The pack bitmap file gained a bitmap-lookup table to speed up
locating the necessary bitmap for a given commit.
* ac/bitmap-lookup-table:
pack-bitmap-write: drop unused pack_idx_entry parameters
bitmap-lookup-table: add performance tests for lookup table
pack-bitmap: prepare to read lookup table extension
pack-bitmap-write: learn pack.writeBitmapLookupTable and add tests
pack-bitmap-write.c: write lookup table extension
bitmap: move `get commit positions` code to `bitmap_writer_finish`
Documentation/technical: describe bitmap lookup table extension
Multi-pack index got corrupted when preferred pack changed from one
pack to another in a certain way, which has been corrected.
* tb/midx-with-changing-preferred-pack-fix:
midx.c: avoid adding preferred objects twice
midx.c: include preferred pack correctly with existing MIDX
midx.c: extract `midx_fanout_add_pack_fanout()`
midx.c: extract `midx_fanout_add_midx_fanout()`
midx.c: extract `struct midx_fanout`
t/lib-bitmap.sh: avoid silencing stderr
t5326: demonstrate potential bitmap corruption
Teach Git to provide a way for users to enable/disable bitmap lookup
table extension by providing a config option named 'writeBitmapLookupTable'.
Default is false.
Also add test to verify writting of lookup table.
Mentored-by: Taylor Blau <me@ttaylorr.com>
Co-Mentored-by: Kaartic Sivaraam <kaartic.sivaraam@gmail.com>
Co-Authored-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Abhradeep Chakraborty <chakrabortyabhradeep79@gmail.com>
Reviewed-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The last commit changes the behavior of midx.c's `get_sorted_objects()`
function to handle the case of writing a MIDX bitmap while reusing an
existing MIDX and changing the identity of the preferred pack
separately.
As part of this change, all objects from the (new) preferred pack are
added to the fanout table in a separate pass. Since these copies of the
objects all have their preferred bits set, any duplicates will be
resolved in their favor.
Importantly, this includes any copies of those same objects that come
from the existing MIDX. We know at the time of adding them that they'll
be redundant if their source pack is the (new) preferred one, so we can
avoid adding them to the list in this case.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This patch resolves an issue where the object order used to generate a
MIDX bitmap would violate an invariant that all of the preferred pack's
objects are represented by that pack in the MIDX.
The problem arises when reusing an existing MIDX while generating a new
one, and occurs specifically when the identity of the preferred pack
changes from one MIDX to another, along with a few other conditions:
- the new preferred pack must also be present in the existing MIDX
- the new preferred pack must *not* have been the preferred pack in
the existing MIDX
- most importantly, there must be at least one object present in the
physical preferred pack (ie., it shows up in that pack's index)
but was selected from a *different* pack when the previous MIDX
was generated
When the above conditions are all met, we end up (incorrectly)
discarding copies of some objects in the pack selected as the preferred
pack. This is because `get_sorted_entries()` adds objects to its list
by doing the following at each fanout level:
- first, adding all objects from that fanout level from an existing
MIDX
- then, adding all objects from that fanout level in each pack *not*
included in the existing MIDX
So if some object was not selected from the to-be-preferred pack when
writing the previous MIDX, then we will never consider it as a candidate
when generating the new MIDX. This means that it's possible for the
preferred pack to not include all of its objects in the MIDX's
pseudo-pack object order, which is an invariant violation of that order.
Resolve this by adding all objects from the preferred pack separately
when it appears in the existing MIDX (if one was present). This will
duplicate objects from that pack that *did* appear in the MIDX, but this
is fine, since get_sorted_entries() already handles duplicates. (A
future optimization in this area could avoid adding copies of objects
that we know already existing in the MIDX.)
Note that we no longer need to compute the preferred-ness of objects
added from the MIDX, since we only want to select the preferred objects
from a single source. (We could still mark these preferred bits, but
doing so is redundant and unnecessary).
This resolves the bug demonstrated by t5326.174 ("preferred pack change
with existing MIDX bitmap").
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Extract a routine to add all objects whose object ID's first byte is
`cur_fanout` from a given pack (identified by its index into the `struct
pack_info` array maintained by the MIDX writing routine).
Unlike the previous extraction (for `midx_fanout_add_midx_fanout()`),
this function will be called twice, once for all new packs, and again
for the preferred pack (if it appears in an existing MIDX). The latter
change is to resolve the bug described a few patches ago, and will be
made in the subsequent commit.
Similar to the previous refactoring, this function also enhances the
readability of its caller in `get_sorted_entries()`.
Its functionality is unchanged in this commit.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Extract a routine to add all objects whose object ID's first byte is
`cur_fanout` from an existing MIDX.
This function will only be called once, so extracting it is purely
cosmetic to improve the readability of `get_sorted_entries()` (its sole
caller) below.
The functionality is unchanged in this commit.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
To build up a list of objects (along with their packs, and the offsets
within those packs that each object appears at), the MIDX code
implements `get_sorted_entries()` which builds up a list of candidates,
sorts them, and then removes duplicate entries.
To do this, it keeps an array of `pack_midx_entry` structures that it
builds up once for each fanout level (ie., for all possible values of
the first byte of each object's ID).
This array is a function-local variable of `get_sorted_entries()`. Since
it uses the ALLOC_GROW() macro, having the `alloc_fanout` variable also
be local to that function, and only modified within that function is
convenient.
However, subsequent changes will extract the two ways this array is
filled (from a pack at some fanout value, and from an existing MIDX at
some fanout value) into separate functions. Instead of passing around
pointers to the entries array, along with `nr_fanout` and
`alloc_fanout`, encapsulate these three into a structure instead. Then
pass around a pointer to this structure instead.
This patch does not yet extract the above two functions, but sets us up
to begin doing so in the following commit. For now, the implementation
of get_sorted_entries() is only modified to replace `entries_by_fanout`
with `fanout.entries`, `nr_fanout` with `fanout.nr`, and so on.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The refactoring in 90b2bb710d (midx: extract bitmap write setup,
2022-07-19) hoisted our call to find_commits_for_midx_bitmap() into the
caller, which means we no longer need to see the refs_snapshot at all.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The write_midx_bitmap() method is a long method that does a lot of
steps. It requires the write_midx_context struct for use in
prepare_midx_packing_data() and find_commits_for_midx_bitmap(), but
after that only needs the pack_order array.
This is a messy, but completely non-functional refactoring. The code is
only being moved around to reduce visibility of the write_midx_context
during the longest part of computing reachability bitmaps.
Signed-off-by: Derrick Stolee <derrickstolee@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Plug the memory leaks from the trickiest API of all, the revision
walker.
* ab/plug-leak-in-revisions: (27 commits)
revisions API: add a TODO for diff_free(&revs->diffopt)
revisions API: have release_revisions() release "topo_walk_info"
revisions API: have release_revisions() release "date_mode"
revisions API: call diff_free(&revs->pruning) in revisions_release()
revisions API: release "reflog_info" in release revisions()
revisions API: clear "boundary_commits" in release_revisions()
revisions API: have release_revisions() release "prune_data"
revisions API: have release_revisions() release "grep_filter"
revisions API: have release_revisions() release "filter"
revisions API: have release_revisions() release "cmdline"
revisions API: have release_revisions() release "mailmap"
revisions API: have release_revisions() release "commits"
revisions API users: use release_revisions() for "prune_data" users
revisions API users: use release_revisions() with UNLEAK()
revisions API users: use release_revisions() in builtin/log.c
revisions API users: use release_revisions() in http-push.c
revisions API users: add "goto cleanup" for release_revisions()
stash: always have the owner of "stash_info" free it
revisions API users: use release_revisions() needing REV_INFO_INIT
revision.[ch]: document and move code declared around "init"
...
A mechanism to pack unreachable objects into a "cruft pack",
instead of ejecting them into loose form to be reclaimed later, has
been introduced.
* tb/cruft-packs:
sha1-file.c: don't freshen cruft packs
builtin/gc.c: conditionally avoid pruning objects via loose
builtin/repack.c: add cruft packs to MIDX during geometric repack
builtin/repack.c: use named flags for existing_packs
builtin/repack.c: allow configuring cruft pack generation
builtin/repack.c: support generating a cruft pack
builtin/pack-objects.c: --cruft with expiration
reachable: report precise timestamps from objects in cruft packs
reachable: add options to add_unseen_recent_objects_to_traversal
builtin/pack-objects.c: --cruft without expiration
builtin/pack-objects.c: return from create_object_entry()
t/helper: add 'pack-mtimes' test-tool
pack-mtimes: support writing pack .mtimes files
chunk-format.h: extract oid_version()
pack-write: pass 'struct packing_data' to 'stage_tmp_packfiles'
pack-mtimes: support reading .mtimes files
Documentation/technical: add cruft-packs.txt
There are three definitions of an identical function which converts
`the_hash_algo` into either 1 (for SHA-1) or 2 (for SHA-256). There is a
copy of this function for writing both the commit-graph and
multi-pack-index file, and another inline definition used to write the
.rev header.
Consolidate these into a single definition in chunk-format.h. It's not
clear that this is the best header to define this function in, but it
should do for now.
(Worth noting, the .rev caller expects a 4-byte unsigned, but the other
two callers work with a single unsigned byte. The consolidated version
uses the latter type, and lets the compiler widen it when required).
Another caller will be added in a subsequent patch.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The path taken by "git multi-pack-index" command from the end user
was compared with path internally prepared by the tool withut first
normalizing, which lead to duplicated paths not being noticed,
which has been corrected.
* ds/midx-normalize-pathname-before-comparison:
cache: use const char * for get_object_directory()
multi-pack-index: use --object-dir real path
midx: use real paths in lookup_multi_pack_index()
This helper looks for a parsed multi-pack-index whose object directory
matches the given object_dir. Before going into the loop over the parsed
multi-pack-indexes, it calls find_odb() to ensure that the given
object_dir is actually a known object directory.
However, find_odb() uses real-path manipulations to compare the input to
the alternate directories. This same real-path comparison is not used in
the loop, leading to potential issues with the strcmp().
Update the method to use the real-path values instead.
Signed-off-by: Derrick Stolee <derrickstolee@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a release_revisions() to various users of "struct rev_list" in
those straightforward cases where we only need to add the
release_revisions() call to the end of a block, and don't need to
e.g. refactor anything to use a "goto cleanup" pattern.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This commit introduces the infrastructure for the core.fsync
configuration knob. The repository components we want to sync
are identified by flags so that we can turn on or off syncing
for specific components.
If core.fsyncObjectFiles is set and the core.fsync configuration
also includes FSYNC_COMPONENT_LOOSE_OBJECT, we will fsync any
loose objects. This picks the strictest data integrity behavior
if core.fsync and core.fsyncObjectFiles are set to conflicting values.
This change introduces the currently unused fsync_component
helper, which will be used by a later patch that adds fsyncing to
the refs backend.
Actual configuration and documentation of the fsync components
list are in other patches in the series to separate review of
the underlying mechanism from the policy of how it's configured.
Helped-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Neeraj Singh <neerajsi@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When there is no object to write .bitmap file for, "git
multi-pack-index" triggered an error, instead of just skipping,
which has been corrected.
* tb/midx-no-bitmap-for-no-objects:
midx: prevent writing a .bitmap without any objects
When trying to write a MIDX, we already prevent the case where there
weren't any packs present, and thus we would have written an empty MIDX.
But there is another "empty" case, which is more interesting, and we
don't yet handle. If we try to write a MIDX which has at least one pack,
but those packs together don't contain any objects, we will encounter a
BUG() when trying to use the bitmap corresponding to that MIDX, like so:
$ git rev-parse HEAD | git pack-objects --revs --use-bitmap-index --stdout >/dev/null
BUG: pack-revindex.c:394: pack_pos_to_midx: out-of-bounds object at 0
(note that in the above reproduction, both `--use-bitmap-index` and
`--stdout` are important, since without the former we won't even both to
load the .bitmap, and without the latter we wont attempt pack reuse).
The problem occurs when we try to discover the identity of the
preferred pack to determine which range if any of existing packs we can
reuse verbatim. This path is: `reuse_packfile_objects()` ->
`reuse_partial_packfile_from_bitmap()` -> `midx_preferred_pack()`.
#4 0x000055555575401f in pack_pos_to_midx (m=0x555555997160, pos=0) at pack-revindex.c:394
#5 0x00005555557502c8 in midx_preferred_pack (bitmap_git=0x55555599c280) at pack-bitmap.c:1431
#6 0x000055555575036c in reuse_partial_packfile_from_bitmap (bitmap_git=0x55555599c280, packfile_out=0x5555559666b0 <reuse_packfile>,
entries=0x5555559666b8 <reuse_packfile_objects>, reuse_out=0x5555559666c0 <reuse_packfile_bitmap>) at pack-bitmap.c:1452
#7 0x00005555556041f6 in get_object_list_from_bitmap (revs=0x7fffffffcbf0) at builtin/pack-objects.c:3658
#8 0x000055555560465c in get_object_list (ac=2, av=0x555555997050) at builtin/pack-objects.c:3765
#9 0x0000555555605e4e in cmd_pack_objects (argc=0, argv=0x7fffffffe920, prefix=0x0) at builtin/pack-objects.c:4154
Since neither the .bitmap or MIDX stores the identity of the
preferred pack, we infer it by trying to load the first object in
pseudo-pack order, and then asking the MIDX which pack was chosen to
represent that object.
But this fails our bounds check, since there are zero objects in the
MIDX to begin with, which results in the BUG().
We could catch this more carefully in `midx_preferred_pack()`, but
signaling the absence of a preferred pack out to all of its callers is
somewhat awkward.
Instead, let's avoid writing a MIDX .bitmap without any objects
altogether. We catch this case in `write_midx_internal()`, and emit a
warning if the caller indicated they wanted to write a bitmap before
clearing out the relevant flags. If we somehow got to
write_midx_bitmap(), then we will call BUG(), but this should now be an
unreachable path.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a MIDX contains the new `RIDX` chunk, ensure that the reverse index
is read from it instead of the on-disk .rev file. Since we need to
encode the object order in the MIDX itself for correctness reasons,
there is no point in storing the same data again outside of the MIDX.
So, this patch stops writing separate .rev files, and reads it out of
the MIDX itself. This is possible to do with relatively little new code,
since the format of the RIDX chunk is identical to the data in the .rev
file. In other words, we can implement this by pointing the
`revindex_data` field at the reverse index chunk of the MIDX instead of
the .rev file without any other changes.
Note that we have two knobs that are adjusted for the new tests:
GIT_TEST_MIDX_WRITE_REV and GIT_TEST_MIDX_READ_RIDX. The former controls
whether the MIDX .rev is written at all, and the latter controls whether
we read the MIDX's RIDX chunk.
Both are necessary to ensure that the test added at the beginning of
this series continues to work. This is because we always need to write
the RIDX chunk in the MIDX in order to change its checksum, but we want
to make sure reading the existing .rev file still works (since the RIDX
chunk takes precedence by default).
Arguably this isn't a very interesting mode to test, because the
precedence rules mean that we'll always read the RIDX chunk over the
.rev file. But it makes it impossible for a user to induce corruption in
their repository by adjusting the test knobs (since if we had an
either/or knob they could stop writing the RIDX chunk, allowing them to
tweak the MIDX's object order without changing its checksum).
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>
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>
To ask for the name of a MIDX and its corresponding .rev file, callers
invoke get_midx_filename() and get_midx_rev_filename(), respectively.
These both invoke xstrfmt(), allocating a chunk of memory which must be
freed later on.
This makes callers in pack-bitmap.c somewhat awkward. Specifically,
midx_bitmap_filename(), which is implemented like:
return xstrfmt("%s-%s.bitmap",
get_midx_filename(midx->object_dir),
hash_to_hex(get_midx_checksum(midx)));
this leaks the second argument to xstrfmt(), which itself was allocated
with xstrfmt(). This caller could assign both the result of
get_midx_filename() and the outer xstrfmt() to a temporary variable,
remembering to free() the former before returning. But that involves a
wasteful copy.
Instead, get_midx_filename() and get_midx_rev_filename() take a strbuf
as an output parameter. This way midx_bitmap_filename() can manipulate
and pass around a temporary buffer which it detaches back to its caller.
That allows us to implement the function without copying or open-coding
get_midx_filename() in a way that doesn't leak.
Update the other callers of get_midx_filename() and
get_midx_rev_filename() accordingly.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The function midx.c:verify_midx_file() allocates a MIDX struct by
calling load_multi_pack_index(). But when cleaning up, it calls free()
without freeing any resources associated with the MIDX.
Call the more appropriate close_midx() which does free those resources,
which causes t5319.3 to pass when Git is compiled with SANITIZE=leak.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In order to read the contents of a MIDX, we initialize a chunkfile
structure which can read the table of contents and assign pointers into
different sections of the file for us.
We do call free(), since the chunkfile struct is heap allocated, but not
the more appropriate free_chunkfile(), which also frees memory that the
structure itself owns.
Call that instead to avoid leaking memory in this function.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git repack" has been taught to generate multi-pack reachability
bitmaps.
* tb/repack-write-midx:
test-read-midx: fix leak of bitmap_index struct
builtin/repack.c: pass `--refs-snapshot` when writing bitmaps
builtin/repack.c: make largest pack preferred
builtin/repack.c: support writing a MIDX while repacking
builtin/repack.c: extract showing progress to a variable
builtin/repack.c: rename variables that deal with non-kept packs
builtin/repack.c: keep track of existing packs unconditionally
midx: preliminary support for `--refs-snapshot`
builtin/multi-pack-index.c: support `--stdin-packs` mode
midx: expose `write_midx_file_only()` publicly
When writing a MIDX, we atomically move the new MIDX into place via
commit_lock_file(), but do not check to see if that call was successful.
Make sure that we do check in order to prevent us from incorrectly
reporting that we wrote a new MIDX if we actually encountered an error.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Apply similar treatment as in the last commit to the MIDX `repack`
operation.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Before a new MIDX can be written, expire_midx_packs() first loads the
existing MIDX, figures out which packs can be expired, and then writes a
new MIDX based on that information.
In order to load the existing MIDX, it uses load_multi_pack_index(),
which mmaps the multi-pack-index file, but does not store the resulting
`struct multi_pack_index *` in the object store.
write_midx_internal() also needs to open the existing MIDX, and it does
so by iterating the results of get_multi_pack_index(), so that it reuses
the same pointer held by the object store. But before it can move the
new MIDX into place, it close_object_store() to munmap() the
multi-pack-index file to accommodate platforms like Windows which don't
allow overwriting files which are memory mapped.
That's where things get weird. Since expire_midx_packs has its own
*separate* memory mapped copy of the MIDX, the MIDX file is still memory
mapped! Interestingly, this doesn't seem to cause a problem in our
tests. (I believe that this has much more to do with my own lack of
familiarity with Windows than it does a lack of coverage in our tests).
In any case, we can side-step the whole issue by teaching
expire_midx_packs() to use the `struct multi_pack_index` pointer it
found via the object store instead of maintain its own copy. That way,
when write_midx_internal() calls `close_object_store()`, we know that
there are no memory mapped copies of the MIDX laying around.
A couple of other small notes about this patch:
- As far as I can tell, passing `local == 1` to the call to
load_multi_pack_index() was an error, since object_dir could be an
alternate. But it doesn't matter, since even though we write
`m->local = 1`, we never read that field back later on.
- Setting `m = NULL` after write_midx_internal() was likely to prevent
a double-free back from when that function took a `struct
multi_pack_index *` that it called close_midx() on itself. We can
rely on write_midx_internal() to call that for us now.
Finally, this enforces the same "the value of --object-dir must be the
local object store, or an alternate" rule from f57a739691 (midx: avoid
opening multiple MIDXs when writing, 2021-09-01) to the `expire`
sub-command, too.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The first thing that write_midx_internal() does is load the MIDX
corresponding to the given object directory, if one is present.
Prepare for other functions in midx.c to do the same thing by extracting
that operation out to a small helper function.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* tb/repack-write-midx:
test-read-midx: fix leak of bitmap_index struct
builtin/repack.c: pass `--refs-snapshot` when writing bitmaps
builtin/repack.c: make largest pack preferred
builtin/repack.c: support writing a MIDX while repacking
builtin/repack.c: extract showing progress to a variable
builtin/repack.c: rename variables that deal with non-kept packs
builtin/repack.c: keep track of existing packs unconditionally
midx: preliminary support for `--refs-snapshot`
builtin/multi-pack-index.c: support `--stdin-packs` mode
midx: expose `write_midx_file_only()` publicly
"git multi-pack-index write --bitmap" learns to propagate the
hashcache from original bitmap to resulting bitmap.
* tb/midx-write-propagate-namehash:
t5326: test propagating hashcache values
p5326: generate pack bitmaps before writing the MIDX bitmap
p5326: don't set core.multiPackIndex unnecessarily
p5326: create missing 'perf-tag' tag
midx.c: respect 'pack.writeBitmapHashcache' when writing bitmaps
pack-bitmap.c: propagate namehash values from existing bitmaps
t/helper/test-bitmap.c: add 'dump-hashes' mode
To figure out which commits we can write a bitmap for, the multi-pack
index/bitmap code does a reachability traversal, marking any commit
which can be found in the MIDX as eligible to receive a bitmap.
This approach will cause a problem when multi-pack bitmaps are able to
be generated from `git repack`, since the reference tips can change
during the repack. Even though we ignore commits that don't exist in
the MIDX (when doing a scan of the ref tips), it's possible that a
commit in the MIDX reaches something that isn't.
This can happen when a multi-pack index contains some pack which refers
to loose objects (e.g., if a pack was pushed after starting the repack
but before generating the MIDX which depends on an object which is
stored as loose in the repository, and by definition isn't included in
the multi-pack index).
By taking a snapshot of the references before we start repacking, we can
close that race window. In the above scenario (where we have a packed
object pointing at a loose one), we'll either (a) take a snapshot of the
references before seeing the packed one, or (b) take it after, at which
point we can guarantee that the loose object will be packed and included
in the MIDX.
This patch does just that. It writes a temporary "reference snapshot",
which is a list of OIDs that are at the ref tips before writing a
multi-pack bitmap. References that are "preferred" (i.e,. are a suffix
of at least one value of the 'pack.preferBitmapTips' configuration) are
marked with a special '+'.
The format is simple: one line per commit at each tip, with an optional
'+' at the beginning (for preferred references, as described above).
When provided, the reference snapshot is used to drive bitmap selection
instead of the MIDX code doing its own traversal. When it isn't
provided, the usual traversal takes place instead.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Expose a variant of the write_midx_file() function which ignores packs
that aren't included in an explicit "allow" list.
This will be used in an upcoming patch to power a new `--stdin-packs`
mode of `git multi-pack-index write` for callers that only want to
include certain packs in a MIDX (and ignore any packs which may have
happened to enter the repository independently, e.g., from pushes).
Those patches will provide test coverage for this new function.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Replace a handcrafted data structure used to keep track of bad
objects in the packfile API by an oidset.
* rs/packfile-bad-object-list-in-oidset:
packfile: use oidset for bad objects
packfile: convert has_packed_and_bad() to object_id
packfile: convert mark_bad_packed_object() to object_id
midx: inline nth_midxed_pack_entry()
oidset: make oidset_size() an inline function
In the previous commit, the bitmap writing code learned to propagate
values from an existing hash-cache extension into the bitmap that it is
writing.
Now that that functionality exists, let's expose it by teaching the 'git
multi-pack-index' builtin to respect the `pack.writeBitmapHashCache`
option so that the hash-cache may be written at all.
Two minor points worth noting here:
- The 'git multi-pack-index write' sub-command didn't previously read
any configuration (instead this is handled in the base command). A
separate handler is added here to respect this write-specific
config option.
- I briefly considered adding a 'bitmap_flags' field to the static
options struct, but decided against it since it would require
plumbing through a new parameter to the write_midx_file() function.
Instead, a new MIDX-specific flag is added, which is translated to
the corresponding bitmap one.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Store the object ID of broken pack entries in an oidset instead of
keeping only their hashes in an unsorted array. The resulting code is
shorter and easier to read. It also handles the (hopefully) very rare
case of having a high number of bad objects better.
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: René Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
fill_midx_entry() finds the position of an object ID and passes it to
nth_midxed_pack_entry(), which uses the position to look up the object
ID for its own purposes. Inline the latter into the former to avoid
that lookup.
Signed-off-by: René Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We never look at the repository argument which is passed. This makes
sense, since the multi_pack_index struct already tells us everything we
need to access the files in its associated object directory.
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Write multi-pack bitmaps in the format described by
Documentation/technical/bitmap-format.txt, inferring their presence with
the absence of '--bitmap'.
To write a multi-pack bitmap, this patch attempts to reuse as much of
the existing machinery from pack-objects as possible. Specifically, the
MIDX code prepares a packing_data struct that pretends as if a single
packfile has been generated containing all of the objects contained
within the MIDX.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This prepares the code in pack-bitmap to interpret the new multi-pack
bitmaps described in Documentation/technical/bitmap-format.txt, which
mostly involves converting bit positions to accommodate looking them up
in a MIDX.
Note that there are currently no writers who write multi-pack bitmaps,
and that this will be implemented in the subsequent commit. Note also
that get_midx_checksum() and get_midx_filename() are made non-static so
they can be called from pack-bitmap.c.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Opening multiple instance of the same MIDX can lead to problems like two
separate packed_git structures which represent the same pack being added
to the repository's object store.
The above scenario can happen because prepare_midx_pack() checks if
`m->packs[pack_int_id]` is NULL in order to determine if a pack has been
opened and installed in the repository before. But a caller can
construct two copies of the same MIDX by calling get_multi_pack_index()
and load_multi_pack_index() since the former manipulates the
object store directly but the latter is a lower-level routine which
allocates a new MIDX for each call.
So if prepare_midx_pack() is called on multiple MIDXs with the same
pack_int_id, then that pack will be installed twice in the object
store's packed_git pointer.
This can lead to problems in, for e.g., the pack-bitmap code, which does
something like the following (in pack-bitmap.c:open_pack_bitmap()):
struct bitmap_index *bitmap_git = ...;
for (p = get_all_packs(r); p; p = p->next) {
if (open_pack_bitmap_1(bitmap_git, p) == 0)
ret = 0;
}
which is a problem if two copies of the same pack exist in the
packed_git list because pack-bitmap.c:open_pack_bitmap_1() contains a
conditional like the following:
if (bitmap_git->pack || bitmap_git->midx) {
/* ignore extra bitmap file; we can only handle one */
warning("ignoring extra bitmap file: %s", packfile->pack_name);
close(fd);
return -1;
}
Avoid this scenario by not letting write_midx_internal() open a MIDX
that isn't also pointed at by the object store. So long as this is the
case, other routines should prefer to open MIDXs with
get_multi_pack_index() or reprepare_packed_git() instead of creating
instances on their own. Because get_multi_pack_index() returns
`r->object_store->multi_pack_index` if it is non-NULL, we'll only have
one instance of a MIDX open at one time, avoiding these problems.
To encourage this, drop the `struct multi_pack_index *` parameter from
`write_midx_internal()`, and rely instead on the `object_dir` to find
(or initialize) the correct MIDX instance.
Likewise, replace the call to `close_midx()` with
`close_object_store()`, since we're about to replace the MIDX with a new
one and should invalidate the object store's memory of any MIDX that
might have existed beforehand.
Note that this now forbids passing object directories that don't belong
to alternate repositories over `--object-dir`, since before we would
have happily opened a MIDX in any directory, but now restrict ourselves
to only those reachable by `r->objects->multi_pack_index` (and alternate
MIDXs that we can see by walking the `next` pointer).
As far as I can tell, supporting arbitrary directories with
`--object-dir` was a historical accident, since even the documentation
says `<alt>` when referring to the value passed to this option.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a repository has at least one alternate, the MIDX belonging to each
alternate is accessed through the `next` pointer on the main object
store's copy of the MIDX. close_midx() didn't bother to close any
of the linked MIDXs. It likewise didn't free the memory pointed to by
`m`, leaving uninitialized bytes with live pointers to them left around
in the heap.
Clean this up by closing linked MIDXs, and freeing up the memory pointed
to by each of them. When callers call close_midx(), then they can
discard the entire linked list of MIDXs and set their pointer to the
head of that list to NULL.
This isn't strictly required for the upcoming patches, but it makes it
much more difficult (though still possible, for e.g., by calling
`close_midx(m->next)` which leaves `m->next` pointing at uninitialized
bytes) to have pointers to uninitialized memory.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 9218c6a40c (midx: allow marking a pack as preferred, 2021-03-30), the
multi-pack index code learned how to select a pack which all duplicate
objects are selected from. That is, if an object appears in multiple
packs, select the copy in the preferred pack before breaking ties
according to the other rules like pack mtime and readdir() order.
Not specifying a preferred pack can cause serious problems with
multi-pack reachability bitmaps, because these bitmaps rely on having at
least one pack from which all duplicates are selected. Not having such a
pack causes problems with the code in pack-objects to reuse packs
verbatim (e.g., that code assumes that a delta object in a chunk of pack
sent verbatim will have its base object sent from the same pack).
So why does not marking a pack preferred cause problems here? The reason
is roughly as follows:
- Ties are broken (when handling duplicate objects) by sorting
according to midx_oid_compare(), which sorts objects by OID,
preferred-ness, pack mtime, and finally pack ID (more on that
later).
- The psuedo pack-order (described in
Documentation/technical/pack-format.txt under the section
"multi-pack-index reverse indexes") is computed by
midx_pack_order(), and sorts by pack ID and pack offset, with
preferred packs sorting first.
- But! Pack IDs come from incrementing the pack count in
add_pack_to_midx(), which is a callback to
for_each_file_in_pack_dir(), meaning that pack IDs are assigned in
readdir() order.
When specifying a preferred pack, all of that works fine, because
duplicate objects are correctly resolved in favor of the copy in the
preferred pack, and the preferred pack sorts first in the object order.
"Sorting first" is critical, because the bitmap code relies on finding
out which pack holds the first object in the MIDX's pseudo pack-order to
determine which pack is preferred.
But if we didn't specify a preferred pack, and the pack which comes
first in readdir() order does not also have the lowest timestamp, then
it's possible that that pack (the one that sorts first in pseudo-pack
order, which the bitmap code will treat as the preferred one) did *not*
have all duplicate objects resolved in its favor, resulting in breakage.
The fix is simple: pick a (semi-arbitrary, non-empty) preferred pack
when none was specified. This forces that pack to have duplicates
resolved in its favor, and (critically) to sort first in pseudo-pack
order. Unfortunately, testing this behavior portably isn't possible,
since it depends on readdir() order which isn't guaranteed by POSIX.
(Note that multi-pack reachability bitmaps have yet to be implemented;
so in that sense this patch is fixing a bug which does not yet exist.
But by having this patch beforehand, we can prevent the bug from ever
materializing.)
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The soon-to-be-implemented multi-pack bitmap treats object in the first
bit position specially by assuming that all objects in the pack it was
selected from are also represented from that pack in the MIDX. In other
words, the pack from which the first object was selected must also have
all of its other objects selected from that same pack in the MIDX in
case of any duplicates.
But this assumption relies on the fact that there is at least one object
in that pack to begin with; otherwise the object in the first bit
position isn't from a preferred pack, in which case we can no longer
assume that all objects in that pack were also selected from the same
pack.
Guard this assumption by checking the number of objects in the given
preferred pack, and failing if the given pack is empty.
To make sure we can safely perform this check, open any packs which are
contained in an existing MIDX via prepare_midx_pack(). The same is done
for new packs via the add_pack_to_midx() callback, but packs picked up
from a previous MIDX will not yet have these opened.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When writing a new multi-pack index, write_midx_internal() attempts to
clean up any auxiliary files (currently just the MIDX's `.rev` file, but
soon to include a `.bitmap`, too) corresponding to the MIDX it's
replacing.
This step should happen after the new MIDX is written into place, since
doing so beforehand means that the old MIDX could be read without its
corresponding .rev file.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If using --object-dir to point into an object directory which belongs to
a different repository than the one in the current working directory,
such as:
git init repo
git -C repo ... # add some objects
cd alternate
git multi-pack-index --object-dir ../repo/.git/objects write
the binary will segfault trying to access the object-dir via the repo it
found, but that's not fully initialized. Worse, if we later call
clear_midx_files_ext(), we will use `the_repository` and remove files
out of the wrong object directory.
Fix this by using the given object_dir (or the object directory of
`the_repository` if `--object-dir` wasn't given) to properly to clean up
the *.rev files, avoiding the crash.
Original-patch-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Many "printf"-like helper functions we have have been annotated
with __attribute__() to catch placeholder/parameter mismatches.
* ab/attribute-format:
advice.h: add missing __attribute__((format)) & fix usage
*.h: add a few missing __attribute__((format))
*.c static functions: add missing __attribute__((format))
sequencer.c: move static function to avoid forward decl
*.c static functions: don't forward-declare __attribute__
Add missing __attribute__((format)) function attributes to various
"static" functions that take printf arguments.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Junio C Hamano
Abhradeep Chakraborty
Taylor Blau
Jeff King
Derrick Stolee
Ævar Arnfjörð Bjarmason
Neeraj Singh
René Scharfe