Internal API clean-up to allow write_locked_index() optionally skip
writing the in-core index when it is not modified.
* ma/skip-writing-unchanged-index:
write_locked_index(): add flag to avoid writing unchanged index
We have several callers like
if (active_cache_changed && write_locked_index(...))
handle_error();
rollback_lock_file(...);
where the final rollback is needed because "!active_cache_changed"
shortcuts the if-expression. There are also a few variants of this,
including some if-else constructs that make it more clear when the
explicit rollback is really needed.
Teach `write_locked_index()` to take a new flag SKIP_IF_UNCHANGED and
simplify the callers. Leave the most complicated of the callers (in
builtin/update-index.c) unchanged. Rewriting it to use this new flag
would end up duplicating logic.
We could have made the new flag behave the other way round
("FORCE_WRITE"), but that could break existing users behind their backs.
Let's take the more conservative approach. We can still migrate existing
callers to use our new flag. Later we might even be able to flip the
default, possibly without entirely ignoring the risk to in-flight or
out-of-tree topics.
Suggested-by: Jeff King <peff@peff.net>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Rename C++ keyword in order to bring the codebase closer to being able
to be compiled with a C++ compiler.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
More abstraction of hash function from the codepath.
* bc/hash-algo:
hash: update obsolete reference to SHA1_HEADER
bulk-checkin: abstract SHA-1 usage
csum-file: abstract uses of SHA-1
csum-file: rename sha1file to hashfile
read-cache: abstract away uses of SHA-1
pack-write: switch various SHA-1 values to abstract forms
pack-check: convert various uses of SHA-1 to abstract forms
fast-import: switch various uses of SHA-1 to the_hash_algo
sha1_file: switch uses of SHA-1 to the_hash_algo
builtin/unpack-objects: switch uses of SHA-1 to the_hash_algo
builtin/index-pack: improve hash function abstraction
hash: create union for hash context allocation
hash: move SHA-1 macros to hash.h
Code clean-up.
* nd/shared-index-fix:
read-cache: don't write index twice if we can't write shared index
read-cache.c: move tempfile creation/cleanup out of write_shared_index
read-cache.c: change type of "temp" in write_shared_index()
The split-index mode had a few corner case bugs fixed.
* tg/split-index-fixes:
travis: run tests with GIT_TEST_SPLIT_INDEX
split-index: don't write cache tree with null oid entries
read-cache: fix reading the shared index for other repos
Convert various uses of direct calls to SHA-1 and 20- and 40-based
constants to use the_hash_algo instead. Don't yet convert the on-disk
data structures, which will be handled in a future commit.
Adjust some comments so as not to refer explicitly to SHA-1.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
All the known heavy code blocks are measured (except object database
access). This should help identify if an optimization is effective or
not. An unoptimized git-status would give something like below:
0.001791141 s: read cache ...
0.004011363 s: preload index
0.000516161 s: refresh index
0.003139257 s: git command: ... 'status' '--porcelain=2'
0.006788129 s: diff-files
0.002090267 s: diff-index
0.001885735 s: initialize name hash
0.032013138 s: read directory
0.051781209 s: git command: './git' 'status'
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert the definition and declaration of write_sha1_file to
struct object_id and adjust usage of this function.
This commit also converts static function write_sha1_file_prepare, as it
is closely related.
Rename these functions to write_object_file and
write_object_file_prepare respectively.
Replace sha1_to_hex, hashcpy and hashclr with their oid equivalents
wherever possible.
Signed-off-by: Patryk Obara <patryk.obara@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In a0a967568e ("update-index --split-index: do not split if $GIT_DIR is
read only", 2014-06-13), we tried to make sure we can still write an
index, even if the shared index can not be written.
We did so by just calling 'do_write_locked_index()' just before
'write_shared_index()'. 'do_write_locked_index()' always at least
closes the tempfile nowadays, and used to close or commit the lockfile
if COMMIT_LOCK or CLOSE_LOCK were given at the time this feature was
introduced. COMMIT_LOCK or CLOSE_LOCK is passed in by most callers of
'write_locked_index()'.
After calling 'write_shared_index()', we call 'write_split_index()',
which calls 'do_write_locked_index()' again, which then tries to use the
closed lockfile again, but in fact fails to do so as it's already
closed. This eventually leads to a segfault.
Make sure to write the main index only once.
[nd: most of the commit message and investigation done by Thomas, I only
tweaked the solution a bit]
Helped-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Use the helper macro MOVE_ARRAY to move arrays. This is shorter and
safer, as it automatically infers the size of elements.
Patch generated by Coccinelle and contrib/coccinelle/array.cocci in
Travis CI's static analysis build job.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In a96d3cc3f6 ("cache-tree: reject entries with null sha1", 2017-04-21)
we made sure that broken cache entries do not get propagated to new
trees. Part of that was making sure not to re-use an existing cache
tree that includes a null oid.
It did so by dropping the cache tree in 'do_write_index()' if one of
the entries contains a null oid. In split index mode however, there
are two invocations to 'do_write_index()', one for the shared index
and one for the split index. The cache tree is only written once, to
the split index.
As we only loop through the elements that are effectively being
written by the current invocation, that may not include the entry with
a null oid in the split index (when it is already written to the
shared index), where we write the cache tree. Therefore in split
index mode we may still end up writing the cache tree, even though
there is an entry with a null oid in the index.
Fix this by checking for null oids in prepare_to_write_split_index,
where we loop the entries of the shared index as well as the entries for
the split index.
This fixes t7009 with GIT_TEST_SPLIT_INDEX. Also add a new test that's
more specifically showing the problem.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
read_index_from() takes a path argument for the location of the index
file. For reading the shared index in split index mode however it just
ignores that path argument, and reads it from the gitdir of the current
repository.
This works as long as an index in the_repository is read. Once that
changes, such as when we read the index of a submodule, or of a
different working tree than the current one, the gitdir of
the_repository will no longer contain the appropriate shared index,
and git will fail to read it.
For example t3007-ls-files-recurse-submodules.sh was broken with
GIT_TEST_SPLIT_INDEX set in 188dce131f ("ls-files: use repository
object", 2017-06-22), and t7814-grep-recurse-submodules.sh was also
broken in a similar manner, probably by introducing struct repository
there, although I didn't track down the exact commit for that.
be489d02d2 ("revision.c: --indexed-objects add objects from all
worktrees", 2017-08-23) breaks with split index mode in a similar
manner, not erroring out when it can't read the index, but instead
carrying on with pruning, without taking the index of the worktree into
account.
Fix this by passing an additional gitdir parameter to read_index_from,
to indicate where it should look for and read the shared index from.
read_cache_from() defaults to using the gitdir of the_repository. As it
is mostly a convenience macro, having to pass get_git_dir() for every
call seems overkill, and if necessary users can have more control by
using read_index_from().
Helped-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For one thing, we have more consistent cleanup procedure now and always
keep errno intact.
The real purpose is the ability to break out of write_locked_index()
early when mks_tempfile() fails in the next patch. It's more awkward to
do it if this mks_tempfile() is still inside write_shared_index().
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This local variable 'temp' will be passed in from the caller in the next
patch. To reduce patch noise, let's change its type now while it's still
a local variable and get all the trival conversion out of the next patch.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git add --renormalize ." is a new and safer way to record the fact
that you are correcting the end-of-line convention and other
"convert_to_git()" glitches in the in-repository data.
* tb/add-renormalize:
add: introduce "--renormalize"
Various fixes to bp/fsmonitor topic.
* av/fsmonitor:
fsmonitor: simplify determining the git worktree under Windows
fsmonitor: store fsmonitor bitmap before splitting index
fsmonitor: read from getcwd(), not the PWD environment variable
fsmonitor: delay updating state until after split index is merged
fsmonitor: document GIT_TRACE_FSMONITOR
fsmonitor: don't bother pretty-printing JSON from watchman
fsmonitor: set the PWD to the top of the working tree
We learned to talk to watchman to speed up "git status" and other
operations that need to see which paths have been modified.
* bp/fsmonitor:
fsmonitor: preserve utf8 filenames in fsmonitor-watchman log
fsmonitor: read entirety of watchman output
fsmonitor: MINGW support for watchman integration
fsmonitor: add a performance test
fsmonitor: add a sample integration script for Watchman
fsmonitor: add test cases for fsmonitor extension
split-index: disable the fsmonitor extension when running the split index test
fsmonitor: add a test tool to dump the index extension
update-index: add fsmonitor support to update-index
ls-files: Add support in ls-files to display the fsmonitor valid bit
fsmonitor: add documentation for the fsmonitor extension.
fsmonitor: teach git to optionally utilize a file system monitor to speed up detecting new or changed files.
update-index: add a new --force-write-index option
preload-index: add override to enable testing preload-index
bswap: add 64 bit endianness helper get_be64
Make it safer to normalize the line endings in a repository.
Files that had been commited with CRLF will be commited with LF.
The old way to normalize a repo was like this:
# Make sure that there are not untracked files
$ echo "* text=auto" >.gitattributes
$ git read-tree --empty
$ git add .
$ git commit -m "Introduce end-of-line normalization"
The user must make sure that there are no untracked files,
otherwise they would have been added and tracked from now on.
The new "add --renormalize" does not add untracked files:
$ echo "* text=auto" >.gitattributes
$ git add --renormalize .
$ git commit -m "Introduce end-of-line normalization"
Note that "git add --renormalize <pathspec>" is the short form for
"git add -u --renormalize <pathspec>".
While at it, document that the same renormalization may be needed,
whenever a clean filter is added or changed.
Helped-By: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Torsten Bögershausen <tboegi@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Drop (perhaps overly cautious) sanity check before using the index
read from the filesystem at runtime.
* bp/read-index-from-skip-verification:
read_index_from(): speed index loading by skipping verification of the entry order
ba1b9cac ("fsmonitor: delay updating state until after split index
is merged", 2017-10-27) resolved the problem of the fsmonitor data
being applied to the non-base index when reading; however, a similar
problem exists when writing the index. Specifically, writing of the
fsmonitor extension happens only after the work to split the index
has been applied -- as such, the information in the index is only
for the non-"base" index, and thus the extension information
contains only partial data.
When saving, compute the ewah bitmap before the index is split, and
store it in the fsmonitor_dirty field, mirroring the behavior that
occurred during reading. fsmonitor_dirty is kept from being leaked by
being freed when the extension data is written -- which always happens
precisely once, no matter the split index configuration.
Signed-off-by: Alex Vandiver <alexmv@dropbox.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
There is code in post_read_index_from() to catch out of order
entries when reading an index file. This order verification is ~13%
of the cost of every call to read_index_from().
Update check_ce_order() so that it skips this verification unless
the "verify_ce_order" global variable is set.
Teach fsck to force this verification.
The effect can be seen using t/perf/p0002-read-cache.sh:
Test HEAD HEAD~1
--------------------------------------------------------------------------------------
0002.1: read_cache/discard_cache 1000 times 0.41(0.04+0.04) 0.50(0.00+0.10) +22.0%
Signed-off-by: Ben Peart <benpeart@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Conversion from uchar[20] to struct object_id continues.
* bc/object-id: (25 commits)
refs/files-backend: convert static functions to object_id
refs: convert read_raw_ref backends to struct object_id
refs: convert peel_object to struct object_id
refs: convert resolve_ref_unsafe to struct object_id
worktree: convert struct worktree to object_id
refs: convert resolve_gitlink_ref to struct object_id
Convert remaining callers of resolve_gitlink_ref to object_id
sha1_file: convert index_path and index_fd to struct object_id
refs: convert reflog_expire parameter to struct object_id
refs: convert read_ref_at to struct object_id
refs: convert peel_ref to struct object_id
builtin/pack-objects: convert to struct object_id
pack-bitmap: convert traverse_bitmap_commit_list to object_id
refs: convert dwim_log to struct object_id
builtin/reflog: convert remaining unsigned char uses to object_id
refs: convert dwim_ref and expand_ref to struct object_id
refs: convert read_ref and read_ref_full to object_id
refs: convert resolve_refdup and refs_resolve_refdup to struct object_id
Convert check_connected to use struct object_id
refs: update ref transactions to use struct object_id
...
Convert the declaration and definition of resolve_gitlink_ref to use
struct object_id and apply the following semantic patch:
@@
expression E1, E2, E3;
@@
- resolve_gitlink_ref(E1, E2, E3.hash)
+ resolve_gitlink_ref(E1, E2, &E3)
@@
expression E1, E2, E3;
@@
- resolve_gitlink_ref(E1, E2, E3->hash)
+ resolve_gitlink_ref(E1, E2, E3)
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
`update_index_if_able()` used to always commit the lock or roll it back.
Commit 03b866477 (read-cache: new API write_locked_index instead of
write_index/write_cache, 2014-06-13) stopped rolling it back in case a
write was not even attempted. This change in behavior is not motivated
in the commit message and appears to be accidental: the `else`-path was
removed, although that changed the behavior in case the `if` shortcuts.
Reintroduce the rollback and document this behavior. While at it, move
the documentation on this function from the function definition to the
function declaration in cache.h.
If `write_locked_index(..., COMMIT_LOCK)` fails, it will roll back the
lock for us (see the previous commit).
Noticed-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If the original version of `write_locked_index()` returned with an
error, it didn't roll back the lockfile unless the error occured at the
very end, during closing/committing. See commit 03b866477 (read-cache:
new API write_locked_index instead of write_index/write_cache,
2014-06-13).
In commit 9f41c7a6b (read-cache: close index.lock in do_write_index,
2017-04-26), we learned to close the lock slightly earlier in the
callstack. That was mostly a side-effect of lockfiles being implemented
using temporary files, but didn't cause any real harm.
Recently, commit 076aa2cbd (tempfile: auto-allocate tempfiles on heap,
2017-09-05) introduced a subtle bug. If the temporary file is deleted
(i.e., the lockfile is rolled back), the tempfile-pointer in the `struct
lock_file` will be left dangling. Thus, an attempt to reuse the
lockfile, or even just to roll it back, will induce undefined behavior
-- most likely a crash.
Besides not crashing, we clearly want to make things consistent. The
guarantees which the lockfile-machinery itself provides is A) if we ask
to commit and it fails, roll back, and B) if we ask to close and it
fails, do _not_ roll back. Let's do the same for consistency.
Do not delete the temporary file in `do_write_index()`. One of its
callers, `write_locked_index()` will thereby avoid rolling back the
lock. The other caller, `write_shared_index()`, will delete its
temporary file anyway. Both of these callers will avoid undefined
behavior (crashing).
Teach `write_locked_index(..., COMMIT_LOCK)` to roll back the lock
before returning. If we have already succeeded and committed, it will be
a noop. Simplify the existing callers where we now have a superfluous
call to `rollback_lockfile()`. That should keep future readers from
wondering why the callers are inconsistent.
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
`write_locked_index()` takes two flags: `COMMIT_LOCK` and `CLOSE_LOCK`.
At most one is allowed. But it is also possible to use no flag, i.e.,
`0`. But when `write_locked_index()` calls `do_write_index()`, the
temporary file, a.k.a. the lockfile, will be closed. So passing `0` is
effectively the same as `CLOSE_LOCK`, which seems like a bug.
We might feel tempted to restructure the code in order to close the file
later, or conditionally. It also feels a bit unfortunate that we simply
"happen" to close the lock by way of an implementation detail of
lockfiles. But note that we need to close the temporary file before
`stat`-ing it, at least on Windows. See 9f41c7a6b (read-cache: close
index.lock in do_write_index, 2017-04-26).
Drop `CLOSE_LOCK` and make it explicit that `write_locked_index()`
always closes the lock. Whether it is also committed is governed by the
remaining flag, `COMMIT_LOCK`.
This means we neither have nor suggest that we have a mode to write the
index and leave the file open. Whatever extra contents we might
eventually want to write, we should probably write it from within
`write_locked_index()` itself anyway.
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When the index is read from disk, the fsmonitor index extension is used
to flag the last known potentially dirty index entries. The registered
core.fsmonitor command is called with the time the index was last
updated and returns the list of files changed since that time. This list
is used to flag any additional dirty cache entries and untracked cache
directories.
We can then use this valid state to speed up preload_index(),
ie_match_stat(), and refresh_cache_ent() as they do not need to lstat()
files to detect potential changes for those entries marked
CE_FSMONITOR_VALID.
In addition, if the untracked cache is turned on valid_cached_dir() can
skip checking directories for new or changed files as fsmonitor will
invalidate the cache only for those directories that have been
identified as having potential changes.
To keep the CE_FSMONITOR_VALID state accurate during git operations;
when git updates a cache entry to match the current state on disk,
it will now set the CE_FSMONITOR_VALID bit.
Inversely, anytime git changes a cache entry, the CE_FSMONITOR_VALID bit
is cleared and the corresponding untracked cache directory is marked
invalid.
Signed-off-by: Ben Peart <benpeart@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Many codepaths have been updated to squelch -Wimplicit-fallthrough
warnings from Gcc 7 (which is a good code hygiene).
* jk/fallthrough:
consistently use "fallthrough" comments in switches
curl_trace(): eliminate switch fallthrough
test-line-buffer: simplify command parsing
Many codepaths did not diagnose write failures correctly when disks
go full, due to their misuse of write_in_full() helper function,
which have been corrected.
* jk/write-in-full-fix:
read_pack_header: handle signed/unsigned comparison in read result
config: flip return value of store_write_*()
notes-merge: use ssize_t for write_in_full() return value
pkt-line: check write_in_full() errors against "< 0"
convert less-trivial versions of "write_in_full() != len"
avoid "write_in_full(fd, buf, len) != len" pattern
get-tar-commit-id: check write_in_full() return against 0
config: avoid "write_in_full(fd, buf, len) < len" pattern
A hotfix to a topic already in 'master'.
* kw/write-index-reduce-alloc:
read-cache: fix index corruption with index v4
Add t/helper/test-write-cache to .gitignore
Gcc 7 adds -Wimplicit-fallthrough, which can warn when a
switch case falls through to the next case. The general idea
is that the compiler can't tell if this was intentional or
not, so you should annotate any intentional fall-throughs as
such, leaving it to complain about any unannotated ones.
There's a GNU __attribute__ which can be used for
annotation, but of course we'd have to #ifdef it away on
non-gcc compilers. Gcc will also recognize
specially-formatted comments, which matches our current
practice. Let's extend that practice to all of the
unannotated sites (which I did look over and verify that
they were behaving as intended).
Ideally in each case we'd actually give some reasons in the
comment about why we're falling through, or what we're
falling through to. And gcc does support that with
-Wimplicit-fallthrough=2, which relaxes the comment pattern
matching to anything that contains "fallthrough" (or a
variety of spelling variants). However, this isn't the
default for -Wimplicit-fallthrough, nor for -Wextra. In the
name of simplicity, it's probably better for us to support
the default level, which requires "fallthrough" to be the
only thing in the comment (modulo some window dressing like
"else" and some punctuation; see the gcc manual for the
complete set of patterns).
This patch suppresses all warnings due to
-Wimplicit-fallthrough. We might eventually want to add that
to the DEVELOPER Makefile knob, but we should probably wait
until gcc 7 is more widely adopted (since earlier versions
will complain about the unknown warning type).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The return value of write_in_full() is either "-1", or the
requested number of bytes[1]. If we make a partial write
before seeing an error, we still return -1, not a partial
value. This goes back to f6aa66cb95 (write_in_full: really
write in full or return error on disk full., 2007-01-11).
So checking anything except "was the return value negative"
is pointless. And there are a couple of reasons not to do
so:
1. It can do a funny signed/unsigned comparison. If your
"len" is signed (e.g., a size_t) then the compiler will
promote the "-1" to its unsigned variant.
This works out for "!= len" (unless you really were
trying to write the maximum size_t bytes), but is a
bug if you check "< len" (an example of which was fixed
recently in config.c).
We should avoid promoting the mental model that you
need to check the length at all, so that new sites are
not tempted to copy us.
2. Checking for a negative value is shorter to type,
especially when the length is an expression.
3. Linus says so. In d34cf19b89 (Clean up write_in_full()
users, 2007-01-11), right after the write_in_full()
semantics were changed, he wrote:
I really wish every "write_in_full()" user would just
check against "<0" now, but this fixes the nasty and
stupid ones.
Appeals to authority aside, this makes it clear that
writing it this way does not have an intentional
benefit. It's a historical curiosity that we never
bothered to clean up (and which was undoubtedly
cargo-culted into new sites).
So let's convert these obviously-correct cases (this
includes write_str_in_full(), which is just a wrapper for
write_in_full()).
[1] A careful reader may notice there is one way that
write_in_full() can return a different value. If we ask
write() to write N bytes and get a return value that is
_larger_ than N, we could return a larger total. But
besides the fact that this would imply a totally broken
version of write(), it would already invoke undefined
behavior. Our internal remaining counter is an unsigned
size_t, which means that subtracting too many byte will
wrap it around to a very large number. So we'll instantly
begin reading off the end of the buffer, trying to write
gigabytes (or petabytes) of data.
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
ce012deb98 ("read-cache: avoid allocating every ondisk entry when
writing", 2017-08-21) changed the way cache entries are written to the
index file. While previously it wrote the name to an struct that was
allocated using xcalloc(), it now uses ce_write() directly. Previously
ce_namelen - common bytes were written to the cache entry, which would
automatically make it nul terminated, as it was allocated using calloc.
Now we are writing ce_namelen - common + 1 bytes directly from the
ce->name to the index. If CE_STRIP_NAME however gets set in the split
index case ce->ce_namelen is set to 0 without changing the actual
ce->name buffer. When index-v4, this results in the first character of
ce->name being written out instead of just a terminating nul charcter.
As index-v4 requires the terminating nul character as terminator of
the name when reading it back, this results in a corrupted index.
Fix that by only writing ce_namelen - common bytes directly from
ce->name to the index, and adding the nul terminator in an extra call to
ce_write.
This bug was turned up by setting TEST_GIT_INDEX_VERSION = 4 in
config.mak and running the test suite (t1700 specifically broke).
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The previous commit taught the tempfile code to give up
ownership over tempfiles that have been renamed or deleted.
That makes it possible to use a stack variable like this:
struct tempfile t;
create_tempfile(&t, ...);
...
if (!err)
rename_tempfile(&t, ...);
else
delete_tempfile(&t);
But doing it this way has a high potential for creating
memory errors. The tempfile we pass to create_tempfile()
ends up on a global linked list, and it's not safe for it to
go out of scope until we've called one of those two
deactivation functions.
Imagine that we add an early return from the function that
forgets to call delete_tempfile(). With a static or heap
tempfile variable, the worst case is that the tempfile hangs
around until the program exits (and some functions like
setup_shallow_temporary rely on this intentionally, creating
a tempfile and then leaving it for later cleanup).
But with a stack variable as above, this is a serious memory
error: the variable goes out of scope and may be filled with
garbage by the time the tempfile code looks at it. Let's
see if we can make it harder to get this wrong.
Since many callers need to allocate arbitrary numbers of
tempfiles, we can't rely on static storage as a general
solution. So we need to turn to the heap. We could just ask
all callers to pass us a heap variable, but that puts the
burden on them to call free() at the right time.
Instead, let's have the tempfile code handle the heap
allocation _and_ the deallocation (when the tempfile is
deactivated and removed from the list).
This changes the return value of all of the creation
functions. For the cleanup functions (delete and rename),
we'll add one extra bit of safety: instead of taking a
tempfile pointer, we'll take a pointer-to-pointer and set it
to NULL after freeing the object. This makes it safe to
double-call functions like delete_tempfile(), as the second
call treats the NULL input as a noop. Several callsites
follow this pattern.
The resulting patch does have a fair bit of noise, as each
caller needs to be converted to handle:
1. Storing a pointer instead of the struct itself.
2. Passing the pointer instead of taking the struct
address.
3. Handling a "struct tempfile *" return instead of a file
descriptor.
We could play games to make this less noisy. For example, by
defining the tempfile like this:
struct tempfile {
struct heap_allocated_part_of_tempfile {
int fd;
...etc
} *actual_data;
}
Callers would continue to have a "struct tempfile", and it
would be "active" only when the inner pointer was non-NULL.
But that just makes things more awkward in the long run.
There aren't that many callers, so we can simply bite
the bullet and adjust all of them. And the compiler makes it
easy for us to find them all.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since the lockfile code is based on the tempfile code, it
has some of the same problems, including that close_lock_file()
erases the tempfile's filename buf, making it hard for the
caller to write a good error message.
In practice this comes up less for lockfiles than for
straight tempfiles, since we usually just report the
refname. But there is at least one buggy case in
write_ref_to_lockfile(). Besides, given the coupling between
the lockfile and tempfile modules, it's less confusing if
their close() functions have the same semantics.
Just as the previous commit did for close_tempfile(), let's
teach close_lock_file() and its wrapper close_ref() not to
rollback on error. And just as before, we'll give them new
"gently" names to catch any new callers that are added.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When close_tempfile() fails, we delete the tempfile and
reset the fields of the tempfile struct. This makes it
easier for callers to return without cleaning up, but it
also makes this common pattern:
if (close_tempfile(tempfile))
return error_errno("error closing %s", tempfile->filename.buf);
wrong, because the "filename" field has been reset after the
failed close. And it's not easy to fix, as in many cases we
don't have another copy of the filename (e.g., if it was
created via one of the mks_tempfile functions, and we just
have the original template string).
Let's drop the feature that a failed close automatically
deletes the file. This puts the burden on the caller to do
the deletion themselves, but this isn't that big a deal.
Callers which do:
if (write(...) || close_tempfile(...)) {
delete_tempfile(...);
return -1;
}
already had to call delete when the write() failed, and so
aren't affected. Likewise, any caller which just calls die()
in the error path is OK; we'll delete the tempfile during
the atexit handler.
Because this patch changes the semantics of close_tempfile()
without changing its signature, all callers need to be
manually checked and converted to the new scheme. This patch
covers all in-tree callers, but there may be others for
not-yet-merged topics. To catch these, we rename the
function to close_tempfile_gently(), which will attract
compile-time attention to new callers. (Technically the
original could be considered "gentle" already in that it
didn't die() on errors, but this one is even more so).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We used to spend more than necessary cycles allocating and freeing
piece of memory while writing each index entry out. This has been
optimized.
* kw/write-index-reduce-alloc:
read-cache: avoid allocating every ondisk entry when writing
read-cache: fix memory leak in do_write_index
perf: add test for writing the index
When writing the index for each entry an ondisk struct will be
allocated and freed in ce_write_entry. We can do better by
using a ondisk struct on the stack for each entry.
This is accomplished by using a stack ondisk_cache_entry_extended
outside looping through the entries in do_write_index. Only the
fixed fields of this struct are used when writing and depending on
whether it is extended or not the flags2 field will be written.
The name field is not used and instead the cache_entry name field
is used directly when writing out the name. Because ce_write is
using a buffer and memcpy to fill the buffer before flushing to disk,
we don't have to worry about doing multiple ce_write calls.
Running the p0007-write-cache.sh tests would save anywhere
between 3-7% when the index had over a million entries with no
performance degradation on small repos.
Signed-off-by: Kevin Willford <kewillf@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The previous_name_buf was never getting released when there
was an error in ce_write_entry or allow was false and execution
was returned to the caller.
Signed-off-by: Kevin Willford <kewillf@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Simplify the code for moving members inside of an array and make it more
robust by using the helper macro MOVE_ARRAY. It calculates the size
based on the specified number of elements for us and supports NULL
pointers when that number is zero. Raw memmove(3) calls with NULL can
cause the compiler to (over-eagerly) optimize out later NULL checks.
This patch was generated with contrib/coccinelle/array.cocci and spatch
(Coccinelle).
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The split index code did not honor core.sharedrepository setting
correctly.
* cc/shared-index-permfix:
t1700: make sure split-index respects core.sharedrepository
t1301: move modebits() to test-lib-functions.sh
read-cache: use shared perms when writing shared index
