aboutsummaryrefslogtreecommitdiff
path: root/src/runtime
AgeCommit message (Collapse)Author
18 hoursall: prealloc slice with possible minimum capabilitiesShulhan
41 hoursruntime/pprof: using buffer writer instead of original writerWeixie Cui
Change-Id: Ibff8e7a08090d67d9a0625dee068c8d2eefbd190 GitHub-Last-Rev: b90b986121314615cf5d259d76a7be59303de1b0 GitHub-Pull-Request: golang/go#78630 Reviewed-on: https://go-review.googlesource.com/c/go/+/765480 Reviewed-by: Florian Lehner <lehner.florian86@gmail.com> Reviewed-by: David Chase <drchase@google.com> Auto-Submit: Keith Randall <khr@google.com> Reviewed-by: Keith Randall <khr@google.com> Reviewed-by: Ian Lance Taylor <iant@golang.org> LUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
5 daysruntime: fix AIX library initializationqmuntal
On AIX, libInit passes a function descriptor instead of the function pointer to pthread_create. This is a regression from CL 706417. Change-Id: I660175eb992a41ef61b1927c51392887a724cd76 Reviewed-on: https://go-review.googlesource.com/c/go/+/761780 Reviewed-by: Ian Lance Taylor <iant@golang.org> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: David Chase <drchase@google.com>
8 daysruntime: remove gopanicFP field from panic recordKeith Randall
We can use SP instead of FP for the single thing this was used for (finding the gopanic frame). Change-Id: Iad2b406705abc3cdb0dbcdd069f7f9a330509164 Reviewed-on: https://go-review.googlesource.com/c/go/+/738042 Reviewed-by: Keith Randall <khr@google.com> Reviewed-by: David Chase <drchase@google.com> Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
8 daysruntime: save and restore fcsr0 registers in async preempt on loong64Guoqi Chen
Currently, all Op implementations on loong64 do not modify the FCSR (floating point control status register), but users can modify the FCSR for special floating point computing needs. FCSR is also saved and restored in the fpu context of the Linux kernel [1]. [1]: https://github.com/torvalds/linux/blob/master/arch/loongarch/kernel/fpu.S#L565C1-L566C1 Change-Id: I776e72dfcc1f027bfe8d22aa2eae453100a6ee1c Reviewed-on: https://go-review.googlesource.com/c/go/+/761720 Reviewed-by: Meidan Li <limeidan@loongson.cn> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Junyang Shao <shaojunyang@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
9 daysruntime: fix timespec definition on 32bits systemsJorropo
The nsec field of timespec is a C long even when using 64bits time on 32bits systems. This is because by timespec API if nsec never holds more than a second worth of nanoseconds. If it would theses would increment the sec field while the nsec field would get the amount of nanoseconds modulus a second. Fixes #77934 Change-Id: I9803998ba70123eb3b226379bd72b11cae972c38 Reviewed-on: https://go-review.googlesource.com/c/go/+/751341 Reviewed-by: Michael Pratt <mpratt@google.com> Auto-Submit: Jorropo <jorropo.pgm@gmail.com> Reviewed-by: David Chase <drchase@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
12 daysruntime: add sysUnreserve to undo sysReserveMichael Pratt
This is inspired by CL 724560 by Bobby Powers, particularly their great commit message. When using address sanitizer with leak detection, sysReserve registers memory regions with LSAN via lsanregisterrootregion. However, several code paths release this memory using sysFreeOS without first unregistering from LSAN. This leaves LSAN with stale root region entries pointing to memory that has been unmapped and may be reallocated for other purposes. This bug was latent until glibc 2.42, which changed pthread stack guard pages from mprotect(PROT_NONE) to madvise(MADV_GUARD_INSTALL). The difference matters because LSAN filters root region scanning by intersecting registered regions with readable mappings from /proc/self/maps: - mprotect(PROT_NONE) splits the VMA, creating a separate entry with ---p permissions. LSAN's IsReadable() check excludes it from scanning. - MADV_GUARD_INSTALL operates at the page table level without modifying the VMA. The region still appears as rw-p in /proc/self/maps, so LSAN includes it in the scan and crashes with SIGSEGV when accessing the guard pages. Address this by adding sysUnreserve to undo sysReserve. sysUnreserve unregisters the region from LSAN and frees the mapping. With the addition of sysUnreserve, we have complete coverage of LSAN unregister in the mem.go abstract: sysFree unregisters Ready memory. sysUnreserve unregisters Reserved memory. And there is no way to free Prepared memory at all (it must transition to Ready or Reserved first). The implementation of lsanunregisterrootregion [1] finds the region by exact match of start and end address. It therefore does not support splitting a region, and we must extend this requirement to sysUnreserve and sysFree. I am not completely confident that we always pass the full region to sysFree, but LSAN aborts if it can't find the region, so we must not be blatantly violating this. sysReserveAligned does need to unreserve a subset of a region, so it cannot use sysUnreserve directly. Rather than breaking the mem.go abstract, move sysReserveAligned into mem.go, adding it to the abstraction. We should not have any calls to sysFreeOS outside of the mem.go abstraction. That is now true with this CL. Fixes #74476. [1] https://github.com/llvm/llvm-project/blob/3e3e362648fa062038b90ccc21f46a09d6902288/compiler-rt/lib/lsan/lsan_common.cpp#L1157 Change-Id: I8c46a62154b2f23456ffd5086a7b91156a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/762381 Reviewed-by: Michael Knyszek <mknyszek@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
13 daysruntime: clarify sysFree documentationMichael Pratt
sysFree decrements mappedReady since CL 393402, meaning it can only be used on Ready memory, not any memory as stated. The example uses are not relevant as those cases would not necessarily be Ready memory. The no-op note is meant to be relevant to sysReserveAligned, but that doesn't use sysFree anyway. Change-Id: I3aeb0e703b0a57202a01317e0904cd966a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/762380 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Michael Pratt <mpratt@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>
13 daysall: add export linknames for assembly symbols accessed from other packageCherry Mui
For Go symbols accessed from other package via linkname or assembly, we have an export linkname from the definition side. We currently don't always have the linkname directive for assembly functions, for which external accesses are allowed. We may want to tighten up the restriction. So add export linknames for the ones that are needed. Change-Id: If664634c81580edd49086d916024f23f86871092 Reviewed-on: https://go-review.googlesource.com/c/go/+/749981 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Pratt <mpratt@google.com>
13 daysruntime: use asmcgocall_no_g in libInitMichael Pratt
libInit runs before rt0_go, which is where TLS setup occurs. Thus the contents of the TLS may not be defined, so the g lookup in asmcgocall is not safe. Concretely, android-386 c-shared builds crash without this change because asmcgocall reads an invalid non-zero g from the TLS. Move libInit to a file limited to GOARCH that support c-archive or c-shared so that only those require asmcgocall_no_g. In addition, loong64 and s390x need asmcgocall_no_g implementations. Fixes #78480. Cq-Include-Trybots: luci.golang.try:gotip-linux-loong64,gotip-linux-s390x Change-Id: I175e6d020339af89c9b576535d79c1e76a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/761541 Reviewed-by: Quim Muntal <quimmuntal@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
13 daysruntime: mention linknamestd in HACKING.mdCherry Mui
Change-Id: Icc89533e4cfc737bfee8b0c7a72006bdc79746c2 Reviewed-on: https://go-review.googlesource.com/c/go/+/762280 Reviewed-by: Michael Pratt <mpratt@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
13 daysruntime: rename _mkmalloc's module to avoid GOROOT conflictMichael Matloob
Often we want to run _mkmalloc using a different toolchain. The previous module path runtime/_mkmalloc would produce an ambiguous import path error because the directory runtime/_mkmalloc exists in a Go 1.26+ toolchain. Rename the module to avoid the conflict. Change-Id: Id73a3ca102081d8b51da58879aee851b6a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/761880 Reviewed-by: Michael Matloob <matloob@google.com> Auto-Submit: Michael Matloob <matloob@google.com> Reviewed-by: Michael Pratt <mpratt@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
14 daysruntime/secret: replace stub reference in test assemblyCherry Mui
In runtime/secret's TestCore, it copies some files from the runtime/secret package to a test main package, including assembly files. The assembly function references a stub function in Go, which is also copied. Make it reference the copied one, not the original one, to avoid cross-package assembly references. We may want to rewrite the test to avoid copy, e.g. including the test functions in the test binary itself and having the test run itself in a subprocess. Leave that for later. Change-Id: I5187edcc41d1d0469fc6b1b5f2acc60df37e5c1e Reviewed-on: https://go-review.googlesource.com/c/go/+/761520 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: David Chase <drchase@google.com>
14 dayscmd/compile, cmd/link: add linknamestd directive for std-only linknamesCherry Mui
In the standard library, there are a number of linknames, for sharing symbols within the standard library. They are not supposed to be accessed externally. But currently there is no good mechanism to prevent that. In the linker we have a blocklist of linknames, which forbids linkname references other than explicitly allowed packages. The blocklist is manually maintained, requiring periodic manual update. To move away from that manually maintained blocklist, this CL introduces a new directive, linknamestd, that marks a linkname for use within the standard library only. The linker will allow references within the standard library and forbid others. For a proof of concept, runtime.coroswitch is removed from the blocklist, and replaced with linknamestd. An external reference to it is still disallowed by the linker, as tested with cmd/link.TestCheckLinkname with testdata/linkname/coro.go. Change-Id: I0d0f8746b8835d8cdcfc3ff835d22a551da5f038 Reviewed-on: https://go-review.googlesource.com/c/go/+/749942 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: David Chase <drchase@google.com>
2026-03-31runtime/cgo: remove SIGSEGV emulation on iOSqmuntal
The misc/ios exec wrapper doesn't support lldb since https://go.dev/cl/573175. This means that the SIGSEGV emulation code in the cgo signal handling on iOS is no longer needed, and can be removed. Cq-Include-Trybots: luci.golang.try:gotip-darwin-arm64_26 Change-Id: I39827cb20756e4730352d87cb3514bb6a3f1cee8 Reviewed-on: https://go-review.googlesource.com/c/go/+/756800 Reviewed-by: Junyang Shao <shaojunyang@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
2026-03-31runtime: unix: sane exit in dieFromSignal for pid 1Kir Kolyshkin
A curious bug was reported to kubernetes[1] and runc[2] recently: sometimes runc init reports exit status of 2. Turns out, Go runtime assumes that on any UNIX system signals such as SIGTERM (or any other that has _sigKill flag set in sigtable) with no signal handler set up, will result in kernel terminating the program. This is true, except for PID 1 which gets a custom treatment from the kernel. As a result, when a Go program that runs as PID 1 (which is easy to achieve in Linux by using a new PID namespace) receives such a signal, Go runtime calls dieFromSignal which falls through all the way to exit(2), which is very confusing to a user. This issue can be worked around by the program by adding custom handlers for SIGTERM/SIGINT/SIGHUP, but that requires a goroutine to handle those signals, which, in case of runc, unnecessarily raises its NPROC/pid.max requirement (see discussion at [2]). Since practically exit(2) in dieFromSignal can only happen when the process is running as PID 1, replace it with exit(128+sig) to mimic the shell convention when a child is terminated by a signal. Add a test case which demonstrates the issue and validates the fix (albeit only on Linux). [An earlier version of this patch used to do nothing in dieFromSignal for PID 1 case, but such behavior might be a breaking change for a Go program running in a Linux container as PID 1.] Fixes #78442 [1]: https://github.com/kubernetes/kubernetes/issues/135713 [2]: https://github.com/opencontainers/runc/pull/5189 Change-Id: I196e09e4b5ce84ce2c747a0c2d1fc6e9cf3a6131 Reviewed-on: https://go-review.googlesource.com/c/go/+/759040 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Junyang Shao <shaojunyang@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
2026-03-31cmd/link, runtime: record size of itabsIan Lance Taylor
We were depending on runtime.etypes immediately following the itabs. However, on AIX, runtime.etypes, as a zero-sized symbol, can float when using external linking. It won't necessarily stay right at the end of the itabs. Rather than worry about this, just record the size of the itab data. In practice it almost always works on AIX, but it fails the runtime test TestSchedPauseMetrics/runtime/debug.WriteHeapDump, which fails when iterating over all the itabs. Tested on AIX. This should fix AIX on the build dashboard. Change-Id: Id3a113b75b93fa8440c047e92f764ab81423df48 Reviewed-on: https://go-review.googlesource.com/c/go/+/760203 Auto-Submit: Ian Lance Taylor <iant@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Junyang Shao <shaojunyang@google.com>
2026-03-30runtime/race: allow TestOutput output in either orderMichael Pratt
We have only seen the wrappersym case reported in the opposite order in practice, but since these are intentionally racy, they could theoretically all occur in the opposite order. It would be nice to have a bit more structure to these tests so the test itself could easily flip the order. However, since they are just regular expressions, for now I've simply listed both orders. Fixes #78394. Change-Id: I92e04127f275c2394a12d63d2f2a3ce56a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/761161 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Michael Pratt <mpratt@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
2026-03-30runtime/cgo: remove unused _cgo_openbsd_try_pthread_create prototypeTobias Klauser
The function is no longer defined since CL 40331. Remove its prototype from libcgo_unix.h as well. Change-Id: I155ca632cd61bddcd798def691856c04b03040d7 Reviewed-on: https://go-review.googlesource.com/c/go/+/760163 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Florian Lehner <lehner.florian86@gmail.com> Auto-Submit: Tobias Klauser <tobias.klauser@gmail.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2026-03-28runtime: remove redundant fields from memory profile recordsNick Ripley
The memProfCycle struct holds allocation counts and bytes allocated, and frees and bytes freed. But the memory profile records are already aggregated by allocation size, which is stored in the "size" field of the bucket struct. We can derive the bytes allocated/freed using the counts and the size we already store. Thus we can delete the bytes fields from memProfCycle, saving 64 bytes per memRecord. We can do something similar for the profilerecord.MemProfileRecord type. We just need to know the object size and we can derive the allocated and freed bytes accordingly. Change-Id: I103885c2f29471b25283e330674fc16d6a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/760140 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Michael Pratt <mpratt@google.com>
2026-03-27runtime/race: use subtests for TestOutputMichael Pratt
For #78394. Change-Id: I4ef1a299f37e2b6532e2008378aca3aa6a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/760308 Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Michael Pratt <mpratt@google.com>
2026-03-26reflect, runtime: replace reflect.typelinks to return types, not offsetsIan Lance Taylor
Change reflect to call a new function to get the compiled type descriptors, returning the type pointers directly rather than the type offsets. We have to keep the existing reflect.typelinks for third party packages that break the rules and call it directly using linkname. Change-Id: I476efb6bd7836a7a5f396c97bbe4b2c1a2428990 Reviewed-on: https://go-review.googlesource.com/c/go/+/729502 Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Antonio Camacho <antoniocho444@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Ian Lance Taylor <iant@golang.org> Reviewed-by: Keith Randall <khr@google.com>
2026-03-26runtime: truncate trace strings before inserting into trace mapTobias Grieger
traceStringTable.put inserted the full user-supplied string into the trace map, then only truncated it to MaxEventTrailerDataSize (1024 bytes) when writing to the trace buffer. If the string exceeded the traceRegionAlloc block size (~64KB), this caused a fatal "traceRegion: alloc too large" crash. Move the truncation to the top of put, before the map insertion, so that the map key, map entry, and written output are all consistent and bounded. The existing truncation in writeString is retained: the emit method also calls writeString without going through the map, so writeString still needs its own guard. TestStartRegionLongString reproduces the crash before the fix. Observed in production at CockroachDB: Stopper.RunTask passes singleflight keys (up to ~450KB) as trace region names via trace.StartRegion. See: https://github.com/cockroachdb/cockroach/pull/166669 for context on the trigger. Change-Id: I95d0b2f0bd2e806840b83a0b675ce6d2f0e2c2c5 GitHub-Last-Rev: 3c53061685d5237f9f2fc4522fce6d774776fede GitHub-Pull-Request: golang/go#78348 Reviewed-on: https://go-review.googlesource.com/c/go/+/759140 Auto-Submit: Michael Knyszek <mknyszek@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-26runtime, cmd/link: remove itablinksIan Lance Taylor
Instead of keeping a separate list of pointers to itabs, just walk through the itabs themselves. For #6853 Change-Id: If030bd64fbd01d73b0bf8495f6c9826ed2e61568 Reviewed-on: https://go-review.googlesource.com/c/go/+/729201 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Ian Lance Taylor <iant@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
2026-03-26runtime: std FDs are always opened, not only when AT_SECURE is setMateusz Poliwczak
Updates #60641 Change-Id: I0340a561690a2b45e27a82dc15479b846a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/754400 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Keith Randall <khr@google.com> Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
2026-03-25runtime: use uname version check for 64-bit time on 32-bit arch codepathsBrad Fitzpatrick
The previous fallback-on-ENOSYS logic causes issues on forks of Linux. Android: #77621 (CL 750040 added a workaround with a TODO, this fixes that TODO) Causes the OS to terminate the program when running on Android versions <=10 since the seccomp jail does not know about the 64-bit time syscall and is configured to terminate the program on any unknown syscall. Synology's Linux: #77930 On old versions of Synology's Linux they added custom vendor syscalls without adding a gap in the syscall numbers, that means when we call the newer Linux syscall which was added later, Synology's Linux interprets it as a completely different vendor syscall. Originally by Jorropo in CL 751340. Fixes #77930 Co-authored-by: Jorropo <jorropo.pgm@gmail.com> Change-Id: I90e15495d9249fd7f6e112f9e3ae8ad1322f56e0 Reviewed-on: https://go-review.googlesource.com/c/go/+/758902 Reviewed-by: Damien Neil <dneil@google.com> Reviewed-by: Michael Pratt <mpratt@google.com> Reviewed-by: Jorropo <jorropo.pgm@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-24runtime: zero-extend Windows error code in syscall_syscallnChristian Hoeppner
asmstdcall now returns the 32-bit Windows error code in the return register. cgocall captures this as a signed int32. On 64-bit systems, casting this directly to uintptr results in sign extension for error codes with the high bit set (e.g., 0x80092004 becomes 0xffffffff80092004), breaking equality checks against zero-extended constants. This CL ensures the error code is zero-extended by casting to uint32 before converting to uintptr. Fixes #78324 Change-Id: Ia75fe32965ccc7fb836f7caff8bbb8575af11e2e Reviewed-on: https://go-review.googlesource.com/c/go/+/758800 Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Carlos Amedee <carlos@golang.org>
2026-03-24internal/runtime/maps: add GOEXPERIMENT=mapsplitgroup for KKKKVVVV slot orderJake Bailey
Map groups are currently: type group struct { ctrl uint64 slots [8]slot } type slot struct { key K elem E } If the element type is struct{}, the slot will be padded so that the address of the elem is unique rather than pointing outside the alloc. This has the effect of map[K]struct{} wasting space due to the extra byte and padding, making it no better than map[K]bool. This CL changes the group layout to instead place keys and elems together, as they used to be before swiss maps: type group struct { ctrl uint64 keys [8]K elems [8]V } This is an alternative to CL 701976, which I suspect will have better performance. Keys placed together should lead to better cache behavior, at the cost of more expensive elem lookups, since the elems are not a fixed offset from their keys. This change is locked behind GOEXPERIMENT=mapsplitgroup. Updates #70835 Updates #71368 Change-Id: Ide8d1406ae4ab636f86edc40e0640cc80653197c Reviewed-on: https://go-review.googlesource.com/c/go/+/711560 Reviewed-by: Michael Pratt <mpratt@google.com> Auto-Submit: Michael Pratt <mpratt@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
2026-03-24cmd/link: raise minimum windows version to 10Jason A. Donenfeld
The minimum Windows version has been 10 for a few releases, but the PE headers weren't updated. Windows sometimes can use these in determining what kind of subsystem compatibility hacks to apply, which of course we don't want now, since Go targets Windows 10. This also causes older OSes to refuse to run the executables, rather than having them crash in some undefined way. This isn't trivial to do, because subsystem ≥ 10.0 means that the Windows loader expects to see either _load_config_used.SecurityCookie set to the initial magic value, or for IMAGE_GUARD_SECURITY_COOKIE_UNUSED to be set. Go obviously isn't making use of these features, and neither does clang/gcc for that matter; libssp doesn't even use SecurityCookie. Rather, it's exclusively for MSVC's /GS protection. So it seems like the proper thing to do is signal to the OS that it doesn't need to initialize SecurityCookie. This check lives in ntdll!LdrInitSecurityCookie. So, add the _load_config_used structure to the right PE section and give it the right flag. This lets the Windows 10-marked binaries actually run. Change-Id: I91887073c7ad01aeb0237906aafa4ea5574ac8fa Reviewed-on: https://go-review.googlesource.com/c/go/+/756680 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Auto-Submit: Jason Donenfeld <Jason@zx2c4.com> Reviewed-by: Jason Donenfeld <Jason@zx2c4.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Quim Muntal <quimmuntal@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-24runtime: return the error code as a return value in syscallN_trampolineqmuntal
Getting errno from assembly code is cheap. There is no need to overcomplicate syscall_rawsyscalln to get errno from the cached errno address pointer stored in the M struct. This also better aligns syscallN_trampoline with the cgocall convention of returning the error code as a return value. Cq-Include-Trybots: luci.golang.try:gotip-darwin-amd64_15,gotip-darwin-arm64_26 Change-Id: I05d5177e7c1471942e8ecafb4fb05594b4b18e2b Reviewed-on: https://go-review.googlesource.com/c/go/+/753540 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Pratt <mpratt@google.com>
2026-03-23runtime/race: apply LLVM zero-initialization fixMichael Pratt
Upstream TSAN had bug that could result in use of uninitialized memory on Go threads that don't have any TSAN events. For example, if the thread only ever runs the GC. This bug was fixed upstream in https://github.com/llvm/llvm-project/commit/cdfdb06c9155080ec97d6e4f4dd90b6e7cefb0ca. In https://go.dev/issue/78059 we have reports of actual Go crashes due to this bug. Update the prebuilt race sysos to incorporate this fix. The fix is applied as a single patch on top of the existing LLVM revisions to minimize risk of this CL, making it safe to backport. A later CL can update to a newer version of LLVM. Note that all of the patch files are identical. CL 756620 makes racebuild add a unique patch file for each architecture in the event that some arches need distinct patches. linux-loong64 failed race.bash when building the new syso, though they were just timeouts, perhaps from a slow builder. linux-riscv64 is not updated because its builder is too slow (https://go.dev/issue/78258). linux-ppc64le is not updated because its builder is missing curl (https://go.dev/issue/78210). openbsd-amd64 is not updated because its builder is missing unzip (https://go.dev/issue/78212). netbsd-amd64 is not updated because it does not have a LUCI builder (https://go.dev/issue/61121). Fixes #78059. Cq-Include-Trybots: luci.golang.try:gotip-linux-amd64-longtest-race,gotip-darwin-arm64-race,gotip-darwin-amd64-race,gotip-freebsd-amd64-race,gotip-windows-amd64-race,gotip-linux-s390x-race,gotip-linux-arm64-race,gotip-linux-loong64 Change-Id: I5404cb88af9d86b56b385801f8a9ed106a6a6964 Reviewed-on: https://go-review.googlesource.com/c/go/+/757521 Reviewed-by: Carlos Amedee <carlos@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> TryBot-Bypass: Michael Pratt <mpratt@google.com>
2026-03-20Revert "runtime, cmd/compile: use preemptible memclr for large pointer-free ↵Michael Pratt
clears" This reverts CL 750480. Reason: Adding preemptible memclrNoHeapPointers exposes existing unsafe use of notInHeapSlice, causing crashes. Revert the memclr stack until the underlying issue is fixed. We keep the test added in CL 755942, which is useful regardless. For #78254. Change-Id: I8be3f9a20292b7f294e98e74e5a86c6a204406ae Reviewed-on: https://go-review.googlesource.com/c/go/+/757343 Reviewed-by: Keith Randall <khr@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Keith Randall <khr@golang.org>
2026-03-20Revert "runtime: fix memclrNoHeapPointersPreemptible"Michael Pratt
This reverts CL 756122. Reason: Adding preemptible memclrNoHeapPointers exposes existing unsafe use of notInHeapSlice, causing crashes. Revert the memclr stack until the underlying issue is fixed. For #78254. Change-Id: I8a234d1a6dddf70d9aa5ecda1ac8bb25deb08248 Reviewed-on: https://go-review.googlesource.com/c/go/+/757342 Reviewed-by: Keith Randall <khr@golang.org> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Keith Randall <khr@google.com>
2026-03-18syscall: remove the 42 args limit in SyscallNqmuntal
SyscallN already supports more than 42 arguments on 386 and arm64. Lift that limit on amd64 as well. Change-Id: I3f5c6ebb959aa9b0d367903cac119fc27cb93064 Reviewed-on: https://go-review.googlesource.com/c/go/+/751862 Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Alex Brainman <alex.brainman@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Carlos Amedee <carlos@golang.org>
2026-03-18runtime: return the error code as a return value in asmstdcallqmuntal
This shaves off 8 bytes from the syscall_syscalln stack frame, which is significant as that call path is almost over the nosplit limit. Also, it follows the cgocall convention of returning the error code as a return value, making it easier to reason about. Cq-Include-Trybots: luci.golang.try:gotip-windows-amd64-longtest,gotip-windows-arm64,gotip-windows-386 Change-Id: I62acdb7c0d4cf9cb928bf3974d3300dd752f6c29 Reviewed-on: https://go-review.googlesource.com/c/go/+/751861 Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Carlos Amedee <carlos@golang.org> Reviewed-by: Alex Brainman <alex.brainman@gmail.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-18runtime,runtime/cgo: do cgo thread initialization in Go on Windowsqmuntal
Windows doesn't require any special handling for cgo threads. They can be created in the same way as in non-cgo code. In fact, the code to create threads in runtime and in runtime/cgo is basically the same, except that the latter does some retries on failure. Cq-Include-Trybots: luci.golang.try:gotip-windows-amd64-longtest,gotip-windows-amd64-race,gotip-windows-arm64 Change-Id: I49d4de93d4d3b07a4c89e2bfb6b7302c6dfb9877 Reviewed-on: https://go-review.googlesource.com/c/go/+/746300 Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Michael Pratt <mpratt@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-18runtime: implement part of library initialization in Goqmuntal
All architectures supporting c-shared and c-archive share the same initialization code in assembly, and most of it can be implemented in pure Go. Cq-Include-Trybots: luci.golang.try:gotip-darwin-arm64-longtest,gotip-linux-ppc64le_power10,gotip-linux-riscv64,gotip-linux-loong64,gotip-linux-s390x Change-Id: Iaa9fb7d6f9ca8785f1098461646d607ef6b00d47 Reviewed-on: https://go-review.googlesource.com/c/go/+/706417 Auto-Submit: Quim Muntal <quimmuntal@gmail.com> Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Carlos Amedee <carlos@golang.org>
2026-03-18runtime: make asmcgocall more robust to missing Gqmuntal
Being able to call asmcgocall without a G is useful for code shared between different stages of the runtime initialization and thread creation. Cq-Include-Trybots: luci.golang.try:gotip-darwin-arm64_15,gotip-linux-mips64le,gotip-linux-ppc64le_power10,gotip-linux-riscv64,gotip-openbsd-ppc64,gotip-openbsd-amd64 Change-Id: Ic427764de197e648e8b9987c98c3b7521512cc5c Reviewed-on: https://go-review.googlesource.com/c/go/+/750541 Reviewed-by: Carlos Amedee <carlos@golang.org> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
2026-03-17runtime: fix memclrNoHeapPointersPreemptibleKeith Randall
memclrNoHeapPointersChunked was originally written for clearing fresh allocations. It converts to uintptr early and thus doesn't handle the case where the pointer points to the stack. At the preemption point, the buffer being pointed to might be on the stack and copied to a new location. Then we'd end up zeroing random heap memory when we return from preemption. Fix it to keep the pointer to the allocation in an unsafe.Pointer, so it gets updated correctly on a stack copy. Update #78081 (Fixes?) Change-Id: Id25c7faf45f201929cb8c8bbcc7c8bd4f67c31ac Reviewed-on: https://go-review.googlesource.com/c/go/+/756122 Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: Michael Pratt <mpratt@google.com> Reviewed-by: Keith Randall <khr@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-11runtime/cgo: clarify pinning requirement in Handle examplekovan
Add a comment to the void* example in the Handle documentation highlighting that if the C code keeps the pointer after the call returns, runtime.Pinner should be used to pin it. The existing documentation already mentions this requirement in prose (lines 69-73), but adding a comment directly in the example makes it more visible and less likely to be missed when copying the example. Fixes #68044 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> Change-Id: I20a5b1ad10de1b441980dc7ed3185720e77912bb Reviewed-on: https://go-review.googlesource.com/c/go/+/741443 Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-10syscall,runtime/cgo: use cgo_ldflag to link against -lsocket and -lxnet on ↵qmuntal
Solaris The -lsocket and -lxnet flags are needed for accept4 syscall on Solaris. The runtime/cgo package doens't use them, so it doesn't make sense for it to explicitly link against those libraries. Instead, use the //go:cgo_ldflag in the syscall package. Cq-Include-Trybots: luci.golang.try:gotip-solaris-amd64 Change-Id: I10db524ebf1c720a460515d8c1f362b0070bd771 Reviewed-on: https://go-review.googlesource.com/c/go/+/751760 Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Mark Freeman <markfreeman@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Ian Lance Taylor <iant@golang.org> Auto-Submit: Quim Muntal <quimmuntal@gmail.com>
2026-03-10runtime: GODEBUG=tracebacklabels=1 for 1.27+David Finkel
https://go.dev/cl/742580 updated the goroutine label format in tracebacks to only quote the values under some circumstances, and use a more broadly acceptable format. Flip the GODEBUG=tracebacklabels default for go 1.27+ modules. Notably, tracebacklabels is marked as Opaque to avoid some diciness in maintaining a usage metric in signal handlers, etc. (where mutex acquisition may be problematic) Also, update the go 1.27 release notes to call out the new feature. Updates #76349 Change-Id: Iedd3f6b17d83e99da20365b952a565022d20ff13 Reviewed-on: https://go-review.googlesource.com/c/go/+/751580 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Alan Donovan <adonovan@google.com> Reviewed-by: Alan Donovan <adonovan@google.com> Reviewed-by: Mark Freeman <markfreeman@google.com>
2026-03-06runtime: when panicking, skip ahead to previous panicKeith Randall
While looking up the stack for a defer to run, if we come across a panic frame we can skip ahead (up) to where the previous panic was looking for a defer to run. Switch from keeping LR (the caller's pc) to PC (the frame's PC). Seems easier to reason about. Fixes #77062 Change-Id: Idb39411ebad8c072c8f65c62a518da848bddbd61 Reviewed-on: https://go-review.googlesource.com/c/go/+/738041 Reviewed-by: Keith Randall <khr@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
2026-03-05html/template: properly escape URLs in meta content attributesRoland Shoemaker
The meta tag can include a content attribute that contains URLs, which we currently don't escape if they are inserted via a template action. This can plausibly lead to XSS vulnerabilities if untrusted data is inserted there, the http-equiv attribute is set to "refresh", and the content attribute contains an action like `url={{.}}`. Track whether we are inside of a meta element, if we are inside of a content attribute, _and_ if the content attribute contains "url=". If all of those are true, then we will apply the same URL escaping that we use elsewhere. Also add a new GODEBUG, htmlmetacontenturlescape, to allow disabling this escaping for cases where this behavior is considered safe. The behavior can be disabled by setting htmlmetacontenturlescape=0. Fixes CVE-2026-27142 Fixes #77954 Change-Id: I9bbca263be9894688e6ef1e9a8f8d2f4304f5873 Reviewed-on: https://go-internal-review.googlesource.com/c/go/+/3360 Reviewed-by: Neal Patel <nealpatel@google.com> Reviewed-by: Nicholas Husin <husin@google.com> Reviewed-on: https://go-review.googlesource.com/c/go/+/752181 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-03-04runtime, cmd/compile: use preemptible memclr for large pointer-free clears“Muhammad
Large memory clearing operations (via clear() or large slice allocation) currently use non-preemptible assembly loops. This blocks the Garbage Collector from performing a Stop The World (STW) event, leading to significant tail latency or even indefinite hangs in tight loops. This change introduces memclrNoHeapPointersPreemptible, which chunks clears into 256KB blocks with preemption checks. The compiler's walk phase is updated to emit this call for large pointer-free clears. To prevent regressions, SSA rewrite rules are added to ensure that constant-size clears (which are common and small) continue to be inlined into OpZero assembly. Benchmarks on darwin/arm64: - STW with 50MB clear: Improved from 'Hung' to ~500µs max pause. - Small clears (5-64B): No measurable regression. - Large clears (1M-64M): No measurable regression. Fixes #69327 Change-Id: Ide14d6bcdca1f60d6ac95443acb57da9a8822538 Reviewed-on: https://go-review.googlesource.com/c/go/+/750480 Reviewed-by: Keith Randall <khr@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Keith Randall <khr@golang.org> Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: Robert Griesemer <gri@google.com>
2026-03-04runtime: fix value of ENOSYS on mips from 38 to 89Jorropo
Fixes #77730 Change-Id: I8b68b2d4353344e70f970d48a2d50c1a3bd7f273 Reviewed-on: https://go-review.googlesource.com/c/go/+/747664 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Rongrong <rongrong-for-oss@oss.cipunited.com> Reviewed-by: Michael Pratt <mpratt@google.com>
2026-03-02runtime: fix printfloat, printcomplex buffer sizesMichael Pratt
The buffers added in CL 716002 for printfloat64 and printcomplex128 are too small to fit the longest formatted values. For values that are too long, AppendFloat allocates, which may cause a crash for prints in places in the runtime where allocation is not allowed. Fixes #77854. Change-Id: I6a6a636cc2fc5cae9fda25f10b28fd641aa1ff28 Reviewed-on: https://go-review.googlesource.com/c/go/+/749947 Reviewed-by: Russ Cox <rsc@golang.org> Auto-Submit: Michael Pratt <mpratt@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>
2026-03-02runtime: skip futex_time64 and timer_settime64 on 32-bit Androiddyhkwong
In Android versions 8.0-10 (API levels 26-29), futex_time64 and timer_settime64 are not in the allowlist of the seccomp filter and will lead to a runtime crash. Fixes #77621 Change-Id: I99607d7128c395edf93738440c2928b6819d8a21 Reviewed-on: https://go-review.googlesource.com/c/go/+/750040 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: David Chase <drchase@google.com> Auto-Submit: Jorropo <jorropo.pgm@gmail.com> Reviewed-by: Jorropo <jorropo.pgm@gmail.com> Reviewed-by: Michael Pratt <mpratt@google.com>
2026-02-28runtime/cgo: build with C compiler option -std=gnu90Ian Lance Taylor
This will ensure we remain portable. We use gnu90 rather than c90 because c90 doesn't permit C++ style line comments, and that is just too painful. Change-Id: Ia0e52c4aa75493ceb3e8ce383cdd2ec85afa0bd2 Reviewed-on: https://go-review.googlesource.com/c/go/+/653138 Reviewed-by: Cherry Mui <cherryyz@google.com> Auto-Submit: Ian Lance Taylor <iant@golang.org> Reviewed-by: Junyang Shao <shaojunyang@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
2026-02-28test: skip TestScanAllocIssue77573 with asan modeWang Deyu
In ASAN mode, extra memory allocations alter the object size and impact the scansize computation. Skip this test when ASAN is enabled. Fixes #77857. Fixes #77858. Fixes #77859. Change-Id: I21027ff20c411a0fda7a50446b2202871baa2262 Reviewed-on: https://go-review.googlesource.com/c/go/+/750140 Reviewed-by: Keith Randall <khr@golang.org> Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: David Chase <drchase@google.com> Reviewed-by: Keith Randall <khr@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
generated by cgit v1.3 (git 2.53.0) at 2026-04-15 09:09:26 +0000