By replacing bytes-reversing routines with bits.ReverseBytes64 calls.

name                     old time/op  new time/op  delta
EncodeComplex128Slice-4  35.1µs ± 1%  23.2µs ± 2%  -33.94%  (p=0.000 n=20+20)
EncodeFloat64Slice-4     17.9µs ± 1%  11.0µs ± 1%  -38.36%  (p=0.000 n=17+18)

name                     old time/op  new time/op  delta
DecodeComplex128Slice-4  79.7µs ± 0%  69.9µs ± 1%  -12.31%  (p=0.000 n=20+20)
DecodeFloat64Slice-4     47.3µs ± 1%  42.2µs ± 1%  -10.65%  (p=0.000 n=17+17)

Change-Id: I91a6401c6009b5712fca6258dd1e57c6fe68ea64
Reviewed-on: https://go-review.googlesource.com/38352
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Change-Id: Iae41e14ee55eb4068fcb2189a77b345a7c5468b4
Reviewed-on: https://go-review.googlesource.com/38333
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Fixes #19592.

Change-Id: I8946b33fd36ae1f39bdcc4bf0bd4b5b99618efe8
Reviewed-on: https://go-review.googlesource.com/38300Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>

While we're here, also eliminate a few more Curfn uses.

Passes toolstash -cmp. No compiler performance impact.

Updates #15756

Change-Id: Ib8db9e23467bbaf16cc44bf62d604910f733d6b8
Reviewed-on: https://go-review.googlesource.com/38331
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Passes toolstash -cmp. No compiler performance impact.

Updates #15756

Change-Id: I85b45244453ae28d4da76be4313badddcbf3f5dc
Reviewed-on: https://go-review.googlesource.com/38330
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Passes toolstash -cmp. No compiler performance impact.

Updates #15756

Change-Id: I1294058716d83dd1be495d399ed7ab2277754dc6
Reviewed-on: https://go-review.googlesource.com/38329
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Preparation for de-globalizing Stksize and MaxArg.

Passes toolstash -cmp. No compiler performance impact.

Updates #15756

Change-Id: I312f0bbd15587a6aebf472cd66c8e62b89e55c8a
Reviewed-on: https://go-review.googlesource.com/38328
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

This reverts commit b744a11a.

Reason for revert: Broke trybots. (misc-vetall builder is busted)

Change-Id: I651d1c18db2fb3cb6ec12c2ae62024627baf8d77
Reviewed-on: https://go-review.googlesource.com/38332Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This is a first step towards eliminating the
Curfn global in the backend.
There's more to do.

Passes toolstash -cmp. No compiler performance impact.

Updates #15756

Change-Id: Ib09f550a001e279a5aeeed0f85698290f890939c
Reviewed-on: https://go-review.googlesource.com/38232
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

This makes fewer Values around until decomposing, reducing
allocation in compiler.

name       old alloc/op    new alloc/op    delta
Template      41.4MB ± 0%     40.8MB ± 0%  -1.29%  (p=0.000 n=10+10)
Unicode       30.3MB ± 0%     30.2MB ± 0%  -0.24%  (p=0.000 n=10+10)
GoTypes        118MB ± 0%      115MB ± 0%  -2.23%  (p=0.000 n=10+10)
Compiler       505MB ± 0%      493MB ± 0%  -2.47%  (p=0.000 n=10+10)
SSA            881MB ± 0%      872MB ± 0%  -1.03%  (p=0.000 n=10+10)

name       old allocs/op   new allocs/op   delta
Template        401k ± 1%       400k ± 1%    ~     (p=0.631 n=10+10)
Unicode         321k ± 0%       321k ± 1%    ~     (p=0.684 n=10+10)
GoTypes        1.18M ± 0%      1.17M ± 0%  -0.34%  (p=0.000 n=10+10)
Compiler       4.63M ± 0%      4.61M ± 0%  -0.43%  (p=0.000 n=10+10)
SSA            7.83M ± 0%      7.82M ± 0%  -0.13%  (p=0.000 n=10+10)

Change-Id: I8f736396294444248a439bd4c90be1357024ce88
Reviewed-on: https://go-review.googlesource.com/38294
Run-TryBot: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

Suggested by mdempsky in CL 38232.
This allows us to use the Frontend field
to associate frontend state and information
with a function.
See the following CL in the series for examples.

This is a giant CL, but it is almost entirely routine refactoring.

The ssa test API is starting to feel a bit unwieldy.
I will clean it up separately, once the dust has settled.

Passes toolstash -cmp.

Updates #15756

Change-Id: I71c573bd96ff7251935fce1391b06b1f133c3caf
Reviewed-on: https://go-review.googlesource.com/38327
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Prior to this CL, config was an explicit argument
to the SSA rewrite rules, and rules that needed
a Frontend got at it via config.
An upcoming CL moves Frontend from Config to Func,
so rules can no longer reach Frontend via Config.
Passing a Frontend as an argument to the rewrite rules
causes a 2-3% regression in compile times.
This CL takes a different approach:
It treats the variable names "config" and "fe"
as special and calculates them as needed.
The "as needed part" is also important to performance:
If they are calculated eagerly, the nilchecks themselves
cause a regression.

This introduces a little bit of magic into the rewrite
generator. However, from the perspective of the rules,
the config variable was already more or less magic.
And it makes the upcoming changes much clearer.

Passes toolstash -cmp.

Change-Id: I173f2bcc124cba43d53138bfa3775e21316a9107
Reviewed-on: https://go-review.googlesource.com/38326
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Prepared using gorename.

Change-Id: Id55dac9ae5446a8bfeac06e7995b35f4c249eeca
Reviewed-on: https://go-review.googlesource.com/38302
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

This CL changes the GOARCH.Init functions to take gc.Thearch as a
parameter, which gc.Main supplies.

Additionally, the x86 backend is refactored to decide within Init
whether to use the 387 or SSE2 instruction generators, rather than for
each individual SSA Value/Block.

Passes toolstash-check -all.

Change-Id: Ie6305a6cd6f6ab4e89ecbb3cbbaf5ffd57057a24
Reviewed-on: https://go-review.googlesource.com/38301
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

This reduces memory use yet still provides the significant
performance gain seen when using a fast path for small integers.

Improvement of this CL comparing to code without fast path:

name              old time/op  new time/op  delta
FormatIntSmall-8  35.6ns ± 1%   4.5ns ± 1%  -87.30%  (p=0.008 n=5+5)
AppendIntSmall-8  17.4ns ± 1%   9.4ns ± 3%  -45.70%  (p=0.008 n=5+5)

For comparison, here's the improvement before this CL to code without
fast path (1% better for FormatIntSmall):

name              old time/op  new time/op  delta
FormatIntSmall-8  35.6ns ± 1%   4.0ns ± 3%  -88.64%  (p=0.008 n=5+5)
AppendIntSmall-8  17.4ns ± 1%   8.2ns ± 1%  -52.80%  (p=0.008 n=5+5)

Thus, the code in this CL performs slower for small integers using fast
path then the prior version, but this is relative to an already very fast
version:

name              old time/op  new time/op  delta
FormatIntSmall-8  4.05ns ± 3%  4.52ns ± 1%  +11.81%  (p=0.008 n=5+5)
AppendIntSmall-8  8.21ns ± 1%  9.45ns ± 3%  +15.05%  (p=0.008 n=5+5)

Measured on 2.3 GHz Intel Core i7 running macOS Sierra 10.12.3.

Overall, it's still ~88% faster than without fast path for small integers,
so probably worth it as it removes 100 global string slices in favor of
a single string.

Credits: This is based on the original (but cleaned up) version of the
code by Aliaksandr Valialkin (https://go-review.googlesource.com/c/37963/).

Change-Id: Icda78679c8c14666d46257894e9fa3d7f35e58b8
Reviewed-on: https://go-review.googlesource.com/38319Reviewed-by: Martin Möhrmann <moehrmann@google.com>

Its godoc says that path must not be empty or dot, while the existing
implementation happily accepts both.

Change-Id: I64766271c35152dc7adb21ff60eb05c52237e6b6
Reviewed-on: https://go-review.googlesource.com/38262Reviewed-by: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>

name                old time/op  new time/op  delta
EncodeInt32Slice-4  14.6µs ± 2%  12.2µs ± 1%  -16.65%  (p=0.000 n=19+18)

Change-Id: I078a171f1633ff81d7e3f981dc9a398309ecb2c0
Reviewed-on: https://go-review.googlesource.com/38269Reviewed-by: Rob Pike <r@golang.org>

I don't know that it exists for any other architectures.

Update #18616

Change-Id: Idfe5dee251764d32787915889ec0be4bebc5be24
Reviewed-on: https://go-review.googlesource.com/38323
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>

`go test -i -race` adds the "sync/atomic" package to every package dependency tree
that makes buildIDs different from packages installed with `go install -race`
and causes cache rebuilding.

Fixes #19133
Fixes #19151

Change-Id: I0536c6fa41b0d20fe361b5d35b3c0937b146d07d
Reviewed-on: https://go-review.googlesource.com/37598Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Sometimes it's necessary to deal with emails that do not follow the specification; in particular, it's possible to download such email via gmail.
When the existing implementation handle invalid mime media parameters, it returns nils and error, although there is a valid media type, which may be returned.
If this behavior changes, it may not affect any existing programs, but it will help to parse some emails.

Fixes #19498

Change-Id: Ieb2fdbddfd93857faee941d2aa49d59e286d57fd
Reviewed-on: https://go-review.googlesource.com/38190Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

This change improves the performance of crc32 for ppc64le by using
vpmsum and other vector instructions in the algorithm.

The testcase was updated to test more sizes.

Fixes #19570

BenchmarkCRC32/poly=IEEE/size=15/align=0-8             90.5          81.8          -9.61%
BenchmarkCRC32/poly=IEEE/size=15/align=1-8             89.7          81.7          -8.92%
BenchmarkCRC32/poly=IEEE/size=40/align=0-8             93.2          61.1          -34.44%
BenchmarkCRC32/poly=IEEE/size=40/align=1-8             92.8          60.9          -34.38%
BenchmarkCRC32/poly=IEEE/size=512/align=0-8            501           55.8          -88.86%
BenchmarkCRC32/poly=IEEE/size=512/align=1-8            502           132           -73.71%
BenchmarkCRC32/poly=IEEE/size=1kB/align=0-8            947           69.9          -92.62%
BenchmarkCRC32/poly=IEEE/size=1kB/align=1-8            946           144           -84.78%
BenchmarkCRC32/poly=IEEE/size=4kB/align=0-8            3602          186           -94.84%
BenchmarkCRC32/poly=IEEE/size=4kB/align=1-8            3603          263           -92.70%
BenchmarkCRC32/poly=IEEE/size=32kB/align=0-8           28404         1338          -95.29%
BenchmarkCRC32/poly=IEEE/size=32kB/align=1-8           28856         1405          -95.13%
BenchmarkCRC32/poly=Castagnoli/size=15/align=0-8       89.7          81.8          -8.81%
BenchmarkCRC32/poly=Castagnoli/size=15/align=1-8       89.8          81.9          -8.80%
BenchmarkCRC32/poly=Castagnoli/size=40/align=0-8       93.8          61.4          -34.54%
BenchmarkCRC32/poly=Castagnoli/size=40/align=1-8       94.3          61.3          -34.99%
BenchmarkCRC32/poly=Castagnoli/size=512/align=0-8      503           56.4          -88.79%
BenchmarkCRC32/poly=Castagnoli/size=512/align=1-8      502           132           -73.71%
BenchmarkCRC32/poly=Castagnoli/size=1kB/align=0-8      941           70.2          -92.54%
BenchmarkCRC32/poly=Castagnoli/size=1kB/align=1-8      943           145           -84.62%
BenchmarkCRC32/poly=Castagnoli/size=4kB/align=0-8      3588          186           -94.82%
BenchmarkCRC32/poly=Castagnoli/size=4kB/align=1-8      3595          264           -92.66%
BenchmarkCRC32/poly=Castagnoli/size=32kB/align=0-8     28266         1323          -95.32%
BenchmarkCRC32/poly=Castagnoli/size=32kB/align=1-8     28344         1404          -95.05%

Change-Id: Ic4d8274c66e0e87bfba5f609f508a3877aee6bb5
Reviewed-on: https://go-review.googlesource.com/38184Reviewed-by: David Chase <drchase@google.com>

Fixes #19553.

Change-Id: I414cb3b1c2dab20f41a7f4e7aba49c534ff19942
Reviewed-on: https://go-review.googlesource.com/38271Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This is a follow-up to CL 38167.
Pure code movement.

Change-Id: I13e58f7eac6718c77076d89e13fc721a5205ec57
Reviewed-on: https://go-review.googlesource.com/38322
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This makes ssa.Func, ssa.Cache, and ssa.Config fulfill
the roles laid out for them in CL 38160.

The only non-trivial change in this CL is how cached
values and blocks get IDs. Prior to this CL, their IDs were
assigned as part of resetting the cache, and only modified
IDs were reset. This required knowing how many values and
blocks were modified, which required a tight coupling between
ssa.Func and ssa.Config. To eliminate that coupling,
we now zero values and blocks during reset,
and assign their IDs when they are used.
Since unused values and blocks have ID == 0,
we can efficiently find the last used value/block,
to avoid zeroing everything.
Bulk zeroing is efficient, but not efficient enough
to obviate the need to avoid zeroing everything every time.
As a happy side-effect, ssa.Func.Free is no longer necessary.

DebugHashMatch and friends now belong in func.go.
They have been left in place for clarity and review.
I will move them in a subsequent CL.

Passes toolstash -cmp. No compiler performance impact.
No change in 'go test cmd/compile/internal/ssa' execution time.

Change-Id: I2eb7af58da067ef6a36e815a6f386cfe8634d098
Reviewed-on: https://go-review.googlesource.com/38167
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>

Minor cleanup only.

No reason to go through String() when it is
just as easy to do a direct string comparison.

Eliminates a surprising number of allocations.

name       old alloc/op    new alloc/op    delta
Template      40.9MB ± 0%     40.9MB ± 0%    ~     (p=0.190 n=10+10)
Unicode       30.3MB ± 0%     30.3MB ± 0%    ~     (p=0.218 n=10+10)
GoTypes        116MB ± 0%      116MB ± 0%  -0.09%  (p=0.000 n=10+10)
SSA            871MB ± 0%      869MB ± 0%  -0.14%  (p=0.000 n=10+9)
Flate         26.2MB ± 0%     26.2MB ± 0%  -0.15%  (p=0.002 n=10+10)
GoParser      32.5MB ± 0%     32.5MB ± 0%    ~     (p=0.165 n=10+10)
Reflect       80.5MB ± 0%     80.4MB ± 0%  -0.12%  (p=0.003 n=9+10)
Tar           27.3MB ± 0%     27.3MB ± 0%  -0.13%  (p=0.008 n=10+9)
XML           43.1MB ± 0%     43.1MB ± 0%    ~     (p=0.218 n=10+10)

name       old allocs/op   new allocs/op   delta
Template        402k ± 1%       400k ± 1%  -0.64%  (p=0.002 n=10+10)
Unicode         322k ± 1%       321k ± 1%    ~     (p=0.075 n=10+10)
GoTypes        1.19M ± 0%      1.18M ± 0%  -0.90%  (p=0.000 n=10+10)
SSA            7.94M ± 0%      7.81M ± 0%  -1.66%  (p=0.000 n=10+9)
Flate           246k ± 0%       242k ± 1%  -1.42%  (p=0.000 n=10+10)
GoParser        325k ± 1%       323k ± 1%  -0.84%  (p=0.000 n=10+10)
Reflect        1.02M ± 0%      1.01M ± 0%  -0.99%  (p=0.000 n=10+10)
Tar             259k ± 0%       257k ± 1%  -0.72%  (p=0.009 n=10+10)
XML             406k ± 1%       403k ± 1%  -0.69%  (p=0.001 n=10+10)

Change-Id: Ia129a4cd272027d627e1f3b27e9f07f93e3aa27e
Reviewed-on: https://go-review.googlesource.com/38230Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Passes toolstash -cmp.

Updates #15756

Change-Id: Ia071dbbd7f2ee0f8433d8c37af4f7b588016244e
Reviewed-on: https://go-review.googlesource.com/38231Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Stack barriers were removed in CL 36620.

Change-Id: If124d65a73a7b344a42be2a4b386a14d7a0a428b
Reviewed-on: https://go-review.googlesource.com/38169Reviewed-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
Reviewed-by: David Chase <drchase@google.com>

This matches the error message of cmd/compile (for assignments).

Change-Id: I42a428f5d72f034e7b7e97b090a929e317e812af
Reviewed-on: https://go-review.googlesource.com/38315
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>

See https://go-review.googlesource.com/#/c/38313/ for background.
It turns out that only a few tests checked for this.

The new error message is shorter and very clear.

Change-Id: I8ab4ad59fb023c8b54806339adc23aefd7dc7b07
Reviewed-on: https://go-review.googlesource.com/38314
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

This is an evolution of https://go-review.googlesource.com/33616, as discussed
via email with Robert (gri):

$ cat foobar.go
package main

func main() {
        a := "foo", "bar"
}

before:
./foobar.go:4:4: assignment count mismatch: want 1 values, got 2

after:
./foobar.go:4:4: assignment count mismatch: cannot assign 2 values to 1 variables

We could likely also eliminate the "assignment count mismatch" prefix now
without losing any information, but that string is matched by a number of
tests.

Change-Id: Ie6fc8a7bbd0ebe841d53e66e5c2f49868decf761
Reviewed-on: https://go-review.googlesource.com/38313Reviewed-by: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

name              old time/op  new time/op  delta
LeadingZeros-4    2.00ns ± 0%  1.34ns ± 1%  -33.02%  (p=0.000 n=8+10)
LeadingZeros16-4  1.62ns ± 0%  1.57ns ± 0%   -3.09%  (p=0.001 n=8+9)
LeadingZeros32-4  2.14ns ± 0%  1.48ns ± 0%  -30.84%  (p=0.002 n=8+10)
LeadingZeros64-4  2.06ns ± 1%  1.33ns ± 0%  -35.08%  (p=0.000 n=8+8)

8-bit args is a special case - the Go code is really fast because
it is just a single table lookup.  So I've disabled that for now.
Intrinsics were actually slower:
LeadingZeros8-4   1.22ns ± 3%  1.58ns ± 1%  +29.56%  (p=0.000 n=10+10)

Update #18616

Change-Id: Ia9c289b9ba59c583ea64060470315fd637e814cf
Reviewed-on: https://go-review.googlesource.com/38311
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>

Updates #17802

Change-Id: I65ea0f4cde973604c04051e7eb25d12e4facecd3
Reviewed-on: https://go-review.googlesource.com/36626Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Chris Broadfoot <cbro@golang.org>

Avoid memory allocations by returning pre-calculated strings
for decimal ints in the range 0..99.

Benchmark results:

name              old time/op    new time/op    delta
FormatInt-4         2.45µs ± 1%    2.40µs ± 1%    -1.86%  (p=0.000 n=8+9)
AppendInt-4         1.67µs ± 1%    1.65µs ± 0%    -0.92%  (p=0.000 n=10+10)
FormatUint-4         676ns ± 3%     669ns ± 1%      ~     (p=0.146 n=10+10)
AppendUint-4         467ns ± 2%     474ns ± 0%    +1.58%  (p=0.000 n=10+10)
FormatIntSmall-4    29.6ns ± 2%     3.3ns ± 0%   -88.98%  (p=0.000 n=10+9)
AppendIntSmall-4    16.0ns ± 1%     8.5ns ± 0%   -46.98%  (p=0.000 n=10+9)

name              old alloc/op   new alloc/op   delta
FormatInt-4           576B ± 0%      576B ± 0%      ~     (all equal)
AppendInt-4          0.00B          0.00B           ~     (all equal)
FormatUint-4          224B ± 0%      224B ± 0%      ~     (all equal)
AppendUint-4         0.00B          0.00B           ~     (all equal)
FormatIntSmall-4     2.00B ± 0%     0.00B       -100.00%  (p=0.000 n=10+10)
AppendIntSmall-4     0.00B          0.00B           ~     (all equal)

name              old allocs/op  new allocs/op  delta
FormatInt-4           37.0 ± 0%      35.0 ± 0%    -5.41%  (p=0.000 n=10+10)
AppendInt-4           0.00           0.00           ~     (all equal)
FormatUint-4          6.00 ± 0%      6.00 ± 0%      ~     (all equal)
AppendUint-4          0.00           0.00           ~     (all equal)
FormatIntSmall-4      1.00 ± 0%      0.00       -100.00%  (p=0.000 n=10+10)
AppendIntSmall-4      0.00           0.00           ~     (all equal)

Fixes #19445

Change-Id: Ib1f8922f2e0b13743c847ee9e703d1dab77f705c
Reviewed-on: https://go-review.googlesource.com/37963Reviewed-by: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Update #18616

Change-Id: I0c2d643cbbeb131b4c9b12194697afa4af48e1d2
Reviewed-on: https://go-review.googlesource.com/38166
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>

A copy-paste error in CL 38150. Fix build.

Change-Id: Ib2afc83564ebe7dab934d45522803e1a191dea18
Reviewed-on: https://go-review.googlesource.com/38292
Run-TryBot: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Keith Randall <khr@golang.org>

Fixes #19576.

Change-Id: I11034fb08e989f6eb7d54bde873b92804223598d
Reviewed-on: https://go-review.googlesource.com/38291
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Remove size AuxInt in Store, and alignment in Move/Zero. We still
pass size AuxInt to Move/Zero, as it is used for partial Move/Zero
lowering (e.g. cmd/compile/internal/ssa/gen/386.rules:288).
SizeAndAlign is gone.

Passes "toolstash -cmp" on std.

Change-Id: I1ca34652b65dd30de886940e789fcf41d521475d
Reviewed-on: https://go-review.googlesource.com/38150
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>

The old writebarrier implementation fails to handle single-block
loop where a memory Phi value depends on the write barrier store
in the same block. The new implementation (CL 36834) doesn't have
this problem. Add a test to ensure it.

Fix #19067.

Change-Id: Iab13c6817edc12be8a048d18699b4450fa7ed712
Reviewed-on: https://go-review.googlesource.com/36940Reviewed-by: David Chase <drchase@google.com>

Now that the write barrier insertion is moved to SSA, the SSA
building code can be simplified.

Updates #17583.

Change-Id: I5cacc034b11aa90b0abe6f8dd97e4e3994e2bc25
Reviewed-on: https://go-review.googlesource.com/36840
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>

When the compiler insert write barriers, the frontend makes
conservative decisions at an early stage. This sometimes have
false positives because of the lack of information, for example,
writes on stack. SSA's writebarrier pass identifies writes on
stack and eliminates write barriers for them.

This CL moves write barrier insertion into SSA. The frontend no
longer makes decisions about write barriers, and simply does
normal assignments and emits normal Store ops when building SSA.
SSA writebarrier pass inserts write barrier for Stores when needed.
There, it has better information about the store because Phi and
Copy propagation are done at that time.

This CL only changes StoreWB to Store in gc/ssa.go. A followup CL
simplifies SSA building code.

Updates #17583.

Change-Id: I4592d9bc0067503befc169c50b4e6f4765673bec
Reviewed-on: https://go-review.googlesource.com/36839
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>