Enabling pack.writebitmaphashcache should always be a performance win.
It costs only 4 bytes per object on disk, and the timings in ae4f07fbcc
(pack-bitmap: implement optional name_hash cache, 2013-12-21) show it
improving fetch and partial-bitmap clone times by 40-50%.
The only reason we didn't enable it by default at the time is that early
versions of JGit's bitmap reader complained about the presence of
optional header bits it didn't understand. But that was changed in
JGit's d2fa3987a (Use bitcheck to check for presence of OPT_FULL option,
2013-10-30), which made it into JGit v3.5.0 in late 2014.
So let's turn this option on by default. It's backwards-compatible with
all versions of Git, and if you are also using JGit on the same
repository, you'd only run into problems using a version that's almost 5
years old.
We'll drop the manual setting from all of our test scripts, including
perf tests. This isn't strictly necessary, but it has two advantages:
1. If the hash-cache ever stops being enabled by default, our perf
regression tests will notice.
2. We can use the modified perf tests to show off the behavior of an
otherwise unconfigured repo, as shown below.
These are the results of a few of a perf tests against linux.git that
showed interesting results. You can see the expected speedup in 5310.4,
which was noted in ae4f07fbcc. Curiously, 5310.8 did not improve (and
actually got slower), despite seeing the opposite in ae4f07fbcc.
I don't have an explanation for that.
The tests from p5311 did not exist back then, but do show improvements
(a smaller pack due to better deltas, which we found in less time).
Test HEAD^ HEAD
-------------------------------------------------------------------------------------
5310.4: simulated fetch 7.39(22.70+0.25) 5.64(11.43+0.22) -23.7%
5310.8: clone (partial bitmap) 18.45(24.83+1.19) 19.94(28.40+1.36) +8.1%
5311.31: server (128 days) 0.41(1.13+0.05) 0.34(0.72+0.02) -17.1%
5311.32: size (128 days) 7.4M 7.0M -4.8%
5311.33: client (128 days) 1.33(1.49+0.06) 1.29(1.37+0.12) -3.0%
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When studying the performance of 'git push' we would like to know
how much time is spent at various parts of the command. One area
that could cause performance trouble is 'git pack-objects'.
Add trace2 regions around the three main actions taken in this
command:
1. Enumerate objects.
2. Prepare pack.
3. Write pack-file.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git pack-objects" learned another algorithm to compute the set of
objects to send, that trades the resulting packfile off to save
traversal cost to favor small pushes.
* ds/push-sparse-tree-walk:
pack-objects: create GIT_TEST_PACK_SPARSE
pack-objects: create pack.useSparse setting
revision: implement sparse algorithm
list-objects: consume sparse tree walk
revision: add mark_tree_uninteresting_sparse
The assumption to work on the single "in-core index" instance has
been reduced from the library-ish part of the codebase.
* nd/the-index-final:
cache.h: flip NO_THE_INDEX_COMPATIBILITY_MACROS switch
read-cache.c: remove the_* from index_has_changes()
merge-recursive.c: remove implicit dependency on the_repository
merge-recursive.c: remove implicit dependency on the_index
sha1-name.c: remove implicit dependency on the_index
read-cache.c: replace update_index_if_able with repo_&
read-cache.c: kill read_index()
checkout: avoid the_index when possible
repository.c: replace hold_locked_index() with repo_hold_locked_index()
notes-utils.c: remove the_repository references
grep: use grep_opt->repo instead of explict repo argument
"git pack-objects" incorrectly used uninitialized mutex, which has
been corrected.
* ph/pack-objects-mutex-fix:
pack-objects: merge read_lock and lock in packing_data struct
pack-objects: move read mutex to packing_data struct
The code to walk tree objects has been taught that we may be
working with object names that are not computed with SHA-1.
* bc/tree-walk-oid:
cache: make oidcpy always copy GIT_MAX_RAWSZ bytes
tree-walk: store object_id in a separate member
match-trees: use hashcpy to splice trees
match-trees: compute buffer offset correctly when splicing
tree-walk: copy object ID before use
Rename the packing_data lock to obd_lock and upgrade it to a recursive
mutex to make it suitable for current read_lock usages. Additionally
remove the superfluous #ifndef NO_PTHREADS guard around mutex
initialization in prepare_packing_data as the mutex functions
themselves are already protected.
Signed-off-by: Patrick Hogg <phogg@novamoon.net>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
ac77d0c37 ("pack-objects: shrink size field in struct object_entry",
2018-04-14) added an extra usage of read_lock/read_unlock in the newly
introduced oe_get_size_slow for thread safety in parallel calls to
try_delta(). Unfortunately oe_get_size_slow is also used in serial
code, some of which is called before the first invocation of
ll_find_deltas. As such the read mutex is not guaranteed to be
initialized.
Resolve this by moving the read mutex to packing_data and initializing
it in prepare_packing_data which is initialized in cmd_pack_objects.
Signed-off-by: Patrick Hogg <phogg@novamoon.net>
Reviewed-by: Duy Nguyen <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
By default, index compat macros are off from now on, because they
could hide the_index dependency.
Only those in builtin can use it.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Create a test variable GIT_TEST_PACK_SPARSE to enable the sparse
object walk algorithm by default during the test suite. Enabling
this variable ensures coverage in many interesting cases, such as
shallow clones, partial clones, and missing objects.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The '--sparse' flag in 'git pack-objects' changes the algorithm
used to enumerate objects to one that is faster for individual
users pushing new objects that change only a small cone of the
working directory. The sparse algorithm is not recommended for a
server, which likely sends new objects that appear across the
entire working directory.
Create a 'pack.useSparse' setting that enables this new algorithm.
This allows 'git push' to use this algorithm without passing a
'--sparse' flag all the way through four levels of run_command()
calls.
If the '--no-sparse' flag is set, then this config setting is
overridden.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When creating a pack-file using 'git pack-objects --revs' we provide
a list of interesting and uninteresting commits. For example, a push
operation would make the local topic branch be interesting and the
known remote refs as uninteresting. We want to discover the set of
new objects to send to the server as a thin pack.
We walk these commits until we discover a frontier of commits such
that every commit walk starting at interesting commits ends in a root
commit or unintersting commit. We then need to discover which
non-commit objects are reachable from uninteresting commits. This
commit walk is not changing during this series.
The mark_edges_uninteresting() method in list-objects.c iterates on
the commit list and does the following:
* If the commit is UNINTERSTING, then mark its root tree and every
object it can reach as UNINTERESTING.
* If the commit is interesting, then mark the root tree of every
UNINTERSTING parent (and all objects that tree can reach) as
UNINTERSTING.
At the very end, we repeat the process on every commit directly
given to the revision walk from stdin. This helps ensure we properly
cover shallow commits that otherwise were not included in the
frontier.
The logic to recursively follow trees is in the
mark_tree_uninteresting() method in revision.c. The algorithm avoids
duplicate work by not recursing into trees that are already marked
UNINTERSTING.
Add a new 'sparse' option to the mark_edges_uninteresting() method
that performs this logic in a slightly different way. As we iterate
over the commits, we add all of the root trees to an oidset. Then,
call mark_trees_uninteresting_sparse() on that oidset. Note that we
include interesting trees in this process. The current implementation
of mark_trees_unintersting_sparse() will walk the same trees as
the old logic, but this will be replaced in a later change.
Add a '--sparse' flag in 'git pack-objects' to call this new logic.
Add a new test script t/t5322-pack-objects-sparse.sh that tests this
option. The tests currently demonstrate that the resulting object
list is the same as the old algorithm. This includes a case where
both algorithms pack an object that is not needed by a remote due to
limits on the explored set of trees. When the sparse algorithm is
changed in a later commit, we will add a test that demonstrates a
change of behavior in some cases.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When parsing a tree, we read the object ID directly out of the tree
buffer. This is normally fine, but such an object ID cannot be used with
oidcpy, which copies GIT_MAX_RAWSZ bytes, because if we are using SHA-1,
there may not be that many bytes to copy.
Instead, store the object ID in a separate struct member. Since we can
no longer efficiently compute the path length, store that information as
well in struct name_entry. Ensure we only copy the object ID into the
new buffer if the path length is nonzero, as some callers will pass us
an empty path with no object ID following it, and we will not want to
read past the end of the buffer.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit abef9020e3 (sha1_file: convert sha1_object_info* to object_id,
2018-03-12) renamed the function to oid_object_info(), but missed some
comments which mention it.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Put the allow_exclude_promisor_objects flag in setup_revision_opt. When
it was in rev_info, it was unclear when it was used, since rev_info is
passed to functions that don't use the flag. This resulted in
unnecessary setting of the flag in prune.c, so fix that as well.
Signed-off-by: Matthew DeVore <matvore@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A few issues in the implementation of "delta-islands" feature has
been corrected.
* cc/delta-islands:
pack-objects: fix off-by-one in delta-island tree-depth computation
pack-objects: zero-initialize tree_depth/layer arrays
pack-objects: fix tree_depth and layer invariants
When delta-islands are in use, we need to record the deepest path at
which we find each tree and blob. Our loop to do so counts slashes, so
"foo" is depth 0, "foo/bar" is depth 1, and so on.
However, this neglects root trees, which are represented by the empty
string. Those also have depth 0, but are at a layer above "foo". Thus,
"foo" should be 1, "foo/bar" at 2, and so on. We use this depth to
topo-sort the trees in resolve_tree_islands(). As a result, we may fail
to visit a root tree before the sub-trees it contains, and therefore not
correctly pass down the island marks.
That in turn could lead to missing some delta opportunities (objects are
in the same island, but we didn't realize it) or creating unwanted
cross-island deltas (one object is in an island another isn't, but we
don't realize). In practice, it seems to have only a small effect. Some
experiments on the real-world git/git fork network at GitHub showed an
improvement of only 0.14% in the resulting clone size.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code preparation to replace ulong vars with size_t vars where
appropriate.
* tb/print-size-t-with-uintmax-format:
Upcast size_t variables to uintmax_t when printing
"git push" used to check ambiguities between object-names and
refnames while processing the list of refs' old and new values,
which was unnecessary (as it knew that it is feeding raw object
names). This has been optimized out.
* ds/push-squelch-ambig-warning:
pack-objects: ignore ambiguous object warnings
Various functions have been audited for "-Wunused-parameter" warnings
and bugs in them got fixed.
* jk/unused-parameter-fixes:
midx: double-check large object write loop
assert NOARG/NONEG behavior of parse-options callbacks
parse-options: drop OPT_DATE()
apply: return -1 from option callback instead of calling exit(1)
cat-file: report an error on multiple --batch options
tag: mark "--message" option with NONEG
show-branch: mark --reflog option as NONEG
format-patch: mark "--no-numbered" option with NONEG
status: mark --find-renames option with NONEG
cat-file: mark batch options with NONEG
pack-objects: mark index-version option as NONEG
ls-files: mark exclude options as NONEG
am: handle --no-patch-format option
apply: mark include/exclude options as NONEG
The codebase has been cleaned up to reduce "#ifndef NO_PTHREADS".
* nd/pthreads:
Clean up pthread_create() error handling
read-cache.c: initialize copy_len to shut up gcc 8
read-cache.c: reduce branching based on HAVE_THREADS
read-cache.c: remove #ifdef NO_PTHREADS
pack-objects: remove #ifdef NO_PTHREADS
preload-index.c: remove #ifdef NO_PTHREADS
grep: clean up num_threads handling
grep: remove #ifdef NO_PTHREADS
attr.c: remove #ifdef NO_PTHREADS
name-hash.c: remove #ifdef NO_PTHREADS
index-pack: remove #ifdef NO_PTHREADS
send-pack.c: move async's #ifdef NO_PTHREADS back to run-command.c
run-command.h: include thread-utils.h instead of pthread.h
thread-utils: macros to unconditionally compile pthreads API
When printing variables which contain a size, today "unsigned long"
is used at many places.
In order to be able to change the type from "unsigned long" into size_t
some day in the future, we need to have a way to print 64 bit variables
on a system that has "unsigned long" defined to be 32 bit, like Win64.
Upcast all those variables into uintmax_t before they are printed.
This is to prepare for a bigger change, when "unsigned long"
will be converted into size_t for variables which may be > 4Gib.
Signed-off-by: Torsten Bögershausen <tboegi@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A git push process runs several processes during its run, but one
includes git send-pack which calls git pack-objects and passes
the known have/wants into stdin using object ids. However, the
default setting for core.warnAmbiguousRefs requires git pack-objects
to check for ref names matching the ref_rev_parse_rules array in
refs.c. This means that every object is triggering at least six
"file exists?" queries. When there are a lot of refs, this can
add up significantly! I observed a simple push spending three
seconds checking these paths.
The fix here is similar to 4c30d50 "rev-list: disable object/refname
ambiguity check with --stdin". While the get_object_list() method
reads the objects from stdin, turn warn_on_object_refname_ambiguity
flag (which is usually true) to false. Just for code hygiene, save
away the original at the beginning and restore it once we are done.
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Operations on promisor objects make sense in the context of only a
small subset of the commands that internally use the revisions
machinery, but the "--exclude-promisor-objects" option were taken
and led to nonsense results by commands like "log", to which it
didn't make much sense. This has been corrected.
* md/exclude-promisor-objects-fix:
exclude-promisor-objects: declare when option is allowed
Documentation/git-log.txt: do not show --exclude-promisor-objects
When we define a parse-options callback, the flags we put in the option
struct must match what the callback expects. For example, a callback
which does not handle the "unset" parameter should only be used with
PARSE_OPT_NONEG. But since the callback and the option struct are not
defined next to each other, it's easy to get this wrong (as earlier
patches in this series show).
Fortunately, the compiler can help us here: compiling with
-Wunused-parameters can show us which callbacks ignore their "unset"
parameters (and likewise, ones that ignore "arg" expect to be triggered
with PARSE_OPT_NOARG).
But after we've inspected a callback and determined that all of its
callers use the right flags, what do we do next? We'd like to silence
the compiler warning, but do so in a way that will catch any wrong calls
in the future.
We can do that by actually checking those variables and asserting that
they match our expectations. Because this is such a common pattern,
we'll introduce some helper macros. The resulting messages aren't
as descriptive as we could make them, but the file/line information from
BUG() is enough to identify the problem (and anyway, the point is that
these should never be seen).
Each of the annotated callbacks in this patch triggers
-Wunused-parameters, and was manually inspected to make sure all callers
use the correct options (so none of these BUGs should be triggerable).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Running "git pack-objects --no-index-version" will segfault, since the
callback is not prepared to handle the "unset" flag.
In theory this might be used to counteract an earlier "--index-version",
or override a pack.indexversion config setting. But the semantics aren't
immediately obvious, and it's unlikely anybody wants this. Let's just
disable the broken option for now.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A mutex used in "git pack-objects" were not correctly initialized
and this caused "git repack" to dump core on Windows.
* js/pack-objects-mutex-init-fix:
pack-objects (mingw): initialize `packing_data` mutex in the correct spot
pack-objects (mingw): demonstrate a segmentation fault with large deltas
pack-objects: fix typo 'detla' -> 'delta'
The --exclude-promisor-objects option causes some funny behavior in at
least two commands: log and blame. It causes a BUG crash:
$ git log --exclude-promisor-objects
BUG: revision.c:2143: exclude_promisor_objects can only be used
when fetch_if_missing is 0
Aborted
[134]
Fix this such that the option is treated like any other unknown option.
The commands that must support it are limited, so declare in those
commands that the flag is supported. In particular:
pack-objects
prune
rev-list
The commands were found by searching for logic which parses
--exclude-promisor-objects outside of revision.c. Extra logic outside of
revision.c is needed because fetch_if_missing must be turned on before
revision.c sees the option or it will BUG-crash. The above list is
supported by the fact that no other command is introspectively invoked
by another command passing --exclude-promisor-object.
Signed-off-by: Matthew DeVore <matvore@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 9ac3f0e5b3 (pack-objects: fix performance issues on packing large
deltas, 2018-07-22), a mutex was introduced that is used to guard the
call to set the delta size. This commit even added code to initialize
it, but at an incorrect spot: in `init_threaded_search()`, while the
call to `oe_set_delta_size()` (and hence to `packing_data_lock()`) can
happen in the call chain `check_object()` <- `get_object_details()` <-
`prepare_pack()` <- `cmd_pack_objects()`, which is long before the
`prepare_pack()` function calls `ll_find_deltas()` (which initializes
the threaded search).
Another tell-tale that the mutex was initialized in an incorrect spot is
that the function to initialize it lives in builtin/, while the code
that uses the mutex is defined in a libgit.a header file.
Let's use a more appropriate function: `prepare_packing_data()`, which
not only lives in libgit.a, but *has* to be called before the
`packing_data` struct is used that contains that mutex.
This fixes https://github.com/git-for-windows/git/issues/1839.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Various codepaths in the core-ish part learn to work on an
arbitrary in-core index structure, not necessarily the default
instance "the_index".
* nd/the-index: (23 commits)
revision.c: reduce implicit dependency the_repository
revision.c: remove implicit dependency on the_index
ws.c: remove implicit dependency on the_index
tree-diff.c: remove implicit dependency on the_index
submodule.c: remove implicit dependency on the_index
line-range.c: remove implicit dependency on the_index
userdiff.c: remove implicit dependency on the_index
rerere.c: remove implicit dependency on the_index
sha1-file.c: remove implicit dependency on the_index
patch-ids.c: remove implicit dependency on the_index
merge.c: remove implicit dependency on the_index
merge-blobs.c: remove implicit dependency on the_index
ll-merge.c: remove implicit dependency on the_index
diff-lib.c: remove implicit dependency on the_index
read-cache.c: remove implicit dependency on the_index
diff.c: remove implicit dependency on the_index
grep.c: remove implicit dependency on the_index
diff.c: remove the_index dependency in textconv() functions
blame.c: rename "repo" argument to "r"
combine-diff.c: remove implicit dependency on the_index
...
Fix interactions between two recent topics.
* jk/delta-islands-with-bitmap-reuse-delta-fix:
pack-objects: handle island check for "external" delta base
Two recent topics, jk/pack-delta-reuse-with-bitmap and
cc/delta-islands, can have a funny interaction. When
checking if we can reuse an on-disk delta, the first topic
allows base_entry to be NULL when we find an object that's
not in the packing list. But the latter topic introduces a
call to in_same_island(), which needs to look at
base_entry->idx.oid. When these two features are used
together, we might try to dereference a NULL base_entry.
In practice, this doesn't really happen. We'd generally only
use delta islands when packing to disk, since the whole
point is to optimize the pack for serving fetches later. And
the new delta-reuse code relies on having used reachability
bitmaps to determine the set of objects, which we would
typically only do when serving an actual fetch.
However, it is technically possible to combine these
features. And even without doing so, building with
"SANITIZE=address,undefined" will cause t5310.46 to
complain. Even though that test does not have delta islands
enabled, we still take the address of the NULL entry to pass
to in_same_island(). That function then promptly returns
without dereferencing the value when it sees that islands
are not enabled, but it's enough to trigger a sanitizer
error.
The solution is straight-forward: when both features are
used together, we should pass the oid of the found base to
in_same_island().
This is tricky to do inside a single "if" statement. And
after the merge in f3504ea3dd (Merge branch
'cc/delta-islands', 2018-09-17), that "if" condition is
already getting pretty unwieldy. So this patch moves the
logic into a helper function, where we can easily use
multiple return paths. The result is a bit longer, but the
logic should be much easier to follow.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
spatch transformation to replace boolean uses of !hashcmp() to
newly introduced oideq() is added, and applied, to regain
performance lost due to support of multiple hash algorithms.
* jk/cocci:
show_dirstat: simplify same-content check
read-cache: use oideq() in ce_compare functions
convert hashmap comparison functions to oideq()
convert "hashcmp() != 0" to "!hasheq()"
convert "oidcmp() != 0" to "!oideq()"
convert "hashcmp() == 0" to hasheq()
convert "oidcmp() == 0" to oideq()
introduce hasheq() and oideq()
coccinelle: use <...> for function exclusion
Lift code from GitHub to restrict delta computation so that an
object that exists in one fork is not made into a delta against
another object that does not appear in the same forked repository.
* cc/delta-islands:
pack-objects: move 'layer' into 'struct packing_data'
pack-objects: move tree_depth into 'struct packing_data'
t5320: tests for delta islands
repack: add delta-islands support
pack-objects: add delta-islands support
pack-objects: refactor code into compute_layer_order()
Add delta-islands.{c,h}
When creating a thin pack, which allows objects to be made into a
delta against another object that is not in the resulting pack but
is known to be present on the receiving end, the code learned to
take advantage of the reachability bitmap; this allows the server
to send a delta against a base beyond the "boundary" commit.
* jk/pack-delta-reuse-with-bitmap:
pack-objects: reuse on-disk deltas for thin "have" objects
pack-bitmap: save "have" bitmap from walk
t/perf: add perf tests for fetches from a bitmapped server
t/perf: add infrastructure for measuring sizes
t/perf: factor out percent calculations
t/perf: factor boilerplate out of test_perf
When there are too many packfiles in a repository (which is not
recommended), looking up an object in these would require
consulting many pack .idx files; a new mechanism to have a single
file that consolidates all of these .idx files is introduced.
* ds/multi-pack-index: (32 commits)
pack-objects: consider packs in multi-pack-index
midx: test a few commands that use get_all_packs
treewide: use get_all_packs
packfile: add all_packs list
midx: fix bug that skips midx with alternates
midx: stop reporting garbage
midx: mark bad packed objects
multi-pack-index: store local property
multi-pack-index: provide more helpful usage info
midx: clear midx on repack
packfile: skip loading index if in multi-pack-index
midx: prevent duplicate packfile loads
midx: use midx in approximate_object_count
midx: use existing midx when writing new one
midx: use midx in abbreviation calculations
midx: read objects from multi-pack-index
config: create core.multiPackIndex setting
midx: write object offsets
midx: write object id fanout chunk
midx: write object ids in a chunk
...
git_check_attr() returns always 0.
Remove all the error handling code of the callers, which is never executed.
Change git_check_attr() to be a void function.
Signed-off-by: Torsten Bögershausen <tboegi@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Using the more restrictive oideq() should, in the long run,
give the compiler more opportunities to optimize these
callsites. For now, this conversion should be a complete
noop with respect to the generated code.
The result is also perhaps a little more readable, as it
avoids the "zero is equal" idiom. Since it's so prevalent in
C, I think seasoned programmers tend not to even notice it
anymore, but it can sometimes make for awkward double
negations (e.g., we can drop a few !!oidcmp() instances
here).
This patch was generated almost entirely by the included
coccinelle patch. This mechanical conversion should be
completely safe, because we check explicitly for cases where
oidcmp() is compared to 0, which is what oideq() is doing
under the hood. Note that we don't have to catch "!oidcmp()"
separately; coccinelle's standard isomorphisms make sure the
two are treated equivalently.
I say "almost" because I did hand-edit the coccinelle output
to fix up a few style violations (it mostly keeps the
original formatting, but sometimes unwraps long lines).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In a recent update in 2.18 era, "git pack-objects" started
producing a larger than necessary packfiles by missing
opportunities to use large deltas.
* nd/pack-deltify-regression-fix:
pack-objects: fix performance issues on packing large deltas
When we serve a fetch, we pass the "wants" and "haves" from
the fetch negotiation to pack-objects. That tells us not
only which objects we need to send, but we also use the
boundary commits as "preferred bases": their trees and blobs
are candidates for delta bases, both for reusing on-disk
deltas and for finding new ones.
However, this misses some opportunities. Modulo some special
cases like shallow or partial clones, we know that every
object reachable from the "haves" could be a preferred base.
We don't use all of them for two reasons:
1. It's expensive to traverse the whole history and
enumerate all of the objects the other side has.
2. The delta search is expensive, so we want to keep the
number of candidate bases sane. The boundary commits
are the most likely to work.
When we have reachability bitmaps, though, reason 1 no
longer applies. We can efficiently compute the set of
reachable objects on the other side (and in fact already did
so as part of the bitmap set-difference to get the list of
interesting objects). And using this set conveniently
covers the shallow and partial cases, since we have to
disable the use of bitmaps for those anyway.
The second reason argues against using these bases in the
search for new deltas. But there's one case where we can use
this information for free: when we have an existing on-disk
delta that we're considering reusing, we can do so if we
know the other side has the base object. This in fact saves
time during the delta search, because it's one less delta we
have to compute.
And that's exactly what this patch does: when we're
considering whether to reuse an on-disk delta, if bitmaps
tell us the other side has the object (and we're making a
thin-pack), then we reuse it.
Here are the results on p5311 using linux.git, which
simulates a client fetching after `N` days since their last
fetch:
Test origin HEAD
--------------------------------------------------------------------------
5311.3: server (1 days) 0.27(0.27+0.04) 0.12(0.09+0.03) -55.6%
5311.4: size (1 days) 0.9M 237.0K -73.7%
5311.5: client (1 days) 0.04(0.05+0.00) 0.10(0.10+0.00) +150.0%
5311.7: server (2 days) 0.34(0.42+0.04) 0.13(0.10+0.03) -61.8%
5311.8: size (2 days) 1.5M 347.7K -76.5%
5311.9: client (2 days) 0.07(0.08+0.00) 0.16(0.15+0.01) +128.6%
5311.11: server (4 days) 0.56(0.77+0.08) 0.13(0.10+0.02) -76.8%
5311.12: size (4 days) 2.8M 566.6K -79.8%
5311.13: client (4 days) 0.13(0.15+0.00) 0.34(0.31+0.02) +161.5%
5311.15: server (8 days) 0.97(1.39+0.11) 0.30(0.25+0.05) -69.1%
5311.16: size (8 days) 4.3M 1.0M -76.0%
5311.17: client (8 days) 0.20(0.22+0.01) 0.53(0.52+0.01) +165.0%
5311.19: server (16 days) 1.52(2.51+0.12) 0.30(0.26+0.03) -80.3%
5311.20: size (16 days) 8.0M 2.0M -74.5%
5311.21: client (16 days) 0.40(0.47+0.03) 1.01(0.98+0.04) +152.5%
5311.23: server (32 days) 2.40(4.44+0.20) 0.31(0.26+0.04) -87.1%
5311.24: size (32 days) 14.1M 4.1M -70.9%
5311.25: client (32 days) 0.70(0.90+0.03) 1.81(1.75+0.06) +158.6%
5311.27: server (64 days) 11.76(26.57+0.29) 0.55(0.50+0.08) -95.3%
5311.28: size (64 days) 89.4M 47.4M -47.0%
5311.29: client (64 days) 5.71(9.31+0.27) 15.20(15.20+0.32) +166.2%
5311.31: server (128 days) 16.15(36.87+0.40) 0.91(0.82+0.14) -94.4%
5311.32: size (128 days) 134.8M 100.4M -25.5%
5311.33: client (128 days) 9.42(16.86+0.49) 25.34(25.80+0.46) +169.0%
In all cases we save CPU time on the server (sometimes
significant) and the resulting pack is smaller. We do spend
more CPU time on the client side, because it has to
reconstruct more deltas. But that's the right tradeoff to
make, since clients tend to outnumber servers. It just means
the thin pack mechanism is doing its job.
From the user's perspective, the end-to-end time of the
operation will generally be faster. E.g., in the 128-day
case, we saved 15s on the server at a cost of 16s on the
client. Since the resulting pack is 34MB smaller, this is a
net win if the network speed is less than 270Mbit/s. And
that's actually the worst case. The 64-day case saves just
over 11s at a cost of just under 11s. So it's a slight win
at any network speed, and the 40MB saved is pure bonus. That
trend continues for the smaller fetches.
The implementation itself is mostly straightforward, with
the new logic going into check_object(). But there are two
tricky bits.
The first is that check_object() needs access to the
relevant information (the thin flag and bitmap result). We
can do this by pushing these into program-lifetime globals.
The second is that the rest of the code assumes that any
reused delta will point to another "struct object_entry" as
its base. But of course the case we are interested in here
is the one where don't have such an entry!
I looked at a number of options that didn't quite work:
- we could use a flag to signal a reused delta, but it's
not a single bit. We have to actually store the oid of
the base, which is normally done by pointing to the
existing object_entry. And we'd have to modify all the
code which looks at deltas.
- we could add the reused bases to the end of the existing
object_entry array. While this does create some extra
work as later stages consider the extra entries, it's
actually not too bad (we're not sending them, so they
don't cost much in the delta search, and at most we'd
have 2*N of them).
But there's a more subtle problem. Adding to the existing
array means we might need to grow it with realloc, which
could move the earlier entries around. While many of the
references to other entries are done by integer index,
some (including ones on the stack) use pointers, which
would become invalidated.
This isn't insurmountable, but it would require quite a
bit of refactoring (and it's hard to know that you've got
it all, since it may work _most_ of the time and then
fail subtly based on memory allocation patterns).
- we could allocate a new one-off entry for the base. In
fact, this is what an earlier version of this patch did.
However, since the refactoring brought in by ad635e82d6
(Merge branch 'nd/pack-objects-pack-struct', 2018-05-23),
the delta_idx code requires that both entries be in the
main packing list.
So taking all of those options into account, what I ended up
with is a separate list of "external bases" that are not
part of the main packing list. Each delta entry that points
to an external base has a single-bit flag to do so; we have a
little breathing room in the bitfield section of
object_entry.
This lets us limit the change primarily to the oe_delta()
and oe_set_delta_ext() functions. And as a bonus, most of
the rest of the code does not consider these dummy entries
at all, saving both runtime CPU and code complexity.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
