Transactions to update multiple references that involves a deletion
was quite broken in an error codepath and did not abort everything
correctly.
* mh/ref-locking-fix:
files_transaction_prepare(): fix handling of ref lock failure
t1404: add a bunch of tests of D/F conflicts
Since dc39e09942 (files_ref_store: use a transaction to update packed
refs, 2017-09-08), failure to lock a reference has been handled
incorrectly by `files_transaction_prepare()`. If
`lock_ref_for_update()` fails in the lock-acquisition loop of that
function, it sets `ret` then breaks out of that loop. Prior to
dc39e09942, that was OK, because the only thing following the loop was
the cleanup code. But dc39e09942 added another blurb of code between
the loop and the cleanup. That blurb sometimes resets `ret` to zero,
making the cleanup code think that the locking was successful.
Specifically, whenever
* One or more reference deletions have been processed successfully in
the lock-acquisition loop. (Processing the first such reference
causes a packed-ref transaction to be initialized.)
* Then `lock_ref_for_update()` fails for a subsequent reference. Such
a failure can happen for a number of reasons, such as the old SHA-1
not being correct, lock contention, etc. This causes a `break` out
of the lock-acquisition loop.
* The `packed-refs` lock is acquired successfully and
`ref_transaction_prepare()` succeeds for the packed-ref transaction.
This has the effect of resetting `ret` back to 0, and making the
cleanup code think that lock acquisition was successful.
In that case, any reference updates that were processed prior to
breaking out of the loop would be carried out (loose and packed), but
the reference that couldn't be locked and any subsequent references
would silently be ignored.
This can easily cause data loss if, for example, the user was trying
to push a new name for an existing branch while deleting the old name.
After the push, the branch could be left unreachable, and could even
subsequently be garbage-collected.
This problem was noticed in the context of deleting one reference and
creating another in a single transaction, when the two references D/F
conflict with each other, like
git update-ref --stdin <<EOF
delete refs/foo
create refs/foo/bar HEAD
EOF
This triggers the above bug because the deletion is processed
successfully for `refs/foo`, then the D/F conflict causes
`lock_ref_for_update()` to fail when `refs/foo/bar` is processed. In
this case the transaction *should* fail, but instead it causes
`refs/foo` to be deleted without creating `refs/foo`. This could
easily result in data loss.
The fix is simple: instead of just breaking out of the loop, jump
directly to the cleanup code. This fixes some tests in t1404 that were
added in the previous commit.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code clean-up.
* rs/resolve-ref-optional-result:
refs: pass NULL to resolve_refdup() if hash is not needed
refs: pass NULL to refs_resolve_refdup() if hash is not needed
Operations that do not touch (majority of) packed refs have been
optimized by making accesses to packed-refs file lazy; we no longer
pre-parse everything, and an access to a single ref in the
packed-refs does not touch majority of irrelevant refs, either.
* mh/mmap-packed-refs: (21 commits)
packed-backend.c: rename a bunch of things and update comments
mmapped_ref_iterator: inline into `packed_ref_iterator`
ref_cache: remove support for storing peeled values
packed_ref_store: get rid of the `ref_cache` entirely
ref_store: implement `refs_peel_ref()` generically
packed_read_raw_ref(): read the reference from the mmapped buffer
packed_ref_iterator_begin(): iterate using `mmapped_ref_iterator`
read_packed_refs(): ensure that references are ordered when read
packed_ref_cache: keep the `packed-refs` file mmapped if possible
packed-backend.c: reorder some definitions
mmapped_ref_iterator_advance(): no peeled value for broken refs
mmapped_ref_iterator: add iterator over a packed-refs file
packed_ref_cache: remember the file-wide peeling state
read_packed_refs(): read references with minimal copying
read_packed_refs(): make parsing of the header line more robust
read_packed_refs(): only check for a header at the top of the file
read_packed_refs(): use mmap to read the `packed-refs` file
die_unterminated_line(), die_invalid_line(): new functions
packed_ref_cache: add a backlink to the associated `packed_ref_store`
prefix_ref_iterator: break when we leave the prefix
...
Code clean-up to prevent future mistakes by copying and pasting
code that checks the result of read_in_full() function.
* jk/read-in-full:
worktree: check the result of read_in_full()
worktree: use xsize_t to access file size
distinguish error versus short read from read_in_full()
avoid looking at errno for short read_in_full() returns
prefer "!=" when checking read_in_full() result
notes-merge: drop dead zero-write code
files-backend: prefer "0" for write_in_full() error check
"git branch" learned "-c/-C" to create a new branch by copying an
existing one.
* sd/branch-copy:
branch: fix "copy" to never touch HEAD
branch: add a --copy (-c) option to go with --move (-m)
branch: add test for -m renaming multiple config sections
config: create a function to format section headers
Code clean-up.
* rs/resolve-ref-optional-result:
refs: pass NULL to resolve_ref_unsafe() if hash is not needed
refs: pass NULL to refs_resolve_ref_unsafe() if hash is not needed
refs: make sha1 output parameter of refs_resolve_ref_unsafe() optional
Commit 06f46f237a (avoid "write_in_full(fd, buf, len) !=
len" pattern, 2017-09-13) converted this callsite from:
write_in_full(...) != 1
to
write_in_full(...) < 0
But during the conflict resolution in c50424a6f0 (Merge
branch 'jk/write-in-full-fix', 2017-09-25), this morphed
into
write_in_full(...) < 1
This behaves as we want, but we prefer to avoid modeling the
"less than length" error-check which can be subtly buggy, as
shown in efacf609c8 (config: avoid "write_in_full(fd, buf,
len) < len" pattern, 2017-09-13).
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We're about to stop storing packed refs in a `ref_cache`. That means
that the only way we have left to optimize `peel_ref()` is by checking
whether the reference being peeled is the one currently being iterated
over (in `current_ref_iter`), and if so, using `ref_iterator_peel()`.
But this can be done generically; it doesn't have to be implemented
per-backend.
So implement `refs_peel_ref()` in `refs.c` and remove the `peel_ref()`
method from the refs API.
This removes the last callers of a couple of functions, so delete
them. More cleanup to come...
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
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
This allows us to get rid of two write-only variables, one of them
being a SHA1 buffer.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Implement transactional update to the packed-ref representation of
references.
* mh/packed-ref-transactions:
files_transaction_finish(): delete reflogs before references
packed-backend: rip out some now-unused code
files_ref_store: use a transaction to update packed refs
t1404: demonstrate two problems with reference transactions
files_initial_transaction_commit(): use a transaction for packed refs
prune_refs(): also free the linked list
files_pack_refs(): use a reference transaction to write packed refs
packed_delete_refs(): implement method
packed_ref_store: implement reference transactions
struct ref_transaction: add a place for backends to store data
packed-backend: don't adjust the reference count on lock/unlock
The long-standing rule that an in-core lockfile instance, once it
is used, must not be freed, has been lifted and the lockfile and
tempfile APIs have been updated to reduce the chance of programming
errors.
* jk/incore-lockfile-removal:
stop leaking lock structs in some simple cases
ref_lock: stop leaking lock_files
lockfile: update lifetime requirements in documentation
tempfile: auto-allocate tempfiles on heap
tempfile: remove deactivated list entries
tempfile: use list.h for linked list
tempfile: release deactivated strbufs instead of resetting
tempfile: robustify cleanup handler
tempfile: factor out deactivation
tempfile: factor out activation
tempfile: replace die("BUG") with BUG()
tempfile: handle NULL tempfile pointers gracefully
tempfile: prefer is_tempfile_active to bare access
lockfile: do not rollback lock on failed close
tempfile: do not delete tempfile on failed close
always check return value of close_tempfile
verify_signed_buffer: prefer close_tempfile() to close()
setup_temporary_shallow: move tempfile struct into function
setup_temporary_shallow: avoid using inactive tempfile
write_index_as_tree: cleanup tempfile on error
"git gc" and friends when multiple worktrees are used off of a
single repository did not consider the index and per-worktree refs
of other worktrees as the root for reachability traversal, making
objects that are in use only in other worktrees to be subject to
garbage collection.
* nd/prune-in-worktree:
refs.c: reindent get_submodule_ref_store()
refs.c: remove fallback-to-main-store code get_submodule_ref_store()
rev-list: expose and document --single-worktree
revision.c: --reflog add HEAD reflog from all worktrees
files-backend: make reflog iterator go through per-worktree reflog
revision.c: --all adds HEAD from all worktrees
refs: remove dead for_each_*_submodule()
refs.c: move for_each_remote_ref_submodule() to submodule.c
revision.c: use refs_for_each*() instead of for_each_*_submodule()
refs: add refs_head_ref()
refs: move submodule slash stripping code to get_submodule_ref_store
refs.c: refactor get_submodule_ref_store(), share common free block
revision.c: --indexed-objects add objects from all worktrees
revision.c: refactor add_index_objects_to_pending()
refs.c: use is_dir_sep() in resolve_gitlink_ref()
revision.h: new flag in struct rev_info wrt. worktree-related refs
A leakfix.
* ma/split-symref-update-fix:
refs/files-backend: add `refname`, not "HEAD", to list
refs/files-backend: correct return value in lock_ref_for_update
refs/files-backend: fix memory leak in lock_ref_for_update
refs/files-backend: add longer-scoped copy of string to list
References are iterated over in order by refname, but reflogs are not.
Some consumers of reference iteration care about the difference. Teach
each `ref_iterator` to keep track of whether its output is ordered.
`overlay_ref_iterator` is one of the picky consumers. Add a sanity
check in `overlay_ref_iterator_begin()` to verify that its inputs are
ordered.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The prior commit converted many sites to check the return
value of write_in_full() for negativity, rather than a
mismatch with the input length. This patch covers similar
cases, but where the return value is stored in an
intermediate variable. These should get the same treatment,
but they need to be reviewed more carefully since it would
be a bug if the return value is stored in an unsigned type
(which indeed, it is in one of the cases).
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
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>
An earlier patch rewrote `split_symref_update()` to add a copy of a
string to a string list instead of adding the original string. That was
so that the original string could be freed in a later patch, but it is
also conceptually cleaner, since now all calls to `string_list_insert()`
and `string_list_append()` add `update->refname`. --- Except a literal
"HEAD" is added in `split_head_update()`.
Restructure `split_head_update()` in the same way as the earlier patch
did for `split_symref_update()`. This does not correct any practical
problem, but makes things conceptually cleaner. The downside is a call
to `string_list_has_string()`, which should be relatively cheap.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In one code path we return a literal -1 and not a symbolic constant. The
value -1 would be interpreted as TRANSACTION_NAME_CONFLICT, which is
wrong. Use TRANSACTION_GENERIC_ERROR instead (that is the only other
return value we have to choose from).
Noticed-by: Michael Haggerty <mhagger@alum.mit.edu>
Reviewed-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
After the previous patch, none of the functions we call hold on to
`referent.buf`, so we can safely release the string buffer before
returning.
Reviewed-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
split_symref_update() receives a string-pointer `referent` and adds it
to the list of `affected_refnames`. The list simply holds on to the
pointers it is given, it does not copy the strings and it does not ever
free them. The `referent` string in split_symref_update() belongs to a
string buffer in the caller. After we return, the string will be leaked.
In the next patch, we want to properly release the string buffer in the
caller, but we can't safely do so until we've made sure that
`affected_refnames` will not be holding on to a pointer to the string.
We could configure the list to handle its own resources, but it would
mean some alloc/free-churning. The list is already handling other
strings (through other code paths) which we do not need to worry about,
and we'd be memory-churning those strings too, completely unnecessary.
Observe that split_symref_update() creates a `new_update`-object through
ref_transaction_add_update(), after which `new_update->refname` is a
copy of `referent`. The difference is, this copy will be freed, and it
will be freed *after* `affected_refnames` has been cleared.
Rearrange the handling of `referent`, so that we don't add it directly
to `affected_refnames`. Instead, first just check whether `referent`
exists in the string list, and later add `new_update->refname`.
Helped-by: Michael Haggerty <mhagger@alum.mit.edu>
Reviewed-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If the deletion steps unexpectedly fail, it is less bad to leave a
reference without its reflog than it is to leave a reflog without its
reference, since the latter is an invalid repository state.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When processing a `files_ref_store` transaction, it is sometimes
necessary to delete some references from the "packed-refs" file. Do
that using a reference transaction conducted against the
`packed_ref_store`.
This change further decouples `files_ref_store` from
`packed_ref_store`. It also fixes multiple problems, including the two
revealed by test cases added in the previous commit.
First, the old code didn't obtain the `packed-refs` lock until
`files_transaction_finish()`. This means that a failure to acquire the
`packed-refs` lock (e.g., due to contention with another process)
wasn't detected until it was too late (problems like this are supposed
to be detected in the "prepare" phase). The new code acquires the
`packed-refs` lock in `files_transaction_prepare()`, the same stage of
the processing when the loose reference locks are being acquired,
removing another reason why the "prepare" phase might succeed and the
"finish" phase might nevertheless fail.
Second, the old code deleted the loose version of a reference before
deleting any packed version of the same reference. This left a moment
when another process might think that the packed version of the
reference is current, which is incorrect. (Even worse, the packed
version of the reference can be arbitrarily old, and might even point
at an object that has since been garbage-collected.)
Third, if a reference deletion fails to acquire the `packed-refs` lock
altogether, then the old code might leave the repository in the
incorrect state (possibly corrupt) described in the previous
paragraph.
Now we activate the new "packed-refs" file (sans any references that
are being deleted) *before* deleting the corresponding loose
references. But we hold the "packed-refs" lock until after the loose
references have been finalized, thus preventing a simultaneous
"pack-refs" process from packing the loose version of the reference in
the time gap, which would otherwise defeat our attempt to delete it.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Use a `packed_ref_store` transaction in the implementation of
`files_initial_transaction_commit()` rather than using internal
features of the packed ref store. This further decouples
`files_ref_store` from `packed_ref_store`.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
At least since v1.7, the elements of the `refs_to_prune` linked list
have been leaked. Fix the leak by teaching `prune_refs()` to free the
list elements as it processes them.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Now that the packed reference store supports transactions, we can use
a transaction to write the packed versions of references that we want
to pack. This decreases the coupling between `files_ref_store` and
`packed_ref_store`.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Implement `packed_delete_refs()` using a reference transaction. This
means that `files_delete_refs()` can use `refs_delete_refs()` instead
of `repack_without_refs()` to delete any packed references, decreasing
the coupling between the classes.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since the tempfile code recently relaxed the rule that
tempfile structs (and thus locks) need to hang around
forever, we no longer have to leak our lock_file structs.
In fact, we don't even need to heap-allocate them anymore,
since their lifetime can just match that of the surrounding
ref_lock (and if we forget to delete a lock, the effect is
the same as before: it will eventually go away at program
exit).
Note that there is a check in unlock_ref() to only rollback
a lock file if it has been allocated. We don't need that
check anymore; we zero the ref_lock (and thus the
lock_file), so at worst we pass a NULL pointer to
delete_tempfile(), which considers that a noop.
Signed-off-by: Jeff King <peff@peff.net>
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>
The code to acquire a lock on a reference (e.g. while accepting a
push from a client) used to immediately fail when the reference is
already locked---now it waits for a very short while and retries,
which can make it succeed if the lock holder was holding it during
a read-only operation.
* mh/ref-lock-entry:
refs: retry acquiring reference locks for 100ms
refs/bisect is unfortunately per-worktree, so we need to look in
per-worktree logs/refs/bisect in addition to per-repo logs/refs. The
current iterator only goes through per-repo logs/refs.
Use merge iterator to walk two ref stores at the same time and pick
per-worktree refs from the right iterator.
PS. Note the unsorted order of for_each_reflog in the test. This is
supposed to be OK, for now. If we enforce order on for_each_reflog()
then some more work will be required.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The philosophy of reference locking has been, "if another process is
changing a reference, then whatever I'm trying to do to it will
probably fail anyway because my old-SHA-1 value is probably no longer
current". But this argument falls down if the other process has locked
the reference to do something that doesn't actually change the value
of the reference, such as `pack-refs` or `reflog expire`. There
actually *is* a decent chance that a planned reference update will
still be able to go through after the other process has released the
lock.
So when trying to lock an individual reference (e.g., when creating
"refs/heads/master.lock"), if it is already locked, then retry the
lock acquisition for approximately 100 ms before giving up. This
should eliminate some unnecessary lock conflicts without wasting a lot
of time.
Add a configuration setting, `core.filesRefLockTimeout`, to allow this
setting to be tweaked.
Note: the function `get_files_ref_lock_timeout_ms()` cannot be private
to the files backend because it is also used by `write_pseudoref()`
and `delete_pseudoref()`, which are defined in `refs.c` so that they
can be used by other reference backends.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "ref-store" code reorganization continues.
* mh/packed-ref-store: (32 commits)
files-backend: cheapen refname_available check when locking refs
packed_ref_store: handle a packed-refs file that is a symlink
read_packed_refs(): die if `packed-refs` contains bogus data
t3210: add some tests of bogus packed-refs file contents
repack_without_refs(): don't lock or unlock the packed refs
commit_packed_refs(): remove call to `packed_refs_unlock()`
clear_packed_ref_cache(): don't protest if the lock is held
packed_refs_unlock(), packed_refs_is_locked(): new functions
packed_refs_lock(): report errors via a `struct strbuf *err`
packed_refs_lock(): function renamed from lock_packed_refs()
commit_packed_refs(): use a staging file separate from the lockfile
commit_packed_refs(): report errors rather than dying
packed_ref_store: make class into a subclass of `ref_store`
packed-backend: new module for handling packed references
packed_read_raw_ref(): new function, replacing `resolve_packed_ref()`
packed_ref_store: support iteration
packed_peel_ref(): new function, extracted from `files_peel_ref()`
repack_without_refs(): take a `packed_ref_store *` parameter
get_packed_ref(): take a `packed_ref_store *` parameter
rollback_packed_refs(): take a `packed_ref_store *` parameter
...
When locking references in preparation for updating them, we need to
check that none of the newly added references D/F conflict with
existing references (e.g., we don't allow `refs/foo` to be added if
`refs/foo/bar` already exists, or vice versa).
Prior to 524a9fdb51 (refs_verify_refname_available(): use function in
more places, 2017-04-16), conflicts with existing loose references
were checked by looking directly in the filesystem, and then conflicts
with existing packed references were checked by running
`verify_refname_available_dir()` against the packed-refs cache.
But that commit changed the final check to call
`refs_verify_refname_available()` against the *whole* files ref-store,
including both loose and packed references, with the following
comment:
> This means that those callsites now check for conflicts with all
> references rather than just packed refs, but the performance cost
> shouldn't be significant (and will be regained later).
That comment turned out to be too sanguine. User s@kazlauskas.me
reported that fetches involving a very large number of references in
neighboring directories were slowed down by that change.
The problem is that when fetching, each reference is updated
individually, within its own reference transaction. This is done
because some reference updates might succeed even though others fail.
But every time a reference update transaction is finished,
`clear_loose_ref_cache()` is called. So when it is time to update the
next reference, part of the loose ref cache has to be repopulated for
the `refs_verify_refname_available()` call. If the references are all
in neighboring directories, then the cost of repopulating the
reference cache increases with the number of references, resulting in
O(N²) effort.
The comment above also claims that the performance cost "will be
regained later". The idea was that once the packed-refs were finished
being split out into a separate ref-store, we could limit the
`refs_verify_refname_available()` call to the packed references again.
That is what we do now.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change `repack_without_refs()` to expect the packed-refs lock to be
held already, and not to release the lock before returning. Change the
callers to deal with lock management.
This change makes it possible for callers to hold the packed-refs lock
for a longer span of time, a possibility that will eventually make it
possible to fix some longstanding races.
The only semantic change here is that `repack_without_refs()` used to
forget to release the lock in the `if (!removed)` exit path. That
omission is now fixed.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Bugfix for a topic that is (only) in 'master'.
* mh/packed-ref-store-prep:
for_each_bisect_ref(): don't trim refnames
lock_packed_refs(): fix cache validity check
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
Instead, change the callers of `commit_packed_refs()` to call
`packed_refs_unlock()`.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
That way the callers don't have to come up with error messages
themselves.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Rename `lock_packed_refs()` to `packed_refs_lock()` for consistency
with how other methods are named. Also, it's about to get some
companions.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Report errors via a `struct strbuf *err` rather than by calling
`die()`. To enable this goal, change `write_packed_entry()` to report
errors via a return value and `errno` rather than dying.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add the infrastructure to make `packed_ref_store` implement
`ref_store`, at least formally (few of the methods are actually
implemented yet). Change the functions in its interface to take
`ref_store *` arguments. Change `files_ref_store` to store a pointer
to `ref_store *` and to call functions via the virtual `ref_store`
interface where possible. This also means that a few
`packed_ref_store` functions can become static.
This is a work in progress. Some more `ref_store` methods will soon be
implemented (e.g., those having to do with reference transactions).
But some of them will never be implemented (e.g., those having to do
with symrefs or reflogs).
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Now that the interface between `files_ref_store` and
`packed_ref_store` is relatively narrow, move the latter into a new
module, "refs/packed-backend.h" and "refs/packed-backend.c". It still
doesn't quite implement the `ref_store` interface, but it will soon.
This commit moves code around and adjusts its visibility, but doesn't
change anything.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a new function, `packed_read_raw_ref()`, which is nearly a
`read_raw_ref_fn`. Use it in place of `resolve_packed_ref()`.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add the infrastructure to iterate over a `packed_ref_store`. It's a
lot of boilerplate, but it's all part of a campaign to make
`packed_ref_store` implement `ref_store`. In the future, this iterator
will work much differently.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This will later become a method of `packed_ref_store`.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
