-06-22 15:45:48 by athena] Chris@10: Minor fixes Chris@10: Chris@10: M ./configure.ac -1 +1 Chris@10: M ./kernel/cycle.h -15 +36 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: Chris@10: Sat Jun 22 11:11:46 EDT 2002 athena Chris@10: * [project @ 2002-06-22 15:11:46 by athena] Chris@10: Fixed ev67 detection Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Sat Jun 22 10:19:11 EDT 2002 athena Chris@10: * [project @ 2002-06-22 14:19:11 by athena] Chris@10: Print flops Chris@10: Chris@10: M ./tests/bench.c +2 Chris@10: Chris@10: Sat Jun 22 09:36:47 EDT 2002 athena Chris@10: * [project @ 2002-06-22 13:36:47 by athena] Chris@10: Nothing really Chris@10: Chris@10: M ./genfft/simd.ml -7 +3 Chris@10: Chris@10: Fri Jun 21 22:19:20 EDT 2002 athena Chris@10: * [project @ 2002-06-22 02:19:20 by athena] Chris@10: More simd work Chris@10: Chris@10: M ./dft/codelet-k7.h -2 +2 Chris@10: M ./dft/codelet.h -3 +3 Chris@10: M ./genfft/Makefile.am -2 +2 Chris@10: M ./genfft/c.ml -19 +1 Chris@10: M ./genfft/c.mli -2 +2 Chris@10: M ./genfft/gen_conv.ml -3 +3 Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -3 +3 Chris@10: M ./genfft/gen_trig.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -3 +3 Chris@10: M ./genfft/gen_twidsq.ml -3 +3 Chris@10: M ./genfft/genutil.ml -2 +20 Chris@10: M ./genfft/magic.ml -1 +5 Chris@10: M ./genfft/simd.ml -47 +39 Chris@10: M ./genfft/simd.mli -1 +3 Chris@10: M ./genfft/simdmagic.ml -4 +2 Chris@10: M ./genfft/to_alist.ml -2 +5 Chris@10: M ./genfft-k7/magic.ml +4 Chris@10: M ./genfft-k7/to_alist.ml -2 +5 Chris@10: M ./genfft-k7/to_alist.mli -1 +1 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: Chris@10: Thu Jun 20 21:22:41 EDT 2002 athena Chris@10: * [project @ 2002-06-21 01:22:41 by athena] Chris@10: More simd work Chris@10: Chris@10: A ./genfft/simd.mli Chris@10: A ./genfft/simdmagic.ml Chris@10: M ./genfft/Makefile.am -7 +8 Chris@10: M ./genfft/annotate.ml -13 +13 Chris@10: M ./genfft/annotate.mli -1 +2 Chris@10: M ./genfft/magic.ml -21 +1 Chris@10: M ./genfft/simd.ml -21 +10 Chris@10: M ./genfft/simd.mli +21 Chris@10: M ./genfft/simdmagic.ml +43 Chris@10: Chris@10: Thu Jun 20 18:51:33 EDT 2002 athena Chris@10: * [project @ 2002-06-20 22:51:33 by athena] Chris@10: More simd work Chris@10: Chris@10: A ./genfft/simd.ml Chris@10: M ./genfft/Makefile.am -23 +7 Chris@10: M ./genfft/annotate.ml -18 +2 Chris@10: M ./genfft/c.ml -6 +1 Chris@10: M ./genfft/c.mli -5 +8 Chris@10: M ./genfft/magic.ml -1 +6 Chris@10: M ./genfft/simd.ml +288 Chris@10: M ./genfft/variable.ml -1 +18 Chris@10: M ./genfft/variable.mli -1 +4 Chris@10: Chris@10: Thu Jun 20 15:04:37 EDT 2002 athena Chris@10: * [project @ 2002-06-20 19:04:37 by athena] Chris@10: Moving towards incorporation of simd stuff Chris@10: Chris@10: M ./genfft/annotate.ml -11 +144 Chris@10: M ./genfft/annotate.mli -2 +16 Chris@10: M ./genfft/gen_athnotw.ml -5 +5 Chris@10: M ./genfft/gen_athtw.ml -6 +6 Chris@10: M ./genfft/gen_conv.ml -3 +3 Chris@10: M ./genfft/gen_hc2r.ml -3 +3 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_r2hc.ml -3 +3 Chris@10: M ./genfft/gen_trig.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -6 +5 Chris@10: M ./genfft/genutil.ml -26 +17 Chris@10: M ./genfft/magic.ml -2 +17 Chris@10: M ./genfft/twiddle.ml -4 +4 Chris@10: M ./genfft/twiddle.mli -2 +2 Chris@10: M ./genfft/variable.ml -12 +22 Chris@10: M ./genfft/variable.mli -3 +6 Chris@10: Chris@10: Wed Jun 19 18:47:55 EDT 2002 athena Chris@10: * [project @ 2002-06-19 22:47:55 by athena] Chris@10: Reorganized k7 stuff into own directory Chris@10: Chris@10: A ./dft/k7/ Chris@10: A ./dft/k7/Makefile.am Chris@10: A ./dft/k7/codelets/ Chris@10: A ./dft/k7/codelets/Makefile.am Chris@10: A ./dft/k7/ct-dif.c Chris@10: A ./dft/k7/ct-dit.c Chris@10: A ./dft/k7/ct-ditbuf.c Chris@10: A ./dft/k7/direct.c Chris@10: A ./dft/k7/kdft-dif.c Chris@10: A ./dft/k7/kdft-dit.c Chris@10: A ./dft/k7/kdft.c Chris@10: M ./Makefile.am -1 +2 Chris@10: M ./configure.ac -2 +3 Chris@10: M ./dft/Makefile.am -6 +5 Chris@10: M ./dft/codelets/Makefile.am -1 +1 Chris@10: R ./dft/ct-dif-k7.c Chris@10: R ./dft/ct-dit-k7.c Chris@10: R ./dft/ct-ditbuf-k7.c Chris@10: R ./dft/direct-k7.c Chris@10: M ./dft/k7/Makefile.am +7 Chris@10: M ./dft/k7/codelets/Makefile.am +74 Chris@10: M ./dft/k7/ct-dif.c +129 Chris@10: M ./dft/k7/ct-dit.c +137 Chris@10: M ./dft/k7/ct-ditbuf.c +176 Chris@10: M ./dft/k7/direct.c +172 Chris@10: M ./dft/k7/kdft-dif.c +30 Chris@10: M ./dft/k7/kdft-dit.c +31 Chris@10: M ./dft/k7/kdft.c +31 Chris@10: R ./dft/kdft-dif-k7.c Chris@10: R ./dft/kdft-dit-k7.c Chris@10: R ./dft/kdft-k7.c Chris@10: Chris@10: Wed Jun 19 13:21:13 EDT 2002 athena Chris@10: * [project @ 2002-06-19 17:20:37 by athena] Chris@10: Minor experimental stuff Chris@10: Chris@10: M ./genfft/expr.ml -3 +2 Chris@10: M ./genfft/expr.mli -2 +2 Chris@10: M ./genfft/genutil.ml -2 +11 Chris@10: M ./genfft/magic.ml -1 +3 Chris@10: M ./genfft-k7/expr.ml -1 +3 Chris@10: M ./genfft-k7/expr.mli -1 +3 Chris@10: Chris@10: Wed Jun 19 11:20:29 EDT 2002 athena Chris@10: * [project @ 2002-06-19 15:20:29 by athena] Chris@10: Cosmetic changes Chris@10: Chris@10: M ./genfft/expr.ml -1 +4 Chris@10: M ./genfft/expr.mli -1 +3 Chris@10: M ./genfft/genutil.ml -9 +2 Chris@10: Chris@10: Wed Jun 19 01:43:31 EDT 2002 fftw Chris@10: * [project @ 2002-06-19 05:43:31 by fftw] Chris@10: allocate buffers on the fly Chris@10: Chris@10: M ./dft/buffered.c -18 +7 Chris@10: M ./dft/rader.c -17 +20 Chris@10: Chris@10: Tue Jun 18 17:48:41 EDT 2002 athena Chris@10: * [project @ 2002-06-18 21:48:41 by athena] Chris@10: Added ct-ditbuf-k7.c . Major changes required in generator. Chris@10: Chris@10: A ./dft/ct-ditbuf-k7.c Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/codelet-k7.h -1 +2 Chris@10: M ./dft/ct-dif-k7.c -17 +3 Chris@10: M ./dft/ct-dif.c -16 +2 Chris@10: M ./dft/ct-dit-k7.c -18 +10 Chris@10: M ./dft/ct-dit.c -16 +2 Chris@10: M ./dft/ct-ditbuf-k7.c +176 Chris@10: M ./dft/ct-ditbuf.c -16 +2 Chris@10: M ./dft/ct.c -1 +30 Chris@10: M ./dft/ct.h -1 +3 Chris@10: M ./dft/kdft-dit-k7.c -1 +2 Chris@10: M ./dft/rader.c -1 +1 Chris@10: M ./genfft/number.ml -2 +2 Chris@10: M ./genfft/to_alist.ml -2 +2 Chris@10: M ./genfft-k7/Makefile.am -4 +4 Chris@10: M ./genfft-k7/assignmentsToVfpinstrs.ml -1 +4 Chris@10: M ./genfft-k7/gen_twiddle.ml -2 +5 Chris@10: M ./genfft-k7/k7Basics.ml -1 +1 Chris@10: M ./genfft-k7/k7Basics.mli +1 Chris@10: M ./genfft-k7/k7RegisterAllocationBasics.ml +3 Chris@10: M ./genfft-k7/k7RegisterAllocationBasics.mli -2 +4 Chris@10: M ./genfft-k7/k7RegisterAllocator.ml -15 +28 Chris@10: M ./genfft-k7/k7RegisterAllocatorInit.ml -6 +12 Chris@10: M ./genfft-k7/number.ml -2 +2 Chris@10: M ./genfft-k7/to_alist.ml -2 +2 Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner.c -3 +5 Chris@10: M ./kernel/primes.c -2 +2 Chris@10: Chris@10: Tue Jun 18 11:55:57 EDT 2002 athena Chris@10: * [project @ 2002-06-18 15:55:57 by athena] Chris@10: Nothing, really Chris@10: Chris@10: M ./genfft-k7/gen_twiddle.ml -2 +2 Chris@10: M ./kernel/ifftw.h -18 +19 Chris@10: Chris@10: Tue Jun 18 11:19:59 EDT 2002 athena Chris@10: * [project @ 2002-06-18 15:19:59 by athena] Chris@10: !SINGLE ==> !K7_MODE (for some reason the contrapositive sounds wrong) Chris@10: Chris@10: M ./configure.ac +1 Chris@10: Chris@10: Tue Jun 18 11:07:13 EDT 2002 athena Chris@10: * [project @ 2002-06-18 15:07:13 by athena] Chris@10: Buffer is now symmetric wrt forward/backward transform Chris@10: Chris@10: M ./dft/buffered.c -12 +22 Chris@10: Chris@10: Tue Jun 18 10:33:58 EDT 2002 athena Chris@10: * [project @ 2002-06-18 14:33:58 by athena] Chris@10: Fixed applicable() in indirect.c Chris@10: Chris@10: A ./kernel/debug.c Chris@10: M ./dft/ct-dif.c -2 +1 Chris@10: M ./dft/indirect.c -3 +3 Chris@10: M ./dft/vrank2-transpose.c -10 +14 Chris@10: M ./dft/vrank3-transpose.c -10 +13 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/debug.c +39 Chris@10: M ./kernel/ifftw.h -1 +5 Chris@10: M ./kernel/print.c -4 +10 Chris@10: Chris@10: Tue Jun 18 08:41:18 EDT 2002 athena Chris@10: * [project @ 2002-06-18 12:41:18 by athena] Chris@10: Fixed attempt to free() uninitialized pointer. Chris@10: Chris@10: M ./dft/rader.c -2 +3 Chris@10: Chris@10: Tue Jun 18 08:13:55 EDT 2002 athena Chris@10: * [project @ 2002-06-18 12:13:55 by athena] Chris@10: Added reference counts for awake() Chris@10: Chris@10: M ./CONVENTIONS -1 +1 Chris@10: M ./TODO +5 Chris@10: M ./dft/rader.c -11 +3 Chris@10: M ./kernel/ifftw.h -4 +7 Chris@10: M ./kernel/plan.c -2 +15 Chris@10: M ./kernel/planner.c -3 +3 Chris@10: M ./tests/bench.c -2 +3 Chris@10: Chris@10: Mon Jun 17 20:49:05 EDT 2002 stevenj Chris@10: * [project @ 2002-06-18 00:49:05 by stevenj] Chris@10: updated comment Chris@10: Chris@10: M ./dft/rader.c -2 +2 Chris@10: Chris@10: Mon Jun 17 20:46:45 EDT 2002 stevenj Chris@10: * [project @ 2002-06-18 00:46:45 by stevenj] Chris@10: slight update Chris@10: Chris@10: M ./TODO -1 +3 Chris@10: Chris@10: Mon Jun 17 02:30:16 EDT 2002 fftw Chris@10: * [project @ 2002-06-17 06:30:16 by fftw] Chris@10: moved prime-number stuff into primes.c, so it can be shared with generic codelet and with rfftw rader Chris@10: Chris@10: A ./kernel/primes.c Chris@10: M ./dft/rader.c -104 +5 Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -1 +24 Chris@10: M ./kernel/primes.c +112 Chris@10: Chris@10: Mon Jun 17 02:01:58 EDT 2002 fftw Chris@10: * [project @ 2002-06-17 06:01:58 by fftw] Chris@10: added comment Chris@10: Chris@10: M ./dft/rader.c -6 +9 Chris@10: Chris@10: Mon Jun 17 01:39:55 EDT 2002 fftw Chris@10: * [project @ 2002-06-17 05:39:55 by fftw] Chris@10: added rader-dit Chris@10: Chris@10: M ./dft/rader.c -66 +308 Chris@10: M ./kernel/ifftw.h -1 +10 Chris@10: M ./kernel/twiddle.c -8 +1 Chris@10: Chris@10: Sun Jun 16 23:50:16 EDT 2002 fftw Chris@10: * [project @ 2002-06-17 03:50:16 by fftw] Chris@10: added initial Rader (no DIT yet) Chris@10: Chris@10: A ./dft/rader.c Chris@10: M ./configure.ac +8 Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/rader.c +457 Chris@10: Chris@10: Sun Jun 16 22:29:51 EDT 2002 fftw Chris@10: * [project @ 2002-06-17 02:29:51 by fftw] Chris@10: don't warn about long long Chris@10: Chris@10: M ./acinclude.m4 -1 +1 Chris@10: Chris@10: Sun Jun 16 21:30:42 EDT 2002 athena Chris@10: * [project @ 2002-06-17 01:30:42 by athena] Chris@10: Added k7 DIF codelets Chris@10: Chris@10: A ./dft/ct-dif-k7.c Chris@10: A ./dft/kdft-dif-k7.c Chris@10: M ./dft/Makefile.am -5 +6 Chris@10: M ./dft/codelet-k7.h -5 +10 Chris@10: M ./dft/ct-dif-k7.c +143 Chris@10: M ./dft/ct.h -1 +2 Chris@10: M ./dft/kdft-dif-k7.c +30 Chris@10: M ./genfft-k7/gen_twiddle.ml -4 +25 Chris@10: M ./kernel/planner.c -3 +2 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Sun Jun 16 19:13:31 EDT 2002 athena Chris@10: * [project @ 2002-06-16 23:13:31 by athena] Chris@10: Added stuff to do Chris@10: Chris@10: M ./TODO +6 Chris@10: Chris@10: Sun Jun 16 19:05:58 EDT 2002 athena Chris@10: * [project @ 2002-06-16 23:05:58 by athena] Chris@10: Handle dual case R = I + 1 Chris@10: Chris@10: M ./dft/rank0.c -6 +30 Chris@10: Chris@10: Sun Jun 16 18:54:31 EDT 2002 athena Chris@10: * [project @ 2002-06-16 22:54:31 by athena] Chris@10: Removed useless flag Chris@10: Chris@10: M ./bootstrap.sh -1 +1 Chris@10: Chris@10: Sun Jun 16 18:30:32 EDT 2002 athena Chris@10: * [project @ 2002-06-16 22:30:32 by athena] Chris@10: Removed useless file Chris@10: Chris@10: R ./mkdist.sh Chris@10: Chris@10: Sun Jun 16 18:30:18 EDT 2002 athena Chris@10: * [project @ 2002-06-16 22:30:18 by athena] Chris@10: More k7 work. Switched to runtime CLASSIC mode. Chris@10: Chris@10: A ./dft/ct-dit-k7.c Chris@10: A ./dft/kdft-dit-k7.c Chris@10: R ./CLASSIC-MODE Chris@10: M ./Makefile.am -10 +1 Chris@10: M ./configure.ac -21 Chris@10: M ./dft/Makefile.am -5 +5 Chris@10: M ./dft/buffered.c -2 +4 Chris@10: M ./dft/codelet-k7.h -5 +7 Chris@10: M ./dft/codelet.h -1 +2 Chris@10: M ./dft/codelets/inplace/Makefile.am -2 Chris@10: M ./dft/ct-dif.c -3 +4 Chris@10: M ./dft/ct-dit-k7.c +145 Chris@10: M ./dft/ct-dit.c -6 +9 Chris@10: M ./dft/ct-ditbuf.c -9 +11 Chris@10: M ./dft/ct-ditf.c -2 +3 Chris@10: M ./dft/ct.h -5 +3 Chris@10: M ./dft/direct-k7.c -2 +3 Chris@10: M ./dft/direct.c -2 +3 Chris@10: M ./dft/indirect.c -2 +3 Chris@10: M ./dft/kdft-dit-k7.c +30 Chris@10: M ./dft/kdft-dit.c -4 +2 Chris@10: M ./dft/nop.c -2 +3 Chris@10: M ./dft/rank-geq2.c -8 +8 Chris@10: M ./dft/rank0.c -2 +3 Chris@10: M ./dft/vrank-geq1.c -6 +7 Chris@10: M ./dft/vrank2-transpose.c -3 +3 Chris@10: M ./dft/vrank3-transpose.c -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./genfft/twiddle.ml -7 +28 Chris@10: M ./genfft/twiddle.mli -2 +5 Chris@10: M ./genfft-k7/Makefile.am -25 +24 Chris@10: M ./genfft-k7/genUtil.ml -2 +3 Chris@10: M ./genfft-k7/gen_notw.ml -1 +2 Chris@10: M ./genfft-k7/gen_twiddle.ml -23 +29 Chris@10: M ./genfft-k7/twiddle.ml -7 +36 Chris@10: M ./genfft-k7/twiddle.mli -3 +7 Chris@10: M ./kernel/ifftw.h -15 +13 Chris@10: M ./kernel/planner-naive.c -3 +3 Chris@10: M ./kernel/planner-score.c -5 +6 Chris@10: M ./kernel/planner.c -6 +5 Chris@10: M ./kernel/version.c -7 +2 Chris@10: M ./support/Makefile.codelets +1 Chris@10: M ./tests/bench.c -12 +2 Chris@10: Chris@10: Sun Jun 16 17:15:18 EDT 2002 stevenj Chris@10: * [project @ 2002-06-16 21:15:18 by stevenj] Chris@10: spelling Chris@10: Chris@10: M ./kernel/tensor.c -2 +2 Chris@10: Chris@10: Sun Jun 16 15:51:44 EDT 2002 athena Chris@10: * [project @ 2002-06-16 19:51:44 by athena] Chris@10: Do not compile if not K7_MODE Chris@10: Chris@10: M ./dft/kdft-k7.c -1 +3 Chris@10: Chris@10: Sun Jun 16 15:35:02 EDT 2002 athena Chris@10: * [project @ 2002-06-16 19:35:02 by athena] Chris@10: Do not require K7 definitions to compile Chris@10: Chris@10: M ./dft/codelet-k7.h -1 +2 Chris@10: M ./dft/dft.h -2 +1 Chris@10: Chris@10: Sun Jun 16 08:05:17 EDT 2002 athena Chris@10: * [project @ 2002-06-16 12:05:17 by athena] Chris@10: More k7 stuff Chris@10: Chris@10: A ./dft/codelet-k7.h Chris@10: A ./genfft-k7/gen_twiddle.ml Chris@10: M ./dft/Makefile.am -4 +4 Chris@10: M ./dft/codelet-k7.h +40 Chris@10: M ./dft/codelet.h -20 +5 Chris@10: M ./dft/direct-k7.c -1 +5 Chris@10: M ./genfft-k7/Makefile.am -1 +9 Chris@10: M ./genfft-k7/complex.ml -1 +2 Chris@10: M ./genfft-k7/complex.mli -1 +2 Chris@10: M ./genfft-k7/genUtil.ml -1 +12 Chris@10: M ./genfft-k7/gen_notw.ml -5 +1 Chris@10: M ./genfft-k7/gen_twiddle.ml +149 Chris@10: Chris@10: Sat Jun 15 18:30:43 EDT 2002 athena Chris@10: * [project @ 2002-06-15 22:30:43 by athena] Chris@10: Try to be compatible with automake-1.6 Chris@10: Chris@10: M ./acinclude.m4 -1 +5 Chris@10: Chris@10: Sat Jun 15 18:23:40 EDT 2002 athena Chris@10: * [project @ 2002-06-15 22:23:40 by athena] Chris@10: More merging of Stefan's generator with main genfft branch Chris@10: Chris@10: A ./genfft-k7/algsimp.ml Chris@10: A ./genfft-k7/algsimp.mli Chris@10: A ./genfft-k7/assoctable.ml Chris@10: A ./genfft-k7/assoctable.mli Chris@10: A ./genfft-k7/littlesimp.ml Chris@10: A ./genfft-k7/littlesimp.mli Chris@10: A ./genfft-k7/monads.ml Chris@10: A ./genfft-k7/oracle.ml Chris@10: A ./genfft-k7/oracle.mli Chris@10: A ./genfft-k7/to_alist.ml Chris@10: A ./genfft-k7/to_alist.mli Chris@10: A ./genfft-k7/twiddle.mli Chris@10: A ./support/codelet_asmprelude Chris@10: M ./acinclude.m4 +8 Chris@10: M ./configure.ac +1 Chris@10: M ./genfft/number.ml -7 +8 Chris@10: M ./genfft-k7/Makefile.am -23 +27 Chris@10: M ./genfft-k7/algsimp.ml +517 Chris@10: M ./genfft-k7/algsimp.mli +24 Chris@10: M ./genfft-k7/assignmentsToVfpinstrs.ml -11 +10 Chris@10: M ./genfft-k7/assoctable.ml +66 Chris@10: M ./genfft-k7/assoctable.mli +30 Chris@10: M ./genfft-k7/complex.ml -4 +19 Chris@10: M ./genfft-k7/complex.mli -5 +6 Chris@10: M ./genfft-k7/expr.ml -27 +58 Chris@10: M ./genfft-k7/expr.mli -4 +11 Chris@10: R ./genfft-k7/exprdag.ml Chris@10: R ./genfft-k7/exprdag.mli Chris@10: M ./genfft-k7/genUtil.ml -13 +18 Chris@10: M ./genfft-k7/gen_notw.ml -16 +31 Chris@10: M ./genfft-k7/k7Unparsing.ml -2 +2 Chris@10: M ./genfft-k7/littlesimp.ml +72 Chris@10: M ./genfft-k7/littlesimp.mli +26 Chris@10: M ./genfft-k7/magic.ml -16 +41 Chris@10: M ./genfft-k7/monads.ml +76 Chris@10: M ./genfft-k7/number.ml -9 +5 Chris@10: M ./genfft-k7/number.mli -8 +4 Chris@10: M ./genfft-k7/oracle.ml +138 Chris@10: M ./genfft-k7/oracle.mli +25 Chris@10: M ./genfft-k7/to_alist.ml +258 Chris@10: M ./genfft-k7/to_alist.mli +24 Chris@10: M ./genfft-k7/twiddle.ml -5 +22 Chris@10: M ./genfft-k7/twiddle.mli +29 Chris@10: M ./genfft-k7/util.ml -4 +4 Chris@10: M ./genfft-k7/util.mli -2 +1 Chris@10: M ./genfft-k7/vFpUnparsing.ml -1 +1 Chris@10: M ./genfft-k7/vSimdBasics.ml -2 +3 Chris@10: M ./genfft-k7/vSimdUnparsing.ml -1 +1 Chris@10: M ./genfft-k7/variable.ml +1 Chris@10: M ./genfft-k7/variable.mli +1 Chris@10: M ./support/Makefile.am -1 +1 Chris@10: M ./support/Makefile.codelets -1 +13 Chris@10: M ./support/codelet_asmprelude +8 Chris@10: Chris@10: Sat Jun 15 13:51:39 EDT 2002 athena Chris@10: * [project @ 2002-06-15 17:51:39 by athena] Chris@10: Slowly merging genfft-k7 with main genfft branch Chris@10: Chris@10: A ./genfft-k7/gen_notw.ml Chris@10: M ./genfft/expr.ml -1 +3 Chris@10: M ./genfft/expr.mli -1 +2 Chris@10: M ./genfft/genutil.ml -1 +9 Chris@10: M ./genfft-k7/Makefile.am -26 +23 Chris@10: M ./genfft-k7/complex.ml -34 +92 Chris@10: M ./genfft-k7/complex.mli -1 +11 Chris@10: M ./genfft-k7/expr.ml -1 +19 Chris@10: M ./genfft-k7/expr.mli -4 +3 Chris@10: M ./genfft-k7/exprdag.ml -73 +103 Chris@10: M ./genfft-k7/exprdag.mli -1 Chris@10: M ./genfft-k7/fft.ml -167 +83 Chris@10: M ./genfft-k7/fft.mli -29 +1 Chris@10: M ./genfft-k7/genUtil.ml -56 +33 Chris@10: R ./genfft-k7/gen_hc2hc.ml Chris@10: R ./genfft-k7/gen_hc2real.ml Chris@10: M ./genfft-k7/gen_notw.ml +143 Chris@10: R ./genfft-k7/gen_notwiddle.ml Chris@10: R ./genfft-k7/gen_notwiddle_fixedstride.ml Chris@10: R ./genfft-k7/gen_real2hc.ml Chris@10: R ./genfft-k7/gen_realeven.ml Chris@10: R ./genfft-k7/gen_realeven2.ml Chris@10: R ./genfft-k7/gen_realodd.ml Chris@10: R ./genfft-k7/gen_realodd2.ml Chris@10: R ./genfft-k7/gen_twiddle.ml Chris@10: M ./genfft-k7/magic.ml -11 +35 Chris@10: R ./genfft-k7/symmetry.ml Chris@10: M ./genfft-k7/twiddle.ml -63 +95 Chris@10: M ./genfft-k7/util.ml +15 Chris@10: M ./genfft-k7/util.mli +3 Chris@10: M ./genfft-k7/variable.ml +12 Chris@10: M ./genfft-k7/variable.mli +3 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Fri Jun 14 21:33:02 EDT 2002 athena Chris@10: * [project @ 2002-06-15 01:33:02 by athena] Chris@10: Fixed, really Chris@10: Chris@10: M ./genfft-k7/Makefile.am -1 +1 Chris@10: M ./genfft-k7/genUtil.ml -25 Chris@10: M ./genfft-k7/magic.ml -1 +1 Chris@10: R ./genfft-k7/magic.mli Chris@10: M ./genfft-k7/twiddle.ml -2 +2 Chris@10: M ./support/Makefile.codelets -4 +8 Chris@10: Chris@10: Fri Jun 14 21:27:12 EDT 2002 athena Chris@10: * [project @ 2002-06-15 01:27:12 by athena] Chris@10: Oops... Chris@10: Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Fri Jun 14 21:25:34 EDT 2002 athena Chris@10: * [project @ 2002-06-15 01:25:34 by athena] Chris@10: Work properly when $(ALL_CODELETS) = "" Chris@10: Chris@10: M ./support/Makefile.codelets -4 +4 Chris@10: Chris@10: Fri Jun 14 21:11:16 EDT 2002 athena Chris@10: * [project @ 2002-06-15 01:11:16 by athena] Chris@10: Fixed k7 build machinery Chris@10: Chris@10: M ./Makefile.am -1 +2 Chris@10: M ./configure.ac +8 Chris@10: M ./dft/codelet.h -1 +4 Chris@10: M ./dft/codelets/Makefile.am -1 +1 Chris@10: M ./dft/conf.c -1 +4 Chris@10: M ./genfft-k7/gen_notwiddle.ml -1 +3 Chris@10: M ./kernel/ifftw.h -1 +10 Chris@10: M ./support/Makefile.codelets +2 Chris@10: Chris@10: Fri Jun 14 17:42:35 EDT 2002 athena Chris@10: * [project @ 2002-06-14 21:42:35 by athena] Chris@10: More work on k7 stuff Chris@10: Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +1 Chris@10: M ./dft/codelet.h -1 +4 Chris@10: M ./dft/direct-k7.c -3 +3 Chris@10: M ./genfft-k7/Makefile.am -24 +24 Chris@10: R ./genfft-k7/codeletMisc.ml Chris@10: R ./genfft-k7/codeletMisc.mli Chris@10: M ./genfft-k7/genUtil.ml -10 +128 Chris@10: R ./genfft-k7/genUtil.mli Chris@10: R ./genfft-k7/gen_hc2hc.mli Chris@10: R ./genfft-k7/gen_hc2real.mli Chris@10: M ./genfft-k7/gen_notwiddle.ml -3 +33 Chris@10: R ./genfft-k7/gen_notwiddle.mli Chris@10: R ./genfft-k7/gen_real2hc.mli Chris@10: R ./genfft-k7/gen_realeven.mli Chris@10: R ./genfft-k7/gen_realeven2.mli Chris@10: R ./genfft-k7/gen_realodd.mli Chris@10: R ./genfft-k7/gen_realodd2.mli Chris@10: R ./genfft-k7/gen_twiddle.mli Chris@10: R ./genfft-k7/genfft.ml Chris@10: M ./genfft-k7/k7Basics.ml +19 Chris@10: M ./genfft-k7/k7Basics.mli +2 Chris@10: M ./genfft-k7/k7Unparsing.ml -1 +1 Chris@10: M ./genfft-k7/magic.ml -1 Chris@10: M ./genfft-k7/magic.mli -1 Chris@10: M ./kernel/ifftw.h -1 +4 Chris@10: M ./libbench/bench-user.h -1 +3 Chris@10: M ./support/Makefile.am -1 +2 Chris@10: Chris@10: Fri Jun 14 15:54:29 EDT 2002 athena Chris@10: * [project @ 2002-06-14 19:54:29 by athena] Chris@10: More work on k7 stuff Chris@10: Chris@10: A ./dft/direct-k7.c Chris@10: A ./dft/kdft-k7.c Chris@10: M ./dft/Makefile.am -1 +2 Chris@10: M ./dft/codelet.h -1 +13 Chris@10: M ./dft/conf.c -9 +15 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/direct-k7.c +167 Chris@10: M ./dft/kdft-k7.c +29 Chris@10: M ./genfft-k7/codeletMisc.ml -91 Chris@10: M ./genfft-k7/codeletMisc.mli -13 +1 Chris@10: M ./genfft-k7/genUtil.ml -24 +19 Chris@10: M ./genfft-k7/gen_notwiddle.ml -2 +44 Chris@10: M ./genfft-k7/genfft.ml +4 Chris@10: M ./genfft-k7/magic.ml +1 Chris@10: M ./genfft-k7/magic.mli +1 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Fri Jun 14 14:18:15 EDT 2002 athena Chris@10: * [project @ 2002-06-14 18:18:15 by athena] Chris@10: Changed my mind again Chris@10: Chris@10: M ./dft/codelet.h -4 +1 Chris@10: M ./dft/direct.c -7 +4 Chris@10: M ./genfft/gen_notw.ml -5 +4 Chris@10: Chris@10: Fri Jun 14 11:53:09 EDT 2002 athena Chris@10: * [project @ 2002-06-14 15:53:09 by athena] Chris@10: Removed some useless stuff. Chris@10: Chris@10: M ./genfft-k7/gen_notwiddle.ml -8 +3 Chris@10: Chris@10: Fri Jun 14 11:01:39 EDT 2002 athena Chris@10: * [project @ 2002-06-14 15:01:39 by athena] Chris@10: Hmm... Chris@10: Chris@10: M ./genfft-k7/gen_notwiddle.ml -3 +8 Chris@10: Chris@10: Fri Jun 14 10:28:12 EDT 2002 athena Chris@10: * [project @ 2002-06-14 14:28:12 by athena] Chris@10: More work in preparation for k7 stuff Chris@10: Chris@10: M ./dft/codelet.h -1 +2 Chris@10: M ./dft/direct.c -3 +4 Chris@10: M ./genfft/gen_notw.ml -4 +5 Chris@10: M ./genfft-k7/gen_notwiddle.ml -18 +13 Chris@10: Chris@10: Fri Jun 14 07:25:28 EDT 2002 athena Chris@10: * [project @ 2002-06-14 11:25:28 by athena] Chris@10: Still preparing to include k7 stuff Chris@10: Chris@10: M ./TODO +4 Chris@10: M ./dft/codelet.h -4 +6 Chris@10: M ./dft/direct.c -1 +5 Chris@10: M ./genfft/gen_notw.ml -7 +9 Chris@10: Chris@10: Fri Jun 14 07:06:02 EDT 2002 athena Chris@10: * [project @ 2002-06-14 11:06:02 by athena] Chris@10: Create .depend Chris@10: Chris@10: M ./bootstrap.sh -2 +2 Chris@10: Chris@10: Fri Jun 14 06:56:15 EDT 2002 athena Chris@10: * [project @ 2002-06-14 10:56:14 by athena] Chris@10: Imported Stefan's K7 generator Chris@10: Chris@10: A ./genfft-k7/ Chris@10: A ./genfft-k7/Makefile.am Chris@10: A ./genfft-k7/assignmentsToVfpinstrs.ml Chris@10: A ./genfft-k7/assignmentsToVfpinstrs.mli Chris@10: A ./genfft-k7/balanceVfpinstrs.ml Chris@10: A ./genfft-k7/balanceVfpinstrs.mli Chris@10: A ./genfft-k7/codeletMisc.ml Chris@10: A ./genfft-k7/codeletMisc.mli Chris@10: A ./genfft-k7/complex.ml Chris@10: A ./genfft-k7/complex.mli Chris@10: A ./genfft-k7/expr.ml Chris@10: A ./genfft-k7/expr.mli Chris@10: A ./genfft-k7/exprdag.ml Chris@10: A ./genfft-k7/exprdag.mli Chris@10: A ./genfft-k7/fft.ml Chris@10: A ./genfft-k7/fft.mli Chris@10: A ./genfft-k7/genUtil.ml Chris@10: A ./genfft-k7/genUtil.mli Chris@10: A ./genfft-k7/gen_hc2hc.ml Chris@10: A ./genfft-k7/gen_hc2hc.mli Chris@10: A ./genfft-k7/gen_hc2real.ml Chris@10: A ./genfft-k7/gen_hc2real.mli Chris@10: A ./genfft-k7/gen_notwiddle.ml Chris@10: A ./genfft-k7/gen_notwiddle.mli Chris@10: A ./genfft-k7/gen_notwiddle_fixedstride.ml Chris@10: A ./genfft-k7/gen_real2hc.ml Chris@10: A ./genfft-k7/gen_real2hc.mli Chris@10: A ./genfft-k7/gen_realeven.ml Chris@10: A ./genfft-k7/gen_realeven.mli Chris@10: A ./genfft-k7/gen_realeven2.ml Chris@10: A ./genfft-k7/gen_realeven2.mli Chris@10: A ./genfft-k7/gen_realodd.ml Chris@10: A ./genfft-k7/gen_realodd.mli Chris@10: A ./genfft-k7/gen_realodd2.ml Chris@10: A ./genfft-k7/gen_realodd2.mli Chris@10: A ./genfft-k7/gen_twiddle.ml Chris@10: A ./genfft-k7/gen_twiddle.mli Chris@10: A ./genfft-k7/genfft.ml Chris@10: A ./genfft-k7/id.ml Chris@10: A ./genfft-k7/id.mli Chris@10: A ./genfft-k7/k7Basics.ml Chris@10: A ./genfft-k7/k7Basics.mli Chris@10: A ./genfft-k7/k7ExecutionModel.ml Chris@10: A ./genfft-k7/k7ExecutionModel.mli Chris@10: A ./genfft-k7/k7FlatInstructionScheduling.ml Chris@10: A ./genfft-k7/k7FlatInstructionScheduling.mli Chris@10: A ./genfft-k7/k7InstructionSchedulingBasics.ml Chris@10: A ./genfft-k7/k7InstructionSchedulingBasics.mli Chris@10: A ./genfft-k7/k7RegisterAllocationBasics.ml Chris@10: A ./genfft-k7/k7RegisterAllocationBasics.mli Chris@10: A ./genfft-k7/k7RegisterAllocator.ml Chris@10: A ./genfft-k7/k7RegisterAllocator.mli Chris@10: A ./genfft-k7/k7RegisterAllocatorEATranslation.ml Chris@10: A ./genfft-k7/k7RegisterAllocatorEATranslation.mli Chris@10: A ./genfft-k7/k7RegisterAllocatorInit.ml Chris@10: A ./genfft-k7/k7RegisterAllocatorInit.mli Chris@10: A ./genfft-k7/k7RegisterReallocation.ml Chris@10: A ./genfft-k7/k7RegisterReallocation.mli Chris@10: A ./genfft-k7/k7Translate.ml Chris@10: A ./genfft-k7/k7Translate.mli Chris@10: A ./genfft-k7/k7Unparsing.ml Chris@10: A ./genfft-k7/k7Unparsing.mli Chris@10: A ./genfft-k7/k7Vectorization.ml Chris@10: A ./genfft-k7/k7Vectorization.mli Chris@10: A ./genfft-k7/magic.ml Chris@10: A ./genfft-k7/magic.mli Chris@10: A ./genfft-k7/memoMonad.ml Chris@10: A ./genfft-k7/memoMonad.mli Chris@10: A ./genfft-k7/nonDetMonad.ml Chris@10: A ./genfft-k7/nonDetMonad.mli Chris@10: A ./genfft-k7/nullVectorization.ml Chris@10: A ./genfft-k7/nullVectorization.mli Chris@10: A ./genfft-k7/number.ml Chris@10: A ./genfft-k7/number.mli Chris@10: A ./genfft-k7/stateMonad.ml Chris@10: A ./genfft-k7/stateMonad.mli Chris@10: A ./genfft-k7/symmetry.ml Chris@10: A ./genfft-k7/twiddle.ml Chris@10: A ./genfft-k7/util.ml Chris@10: A ./genfft-k7/util.mli Chris@10: A ./genfft-k7/vAnnotatedScheduler.ml Chris@10: A ./genfft-k7/vAnnotatedScheduler.mli Chris@10: A ./genfft-k7/vDag.ml Chris@10: A ./genfft-k7/vDag.mli Chris@10: A ./genfft-k7/vFpBasics.ml Chris@10: A ./genfft-k7/vFpBasics.mli Chris@10: A ./genfft-k7/vFpUnparsing.ml Chris@10: A ./genfft-k7/vFpUnparsing.mli Chris@10: A ./genfft-k7/vImproveSchedule.ml Chris@10: A ./genfft-k7/vImproveSchedule.mli Chris@10: A ./genfft-k7/vK7Optimization.ml Chris@10: A ./genfft-k7/vK7Optimization.mli Chris@10: A ./genfft-k7/vScheduler.ml Chris@10: A ./genfft-k7/vScheduler.mli Chris@10: A ./genfft-k7/vSimdBasics.ml Chris@10: A ./genfft-k7/vSimdBasics.mli Chris@10: A ./genfft-k7/vSimdIndexing.ml Chris@10: A ./genfft-k7/vSimdIndexing.mli Chris@10: A ./genfft-k7/vSimdUnparsing.ml Chris@10: A ./genfft-k7/vSimdUnparsing.mli Chris@10: A ./genfft-k7/variable.ml Chris@10: A ./genfft-k7/variable.mli Chris@10: M ./AUTHORS -2 +5 Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./bootstrap.sh +2 Chris@10: M ./configure.ac +1 Chris@10: M ./genfft-k7/Makefile.am +108 Chris@10: M ./genfft-k7/assignmentsToVfpinstrs.ml +225 Chris@10: M ./genfft-k7/assignmentsToVfpinstrs.mli +38 Chris@10: M ./genfft-k7/balanceVfpinstrs.ml +195 Chris@10: M ./genfft-k7/balanceVfpinstrs.mli +29 Chris@10: M ./genfft-k7/codeletMisc.ml +127 Chris@10: M ./genfft-k7/codeletMisc.mli +45 Chris@10: M ./genfft-k7/complex.ml +142 Chris@10: M ./genfft-k7/complex.mli +44 Chris@10: M ./genfft-k7/expr.ml +46 Chris@10: M ./genfft-k7/expr.mli -1 +34 Chris@10: M ./genfft-k7/exprdag.ml +879 Chris@10: M ./genfft-k7/exprdag.mli +42 Chris@10: M ./genfft-k7/fft.ml +311 Chris@10: M ./genfft-k7/fft.mli +50 Chris@10: M ./genfft-k7/genUtil.ml +377 Chris@10: M ./genfft-k7/genUtil.mli +41 Chris@10: M ./genfft-k7/gen_hc2hc.ml +133 Chris@10: M ./genfft-k7/gen_hc2hc.mli +25 Chris@10: M ./genfft-k7/gen_hc2real.ml +77 Chris@10: M ./genfft-k7/gen_hc2real.mli +24 Chris@10: M ./genfft-k7/gen_notwiddle.ml +69 Chris@10: M ./genfft-k7/gen_notwiddle.mli +25 Chris@10: M ./genfft-k7/gen_notwiddle_fixedstride.ml +45 Chris@10: M ./genfft-k7/gen_real2hc.ml +72 Chris@10: M ./genfft-k7/gen_real2hc.mli +24 Chris@10: M ./genfft-k7/gen_realeven.ml +55 Chris@10: M ./genfft-k7/gen_realeven.mli +24 Chris@10: M ./genfft-k7/gen_realeven2.ml +57 Chris@10: M ./genfft-k7/gen_realeven2.mli +24 Chris@10: M ./genfft-k7/gen_realodd.ml +60 Chris@10: M ./genfft-k7/gen_realodd.mli +24 Chris@10: M ./genfft-k7/gen_realodd2.ml +60 Chris@10: M ./genfft-k7/gen_realodd2.mli +24 Chris@10: M ./genfft-k7/gen_twiddle.ml +98 Chris@10: M ./genfft-k7/gen_twiddle.mli +25 Chris@10: M ./genfft-k7/genfft.ml +282 Chris@10: M ./genfft-k7/id.ml +61 Chris@10: M ./genfft-k7/id.mli +79 Chris@10: M ./genfft-k7/k7Basics.ml +524 Chris@10: M ./genfft-k7/k7Basics.mli +219 Chris@10: M ./genfft-k7/k7ExecutionModel.ml +111 Chris@10: M ./genfft-k7/k7ExecutionModel.mli +24 Chris@10: M ./genfft-k7/k7FlatInstructionScheduling.ml +177 Chris@10: M ./genfft-k7/k7FlatInstructionScheduling.mli +22 Chris@10: M ./genfft-k7/k7InstructionSchedulingBasics.ml +205 Chris@10: M ./genfft-k7/k7InstructionSchedulingBasics.mli +74 Chris@10: M ./genfft-k7/k7RegisterAllocationBasics.ml +162 Chris@10: M ./genfft-k7/k7RegisterAllocationBasics.mli +88 Chris@10: M ./genfft-k7/k7RegisterAllocator.ml +516 Chris@10: M ./genfft-k7/k7RegisterAllocator.mli +23 Chris@10: M ./genfft-k7/k7RegisterAllocatorEATranslation.ml +134 Chris@10: M ./genfft-k7/k7RegisterAllocatorEATranslation.mli +27 Chris@10: M ./genfft-k7/k7RegisterAllocatorInit.ml +97 Chris@10: M ./genfft-k7/k7RegisterAllocatorInit.mli +30 Chris@10: M ./genfft-k7/k7RegisterReallocation.ml +259 Chris@10: M ./genfft-k7/k7RegisterReallocation.mli +22 Chris@10: M ./genfft-k7/k7Translate.ml +142 Chris@10: M ./genfft-k7/k7Translate.mli +87 Chris@10: M ./genfft-k7/k7Unparsing.ml +326 Chris@10: M ./genfft-k7/k7Unparsing.mli +53 Chris@10: M ./genfft-k7/k7Vectorization.ml +739 Chris@10: M ./genfft-k7/k7Vectorization.mli +25 Chris@10: M ./genfft-k7/magic.ml +80 Chris@10: M ./genfft-k7/magic.mli +55 Chris@10: M ./genfft-k7/memoMonad.ml +32 Chris@10: M ./genfft-k7/memoMonad.mli +24 Chris@10: M ./genfft-k7/nonDetMonad.ml +136 Chris@10: M ./genfft-k7/nonDetMonad.mli +94 Chris@10: M ./genfft-k7/nullVectorization.ml +78 Chris@10: M ./genfft-k7/nullVectorization.mli +24 Chris@10: M ./genfft-k7/number.ml +158 Chris@10: M ./genfft-k7/number.mli +53 Chris@10: M ./genfft-k7/stateMonad.ml +71 Chris@10: M ./genfft-k7/stateMonad.mli +41 Chris@10: M ./genfft-k7/symmetry.ml +314 Chris@10: M ./genfft-k7/twiddle.ml +116 Chris@10: M ./genfft-k7/util.ml -1 +322 Chris@10: M ./genfft-k7/util.mli +124 Chris@10: M ./genfft-k7/vAnnotatedScheduler.ml +182 Chris@10: M ./genfft-k7/vAnnotatedScheduler.mli +39 Chris@10: M ./genfft-k7/vDag.ml +123 Chris@10: M ./genfft-k7/vDag.mli +52 Chris@10: M ./genfft-k7/vFpBasics.ml +203 Chris@10: M ./genfft-k7/vFpBasics.mli +106 Chris@10: M ./genfft-k7/vFpUnparsing.ml +79 Chris@10: M ./genfft-k7/vFpUnparsing.mli +28 Chris@10: M ./genfft-k7/vImproveSchedule.ml +155 Chris@10: M ./genfft-k7/vImproveSchedule.mli +22 Chris@10: M ./genfft-k7/vK7Optimization.ml +626 Chris@10: M ./genfft-k7/vK7Optimization.mli +22 Chris@10: M ./genfft-k7/vScheduler.ml +166 Chris@10: M ./genfft-k7/vScheduler.mli +31 Chris@10: M ./genfft-k7/vSimdBasics.ml +296 Chris@10: M ./genfft-k7/vSimdBasics.mli +247 Chris@10: M ./genfft-k7/vSimdIndexing.ml +119 Chris@10: M ./genfft-k7/vSimdIndexing.mli +68 Chris@10: M ./genfft-k7/vSimdUnparsing.ml +104 Chris@10: M ./genfft-k7/vSimdUnparsing.mli +33 Chris@10: M ./genfft-k7/variable.ml -1 +119 Chris@10: M ./genfft-k7/variable.mli +53 Chris@10: Chris@10: Thu Jun 13 15:30:41 EDT 2002 athena Chris@10: * [project @ 2002-06-13 19:30:41 by athena] Chris@10: Generator for real->halfcomplex and halfcomplex->real codelets Chris@10: Chris@10: A ./genfft/gen_hc2r.ml Chris@10: A ./genfft/gen_r2hc.ml Chris@10: M ./genfft/Makefile.am -9 +27 Chris@10: M ./genfft/c.ml -2 +2 Chris@10: M ./genfft/complex.ml -1 +8 Chris@10: M ./genfft/complex.mli -1 +3 Chris@10: M ./genfft/gen_hc2r.ml +121 Chris@10: M ./genfft/gen_r2hc.ml +121 Chris@10: M ./genfft/genutil.ml -8 +9 Chris@10: M ./genfft/trig.ml -2 +2 Chris@10: Chris@10: Thu Jun 13 11:54:02 EDT 2002 athena Chris@10: * [project @ 2002-06-13 15:54:02 by athena] Chris@10: Improved hash functions, printers Chris@10: Chris@10: M ./dft/problem.c -5 +10 Chris@10: M ./kernel/planner.c -11 +5 Chris@10: M ./kernel/tensor.c -7 +6 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Thu Jun 13 11:17:31 EDT 2002 athena Chris@10: * [project @ 2002-06-13 15:17:31 by athena] Chris@10: Only regenerate codlist.c in maintainer mode Chris@10: Chris@10: M ./support/Makefile.codelets -18 +18 Chris@10: Chris@10: Thu Jun 13 11:04:24 EDT 2002 athena Chris@10: * [project @ 2002-06-13 15:04:24 by athena] Chris@10: Planner can export solution list Chris@10: Chris@10: M ./dft/problem.c -10 +21 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./kernel/ifftw.h -11 +13 Chris@10: M ./kernel/planner-naive.c -7 +7 Chris@10: M ./kernel/planner-score.c -8 +7 Chris@10: M ./kernel/planner.c -42 +47 Chris@10: M ./kernel/print.c -9 +27 Chris@10: M ./kernel/tensor.c -4 +4 Chris@10: M ./tests/bench.c -5 +6 Chris@10: Chris@10: Thu Jun 13 08:59:53 EDT 2002 athena Chris@10: * [project @ 2002-06-13 12:59:53 by athena] Chris@10: Fixed for intel compiler Chris@10: Chris@10: M ./dft/ct-ditbuf.c -2 +3 Chris@10: M ./dft/dft.h -2 +1 Chris@10: M ./dft/direct.c -3 +1 Chris@10: M ./kernel/cycle.h -2 +2 Chris@10: M ./libbench/bench-user.h -12 +1 Chris@10: Chris@10: Thu Jun 13 08:48:51 EDT 2002 athena Chris@10: * [project @ 2002-06-13 12:48:51 by athena] Chris@10: Revised strategy for constants in codelets Chris@10: Chris@10: M ./dft/codelet.h -13 +2 Chris@10: M ./genfft/c.ml -50 +43 Chris@10: M ./genfft/c.mli -1 +4 Chris@10: M ./genfft/gen_notw.ml -5 +5 Chris@10: M ./genfft/gen_trig.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -3 +3 Chris@10: M ./genfft/gen_twidsq.ml -3 +3 Chris@10: M ./genfft/magic.ml -5 +1 Chris@10: Chris@10: Thu Jun 13 06:21:31 EDT 2002 athena Chris@10: * [project @ 2002-06-13 10:21:31 by athena] Chris@10: Enable score planner in classic mode, naive planner in pro mode. Chris@10: Chris@10: M ./tests/bench.c -3 +3 Chris@10: Chris@10: Wed Jun 12 19:18:18 EDT 2002 athena Chris@10: * [project @ 2002-06-12 23:18:18 by athena] Chris@10: Report classic/pro Chris@10: Chris@10: M ./tests/bench.c -2 +7 Chris@10: Chris@10: Wed Jun 12 19:07:48 EDT 2002 athena Chris@10: * [project @ 2002-06-12 23:07:48 by athena] Chris@10: Fixed behavior of buffered solver for large buffers. Chris@10: Chris@10: M ./dft/buffered.c -5 +5 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Wed Jun 12 18:57:19 EDT 2002 athena Chris@10: * [project @ 2002-06-12 22:57:19 by athena] Chris@10: Make assumption COST(vector) = length * COST(scalar) in classic mode. Chris@10: Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/vrank-geq1.c -1 +2 Chris@10: M ./kernel/ifftw.h -4 +7 Chris@10: M ./kernel/planner-naive.c -4 +2 Chris@10: M ./kernel/planner-score.c -5 +3 Chris@10: M ./kernel/planner.c -13 +20 Chris@10: M ./kernel/timer.c -8 +2 Chris@10: M ./libbench/timer.c -2 +2 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Wed Jun 12 18:19:48 EDT 2002 athena Chris@10: * [project @ 2002-06-12 22:19:48 by athena] Chris@10: Revised planner implementation in preparation for wisdom. Chris@10: Chris@10: M ./kernel/ifftw.h -5 +6 Chris@10: M ./kernel/plan.c -1 +2 Chris@10: M ./kernel/planner-naive.c -3 +7 Chris@10: M ./kernel/planner-score.c -3 +7 Chris@10: M ./kernel/planner.c -46 +81 Chris@10: M ./support/Makefile.codelets -19 +19 Chris@10: Chris@10: Wed Jun 12 08:27:36 EDT 2002 athena Chris@10: * [project @ 2002-06-12 12:27:36 by athena] Chris@10: Manually hoist loop invariants. Chris@10: Chris@10: M ./dft/ct-ditbuf.c -5 +11 Chris@10: Chris@10: Wed Jun 12 07:47:41 EDT 2002 athena Chris@10: * [project @ 2002-06-12 11:47:41 by athena] Chris@10: Revised loop to compile better with gcc -O Chris@10: Chris@10: M ./dft/rank-geq2.c -16 +8 Chris@10: M ./dft/rank0.c -23 +22 Chris@10: M ./dft/vrank-geq1.c -2 +2 Chris@10: Chris@10: Tue Jun 11 17:24:09 EDT 2002 athena Chris@10: * [project @ 2002-06-11 21:24:09 by athena] Chris@10: Changed tensor syntax Chris@10: Chris@10: M ./kernel/tensor.c -4 +5 Chris@10: Chris@10: Tue Jun 11 16:39:45 EDT 2002 athena Chris@10: * [project @ 2002-06-11 20:39:45 by athena] Chris@10: Added stuff to do. Chris@10: Chris@10: M ./TODO +15 Chris@10: Chris@10: Tue Jun 11 16:28:14 EDT 2002 athena Chris@10: * [project @ 2002-06-11 20:28:14 by athena] Chris@10: Report classic/pro in version number Chris@10: Chris@10: M ./kernel/version.c -2 +7 Chris@10: Chris@10: Tue Jun 11 14:22:49 EDT 2002 athena Chris@10: * [project @ 2002-06-11 18:22:49 by athena] Chris@10: Renamed versions into classic/pro Chris@10: Chris@10: A ./CLASSIC-MODE Chris@10: M ./Makefile.am -7 +7 Chris@10: R ./RESEARCH-MODE Chris@10: M ./bootstrap.sh -1 +1 Chris@10: M ./configure.ac -13 +13 Chris@10: M ./dft/codelets/inplace/Makefile.am -1 +1 Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/ct-ditbuf.c -2 +2 Chris@10: M ./dft/kdft-dit.c -2 +2 Chris@10: M ./dft/rank-geq2.c -4 +4 Chris@10: M ./dft/vrank-geq1.c -4 +4 Chris@10: M ./kernel/ifftw.h -3 +3 Chris@10: M ./mkdist.sh -2 +2 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Tue Jun 11 14:06:06 EDT 2002 athena Chris@10: * [project @ 2002-06-11 18:06:06 by athena] Chris@10: Revised planners, estimator Chris@10: Chris@10: M ./kernel/Makefile.am -4 +4 Chris@10: M ./kernel/ifftw.h -6 +7 Chris@10: R ./kernel/planner-estimate.c Chris@10: M ./kernel/planner-naive.c -4 +4 Chris@10: M ./kernel/planner-score.c -4 +4 Chris@10: M ./kernel/planner.c -2 +17 Chris@10: M ./tests/bench.c -5 +5 Chris@10: Chris@10: Tue Jun 11 11:45:41 EDT 2002 athena Chris@10: * [project @ 2002-06-11 15:45:41 by athena] Chris@10: I don't know what I am doing. Chris@10: Chris@10: M ./Makefile.am -2 +2 Chris@10: M ./dft/buffered.c -3 +2 Chris@10: M ./dft/ct-dit.c -4 +5 Chris@10: M ./dft/ct-ditbuf.c -4 +6 Chris@10: M ./dft/kdft-dif.c -3 +2 Chris@10: M ./dft/kdft-difsq.c -3 +2 Chris@10: M ./dft/kdft-dit.c -7 +4 Chris@10: M ./kernel/ifftw.h -3 +2 Chris@10: Chris@10: Tue Jun 11 10:35:52 EDT 2002 athena Chris@10: * [project @ 2002-06-11 14:35:52 by athena] Chris@10: Massive revision of estimator Chris@10: Chris@10: A ./kernel/ops.c Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./dft/buffered.c -8 +7 Chris@10: M ./dft/codelet.h -3 +3 Chris@10: M ./dft/ct-dif.c -5 +4 Chris@10: M ./dft/ct-dit.c -5 +7 Chris@10: M ./dft/ct-ditbuf.c -4 +9 Chris@10: M ./dft/ct-ditf.c -4 +3 Chris@10: M ./dft/ct.c -6 +1 Chris@10: M ./dft/dft.h -1 +3 Chris@10: M ./dft/direct.c -3 +2 Chris@10: M ./dft/indirect.c -3 +2 Chris@10: M ./dft/nop.c -1 +3 Chris@10: M ./dft/rank-geq2.c -3 +7 Chris@10: M ./dft/rank0.c -3 +4 Chris@10: M ./dft/vrank-geq1.c -10 +18 Chris@10: M ./dft/vrank2-transpose.c -3 +3 Chris@10: M ./dft/vrank3-transpose.c -3 +4 Chris@10: M ./genfft/c.ml -2 +2 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/cycle.h -3 +3 Chris@10: R ./kernel/flops.c Chris@10: M ./kernel/ifftw.h -13 +21 Chris@10: M ./kernel/ops.c +58 Chris@10: M ./kernel/plan.c -2 +2 Chris@10: M ./kernel/planner-estimate.c -2 +9 Chris@10: M ./kernel/planner-naive.c -3 +3 Chris@10: M ./kernel/planner-score.c -4 +4 Chris@10: M ./tests/bench.c -1 +3 Chris@10: Chris@10: Tue Jun 11 07:32:20 EDT 2002 athena Chris@10: * [project @ 2002-06-11 11:32:20 by athena] Chris@10: Many changes Chris@10: Chris@10: A ./dft/vrank-geq1.c Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/buffered.c -9 +8 Chris@10: M ./dft/codelets/inplace/Makefile.am -2 +3 Chris@10: M ./dft/codelets/standard/Makefile.am -2 Chris@10: M ./dft/conf.c -2 +2 Chris@10: M ./dft/ct-dit.c -2 +5 Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/dft.h -6 +4 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/problem.c -1 +13 Chris@10: M ./dft/rank-geq2.c -5 +4 Chris@10: R ./dft/vecloop.c Chris@10: M ./dft/vrank-geq1.c +243 Chris@10: M ./kernel/ifftw.h -2 +6 Chris@10: M ./kernel/planner-estimate.c -2 +2 Chris@10: M ./kernel/planner-naive.c -2 +2 Chris@10: M ./kernel/planner-score.c -4 +17 Chris@10: M ./kernel/planner.c -4 +6 Chris@10: M ./kernel/print.c -6 +27 Chris@10: M ./kernel/tensor.c -1 +15 Chris@10: M ./kernel/timer.c -2 +2 Chris@10: M ./libbench/bench-main.c -3 +6 Chris@10: M ./tests/bench.c -15 +15 Chris@10: Chris@10: Mon Jun 10 21:35:29 EDT 2002 athena Chris@10: * [project @ 2002-06-11 01:35:29 by athena] Chris@10: Keep it simple, stupid. Chris@10: Chris@10: M ./dft/ct-ditbuf.c -21 +4 Chris@10: Chris@10: Mon Jun 10 19:24:28 EDT 2002 athena Chris@10: * [project @ 2002-06-10 23:24:28 by athena] Chris@10: Fixed when #undef PRECOMPUTE_ARRAY_INDICES Chris@10: Chris@10: M ./kernel/ifftw.h -3 +5 Chris@10: Chris@10: Mon Jun 10 17:58:13 EDT 2002 athena Chris@10: * [project @ 2002-06-10 21:58:13 by athena] Chris@10: Minor changes Chris@10: Chris@10: M ./dft/vrank3-transpose.c -5 +3 Chris@10: M ./kernel/print.c -2 +2 Chris@10: Chris@10: Mon Jun 10 16:30:37 EDT 2002 athena Chris@10: * [project @ 2002-06-10 20:30:37 by athena] Chris@10: Added ct-ditbuf.c, many changes everywhere Chris@10: Chris@10: A ./dft/ct-ditbuf.c Chris@10: M ./CONVENTIONS -2 +3 Chris@10: M ./configure.ac +2 Chris@10: M ./dft/Makefile.am -3 +3 Chris@10: M ./dft/buffered.c -11 +11 Chris@10: M ./dft/ct-dif.c -5 +3 Chris@10: M ./dft/ct-dit.c -6 +4 Chris@10: M ./dft/ct-ditbuf.c +187 Chris@10: M ./dft/ct.c -3 +3 Chris@10: M ./dft/ct.h -1 +3 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/direct.c -6 +6 Chris@10: M ./dft/indirect.c -6 +6 Chris@10: M ./dft/kdft-dif.c -2 +2 Chris@10: M ./dft/kdft-difsq.c -2 +2 Chris@10: M ./dft/kdft-dit.c -3 +8 Chris@10: M ./dft/kdft.c -2 +2 Chris@10: M ./dft/nop.c -2 +2 Chris@10: M ./dft/rank-geq2.c -6 +6 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: M ./dft/vecloop.c -10 +8 Chris@10: M ./dft/vrank2-transpose.c -2 +2 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: M ./kernel/ifftw.h -3 +25 Chris@10: M ./kernel/plan.c -5 +4 Chris@10: M ./kernel/problem.c -2 +2 Chris@10: M ./kernel/timer.c -3 +3 Chris@10: M ./tests/bench.c -7 +8 Chris@10: Chris@10: Mon Jun 10 10:55:40 EDT 2002 athena Chris@10: * [project @ 2002-06-10 14:55:40 by athena] Chris@10: More name mangling Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./tests/bench.c -58 +47 Chris@10: Chris@10: Mon Jun 10 10:08:27 EDT 2002 athena Chris@10: * [project @ 2002-06-10 14:08:27 by athena] Chris@10: Fixed build system for single/double precision Chris@10: Chris@10: M ./Makefile.am -4 +3 Chris@10: M ./acinclude.m4 -13 Chris@10: M ./configure.ac +4 Chris@10: M ./tests/Makefile.am -1 +1 Chris@10: Chris@10: Mon Jun 10 09:04:21 EDT 2002 athena Chris@10: * [project @ 2002-06-10 13:04:21 by athena] Chris@10: Massive renaming to support both single and double precision. Chris@10: (Must recompile everything twice). Chris@10: Chris@10: A ./kernel/dfftw3.h Chris@10: A ./kernel/fftw3.h Chris@10: A ./kernel/sfftw3.h Chris@10: M ./CONVENTIONS -3 +9 Chris@10: M ./configure.ac -1 +1 Chris@10: M ./dft/buffered.c -96 +102 Chris@10: M ./dft/codelet.h -16 +16 Chris@10: M ./dft/codelets/inplace/Makefile.am -1 +1 Chris@10: M ./dft/codelets/standard/Makefile.am -1 +1 Chris@10: M ./dft/conf.c -12 +12 Chris@10: M ./dft/ct-dif.c -31 +33 Chris@10: M ./dft/ct-dit.c -29 +31 Chris@10: M ./dft/ct-ditf.c -27 +29 Chris@10: M ./dft/ct.c -40 +41 Chris@10: M ./dft/ct.h -10 +12 Chris@10: M ./dft/dft.h -25 +27 Chris@10: M ./dft/direct.c -26 +26 Chris@10: M ./dft/indirect.c -50 +63 Chris@10: M ./dft/kdft-dif.c -3 +3 Chris@10: M ./dft/kdft-difsq.c -3 +3 Chris@10: M ./dft/kdft-dit.c -3 +3 Chris@10: M ./dft/kdft.c -3 +3 Chris@10: M ./dft/nop.c -11 +11 Chris@10: M ./dft/plan.c -3 +3 Chris@10: M ./dft/problem.c -39 +40 Chris@10: M ./dft/rank-geq2.c -57 +59 Chris@10: M ./dft/rank0.c -54 +62 Chris@10: M ./dft/solve.c -2 +2 Chris@10: M ./dft/vecloop.c -49 +49 Chris@10: M ./dft/vrank2-transpose.c -27 +28 Chris@10: M ./dft/vrank3-transpose.c -44 +47 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./genfft/genutil.ml -2 +2 Chris@10: M ./kernel/Makefile.am -1 +2 Chris@10: M ./kernel/alloc.c -76 +76 Chris@10: M ./kernel/assert.c -2 +2 Chris@10: M ./kernel/awake.c -2 +2 Chris@10: M ./kernel/cycle.h -11 +11 Chris@10: M ./kernel/dfftw3.h +24 Chris@10: R ./kernel/fftw.h Chris@10: M ./kernel/fftw3.h +39 Chris@10: M ./kernel/flops.c -4 +6 Chris@10: M ./kernel/ifftw.h -90 +89 Chris@10: M ./kernel/minmax.c -5 +5 Chris@10: M ./kernel/plan.c -6 +6 Chris@10: M ./kernel/planner-estimate.c -6 +6 Chris@10: M ./kernel/planner-naive.c -7 +7 Chris@10: M ./kernel/planner-score.c -11 +11 Chris@10: M ./kernel/planner.c -73 +71 Chris@10: M ./kernel/print.c -12 +14 Chris@10: M ./kernel/problem.c -5 +5 Chris@10: M ./kernel/sfftw3.h +24 Chris@10: M ./kernel/solver.c -5 +5 Chris@10: M ./kernel/solvtab.c -4 +4 Chris@10: M ./kernel/square.c -2 +2 Chris@10: M ./kernel/stride.c -6 +6 Chris@10: M ./kernel/tensor.c -96 +96 Chris@10: M ./kernel/timer.c -39 +40 Chris@10: M ./kernel/twiddle.c -33 +38 Chris@10: M ./kernel/version.c -3 +4 Chris@10: M ./support/Makefile.codelets -3 +3 Chris@10: M ./tests/bench.c -39 +53 Chris@10: Chris@10: Mon Jun 10 06:49:55 EDT 2002 athena Chris@10: * [project @ 2002-06-10 10:49:55 by athena] Chris@10: Preliminary crude support for vector transforms in benchmark library. Chris@10: Chris@10: M ./libbench/allocate.c -7 +8 Chris@10: M ./libbench/bench-user.h -2 +5 Chris@10: M ./libbench/mflops.c -2 +2 Chris@10: M ./libbench/problem.c -2 +31 Chris@10: M ./tests/bench.c -1 +2 Chris@10: Chris@10: Sun Jun 9 21:11:51 EDT 2002 athena Chris@10: * [project @ 2002-06-10 01:11:51 by athena] Chris@10: Wrong cast Chris@10: Chris@10: M ./kernel/tensor.c -3 +3 Chris@10: Chris@10: Sun Jun 9 16:48:54 EDT 2002 athena Chris@10: * [project @ 2002-06-09 20:48:54 by athena] Chris@10: Added things to do. Chris@10: Chris@10: M ./TODO +1 Chris@10: Chris@10: Sun Jun 9 16:07:12 EDT 2002 athena Chris@10: * [project @ 2002-06-09 20:07:12 by athena] Chris@10: twlen0: make static Chris@10: Chris@10: M ./kernel/twiddle.c -2 +2 Chris@10: Chris@10: Sun Jun 9 15:34:54 EDT 2002 athena Chris@10: * [project @ 2002-06-09 19:34:54 by athena] Chris@10: Nothing Chris@10: Chris@10: M ./dft/buffered.c -3 +3 Chris@10: Chris@10: Sun Jun 9 15:30:13 EDT 2002 athena Chris@10: * [project @ 2002-06-09 19:30:13 by athena] Chris@10: Forgot break in switch statement. Chris@10: Chris@10: M ./kernel/print.c -1 +2 Chris@10: Chris@10: Sun Jun 9 15:27:24 EDT 2002 athena Chris@10: * [project @ 2002-06-09 19:27:24 by athena] Chris@10: Fix for c++ compatibility Chris@10: Chris@10: M ./kernel/print.c -4 +6 Chris@10: Chris@10: Sun Jun 9 15:16:43 EDT 2002 athena Chris@10: * [project @ 2002-06-09 19:16:43 by athena] Chris@10: Added printer, changed everything Chris@10: Chris@10: A ./kernel/print.c Chris@10: M ./TODO -4 Chris@10: M ./dft/buffered.c -26 +22 Chris@10: M ./dft/ct.c -10 +5 Chris@10: M ./dft/direct.c -8 +3 Chris@10: M ./dft/indirect.c -20 +16 Chris@10: M ./dft/nop.c -4 +6 Chris@10: M ./dft/rank-geq2.c -7 +3 Chris@10: M ./dft/rank0.c -6 +3 Chris@10: M ./dft/vecloop.c -5 +3 Chris@10: M ./dft/vrank2-transpose.c -3 +3 Chris@10: M ./dft/vrank3-transpose.c -3 +3 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -4 +14 Chris@10: M ./kernel/planner.c -3 +10 Chris@10: M ./kernel/print.c +122 Chris@10: M ./tests/bench.c -2 +22 Chris@10: Chris@10: Sun Jun 9 11:37:07 EDT 2002 athena Chris@10: * [project @ 2002-06-09 15:37:07 by athena] Chris@10: Removed redundant nop solver Chris@10: Chris@10: M ./dft/buffered.c -8 +3 Chris@10: M ./dft/nop.c -2 +12 Chris@10: M ./dft/rank0.c -25 +3 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sun Jun 9 11:06:31 EDT 2002 athena Chris@10: * [project @ 2002-06-09 15:06:31 by athena] Chris@10: More things to do Chris@10: Chris@10: M ./TODO +10 Chris@10: Chris@10: Sun Jun 9 11:01:41 EDT 2002 athena Chris@10: * [project @ 2002-06-09 15:01:41 by athena] Chris@10: Introduced idea of rank -infinity and associated NOP plans Chris@10: Chris@10: A ./TODO Chris@10: A ./dft/nop.c Chris@10: M ./TODO +4 Chris@10: M ./dft/Makefile.am -4 +4 Chris@10: M ./dft/buffered.c -88 +66 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/direct.c -2 +3 Chris@10: M ./dft/indirect.c -1 +2 Chris@10: M ./dft/nop.c +84 Chris@10: M ./dft/problem.c -2 +5 Chris@10: M ./dft/rank0.c -3 +12 Chris@10: M ./dft/vecloop.c -1 +2 Chris@10: M ./kernel/ifftw.h -1 +11 Chris@10: M ./kernel/tensor.c -11 +39 Chris@10: Chris@10: Sun Jun 9 08:36:27 EDT 2002 athena Chris@10: * [project @ 2002-06-09 12:36:27 by athena] Chris@10: Fixed comment Chris@10: Chris@10: M ./dft/buffered.c -2 +2 Chris@10: Chris@10: Sun Jun 9 08:20:13 EDT 2002 athena Chris@10: * [project @ 2002-06-09 12:20:13 by athena] Chris@10: Removed useless assertions. Chris@10: Chris@10: M ./kernel/tensor.c -4 +3 Chris@10: Chris@10: Sun Jun 9 08:19:26 EDT 2002 athena Chris@10: * [project @ 2002-06-09 12:19:26 by athena] Chris@10: Don't malloc(0). Chris@10: Chris@10: M ./kernel/tensor.c -3 +7 Chris@10: Chris@10: Sun Jun 9 08:08:13 EDT 2002 athena Chris@10: * [project @ 2002-06-09 12:08:13 by athena] Chris@10: Fixed signed/unsigned puns Chris@10: Chris@10: M ./dft/buffered.c -11 +11 Chris@10: Chris@10: Sun Jun 9 07:52:22 EDT 2002 athena Chris@10: * [project @ 2002-06-09 11:52:22 by athena] Chris@10: Added buffered.c Chris@10: Chris@10: A ./dft/buffered.c Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/buffered.c +374 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/ct-dif.c -2 +2 Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/ct-ditf.c -2 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/indirect.c -3 +3 Chris@10: M ./dft/rank-geq2.c -2 +2 Chris@10: M ./dft/rank0.c -6 +6 Chris@10: M ./dft/vecloop.c -2 +2 Chris@10: M ./dft/vrank2-transpose.c -2 +2 Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: M ./libbench/bench-main.c -2 +7 Chris@10: M ./libbench/bench-user.h -1 +3 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sat Jun 8 16:57:54 EDT 2002 athena Chris@10: * [project @ 2002-06-08 20:57:54 by athena] Chris@10: Fixed printout Chris@10: Chris@10: M ./dft/ct.c -6 +5 Chris@10: Chris@10: Sat Jun 8 16:42:52 EDT 2002 athena Chris@10: * [project @ 2002-06-08 20:42:52 by athena] Chris@10: Fixed comment Chris@10: Chris@10: M ./dft/vrank3-transpose.c -2 +2 Chris@10: Chris@10: Sat Jun 8 16:40:58 EDT 2002 athena Chris@10: * [project @ 2002-06-08 20:40:58 by athena] Chris@10: Added vrank3-transpose, renamed vrank0-transpose -> vrank2-transpose Chris@10: Chris@10: A ./dft/vrank2-transpose.c Chris@10: A ./dft/vrank3-transpose.c Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/conf.c -2 +3 Chris@10: M ./dft/dft.h -1 +3 Chris@10: R ./dft/vrank0-transpose.c Chris@10: M ./dft/vrank2-transpose.c +132 Chris@10: M ./dft/vrank3-transpose.c +173 Chris@10: M ./tests/bench.c -1 +1 Chris@10: Chris@10: Sat Jun 8 15:51:46 EDT 2002 athena Chris@10: * [project @ 2002-06-08 19:51:46 by athena] Chris@10: Added vrank0-transpose Chris@10: Chris@10: A ./dft/rank-geq2.c Chris@10: A ./dft/vrank0-transpose.c Chris@10: M ./bootstrap.sh -1 +1 Chris@10: M ./dft/Makefile.am -1 +2 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/direct.c -4 +4 Chris@10: M ./dft/rank-geq2.c +257 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: R ./dft/rank_geq2.c Chris@10: M ./dft/vrank0-transpose.c +132 Chris@10: M ./tests/bench.c -1 +2 Chris@10: Chris@10: Sat Jun 8 15:11:09 EDT 2002 athena Chris@10: * [project @ 2002-06-08 19:11:09 by athena] Chris@10: Added planner-score.c Chris@10: Chris@10: A ./kernel/planner-score.c Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/ct-dif.c -2 +2 Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/ct-ditf.c -2 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/indirect.c -2 +2 Chris@10: M ./dft/rank0.c -2 +2 Chris@10: M ./dft/rank_geq2.c -2 +2 Chris@10: M ./dft/vecloop.c -2 +2 Chris@10: M ./kernel/Makefile.am -3 +3 Chris@10: M ./kernel/ifftw.h -4 +5 Chris@10: M ./kernel/planner-score.c +70 Chris@10: M ./kernel/planner.c -12 +13 Chris@10: M ./tests/bench.c -1 +4 Chris@10: Chris@10: Sat Jun 8 11:10:44 EDT 2002 athena Chris@10: * [project @ 2002-06-08 15:10:44 by athena] Chris@10: Added indirect.c Chris@10: Chris@10: A ./dft/indirect.c Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/ct-dif.c -2 +2 Chris@10: M ./dft/ct-dit.c -2 +2 Chris@10: M ./dft/ct-ditf.c -4 +6 Chris@10: M ./dft/dft.h -1 +3 Chris@10: M ./dft/indirect.c +227 Chris@10: M ./dft/rank_geq2.c -5 +5 Chris@10: M ./dft/vecloop.c -2 +1 Chris@10: Chris@10: Sat Jun 8 09:34:58 EDT 2002 athena Chris@10: * [project @ 2002-06-08 13:34:58 by athena] Chris@10: dif, ditf solvers Chris@10: Chris@10: A ./mkdist.sh Chris@10: A ./dft/ct-dif.c Chris@10: A ./dft/ct-ditf.c Chris@10: A ./dft/kdft-dif.c Chris@10: A ./dft/kdft-difsq.c Chris@10: M ./Makefile.am -4 +6 Chris@10: M ./dft/Makefile.am -5 +3 Chris@10: M ./dft/codelet.h -2 +3 Chris@10: M ./dft/codelets/Makefile.am -5 Chris@10: M ./dft/codelets/inplace/Makefile.am -2 +4 Chris@10: M ./dft/conf.c -2 +3 Chris@10: M ./dft/ct-dif.c +120 Chris@10: M ./dft/ct-dit.c -2 +1 Chris@10: M ./dft/ct-ditf.c +106 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/direct.c -2 +2 Chris@10: M ./dft/kdft-dif.c +29 Chris@10: M ./dft/kdft-difsq.c +29 Chris@10: M ./mkdist.sh +6 Chris@10: M ./tests/Makefile.am -2 +2 Chris@10: Chris@10: Fri Jun 7 18:07:53 EDT 2002 athena Chris@10: * [project @ 2002-06-07 22:07:53 by athena] Chris@10: Implemented rank_geq2. Revised build system Chris@10: Chris@10: A ./RESEARCH-MODE Chris@10: A ./dft/rank_geq2.c Chris@10: M ./Makefile.am -3 +12 Chris@10: M ./RESEARCH-MODE +1 Chris@10: M ./bootstrap.sh -1 +1 Chris@10: M ./configure.ac -1 +10 Chris@10: M ./dft/Makefile.am -1 +3 Chris@10: M ./dft/conf.c -1 +2 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/rank_geq2.c +257 Chris@10: M ./dft/vecloop.c -42 +48 Chris@10: M ./kernel/ifftw.h -1 +3 Chris@10: M ./kernel/minmax.c -1 +11 Chris@10: M ./kernel/planner.c -1 +2 Chris@10: M ./support/Makefile.codelets -1 +1 Chris@10: Chris@10: Fri Jun 7 07:12:25 EDT 2002 athena Chris@10: * [project @ 2002-06-07 11:12:25 by athena] Chris@10: Fixed printout Chris@10: Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: Chris@10: Fri Jun 7 07:07:46 EDT 2002 athena Chris@10: * [project @ 2002-06-07 11:07:46 by athena] Chris@10: Added rank0. Revised codelet organization. Chris@10: Chris@10: A ./dft/codelet.h Chris@10: A ./dft/conf.c Chris@10: A ./dft/codelets/ Chris@10: A ./dft/codelets/Makefile.am Chris@10: A ./dft/codelets/inplace/ Chris@10: A ./dft/codelets/inplace/Makefile.am Chris@10: A ./dft/codelets/standard/ Chris@10: A ./dft/codelets/standard/Makefile.am Chris@10: A ./dft/rank0.c Chris@10: M ./Makefile.am -3 +2 Chris@10: M ./bootstrap.sh -5 +1 Chris@10: M ./configure.ac -3 +18 Chris@10: M ./dft/Makefile.am -3 +4 Chris@10: M ./dft/codelet.h +118 Chris@10: M ./dft/codelets/Makefile.am +7 Chris@10: M ./dft/codelets/inplace/Makefile.am -1 +35 Chris@10: M ./dft/codelets/standard/Makefile.am -1 +37 Chris@10: M ./dft/conf.c +30 Chris@10: M ./dft/dft.h -1 +5 Chris@10: M ./dft/rank0.c +263 Chris@10: M ./genfft/gen_notw.ml -3 +3 Chris@10: M ./tests/Makefile.am -4 +5 Chris@10: M ./tests/bench.c -6 +1 Chris@10: Chris@10: Thu Jun 6 18:03:17 EDT 2002 athena Chris@10: * [project @ 2002-06-06 22:03:17 by athena] Chris@10: Added memoization Chris@10: Chris@10: M ./dft/ct.c -2 +2 Chris@10: M ./dft/vecloop.c -2 +2 Chris@10: M ./genfft/trig.ml -5 +5 Chris@10: M ./kernel/ifftw.h -2 +13 Chris@10: M ./kernel/planner-estimate.c -2 +2 Chris@10: M ./kernel/planner-naive.c -2 +2 Chris@10: M ./kernel/planner.c -12 +191 Chris@10: M ./libbench/bench-user.h -1 +2 Chris@10: M ./tests/bench.c -2 +16 Chris@10: Chris@10: Thu Jun 6 08:07:33 EDT 2002 athena Chris@10: * [project @ 2002-06-06 12:07:33 by athena] Chris@10: Added vecloop Chris@10: Chris@10: A ./dft/vecloop.c Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/dft.h -1 +3 Chris@10: M ./dft/direct.c -9 +7 Chris@10: M ./dft/vecloop.c +242 Chris@10: M ./kernel/alloc.c -11 +38 Chris@10: M ./kernel/ifftw.h -2 +4 Chris@10: M ./kernel/planner.c -2 +2 Chris@10: M ./tests/bench.c +5 Chris@10: Chris@10: Wed Jun 5 19:02:56 EDT 2002 athena Chris@10: * [project @ 2002-06-05 23:02:56 by athena] Chris@10: First DIT solver/plan Chris@10: Chris@10: A ./dft/ct-dit.c Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/ct-dit.c +118 Chris@10: M ./dft/ct.c -9 +16 Chris@10: M ./dft/ct.h -3 +14 Chris@10: M ./dft/dft.h -1 +2 Chris@10: M ./dft/direct.c -10 +6 Chris@10: M ./dft/kdft-dit.c -2 +3 Chris@10: M ./kernel/alloc.c -1 +2 Chris@10: M ./kernel/twiddle.c -3 +11 Chris@10: Chris@10: Wed Jun 5 16:03:44 EDT 2002 athena Chris@10: * [project @ 2002-06-05 20:03:44 by athena] Chris@10: More work on ct Chris@10: Chris@10: A ./dft/ct.c Chris@10: A ./dft/ct.h Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/ct.c +166 Chris@10: M ./dft/ct.h +56 Chris@10: M ./kernel/ifftw.h -1 +2 Chris@10: M ./kernel/stride.c -2 +3 Chris@10: M ./kernel/twiddle.c -16 +26 Chris@10: Chris@10: Wed Jun 5 11:28:09 EDT 2002 athena Chris@10: * [project @ 2002-06-05 15:28:09 by athena] Chris@10: Only use cycle counters Chris@10: Chris@10: M ./kernel/ifftw.h -2 +2 Chris@10: M ./kernel/planner-naive.c -2 +2 Chris@10: M ./kernel/timer.c -166 +18 Chris@10: Chris@10: Tue Jun 4 20:22:23 EDT 2002 athena Chris@10: * [project @ 2002-06-05 00:22:23 by athena] Chris@10: Signed/unsigned fixup Chris@10: Chris@10: M ./CONVENTIONS +2 Chris@10: M ./bootstrap.sh -1 +1 Chris@10: M ./kernel/ifftw.h -7 +6 Chris@10: M ./kernel/twiddle.c -28 +16 Chris@10: Chris@10: Tue Jun 4 20:03:56 EDT 2002 athena Chris@10: * [project @ 2002-06-05 00:03:56 by athena] Chris@10: New file twiddle.c Chris@10: Chris@10: A ./kernel/twiddle.c Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -1 +24 Chris@10: M ./kernel/twiddle.c +137 Chris@10: Chris@10: Tue Jun 4 17:49:39 EDT 2002 athena Chris@10: * [project @ 2002-06-04 21:49:39 by athena] Chris@10: Made tensor ranks and vector lengths unsigned. Hopefully fixed Chris@10: all places where it matters. Chris@10: Chris@10: A ./dft/kdft-dit.c Chris@10: M ./configure.ac +1 Chris@10: M ./dft/Makefile.am -1 +2 Chris@10: M ./dft/direct.c -3 +4 Chris@10: M ./dft/kdft-dit.c +28 Chris@10: M ./dft/problem.c -5 +7 Chris@10: M ./genfft/gen_notw.ml -4 +4 Chris@10: M ./genfft/gen_twiddle.ml -4 +4 Chris@10: M ./genfft/gen_twidsq.ml -4 +4 Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: R ./kernel/codelet.h Chris@10: M ./kernel/flops.c -2 +2 Chris@10: M ./kernel/ifftw.h -17 +20 Chris@10: M ./kernel/tensor.c -35 +38 Chris@10: M ./kernel/timer.c -3 +2 Chris@10: M ./tests/bench.c +1 Chris@10: Chris@10: Tue Jun 4 16:28:58 EDT 2002 athena Chris@10: * [project @ 2002-06-04 20:28:58 by athena] Chris@10: System is in working state now (but very incomplete) Chris@10: Chris@10: A ./dft/kdft.c Chris@10: A ./kernel/planner-estimate.c Chris@10: A ./kernel/solvtab.c Chris@10: A ./tests/ Chris@10: A ./tests/Makefile.am Chris@10: A ./tests/bench.c Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac +2 Chris@10: M ./dft/Makefile.am -2 +2 Chris@10: M ./dft/dft.h -1 +10 Chris@10: M ./dft/direct.c -3 +2 Chris@10: M ./dft/kdft.c +29 Chris@10: M ./kernel/Makefile.am -3 +4 Chris@10: M ./kernel/codelet.h -1 +3 Chris@10: M ./kernel/fftw.h -1 +5 Chris@10: M ./kernel/ifftw.h -2 +21 Chris@10: M ./kernel/planner-estimate.c +55 Chris@10: M ./kernel/planner-naive.c -9 +9 Chris@10: M ./kernel/solvtab.c +30 Chris@10: M ./libbench/Makefile.am -2 +2 Chris@10: M ./support/Makefile.codelets +23 Chris@10: M ./tests/Makefile.am -1 +7 Chris@10: M ./tests/bench.c +92 Chris@10: Chris@10: Mon Jun 3 18:10:12 EDT 2002 athena Chris@10: * [project @ 2002-06-03 22:10:12 by athena] Chris@10: Started implementing planners Chris@10: Chris@10: A ./kernel/planner-naive.c Chris@10: A ./kernel/planner.c Chris@10: M ./CONVENTIONS +2 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/ifftw.h -6 +55 Chris@10: M ./kernel/planner-naive.c +58 Chris@10: M ./kernel/planner.c +113 Chris@10: Chris@10: Mon Jun 3 11:44:18 EDT 2002 athena Chris@10: * [project @ 2002-06-03 15:44:18 by athena] Chris@10: Imported libbench from the new benchfft. We will use libbench Chris@10: for benchmarking and testing. Chris@10: Chris@10: A ./libbench/ Chris@10: A ./libbench/Makefile.am Chris@10: A ./libbench/accopy-from.c Chris@10: A ./libbench/accopy-to.c Chris@10: A ./libbench/acopy.c Chris@10: A ./libbench/allocate.c Chris@10: A ./libbench/ascale.c Chris@10: A ./libbench/aset.c Chris@10: A ./libbench/bench-main.c Chris@10: A ./libbench/bench-user.h Chris@10: A ./libbench/bench.h Chris@10: A ./libbench/caadd.c Chris@10: A ./libbench/cacopy.c Chris@10: A ./libbench/can-do.c Chris@10: A ./libbench/cascale.c Chris@10: A ./libbench/caset.c Chris@10: A ./libbench/casub.c Chris@10: A ./libbench/ccopy-from.c Chris@10: A ./libbench/ccopy-to.c Chris@10: A ./libbench/copy-c2c-from.c Chris@10: A ./libbench/copy-c2c-to.c Chris@10: A ./libbench/copy-c2h-1d-fftpack.c Chris@10: A ./libbench/copy-c2h-1d-halfcomplex.c Chris@10: A ./libbench/copy-c2h-1d-packed.c Chris@10: A ./libbench/copy-c2h-1d-unpacked-ri.c Chris@10: A ./libbench/copy-c2h-unpacked.c Chris@10: A ./libbench/copy-c2h.c Chris@10: A ./libbench/copy-c2r-packed.c Chris@10: A ./libbench/copy-c2r-unpacked.c Chris@10: A ./libbench/copy-c2r.c Chris@10: A ./libbench/copy-c2ri.c Chris@10: A ./libbench/copy-h2c-1d-fftpack.c Chris@10: A ./libbench/copy-h2c-1d-halfcomplex.c Chris@10: A ./libbench/copy-h2c-1d-packed.c Chris@10: A ./libbench/copy-h2c-1d-unpacked-ri.c Chris@10: A ./libbench/copy-h2c-unpacked.c Chris@10: A ./libbench/copy-h2c.c Chris@10: A ./libbench/copy-r2c-packed.c Chris@10: A ./libbench/copy-r2c-unpacked.c Chris@10: A ./libbench/copy-r2c.c Chris@10: A ./libbench/copy-ri2c.c Chris@10: A ./libbench/deallocate.c Chris@10: A ./libbench/getopt-utils.c Chris@10: A ./libbench/getopt.c Chris@10: A ./libbench/getopt.h Chris@10: A ./libbench/getopt1.c Chris@10: A ./libbench/info.c Chris@10: A ./libbench/log2.c Chris@10: A ./libbench/main.c Chris@10: A ./libbench/mflops.c Chris@10: A ./libbench/ovtpvt.c Chris@10: A ./libbench/pow2.c Chris@10: A ./libbench/prime.c Chris@10: A ./libbench/problem.c Chris@10: A ./libbench/report.c Chris@10: A ./libbench/speed.c Chris@10: A ./libbench/timer.c Chris@10: A ./libbench/unnormalize.c Chris@10: A ./libbench/util.c Chris@10: A ./libbench/verify.c Chris@10: A ./libbench/zero.c Chris@10: M ./Makefile.am -1 +1 Chris@10: M ./configure.ac -1 +4 Chris@10: M ./libbench/Makefile.am +18 Chris@10: M ./libbench/accopy-from.c +10 Chris@10: M ./libbench/accopy-to.c +10 Chris@10: M ./libbench/acopy.c +11 Chris@10: M ./libbench/allocate.c +40 Chris@10: M ./libbench/ascale.c +13 Chris@10: M ./libbench/aset.c +10 Chris@10: M ./libbench/bench-main.c +218 Chris@10: M ./libbench/bench-user.h +247 Chris@10: M ./libbench/bench.h +66 Chris@10: M ./libbench/caadd.c +15 Chris@10: M ./libbench/cacopy.c +11 Chris@10: M ./libbench/can-do.c +33 Chris@10: M ./libbench/cascale.c +15 Chris@10: M ./libbench/caset.c +10 Chris@10: M ./libbench/casub.c +15 Chris@10: M ./libbench/ccopy-from.c +21 Chris@10: M ./libbench/ccopy-to.c +20 Chris@10: M ./libbench/copy-c2c-from.c +7 Chris@10: M ./libbench/copy-c2c-to.c +7 Chris@10: M ./libbench/copy-c2h-1d-fftpack.c +29 Chris@10: M ./libbench/copy-c2h-1d-halfcomplex.c +29 Chris@10: M ./libbench/copy-c2h-1d-packed.c +40 Chris@10: M ./libbench/copy-c2h-1d-unpacked-ri.c +25 Chris@10: M ./libbench/copy-c2h-unpacked.c +32 Chris@10: M ./libbench/copy-c2h.c +10 Chris@10: M ./libbench/copy-c2r-packed.c +11 Chris@10: M ./libbench/copy-c2r-unpacked.c +24 Chris@10: M ./libbench/copy-c2r.c +6 Chris@10: M ./libbench/copy-c2ri.c +15 Chris@10: M ./libbench/copy-h2c-1d-fftpack.c +31 Chris@10: M ./libbench/copy-h2c-1d-halfcomplex.c +34 Chris@10: M ./libbench/copy-h2c-1d-packed.c +44 Chris@10: M ./libbench/copy-h2c-1d-unpacked-ri.c +35 Chris@10: M ./libbench/copy-h2c-unpacked.c +51 Chris@10: M ./libbench/copy-h2c.c +10 Chris@10: M ./libbench/copy-r2c-packed.c +13 Chris@10: M ./libbench/copy-r2c-unpacked.c +26 Chris@10: M ./libbench/copy-r2c.c +6 Chris@10: M ./libbench/copy-ri2c.c +15 Chris@10: M ./libbench/deallocate.c +14 Chris@10: M ./libbench/getopt-utils.c +104 Chris@10: M ./libbench/getopt.c +1062 Chris@10: M ./libbench/getopt.h +180 Chris@10: M ./libbench/getopt1.c +188 Chris@10: M ./libbench/info.c +57 Chris@10: M ./libbench/log2.c +13 Chris@10: M ./libbench/main.c +40 Chris@10: M ./libbench/mflops.c +22 Chris@10: M ./libbench/ovtpvt.c +13 Chris@10: M ./libbench/pow2.c +6 Chris@10: M ./libbench/prime.c +49 Chris@10: M ./libbench/problem.c +126 Chris@10: M ./libbench/report.c +159 Chris@10: M ./libbench/speed.c +73 Chris@10: M ./libbench/timer.c +218 Chris@10: M ./libbench/unnormalize.c +13 Chris@10: M ./libbench/util.c +189 Chris@10: M ./libbench/verify.c +408 Chris@10: M ./libbench/zero.c +37 Chris@10: Chris@10: Mon Jun 3 09:18:46 EDT 2002 athena Chris@10: * [project @ 2002-06-03 13:18:46 by athena] Chris@10: Removed useless rand.c Chris@10: Chris@10: M ./kernel/Makefile.am -2 +2 Chris@10: M ./kernel/ifftw.h -4 +1 Chris@10: R ./kernel/rand.c Chris@10: M ./kernel/timer.c -2 +2 Chris@10: Chris@10: Mon Jun 3 08:09:05 EDT 2002 athena Chris@10: * [project @ 2002-06-03 12:09:05 by athena] Chris@10: Added timer Chris@10: Chris@10: A ./kernel/cycle.h Chris@10: A ./kernel/timer.c Chris@10: M ./CONVENTIONS +1 Chris@10: M ./dft/problem.c -2 +2 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/alloc.c -2 +2 Chris@10: M ./kernel/cycle.h +204 Chris@10: M ./kernel/ifftw.h -2 +10 Chris@10: M ./kernel/plan.c -2 +4 Chris@10: M ./kernel/timer.c +293 Chris@10: Chris@10: Sun Jun 2 21:03:51 EDT 2002 athena Chris@10: * [project @ 2002-06-03 01:03:51 by athena] Chris@10: Split codelets into standard and inplace Chris@10: Chris@10: M ./configure.ac -2 +4 Chris@10: Chris@10: Sun Jun 2 19:49:03 EDT 2002 athena Chris@10: * [project @ 2002-06-02 23:49:03 by athena] Chris@10: Many many changes Chris@10: Chris@10: A ./CONVENTIONS Chris@10: A ./dft/direct.c Chris@10: A ./dft/plan.c Chris@10: A ./dft/solve.c Chris@10: A ./kernel/awake.c Chris@10: A ./kernel/square.c Chris@10: M ./CONVENTIONS +49 Chris@10: M ./Makefile.am -4 +4 Chris@10: M ./dft/Makefile.am -1 +1 Chris@10: M ./dft/dft.h -3 +22 Chris@10: M ./dft/direct.c +171 Chris@10: M ./dft/plan.c +33 Chris@10: M ./dft/problem.c -10 +10 Chris@10: M ./dft/solve.c +31 Chris@10: M ./kernel/Makefile.am -1 +1 Chris@10: M ./kernel/awake.c +30 Chris@10: M ./kernel/ifftw.h -13 +33 Chris@10: M ./kernel/square.c +28 Chris@10: Chris@10: Sun Jun 2 15:00:11 EDT 2002 athena Chris@10: * [project @ 2002-06-02 19:00:11 by athena] Chris@10: Fixed anachronism Chris@10: Chris@10: M ./kernel/codelet.h -4 +4 Chris@10: Chris@10: Sat Aug 12 21:43:16 EDT 2006 Unknown tagger Chris@10: tagged bar Chris@10: Chris@10: Chris@10: Sun Jun 2 14:42:32 EDT 2002 athena Chris@10: * [project @ 2002-06-02 18:42:32 by athena] Chris@10: Initial revision Chris@10: Chris@10: A ./AUTHORS Chris@10: A ./COPYRIGHT Chris@10: A ./ChangeLog Chris@10: A ./Makefile.am Chris@10: A ./NEWS Chris@10: A ./README Chris@10: A ./acinclude.m4 Chris@10: A ./bootstrap.sh Chris@10: A ./configure.ac Chris@10: A ./dft/ Chris@10: A ./dft/Makefile.am Chris@10: A ./dft/dft.h Chris@10: A ./dft/problem.c Chris@10: A ./genfft/ Chris@10: A ./genfft/Makefile.am Chris@10: A ./genfft/algsimp.ml Chris@10: A ./genfft/algsimp.mli Chris@10: A ./genfft/annotate.ml Chris@10: A ./genfft/annotate.mli Chris@10: A ./genfft/assoctable.ml Chris@10: A ./genfft/assoctable.mli Chris@10: A ./genfft/c.ml Chris@10: A ./genfft/c.mli Chris@10: A ./genfft/complex.ml Chris@10: A ./genfft/complex.mli Chris@10: A ./genfft/conv.ml Chris@10: A ./genfft/conv.mli Chris@10: A ./genfft/dag.ml Chris@10: A ./genfft/dag.mli Chris@10: A ./genfft/expr.ml Chris@10: A ./genfft/expr.mli Chris@10: A ./genfft/fft.ml Chris@10: A ./genfft/fft.mli Chris@10: A ./genfft/gen_athnotw.ml Chris@10: A ./genfft/gen_athtw.ml Chris@10: A ./genfft/gen_conv.ml Chris@10: A ./genfft/gen_notw.ml Chris@10: A ./genfft/gen_trig.ml Chris@10: A ./genfft/gen_twiddle.ml Chris@10: A ./genfft/gen_twidsq.ml Chris@10: A ./genfft/genutil.ml Chris@10: A ./genfft/littlesimp.ml Chris@10: A ./genfft/littlesimp.mli Chris@10: A ./genfft/magic.ml Chris@10: A ./genfft/monads.ml Chris@10: A ./genfft/number.ml Chris@10: A ./genfft/number.mli Chris@10: A ./genfft/oracle.ml Chris@10: A ./genfft/oracle.mli Chris@10: A ./genfft/schedule.ml Chris@10: A ./genfft/schedule.mli Chris@10: A ./genfft/to_alist.ml Chris@10: A ./genfft/to_alist.mli Chris@10: A ./genfft/trig.ml Chris@10: A ./genfft/trig.mli Chris@10: A ./genfft/twiddle.ml Chris@10: A ./genfft/twiddle.mli Chris@10: A ./genfft/unique.ml Chris@10: A ./genfft/unique.mli Chris@10: A ./genfft/util.ml Chris@10: A ./genfft/util.mli Chris@10: A ./genfft/variable.ml Chris@10: A ./genfft/variable.mli Chris@10: A ./kernel/ Chris@10: A ./kernel/Makefile.am Chris@10: A ./kernel/alloc.c Chris@10: A ./kernel/assert.c Chris@10: A ./kernel/codelet.h Chris@10: A ./kernel/fftw.h Chris@10: A ./kernel/flops.c Chris@10: A ./kernel/ifftw.h Chris@10: A ./kernel/minmax.c Chris@10: A ./kernel/plan.c Chris@10: A ./kernel/problem.c Chris@10: A ./kernel/rand.c Chris@10: A ./kernel/solver.c Chris@10: A ./kernel/stride.c Chris@10: A ./kernel/tensor.c Chris@10: A ./kernel/version.c Chris@10: A ./support/ Chris@10: A ./support/Makefile.am Chris@10: A ./support/Makefile.codelets Chris@10: A ./support/codelet_prelude Chris@10: M ./AUTHORS +4 Chris@10: M ./COPYRIGHT +19 Chris@10: M ./Makefile.am +21 Chris@10: M ./acinclude.m4 +356 Chris@10: M ./bootstrap.sh +26 Chris@10: M ./configure.ac +92 Chris@10: M ./dft/Makefile.am +6 Chris@10: M ./dft/dft.h +30 Chris@10: M ./dft/problem.c +121 Chris@10: M ./genfft/Makefile.am +136 Chris@10: M ./genfft/algsimp.ml +517 Chris@10: M ./genfft/algsimp.mli +24 Chris@10: M ./genfft/annotate.ml +180 Chris@10: M ./genfft/annotate.mli +37 Chris@10: M ./genfft/assoctable.ml +66 Chris@10: M ./genfft/assoctable.mli +30 Chris@10: M ./genfft/c.ml +445 Chris@10: M ./genfft/c.mli +65 Chris@10: M ./genfft/complex.ml +198 Chris@10: M ./genfft/complex.mli +61 Chris@10: M ./genfft/conv.ml +130 Chris@10: M ./genfft/conv.mli +23 Chris@10: M ./genfft/dag.ml +110 Chris@10: M ./genfft/dag.mli +44 Chris@10: M ./genfft/expr.ml +93 Chris@10: M ./genfft/expr.mli +38 Chris@10: M ./genfft/fft.ml +227 Chris@10: M ./genfft/fft.mli +24 Chris@10: M ./genfft/gen_athnotw.ml +80 Chris@10: M ./genfft/gen_athtw.ml +108 Chris@10: M ./genfft/gen_conv.ml +89 Chris@10: M ./genfft/gen_notw.ml +126 Chris@10: M ./genfft/gen_trig.ml +159 Chris@10: M ./genfft/gen_twiddle.ml +144 Chris@10: M ./genfft/gen_twidsq.ml +172 Chris@10: M ./genfft/genutil.ml +256 Chris@10: M ./genfft/littlesimp.ml +72 Chris@10: M ./genfft/littlesimp.mli +26 Chris@10: M ./genfft/magic.ml +121 Chris@10: M ./genfft/monads.ml +76 Chris@10: M ./genfft/number.ml +153 Chris@10: M ./genfft/number.mli +49 Chris@10: M ./genfft/oracle.ml +138 Chris@10: M ./genfft/oracle.mli +25 Chris@10: M ./genfft/schedule.ml +186 Chris@10: M ./genfft/schedule.mli +30 Chris@10: M ./genfft/to_alist.ml +258 Chris@10: M ./genfft/to_alist.mli +24 Chris@10: M ./genfft/trig.ml +158 Chris@10: M ./genfft/trig.mli +35 Chris@10: M ./genfft/twiddle.ml +165 Chris@10: M ./genfft/twiddle.mli +29 Chris@10: M ./genfft/unique.ml +39 Chris@10: M ./genfft/unique.mli +25 Chris@10: M ./genfft/util.ml +177 Chris@10: M ./genfft/util.mli +50 Chris@10: M ./genfft/variable.ml +99 Chris@10: M ./genfft/variable.mli +35 Chris@10: M ./kernel/Makefile.am +5 Chris@10: M ./kernel/alloc.c +217 Chris@10: M ./kernel/assert.c +32 Chris@10: M ./kernel/codelet.h +126 Chris@10: M ./kernel/fftw.h +29 Chris@10: M ./kernel/flops.c +41 Chris@10: M ./kernel/ifftw.h +215 Chris@10: M ./kernel/minmax.c +33 Chris@10: M ./kernel/plan.c +50 Chris@10: M ./kernel/problem.c +47 Chris@10: M ./kernel/rand.c +36 Chris@10: M ./kernel/solver.c +43 Chris@10: M ./kernel/stride.c +41 Chris@10: M ./kernel/tensor.c +318 Chris@10: M ./kernel/version.c +26 Chris@10: M ./support/Makefile.am -1 +1 Chris@10: M ./support/Makefile.codelets +99 Chris@10: M ./support/codelet_prelude +8