Change-Id: Ic507cb740395e76d1d011a5a2f395b96c3d172a2
Reviewed-on: https://go-review.googlesource.com/105915Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

A test for CL 102036.

Change-Id: Ief6dcb4f478670813fbe22ea75a06815a4b201a3
Reviewed-on: https://go-review.googlesource.com/105875
Run-TryBot: Keith Randall <khr@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Insert appropriate race/msan calls before each memory operation during
SSA construction.

This is conceptually simple, but subtle because we need to be careful
that inserted instrumentation calls don't clobber arguments that are
currently being prepared for a user function call.

reorder1 already handles introducing temporary variables for arguments
in some cases. This CL changes it to use them for all arguments when
instrumenting.

Also, we can't SSA struct types with more than one field while
instrumenting. Otherwise, concurrent uses of disjoint fields within an
SSA-able struct can introduce false races.

This is both somewhat better and somewhat worse than the old racewalk
instrumentation pass. We're now able to easily recognize cases like
constructing non-escaping closures on the stack or accessing closure
variables don't need instrumentation calls. On the other hand,
spilling escaping parameters to the heap now results in an
instrumentation call.

Overall, this CL results in a small net reduction in the number of
instrumentation calls, but a small net increase in binary size for
instrumented executables. cmd/go ends up with 5.6% fewer calls, but a
2.4% larger binary.

Fixes #19054.

Change-Id: I70d1dd32ad6340e6fdb691e6d5a01452f58e97f3
Reviewed-on: https://go-review.googlesource.com/102817Reviewed-by: Cherry Zhang <cherryyz@google.com>

The patch adds support for LDR(register offset) instruction.
And add the test cases and negative tests.

Change-Id: I5b32c6a5065afc4571116d4896f7ebec3c0416d3
Reviewed-on: https://go-review.googlesource.com/87955Reviewed-by: Cherry Zhang <cherryyz@google.com>

ssa's pos parameter on the Const* funcs is unused, so remove it.

ld's alloc parameter on elfnote is always true, so remove the arguments
and simplify the code.

Finally, arm's addpltreloc never has its return parameter used, so
remove it.

Change-Id: I63387ecf6ab7b5f7c20df36be823322bb98427b8
Reviewed-on: https://go-review.googlesource.com/104456
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>

And delete them from asm_test.

Change-Id: I0e33d58274951ab5acb67b0117b60ef617ea887a
Reviewed-on: https://go-review.googlesource.com/105735
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Daniel Martí <mvdan@mvdan.cc>

Change-Id: Ibb793158a190758cebc00b58c12e3b8b106c0548
GitHub-Last-Rev: 6a6d949eb252ec6d3e76920fc2f50ef5ac3cdb5a
GitHub-Pull-Request: golang/go#24702
Reviewed-on: https://go-review.googlesource.com/104896Reviewed-by: Daniel Martí <mvdan@mvdan.cc>
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Fixes #24753

Change-Id: Ic6049da98b8eee41223df71ffafa71a894915bd7
Reviewed-on: https://go-review.googlesource.com/105455Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

NumComponents is used by racewalk to decide whether reads and writes
might occur to subobjects of an address. For that purpose,
blank fields matter.

It is also used to decide whether to inline == and != for a type.
For that purpose, blank fields may be ignored.

Add a parameter to NumComponents to support this distinction.
While we're here, document NumComponents, as requested in CL 59334.

Change-Id: I8c2021b172edadd6184848a32a74774dde1805c8
Reviewed-on: https://go-review.googlesource.com/103755
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Found with megacheck (S1007).

Change-Id: Icb15fd5bfefa8e0b39a1bfa9ec3e9af3eff6b390
Reviewed-on: https://go-review.googlesource.com/105415
Run-TryBot: Iskander Sharipov <iskander.sharipov@intel.com>
Reviewed-by: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Finish porting arm64 slice zeroing codegen tests; delete them from
asm_test.

Change-Id: Id2532df8ba1c340fa662a6b5238daa3de30548be
Reviewed-on: https://go-review.googlesource.com/105136
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Giovanni Bajo <rasky@develer.com>

This ensures that gdb tests run on Windows by ignoring any line ending.

Works with gdb 7.7, however with gdb 7.9 and 7.12 gets an error
error:

    internal-error: buildsym_init: Assertion `free_pendings == NULL'
    failed.

Updates #21380

Change-Id: I6a6e5b2a1b5efdca4dfce009fcb9c134c87497d6
Reviewed-on: https://go-review.googlesource.com/102419
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>

In particular, to pull a few fixes that were causing some tests to be
flaky on our build dashboard.

Fixes #24405.
Fixes #24508.
Fixes #24611.

Change-Id: I713156ad11c924e4a4b603144d10395523d526ed
Reviewed-on: https://go-review.googlesource.com/105275
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This change replaces the vendored socks client implementation with the
bundle of golang.org/x/net/internal/socks package which contains fixes
for 19354 and 11682.

golang.org/x/net/internal/socks becomes socks_bundle.go.

At git rev 61147c4. (golang.org/cl/38278)

Updates #11682.
Updates #18508.
Updates #19354.
Fixes #19688.
Updates #21333.

Change-Id: I8cf6c3f5eb87c24685a7592be015729f84fbed77
Reviewed-on: https://go-review.googlesource.com/41031
Run-TryBot: Mikio Hara <mikioh.mikioh@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This provides a way to enforce pure Go implementation of os/user
lookup functions on UNIX platforms by means of "osusergo" build tag,
in a manner similar to netgo/netcgo tags in the net package.

If "osusergo" build tag is set, Go implementation is selected.

If "osusergo" build tag is NOT set, the old behavior is retained,
that is to use cgo (libc-backed) implementation if both cgo and such
and such implementation are available.

The reason behind this change is to make it possible to build proper
static binaries on Linux. The problem is, glibc implementation of
getpw*, getgrp* and getgrouplist functions relies on presense of
libnss*.so libraries during runtime, making it impossible to build
a self-contained static binary which uses both cgo and os/user.
In such case, linker warnings like this are shown:

> warning: Using 'getgrouplist' in statically linked applications
> requires at runtime the shared libraries from the glibc version
> used for linking

While this can be solved by recompiling glibc with --enable-static-nss
flag or using a different libc implementation (like musl on Alpine Linux),
it is not always practical or even possible.

Fixes #23265

Change-Id: I383a448a2ecf15493ec93dbd5d076b6330cb14cb
Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
Reviewed-on: https://go-review.googlesource.com/92456
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Found using go-simple.

Change-Id: I349e80a8c083688539bb4267564f02e3d7913da5
Reviewed-on: https://go-review.googlesource.com/105195Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Passes toolstash-check.

Change-Id: Icae55fe4fa1bb8e4f2f83b7c69e08d30a5559d9e
Reviewed-on: https://go-review.googlesource.com/105047
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>

There are three places where we wait for the GC mark phase to
complete. Factor these all into a single helper function.

Fixes #24362.

Change-Id: I47f6a7147974f5b9a2869c527a024519070ba6f3
Reviewed-on: https://go-review.googlesource.com/102605
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rick Hudson <rlh@golang.org>

In particular, avoid exponential memory usage from growing it in
proportion to its current size.

Fixes #24731

Change-Id: I277d2fbac2ef7b00ae4b83d6d1dcd7f2e630a5cd
Reviewed-on: https://go-review.googlesource.com/105155Reviewed-by: Daniel Martí <mvdan@mvdan.cc>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Updates #24362.

Change-Id: Ided1ab31792f05d9d7a86f17c1bcbd9e9b80052c
Reviewed-on: https://go-review.googlesource.com/102606
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

Replace int variables with 0/1 as only possible values with bools,
where possible.

Change-Id: I958c082e703bbc1540309da3e17612fc8e247932
Reviewed-on: https://go-review.googlesource.com/105036
Run-TryBot: Ilya Tocar <ilya.tocar@intel.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

In gdb, "b f" gets confused if the first instruction of "f"
is not marked as a statement in the DWARF line table.

To ensure gdb is not confused, move the first statement
marker in "f" to its first instruction.

The screwy-looking conditional for "what's the first
instruction with a statement marker" will become simpler in
the future.

Fixes #24695.

Change-Id: I2eef81676b64d1bd9bff5da03b89b9dc0c18f44f
Reviewed-on: https://go-review.googlesource.com/104955
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>

Passes toolstash-check.

Change-Id: I804e73447b6fdbb75af6235c193c4ee7cbcf8d3a
Reviewed-on: https://go-review.googlesource.com/105045
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Method expressions with anonymous receiver types like "struct { T }.m"
require wrapper functions, which we weren't always creating. This in
turn resulted in linker errors.

This CL ensures that we generate wrapper functions for any anonymous
receiver types used in a method expression.

Fixes #22444.

Change-Id: Ia8ac27f238c2898965e57b82a91d959792d2ddd4
Reviewed-on: https://go-review.googlesource.com/105044
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

FreeBSD/Dragonfly and Solaris have identical implementations of
sendFile. Keep one and adjust the comments accordingly.

Change-Id: I77b0f88a4816dd6e40f5cb33919c44606401ac6b
Reviewed-on: https://go-review.googlesource.com/104915
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Although these changes have no essential influence, I think this is a better point.

Change-Id: I571d3a14c948d2fd7bc9561f47f33e9e4c90683f
GitHub-Last-Rev: d8c5d180069bb91b6eb1247769bd14f728922fc0
GitHub-Pull-Request: golang/go#24697
Reviewed-on: https://go-review.googlesource.com/104895Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

ParseARM64Suffix is not used outside cmd/asm/internal/arch.

Change-Id: I8e7782dce11cf8cd2fd08dd17e555ced8d87ba24
Reviewed-on: https://go-review.googlesource.com/105115
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Daniel Martí <mvdan@mvdan.cc>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

* Remove some redundant returns
* Replace HasPrefix with TrimPrefix
* Remove some obviously dead code

Passes toolstash -cmp on std cmd.

Change-Id: Ifb0d70a45cbb8a8553758a8c4878598b7fe932bc
Reviewed-on: https://go-review.googlesource.com/105017
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Change-Id: If498d1fc6e8c0c4e8cf7ed38c4997adf05e003a6
Reviewed-on: https://go-review.googlesource.com/105043
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Dave Cheney <dave@cheney.net>

These rules trigger 190 times during make.bash.

Change-Id: I20d1688db5d8c904a7237c08635c6c9d8bd58b1c
Reviewed-on: https://go-review.googlesource.com/105037
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Giovanni Bajo <rasky@develer.com>

Updates #22830

Due to not checking if the output slices alias in divLarge,
calls of the form z.div(z, x, y) caused the slice z
to attempt to be used to store both the quotient and the
remainder of the division.  CL 78995 applies an alias
check to correct that error.  This CL cleans up the
additional div calls that attempt to supply the same slice
to hold both the quotient and remainder.

Note that the call in expNN was responsible for the reported
error in r.Exp(x, 1, m) when r was initialized to a non-zero value.

The second instance in expNNMontgomery did not result in an error
due to the size of the arguments.

	// RR = 2**(2*_W*len(m)) mod m
	RR := nat(nil).setWord(1)
	zz := nat(nil).shl(RR, uint(2*numWords*_W))
	_, RR = RR.div(RR, zz, m)

Specifically,

cap(RR) == 5 after setWord(1) due to const e = 4 in z.make(1)
len(zz) == 2*len(m) + 1 after shifting left, numWords = len(m)

Reusing the backing array for z and z2 in div was only triggered if
cap(RR) >= len(zz) + 1 and len(m) > 1 so that divLarge was called.

But, 5 < 2*len(m) + 2 if len(m) > 1, so new arrays were allocated
and the error was never triggered in this case.

Change-Id: Iedac80dbbde13216c94659e84d28f6f4be3aaf24
Reviewed-on: https://go-review.googlesource.com/81055
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>

There were multiple ad hoc ways to create method symbols, with subtle
and confusing differences between them. This CL unifies them into a
single well-documented encoding and implementation.

This introduces some inconsequential changes to symbol format for the
sake of simplicity and consistency. Two notable changes:

1) Symbol construction is now insensitive to the package currently
being compiled. Previously, non-exported methods on anonymous types
received different method symbols depending on whether the method was
local or imported.

2) Symbols for method values parenthesized non-pointer receiver types
and non-exported method names, and also always package-qualified
non-exported method names. Now they use the same rules as normal
method symbols.

The methodSym function is also now stricter about rejecting
non-sensical method/receiver combinations. Notably, this means that
typecheckfunc needs to call addmethod to validate the method before
calling declare, which also means we no longer emit errors about
redeclaring bogus methods.

Change-Id: I9501c7a53dd70ef60e5c74603974e5ecc06e2003
Reviewed-on: https://go-review.googlesource.com/104876Reviewed-by: Robert Griesemer <gri@golang.org>

adjustpointers loops over a bitmap.
If the length of that bitmap is zero,
we can skip making the call entirely.
This speeds up stack copying when there are
no pointers present in either args or locals.

name                old time/op  new time/op  delta
StackCopyPtr-8       101ms ± 4%    90ms ± 4%  -10.95%  (p=0.000 n=87+93)
StackCopy-8         80.1ms ± 4%  72.6ms ± 4%   -9.41%  (p=0.000 n=98+100)
StackCopyNoCache-8   121ms ± 3%   113ms ± 3%   -6.57%  (p=0.000 n=98+97)

Change-Id: I7a272e19bc9a14fa3e3318771ebd082dc6247d25
Reviewed-on: https://go-review.googlesource.com/104737
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>

This commit does not change the semantics of the Call method. Its
purpose is to avoid duplication of code by making PrepareCall available
for separate use by the wasm backend.

Updates #18892

Change-Id: I04a3098f56ebf0d995791c5375dd4c03b6a202a3
Reviewed-on: https://go-review.googlesource.com/103275Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

Change-Id: Id4e3407ba497a018d5ace92813ba8e9653d0ac7d
Reviewed-on: https://go-review.googlesource.com/104976Reviewed-by: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>

This change improves performance of addVV, subVV and mulAddVWW
by unrolling the loops, with improvements up to 1.45x.

benchmark                    old ns/op     new ns/op     delta
BenchmarkAddVV/1-16          5.79          5.85          +1.04%
BenchmarkAddVV/2-16          6.41          6.62          +3.28%
BenchmarkAddVV/3-16          6.89          7.35          +6.68%
BenchmarkAddVV/4-16          7.47          8.26          +10.58%
BenchmarkAddVV/5-16          8.04          8.18          +1.74%
BenchmarkAddVV/10-16         10.9          11.2          +2.75%
BenchmarkAddVV/100-16        81.7          57.0          -30.23%
BenchmarkAddVV/1000-16       714           500           -29.97%
BenchmarkAddVV/10000-16      7088          4946          -30.22%
BenchmarkAddVV/100000-16     71514         49364         -30.97%
BenchmarkSubVV/1-16          5.94          5.89          -0.84%
BenchmarkSubVV/2-16          12.9          6.82          -47.13%
BenchmarkSubVV/3-16          7.03          7.34          +4.41%
BenchmarkSubVV/4-16          7.58          8.23          +8.58%
BenchmarkSubVV/5-16          8.15          8.19          +0.49%
BenchmarkSubVV/10-16         11.2          11.4          +1.79%
BenchmarkSubVV/100-16        82.4          57.0          -30.83%
BenchmarkSubVV/1000-16       715           499           -30.21%
BenchmarkSubVV/10000-16      7089          4947          -30.22%
BenchmarkSubVV/100000-16     71568         49378         -31.01%

benchmark                    old MB/s     new MB/s      speedup
BenchmarkAddVV/1-16          11048.49     10939.92      0.99x
BenchmarkAddVV/2-16          19973.41     19323.60      0.97x
BenchmarkAddVV/3-16          27847.09     26123.06      0.94x
BenchmarkAddVV/4-16          34276.46     30976.54      0.90x
BenchmarkAddVV/5-16          39781.92     39140.68      0.98x
BenchmarkAddVV/10-16         58559.29     56894.68      0.97x
BenchmarkAddVV/100-16        78354.88     112243.69     1.43x
BenchmarkAddVV/1000-16       89592.74     127889.04     1.43x
BenchmarkAddVV/10000-16      90292.39     129387.06     1.43x
BenchmarkAddVV/100000-16     89492.92     129647.78     1.45x
BenchmarkSubVV/1-16          10781.03     10861.22      1.01x
BenchmarkSubVV/2-16          9949.27      18760.21      1.89x
BenchmarkSubVV/3-16          27319.40     26166.01      0.96x
BenchmarkSubVV/4-16          33764.35     31123.02      0.92x
BenchmarkSubVV/5-16          39272.40     39050.31      0.99x
BenchmarkSubVV/10-16         57262.87     56206.33      0.98x
BenchmarkSubVV/100-16        77641.78     112280.86     1.45x
BenchmarkSubVV/1000-16       89486.27     128064.08     1.43x
BenchmarkSubVV/10000-16      90274.37     129356.59     1.43x
BenchmarkSubVV/100000-16     89424.42     129610.50     1.45x

Change-Id: I2795a82134d1e3b75e2634c76b8ca165a723ec7b
Reviewed-on: https://go-review.googlesource.com/103495
Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>

OpenBSD's __threxit syscall takes a pointer to a 32-bit value that will be
zeroed immediately before the thread exits. Make use of this instead of
zeroing freeWait from the exitThread assembly and using hacks like switching
to a static stack, so this works on 386.

Change-Id: I3ec5ead82b6496404834d148f713794d5d9da723
Reviewed-on: https://go-review.googlesource.com/105055Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Don't generate FP ops with 1 operand in memory for 387.

Change-Id: I23b49dfa2a1e60c8778c920230e64785a3ddfbd1
Reviewed-on: https://go-review.googlesource.com/105035
Run-TryBot: Ilya Tocar <ilya.tocar@intel.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Originally, scalar values were directly stored within interface values
as long as they fit into a pointer-sized slot of memory. And since
interface method calls always pass the full pointer-sized value as the
receiver argument, value-narrowing wrappers were necessary to adapt to
the calling convention for methods with smaller receiver types.

However, for precise garbage collection, we now only store actual
pointers within interface values, so these wrappers are no longer
necessary.

Passes toolstash-check.

Change-Id: I5303bfeb8d0f11db619b5a5d06b37ac898588670
Reviewed-on: https://go-review.googlesource.com/104875
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Austin Clements <austin@google.com>

Mostly replacing C-Style loops with range expressions, but also other
simplifications like the introduction of writeBool and unindenting some
code.

Passes toolstash -cmp on std cmd.

Change-Id: I799bccd4e5d411428dcf122b8588a564a9217e7c
Reviewed-on: https://go-review.googlesource.com/104936
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Marvin Stenger <marvin.stenger94@gmail.com>
Reviewed-by: Ilya Tocar <ilya.tocar@intel.com>