view tests/killdaemons.py @ 38781:0a57945aaf7f

manifest: persist the manifestfulltext cache

Reconstructing the manifest from the revlog takes time, so much so that there
already is a LRU cache to avoid having to load a manifest multiple times.

This patch persists that LRU cache in the .hg/cache directory, so we can re-use
this cache across hg commands. Commit benchmark (run on Macos 10.13 on a
2017-model Macbook Pro with Core i7 2.9GHz and flash drive), testing without
and with patch run 5 times, baseline is r2a227782e754:

* committing to an existing file, against the mozilla-central repository.
  Baseline real time average 1.9692, with patch 1.3786.

A new debugcommand "hg debugmanifestfulltextcache" lets you inspect the cache,
clear it, or add specific manifest nodeids to it. When calling
repo.updatecaches(), the manifest(s) for the working copy parents are added to
the cache.

The hg perfmanifest command has an additional --clear-disk switch to clear this
cache when testing manifest loading performance.

Using this command to test performance on the firefox repository for revision
f947d902ed91, whose manifest has a delta chain length of 60540, we see:

$ hg perfmanifest f947d902ed91 --clear-disk
! wall 0.972253 comb 0.970000 user 0.850000 sys 0.120000 (best of 10)
$ hg debugmanifestfulltextcache -a `hg log --debug -r f947d902ed91 | grep manifest | cut -d: -f3`
Cache contains 1 manifest entries, in order of most to least recent:
id: 0294517df4aad07c70701db43bc7ff24c3ce7dbc, size 25.6 MB
Total cache data size 25.6 MB, on-disk 0 bytes
$ hg perfmanifest f947d902ed91
! wall 0.036748 comb 0.040000 user 0.020000 sys 0.020000 (best of 100)

Worst-case scenario: a manifest text loaded from a single delta; in the firefox
repository manifest node 9a1246ff762e is the chain base for the manifest
attached to revision f947d902ed91. Loading this from a full cache file is just
as fast as without the cache; the extra node ids ensure a big full cache:

$ for node in 9a1246ff762e 1a1922c14a3e 54a31d11a36a 0294517df4aa; do
>   hgd debugmanifestfulltextcache -a $node > /dev/null
> done
$ hgd perfmanifest -m 9a1246ff762e
! wall 0.077513 comb 0.080000 user 0.030000 sys 0.050000 (best of 100)
$ hgd perfmanifest -m 9a1246ff762e --clear-disk
! wall 0.078547 comb 0.080000 user 0.070000 sys 0.010000 (best of 100)

#!/usr/bin/env python

from __future__ import absolute_import
import errno
import os
import signal
import sys
import time

if os.name =='nt':
    import ctypes

    _BOOL = ctypes.c_long
    _DWORD = ctypes.c_ulong
    _UINT = ctypes.c_uint
    _HANDLE = ctypes.c_void_p

    ctypes.windll.kernel32.CloseHandle.argtypes = [_HANDLE]
    ctypes.windll.kernel32.CloseHandle.restype = _BOOL

    ctypes.windll.kernel32.GetLastError.argtypes = []
    ctypes.windll.kernel32.GetLastError.restype = _DWORD

    ctypes.windll.kernel32.OpenProcess.argtypes = [_DWORD, _BOOL, _DWORD]
    ctypes.windll.kernel32.OpenProcess.restype = _HANDLE

    ctypes.windll.kernel32.TerminateProcess.argtypes = [_HANDLE, _UINT]
    ctypes.windll.kernel32.TerminateProcess.restype = _BOOL

    ctypes.windll.kernel32.WaitForSingleObject.argtypes = [_HANDLE, _DWORD]
    ctypes.windll.kernel32.WaitForSingleObject.restype = _DWORD

    def _check(ret, expectederr=None):
        if ret == 0:
            winerrno = ctypes.GetLastError()
            if winerrno == expectederr:
                return True
            raise ctypes.WinError(winerrno)

    def kill(pid, logfn, tryhard=True):
        logfn('# Killing daemon process %d' % pid)
        PROCESS_TERMINATE = 1
        PROCESS_QUERY_INFORMATION = 0x400
        SYNCHRONIZE = 0x00100000
        WAIT_OBJECT_0 = 0
        WAIT_TIMEOUT = 258
        WAIT_FAILED = _DWORD(0xFFFFFFFF).value
        handle = ctypes.windll.kernel32.OpenProcess(
                PROCESS_TERMINATE|SYNCHRONIZE|PROCESS_QUERY_INFORMATION,
                False, pid)
        if handle is None:
            _check(0, 87) # err 87 when process not found
            return # process not found, already finished
        try:
            r = ctypes.windll.kernel32.WaitForSingleObject(handle, 100)
            if r == WAIT_OBJECT_0:
                pass # terminated, but process handle still available
            elif r == WAIT_TIMEOUT:
                _check(ctypes.windll.kernel32.TerminateProcess(handle, -1))
            elif r == WAIT_FAILED:
                _check(0)  # err stored in GetLastError()

            # TODO?: forcefully kill when timeout
            #        and ?shorter waiting time? when tryhard==True
            r = ctypes.windll.kernel32.WaitForSingleObject(handle, 100)
                                                       # timeout = 100 ms
            if r == WAIT_OBJECT_0:
                pass # process is terminated
            elif r == WAIT_TIMEOUT:
                logfn('# Daemon process %d is stuck')
            elif r == WAIT_FAILED:
                _check(0)  # err stored in GetLastError()
        except: #re-raises
            ctypes.windll.kernel32.CloseHandle(handle) # no _check, keep error
            raise
        _check(ctypes.windll.kernel32.CloseHandle(handle))

else:
    def kill(pid, logfn, tryhard=True):
        try:
            os.kill(pid, 0)
            logfn('# Killing daemon process %d' % pid)
            os.kill(pid, signal.SIGTERM)
            if tryhard:
                for i in range(10):
                    time.sleep(0.05)
                    os.kill(pid, 0)
            else:
                time.sleep(0.1)
                os.kill(pid, 0)
            logfn('# Daemon process %d is stuck - really killing it' % pid)
            os.kill(pid, signal.SIGKILL)
        except OSError as err:
            if err.errno != errno.ESRCH:
                raise

def killdaemons(pidfile, tryhard=True, remove=False, logfn=None):
    if not logfn:
        logfn = lambda s: s
    # Kill off any leftover daemon processes
    try:
        pids = []
        with open(pidfile) as fp:
            for line in fp:
                try:
                    pid = int(line)
                    if pid <= 0:
                        raise ValueError
                except ValueError:
                    logfn('# Not killing daemon process %s - invalid pid'
                          % line.rstrip())
                    continue
                pids.append(pid)
        for pid in pids:
            kill(pid, logfn, tryhard)
        if remove:
            os.unlink(pidfile)
    except IOError:
        pass

if __name__ == '__main__':
    if len(sys.argv) > 1:
        path, = sys.argv[1:]
    else:
        path = os.environ["DAEMON_PIDS"]

    killdaemons(path, remove=True)