cext: fix memory leak in phases computation
Without this a buffer whose size in bytes is the number of
changesets in the repository is leaked each time the repository is
opened and changeset phases are computed.
Impact: the current code in hgwebdir creates a new `localrepository`
instance for each HTTP request. Since any pull or push is made of several
requests, a team of 100 people can easily produce thousands of such
requests per day.
Being a low-level malloc, this leak can't be seen with the gc module and
tools relying on that, but was spotted by valgrind immediately.
Reproduction
------------
for i in range(cl_args.iterations):
repo = hg.repository(baseui, repo_path)
rev = repo.revs(rev).first()
ctx = repo[rev]
del ctx
del repo
# avoid any pollution by other type of leak
# (that should be fixed in 5.8)
repoview._filteredrepotypes.clear()
gc.collect()
Measurements
------------
Resident Set Size (RSS), taken on a clone of
mozilla-central for performance analysis (440 000
changesets).
before:
5.8+hg19.5ac0f2a8ba72 1000 iterations: 1606MB
5.8+hg19.5ac0f2a8ba72 10000 iterations: 5723MB
after:
5.8+hg20.e2084d39e145 1000 iterations: 555MB
5.8+hg20.e2084d39e145 10000 iterations: 555MB
(double checked, not a copy/paste error)
(e2084d39e14 is the present changeset, before amendment
of the message to add the measurements)
$ hg init test
$ cd test
$ hg debugbuilddag '+2'
$ hg phase --public 0
$ hg serve -p $HGPORT -d --pid-file=hg.pid -E errors.log
$ cat hg.pid >> $DAEMON_PIDS
$ cd ..
$ hg init test2
$ cd test2
$ hg incoming http://foo:xyzzy@localhost:$HGPORT/
comparing with http://foo:***@localhost:$HGPORT/
changeset: 0:1ea73414a91b
user: debugbuilddag
date: Thu Jan 01 00:00:00 1970 +0000
summary: r0
changeset: 1:66f7d451a68b
tag: tip
user: debugbuilddag
date: Thu Jan 01 00:00:01 1970 +0000
summary: r1
$ killdaemons.py
$ cd ..
$ hg -R test --config server.view=immutable serve -p $HGPORT -d --pid-file=hg.pid -E errors.log
$ cat hg.pid >> $DAEMON_PIDS
$ hg -R test2 incoming http://foo:xyzzy@localhost:$HGPORT/
comparing with http://foo:***@localhost:$HGPORT/
changeset: 0:1ea73414a91b
tag: tip
user: debugbuilddag
date: Thu Jan 01 00:00:00 1970 +0000
summary: r0
Check same result using `experimental.extra-filter-revs`
$ hg -R test --config experimental.extra-filter-revs='not public()' serve -p $HGPORT1 -d --pid-file=hg2.pid -E errors.log
$ cat hg2.pid >> $DAEMON_PIDS
$ hg -R test2 incoming http://foo:xyzzy@localhost:$HGPORT1/
comparing with http://foo:***@localhost:$HGPORT1/
changeset: 0:1ea73414a91b
tag: tip
user: debugbuilddag
date: Thu Jan 01 00:00:00 1970 +0000
summary: r0
$ hg -R test --config experimental.extra-filter-revs='not public()' debugupdatecache
$ ls -1 test/.hg/cache/
branch2-base%89c45d2fa07e
branch2-immutable%89c45d2fa07e
branch2-served
branch2-served%89c45d2fa07e
branch2-served.hidden%89c45d2fa07e
branch2-visible%89c45d2fa07e
branch2-visible-hidden%89c45d2fa07e
hgtagsfnodes1
rbc-names-v1
rbc-revs-v1
tags2
tags2-served%89c45d2fa07e
cleanup
$ cat errors.log
$ killdaemons.py