The code to drive GIT_EXTERNAL_DIFF command relied on the string
returned from getenv() to be non-volatile, which is not true, that
has been corrected.
* kg/external-diff-save-env:
diff: ensure correct lifetime of external_diff_cmd
This backports several patch series from Git for Windows' `master` to
make things compile with GCC v8.x, as that will soon be the default for
Git for Windows (and is already the current setup on this developer's
machine).
The kwset functionality makes use of the obstack code, which expects to
be handed a function that can allocate large chunks of data. It expects
that function to accept a `size` parameter of type `long`.
This upsets GCC 8 on Windows, because `long` does not have the same
bit size as `size_t` there.
Now, the proper thing to do would be to switch to `size_t`. But this
would make us deviate from the "upstream" code even further, making it
hard to synchronize with newer versions, and also it would be quite
involved because that `long` type is so invasive in that code.
Let's punt, and instead provide a super small wrapper around
`xmalloc()`.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
The return type of the `GetProcAddress()` function is `FARPROC` which
evaluates to `long long int (*)()`, i.e. it cannot be cast to the
correct function signature by GCC 8.
To work around that, we first cast to `void *` and go on with our merry
lives.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
GCC 8 introduced the new -Wcast-function-type warning, which is
implied by -Wextra (which, in turn is enabled in our DEVELOPER flags).
When building Git with GCC 8 and this warning enabled on a non-glibc
platform [1], one is greeted with a screenful of compiler
warnings/errors:
compat/obstack.c: In function '_obstack_begin':
compat/obstack.c:162:17: error: cast between incompatible function types from 'void * (*)(long int)' to 'struct _obstack_chunk * (*)(void *, long int)' [-Werror=cast-function-type]
h->chunkfun = (struct _obstack_chunk * (*)(void *, long)) chunkfun;
^
compat/obstack.c:163:16: error: cast between incompatible function types from 'void (*)(void *)' to 'void (*)(void *, struct _obstack_chunk *)' [-Werror=cast-function-type]
h->freefun = (void (*) (void *, struct _obstack_chunk *)) freefun;
^
compat/obstack.c:116:8: error: cast between incompatible function types from 'struct _obstack_chunk * (*)(void *, long int)' to 'struct _obstack_chunk * (*)(long int)' [-Werror=cast-function-type]
: (*(struct _obstack_chunk *(*) (long)) (h)->chunkfun) ((size)))
^
compat/obstack.c:168:22: note: in expansion of macro 'CALL_CHUNKFUN'
chunk = h->chunk = CALL_CHUNKFUN (h, h -> chunk_size);
^~~~~~~~~~~~~
<snip>
'struct obstack' stores pointers to two functions to allocate and free
"chunks", and depending on how obstack is used, these functions take
either one parameter (like standard malloc() and free() do; this is
how we use it in 'kwset.c') or two parameters. Presumably to reduce
memory footprint, a single field is used to store the function pointer
for both signatures, and then it's casted to the appropriate signature
when the function pointer is accessed. These casts between function
pointers with different number of parameters are what trigger those
compiler errors.
Modify 'struct obstack' to use unions to store function pointers with
different signatures, and then use the union member with the
appropriate signature when accessing these function pointers. This
eliminates the need for those casts, and thus avoids this compiler
error.
[1] Compiling 'compat/obstack.c' on a platform with glibc is sort of
a noop, see the comment before '# define ELIDE_CODE', so this is
not an issue on common Linux distros.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
While working on parallelizing the tests in Azure Pipelines, an issue
was discovered with the `is_msys2_sh()` function: it expects the path
components to be separated by exactly one dir separator. That does not
need to be the case, though, e.g. when the components in the `PATH`
variable have trailing slashes.
Let's make the code much more robust in this respect.
This fixes t5580.5.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
It just takes forever, or at least over an hour, to run them in a
non-parallel way. That's why we switched to parallel testing in v2.21.0,
and hereby we bring that goodness back to vfs-2.20.1.
This is a (partial) backport of the improvements of the Azure Pipeline
made in Git for Windows v2.21.0.
The major benefit is that it makes the task to initialize the Git for
Windows SDK *a lot* faster.
It would also have worked around a problem where an MSYS2 update breaks
t6500.13, and it will work around similar problems in the future, as the
Azure Pipeline that updates git-sdk-64-minimal will not publish a new
artifact unless Git's test suite (as of upstream's current `master`)
passes.
In Git for Windows, we use the MSYS2 Bash which inherits a non-standard
PID model from Cygwin's POSIX emulation layer: every MSYS2 process has a
regular Windows PID, and in addition it has an MSYS2 PID (which
corresponds to a shadow process that emulates Unix-style signal
handling).
With the upgrade to the MSYS2 runtime v3.x, this shadow process cannot
be accessed via `OpenProcess()` any longer, and therefore t6500 thought
incorrectly that the process referenced in `gc.pid` (which is not
actually a real `gc` process in this context, but the current shell) no
longer exists.
Let's fix this by making sure that the Windows PID is written into
`gc.pid` in this test script soo that `git.exe` is able to understand
that that process does indeed still exist.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Recently the Git for Windows project started the upgrade process to
a MSYS2 runtime version based on Cygwin v3.x.
This has the very notable consequence that `$(uname -r)` no longer
reports a version starting with "2", but a version with "3".
That breaks our build, as df5218b4c3 (config.mak.uname: support MSys2,
2016-01-13) simply did not expect the version reported by `uname -r` to
depend on the underlying Cygwin version: it expected the reported
version to match the "2" in "MSYS2".
So let's invert that test case to test for *anything else* than a
version starting with "1" (for MSys). That should safeguard us for the
future, even if Cygwin ends up releasing versionsl like 314.272.65536.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Most notably, it seems that macOS' APFS does not allow that.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Due to reasons (some XCode versions seem to include gettext, some
don't?), Homebrew does not expose the libraries and headers in
/usr/local/ by default anymore.
Let's help find them again.
Note: for some reason, this is a change of behavior caused by the
upgrade to Mojave, identified in our Azure Pipeline; it seems that
Homebrew used to add the /usr/local/ directories to the include and link
search path before, but now it no longer does.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The default SIGPIPE behavior can be useful for a command that generates
a lot of output: if the receiver of our output goes away, we'll be
notified asynchronously to stop generating it (typically by killing the
program).
But for a command like fetch, which is primarily concerned with
receiving data and writing it to disk, an unexpected SIGPIPE can be
awkward. We're already checking the return value of all of our write()
calls, and dying due to the signal takes away our chance to gracefully
handle the error.
On Linux, we wouldn't generally see SIGPIPE at all during fetch. If the
other side of the network connection hangs up, we'll see ECONNRESET. But
on OS X, we get a SIGPIPE, and the process is killed. This causes t5570
to racily fail, as we sometimes die by signal (instead of the expected
die() call) when the server side hangs up.
Let's ignore SIGPIPE during the network portion of the fetch, which will
cause our write() to return EPIPE, giving us consistent behavior across
platforms.
This fixes the test flakiness, but note that it stops short of fixing
the larger problem. The server side hit a fatal error, sent us an "ERR"
packet, and then hung up. We notice the failure because we're trying to
write to a closed socket. But by dying immediately, we never actually
read the ERR packet and report its content to the user. This is a (racy)
problem on all platforms. So this patch lays the groundwork from which
that problem might be fixed consistently, but it doesn't actually fix
it.
Note the placement of the SIGPIPE handling. The absolute minimal change
would be to ignore SIGPIPE only when we're writing. But twiddling the
signal handler for each write call is inefficient and maintenance
burden. On the opposite end of the spectrum, we could simply declare
that fetch does not need SIGPIPE handling, since it doesn't generate a
lot of output, and we could just ignore it at the start of cmd_fetch().
This patch takes a middle ground. It ignores SIGPIPE during the network
operation (which is admittedly most of the program, since the actual
network operations are all done under the hood by the transport code).
So it's still pretty coarse.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The write_or_die() function has one quirk that a caller might not
expect: when it sees EPIPE from the write() call, it translates that
into a death by SIGPIPE. This doesn't change the overall behavior (the
program exits either way), but it does potentially confuse test scripts
looking for a non-signal exit code.
Let's switch away from using write_or_die() in a few code paths, which
will give us more consistent exit codes. It also gives us the
opportunity to write more descriptive error messages, since we have
context that write_or_die() does not.
Note that this won't do much by itself, since we'd typically be killed
by SIGPIPE before write_or_die() even gets a chance to do its thing.
That will be addressed in the next patch.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ensure writing out files properly blocks expanding $Id: when the ident
attribute is set for a path under GVFS.
Signed-off-by: Ben Peart <benpeart@microsoft.com>
Sort the set of objects by packfile so that only one packfile needs
to be open at a time.
This is a performance improvement. Previously, objects were
verified in OID order. This essentially requires all packfiles
to be open at the same time. If the number of packfiles exceeds
the open file limit, packfiles would be closed and re-opened
many times.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Fix "git multi-pack-index verify" to handle repos with thousands
of packfiles.
Midx verify adds the individual "packed_git" structures to the
multi_pack_index.packs array, but it does not add them to the
"repository.objects.packed_git" list. During the verification
code, each packfile is opened and scanned. And "pack_open_fds"
is incremented. If "pack_open_fds" equals the "pack_max_fds"
open_packed_git_1() calls close_one_pack() to LRU-style close
an already open packfile. But because the packfiles were never
added to the "packed_git" list, close_one_pack() does nothing.
If there are very many packfiles, Git runs out of file descriptors
and fails.
Note that this was observed on Windows when build with GCC and
in a repository with more than (2048-25) packfiles.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Guard against infinite loop while computing the parent process hierarchy.
CreateToolhelp32Snapshot() is used to get a list of all processes on the
system. Each process entry contains the process PID and PPID (alive at the
time of the snapshot). We compute the set of ancestors of the current process
by repeated searches on this list.
Testing revealed that the snapshot can contain PPID cycles. This causes an
infinite loop during the set construction and causes the git.exe command to
hang.
Testing found an instance where 3 processes were in a PPID cycle. The
snapshot implied that each of these processes was its own great-grandparent.
This should not be possible unless a PID was recycled and just happened to
match up.
For full disclosure, the Windows "System Idle Process" has PID and PPID 0.
If it were to launch a Git command, it could cause a similar infinite loop.
Or more properly, if any ancestor of the current Git command has PPID 0, it
will appear to be a descendant of the idle process and trigger the problem.
This commit fixes both cases by maintaining a list of the PIDs seen during
the ancestor walk and stopping if a cycle is detected.
Additionally, code was added to truncate the search after a reasonable fixed
depth limit.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
The git-p4 login ticket expiry test causes unreliable test
runs. Since the handling of ticket expiry in git-p4 is far
from polished anyway, let's remove it for now.
A better way to actually run the test is to create a python
"fake" version of "p4" which returns whatever expiry results
the test requires.
Ideally git-p4 would look at the expiry time before starting
any long operations, and cleanup gracefully if there is not
enough time left. But that's quite hard to do.
Signed-off-by: Luke Diamand <luke@diamand.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add trace2 region and data events describing attempts to deserialize
status data using a status cache.
A category:status, label:deserialize region is pushed around the
deserialize code.
Deserialization results when reading from a file are:
category:status, path = <path>
category:status, polled = <number_of_attempts>
category:status, result = "ok" | "reject"
When reading from STDIN are:
category:status, path = "STDIN"
category:status, result = "ok" | "reject"
Status will fallback and run a normal status scan when a "reject"
is reported (unless "--deserialize-wait=fail").
If "ok" is reported, status was able to use the status cache and
avoid scanning the workdir.
Additionally, a cmd_mode is emitted for each step: collection,
deserialization, and serialization. For example, if deserialization
is attempted and fails and status falls back to actually computing
the status, a cmd_mode message containing "deserialize" is issued
and then a cmd_mode for "collect" is issued.
Also, if deserialization fails, a data message containing the
rejection reason is emitted.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Add trace2 region around read_object_process to collect
time spent waiting for missing objects to be dynamically
fetched.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Temporarily fix wt-status.c in GVFS 2.20.* to use "the_repository"
rather than "s->repo" which does not yet appear in the this fork.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Johannes Schindelin
Junio C Hamano
Kevin Willford
Johannes Schindelin
SZEDER Gábor
Johannes Schindelin
Jeff Hostetler
Jeff Hostetler
Jeff King
Ben Peart
Luke Diamand