]> Cypherpunks repositories - gostls13.git/log
gostls13.git
4 years agoos/exec: constrain thread usage in leaked descriptor test on illumos
Joshua M. Clulow [Mon, 30 Nov 2020 01:18:51 +0000 (17:18 -0800)]
os/exec: constrain thread usage in leaked descriptor test on illumos

On illumos systems, libc can under some conditions make use of files
from /proc.  In the case of this test, the creation of new threads was
(in the target thread) causing libc to open and close
"/proc/self/lwp/5/lwpname" to set the thread name, which raced with the
leaking descriptor check (see detailed analysis in #42431).

This change requests that the Go runtime use less threads in the child
process used to check for leaked descriptors, without just disabling the
test.  After a thousand repeated trials, the test no longer fails on
illumos.

Fixes #42431.

Change-Id: Iefda26134fc91f7cb205754676e9845d9b7205cc
Reviewed-on: https://go-review.googlesource.com/c/go/+/273966
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agoruntime/pprof: ignore test failures on windows/arm
Jason A. Donenfeld [Sat, 28 Nov 2020 13:38:29 +0000 (14:38 +0100)]
runtime/pprof: ignore test failures on windows/arm

This is blocking forward progress of the de-bitrotting work, and I don't
know off hand how to fix this. Seeing as its disabled on other
platforms, I suspect pprof might not be a very reliable feature, so just
allow for the tests to fail for now, until somebody more motivated comes
along to fix it.

Updates #42862.

Change-Id: Ibc5cd1d82d97b9c2f887d7f3565f2fa70207c8b0
Reviewed-on: https://go-review.googlesource.com/c/go/+/273826
Run-TryBot: Jason A. Donenfeld <Jason@zx2c4.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>

4 years agobufio: make string(int) conversion safer
smasher164 [Thu, 26 Nov 2020 19:16:33 +0000 (14:16 -0500)]
bufio: make string(int) conversion safer

Updates #42792.

Change-Id: I7e53426c41e5609d9dadceb300f7983ba7ad6577
Reviewed-on: https://go-review.googlesource.com/c/go/+/273526
Run-TryBot: Akhil Indurti <aindurti@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
4 years agotest: match gofrontend error messages
Ian Lance Taylor [Thu, 26 Nov 2020 20:26:02 +0000 (12:26 -0800)]
test: match gofrontend error messages

These changes match the following gofrontend error messages:

blank1.go:16:1: error: may not define methods on non-local type

chan/perm.go:28:9: error: expected channel
chan/perm.go:29:11: error: left operand of ‘<-’ must be channel
chan/perm.go:69:9: error: argument must be channel

complit1.go:25:16: error: attempt to slice object that is not array, slice, or string
complit1.go:26:16: error: attempt to slice object that is not array, slice, or string
complit1.go:27:17: error: attempt to slice object that is not array, slice, or string
complit1.go:49:41: error: may only omit types within composite literals of slice, array, or map type
complit1.go:50:14: error: expected struct, slice, array, or map type for composite literal

convlit.go:24:9: error: invalid type conversion (cannot use type unsafe.Pointer as type string)
convlit.go:25:9: error: invalid type conversion (cannot use type unsafe.Pointer as type float64)
convlit.go:26:9: error: invalid type conversion (cannot use type unsafe.Pointer as type int)

ddd1.go:63:9: error: invalid use of ‘...’ calling non-variadic function

fixedbugs/bug176.go:12:18: error: index expression is not integer constant

fixedbugs/bug332.go:17:10: error: use of undefined type ‘T’

fixedbugs/issue4232.go:22:16: error: integer constant overflow
fixedbugs/issue4232.go:33:16: error: integer constant overflow
fixedbugs/issue4232.go:44:25: error: integer constant overflow
fixedbugs/issue4232.go:55:16: error: integer constant overflow

fixedbugs/issue4458.go:19:14: error: type has no method ‘foo’

fixedbugs/issue5172.go:24:14: error: too many expressions for struct

init.go:17:9: error: reference to undefined name ‘runtime’

initializerr.go:26:29: error: duplicate value for index 1

interface/explicit.go:60:14: error: type assertion only valid for interface types

label.go:64:9: error: reference to undefined label ‘go2’

label1.go:18:97: error: continue statement not within for
label1.go:22:97: error: continue statement not within for
label1.go:106:89: error: continue statement not within for
label1.go:108:26: error: invalid continue label ‘on’
label1.go:111:118: error: break statement not within for or switch or select
label1.go:113:23: error: invalid break label ‘dance’

map1.go:64:9: error: not enough arguments
map1.go:65:9: error: not enough arguments
map1.go:67:9: error: argument 1 must be a map

method2.go:36:11: error: reference to undefined field or method ‘val’
method2.go:37:11: error: reference to undefined field or method ‘val’
method2.go:41:12: error: method requires pointer (use ‘(*T).g’)

syntax/chan1.go:13:19: error: send statement used as value; use select for non-blocking send
syntax/chan1.go:17:11: error: send statement used as value; use select for non-blocking send

Change-Id: I98047b60a376e3d2788836300f7fcac3f2c285cb
Reviewed-on: https://go-review.googlesource.com/c/go/+/273527
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agocmd/link: mark windows/arm as all PIE
Jason A. Donenfeld [Thu, 26 Nov 2020 21:38:45 +0000 (22:38 +0100)]
cmd/link: mark windows/arm as all PIE

If the linker thinks that it's in exe mode instead of pie mode, it
won't emit relocations when generating the pcln table, and we wind
up with crashes like this on windows/arm, where all binaries are
in fact relocated:

    Building Go toolchain2 using go_bootstrap and Go toolchain1.
    fatal error: minpc or maxpc invalid
    runtime: panic before malloc heap initialized

This problem was already solved by darwin/arm64, so solve it the same
way here for windows/arm.

Fixes CL 228478.
Fixes #42786.

Change-Id: I6d1db6907c131183649fc263ccca06783188f344
Reviewed-on: https://go-review.googlesource.com/c/go/+/273566
Run-TryBot: Jason A. Donenfeld <Jason@zx2c4.com>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Trust: Alex Brainman <alex.brainman@gmail.com>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>

4 years agoruntime: adjust address calculation in identifying abort on windows/arm
Jason A. Donenfeld [Fri, 27 Nov 2020 21:07:23 +0000 (22:07 +0100)]
runtime: adjust address calculation in identifying abort on windows/arm

Apparently we're being called on arm 1 byte off, just like on 386 and
amd64, so unify the handler for isAbortPC.

Fixes #42859.
Updates #29050.

Change-Id: I97fffeb4a33d93ca3397ce1c9ba2b05137f391ca
Reviewed-on: https://go-review.googlesource.com/c/go/+/273727
Run-TryBot: Jason A. Donenfeld <Jason@zx2c4.com>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Trust: Alex Brainman <alex.brainman@gmail.com>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>

4 years agoruntime: return 0 from C function in test
Jason A. Donenfeld [Fri, 27 Nov 2020 19:45:15 +0000 (20:45 +0100)]
runtime: return 0 from C function in test

This function's prototype includes a return value, so return a value.
Otherwise clang gets upset:

    --- FAIL: TestDLLPreloadMitigation (1.40s)
        syscall_windows_test.go:986: failed to build dll: exit status 1 - nojack.c:7:1: error: non-void function does not return a value [-Werror,-Wreturn-type]
            }
            ^
            1 error generated.

Fixes #42860.

Change-Id: I65b8eb9ccb502692c5b65bd34829f331cd86eef0
Reviewed-on: https://go-review.googlesource.com/c/go/+/273726
Trust: Jason A. Donenfeld <Jason@zx2c4.com>
Trust: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
4 years agolog: make Default doc comment consistent with package doc
Rodolfo Carvalho [Thu, 26 Nov 2020 16:01:35 +0000 (17:01 +0100)]
log: make Default doc comment consistent with package doc

None of the other, older, doc comments use the '*Logger' form, and while
'Logger' and 'logger' are both used in the package doc comment, the
common term used with the intended meaning is 'standard logger', which
appears another eleven times in doc comments.

Change-Id: I089103198fc82390517615eb27bbe7ef77107d34
Reviewed-on: https://go-review.googlesource.com/c/go/+/273486
Reviewed-by: Rob Pike <r@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Ian Lance Taylor <iant@golang.org>

4 years agoreflect: fix Value.Convert for int-to-string conversions (regression)
Robert Griesemer [Wed, 25 Nov 2020 21:47:42 +0000 (13:47 -0800)]
reflect: fix Value.Convert for int-to-string conversions (regression)

The bug was introduced by https://golang.org/cl/220844.

Updates #42792.
Fixes #42835.

Change-Id: I03065c7526488aded35ef2f800b7162e1606877a
Reviewed-on: https://go-review.googlesource.com/c/go/+/273326
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agodoc/go1.16: consolidate stdlib changes in "Minor changes" section
Dmitri Shuralyov [Wed, 25 Nov 2020 19:07:30 +0000 (14:07 -0500)]
doc/go1.16: consolidate stdlib changes in "Minor changes" section

Many of the standard library changes that were added before CL 272871
ended up in the "Core library" section. That section is meant for
major changes like new packages, and most of these aren't.

Consolidate all changes in the "Minor changes to the library" section
for now, so that it's easier to get a complete picture of changes for
each package, along with the remaining TODOs. Add a TODO to read them
over at the end and factor out items that are worth highlighting.

Apply minor other fixups to improve consistency.

For #40700.

Change-Id: I7dc2e7ebf2ea3385fce0c207bae4ce467998a717
Reviewed-on: https://go-review.googlesource.com/c/go/+/273267
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/dist: restore GOARM=7 default for android/arm
Elias Naur [Wed, 25 Nov 2020 18:12:13 +0000 (19:12 +0100)]
cmd/dist: restore GOARM=7 default for android/arm

Fixes the android/arm builder. Without it, the builder reported
unexpected stale targets during bootstrap:

https://build.golang.org/log/b951f1171be54cf4a12c2a0720ffaf07f8a11377

Tighten the GOARM=7 default in cmd/internal/objabi while here.

Change-Id: I944744910193e72e91bc37b5bf0783076b45e579
Reviewed-on: https://go-review.googlesource.com/c/go/+/273167
Run-TryBot: Elias Naur <mail@eliasnaur.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Elias Naur <mail@eliasnaur.com>

4 years agodoc/go1.16: add time/tzdata release note for CL 261877
Tobias Klauser [Wed, 25 Nov 2020 11:57:38 +0000 (12:57 +0100)]
doc/go1.16: add time/tzdata release note for CL 261877

For #40700

Change-Id: I056cef20a5f071977d0ae589c7a50d5f69af3283
Reviewed-on: https://go-review.googlesource.com/c/go/+/273166
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
4 years agocmd/internal/objabi: assume GOARM=7 on Android
Elias Naur [Tue, 24 Nov 2020 16:48:38 +0000 (17:48 +0100)]
cmd/internal/objabi: assume GOARM=7 on Android

CL 34641 changed the Go runtime to assume GOARM=7 support on Android.
This change completes that by assuming GOARM=7 in the toolchain, fixing
the gotcha of inexplicably slow performance on non-arm64 Android devices.

There is already code in cmd/dist to force GOARM to 7 on GOOS=android. However,
dist is most likely run with GOOS != android.

Change-Id: I5e2bf11c3ecd0f6c193229eaa8ddc570722799d1
Reviewed-on: https://go-review.googlesource.com/c/go/+/272846
Run-TryBot: Elias Naur <mail@eliasnaur.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Trust: Elias Naur <mail@eliasnaur.com>

4 years agoruntime: check channel's elemsize before calling race detector
Daniel S Fava [Fri, 20 Nov 2020 20:23:45 +0000 (21:23 +0100)]
runtime: check channel's elemsize before calling race detector

When c.elemsize==0 we call raceacquire() and racerelease()
as opposed to calling racereleaseacquire()

The reason for this change is that, when elemsize==0, we don't
allocate a full buffer for the channel.  Instead of individual
buffer entries, the race detector uses the c.buf as the only
buffer entry.  This simplification prevents us following the
memory model's happens-before rules implemented in racereleaseacquire().
So, instead of calling racereleaseacquire(), we accumulate
happens-before information in the synchronization object associated
with c.buf.

The functionality in this change is implemented in a new function
called racenotify()

Fixes #42598

Change-Id: I75b92708633fdfde658dc52e06264e2171824e51
Reviewed-on: https://go-review.googlesource.com/c/go/+/271987
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Dmitry Vyukov <dvyukov@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Ian Lance Taylor <iant@golang.org>

4 years agoregexp/syntax: add note about Unicode character classes
Tom Payne [Fri, 23 Oct 2020 22:40:00 +0000 (23:40 +0100)]
regexp/syntax: add note about Unicode character classes

As proposed on golang-nuts:
https://groups.google.com/g/golang-nuts/c/M3lmSUptExQ/m/hRySV9GsCAAJ

Includes the latest updates from re2's mksyntaxgo:
https://code.googlesource.com/re2/+/refs/heads/master/doc/mksyntaxgo

Change-Id: Ib7b79aa6531f473feabd0a7f1d263cd65c4388e4
Reviewed-on: https://go-review.googlesource.com/c/go/+/264678
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Emmanuel Odeke <emmanuel@orijtech.com>

4 years agogo/constant: MakeFloat64(0) must return a value of Float kind
Robert Griesemer [Wed, 25 Nov 2020 04:28:20 +0000 (20:28 -0800)]
go/constant: MakeFloat64(0) must return a value of Float kind

Fixes #42641.

Change-Id: I10fdc7c90054b37ab5b303999015262691c12927
Reviewed-on: https://go-review.googlesource.com/c/go/+/273126
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agocmd/link: add relocation type R_AARCH64_LDST16_ABS_LO12_NC for arm64
eric fang [Wed, 18 Nov 2020 04:00:57 +0000 (04:00 +0000)]
cmd/link: add relocation type R_AARCH64_LDST16_ABS_LO12_NC for arm64

The linker already has R_AARCH64_LDST{8,32,64,128}_ABS_LO12_NC, some cgo tests require
 R_AARCH64_LDST16_ABS_LO12_NC, this CL adds this relocation type.

Fixes #42660

Change-Id: I9a5120cd872f5095c61175cb602427c6ab3225cc
Reviewed-on: https://go-review.googlesource.com/c/go/+/271017
Reviewed-by: eric fang <eric.fang@arm.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Run-TryBot: eric fang <eric.fang@arm.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: eric fang <eric.fang@arm.com>
Trust: Benny Siegert <bsiegert@gmail.com>

4 years agogo/constant: make constant.Make produce "smallest" const representation
Robert Griesemer [Wed, 25 Nov 2020 01:26:22 +0000 (17:26 -0800)]
go/constant: make constant.Make produce "smallest" const representation

Fixes #42640.

Change-Id: I22b8142b0a47a0f957d1bda28cdfdbb8388cffc4
Reviewed-on: https://go-review.googlesource.com/c/go/+/273086
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agoruntime: use inlined function name for traceback elision
Austin Clements [Fri, 20 Nov 2020 22:32:46 +0000 (17:32 -0500)]
runtime: use inlined function name for traceback elision

Currently, gentraceback decides which frames to print or elide when
unwinding inlined frames using only the name of the outermost
function. If the outermost function should be elided, then inlined
functions will also be elided, even if they shouldn't be.

This happens in practice in at least one situation. As of CL 258938,
exported Go functions (and functions they call) can now be inlined
into the generated _cgoexp_HASH_FN function. The runtime elides
_cgoexp_HASH_FN from tracebacks because it doesn't contain a ".".
Because of this bug, it also elides anything that was inlined into it.

This CL fixes this by synthesizing a funcInfo for the inlined
functions to pass to showframe.

Fixes #42754.

Change-Id: Ie6c663a4a1ac7f0d4beb1aa60bc26fc8cddd0f9d
Reviewed-on: https://go-review.googlesource.com/c/go/+/272131
Trust: Austin Clements <austin@google.com>
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agoruntime/testdata/testprogcgo: refactor CrashTraceback
Austin Clements [Fri, 20 Nov 2020 22:09:23 +0000 (17:09 -0500)]
runtime/testdata/testprogcgo: refactor CrashTraceback

This moves the C part of the CrashTraceback test into its own file in
preparation for adding a test that transitions back into Go.

Change-Id: I9560dcfd80bf8a1d30809fd360f958f5261ebb01
Reviewed-on: https://go-review.googlesource.com/c/go/+/272130
Trust: Austin Clements <austin@google.com>
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agodoc/go1.16: fill in Go 1.16 release note TODOs using relnote
Dmitri Shuralyov [Tue, 24 Nov 2020 18:23:10 +0000 (13:23 -0500)]
doc/go1.16: fill in Go 1.16 release note TODOs using relnote

The additions were generated using golang.org/x/build/cmd/relnote
at CL 272907. It was modified to find previously-missed entries
by querying the Gerrit API in addition to the maintner corpus.

For #40700.
Updates #41849.

Change-Id: If575984fe40e0133ad5e8fc5411ea5063457250d
Reviewed-on: https://go-review.googlesource.com/c/go/+/272871
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Carlos Amedee <carlos@golang.org>
4 years agoruntime: allow for usleep2HighRes to run without TLS setup
Alex Brainman [Sat, 21 Nov 2020 03:56:26 +0000 (14:56 +1100)]
runtime: allow for usleep2HighRes to run without TLS setup

This change adjusts usleep2HighRes so it does not crash when TLS is
not configured. When g is not available, usleep2HighRes just calls
usleep2 instead.

Updates #8687

Change-Id: Idbb80f7b71d1da350a6a7df7c49154eb1ffe29a8
Reviewed-on: https://go-review.googlesource.com/c/go/+/271907
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jason A. Donenfeld <Jason@zx2c4.com>
Reviewed-by: Simon Rozman <simon@rozman.si>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>
Trust: Alex Brainman <alex.brainman@gmail.com>

4 years agocmd/compile: set OpLoad argument type interface{} correctly
Cuong Manh Le [Mon, 23 Nov 2020 17:58:00 +0000 (00:58 +0700)]
cmd/compile: set OpLoad argument type interface{} correctly

CL 271906 allows loading single field of typed-interface{} OpIData, but
it does not update the corresponding selector type. So the generated
OpLoad has the named type instead, prevent it from being lowered by
lower pass.

Fixes #42784

Change-Id: Idf32e4f711731be09d508dd712b60bc8c58309bd
Reviewed-on: https://go-review.googlesource.com/c/go/+/272466
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agogo/types: fix incorrect string(int) conversion (regression)
Robert Griesemer [Mon, 23 Nov 2020 23:48:37 +0000 (15:48 -0800)]
go/types: fix incorrect string(int) conversion (regression)

The bug was introduced by https://golang.org/cl/220844.

Fixes #42790.

Change-Id: I44d619a1a4d3f2aee1c5575d5cfddcc4ba10895f
Reviewed-on: https://go-review.googlesource.com/c/go/+/272666
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agoruntime/metrics: tweak wording of stack and unused memory metrics
Michael Anthony Knyszek [Mon, 23 Nov 2020 19:25:32 +0000 (19:25 +0000)]
runtime/metrics: tweak wording of stack and unused memory metrics

This change tweaks and simplifies the descriptions of a couple metrics
to make them easier to parse (for humans).

Change-Id: I852654c7e7042c662ebdfa6334e3baf49ca4b33c
Reviewed-on: https://go-review.googlesource.com/c/go/+/272566
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agosync: use 386 instead of x86-32 to refer to the 32 bit x86 architecture
Martin Möhrmann [Sat, 21 Nov 2020 14:53:18 +0000 (15:53 +0100)]
sync: use 386 instead of x86-32 to refer to the 32 bit x86 architecture

This aligns the naming with GOARCH using 386 as a build target for
this architecture and makes it more easily found when searching
for documentation related to the build target.

Change-Id: I393bb89dd2f71e568124107b13e1b288fbd0c76a
Reviewed-on: https://go-review.googlesource.com/c/go/+/271988
Trust: Martin Möhrmann <moehrmann@google.com>
Run-TryBot: Martin Möhrmann <martisch@uos.de>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agocmd/compile: add test for 42753
Keith Randall [Fri, 20 Nov 2020 21:59:29 +0000 (13:59 -0800)]
cmd/compile: add test for 42753

This issue was already fixed at tip. Just adding the test that
failed on 1.14/1.15.

Update #42753

Change-Id: I00d13ade476b9c17190d762d7fdcb30cf6c83954
Reviewed-on: https://go-review.googlesource.com/c/go/+/272029
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
4 years agosyscall: add DLLError.Unwrap function
Jason A. Donenfeld [Fri, 13 Nov 2020 14:48:05 +0000 (15:48 +0100)]
syscall: add DLLError.Unwrap function

Because we're expecting for future functions to be unavailable, we
should add an Unwrap() function to the DLLError struct, so that people
can test for this situation easily via:

    if errors.Is(err, syscall.ERROR_PROC_NOT_FOUND) { ... }

DLLError already was wrapping the underlying Errno error, but never got
the Go 1.13 helper method.

Fixes golang/go#42584

Change-Id: I0f32a5146946b1b37a30897ba825a56faefc792c
Reviewed-on: https://go-review.googlesource.com/c/go/+/269761
Run-TryBot: Jason A. Donenfeld <Jason@zx2c4.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Trust: Alex Brainman <alex.brainman@gmail.com>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>

4 years agocmd/go/internal/modload: remove the Reqs function
Bryan C. Mills [Fri, 20 Nov 2020 22:03:42 +0000 (17:03 -0500)]
cmd/go/internal/modload: remove the Reqs function

The Reqs function returns an mvs.Reqs implemention for the global
build list. The API that it presents assumes that the build list is
globally consistent (problematic for #40775) and readily available
(problematic for #36460).

Fortunately, it is no longer used outside of the modload package.
We can instead use individual instances of the unexported mvsReqs
struct, making the dependency on the global build list more explicit.

For #36460
For #40775

Change-Id: I8674442f2a86416b0bf9c3395cb591c1e724c9d2
Reviewed-on: https://go-review.googlesource.com/c/go/+/272129
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go/internal/modload: remove a stale comment for EditBuildList
Bryan C. Mills [Fri, 20 Nov 2020 21:40:49 +0000 (16:40 -0500)]
cmd/go/internal/modload: remove a stale comment for EditBuildList

For #36460
Updates #37438

Change-Id: I1626d40e78b110035a893b1b80dbd2279bf50ffe
Reviewed-on: https://go-review.googlesource.com/c/go/+/272128
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go: support the -overlay flag for go mod commands
Michael Matloob [Fri, 20 Nov 2020 19:09:03 +0000 (14:09 -0500)]
cmd/go: support the -overlay flag for go mod commands

Move the declaration of the -overlay flag to base.AddModCommonFlags,
where other flags that are needed for go mod commands and for builds
are declared. The flag's already initialized in modload.Init so
there's no additional work needed to be done to support it in the go
mod commands.

For #39958

Change-Id: I70725d620cc69cb820f6ed923d626f4fe041b1c5
Reviewed-on: https://go-review.googlesource.com/c/go/+/272126
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/cgo, cmd/internal/pkgpath: support gofrontend mangler v3
Ian Lance Taylor [Fri, 20 Nov 2020 20:54:18 +0000 (12:54 -0800)]
cmd/cgo, cmd/internal/pkgpath: support gofrontend mangler v3

The gofrontend mangling scheme used by gccgo and GoLLVM has changed again.
Support the new version. This is a port of the relevant parts of
https://golang.org/cl/271726.

For #41862

Change-Id: I9c961c8e17ec960a83a23e1d49ea900962b63393
Reviewed-on: https://go-review.googlesource.com/c/go/+/272127
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agodoc: fix misspelling of “initialization” in diagnostics.html
Hollow Man [Fri, 20 Nov 2020 16:44:15 +0000 (16:44 +0000)]
doc: fix misspelling of “initialization” in diagnostics.html

initilization -> initialization

Change-Id: Ie5edd30559941f2d044280d8d586c2c2692d5b69
GitHub-Last-Rev: 7495a8c7227bc9c574c93861e5fedc1bada0397c
GitHub-Pull-Request: golang/go#42749
Reviewed-on: https://go-review.googlesource.com/c/go/+/272026
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
Trust: Ian Lance Taylor <iant@golang.org>

4 years agocmd/go: support overlaying go.mod files
Michael Matloob [Fri, 30 Oct 2020 21:11:36 +0000 (17:11 -0400)]
cmd/go: support overlaying go.mod files

This change updates the lockedfile package to open files using the
new fsys.OpenFile function. The logic of fsys.Open has been moved into
fsys.OpenFile, and fsys.Open is now just a light wrapper around it.

For #39958

Change-Id: I552f1a45ac00ac06b5812008d17a61e610b4b113
Reviewed-on: https://go-review.googlesource.com/c/go/+/266797
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/go: recommend 'go get' command to switch from retracted versions
Jay Conrod [Fri, 20 Nov 2020 16:18:15 +0000 (11:18 -0500)]
cmd/go: recommend 'go get' command to switch from retracted versions

This CL restores a message unintentionally removed in CL 270858.

For #24031

Change-Id: I957c5c59e624df98e72dfff351298bfc5bf9a9e7
Reviewed-on: https://go-review.googlesource.com/c/go/+/272066
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/compile: allow loading single field of typed-interface{} OpIData
Cuong Manh Le [Fri, 20 Nov 2020 07:09:01 +0000 (14:09 +0700)]
cmd/compile: allow loading single field of typed-interface{} OpIData

Same reason as CL 270057, but for OpLoad.

Fixes #42727

Change-Id: Iebb1a8110f29427a0aed3b5e3e84f0540de3d1b7
Reviewed-on: https://go-review.googlesource.com/c/go/+/271906
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
4 years agocmd/go: report changes and resolved versions in 'go get'
Jay Conrod [Wed, 18 Nov 2020 22:07:30 +0000 (17:07 -0500)]
cmd/go: report changes and resolved versions in 'go get'

Fixes #33284

Change-Id: I33daa5eb518985bc7308f29655e04c57e244b479
Reviewed-on: https://go-review.googlesource.com/c/go/+/269018
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/go/internal/modload: ignore selected version in checkRetractions
Jay Conrod [Fri, 13 Nov 2020 22:14:46 +0000 (17:14 -0500)]
cmd/go/internal/modload: ignore selected version in checkRetractions

Fixes #42601

Change-Id: I58d817ed34ccbd39591326c4bc23569f94028412
Reviewed-on: https://go-review.googlesource.com/c/go/+/272006
Run-TryBot: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Jay Conrod <jayconrod@google.com>

4 years agocmd/go: remove ListModules call in runGet
Jay Conrod [Thu, 19 Nov 2020 15:56:35 +0000 (10:56 -0500)]
cmd/go: remove ListModules call in runGet

ListModules was used to download .info files so that 'go list -m all'
would succeed later when offline. However, 'go list -m all' may
already fail when offline after 'go mod tidy', so it doesn't make
sense to add complexity to 'go get'.

Instead, remove the ListModules call and fix the test that
accidentally depended on it.

For #42723

Change-Id: I692597cf5ca15c23fa6fc9d2bac4b6e044299482
Reviewed-on: https://go-review.googlesource.com/c/go/+/271577
Trust: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/go/internal/modload: eliminate LoadedModules
Bryan C. Mills [Thu, 19 Nov 2020 18:58:52 +0000 (13:58 -0500)]
cmd/go/internal/modload: eliminate LoadedModules

As of CL 271646, all external callers have been eliminated. Replace
the remaining internal caller with a direct reference to the buildList
variable and remove the exported function to prevent backsliding.

For #36460

Change-Id: Iea82df1e3e604ada602dda3e830c06d441eee2a7
Reviewed-on: https://go-review.googlesource.com/c/go/+/271647
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go/internal/work: remove a redundant call to modload.LoadedModules
Bryan C. Mills [Thu, 19 Nov 2020 18:26:51 +0000 (13:26 -0500)]
cmd/go/internal/work: remove a redundant call to modload.LoadedModules

The modload.EditBuildList call added in CL 270980 already ensures that
installMod does not require a newer version of itself, so the condition
that this loop is checking for is redundant.

(I had meant for this change to be included in CL 270980, but
apparently somehow reverted it prior to mailing.)

For #36460

Change-Id: I4dd746b927f7012d950187cac9c510cd6fec8fd9
Reviewed-on: https://go-review.googlesource.com/c/go/+/271646
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
Trust: Bryan C. Mills <bcmills@google.com>

4 years agonet, internal/poll: reset value before adding in minor kernel version
Ian Lance Taylor [Fri, 20 Nov 2020 03:15:57 +0000 (19:15 -0800)]
net, internal/poll: reset value before adding in minor kernel version

Fixes #42733

Change-Id: I5446aeb5de13cd70212755fb12c9bc484f343c74
Reviewed-on: https://go-review.googlesource.com/c/go/+/271846
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agogo/types: use correct error position
Robert Griesemer [Fri, 20 Nov 2020 02:00:13 +0000 (18:00 -0800)]
go/types: use correct error position

Follow-up on https://golang.org/cl/271706 .

Change-Id: I90339987aed88b0de3ee7ebe7d413282055c260c
Reviewed-on: https://go-review.googlesource.com/c/go/+/271789
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agogo/types: fix error message for consistency
Robert Griesemer [Fri, 20 Nov 2020 00:38:11 +0000 (16:38 -0800)]
go/types: fix error message for consistency

Follow-up on https://golang.org/cl/271706 .
(Missed a review comment.)

Change-Id: Ibff542f43d721600a2452907c0a20941961e793f
Reviewed-on: https://go-review.googlesource.com/c/go/+/271766
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agogo/types, go/constant: handle infinities as unknown values
Robert Griesemer [Thu, 19 Nov 2020 20:40:19 +0000 (12:40 -0800)]
go/types, go/constant: handle infinities as unknown values

With this change, constant literals (and results of constant
operations) that internally become infinities are represented
externally (to go/constant) as "unknown" values.

The language has no provisions to deal with infinite constants,
and producing unknown values allows the typechecker to report
errors and avoid invalid operations (such as multiplication of
zero with infinity).

Fixes #20583.

Change-Id: I12f36a17d262ff7957b0d3880241b5a8b2984777
Reviewed-on: https://go-review.googlesource.com/c/go/+/271706
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agoio/fs: make WalkDirFunc parameter name consistent with doc comment
fzipp [Sun, 8 Nov 2020 13:50:30 +0000 (13:50 +0000)]
io/fs: make WalkDirFunc parameter name consistent with doc comment

The the DirEntry parameter of WalkDirFunc is referred to as `d` in the doc comment.

Change-Id: Ibfcf7908eaa0ef1309898150e8fd71101e7de09b
GitHub-Last-Rev: e858c52d81b93d293621d7e744bdcb7d6cbd412c
GitHub-Pull-Request: golang/go#42447
Reviewed-on: https://go-review.googlesource.com/c/go/+/268277
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years agoruntime/metrics: clarify Read's documentation
Michael Anthony Knyszek [Thu, 19 Nov 2020 15:07:06 +0000 (15:07 +0000)]
runtime/metrics: clarify Read's documentation

Change-Id: Idbcbc304f1568399a82af9dcd51e511393ed5ee0
Reviewed-on: https://go-review.googlesource.com/c/go/+/271558
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
4 years agoruntime/metrics: clarify memory and GC metrics documentation
Michael Anthony Knyszek [Thu, 19 Nov 2020 15:01:34 +0000 (15:01 +0000)]
runtime/metrics: clarify memory and GC metrics documentation

Change-Id: I8940990a591a808ddd4b8613531f52453f85bde1
Reviewed-on: https://go-review.googlesource.com/c/go/+/271557
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
4 years agocmd/go/internal/work: avoid modload.Selected in 'go install pkg@version'
Bryan C. Mills [Wed, 18 Nov 2020 18:44:39 +0000 (13:44 -0500)]
cmd/go/internal/work: avoid modload.Selected in 'go install pkg@version'

At this point in installOutsideModule the build list is empty, so
Selected trivially returns "none" for all modules.

(This change could have been made in CL 266657, but it was a bit
simpler to update the QueryPattern call sites mechanically to ensure
that there would be no unintentional semantic drift.)

For #36460

Change-Id: I44fb73794985bfeebb1dde0c092313f319c2945a
Reviewed-on: https://go-review.googlesource.com/c/go/+/271419
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go: fix failing gccgo cases in TestScript/build_overlay
Bryan C. Mills [Wed, 18 Nov 2020 03:54:13 +0000 (22:54 -0500)]
cmd/go: fix failing gccgo cases in TestScript/build_overlay

The 'go install' command does not support the -gccgo flag.
(I'm not sure why, but it doesn't.)

gccgo also uses system-native assembly syntax instead of cmd/compile's
Plan 9 derivative. I've added an assembly file that seems to work on
Linux, but I haven't tested it on other platforms; if it fails on
other platforms, we can refine the test as needed.

Fixes #42688

Change-Id: I0693a6a9eb58975f20cdc4160ef5f9a948563c88
Reviewed-on: https://go-review.googlesource.com/c/go/+/270978
Trust: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Michael Matloob <matloob@golang.org>
4 years agocmd/compile,cmd/asm: fix function pointer call perf regression on ppc64
Paul E. Murphy [Fri, 13 Nov 2020 20:44:25 +0000 (14:44 -0600)]
cmd/compile,cmd/asm: fix function pointer call perf regression on ppc64

by inserting hint when using bclrl.

Using this instruction as subroutine call is not the expected
default behavior, and as a result confuses the branch predictor.

The default expected behavior is a conditional return from a
subroutine.

We can change this assumption by encoding a hint this is not a
subroutine return.

The regex benchmarks are a pretty good example of how much this
hint can help generic ppc64le code on a power9 machine:

name                          old time/op    new time/op     delta
Find                             606ns ± 0%      447ns ± 0%  -26.27%
FindAllNoMatches                 309ns ± 0%      205ns ± 0%  -33.72%
FindString                       609ns ± 0%      451ns ± 0%  -26.04%
FindSubmatch                     734ns ± 0%      594ns ± 0%  -19.07%
FindStringSubmatch               706ns ± 0%      574ns ± 0%  -18.83%
Literal                          177ns ± 0%      136ns ± 0%  -22.89%
NotLiteral                      4.69µs ± 0%     2.34µs ± 0%  -50.14%
MatchClass                      6.05µs ± 0%     3.26µs ± 0%  -46.08%
MatchClass_InRange              5.93µs ± 0%     3.15µs ± 0%  -46.86%
ReplaceAll                      3.15µs ± 0%     2.18µs ± 0%  -30.77%
AnchoredLiteralShortNonMatch     156ns ± 0%      109ns ± 0%  -30.61%
AnchoredLiteralLongNonMatch      192ns ± 0%      136ns ± 0%  -29.34%
AnchoredShortMatch               268ns ± 0%      209ns ± 0%  -22.00%
AnchoredLongMatch                472ns ± 0%      357ns ± 0%  -24.30%
OnePassShortA                   1.16µs ± 0%     0.87µs ± 0%  -25.03%
NotOnePassShortA                1.34µs ± 0%     1.20µs ± 0%  -10.63%
OnePassShortB                    940ns ± 0%      655ns ± 0%  -30.29%
NotOnePassShortB                 873ns ± 0%      703ns ± 0%  -19.52%
OnePassLongPrefix                258ns ± 0%      155ns ± 0%  -40.13%
OnePassLongNotPrefix             943ns ± 0%      529ns ± 0%  -43.89%
MatchParallelShared              591ns ± 0%      436ns ± 0%  -26.31%
MatchParallelCopied              596ns ± 0%      435ns ± 0%  -27.10%
QuoteMetaAll                     186ns ± 0%      186ns ± 0%   -0.16%
QuoteMetaNone                   55.9ns ± 0%     55.9ns ± 0%   +0.02%
Compile/Onepass                 9.64µs ± 0%     9.26µs ± 0%   -3.97%
Compile/Medium                  21.7µs ± 0%     20.6µs ± 0%   -4.90%
Compile/Hard                     174µs ± 0%      174µs ± 0%   +0.07%
Match/Easy0/16                  7.35ns ± 0%     7.34ns ± 0%   -0.11%
Match/Easy0/32                   116ns ± 0%       97ns ± 0%  -16.27%
Match/Easy0/1K                   592ns ± 0%      562ns ± 0%   -5.04%
Match/Easy0/32K                 12.6µs ± 0%     12.5µs ± 0%   -0.64%
Match/Easy0/1M                   556µs ± 0%      556µs ± 0%   -0.00%
Match/Easy0/32M                 17.7ms ± 0%     17.7ms ± 0%   +0.05%
Match/Easy0i/16                 7.34ns ± 0%     7.35ns ± 0%   +0.10%
Match/Easy0i/32                 2.82µs ± 0%     1.64µs ± 0%  -41.71%
Match/Easy0i/1K                 83.2µs ± 0%     48.2µs ± 0%  -42.06%
Match/Easy0i/32K                2.13ms ± 0%     1.80ms ± 0%  -15.34%
Match/Easy0i/1M                 68.1ms ± 0%     57.6ms ± 0%  -15.31%
Match/Easy0i/32M                 2.18s ± 0%      1.80s ± 0%  -17.52%
Match/Easy1/16                  7.36ns ± 0%     7.34ns ± 0%   -0.24%
Match/Easy1/32                   118ns ± 0%       96ns ± 0%  -18.72%
Match/Easy1/1K                  2.46µs ± 0%     1.58µs ± 0%  -35.65%
Match/Easy1/32K                 80.2µs ± 0%     54.6µs ± 0%  -31.92%
Match/Easy1/1M                  2.75ms ± 0%     1.88ms ± 0%  -31.66%
Match/Easy1/32M                 87.5ms ± 0%     59.8ms ± 0%  -31.62%
Match/Medium/16                 7.34ns ± 0%     7.34ns ± 0%   +0.01%
Match/Medium/32                 2.60µs ± 0%     1.50µs ± 0%  -42.61%
Match/Medium/1K                 78.1µs ± 0%     43.7µs ± 0%  -44.06%
Match/Medium/32K                2.08ms ± 0%     1.52ms ± 0%  -27.11%
Match/Medium/1M                 66.5ms ± 0%     48.6ms ± 0%  -26.96%
Match/Medium/32M                 2.14s ± 0%      1.60s ± 0%  -25.18%
Match/Hard/16                   7.35ns ± 0%     7.35ns ± 0%   +0.03%
Match/Hard/32                   3.58µs ± 0%     2.44µs ± 0%  -31.82%
Match/Hard/1K                    108µs ± 0%       75µs ± 0%  -31.04%
Match/Hard/32K                  2.79ms ± 0%     2.25ms ± 0%  -19.30%
Match/Hard/1M                   89.4ms ± 0%     72.2ms ± 0%  -19.26%
Match/Hard/32M                   2.91s ± 0%      2.37s ± 0%  -18.60%
Match/Hard1/16                  11.1µs ± 0%      8.3µs ± 0%  -25.07%
Match/Hard1/32                  21.4µs ± 0%     16.1µs ± 0%  -24.85%
Match/Hard1/1K                   658µs ± 0%      498µs ± 0%  -24.27%
Match/Hard1/32K                 12.2ms ± 0%     11.7ms ± 0%   -4.60%
Match/Hard1/1M                   391ms ± 0%      374ms ± 0%   -4.40%
Match/Hard1/32M                  12.6s ± 0%      12.0s ± 0%   -4.68%
Match_onepass_regex/16           870ns ± 0%      611ns ± 0%  -29.79%
Match_onepass_regex/32          1.58µs ± 0%     1.08µs ± 0%  -31.48%
Match_onepass_regex/1K          45.7µs ± 0%     30.3µs ± 0%  -33.58%
Match_onepass_regex/32K         1.45ms ± 0%     0.97ms ± 0%  -33.20%
Match_onepass_regex/1M          46.2ms ± 0%     30.9ms ± 0%  -33.01%
Match_onepass_regex/32M          1.46s ± 0%      0.99s ± 0%  -32.02%

name                          old alloc/op   new alloc/op    delta
Find                             0.00B           0.00B         0.00%
FindAllNoMatches                 0.00B           0.00B         0.00%
FindString                       0.00B           0.00B         0.00%
FindSubmatch                     48.0B ± 0%      48.0B ± 0%    0.00%
FindStringSubmatch               32.0B ± 0%      32.0B ± 0%    0.00%
Compile/Onepass                 4.02kB ± 0%     4.02kB ± 0%    0.00%
Compile/Medium                  9.39kB ± 0%     9.39kB ± 0%    0.00%
Compile/Hard                    84.7kB ± 0%     84.7kB ± 0%    0.00%
Match_onepass_regex/16           0.00B           0.00B         0.00%
Match_onepass_regex/32           0.00B           0.00B         0.00%
Match_onepass_regex/1K           0.00B           0.00B         0.00%
Match_onepass_regex/32K          0.00B           0.00B         0.00%
Match_onepass_regex/1M           5.00B ± 0%      3.00B ± 0%  -40.00%
Match_onepass_regex/32M           136B ± 0%        68B ± 0%  -50.00%

name                          old allocs/op  new allocs/op   delta
Find                              0.00            0.00         0.00%
FindAllNoMatches                  0.00            0.00         0.00%
FindString                        0.00            0.00         0.00%
FindSubmatch                      1.00 ± 0%       1.00 ± 0%    0.00%
FindStringSubmatch                1.00 ± 0%       1.00 ± 0%    0.00%
Compile/Onepass                   52.0 ± 0%       52.0 ± 0%    0.00%
Compile/Medium                     112 ± 0%        112 ± 0%    0.00%
Compile/Hard                       424 ± 0%        424 ± 0%    0.00%
Match_onepass_regex/16            0.00            0.00         0.00%
Match_onepass_regex/32            0.00            0.00         0.00%
Match_onepass_regex/1K            0.00            0.00         0.00%
Match_onepass_regex/32K           0.00            0.00         0.00%
Match_onepass_regex/1M            0.00            0.00         0.00%
Match_onepass_regex/32M           2.00 ± 0%       1.00 ± 0%  -50.00%

name                          old speed      new speed       delta
QuoteMetaAll                  75.2MB/s ± 0%   75.3MB/s ± 0%   +0.15%
QuoteMetaNone                  465MB/s ± 0%    465MB/s ± 0%   -0.02%
Match/Easy0/16                2.18GB/s ± 0%   2.18GB/s ± 0%   +0.10%
Match/Easy0/32                 276MB/s ± 0%    330MB/s ± 0%  +19.46%
Match/Easy0/1K                1.73GB/s ± 0%   1.82GB/s ± 0%   +5.29%
Match/Easy0/32K               2.60GB/s ± 0%   2.62GB/s ± 0%   +0.64%
Match/Easy0/1M                1.89GB/s ± 0%   1.89GB/s ± 0%   +0.00%
Match/Easy0/32M               1.89GB/s ± 0%   1.89GB/s ± 0%   -0.05%
Match/Easy0i/16               2.18GB/s ± 0%   2.18GB/s ± 0%   -0.10%
Match/Easy0i/32               11.4MB/s ± 0%   19.5MB/s ± 0%  +71.48%
Match/Easy0i/1K               12.3MB/s ± 0%   21.2MB/s ± 0%  +72.62%
Match/Easy0i/32K              15.4MB/s ± 0%   18.2MB/s ± 0%  +18.12%
Match/Easy0i/1M               15.4MB/s ± 0%   18.2MB/s ± 0%  +18.12%
Match/Easy0i/32M              15.4MB/s ± 0%   18.6MB/s ± 0%  +21.21%
Match/Easy1/16                2.17GB/s ± 0%   2.18GB/s ± 0%   +0.24%
Match/Easy1/32                 271MB/s ± 0%    333MB/s ± 0%  +23.07%
Match/Easy1/1K                 417MB/s ± 0%    648MB/s ± 0%  +55.38%
Match/Easy1/32K                409MB/s ± 0%    600MB/s ± 0%  +46.88%
Match/Easy1/1M                 381MB/s ± 0%    558MB/s ± 0%  +46.33%
Match/Easy1/32M                383MB/s ± 0%    561MB/s ± 0%  +46.25%
Match/Medium/16               2.18GB/s ± 0%   2.18GB/s ± 0%   -0.01%
Match/Medium/32               12.3MB/s ± 0%   21.4MB/s ± 0%  +74.13%
Match/Medium/1K               13.1MB/s ± 0%   23.4MB/s ± 0%  +78.73%
Match/Medium/32K              15.7MB/s ± 0%   21.6MB/s ± 0%  +37.23%
Match/Medium/1M               15.8MB/s ± 0%   21.6MB/s ± 0%  +36.93%
Match/Medium/32M              15.7MB/s ± 0%   21.0MB/s ± 0%  +33.67%
Match/Hard/16                 2.18GB/s ± 0%   2.18GB/s ± 0%   -0.03%
Match/Hard/32                 8.93MB/s ± 0%  13.10MB/s ± 0%  +46.70%
Match/Hard/1K                 9.48MB/s ± 0%  13.74MB/s ± 0%  +44.94%
Match/Hard/32K                11.7MB/s ± 0%   14.5MB/s ± 0%  +23.87%
Match/Hard/1M                 11.7MB/s ± 0%   14.5MB/s ± 0%  +23.87%
Match/Hard/32M                11.6MB/s ± 0%   14.2MB/s ± 0%  +22.86%
Match/Hard1/16                1.44MB/s ± 0%   1.93MB/s ± 0%  +34.03%
Match/Hard1/32                1.49MB/s ± 0%   1.99MB/s ± 0%  +33.56%
Match/Hard1/1K                1.56MB/s ± 0%   2.05MB/s ± 0%  +31.41%
Match/Hard1/32K               2.68MB/s ± 0%   2.80MB/s ± 0%   +4.48%
Match/Hard1/1M                2.68MB/s ± 0%   2.80MB/s ± 0%   +4.48%
Match/Hard1/32M               2.66MB/s ± 0%   2.79MB/s ± 0%   +4.89%
Match_onepass_regex/16        18.4MB/s ± 0%   26.2MB/s ± 0%  +42.41%
Match_onepass_regex/32        20.2MB/s ± 0%   29.5MB/s ± 0%  +45.92%
Match_onepass_regex/1K        22.4MB/s ± 0%   33.8MB/s ± 0%  +50.54%
Match_onepass_regex/32K       22.6MB/s ± 0%   33.9MB/s ± 0%  +49.67%
Match_onepass_regex/1M        22.7MB/s ± 0%   33.9MB/s ± 0%  +49.27%
Match_onepass_regex/32M       23.0MB/s ± 0%   33.9MB/s ± 0%  +47.14%

Fixes #42709

Change-Id: Ice07fec2de4c5b1302febf8c2978ae8c1e4fd3e5
Reviewed-on: https://go-review.googlesource.com/c/go/+/271337
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
Trust: Carlos Eduardo Seo <carlos.seo@linaro.org>

4 years agocmd/compile: flag "-d=dumpptrs" to print Node ptrs in Dump output
Dan Scales [Thu, 19 Nov 2020 00:42:31 +0000 (16:42 -0800)]
cmd/compile:  flag "-d=dumpptrs" to print Node ptrs in Dump output

The printing of the ptr values can mean that two dump outputs can't easily be
compared for the identical structure, so adding the "-d=dumpptrs" option to make
printing of Node pointer values be an option.

Change-Id: I0e92b02f069e9de2e6fa036a7841645d13cdd7a9
Reviewed-on: https://go-review.googlesource.com/c/go/+/271339
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agoruntime: support new callbackasm1 calling convention on windows/arm
Jason A. Donenfeld [Wed, 18 Nov 2020 20:38:57 +0000 (21:38 +0100)]
runtime: support new callbackasm1 calling convention on windows/arm

This updates the callbacks implementation on windows/arm for the
changes made in CL 258938. At the time, that was left as a TODO.

At the same time, it also extends the previous support for only 4
arguments to also support additional arguments on the stack. This is
required for functions like SetWinEventHook, which take 7 arguments. It
does this by pushing r0-r3 onto the stack before the normal prologue,
and then pointing the args struct to that location.

This is derived from CL 270077 and CL 270078.

Updates #40724.
Fixes #42591.

Change-Id: Icc199e7f2c24205e41be4e00015283c7e2a9b797
Reviewed-on: https://go-review.googlesource.com/c/go/+/271178
Run-TryBot: Jason A. Donenfeld <Jason@zx2c4.com>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>
Reviewed-by: Austin Clements <austin@google.com>
4 years agocmd/go/internal/modload: remove SetBuildList
Bryan C. Mills [Wed, 18 Nov 2020 04:16:30 +0000 (23:16 -0500)]
cmd/go/internal/modload: remove SetBuildList

For the last remaining call site (in cmd/go/internal/work, added for
the new 'go install pkg@version' codepath in CL 254365), use
EditBuildList instead.

SetBuildList assumes that the caller has enough information to produce
a complete, coherent build list. With lazy loading, producing a
complete, coherent build list is no longer quite so trivial.

In CL 263267, I rewrote the main caller of SetBuildList (the 'go get'
command), and in the process added a more targeted modload hook
(EditBuildList). That hook also suffices for 'go install pkg@version'.
The resulting error messages are perhaps not as smooth as they ought
to be, but if they are too awkward we should probably fix them for
'go get' too, and the commands can continue to share the edit hook.

For #36460
Updates #40276

Change-Id: I698a9dcd2efe6378a4d91f21362880aa8e50001b
Reviewed-on: https://go-review.googlesource.com/c/go/+/270980
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go/internal/modcmd: eliminate a call to modload.LoadedModules
Bryan C. Mills [Wed, 18 Nov 2020 03:13:35 +0000 (22:13 -0500)]
cmd/go/internal/modcmd: eliminate a call to modload.LoadedModules

modload.LoadedModules reveals more information than necessary about
whether modules have been loaded lazily. The 'vendor' subcommand
doesn't actually need that much information: it has all of the
information that it needs from prior calls to LoadPackages and
ModFile.

For #36460

Change-Id: If08733cca930b2b80616b037b63985ecfd6a320b
Reviewed-on: https://go-review.googlesource.com/c/go/+/270979
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agointernal/fmtsort: sort the unsafe pointers in map
Hanlin Shi [Mon, 16 Nov 2020 17:29:33 +0000 (17:29 +0000)]
internal/fmtsort: sort the unsafe pointers in map

Currently storing keys that contain unsafe.
Pointer in a map could result inruntime panic when printing the map.
The root cause is that unsafe.Pointer is not comparable.

Fixes #42622.

Change-Id: Ie3bae7ee4945041843b66514de6227212a3da73e
GitHub-Last-Rev: d12d41302e6118cb457aafb05f7aaed9df259b56
GitHub-Pull-Request: golang/go#42623
Reviewed-on: https://go-review.googlesource.com/c/go/+/270277
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agogo/types: report an error for invalid constant values
Robert Griesemer [Thu, 19 Nov 2020 01:05:02 +0000 (17:05 -0800)]
go/types: report an error for invalid constant values

The parser reports syntactic errors in constant literals.
The go/constant package produces an "unknown" value for
syntactically correct numeric constants that are too small
or too large. Check for the unknown value and report an
error rather than silently continuing.

Fixes #42695.

Change-Id: I414214559a285d67ed50184dc750f106960b5620
Reviewed-on: https://go-review.googlesource.com/c/go/+/271377
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/compile: fix miscompilation during inlining
Matthew Dempsky [Wed, 18 Nov 2020 20:50:46 +0000 (12:50 -0800)]
cmd/compile: fix miscompilation during inlining

When inlining a function call expression, it's possible that the
function callee subexpression has side effects that need to be
preserved. This used to not be an issue, because inlining wouldn't
recognize these as inlinable anyway. But golang.org/cl/266199 extended
the inlining logic to recognize more cases, but did not notice that
the actual inlining code was discarding side effects.

Issue identified by danscales@.

Fixes #42703.

Change-Id: I95f8fc076b6ca4e9362e80ec26dad9d87a5bc44a
Reviewed-on: https://go-review.googlesource.com/c/go/+/271219
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years agocmd/compile: fix panic in field tracking logic
Matthew Dempsky [Wed, 18 Nov 2020 20:08:59 +0000 (12:08 -0800)]
cmd/compile: fix panic in field tracking logic

Within the frontend, we generally don't guarantee uniqueness of
anonymous types. For example, each struct type literal gets
represented by its own types.Type instance.

However, the field tracking code was using the struct type as a map
key. This broke in golang.org/cl/256457, because that CL started
changing the inlined parameter variables from using the types.Type of
the declared parameter to that of the call site argument. These are
always identical types (e.g., types.Identical would report true), but
they can be different pointer values, causing the map lookup to fail.

The easiest fix is to simply get rid of the map and instead use
Node.Opt for tracking the types.Field. To mitigate against more latent
field tracking failures (e.g., if any other code were to start trying
to use Opt on ODOT/ODOTPTR fields), we store this field
unconditionally. I also expect having the types.Field will be useful
to other frontend code in the future.

Finally, to make it easier to test field tracking without having to
run make.bash with GOEXPERIMENT=fieldtrack, this commit adds a
-d=fieldtrack flag as an alternative way to enable field tracking
within the compiler. See also #42681.

Fixes #42686.

Change-Id: I6923d206d5e2cab1e6798cba36cae96c1eeaea55
Reviewed-on: https://go-review.googlesource.com/c/go/+/271217
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years agosync: document RWMutex race semantics
Michael Pratt [Tue, 17 Nov 2020 21:47:08 +0000 (16:47 -0500)]
sync: document RWMutex race semantics

RWMutex provides explicit acquire/release synchronization events to the
race detector to model the mutex. It disables sync events within the
methods to avoid e.g., the atomics from adding false synchronization
events, which could cause false negatives in the race detector.

Change-Id: I5126ce2efaab151811ac264864aab1fa025a4aaf
Reviewed-on: https://go-review.googlesource.com/c/go/+/270865
Run-TryBot: Michael Pratt <mpratt@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Trust: Michael Pratt <mpratt@google.com>

4 years agoruntime: give test child time to block
Michael Pratt [Mon, 9 Nov 2020 21:37:05 +0000 (16:37 -0500)]
runtime: give test child time to block

The child in TestPanicSystemstack prints "x\n" and then blocks on a
lock. Receiving those bytes only indicates that the child is _about to
block_. Since we don't have a way to know when it is fully blocked,
sleep a bit to give it time to block. This makes us less likely to lose
the race and signal before the child blocks, which will fail the test as
the stack trace cannot be read from a running G.

Fixes #33626

Change-Id: I8a27b1b114bf75e1e5bcb2a7a33aa69cdbc22f40
Reviewed-on: https://go-review.googlesource.com/c/go/+/268578
Trust: Michael Pratt <mpratt@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
4 years agoruntime: use clock_gettime instead of gettimeofday on darwin
Tobias Klauser [Wed, 18 Nov 2020 10:08:43 +0000 (11:08 +0100)]
runtime: use clock_gettime instead of gettimeofday on darwin

clock_gettime has higher resolution than gettimeofday and is available
since macOS 10.12. Go 1.15 already requires at least macOS 10.12 and
thus clock_gettime can be used unconditionally (also see
https://golang.org/doc/go1.15#darwin)

Fixes #25633

Change-Id: I46305387212735e5d3a13e5f02ec90f3e6d546a4
Reviewed-on: https://go-review.googlesource.com/c/go/+/270918
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agoruntime: use pipe2 syscall for Pipe in tests
Tobias Klauser [Wed, 18 Nov 2020 07:53:00 +0000 (08:53 +0100)]
runtime: use pipe2 syscall for Pipe in tests

On FreeBSD >= 11 with a kernel built with COMPAT_FREEBSD11 but not
COMPAT_FREEBSD10, the pipe syscall is not available. Thus, tests using
runtime.pipe fail with ENOSYS. As suggested by Ian, fix this by calling
pipe2(0) in these tests and fall back to pipe() in case of ENOSYS.

Fixes #42659

Change-Id: Ifbb8008884b7901fe87830d162ad326122c5fab9
Reviewed-on: https://go-review.googlesource.com/c/go/+/270917
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agocmd/go: in 'go get', only load retractions for resolved versions
Jay Conrod [Tue, 10 Nov 2020 20:48:37 +0000 (15:48 -0500)]
cmd/go: in 'go get', only load retractions for resolved versions

Previously, 'go get' loaded retractions for every module in the build
list, which took a long time and usually wasn't helpful.

This rolls forward CL 269019, which was reverted in CL 270521. The new
revision adds a call to modload.ListModules at the end of 'go get' to
ensure .info files are cached for everything in the build list.

Fixes #42185

Change-Id: I684f66c5e674384d5a0176fbc8317e5530b8a915
Reviewed-on: https://go-review.googlesource.com/c/go/+/270858
Trust: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/link: recompute heapPos after copyHeap
Cherry Zhang [Wed, 18 Nov 2020 00:10:51 +0000 (19:10 -0500)]
cmd/link: recompute heapPos after copyHeap

Immediately after a forward Seek, the offset we're writing to is
beyond len(buf)+len(heap):

|<--- buf --->|<--- heap --->|
                                    ^
                                    off

If we do a copyHeap at this point, the new heapPos should not be
0:

|<---------- buf ----------->|<-heap->|
                                    ^
                                    off

Recompute it.

For #42082.

Change-Id: Icb3e4e1c7bf7d1fd3d76a2e0d7dfcb319c661534
Reviewed-on: https://go-review.googlesource.com/c/go/+/270941
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
4 years agocmd/go: fix retract interval syntax in 'go help mod edit'
Jay Conrod [Mon, 16 Nov 2020 23:31:22 +0000 (18:31 -0500)]
cmd/go: fix retract interval syntax in 'go help mod edit'

For #24031

Change-Id: I70461431aac24c9465b9bdab082bcc34343a53a8
Reviewed-on: https://go-review.googlesource.com/c/go/+/270557
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Trust: Jay Conrod <jayconrod@google.com>

4 years agocmd/compile: stop MOVW-ing -1 as SRA shift amount in mips
Alberto Donizetti [Tue, 17 Nov 2020 09:13:29 +0000 (10:13 +0100)]
cmd/compile: stop MOVW-ing -1 as SRA shift amount in mips

The shift amount in SRAconst needs to be in the [0,31] range, so stop
MOVWing -1 to SRA in the Rsh lowering rules.

Also see CL 270117.

Passes

  $ GOARCH=mips go build -toolexec 'toolstash -cmp' -a std
  $ GOARCH=mipsle go build -toolexec 'toolstash -cmp' -a std

Updates #42587

Change-Id: Ib5eb99b82310e404cc2d6f0c619b21b8a15406ce
Reviewed-on: https://go-review.googlesource.com/c/go/+/270558
Trust: Alberto Donizetti <alb.donizetti@gmail.com>
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agogo/ast: document invalid comment end positions with CRLF line endings
Rebecca Stambler [Tue, 17 Nov 2020 23:28:38 +0000 (18:28 -0500)]
go/ast: document invalid comment end positions with CRLF line endings

We've decided that issues like golang/go#41197 are unfixable, so
instead, document the bug.

Fixes golang/go#41197

Change-Id: I5649027f6e2445eec765516f2f642db0d601ea20
Reviewed-on: https://go-review.googlesource.com/c/go/+/270938
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
4 years agotime: in NewTicker, document that the 1st tick comes after d
Alberto Donizetti [Wed, 28 Oct 2020 09:48:17 +0000 (10:48 +0100)]
time: in NewTicker, document that the 1st tick comes after d

Fixes #42245

Change-Id: I3b298ab6be65569389873d68bd3c6e49cf892c69
Reviewed-on: https://go-review.googlesource.com/c/go/+/265818
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Rob Pike <r@golang.org>
Trust: Alberto Donizetti <alb.donizetti@gmail.com>

4 years agogo/types: add test case for incorrect map index expression
Robert Griesemer [Tue, 17 Nov 2020 06:23:48 +0000 (22:23 -0800)]
go/types: add test case for incorrect map index expression

The existing code for map index expressions checked the
wrong variable (x rather than key) to see if the index
assignment was correct. Since x.mode was always valid in
that case, type-checking didn't follow the error exit in
case of an incorrect map index expression.

However, since we know the correct map element type
irrespective of the validity of the map key, the existing
code path is preferrable over exiting early via an error
because the map index expression returns a valid type which
then can be used for further type-checking.

Removed the unneeded 'if' statement and added a test case
producing the expected two errors (rather than only one if
we would "correct" the 'if' statement instead).

In summary, this commit adds a test but doesn't change the
behavior of type-checking of map index expressions.

Change-Id: I67845bfaa03600c9400f9a1462d7a68a66921ad4
Reviewed-on: https://go-review.googlesource.com/c/go/+/270658
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
4 years agocmd/compile: clean up buggy DWARF inlined info PC ranges
Than McIntosh [Mon, 17 Aug 2020 18:17:07 +0000 (14:17 -0400)]
cmd/compile: clean up buggy DWARF inlined info PC ranges

Repair the code that generates PC ranges for DWARF inlined routine
instances to insure that if II Y is a child of II X within the inline
tree, X's ranges include the ranges from Y. This is similar to what
we're already doing for DWARF scopes.

Updates #33188.

Change-Id: I9bb552777fcd1ae93dc01872707667ad092b1dd9
Reviewed-on: https://go-review.googlesource.com/c/go/+/248724
Run-TryBot: Than McIntosh <thanm@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: David Chase <drchase@google.com>
Trust: Than McIntosh <thanm@google.com>

4 years agocmd/go: allow querying other versions of the main module
Jay Conrod [Mon, 16 Nov 2020 21:27:19 +0000 (16:27 -0500)]
cmd/go: allow querying other versions of the main module

'go mod download' and a few other commands can now query specific
versions of the main module.

'go get' still reports an error when attempting to update the main
module.

Fixes #42524

Change-Id: Ia93ef8f5f34443e938667c48a0db432200108c63
Reviewed-on: https://go-review.googlesource.com/c/go/+/270520
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/go/internal/modget: clarify error for 'go get' without arguments
Jay Conrod [Wed, 20 May 2020 17:51:53 +0000 (13:51 -0400)]
cmd/go/internal/modget: clarify error for 'go get' without arguments

If the current directory doesn't contain a package, 'go get' will say
that without additional detail.

If there were no arguments, errors will start with "go get:" instead
of "go get .:".

Fixes #39080

Change-Id: I47366f2a27bce17bd8b79344ad15b8b934a888c9
Reviewed-on: https://go-review.googlesource.com/c/go/+/234681
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Trust: Jay Conrod <jayconrod@google.com>

4 years agocmd/go: revert "in 'go get', only load retractions for resolved versions"
Dmitri Shuralyov [Tue, 17 Nov 2020 02:04:34 +0000 (02:04 +0000)]
cmd/go: revert "in 'go get', only load retractions for resolved versions"

This reverts CL 269019.

Reason for revert: The TestScript/mod_gonoproxy test is failing
on linux-386-longtest and linux-amd64-longtest builders.

Change-Id: I7e132fb4fb5a9c00add28e5100a0e96a9250282c
Reviewed-on: https://go-review.googlesource.com/c/go/+/270521
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/compile: fix rules regression with shifts on PPC64
Lynn Boger [Mon, 16 Nov 2020 14:40:45 +0000 (09:40 -0500)]
cmd/compile: fix rules regression with shifts on PPC64

Some rules for PPC64 were checking for a case
where a shift followed by an 'and' of a mask could
be lowered, depending on the format of the mask. The
function to verify if the mask was valid for this purpose
was not checking if the mask was 0 which we don't want to
allow. This case can happen if previous optimizations
resulted in that mask value.

This fixes isPPC64ValidShiftMask to check for a mask of 0 and return
false.

This also adds a codegen testcase to verify it doesn't try to
match the rules in the future.

Fixes #42610

Change-Id: I565d94e88495f51321ab365d6388c01e791b4dbb
Reviewed-on: https://go-review.googlesource.com/c/go/+/270358
Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Paul Murphy <murp@ibm.com>
Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.org>
Trust: Lynn Boger <laboger@linux.vnet.ibm.com>

4 years agocmd/go: update 'go help mod init'
Jay Conrod [Mon, 12 Oct 2020 16:03:43 +0000 (12:03 -0400)]
cmd/go: update 'go help mod init'

'go help mod init' now mentions that the module path can be derived
from the directory within GOPATH. We no longer mention version
control, since that's now ignored.

Fixes #36775

Change-Id: Ia5559ecb537fccd838eeab84517e76aa01989292
Reviewed-on: https://go-review.googlesource.com/c/go/+/261539
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Michael Matloob <matloob@golang.org>
4 years agocmd/go: in 'go get', only load retractions for resolved versions
Jay Conrod [Tue, 10 Nov 2020 20:48:37 +0000 (15:48 -0500)]
cmd/go: in 'go get', only load retractions for resolved versions

Previously, 'go get' loaded retractions for every module in the build
list, which took a long time and usually wasn't helpful.

Fixes #42185

Change-Id: I64294585db141106b63ec74aafa0d266b7536ef2
Reviewed-on: https://go-review.googlesource.com/c/go/+/269019
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Trust: Jay Conrod <jayconrod@google.com>

4 years agocmd/link/internal/ld: dedup shared libraries on openbsd
Joel Sing [Wed, 11 Nov 2020 17:30:15 +0000 (04:30 +1100)]
cmd/link/internal/ld: dedup shared libraries on openbsd

When linking internally on OpenBSD, dedup libraries treating versioned
and unversioned libraries as equivalents. Versioned libraries are preferred
and are retained over unversioned libraries.

This avoids the situation where the use of cgo results in a DT_NEEDED for a
versioned library (for example, libc.so.96.1), while a dynamic import
specifies an unversioned library (for example, libc.so). Without deduplication
this would result in two DT_NEEDED entries, causing a failure when ld.so
attempts to load the Go binrary.

Updates #36435
Fixes #39257

Change-Id: I4a4942f259dece01d97bb51df9e13d67c9f94d34
Reviewed-on: https://go-review.googlesource.com/c/go/+/249978
Trust: Joel Sing <joel@sing.id.au>
Run-TryBot: Joel Sing <joel@sing.id.au>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agoruntime/race: reject runtime fatal error in tests
Cherry Zhang [Sat, 14 Nov 2020 19:29:31 +0000 (14:29 -0500)]
runtime/race: reject runtime fatal error in tests

We expect those tests to fail with non-zero exit code, due to
intentional races, but we don't expect the runtime to crash.
Reject that.

Change-Id: Ic37987dabecde5f0703c031c49ce7f884a7b06a5
Reviewed-on: https://go-review.googlesource.com/c/go/+/270398
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agoruntime: declare arg size/map for race version of sync/atomic functions
Cherry Zhang [Sat, 14 Nov 2020 02:08:26 +0000 (21:08 -0500)]
runtime: declare arg size/map for race version of sync/atomic functions

The argument size and map are used in stack scanning if those
functions are deferred. Declare the right argument size and map
so they can be scanned correctly.

Fixes #42599.

Change-Id: I74f9409d574cf7c383f4d8f83e38521026b48861
Reviewed-on: https://go-review.googlesource.com/c/go/+/270079
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
4 years agocmd/go/internal/work: add missing newline to go version note
Bryan C. Mills [Thu, 1 Oct 2020 14:53:41 +0000 (10:53 -0400)]
cmd/go/internal/work: add missing newline to go version note

A missed newline was added for one case in CL 162957, but
the parallel no-output case was missed.

Add the missed newline for the second case and update the test to
cover the full line for both cases.

Updates #30263

Change-Id: I02aa523290295a6d409cd68066b45c6990e6fb6e
Reviewed-on: https://go-review.googlesource.com/c/go/+/258758
Reviewed-by: Jay Conrod <jayconrod@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Bryan C. Mills <bcmills@google.com>

4 years agocmd/go: permit wrongly rejected -Wl,-O... linker flags
Kai Lüke [Mon, 16 Nov 2020 12:05:13 +0000 (12:05 +0000)]
cmd/go: permit wrongly rejected -Wl,-O... linker flags

A typo caused the validation rule to check against -WL,-O... which is
not a regular flag because the L should be lowercase as in the other
rules. This caused valid linker flags to be rejected and people had to
work around this by filtering their default flags that include, e.g.,
-Wl,-O1 for a simple link optimization.
Fix the typo that wrongly rejected -Wl,-O... but allowed a non-existing
-WL,-O flag.

Change-Id: Ia3bf730f16f5ad98a39d7f17159de17b44075462
GitHub-Last-Rev: 2ec7f2a2b92c9c76e707a71f6c5273aa1d512006
GitHub-Pull-Request: golang/go#42631
Reviewed-on: https://go-review.googlesource.com/c/go/+/270278
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Jay Conrod <jayconrod@google.com>

4 years agocmd/compile: mask SLL,SRL,SRAconst shift amount
Alberto Donizetti [Sat, 14 Nov 2020 12:33:32 +0000 (13:33 +0100)]
cmd/compile: mask SLL,SRL,SRAconst shift amount

mips SRA/SLL/SRL shift amounts are used mod 32; this change aligns the
XXXconst rules to mask the shift amount by &31.

Passes

  $ GOARCH=mips go build -toolexec 'toolstash -cmp' -a std
  $ GOARCH=mipsle go build -toolexec 'toolstash -cmp' -a std

Fixes #42587

Change-Id: I6003ebd0bc500fba4cf6fb10254e1b557bf8c48f
Reviewed-on: https://go-review.googlesource.com/c/go/+/270117
Trust: Alberto Donizetti <alb.donizetti@gmail.com>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agocmd/compile: fix load of interface{}-typed OpIData in expand_calls
David Chase [Fri, 13 Nov 2020 21:54:48 +0000 (16:54 -0500)]
cmd/compile: fix load of interface{}-typed OpIData in expand_calls

In certain cases, the declkared type of an OpIData is interface{}.
This was not expected (since interface{} is a pair, right?) and
thus caused a crash.  What is intended is that these be treated as
a byteptr, so do that instead (this is what happens in 1.15).

Fixes #42568.

Change-Id: Id7c9e5dc2cbb5d7c71c6748832491ea62b0b339f
Reviewed-on: https://go-review.googlesource.com/c/go/+/270057
Trust: David Chase <drchase@google.com>
Run-TryBot: David Chase <drchase@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
4 years agocmd/go: permit CGO_LDFLAGS to appear in //go:ldflag
Ian Lance Taylor [Fri, 13 Nov 2020 19:05:37 +0000 (11:05 -0800)]
cmd/go: permit CGO_LDFLAGS to appear in //go:ldflag

Fixes #42565

Change-Id: If7cf39905d124dbd54dfac6a53ee38270498efed
Reviewed-on: https://go-review.googlesource.com/c/go/+/269818
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/compile: update comments only for Node types and some functions
Dan Scales [Thu, 12 Nov 2020 17:33:34 +0000 (09:33 -0800)]
cmd/compile:  update comments only for Node types and some functions

Improve the comments in syntax.go on Node structs and constants. Also, updated a
few function header comments.

Change-Id: I3e6e4a3c5678fc0b4e18844507b3460303ce1240
Reviewed-on: https://go-review.googlesource.com/c/go/+/269538
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Trust: Dan Scales <danscales@google.com>

4 years agocmd/compile: mark plugin-exported types as used in interface
Cherry Zhang [Fri, 13 Nov 2020 19:42:45 +0000 (14:42 -0500)]
cmd/compile: mark plugin-exported types as used in interface

Plugin exports symbols as interfaces. Mark their types as used in
interfaces, so their methods will be kept alive by the linker.

Fixes #42579.

Change-Id: If1b5aacc21510c20c25f88bb131bca61db6f1d56
Reviewed-on: https://go-review.googlesource.com/c/go/+/269819
Trust: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
4 years agocmd/cgo: fix initialization of empty argument types
Austin Clements [Wed, 11 Nov 2020 20:33:39 +0000 (15:33 -0500)]
cmd/cgo: fix initialization of empty argument types

CL 258938 changed the way C to Go calls work such that they now
construct a C struct on the C side for the arguments and space for the
results. Any pointers in the result space must be zeroed, so we just
zero the whole struct.

However, C makes it surprisingly hard to robustly zero any struct
type. We had used a "{0}" initializer, which works in the vast
majority of cases, but fails if the type is empty or effectively
empty.

This CL fixes this by changing how the cgo tool zero-initializes the
argument struct to be more robust.

Fixes #42495.

Change-Id: Id1749b9d751e59eb7a02a9d44fec0698a2bf63cd
Reviewed-on: https://go-review.googlesource.com/c/go/+/269337
Trust: Austin Clements <austin@google.com>
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agoruntime: swap the order of raceacquire() and racerelease()
Daniel S Fava [Thu, 12 Nov 2020 16:25:47 +0000 (17:25 +0100)]
runtime: swap the order of raceacquire() and racerelease()

In chansend() and chanrecv() of chan.go, the order of calls to
raceacquire() and racerelease() was swapped, which meant that the
code was not following the memory model "by the letter of the law."
Similar for bufrecv and bufsend in select.go

The memory model says:

- A send happens before the corresponding receive completes, and
- the kth receive on a channel with capacity C happens before the
k+C send on that channel completes.

The operative word here is "completes."  For example, a sender obtains
happens-before information on completion of the send-operation, which
means, after the sender has deposited its message onto the channel.
Similarly for receives.

If the order of raceacquire() and racerelease() is incorrect, the race
detector may fail to report some race conditions.

The fix is minimal from the point of view of Go.  The fix does, however,
rely on a new function added to TSan:

https://reviews.llvm.org/D76322

This commit only affects execution when race detection is enabled.

Added two tests into `runtime/race/output_test.go`:

- `chanmm` tests for the issue addressed by this patch
- `mutex` is a test for inverted semaphores, which must not be broken
  by this (or any other) patch

Fixes #37355

Change-Id: I5e886879ead2bd456a4b7dd1d17253641b767f63
Reviewed-on: https://go-review.googlesource.com/c/go/+/220419
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
4 years agosyscall: use correct type for TIOCSPGRP/TIOCGPGRP
Joel Sing [Mon, 2 Nov 2020 15:11:51 +0000 (02:11 +1100)]
syscall: use correct type for TIOCSPGRP/TIOCGPGRP

These ioctls take a pid_t (generally a C integer aka int32) and not an int64 - we
currently get away with this on little endian 64 bit platforms, since the bytes
fall into the correct place, however this breaks on big endian 64 bit platforms
(like openbsd/mips64).

Update #40995

Change-Id: I622a0543fd562d97f76a7376a84fd2641e6d6a24
Reviewed-on: https://go-review.googlesource.com/c/go/+/267605
Trust: Joel Sing <joel@sing.id.au>
Run-TryBot: Joel Sing <joel@sing.id.au>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agocmd/go: use overlaid path contents in build cache
Michael Matloob [Fri, 30 Oct 2020 20:47:58 +0000 (16:47 -0400)]
cmd/go: use overlaid path contents in build cache

When caching actions, use the overlaid file contents, because those
are the ones actually used to produce the outputs.

For #39958

Change-Id: Ia1f85b2fcf1f26e3b5be82f4d35c2726b134a36b
Reviewed-on: https://go-review.googlesource.com/c/go/+/266720
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go: pass in overlaid paths for .s files
Michael Matloob [Thu, 29 Oct 2020 23:28:07 +0000 (19:28 -0400)]
cmd/go: pass in overlaid paths for .s files

This change adds support for adding overlays on assembly files.

For #39958

Change-Id: I1a328656199cc836f48e16de1ffd944fdd07fb39
Reviewed-on: https://go-review.googlesource.com/c/go/+/266417
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go: pass in overlaid file paths to C compiler
Michael Matloob [Thu, 29 Oct 2020 23:05:56 +0000 (19:05 -0400)]
cmd/go: pass in overlaid file paths to C compiler

This change moves the code in work.(*Builder).cgo that, when there is
an overlay, copies non-Go files to objdir into work.(*Builder).Build,
and creates an overlay structure mapping from the nominal file paths
into the copies in objdir. That's propagated through to
work.(*Builder).ccompile, which will use it to pass in the path to the
overlaid contents in objdir when calling the compiler.

This allows for overlays of C/C++/Fortran files.

For #39958

Change-Id: I9a2e3d3ba6afdf7ce19be1dbf4eee34805cdc05f
Reviewed-on: https://go-review.googlesource.com/c/go/+/266376
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/go, cmd/cgo: don't let bogus symbol set cgo_ldflag
Ian Lance Taylor [Fri, 6 Nov 2020 17:38:38 +0000 (09:38 -0800)]
cmd/go, cmd/cgo: don't let bogus symbol set cgo_ldflag

A hand-edited object file can have a symbol name that uses newline and
other normally invalid characters. The cgo tool will generate Go files
containing symbol names, unquoted. That can permit those symbol names
to inject Go code into a cgo-generated file. If that Go code uses the
//go:cgo_ldflag pragma, it can cause the C linker to run arbitrary
code when building a package. If you build an imported package we
permit arbitrary code at run time, but we don't want to permit it at
package build time. This CL prevents this in two ways.

In cgo, reject invalid symbols that contain non-printable or space
characters, or that contain anything that looks like a Go comment.

In the go tool, double check all //go:cgo_ldflag directives in
generated code, to make sure they follow the existing LDFLAG restrictions.

Thanks to Imre Rad / https://www.linkedin.com/in/imre-rad-2358749b for
reporting this.

Fixes CVE-2020-28367

Change-Id: Ia1ad8f3791ea79612690fa7d26ac451d0f6df7c1
Reviewed-on: https://team-review.git.corp.google.com/c/golang/go-private/+/895832
Reviewed-by: Than McIntosh <thanm@google.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/269658
Trust: Katie Hockman <katie@golang.org>
Trust: Roland Shoemaker <roland@golang.org>
Run-TryBot: Katie Hockman <katie@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Roland Shoemaker <roland@golang.org>
4 years agomath/big: fix shift for recursive division
Katie Hockman [Tue, 10 Nov 2020 20:54:12 +0000 (15:54 -0500)]
math/big: fix shift for recursive division

The previous s value could cause a crash
for certain inputs.

Will check in tests and documentation improvements later.

Thanks to the Go Ethereum team and the OSS-Fuzz project for reporting this.
Thanks to Rémy Oudompheng and Robert Griesemer for their help
developing and validating the fix.

Fixes CVE-2020-28362

Change-Id: Ibbf455c4436bcdb07c84a34fa6551fb3422356d3
Reviewed-on: https://team-review.git.corp.google.com/c/golang/go-private/+/899974
Reviewed-by: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Filippo Valsorda <valsorda@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/269657
Trust: Katie Hockman <katie@golang.org>
Trust: Roland Shoemaker <roland@golang.org>
Run-TryBot: Katie Hockman <katie@golang.org>
Reviewed-by: Roland Shoemaker <roland@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/go: skip TestScript/build_plugin_non_main on platforms that do not support -build...
xiaodong liu [Thu, 12 Nov 2020 08:12:08 +0000 (08:12 +0000)]
cmd/go: skip TestScript/build_plugin_non_main on platforms that do not support -buildmode=plugin

Fixes #42474

Change-Id: I1550b44b92cd272854e2f17493245a14e3d39f41
GitHub-Last-Rev: 948d01716eff41c25515dfb6135769862977aba1
GitHub-Pull-Request: golang/go#42475
Reviewed-on: https://go-review.googlesource.com/c/go/+/268737
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
Trust: Jay Conrod <jayconrod@google.com>

4 years agocmd/compile: gofmt
Cherry Zhang [Thu, 12 Nov 2020 13:44:29 +0000 (08:44 -0500)]
cmd/compile: gofmt

Change-Id: I2aa44b9976bdac577db44ef76e18fdf5748df8ca
Reviewed-on: https://go-review.googlesource.com/c/go/+/269477
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agocmd/go: migrate away from semver.Max
Jay Conrod [Wed, 11 Nov 2020 21:59:55 +0000 (16:59 -0500)]
cmd/go: migrate away from semver.Max

For #32700

Change-Id: Ib5cd7004e4558bebebc5f9e7c9263d720c590845
Reviewed-on: https://go-review.googlesource.com/c/go/+/269338
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
4 years agocmd/compile: do not emit an extra debug_line entry for the end of seq addr
Alessandro Arzilli [Tue, 10 Nov 2020 16:02:27 +0000 (17:02 +0100)]
cmd/compile: do not emit an extra debug_line entry for the end of seq addr

Uses DW_LNS_advance_pc directly, instead of calling putpclcdelta
because the latter will create a new debug_line entry for the end of
sequence address.

Fixes #42484

Change-Id: Ib6355605cac101b9bf37a3b4961ab0cee678a839
Reviewed-on: https://go-review.googlesource.com/c/go/+/268937
Trust: Than McIntosh <thanm@google.com>
Run-TryBot: Than McIntosh <thanm@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agocmd/link: put DYLD_INFO at beginning of LINKEDIT segment on darwin
Cherry Zhang [Thu, 12 Nov 2020 00:43:26 +0000 (19:43 -0500)]
cmd/link: put DYLD_INFO at beginning of LINKEDIT segment on darwin

Apparently, code signing requires DYLD_INFO tables are at the
beginning of the LINKEDIT segment. Put it there.

May fix #42507.

Change-Id: I1836e0f495719cf75f66d0831fe1544bbe3ff1a8
Reviewed-on: https://go-review.googlesource.com/c/go/+/269377
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>