| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The previous patch demonstrates a bug where a MIDX's auxiliary object
order can become out of sync with a MIDX bitmap.
This is because of two confounding factors:
- First, the object order is stored in a file which is named according
to the multi-pack index's checksum, and the MIDX does not store the
object order. This means that the object order can change without
altering the checksum.
- But the .rev file is moved into place with finalize_object_file(),
which link(2)'s the file into place instead of renaming it. For us,
that means that a modified .rev file will not be moved into place if
MIDX's checksum was unchanged.
This fix is to force the MIDX's checksum to change when the preferred
pack changes but the set of packs contained in the MIDX does not. In
other words, when the object order changes, the MIDX's checksum needs to
change with it (regardless of whether the MIDX is tracking the same or
different packs).
This prevents a race whereby changing the object order (but not the
packs themselves) enables a reader to see the new .rev file with the old
MIDX, or similarly seeing the new bitmap with the old object order.
But why can't we just stop hardlinking the .rev into place instead
adding additional data to the MIDX? Suppose that's what we did. Then
when we go to generate the new bitmap, we'll load the old MIDX bitmap,
along with the MIDX that it references. That's fine, since the new MIDX
isn't moved into place until after the new bitmap is generated. But the
new object order *has* been moved into place. So we'll read the old
bitmaps in the new order when generating the new bitmap file, meaning
that without this secondary change, bitmap generation itself would
become a victim of the race described here.
This can all be prevented by forcing the MIDX's checksum to change when
the object order does. By embedding the entire object order into the
MIDX, we do just that. That is, the MIDX's checksum will change in
response to any perturbation of the underlying object order. In t5326,
this will cause the MIDX's checksum to update (even without changing the
set of packs in the MIDX), preventing the stale read problem.
Note that this makes it safe to continue to link(2) the MIDX .rev file
into place, since it is now impossible to have a .rev file that is
out-of-sync with the MIDX whose checksum it references. (But we will do
away with MIDX .rev files later in this series anyway, so this is
somewhat of a moot point).
In theory, it is possible to store a "fingerprint" of the full object
order here, so long as that fingerprint changes at least as often as the
full object order does. Some possibilities here include storing the
identity of the preferred pack, along with the mtimes of the
non-preferred packs in a consistent order. But storing a limited part of
the information makes it difficult to reason about whether or not there
are gaps between the two that would cause us to get bitten by this bug
again.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Reviewed-by: Derrick Stolee <dstolee@microsoft.com>
Reviewed-by: Jonathan Tan <jonathantanmy@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
As a prerequisite to implementing multi-pack bitmaps, motivate and
describe the format and ordering of the multi-pack reverse index.
The subsequent patch will implement reading this format, and the patch
after that will implement writing it while producing a multi-pack index.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
* ds/chunked-file-api:
commit-graph.c: display correct number of chunks when writing
chunk-format: add technical docs
chunk-format: restore duplicate chunk checks
midx: use 64-bit multiplication for chunk sizes
midx: use chunk-format read API
commit-graph: use chunk-format read API
chunk-format: create read chunk API
midx: use chunk-format API in write_midx_internal()
midx: drop chunk progress during write
midx: return success/failure in chunk write methods
midx: add num_large_offsets to write_midx_context
midx: add pack_perm to write_midx_context
midx: add entries to write_midx_context
midx: use context in write_midx_pack_names()
midx: rename pack_info to write_midx_context
commit-graph: use chunk-format write API
chunk-format: create chunk format write API
commit-graph: anonymize data in chunk_write_fn
|
|
The chunk-based file format is now an API in the code, but we should
also take time to document it as a file format. Specifically, it matches
the CHUNK LOOKUP sections of the commit-graph and multi-pack-index
files, but there are some commonalities that should be grouped in this
document.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Specify the format of the on-disk reverse index 'pack-*.rev' file, as
well as prepare the code for the existence of such files.
The reverse index maps from pack relative positions (i.e., an index into
the array of object which is sorted by their offsets within the
packfile) to their position within the 'pack-*.idx' file. Today, this is
done by building up a list of (off_t, uint32_t) tuples for each object
(the off_t corresponding to that object's offset, and the uint32_t
corresponding to its position in the index). To convert between pack and
index position quickly, this array of tuples is radix sorted based on
its offset.
This has two major drawbacks:
First, the in-memory cost scales linearly with the number of objects in
a pack. Each 'struct revindex_entry' is sizeof(off_t) +
sizeof(uint32_t) + padding bytes for a total of 16.
To observe this, force Git to load the reverse index by, for e.g.,
running 'git cat-file --batch-check="%(objectsize:disk)"'. When asking
for a single object in a fresh clone of the kernel, Git needs to
allocate 120+ MB of memory in order to hold the reverse index in memory.
Second, the cost to sort also scales with the size of the pack.
Luckily, this is a linear function since 'load_pack_revindex()' uses a
radix sort, but this cost still must be paid once per pack per process.
As an example, it takes ~60x longer to print the _size_ of an object as
it does to print that entire object's _contents_:
Benchmark #1: git.compile cat-file --batch <obj
Time (mean ± σ): 3.4 ms ± 0.1 ms [User: 3.3 ms, System: 2.1 ms]
Range (min … max): 3.2 ms … 3.7 ms 726 runs
Benchmark #2: git.compile cat-file --batch-check="%(objectsize:disk)" <obj
Time (mean ± σ): 210.3 ms ± 8.9 ms [User: 188.2 ms, System: 23.2 ms]
Range (min … max): 193.7 ms … 224.4 ms 13 runs
Instead, avoid computing and sorting the revindex once per process by
writing it to a file when the pack itself is generated.
The format is relatively straightforward. It contains an array of
uint32_t's, the length of which is equal to the number of objects in the
pack. The ith entry in this table contains the index position of the
ith object in the pack, where "ith object in the pack" is determined by
pack offset.
One thing that the on-disk format does _not_ contain is the full (up to)
eight-byte offset corresponding to each object. This is something that
the in-memory revindex contains (it stores an off_t in 'struct
revindex_entry' along with the same uint32_t that the on-disk format
has). Omit it in the on-disk format, since knowing the index position
for some object is sufficient to get a constant-time lookup in the
pack-*.idx file to ask for an object's offset within the pack.
This trades off between the on-disk size of the 'pack-*.rev' file for
runtime to chase down the offset for some object. Even though the lookup
is constant time, the constant is heavier, since it can potentially
involve two pointer walks in v2 indexes (one to access the 4-byte offset
table, and potentially a second to access the double wide offset table).
Consider trying to map an object's pack offset to a relative position
within that pack. In a cold-cache scenario, more page faults occur while
switching between binary searching through the reverse index and
searching through the *.idx file for an object's offset. Sure enough,
with a cold cache (writing '3' into '/proc/sys/vm/drop_caches' after
'sync'ing), printing out the entire object's contents is still
marginally faster than printing its size:
Benchmark #1: git.compile cat-file --batch-check="%(objectsize:disk)" <obj >/dev/null
Time (mean ± σ): 22.6 ms ± 0.5 ms [User: 2.4 ms, System: 7.9 ms]
Range (min … max): 21.4 ms … 23.5 ms 41 runs
Benchmark #2: git.compile cat-file --batch <obj >/dev/null
Time (mean ± σ): 17.2 ms ± 0.7 ms [User: 2.8 ms, System: 5.5 ms]
Range (min … max): 15.6 ms … 18.2 ms 45 runs
(Numbers taken in the kernel after cheating and using the next patch to
generate a reverse index). There are a couple of approaches to improve
cold cache performance not pursued here:
- We could include the object offsets in the reverse index format.
Predictably, this does result in fewer page faults, but it triples
the size of the file, while simultaneously duplicating a ton of data
already available in the .idx file. (This was the original way I
implemented the format, and it did show
`--batch-check='%(objectsize:disk)'` winning out against `--batch`.)
On the other hand, this increase in size also results in a large
block-cache footprint, which could potentially hurt other workloads.
- We could store the mapping from pack to index position in more
cache-friendly way, like constructing a binary search tree from the
table and writing the values in breadth-first order. This would
result in much better locality, but the price you pay is trading
O(1) lookup in 'pack_pos_to_index()' for an O(log n) one (since you
can no longer directly index the table).
So, neither of these approaches are taken here. (Thankfully, the format
is versioned, so we are free to pursue these in the future.) But, cold
cache performance likely isn't interesting outside of one-off cases like
asking for the size of an object directly. In real-world usage, Git is
often performing many operations in the revindex (i.e., asking about
many objects rather than a single one).
The trade-off is worth it, since we will avoid the vast majority of the
cost of generating the revindex that the extra pointer chase will look
like noise in the following patch's benchmarks.
This patch describes the format and prepares callers (like in
pack-revindex.c) to be able to read *.rev files once they exist. An
implementation of the writer will appear in the next patch, and callers
will gradually begin to start using the writer in the patches that
follow after that.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
We document the delta data as a set of instructions, but forget to
document the two sizes that precede those instructions: the size of the
base object and the size of the object to be reconstructed. Fix this
omission.
Rather than cramming all the details about the encoding into the running
text, introduce a separate section detailing our "size encoding" and
refer to it.
Reported-by: Ross Light <ross@zombiezen.com>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
midx and commit-graph files now use the byte defined in their file
format specification for identifying the hash function used for
object names.
* ds/sha256-leftover-bits:
multi-pack-index: use hash version byte
commit-graph: use the "hash version" byte
t/README: document GIT_TEST_DEFAULT_HASH
|
|
Similar to the commit-graph format, the multi-pack-index format has a
byte in the header intended to track the hash version used to write the
file. This allows one to interpret the hash length without having the
context of the repository config specifying the hash length. This was
not modified as part of the SHA-256 work because the hash length was
automatically up-shifted due to that config.
Since we have this byte available, we can make the file formats more
obviously incompatible instead of relying on other context from the
repository.
Add a new oid_version() method in midx.c similar to the one in
commit-graph.c. This is specifically made separate from that
implementation to avoid artificially linking the formats.
The test impact requires a few more things than the corresponding change
in the commit-graph format. Specifically, 'test-tool read-midx' was not
writing anything about this header value to output. Since the value
available in 'struct multi_pack_index' is hash_len instead of a version
value, we output "20" or "32" instead of "1" or "2".
Since we want a user to not have their Git commands fail if their
multi-pack-index has the incorrect hash version compared to the
repository's hash version, we relax the die() to an error() in
load_multi_pack_index(). This has some effect on 'git multi-pack-index
verify' as we need to check that a failed parse of a file that exists is
actually a verify error. For that test that checks the hash version
matches, we change the corrupted byte from "2" to "3" to ensure the test
fails for both hash algorithms.
Helped-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Reviewed-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Now that we have SHA-256 support for packs and indices, let's document
that in SHA-256 repositories, we use SHA-256 instead of SHA-1 for object
names and checksums. Instead of duplicating this information throughout
the document, let's just document that in SHA-1 repositories, we use
SHA-1 for these purposes, and in SHA-256 repositories, we use SHA-256.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
The description of the multi-pack-index contains a small bug,
if all offsets are < 2^32 then there will be no LOFF chunk,
not only if they're all < 2^31 (since the highest bit is only
needed as the "LOFF-escape" when that's actually needed.)
Correct this, and clarify that in that case only offsets up
to 2^31-1 can be stored in the OOFF chunk.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
The final pair of chunks for the multi-pack-index file stores the object
offsets. We default to using 32-bit offsets as in the pack-index version
1 format, but if there exists an offset larger than 32-bits, we use a
trick similar to the pack-index version 2 format by storing all offsets
at least 2^31 in a 64-bit table; we use the 32-bit table to point into
that 64-bit table as necessary.
We only store these 64-bit offsets if necessary, so create a test that
manipulates a version 2 pack-index to fake a large offset. This allows
us to test that the large offset table is created, but the data does not
match the actual packfile offsets. The multi-pack-index offset does match
the (corrupted) pack-index offset, so a future feature will compare these
offsets during a 'verify' step.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
The multi-pack-index needs to track which packfiles it indexes. Store
these in our first required chunk. Since filenames are not well
structured, add padding to keep good alignment in later chunks.
Modify the 'git multi-pack-index read' subcommand to output the
existence of the pack-file name chunk. Modify t5319-multi-pack-index.sh
to reflect this new output and the new expected number of chunks.
Defense in depth: A pattern we are using in the multi-pack-index feature
is to verify the data as we write it. We want to ensure we never write
invalid data to the multi-pack-index. There are many checks that verify
that the values we are writing fit the format definitions. This mainly
helps developers while working on the feature, but it can also identify
issues that only appear when dealing with very large data sets. These
large sets are hard to encode into test cases.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
The multi-pack-index feature generalizes the existing pack-index
feature by indexing objects across multiple pack-files.
Describe the basic file format, using a 12-byte header followed by
a lookup table for a list of "chunks" which will be described later.
The file ends with a footer containing a checksum using the hash
algorithm.
The header allows later versions to create breaking changes by
advancing the version number. We can also change the hash algorithm
using a different version value.
We will add the individual chunk format information as we introduce
the code that writes that information.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
The current document mentions OBJ_* constants without their actual
values. A git developer would know these are from cache.h but that's
not very friendly to a person who wants to read this file to implement
a pack file parser.
Similarly, the deltified representation is not documented at all (the
"document" is basically patch-delta.c). Translate that C code to
English with a bit more about what ofs-delta and ref-delta mean.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Use "SHA-1" instead of "SHA1" whenever we talk about the hash function.
When used as a programming symbol, we keep "SHA1".
Signed-off-by: Thomas Ackermann <th.acker@arcor.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
* maint-1.8.1:
fast-export: fix argument name in error messages
Documentation: distinguish between ref and offset deltas in pack-format
|
|
eb32d236 introduced the OBJ_OFS_DELTA object that uses a relative offset to
identify the base object instead of the 20-byte SHA1 reference. The pack file
documentation only mentions the SHA1 based reference in its description of the
deltified object entry.
Update the pack format documentation to clarify that the deltified object
representation refers to its base using either a relative negative offset or
the absolute SHA1 identifier.
Signed-off-by: Stefan Saasen <ssaasen@atlassian.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
In the earlier days, we used to spell the name of the system as GIT,
to simulate as if it were typeset with capital G and IT in small
caps. Later we stopped doing so at around 1.6.5 days.
Let's stop doing so throughout the documentation. The name to refer
to the whole system (and the concept it embodies) is "Git"; the
command end-users type is "git". And document this in the coding
guideline.
Signed-off-by: Thomas Ackermann <th.acker@arcor.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
These were not originally meant for asciidoc, but they are already
so close. Mark them up in asciidoc.
Signed-off-by: Thomas Ackermann <th.acker@arcor.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Signed-off-by: Peter Eriksen <s022018@student.dtu.dk>
Acked-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Signed-off-by: Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Lifted from the log message of c553ca25bd60dc9fd50b8bc7bd329601b81cee66
(pack-objects: learn about pack index version 2).
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
This uses "git-apply --whitespace=strip" to fix whitespace errors that have
crept in to our source files over time. There are a few files that need
to have trailing whitespaces (most notably, test vectors). The results
still passes the test, and build result in Documentation/ area is unchanged.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
|
|
Signed-off-by: Peter Eriksen <s022018@student.dtu.dk>
Signed-off-by: Junio C Hamano <junkio@cox.net>
|
|
While trying to implement a pack reader in Java I was mislead by
some facts listed in this documentation as well as found a few
details to be missing about the pack header.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
|
|
... along with the previous one, pack-heuristics, by popular
demand.
Signed-off-by: Junio C Hamano <junkio@cox.net>
|
Taylor Blau
Junio C Hamano
Derrick Stolee
Martin Ågren
brian m. carlson
Johannes Berg
Derrick Stolee
Nguyễn Thái Ngọc Duy
Thomas Ackermann
Stefan Saasen
Thomas Ackermann
Peter Eriksen
Ralf Wildenhues
linux@horizon.com
Shawn Pearce
Junio C Hamano