Dan Scales [Fri, 15 Oct 2021 18:54:25 +0000 (11:54 -0700)]
cmd/compile: convert to using a map in getInstInfo, rather than SetImplicit()
SetImplicit() has an explicit meaning and really shouldn't be used in
this way - its use is left over from early prototype of the dictionary
code. Convert from using SetImplicit to just using a map during
traversal.
Change-Id: I3d257c101a859f000e159d7ced307d1b7cf990d4
Reviewed-on: https://go-review.googlesource.com/c/go/+/356310
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Michael Pratt [Thu, 14 Oct 2021 22:18:49 +0000 (18:18 -0400)]
runtime: consistently access pollDesc r/w Gs with atomics
Both netpollblock and netpollunblock read gpp using a non-atomic load.
When consuming a ready event, netpollblock clears gpp using a non-atomic
store, thus skipping a barrier.
Thus on systems with weak memory ordering, a sequence like so this is
possible:
i.e., without a happens-before edge between (2) and (3), netpollunblock
may read the stale value of gpp.
Switch these access to use atomic loads and stores in order to create
these edges.
For ease of future maintainance, I've simply changed rg and wg to always
be accessed atomically, though I don't believe pollOpen or pollClose
require atomics today.
Fixes #48925
Change-Id: I903ea667eea320277610b4f969129935731520c3
Reviewed-on: https://go-review.googlesource.com/c/go/+/355952
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> Reviewed-by: David Chase <drchase@google.com>
Jay Conrod [Fri, 15 Oct 2021 18:57:58 +0000 (11:57 -0700)]
cmd/go: allow nested VCS repositories when preparing build stamp
The go command no longer reports an error when invoked in a repository
nested inside another. This check is still used by 'go get' in GOPATH
mode when locating a repository, but it's not needed when preparing
the build stamp.
Fixes #49004
Change-Id: I4ed4dcc04174d2d42da8651d47e52ab1d7d66e35
Reviewed-on: https://go-review.googlesource.com/c/go/+/356309
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>
Keith Randall [Fri, 15 Oct 2021 15:06:58 +0000 (08:06 -0700)]
cmd/compile: make for loops with range statements not terminating
Fixes #49003
Change-Id: If09c6f028dce5440b1be238612653ffdd626113a
Reviewed-on: https://go-review.googlesource.com/c/go/+/356189
Trust: Keith Randall <khr@golang.org> Reviewed-by: roger peppe <rogpeppe@gmail.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Robert Griesemer <gri@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Katie Hockman [Wed, 13 Oct 2021 20:49:27 +0000 (16:49 -0400)]
internal/fuzz: fix bugs with minimization
This pulls in some code and tests from CL 353355.
This change makes some refactors for when we read
to and write from memory during minimization.
That fixes a bug when minimizing interesting inputs.
Now, if an error occurs while minimizing an interesting
input, that value will continue to be minimized as a
crash, and returned to the user.
This change also allows minimization of a crash that
occurred during the warmup phase. We don't want to
minimize failures in the seed corpus, but if an entry
in the cache causes a new failure, then there's no
compelling reason why we shouldn't try to minimize it.
Fixes #48731
Change-Id: I7262cecd8ea7ae6fdf932f3a36db55fb062a1f2a
Reviewed-on: https://go-review.googlesource.com/c/go/+/355691
Trust: Katie Hockman <katie@golang.org>
Run-TryBot: Katie Hockman <katie@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Roland Shoemaker <roland@golang.org>
Keith Randall [Wed, 15 Sep 2021 16:56:09 +0000 (09:56 -0700)]
reflect: make Elem panic on bad notinheap pointers
This CL fixes the subtle issue that Elem can promote a
not-in-heap pointer, which could be any bit pattern, into an
unsafe.Pointer, which the garbage collector can see. If that
resulting value is bad, it can crash the GC.
Make sure that we don't introduce bad pointers that way. We can
make Elem() panic, because any such bad pointers are in the Go heap,
and not-in-heap pointers are not allowed to point into the Go heap.
Update #48399
Change-Id: Ieaf35a611b16b4dfb5e907e229ed4a2aed30e18c
Reviewed-on: https://go-review.googlesource.com/c/go/+/350153
Trust: Keith Randall <khr@golang.org>
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Robert Griesemer [Thu, 14 Oct 2021 01:50:06 +0000 (18:50 -0700)]
cmd/compile/internal/types2: add debugging support for delayed actions
Add a simple mechanism to provide formatted descriptions for
delayed actions. The comment strings are printed when tracing
is enabled and the delayed action is executed. This results
in more easily decipherable tracing output.
Requires debug mode in order to minimize the overhead during normal
execution. Use the mechanism in a few places to show typical use.
Also cleaned up a few unrelated comments.
Change-Id: Ic273c380c3963341500396ec62b694d143c25de2
Reviewed-on: https://go-review.googlesource.com/c/go/+/355871
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org> Reviewed-by: Robert Findley <rfindley@google.com>
Dan Scales [Sun, 19 Sep 2021 16:13:47 +0000 (09:13 -0700)]
cmd/compile: support new fully-inst types referenced during inlining
Modify the phase for creating needed function/method instantiations and
modifying functions to use those instantiations, so that the phase is
self-contained and can be called again after inlining. This is to deal
with the issue that inlining may reveal new fully-instantiated types
whose methods must be instantiated.
With this change, we have an extra phase for instantiation after
inlining, to take care of the new fully-instantiated types that have
shown up during inlining. We call inline.InlineCalls() for any new
instantiated functions that are created.
Change-Id: I4ddf0b1907e5f1f7d45891db7876455a99381133
Reviewed-on: https://go-review.googlesource.com/c/go/+/352870
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
Jay Conrod [Thu, 14 Oct 2021 23:47:47 +0000 (16:47 -0700)]
cmd/go: don't stamp build or vcs info for GOROOT binaries
Fixes a test failure in cmd/go TestScript/mod_outside.
make.bash (cmd/dist) builds everything with -gcflags=all= -ldflags=all=
by default. If those no-op flags aren't used, all GOROOT binaries
appear stale.
It's likely safe to omit those flags in cmd/dist if they're
empty. Checking out a new commit in GOROOT would always cause
staleness since the VCS info would change.
For #37475
Change-Id: Ic9aa0f3b7318e05fbb2f7d2c008ad07a4c61952f
Reviewed-on: https://go-review.googlesource.com/c/go/+/356014
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>
Cuong Manh Le [Wed, 13 Oct 2021 10:11:16 +0000 (17:11 +0700)]
cmd/compile: fix irgen reports wrong error message for misuse of //go:embed
Fixes #48230
Change-Id: Ic6490e065e7e79793faa0d0201dc94f5fcea694a
Reviewed-on: https://go-review.googlesource.com/c/go/+/355529
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Dan Scales <danscales@google.com>
Robert Findley [Thu, 14 Oct 2021 15:34:00 +0000 (11:34 -0400)]
cmd/go: produce a better error for generic test functions
Test functions with type parameters cannot be invoked without type
arguments, so cmd/go would previously fail with a type-checking error in
the test harness.
Avoid this by producing an error explaining that test functions cannot
have type parameters.
Fixes #48953
Change-Id: I4e39c9b7a06c964fad5f12d74d262aa090c25d79
Reviewed-on: https://go-review.googlesource.com/c/go/+/355889
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com> Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Jay Conrod [Tue, 12 Oct 2021 22:08:54 +0000 (15:08 -0700)]
cmd/go: stamp tags and flags in build info
Toolchain flags (like -gcflags), build tags (including race and msan),
and cgo variables (including CGO_ENABLED, CGO_CPPFLAGS and others) are
now stamped into binaries.
For #37475
Change-Id: I9023e682c0618f91805434946c6bc937536b69bb
Reviewed-on: https://go-review.googlesource.com/c/go/+/355493
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>
Jay Conrod [Fri, 1 Oct 2021 18:21:49 +0000 (11:21 -0700)]
cmd/go: stamp VCS revision and uncommitted status into binaries
When the go command builds a binary, it will now stamp the current
revision from the local Git or Mercurial repository, and it will also
stamp whether there are uncommitted edited or untracked files. Only
Git and Mercurial are supported for now.
If no repository is found containing the current working directory
(where the go command was started), or if either the main package
directory or the containing module's root directory is outside the
repository, no VCS information will be stamped. If the VCS tool is
missing or returns an error, that error is reported on the main
package (hinting that -buildvcs may be disabled).
This change introduces the -buildvcs flag, which is enabled by
default. When disabled, VCS information won't be stamped when it would
be otherwise.
Stamped information may be read using 'go version -m file' or
debug.ReadBuildInfo.
For #37475
Change-Id: I4e7d3159e1c270d85869ad99f10502e546e7582d
Reviewed-on: https://go-review.googlesource.com/c/go/+/353930
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>
Jay Conrod [Thu, 30 Sep 2021 17:25:49 +0000 (10:25 -0700)]
debug/buildinfo: new package with Read and ReadFile
These functions provide access to module information stamped into Go
binaries. In the future, they'll provide access to other information
(like VCS info).
These functions are added in a new package instead of runtime/debug
since they use binary parsing packages like debug/elf, which would
make runtime/debug an unacceptably heavy dependency. The types in
runtime/debug are still used; debug/buildinfo uses them via type
aliases.
This information is already available for the running binary through
debug.ReadBuildInfo and for other binaries with 'go version -m', but
until now, there hasn't been a way to get it for other binaries
without installing cmd/go.
This change copies most of the code in cmd/go/internal/version. A
later CL will migrate 'go version -m' to use this package.
For #37475
Fixes #39301
Change-Id: I0fbe0896e04f12ef81c6d79fb61b20daede86159
Reviewed-on: https://go-review.googlesource.com/c/go/+/353887
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com> Reviewed-by: Bryan C. Mills <bcmills@google.com>
Jay Conrod [Thu, 30 Sep 2021 21:37:30 +0000 (14:37 -0700)]
cmd/go: move module build info formatting into runtime/debug
Previously, modload.PackageBuildInfo returned a string containing
information about modules used to build an executable. This string is
embedded in the binary and can be read with debug.ReadBuildInfo or
'go version -m'.
With this change, debug.BuildInfo now has a MarshalText method that
returns a string in the same format as modload.PackageBuildInfo.
Package.load now calls Package.setBuildInfo, which constructs a
debug.BuildInfo, formats it with MarshalText, then sets
Package.Internal.BuildInfo. This is equivalent to what
modload.PackageBuildInfo did.
modload.PackageBuildInfo is deleted, since it's no longer used.
For #37475
Change-Id: I5875a98cb64737637fec2a450ab2ffa7f1805707
Reviewed-on: https://go-review.googlesource.com/c/go/+/353886
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com> Reviewed-by: Bryan C. Mills <bcmills@google.com>
Damien Neil [Thu, 5 Aug 2021 18:32:43 +0000 (11:32 -0700)]
sync: remove TestWaitGroupMisuse2 and TestWaitGroupMisuse3
These tests are inherently nondeterministic: They exercise a racy
code path for up to one million iterations, and require that an
error occur at least once.
TestWaitGroupMisuse2 in particular is an ongoing source of trybot
flakiness.
Fixes #38163.
Change-Id: Ibbbda2c998c915333487ad262d3df6829de01c2b
Reviewed-on: https://go-review.googlesource.com/c/go/+/340249
Trust: Damien Neil <dneil@google.com>
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Run-TryBot: Damien Neil <dneil@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Robert Griesemer [Thu, 14 Oct 2021 04:13:05 +0000 (21:13 -0700)]
cmd/compile/internal/types2: avoid infinite expansion for invalid recursive generic types
The algorithm for detecting invalid recursive types that
expand indefinitely suffered from the exact problem is was
intended to detect: if the indefinite expansion is happening
through type parameters, the algorithm ended up in an infinite
sequence of instantiations. (This is only a problem for generic
types).
Changed the algorithm to always only consider the "original"
uninstantiated types. This avoids the problem but it will also
not detect some invalid recursive generic types anymore. That
requires a more sophisticated type flow analysis.
Opened #48962 to track.
Addressed with help from @findleyr.
For #48951.
Change-Id: Ie29cea8f810dae55153dbb1b17c9390cd823c2d9
Reviewed-on: https://go-review.googlesource.com/c/go/+/355732
Trust: Robert Griesemer <gri@golang.org> Reviewed-by: Robert Findley <rfindley@google.com>
Ian Lance Taylor [Wed, 13 Oct 2021 19:46:33 +0000 (12:46 -0700)]
cmd/link/internal/ld: don't use linkname before 1.12
Before the 1.12 release the use of linkname did not prevent a compiler
error for an empty function body (see #23311). Add some build tags so
that cmd/link will build with earlier releases.
It's true that we currently require Go 1.16 as the bootstrap compiler (#44505).
But for this simple case keep things working with older compilers for now.
Change-Id: I67fe021406096c64c01d6e2c9adbcc4388988a6e
Reviewed-on: https://go-review.googlesource.com/c/go/+/355690
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
Dan Scales [Wed, 13 Oct 2021 17:27:01 +0000 (10:27 -0700)]
cmd/compile: fix inst_test.go for riscv5
On riscv5, apparently extra wrappers (trampolines) are created for some
functions with the suffix "-tramp". Modify inst_test.go to not match
these "-tramp" wrappers.
Change-Id: I754c724f03555c30f8e1b5cfc08152555bbbcf17
Reviewed-on: https://go-review.googlesource.com/c/go/+/355650
Run-TryBot: Dan Scales <danscales@google.com> Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Matthew Dempsky [Tue, 12 Oct 2021 22:38:52 +0000 (15:38 -0700)]
unsafe: optimize Slice bounds checking
This reduces the number of branches to bounds check non-empty slices
from 5 to 3. It does also increase the number of branches to handle
empty slices from 1 to 3; but for non-panicking calls, they should all
be predictable.
Updates #48798.
Change-Id: I3ffa66857096486f4dee417e1a66eb8fdf7a3777
Reviewed-on: https://go-review.googlesource.com/c/go/+/355490
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Matthew Dempsky [Tue, 12 Oct 2021 22:31:25 +0000 (15:31 -0700)]
unsafe: allow unsafe.Slice up to end of address space
Allow the user to construct slices that are larger than the Go heap as
long as they don't overflow the address space.
Updates #48798.
Change-Id: I659c8334d04676e1f253b9c3cd499eab9b9f989a
Reviewed-on: https://go-review.googlesource.com/c/go/+/355489
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
James Harris [Fri, 17 Sep 2021 02:29:00 +0000 (02:29 +0000)]
net/http/httputil: allow MIME parameters when detecting SSE in ReverseProxy
This change allows httputil.ReverseProxy to detect SSE (server-sent events)
content when the response's Content-Type header includes MIME parameters,
such as "text/event-stream;charset=utf-8".
Prior to this change the value of the Content-Type header was compared
directly to the literal "text/event-stream". This caused a false-negative
which failed to set the FlushInterval correctly when MIME parameters were
present.
Change-Id: If8bb43efb78787b6519d7fe7599ca018a0da0023
GitHub-Last-Rev: 224518c5eb9686ee050c79f5f853ebacfdf6fc42
GitHub-Pull-Request: golang/go#48427
Reviewed-on: https://go-review.googlesource.com/c/go/+/350509
Trust: Alexander Rakoczy <alex@golang.org>
Trust: Damien Neil <dneil@google.com>
Run-TryBot: Alexander Rakoczy <alex@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Damien Neil <dneil@google.com>
Tamir Duberstein [Tue, 12 Oct 2021 19:03:25 +0000 (15:03 -0400)]
net: remove timeout in TestDialTimeoutMaxDuration
This test seems only to be testing that Dial does not time out
immediately as a result of integer overflow; the precise time taken to
connect is immaterial. Replace naked loop with sub-tests.
Fixes #43069.
Change-Id: Ib5e38a1d8cd191b74c2bc7c26bef57b180e16f68
Reviewed-on: https://go-review.googlesource.com/c/go/+/355390
Trust: Michael Pratt <mpratt@google.com>
Run-TryBot: Michael Pratt <mpratt@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Damien Neil <dneil@google.com>
Koichi Shiraishi [Mon, 11 Oct 2021 15:33:43 +0000 (00:33 +0900)]
internal/poll,net: support poll.Sendfile for darwin
darwin already supports syscall.Sendfile.
Change-Id: Id3db06591ffad0550b4173bacddeb0acfe355f9e
Reviewed-on: https://go-review.googlesource.com/c/go/+/355109 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
kjgorman [Sun, 10 Oct 2021 13:30:26 +0000 (14:30 +0100)]
bytes: fix Cut godoc typo
If sep does not appear in s, we return nil here
rather than the empty string. Presumably the docs
were copied from the strings package implementation
and brought that along.
Change-Id: I3706c94fe04abd9e2a2c8840f7cc6116386965c8
Reviewed-on: https://go-review.googlesource.com/c/go/+/354969 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
Amelia Downs [Mon, 30 Aug 2021 21:05:36 +0000 (17:05 -0400)]
net: add examples for remaining IP functions
Fixes #48848
Change-Id: If85fc16ed10a26eca840ed128ab47c5f00776ddd
Reviewed-on: https://go-review.googlesource.com/c/go/+/346289 Reviewed-by: Amelia Downs <adowns@vmware.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
When a process is forcefully killed (for example, with SIGKILL on unix), its ProccessState.Exited() will return false.
Change-Id: I8cebc8d28f2ba1b687c145f6d941647fc3a10665
GitHub-Last-Rev: 414ae1980fe61dd530c46cc2c7f4de3a5bc83ed5
GitHub-Pull-Request: golang/go#48871
Reviewed-on: https://go-review.googlesource.com/c/go/+/354698 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
Run-TryBot: Alexander Rakoczy <alex@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
cmd/link: disable weak reference in itab if build with "-linkshared"
When build with "-linkshared", we can't tell if the interface method will be used or not. It can be used in shared library.
Fixes #47873
Change-Id: Iba12812f199b7679cf2fd41a304268d6d6dd03c6
Reviewed-on: https://go-review.googlesource.com/c/go/+/350189 Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
helbing [Tue, 12 Oct 2021 23:33:50 +0000 (23:33 +0000)]
embed: add example
Change-Id: I4e50e469047ac7efbf4ed464e238000dbdf53d6b
GitHub-Last-Rev: 8d29b73d1160b4498a38aa3ef6530ee5c9353186
GitHub-Pull-Request: golang/go#48785
Reviewed-on: https://go-review.googlesource.com/c/go/+/353936
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
nicksherron [Wed, 13 Oct 2021 01:39:13 +0000 (01:39 +0000)]
all: fix spelling mistakes
Corrections were only made to comments and can be reproduced with the
following sed.
sed -i 's/communciation/communication/g' src/internal/fuzz/sys_windows.go
sed -i 's/communciation/communication/g' src/internal/fuzz/sys_posix.go
sed -i 's/substitued/substituted/g' src/cmd/compile/internal/noder/irgen.go
Change-Id: I435aee2fdb41e73b694e6233d52742559d32cf7e
GitHub-Last-Rev: a471e3d123aa6c8bb9b091e65de5d16609fb8db1
GitHub-Pull-Request: golang/go#48932
Reviewed-on: https://go-review.googlesource.com/c/go/+/355494 Reviewed-by: Robert Griesemer <gri@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Dan Scales [Tue, 12 Oct 2021 19:52:59 +0000 (12:52 -0700)]
cmd/compile: allow spaces in types.(*Type).LinkString()
Go back to allowing spaces in types.(*Type).LinkSring().
Delve folks prefer that there are spaces in type names, if needed, since
DWARF expects type names to be "a string representing the name as it
appears in the source program". At this point, it doesn't necessarily
seem worth having a separate function that removes spaces, only for use
when printing the type names in a function/method instantiation.
Most names of function/method instantiations will still not have spaces,
since they most type args are named or builtin types. I confirmed that
we are back to the original definition of LinkString(), except for some
comment changes.
Change-Id: Iab7143f659ddea4f099c6230cb134edf8aaef868
Reviewed-on: https://go-review.googlesource.com/c/go/+/355354
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Dan Scales <danscales@google.com> Reviewed-by: Keith Randall <khr@golang.org>
Robert Findley [Tue, 12 Oct 2021 15:54:09 +0000 (11:54 -0400)]
cmd/api: use placeholder names for type parameters
Changing type parameter names is not a breaking API change, so we should
not include these names in the output of cmd/api. Instead print a
placeholder '$<index>' wherever type parameters are referenced.
This is valid for cmd/api as there is at most one type parameter list in
scope for any exported declaration. If we ever support method type
parameters, we'll need to revisit this syntax.
Change-Id: I7e677b1dab6ffeb0b79afefdb8d2580bef93891c
Reviewed-on: https://go-review.googlesource.com/c/go/+/355389
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Tamir Duberstein [Tue, 12 Oct 2021 20:00:23 +0000 (16:00 -0400)]
os: Simplify size using io.Discard.
Change-Id: Ib7cc86643a3dcae788a94472e54de171e0d655fc
Reviewed-on: https://go-review.googlesource.com/c/go/+/355449
Trust: Michael Pratt <mpratt@google.com>
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Michael Pratt <mpratt@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Michael Pratt <mpratt@google.com>
Bryan C. Mills [Wed, 18 Mar 2020 01:29:54 +0000 (21:29 -0400)]
cmd/dist: run a checkNotStale on the builders before running tests
This should be a no-op, but if the sharded builders are for some
reason distributing stale snapshots — or testing them with mismatched
environments — this should catch them out at a relatively low cost
(#24300 notwithstanding).
Given the frequently at which (*tester).runPending already checks for
staleness, we do not expect the impact of this extra check to be
significant for most builders.
For #33598
Updates #24300
Change-Id: I197d6a69c72e2eec9e4563b459206de76c89e8a1
Reviewed-on: https://go-review.googlesource.com/c/go/+/223755
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Dan Scales [Tue, 10 Aug 2021 21:51:57 +0000 (14:51 -0700)]
[dev.typeparams] cmd/compile: add test for number of instantiations
Add a test for a generic sort function, operating on several different
pointer types (across two packages), so they should all share the same
shape-based instantiation. Actually check that only one instantiation of
Sort is created using 'go tool nm', and also check that the output is
correct.
In order to do the test on the executable using 'go nm', added this as a
'go test' in cmd/compile/internal/test.
Added the genembed.go test that I meant to include with a previous CL.
Change-Id: I9962913c2f1809484c2b1dfef3b07e4c8770731c
Reviewed-on: https://go-review.googlesource.com/c/go/+/354696
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Jay Conrod [Wed, 6 Oct 2021 21:58:22 +0000 (14:58 -0700)]
testing: don't create unique subtest names while fuzzing
T.Run uses a map[string]int64 to keep track of subtest names that may
be returned through T.Name. T.Name can't return duplicate names for
subtests started with T.Run.
If a fuzz target calls T.Run, this map takes a large amount of memory,
since there are a very large number of subtests that would
otherwise have duplicate names, and the map stores one entry per subtest.
The unique suffixes are not useful (and may be confusing) since the
full sequence of tests cannot be re-run deterministically.
This change deletes all entries in the map before each call to the
function being fuzzed. There is a slight change in the contract of
T.Name while fuzzing.
This change was discussed in CL 351452.
Fixes #44517
Change-Id: I3093a2e5568099ce54aca1006fac84a6fd2c3ddf
Reviewed-on: https://go-review.googlesource.com/c/go/+/354430
Trust: Jay Conrod <jayconrod@google.com>
Trust: Katie Hockman <katie@golang.org>
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com>
Run-TryBot: Katie Hockman <katie@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Katie Hockman <katie@golang.org>
Dan Scales [Mon, 4 Oct 2021 18:56:12 +0000 (11:56 -0700)]
cmd/compile: some fixes in type substituter for Instantiate
In the case in (*TSubster).Type() that we were running into an
incomplete underlying type (TFORW), we should just be immediately
returning the type returned by ts.SubstForwFunc(forw), since that call
returns a proper type node, and has set up any remaining work that has
to be done when we get done with the current top-level type definition.
(For import, that function is doInst, which does an Instantiate of the
new substituted type, with the delayed part via deferredInstStack.) We
should not continue doing the later parts of (*TSubster).Type(), since
the underlying type may not yet have its methods filled in, etc.
Also, in Instantiate(), we need to put the desired new type on
deferredInstStack, even if the base type node already exists, if the
type node is in TFORW state. This is now exactly the case when
Instantiate is called from (*TSubster).Type via doInst, since
(*TSubster).Type has already called NewIncompleteNamedType().
Fixes #48716
Fixes #48889
Change-Id: Icd6be5721c4ac75bf8869b8bbdeca50069d632ec
Reviewed-on: https://go-review.googlesource.com/c/go/+/355250
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Robert Findley [Wed, 6 Oct 2021 17:51:21 +0000 (13:51 -0400)]
go/types: export TypeParam.Index and remove TypeParam._SetId
This change resolves a TODO regarding a couple uncertain APIs for
types.TypeParam. In the case of TypeParam._Index, we've decided it is
worth exporting. In the case of TypeParam._SetId, we've decided it is
unnecessary.
This aligns go/types with types2 (a doc comment in types2 is also
updated).
Updates #47916
Change-Id: I705e8b3437d014775c473e2f8be6f32b1540bb0e
Reviewed-on: https://go-review.googlesource.com/c/go/+/354370
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Jay Conrod [Mon, 11 Oct 2021 22:51:10 +0000 (15:51 -0700)]
cmd/go: adjust documentation mentioning 'go get'
In module-aware mode, 'go get' no longer builds or installs packages.
- 'go generate' explains build commands do not run generate
commands. 'go get' is no longer a build command, so this CL removes
mention of it.
- 'go get' will continue to accept build flags, but they're
ignored. The documentation no longer mentions them, though it does
mention -x for printing VCS commands.
For #43684
Change-Id: I1eea7241eecf72ba9f98238b729d91cc77ec7665
Reviewed-on: https://go-review.googlesource.com/c/go/+/355209
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>
Cherry Mui [Tue, 12 Oct 2021 16:38:42 +0000 (12:38 -0400)]
cmd/compile: do not reuse dead value in expand_calls pass
We reuse a value for the same selector on the same arg. But if the
value is already marked dead, don't reuse it. A use of an
OpInvalid will confuse the compiler.
Fixes #48916.
Change-Id: I15b9e15b49f6e1991fe91df246cd12a193385e85
Reviewed-on: https://go-review.googlesource.com/c/go/+/355409
Trust: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com> Reviewed-by: David Chase <drchase@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Katie Hockman [Thu, 7 Oct 2021 20:26:36 +0000 (16:26 -0400)]
testing: fix -run behavior with fuzz tests
This change fixes some issues with -run, and
the subsequent command line output when running
in verbose mode. It replaces CorpusEntry.Name
with CorpusEntry.Path, and refactors the code
accordingly.
This change also adds a lot of additional tests
which check explicit command line output when
fuzz targets are run without fuzzing. This will
be important to avoid regressions.
Updates #48149
Change-Id: If34b1f51db646317b7b51c3c38ae53231d01f568
Reviewed-on: https://go-review.googlesource.com/c/go/+/354632
Trust: Katie Hockman <katie@golang.org>
Run-TryBot: Katie Hockman <katie@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Jay Conrod <jayconrod@google.com>
Paul E. Murphy [Tue, 9 Mar 2021 22:55:20 +0000 (16:55 -0600)]
cmd/internal/obj/ppc64: support alignment of prefixed insn
Insert machine NOPs when a prefixed instruction crosses a 64B boundary.
ISA 3.1 prohibits prefixed instructions being placed across them. Such
instructions generate SIGILL if executed.
Likewise, adjust the function alignment to guarantee such instructions
can never cross one. And, don't pad the PC based on alignment. The
linker can fit these more optimally.
Likewise, include the function alignment when printing function debug
information. This is needed to verify function alignment happens.
Updates #44549
Change-Id: I434fb0ee4e984ca00dc4566f7569c3bcdf93f910
Reviewed-on: https://go-review.googlesource.com/c/go/+/347050
Run-TryBot: Paul Murphy <murp@ibm.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Change-Id: I35ac0ed862ba6ee921ba9aee257bc19828abaa82
Reviewed-on: https://go-review.googlesource.com/c/go/+/354710
Trust: Nigel Tao <nigeltao@golang.org>
Run-TryBot: Nigel Tao <nigeltao@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Andrew Gerrand <adg@golang.org>
Ian Lance Taylor [Thu, 7 Oct 2021 20:05:16 +0000 (13:05 -0700)]
cmd/api: support type parameters
Fixes #48706
Change-Id: If0f8d0b49300027e3b2b46f6870302acf2e00f4e
Reviewed-on: https://go-review.googlesource.com/c/go/+/354612
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Robert Findley [Mon, 11 Oct 2021 13:59:40 +0000 (09:59 -0400)]
cmd/compile/internal/types2: avoid duplicate errors for invalid bounds
Resolve a TODO from an earlier CL: we should only check type parameter
bounds once in collectTypeParams.
Change-Id: Icf6053ec359f8ac8143cf68ee2defd504f8f86e6
Reviewed-on: https://go-review.googlesource.com/c/go/+/355069
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Sun, 10 Oct 2021 14:53:06 +0000 (10:53 -0400)]
go/types: mark implicit interfaces as such
This is a straightforward port of CL 353396 to go/types.
For #48424
Change-Id: I3040c2ad3a8c9573026277de01deb9c47953cec8
Reviewed-on: https://go-review.googlesource.com/c/go/+/354991
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
This is a partial revert of CL 353389, now that go/types supports
eliding interface in constraints.
For #48424
Change-Id: Ibde26fb66bf6bd5a4c919acffd045a57783d1b47
Reviewed-on: https://go-review.googlesource.com/c/go/+/354990
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Sun, 10 Oct 2021 14:43:20 +0000 (10:43 -0400)]
go/types: accept constraint literals with elided interfaces
This is a port of CL 353139 to go/types, adjusted for error reporting
and for the different representation of field lists in go/ast. A TODO is
added to verify if types2 produces redundant error messages for type
parameters sharing a bound.
For #48424
Change-Id: I3549942be0328de616d1d87d0ba621311fc53576
Reviewed-on: https://go-review.googlesource.com/c/go/+/354989
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Fri, 8 Oct 2021 20:28:35 +0000 (16:28 -0400)]
go/parser: allow eliding interface in constraint literals
This is a port of CL 353133 from cmd/compile/internal/syntax, with
significant adjustments for the mechanics of go/parser.
Some additional cleanup is made along the way: parseParameterList can
call parseParamDecl without indirection, and the tparams argument is
redundant with the closing token. Also, the error that "all type
parameters must be named" is positioned on the first unnamed type
parameter.
Error recovery in go/parser is notably worse here than the compiler
parser, so the test data had to be adjusted to synchronize positions.
Fixing this error recovery will have to wait for a later CL.
As with the compiler changes, these changes are guarded behind a flag so
that they may be easily removed if #48424 is not accepted.
For #48424
Change-Id: If87925d246f36aaab11a95442f75f659462d4286
Reviewed-on: https://go-review.googlesource.com/c/go/+/354870
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
In particular, the decoder would return io.EOF after reading a valid
message for a type specification, and then hit EOF before reading a data
item message.
Fix that by only allowing a Decode call to return io.EOF if the reader
hits EOF immediately, without successfully reading any message.
Otherwise, hitting EOF is an ErrUnexpectedEOF, like in other cases.
Also fix a net/rpc test that, coincidentally, expected an io.EOF
as an error when feeding bad non-zero data to a gob decoder.
An io.ErrUnexpectedEOF is clearly better in that scenario.
Fixes #48905.
Change-Id: Ied6a0d8ac8377f89646319a18c0380c4f2b09b85
Reviewed-on: https://go-review.googlesource.com/c/go/+/354972
Trust: Daniel Martí <mvdan@mvdan.cc>
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Rob Pike <r@golang.org>
Dan Scales [Thu, 7 Oct 2021 19:21:01 +0000 (12:21 -0700)]
cmd/compile: adjust debug/gosym to deal with instantiated types/funcs/methods
This changes debug/gosym so it can deal with instantiated
types/funcs/methods. I also added tests for instantiated names. My
assumption is that the concatenation of PackageName, ReceiverName, and
BaseName in order should cover the entire symbol name, so either the
ReceiverName or the BaseName should include any bracketed information
(either the instantiation of the receiver of a method or the
instantiation of function).
This can provide a model for how to parse instantiated functions and
method names.
Fixes #48032
Change-Id: I476781de2d6fc096efbb4be85c197d6f1cafac21
Reviewed-on: https://go-review.googlesource.com/c/go/+/354689
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Bryan C. Mills [Fri, 1 Oct 2021 15:52:56 +0000 (11:52 -0400)]
cmd/go: insert goroot to the hash of build cache when the packages include C files
This reverts CL 351851, which itself reverted CL 348991.
The problem with the original CL, as far as I can tell, was due to a
bug in the Go project's builder infrastructure (#33598) and not the
change itself. Once the build infrastructure is fixed, this change
can be resubmitted.
Fixes #48319
Updates #33598
Change-Id: I0fdbcc241eb2bdeb350944aad58bf58774fb591e
Reviewed-on: https://go-review.googlesource.com/c/go/+/353352
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>
Dan Scales [Mon, 11 Oct 2021 16:15:02 +0000 (09:15 -0700)]
cmd/compile: deal with TODO related to generic wrappers with embedded fields
It turns out there is no real TODO here - things are working fine. If we
are generating a wrapper for a method on a generic type that is actually
the method on embedded type, then we should just just generate the
normal embedded wrapper, which calls the wrapper for the real receiver
type on that method. There is no need to do the generic path where we
add in the dictionary argument. So, just updated that TODO comment with
this explanation.
Added a new test case embedded.go, which specifically tests various
situations involving converting to empty and non-empty interfaces.
issue44688.go already tests a bunch of these situations as well.
Also made some other cleanups in reflect.go:
- The shape test (that I had added) at the top of imethods is useless
(never true), since it is always an interface type, so removed it.
- Added usual helper function deref() to make code clearer in several
places.
- The shape test in methodWrapper() doesn't have to check HasShape() on
each targ - it can just check HasShape() on the whole receiver.
- The comment about disabling the tail call optimization for RegABI is
no longer true.
- Simplified code in several places by using the value of existing
variable 'methodrcvr'.
Change-Id: I8b1a5dc518dad37585824a1f73ceebb7627a9f82
Reviewed-on: https://go-review.googlesource.com/c/go/+/355129
Trust: Dan Scales <danscales@google.com> Reviewed-by: Keith Randall <khr@golang.org>
Robert Findley [Mon, 11 Oct 2021 16:22:52 +0000 (12:22 -0400)]
all: update vendored golang.org/x/tools
Now that x/tools/go/types/objectpath has been updates to support type
parameters, I ran the following commands to update x/tools inside the
cmd module:
Change-Id: Ibebc1ac2f721d6e2eb1a4f6c610918770d4879ed
Reviewed-on: https://go-review.googlesource.com/c/go/+/355012
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Paul E. Murphy [Thu, 26 Aug 2021 22:14:12 +0000 (17:14 -0500)]
cmd/internal/obj/ppc64: fix mtocrf, cleanup other CR ops
Fix "MOVW CRx, Rx" and "MOVFL Rx, constant", The FXM field was not
encoded correctly.
Generate mtocrf instead of mtcrf when a CRx argument is used. This
form is much faster.
Simplify several conditional statements which test if the register
argument is REG_CR or one of REG_CRx if the tested argument is known
to be matched as C_CREG. Likewise, a4 is (the From3 arg) is always
TYPE_NONE in the existing optab entries for type_ 69.
Cherry Mui [Fri, 8 Oct 2021 23:55:17 +0000 (19:55 -0400)]
cmd/link: do not mark holes in functab
With multiple text sections, there may be holes (non-Go code) in
the PC range of Go code and covered by the functab. Previously, we
use a linear search with actual PCs to find the functab entry. We
need to use special entries to mark holes, so a PC in the hole can
be distinguished from the previous function.
Now, with the previous CL we find if the PC is in between of the
sections upfront in textOff. There is no need to mark holes in the
functab.
Cherry Mui [Fri, 8 Oct 2021 23:24:53 +0000 (19:24 -0400)]
runtime: simplify multiple text section handling in findfunc
In findfunc, we first us the relative PC to find the function's
index in functab. When we split text sections, as the external
linker may shift the sections, and the PC may not match the
(virtual) PC we used to build the functab. So the index may be
inaccurate, and we need to do a (forward or backward) linear
search to find the actual entry.
Instead of using the PC directly, we can first compute the
(pre-external-link virtual) relative PC and use that to find the
index in functab. This way, the index will be accurate and we will
not need to do the special backward linear search.
Change-Id: I8ab11c66b7a5a3d79aae00198b98780e10db27b0
Reviewed-on: https://go-review.googlesource.com/c/go/+/354873
Trust: Cherry Mui <cherryyz@google.com>
Trust: Josh Bleecher Snyder <josharian@gmail.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
Russ Cox [Thu, 7 Oct 2021 13:56:29 +0000 (09:56 -0400)]
regexp: document and implement that invalid UTF-8 bytes are the same as U+FFFD
What should it mean to run a regexp match on invalid UTF-8 bytes?
The coherent behavior options are:
1. Invalid UTF-8 does not match any character classes,
nor a U+FFFD literal (nor \x{fffd}).
2. Each byte of invalid UTF-8 is treated identically to a U+FFFD in the input,
as a utf8.DecodeRune loop might.
RE2 uses Rule 1.
Because it works byte at a time, it can also provide \C to match any
single byte of input, which matches invalid UTF-8 as well.
This provides the nice property that a match for a regexp without \C
is guaranteed to be valid UTF-8.
Unfortunately, today Go has an incoherent mix of these two, although
mostly Rule 2. This is a deviation from RE2, and it gives up the nice
property, but we probably can't correct that at this point.
In particular .* already matches entire inputs today, valid UTF-8 or
not, and I doubt we can break that.
This CL adopts Rule 2 officially, fixing the few places that deviate from it.
Fixes #48749.
Change-Id: I96402527c5dfb1146212f568ffa09dde91d71244
Reviewed-on: https://go-review.googlesource.com/c/go/+/354569
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Rob Pike <r@golang.org>
Robert Findley [Fri, 8 Oct 2021 15:14:35 +0000 (11:14 -0400)]
go/parser: clean up unnecessary arguments and replace an if statement
Eliminate an unnecessary argument from parseGenericType, and replace an
if statement with a switch.
Change-Id: Iaa8afeface929332579f183c8e523961cca9aca4
Reviewed-on: https://go-review.googlesource.com/c/go/+/354869
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Alejandro García Montoro [Fri, 14 May 2021 16:42:16 +0000 (18:42 +0200)]
cmd/compile: eliminate successive swaps
The code generated when storing eight bytes loaded from memory in big
endian introduced two successive byte swaps that did not actually
modified the data.
The new rules match this specific pattern both for amd64 and for arm64,
eliminating the double swap.
Fixes #41684
Change-Id: Icb6dc20b68e4393cef4fe6a07b33aba0d18c3ff3
Reviewed-on: https://go-review.googlesource.com/c/go/+/320073 Reviewed-by: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Daniel Martí <mvdan@mvdan.cc>
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Matthew Dempsky [Fri, 8 Oct 2021 19:19:57 +0000 (12:19 -0700)]
cmd/internal/obj: rename MOVBE{LL,QQ,WW} to just MOVBE{L,Q,W}
The double suffix doesn't seem to serve any purpose, and we can keep
the old spelling as a backwards compatible alias in cmd/asm.
Change-Id: I3f01fc7249fb093ac1b25bd75c1cb9f39b8f62a9
Reviewed-on: https://go-review.googlesource.com/c/go/+/354700
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>
Matthew Dempsky [Fri, 8 Oct 2021 18:23:38 +0000 (11:23 -0700)]
cmd/cgo: use "!compiler_bootstrap" tag instead of "go1.18"
The illumos builder is using a Go 1.18 prerelease toolchain for
bootstrapping, which doesn't have ast.IndexListExpr. But we can
instead check for the "compiler_bootstrap" build tag that's already
used for distinguishing binaries built for toolchain1.
Fixes #48863.
Change-Id: I87b672322eee22a60ab8d0bb3be5f76cffc97545
Reviewed-on: https://go-review.googlesource.com/c/go/+/354695
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Cherry Mui [Fri, 8 Oct 2021 17:10:08 +0000 (13:10 -0400)]
cmd/link: put gcprog symbols along with gcbits
A stack object record may refer to the object's type's GC mask or
GC program (for very large types). For the latter, currently the GC
program symbol is named "type..gcprog.XXX" which is then laid out
along with type symbols at link time. When relro is used, the type
symbols end up in a different section.
As we now use relative addressing for stack object records to refer
to GC masks or GC programs, it is important that it is laid out in
the rodata section (not rodata.rel.ro). Move GC program symbols to
be along with GC masks, as they are similar and accessed the same
way. They don't have relocations so they don't need to be laid to a
relro section.
Keith Randall [Thu, 7 Oct 2021 20:29:09 +0000 (13:29 -0700)]
runtime: fix uint64->float32 conversion for softfloat
The fix for #48807 in CL 354429 forgot that we also need to fix
the softfloat implementation.
Update #48807
Change-Id: I596fb4e14e78145d1ad43c130b2cc5122b73655c
Reviewed-on: https://go-review.googlesource.com/c/go/+/354613
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>
Robert Griesemer [Fri, 8 Oct 2021 16:49:22 +0000 (09:49 -0700)]
cmd/compile/internal/types2: use an identifier map rather than isubst for recv type params
This is a port of CL 354643 from go/types to types2 with adjustments:
- use of syntax rather than go/ast package as needed
- adjustments due to the different code for type parameter declarations
- rename of Checker.rparamMap to Checker.recvTParamMap, which seems clearer
Change-Id: I5311a0c05a13c6b87ea1422b250b90c3d05c5dce
Reviewed-on: https://go-review.googlesource.com/c/go/+/354693
Trust: Robert Griesemer <gri@golang.org> Reviewed-by: Robert Findley <rfindley@google.com>
Dan Scales [Wed, 22 Sep 2021 17:05:33 +0000 (10:05 -0700)]
cmd/compile: allow delaying of transformCompLit, new transformAddr
For this unusual case, where a constraint specifies exactly one type, we
can have a COMPLIT expression with a type that is/has typeparams.
Therefore, we add code to delay transformCompLit for generic functions.
We also need to break out transformAddr (which corresponds to tcAddr),
and added code for delaying it as well. Also, we now need to export
generic functions containing untransformed OCOMPLIT and OKEY nodes, so
added support for that in iexport.go/iimport.go. Untransformed OKEY
nodes include an ir.Ident/ONONAME which we can now export.
Had to adjust some code/asserts in transformCompLit(), since we may now
be transforming an OCOMPLIT from an imported generic function (i.e. from
a non-local package).
Fixes #48537
Change-Id: I09e1b3bd08b4e013c0b098b8a25d082efa1fef51
Reviewed-on: https://go-review.googlesource.com/c/go/+/354354
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Dan Scales <danscales@google.com> Reviewed-by: Keith Randall <khr@golang.org>
Robert Findley [Fri, 8 Oct 2021 14:37:54 +0000 (10:37 -0400)]
go/ast, go/types: remove some stale TODOs
We've decided to leave TParams on FuncType, and type list syntax is
no longer part of the proposal.
Change-Id: Id34f6495a358d76994df331384a4b93487275c4d
Reviewed-on: https://go-review.googlesource.com/c/go/+/354751
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Change-Id: I0ed80dad99d6920a187af4339e437a08fbe91ade
Reviewed-on: https://go-review.googlesource.com/c/go/+/354572
Trust: Bryan C. Mills <bcmills@google.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Robert Findley [Thu, 7 Oct 2021 23:12:16 +0000 (19:12 -0400)]
go/types: use an identifier map rather than isubst for recv type params
Receiver type parameters are the only type expressions allowed to be
blank. Previously this was handled by substitution of synthetic
non-blank names in the receiver type expression, but that introduced
problems related to AST mangling: the scope had extra elements,
Object.Name() was inaccurate, and synthetic nodes were recorded in
types.Info.
Fix this instead by recording a map of *ast.Ident->*TypeParam on the
Checker, which is read in Checker.ident to resolve blank identifiers
denoting receiver type parameters.
Change-Id: I6a7a86b823409f54778c0f141e8bd269a2cc85d9
Reviewed-on: https://go-review.googlesource.com/c/go/+/354643
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Nigel Tao [Thu, 7 Oct 2021 23:33:47 +0000 (10:33 +1100)]
image/png: fix interlaced palette out-of-bounds
PNG images can be paletted, where each pixel value (a uint8) indexes a
slice of colors. In terms of wire format, the PLTE chunk explicitly
contains the palette length. However, in practice, some arguably
malformed images contain pixel values greater than or equal to the
explicit PLTE length.
Go's image/png decoder accomodates such images by lengthening the
decoded image's palette if the implicit maximum is larger than the
explicit maximum. This was already done, prior to this commit, by the
"if len(paletted.Palette) <= int(idx)" lines in decoder.readImagePass.
Separately, PNG images can also be interlaced, where the final image is
the result of merging multiple partial images, also called passes. Prior
to this commit, we applied the palette lengthening to the pass images
but not the final image. This commit fixes that.
Fixes #48612
Change-Id: I77606538cc9a504fbd726071756ebcd10c9da73f
Reviewed-on: https://go-review.googlesource.com/c/go/+/354709
Trust: Nigel Tao <nigeltao@golang.org> Reviewed-by: Andrew Gerrand <adg@golang.org>