comparison src/fftw-3.3.5/doc/FAQ/fftw-faq.html/section5.html @ 42:2cd0e3b3e1fd

Current fftw source
author Chris Cannam
date Tue, 18 Oct 2016 13:40:26 +0100
parents
children
comparison
equal deleted inserted replaced
41:481f5f8c5634 42:2cd0e3b3e1fd
1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
2 <html>
3 <head><title>
4 FFTW FAQ - Section 5
5 </title>
6 <link rev="made" href="mailto:fftw@fftw.org">
7 <link rel="Contents" href="index.html">
8 <link rel="Start" href="index.html">
9 <link rel="Previous" href="section4.html"><link rel="Bookmark" title="FFTW FAQ" href="index.html">
10 </head><body text="#000000" bgcolor="#FFFFFF"><h1>
11 FFTW FAQ - Section 5 <br>
12 Known bugs
13 </h1>
14
15 <ul>
16 <li><a href="#rfftwndbug" rel=subdocument>Q5.1. FFTW 1.1 crashes in rfftwnd on Linux.</a>
17 <li><a href="#fftwmpibug" rel=subdocument>Q5.2. The MPI transforms in FFTW 1.2 give incorrect results/leak
18 memory.</a>
19 <li><a href="#testsingbug" rel=subdocument>Q5.3. The test programs in FFTW 1.2.1 fail when I change FFTW to use single
20 precision.</a>
21 <li><a href="#teststoobig" rel=subdocument>Q5.4. The test program in FFTW 1.2.1 fails for n &gt;
22 46340.</a>
23 <li><a href="#linuxthreads" rel=subdocument>Q5.5. The threaded code fails on Linux Redhat 5.0</a>
24 <li><a href="#bigrfftwnd" rel=subdocument>Q5.6. FFTW 2.0's rfftwnd fails for rank &gt; 1 transforms with a final
25 dimension &gt;= 65536.</a>
26 <li><a href="#primebug" rel=subdocument>Q5.7. FFTW 2.0's complex transforms give the wrong results with prime
27 factors 17 to 97.</a>
28 <li><a href="#mpichbug" rel=subdocument>Q5.8. FFTW 2.1.1's MPI test programs crash with
29 MPICH.</a>
30 <li><a href="#aixthreadbug" rel=subdocument>Q5.9. FFTW 2.1.2's multi-threaded transforms don't work on
31 AIX.</a>
32 <li><a href="#bigprimebug" rel=subdocument>Q5.10. FFTW 2.1.2's complex transforms give incorrect results for large prime
33 sizes.</a>
34 <li><a href="#solaristhreadbug" rel=subdocument>Q5.11. FFTW 2.1.3's multi-threaded transforms don't give any speedup on
35 Solaris.</a>
36 <li><a href="#aixflags" rel=subdocument>Q5.12. FFTW 2.1.3 crashes on AIX.</a>
37 </ul><hr>
38
39 <h2><A name="rfftwndbug">
40 Question 5.1. FFTW 1.1 crashes in rfftwnd on
41 Linux.
42 </A></h2>
43
44 This bug was fixed in FFTW 1.2. There was a bug in
45 <code>rfftwnd</code> causing an incorrect amount of memory to be allocated. The bug showed
46 up in Linux with libc-5.3.12 (and nowhere else that we know of).
47
48 <h2><A name="fftwmpibug">
49 Question 5.2. The MPI transforms in FFTW 1.2 give incorrect
50 results/leak memory.
51 </A></h2>
52
53 These bugs were corrected in FFTW 1.2.1. The MPI transforms (really,
54 just the transpose routines) in FFTW 1.2 had bugs that could cause
55 errors in some situations.
56 <h2><A name="testsingbug">
57 Question 5.3. The test programs in FFTW 1.2.1 fail when I change FFTW
58 to use single precision.
59 </A></h2>
60
61 This bug was fixed in FFTW 1.3. (Older versions of FFTW did
62 work in single precision, but the test programs didn't--the error
63 tolerances in the tests were set for double precision.)
64
65 <h2><A name="teststoobig">
66 Question 5.4. The test program in FFTW 1.2.1 fails for n &gt;
67 46340.
68 </A></h2>
69
70 This bug was fixed in FFTW 1.3. FFTW 1.2.1 produced the right answer,
71 but the test program was wrong. For large n, n*n in the naive
72 transform that we used for comparison overflows 32 bit integer
73 precision, breaking the test.
74 <h2><A name="linuxthreads">
75 Question 5.5. The threaded code fails on Linux Redhat
76 5.0
77 </A></h2>
78
79 We had problems with glibc-2.0.5. The code should work with
80 glibc-2.0.7.
81 <h2><A name="bigrfftwnd">
82 Question 5.6. FFTW 2.0's rfftwnd fails for rank &gt; 1 transforms
83 with a final dimension &gt;= 65536.
84 </A></h2>
85
86 This bug was fixed in FFTW 2.0.1. (There was a 32-bit integer
87 overflow due to a poorly-parenthesized expression.)
88 <h2><A name="primebug">
89 Question 5.7. FFTW 2.0's complex transforms give the wrong results
90 with prime factors 17 to 97.
91 </A></h2>
92
93 There was a bug in the complex transforms that could cause incorrect
94 results under (hopefully rare) circumstances for lengths with
95 intermediate-size prime factors (17-97). This bug was fixed in FFTW
96 2.1.1.
97 <h2><A name="mpichbug">
98 Question 5.8. FFTW 2.1.1's MPI test programs crash with
99 MPICH.
100 </A></h2>
101
102 This bug was fixed in FFTW 2.1.2. The 2.1/2.1.1 MPI test programs
103 crashed when using the MPICH implementation of MPI with the
104 <code>ch_p4</code> device (TCP/IP); the transforms themselves worked fine.
105
106 <h2><A name="aixthreadbug">
107 Question 5.9. FFTW 2.1.2's multi-threaded transforms don't work on
108 AIX.
109 </A></h2>
110
111 This bug was fixed in FFTW 2.1.3. The multi-threaded transforms in
112 previous versions didn't work with AIX's
113 <code>pthreads</code> implementation, which idiosyncratically creates threads in detached
114 (non-joinable) mode by default.
115 <h2><A name="bigprimebug">
116 Question 5.10. FFTW 2.1.2's complex transforms give incorrect results
117 for large prime sizes.
118 </A></h2>
119
120 This bug was fixed in FFTW 2.1.3. FFTW's complex-transform algorithm
121 for prime sizes (in versions 2.0 to 2.1.2) had an integer overflow
122 problem that caused incorrect results for many primes greater than
123 32768 (on 32-bit machines). (Sizes without large prime factors are
124 not affected.)
125 <h2><A name="solaristhreadbug">
126 Question 5.11. FFTW 2.1.3's multi-threaded transforms don't give any
127 speedup on Solaris.
128 </A></h2>
129
130 This bug was fixed in FFTW 2.1.4. (By default, Solaris creates
131 threads that do not parallelize over multiple processors, so one has
132 to request the proper behavior specifically.)
133
134 <h2><A name="aixflags">
135 Question 5.12. FFTW 2.1.3 crashes on AIX.
136 </A></h2>
137
138 The FFTW 2.1.3 <code>configure</code> script picked incorrect compiler flags for the <code>xlc</code> compiler on newer IBM processors. This
139 is fixed in FFTW 2.1.4. <hr>
140 Back: <a href="section4.html" rev=precedes>Internals of FFTW</a>.<br>
141 <a href="index.html" rev=subdocument>Return to contents</a>.<p>
142 <address>
143 <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
144 - 30 July 2016
145 </address><br>
146 Extracted from FFTW Frequently Asked Questions with Answers,
147 Copyright &copy; 2016 Matteo Frigo and Massachusetts Institute of Technology.
148 </body></html>