[wasm]implement simd lowering for simple F32x4 and I32x4 unops

[wasm]implement simd lowering for simple F32x4 and I32x4 unops

BUG=v8:4124
R=bradnelson@chromium.org,bbudge@chromium.org,gdeepti@chromium.org,mtrofin@chromium.org,titzer@chromium.org

Review-Url: https://codereview.chromium.org/2726453003
Cr-Commit-Position: refs/heads/master@{#43509}
Committed: https://chromium.googlesource.com/v8/v8/+/a727f9e8c0c8c9984961e1a755a6921122907034

[aseemgarg, 2017-02-28 09:11:08]

The CQ bit was checked by aseemgarg@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-02-28 09:11:12]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-02-28 09:39:46]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-02-28 09:39:47]

Dry run: Try jobs failed on following builders:
  v8_linux_arm64_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel_ng/build...)
  v8_linux_arm64_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel_ng_trigg...)

[gdeepti, 2017-02-28 09:50:49]

https://codereview.chromium.org/2726453003/diff/1/test/cctest/wasm/test-run-w...
File test/cctest/wasm/test-run-wasm-simd.cc (right):

https://codereview.chromium.org/2726453003/diff/1/test/cctest/wasm/test-run-w...
test/cctest/wasm/test-run-wasm-simd.cc:5: #include <cmath>
Header not needed here, already included in cctest.h (utils).

[aseemgarg, 2017-02-28 12:54:08]

https://codereview.chromium.org/2726453003/diff/1/test/cctest/wasm/test-run-w...
File test/cctest/wasm/test-run-wasm-simd.cc (right):

https://codereview.chromium.org/2726453003/diff/1/test/cctest/wasm/test-run-w...
test/cctest/wasm/test-run-wasm-simd.cc:5: #include <cmath>
On 2017/02/28 09:50:49, gdeepti wrote:
> Header not needed here, already included in cctest.h (utils). 

Done.

[aseemgarg, 2017-02-28 12:54:16]

The CQ bit was checked by aseemgarg@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-02-28 12:54:25]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-02-28 13:37:50]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-02-28 13:37:51]

Dry run: This issue passed the CQ dry run.

[titzer, 2017-03-01 09:57:58]

lgtm

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
File src/compiler/simd-scalar-lowering.cc (right):

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
src/compiler/simd-scalar-lowering.cc:431: machine()->Int32Sub(),
graph()->NewNode(common()->Int32Constant(0)),
Can you pull the common()->Int32Constant() out of the loop?

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
src/compiler/simd-scalar-lowering.cc:444:
graph()->NewNode(common()->Int32Constant(0xffffffff)));
Same here.

[aseemgarg, 2017-03-01 10:35:46]

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
File src/compiler/simd-scalar-lowering.cc (right):

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
src/compiler/simd-scalar-lowering.cc:431: machine()->Int32Sub(),
graph()->NewNode(common()->Int32Constant(0)),
On 2017/03/01 09:57:58, titzer wrote:
> Can you pull the common()->Int32Constant() out of the loop?

pulled the whole node out.

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
src/compiler/simd-scalar-lowering.cc:444:
graph()->NewNode(common()->Int32Constant(0xffffffff)));
On 2017/03/01 09:57:58, titzer wrote:
> Same here.

Done.

[aseemgarg, 2017-03-01 10:35:53]

The CQ bit was checked by aseemgarg@chromium.org

[aseemgarg, 2017-03-01 10:35:53]

The patchset sent to the CQ was uploaded after l-g-t-m from titzer@chromium.org
Link to the patchset: https://codereview.chromium.org/2726453003/#ps40001
(title: "comments")

[commit-bot: I haz the power, 2017-03-01 10:35:59]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[titzer, 2017-03-01 10:58:08]

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
File src/compiler/simd-scalar-lowering.cc (right):

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
src/compiler/simd-scalar-lowering.cc:431: machine()->Int32Sub(),
graph()->NewNode(common()->Int32Constant(0)),
On 2017/03/01 10:35:46, aseemgarg wrote:
> On 2017/03/01 09:57:58, titzer wrote:
> > Can you pull the common()->Int32Constant() out of the loop?
> 
> pulled the whole node out.

Oh darn, realized too late. If you had access to the JSGraph (which at this
point, is probably poorly named), then you could use its constant-generating
methods, which cache the nodes.

[aseemgarg, 2017-03-01 11:03:09]

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
File src/compiler/simd-scalar-lowering.cc (right):

https://codereview.chromium.org/2726453003/diff/20001/src/compiler/simd-scala...
src/compiler/simd-scalar-lowering.cc:431: machine()->Int32Sub(),
graph()->NewNode(common()->Int32Constant(0)),
On 2017/03/01 10:58:07, titzer wrote:
> On 2017/03/01 10:35:46, aseemgarg wrote:
> > On 2017/03/01 09:57:58, titzer wrote:
> > > Can you pull the common()->Int32Constant() out of the loop?
> > 
> > pulled the whole node out.
> 
> Oh darn, realized too late. If you had access to the JSGraph (which at this
> point, is probably poorly named), then you could use its constant-generating
> methods, which cache the nodes.

Will take care of it in the next patch.

[commit-bot: I haz the power, 2017-03-01 11:13:01]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1488364553329160,
"parent_rev": "bff9564204c77687189c1271a5eccbeb63d680ba", "commit_rev":
"a727f9e8c0c8c9984961e1a755a6921122907034"}

[commit-bot: I haz the power, 2017-03-01 11:13:06]

Description was changed from

==========
[wasm]implement simd lowering for simple F32x4 and I32x4 unops

BUG=v8:4124
R=bradnelson@chromium.org,bbudge@chromium.org,gdeepti@chromium.org,mtrofin@ch...
==========

to

==========
[wasm]implement simd lowering for simple F32x4 and I32x4 unops

BUG=v8:4124
R=bradnelson@chromium.org,bbudge@chromium.org,gdeepti@chromium.org,mtrofin@ch...

Review-Url: https://codereview.chromium.org/2726453003
Cr-Commit-Position: refs/heads/master@{#43509}
Committed:
https://chromium.googlesource.com/v8/v8/+/a727f9e8c0c8c9984961e1a755a69211229...
==========

[commit-bot: I haz the power, 2017-03-01 11:13:06]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/v8/v8/+/a727f9e8c0c8c9984961e1a755a69211229...

[ahaas, 2017-03-01 12:32:57]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.chromium.org/2721223003/ by ahaas@chromium.org.

The reason for reverting is: There is a crash on a bot
https://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20debug/builds....

[bbudge, 2017-03-01 17:36:42]

https://codereview.chromium.org/2726453003/diff/40001/src/compiler/wasm-compi...
File src/compiler/wasm-compiler.cc (right):

https://codereview.chromium.org/2726453003/diff/40001/src/compiler/wasm-compi...
src/compiler/wasm-compiler.cc:3387: return
graph()->NewNode(jsgraph()->machine()->Float32x4Sqrt(), inputs[0]);
We're rethinking sqrt support. On some platforms (ARM32) this has to be
simulated by a very slow sequence of scalar square root instructions which can't
be rescheduled. Users would be much better off using the recip and rsqrt
approximation and refinement operations instead, since they can schedule those
to avoid stalls.