phases: rework the logic of _pushdiscoveryphase to bound complexity
This rework the various graph traversal in _pushdiscoveryphase to keep the
complexity in check.
This is done though a couple of things:
- first, limiting the space we have to explore, for example, if we are not in
publishing push, we don't need to consider remote draft roots that are also
draft locally, as there is nothing to be moved there.
- avoid unbounded descendant computation, and use the faster "rev between"
computation.
This provide a massive boost to performance when exchanging with repository with
a massive amount of draft, like mozilla-try:
### data-env-vars.name = mozilla-try-2023-03-22-zstd-sparse-revlog
# benchmark.name = hg.command.push
# bin-env-vars.hg.flavor = default
# bin-env-vars.hg.py-re2-module = default
# benchmark.variants.explicit-rev = all-out-heads
# benchmark.variants.issue6528 = disabled
# benchmark.variants.protocol = ssh
# benchmark.variants.reuse-external-delta-parent = default
## benchmark.variants.revs = any-1-extra-rev
before: 20.346590 seconds
after: 11.232059 seconds (-38.15%, -7.48 seconds)
## benchmark.variants.revs = any-100-extra-rev
before: 24.752051 seconds
after: 15.367412 seconds (-37.91%, -9.38 seconds)
After this changes, the push operation is still quite too slow. Some of this
can be attributed to general phases slowness (reading all the roots from disk
for example) and other know slowness (not using persistent-nodemap, branchmap,
tags, etc. We are also working on them, but with this series, phase discovery
during push no longer showing up in profile and this is a pretty nice and bit
low-hanging fruit out of the way.
### (same case as the above)
# benchmark.variants.revs = any-1-extra-rev
pre-%ln-change: 44.235070
this-changeset: 11.232059 seconds (-74.61%, -33.00 seconds)
# benchmark.variants.revs = any-100-extra-rev
pre-%ln-change: 49.234697
this-changeset: 15.367412 seconds (-68.79%, -33.87 seconds)
Note that with this change, the `hg push` performance is now much closer to the
`hg pull` performance, even it still lagging behind a bit. (and the overall
performance are still too slow).
### data-env-vars.name = mozilla-try-2023-03-22-ds2-pnm
# benchmark.variants.explicit-rev = all-out-heads
# benchmark.variants.issue6528 = disabled
# benchmark.variants.protocol = ssh
# benchmark.variants.pulled-delta-reuse-policy = default
# bin-env-vars.hg.flavor = rust
## benchmark.variants.revs = any-1-extra-rev
hg.command.pull: 6.517450
hg.command.push: 11.219888
## benchmark.variants.revs = any-100-extra-rev
hg.command.pull: 10.160991
hg.command.push: 14.251107
### data-env-vars.name = mozilla-try-2023-03-22-zstd-sparse-revlog
# bin-env-vars.hg.py-re2-module = default
# benchmark.variants.explicit-rev = all-out-heads
# benchmark.variants.issue6528 = disabled
# benchmark.variants.protocol = ssh
# benchmark.variants.pulled-delta-reuse-policy = default
## bin-env-vars.hg.flavor = default
## benchmark.variants.revs = any-1-extra-rev
hg.command.pull: 8.577772
hg.command.push: 11.232059
## bin-env-vars.hg.flavor = default
## benchmark.variants.revs = any-100-extra-rev
hg.command.pull: 13.152976
hg.command.push: 15.367412
## bin-env-vars.hg.flavor = rust
## benchmark.variants.revs = any-1-extra-rev
hg.command.pull: 8.731982
hg.command.push: 11.178751
## bin-env-vars.hg.flavor = rust
## benchmark.variants.revs = any-100-extra-rev
hg.command.pull: 13.184236
hg.command.push: 15.620843
# A minimal client for Mercurial's command server
import io
import os
import re
import signal
import socket
import struct
import subprocess
import sys
import time
if sys.version_info[0] >= 3:
stdout = sys.stdout.buffer
stderr = sys.stderr.buffer
stringio = io.BytesIO
def bprint(*args):
# remove b'' as well for ease of test migration
pargs = [re.sub(br'''\bb(['"])''', br'\1', b'%s' % a) for a in args]
stdout.write(b' '.join(pargs) + b'\n')
else:
import cStringIO
stdout = sys.stdout
stderr = sys.stderr
stringio = cStringIO.StringIO
bprint = print
def connectpipe(path=None, extraargs=()):
cmdline = [b'hg', b'serve', b'--cmdserver', b'pipe']
if path:
cmdline += [b'-R', path]
cmdline.extend(extraargs)
def tonative(cmdline):
if os.name != 'nt':
return cmdline
return [arg.decode("utf-8") for arg in cmdline]
server = subprocess.Popen(
tonative(cmdline), stdin=subprocess.PIPE, stdout=subprocess.PIPE
)
return server
class unixconnection:
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:
def __init__(self, sockpath, logpath=None, repopath=None):
self.sockpath = sockpath
cmdline = [b'hg', b'serve', b'--cmdserver', b'unix', b'-a', sockpath]
if repopath:
cmdline += [b'-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(b'>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 b'IL':
return channel, length
else:
return channel, server.stdout.read(length)
def sep(text):
return text.replace(b'\\', b'/')
def runcommand(
server, args, output=stdout, error=stderr, input=None, outfilter=lambda x: x
):
bprint(b'*** runcommand', b' '.join(args))
stdout.flush()
server.stdin.write(b'runcommand\n')
writeblock(server, b'\0'.join(args))
if not input:
input = stringio()
while True:
ch, data = readchannel(server)
if ch == b'o':
output.write(outfilter(data))
output.flush()
elif ch == b'e':
error.write(data)
error.flush()
elif ch == b'I':
writeblock(server, input.read(data))
elif ch == b'L':
writeblock(server, input.readline(data))
elif ch == b'm':
bprint(b"message: %r" % data)
elif ch == b'r':
(ret,) = struct.unpack('>i', data)
if ret != 0:
bprint(b' [%d]' % ret)
return ret
else:
bprint(b"unexpected channel %c: %r" % (ch, data))
if ch.isupper():
return
def check(func, connect=connectpipe):
stdout.flush()
server = connect()
try:
return func(server)
finally:
server.stdin.close()
server.wait()
def checkwith(connect=connectpipe, **kwargs):
def wrap(func):
return check(func, lambda: connect(**kwargs))
return wrap