contrib/hgclient.py

streamclone: use backgroundfilecloser (issue4889)

Closing files that have been appended to is slow on Windows/NTFS.
CloseHandle() calls on this platform often take 1-10ms - and that's
on my i7-6700K Skylake processor with a modern and fast SSD. Contrast
with other I/O operations, such as writing data, which take <100us.

This means that creating/appending thousands of files can add
significant overhead. For example, cloning mozilla-central creates
~232,000 revlog files. Assuming 1ms per CloseHandle(), that yields
232s (3:52) of wall time waiting for file closes!

The impact of this overhead can be measured most directly when applying
stream clone bundles. Applying these files is effectively uncompressing
a tar archive (read: it's very fast).

Using a RAM disk (read: no I/O wait), the difference in wall time for a
`hg debugapplystreamclonebundle` for a ~1731 MB mozilla-central bundle
between Windows and Linux from the same machine is drastic:

Linux:    ~12.8s (128MB/s)
Windows: ~352.0s (4.7MB/s)

Windows is ~27.5x slower. Yikes!

After this patch:

Linux:    ~12.8s (128MB/s)
Windows: ~102.1s (16.1MB/s)

Windows is now ~3.4x faster. Unfortunately, it is still ~8x slower than
Linux. Profiling reveals a few hot code paths that could likely be
improved. But those are for other patches.

This patch introduces test-clone-uncompressed.t because existing tests
of `clone --uncompressed` are scattered about and adding a variation for
background thread closing to e.g. test-http.t doesn't feel correct.

# A minimal client for Mercurial's command server

import os, sys, signal, struct, socket, subprocess, time, cStringIO

def connectpipe(path=None):
    cmdline = ['hg', 'serve', '--cmdserver', 'pipe']
    if path:
        cmdline += ['-R', path]

    server = subprocess.Popen(cmdline, stdin=subprocess.PIPE,
                              stdout=subprocess.PIPE)

    return server

class unixconnection(object):
    def __init__(self, sockpath):
        self.sock = sock = socket.socket(socket.AF_UNIX)
        sock.connect(sockpath)
        self.stdin = sock.makefile('wb')
        self.stdout = sock.makefile('rb')

    def wait(self):
        self.stdin.close()
        self.stdout.close()
        self.sock.close()

class unixserver(object):
    def __init__(self, sockpath, logpath=None, repopath=None):
        self.sockpath = sockpath
        cmdline = ['hg', 'serve', '--cmdserver', 'unix', '-a', sockpath]
        if repopath:
            cmdline += ['-R', repopath]
        if logpath:
            stdout = open(logpath, 'a')
            stderr = subprocess.STDOUT
        else:
            stdout = stderr = None
        self.server = subprocess.Popen(cmdline, stdout=stdout, stderr=stderr)
        # wait for listen()
        while self.server.poll() is None:
            if os.path.exists(sockpath):
                break
            time.sleep(0.1)

    def connect(self):
        return unixconnection(self.sockpath)

    def shutdown(self):
        os.kill(self.server.pid, signal.SIGTERM)
        self.server.wait()

def writeblock(server, data):
    server.stdin.write(struct.pack('>I', len(data)))
    server.stdin.write(data)
    server.stdin.flush()

def readchannel(server):
    data = server.stdout.read(5)
    if not data:
        raise EOFError
    channel, length = struct.unpack('>cI', data)
    if channel in 'IL':
        return channel, length
    else:
        return channel, server.stdout.read(length)

def sep(text):
    return text.replace('\\', '/')

def runcommand(server, args, output=sys.stdout, error=sys.stderr, input=None,
               outfilter=lambda x: x):
    print '*** runcommand', ' '.join(args)
    sys.stdout.flush()
    server.stdin.write('runcommand\n')
    writeblock(server, '\0'.join(args))

    if not input:
        input = cStringIO.StringIO()

    while True:
        ch, data = readchannel(server)
        if ch == 'o':
            output.write(outfilter(data))
            output.flush()
        elif ch == 'e':
            error.write(data)
            error.flush()
        elif ch == 'I':
            writeblock(server, input.read(data))
        elif ch == 'L':
            writeblock(server, input.readline(data))
        elif ch == 'r':
            ret, = struct.unpack('>i', data)
            if ret != 0:
                print ' [%d]' % ret
            return ret
        else:
            print "unexpected channel %c: %r" % (ch, data)
            if ch.isupper():
                return

def check(func, connect=connectpipe):
    sys.stdout.flush()
    server = connect()
    try:
        return func(server)
    finally:
        server.stdin.close()
        server.wait()