Mercurial > hg > audiodb
view tests/0042/run-test.sh @ 400:8c7453fb5bd9 api-inversion
Invert audioDB::power_flag / audiodb_power()
Here the exciting discovery is that the mmap(), memcpy(), munmap()
sequence is in fact not safe. In principle an msync() call should be
inserted before unmapping for in-core changes to mmap()ed files to be
flushed to disk.
In this case we work around the problem entirely, by not mmap()ing
anything and doing everything with file descriptors. Amusingly, that's
probably not desperately safe either, this time because we have to move
the file descriptor position (which is also a shared resource). dup()
doesn't save us, as the duplicate file descriptor shares a file
position.
This applies also to the filling of data_buffer in the query loop, and
in fact basically any call to lseek(), which is why I'm not fixing it
now. Solution: if you have multiple threads all acting at once on a
single database, do one audiodb_open() per thread, for now at least.
author | mas01cr |
---|---|
date | Thu, 27 Nov 2008 16:22:52 +0000 |
parents | da901c62e569 |
children |
line wrap: on
line source
#! /bin/bash . ../test-utils.sh if [ -f testdb ]; then rm -f testdb; fi # Make LARGE_ADB ${AUDIODB} -d testdb -N --ntracks 50000 ${AUDIODB} -d testdb -P ${AUDIODB} -d testdb -L ${AUDIODB} -d testdb -S | grep "flags:" > testoutput echo "flags: l2norm[on] minmax[off] power[on] times[off] largeADB[on]" > test-expected-output cmp testoutput test-expected-output intstring 2 > testfeature floatstring 0 1 >> testfeature floatstring 1 0 >> testfeature floatstring 1 0 >> testfeature floatstring 0 1 >> testfeature intstring 1 > testpower floatstring -0.5 >> testpower floatstring -1 >> testpower floatstring -1 >> testpower floatstring -0.5 >> testpower echo testfeature > testList.txt echo testpower > pwrList.txt echo key1 > keyList.txt echo testfeature >> testList.txt echo testpower >> pwrList.txt echo key2 >> keyList.txt ${AUDIODB} -d testdb -B -F testList.txt -W pwrList.txt -K keyList.txt # Cleanup rm testList.txt pwrList.txt keyList.txt echo "query points (0.0,0.5),(0.0,0.5),(0.5,0.0)" intstring 2 > testquery floatstring 0 0.5 >> testquery floatstring 0 0.5 >> testquery floatstring 0.5 0 >> testquery # LARGE_ADB requires an INDEX ${AUDIODB} -d testdb -X -R 1 -l 1 # LARGE_ADB query from key ${AUDIODB} -d testdb -Q sequence -l 1 -k key1 -R 1 --absolute-threshold -4.5 > testoutput echo key2 1 > test-expected-output cmp testoutput test-expected-output # LARGE_ADB query from feature file and power file ${AUDIODB} -d testdb -Q sequence -l 1 -f testquery -w testpower -R 1 --absolute-threshold -4.5 > testoutput echo key1 1 > test-expected-output echo key2 1 >> test-expected-output cmp testoutput test-expected-output # WS WSPORT=10020 start_server ${AUDIODB} ${WSPORT} # LARGE_ADB WS query from key ${AUDIODB} -c localhost:${WSPORT} -d testdb -Q sequence -l 1 -k key1 -R 1 --absolute-threshold -4.5 -n 1 --lsh_exact > testoutput echo key2 0 0 0 > test-expected-output cmp testoutput test-expected-output # LARGE_ADB WS query from feature file and power file tests ${AUDIODB} -c localhost:${WSPORT} -d testdb -Q sequence -l 1 -f testquery -w testpower -R 1 --absolute-threshold -4.5 -n 1 > testoutput echo key1 1 > test-expected-output echo key2 1 >> test-expected-output cmp testoutput test-expected-output stop_server $! exit 104