mirror of
https://git.notmuchmail.org/git/notmuch
synced 2024-12-12 04:24:51 +01:00
de23409704
Today someone asked me the (reasonable) question of how much performance impact there is from synching tags to maildir flags. It turns out it is noticeable, about a 50% overhead compared to non-synched tags (according to these tests). In practice I don't know if it's a big problem for users, since I don't know what fraction of tagging operations involve "special" tags. |
||
---|---|---|
.. | ||
download | ||
.gitignore | ||
M00-new.sh | ||
M01-dump-restore.sh | ||
M02-show.sh | ||
M03-search.sh | ||
M04-reply.sh | ||
M05-reindex.sh | ||
M06-insert.sh | ||
Makefile | ||
Makefile.local | ||
notmuch-memory-test | ||
notmuch-time-test | ||
perf-test-lib.sh | ||
README | ||
T00-new.sh | ||
T01-dump-restore.sh | ||
T02-tag.sh | ||
T03-reindex.sh | ||
T04-thread-subquery.sh | ||
T05-ruby.sh | ||
T06-emacs.sh | ||
T07-git.sh | ||
version.sh |
Performance Tests ----------------- This directory contains two kinds of performance tests: time tests, and memory tests. The former use gnu time, and the latter use valgrind. Pre-requisites -------------- In addition to having notmuch, you need: - gpg - gnu tar - gnu time (for the time tests) - xz. Some speedup can be gotten by installing "pixz", but this is probably only worthwhile if you are debugging the tests. - valgrind (for the memory tests) - perf (optional, for more fine-grained timing) Getting set up to run tests: ---------------------------- First, you need to get the corpus. If you don't already have the gpg key for David Bremner, run % gpg --locate-external-key 'david@tethera.net' This should get you a key with fingerprint 7A18 807F 100A 4570 C596 8420 7E4E 65C8 720B 706B (the last 8 digits are printed as the "key id"). To fetch the actual corpus it should work to run % make download-corpus In case that fails or is too slow, check https://notmuchmail.org/corpus for a list of mirrors. Running tests ------------- The easiest way to run performance tests is to say "make perf-test". This will run all time and memory tests. Be aware that the memory tests are quite time consuming when run on the full corpus, and that depending on your interests it may be more sensible to run "make time-test" or "make memory-test". You can also invoke one of the scripts notmuch-time-test or notmuch-memory-test or run a more specific subset of tests by simply invoking one of the executable scripts in this directory, (such as ./T00-new). Each test script supports the following arguments --small / --medium / --large Choose corpus size. --debug Enable debugging. In particular don't delete temporary directories. --perf Run perf record in place of /usr/bin/time. Perf output can be found in a log directory. --call-graph {fp,lbr,dwarf} Call graph option for perf record. Default is 'lbr'. When using the make targets, you can pass arguments to all test scripts by defining the make variable OPTIONS. Log Directory ------------- The memory tests, and the time tests when option '--perf' is given save their output in a directory named as follows log.$test_name-$corpus_size-$timestamp These directories are removed by "make clean". Writing tests ------------- Have a look at "T01-dump-restore" for an example time test and "M00-new" for an example memory test. In both cases sourcing "perf-test-lib.sh" is mandatory. Basics: - '(time|memory)_start' unpacks the mail corpus and calls notmuch new if it cannot find a cache of the appropriate corpus. - '(time|memory)_run' runs the command under time or valgrind. Currently "memory_run" does not support i/o redirection in the command. - '(time|memory)_done' does the cleanup; comment it out or pass --debug to the script to leave the temporary files around. Utility functions include - 'add_email_corpus' unpacks a set of messages and tags - 'cache_database': makes a snapshot of the current database - 'uncache_database': forces the next '(time|memory)_start' to rebuild the database. Scripts are run in the order specified in notmuch-perf-test. In the future this order might be chosen automatically so please follow the convention of starting the name with 'T' or 'M' followed by two digits to specify the order.