##// END OF EJS Templates
upstream » mercurial-mirror
RSS

Clone from https://www.mercurial-scm.org/repo/hg-all

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...
Gregory Szorc - - Load All Authors

File last commit:

r29747:aba2bb2a default
r30818:4c0a5a25 default
Show More
changegroups.txt
153 lines | 6.1 KiB | text/plain | TextLexer
/ mercurial / help / internals / changegroups.txt
History | Annotation | Raw |Copy content |Copy permalink
Changegroups are representations of repository revlog data, specifically
the changelog, manifest, and filelogs.
There are 3 versions of changegroups: ``1``, ``2``, and ``3``. From a
high-level, versions ``1`` and ``2`` are almost exactly the same, with
the only difference being a header on entries in the changeset
segment. Version ``3`` adds support for exchanging treemanifests and
includes revlog flags in the delta header.
Changegroups consists of 3 logical segments::
+---------------------------------+
| | | |
| changeset | manifest | filelogs |
| | | |
+---------------------------------+
The principle building block of each segment is a *chunk*. A *chunk*
is a framed piece of data::
+---------------------------------------+
| | |
| length | data |
| (32 bits) | <length> bytes |
| | |
+---------------------------------------+
Each chunk starts with a 32-bit big-endian signed integer indicating
the length of the raw data that follows.
There is a special case chunk that has 0 length (``0x00000000``). We
call this an *empty chunk*.
Delta Groups
============
A *delta group* expresses the content of a revlog as a series of deltas,
or patches against previous revisions.
Delta groups consist of 0 or more *chunks* followed by the *empty chunk*
to signal the end of the delta group::
+------------------------------------------------------------------------+
| | | | | |
| chunk0 length | chunk0 data | chunk1 length | chunk1 data | 0x0 |
| (32 bits) | (various) | (32 bits) | (various) | (32 bits) |
| | | | | |
+------------------------------------------------------------+-----------+
Each *chunk*'s data consists of the following::
+-----------------------------------------+
| | | |
| delta header | mdiff header | delta |
| (various) | (12 bytes) | (various) |
| | | |
+-----------------------------------------+
The *length* field is the byte length of the remaining 3 logical pieces
of data. The *delta* is a diff from an existing entry in the changelog.
The *delta header* is different between versions ``1``, ``2``, and
``3`` of the changegroup format.
Version 1::
+------------------------------------------------------+
| | | | |
| node | p1 node | p2 node | link node |
| (20 bytes) | (20 bytes) | (20 bytes) | (20 bytes) |
| | | | |
+------------------------------------------------------+
Version 2::
+------------------------------------------------------------------+
| | | | | |
| node | p1 node | p2 node | base node | link node |
| (20 bytes) | (20 bytes) | (20 bytes) | (20 bytes) | (20 bytes) |
| | | | | |
+------------------------------------------------------------------+
Version 3::
+------------------------------------------------------------------------------+
| | | | | | |
| node | p1 node | p2 node | base node | link node | flags |
| (20 bytes) | (20 bytes) | (20 bytes) | (20 bytes) | (20 bytes) | (2 bytes) |
| | | | | | |
+------------------------------------------------------------------------------+
The *mdiff header* consists of 3 32-bit big-endian signed integers
describing offsets at which to apply the following delta content::
+-------------------------------------+
| | | |
| offset | old length | new length |
| (32 bits) | (32 bits) | (32 bits) |
| | | |
+-------------------------------------+
In version 1, the delta is always applied against the previous node from
the changegroup or the first parent if this is the first entry in the
changegroup.
In version 2, the delta base node is encoded in the entry in the
changegroup. This allows the delta to be expressed against any parent,
which can result in smaller deltas and more efficient encoding of data.
Changeset Segment
=================
The *changeset segment* consists of a single *delta group* holding
changelog data. It is followed by an *empty chunk* to denote the
boundary to the *manifests segment*.
Manifest Segment
================
The *manifest segment* consists of a single *delta group* holding
manifest data. It is followed by an *empty chunk* to denote the boundary
to the *filelogs segment*.
Filelogs Segment
================
The *filelogs* segment consists of multiple sub-segments, each
corresponding to an individual file whose data is being described::
+--------------------------------------+
| | | | |
| filelog0 | filelog1 | filelog2 | ... |
| | | | |
+--------------------------------------+
In version ``3`` of the changegroup format, filelogs may include
directory logs when treemanifests are in use. directory logs are
identified by having a trailing '/' on their filename (see below).
The final filelog sub-segment is followed by an *empty chunk* to denote
the end of the segment and the overall changegroup.
Each filelog sub-segment consists of the following::
+------------------------------------------+
| | | |
| filename size | filename | delta group |
| (32 bits) | (various) | (various) |
| | | |
+------------------------------------------+
That is, a *chunk* consisting of the filename (not terminated or padded)
followed by N chunks constituting the *delta group* for this file.