On macOS, fsmonitor can fall into a race condition that results in
a client waiting forever to be notified for an event that have
already happened. This problem has been corrected.
* jk/fsmonitor-event-listener-race-fix:
fsmonitor: initialize fs event listener before accepting clients
simple-ipc: split async server initialization and running
A new configuration variable remote.<name>.serverOption makes the
transport layer act as if the --serverOption=<value> option is
given from the command line.
* xx/remote-server-option-config:
ls-remote: leakfix for not clearing server_options
fetch: respect --server-option when fetching multiple remotes
transport.c:🤝 make use of server options from remote
remote: introduce remote.<name>.serverOption configuration
transport: introduce parse_transport_option() method
Code clean-up.
* jk/output-prefix-cleanup:
diff: store graph prefix buf in git_graph struct
diff: return line_prefix directly when possible
diff: return const char from output_prefix callback
diff: drop line_prefix_length field
line-log: use diff_line_prefix() instead of custom helper
Use after free and double freeing at the end in "git log -L... -p"
had been identified and fixed.
* ds/line-log-asan-fix:
line-log: protect inner strbuf from free
Doc update to clarify how periodical maintenance are scheduled,
spread across time to avoid thundering hurds.
* sk/doc-maintenance-schedule:
doc: add a note about staggering of maintenance
The reftable library is now prepared to expect that the memory
allocation function given to it may fail to allocate and to deal
with such an error.
* ps/reftable-alloc-failures: (26 commits)
reftable/basics: fix segfault when growing `names` array fails
reftable/basics: ban standard allocator functions
reftable: introduce `REFTABLE_FREE_AND_NULL()`
reftable: fix calls to free(3P)
reftable: handle trivial allocation failures
reftable/tree: handle allocation failures
reftable/pq: handle allocation failures when adding entries
reftable/block: handle allocation failures
reftable/blocksource: handle allocation failures
reftable/iter: handle allocation failures when creating indexed table iter
reftable/stack: handle allocation failures in auto compaction
reftable/stack: handle allocation failures in `stack_compact_range()`
reftable/stack: handle allocation failures in `reftable_new_stack()`
reftable/stack: handle allocation failures on reload
reftable/reader: handle allocation failures in `reader_init_iter()`
reftable/reader: handle allocation failures for unindexed reader
reftable/merged: handle allocation failures in `merged_table_init_iter()`
reftable/writer: handle allocation failures in `reftable_new_writer()`
reftable/writer: handle allocation failures in `writer_index_hash()`
reftable/record: handle allocation failures when decoding records
...
The way AsciiDoc is used for SYNOPSIS part of the manual pages has
been revamped. The sources, at least for the simple cases, got
vastly pleasant to work with.
* ja/doc-synopsis-markup:
doc: apply synopsis simplification on git-clone and git-init
doc: update the guidelines to reflect the current formatting rules
doc: introduce a synopsis typesetting
The synopsis for `git config unset` mentions two positional arguments:
`<name>` and `<value>`. While the first argument is correct, the second
is not. Users are expected to provide the value via `--value=<value>`.
Remove the positional argument. The `--value=<value>` option is already
documented correctly, so this is all we need to do to fix the
documentation.
Signed-off-by: Josh Heinrichs <joshiheinrichs@gmail.com>
Reviewed-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
There's a racy hang in fsmonitor on macOS that we sometimes see in CI.
When we serve a client, what's supposed to happen is:
1. The client thread calls with_lock__wait_for_cookie() in which we
create a cookie file and then wait for a pthread_cond event
2. The filesystem event listener sees the cookie file creation, does
some internal book-keeping, and then triggers the pthread_cond.
But there's a problem: we start the listener that accepts client threads
before we start the fs event thread. So it's possible for us to accept a
client which creates the cookie file and starts waiting before the fs
event thread is initialized, and we miss those filesystem events
entirely. That leaves the client thread hanging forever.
In CI, the symptom is that t9210 (which is testing scalar, which always
enables fsmonitor under the hood) may hang forever in "scalar clone". It
is waiting on "git fetch" which is waiting on the fsmonitor daemon.
The race happens more frequently under load, but you can trigger it
predictably with a sleep like this, which delays the start of the fs
event thread:
--- a/compat/fsmonitor/fsm-listen-darwin.c
+++ b/compat/fsmonitor/fsm-listen-darwin.c
@@ -510,6 +510,7 @@ void fsm_listen__loop(struct fsmonitor_daemon_state *state)
FSEventStreamSetDispatchQueue(data->stream, data->dq);
data->stream_scheduled = 1;
+ sleep(1);
if (!FSEventStreamStart(data->stream)) {
error(_("Failed to start the FSEventStream"));
goto force_error_stop_without_loop;
One solution might be to reverse the order of initialization: start the
fs event thread before we start the thread listening for clients. But
the fsmonitor code explicitly does it in the opposite direction. The fs
event thread wants to refer to the ipc_server_data struct, so we need it
to be initialized first.
A further complication is that we need a signal from the fs event thread
that it is actually ready and listening. And those details happen within
backend-specific fsmonitor code, whereas the initialization is in the
shared code.
So instead, let's use the ipc_server init/start split added in the
previous commit. The generic fsmonitor code will init the ipc_server but
_not_ start it, leaving that to the backend specific code, which now
needs to call ipc_server_start_async() at the right time.
For macOS, that is right after we start the FSEventStream that you can
see in the diff above.
It's not clear to me if Windows suffers from the same problem (and we
simply don't trigger it in CI), or if it is immune. Regardless, the
obvious place to start accepting clients there is right after we've
established the ReadDirectoryChanges watch.
This makes the hangs go away in our macOS CI environment, even when
compiled with the sleep() above.
Helped-by: Koji Nakamaru <koji.nakamaru@gree.net>
Signed-off-by: Jeff King <peff@peff.net>
Acked-by: Koji Nakamaru <koji.nakamaru@gree.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
To start an async ipc server, you call ipc_server_run_async(). That
initializes the ipc_server_data object, and starts all of the threads
running, which may immediately start serving clients.
This can create some awkward timing problems, though. In the fsmonitor
daemon (the sole user of the simple-ipc system), we want to create the
ipc server early in the process, which means we may start serving
clients before the rest of the daemon is fully initialized.
To solve this, let's break run_async() into two parts: an initialization
which allocates all data and spawns the threads (without letting them
run), and a start function which actually lets them begin work. Since we
have two simple-ipc implementations, we have to handle this twice:
- in ipc-unix-socket.c, we have a central listener thread which hands
connections off to worker threads using a work_available mutex. We
can hold that mutex after init, and release it when we're ready to
start.
We do need an extra "started" flag so that we know whether the main
thread is holding the mutex or not (e.g., if we prematurely stop the
server, we want to make sure all of the worker threads are released
to hear about the shutdown).
- in ipc-win32.c, we don't have a central mutex. So we'll introduce a
new startup_barrier mutex, which we'll similarly hold until we're
ready to let the threads proceed.
We again need a "started" flag here to make sure that we release the
barrier mutex when shutting down, so that the sub-threads can
proceed to the finish.
I've renamed the run_async() function to init_async() to make sure we
catch all callers, since they'll now need to call the matching
start_async().
We could leave run_async() as a wrapper that does both, but there's not
much point. There are only two callers, one of which is fsmonitor, which
will want to actually do work between the two calls. And the other is
just a test-tool wrapper.
For now I've added the start_async() calls in fsmonitor where they would
otherwise have happened, so there should be no behavior change with this
patch.
Signed-off-by: Jeff King <peff@peff.net>
Acked-by: Koji Nakamaru <koji.nakamaru@gree.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ensure `server_options` is properly cleared using `string_list_clear()`
in `builtin/ls-remote.c:cmd_ls_remote`.
Although we cannot yet enable `TEST_PASSES_SANITIZE_LEAK=true` for
`t/t5702-protocol-v2.sh` due to other existing leaks, this fix ensures
that "git-ls-remote" related server options tests pass the sanitize leak
check:
...
ok 12 - server-options are sent when using ls-remote
ok 13 - server-options from configuration are used by ls-remote
...
Signed-off-by: Xing Xin <xingxin.xx@bytedance.com>
Reviewed-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Fix an issue where server options specified via the command line
(`--server-option` or `-o`) were not sent when fetching from multiple
remotes using Git protocol v2.
To reproduce the issue with a repository containing multiple remotes:
GIT_TRACE_PACKET=1 git -c protocol.version=2 fetch --server-option=demo --all
Observe that no server options are sent to any remote.
The root cause was identified in `builtin/fetch.c:fetch_multiple`, which
is invoked when fetching from more than one remote. This function forks
a `git-fetch` subprocess for each remote but did not include the
specified server options in the subprocess arguments.
This commit ensures that command-line specified server options are
properly passed to each subprocess. Relevant tests have been added.
Signed-off-by: Xing Xin <xingxin.xx@bytedance.com>
Reviewed-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Utilize the `server_options` from the corresponding remote during the
handshake in `transport.c` when Git protocol v2 is detected. This helps
initialize the `server_options` in `transport.h:transport` if no server
options are set for the transport (typically via `--server-option` or
`-o`).
While another potential place to incorporate server options from the
remote is in `transport.c:transport_get`, setting server options for a
transport using a protocol other than v2 could lead to unexpected errors
(see `transport.c:die_if_server_options`).
Relevant tests and documentation have been updated accordingly.
Signed-off-by: Xing Xin <xingxin.xx@bytedance.com>
Reviewed-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently, server options for Git protocol v2 can only be specified via
the command line option "--server-option" or "-o", which is inconvenient
when users want to specify a list of default options to send. Therefore,
we are introducing a new configuration to hold a list of default server
options, akin to the `push.pushOption` configuration for push options.
Initially, I named the new configuration `fetch.serverOption` to align
with `push.pushOption`. However, after discussing with Patrick, it was
renamed to `remote.<name>.serverOption` as suggested, because:
1. Server options are designed to be server-specific, making it more
logical to use a per-remote configuration.
2. Using "fetch." prefixed configurations in git-clone or git-ls-remote
seems out of place and inconsistent in design.
The parsing logic for `remote.<name>.serverOption` also relies on
`transport.c:parse_transport_option`, similar to `push.pushOption`, and
they follow the same priority design:
1. Server options set in lower-priority configuration files (e.g.,
/etc/gitconfig or $HOME/.gitconfig) can be overridden or unset in
more specific repository configurations using an empty string.
2. Command-line specified server options take precedence over those from
the configuration.
Server options from configuration are stored to the corresponding
`remote.h:remote` as a new field `server_options`. The field will be
utilized in the subsequent commit to help initialize the
`server_options` of `transport.h:transport`.
And documentation have been updated accordingly.
Helped-by: Patrick Steinhardt <ps@pks.im>
Helped-by: Junio C Hamano <gitster@pobox.com>
Reported-by: Liu Zhongbo <liuzhongbo.6666@bytedance.com>
Signed-off-by: Xing Xin <xingxin.xx@bytedance.com>
Reviewed-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add the `parse_transport_option()` method to parse the `push.pushOption`
configuration. This method will also be used in the next commit to
handle the new `remote.<name>.serverOption` configuration for setting
server options in Git protocol v2.
Signed-off-by: Xing Xin <xingxin.xx@bytedance.com>
Reviewed-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
These links should point to `.html` files, not to `.txt` ones.
Compare also to 4945f046c7 (api docs: link to html version of
api-trace2, 2022-09-16).
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
macOS with fsmonitor daemon can hang forever when a submodule is
involved, which has been corrected.
* kn/osx-fsmonitor-with-submodules-fix:
fsmonitor OSX: fix hangs for submodules
