Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Release 4.0.54.0.5 | Tavian Barnes | 13 days | 4 | -4/+22 |
| | |||||
* | tests: Add a test for sigmask restoration | Tavian Barnes | 13 days | 2 | -0/+5 |
| | |||||
* | sighook: Don't forget to check `initialized` in sigreset() | Tavian Barnes | 13 days | 1 | -0/+4 |
| | |||||
* | sighook: Fix up siginit() error paths | Tavian Barnes | 13 days | 1 | -8/+14 |
| | |||||
* | xspawn: Unblock signals between fork() and exec() | Tavian Barnes | 13 days | 1 | -2/+15 |
| | | | | | | Otherwise the child process runs with all signals blocked. Fixes: 423cfa4 ("xspawn: Mask signals before fork()") | ||||
* | sighook: New sigreset() function to reset all handlers | Tavian Barnes | 13 days | 2 | -3/+45 |
| | |||||
* | sighook: Convert siglist to a more general rcu_list type | Tavian Barnes | 13 days | 1 | -52/+66 |
| | |||||
* | build: Feature-detect #pragma nounroll support | Tavian Barnes | 14 days | 3 | -2/+13 |
| | |||||
* | list: Don't use leading underscores for globals | Tavian Barnes | 2025-01-16 | 1 | -6/+6 |
| | |||||
* | ioq: Don't unroll the spin loop | Tavian Barnes | 2025-01-16 | 2 | -1/+15 |
| | |||||
* | ioq: Prefetch slots with a write intent | Tavian Barnes | 2025-01-16 | 1 | -1/+1 |
| | |||||
* | build: Check for compound literal storage-class support | Tavian Barnes | 2025-01-07 | 3 | -1/+8 |
| | | | | This lets us use them on GCC >= 13, even before C23. | ||||
* | configure: Suggest --enable-release over RELEASE=y etc. | Tavian Barnes | 2025-01-07 | 1 | -0/+19 |
| | |||||
* | configure: Reduce redundant info in CONFFLAGS for bfs --version | Tavian Barnes | 2025-01-07 | 1 | -1/+8 |
| | |||||
* | sighook: Add some missing fatal signals | Tavian Barnes | 2024-12-24 | 1 | -5/+14 |
| | |||||
* | sighook: Preserve the exact siginfo_t in reraise() on Linux | Tavian Barnes | 2024-12-24 | 1 | -3/+16 |
| | | | | Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=66dd34ad31e5963d72a700ec3f2449291d322921 | ||||
* | sighook: Always re-raise faults on macOS | Tavian Barnes | 2024-12-22 | 2 | -0/+11 |
| | | | | | | | | | | macOS always fills in si_code for SIG{BUS,ILL,SEGV} as if it were a real hardware fault, so returning from the handler is not guaranteed to re- trigger the signal. Fixes: aecdabb ("sighook: Return instead of re-raising for faults") Link: https://chromium-review.googlesource.com/c/crashpad/crashpad/+/2386463/4/util/posix/signals_test.cc Link: https://github.com/chromium/crashpad/commit/e0d8a0aa01ac176804077f1f128ccc894c098f79 | ||||
* | tests/sighook: Add missing #include <sys/wait.h> | Tavian Barnes | 2024-12-19 | 1 | -2/+3 |
| | |||||
* | tests/gnu: Add an -ignore_readdir_race test for removed directories | Tavian Barnes | 2024-12-17 | 2 | -0/+7 |
| | | | | Link: https://savannah.gnu.org/bugs/index.php?45930 | ||||
* | tests/sighook: Add atsigexit() tests | Tavian Barnes | 2024-12-17 | 1 | -1/+77 |
| | |||||
* | sanity: New sanitize_resize() function | Tavian Barnes | 2024-12-16 | 2 | -16/+33 |
| | | | | | This wraps __sanitizer_annotate_contiguous_container() to give byte-precise tracking of usable allocation sizes with ASan. | ||||
* | sanity: Get rid of sanitize_ignore() | Tavian Barnes | 2024-12-16 | 2 | -11/+6 |
| | | | | | It's probably nicer to avoid evaluating expensive arguments when not sanitizing, rather than relying on the optimizer to clean them up. | ||||
* | ci: Update VM versions | Tavian Barnes | 2024-12-10 | 1 | -3/+3 |
| | |||||
* | build(deps): bump cross-platform-actions/action from 0.25.0 to 0.26.0 | dependabot[bot] | 2024-12-09 | 1 | -3/+3 |
| | | | | | | | | | | | | | | | Bumps [cross-platform-actions/action](https://github.com/cross-platform-actions/action) from 0.25.0 to 0.26.0. - [Release notes](https://github.com/cross-platform-actions/action/releases) - [Changelog](https://github.com/cross-platform-actions/action/blob/master/changelog.md) - [Commits](https://github.com/cross-platform-actions/action/compare/v0.25.0...v0.26.0) --- updated-dependencies: - dependency-name: cross-platform-actions/action dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> | ||||
* | docs/bfs.1: Add copyright comments | Tavian Barnes | 2024-12-04 | 1 | -0/+2 |
| | |||||
* | ioq: Refactor ioq_ring_probe_flags() | Tavian Barnes | 2024-12-04 | 1 | -11/+13 |
| | |||||
* | tests/sighook: Test the SH_ONESHOT flag | Tavian Barnes | 2024-12-04 | 1 | -2/+28 |
| | |||||
* | ioq: Try spinning before blocking in ioq_slot_wait() | Tavian Barnes | 2024-12-03 | 1 | -1/+17 |
| | |||||
* | ioq: Submit and pop requests in batches | Tavian Barnes | 2024-12-03 | 5 | -17/+44 |
| | | | | | The new ioq_submit() function is now necessary to call to ensure the pending request batch is flushed. | ||||
* | ioq: Add more io_uring setup flags | Tavian Barnes | 2024-12-03 | 1 | -20/+55 |
| | |||||
* | ioq: Refactor io_uring submission logic | Tavian Barnes | 2024-12-03 | 1 | -94/+145 |
| | |||||
* | ioq: Use ioq_batch for both pushing and popping | Tavian Barnes | 2024-12-03 | 1 | -47/+84 |
| | |||||
* | ioq: Add a hash function between slots and monitors | Tavian Barnes | 2024-12-03 | 1 | -1/+11 |
| | | | | | | This helps avoid situations where multiple waiters block on different slots using the same monitor, which happened more often than expected due to correlations caused by batching. | ||||
* | ioq: Prefetch pointers before popping them | Tavian Barnes | 2024-12-03 | 2 | -8/+10 |
| | | | | | Also, cache-align struct ioq_ent to avoid false sharing when two workers are handling neighbouring requests. | ||||
* | bench/ioq: New ioq microbenchmark | Tavian Barnes | 2024-12-03 | 2 | -1/+332 |
| | |||||
* | ioq: Add an ioq_nop() operation for benchmarking | Tavian Barnes | 2024-12-02 | 3 | -0/+62 |
| | |||||
* | xtime: New timespec utility functions | Tavian Barnes | 2024-12-02 | 3 | -23/+95 |
| | |||||
* | sighook: New SH_ONESHOT flag | Tavian Barnes | 2024-12-02 | 2 | -1/+21 |
| | |||||
* | bfstd: Add more strto*() wrappers | Tavian Barnes | 2024-12-02 | 2 | -24/+69 |
| | |||||
* | ioq: Set the worker thread names to ioq-%d | Tavian Barnes | 2024-11-27 | 1 | -2/+8 |
| | |||||
* | thread: New thread_setname() function | Tavian Barnes | 2024-11-27 | 5 | -0/+37 |
| | |||||
* | sighook: Return instead of re-raising for faults | Tavian Barnes | 2024-11-23 | 1 | -5/+11 |
| | | | | | | | | | | | This makes the death look exactly like it would if we didn't handle the signal at all. Coredumps will point at the right instruction, segfaults will get logged in dmesg again, etc. Technically POSIX says this is undefined, but if we get a fault, we've already done something undefined anyway ;) Link: https://github.com/rust-lang/rust/issues/26458 | ||||
* | sighook: Check for SA_RESTART before using it | Tavian Barnes | 2024-11-23 | 1 | -1/+7 |
| | | | | QNX doesn't support SA_RESTART. | ||||
* | bfstd: makedev() takes 3 arguments on QNX | Tavian Barnes | 2024-11-23 | 1 | -1/+3 |
| | |||||
* | prelude: Define _QNX_SOURCE | Tavian Barnes | 2024-11-23 | 1 | -0/+5 |
| | |||||
* | build: Check for -pthread before using it | Tavian Barnes | 2024-11-23 | 2 | -2/+12 |
| | | | | | | QNX's qcc, for example, will complain that cc1: error: command-line option '-pthread' is valid for the driver but not for C | ||||
* | tests/getopts: Remove unused variable | Tavian Barnes | 2024-11-21 | 1 | -1/+0 |
| | |||||
* | docs/bfs.1: Simplify the formatting | Tavian Barnes | 2024-11-20 | 1 | -64/+105 |
| | |||||
* | ioq: Enable IORING_SETUP_SUBMIT_ALL | Tavian Barnes | 2024-11-20 | 1 | -2/+18 |
| | |||||
* | color: Squelch a -Wmaybe-uninitialized warning on non-LTO builds | Tavian Barnes | 2024-11-20 | 1 | -1/+1 |
| |