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
Since f6ecc62dbf (write_shared_index(): use tempfile module, 2015-08-10)
write_shared_index() has been using mks_tempfile() to create the
temporary file that will become the shared index.
But even before that, it looks like the functions used to create this
file didn't call adjust_shared_perm(), which means that the shared
index file has always been created with 600 permissions regardless
of the shared permission settings.
Because of that, on repositories created with `git init --shared=all`
and using the split index feature, one gets an error like:
fatal: .git/sharedindex.a52f910b489bc462f187ab572ba0086f7b5157de: index file open failed: Permission denied
when another user performs any operation that reads the shared index.
Call adjust_shared_perm() on the temporary file created by
mks_tempfile() ourselves to adjust the permission bits.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A common pattern to free a piece of memory and assign NULL to the
pointer that used to point at it has been replaced with a new
FREE_AND_NULL() macro.
* ab/free-and-null:
*.[ch] refactoring: make use of the FREE_AND_NULL() macro
coccinelle: make use of the "expression" FREE_AND_NULL() rule
coccinelle: add a rule to make "expression" code use FREE_AND_NULL()
coccinelle: make use of the "type" FREE_AND_NULL() rule
coccinelle: add a rule to make "type" code use FREE_AND_NULL()
git-compat-util: add a FREE_AND_NULL() wrapper around free(ptr); ptr = NULL
Fix configuration codepath to pay proper attention to commondir
that is used in multi-worktree situation, and isolate config API
into its own header file.
* bw/config-h:
config: don't implicitly use gitdir or commondir
config: respect commondir
setup: teach discover_git_directory to respect the commondir
config: don't include config.h by default
config: remove git_config_iter
config: create config.h
Apply the result of the just-added coccinelle rule. This manually
excludes a few occurrences, mostly things that resulted in many
FREE_AND_NULL() on one line, that'll be manually fixed in a subsequent
change.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Stop including config.h by default in cache.h. Instead only include
config.h in those files which require use of the config system.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The timestamp of the index file is now taken after the file is
closed, to help Windows, on which a stale timestamp is reported by
fstat() on a file that is opened for writing and data was written
but not yet closed.
* jh/close-index-before-stat:
read-cache: close index.lock in do_write_index
The timestamp of the index file is now taken after the file is
closed, to help Windows, on which a stale timestamp is reported by
fstat() on a file that is opened for writing and data was written
but not yet closed.
* jh/close-index-before-stat:
read-cache: close index.lock in do_write_index
When "git checkout", "git merge", etc. manipulates the in-core
index, various pieces of information in the index extensions are
discarded from the original state, as it is usually not the case
that they are kept up-to-date and in-sync with the operation on the
main index. The untracked cache extension is copied across these
operations now, which would speed up "git status" (as long as the
cache is properly invalidated).
* dt/unpack-save-untracked-cache-extension:
unpack-trees: preserve index extensions
Plug some leaks and updates internal API used to implement the
split index feature to make it easier to avoid such a leak in the
future.
* nd/split-index-unshare:
p3400: add perf tests for rebasing many changes
split-index: add and use unshare_split_index()
Make git checkout (and other unpack_tree operations) preserve the
untracked cache. This is valuable for two reasons:
1. Often, an unpack_tree operation will not touch large parts of the
working tree, and thus most of the untracked cache will continue to be
valid.
2. Even if the untracked cache were entirely invalidated by such an
operation, the user has signaled their intention to have such a cache,
and we don't want to throw it away.
[jes: backed out the watchman-specific parts]
Signed-off-by: David Turner <dturner@twopensource.com>
Signed-off-by: Ben Peart <benpeart@microsoft.com>
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code to update the cache-tree has been tightened so that we won't
accidentally write out any 0{40} entry in the tree object.
* jk/no-null-sha1-in-cache-tree:
cache-tree: reject entries with null sha1
When split-index is being used, we have two cache_entry arrays in
index_state->cache[] and index_state->split_index->base->cache[].
index_state->cache[] may share the same entries with base->cache[] so
we can quickly determine what entries are shared. This makes memory
management tricky, we can't free base->cache[] until we know
index_state->cache[] does not point to any of those entries.
unshare_split_index() is added for this purpose, to find shared
entries and either duplicate them in index_state->cache[], or discard
them. Either way it should be safe to free base->cache[] after
unshare_split_index().
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Peter Krefting <peter@softwolves.pp.se>
Signed-off-by: Jean-Noel Avila <jn.avila@free.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach do_write_index() to close the index.lock file
before getting the mtime and updating the istate.timestamp
fields.
On Windows, a file's mtime is not updated until the file is
closed. On Linux, the mtime is set after the last flush.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The split-index code configuration code used an unsafe git_path()
function without copying its result out.
* cc/split-index-config:
read-cache: avoid using git_path() in freshen_shared_index()
"git checkout" that handles a lot of paths has been optimized by
reducing the number of unnecessary checks of paths in the
has_dir_name() function.
* jh/add-index-entry-optim:
read-cache: speed up has_dir_name (part 2)
read-cache: speed up has_dir_name (part 1)
read-cache: speed up add_index_entry during checkout
p0006-read-tree-checkout: perf test to time read-tree
read-cache: add strcmp_offset function
We generally disallow null sha1s from entering the index,
due to 4337b5856 (do not write null sha1s to on-disk index,
2012-07-28). However, we loosened that in 83bd7437c
(write_index: optionally allow broken null sha1s,
2013-08-27) so that tools like filter-branch could be used
to repair broken history.
However, we should make sure that these broken entries do
not get propagated into new trees. For most entries, we'd
catch them with the missing-object check (since presumably
the null sha1 does not exist in our object database). But
gitlink entries do not need reachability, so we may blindly
copy the entry into a bogus tree.
This patch rejects all null sha1s (with the same "invalid
entry" message that missing objects get) when building trees
from the index. It does so even for non-gitlinks, and even
when "write-tree" is given the --missing-ok flag. The null
sha1 is a special sentinel value that is already rejected in
trees by fsck; whether the object exists or not, it is an
error to put it in a tree.
Note that for this to work, we must also avoid reusing an
existing cache-tree that contains the null sha1. This patch
does so by just refusing to write out any cache tree when
the index contains a null sha1. This is blunter than we need
to be; we could just reject the subtree that contains the
offending entry. But it's not worth the complexity. The
behavior is unchanged unless you have a broken index entry,
and even then we'd refuse the whole index write unless the
emergency GIT_ALLOW_NULL_SHA1 is in use. And even then the
end result is only a performance drop (any write-tree will
have to generate the whole cache-tree from scratch).
The tests bear some explanation.
The existing test in t7009 doesn't catch this problem,
because our index-filter runs "git rm --cached", which will
try to rewrite the updated index and barf on the bogus
entry. So we never even make it to write-tree. The new test
there adds a noop index-filter, which does show the problem.
The new tests in t1601 are slightly redundant with what
filter-branch is doing under the hood in t7009. But as
they're much more direct, they're easier to reason about.
And should filter-branch ever change or go away, we'd want
to make sure that these plumbing commands behave sanely.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When performing an interactive rebase in split-index mode,
the commit message that one should rework when squashing commits
can contain some garbage instead of the usual concatenation of
both of the commit messages.
The code uses git_path() to compute the shared index filename, and
passes it to check_and_freshen_file() as its argument; there is no
guarantee that the rotating pathname buffer passed as argument will
stay valid during the life of this call. Make our own copy before
calling the function and pass the copy as its argument to avoid this
risky pattern.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach has_dir_name() to see if the path of the new item
is greater than the last path in the index array before
attempting to search for it.
has_dir_name() is looking for file/directory collisions
in the index and has to consider each sub-directory
prefix in turn. This can cause multiple binary searches
for each path.
During operations like checkout, merge_working_tree()
populates the new index in sorted order, so we expect
to be able to append in many cases.
This commit is part 2 of 2. This commit handles the
additional possible short-cuts as we look at each
sub-directory prefix.
The net-net gains for add_index_entry_with_check() and
both had_dir_name() commits are best seen for very
large repos.
Here are results for an INFLATED version of linux.git
with 1M files.
$ GIT_PERF_REPO=/mnt/test/linux_inflated.git/ ./run upstream/base HEAD ./p0006-read-tree-checkout.sh
Test upstream/base HEAD
0006.2: read-tree br_base br_ballast (1043893) 3.79(3.63+0.15) 2.68(2.52+0.15) -29.3%
0006.3: switch between br_base br_ballast (1043893) 7.55(6.58+0.44) 6.03(4.60+0.43) -20.1%
0006.4: switch between br_ballast br_ballast_plus_1 (1043893) 10.84(9.26+0.59) 8.44(7.06+0.65) -22.1%
0006.5: switch between aliases (1043893) 10.93(9.39+0.58) 10.24(7.04+0.63) -6.3%
Here are results for a synthetic repo with 4.2M files.
$ GIT_PERF_REPO=~/work/gfw/t/perf/repos/gen-many-files-10.4.3.git/ ./run HEAD~3 HEAD ./p0006-read-tree-checkout.sh
Test HEAD~3 HEAD
0006.2: read-tree br_base br_ballast (4194305) 29.96(19.26+10.50) 23.76(13.42+10.12) -20.7%
0006.3: switch between br_base br_ballast (4194305) 56.95(36.08+16.83) 45.54(25.94+15.68) -20.0%
0006.4: switch between br_ballast br_ballast_plus_1 (4194305) 90.94(51.50+31.52) 78.22(39.39+30.70) -14.0%
0006.5: switch between aliases (4194305) 93.72(51.63+34.09) 77.94(39.00+30.88) -16.8%
Results for medium repos (like linux.git) are mixed and have
more variance (probably do to disk IO unrelated to this test.
$ GIT_PERF_REPO=/mnt/test/linux.git/ ./run HEAD~3 HEAD ./p0006-read-tree-checkout.sh
Test HEAD~3 HEAD
0006.2: read-tree br_base br_ballast (57994) 0.25(0.21+0.03) 0.20(0.17+0.02) -20.0%
0006.3: switch between br_base br_ballast (57994) 10.67(6.06+2.92) 10.51(5.94+2.91) -1.5%
0006.4: switch between br_ballast br_ballast_plus_1 (57994) 0.59(0.47+0.16) 0.52(0.40+0.13) -11.9%
0006.5: switch between aliases (57994) 0.59(0.44+0.17) 0.51(0.38+0.14) -13.6%
$ GIT_PERF_REPO=/mnt/test/linux.git/ ./run HEAD~3 HEAD ./p0006-read-tree-checkout.sh
Test HEAD~3 HEAD
0006.2: read-tree br_base br_ballast (57994) 0.24(0.21+0.02) 0.21(0.18+0.02) -12.5%
0006.3: switch between br_base br_ballast (57994) 10.42(5.98+2.91) 10.66(5.86+3.09) +2.3%
0006.4: switch between br_ballast br_ballast_plus_1 (57994) 0.59(0.49+0.13) 0.53(0.37+0.16) -10.2%
0006.5: switch between aliases (57994) 0.59(0.43+0.17) 0.50(0.37+0.14) -15.3%
Results for smaller repos (like git.git) are not significant.
$ ./run HEAD~3 HEAD ./p0006-read-tree-checkout.sh
Test HEAD~3 HEAD
0006.2: read-tree br_base br_ballast (3043) 0.01(0.00+0.00) 0.01(0.00+0.00) +0.0%
0006.3: switch between br_base br_ballast (3043) 0.31(0.17+0.11) 0.29(0.19+0.08) -6.5%
0006.4: switch between br_ballast br_ballast_plus_1 (3043) 0.03(0.02+0.00) 0.03(0.02+0.00) +0.0%
0006.5: switch between aliases (3043) 0.03(0.02+0.00) 0.03(0.02+0.00) +0.0%
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach has_dir_name() to see if the path of the new item
is greater than the last path in the index array before
attempting to search for it.
has_dir_name() is looking for file/directory collisions
in the index and has to consider each sub-directory
prefix in turn. This can cause multiple binary searches
for each path.
During operations like checkout, merge_working_tree()
populates the new index in sorted order, so we expect
to be able to append in many cases.
This commit is part 1 of 2. This commit handles the top
of has_dir_name() and the trivial optimization.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach add_index_entry_with_check() to see if the path
of the new item is greater than the last path in the
index array before attempting to search for it.
During checkout, merge_working_tree() populates the new
index in sorted order, so this change will save a binary
lookups per file. This preserves the original behavior
but simply checks the last element before starting the
search.
This helps performance on very large repositories.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add strcmp_offset() function to also return the offset of the
first change.
Add unit test and helper to verify.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach git to skip verification of the SHA1-1 checksum at the end of
the index file in verify_hdr() which is called from read_index()
unless the "force_verify_index_checksum" global variable is set.
Teach fsck to force this verification.
The checksum verification is for detecting disk corruption, and for
small projects, the time it takes to compute SHA-1 is not that
significant, but for gigantic repositories this calculation adds
significant time to every command.
These effect can be seen using t/perf/p0002-read-cache.sh:
Test HEAD~1 HEAD
--------------------------------------------------------------------------------------
0002.1: read_cache/discard_cache 1000 times 0.66(0.44+0.20) 0.30(0.27+0.02) -54.5%
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The experimental "split index" feature has gained a few
configuration variables to make it easier to use.
* cc/split-index-config: (22 commits)
Documentation/git-update-index: explain splitIndex.*
Documentation/config: add splitIndex.sharedIndexExpire
read-cache: use freshen_shared_index() in read_index_from()
read-cache: refactor read_index_from()
t1700: test shared index file expiration
read-cache: unlink old sharedindex files
config: add git_config_get_expiry() from gc.c
read-cache: touch shared index files when used
sha1_file: make check_and_freshen_file() non static
Documentation/config: add splitIndex.maxPercentChange
t1700: add tests for splitIndex.maxPercentChange
read-cache: regenerate shared index if necessary
config: add git_config_get_max_percent_split_change()
Documentation/git-update-index: talk about core.splitIndex config var
Documentation/config: add information for core.splitIndex
t1700: add tests for core.splitIndex
update-index: warn in case of split-index incoherency
read-cache: add and then use tweak_split_index()
split-index: add {add,remove}_split_index() functions
config: add git_config_get_split_index()
...
This way a share index file will not be garbage collected if
we still read from an index it is based from.
As we need to read the current index before creating a new
one, the tests have to be adjusted, so that we don't expect
an old shared index file to be deleted right away when we
create a new one.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It looks better and is simpler to review when we don't compute
the same things many times in the function.
It will also help make the following commit simpler.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Everytime split index is turned on, it creates a "sharedindex.XXXX"
file in the git directory. This change makes sure that shared index
files that haven't been used for a long time are removed when a new
shared index file is created.
The new "splitIndex.sharedIndexExpire" config variable is created
to tell the delay after which an unused shared index file can be
deleted. It defaults to "2.weeks.ago".
A previous commit made sure that each time a split index file is
created the mtime of the shared index file it references is updated.
This makes sure that recently used shared index file will not be
deleted.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a split-index file is created, let's update the mtime of the
shared index file that the split-index file is referencing.
In a following commit we will make shared index file expire
depending on their mtime, so updating the mtime makes sure that
the shared index file will not be deleted soon.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When writing a new split-index and there is a big number of cache
entries in the split-index compared to the shared index, it is a
good idea to regenerate the shared index.
By default when the ratio reaches 20%, we will push back all
the entries from the split-index into a new shared index file
instead of just creating a new split-index file.
The threshold can be configured using the
"splitIndex.maxPercentChange" config variable.
We need to adjust the existing tests in t1700 by setting
"splitIndex.maxPercentChange" to 100 at the beginning of t1700,
as the existing tests are assuming that the shared index is
regenerated only when `git update-index --split-index` is used.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This will make us use the split-index feature or not depending
on the value of the "core.splitIndex" config variable.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Do this by moving the existing documentation from
read-cache.c to cache.h.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Improve constness of the index_state parameter to the
'read_blob_data_from_index' function.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The codeflow of setting NOATIME and CLOEXEC on file descriptors Git
opens has been simplified.
We may want to drop the tip one, but we'll see.
* jc/git-open-cloexec:
sha1_file: stop opening files with O_NOATIME
git_open_cloexec(): use fcntl(2) w/ FD_CLOEXEC fallback
git_open(): untangle possible NOATIME and CLOEXEC interactions
Callers of the hold_locked_index() function pass 0 when they want to
prepare to write a new version of the index file without wishing to
die or emit an error message when the request fails (e.g. somebody
else already held the lock), and pass 1 when they want the call to
die upon failure.
This option is called LOCK_DIE_ON_ERROR by the underlying lockfile
API, and the hold_locked_index() function translates the paramter to
LOCK_DIE_ON_ERROR when calling the hold_lock_file_for_update().
Replace these hardcoded '1' with LOCK_DIE_ON_ERROR and stop
translating. Callers other than the ones that are replaced with
this change pass '0' to the function; no behaviour change is
intended with this patch.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
Among the callers of hold_locked_index() that passes 0:
- diff.c::refresh_index_quietly() at the end of "git diff" is an
opportunistic update; it leaks the lockfile structure but it is
just before the program exits and nobody should care.
- builtin/describe.c::cmd_describe(),
builtin/commit.c::cmd_status(),
sequencer.c::read_and_refresh_cache() are all opportunistic
updates and they are OK.
- builtin/update-index.c::cmd_update_index() takes a lock upfront
but we may end up not needing to update the index (i.e. the
entries may be fully up-to-date), in which case we do not need to
issue an error upon failure to acquire the lock. We do diagnose
and die if we indeed need to update, so it is OK.
- wt-status.c::require_clean_work_tree() IS BUGGY. It asks
silence, does not check the returned value. Compare with
callsites like cmd_describe() and cmd_status() to notice that it
is wrong to call update_index_if_able() unconditionally.
The way we structured the fallback/retry mechanism for opening with
O_NOATIME and O_CLOEXEC meant that if we failed due to lack of
support to open the file with O_NOATIME option (i.e. EINVAL), we
would still try to drop O_CLOEXEC first and retry, and then drop
O_NOATIME. A platform on which O_NOATIME is defined in the header
without support from the kernel wouldn't have a chance to open with
O_CLOEXEC option due to this code structure.
Arguably, O_CLOEXEC is more important than O_NOATIME, as the latter
is mostly about performance, while the former can affect correctness.
Instead use O_CLOEXEC to open the file, and then use fcntl(2) to set
O_NOATIME on the resulting file descriptor. open(2) itself does not
cause atime to be updated according to Linus [*1*].
The helper to do the former can be usable in the codepath in
ce_compare_data() that was recently added to open a file descriptor
with O_CLOEXEC; use it while we are at it.
*1* <CA+55aFw83E+zOd+z5h-CA-3NhrLjVr-anL6pubrSWttYx3zu8g@mail.gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This fixes "convert: add filter.<driver>.process option" (edcc8581) on
Windows.
Consider the case of a file that requires filtering and is present in
branch A but not in branch B. If A is the current HEAD and we checkout B
then the following happens:
1. ce_compare_data() opens the file
2. index_fd() detects that the file requires to run a clean filter and
calls index_stream_convert_blob()
4. index_stream_convert_blob() calls convert_to_git_filter_fd()
5. convert_to_git_filter_fd() calls apply_filter() which creates a
new long running filter process (in case it is the first file
of this kind to be filtered)
6. The new filter process inherits all file handles. This is the
default on Linux/OSX and is explicitly defined in the
`CreateProcessW` call in `mingw.c` on Windows.
7. ce_compare_data() closes the file
8. Git unlinks the file as it is not present in B
The unlink operation does not work on Windows because the filter process
has still an open handle to the file. On Linux/OSX the unlink operation
succeeds but the file descriptors still leak into the child process.
Fix this problem by opening files in read-cache with the O_CLOEXEC flag
to ensure that the file descriptor does not remain open in a newly
spawned process similar to 05d1ed6148 ("mingw: ensure temporary file
handles are not inherited by child processes", 2016-08-22).
Signed-off-by: Lars Schneider <larsxschneider@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git add --chmod=+x <pathspec>" added recently only toggled the
executable bit for paths that are either new or modified. This has
been corrected to flip the executable bit for all paths that match
the given pathspec.
* tg/add-chmod+x-fix:
t3700-add: do not check working tree file mode without POSIXPERM
t3700-add: create subdirectory gently
add: modify already added files when --chmod is given
read-cache: introduce chmod_index_entry
update-index: add test for chmod flags
When the chmod option was added to git add, it was hooked up to the diff
machinery, meaning that it only works when the version in the index
differs from the version on disk.
As the option was supposed to mirror the chmod option in update-index,
which always changes the mode in the index, regardless of the status of
the file, make sure the option behaves the same way in git add.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
As there are chmod options for both add and update-index, introduce a
new chmod_index_entry function to do the work. Use it in update-index,
while it will be used in add in the next patch.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert struct cache_entry to use struct object_id by applying the
following semantic patch and the object_id transforms from contrib, plus
the actual change to the struct:
@@
struct cache_entry E1;
@@
- E1.sha1
+ E1.oid.hash
@@
struct cache_entry *E1;
@@
- E1->sha1
+ E1->oid.hash
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git merge" with renormalization did not work well with
merge-recursive, due to "safer crlf" conversion kicking in when it
shouldn't.
* jc/renormalize-merge-kill-safer-crlf:
merge: avoid "safer crlf" during recording of merge results
convert: unify the "auto" handling of CRLF
When merge_recursive() decides what the correct blob object merge
result for a path should be, it uses update_file_flags() helper
function to write it out to a working tree file and then calls
add_cacheinfo(). The add_cacheinfo() function in turn calls
make_cache_entry() to create a new cache entry to replace the
higher-stage entries for the path that represents the conflict.
The make_cache_entry() function calls refresh_cache_entry() to fill
in the cached stat information. To mark a cache entry as
up-to-date, the data is re-read from the file in the working tree,
and goes through convert_to_git() conversion to be compared with the
blob object name the new cache entry records.
It is important to note that this happens while the higher-stage
entries, which are going to be replaced with the new entry, are
still in the index. Unfortunately, the convert_to_git() conversion
has a misguided "safer crlf" mechanism baked in, and looks at the
existing cache entry for the path to decide how to convert the
contents in the working tree file. If our side (i.e. stage#2)
records a text blob with CRLF in it, even when the system is
configured to record LF in blobs and convert them to CRLF upon
checkout (and back to LF upon checkin), the "safer crlf" mechanism
stops us doing so.
This especially poses a problem during a renormalizing merge, where
the merge result for the path is computed by first "normalizing" the
blobs involved in the merge by using convert_to_working_tree()
followed by convert_to_git() with "safer crlf" disabled. The merge
result that is computed correctly and fed to add_cacheinfo() via
update_file_flags() does _not_ match what refresh_cache_entry() sees
by converting the working tree file via convert_to_git().
We can work this around by not refreshing the new cache entry in
make_cache_entry() called by add_cacheinfo(). After add_cacheinfo()
adds the new entry, we can call refresh_cache_entry() on that,
knowing that addition of this new cache entry would have removed the
stale cache entries that had CRLF in stage #2 that were carried over
before the renormalizing merge started and will not interfere with
the correct recording of the result.
The test update was taken from a series by Torsten Bögershausen
that attempted to fix this with a different approach.
Signed-off-by: Torsten Bögershausen <tboegi@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Reviewed-by: Torsten Bögershausen <tboegi@web.de>
The executable bit will not be detected (and therefore will not be
set) for paths in a repository with `core.filemode` set to false,
though the users may still wish to add files as executable for
compatibility with other users who _do_ have `core.filemode`
functionality. For example, Windows users adding shell scripts may
wish to add them as executable for compatibility with users on
non-Windows.
Although this can be done with a plumbing command
(`git update-index --add --chmod=+x foo`), teaching the `git-add`
command allows users to set a file executable with a command that
they're already familiar with.
Signed-off-by: Edward Thomson <ethomson@edwardthomson.com>
Helped-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we know that mtime on directory as given by the environment
is usable for the purpose of untracked cache, we may want the
untracked cache to be always used without any mtime test or
kernel name check being performed.
Also when we know that mtime is not usable for the purpose of
untracked cache, for example because the repo is shared over a
network file system, we may want the untracked-cache to be
automatically removed from the index.
Allow the user to express such preference by setting the
'core.untrackedCache' configuration variable, which can take
'keep', 'false', or 'true' and default to 'keep'.
When read_index_from() is called, it now adds or removes the
untracked cache in the index to respect the value of this
variable. So it does nothing if the value is `keep` or if the
variable is unset; it adds the untracked cache if the value is
`true`; and it removes the cache if the value is `false`.
`git update-index --[no-|force-]untracked-cache` still adds the
untracked cache to, or removes it, from the index, but this
shows a warning if it goes against the value of
core.untrackedCache, because the next time the index is read
the untracked cache will be added or removed if the
configuration is set to do so.
Also `--untracked-cache` used to check that the underlying
operating system and file system change `st_mtime` field of a
directory if files are added or deleted in that directory. But
because those tests take a long time, `--untracked-cache` no
longer performs them. Instead, there is now
`--test-untracked-cache` to perform the tests. This change
makes `--untracked-cache` the same as `--force-untracked-cache`.
This last change is backward incompatible and should be
mentioned in the release notes.
Helped-by: Duy Nguyen <pclouds@gmail.com>
Helped-by: Torsten Bögershausen <tboegi@web.de>
Helped-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
read-cache: Duy'sfixup
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Paths that have been told the index about with "add -N" are not
quite yet in the index, but a few commands behaved as if they
already are in a harmful way.
* nd/ita-cleanup:
grep: make it clear i-t-a entries are ignored
add and use a convenience macro ce_intent_to_add()
blame: remove obsolete comment
The name-hash subsystem that is used to cope with case insensitive
filesystems keeps track of directories and their on-filesystem
cases for all the paths in the index by holding a pointer to a
randomly chosen cache entry that is inside the directory (for its
ce->ce_name component). This pointer was not updated even when the
cache entry was removed from the index, leading to use after free.
This was fixed by recording the path for each directory instead of
borrowing cache entries and restructuring the API somewhat.
* dt/name-hash-dir-entry-fix:
name-hash: don't reuse cache_entry in dir_entry
After switching to use the tempfile module in commit f6ecc62d
(write_shared_index(): use tempfile module), no declarations from
sigchain.h are used in read-cache.c anymore. Thus, remove the #include.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Stop reusing cache_entry in dir_entry; doing so causes a
use-after-free bug.
During merges, we free entries that we no longer need in the
destination index. But those entries might have also been stored in
the dir_entry cache, and when a later call to add_to_index found them,
they would be used after being freed.
To prevent this, change dir_entry to store a copy of the name instead
of a pointer to a cache_entry. This entails some refactoring of code
that expects the cache_entry.
Keith McGuigan <kmcguigan@twitter.com> diagnosed this bug and wrote
the initial patch, but this version does not use any of Keith's code.
Helped-by: Keith McGuigan <kmcguigan@twitter.com>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: David Turner <dturner@twopensource.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "lockfile" API has been rebuilt on top of a new "tempfile" API.
* mh/tempfile:
credential-cache--daemon: use tempfile module
credential-cache--daemon: delete socket from main()
gc: use tempfile module to handle gc.pid file
lock_repo_for_gc(): compute the path to "gc.pid" only once
diff: use tempfile module
setup_temporary_shallow(): use tempfile module
write_shared_index(): use tempfile module
register_tempfile(): new function to handle an existing temporary file
tempfile: add several functions for creating temporary files
prepare_tempfile_object(): new function, extracted from create_tempfile()
tempfile: a new module for handling temporary files
commit_lock_file(): use get_locked_file_path()
lockfile: add accessor get_lock_file_path()
lockfile: add accessors get_lock_file_fd() and get_lock_file_fp()
create_bundle(): duplicate file descriptor to avoid closing it twice
lockfile: move documentation to lockfile.h and lockfile.c
We are about to move those members, so change client code to read them
through accessor functions.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Hotfix for the 'untracked-cache' topic that is already in 'master'.
* nd/untracked-cache:
read-cache: fix untracked cache invalidation when split-index is used
The configuration reader/writer uses mmap(2) interface to access
the files; when we find a directory, it barfed with "Out of memory?".
* jk/diagnose-config-mmap-failure:
xmmap(): drop "Out of memory?"
config.c: rewrite ENODEV into EISDIR when mmap fails
config.c: avoid xmmap error messages
config.c: fix mmap leak when writing config
read-cache.c: drop PROT_WRITE from mmap of index
Before this change, t7063.17 fails. The actual action though happens at
t7063.16 where the entry "two" is added back to index after being
removed in the .13. Here we expect a directory invalidate at .16 and
none at .17 where untracked cache is refreshed. But things do not go as
expected when GIT_TEST_SPLIT_INDEX is set.
The different behavior that happens at .16 when split index is used: the
entry "two", when deleted at .13, is simply marked "deleted". When .16
executes, the entry resurfaces from the version in base index. This
happens in merge_base_index() where add_index_entry() is called to add
"two" back from the base index.
This is where the bug comes from. The add_index_entry() is called with
ADD_CACHE_KEEP_CACHE_TREE flag because this version of "two" is not new,
it does not break either cache-tree or untracked cache. The code should
check this flag and not invalidate untracked cache. This causes a second
invalidation violates test expectation. The fix is obvious.
Noticed-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>
Once upon a time, git's in-memory representation of a cache
entry actually pointed to the mmap'd on-disk data. So in
520fc24 (Allow writing to the private index file mapping.,
2005-04-26), we specified PROT_WRITE so that we could tweak
the entries while we run (in our own MAP_PRIVATE copy-on-write
version, of course).
Later, 7a51ed6 (Make on-disk index representation separate
from in-core one, 2008-01-14) stopped doing this; we copy
the data into our in-core representation, and then drop the
mmap immediately. We can therefore drop the PROT_WRITE flag.
It's probably not hurting anything as it is, but it's
potentially confusing.
Note that we could also mark the mapping as "const" to
verify that we never write to it. However, we don't
typically do that for our other maps, as it then requires
casting to munmap() it.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach the index to optionally remember already seen untracked files
to speed up "git status" in a working tree with tons of cruft.
* nd/untracked-cache: (24 commits)
git-status.txt: advertisement for untracked cache
untracked cache: guard and disable on system changes
mingw32: add uname()
t7063: tests for untracked cache
update-index: test the system before enabling untracked cache
update-index: manually enable or disable untracked cache
status: enable untracked cache
untracked-cache: temporarily disable with $GIT_DISABLE_UNTRACKED_CACHE
untracked cache: mark index dirty if untracked cache is updated
untracked cache: print stats with $GIT_TRACE_UNTRACKED_STATS
untracked cache: avoid racy timestamps
read-cache.c: split racy stat test to a separate function
untracked cache: invalidate at index addition or removal
untracked cache: load from UNTR index extension
untracked cache: save to an index extension
ewah: add convenient wrapper ewah_serialize_strbuf()
untracked cache: don't open non-existent .gitignore
untracked cache: mark what dirs should be recursed/saved
untracked cache: record/validate dir mtime and reuse cached output
untracked cache: make a wrapper around {open,read,close}dir()
...
* sb/leaks:
http: release the memory of a http pack request as well
read-cache: fix memleak
add_to_index(): free unused cache-entry
commit.c: fix a memory leak
http-push: remove unneeded cleanup
merge-recursive: fix memleaks
merge-blobs.c: fix a memleak
builtin/apply.c: fix a memleak
update-index: fix a memleak
read-cache: free cache entry in add_to_index in case of early return
The split-index mode introduced at v2.3.0-rc0~41 was broken in the
codepath to protect us against a broken reimplementation of Git
that writes an invalid index with duplicated index entries, etc.
* tg/fix-check-order-with-split-index:
read-cache: fix reading of split index
`ce` is allocated in make_cache_entry and should be freed if it is not
used any more. refresh_cache_entry as a wrapper around refresh_cache_ent
will either return
- the `ce` given as the parameter, when it was up-to-date;
- a new updated cache entry which is allocated to new memory; or
- a NULL when refreshing failed.
In the latter two cases, the original cache-entry `ce` is not used
and needs to be freed. The rule can be expressed as "if the return
value from refresh is different from the original ce, ce is no
longer used."
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We allocate a cache-entry pretty early in the function and then
decide either not to do anything when we are pretending to add, or
add it and then get an error (another possibility is obviously to
succeed).
When pretending or failing to add, we forgot to free the
cache-entry.
Noticed during a discussion on Stefan's patch to change the coding
style without fixing the issue ;-)
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This frees `ce` would be leaking in the error path.
Additionally a free is moved towards the return. This helps code
readability as we often have this pattern of freeing resources just
before return/exit and not in between the code.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The split index extension uses ewah bitmaps to mark index entries as
deleted, instead of removing them from the index directly. This can
result in an on-disk index, in which entries of stage #0 and higher
stages appear, which are removed later when the index bases are merged.
15999d0 read_index_from(): catch out of order entries when reading an
index file introduces a check which checks if the entries are in order
after each index entry is read in do_read_index. This check may however
fail when a split index is read.
Fix this by moving checking the index after we know there is no split
index or after the split index bases are successfully merged instead.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a directory is updated within the same second that its timestamp
is last saved, we cannot realize the directory has been updated by
checking timestamps. Assume the worst (something is update). See
29e4d36 (Racy GIT - 2005-12-20) for more information.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ideally we should implement untracked_cache_remove_from_index() and
untracked_cache_add_to_index() so that they update untracked cache
right away instead of invalidating it and wait for read_directory()
next time to deal with it. But that may need some more work in
unpack-trees.c. So stay simple as the first step.
The new call in add_index_entry_with_check() may look strange because
new calls usually stay close to cache_tree_invalidate_path(). We do it
a bit later than c_t_i_p() in this function because if it's about
replacing the entry with the same name, we don't care (but cache-tree
does).
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"update-index --refresh" used to leak when an entry cannot be
refreshed for whatever reason.
* sb/plug-leak-in-make-cache-entry:
read-cache.c: free cache entry when refreshing fails
This fixes a memory leak when building the cache entries as
refresh_cache_entry may decide to return NULL, but it does not
free the cache entry structure which was passed in as an argument.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The point of disallowing ".git" in the index is that we
would never want to accidentally overwrite files in the
repository directory. But this means we need to respect the
filesystem's idea of when two paths are equal. The prior
commit added a helper to make such a comparison for NTFS
and FAT32; let's use it in verify_path().
We make this check optional for two reasons:
1. It restricts the set of allowable filenames, which is
unnecessary for people who are not on NTFS nor FAT32.
In practice this probably doesn't matter, though, as
the restricted names are rather obscure and almost
certainly would never come up in practice.
2. It has a minor performance penalty for every path we
insert into the index.
This patch ties the check to the core.protectNTFS config
option. Though this is expected to be most useful on Windows,
we allow it to be set everywhere, as NTFS may be mounted on
other platforms. The variable does default to on for Windows,
though.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The point of disallowing ".git" in the index is that we
would never want to accidentally overwrite files in the
repository directory. But this means we need to respect the
filesystem's idea of when two paths are equal. The prior
commit added a helper to make such a comparison for HFS+;
let's use it in verify_path.
We make this check optional for two reasons:
1. It restricts the set of allowable filenames, which is
unnecessary for people who are not on HFS+. In practice
this probably doesn't matter, though, as the restricted
names are rather obscure and almost certainly would
never come up in practice.
2. It has a minor performance penalty for every path we
insert into the index.
This patch ties the check to the core.protectHFS config
option. Though this is expected to be most useful on OS X,
we allow it to be set everywhere, as HFS+ may be mounted on
other platforms. The variable does default to on for OS X,
though.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We do not allow ".git" to enter into the index as a path
component, because checking out the result to the working
tree may causes confusion for subsequent git commands.
However, on case-insensitive file systems, ".Git" or ".GIT"
is the same. We should catch and prevent those, too.
Note that technically we could allow this for repos on
case-sensitive filesystems. But there's not much point. It's
unlikely that anybody cares, and it creates a repository
that is unexpectedly non-portable to other systems.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move the interface declaration for the functions in lockfile.c from
cache.h to a new file, lockfile.h. Add #includes where necessary (and
remove some redundant includes of cache.h by files that already
include builtin.h).
Move the documentation of the lock_file state diagram from lockfile.c
to the new header file.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
lockfile.c contains the general API for locking any file. Code
specifically about the index file doesn't belong here.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit_locked_index(), when writing to an alternate index file,
duplicates (poorly) the code in commit_lock_file(). And anyway, it
shouldn't have to know so much about the internal workings of lockfile
objects. So extract a new function commit_lock_file_to() that does the
work common to the two functions, and call it from both
commit_lock_file() and commit_locked_index().
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For now, we still make sure to allocate at least PATH_MAX characters
for the strbuf because resolve_symlink() doesn't know how to expand
the space for its return value. (That will be fixed in a moment.)
Another alternative would be to just use a strbuf as scratch space in
lock_file() but then store a pointer to the naked string in struct
lock_file. But lock_file objects are often reused. By reusing the
same strbuf, we can avoid having to reallocate the string most times
when a lock_file object is reused.
Helped-by: Torsten Bögershausen <tboegi@web.de>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Because remove_lock_file() can be called any time by the signal
handler, it is important that any lock_file objects that are in the
lock_file_list are always in a valid state. And since lock_file
objects are often reused (but are never removed from lock_file_list),
that means we have to be careful whenever mutating a lock_file object
to always keep it in a well-defined state.
This was formerly not the case, because part of the state was encoded
by setting lk->filename to the empty string vs. a valid filename. It
is wrong to assume that this string can be updated atomically; for
example, even
strcpy(lk->filename, value)
is unsafe. But the old code was even more reckless; for example,
strcpy(lk->filename, path);
if (!(flags & LOCK_NODEREF))
resolve_symlink(lk->filename, max_path_len);
strcat(lk->filename, ".lock");
During the call to resolve_symlink(), lk->filename contained the name
of the file that was being locked, not the name of the lockfile. If a
signal were raised during that interval, then the signal handler would
have deleted the valuable file!
We could probably continue to use the filename field to encode the
state by being careful to write characters 1..N-1 of the filename
first, and then overwrite the NUL at filename[0] with the first
character of the filename, but that would be awkward and error-prone.
So, instead of using the filename field to determine whether the
lock_file object is active, add a new field "lock_file::active" for
this purpose. Be careful to set this field only when filename really
contains the name of a file that should be deleted on cleanup.
Helped-by: Johannes Sixt <j6t@kdbg.org>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Suggested-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A broken reimplementation of Git could write an invalid index that
records both stage #0 and higher stage entries for the same path.
Notice and reject such an index, as there is no sensible fallback
(we do not know if the broken tool wanted to resolve and forgot to
remove higher stage entries, or if it wanted to unresolve and
forgot to remove the stage#0 entry).
* jp/index-with-corrupt-stages:
read_index_unmerged(): remove unnecessary loop index adjustment
read_index_from(): catch out of order entries when reading an index file
Update git_config() users with callback functions for a very narrow
scope with calls to config-set API that lets us query a single
variable.
* ta/config-set-2:
builtin/apply.c: replace `git_config()` with `git_config_get_string_const()`
merge-recursive.c: replace `git_config()` with `git_config_get_int()`
ll-merge.c: refactor `read_merge_config()` to use `git_config_string()`
fast-import.c: replace `git_config()` with `git_config_get_*()` family
branch.c: replace `git_config()` with `git_config_get_string()
alias.c: replace `git_config()` with `git_config_get_string()`
imap-send.c: replace `git_config()` with `git_config_get_*()` family
pager.c: replace `git_config()` with `git_config_get_value()`
builtin/gc.c: replace `git_config()` with `git_config_get_*()` family
rerere.c: replace `git_config()` with `git_config_get_*()` family
fetchpack.c: replace `git_config()` with `git_config_get_*()` family
archive.c: replace `git_config()` with `git_config_get_bool()` family
read-cache.c: replace `git_config()` with `git_config_get_*()` family
http-backend.c: replace `git_config()` with `git_config_get_bool()` family
daemon.c: replace `git_config()` with `git_config_get_bool()` family
"git add x" where x that used to be a directory has become a
symbolic link to a directory misbehaved.
* rs/refresh-beyond-symlink:
read-cache: check for leading symlinks when refreshing index
Don't add paths with leading symlinks to the index while refreshing; we
only track those symlinks themselves. We already ignore them while
preloading (see read_index_preload.c).
Reported-by: Nikolay Avdeev <avdeev@math.vsu.ru>
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Use `git_config_get_*()` family instead of `git_config()` to take
advantage of the config-set API which provides a cleaner control flow.
Use an intermediate value, as `version` can not be used directly in
git_config_get_int() due to incompatible type.
Signed-off-by: Tanay Abhra <tanayabh@gmail.com>
Reviewed-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
An experiment to use two files (the base file and incremental
changes relative to it) to represent the index to reduce I/O cost
of rewriting a large index when only small part of the working tree
changes.
* nd/split-index: (32 commits)
t1700: new tests for split-index mode
t2104: make sure split index mode is off for the version test
read-cache: force split index mode with GIT_TEST_SPLIT_INDEX
read-tree: note about dropping split-index mode or index version
read-tree: force split-index mode off on --index-output
rev-parse: add --shared-index-path to get shared index path
update-index --split-index: do not split if $GIT_DIR is read only
update-index: new options to enable/disable split index mode
split-index: strip pathname of on-disk replaced entries
split-index: do not invalidate cache-tree at read time
split-index: the reading part
split-index: the writing part
read-cache: mark updated entries for split index
read-cache: save deleted entries in split index
read-cache: mark new entries for split index
read-cache: split-index mode
read-cache: save index SHA-1 after reading
entry.c: update cache_changed if refresh_cache is set in checkout_entry()
cache-tree: mark istate->cache_changed on prime_cache_tree()
cache-tree: mark istate->cache_changed on cache tree update
...
"git status", even though it is a read-only operation, tries to
update the index with refreshed lstat(2) info to optimize future
accesses to the working tree opportunistically, but this could
race with a "read-write" operation that modify the index while it
is running. Detect such a race and avoid overwriting the index.
* ym/fix-opportunistic-index-update-race:
read-cache.c: verify index file before we opportunistically update it
wrapper.c: add xpread() similar to xread()
We often represent our strings as a counted string, i.e. a pair of
the pointer to the beginning of the string and its length, and the
string may not be NUL terminated to that length.
To compare a pair of such counted strings, unpack-trees.c and
read-cache.c implement their own name_compare() functions
identically. In addition, the cache_name_compare() function in
read-cache.c is nearly identical. The only difference is when one
string is the prefix of the other string, in which case
name_compare() returns -1/+1 to show which one is longer, and
cache_name_compare() returns the difference of the lengths to show
the same information.
Unify these three functions by using the implementation from
cache_name_compare(). This does not make any difference to the
existing and future callers, as they must be paying attention only
to the sign of the returned value (and not the magnitude) because
the original implementations of these two functions return values
returned by memcmp(3) when the one string is not a prefix of the
other string, and the only thing memcmp(3) guarantees its callers is
the sign of the returned value, not the magnitude.
Signed-off-by: Jeremiah Mahler <jmmahler@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This could be used to run the whole test suite with split
indexes. Index splitting is carried out at random. "git read-tree"
also resets the index and forces splitting at the next update.
I had a lot of headaches with the test suite, which proves it
exercises split index pretty good.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If $GIT_DIR is read only, we can't write $GIT_DIR/sharedindex. This
could happen when $GIT_INDEX_FILE is set to somehwere outside
$GIT_DIR.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If you have a large work tree but only make changes in a subset, then
$GIT_DIR/index's size should be stable after a while. If you change
branches that touch something else, $GIT_DIR/index's size may grow
large that it becomes as slow as the unified index. Do --split-index
again occasionally to force all changes back to the shared index and
keep $GIT_DIR/index small.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We know the positions of replaced entries via the replace bitmap in
"link" extension, so the "name" path does not have to be stored (it's
still in the shared index). With this, we also have a way to
distinguish additions vs replacements at load time and can catch
broken "link" extensions.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We are sure that after merge_base_index() is done. cache-tree can
still be used with the final index. So don't destroy cache tree.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
CE_REMOVE'd entries are removed here because only parts of the code
base (unpack_trees in fact) test this bit when they look for the
presence of an entry. Leaving them may confuse the code ignores this
bit and expects to see a real entry.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The large part of this patch just follows CE_ENTRY_CHANGED
marks. replace_index_entry() is updated to update
split_index->base->cache[] as well so base->cache[] does not reference
to a freed entry.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Entries that belong to the base index should not be freed. Mark
CE_REMOVE to track them.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Make sure entry addition does not lead to unifying the index. We don't
need to explicitly keep track of new entries. If ce->index is zero,
they're new. Otherwise it's unlikely that they are new, but we'll do a
thorough check later at writing time.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This split-index mode is designed to keep write cost proportional to
the number of changes the user has made, not the size of the work
tree. (Read cost is another matter, to be dealt separately.)
This mode stores index info in a pair of $GIT_DIR/index and
$GIT_DIR/sharedindex.<SHA-1>. sharedindex is large and unchanged over
time while "index" is smaller and updated often. Format details are in
index-format.txt, although not everything is implemented in this
patch.
Shared indexes are not automatically removed, because it's unclear if
the shared index is needed by any (even temporary) indexes by just
looking at it. After a while you'll collect stale shared indexes. The
good news is one shared index is useable for long, until
$GIT_DIR/index becomes too big and sluggish that the new shared index
must be created.
The safest way to clean shared indexes is to turn off split index
mode, so shared files are all garbage, delete them all, then turn on
split index mode again.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also update SHA-1 after writing. If we do not do that, the second
read_index() will see "initialized" variable already set and not read
.git/index again, which is fine, except istate->sha1 now has a stale
value.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
cache entry additions, removals and modifications are separated
out. The rest of changes are still in the catch-all flag
SOMETHING_CHANGED, which would be more specific later.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
remove_marked_cache_entries() deletes entries marked with
CE_REMOVE. But if there is no such entry, do not mark the index as
"changed" because that could trigger an index update unnecessarily.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We're running out of room for in-memory flags. But since b60e188
(Strip namelen out of ce_flags into a ce_namelen field - 2012-07-11),
we copy the namelen (first 12 bits) to ce_namelen field. So those bits
are free to use. Just make sure we do not accidentally write any
in-memory flags back.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function is now only used by write_locked_index(). Move it to
read-cache.c (because read-cache.c will need to be aware of
alternate_index_output later) and unexport it.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Read-only operations such as "git status" that internally refreshes
the index write out the refreshed index to the disk to optimize
future accesses to the working tree, but this could race with a
"read-write" operation that modify the index while it is running.
Detect such a race and avoid overwriting the index.
Duy raised a good point that we may need to do the same for the
normal writeout codepath, not just the "opportunistic" update
codepath. While that is true, nobody sane would be running two
simultaneous operations that are clearly write-oriented competing
with each other against the same index file. So in that sense that
can be done as a less urgent follow-up for this topic.
* ym/fix-opportunistic-index-update-race:
read-cache.c: verify index file before we opportunistically update it
wrapper.c: add xpread() similar to xread()
Before we proceed to opportunistically update the index (often done
by an otherwise read-only operation like "git status" and "git diff"
that internally refreshes the index), we must verify that the
current index file is the same as the one that we read earlier
before we took the lock on it, in order to avoid a possible race.
In the example below git-status does "opportunistic update" and
git-rebase updates the index, but the race can happen in general.
1. process A calls git-rebase (or does anything that uses the index)
2. process A applies 1st commit
3. process B calls git-status (or does anything that updates the index)
4. process B reads index
5. process A applies 2nd commit
6. process B takes the lock, then overwrites process A's changes.
7. process A applies 3rd commit
As an end result the 3rd commit will have a revert of the 2nd commit.
When process B takes the lock, it needs to make sure that the index
hasn't changed since step 4.
Signed-off-by: Yiannis Marangos <yiannis.marangos@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"merge-recursive" was broken in 1.7.7 era and stopped working in an
empty (temporary) working tree, when there are renames involved.
This has been corrected.
* bk/refresh-missing-ok-in-merge-recursive:
merge-recursive.c: tolerate missing files while refreshing index
read-cache.c: extend make_cache_entry refresh flag with options
read-cache.c: refactor --ignore-missing implementation
t3030-merge-recursive: test known breakage with empty work tree
Replace open-coded reallocation with ALLOC_GROW() macro.
* dd/use-alloc-grow:
sha1_file.c: use ALLOC_GROW() in pretend_sha1_file()
read-cache.c: use ALLOC_GROW() in add_index_entry()
builtin/mktree.c: use ALLOC_GROW() in append_to_tree()
attr.c: use ALLOC_GROW() in handle_attr_line()
dir.c: use ALLOC_GROW() in create_simplify()
reflog-walk.c: use ALLOC_GROW()
replace_object.c: use ALLOC_GROW() in register_replace_object()
patch-ids.c: use ALLOC_GROW() in add_commit()
diffcore-rename.c: use ALLOC_GROW()
diff.c: use ALLOC_GROW()
commit.c: use ALLOC_GROW() in register_commit_graft()
cache-tree.c: use ALLOC_GROW() in find_subtree()
bundle.c: use ALLOC_GROW() in add_to_ref_list()
builtin/pack-objects.c: use ALLOC_GROW() in check_pbase_path()
* tg/index-v4-format:
read-cache: add index.version config variable
test-lib: allow setting the index format version
introduce GIT_INDEX_VERSION environment variable
Borrow the bitmap index into packfiles from JGit to speed up
enumeration of objects involved in a commit range without having to
fully traverse the history.
* jk/pack-bitmap: (26 commits)
ewah: unconditionally ntohll ewah data
ewah: support platforms that require aligned reads
read-cache: use get_be32 instead of hand-rolled ntoh_l
block-sha1: factor out get_be and put_be wrappers
do not discard revindex when re-preparing packfiles
pack-bitmap: implement optional name_hash cache
t/perf: add tests for pack bitmaps
t: add basic bitmap functionality tests
count-objects: recognize .bitmap in garbage-checking
repack: consider bitmaps when performing repacks
repack: handle optional files created by pack-objects
repack: turn exts array into array-of-struct
repack: stop using magic number for ARRAY_SIZE(exts)
pack-objects: implement bitmap writing
rev-list: add bitmap mode to speed up object lists
pack-objects: use bitmaps when packing objects
pack-objects: split add_object_entry
pack-bitmap: add support for bitmap indexes
documentation: add documentation for the bitmap format
ewah: compressed bitmap implementation
...
Allow "git cmd path/", when the 'path' is where a submodule is
bound to the top-level working tree, to match 'path', despite the
extra and unnecessary trailing slash.
* nd/submodule-pathspec-ending-with-slash:
clean: use cache_name_is_other()
clean: replace match_pathspec() with dir_path_match()
pathspec: pass directory indicator to match_pathspec_item()
match_pathspec: match pathspec "foo/" against directory "foo"
dir.c: prepare match_pathspec_item for taking more flags
pathspec: rename match_pathspec_depth() to match_pathspec()
pathspec: convert some match_pathspec_depth() to dir_path_match()
pathspec: convert some match_pathspec_depth() to ce_path_match()
Allow "merge-recursive" to work in an empty (temporary) working
tree again when there are renames involved, correcting an old
regression in 1.7.7 era.
* bk/refresh-missing-ok-in-merge-recursive:
merge-recursive.c: tolerate missing files while refreshing index
read-cache.c: extend make_cache_entry refresh flag with options
read-cache.c: refactor --ignore-missing implementation
t3030-merge-recursive: test known breakage with empty work tree
This helps reduce the number of match_pathspec_depth() call sites and
show how match_pathspec_depth() is used.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert the make_cache_entry boolean 'refresh' argument to a more
general 'refresh_options' argument. Pass the value through to the
underlying refresh_cache_ent call. Add option CE_MATCH_REFRESH to
enable stat refresh. Update call sites to use the new signature.
Signed-off-by: Brad King <brad.king@kitware.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move lstat ENOENT handling from refresh_index to refresh_cache_ent and
activate it with a new CE_MATCH_IGNORE_MISSING option. This will allow
other call paths into refresh_cache_ent to use the feature.
Signed-off-by: Brad King <brad.king@kitware.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a config variable that allows setting the default index version when
initializing a new index file. Similar to the GIT_INDEX_VERSION
environment variable this only affects new index files.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Respect a GIT_INDEX_VERSION environment variable, when a new index is
initialized. Setting the environment variable will not cause existing
index files to be converted to another format, but will only affect
newly written index files. This can be used to initialize repositories
with index-v4.
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When --mixed is used, entries could be removed from index if the
target ref does not have them. When "reset" is used in preparation for
commit spliting (in a dirty worktree), it could be hard to track what
files to be added back. The new option --intent-to-add simplifies it
by marking all removed files intent-to-add.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Commit d60c49c (read-cache.c: allow unaligned mapping of the
index file, 2012-04-03) introduced helpers to access
unaligned data. However, we already have get_be32, which has
a few advantages:
1. It's already written, so we avoid duplication.
2. It's probably faster, since it does the endian
conversion and the alignment fix at the same time.
3. The get_be32 code is well-tested, having been in
block-sha1 for a long time. By contrast, our custom
helpers were probably almost never used, since the user
needed to manually define a macro to enable them.
We have to add a get_be16 implementation to the existing
get_be32, but that is very simple to do.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When cache_entry structs are removed from index_state.cache, they are not
properly freed. Freeing those entries wasn't possible before because we
couldn't remove them from index_state.name_hash.
Now that we _do_ remove the entries from name_hash, we can also free them.
Add 'free(cache_entry)' to all call sites of name-hash.c::remove_name_hash
in read-cache.c (we could free() directly in remove_name_hash(), but
name-hash.c isn't concerned with cache_entry allocation at all).
Accessing a cache_entry after removing it from the index is now no longer
allowed, as the memory has been freed. The following functions need minor
fixes (typically by copying ce->name before use):
- builtin/rm.c::cmd_rm
- builtin/update-index.c::do_reupdate
- read-cache.c::read_index_unmerged
- resolve-undo.c::unmerge_index_entry_at
Signed-off-by: Karsten Blees <blees@dcon.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The new hashmap implementation supports remove, so really remove unused
cache entries from the name hashmap instead of just marking them.
The CE_UNHASHED flag and CE_STATE_MASK are no longer needed.
Keep the CE_HASHED flag to prevent adding entries twice.
Signed-off-by: Karsten Blees <blees@dcon.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Clean up the internal of the name-hash mechanism used to work
around case insensitivity on some filesystems to cleanly fix a
long-standing API glitch where the caller of cache_name_exists()
that ask about a directory with a counted string was required to
have '/' at one location past the end of the string.
* es/name-hash-no-trailing-slash-in-dirs:
dir: revert work-around for retired dangerous behavior
name-hash: stop storing trailing '/' on paths in index_state.dir_hash
employ new explicit "exists in index?" API
name-hash: refactor polymorphic index_name_exists()
Earlier we started rejecting an attempt to add 0{40} object name to
the index and to tree objects, but it sometimes is necessary to
allow so to be able to use tools like filter-branch to correct such
broken tree objects.
* jk/write-broken-index-with-nul-sha1:
write_index: optionally allow broken null sha1s
When 5102c617 (Add case insensitivity support for directories when using
git status, 2010-10-03) added directories to the name-hash there was
only a single hash table in which both real cache entries and leading
directory prefixes were registered. To distinguish between the two types
of entries, directories were stored with a trailing '/'.
2092678c (name-hash.c: fix endless loop with core.ignorecase=true,
2013-02-28), however, moved directories to a separate hash table
(index_state.dir_hash) but retained the (now) redundant trailing '/',
thus callers continue to bear the burden of ensuring the slash's
presence before searching the index for a directory. Eliminate this
redundancy by storing paths in the dir-hash without the trailing '/'.
An important benefit of this change is that it eliminates undocumented
and dangerous behavior of dir.c:directory_exists_in_index_icase() in
which it assumes not only that it can validly access one character
beyond the end of its incoming directory argument, but also that that
character will unconditionally be a '/'. This perilous behavior was
"tolerated" because the string passed in by its lone caller always had a
'/' in that position, however, things broke [1] when 2eac2a4c (ls-files
-k: a directory only can be killed if the index has a non-directory,
2013-08-15) added a new caller which failed to respect the undocumented
assumption.
[1]: http://thread.gmane.org/gmane.comp.version-control.git/232727
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Each caller of index_name_exists() knows whether it is looking for a
directory or a file, and can avoid the unnecessary indirection of
index_name_exists() by instead calling index_dir_exists() or
index_file_exists() directly.
Invoking the appropriate search function explicitly will allow a
subsequent patch to relieve callers of the artificial burden of having
to add a trailing '/' to the pathname given to index_dir_exists().
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The code that reads from a region that mmaps an on-disk index
assumed that "int"/"short" are always 32/16 bits.
* tg/index-struct-sizes:
read-cache: use fixed width integer types
"git mv A B" when moving a submodule A does "the right thing",
inclusing relocating its working tree and adjusting the paths in
the .gitmodules file.
* jl/submodule-mv: (53 commits)
rm: delete .gitmodules entry of submodules removed from the work tree
mv: update the path entry in .gitmodules for moved submodules
submodule.c: add .gitmodules staging helper functions
mv: move submodules using a gitfile
mv: move submodules together with their work trees
rm: do not set a variable twice without intermediate reading.
t6131 - skip tests if on case-insensitive file system
parse_pathspec: accept :(icase)path syntax
pathspec: support :(glob) syntax
pathspec: make --literal-pathspecs disable pathspec magic
pathspec: support :(literal) syntax for noglob pathspec
kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
parse_pathspec: make sure the prefix part is wildcard-free
rename field "raw" to "_raw" in struct pathspec
tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
remove match_pathspec() in favor of match_pathspec_depth()
remove init_pathspec() in favor of parse_pathspec()
remove diff_tree_{setup,release}_paths
convert common_prefix() to use struct pathspec
...
Commit 4337b58 (do not write null sha1s to on-disk index,
2012-07-28) added a safety check preventing git from writing
null sha1s into the index. The intent was to catch errors in
other parts of the code that might let such an entry slip
into the index (or worse, a tree).
Some existing repositories may have invalid trees that
contain null sha1s already, though. Until 4337b58, a common
way to clean this up would be to use git-filter-branch's
index-filter to repair such broken entries. That now fails
when filter-branch tries to write out the index.
Introduce a GIT_ALLOW_NULL_SHA1 environment variable to
relax this check and make it easier to recover from such a
history.
It is tempting to not involve filter-branch in this commit
at all, and instead require the user to manually invoke
GIT_ALLOW_NULL_SHA1=1 git filter-branch ...
to perform an index-filter on a history with trees with null
sha1s. That would be slightly safer, but requires some
specialized knowledge from the user. So let's set the
GIT_ALLOW_NULL_SHA1 variable automatically when checking out
the to-be-filtered trees. Advice on using filter-branch to
remove such entries already exists on places like
stackoverflow, and this patch makes it Just Work again on
recent versions of git.
Further commands that touch the index will still notice and
fail, unless they actually remove the broken entries. A
filter-branch whose filters do not touch the index at all
will not error out (since we complain of the null sha1 only
on writing, not when making a tree out of the index), but
this is acceptable, as we still print a loud warning, so the
problem is unlikely to go unnoticed.
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Use the fixed width integer types uint16_t and uint32_t for on-disk
structures; unsigned short and unsigned int do not have a guaranteed
size.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
I attempted to make index_state->cache[] a "const struct cache_entry **"
to find out how existing entries in index are modified and where. The
question I have is what do we do if we really need to keep track of on-disk
changes in the index. The result is
- diff-lib.c: setting CE_UPTODATE
- name-hash.c: setting CE_HASHED
- preload-index.c, read-cache.c, unpack-trees.c and
builtin/update-index: obvious
- entry.c: write_entry() may refresh the checked out entry via
fill_stat_cache_info(). This causes "non-const struct cache_entry
*" in builtin/apply.c, builtin/checkout-index.c and
builtin/checkout.c
- builtin/ls-files.c: --with-tree changes stagemask and may set
CE_UPDATE
Of these, write_entry() and its call sites are probably most
interesting because it modifies on-disk info. But this is stat info
and can be retrieved via refresh, at least for porcelain
commands. Other just uses ce_flags for local purposes.
So, keeping track of "dirty" entries is just a matter of setting a
flag in index modification functions exposed by read-cache.c. Except
unpack-trees, the rest of the code base does not do anything funny
behind read-cache's back.
The actual patch is less valueable than the summary above. But if
anyone wants to re-identify the above sites. Applying this patch, then
this:
diff --git a/cache.h b/cache.h
index 430d021..1692891 100644
--- a/cache.h
+++ b/cache.h
@@ -267,7 +267,7 @@ static inline unsigned int canon_mode(unsigned int mode)
#define cache_entry_size(len) (offsetof(struct cache_entry,name) + (len) + 1)
struct index_state {
- struct cache_entry **cache;
+ const struct cache_entry **cache;
unsigned int version;
unsigned int cache_nr, cache_alloc, cache_changed;
struct string_list *resolve_undo;
will help quickly identify them without bogus warnings.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git pack-refs" that races with new ref creation or deletion have
been susceptible to lossage of refs under right conditions, which
has been tightened up.
* mh/ref-races:
for_each_ref: load all loose refs before packed refs
get_packed_ref_cache: reload packed-refs file when it changes
add a stat_validity struct
Extract a struct stat_data from cache_entry
packed_ref_cache: increment refcount when locked
do_for_each_entry(): increment the packed refs cache refcount
refs: manage lifetime of packed refs cache via reference counting
refs: implement simple transactions for the packed-refs file
refs: wrap the packed refs cache in a level of indirection
pack_refs(): split creation of packed refs and entry writing
repack_without_ref(): split list curation and entry writing
It can sometimes be useful to know whether a path in the
filesystem has been updated without going to the work of
opening and re-reading its content. We trust the stat()
information on disk already to handle index updates, and we
can use the same trick here.
This patch introduces a "stat_validity" struct which
encapsulates the concept of checking the stat-freshness of a
file. It is implemented on top of "struct stat_data" to
reuse the logic about which stat entries to trust for a
particular platform, but hides the complexity behind two
simple functions: check and update.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add public functions fill_stat_data() and match_stat_data() to work
with it. This infrastructure will later be used to check the validity
of other types of file.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
discard_cache doesn't have to free the array of cache entries, because
the next call of read_cache can simply reuse it, as they all operate on
the global variable the_index.
discard_index on the other hand does have to free it, because it can be
used e.g. with index_state variables on the stack, in which case a
missing free would cause an unrecoverable leak. This patch releases the
memory and removes a comment that was relevant for discard_cache but has
become outdated.
Since discard_cache is just a wrapper around discard_index nowadays, we
lose the optimization that avoids reallocation of that array within
loops of read_cache and discard_cache. That doesn't cause a performance
regression for me, however (HEAD = this patch, HEAD^ = master + p0002):
Test // HEAD^ HEAD
---------------\\-----------------------------------------------------
0002.1: read_ca// 1000 times 0.62(0.58+0.04) 0.61(0.58+0.02) -1.6%
Suggested-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: René Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We are dealing with the 'istate' index, not 'the_index'.
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
ie_match_stat and ie_modified only derefence their struct cache_entry
pointers for reading. Add const to the parameter declaration here and
do the same for the static helper function used by them, as it's the
same there as well. This allows callers to pass in const pointers.
Signed-off-by: René Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Reduce duplicated code between convert.c and attr.c.
* lf/read-blob-data-from-index:
convert.c: remove duplicate code
read_blob_data_from_index(): optionally return the size of blob data
attr.c: extract read_index_data() as read_blob_data_from_index()
This allows for optionally getting the size of the returned data and
will be used in a follow-up patch.
Signed-off-by: Lukas Fleischer <git@cryptocrack.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Extract the read_index_data() function from attr.c and move it to
read-cache.c; rename it to read_blob_data_from_index() and update
the function signature of it to align better with index/cache API
functions.
This allows for reusing the function in convert.c later.
Signed-off-by: Lukas Fleischer <git@cryptocrack.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The code to keep track of what directory names are known to Git on
platforms with case insensitive filesystems can get confused upon
a hash collision between these pathnames and looped forever.
* kb/name-hash:
name-hash.c: fix endless loop with core.ignorecase=true
Update the index format documentation to mention the v4 format.
* nd/doc-index-format:
update-index: list supported idx versions and their features
read-cache.c: use INDEX_FORMAT_{LB,UB} in verify_hdr()
index-format.txt: mention of v4 is missing in some places
With core.ignorecase=true, name-hash.c builds a case insensitive index of
all tracked directories. Currently, the existing cache entry structures are
added multiple times to the same hashtable (with different name lengths and
hash codes). However, there's only one dir_next pointer, which gets
completely messed up in case of hash collisions. In the worst case, this
causes an endless loop if ce == ce->dir_next (see t7062).
Use a separate hashtable and separate structures for the directory index
so that each directory entry has its own next pointer. Use reference
counting to track which directory entry contains files.
There are only slight changes to the name-hash.c API:
- new free_name_hash() used by read_cache.c::discard_index()
- remove_name_hash() takes an additional index_state parameter
- index_name_exists() for a directory (trailing '/') may return a cache
entry that has been removed (CE_UNHASHED). This is not a problem as the
return value is only used to check if the directory exists (dir.c) or to
normalize casing of directory names (read-cache.c).
Getting rid of cache_entry.dir_next reduces memory consumption, especially
with core.ignorecase=false (which doesn't use that member at all).
With core.ignorecase=true, building the directory index is slightly faster
as we add / check the parent directory first (instead of going through all
directory levels for each file in the index). E.g. with WebKit (~200k
files, ~7k dirs), time spent in lazy_init_name_hash is reduced from 176ms
to 130ms.
Signed-off-by: Karsten Blees <blees@dcon.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
9d22778 (read-cache.c: write prefix-compressed names in the index -
2012-04-04) defined these. Interestingly, they were not used by
read-cache.c, or anywhere in that patch. They were used in
builtin/update-index.c later for checking supported index
versions. Use them here too.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Specifically the fields uid, gid, ctime, ino and dev are set to zero
by JGit. Other implementations, eg. Git in cygwin are allegedly also
somewhat incompatible with Git For Windows and on *nix platforms
the resolution of the timestamps may differ.
Any stat checking by git will then need to check content, which may
be very slow, particularly on Windows. Since mtime and size
is typically enough we should allow the user to tell git to avoid
checking these fields if they are set to zero in the index.
This change introduces a core.checkstat config option where the
the user can select to check all fields (default), or just size
and the whole second part of mtime (minimal).
Signed-off-by: Robin Rosenberg <robin.rosenberg@dewire.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We do not want a link to 0{40} object stored anywhere in our objects.
* jk/maint-null-in-trees:
fsck: detect null sha1 in tree entries
do not write null sha1s to on-disk index
diff: do not use null sha1 as a sentinel value
Assignments to errno before calling system functions that used to
matter in the old code were left behind after the code structure
changed sufficiently to make them useless.
* nd/index-errno:
read_index_from: remove bogus errno assignments
These assignments comes from the very first commit e83c516 (Initial
revision of "git", the information manager from hell - 2005-04-07).
Back then we did not die() when errors happened so correct errno was
required.
Since 5d1a5c0 ([PATCH] Better error reporting for "git status" -
2005-10-01), read_index_from() learned to die rather than just return
-1 and these assignments became irrelevant. Remove them.
While at it, move die_errno() next to xmmap() call because it's the
mmap's error code that we care about. Otherwise if close(fd); fails,
it could overwrite mmap's errno.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We should never need to write the null sha1 into an index
entry (short of the 1 in 2^160 chance that somebody actually
has content that hashes to it). If we attempt to do so, it
is much more likely that it is a bug, since we use the null
sha1 as a sentinel value to mean "not valid".
The presence of null sha1s in the index (which can come
from, among other things, "update-index --cacheinfo", or by
reading a corrupted tree) can cause problems for later
readers, because they cannot distinguish the literal null
sha1 from its use a sentinel value. For example, "git
diff-files" on such an entry would make it appear as if it
is stat-dirty, and until recently, the diff code assumed
such an entry meant that we should be diffing a working tree
file rather than a blob.
Ideally, we would stop such entries from entering even our
in-core index. However, we do sometimes legitimately add
entries with null sha1s in order to represent these sentinel
situations; simply forbidding them in add_index_entry breaks
a lot of the existing code. However, we can at least make
sure that our in-core sentinel representation never makes it
to disk.
To be thorough, we will test an attempt to add both a blob
and a submodule entry. In the former case, we might run into
problems anyway because we will be missing the blob object.
But in the latter case, we do not enforce connectivity
across gitlink entries, making this our only point of
enforcement. The current implementation does not care which
type of entry we are seeing, but testing both cases helps
future-proof the test suite in case that changes.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Split lower bits of ce_flags field and creates a new ce_namelen
field in the in-core index structure.
* tg/ce-namelen-field:
Strip namelen out of ce_flags into a ce_namelen field
Even though the index can record pathnames longer than 1<<12 bytes,
in some places we were not comparing them in full, potentially
replacing index entries instead of adding.
* tg/maint-cache-name-compare:
cache_name_compare(): do not truncate while comparing paths
Strip the name length from the ce_flags field and move it
into its own ce_namelen field in struct cache_entry. This
will both give us a tiny bit of a performance enhancement
when working with long pathnames and is a refactoring for
more readability of the code.
It enhances readability, by making it more clear what
is a flag, and where the length is stored and make it clear
which functions use stages in comparisions and which only
use the length.
It also makes CE_NAMEMASK private, so that users don't
mistakenly write the name length in the flags.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We failed to use ce_namelen() equivalent and instead only compared
up to the CE_NAMEMASK bytes by mistake. Adding an overlong path
that shares the same common prefix as an existing entry in the index
did not add a new entry, but instead replaced the existing one, as
the result.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Replace strlen(ce->name) with ce_namelen() in a couple
of places which gives us some additional bits of
performance.
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Trivially shrinks the on-disk size of the index file to save both I/O and
checksum overhead.
The topic should give a solid base to build on further updates, with the
code refactoring in its earlier parts, and the backward compatibility
mechanism in its later parts.
* jc/index-v4:
index-v4: document the entry format
unpack-trees: preserve the index file version of original
update-index: upgrade/downgrade on-disk index version
read-cache.c: write prefix-compressed names in the index
read-cache.c: read prefix-compressed names in index on-disk version v4
read-cache.c: move code to copy incore to ondisk cache to a helper function
read-cache.c: move code to copy ondisk to incore cache to a helper function
read-cache.c: report the header version we do not understand
read-cache.c: make create_from_disk() report number of bytes it consumed
read-cache.c: allow unaligned mapping of the index file
cache.h: hide on-disk index details
varint: make it available outside the context of pack
Teach the code to write the index in the v4 on-disk format.
Record the format version of the on-disk index we read from in the
index_state, and use the format when writing the new index out.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Because the entries are sorted by path, adjacent entries in the index tend
to share the leading components of them, and it makes sense to only store
the differences in later entries. In the v4 on-disk format of the index,
each on-disk cache entry stores the number of bytes to be stripped from
the end of the previous name, and the bytes to append to the result, to
come up with its name.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The function is the one that is reading from the data stream. It only is
natural to make it responsible for reporting this number, not the caller.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Both the on-disk format v2 and v3 pads the "name" field to the multiple of
eight to make sure that various quantities in network long/short type can
be accessed with ntohl/ntohs without having to worry about alignment, but
this forces us to waste disk I/O bandwidth.
Introduce ntoh_s()/ntoh_l() macros that the callers can use as if they were
the regular ntohs()/ntohl() on a field that may not be aligned correctly.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The on-disk format of the index file is a detail whose implementation is
neatly encapsulated in read-cache.c; there is no need to expose it to the
general public that include the cache.h header file.
Also add a prominent mark to read-cache.c to delineate the parts that deal
with the index file I/O routines from the remainder of the file.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The read-cache implementation defines this static function,
but it is a generally useful concept in git. Let's give
the empty blob the same treatment as the empty tree,
providing both hex and binary forms of the sha1.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When running "git add --refresh <pathspec>", we incorrectly showed the
path that is unmerged even if it is outside the specified pathspec, even
though we did honor pathspec and refreshed only the paths that matched.
Note that this cange does not affect "git update-index --refresh"; for
hysterical raisins, it does not take a pathspec (it takes real paths) and
more importantly itss command line options are parsed and executed one by
one as they are encountered, so "git update-index --refresh foo" means
"first refresh the index, and then update the entry 'foo' by hashing the
contents in file 'foo'", not "refresh only entry 'foo'".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* rs/allocate-cache-entry-individually:
cache.h: put single NUL at end of struct cache_entry
read-cache.c: allocate index entries individually
Conflicts:
read-cache.c
If you have a deleted file and a porcelain refreshes the
cache, we print:
Unstaged changes after reset:
M file
This is technically correct, in that the file is modified,
but it's friendlier to the user if we further differentiate
the case of a deleted file (especially because this output
looks a lot like "diff --name-status", which would also make
the distinction).
Similarly, we can distinguish typechanges ("T") and
intent-to-add files ("A"), both of which appear as just "M"
in the current output.
The plumbing output for all cases remains "needs update" for
historical compatibility.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When refreshing the index, for modified (or unmerged) files we will print
"needs update" (or "needs merge") for plumbing, or line similar to the
output from "diff --name-status" for porcelain.
The variables holding which type of message to show are named after the
plumbing messages. However, as we begin to differentiate more cases at the
porcelain level (with the plumbing message staying the same), that naming
scheme will become awkward.
Instead, name the variables after which case we found (modified or
unmerged), not what we will output.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This will enable refresh_cache to differentiate more cases
of modification (such as typechange) when telling the user
what isn't fresh.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The code to estimate the in-memory size of the index based on its on-disk
representation is subtly wrong for certain architecture-dependent struct
layouts. Instead of fixing it, replace the code to keep the index entries
in a single large block of memory and allocate each entry separately
instead. This is both simpler and more flexible, as individual entries
can now be freed. Actually using that added flexibility is left for a
later patch.
Suggested-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
estimate_cache_size() tries to guess how much memory is needed for the
in-memory representation of an index file. It does that by using the
file size, the number of entries and the difference of the sizes of the
on-disk and in-memory structs -- without having to check the length of
the name of each entry, which varies for each entry, but their sums are
the same no matter the representation.
Except there can be a difference. First of all, the size is really
calculated by ce_size and ondisk_ce_size based on offsetof(..., name),
not sizeof, which can be different. And entries are padded with 1 to 8
NULs at the end (after the variable name) to make their total length a
multiple of eight.
So in order to allocate enough memory to hold the index, change the
delta calculation to be based on offsetof(..., name) and round up to
the next multiple of eight.
On a 32-bit Linux, this delta was used before:
sizeof(struct cache_entry) == 72
sizeof(struct ondisk_cache_entry) == 64
---
8
The actual difference for an entry with a filename length of one was,
however (find the definitions are in cache.h):
offsetof(struct cache_entry, name) == 72
offsetof(struct ondisk_cache_entry, name) == 62
ce_size == (72 + 1 + 8) & ~7 == 80
ondisk_ce_size == (62 + 1 + 8) & ~7 == 64
---
16
So eight bytes less had been allocated for such entries. The new
formula yields the correct delta:
(72 - 62 + 7) & ~7 == 16
Reported-by: John Hsing <tsyj2007@gmail.com>
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* ef/maint-win-verify-path:
verify_dotfile(): do not assume '/' is the path seperator
verify_path(): simplify check at the directory boundary
verify_path: consider dos drive prefix
real_path: do not assume '/' is the path seperator
A Windows path starting with a backslash is absolute
verify_dotfile() currently assumes that the path seperator is '/', but on
Windows it can also be '\\', so use is_dir_sep() instead.
Signed-off-by: Theo Niessink <theo@taletn.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We simply want to say "At a directory boundary, be careful with a name
that begins with a dot, forbid a name that ends with the boundary
character or has duplicated bounadry characters".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If someone manage to create a repo with a 'C:' entry in the
root-tree, files can be written outside of the working-dir. This
opens up a can-of-worms of exploits.
Fix it by explicitly checking for a dos drive prefix when verifying
a paht. While we're at it, make sure that paths beginning with '\' is
considered absolute as well.
Noticed-by: Theo Niessink <theo@taletn.com>
Signed-off-by: Erik Faye-Lund <kusmabite@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "format_check" parameter tucked after the existing parameters is too
ugly an afterthought to live in any reasonable API.
Combine it with the other boolean parameter "write_object" into a single
"flags" parameter.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Traditional "opportunistic index update" done by read-only "diff" and
"status" was about updating cached lstat(2) information in the index for
the next round. We missed another obvious optimization opportunity: when
there are racily clean entries that will cease to be racily clean by
updating $GIT_INDEX_FILE. Detect that case and write $GIT_INDEX_FILE out
to give it a newer timestamp.
Noticed by Lasse Makholm by stracing "git status" in a fresh checkout and
counting the number of open(2) calls.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we had to refresh the index internally before running diff or status,
we opportunistically updated the $GIT_INDEX_FILE so that later invocation
of git can use the lstat(2) we already did in this invocation.
Make them share a helper function to do so.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* nd/struct-pathspec: (22 commits)
t6004: add pathspec globbing test for log family
t7810: overlapping pathspecs and depth limit
grep: drop pathspec_matches() in favor of tree_entry_interesting()
grep: use writable strbuf from caller for grep_tree()
grep: use match_pathspec_depth() for cache/worktree grepping
grep: convert to use struct pathspec
Convert ce_path_match() to use match_pathspec_depth()
Convert ce_path_match() to use struct pathspec
struct rev_info: convert prune_data to struct pathspec
pathspec: add match_pathspec_depth()
tree_entry_interesting(): optimize wildcard matching when base is matched
tree_entry_interesting(): support wildcard matching
tree_entry_interesting(): fix depth limit with overlapping pathspecs
tree_entry_interesting(): support depth limit
tree_entry_interesting(): refactor into separate smaller functions
diff-tree: convert base+baselen to writable strbuf
glossary: define pathspec
Move tree_entry_interesting() to tree-walk.c and export it
tree_entry_interesting(): remove dependency on struct diff_options
Convert struct diff_options to use struct pathspec
...
Commits, trees and tags have structure. Don't let users feed git
with malformed ones. Sooner or later git will die() when
encountering them.
Note that this patch does not check semantics. A tree that points
to non-existent objects is perfectly OK (and should be so, users
may choose to add commit first, then its associated tree for example).
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jj/icase-directory:
Support case folding in git fast-import when core.ignorecase=true
Support case folding for git add when core.ignorecase=true
Add case insensitivity support when using git ls-files
Add case insensitivity support for directories when using git status
Case insensitivity support for .gitignore via core.ignorecase
Add string comparison functions that respect the ignore_case variable.
Makefile & configure: add a NO_FNMATCH_CASEFOLD flag
Makefile & configure: add a NO_FNMATCH flag
Conflicts:
Makefile
config.mak.in
configure.ac
fast-import.c
When MyDir/ABC/filea.txt is added to Git, the disk directory MyDir/ABC/
is renamed to mydir/aBc/, and then mydir/aBc/fileb.txt is added, the
index will contain MyDir/ABC/filea.txt and mydir/aBc/fileb.txt. Although
the earlier portions of this patch series account for those differences
in case, this patch makes the pathing consistent by folding the case of
newly added files against the first file added with that path.
In read-cache.c's add_to_index(), the index_name_exists() support used
for git status's case insensitive directory lookups is used to find the
proper directory case according to what the user already checked in.
That is, MyDir/ABC/'s case is used to alter the stored path for
fileb.txt to MyDir/ABC/fileb.txt (instead of mydir/aBc/fileb.txt).
This is especially important when cloning a repository to a case
sensitive file system. MyDir/ABC/ and mydir/aBc/ exist in the same
directory on a Windows machine, but on Linux, the files exist in two
separate directories. The update to add_to_index(), in effect, treats a
Windows file system as case sensitive by making path case consistent.
Signed-off-by: Joshua Jensen <jjensen@workspacewhiz.com>
Signed-off-by: Johannes Sixt <j6t@kdbg.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The new dircache extension CACHE_EXT_RESOLVE_UNDO, whose value is
0x52455543, is actually the ASCII sequence 'REUC', not the ASCII
sequence 'REUN'.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The rule has always been that a cache entry that is ce_uptodate(ce)
means that we already have checked the work tree entity and we know
there is no change in the work tree compared to the index, and nobody
should have to double check. Note that false ce_uptodate(ce) does not
mean it is known to be dirty---it only means we don't know if it is
clean.
There are a few codepaths (refresh-index and preload-index are among
them) that mark a cache entry as up-to-date based solely on the return
value from ie_match_stat(); this function uses lstat() to see if the
work tree entity has been touched, and for a submodule entry, if its
HEAD points at the same commit as the commit recorded in the index of
the superproject (a submodule that is not even cloned is considered
clean).
A submodule is no longer considered unmodified merely because its HEAD
matches the index of the superproject these days, in order to prevent
people from forgetting to commit in the submodule and updating the
superproject index with the new submodule commit, before commiting the
state in the superproject. However, the patch to do so didn't update
the codepath that marks cache entries up-to-date based on the updated
definition and instead worked it around by saying "we don't trust the
return value of ce_uptodate() for submodules."
This makes ce_uptodate() trustworthy again by not marking submodule
entries up-to-date.
The next step _could_ be to introduce a few "in-core" flag bits to
cache_entry structure to record "this entry is _known_ to be dirty",
call is_submodule_modified() from ie_match_stat(), and use these new
bits to avoid running this rather expensive check more than once, but
that can be a separate patch.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Exal Sibeaz pointed out that some git files are way too big, and that
add_files_to_cache() brings in all the diff machinery to any git binary
that needs the basic git SHA1 object operations from read-cache.c. Which
is pretty much all of them.
It's doubly silly, since add_files_to_cache() is only used by builtin
programs (add, checkout and commit), so it's fairly easily fixed by just
moving the thing to builtin-add.c, and avoiding the dependency entirely.
I initially argued to Exal that it would probably be best to try to depend
on smart compilers and linkers, but after spending some time trying to
make -ffunction-sections work and giving up, I think Exal was right, and
the fix is to just do some trivial cleanups like this.
This trivial cleanup results in pretty stunning file size differences.
The diff machinery really is mostly used by just the builtin programs, and
you have things like these trivial before-and-after numbers:
-rwxr-xr-x 1 torvalds torvalds 1727420 2010-01-21 10:53 git-hash-object
-rwxrwxr-x 1 torvalds torvalds 940265 2010-01-21 11:16 git-hash-object
Now, I'm not saying that 940kB is good either, but that's mostly all the
debug information - you can see the real code with 'size':
text data bss dec hex filename
418675 3920 127408 550003 86473 git-hash-object (before)
230650 2288 111728 344666 5425a git-hash-object (after)
ie we have a nice 24% size reduction from this trivial cleanup.
It's not just that one file either. I get:
[torvalds@nehalem git]$ du -s /home/torvalds/libexec/git-core
45640 /home/torvalds/libexec/git-core (before)
33508 /home/torvalds/libexec/git-core (after)
so we're talking 12MB of diskspace here.
(Of course, stripping all the binaries brings the 33MB down to 9MB, so the
whole debug information thing is still the bulk of it all, but that's a
separate issue entirely)
Now, I'm sure there are other things we should do, and changing our
compiler flags from -O2 to -Os would bring the text size down by an
additional almost 20%, but this thing Exal pointed out seems to be some
good low-hanging fruit.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/cache-unmerge:
rerere forget path: forget recorded resolution
rerere: refactor rerere logic to make it independent from I/O
rerere: remove silly 1024-byte line limit
resolve-undo: teach "update-index --unresolve" to use resolve-undo info
resolve-undo: "checkout -m path" uses resolve-undo information
resolve-undo: allow plumbing to clear the information
resolve-undo: basic tests
resolve-undo: record resolved conflicts in a new index extension section
builtin-merge.c: use standard active_cache macros
Conflicts:
builtin-ls-files.c
builtin-merge.c
builtin-rerere.c
* jc/symbol-static:
date.c: mark file-local function static
Replace parse_blob() with an explanatory comment
symlinks.c: remove unused functions
object.c: remove unused functions
strbuf.c: remove unused function
sha1_file.c: remove unused function
mailmap.c: remove unused function
utf8.c: mark file-local function static
submodule.c: mark file-local function static
quote.c: mark file-local function static
remote-curl.c: mark file-local function static
read-cache.c: mark file-local functions static
parse-options.c: mark file-local function static
entry.c: mark file-local function static
http.c: mark file-local functions static
pretty.c: mark file-local function static
builtin-rev-list.c: mark file-local function static
bisect.c: mark file-local function static
* cc/reset-more:
t7111: check that reset options work as described in the tables
Documentation: reset: add some missing tables
Fix bit assignment for CE_CONFLICTED
"reset --merge": fix unmerged case
reset: use "unpack_trees()" directly instead of "git read-tree"
reset: add a few tests for "git reset --merge"
Documentation: reset: add some tables to describe the different options
reset: improve mixed reset error message when in a bare repo
* nd/sparse: (25 commits)
t7002: test for not using external grep on skip-worktree paths
t7002: set test prerequisite "external-grep" if supported
grep: do not do external grep on skip-worktree entries
commit: correctly respect skip-worktree bit
ie_match_stat(): do not ignore skip-worktree bit with CE_MATCH_IGNORE_VALID
tests: rename duplicate t1009
sparse checkout: inhibit empty worktree
Add tests for sparse checkout
read-tree: add --no-sparse-checkout to disable sparse checkout support
unpack-trees(): ignore worktree check outside checkout area
unpack_trees(): apply $GIT_DIR/info/sparse-checkout to the final index
unpack-trees(): "enable" sparse checkout and load $GIT_DIR/info/sparse-checkout
unpack-trees.c: generalize verify_* functions
unpack-trees(): add CE_WT_REMOVE to remove on worktree alone
Introduce "sparse checkout"
dir.c: export excluded_1() and add_excludes_from_file_1()
excluded_1(): support exclude files in index
unpack-trees(): carry skip-worktree bit over in merged_entry()
Read .gitignore from index if it is skip-worktree
Avoid writing to buffer in add_excludes_from_file_1()
...
Conflicts:
.gitignore
Documentation/config.txt
Documentation/git-update-index.txt
Makefile
entry.c
t/t7002-grep.sh
Junio C Hamano
brian m. carlson
Martin Ågren
Ben Peart
Jeff King
Thomas Gummerer
Kevin Willford
Patryk Obara
René Scharfe
Christian Couder
Ævar Arnfjörð Bjarmason
Brandon Williams
David Turner
Nguyễn Thái Ngọc Duy
Peter Krefting
Jeff Hostetler
Stefan Beller
Lars Schneider
Edward Thomson
Tobias Klauser
Michael Haggerty
Johannes Schindelin
Jaime Soriano Pastor
Tanay Abhra
Jeremiah Mahler
Yiannis Marangos
Dmitry S. Dolzhenko
Brad King
Karsten Blees
Eric Sunshine
Ondřej Bílka
René Scharfe
Felipe Contreras
Lukas Fleischer
Robin Rosenberg
Theo Niessink
Erik Faye-Lund
Jonathan Nieder
Joshua Jensen
Shawn O. Pearce
Linus Torvalds