##// END OF EJS Templates
localrepo: experimental support for non-zlib revlog compression...
localrepo: experimental support for non-zlib revlog compression The final part of integrating the compression manager APIs into revlog storage is the plumbing for repositories to advertise they are using non-zlib storage and for revlogs to instantiate a non-zlib compression engine. The main intent of the compression manager work was to zstd all of the things. Adding zstd to revlogs has proved to be more involved than other places because revlogs are... special. Very small inputs and the use of delta chains (which are themselves a form of compression) are a completely different use case from streaming compression, which bundles and the wire protocol employ. I've conducted numerous experiments with zstd in revlogs and have yet to formalize compression settings and a storage architecture that I'm confident I won't regret later. In other words, I'm not yet ready to commit to a new mechanism for using zstd - or any other compression format - in revlogs. That being said, having some support for zstd (and other compression formats) in revlogs in core is beneficial. It can allow others to conduct experiments. This patch introduces *highly experimental* support for non-zlib compression formats in revlogs. Introduced is a config option to control which compression engine to use. Also introduced is a namespace of "exp-compression-*" requirements to denote support for non-zlib compression in revlogs. I've prefixed the namespace with "exp-" (short for "experimental") because I'm not confident of the requirements "schema" and in no way want to give the illusion of supporting these requirements in the future. I fully intend to drop support for these requirements once we figure out what we're doing with zstd in revlogs. A good portion of the patch is teaching the requirements system about registered compression engines and passing the requested compression engine as an opener option so revlogs can instantiate the proper compression engine for new operations. That's a verbose way of saying "we can now use zstd in revlogs!" On an `hg pull` conversion of the mozilla-unified repo with no extra redelta settings (like aggressivemergedeltas), we can see the impact of zstd vs zlib in revlogs: $ hg perfrevlogchunks -c ! chunk ! wall 2.032052 comb 2.040000 user 1.990000 sys 0.050000 (best of 5) ! wall 1.866360 comb 1.860000 user 1.820000 sys 0.040000 (best of 6) ! chunk batch ! wall 1.877261 comb 1.870000 user 1.860000 sys 0.010000 (best of 6) ! wall 1.705410 comb 1.710000 user 1.690000 sys 0.020000 (best of 6) $ hg perfrevlogchunks -m ! chunk ! wall 2.721427 comb 2.720000 user 2.640000 sys 0.080000 (best of 4) ! wall 2.035076 comb 2.030000 user 1.950000 sys 0.080000 (best of 5) ! chunk batch ! wall 2.614561 comb 2.620000 user 2.580000 sys 0.040000 (best of 4) ! wall 1.910252 comb 1.910000 user 1.880000 sys 0.030000 (best of 6) $ hg perfrevlog -c -d 1 ! wall 4.812885 comb 4.820000 user 4.800000 sys 0.020000 (best of 3) ! wall 4.699621 comb 4.710000 user 4.700000 sys 0.010000 (best of 3) $ hg perfrevlog -m -d 1000 ! wall 34.252800 comb 34.250000 user 33.730000 sys 0.520000 (best of 3) ! wall 24.094999 comb 24.090000 user 23.320000 sys 0.770000 (best of 3) Only modest wins for the changelog. But manifest reading is significantly faster. What's going on? One reason might be data volume. zstd decompresses faster. So given more bytes, it will put more distance between it and zlib. Another reason is size. In the current design, zstd revlogs are *larger*: debugcreatestreamclonebundle (size in bytes) zlib: 1,638,852,492 zstd: 1,680,601,332 I haven't investigated this fully, but I reckon a significant cause of larger revlogs is that the zstd frame/header has more bytes than zlib's. For very small inputs or data that doesn't compress well, we'll tend to store more uncompressed chunks than with zlib (because the compressed size isn't smaller than original). This will make revlog reading faster because it is doing less decompression. Moving on to bundle performance: $ hg bundle -a -t none-v2 (total CPU time) zlib: 102.79s zstd: 97.75s So, marginal CPU decrease for reading all chunks in all revlogs (this is somewhat disappointing). $ hg bundle -a -t <engine>-v2 (total CPU time) zlib: 191.59s zstd: 115.36s This last test effectively measures the difference between zlib->zlib and zstd->zstd for revlogs to bundle. This is a rough approximation of what a server does during `hg clone`. There are some promising results for zstd. But not enough for me to feel comfortable advertising it to users. We'll get there...

File last commit:

r29133:25527471 default
r30818:4c0a5a25 default
Show More
parsers.py
178 lines | 5.5 KiB | text/x-python | PythonLexer
# parsers.py - Python implementation of parsers.c
#
# Copyright 2009 Matt Mackall <mpm@selenic.com> and others
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
from __future__ import absolute_import
import struct
import zlib
from .node import nullid
from . import pycompat
stringio = pycompat.stringio
_pack = struct.pack
_unpack = struct.unpack
_compress = zlib.compress
_decompress = zlib.decompress
# Some code below makes tuples directly because it's more convenient. However,
# code outside this module should always use dirstatetuple.
def dirstatetuple(*x):
# x is a tuple
return x
indexformatng = ">Qiiiiii20s12x"
indexfirst = struct.calcsize('Q')
sizeint = struct.calcsize('i')
indexsize = struct.calcsize(indexformatng)
def gettype(q):
return int(q & 0xFFFF)
def offset_type(offset, type):
return long(long(offset) << 16 | type)
class BaseIndexObject(object):
def __len__(self):
return self._lgt + len(self._extra) + 1
def insert(self, i, tup):
assert i == -1
self._extra.append(tup)
def _fix_index(self, i):
if not isinstance(i, int):
raise TypeError("expecting int indexes")
if i < 0:
i = len(self) + i
if i < 0 or i >= len(self):
raise IndexError
return i
def __getitem__(self, i):
i = self._fix_index(i)
if i == len(self) - 1:
return (0, 0, 0, -1, -1, -1, -1, nullid)
if i >= self._lgt:
return self._extra[i - self._lgt]
index = self._calculate_index(i)
r = struct.unpack(indexformatng, self._data[index:index + indexsize])
if i == 0:
e = list(r)
type = gettype(e[0])
e[0] = offset_type(0, type)
return tuple(e)
return r
class IndexObject(BaseIndexObject):
def __init__(self, data):
assert len(data) % indexsize == 0
self._data = data
self._lgt = len(data) // indexsize
self._extra = []
def _calculate_index(self, i):
return i * indexsize
def __delitem__(self, i):
if not isinstance(i, slice) or not i.stop == -1 or not i.step is None:
raise ValueError("deleting slices only supports a:-1 with step 1")
i = self._fix_index(i.start)
if i < self._lgt:
self._data = self._data[:i * indexsize]
self._lgt = i
self._extra = []
else:
self._extra = self._extra[:i - self._lgt]
class InlinedIndexObject(BaseIndexObject):
def __init__(self, data, inline=0):
self._data = data
self._lgt = self._inline_scan(None)
self._inline_scan(self._lgt)
self._extra = []
def _inline_scan(self, lgt):
off = 0
if lgt is not None:
self._offsets = [0] * lgt
count = 0
while off <= len(self._data) - indexsize:
s, = struct.unpack('>i',
self._data[off + indexfirst:off + sizeint + indexfirst])
if lgt is not None:
self._offsets[count] = off
count += 1
off += indexsize + s
if off != len(self._data):
raise ValueError("corrupted data")
return count
def __delitem__(self, i):
if not isinstance(i, slice) or not i.stop == -1 or not i.step is None:
raise ValueError("deleting slices only supports a:-1 with step 1")
i = self._fix_index(i.start)
if i < self._lgt:
self._offsets = self._offsets[:i]
self._lgt = i
self._extra = []
else:
self._extra = self._extra[:i - self._lgt]
def _calculate_index(self, i):
return self._offsets[i]
def parse_index2(data, inline):
if not inline:
return IndexObject(data), None
return InlinedIndexObject(data, inline), (0, data)
def parse_dirstate(dmap, copymap, st):
parents = [st[:20], st[20: 40]]
# dereference fields so they will be local in loop
format = ">cllll"
e_size = struct.calcsize(format)
pos1 = 40
l = len(st)
# the inner loop
while pos1 < l:
pos2 = pos1 + e_size
e = _unpack(">cllll", st[pos1:pos2]) # a literal here is faster
pos1 = pos2 + e[4]
f = st[pos2:pos1]
if '\0' in f:
f, c = f.split('\0')
copymap[f] = c
dmap[f] = e[:4]
return parents
def pack_dirstate(dmap, copymap, pl, now):
now = int(now)
cs = stringio()
write = cs.write
write("".join(pl))
for f, e in dmap.iteritems():
if e[0] == 'n' and e[3] == now:
# The file was last modified "simultaneously" with the current
# write to dirstate (i.e. within the same second for file-
# systems with a granularity of 1 sec). This commonly happens
# for at least a couple of files on 'update'.
# The user could change the file without changing its size
# within the same second. Invalidate the file's mtime in
# dirstate, forcing future 'status' calls to compare the
# contents of the file if the size is the same. This prevents
# mistakenly treating such files as clean.
e = dirstatetuple(e[0], e[1], e[2], -1)
dmap[f] = e
if f in copymap:
f = "%s\0%s" % (f, copymap[f])
e = _pack(">cllll", e[0], e[1], e[2], e[3], len(f))
write(e)
write(f)
return cs.getvalue()