Mercurial > hg
view mercurial/lsprofcalltree.py @ 38732:be4984261611
merge: mark file gets as not thread safe (issue5933)
In default installs, this has the effect of disabling the thread-based
worker on Windows when manifesting files in the working directory. My
measurements have shown that with revlog-based repositories, Mercurial
spends a lot of CPU time in revlog code resolving file data. This ends
up incurring a lot of context switching across threads and slows down
`hg update` operations when going from an empty working directory to
the tip of the repo.
On mozilla-unified (246,351 files) on an i7-6700K (4+4 CPUs):
before: 487s wall
after: 360s wall (equivalent to worker.enabled=false)
cpus=2: 379s wall
Even with only 2 threads, the thread pool is still slower.
The introduction of the thread-based worker (02b36e860e0b) states that
it resulted in a "~50%" speedup for `hg sparse --enable-profile` and
`hg sparse --disable-profile`. This disagrees with my measurement
above. I theorize a few reasons for this:
1) Removal of files from the working directory is I/O - not CPU - bound
and should benefit from a thread pool (unless I/O is insanely fast
and the GIL release is near instantaneous). So tests like `hg sparse
--enable-profile` may exercise deletion throughput and aren't good
benchmarks for worker tasks that are CPU heavy.
2) The patch was authored by someone at Facebook. The results were
likely measured against a repository using remotefilelog. And I
believe that revision retrieval during working directory updates with
remotefilelog will often use a remote store, thus being I/O and not
CPU bound. This probably resulted in an overstated performance gain.
Since there appears to be a need to enable the thread-based worker with
some stores, I've made the flagging of file gets as thread safe
configurable. I've made it experimental because I don't want to formalize
a boolean flag for this option and because this attribute is best
captured against the store implementation. But we don't have a proper
store API for this yet. I'd rather cross this bridge later.
It is possible there are revlog-based repositories that do benefit from
a thread-based worker. I didn't do very comprehensive testing. If there
are, we may want to devise a more proper algorithm for whether to use
the thread-based worker, including possibly config options to limit the
number of threads to use. But until I see evidence that justifies
complexity, simplicity wins.
Differential Revision: https://phab.mercurial-scm.org/D3963
author | Gregory Szorc <gregory.szorc@gmail.com> |
---|---|
date | Wed, 18 Jul 2018 09:49:34 -0700 |
parents | 5a988b3c9645 |
children | 1ae0faa14797 |
line wrap: on
line source
""" lsprofcalltree.py - lsprof output which is readable by kcachegrind Authors: * David Allouche <david <at> allouche.net> * Jp Calderone & Itamar Shtull-Trauring * Johan Dahlin This software may be used and distributed according to the terms of the GNU General Public License, incorporated herein by reference. """ from __future__ import absolute_import, print_function def label(code): if isinstance(code, str): return '~' + code # built-in functions ('~' sorts at the end) else: return '%s %s:%d' % (code.co_name, code.co_filename, code.co_firstlineno) class KCacheGrind(object): def __init__(self, profiler): self.data = profiler.getstats() self.out_file = None def output(self, out_file): self.out_file = out_file print('events: Ticks', file=out_file) self._print_summary() for entry in self.data: self._entry(entry) def _print_summary(self): max_cost = 0 for entry in self.data: totaltime = int(entry.totaltime * 1000) max_cost = max(max_cost, totaltime) print('summary: %d' % max_cost, file=self.out_file) def _entry(self, entry): out_file = self.out_file code = entry.code if isinstance(code, str): print('fi=~', file=out_file) else: print('fi=%s' % code.co_filename, file=out_file) print('fn=%s' % label(code), file=out_file) inlinetime = int(entry.inlinetime * 1000) if isinstance(code, str): print('0 ', inlinetime, file=out_file) else: print('%d %d' % (code.co_firstlineno, inlinetime), file=out_file) # recursive calls are counted in entry.calls if entry.calls: calls = entry.calls else: calls = [] if isinstance(code, str): lineno = 0 else: lineno = code.co_firstlineno for subentry in calls: self._subentry(lineno, subentry) print(file=out_file) def _subentry(self, lineno, subentry): out_file = self.out_file code = subentry.code print('cfn=%s' % label(code), file=out_file) if isinstance(code, str): print('cfi=~', file=out_file) print('calls=%d 0' % subentry.callcount, file=out_file) else: print('cfi=%s' % code.co_filename, file=out_file) print('calls=%d %d' % ( subentry.callcount, code.co_firstlineno), file=out_file) totaltime = int(subentry.totaltime * 1000) print('%d %d' % (lineno, totaltime), file=out_file)