]> Cypherpunks repositories - gostls13.git/log
gostls13.git
3 years agoruntime: improve asanread/asanwrite nosplit comment
Austin Clements [Thu, 6 Jan 2022 20:25:01 +0000 (15:25 -0500)]
runtime: improve asanread/asanwrite nosplit comment

Explain the conditions under which they are called on stacks that
cannot grow.

Change-Id: I08ee5480face7fbedeccc09e55b8149c5a793c2e
Reviewed-on: https://go-review.googlesource.com/c/go/+/376036
Trust: Austin Clements <austin@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Fannie Zhang <Fannie.Zhang@arm.com>
3 years agotest: workaround SIGILL on issue11656 on aix
Paul E. Murphy [Mon, 10 Jan 2022 16:57:16 +0000 (10:57 -0600)]
test: workaround SIGILL on issue11656 on aix

For some reason, aix sometimes executes the bogus function body. This
should never happen as it lives in a no-execute section. It might be
a transient permission blip as the heap grows.

Add a small function to cleanup and synchronize the icache before
jumping to the bogus function to ensure it causes a panic, not SIGILL.

Fixes #44583

Change-Id: Iadca62d82bfb70fc62088705dac42a880a1208fa
Reviewed-on: https://go-review.googlesource.com/c/go/+/377314
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Paul Murphy <murp@ibm.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agocmd/go: skip TestScript/test_fuzz_test_race on !race
Tobias Klauser [Tue, 11 Jan 2022 09:22:40 +0000 (10:22 +0100)]
cmd/go: skip TestScript/test_fuzz_test_race on !race

Skip the test on targets which don't support the race detector. This
fixes the linux-386-longtest builder after CL 376554.

Updates #50488

Change-Id: I08bf6f72cc0731761d49121eb7cfaa8b53906d37
Reviewed-on: https://go-review.googlesource.com/c/go/+/377634
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
Trust: Katie Hockman <katie@golang.org>
Run-TryBot: Katie Hockman <katie@golang.org>

3 years agotest: re-enable most go/tests that were disabled because of types2 differences
Dan Scales [Sat, 4 Dec 2021 00:10:10 +0000 (16:10 -0800)]
test: re-enable most go/tests that were disabled because of types2 differences

I made the default be that, where there are differences between types2
and -G=0 error messages, we want errorcheck tests to pass types2.
Typically, we can get errorcheck to pass on types2 and -G=0 if they give
the same number of error messages on the same lines, just different
wording. If they give a different number of error messages, then I made
types2 pass. I added an exception list for -G=0 to cover those cases
where -G=0 and types give different numbers of error messages.

Because types2 does not run if there are syntax errors, for several
tests, I had to split the tests into two parts in order to get all the
indicated errors to be reported in types2 (bug228.go, bug388.go,
issue11610.go, issue14520.go)

I tried to preserve the GCCGO labeling correctly (but may have gotten
some wrong). When types2 now matches where a GCCGO error previously
occurred, I transformed GCCGO_ERROR -> ERROR. When types2 no longer
reports an error in a certain place, I transformed ERROR -> GCCGO_ERROR.
When types2 reports an error in a new place, I used GC_ERROR.

The remaining entries in types2Failures are things that I think we
probably still need to fix - either actually missing errors in types2,
or cases where types2 gives worse errors than -G=0.

Change-Id: I7f01e82b322b16094096b67d7ed2bb39b410c34f
Reviewed-on: https://go-review.googlesource.com/c/go/+/372854
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
3 years agogo/types, types2: refer to type parameter if so for interface pointer errors
Robert Griesemer [Mon, 10 Jan 2022 18:33:16 +0000 (10:33 -0800)]
go/types, types2: refer to type parameter if so for interface pointer errors

Follow-up on comment in CL 376914.

Also:
- add missing check != nil test in assignableTo
- use check.sprintf rather than fmt.Sprintf in missingMethodReason

For #48312.

Change-Id: Ie209b4101a7f2c279e42a59987d0068079c8b69f
Reviewed-on: https://go-review.googlesource.com/c/go/+/377375
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agogo/types, types2: better error message for type parameter field access
Robert Griesemer [Sat, 8 Jan 2022 22:39:43 +0000 (14:39 -0800)]
go/types, types2: better error message for type parameter field access

Fixes #50516.

Also call DefPredeclaredTestFuncs in TestFixedbugs so it can be
run independently again.

Change-Id: I78d4cc11790b1543a2545a7ab297a223b3d5e3c8
Reviewed-on: https://go-review.googlesource.com/c/go/+/376954
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agogo/types, types2: better error message when using *interface instead of interface
Robert Griesemer [Sat, 8 Jan 2022 21:01:37 +0000 (13:01 -0800)]
go/types, types2: better error message when using *interface instead of interface

- detect *interface case and report specific error
- replaced switch with sequence of if's for more clarity
- fixed isInterfacePtr: it applies to all interfaces, incl.
  type parameters
- reviewed/fixed all uses of isInterfacePtr
- adjusted error messages to be consistently of the format
  "type %s is pointer to interface, not interface"

Fixes #48312.

Change-Id: Ic3c8cfcf93ad57ecdb60f6a727cce9e1aa4afb5d
Reviewed-on: https://go-review.googlesource.com/c/go/+/376914
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agotesting: only snapshot coverage during fuzzing
Roland Shoemaker [Fri, 7 Jan 2022 20:18:30 +0000 (12:18 -0800)]
testing: only snapshot coverage during fuzzing

Only snapshot/reset coverage counters when we are actually fuzzing.
This prevents a race when running corpus/seed values during the testing
phase.

Fixes #50488

Change-Id: I7dd5a0353a296c0b13eede29ad9af7c78814fa2d
Reviewed-on: https://go-review.googlesource.com/c/go/+/376554
Trust: Katie Hockman <katie@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
Trust: Roland Shoemaker <roland@golang.org>
Run-TryBot: Roland Shoemaker <roland@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile: use exact constant in go_asm.h
Ian Lance Taylor [Mon, 10 Jan 2022 04:20:43 +0000 (20:20 -0800)]
cmd/compile: use exact constant in go_asm.h

Fixes #50523

Change-Id: Idab1b44d106250e9301d90ee6571f0ea51242dd9
Reviewed-on: https://go-review.googlesource.com/c/go/+/377074
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Liz Fong-Jones <lizf@honeycomb.io>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agoRevert "cmd/compile: unique LinkString for renamed, embedded fields"
Matthew Dempsky [Mon, 10 Jan 2022 19:18:39 +0000 (19:18 +0000)]
Revert "cmd/compile: unique LinkString for renamed, embedded fields"

This reverts CL 372914.

Reason for revert: missing synchronization

Change-Id: I7ebb6de082cebb73741d803ff00e3465bbafab81
Reviewed-on: https://go-review.googlesource.com/c/go/+/377379
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Bryan Mills <bcmills@google.com>
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Matthew Dempsky <mdempsky@google.com>

3 years agocmd/compile, test: updated comments in crawler.go, added test
Dan Scales [Wed, 29 Dec 2021 02:40:12 +0000 (18:40 -0800)]
cmd/compile, test: updated comments in crawler.go, added test

Added a test to make sure that the private methods of a local generic
type are properly exported, if there is a global variable with that
type.

Added comments in crawler.go, to give more detail and to give more about
the overall purpose.

Fixed one place where t.isFullyInstantiated() should be replaced by
isPtrFullyInstantiated(t), so that we catch pointers to generic types
that may be used as a method receiver.

Change-Id: I9c42d14eb6ebe14d249df7c8fa39e889f7cd3f22
Reviewed-on: https://go-review.googlesource.com/c/go/+/374754
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile: unique LinkString for renamed, embedded fields
Matthew Dempsky [Thu, 16 Dec 2021 20:55:15 +0000 (12:55 -0800)]
cmd/compile: unique LinkString for renamed, embedded fields

Using type aliases, it's possible to create structs with embedded
fields that have no corresponding type literal notation. However, we
still need to generate a unique name for these types to use for linker
symbols. This CL introduces a new "struct{ Name = Type }" syntax for
use in LinkString formatting to represent these types.

Fixes #50190.

Change-Id: I025ceb09a86e00b7583d3b9885d612f5d6cb44fe
Reviewed-on: https://go-review.googlesource.com/c/go/+/372914
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Trust: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Robert Griesemer <gri@golang.org>
3 years agoruntime: expand TestGdbPythonCgo skip to include mips64le
Bryan C. Mills [Fri, 7 Jan 2022 21:35:31 +0000 (16:35 -0500)]
runtime: expand TestGdbPythonCgo skip to include mips64le

The failure mode in #37794 does not match the failure mode described
in #18784. However, since the test is currently skipped on all other
MIPS variants, it may be that they suffer from the same underlying GDB
bug. Ideally one of the Go MIPS maintainers should file an upstream
bug and remove the skip once it is fixed; in the meantime, there is no
point in continuing to let the test fail on just one of the four MIPS
variants.

For #37794

Change-Id: I570f51cc04cbb7ef1ed7efd526e26886af53bfb6
Reviewed-on: https://go-review.googlesource.com/c/go/+/376654
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile: fix interaction between generics and inlining
Dan Scales [Wed, 29 Dec 2021 15:08:55 +0000 (07:08 -0800)]
cmd/compile: fix interaction between generics and inlining

Finally figured out how to deal with the interaction between generics
and inlining. The problem has been: what to do if you inline a function
that uses a new instantiated type that hasn't been seen in the current
package? This might mean that you need to do another round of
function/method instantiatiations after inlining, which might lead to
more inlining, etc. (which is what we currently do, but it's not clear
when you can stop the inlining/instantiation loop).

We had thought that one solution was to export instantiated types (even
if not marked as exportable) if they are referenced in exported
inlineable functions. But that was quite complex and required changing
the export format. But I realized that we really only need to make sure
the relevant dictionaries and shape instantiations for the instantiated
types are exported, not the instantiated type itself and its wrappers.
The instantiated type is naturally created as needed, and the wrappers
are generated automatically while writing out run-time type (making use
of the exported dictionaries and shape instantiations).

So, we just have to make sure that those dictionaries and shape
instantiations are exported, and then they will be available without any
extra round of instantiations after inlining. We now do this in
crawler.go. This is especially needed when the instantiated type is only
put in an interface, so relevant dictionaries/shape instantiations are
not directly referenced and therefore exported, but are still needed for
the itab.

This fix avoids the phase ordering problem where we might have to keep
creating new type instantiations and instantiated methods after each
round of inlining we do.

Removed the extra round of instantiation/inlining that were added in the
previous fix. The existing tests
test/typeparam{geninline.go,structinit.go} already test this situation
of inlining a function referencing a new instantiated type.

Added the original example from issue 50121 as test (has 5 packages),
since it found a problem with this code that the current simpler test
for 50121 did not find.

Change-Id: Iac5d0dddf4be19376f6de36ee20a83f0d8f213b5
Reviewed-on: https://go-review.googlesource.com/c/go/+/375494
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>

3 years agoruntime/pprof: run TestCPUProfileMultithreadMagnitude subtests separately
Michael Pratt [Thu, 16 Dec 2021 20:30:47 +0000 (15:30 -0500)]
runtime/pprof: run TestCPUProfileMultithreadMagnitude subtests separately

Currently TestCPUProfileMultithreadMagnitude runs two CPU consumption
functions in a single profile and then analyzes the results as separate
subtests.

This works fine, but when debugging failures it makes manual analysis of
the profile dump a bit annoying.

Refactor the test to collect separate profiles for each subtest for
easier future analysis.

For #50097.
For #50232.

Change-Id: Ia1c8bb86aaaf652e64c5e660dcc2da47d2194c2b
Reviewed-on: https://go-review.googlesource.com/c/go/+/372800
Trust: Michael Pratt <mpratt@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Rhys Hiltner <rhys@justin.tv>
Reviewed-by: Bryan Mills <bcmills@google.com>
3 years agonet/http: map FS Open errors just like Dir
Jonathan Amsterdam [Sat, 8 Jan 2022 14:18:24 +0000 (09:18 -0500)]
net/http: map FS Open errors just like Dir

When an http.FileServer is given a path like file1/file2 where file1
exists but file2 does not, the proper HTTP status should be
NotFound. Some OSes return a "not a directory" error instead, so this
must be mapped to NotFound.

That mapping was already being done for the Dir FileSystem
implementation, as discussed in #18984. But it wasn't for the
FS implementation.

This CL does the same mapping for FS, by generalizing the function
that did it for Dir.

Fixes #49552

Change-Id: I61d6aa8ef101158e9674707d44e653f5dedbd040
Reviewed-on: https://go-review.googlesource.com/c/go/+/376874
Trust: Jonathan Amsterdam <jba@google.com>
Run-TryBot: Jonathan Amsterdam <jba@google.com>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agogo/types, types2: better error message for invalid == on type parameters
Robert Griesemer [Thu, 16 Dec 2021 05:25:50 +0000 (21:25 -0800)]
go/types, types2: better error message for invalid == on type parameters

Fixes #48712.

Change-Id: I6f214cdfdd1815493f2a04828e8f0097f1d8c124
Reviewed-on: https://go-review.googlesource.com/c/go/+/372734
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agonet/http/internal/testcert: use FIPS-compliant certificate
Damien Neil [Mon, 4 Oct 2021 17:20:22 +0000 (10:20 -0700)]
net/http/internal/testcert: use FIPS-compliant certificate

Upgrade the test certificate from RSA 1024 (not FIPS-approved)
to RSA 2048 (FIPS-approved), allowing tests to pass when
the dev.boringcrypto branch FIPS-only mode is enabled.

Fixes #48674.

Change-Id: I613d2f8d0207bf3683fd0df256bf0167604996c5
Reviewed-on: https://go-review.googlesource.com/c/go/+/353869
Trust: Damien Neil <dneil@google.com>
Run-TryBot: Damien Neil <dneil@google.com>
Reviewed-by: Filippo Valsorda <filippo@golang.org>
3 years agoruntime: skip TestSegv traceback check on 386
Michael Pratt [Fri, 7 Jan 2022 22:11:41 +0000 (17:11 -0500)]
runtime: skip TestSegv traceback check on 386

The VDSO (__kernel_vsyscall) is reachable via
asmcgocall(cgo_start_thread) on linux-386, which causes traceback to
throw.

Fixes #49182.
For #50504.

Change-Id: Idb78cb8de752203ce0ed63c2dbd2d12847338688
Reviewed-on: https://go-review.googlesource.com/c/go/+/376656
Reviewed-by: Cherry Mui <cherryyz@google.com>
Trust: Michael Pratt <mpratt@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com>

3 years agocmd/compile: accept string|[]byte-constrained 2nd argument in append
Robert Griesemer [Fri, 7 Jan 2022 02:02:30 +0000 (18:02 -0800)]
cmd/compile: accept string|[]byte-constrained 2nd argument in append

Similarly to what we do for the built-in function `copy`,
where we allow a string as 2nd argument to append, also
permit a type parameter constrained by string|[]byte.

While at it, change date in the manual.go2 test files so
that we don't need to constantly correct it when copying
a test case from that file into a proper test file.

Fixes #50281.

Change-Id: I23fed66736aa07bb3c481fe97313e828425ac448
Reviewed-on: https://go-review.googlesource.com/c/go/+/376214
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agosyscall: in TestDirent, make as many ReadDirent calls as are needed
Bryan C. Mills [Fri, 7 Jan 2022 14:54:44 +0000 (09:54 -0500)]
syscall: in TestDirent, make as many ReadDirent calls as are needed

ReadDirent returns only as many directory entries as will fit in the
buffer, and each entry is variable-length — so we have no guarantee in
general that a buffer of a given arbitrary size can hold even one
entry, let alone all ten entries expected by the test.

Instead, iterate calls to ReadDirent until one of the calls returns
zero entries and no error, indicating that the directory has been read
to completion.

Fixes #37323

Change-Id: I7f1cedde7666107256604e4ea1ac13c71f22151a
Reviewed-on: https://go-review.googlesource.com/c/go/+/376334
Trust: Bryan Mills <bcmills@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile: fix conv of slice of user-define byte type to string
Dan Scales [Thu, 6 Jan 2022 20:39:37 +0000 (12:39 -0800)]
cmd/compile: fix conv of slice of user-define byte type to string

types2 allows the conversion of a slice of a user-defined byte type B
(not builtin uint8 or byte) to string. But runtime.slicebytetostring
requires a []byte argument, so add in a CONVNOP from []B to []byte if
needed. Same for the conversion of a slice of user-defined rune types to
string.

I made the same change in the transformations of the old typechecker, so
as to keep tcConv() and transformConv() in sync. That fixes the bug for
-G=0 mode as well.

Fixes #23536

Change-Id: Ic79364427f27489187f3f8015bdfbf0769a70d69
Reviewed-on: https://go-review.googlesource.com/c/go/+/376056
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agodoc/go1.18: document type parameter name restriction
Robert Griesemer [Fri, 7 Jan 2022 18:06:53 +0000 (10:06 -0800)]
doc/go1.18: document type parameter name restriction

For #47694.

Change-Id: I00862f987a0ff9f71e0295ce4320e6f9a6a4332f
Reviewed-on: https://go-review.googlesource.com/c/go/+/376414
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agocmd/compile: fix instantiation of types referenced during inlining
Cuong Manh Le [Wed, 5 Jan 2022 09:19:19 +0000 (16:19 +0700)]
cmd/compile: fix instantiation of types referenced during inlining

CL 352870 added extra phase for instantiation after inlining, to take
care of the new fully-instantiated types. However, when fetching inlined
body of these types's methods, we need to allow OADDR operations on
untyped expressions, the same as what main inlining phase does.

The problem does not show up, until CL 371554, which made the compiler
do not re-typecheck while importing, thus leaving a OXDOT node to be
marked as address taken when it's not safe to do that.

Fixes #50437

Change-Id: I20076b872182c520075a4f8b84230f5bcb05b341
Reviewed-on: https://go-review.googlesource.com/c/go/+/375574
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
3 years ago.github: remove duplicate security link
Dmitri Shuralyov [Fri, 7 Jan 2022 17:50:37 +0000 (17:50 +0000)]
.github: remove duplicate security link

Since a SECURITY.md file is present in the main Go repository,
GitHub already shows a "Report a security vulnerability" link
in the issue template list. Remove the duplicate custom link.

Fixes #49962.

Change-Id: Ifdf7e93b76ebd9258d907aa9cb4915c0dbc4f93e
Reviewed-on: https://go-review.googlesource.com/c/go/+/376357
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Julie Qiu <julie@golang.org>
3 years agotest/typeparam: adjust test preamble (fix longtests)
Robert Griesemer [Fri, 7 Jan 2022 04:06:40 +0000 (20:06 -0800)]
test/typeparam: adjust test preamble (fix longtests)

For #50481.

Change-Id: I27e6c6499d6abfea6e215d8aedbdd5074ff88291
Reviewed-on: https://go-review.googlesource.com/c/go/+/376216
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Bryan Mills <bcmills@google.com>
3 years agospec: be more precise with rules on specific types
Robert Griesemer [Thu, 6 Jan 2022 05:37:04 +0000 (21:37 -0800)]
spec: be more precise with rules on specific types

Problem pointed out on golang-nuts mailing list.

Change-Id: If1c9b22e1ed7b4ec7ebcaadc80fa450333e6856c
Reviewed-on: https://go-review.googlesource.com/c/go/+/375799
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agodoc/go1.18: point to spec in same directory for release notes
Robert Griesemer [Sat, 18 Dec 2021 01:17:41 +0000 (17:17 -0800)]
doc/go1.18: point to spec in same directory for release notes

The release notes explicitly refer to sections updated
for generics in the spec but then point to the old spec
which is very confusing for beta users.

For #47694

Change-Id: I5b555db3543cc32f088a8b267ec3f1195a52a812
Reviewed-on: https://go-review.googlesource.com/c/go/+/373174
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agotest/typeparam: adjust test preamble (fix longtests)
Robert Griesemer [Fri, 7 Jan 2022 02:23:01 +0000 (18:23 -0800)]
test/typeparam: adjust test preamble (fix longtests)

For #50317.

Change-Id: I24ccf333c380283a36b573ef8fc3e7fcd71bd17f
Reviewed-on: https://go-review.googlesource.com/c/go/+/376215
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>

3 years agotest/typeparam: adjust test preamble (fix longtests)
Robert Griesemer [Fri, 7 Jan 2022 00:50:49 +0000 (16:50 -0800)]
test/typeparam: adjust test preamble (fix longtests)

For #50417.

Change-Id: Ic55727c454ec342354f7fbffd22aa350e0d392c2
Reviewed-on: https://go-review.googlesource.com/c/go/+/376174
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agomath/big: fix spurious race in Rat.Denom, Float.SetRat
Russ Cox [Thu, 6 Jan 2022 17:20:14 +0000 (12:20 -0500)]
math/big: fix spurious race in Rat.Denom, Float.SetRat

Rat maintains the invariant that x.b.neg is always false,
but Rat.Denom was writing x.b.neg = false itself too.
That makes Rat.Denom a writing operation, when it should
be a read-only operation. That in turn makes it unsafe to
use from multiple goroutines, which is highly unexpected.
Make it read-only and therefore race-free again.

Fixes #50473.

Change-Id: I97b87913954511e5200c0665d16b9ed63422e505
Reviewed-on: https://go-review.googlesource.com/c/go/+/375935
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
3 years agogo/types, types2: disallow multiple blank type parameters
Robert Griesemer [Thu, 6 Jan 2022 22:10:45 +0000 (14:10 -0800)]
go/types, types2: disallow multiple blank type parameters

Work-around for #50481: report an error for multiple
blank type parameters. It's always possible to use
non-blank names in those cases.

We expect to lift this restriction for 1.19.

For #50481.

Change-Id: Ifdd2d91340aac1da3387f7d80d46e44f5997c2a8
Reviewed-on: https://go-review.googlesource.com/c/go/+/376058
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Dan Scales <danscales@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile: report type parameter error for methods only once
Robert Griesemer [Thu, 6 Jan 2022 18:58:30 +0000 (10:58 -0800)]
cmd/compile: report type parameter error for methods only once

Move switch to enable method type parameters entirely
to the parser, by adding the mode AllowMethodTypeParams.
Ensure that the error messages are consistent.
Remove unnecessary code in the type checker.

Fixes #50317.

Change-Id: I4f3958722400bdb919efa4c494b85cf62f4002bb
Reviewed-on: https://go-review.googlesource.com/c/go/+/376054
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agotesting: fix deadlock with t.Parallel in testing seed corpus
Katie Hockman [Wed, 22 Dec 2021 15:34:55 +0000 (10:34 -0500)]
testing: fix deadlock with t.Parallel in testing seed corpus

The c.startParallel channel on the testContext is stuck
in t.Parallel() because c.running starts at 1 for the main
fuzz parent test, and is causing a deadlock because it is
never released. It would normally be released by tRunner,
but needs to instead be released by fRunner instead for fuzz
tests.

Fixes #50217

Change-Id: I2d010e9adddfd8e8321ff2f9dd2e43daf46c128f
Reviewed-on: https://go-review.googlesource.com/c/go/+/374054
Reviewed-by: Roland Shoemaker <roland@golang.org>
Reviewed-by: Bryan Mills <bcmills@google.com>
Trust: Katie Hockman <katie@golang.org>
Run-TryBot: Katie Hockman <katie@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agogo/types, types2: implement field access for struct structural constraints
Robert Griesemer [Wed, 5 Jan 2022 23:46:31 +0000 (15:46 -0800)]
go/types, types2: implement field access for struct structural constraints

This change implements field the access p.f where the type of p
is a type parameter with a structural constraint that is a struct
with a field f. This is only the fix for the type checker. The
compiler will need a separate CL.

This makes the behavior consistent with the fact that we can
write struct composite literals for type parameters with a
struct structural type.

For #50417.
For #50233.

Change-Id: I87d07e016f97cbf19c45cde19165eae3ec0bad2b
Reviewed-on: https://go-review.googlesource.com/c/go/+/375795
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agogo/types, types2: remove unused code in lookupFieldOrMethod
Robert Griesemer [Wed, 5 Jan 2022 05:01:21 +0000 (21:01 -0800)]
go/types, types2: remove unused code in lookupFieldOrMethod

The underlying type of a type parameter is an interface,
so we don't need a special case for type parameters anymore.
Simply share the (identical) code for interfaces.

Adjust code in types.NewMethodSet accordingly.

No functional difference.
Preparation for fix of issues below.

For #50233.
For #50417.

Change-Id: Ib2deadd12f89e6918dec224b4ce35583001c3101
Reviewed-on: https://go-review.googlesource.com/c/go/+/375514
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agogo/types, types2: ensure that signature type bounds are interfaces
Robert Griesemer [Wed, 5 Jan 2022 20:25:43 +0000 (12:25 -0800)]
go/types, types2: ensure that signature type bounds are interfaces

Do this by running verification for instantiated signatures
later, after the delayed type parameter set-up had a chance
to wrap type bounds in implicit interfaces where needed.

Fixes #50450

Change-Id: If3ff7dc0be6af14af854830bfddb81112ac575cb
Reviewed-on: https://go-review.googlesource.com/c/go/+/375737
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agotime: skip TestTimerModifiedEarlier on plan9/arm
Bryan C. Mills [Thu, 6 Jan 2022 16:59:09 +0000 (11:59 -0500)]
time: skip TestTimerModifiedEarlier on plan9/arm

This test is observed to be flaky on the plan9-arm builder.
Skip it on that platform until it can be diagnosed and fixed.

For #50470

Change-Id: If626af426d856c377e00ac5baaca52899456556e
Reviewed-on: https://go-review.googlesource.com/c/go/+/375934
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agotest: add test of incorrect gofrontend error
Ian Lance Taylor [Thu, 6 Jan 2022 04:22:15 +0000 (20:22 -0800)]
test: add test of incorrect gofrontend error

For #50439

Change-Id: Ifad6e6f8de42121c695b5a4dc56e0f6606e2917e
Reviewed-on: https://go-review.googlesource.com/c/go/+/375796
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
Trust: Than McIntosh <thanm@google.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agogo/types, types2: eagerly check that constraints are not type params
Robert Findley [Thu, 23 Dec 2021 15:33:17 +0000 (10:33 -0500)]
go/types, types2: eagerly check that constraints are not type params

As a result of the change to the underlying of a type parameter to be
its constraint interface, we had couple inaccuracies that combined to
cause an infinite recursion when type checking the invalid type
parameter list [A A].
 - We deferred tpar.iface() using check.later twice: once in
   newTypeParam, and then again at the end of collectTypeParams.
 - We deferred the check that type parameter constraints are not type
   parameters, even though this is unnecessary: the constraint type is
   known.

With these inaccuracies, tpar.iface() was executing before our guard
against using type parameters as constraints, causing an infinite
recursion through under().

Fix this by eagerly checking whether the constraint is a type
parameter, and marking it invalid if so. Also remove the unnecessary
calls to tpar.iface() at the end of collectTypeParams, as this will
already have been scheduled by newTypeParam.

Fixes #50321

Change-Id: I4eecbecf21656615867cb94be65b520e9e795bd1
Reviewed-on: https://go-review.googlesource.com/c/go/+/374294
Reviewed-by: Robert Griesemer <gri@golang.org>
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile/internal/types2: better error message for invalid range clause
Robert Griesemer [Tue, 4 Jan 2022 23:13:33 +0000 (15:13 -0800)]
cmd/compile/internal/types2: better error message for invalid range clause

Fixes #50372.

Change-Id: I8e4c0020dae42744cce016433e398e0b884bb044
Reviewed-on: https://go-review.googlesource.com/c/go/+/375475
Trust: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agonet: synchronize instead of sleeping in TestDialParallelSpuriousConnection
Bryan C. Mills [Wed, 5 Jan 2022 16:13:27 +0000 (11:13 -0500)]
net: synchronize instead of sleeping in TestDialParallelSpuriousConnection

The arbitrary sleep in this test is empirically not always long enough
on slower builders. However, we know the exact number of connections
that should be dialed: we can wait on that number in the dial hook
instead.

Fixes #34495

Change-Id: I538244ceb75a80271a724304b993309482bd5b41
Reviewed-on: https://go-review.googlesource.com/c/go/+/375694
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agoruntime: crash on netbsd-arm64 when setcontext fails
Benny Siegert [Thu, 6 Jan 2022 11:27:57 +0000 (11:27 +0000)]
runtime: crash on netbsd-arm64 when setcontext fails

Instead of exiting with status 16962 when we fail to call SYS_setcontext
in sigreturn, reference a null pointer and crash. Hopefully, this will
enable grabbing a core dump to debug.

Updates #42422

Change-Id: If02c14a0a37084351f3f00db3dc9766cb68ae4b8
Reviewed-on: https://go-review.googlesource.com/c/go/+/375834
Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Trust: Benny Siegert <bsiegert@gmail.com>

3 years agodoc: improve documentation for GOAMD64
Ulrich Kunitz [Wed, 15 Dec 2021 10:37:36 +0000 (11:37 +0100)]
doc: improve documentation for GOAMD64

The change clarifies in the release notes for go1.18 that the variable
is for compilation and references the microarchitecture description in
the Go Wiki,
https://golang.org/wiki/MinimumRequirements#amd64,
and references the same information in the output of go help
environment.

Fixes #50174

Change-Id: I6a7d5a06f48463a810c96cc9c76fe66113d5147c
Reviewed-on: https://go-review.googlesource.com/c/go/+/372194
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/go: add fuzzing coverage test
Roland Shoemaker [Wed, 5 Jan 2022 20:25:28 +0000 (12:25 -0800)]
cmd/go: add fuzzing coverage test

Adds a test script for fuzzing coverage instrumentation.

Fixes #48654

Change-Id: Ieea7b4146bd5581baae869441cc1c652dd7485f5
Reviewed-on: https://go-review.googlesource.com/c/go/+/375736
Trust: Katie Hockman <katie@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
Trust: Roland Shoemaker <roland@golang.org>
Run-TryBot: Roland Shoemaker <roland@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agonet/http: skip TestClientTimeout_h{1,2} on windows/arm and windows/arm64
Bryan C. Mills [Wed, 5 Jan 2022 19:20:30 +0000 (14:20 -0500)]
net/http: skip TestClientTimeout_h{1,2} on windows/arm and windows/arm64

These tests are empirically flaky on the windows/arm and windows/arm64
builders, with a consistent (but rare) failure mode.

This change skips the test if that particular failure mode is
encountered on those platforms; the skip can be removed if and when
someone has the time to pin down the root cause.

For #43120

Change-Id: Ie3a9a06bf47e3a907c7b07441acc1494a4631135
Reviewed-on: https://go-review.googlesource.com/c/go/+/375635
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agoruntime: ensure that asmsyscall6 follow AIX stack convention
Clément Chigot [Fri, 17 Dec 2021 10:03:21 +0000 (11:03 +0100)]
runtime: ensure that asmsyscall6 follow AIX stack convention

The function asmsyscall6 must follow AIX stack layout. It means
that its first local variable must be stored after its arguments
area, ie after offset 112.

Fixes #50185

Change-Id: I897731ddd2a9faad8218443a4c2f4b204ad7e173
Reviewed-on: https://go-review.googlesource.com/c/go/+/373074
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agosrc/runtime: mark asanread and asanwrite functions as NOSPLIT
fanzha02 [Thu, 30 Dec 2021 03:09:42 +0000 (11:09 +0800)]
src/runtime: mark asanread and asanwrite functions as NOSPLIT

The asan runtime functions may run on stacks that cannot grow, and
they do not have large local variables, so it is safe to mark them
as NOSPLIT.

Add test case.

Fixes #50391

Change-Id: Iadcbf1ae0c837d9b64da5be208c7f424e6ba11de
Reviewed-on: https://go-review.googlesource.com/c/go/+/374398
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
Trust: Fannie Zhang <Fannie.Zhang@arm.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agonet: do not panic on nil IPNet.String()
Jason A. Donenfeld [Tue, 4 Jan 2022 23:16:40 +0000 (00:16 +0100)]
net: do not panic on nil IPNet.String()

The code looks like it was already trying to avoid this but missed a
spot.

Fixes #50271.

Change-Id: I450adac3f618b9535b61a28e6a160eacc351d47c
Reviewed-on: https://go-review.googlesource.com/c/go/+/373075
Trust: Jason Donenfeld <Jason@zx2c4.com>
Run-TryBot: Jason Donenfeld <Jason@zx2c4.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agocmd/compile: save selector/inst info for generic method/function calls
Dan Scales [Tue, 21 Dec 2021 15:59:16 +0000 (07:59 -0800)]
cmd/compile: save selector/inst info for generic method/function calls

In the dict info, we need to save the SelectorExpr of a generic method
call when making its sub-dictionary entry. The generic method call will
eventually be transformed into a function call on the method shape
instantiation, so we may not always have the selector info available
when we need it to create a dictionary. We use this SelectorExpr as
needed if the relevant call node has already been transformed.

Similarly, we save the InstExpr of generic function calls, since the
InstExpr will be dropped when the function call is transformed to a call
to a shape instantiation. We use this InstExpr if the relevant function
call has already been transformed.

Added an extra generic function Some2 and a call to it from Some that
exercises the generic function case. The existing test already tests the
method call case.

Fixes #50264

Change-Id: I2c7c7d79a8e33ca36a5e88e64e913c57500c97f9
Reviewed-on: https://go-review.googlesource.com/c/go/+/373754
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile: pop instantiations of local types when leaving scope
Dan Scales [Thu, 16 Dec 2021 02:42:00 +0000 (18:42 -0800)]
cmd/compile: pop instantiations of local types when leaving scope

Since we use existing instantiations from the symbol table when possible
(to make sure each instantiation is unique), we need to pop
instantiations of local types when leaving the containing scope.
g.stmts() now pushes and pops scope, and we do a Pushdcl() in g.typ0()
when creating an instantiation of a local type.

Non-instantiated local types (generic or not) are translated directly
from types2, so they don't need to be pushed/popped. We don't export
function bodies with local types, so there is no issue during import.

We still don't support local types in generic functions/methods.

Fixes #50177

Change-Id: If2d2fe71aec003d13f0338565c7a0da2c9580a14
Reviewed-on: https://go-review.googlesource.com/c/go/+/372654
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agodoc/go1.18: mention new cmd/go fuzzing flags
Ian Lance Taylor [Thu, 23 Dec 2021 23:57:54 +0000 (15:57 -0800)]
doc/go1.18: mention new cmd/go fuzzing flags

For #47694

Change-Id: I00da9bd39700e938ec492daa71aba2035d911a06
Reviewed-on: https://go-review.googlesource.com/c/go/+/374354
Trust: Ian Lance Taylor <iant@golang.org>
Trust: Katie Hockman <katie@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
3 years agogo/doc: handle generic receiver strings
Jonathan Amsterdam [Sat, 18 Dec 2021 12:28:16 +0000 (07:28 -0500)]
go/doc: handle generic receiver strings

A receiver expression for a type with parameters may be an IndexExpr
or IndexListExpr in addition to an Ident or StarExpr. Add cases to
recvString to account for the new types.

Add tests that compare the fields of Func, and the fields of Type that
hold Funcs. These fields weren't previously tested.

Change-Id: Ia2cef51c85113422e0c5745c77dddcd53507ee51
Reviewed-on: https://go-review.googlesource.com/c/go/+/375095
Trust: Jonathan Amsterdam <jba@google.com>
Run-TryBot: Jonathan Amsterdam <jba@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agogo/doc: use subtests
Jonathan Amsterdam [Fri, 17 Dec 2021 15:30:51 +0000 (10:30 -0500)]
go/doc: use subtests

Change the Test function to use sub-tests for each doc mode and
package.

This will allow more fine-grained test execution.

Change-Id: Ie3dda5791bda2781a60776886dd39fd18e670e24
Reviewed-on: https://go-review.googlesource.com/c/go/+/375094
Trust: Jonathan Amsterdam <jba@google.com>
Run-TryBot: Jonathan Amsterdam <jba@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agoos: simplify ios checks in tests
Tobias Klauser [Thu, 30 Dec 2021 13:06:30 +0000 (14:06 +0100)]
os: simplify ios checks in tests

Go 1.16 renamed the iOS port from darwin/arm64 to ios/arm64 and
darwin/arm64 was repurposed for the macOS ARM64 port (see
https://golang.org/doc/go1.16#darwin).

Change tests to only use GOOS=ios to detect special cases for iOS and
stop treating darwin/arm64 as iOS.

Change-Id: I6a19f16fa9ec159ab1386aeb8fe912585e457171
Reviewed-on: https://go-review.googlesource.com/c/go/+/374399
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agoos: enable TestMkdirAllWithSymlink on darwin/arm64
Tobias Klauser [Tue, 21 Dec 2021 12:20:20 +0000 (13:20 +0100)]
os: enable TestMkdirAllWithSymlink on darwin/arm64

Go 1.16 renamed the iOS port from darwin/arm64 to ios/arm64 and
darwin/arm64 was repurposed for the macOS ARM64 port (see
https://golang.org/doc/go1.16#darwin).

TestMkdirAllWithSymlink ought to run on darwin/arm64, so enable it on
that platform.

For #45696

Change-Id: I2cad6b1dfddf215e6b6cd262bbd22251f48f3d8c
Reviewed-on: https://go-review.googlesource.com/c/go/+/373359
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agomisc/cgo/testsanitizers: accept compilers that don't report location
Ian Lance Taylor [Wed, 29 Dec 2021 19:19:13 +0000 (11:19 -0800)]
misc/cgo/testsanitizers: accept compilers that don't report location

It appears that GCC before version 10 doesn't report file/line
location for asan errors.

Change-Id: I03ee24180ba365636596aa2384961df7ce6ed71f
Reviewed-on: https://go-review.googlesource.com/c/go/+/374874
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agomisc/cgo/testsanitizers: don't fail asan test if no symbolizer
Ian Lance Taylor [Wed, 29 Dec 2021 18:02:27 +0000 (10:02 -0800)]
misc/cgo/testsanitizers: don't fail asan test if no symbolizer

Change-Id: Ic05c641bda3cc8f5292921c9b0c0d3df34f3bc48
Reviewed-on: https://go-review.googlesource.com/c/go/+/374794
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agoruntime: fix the issue that the -asan option cannot print where the error occurred
fanzha02 [Mon, 27 Dec 2021 08:38:48 +0000 (16:38 +0800)]
runtime: fix the issue that the -asan option cannot print where the error occurred

The current -asan option does not print where the error occurred. The
reason is that the current implementation calls incorrect asan runtime
functions, which do not pass sp and pc where asan runtime functions are
called, and report the stack trace from the native code. But asan runtime
functions are called from cgo on a separated stack, so it cannot dump the
Go stack trace correctly.

The correct asan runtime function we should call is __asan_report_error,
which will pass sp and pc, and report where the error occurred correctly.

This patch fixes this issue.

Add the test cases.

Fixes #50362

Change-Id: I12ee1d46c7ae069ddef3d23f2fe86e112db60045
Reviewed-on: https://go-review.googlesource.com/c/go/+/374395
Trust: Fannie Zhang <Fannie.Zhang@arm.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agoruntime: invalid negative frequency while tracing
Meng Zhuo [Fri, 17 Dec 2021 06:15:42 +0000 (14:15 +0800)]
runtime: invalid negative frequency while tracing

The riscv64 Hifive Unmatched is the only platform that
failed on testcase TestAnalyzeAnnotations occasionally
after CL 332954 had merged. The failure happens when
ticks per second (freq) is over 1e12 which causing the timestamps
of two events are same.

There are 2 reasons causing big frequency:
1. RDCYCLE is HART based according to the riscv manual which makes
   negative ticks delta
2. negative float64 -> uint64 is undefined and "lucky" negative float
   is too big to handle for trace

For #46737

Change-Id: I1f3c1ac31aae249969000c719c32aaf5a66d29a5
Reviewed-on: https://go-review.googlesource.com/c/go/+/373034
Trust: Zhuo Meng <mzh@golangcn.org>
Run-TryBot: Zhuo Meng <mzh@golangcn.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
3 years agodoc/go1.18: add section for runtime/debug changes
Ian Lance Taylor [Tue, 21 Dec 2021 21:58:15 +0000 (13:58 -0800)]
doc/go1.18: add section for runtime/debug changes

For #37475
For #47694

Change-Id: If8c1f1b756daf32648110f1a669b2ea60f797a24
Reviewed-on: https://go-review.googlesource.com/c/go/+/373875
Trust: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agodoc/go1.18: mention debug/plan9obj.ErrNoSymbols
Ian Lance Taylor [Tue, 21 Dec 2021 19:59:19 +0000 (11:59 -0800)]
doc/go1.18: mention debug/plan9obj.ErrNoSymbols

For #47694
For #48052

Change-Id: I136be9b498033309d876099aae16bad330555416
Reviewed-on: https://go-review.googlesource.com/c/go/+/373774
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
3 years agocmd/go: remove MallocNanoZone environment variable workaround
Cherry Mui [Thu, 23 Dec 2021 18:09:02 +0000 (13:09 -0500)]
cmd/go: remove MallocNanoZone environment variable workaround

We added a workaround in runtime/race. This should not be
necessary now.

Updates #49138.

Change-Id: Ia2010e4acc95c4ddf5f463ab2919401d893c0bac
Reviewed-on: https://go-review.googlesource.com/c/go/+/374314
Trust: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile/internal/ir: fix a typo in node.go
zhangjian [Thu, 23 Dec 2021 14:21:41 +0000 (14:21 +0000)]
cmd/compile/internal/ir: fix a typo in node.go

README.md -> HACKING.md

Change-Id: I63909b86b2e6f8d9a34622f5d2b05048c79cd698
GitHub-Last-Rev: afaab8f05eff1d66c796909ccf9912e0f4e99754
GitHub-Pull-Request: golang/go#50326
Reviewed-on: https://go-review.googlesource.com/c/go/+/374274
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agodoc/go1.18: fix broken HTML in net/http.MaxBytesHandler
Carl Johnson [Wed, 22 Dec 2021 04:23:26 +0000 (04:23 +0000)]
doc/go1.18: fix broken HTML in net/http.MaxBytesHandler

For #47694

Sorry about that! I guess the autocompleter in VSCode auto-closed the paragraph and I didn't notice.

Change-Id: I1e834e47deb708cd5285d26201a442305f8c3b24
GitHub-Last-Rev: ab8873155191072efd51270c85fc6feaed318cdc
GitHub-Pull-Request: golang/go#50305
Reviewed-on: https://go-review.googlesource.com/c/go/+/374014
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Dmitri Shuralyov <dmitshur@golang.org>

3 years agoruntime: handle stray profiling signal better
Cherry Mui [Wed, 22 Dec 2021 16:43:06 +0000 (11:43 -0500)]
runtime: handle stray profiling signal better

In c-archive mode, when we turn off profiling, we  restore the
previous handler for SIGPROF, and ignore SIGPROF signals if no
handler was installed. So if a pending signal lands after we
remove the Go signal handler, it will not kill the program.

In the current code there is a small window, where we can still
receive signals but we are set to not handling the signal. If a
signal lands in this window (possibly on another thread), it will
see that we are not handling this signal and no previous handler
installed, and kill the program. To avoid this race, we set the
previous handler to SIG_IGN (ignoring the signal) when turning on
profiling. So when turning off profiling we'll ignore the signal
even if a stray signal lands in the small window.

Fixes #43828.

Change-Id: I304bc85a93ca0e63b0c0d8e902b097bfdc8e3f1d
Reviewed-on: https://go-review.googlesource.com/c/go/+/374074
Trust: Cherry Mui <cherryyz@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agodoc/go1.18: list new net/netip and net functions and methods
Ian Lance Taylor [Tue, 21 Dec 2021 20:23:53 +0000 (12:23 -0800)]
doc/go1.18: list new net/netip and net functions and methods

For #46518
For #47694

Change-Id: I4848556674baf85ceec350645d9eddcd83f1b2e0
Reviewed-on: https://go-review.googlesource.com/c/go/+/373834
Trust: Ian Lance Taylor <iant@golang.org>
Trust: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
3 years agodoc/go1.18: mention testing.F in testing package section
Ian Lance Taylor [Tue, 21 Dec 2021 22:33:47 +0000 (14:33 -0800)]
doc/go1.18: mention testing.F in testing package section

For #47694

Change-Id: I27d88d864319bd8dbabfa6675a6abf2f74c50717
Reviewed-on: https://go-review.googlesource.com/c/go/+/373914
Trust: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
Trust: Katie Hockman <katie@golang.org>

3 years agodoc/go1.18: document {text,html}/template {break,continue} commands
Ian Lance Taylor [Tue, 21 Dec 2021 22:44:27 +0000 (14:44 -0800)]
doc/go1.18: document {text,html}/template {break,continue} commands

For #20531
For #47694

Change-Id: Iaefaa0a8982eabf59cd6a53120c8af9124d60d1a
Reviewed-on: https://go-review.googlesource.com/c/go/+/373915
Trust: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agodoc/go1.18: add net/http.MaxBytesHandler
Carl Johnson [Sat, 18 Dec 2021 02:54:22 +0000 (02:54 +0000)]
doc/go1.18: add net/http.MaxBytesHandler

For #47694

Change-Id: Ifab2c2720ec2ccef175a0e14a95f8df0437eb009
GitHub-Last-Rev: f3c2f12c7b6472610428cff5cad2a4497c47150e
GitHub-Pull-Request: golang/go#50250
Reviewed-on: https://go-review.googlesource.com/c/go/+/373015
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agoos: enable TestPipeThreads on FreeBSD
Tobias Klauser [Tue, 21 Dec 2021 12:13:01 +0000 (13:13 +0100)]
os: enable TestPipeThreads on FreeBSD

This test works on FreeBSD since CL 165801 was submitted.

Updates #19093

Change-Id: I45ffeb403c1de4385cdb21b9647f21976061e1ea
Reviewed-on: https://go-review.googlesource.com/c/go/+/373358
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agodoc/go1.18: mention new go/types/Config.Context field
Ian Lance Taylor [Tue, 21 Dec 2021 20:04:57 +0000 (12:04 -0800)]
doc/go1.18: mention new go/types/Config.Context field

For #47694
For #47916

Change-Id: Ieeffaf161da744adfdb4da8aac58a64c109ebcab
Reviewed-on: https://go-review.googlesource.com/c/go/+/373775
Trust: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agoall: update vendored golang.org/x/crypto for cryptobyte fix
Filippo Valsorda [Tue, 21 Dec 2021 16:49:38 +0000 (11:49 -0500)]
all: update vendored golang.org/x/crypto for cryptobyte fix

Fixes #49678

Change-Id: I47dd959a787180a67856e60dfa6eba3ddd045972
Reviewed-on: https://go-review.googlesource.com/c/go/+/373360
Run-TryBot: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Trust: Filippo Valsorda <filippo@golang.org>

3 years agodoc/go1.18: mention new debug/elf.R_PPC64_RELATIVE constant
Ian Lance Taylor [Thu, 16 Dec 2021 00:30:15 +0000 (16:30 -0800)]
doc/go1.18: mention new debug/elf.R_PPC64_RELATIVE constant

For #47694

Change-Id: I9f4838100741d3ba13e9374e70466fa405c6e4d2
Reviewed-on: https://go-review.googlesource.com/c/go/+/372634
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agocmd/link: use SHT_INIT_ARRAY for .init_array section
Ian Lance Taylor [Tue, 21 Dec 2021 18:00:23 +0000 (10:00 -0800)]
cmd/link: use SHT_INIT_ARRAY for .init_array section

Fixes #50295

Change-Id: If55ebcd5f2af724da7c9c744458a56d21a7ddde7
Reviewed-on: https://go-review.googlesource.com/c/go/+/373734
Trust: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agoreflect: update relative path in comment
Jing维 [Mon, 20 Dec 2021 00:36:55 +0000 (00:36 +0000)]
reflect: update relative path in comment

Fix it to avoid misunderstanding.

Change-Id: I2a09cb9edfa8077c5c0c35e07000c0c7dc72755e
GitHub-Last-Rev: ae49d512eb7f58421ff2d3b9a8b04500a96ac831
GitHub-Pull-Request: golang/go#50260
Reviewed-on: https://go-review.googlesource.com/c/go/+/373334
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agonet/http: fix link to ResponseWriter trailer example
Reilly Watson [Mon, 20 Dec 2021 21:31:45 +0000 (16:31 -0500)]
net/http: fix link to ResponseWriter trailer example

The links to this example were using an underscore separator
instead of a hyphen, and incorrect casing.

Fixes #50279

Change-Id: I35d76a8a78cd708b7505ff1a70f7dacddaf43efd
Reviewed-on: https://go-review.googlesource.com/c/go/+/373514
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
3 years agoruntime: typo fix cyle -> cycle
“kinggo” [Mon, 20 Dec 2021 05:54:27 +0000 (05:54 +0000)]
runtime: typo fix cyle -> cycle

Change-Id: I213fa8aa9b9c2537a189677394ddd30c62312518
GitHub-Last-Rev: ccafdee9440b06232cdfca83099bf0aeff62a4c0
GitHub-Pull-Request: golang/go#50268
Reviewed-on: https://go-review.googlesource.com/c/go/+/373336
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Zhuo Meng <mzh@golangcn.org>
3 years agoruntime: check the correct environment variable in TestCgoCallbackGC
Cherry Mui [Tue, 21 Dec 2021 00:08:14 +0000 (19:08 -0500)]
runtime: check the correct environment variable in TestCgoCallbackGC

The test checks RUNTIME_TESTING_SHORT, whereas the test runner
actually set RUNTIME_TEST_SHORT. Check the correct one.

Updates #32023.

Change-Id: Ie8ab00e1f5b8c02112a9aa1ee0e56028185c8a44
Reviewed-on: https://go-review.googlesource.com/c/go/+/373614
Trust: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agonet/netip: add a fuzz test
Andrew LeFevre [Thu, 16 Dec 2021 05:25:09 +0000 (05:25 +0000)]
net/netip: add a fuzz test

This is a pretty straight port of the fuzz test at https://github.com/inetaf/netaddr.

The MarshalText methods of netip.Addr and net.IP, the Is* methods of netip.Addr
and net.IP and the MarshalText and String methods of netip.Addr are also
checked to ensure that they behave the same way.

Fixes #49367

Change-Id: I44abb01f2a7af45f39597992a1fc7ff0305728fa
GitHub-Last-Rev: c2323b0ae119b804a58b810ad7f0e2bb1b3a38ac
GitHub-Pull-Request: golang/go#50108
Reviewed-on: https://go-review.googlesource.com/c/go/+/371055
Trust: Matt Layher <mdlayher@gmail.com>
Trust: Katie Hockman <katie@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
3 years agocmd/doc: don't log on constraint type elements
Ian Lance Taylor [Sat, 18 Dec 2021 23:54:38 +0000 (15:54 -0800)]
cmd/doc: don't log on constraint type elements

Fixes #50256

Change-Id: I2327a0b28f8173c801ed2946bec8083967667027
Reviewed-on: https://go-review.googlesource.com/c/go/+/373314
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agoall: fix spelling errors found by misspell
Dan Kortschak [Mon, 20 Dec 2021 07:17:42 +0000 (17:47 +1030)]
all: fix spelling errors found by misspell

Change-Id: Icedd0c3d49259d5aee249ecb33374e9b78e0c275
Reviewed-on: https://go-review.googlesource.com/c/go/+/373376
Reviewed-by: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agocmd/internal/obj/riscv: mark stack bounds check prologue nonpreemptible
Cherry Mui [Mon, 20 Dec 2021 17:45:04 +0000 (12:45 -0500)]
cmd/internal/obj/riscv: mark stack bounds check prologue nonpreemptible

This is similar to CL 207350, for RISCV64.

May fix #50263.
Updates #35470.

Change-Id: I0d39e195e8254d65fa1aca1cdf1fc553aa8b7cba
Reviewed-on: https://go-review.googlesource.com/c/go/+/373434
Trust: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
Reviewed-by: David Chase <drchase@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agogo/types: better error message when using comparable in union
Robert Griesemer [Fri, 17 Dec 2021 00:28:38 +0000 (16:28 -0800)]
go/types: better error message when using comparable in union

This is a port of CL 372674 from types2 to go/types with
minor adjustments for error handling.

For #49602.

Change-Id: I726081325a2ff2d5690d11ddc8a830bbcbd8ab33
Reviewed-on: https://go-review.googlesource.com/c/go/+/372954
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agogo/types, types2: delay "does not satisfy comparable" error until needed
Robert Griesemer [Thu, 16 Dec 2021 00:13:05 +0000 (16:13 -0800)]
go/types, types2: delay "does not satisfy comparable" error until needed

Fixes #49112.

Change-Id: I8effbca7bcbb257b18fd4d3d1914fd10d4afaaae
Reviewed-on: https://go-review.googlesource.com/c/go/+/372594
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
3 years agoruntime: mgc.go typo fix: becuse -> because
clamyang [Sun, 19 Dec 2021 03:03:38 +0000 (03:03 +0000)]
runtime: mgc.go typo fix: becuse -> because

Change-Id: I5019d5b9520e47a99a6136f615b6c9468073cc3c
GitHub-Last-Rev: 1a5392925a0c4e9b2915620fee3efa79ae14af20
GitHub-Pull-Request: golang/go#50239
Reviewed-on: https://go-review.googlesource.com/c/go/+/373055
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Trust: Brad Fitzpatrick <bradfitz@golang.org>

3 years agocmd/link: force eager binding when using plugins on darwin
Cherry Mui [Thu, 16 Dec 2021 19:33:13 +0000 (14:33 -0500)]
cmd/link: force eager binding when using plugins on darwin

When building/using plugins on darwin, we need to use flat
namespace so the same symbol from the main executable and the
plugin can be resolved to the same address. Apparently, when using
flat namespace the dynamic linker can hang at forkExec when
resolving a lazy binding. Work around it by forcing early bindings.

Fixes #38824.

Change-Id: I983aa0a0960b15bf3f7871382e8231ee244655f4
Reviewed-on: https://go-review.googlesource.com/c/go/+/372798
Trust: Cherry Mui <cherryyz@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agodoc/go1.18: document union element restriction
Ian Lance Taylor [Fri, 17 Dec 2021 01:54:27 +0000 (17:54 -0800)]
doc/go1.18: document union element restriction

For #47694

Change-Id: I9af871a4a45b002e72629904011aac8f076617f1
Reviewed-on: https://go-review.googlesource.com/c/go/+/372974
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
3 years agospec: describe constraint parsing ambiguity and work-around more precisely
Robert Griesemer [Thu, 16 Dec 2021 18:25:07 +0000 (10:25 -0800)]
spec: describe constraint parsing ambiguity and work-around more precisely

The new description matches the implementation (CL 370774).

Also, in the section on type constraints, use "defines" instead of
"determines" because the constraint interface defines the type set
which is precisely the set of acceptable type arguments.

For #49482.

Change-Id: I6f30f49100e8ba8bec0a0f1b450f88cae54312eb
Reviewed-on: https://go-review.googlesource.com/c/go/+/372874
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agospec: clarify that comparable cannot be a union element
Robert Griesemer [Thu, 16 Dec 2021 04:03:11 +0000 (20:03 -0800)]
spec: clarify that comparable cannot be a union element

For #49602.

Change-Id: I0d3ff8f087dffb3409918494147fd1dceff7514d
Reviewed-on: https://go-review.googlesource.com/c/go/+/372694
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agonet: lengthen arbitrary SetDeadline timeout by a few orders of magnitude
Bryan C. Mills [Thu, 16 Dec 2021 17:32:39 +0000 (12:32 -0500)]
net: lengthen arbitrary SetDeadline timeout by a few orders of magnitude

The "someTimeout" constant in the net test is “just to test that
net.Conn implementations don't explode when their SetFooDeadline
methods are called”. It was set to 10 seconds, which is short enough
that it could actually matter on some platforms.

Since the point of the constant is just to make sure methods don't
explode, we should set it to be at least a couple of orders of
magnitude longer than the test: then it is guaranteed not to have any
unintended side-effects.

Fixes #50227

Change-Id: If97ae7bef5e7f16b336d09ccc37f5ea2ea7e70b3
Reviewed-on: https://go-review.googlesource.com/c/go/+/372796
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agoos/exec: skip TestContextCancel on netbsd/arm64
Bryan C. Mills [Thu, 16 Dec 2021 15:40:47 +0000 (10:40 -0500)]
os/exec: skip TestContextCancel on netbsd/arm64

For #42061

Change-Id: I3b4c774ad9e375d4bfef1cfb4336c35ed30a6430
Reviewed-on: https://go-review.googlesource.com/c/go/+/372795
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
3 years agogo/types, types2: use compiler error message for undefined operators
Robert Griesemer [Thu, 16 Dec 2021 18:50:34 +0000 (10:50 -0800)]
go/types, types2: use compiler error message for undefined operators

For #48712.

Change-Id: I1596fe8688f093e0e92cf5b8d5501aac8631324e
Reviewed-on: https://go-review.googlesource.com/c/go/+/372894
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile/internal/types2: better error message when using comparable in union
Robert Griesemer [Thu, 16 Dec 2021 03:50:33 +0000 (19:50 -0800)]
cmd/compile/internal/types2: better error message when using comparable in union

Fixes #49602.

Change-Id: I3499f8a485a2c8ec8afc74c5ef7b20d42c943a05
Reviewed-on: https://go-review.googlesource.com/c/go/+/372674
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/compile/internal/syntax: fix parsing of type parameter lists
Robert Griesemer [Thu, 9 Dec 2021 04:32:29 +0000 (20:32 -0800)]
cmd/compile/internal/syntax: fix parsing of type parameter lists

The parser cannot distinguish a type parameter list of the form

[P *T ] or
[P (T)]

where T is not a type literal from an array length specification
P*T (product) or P(T) (constant-valued function call) and thus
interprets these forms as the start of array types.

This ambiguity must be resolved explicitly by placing *T inside
an interface, adding a trailing comma, or by leaving parentheses
away where possible.

This CL adjusts the parser such that these forms are
interpreted as (the beginning) of type parameter lists
if the token after P*T or P(T) is a comma, or if T is
a type literal.

This CL also adjusts the printer to print a comma if
necessary to avoid this ambiguity, and adds additional
printer tests.

Fixes #49482

Change-Id: I36328e2a7d9439c39ba0349837c445542549e84e
Reviewed-on: https://go-review.googlesource.com/c/go/+/370774
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agocmd/go: update go work docs
Michael Matloob [Wed, 15 Dec 2021 20:07:22 +0000 (15:07 -0500)]
cmd/go: update go work docs

Change-Id: If28cc02a5a6f9bf2c7f2550de77ca0b31c1b35ee
Reviewed-on: https://go-review.googlesource.com/c/go/+/372494
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
Reviewed-by: Bryan Mills <bcmills@google.com>
3 years agocmd/compile: only avoid escaping package paths for "go.shape"
Dan Scales [Thu, 16 Dec 2021 21:24:40 +0000 (13:24 -0800)]
cmd/compile: only avoid escaping package paths for "go.shape"

We have code that intends to avoid escaping the package path for
built-in packages. But it is hard to determine which packages are
built-in from a general rule, and we really only want to avoid escaping
for the "go.shape" package (since that gives ugly shape type names). So,
fix the code to only avoid escaping the package path specifically for
the "go.shape" package.

Fixes #50200

Change-Id: Ibaedd7690b99a173007c608c5dfa783ef82b326d
Reviewed-on: https://go-review.googlesource.com/c/go/+/372934
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
3 years agocmd/go/internal/vcs: prevent Git signatures from breaking commit time parsing
Andre Marianiello [Tue, 14 Dec 2021 22:50:28 +0000 (22:50 +0000)]
cmd/go/internal/vcs: prevent Git signatures from breaking commit time parsing

When a user has showSignature=true set in their Git config and the
commit in question has a signature, the git-show command will output
information about that signature. When this happens, the logic that
tries to parsing a timestamp from the git-show output chokes on this
signature information and the build stamping fails. This change prevents
commit signature information from being displayed even if
showSignature=true, preventing this issue.

Change-Id: I98d0a6fdd1e90dd1b91e0394713b6eb286a69d1a
GitHub-Last-Rev: 610706e23e33a037b9abede2ba0a926c0f336814
GitHub-Pull-Request: golang/go#49790
Reviewed-on: https://go-review.googlesource.com/c/go/+/367034
Trust: Cherry Mui <cherryyz@google.com>
Reviewed-by: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agomisc/cgo/testcarchive: log command output more consistently
Bryan C. Mills [Fri, 3 Dec 2021 15:33:02 +0000 (10:33 -0500)]
misc/cgo/testcarchive: log command output more consistently

Also check that executables exist immediately after building them
in parallel tests.

The parallel tests in this package occasionally fail with
"no such file or directory", implying that either the build
command failed to actually write out the binary or something
concurrently deleted it.

This is purely a shot in the dark, but I'm hoping that perhaps
the stderr output from one of these commands will shed some
light on the underlying failure mode.

For #49693

Change-Id: I2e768190c56053550879b89a3ac88c027d4741dd
Reviewed-on: https://go-review.googlesource.com/c/go/+/369034
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>

3 years agonet/http/pprof: skip TestDeltaProfile on openbsd/arm
Bryan C. Mills [Thu, 16 Dec 2021 15:22:37 +0000 (10:22 -0500)]
net/http/pprof: skip TestDeltaProfile on openbsd/arm

It is observed to be flaky on the only openbsd/arm builder.
Skipping on that platform until someone can investigate.

For #50218

Change-Id: Id3a6dc12b93b3cec67870d8d81bd608c4589c952
Reviewed-on: https://go-review.googlesource.com/c/go/+/372794
Trust: Bryan Mills <bcmills@google.com>
Run-TryBot: Bryan Mills <bcmills@google.com>
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>