git-clone started respecting errors from the transport subsystem in
aab179d937 (builtin/clone.c: don't ignore transport_fetch_refs() errors,
2020-12-03). However, that commit didn't handle the cleanup of the
filesystem quite right.
The cleanup of the directory that cmd_clone() creates is done by an
atexit() handler, which we control with a flag. It starts as
JUNK_LEAVE_NONE ("clean up everything"), then progresses to
JUNK_LEAVE_REPO when we know we have a valid repo but not working tree,
and then finally JUNK_LEAVE_ALL when we have a successful checkout.
Most errors cause us to die(), which then triggers the handler to do the
right thing based on how far into cmd_clone() we got. But the checks
added by aab179d937 instead set the "err" variable and then jump to a
new "cleanup" label, which then returns our non-zero status. However,
the code after the cleanup label includes setting the flag to
JUNK_LEAVE_ALL, and so we accidentally leave the repository and working
tree in place.
One obvious option to fix this is to reorder the end of the function to
set the flag first, before cleanup code, and put the label between them.
But we can observe another small bug: the error return from
transport_fetch_refs() is generally "-1", and we propagate that to the
return value of cmd_clone(), which ultimately becomes the exit code of
the process. And we try to avoid transmitting negative values via exit
codes (only the low 8 bits are passed along as an unsigned value, though
in practice for "-1" this at least retains the property that it's
non-zero).
Instead, let's just die(). That makes us consistent with rest of the
code in the function. It does add a new "fatal:" line to the output, but
I'd argue that's a good thing:
- in the rare case that the transport code didn't say anything, now
the user gets _some_ error message
- even if the transport code said something like "error: ssh died of
signal 9", it's nice to also say "fatal" to indicate that we
considered that to be a show-stopper.
Triggering this in the test suite turns out to be surprisingly
difficult. Almost every error we'd encounter, including ones deep inside
the transport code, cause us to just die() right there! However, one way
is to put a fake wrapper around git-upload-pack that sends the complete
packfile but exits with a failure code.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The text was somewhat confusing between the revision itself and the author.
Signed-off-by: Reuven Yagel <robi@post.jce.ac.il>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
These methods were marked as MAYBE_UNUSED in the previous change to
avoid a complicated diff. Delete them entirely, since we now use the
hashfile API instead of this custom hashing code.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The do_write_index() method in read-cache.c has its own hashing logic
and buffering mechanism. Specifically, the ce_write() method was
introduced by 4990aadc (Speed up index file writing by chunking it
nicely, 2005-04-20) and similar mechanisms were introduced a few months
later in c38138cd (git-pack-objects: write the pack files with a SHA1
csum, 2005-06-26). Based on the timing, in the early days of the Git
codebase, I figured that these roughly equivalent code paths were never
unified only because it got lost in the shuffle. The hashfile API has
since been used extensively in other file formats, such as pack-indexes,
multi-pack-indexes, and commit-graphs. Therefore, it seems prudent to
unify the index writing code to use the same mechanism.
I discovered this disparity while trying to create a new index format
that uses the chunk-format API. That API uses a hashfile as its base, so
it is incompatible with the custom code in read-cache.c.
This rewrite is rather straightforward. It replaces all writes to the
temporary file with writes to the hashfile struct. This takes care of
many of the direct interactions with the_hash_algo.
There are still some git_hash_ctx uses remaining: the extension headers
are hashed for use in the End of Index Entries (EOIE) extension. This
use of the git_hash_ctx is left as-is. There are multiple reasons to not
use a hashfile here, including the fact that the data is not actually
writing to a file, just a hash computation. These hashes do not block
our adoption of the chunk-format API in a future change to the index, so
leave it as-is.
The internals of the algorithms are mostly identical. Previously, the
hashfile API used a smaller 8KB buffer instead of the 128KB buffer from
read-cache.c. The previous change already unified these sizes.
There is one subtle point: we do not pass the CSUM_FSYNC to the
finalize_hashfile() method, which differs from most consumers of the
hashfile API. The extra fsync() call indicated by this flag causes a
significant peformance degradation that is noticeable for quick
commands that write the index, such as "git add". Other consumers can
absorb this cost with their more complicated data structure
organization, and further writing structures such as pack-files and
commit-graphs is rarely in the critical path for common user
interactions.
Some static methods become orphaned in this diff, so I marked them as
MAYBE_UNUSED. The diff is much harder to read if they are deleted during
this change. Instead, they will be deleted in the following change.
In addition to the test suite passing, I computed indexes using the
previous binaries and the binaries compiled after this change, and found
the index data to be exactly equal. Finally, I did extensive performance
testing of "git update-index --force-write" on repos of various sizes,
including one with over 2 million paths at HEAD. These tests
demonstrated less than 1% difference in behavior. As expected, the
performance should be considered unchanged. The previous changes to
increase the hashfile buffer size from 8K to 128K ensured this change
would not create a peformance regression.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The hashfile API uses a hard-coded buffer size of 8KB and has ever since
it was introduced in c38138c (git-pack-objects: write the pack files
with a SHA1 csum, 2005-06-26). It performs a similar function to the
hashing buffers in read-cache.c, but that code was updated from 8KB to
128KB in f279894 (read-cache: make the index write buffer size 128K,
2021-02-18). The justification there was that do_write_index() improves
from 1.02s to 0.72s. Since our end goal is to have the index writing
code use the hashfile API, we need to unify this buffer size to avoid a
performance regression.
There is a buffer, 'check_buffer', that is used to verify the check_fd
file descriptor. When this buffer increases to 128K to fit the data
being flushed, it causes the stack to overflow the limits placed in the
test suite. To avoid issues with stack size, move both 'buffer' and
'check_buffer' to be heap pointers within 'struct hashfile'. The
'check_buffer' member is left as NULL unless check_fd is set in
hashfd_check(). Both buffers are cleared as part of finalize_hashfile()
which also frees the full structure.
Since these buffers are now on the heap, we can adjust their size based
on the needs of the consumer. In particular, callers to
hashfd_throughput() are expecting to report progress indicators as the
buffer flushes. These callers would prefer the smaller 8k buffer to
avoid large delays between updates, especially for users with slower
networks. When the progress indicator is not used, the larger buffer is
preferrable.
By adding a new trace2 region in the chunk-format API, we can see that
the writing portion of 'git multi-pack-index write' lowers from ~1.49s
to ~1.47s on a Linux machine. These effects may be more pronounced or
diminished on other filesystems. The end-to-end timing is too noisy to
have a definitive change either way.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The xsize_t helper aims to safely convert an off_t to a size_t,
erroring out when a file offset is too large to fit into a memory
address. It does this by using two casts:
size_t size = (size_t) len;
if (len != (off_t) size)
... error out ...
On a platform with sizeof(size_t) < sizeof(off_t), this check is safe
and correct. The first cast truncates to a size_t by finding the
remainder modulo SIZE_MAX+1 (see C99 section 6.3.1.3 Signed and
unsigned integers) and the second promotes to an off_t, meaning the
result is true if and only if len is representable as a size_t.
On other platforms, this two-casts strategy still works well (always
succeeds) for len >= 0. But for len < 0, when the first cast succeeds
and produces SIZE_MAX + 1 + len, the resulting value is too large to
be represented as an off_t, so the second cast produces implementation
defined behavior. In practice, it is likely to produce a result of
true despite len not being representable as size_t.
Simplify by replacing with a more straightforward check: compare len
to the relevant bounds and then cast it. (To avoid a -Wsign-compare
warning, after checking that len >= 0, we explicitly convert to a
sufficiently-large unsigned type before comparing to SIZE_MAX.)
In practice, this is not likely to come up since typical callers use
nonnegative len. Still, it's helpful to handle this case to make the
behavior easy to reason about.
Historical note: the original bounds-checking in 46be82dfd0 (xsize_t:
check whether we lose bits, 2010-07-28) did not produce this
implementation-defined behavior, though it still did not handle
negative offsets. It was not until 73560c793a (git-compat-util.h:
xsize_t() - avoid -Wsign-compare warnings, 2017-09-21) introduced the
double cast that the implementation-defined behavior was triggered.
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This reverts commit 1b0d9545bb.
That commit does fix the situation it intended to (avoiding Negotiate
even when the credentials were provided in the URL), but it creates a
more serious regression: we now never hit the conditional for "we had a
username and password, tried them, but the server still gave us a 401".
That has two bad effects:
1. we never call credential_reject(), and thus a bogus credential
stored by a helper will live on forever
2. we never return HTTP_NOAUTH, so the error message the user gets is
"The requested URL returned error: 401", instead of "Authentication
failed".
Doing this correctly seems non-trivial, as we don't know whether the
Negotiate auth was a problem. Since this is a regression in the upcoming
v2.23.0 release (for which we're in -rc0), let's revert for now and work
on a fix separately.
(Note that this isn't a pure revert; the previous commit added a test
showing the regression, so we can now flip it to expect_success).
Reported-by: Ben Humphreys <behumphreys@atlassian.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We test authentication with http, and we independently test that
credential helpers work, but we don't have any tests that cover the
two features working together. Let's add two:
1. Make sure that a successful request asks the helper to save the
credential. This works as expected.
2. Make sure that a failed request asks the helper to forget the
credential. This is marked as expect_failure, as it was recently
regressed by 1b0d9545bb (remote-curl: fall back to basic auth if
Negotiate fails, 2021-03-22). The symptom here is that the second
request should prompt the user, but doesn't.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Sometimes new people are confused by how a revision "range" works,
in that it is not a random collection of commits but a set of
commits that are all connected to each other, and most Git commands
work on a single such "range".
Give an example to clarify it.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The flush() logic in csum-file.c was introduced originally by c38138c
(git-pack-objects: write the pack files with a SHA1 csum, 2005-06-26)
and a portion of the logic performs similar utility to write_in_full()
in wrapper.c. The history of write_in_full() is full of moves and
renames, but was originally introduced by 7230e6d (Add write_or_die(), a
helper function, 2006-08-21).
The point of these sections of code are to flush a write buffer using
xwrite() and report errors in the case of disk space issues or other
generic input/output errors. The logic in flush() can interpret the
output of write_in_full() to provide the correct error messages to
users.
The logic in the hashfile API has an additional set of logic to augment
the progress indicator between calls to xwrite(). This was introduced by
2a128d6 (add throughput display to git-push, 2007-10-30). It seems that
since the hashfile's buffer is only 8KB, these additional progress
indicators might not be incredibly necessary. Instead, update the
progress only when write_in_full() complete.
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
While testing the sparse-index, I verified a test with --valgrind and it
complained about an uninitialized value being used in a jump in the
path_matches_pattern_list() method. The line was this one:
if (*dtype == DT_UNKNOWN)
In the call stack, the culprit was the initialization of the dtype
variable in convert_to_sparse_rec().
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
An object_id storing a SHA-1 name has some unused bytes at the end of
the hash array. Since these bytes are not used, they are usually not
initialized to any value either. However, at
parallel_checkout.c:send_one_item() the object_id of a cache entry is
copied into a buffer which is later sent to a checkout worker through a
pipe write(). This makes Valgrind complain about passing uninitialized
bytes to a syscall. The worker won't use these uninitialized bytes
either, but the warning could confuse someone trying to debug this code;
So instead of using oidcpy(), send_one_item() uses hashcpy() to only
copy the used/initialized bytes of the object_id, and leave the
remaining part with zeros.
However, since cf0983213c ("hash: add an algo member to struct
object_id", 2021-04-26), using hashcpy() is no longer sufficient here as
it won't copy the new algo field from the object_id. Let's add and use a
new function which meets both our requirements of copying all the
important object_id data while still avoiding the uninitialized bytes,
by padding the end of the hash array in the destination object_id. With
this change, we also no longer need the destination buffer from
send_one_item() to be initialized with zeros, so let's switch from
xcalloc() to xmalloc() to make this clear.
Signed-off-by: Matheus Tavares <matheus.bernardino@usp.br>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The C_LOCALE_OUTPUT prerequisite was removed in b1e079807b (tests:
remove last uses of C_LOCALE_OUTPUT, 2021-02-11), where Ævar noted:
I'm not leaving the prerequisite itself in place for in-flight changes
as there currently are none that introduce new tests that rely on it,
and because C_LOCALE_OUTPUT is currently a noop on the master branch
we likely won't have any new submissions that use it.
One more use of C_LOCALE_OUTPUT did creep in with 3d1bda6b5b (t7500: add
tests for --fixup=[amend|reword] options, 2021-03-15). This causes a
number of the tests to be skipped by default:
ok 35 # SKIP --fixup=reword: incompatible with --all (missing C_LOCALE_OUTPUT)
ok 36 # SKIP --fixup=reword: incompatible with --include (missing C_LOCALE_OUTPUT)
ok 37 # SKIP --fixup=reword: incompatible with --only (missing C_LOCALE_OUTPUT)
ok 38 # SKIP --fixup=reword: incompatible with --interactive (missing C_LOCALE_OUTPUT)
ok 39 # SKIP --fixup=reword: incompatible with --patch (missing C_LOCALE_OUTPUT)
Remove the C_LOCALE_OUTPUT prerequisite from these tests so they are
not skipped.
Signed-off-by: Todd Zullinger <tmz@pobox.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
These error messages are intended for the user. Let's touch them up
since we're here from the previous commit.
Signed-off-by: Wolfgang Müller <wolf@oriole.systems>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Calling "git rev-parse --path-format" without an argument segfaults
instead of giving an error message. Commit fac60b8925 (rev-parse: add
option for absolute or relative path formatting, 2020-12-13) added the
argument parsing code but forgot to handle NULL.
Returning an error makes sense here because there is no default value we
could use. Add a test case to verify.
Signed-off-by: Wolfgang Müller <wolf@oriole.systems>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* Eliminated 'Negation of emptiness' of 'nenhum' (not one/none)
* Eliminated 'Negation of emptiness' of 'nada' (nothing)
* Transformed 'Não' (No) into affirmative
* Some other translations
* Transforming 'não' (no) into affirmative
* From junção-de-3 to tri-junção
Signed-off-by: Daniel Santos <hello@brighterdan.com>
Clarify the default length used for the abbreviated form used for
commits in git describe.
The behavior was modified in Git 2.11.0, but the documentation was not
updated to clarify the new behavior.
Signed-off-by: Anders Höckersten <anders@hockersten.se>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
I sometimes receive patches from people with short mononyms, and in my
cultural environment these are not uncommon. To my dismay, git-am
currently discards their names, and replaces them with their email
addresses.
Link: https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/
Signed-off-by: edef <edef@edef.eu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The sendemail.smtpServer configuration option and --smtp-server command
line option both support using a sendmail-like program to send emails by
specifying an absolute file path. However, this is not ideal for the
following reasons:
1. It overloads the meaning of smtpServer (now a program is being used
for the server?)
2. It doesn't allow for non-absolute paths, arguments, or arbitrary
scripting
Requiring an absolute path is bad for portability, as the same program
may be in different locations on different systems. If a user wishes to
pass arguments to their program, they have to use the smtpServerOption
option, which is cumbersome (as it must be repeated for each option) and
doesn't adhere to normal git conventions.
Introduce a new configuration option sendemail.sendmailCmd as well as a
command line option --sendmail-cmd that can be used to specify a command
(with or without arguments) or shell expression to run to send email.
The name of this option is consistent with --to-cmd and --cc-cmd. This
invocation honors the user's $PATH so that absolute paths are not
necessary. Arbitrary shell expressions are also supported, allowing
users to do basic scripting.
Give this option a higher precedence over --smtp-server and
sendemail.smtpServer, as the new interface is more flexible. For
backward compatibility, continue to support absolute paths in
--smtp-server and sendemail.smtpServer.
Signed-off-by: Gregory Anders <greg@gpanders.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The code to handle options recently added to "git stash show"
around untracked part of the stash segfaulted when these options
were used on a stash entry that does not record untracked part.
* dl/stash-show-untracked-fixup:
stash show: fix segfault with --{include,only}-untracked
t3905: correct test title
When "git update-ref -d" removes a ref that is packed, it left
empty directories under $GIT_DIR/refs/ for
* wc/packed-ref-removal-cleanup:
refs: cleanup directories when deleting packed ref
"git mailinfo" (hence "git am") learned the "--quoted-cr" option to
control how lines ending with CRLF wrapped in base64 or qp are
handled.
* dd/mailinfo-quoted-cr:
am: learn to process quoted lines that ends with CRLF
mailinfo: allow stripping quoted CR without warning
mailinfo: allow squelching quoted CRLF warning
mailinfo: warn if CRLF found in decoded base64/QP email
mailinfo: stop parsing options manually
mailinfo: load default metainfo_charset lazily
The final part of "parallel checkout".
* mt/parallel-checkout-part-3:
ci: run test round with parallel-checkout enabled
parallel-checkout: add tests related to .gitattributes
t0028: extract encoding helpers to lib-encoding.sh
parallel-checkout: add tests related to path collisions
parallel-checkout: add tests for basic operations
checkout-index: add parallel checkout support
builtin/checkout.c: complete parallel checkout support
make_transient_cache_entry(): optionally alloc from mem_pool
"git push" learns to discover common ancestor with the receiving
end over protocol v2.
* jt/push-negotiation:
send-pack: support push negotiation
fetch: teach independent negotiation (no packfile)
fetch-pack: refactor command and capability write
fetch-pack: refactor add_haves()
fetch-pack: refactor process_acks()
These strings have not been modified in any translation, nor should they
be.
Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git add -i --dry-run" does not dry-run, which was surprising. The
combination of options has taught to error out.
* ow/no-dryrun-in-add-i:
add: die if both --dry-run and --interactive are given
"git p4" learned to find branch points more efficiently.
* jk/p4-locate-branch-point-optim:
git-p4: speed up search for branch parent
git-p4: ensure complex branches are cloned correctly
Over-the-wire protocol learns a new request type to ask for object
sizes given a list of object names.
* ba/object-info:
object-info: support for retrieving object info
The word-diff mode has been taught to work better with a word
regexp that can match an empty string.
* pw/word-diff-zero-width-matches:
word diff: handle zero length matches
In the original ref-filter design, it will copy the parsed
atom's name and attributes to `used_atom[i].name` in the
atom's parsing step, and use it again for string matching
in the later specific ref attributes filling step. It use
a lot of string matching to determine which atom we need.
Introduce the enum "atom_type", each enum value is named
as `ATOM_*`, which is the index of each corresponding
valid_atom entry. In the first step of the atom parsing,
`used_atom.atom_type` will record corresponding enum value
from valid_atom entry index, and then in specific reference
attribute filling step, only need to compare the value of
the `used_atom[i].atom_type` to check the atom type.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Christian Couder <christian.couder@gmail.com>
Signed-off-by: ZheNing Hu <adlternative@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When the support for "objectsize:disk" was bolted onto the
existing support for "objectsize", it didn't follow the
usual pattern for handling "atomtype:modifier", which reads
the <modifier> part just once while parsing the format
string, and store the parsed result in the union in the
used_atom structure, so that the string form of it does not
have to be parsed over and over at runtime (e.g. in
grab_common_values()).
Add a new member `objectsize` to the union `used_atom.u`,
so that we can separate the check of <modifier> from the
check of <atomtype>, this will bring scalability to atom
`%(objectsize)`.
Signed-off-by: ZheNing Hu <adlternative@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit 878f988350 (t/test-lib: teach --chain-lint to detect broken
&&-chains in subshells, 2018-07-11) introduced additional chain-lint
tests which add an extra "sed" pipeline to each test we run. This has a
measurable impact on runtime. Here are timings with and without a new
environment variable (added by this patch) that lets you disable just
the additional sed-based chain-lint tests:
Benchmark #1: GIT_TEST_CHAIN_LINT_HARDER=1 make test
Time (mean ± σ): 64.202 s ± 1.030 s [User: 622.469 s, System: 301.402 s]
Range (min … max): 61.571 s … 65.662 s 10 runs
Benchmark #2: GIT_TEST_CHAIN_LINT_HARDER=0 make test
Time (mean ± σ): 57.591 s ± 0.333 s [User: 529.368 s, System: 270.618 s]
Range (min … max): 57.143 s … 58.309 s 10 runs
Summary
'GIT_TEST_CHAIN_LINT_HARDER=0 make test' ran
1.11 ± 0.02 times faster than 'GIT_TEST_CHAIN_LINT_HARDER=1 make test'
Of course those extra lint checks are doing something useful, so paying
a few extra seconds (at least on Linux) isn't so bad (though note the
CPU time; we're bounded in our parallel run here by the slowest test, so
it really is ~120s of CPU improvement).
But we can observe that there are some test scripts where they produce a
much stronger effect, and provide less value. In t0027 and t3070 we run
a very large number of small tests, all driven by a series of
functions/loops which are filling in the test bodies. There we get much
less bang for our buck in terms of bug-finding versus CPU cost.
This patch introduces a mechanism for controlling when those extra
lint checks are run, at two levels:
- a user can ask to disable or to force-enable the checks by setting
GIT_TEST_CHAIN_LINT_HARDER
- if the user hasn't specified a preference, individual scripts can
disable the checks by setting GIT_TEST_CHAIN_LINT_HARDER_DEFAULT;
scripts which don't set that get the current behavior of enabling
them.
In addition, this patch flips the default for t0027 and t3070's
mass-generated sections to disable the extra checks. Here are the timing
results for t0027:
Benchmark #1: GIT_TEST_CHAIN_LINT_HARDER=1 ./t0027-auto-crlf.sh
Time (mean ± σ): 17.078 s ± 0.848 s [User: 14.878 s, System: 7.075 s]
Range (min … max): 15.952 s … 18.421 s 10 runs
Benchmark #2: GIT_TEST_CHAIN_LINT_HARDER=0 ./t0027-auto-crlf.sh
Time (mean ± σ): 9.063 s ± 0.759 s [User: 7.890 s, System: 3.362 s]
Range (min … max): 7.747 s … 10.619 s 10 runs
Benchmark #3: ./t0027-auto-crlf.sh
Time (mean ± σ): 9.186 s ± 0.881 s [User: 7.957 s, System: 3.427 s]
Range (min … max): 7.796 s … 10.498 s 10 runs
Summary
'GIT_TEST_CHAIN_LINT_HARDER=0 ./t0027-auto-crlf.sh' ran
1.01 ± 0.13 times faster than './t0027-auto-crlf.sh'
1.88 ± 0.18 times faster than 'GIT_TEST_CHAIN_LINT_HARDER=1 ./t0027-auto-crlf.sh'
We can see that disabling the checks for the whole script buys us an
almost 2x speedup. But the new default behavior, disabling them only for
the mass-generated part, gets us most of that speedup (but still leaves
the checks on for further manual tests people might write).
As a side note, I'd caution about comparing runtimes and CPU seconds
between this timing and the earlier "make test" one. In "make test",
we're running a lot of scripts in parallel, so the CPU is throttling
down (and thus a CPU second saved here would count for more during a
parallel run; the same work takes more CPU seconds there).
We get similar results for t3070:
Benchmark #1: GIT_TEST_CHAIN_LINT_HARDER=1 ./t3070-wildmatch.sh
Time (mean ± σ): 20.054 s ± 3.967 s [User: 16.003 s, System: 8.286 s]
Range (min … max): 11.891 s … 23.671 s 10 runs
Benchmark #2: GIT_TEST_CHAIN_LINT_HARDER=0 ./t3070-wildmatch.sh
Time (mean ± σ): 12.399 s ± 2.256 s [User: 7.542 s, System: 5.342 s]
Range (min … max): 9.606 s … 15.727 s 10 runs
Benchmark #3: ./t3070-wildmatch.sh
Time (mean ± σ): 10.726 s ± 3.476 s [User: 6.790 s, System: 4.365 s]
Range (min … max): 5.444 s … 15.376 s 10 runs
Summary
'./t3070-wildmatch.sh' ran
1.16 ± 0.43 times faster than 'GIT_TEST_CHAIN_LINT_HARDER=0 ./t3070-wildmatch.sh'
1.87 ± 0.71 times faster than 'GIT_TEST_CHAIN_LINT_HARDER=1 ./t3070-wildmatch.sh'
Again, we get almost a 2x speedup disabling these. In this case, there
are no tests not covered by the script's "default to disable" behavior,
so the second two benchmarks should be the same (and while they do
differ, you can see the variance is quite high but they're within one
standard deviation).
So it seems like for these two scripts, at least, disabling the extra
checks is a reasonable tradeoff. Sadly, the overall runtime of "make
test" on my system doesn't get much faster. But that's because we're
mostly limited by the cost of the single biggest test. Here are the
top-5 tests by wall-clock time from a parallel run, before my patch:
57.9192368984222 t9001-send-email.sh
45.6329638957977 t0027-auto-crlf.sh
32.5278220176697 t3070-wildmatch.sh
22.2701289653778 t7610-mergetool.sh
20.8635759353638 t1701-racy-split-index.sh
And after:
57.1476998329163 t9001-send-email.sh
33.776211977005 t0027-auto-crlf.sh
21.3116669654846 t7610-mergetool.sh
20.7748689651489 t1701-racy-split-index.sh
19.6957249641418 t7112-reset-submodule.sh
We dropped 12s from t0027, and t3070 dropped off our list entirely at
around 16s. In both cases we're bound by t9001, but its slowness is
due to the actual tests, so we'll have to deal with it in a different
way. But this reduces overall CPU, and means that dealing with t9001 (by
improving the speed of send-email or splitting it apart) will let us
reduce our overall runtime even on multi-core machines.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
