Robert Griesemer [Thu, 6 Oct 2022 20:51:52 +0000 (13:51 -0700)]
internal/types/errors: add package
The internal/types/errors package defines all error codes used by
the type checkers. This is the 1st step of several that factor out
the error codes from go/types and types2; the package is not yet
used.
- The file codes.go is a copy of go/types/errorcodes.go. The
only change is the updated package name (types -> errors).
- The file codes_test.go is a copy of go/types/errorcodes_test.go
with updated package name (types_test -> errors_test) and minor
changes to walkCodes so that it doesn't require the pkgFiles
helper function (the test only parses a single file).
Change-Id: Idb977b1220737b56b330de1d977f698f022daafc
Reviewed-on: https://go-review.googlesource.com/c/go/+/439560
Run-TryBot: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com> Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@google.com>
Robert Griesemer [Thu, 6 Oct 2022 20:17:53 +0000 (13:17 -0700)]
go/types, types2: better error for generic type decl. with missing constraint
If a generic type declaration is missing a constraint, syntactically
it is an array type declaration with an undefined array length.
Mention the possibility of a missing constraint in the error message
for the undefined array length.
For #56064.
For #55961.
For #51145.
Change-Id: Ic161aeda9ea44faa8aa3bf3e9d62b3b13a95d4c5
Reviewed-on: https://go-review.googlesource.com/c/go/+/439559
Run-TryBot: Robert Griesemer <gri@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@google.com> Reviewed-by: Robert Findley <rfindley@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Motiejus Jakštys [Sat, 8 Oct 2022 05:00:42 +0000 (05:00 +0000)]
cmd/cgo: use -O2 when testing compiler features
Add "-O2" to all compiler/linker tests. This makes compiler/linker
feature probing better resemble actual compiling later.
Why?
----
zig c++ is a clang front-end[1] that accepts, among other things, the
target over the command line. This command:
zig c++ -target x86_64-linux-gnu main.o -o main
Will:
1. Pre-compile libc++.a.
2. Link the program with libc++.a from (1).
Currently Go only is learning about one flag from the linker, that is,
"--no-gc-sections". The resulting command that tests for the flag
support is this:
c++ -Wl,--no-gc-sections -x c - -o
This causes Zig to pre-compile libc++.a in debug mode. Then the actual
compiler+linker command from CGo adds a few more flags, including "-O2":
c++ <...> -Wl,--no-gc-sections -O2 <...>
From Zig perspective, debug-mode libc++.a is different from the
optimized one; that causes Zig to compile a new libc++.a. Specifically,
Zig adds "-fno-omit-frame-pointer" for debug builds, and
"-fomit-frame-pointer" for optimized builds.
As a result, we have to two sets of libc++.a for every arch/os tuple.
That takes CPU time and a bit of disk space.
Zig performance impact
----------------------
First compilation of a simple CGo application is faster by ~2.5 seconds
or ~60%:
$ CC="zig c++ -target x86_64-linux-gnu.2.28" hyperfine \
--warmup 3 --runs 10 \
--prepare 'rm -fr ~/.cache/zig ~/.cache/go-build /tmp/go-*' \
--parameter-list go go1.19,go1.19-O2 \
'/code/go/bin/{go} build .'
Benchmark 1: /code/go/bin/go1.19 build .
Time (mean ± σ): 6.168 s ± 0.059 s [User: 7.465 s, System: 1.578 s]
Range (min … max): 6.111 s … 6.242 s 10 runs
Benchmark 2: /code/go/bin/go1.19-O2 build .
Time (mean ± σ): 3.816 s ± 0.080 s [User: 4.730 s, System: 1.130 s]
Range (min … max): 3.657 s … 3.958 s 10 runs
Summary
'/code/go/bin/go1.19-O2 build .' ran
1.62 ± 0.04 times faster than '/code/go/bin/go1.19 build .'
If we add C++ to the mix, the difference grows to almost ~23 seconds, or
almost 90%:
Benchmark 1: CC="zig c++ -target x86_64-linux-gnu.2.28" /code/go/bin/go1.19 build .
Time (mean ± σ): 51.137 s ± 0.183 s [User: 234.165 s, System: 15.005 s]
Range (min … max): 50.934 s … 51.288 s 3 runs
Benchmark 2: CC="zig c++ -target x86_64-linux-gnu.2.28" /code/go/bin/go1.19-O2 build .
Time (mean ± σ): 27.102 s ± 0.068 s [User: 119.816 s, System: 8.513 s]
Range (min … max): 27.038 s … 27.174 s 3 runs
Summary
'/code/go/bin/go1.19-O2 build .' ran
1.89 ± 0.01 times faster than '/code/go/bin/go1.19 build .'
The difference is just due to the fact that Zig will not be instructed
to compile libc++.a for debug builds; Go doesn't need that.
A.k.a. does "-O2" for this check worsen performance?
No statistically significant performance differences with both clang-15
and gcc-11. Also, it affects only the first compile of a CGo progam, as
the linker tests are cached across invocations. go1.19 binary is the
go1.19 tag; go1.19-O2 is go1.19 + this patch.
$ hyperfine --warmup 3 --runs 20 \
--prepare 'rm -fr ~/.cache/go-build/ /tmp/go-*' \
--parameter-list go go1.19,go1.19-O2 \
--parameter-list cc gcc-11,clang-15 \
'CC={cc} /code/go/bin/{go} build .'
Benchmark 1: CC=gcc-11 /code/go/bin/go1.19 build .
Time (mean ± σ): 681.1 ms ± 13.7 ms [User: 501.6 ms, System: 247.1 ms]
Range (min … max): 654.1 ms … 707.2 ms 20 runs
Benchmark 2: CC=gcc-11 /code/go/bin/go1.19-O2 build .
Time (mean ± σ): 676.8 ms ± 10.2 ms [User: 500.4 ms, System: 245.6 ms]
Range (min … max): 664.4 ms … 696.4 ms 20 runs
Benchmark 3: CC=clang-15 /code/go/bin/go1.19 build .
Time (mean ± σ): 860.1 ms ± 17.1 ms [User: 530.0 ms, System: 394.9 ms]
Range (min … max): 839.4 ms … 920.0 ms 20 runs
Benchmark 4: CC=clang-15 /code/go/bin/go1.19-O2 build .
Time (mean ± σ): 864.5 ms ± 26.6 ms [User: 537.8 ms, System: 390.1 ms]
Range (min … max): 841.9 ms … 955.5 ms 20 runs
Summary
'CC=gcc-11 /code/go/bin/go1.19-O2 build .' ran
1.01 ± 0.03 times faster than 'CC=gcc-11 /code/go/bin/go1.19 build .'
1.27 ± 0.03 times faster than 'CC=clang-15 /code/go/bin/go1.19 build .'
1.28 ± 0.04 times faster than 'CC=clang-15 /code/go/bin/go1.19-O2 build .'
1. Add "-O2" for linker-only tests. That looks like too much catering to
zig alone. If we can add it, then add for everything.
2. Add "-fomit-frame-pointer" instead of "-O2". This flag does not
universally imply debug mode, thus same argument applies as to (1).
3. Add "-O2" for this particular test (`--no-gc-sections`). This is
brittle and not future-proof: a future linker test may omit this
flag.
Hardware
--------
Tested on a 4-core (8 HT) Intel(R) Core(TM) i7-8665U CPU on Debian 11,
Linux 5.10.0-15-amd64.
Zeke Lu [Fri, 7 Oct 2022 00:49:15 +0000 (00:49 +0000)]
cmd/go/internal/modload: improve error message for failing to read module listed in go.work
Run "go build ./x" in this workspace:
-- go.work --
use ./y
-- x/go.mod --
module x
go 1.19
-- x/m.go --
package m
It fails with: "go: open /tmp/foo/y/go.mod: no such file or directory".
It's unclear where the name "y" comes from.
This change will emit error like: "go: cannot load module listed in
go.work file: open /tmp/foo/y/go.mod: no such file or directory"
Ian Lance Taylor [Tue, 4 Oct 2022 21:35:39 +0000 (14:35 -0700)]
os: use poll.fdMutex for Plan 9 files
This permits us to safely support concurrent access to files on Plan 9.
Concurrent access was already safe on other systems.
This does introduce a change: if one goroutine calls a blocking read
on a pipe, and another goroutine closes the pipe, then before this CL
the close would occur. Now the close will be delayed until the blocking
read completes.
Also add tests that concurrent I/O and Close on a pipe are OK.
For #50436
For #56043
Change-Id: I969c869ea3b8c5c2f2ef319e441a56a3c64e7bf5
Reviewed-on: https://go-review.googlesource.com/c/go/+/438347 Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: David du Colombier <0intro@gmail.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Rob Pike <r@golang.org>
eric fang [Thu, 18 Aug 2022 09:59:35 +0000 (09:59 +0000)]
cmd/compile: enable carry chain scheduling for arm64
This is a follow up of CL 393656 on arm64.
This CL puts ScoreCarryChainTail before ScoreMemory and after
ScoreReadFlags, so that the scheduling of the carry chain will not
break the scheduling of ScoreVarDef.
Jeremy Brewer [Tue, 20 Sep 2022 00:09:56 +0000 (20:09 -0400)]
modload: provide a clearer error for standard library packages from newer releases
An older version of go compiling a main module that references a
standard library package from a newer release (e.g. net/netip added in
go 1.18) currently produces a confusing error message. This changes adds
a new error message including go version diagnostics.
Andrew Pogrebnoy [Fri, 7 Oct 2022 07:12:54 +0000 (10:12 +0300)]
runtime: move epoll syscalls to runtime/internal/syscall
This change moves Linux epoll's syscalls implementation to the
"runtime/internal/syscall" package. The intention in this CL was to
minimise behavioural changes but make the code more generalised. This
also will allow adding new syscalls (like epoll_pwait2) without the
need to implement assembly stubs for each arch.
It also drops epoll_create as not all architectures provide this call.
epoll_create1 was added to the kernel in version 2.6.27 and Go requires
Linux kernel version 2.6.32 or later since Go 1.18. So it is safe to
always use epoll_create1.
This is a resubmit as the previous CL 421994 was reverted due to test
failures after the merge with the master. The issue was fixed in
CL 438615
For #53824
For #51087
Change-Id: I1bd0f23a85b4f9b80178c5dd36fd3e95ff4f9648
Reviewed-on: https://go-review.googlesource.com/c/go/+/440115 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Michael Pratt <mpratt@google.com> Reviewed-by: Michael Pratt <mpratt@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com>
Wayne Zuo [Tue, 6 Sep 2022 14:12:16 +0000 (22:12 +0800)]
cmd/compile: intrinsify Add64 on loong64
This is a follow up of CL 420094 on loong64.
Reduce go toolchain size slightly on linux/loong64.
compilecmp HEAD~1 -> HEAD
HEAD~1 (8a32354219): internal/trace: use strings.Builder
HEAD (1767784ac3): cmd/compile: intrinsify Add64 on loong64
platform: linux/loong64
Eddie Lopez [Fri, 30 Sep 2022 20:51:33 +0000 (16:51 -0400)]
testing: add an example showcasing B.RunParallel with B.ReportMetric
This commit was dedicated to adding an example of using B.ReportMetrics
with B.RunParallel called ExampleB_ReportMetric_parallel. In this
example, the same algorithm for ExampleB_ReportMetric was used, instead
with a concurrent for loop using PB.Next instead of a standard one.
There is also notes noting when to use the B.ReportMetric methods when
running concurrent testing.
Damien Neil [Mon, 3 Oct 2022 23:07:48 +0000 (16:07 -0700)]
net/http: refactor tests to run most in HTTP/1 and HTTP/2 modes
Replace the ad-hoc approach to running tests in HTTP/1 and HTTP/2
modes with a 'run' function that executes a test in various modes.
By default, these modes are HTTP/1 and HTTP/2, but tests can
opt-in to HTTPS/1 as well.
The 'run' function also takes care of post-test cleanup (running the
afterTest function).
The 'run' function runs tests in parallel by default. Tests which
can't run in parallel (generally because they use global test hooks)
pass a testNotParallel option to disable parallelism.
Update clientServerTest to use t.Cleanup to clean up after itself,
rather than leaving this up to tests to handle.
Drop an unnecessary mutex in SetReadLoopBeforeNextReadHook.
Test hooks can't be set in parallel, and we want the race detector
to notify us if two simultaneous tests try to set a hook.
Fixes #56032
Change-Id: I16be64913c426fc93d84abc6ad85dbd3bc191224
Reviewed-on: https://go-review.googlesource.com/c/go/+/438137
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Damien Neil <dneil@google.com> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by: David Chase <drchase@google.com>
Tobias Klauser [Tue, 4 Oct 2022 11:53:46 +0000 (13:53 +0200)]
internal/poll, internal/syscall/unix, syscall: move writev definition for solaris
Move the writev definition for solaris from package
internal/syscall/unix to package syscall. This corresponds to where
writev is defined on aix, darwin and openbsd as well and is
go:linkname'ed from internal/poll. This also allows updating the
generated wrappers more easily if needed.
Change-Id: I671ed8232d25319f8e63f549f786d77a17602148
Reviewed-on: https://go-review.googlesource.com/c/go/+/436597
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Carlos Amedee <carlos@golang.org> Reviewed-by: Ian Lance Taylor <iant@google.com>
Auto-Submit: Tobias Klauser <tobias.klauser@gmail.com>
Lynn Boger [Thu, 18 Aug 2022 20:35:38 +0000 (15:35 -0500)]
cmd/compile: leverage cc ops in more cases on ppc64x
This updates some rules to use ops with CC variations to
set the condition code when the result of the operation is
zero. This allows the following compare with zero to be
removed since the equivalent condition code has already been
set.
In addition, a previous rule change to use ANDCCconst was modified
to allow any constant value, not just 1 in some cases.
kxxt [Sat, 1 Oct 2022 00:19:12 +0000 (00:19 +0000)]
all: use grep -E/-F instead of fgrep/egrep
egrep and fgrep are obsolescent now.
This PR updates all egrep and fgrep commands to grep -E and grep -F.
Running egrep/fgrep command with grep v3.8 will output the following warning to stderr:
egrep: warning: egrep is obsolescent; using grep -E
see also:
https://www.phoronix.com/news/GNU-Grep-3.8-Stop-egrep-fgrep
https://lists.gnu.org/archive/html/info-gnu/2022-09/msg00001.html
Change-Id: Iea1ca9ae72264530c67727b5e27cf1b7a362dd97
GitHub-Last-Rev: 3584884bd48cca97271ab86010fce8e4e063c0e4
GitHub-Pull-Request: golang/go#55299
Reviewed-on: https://go-review.googlesource.com/c/go/+/432256
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: Carlos Amedee <carlos@golang.org>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Bryan C. Mills [Wed, 5 Oct 2022 19:21:06 +0000 (15:21 -0400)]
os/exec: parallelize more tests
This cuts the wall duration for 'go test os/exec' and
'go test -race os/exec' roughly in half on my machine,
which is an even more significant speedup with a high '-count'.
For better or for worse, it may also increase the repro rate
of #34988.
Tests that use Setenv or Chdir or check for FDs opened during the test
still cannot be parallelized, but they are only a few of those.
Bryan C. Mills [Wed, 5 Oct 2022 19:26:43 +0000 (15:26 -0400)]
os/exec: delete TestExtraFilesFDShuffle
This test has been disabled for over nine years (since CL 12869049).
Although it still compiles, it seems likely to have rotted since then,
and if it was going to detect a real bug it also seems like that bug
would have been encountered and reported by users since then (and
would presumably have its own regression tests).
To eliminate overhead from mainining it (or skipping over it while
maintaining other tests), let's just delete it.
Zeke Lu [Tue, 13 Sep 2022 18:05:53 +0000 (18:05 +0000)]
debug/buildinfo: correct a typo in calculating next align index
When it calculates the smallest n such that:
n >= i && n % buildInfoAlign == 0
the expression should be
(i+buildInfoAlign-1)&^(buildInfoAlign-1)
instead of
(i+buildInfoAlign-1)&^buildInfoAlign
Fixes #54968.
Change-Id: Ibb7bdf568a521545b2609acc85e2ab4e05da5dae
GitHub-Last-Rev: 479ebc140af9809f0bea039e643cb95b4f857614
GitHub-Pull-Request: golang/go#54971
Reviewed-on: https://go-review.googlesource.com/c/go/+/429815 Reviewed-by: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com>
Auto-Submit: Bryan Mills <bcmills@google.com>
Wayne Zuo [Tue, 6 Sep 2022 03:43:28 +0000 (11:43 +0800)]
cmd/compile: fold constant shift with extension on riscv64
For example:
movb a0, a0
srai $1, a0, a0
the assembler will expand to:
slli $56, a0, a0
srai $56, a0, a0
srai $1, a0, a0
this CL optimize to:
slli $56, a0, a0
srai $57, a0, a0
Remove 270+ instructions from Go binary on linux/riscv64.
Change-Id: I375e19f9d3bd54f2781791d8cbe5970191297dc8
Reviewed-on: https://go-review.googlesource.com/c/go/+/428496 Reviewed-by: Keith Randall <khr@google.com>
Run-TryBot: Wayne Zuo <wdvxdr@golangcn.org> Reviewed-by: Joel Sing <joel@sing.id.au> Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
cui fliter [Wed, 5 Oct 2022 03:15:24 +0000 (03:15 +0000)]
all: remove redundant type conversion
Change-Id: I375233dc700adbc58a6d4af995d07b352bf85b11
GitHub-Last-Rev: ef129205231b892f61b0135c87bb791a5e1a126c
GitHub-Pull-Request: golang/go#55994
Reviewed-on: https://go-review.googlesource.com/c/go/+/437715
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Keith Randall <khr@google.com>
Ian Lance Taylor [Tue, 4 Oct 2022 03:08:11 +0000 (20:08 -0700)]
os: if dirFS.Open fails, undo use of backslashes in error message
This fixes a bug introduced by CL 426094 that caused the
golang.org/x/website/internal/web tests to fail.
Fixes #56034
Change-Id: Ic64967c6d440ad260b7283a18972b20023320ab6
Reviewed-on: https://go-review.googlesource.com/c/go/+/437976
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@google.com>
kijimaD [Wed, 5 Oct 2022 12:56:46 +0000 (12:56 +0000)]
all: fix some typos
Change-Id: I6be77e7b7c919f26bed7b6690cce6741888ba78a
GitHub-Last-Rev: 4ef4a7b425d0b89adf398a1bee04e9f7495813bc
GitHub-Pull-Request: golang/go#56051
Reviewed-on: https://go-review.googlesource.com/c/go/+/438991 Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Run-TryBot: Robert Griesemer <gri@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@google.com>
Michael Matloob [Wed, 5 Oct 2022 16:54:00 +0000 (12:54 -0400)]
cmd/go/internal/modindex: ignore non-source files for index
We were saving non-go file information in the module index files,
leading in an unnecessary increase in memory usage in modules
containing many non-go files. This was a bug because this information
is never used. Don't save that information.
For #54226
Change-Id: I0644064f83f96e3a9f43b7e66ca94d69d9603376
Reviewed-on: https://go-review.googlesource.com/c/go/+/439118
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com>
Than McIntosh [Tue, 4 Oct 2022 19:10:30 +0000 (15:10 -0400)]
runtime/coverage: revise/shorten function names
Use shorter more Go-like names for the new APIs being added in the
runtime/coverage package for writing coverage data under user control
from server programs. Old names were a bit too clunky/verbose.
Cherry Mui [Fri, 30 Sep 2022 23:01:09 +0000 (19:01 -0400)]
runtime: don't jump stack if at entry of systemstack
The traceback code has special "jump stack" logic, to trace back
stack switches through systemstack. If we're at the entry of
systemstack, the stack switch hasn't happened, so don't jump to
user stack.
The jump stack logic is only used if we're on the g0 stack. It can
happen that we're at the entry of a recursive systemstack call on
the g0 stack. In we jump stack here, there will be two problems:
1. There are frames between entering the g0 stack and this
recursive systemstack call. Those frames will be lost.
2. Worse, we switched frame.sp but frame.fp calculation will use
the entry SP delta (0), which will be wrong, which in turn
leads wrong frame.lr and things will go off.
For now, don't jump stack if we're at entry of systemstack (SP
delta is 0).
Using a per-PC SPWRITE marker may be a better fix. If we haven't
written the SP, we haven't switched the stack so we can just
unwind like a normal function.
May fix #55851.
Change-Id: I2b624c8c086b235b34d9c7d3cebd4a37264f00f8
Reviewed-on: https://go-review.googlesource.com/c/go/+/437299
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Cherry Mui <cherryyz@google.com> Reviewed-by: Michael Pratt <mpratt@google.com>
Youlin Feng [Mon, 3 Oct 2022 13:15:12 +0000 (13:15 +0000)]
runtime/cgo: let darwin pthread stacksize follow rlimit
On Mac OS X, the default stack size for non-main threads created by cgo is
fixed at 512KB and cannot be altered by setrlimit. This stack size is too
small for some recursive scenarios. We can solve this problem by explicitly
copying the stack size of the main thread when creating a new thread.
Change-Id: I400d5b2e929a1ee261502914a991e208759f64a8
GitHub-Last-Rev: b29c74599ea1cce4683d25e1ac8a70ffd9b86381
GitHub-Pull-Request: golang/go#53667
Reviewed-on: https://go-review.googlesource.com/c/go/+/415915
Run-TryBot: Cherry Mui <cherryyz@google.com> Reviewed-by: hopehook <hopehook@golangcn.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Paul E. Murphy [Tue, 14 Jun 2022 21:13:51 +0000 (16:13 -0500)]
cmd/link: optimize PPC64 inline plt sequences if local
Indirect branches are much more expensive than direct. If the call is
known to be local, we can replace most of the operations with a nop,
and call directly.
Updates #53345
Change-Id: Icfff9ec1f6c7f8e4181f0f28976033308d2f53eb
Reviewed-on: https://go-review.googlesource.com/c/go/+/412715 Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
eric fang [Wed, 17 Aug 2022 10:01:17 +0000 (10:01 +0000)]
cmd/compile: add late lower pass for last rules to run
Usually optimization rules have corresponding priorities, some need to
be run first, some run next, and some run last, which produces the best
code. But currently our optimization rules have no priority, this CL
adds a late lower pass that runs those rules that need to be run at last,
such as split unreasonable constant folding. This pass can be seen as
the second round of the lower pass.
For example:
func foo(a, b uint64) uint64 {
d := a+0x1234568
d1 := b+0x1234568
return d&d1
}
The code generated by the master branch:
0x0004 00004 ADD $19088744, R0, R2 // movz+movk+add
0x0010 00016 ADD $19088744, R1, R1 // movz+movk+add
0x001c 00028 AND R1, R2, R0
This is because the current constant folding optimization rules do not
take into account the range of constants, causing the constant to be
loaded repeatedly. This CL splits these unreasonable constants folding
in the late lower pass. With this CL the generated code:
0x0004 00004 MOVD $19088744, R2 // movz+movk
0x000c 00012 ADD R0, R2, R3
0x0010 00016 ADD R1, R2, R1
0x0014 00020 AND R1, R3, R0
This CL also adds constant folding optimization for ADDS instruction.
In addition, in order not to introduce the codegen regression, an
optimization rule is added to change the addition of a negative number
into a subtraction of a positive number.
Change-Id: I7c3cc0f48631af73cf34ad3c731c380f46c72359
Reviewed-on: https://go-review.googlesource.com/c/go/+/435257
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: hopehook <hopehook@golangcn.org>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Zeke Lu [Tue, 4 Oct 2022 07:10:09 +0000 (07:10 +0000)]
reflect: avoid unnecessary copy of funcTypes
Imagine that initFuncTypes is called with n=3, funcTypes will be
[nil, nil, nil, **reflect.rtype] afterward, then it's called with n=2.
The current implementation will copy funcTypes because funcTypes[2] is
nil. This is unnecessary. It should make a new slice and copy funcTypes
into it only when n >= len(funcTypes).
Updates #56011.
Change-Id: Ia093d2f550d6924a4c58bcd21325093e32b40baa
GitHub-Last-Rev: a599eae7c2f6a388dfe1ff39cf61fd645885a64d
GitHub-Pull-Request: golang/go#56024
Reviewed-on: https://go-review.googlesource.com/c/go/+/438395
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: Keith Randall <khr@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Auto-Submit: Keith Randall <khr@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Bryan Mills [Tue, 4 Oct 2022 19:55:56 +0000 (19:55 +0000)]
Revert "os/exec: make StdoutPipe and StderrPipe safe to Close concurrently"
This reverts CL 437176.
Reason for revert: broke programs that plumb StdoutPipe from one command to Stdin on another and then call Wait on the former.
os/exec itself uses a type-assertion to *os.File to determine whether to copy stdin using a goroutine or just pass a file descriptor. An early Wait using a *os.File is benign (because closing the pipe doesn't close the child's inherited file descriptor), but an early Wait using a non-*os.File is not.
cmd/cgo, cmd/compile, cmd/link: remove old style build tags
[Roll-forward of CL 436915 by Tobias Klauser, with builtin and gen
directories dropped now that they've been handled separately.]
The minimum bootstrap version for Go ≥ 1.20 is Go 1.17. That version
supports the new style //go:build lines. Thus the old style //+build
lines can be dropped in this part of the tree as well. Leave the
//+build lines in cmd/dist which will ensure the minimum Go version
during bootstrap.
cmd/compile: rename gen and builtin to _gen and _builtin
These two directories are full of //go:build ignore files.
We can ignore them more easily by putting an underscore
at the start of the name. That also works around a bug
in Go 1.17 that was not fixed until Go 1.17.3.
Change-Id: Ia5389b65c79b1e6d08e4fef374d335d776d44ead
Reviewed-on: https://go-review.googlesource.com/c/go/+/435472
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
Than McIntosh [Mon, 3 Oct 2022 22:40:59 +0000 (18:40 -0400)]
runtime/coverage: use atomic access for counter reads
Read counters using atomic ops so as to avoid problems with the race
detector if a goroutine happens to still be executing at the end of a
test run when we're writing out counter data. In theory we could guard
the atomic use on the counter mode, but it's better just to do it in
all cases, leaves us with a simpler implementation.
Fixes #56006.
Change-Id: I81c2234b5a1c3b00cff6c77daf2c2315451b7f6c
Reviewed-on: https://go-review.googlesource.com/c/go/+/438256
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: David Chase <drchase@google.com>
Run-TryBot: Than McIntosh <thanm@google.com>
Than McIntosh [Tue, 4 Oct 2022 13:00:31 +0000 (09:00 -0400)]
cmd/internal/sys: migrate support.go functions to new internal pkg
Separate out the functions from cmd/internal/sys/support.go and
migrate them to a new package internal/platform, so that functions such as
"RaceDetectorSupported" can be called from tests in std as well as in
cmd. This isn't a complete move of everything in cmd/internal/sys;
there are still many functions left.
The original version of this CL (patch set 1) called the new package
"internal/sys", but for packages that needed both "internal/sys" and
"cmd/internal/sys" the import of the former had to be done with a
different name, which was confusing and also required a hack in
cmd/dist.
Updates #56006.
Change-Id: I866d62e75adbf3a640a06e2c7386a6e9e2a18d91
Reviewed-on: https://go-review.googlesource.com/c/go/+/438475 Reviewed-by: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Than McIntosh <thanm@google.com>
Archana R [Thu, 29 Sep 2022 14:53:18 +0000 (09:53 -0500)]
cmd/go: update gcc version check for asan support on ppc64le
Update the minimum version required for asan to be gcc9.
This will ensure that go build -asan is supported only on
systems with the required version of gcc. Update the asan
error message to include the name of the compiler (the
error message is updated to include the name of the compiler
instead of C compiler for other platforms as well).
SecCreatePolicySSL returns null when called from a binary that has a
strange path. This seems to be a weirdo macos bug, but we should be
properly handling those null returns anyway. Also add handling for
SecTrustGetCertificateAtIndex.
Fixes #54590
Change-Id: I251e74f3b0bf65890a80b094b3e88718e13fd3db
Reviewed-on: https://go-review.googlesource.com/c/go/+/438135
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Roland Shoemaker <roland@golang.org> Reviewed-by: Damien Neil <dneil@google.com>
Auto-Submit: Roland Shoemaker <roland@golang.org>
Daniel Martí [Thu, 22 Sep 2022 12:35:08 +0000 (13:35 +0100)]
encoding/gob: prevent a decoder state overflow
When decoding a struct, if a positive delta is large enough to overflow
when added to fieldnum, we would panic due to the resulting negative index.
Instead, catch this problem and produce an error like we do with
negative delta integers. If fieldnum ends up being negative or smaller
than state.fieldnum, the addition overflowed.
While here, remove an unnecessary break after an error call,
since those error functions cause a panic.
Fixes #55337.
Change-Id: I7a0e4f43e5c81a703e79c1597e3bb3714cc832c8
Reviewed-on: https://go-review.googlesource.com/c/go/+/432715 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Bryan Mills <bcmills@google.com>
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gopher Robot <gobot@golang.org>
Cuong Manh Le [Mon, 3 Oct 2022 16:46:13 +0000 (23:46 +0700)]
reflect: fix race condition on funcTypes
CL 425314 made creating funcTypes using StructOf, and using a mutex to
protect read+write to funcTypes. However, after initializing funcTypes,
it is accessed in FuncOf without holding lock, causing a race.
Fixing it by returning the n-th Type directly from initFuncTypes, so the
accessing funcTypes will always be guarded by a mutex.
Fixes #56011
Change-Id: I1b50d1ae342943f16f368b8606f2614076dc90fb
Reviewed-on: https://go-review.googlesource.com/c/go/+/437997 Reviewed-by: Keith Randall <khr@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: Keith Randall <khr@golang.org>
cmd/trace: replace loop with append(slice, slice2...)
Change-Id: I4686f36a8f718fea1a08d816bc14e24e3528bb07
Reviewed-on: https://go-review.googlesource.com/c/go/+/436706
Auto-Submit: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: hopehook <hopehook@golangcn.org>
Run-TryBot: xie cui <523516579@qq.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Ian Lance Taylor <iant@google.com>
miller [Sat, 1 Oct 2022 13:39:47 +0000 (14:39 +0100)]
runtime/coverage: recognise Plan 9 error message in emitToNonexistentDir
In TestCoverageApis/emitToNonexistentDir there is a list of error
messages to match when a nonexistent directory is opened. The list
has message text only for Unix and Windows. Add the corresponding
text for Plan 9.
Fixes #55983
Change-Id: Id32130300cb02394b319e1aeb1229ee147b4afb2
Reviewed-on: https://go-review.googlesource.com/c/go/+/437557 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: David du Colombier <0intro@gmail.com>
Run-TryBot: David du Colombier <0intro@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Than McIntosh <thanm@google.com>
chanxuehong [Sat, 1 Oct 2022 11:37:36 +0000 (11:37 +0000)]
go/token: fix File.AddLineColumnInfo logic
The offset of the line info should be smaller than the file size.
Otherwise, it should be ignored.
Before the change, such an invalid line info won't be ignored when it's
the first one to add.
Félix Dorn [Sat, 1 Oct 2022 21:53:19 +0000 (21:53 +0000)]
log/syslog: return nil directly
Reduce return complexity.
Change-Id: I280a0fe1a49371e231e93e0d3e177730b6f28769
GitHub-Last-Rev: 2ebc10641d1b6706a66132a84c1c5b0f394034c6
GitHub-Pull-Request: golang/go#55989
Reviewed-on: https://go-review.googlesource.com/c/go/+/437516
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Benny Siegert <bsiegert@gmail.com> Reviewed-by: Ian Lance Taylor <iant@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Rob Pike <r@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
cui fliter [Fri, 30 Sep 2022 09:38:39 +0000 (09:38 +0000)]
regexp: fix a few function names on comments
Change-Id: I192dd34c677e52e16f0ef78e1dae58a78f6d1aac
GitHub-Last-Rev: 1638a7468951df72f13fea34855b6a4fcbb08226
GitHub-Pull-Request: golang/go#55967
Reviewed-on: https://go-review.googlesource.com/c/go/+/436885
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Ian Lance Taylor [Sat, 27 Aug 2022 02:22:50 +0000 (19:22 -0700)]
os: use backslashes for DirFS on Windows
Otherwise DirFS of a UNC path does not work.
Fixes #54694
Change-Id: I82c1c436f7c26b3935c2cc4fd238daf094fc4d86
Reviewed-on: https://go-review.googlesource.com/c/go/+/426094 Reviewed-by: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Alex Brainman <alex.brainman@gmail.com> Reviewed-by: hopehook <hopehook@golangcn.org>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Change-Id: Ifaa73b64e5b6a1d37c753e2440b642478d7dfbce
Reviewed-on: https://go-review.googlesource.com/c/go/+/436957
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: hopehook <hopehook@golangcn.org>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
cui fliter [Fri, 30 Sep 2022 03:41:42 +0000 (03:41 +0000)]
syscall: remove redundant type conversion
Change-Id: Iae290216687fd1ce8be720600157fb78cc2446d0
GitHub-Last-Rev: 4fba64ecb14a704d39f6ecc33989522bcac6656f
GitHub-Pull-Request: golang/go#55959
Reviewed-on: https://go-review.googlesource.com/c/go/+/436881
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@google.com>
Daniel Martí [Tue, 5 Jul 2022 14:57:37 +0000 (15:57 +0100)]
cmd/go: remove the -i build flag
The flag is now removed from `go build` and `go test`.
It has been deprecated since Go 1.16, printing a warning message.
The idea was to fully delete it in Go 1.17, but that didn't happen.
First, delete the BuildI variable and its flag declarations,
as well as all the bits of docs that mentioned the flag.
Second, delete or simplify the code paths that used BuildI.
Third, adapt the tests to the removed flag.
Some of them are removed, like test_relative_import_dash_i.txt and
TestGoTestDashIDashOWritesBinary, as they centered around the flag.
The rest are modified to not cover or use the flag.
Finally, change cmd/dist to no longer use `go install -i`.
The purpose of the flag was that, when bootstrapping the toolchain,
all of its dependencies would also be installed as object files.
When removing the use of the -i flags, the checkNotStale call right
after building toolchain3 would fail as expected,
because runtime/internal/sys is now only up to date in the build cache.
Luckily, that's not a problem: we run `go install std cmd` right after,
so all standard library packages will be installed as object files.
Move the checkNotStale call after that install command.
Fixes #41696.
Change-Id: I5d8139f18aaee07da886d483e663f3f2f00d5f3a
Reviewed-on: https://go-review.googlesource.com/c/go/+/416094 Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gopher Robot <gobot@golang.org>
Mikael Urankar [Sun, 18 Sep 2022 15:33:42 +0000 (17:33 +0200)]
cmd/dist: add support for freebsd/riscv64
Updates #53466
Change-Id: I6643b4254dc707351d397018cee485bb508dde94
Reviewed-on: https://go-review.googlesource.com/c/go/+/431659 Reviewed-by: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Joel Sing <joel@sing.id.au> Reviewed-by: Joel Sing <joel@sing.id.au> Reviewed-by: Dmitri Goutnik <dgoutnik@gmail.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
This is the last failed test in Unified IR, since it can inline f5 and
f6 but the old frontend can not. So marking them as //go:noinline, with
a TODO for re-enable once GOEXPERIMENT=nounified is gone.
Fixes #53058
Change-Id: Ifbbc49c87997a53e1b323048f0067f0257655fad
Reviewed-on: https://go-review.googlesource.com/c/go/+/437217 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
The mismatch between Unified IR and the old frontend is not about how
they number the closures, but how they name them. For nested closure,
the old frontend use the immediate function which contains the closure
as the outer function, while Unified IR uses the outer most function as
the outer for all closures.
That said, what important is matching the number of closures, not their
name prefix. So this CL relax the test to match both "main.func1.func2"
and "main.func1.2" to satisfy both Unified IR and the old frontend.
Updates #53058
Change-Id: I66ed816d1968aa68dd3089a4ea5850ba30afd75b
Reviewed-on: https://go-review.googlesource.com/c/go/+/437216 Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
os/exec: make StdoutPipe and StderrPipe safe to Close concurrently
For #50436, I want to be able to close the pipes returned by
StdoutPipe and StderrPipe after the Context has been canceled
and the WaitDelay has subsequently expired.
However, the fact that the exec.onceCloser wrapper for StdinPipe
(added in CL 13329043) was retained in CL 65490 suggests to me that
(*os.File).Close is still not safe to call concurrently.
This may cause type assertions of these ReadClosers to *os.File that
once succeeded to no longer do so. However, the StdoutPipe and
StderrPipe methods return interfaces, not concrete *os.Files, so
callers already should not have been relying on that implementation
detail — and as far as I can tell the closeOnce wrapper does not mask
any (*os.File) methods, so assertions to any interface type that
previously succeeded will continue to do so.
This change is logically part of CL 401835, but since it may expose
fragile type-assertions in callers I want to keep it separate for
clearer bisection of any new test failures.
This change undoes CL 401894, because on further consideration
it turns out not to be needed.
This also makes (*Cmd).closeDescriptors a free function, since it does
not actually use the receiver in any way and is not needed to satisfy
any interfaces.
Ian Lance Taylor [Thu, 29 Sep 2022 22:00:08 +0000 (15:00 -0700)]
debug/dwarf: don't crash on negative range/rnglist offset
No test case because the problem can only happen for invalid data. Let
the fuzzer find cases like this.
Fixes #55948
Change-Id: I7ba40ba928d2a14d4ac5b39f966173f3868d4729
Reviewed-on: https://go-review.googlesource.com/c/go/+/436876
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com> Reviewed-by: Than McIntosh <thanm@google.com>
internal/singleflight: avoid race between multiple Do calls
When the first call to Do finished, it calls c.wg.Done() to signal
others that the call was done. However, that happens without holding
a lock, so if others call to Do complete and be followed by a call to
ForgotUnshared, that then returns false.
Fixing this by moving c.wg.Done() inside the section guarded by g.mu, so
the two operations won't be interrupted.
Thanks bcmills@ for finding and suggesting fix.
Change-Id: I850f5eb3f9751a0aaa65624d4109aeeb59dee42c
Reviewed-on: https://go-review.googlesource.com/c/go/+/436437 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: Bryan Mills <bcmills@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Change-Id: I4698d0fa78108d83ee91732e8d3878dbff7f9c90
Reviewed-on: https://go-review.googlesource.com/c/go/+/436711
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: xie cui <523516579@qq.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com> Reviewed-by: Joseph Tsai <joetsai@digital-static.net>
internal/singleflight: fix duplicate deleting key when ForgetUnshared called
A key may be forgotten while the call is still in flight. So when the
call finished, it should only delete the key if that key is associated
with the call. Otherwise, we may remove the wrong newly created call.
Fixes #55343
Change-Id: I4fa72d79cad006e5884e42d885d193641ef84e0a
Reviewed-on: https://go-review.googlesource.com/c/go/+/433315 Reviewed-by: Bryan Mills <bcmills@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Michael Pratt [Fri, 9 Sep 2022 16:29:48 +0000 (12:29 -0400)]
cmd/link/internal/ld: panic if inlined functions missing FuncInfo
All inlined functions are Go functions, and thus should be capable of
having a FuncInfo. Missing FuncInfo is likely indication of a compiler
bug that dropped the symbol too early, failing to add it to the symbol
list used for writing output. I believe all existing cases have been
fixed; this check will prevent regressions.
The exception is -linkshared mode. There symbols are loaded from the
shared library, and the FuncInfo is not available. This is a bug, as it
can result in incorrect the FuncID in inlinedCall, but it is very
involved to fix.
For #54959.
For #55954.
Change-Id: Ib0dc4f1ea62525b55f68604d6013ff33223fdcdd
Reviewed-on: https://go-review.googlesource.com/c/go/+/429637
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Michael Pratt [Fri, 9 Sep 2022 20:23:39 +0000 (16:23 -0400)]
cmd/compile: eagerly create LSym for closures
The linker needs FuncInfo metadata for all inlined functions. This is
typically handled by gc.enqueueFunc calling ir.InitLSym for all function
declarations in typecheck.Target.Decls (ir.UseClosure adds all closures
to Decls).
However, non-trivial closures in Decls are ignored, and are insteaded
enqueued when walk of the calling function discovers them.
This presents a problem for direct calls to closures. Inlining will
replace the entire closure definition with its body, which hides the
closure from walk and thus suppresses symbol creation.
Explicitly create a symbol early in this edge case to ensure we keep
this metadata.
InitLSym needs to move out of ssagen to avoid a circular dependency (it
doesn't have anything to do with ssa anyway). There isn't a great place
for it, so I placed it in ir, which seemed least objectionable.
The added test triggers one of these inlined direct non-trivial closure
calls, though the test needs CL 429637 to fail, which adds a FuncInfo
assertion to the linker. Note that the test must use "run" instead of
"compile" since the assertion is in the linker, and "compiler" doesn't
run the linker.
Fixes #54959.
Change-Id: I0bd1db4f3539a78da260934cd968372b7aa92546
Reviewed-on: https://go-review.googlesource.com/c/go/+/436240
Run-TryBot: Michael Pratt <mpratt@google.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Andrew Pogrebnoy [Mon, 1 Aug 2022 14:58:17 +0000 (17:58 +0300)]
runtime: move epoll syscalls to runtime/internal/syscall
This change moves Linux epoll's syscalls implementation to the
"runtime/internal/syscall" package. The intention in this CL was to
minimise behavioural changes but make the code more generalised. This
also will allow adding new syscalls (like epoll_pwait2) without the
need to implement assembly stubs for each arch.
It also drops epoll_create as not all architectures provide this call.
epoll_create1 was added to the kernel in version 2.6.27 and Go requires
Linux kernel version 2.6.32 or later since Go 1.18. So it is safe to
always use epoll_create1.
For #53824
For #51087
Change-Id: I9a6a26b7f2075a38e041de1bab4691da0ecb94fc
Reviewed-on: https://go-review.googlesource.com/c/go/+/421994 Reviewed-by: Michael Pratt <mpratt@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com>
Auto-Submit: Michael Pratt <mpratt@google.com>
Change-Id: I268033bfcda34b76ef1d3a3446d6d1d875fc33ea
Reviewed-on: https://go-review.googlesource.com/c/go/+/436716
Run-TryBot: xie cui <523516579@qq.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Change-Id: Ie1e76d2e99bf2af7f064c9073c1fb866086962f6
Reviewed-on: https://go-review.googlesource.com/c/go/+/436715
Run-TryBot: Robert Griesemer <gri@google.com>
Run-TryBot: xie cui <523516579@qq.com>
Auto-Submit: Ian Lance Taylor <iant@google.com> Reviewed-by: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
cmd/cgo, cmd/compile, cmd/link: remove old style build tags
The minimum bootstrap version for Go ≥ 1.20 is Go 1.17. That version
supports the new style //go:build lines. Thus the old style //+build
lines can be dropped in this part of the tree as well. Leave the
//+build lines in cmd/dist which will ensure the minimum Go version
during bootstrap.