changegroup: remove reordering control (BC)
This logic - including the experimental bundle.reorder option -
was originally added in
a8e3931e3fb5 in 2011 and then later ported
to changegroup.py.
The intent of this option and associated logic is to control
the ordering of revisions in deltagroups in changegroups. At the
time it was implemented, only changegroup version 1 existed
and generaldelta revlogs were just coming into the world. Changegroup
version 1 requires that deltas be made against the last revision
sent over the wire. Used with generaldelta, this created an
impedance mismatch of sorts and resulted in changegroup producers
spending a lot of time recomputing deltas.
Revision reordering was introduced so outgoing revisions would be
sent in "generaldelta order" and producers would be able to
reuse internal deltas from storage.
Later on, we introduced changegroup version 2. It supported denoting
which revision a delta was against. So we no longer needed to
sort outgoing revisions to ensure optimal delta generation from the
producer. So, subsequent changegroup versions disabled reordering.
We also later made the changelog not store deltas by default. And
we also made the changelog send out deltas in storage order. Why we
do this for changelog, I'm not sure. Maybe we want to preserve revision
order across clones? It doesn't really matter for this commit.
Fast forward to 2018. We want to abstract storage backends. And having
changegroup code require knowledge about how deltas are stored
internally interferes with that goal.
This commit removes reordering control from changegroup generation.
After this commit, the reordering behavior is:
* The changelog is always sent out in storage order (no behavior
change).
* Non-changelog generaldelta revlogs are reordered to always be in DAG
topological order (previously, generaldelta revlogs would be emitted
in storage order for version 2 and 3 changegroups).
* Non-changelog non-generaldelta revlogs are sent in storage order (no
behavior change).
* There exists no config option to override behavior.
The big difference here is that generaldelta revlogs now *always* have
their revisions sorted in DAG order before going out over the wire. This
behavior was previously only done for changegroup version 1. Version 2
and version 3 changegroups disabled reordering because the interchange
format supported encoding arbitrary delta parents, so reordering wasn't
strictly necessary.
I can think of a few significant implications for this change.
Because changegroup receivers will now see non-changelog revisions
in DAG order instead of storage order, the internal storage order of
manifests and files may differ substantially between producer and
consumer. I don't think this matters that much, since the storage
order of manifests and files is largely hidden from users. Only
the storage order of changelog matters (because `hg log` shows the
changelog in storage order). I don't think there should be any
controversy here.
The reordering of revisions has implications for changegroup producers.
Previously, generaldelta revlogs would be emitted in storage order.
And in the common case, the internally-stored delta could effectively
be copied from disk into the deltagroup delta. This meant that emitting
delta groups for generaldelta revlogs would be mostly linear read I/O.
This is desirable for performance. With us now reordering generaldelta
revlog revisions in DAG order, the read operations may use more random
I/O instead of sequential I/O. This could result in performance
loss. But with the prevalence of SSDs and fast random I/O, I'm not
too worried. (Note: the optimal emission order for revlogs is actually
delta encoding order. But the changegroup code wasn't doing that before
or after this change. We could potentially implement that in a later
commit.)
Changegroups in DAG order will have implications for receivers.
Previously, receiving storage order might mean seeing a number of
interleaved branches. This would mean long delta chains, sparse
I/O, and possibly more fulltext revisions instead of deltas, blowing
up storage storage. (This is the same set of problems that sparse
revlogs aims to address.) With the producer now sending revisions in DAG
order, the receiver also stores revisions in DAG order. That means
revisions for the same DAG branch are all grouped together. And this
should yield better storage outcomes. In other words, sending the
reordered changegroup allows the receiver to have better storage
order and for the producer to not propagate its (possibly sub-optimal)
internal storage order.
On the mozilla-unified repository, this change influences bundle
generation:
$ hg bundle -t none-v2 -a
before: time: real 355.680 secs (user 256.790+0.000 sys 16.820+0.000)
after: time: real 382.950 secs (user 281.700+0.000 sys 17.690+0.000)
before: 7,150,228,967 bytes (uncompressed)
after: 7,041,556,273 bytes (uncompressed)
before: 1,669,063,234 bytes (zstd l=3)
after: 1,628,598,830 bytes (zstd l=3)
$ hg unbundle
before: time: real 511.910 secs (user 466.750+0.000 sys 32.680+0.000)
after: time: real 487.790 secs (user 443.940+0.000 sys 30.840+0.000)
00manifest.d size:
source: 274,924,292 bytes
before: 304,741,626 bytes
after: 245,252,087 bytes
.hg/store total file size:
source: 2,649,133,490
before: 2,680,888,130
after: 2,627,875,673
We see the bundle size drop. That's probably because if a revlog
internally isn't storing a delta, it will choose to delta against
the last emitted revision. And on repos with interleaved branches
(like mozilla-unified), the previous revision could be an unrelated
branch and therefore be a large delta. But with this patch, the
previous revision is likely p1 or p2 and a delta should be small.
We also see the manifest size drop by ~50 MB. It's worth noting that
the manifest actually *increased* in size by ~25 MB in the old
strategy and decreased ~25 MB from its source in the new strategy.
Again, my explanation for this is that the DAG ordering in the
changegroup is resulting in better grouping of revisions in the
receiver, which results in more compact delta chains and higher
storage efficiency.
Unbundle time also dropped. I suspect this is due to the revlog having
to work less to compute deltas since the incoming deltas are more
optimal. i.e. the receiver spends less time resolving fulltext
revisions as incoming deltas bounce around between DAG branches and
delta chains.
We also see bundle generation time increase. This is not desirable.
However, the regression is only significant on the original repository:
if we generate a bundle from the repository created from the new,
always reordered bundles, we're close to baseline (if not at it with
expected noise):
$ hg bundle -t none-v2 -a
before (original): time: real 355.680 secs (user 256.790+0.000 sys 16.820+0.000)
after (original): time: real 382.950 secs (user 281.700+0.000 sys 17.690+0.000)
after (new repo): time: real 362.280 secs (user 260.300+0.000 sys 17.700+0.000)
This regression is a bit worrying because it will impact serving
canonical repositories (that don't have optimal internal storage
unless they are reordered - possibly as part of running
`hg debugupgraderepo`). However, this regression will only be
noticed by very large changegroups. And I'm guessing/hoping that
any repository that large is using clonebundles to mitigate server
load.
Again, sending DAG order isn't the optimal send order for servers:
sending in storage-delta order is. But in order to enable
storage-optimal send order, we'll need a storage API that handles
sorting. Future commits will introduce such an API.
Differential Revision: https://phab.mercurial-scm.org/D4721
# badserverext.py - Extension making servers behave badly
#
# Copyright 2017 Gregory Szorc <gregory.szorc@gmail.com>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
# no-check-code
"""Extension to make servers behave badly.
This extension is useful for testing Mercurial behavior when various network
events occur.
Various config options in the [badserver] section influence behavior:
closebeforeaccept
If true, close() the server socket when a new connection arrives before
accept() is called. The server will then exit.
closeafteraccept
If true, the server will close() the client socket immediately after
accept().
closeafterrecvbytes
If defined, close the client socket after receiving this many bytes.
closeaftersendbytes
If defined, close the client socket after sending this many bytes.
"""
from __future__ import absolute_import
import socket
from mercurial import(
registrar,
)
from mercurial.hgweb import (
server,
)
configtable = {}
configitem = registrar.configitem(configtable)
configitem(b'badserver', b'closeafteraccept',
default=False,
)
configitem(b'badserver', b'closeafterrecvbytes',
default='0',
)
configitem(b'badserver', b'closeaftersendbytes',
default='0',
)
configitem(b'badserver', b'closebeforeaccept',
default=False,
)
# We can't adjust __class__ on a socket instance. So we define a proxy type.
class socketproxy(object):
__slots__ = (
'_orig',
'_logfp',
'_closeafterrecvbytes',
'_closeaftersendbytes',
)
def __init__(self, obj, logfp, closeafterrecvbytes=0,
closeaftersendbytes=0):
object.__setattr__(self, '_orig', obj)
object.__setattr__(self, '_logfp', logfp)
object.__setattr__(self, '_closeafterrecvbytes', closeafterrecvbytes)
object.__setattr__(self, '_closeaftersendbytes', closeaftersendbytes)
def __getattribute__(self, name):
if name in ('makefile',):
return object.__getattribute__(self, name)
return getattr(object.__getattribute__(self, '_orig'), name)
def __delattr__(self, name):
delattr(object.__getattribute__(self, '_orig'), name)
def __setattr__(self, name, value):
setattr(object.__getattribute__(self, '_orig'), name, value)
def makefile(self, mode, bufsize):
f = object.__getattribute__(self, '_orig').makefile(mode, bufsize)
logfp = object.__getattribute__(self, '_logfp')
closeafterrecvbytes = object.__getattribute__(self,
'_closeafterrecvbytes')
closeaftersendbytes = object.__getattribute__(self,
'_closeaftersendbytes')
return fileobjectproxy(f, logfp,
closeafterrecvbytes=closeafterrecvbytes,
closeaftersendbytes=closeaftersendbytes)
# We can't adjust __class__ on socket._fileobject, so define a proxy.
class fileobjectproxy(object):
__slots__ = (
'_orig',
'_logfp',
'_closeafterrecvbytes',
'_closeaftersendbytes',
)
def __init__(self, obj, logfp, closeafterrecvbytes=0,
closeaftersendbytes=0):
object.__setattr__(self, '_orig', obj)
object.__setattr__(self, '_logfp', logfp)
object.__setattr__(self, '_closeafterrecvbytes', closeafterrecvbytes)
object.__setattr__(self, '_closeaftersendbytes', closeaftersendbytes)
def __getattribute__(self, name):
if name in ('read', 'readline', 'write', '_writelog'):
return object.__getattribute__(self, name)
return getattr(object.__getattribute__(self, '_orig'), name)
def __delattr__(self, name):
delattr(object.__getattribute__(self, '_orig'), name)
def __setattr__(self, name, value):
setattr(object.__getattribute__(self, '_orig'), name, value)
def _writelog(self, msg):
msg = msg.replace('\r', '\\r').replace('\n', '\\n')
object.__getattribute__(self, '_logfp').write(msg)
object.__getattribute__(self, '_logfp').write('\n')
object.__getattribute__(self, '_logfp').flush()
def read(self, size=-1):
remaining = object.__getattribute__(self, '_closeafterrecvbytes')
# No read limit. Call original function.
if not remaining:
result = object.__getattribute__(self, '_orig').read(size)
self._writelog('read(%d) -> (%d) (%s) %s' % (size,
len(result),
result))
return result
origsize = size
if size < 0:
size = remaining
else:
size = min(remaining, size)
result = object.__getattribute__(self, '_orig').read(size)
remaining -= len(result)
self._writelog('read(%d from %d) -> (%d) %s' % (
size, origsize, len(result), result))
object.__setattr__(self, '_closeafterrecvbytes', remaining)
if remaining <= 0:
self._writelog('read limit reached, closing socket')
self._sock.close()
# This is the easiest way to abort the current request.
raise Exception('connection closed after receiving N bytes')
return result
def readline(self, size=-1):
remaining = object.__getattribute__(self, '_closeafterrecvbytes')
# No read limit. Call original function.
if not remaining:
result = object.__getattribute__(self, '_orig').readline(size)
self._writelog('readline(%d) -> (%d) %s' % (
size, len(result), result))
return result
origsize = size
if size < 0:
size = remaining
else:
size = min(remaining, size)
result = object.__getattribute__(self, '_orig').readline(size)
remaining -= len(result)
self._writelog('readline(%d from %d) -> (%d) %s' % (
size, origsize, len(result), result))
object.__setattr__(self, '_closeafterrecvbytes', remaining)
if remaining <= 0:
self._writelog('read limit reached; closing socket')
self._sock.close()
# This is the easiest way to abort the current request.
raise Exception('connection closed after receiving N bytes')
return result
def write(self, data):
remaining = object.__getattribute__(self, '_closeaftersendbytes')
# No byte limit on this operation. Call original function.
if not remaining:
self._writelog('write(%d) -> %s' % (len(data), data))
result = object.__getattribute__(self, '_orig').write(data)
return result
if len(data) > remaining:
newdata = data[0:remaining]
else:
newdata = data
remaining -= len(newdata)
self._writelog('write(%d from %d) -> (%d) %s' % (
len(newdata), len(data), remaining, newdata))
result = object.__getattribute__(self, '_orig').write(newdata)
object.__setattr__(self, '_closeaftersendbytes', remaining)
if remaining <= 0:
self._writelog('write limit reached; closing socket')
self._sock.close()
raise Exception('connection closed after sending N bytes')
return result
def extsetup(ui):
# Change the base HTTP server class so various events can be performed.
# See SocketServer.BaseServer for how the specially named methods work.
class badserver(server.MercurialHTTPServer):
def __init__(self, ui, *args, **kwargs):
self._ui = ui
super(badserver, self).__init__(ui, *args, **kwargs)
recvbytes = self._ui.config(b'badserver', b'closeafterrecvbytes')
recvbytes = recvbytes.split(',')
self.closeafterrecvbytes = [int(v) for v in recvbytes if v]
sendbytes = self._ui.config(b'badserver', b'closeaftersendbytes')
sendbytes = sendbytes.split(',')
self.closeaftersendbytes = [int(v) for v in sendbytes if v]
# Need to inherit object so super() works.
class badrequesthandler(self.RequestHandlerClass, object):
def send_header(self, name, value):
# Make headers deterministic to facilitate testing.
if name.lower() == 'date':
value = 'Fri, 14 Apr 2017 00:00:00 GMT'
elif name.lower() == 'server':
value = 'badhttpserver'
return super(badrequesthandler, self).send_header(name,
value)
self.RequestHandlerClass = badrequesthandler
# Called to accept() a pending socket.
def get_request(self):
if self._ui.configbool(b'badserver', b'closebeforeaccept'):
self.socket.close()
# Tells the server to stop processing more requests.
self.__shutdown_request = True
# Simulate failure to stop processing this request.
raise socket.error('close before accept')
if self._ui.configbool('badserver', 'closeafteraccept'):
request, client_address = super(badserver, self).get_request()
request.close()
raise socket.error('close after accept')
return super(badserver, self).get_request()
# Does heavy lifting of processing a request. Invokes
# self.finish_request() which calls self.RequestHandlerClass() which
# is a hgweb.server._httprequesthandler.
def process_request(self, socket, address):
# Wrap socket in a proxy if we need to count bytes.
if self.closeafterrecvbytes:
closeafterrecvbytes = self.closeafterrecvbytes.pop(0)
else:
closeafterrecvbytes = 0
if self.closeaftersendbytes:
closeaftersendbytes = self.closeaftersendbytes.pop(0)
else:
closeaftersendbytes = 0
if closeafterrecvbytes or closeaftersendbytes:
socket = socketproxy(socket, self.errorlog,
closeafterrecvbytes=closeafterrecvbytes,
closeaftersendbytes=closeaftersendbytes)
return super(badserver, self).process_request(socket, address)
server.MercurialHTTPServer = badserver