Michael Pratt [Tue, 23 Jan 2024 18:29:23 +0000 (13:29 -0500)]
runtime: use channels in gcBgMarkStartWorkers
gcBgMarkStartWorkers currently starts workers one at a time, using a
note to communicate readiness back from the worker.
However, this is a pretty standard goroutine, so we can just use a
channel to communicate between the goroutines.
In addition to being conceptually simpler, using channels has the
additional advantage of coordinating with the scheduler. Notes use OS
locks and sleep the entire thread, requiring other threads to run the
other goroutines. Waiting on a channel allows the scheduler to directly
run another goroutine. When the worker sends to the channel, the
scheduler can use runnext to run gcBgMarkStartWorker immediately,
reducing latency.
We could additionally batch start all workers and then wait only once,
however this would defeate runnext switching between the workers and
gcBgMarkStartWorkers, so in a heavily loaded system, we expect the
direct switches to reduce latency.
Change-Id: Iedf0d2ad8ad796b43fd8d32ccb1e815cfe010cb4
Reviewed-on: https://go-review.googlesource.com/c/go/+/558535
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>
Robert Griesemer [Wed, 24 Jan 2024 20:42:11 +0000 (12:42 -0800)]
go/types, types2: use existing case-insensitive lookup (remove TODO)
Rather than implementing a new, less complete mechanism to check
if a selector exists with different capitalization, use the
existing mechanism in lookupFieldOrMethodImpl by making it
available for internal use.
Pass foldCase parameter all the way trough to Object.sameId and
thus make it consistently available where Object.sameId is used.
From sameId, factor out samePkg functionality into stand-alone
predicate.
Do better case distinction when reporting an error for an undefined
selector expression.
Cleanup.
Change-Id: I7be3cecb4976a4dce3264c7e0c49a320c87101e9
Reviewed-on: https://go-review.googlesource.com/c/go/+/558315 Reviewed-by: Robert Griesemer <gri@google.com> Reviewed-by: Robert Findley <rfindley@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Run-TryBot: Robert Griesemer <gri@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Russ Cox [Tue, 23 Jan 2024 17:02:50 +0000 (12:02 -0500)]
all: enable range-over-func in Go 1.23
GOEXPERIMENT=rangefunc still enables it for all Go modules.
Otherwise only enable in Go 1.23 source files.
More work remains but it will be done in follow-up issues.
Fixes #61405.
Change-Id: Icad64942deb152ee65444e4d7be289814a8a0b6b
Reviewed-on: https://go-review.googlesource.com/c/go/+/557835 Reviewed-by: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Robert Findley <rfindley@google.com>
Russ Cox [Tue, 23 Jan 2024 19:42:42 +0000 (14:42 -0500)]
maps: correct use of linkname
Linkname should have uses at both ends,
and then no fake assembly file is needed.
Change-Id: Ia2066be7b1e08d71a6db622d2a167f1b09cb7d9f
Reviewed-on: https://go-review.googlesource.com/c/go/+/558297 Reviewed-by: Mauri de Souza Meneguzzo <mauri870@gmail.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Meng Zhuo [Tue, 1 Aug 2023 14:17:02 +0000 (22:17 +0800)]
cmd/compile: implement float min/max in hardware for riscv64
CL 514596 adds float min/max for amd64, this CL adds it for riscv64.
The behavior of the RISC-V FMIN/FMAX instructions almost match Go's
requirements.
However according to RISCV spec 8.3 "NaN Generation and Propagation"
>> if at least one input is a signaling NaN, or if both inputs are quiet
>> NaNs, the result is the canonical NaN. If one operand is a quiet NaN
>> and the other is not a NaN, the result is the non-NaN operand.
Go using quiet NaN as NaN and according to Go spec
>> if any argument is a NaN, the result is a NaN
This requires the float min/max implementation to check whether one
of operand is qNaN before float mix/max actually execute.
Change-Id: Iab48be6d32b8882044fb8c821438ca8840e5493d
Reviewed-on: https://go-review.googlesource.com/c/go/+/514775 Reviewed-by: Mauri de Souza Meneguzzo <mauri870@gmail.com>
Run-TryBot: M Zhuo <mengzhuo1203@gmail.com> Reviewed-by: Joel Sing <joel@sing.id.au> Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Keith Randall <khr@google.com>
Robert Griesemer [Wed, 24 Jan 2024 21:09:55 +0000 (13:09 -0800)]
go/types, types2: in struct processing, set name position in types2
As a consequence, the positions needed by the Checker.structType
internal helper functions add and addInvalid are always the positions
of the provided identifiers, and we can leave away the extra position
arguments.
Change-Id: Iddc275c83d3781261476b8e1903050e0a049957c
Reviewed-on: https://go-review.googlesource.com/c/go/+/558316
Run-TryBot: Robert Griesemer <gri@google.com> Reviewed-by: Robert Findley <rfindley@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@google.com>
Michael Anthony Knyszek [Thu, 25 Jan 2024 17:23:15 +0000 (17:23 +0000)]
internal/testenv: support the LUCI mobile builders in tests
This change updates the testenv tests to correctly match on future LUCI
builder names for mobile builders. This isn't a problem today because
those haven't been set up yet, but the builder names are structured and
it's clear where the modifiers will appear. Might as well set them up
now.
Change-Id: I244b88a62a90312c0f3ff2360527d58531070362
Reviewed-on: https://go-review.googlesource.com/c/go/+/558597 Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Michael Anthony Knyszek [Thu, 25 Jan 2024 17:13:11 +0000 (17:13 +0000)]
internal/testenv: allow "-noopt" anywhere in builder name in test
testenv's TestHasGoBuild test is supposed to allow noopt builders to not
have go build, but the pattern match is failing on the LUCI builders
where a test shard might have an additional "-test_only" suffix in the
builder name. Furthermore, in the LUCI world, "run mods" (the builder
type suffixes) are supposed to be well-defined and composable, so it
doesn't make sense to restrict "-noopt" to the builder suffix anyway.
This change modifies the test to allow "-noopt" to appear anywhere in
the builder name when checking if it's running on a noopt builder.
Change-Id: I393818e3e8e452c7b0927cbc65726d552aa8ff8e
Reviewed-on: https://go-review.googlesource.com/c/go/+/558596
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com>
Cosmos Nicolaou [Thu, 25 Jan 2024 19:55:38 +0000 (11:55 -0800)]
runtime/pprof: fix retry logic introduced by #858cd8d
The previous attempt to fix this flake was incorrect in that it
examined the vmmap output rather than the detailed error
output for the 'resource shortage' message that triggers the
retry, and hence failed to retry. This PR looks at the
detailed error output.
Fixes #62352
Change-Id: I4218b187528a95842556dc1ea27947ffcbfbc497
Reviewed-on: https://go-review.googlesource.com/c/go/+/558575
Auto-Submit: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Michael Pratt <mpratt@google.com>
Joel Sing [Sat, 17 Sep 2022 16:37:36 +0000 (02:37 +1000)]
cmd/dist,internal/platform,runtime: add cgo support for openbsd/riscv64 port
Updates #55999
Change-Id: Ie9a030130713f3eaf789b91d9033be3335eb9d75
Reviewed-on: https://go-review.googlesource.com/c/go/+/518630 Reviewed-by: Carlos Amedee <carlos@golang.org>
Run-TryBot: Joel Sing <joel@sing.id.au>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
Joel Sing [Fri, 4 Aug 2023 10:40:23 +0000 (20:40 +1000)]
cmd/dist,internal/platform: enable openbsd/riscv64 port
Fixes #55999
Change-Id: If76b61b660db3fefe5f9ca97500782f2a650db25
Reviewed-on: https://go-review.googlesource.com/c/go/+/518629 Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Joel Sing <joel@sing.id.au>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Carlos Amedee <carlos@golang.org>
Mauri de Souza Meneguzzo [Sat, 2 Dec 2023 15:04:52 +0000 (15:04 +0000)]
runtime: add crash stack support for s390x
Change-Id: Ie923f7bbe5ef22e381ae4f421387fbd570622a28
GitHub-Last-Rev: f8f21635025eb6e26c6994679995ade501e870cf
GitHub-Pull-Request: golang/go#63908
Reviewed-on: https://go-review.googlesource.com/c/go/+/539296
Run-TryBot: Joel Sing <joel@sing.id.au>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Carlos Amedee <carlos@golang.org>
Run-TryBot: Mauri de Souza Meneguzzo <mauri870@gmail.com> Reviewed-by: Joel Sing <joel@sing.id.au>
Andy Pan [Fri, 22 Dec 2023 13:49:46 +0000 (21:49 +0800)]
net: verify if internal/poll.Splice has been called during io.Copy
Change-Id: I29ef35b034cd4ec401ac502bf95dbd8c3ef2a2d4
Reviewed-on: https://go-review.googlesource.com/c/go/+/552535
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Damien Neil <dneil@google.com>
Robert Griesemer [Tue, 23 Jan 2024 21:21:19 +0000 (13:21 -0800)]
go/types, types2: better error msg when type inference fails in a call
In Checker.infer, report an error through an (incoming) *error_
so that the error can be reported as desired where infer is called.
Checker.infer is now a pure function.
Fixes #60543.
At call sites of Checker.infer, pass in an *error_ and use it to
report inference errors, together with additional information as
desired.
Fixes #60542.
In go/types, in error_.errorf, pass in a positioner rather than
a token.Pos. Also, introduce noposn, the positioner equivalent
for nopos. Adjust call sites as needed.
Change-Id: I462a7899a77a8bee2a21ba88299df237d74e0672
Reviewed-on: https://go-review.googlesource.com/c/go/+/558035 Reviewed-by: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Robert Griesemer <gri@google.com> Reviewed-by: Robert Findley <rfindley@google.com>
Cherry Mui [Wed, 24 Jan 2024 20:00:47 +0000 (15:00 -0500)]
Revert "archive/tar: add FileInfoNames interface"
This reverts CL 514235. Also reverts CL 518056 which is a followup
fix.
Reason for revert: Proposal #50102 defined an interface that is
too specific to UNIX-y systems and also didn't make much sense.
The proposal is un-accepted, and we'll revisit in Go 1.23.
Fixes (via backport) #65245.
Updates #50102.
Change-Id: I41ba0ee286c1d893e6564a337e5d76418d19435d
Reviewed-on: https://go-review.googlesource.com/c/go/+/558295 Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Bryan C. Mills [Tue, 23 Jan 2024 22:08:50 +0000 (17:08 -0500)]
net: ignore Dial errors in TestAcceptTimeout
Also use DialContext instead of just Dial so that we can ensure
the call returns before we close the listener.
The Dial in this test is intended to complete before the call to
Accept, but there is no synchronization to ensure that and sometimes
it doesn't happen. That's ok and mostly immaterial to the test, but it
does mean we need to ignore Dial errors (which can happen when the
listener is closed), and we need to be a little more careful about not
dialing a port that may have already been reused by some other test.
Jonathan Amsterdam [Mon, 22 Jan 2024 18:20:07 +0000 (13:20 -0500)]
doc/next: initial contents
Initialize the doc/next directory for the next release by copying the
contents of doc/initial into it.
Also, rewrite doc/README.md to add release instructions and to separate
information for developers from information for the release team.
Change-Id: Ie1e5c2ddae7a8d301e9b2f8a68925519648aa84d
Reviewed-on: https://go-review.googlesource.com/c/go/+/557535
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Jonathan Amsterdam <jba@google.com>
Mauri de Souza Meneguzzo [Tue, 28 Nov 2023 18:42:43 +0000 (18:42 +0000)]
runtime: short path for equal pointers in arm64 memequal
If memequal is invoked with the same pointers as arguments it ends up
comparing the whole memory contents, instead of just comparing the pointers.
This effectively makes an operation that could be O(1) into O(n). All the
other architectures already have this optimization in place. For
instance, arm64 also have it, in memequal_varlen.
Such optimization is very specific, one case that it will probably benefit is
programs that rely heavily on interning of strings.
Mark Ryan [Tue, 7 Nov 2023 09:09:49 +0000 (10:09 +0100)]
cmd/go: add GORISCV64 environment variable
The variable represents the RISC-V user-mode application profile for
which to compile. Valid values are rva20u64 (the default) and
rva22u64.
Setting GORISCV64=rva20u64 defines the riscv64.rva20u64 build tag,
sets the internal variable buildcfg.GORISCV64 to 20 and defines the
macro GORISCV64_rva20u64 for use in assembly language code.
Setting GORISCV64=rva22u64 defines the riscv64.rva20u64 and
riscv64.rva22u64 build tags, sets the internal variable
buildcfg.GORISCV64 to 22 and defines the macro GORISCV64_rva22u64
for use in assembly language code.
This patch only provides a mechanism for the compiler and hand-coded
assembly language functions to take advantage of the RISC-V
extensions mandated by the application profiles. Further patches
will be required to get the compiler/assembler and assembly language
functions to actually generate and use these extensions.
Fixes #61476
Change-Id: I9195ae6ee71703cd2112160e89157ab63b8391af
Reviewed-on: https://go-review.googlesource.com/c/go/+/541135 Reviewed-by: M Zhuo <mengzhuo1203@gmail.com> Reviewed-by: Joel Sing <joel@sing.id.au>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Wang Yaduo <wangyaduo@linux.alibaba.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Bryan Mills <bcmills@google.com>
Run-TryBot: M Zhuo <mengzhuo1203@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Russ Cox [Tue, 12 Dec 2023 17:53:32 +0000 (12:53 -0500)]
net: move cgo-using darwin test to separate package
It is a goal / requirement that nothing in the standard library
on darwin require cgo, and this test-only file makes net use
cgo on darwin. Move it elsewhere.
Change-Id: I6c11a8391d3913f73ce0098ba63b29adf5864f93
Reviewed-on: https://go-review.googlesource.com/c/go/+/549255 Reviewed-by: Mauri de Souza Meneguzzo <mauri870@gmail.com> Reviewed-by: Bryan Mills <bcmills@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Russ Cox <rsc@golang.org>
Bryan C. Mills [Mon, 11 Dec 2023 21:45:04 +0000 (16:45 -0500)]
net: unskip and attempt to deflake TestAcceptTimeout
This test has been unconditionally skipped for over five years.
It may be that whatever was causing it to flake has been fixed.
And if it hasn't been fixed, it isn't providing any value.
Let's unskip it for the Go 1.23 development cycle and see what happens.
Let's also use a separate listener for each test case, so that a
leaked Dial goroutine from one case won't interfere with the other.
Fixes #17948 (maybe).
Change-Id: I239f22ca5d5a44388b9aa0ed4d81e451c6342617
Reviewed-on: https://go-review.googlesource.com/c/go/+/548940
Commit-Queue: Bryan Mills <bcmills@google.com>
Auto-Submit: Bryan Mills <bcmills@google.com> Reviewed-by: Damien Neil <dneil@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
qiulaidongfeng [Tue, 23 Jan 2024 00:42:58 +0000 (00:42 +0000)]
reflect: delete TODO pass safe to packEface don't need to copy if safe==true
valueInterface not copy result in the follow incorrect behavior
w1. x := ValueOf(&v).Elem()
r1. iface := Value.Interface()
w2. x.Set() or x.SetT()
The write operation of W2 will be observed by the read operation of r1,
but the existing behavior is not.
The valueInterface in deepValueEqual can, in theory, pass safe==true to not copy the object,
but there is no benchmark to indicate that the memory allocation has changed,
maybe we don't actually need safe==true here.
Change-Id: I55c423fd50adac8822a7fdbfe67af89ee223eace
GitHub-Last-Rev: 4a6386709817f3ea6055711dd39d2694d58b3043
GitHub-Pull-Request: golang/go#64618
Reviewed-on: https://go-review.googlesource.com/c/go/+/548436
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Keith Randall <khr@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Auto-Submit: Keith Randall <khr@golang.org>
Bryan C. Mills [Mon, 22 Jan 2024 23:02:38 +0000 (18:02 -0500)]
net: report timeouts more aggressively in Accept in the fake implementation
This ensures that if the listener has already timed out when Accept
is called, Accept always returns an error instead of instantaneously
accepting a connection that was already pending.
For #17948.
Change-Id: Iabef7121590df3dcc2fe428429d7c2bc2bcb6cd5
Reviewed-on: https://go-review.googlesource.com/c/go/+/557438
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Bryan Mills <bcmills@google.com> Reviewed-by: Damien Neil <dneil@google.com>
qmuntal [Fri, 19 Jan 2024 13:14:17 +0000 (14:14 +0100)]
runtime,internal/poll: move websocket handling out of the runtime on Windows
On Windows, the netpoll is currently coupled with the websocket usage
in the internal/poll package.
This CL moves the websocket handling out of the runtime and puts it into
the internal/poll package, which already contains most of the async I/O
logic for websockets.
This is a good refactor per se, as the Go runtime shouldn't know about
websockets. In addition, it will make it easier (in a future CL) to only
load ws2_32.dll when the Go program actually uses websockets.
Change-Id: Ic820872cf9bdbbf092505ed7f7504edb6687735e
Reviewed-on: https://go-review.googlesource.com/c/go/+/556936
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Olivier Mengué [Mon, 15 Jan 2024 23:32:40 +0000 (00:32 +0100)]
hash/crc32: fix codegen to align with manual changes
Update gen_const_ppc64le.go to match the manual changes applied in
CL 478976.
Change-Id: I79a0d014a2a151750f898517b2771b312f3437bc
Reviewed-on: https://go-review.googlesource.com/c/go/+/555996 Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Auto-Submit: Michael Pratt <mpratt@google.com>
Jorropo [Wed, 10 Jan 2024 06:31:57 +0000 (07:31 +0100)]
cmd/compile: improve integer comparisons with numeric bounds
This do:
- Fold always false or always true comparisons for ints and uint.
- Reduce < and <= where the true set is only one value to == with such value.
Change-Id: Ie9e3f70efd1845bef62db56543f051a50ad2532e
Reviewed-on: https://go-review.googlesource.com/c/go/+/555135
Auto-Submit: Keith Randall <khr@golang.org> Reviewed-by: Keith Randall <khr@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Keith Randall <khr@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Change-Id: I92629a106b75b0526e9f2a8fe3cc4a6f7fc63c8c
Reviewed-on: https://go-review.googlesource.com/c/go/+/518631
Auto-Submit: Dmitri Shuralyov <dmitshur@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Joel Sing <joel@sing.id.au> Reviewed-by: M Zhuo <mzh@golangcn.org> Reviewed-by: Wang Yaduo <wangyaduo@linux.alibaba.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Mark Ryan <markdryan@rivosinc.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Michael Anthony Knyszek [Mon, 22 Jan 2024 16:34:41 +0000 (16:34 +0000)]
runtime: use the correct M ID for syscalling goroutines in traces
Earlier in the development of the new tracer, m.id was used as a the
canonical ID for threads. Later, we switched to m.procid because it
matches the underlying OS resource. However, in that switch, we missed a
spot.
The tracer catches and emits statuses for goroutines that have remained
in either waiting or syscall across a whole generation, and emits a
thread ID for the latter set. The ID being used here, however, was m.id
instead of m.procid, like the rest of the tracer.
This CL also adds a regression test. In order to make the regression
test actually catch the failure, we also have to make the parser a
little less lenient about GoStatus events with GoSyscall: if this isn't
the first generation, then we should've seen the goroutine bound to an
M already when its status is getting emitted for its context. If we emit
the wrong ID, then we'll catch the issue when we emit the right ID when
the goroutine exits the syscall.
Fixes #65196.
Change-Id: I78b64fbea65308de5e1291c478a082a732a8bf9f
Reviewed-on: https://go-review.googlesource.com/c/go/+/557456 Reviewed-by: Michael Pratt <mpratt@google.com>
Auto-Submit: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Jin Lin [Wed, 20 Sep 2023 19:22:26 +0000 (12:22 -0700)]
cmd/preprofile: Add preprocess tool to pre-parse the profile file.
The pgo compilation time is very long if the profile file is large.
We added a preprocess tool to pre-parse profile file in order to
expedite the compile time.
Change-Id: I6f50bbd01f242448e2463607a9b63483c6ca9a12
Reviewed-on: https://go-review.googlesource.com/c/go/+/529738 Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Bryan C. Mills [Mon, 22 Jan 2024 18:54:56 +0000 (13:54 -0500)]
net: delete TestListenCloseListen
In CL 557177, I attempted to fix a logical race in this test (#65175).
However, I introduced a data race in the process (#65209).
The race was reported on the windows-amd64-race builder. When I tried
to reproduce it on linux/amd64, I added a time.Sleep in the Accept
loop. However, that Sleep causes the test to fail outright with
EADDRINUSE, which suggests that my earlier guess about the open Conn
preventing reuse of the port was, in fact, incorrect.
On some platforms we could instead use SO_REUSEPORT and avoid closing
the first Listener entirely, but that wouldn't be even remotely in the
spirit of the original test.
Since I don't see a way to preserve the test in a way that is not
inherently flaky / racy, I suggest that we just delete it. It was
originally added as a regression test for a bug in the nacl port,
which no longer exists anyway. (Some of that code may live on in the
wasm port, but it doesn't seem worth maintaining a flaky
port-independent test to maintain a regression test for a bug specific
to secondary platforms.)
Roland Shoemaker [Wed, 10 Jan 2024 19:02:14 +0000 (11:02 -0800)]
net/mail: properly handle special characters in phrase and obs-phrase
Fixes a couple of misalignments with RFC 5322 which introduce
significant diffs between (mostly) conformant parsers.
This change reverts the changes made in CL50911, which allowed certain
special RFC 5322 characters to appear unquoted in the "phrase" syntax.
It is unclear why this change was made in the first place, and created
a divergence from comformant parsers. In particular this resulted in
treating comments in display names incorrectly.
Additionally properly handle trailing malformed comments in the group
syntax.
Fixes #65083
Change-Id: I00dddc044c6ae3381154e43236632604c390f672
Reviewed-on: https://go-review.googlesource.com/c/go/+/555596 Reviewed-by: Damien Neil <dneil@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Austin Clements [Wed, 3 Jan 2024 15:41:09 +0000 (10:41 -0500)]
runtime: simplify freedefer logic
Currently, freedefer's API forces a subtle and fragile situation. It
requires that the caller unlink the _defer from the G list, but
freedefer itself is responsible for zeroing some _defer fields. In the
window between these two steps, we have to prevent stack growth
because stack growth walks the defer list (which no longer contains
the unlinked defer) to adjust pointers, and would thus leave an
unadjusted and potentially invalid pointer behind in the _defer before
freedefer zeroes it.
This setup puts part of this subtle responsibility on the caller and
also means freedefer must be nosplit, which forces other shenanigans
to avoid nosplit overflows.
We can simplify all of this by replacing freedefer with a new popDefer
function that's responsible for both unlinking and zeroing the _defer,
in addition to freeing it.
Some history: prior to regabi, defer records contained their argument
frame, which deferreturn copied to the stack before freeing the defer
record (and subsequently running the defer). Since that argument frame
didn't have a valid stack map until we ran the deferred function, the
non-preemptible window was much larger and more difficult to isolate.
Now we use normal closure calls to capture defer state and call the
defer, so the non-preemptible window is narrowed to just the unlinking
step.
Change-Id: I7cf95ba18e1e2e7d73f616b9ed9fb38f5e725d72
Reviewed-on: https://go-review.googlesource.com/c/go/+/553696 Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Auto-Submit: Austin Clements <austin@google.com> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Cherry Mui [Mon, 22 Jan 2024 16:47:50 +0000 (11:47 -0500)]
all: update vendored dependencies
The Go 1.23 development tree has opened. This is a time to update
all golang.org/x/... module versions that contribute packages to the
std and cmd modules in the standard library to latest master versions.
Generated with:
go install golang.org/x/build/cmd/updatestd@latest
go install golang.org/x/tools/cmd/bundle@latest
updatestd -goroot=$(pwd) -branch=master
For #36905.
Change-Id: I46a68f27a54f1e3f9e1aa5af4de6ee0b26388f3f
Reviewed-on: https://go-review.googlesource.com/c/go/+/557457 Reviewed-by: Carlos Amedee <carlos@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
qmuntal [Fri, 19 Jan 2024 09:59:39 +0000 (10:59 +0100)]
runtime: avoid loading winmm.dll on newer Windows versions
winmm.dll is only used for timeBeginPeriod and timeEndPeriod, which are
not needed on Windows versions supporting high resolution timers, that
is Windows 10 version 1803, and later.
Updates #56745.
Change-Id: Ie9576638fb8d2b4e648283bec3170aefa76f9f82
Reviewed-on: https://go-review.googlesource.com/c/go/+/556935 Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Bryan Mills <bcmills@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Bryan C. Mills [Sat, 20 Jan 2024 00:05:53 +0000 (19:05 -0500)]
net: attempt to deflake TestListenCloseListen
Also make it flakier in longtest mode by burning through more
ephemeral ports. (Burning through the ports raised the failure rate
for me locally enough to reliably reproduce the failure in #65175 with
-count=10.)
Fixes #65175 (I hope).
Change-Id: I5f5b68b6bf6a6aa92e66f0288078817041656a3e
Reviewed-on: https://go-review.googlesource.com/c/go/+/557177 Reviewed-by: Damien Neil <dneil@google.com>
Auto-Submit: Bryan Mills <bcmills@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Rob Findley [Sat, 6 Jan 2024 00:19:14 +0000 (19:19 -0500)]
go/types, types2: typeparams.IndexExpr must not be an ast.Expr
The typeparams.IndexExpr wrapper type was added as a compatibility layer
to make the go/types code symmetric with types2. However, this type
incidentally implemented the ast.Expr interface, leading to the
accidental misuse that led to golang/go#63933.
Fix this minimally for now, though leave a TODO that this old
compatibility shim really needs to be eliminated.
Also fix a case in types2 where operand.expr was set to a typed nil.
Fixes golang/go#63933
Change-Id: I180d411e52f795a8322ecce6ed8649e88af1c63b
Reviewed-on: https://go-review.googlesource.com/c/go/+/554395 Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Findley <rfindley@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Michael Anthony Knyszek [Thu, 11 Jan 2024 21:55:04 +0000 (21:55 +0000)]
runtime: make a much better effort to emit CPU profile in a trace
Currently the new execution tracer's handling of CPU profile samples is
very best-effort. The same CPU profile buffer is used across
generations, leading to a high probability that CPU samples will bleed
across generations. Also, because the CPU profile buffer (not the trace
buffer the samples get written into) isn't guaranteed to be flushed when
we close out a generation, nor when tracing stops. This has led to test
failures, but can more generally just lead to lost samples.
In general, lost samples are considered OK. The CPU profile buffer is
only read from every 100 ms, so if it fills up too much before then, old
samples will get overwritten. The tests already account for this, and in
that sense the CPU profile samples are already best-effort. But with
actual CPU profiles, this is really the only condition under which
samples are dropped.
This CL aims to align CPU profiles better with traces by eliminating
all best-effort parts of the implementation aside from the possibility
of dropped samples from a full buffer.
To achieve this, this CL adds a second CPU profile buffer and has the
SIGPROF handler pick which CPU profile buffer to use based on the
generation, much like every other part of the tracer. The SIGPROF
handler then reads the trace generation, but not before ensuring it
can't change: it grabs its own thread's trace seqlock. It's possible
that a SIGPROF signal lands while this seqlock is already held by the
thread. Luckily this is detectable and the SIGPROF handler can simply
elide the locking if this happens (the tracer will already wait until
all threads exit their seqlock critical section).
Now that there are two CPU profile buffers written to, the read side
needs to change. Instead of calling traceAcquire/traceRelease for every
single CPU sample event, the trace CPU profile reader goroutine holds
this conceptual lock over the entirety of flushing a buffer. This means
it can pick the CPU profile buffer for the current generation to flush.
With all this machinery in place, we're now at a point where all CPU
profile samples get divided into either the previous generation or the
current generation. This is good, since it means that we're able to
emit profile samples into the correct generation, avoiding surprises in
the final trace. All that's missing is to flush the CPU profile buffer
from the previous generation, once the runtime has moved on from that
generation. That is, when the generation counter updates, there may yet
be CPU profile samples sitting in the last generation's buffer. So,
traceCPUFlush now first flushes the CPU profile buffer, followed by any
trace buffers containing CPU profile samples.
The end result of all this is that no sample gets left behind unless it
gets overwritten in the CPU profile buffer in the first place. CPU
profile samples in the trace will now also get attributed to the right
generation, since the SIGPROF handler now participates in the tracer's
synchronization across trace generations.
Fixes #55317.
Change-Id: I47719fad164c544eef0bb12f99c8f3c15358e344
Reviewed-on: https://go-review.googlesource.com/c/go/+/555495
Auto-Submit: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Pratt <mpratt@google.com>
Bryan C. Mills [Wed, 17 Jan 2024 20:00:40 +0000 (15:00 -0500)]
cmd/go/internal/modfetch: set protocol.version=2 for shallow git fetches
This works around an apparent bug in the Git HTTP backend, introduced
in Git 2.21, that causes responses for the version 1 protocol to
provide incomplete tags.
For Git commands older than 2.18, this configuration flag is ignored.
(Note that Git 2.29 and above already use protocol version 2 by
default.)
Damien Neil [Wed, 17 Jan 2024 19:59:10 +0000 (11:59 -0800)]
doc/go1.22: document net/netip changes
For #61422
Change-Id: Ide818366b035eada4ba04b70b4741fb1891585d6
Reviewed-on: https://go-review.googlesource.com/c/go/+/556396
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Carlos Amedee <carlos@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Auto-Submit: Damien Neil <dneil@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Cuong Manh Le [Mon, 23 Oct 2023 03:56:57 +0000 (10:56 +0700)]
runtime: document GODEBUG panicnil values
Updates #25448
Change-Id: Ia1b7a376f5175f67e14ad4bd065d6e8ad5250d38
Reviewed-on: https://go-review.googlesource.com/c/go/+/536955 Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Keith Randall <khr@google.com> Reviewed-by: Carlos Amedee <carlos@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Keith Randall <khr@golang.org>
Robert Griesemer [Wed, 17 Jan 2024 00:59:31 +0000 (16:59 -0800)]
cmd/compile: call types2.Unalias to be ready for GODEBUG=gotypesalias=1
types2.Unalias is not needed if we know we have a core or underlying
type. Also, types of declared functions (signatures) cannot be aliases
(this includes tuples).
Fixes #65125.
Change-Id: I1faa26b66f6c646719e830dd661136fae86f3775
Reviewed-on: https://go-review.googlesource.com/c/go/+/556036
Run-TryBot: Robert Griesemer <gri@google.com> Reviewed-by: Mauri de Souza Meneguzzo <mauri870@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Robert Griesemer <gri@google.com> Reviewed-by: Robert Griesemer <gri@google.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Dominik Honnef [Wed, 27 Dec 2023 21:01:19 +0000 (22:01 +0100)]
internal/trace/v2: avoid several panics for malformed traces
This addresses some panics (out of bounds slice accesses and nil pointer
dereferences) when parsing malformed data. These were found via light
fuzzing, not by any rigorous means, and more potential panics probably
exist.
Fixes #64878.
Fixes #64879.
Change-Id: I4085788ba7dc91fec62e4abd88f50777577db42f
Reviewed-on: https://go-review.googlesource.com/c/go/+/552995
Auto-Submit: Michael Knyszek <mknyszek@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Most libraries don't consider N secret, but it's arguably useful for
privacy applications. However, E should generally be fixed, and there is
a lot of performance to be gained by using variable-time exponentiation.
Filippo Valsorda [Tue, 26 Dec 2023 22:21:34 +0000 (23:21 +0100)]
crypto/rsa: use E = 65537 in benchmarks
Now, this is embarrassing. For the whole Go 1.20 and Go 1.21 cycles, we
based RSA public key operation (verification and decryption) benchmarks
on the keys in rsa_test.go, which had E = 3. Most keys in use, including
all those generated by GenerateKey, have E = 65537. This significantly
skewed even relative benchmarks, because the new constant-time
algorithms would incur a larger slowdown for larger exponents.
I noticed this only because I got a production profile for an
application that does a lot of RSA verifications, saw ExpShort show up,
made ExpShort faster, and the crypto/rsa profiles didn't move.
We were measuring the wrong thing, and the slowdown was worse than we
thought. My apologies.
(If E had not been parametrized, it would have avoided issues like this
one, too. Grumble. https://words.filippo.io/parameters/#fn9)
Keith Randall [Thu, 11 Jan 2024 20:52:16 +0000 (12:52 -0800)]
runtime: ensure we free unrolled GC bitmaps
CL 555355 has a bug in it - the GC program flag was also used to decide
when to free the unrolled bitmap. After that CL, we just don't free any
unrolled bitmaps, leading to a memory leak.
Use a separate flag to track types that need to be freed when their
corresponding object is freed.
Change-Id: I841b65492561f5b5e1853875fbd8e8a872205a84
Reviewed-on: https://go-review.googlesource.com/c/go/+/555416
Auto-Submit: Keith Randall <khr@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com> Reviewed-by: Keith Randall <khr@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Robert Griesemer [Wed, 10 Jan 2024 23:49:33 +0000 (15:49 -0800)]
go/types, types2: don't lose position info of interface embeddings
Accurate position information for embedded types in interfaces is
crucial to identify the corresponding source file, and with that
the Go language version associated with that file. (The position
information is also important for proper error messages.)
Before this CL, the position information for embedded types was
discarded after type set computation, in the assumption that it
was not needed anymore. However, substitutions that update the
interface may lead to repeated type set computations which then
won't have the correct position information.
This CL does preserve the position information for embedded
types until the end of type checking (cleanup phase), and also
copy the position information during a substitution of the
interface.
The respective bug (#64759) doesn't seem to appear in 1.22 (most
likely because it's hidden by some of the changes made with respect
to the file version logic), but the existing code is still wrong.
The backport of this code to 1.21 and 1.20 fixes the issue in those
releases.
For #64759.
Change-Id: I80f4004c9d79cb02eac6739c324c477706615102
Reviewed-on: https://go-review.googlesource.com/c/go/+/555296
Run-TryBot: Robert Griesemer <gri@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Robert Findley <rfindley@google.com> Reviewed-by: Robert Griesemer <gri@google.com>
Bryan C. Mills [Mon, 8 Jan 2024 22:20:14 +0000 (17:20 -0500)]
os: relax tests and add examples for UserCacheDir and UserConfigDir
Per https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html:
“If, when attempting to write a file, the destination directory is
non-existent an attempt should be made to create it with permission
0700. […] The application should be prepared to handle the case where
the file could not be written […]. In such case it may choose to
present an error message to the user.”
In certain CI environments, these directories have well-defined
locations but do not exist and cannot be created. In that case,
we now choose to log and return from the test without failing it.
To prevent the functions from falling back to being entirely untested,
we still fail the test (and “present an error message to the user”) if
either function returns an empty string without an error, or returns a
path that refers to a non-directory or results in an error other than
ErrNotExist.
In addition, since the tests themselves no longer create subdirectories,
we add examples illustrating the suggested pattern of usage.
Than McIntosh [Tue, 9 Jan 2024 15:06:15 +0000 (15:06 +0000)]
cmd/compile: use hashed symbol name for go.shape types if too long
Shape-based stenciling in the Go compiler's generic instantiation
phase looks up shape types using the underlying type of a given target
type. This has a beneficial effect in most cases (e.g. we can use the
same shape type for two different named types whose underlying type is
"int"), but causes some problems when the underlying type is a very
large structure. The link string for the underlying type of a large
imported struct can be extremely long, since the link string
essentially enumerates the full package path for every field type;
this can produce a "go.shape.struct { ... " symbol name that is
absurdly long.
This patch switches the compiler to use a hash of the underlying type
link string instead of the string itself, which should continue to
provide commoning but keep symbol name lengths reasonable for shape
types based on large imported structs.
Fixes #65030.
Change-Id: I87d602626c43172beb99c186b8ef72327b8227a2
Reviewed-on: https://go-review.googlesource.com/c/go/+/554975
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Than McIntosh <thanm@google.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Michael Pratt [Tue, 9 Jan 2024 18:13:45 +0000 (13:13 -0500)]
runtime: replace rwmutexR/W with per-rwmutex lock rank
CL 549536 intended to decouple the internal implementation of rwmutex
from the semantic meaning of an rwmutex read/write lock in the static
lock ranking.
Unfortunately, it was not thought through well enough. The internals
were represented with the rwmutexR and rwmutexW lock ranks. The idea was
that the internal lock ranks need not model the higher-level ordering,
since those have separate rankings. That is incorrect; rwmutexW is held
for the duration of a write lock, so it must be ranked before any lock
taken while any write lock is held, which is precisely what we were
trying to avoid.
This is visible in violations like:
0 : execW 11 0x0
1 : rwmutexW 51 0x111d9c8
2 : fin 30 0x111d3a0
fatal error: lock ordering problem
execW < fin is modeled, but rwmutexW < fin is missing.
Fix this by eliminating the rwmutexR/W lock ranks shared across
different types of rwmutex. Instead require users to define an
additional "internal" lock rank to represent the implementation details
of rwmutex.rLock. We can avoid an additional "internal" lock rank for
rwmutex.wLock because the existing writeRank has the same semantics for
semantic and internal locking. i.e., writeRank is held for the duration
of a write lock, which is exactly how rwmutex.wLock is used, so we can
use writeRank directly on wLock.
For #64722.
Cq-Include-Trybots: luci.golang.try:gotip-linux-amd64-staticlockranking
Change-Id: Ia572de188a46ba8fe054ae28537648beaa16b12c
Reviewed-on: https://go-review.googlesource.com/c/go/+/555055
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>