[Turbofan] Add other integer SIMD types, add more integer ops.

[Turbofan] Add other integer SIMD types, add more integer ops.
- Adds Int16x8, Int8x16 types.
- Adds neg, abs unary ops.
- Adds add, sub, mul, and signed / unsigned min / max, comparison ops.

LOG=N
BUG=v8:4124
Review-Url: https://codereview.chromium.org/2638133002
Cr-Commit-Position: refs/heads/master@{#42674}
Committed: https://chromium.googlesource.com/v8/v8/+/c5bdbbbebe12ba41e69f5e4cddc443ede2f7576c

[bbudge, 2017-01-18 00:47:20]

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

[commit-bot: I haz the power, 2017-01-18 00:47:23]

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, 2017-01-18 00:49:53]

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

[commit-bot: I haz the power, 2017-01-18 00:49:54]

Dry run: Try jobs failed on following builders:
  v8_linux64_avx2_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_avx2_rel_ng/buil...)
  v8_linux64_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_rel_ng/builds/19370)
  v8_linux_dbg_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_dbg_ng/builds/19430)
  v8_linux_gcc_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)
  v8_linux_mips64el_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_mips64el_compile_r...)
  v8_linux_nodcheck_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_nodcheck_rel_ng/bu...)
  v8_linux_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_rel_ng/builds/19340)
  v8_mac_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel_ng/builds/15813)

[bbudge, 2017-01-18 00:58:41]

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

[commit-bot: I haz the power, 2017-01-18 00:58:44]

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, 2017-01-18 01:00:53]

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

[commit-bot: I haz the power, 2017-01-18 01:00:54]

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_gcc_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)
  v8_mac_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel_ng/builds/15814)

[bbudge, 2017-01-18 01:16:03]

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

[bbudge, 2017-01-18 01:16:07]

Patchset #2 (id:20001) has been deleted

[commit-bot: I haz the power, 2017-01-18 01:16:09]

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, 2017-01-18 01:18:38]

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

[commit-bot: I haz the power, 2017-01-18 01:18:38]

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_gcc_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)

[bbudge, 2017-01-18 01:25:47]

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

[commit-bot: I haz the power, 2017-01-18 01:25:59]

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, 2017-01-18 01:52:05]

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

[commit-bot: I haz the power, 2017-01-18 01:52:06]

Dry run: Try jobs failed on following builders:
  v8_linux_arm_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel_ng/builds/...)
  v8_linux_arm_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel_ng_trigger...)

[bbudge, 2017-01-25 01:05:32]

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

[commit-bot: I haz the power, 2017-01-25 01:05:38]

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, 2017-01-25 01:09:48]

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

[commit-bot: I haz the power, 2017-01-25 01:09:49]

Dry run: Try jobs failed on following builders:
  v8_linux64_avx2_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_avx2_rel_ng/buil...)
  v8_linux64_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_rel_ng/builds/19842)

[bbudge, 2017-01-25 01:52:11]

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

[commit-bot: I haz the power, 2017-01-25 01:52:21]

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

[bbudge, 2017-01-25 01:53:32]

Description was changed from

==========
[Turbofan] Add other integer SIMD types, add more integer ops.

LOG=N
BUG=v8:4124
==========

to

==========
[Turbofan] Add other integer SIMD types, add more integer ops.
- Adds Int16x8, Int8x16 types.
- Adds neg, abs unary ops.
- Adds add, sub, mul, and signed / unsigned min / max, comparison ops.

LOG=N
BUG=v8:4124
==========

[commit-bot: I haz the power, 2017-01-25 01:59:31]

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

[commit-bot: I haz the power, 2017-01-25 01:59:32]

Dry run: Try jobs failed on following builders:
  v8_linux64_gyp_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng/build...)

[bbudge, 2017-01-25 02:21:07]

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

[commit-bot: I haz the power, 2017-01-25 02:21:12]

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, 2017-01-25 02:50:46]

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

[commit-bot: I haz the power, 2017-01-25 02:50:48]

Dry run: This issue passed the CQ dry run.

[bbudge, 2017-01-25 03:14:52]

bbudge@chromium.org changed reviewers:
+ titzer@chromium.org

[bbudge, 2017-01-25 03:14:53]

I'll finish the integer ops in a follow on. That would be the shifts, and
unsigned min/max.

[bbudge, 2017-01-25 03:23:40]

bbudge@chromium.org changed reviewers:
+ martyn.capewell@arm.com

[bbudge, 2017-01-25 03:23:40]

+ Martyn for ARM

[titzer, 2017-01-25 09:47:25]

lgtm if simd ops can be moved out of the common operator builder.

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/arm/instr...
File src/compiler/arm/instruction-selector-arm.cc (right):

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/arm/instr...
src/compiler/arm/instruction-selector-arm.cc:2389: void
InstructionSelector::VisitInt32x4Neg(Node* node) {
I think it's time to macro-ify these methods, but that can be done in a followup
CL.

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/common-op...
File src/compiler/common-operator.h (right):

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/common-op...
src/compiler/common-operator.h:390: const Operator*
Float32x4ExtractLane(int32_t);
Can we move all of these operators to the machine operator builder?

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/graph.h
File src/compiler/graph.h (right):

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/graph.h#n...
src/compiler/graph.h:69: Node* NewNode(const Operator* op, Node* n) { return
NewNode(op, 1, &n); }
This is a nice simplification, but IIRC there was a reason we didn't do this
yet. Might trigger a perf alert. Separate CL?

https://codereview.chromium.org/2638133002/diff/160001/src/wasm/function-body...
File src/wasm/function-body-decoder.cc (right):

https://codereview.chromium.org/2638133002/diff/160001/src/wasm/function-body...
src/wasm/function-body-decoder.cc:351: inline bool Validate(const byte* pc,
LaneOperand& operand,
Can we put the opcode first, just for analogy to their order in the bytecode?

[martyn.capewell, 2017-01-25 11:31:14]

ARM parts lgtm; second Ben's idea on using macros.

[bbudge, 2017-01-25 23:56:53]

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

[commit-bot: I haz the power, 2017-01-25 23:56:57]

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, 2017-01-26 00:01:47]

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

[commit-bot: I haz the power, 2017-01-26 00:01:48]

Dry run: Try jobs failed on following builders:
  v8_linux_gcc_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)

[bbudge, 2017-01-26 00:35:00]

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

[commit-bot: I haz the power, 2017-01-26 00:35:08]

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, 2017-01-26 00:39:58]

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

[commit-bot: I haz the power, 2017-01-26 00:39:59]

Dry run: Try jobs failed on following builders:
  v8_linux_gcc_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)

[bbudge, 2017-01-26 01:13:28]

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

[commit-bot: I haz the power, 2017-01-26 01:13:38]

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

[bbudge, 2017-01-26 01:45:01]

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

[commit-bot: I haz the power, 2017-01-26 01:45:12]

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

[bbudge, 2017-01-26 02:04:43]

With some difficulty, I moved the SIMD lane operators to MachineOperatorBuilder,
caching them like most of the other operators.

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/arm/instr...
File src/compiler/arm/instruction-selector-arm.cc (right):

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/arm/instr...
src/compiler/arm/instruction-selector-arm.cc:2389: void
InstructionSelector::VisitInt32x4Neg(Node* node) {
On 2017/01/25 09:47:25, titzer wrote:
> I think it's time to macro-ify these methods, but that can be done in a
followup
> CL.

Added a TODO.

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/common-op...
File src/compiler/common-operator.h (right):

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/common-op...
src/compiler/common-operator.h:390: const Operator*
Float32x4ExtractLane(int32_t);
On 2017/01/25 09:47:25, titzer wrote:
> Can we move all of these operators to the machine operator builder? 

Moved to MachineOperatorBuilder, and added to cache.

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/graph.h
File src/compiler/graph.h (right):

https://codereview.chromium.org/2638133002/diff/160001/src/compiler/graph.h#n...
src/compiler/graph.h:69: Node* NewNode(const Operator* op, Node* n) { return
NewNode(op, 1, &n); }
On 2017/01/25 09:47:25, titzer wrote:
> This is a nice simplification, but IIRC there was a reason we didn't do this
> yet. Might trigger a perf alert. Separate CL?

Agreed. Changed to explicit overloads up to 17 params.

https://codereview.chromium.org/2638133002/diff/160001/src/wasm/function-body...
File src/wasm/function-body-decoder.cc (right):

https://codereview.chromium.org/2638133002/diff/160001/src/wasm/function-body...
src/wasm/function-body-decoder.cc:351: inline bool Validate(const byte* pc,
LaneOperand& operand,
On 2017/01/25 09:47:25, titzer wrote:
> Can we put the opcode first, just for analogy to their order in the bytecode?

Done. (put opcode before operand)

[commit-bot: I haz the power, 2017-01-26 02:15:12]

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

[commit-bot: I haz the power, 2017-01-26 02:15:13]

Dry run: This issue passed the CQ dry run.

[bbudge, 2017-01-26 02:15:34]

The CQ bit was checked by bbudge@chromium.org

[bbudge, 2017-01-26 02:15:35]

The patchset sent to the CQ was uploaded after l-g-t-m from titzer@chromium.org,
martyn.capewell@arm.com
Link to the patchset: https://codereview.chromium.org/2638133002/#ps240001
(title: "Fix name of static fields.")

[commit-bot: I haz the power, 2017-01-26 02:15:43]

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, 2017-01-26 02:18:02]

CQ is committing da patch.

Bot data: {"patchset_id": 240001, "attempt_start_ts": 1485396934509960,
"parent_rev": "d28c04486bd5c415748cd00f5c21c92fb35dd025", "commit_rev":
"c5bdbbbebe12ba41e69f5e4cddc443ede2f7576c"}

[commit-bot: I haz the power, 2017-01-26 02:18:06]

Description was changed from

==========
[Turbofan] Add other integer SIMD types, add more integer ops.
- Adds Int16x8, Int8x16 types.
- Adds neg, abs unary ops.
- Adds add, sub, mul, and signed / unsigned min / max, comparison ops.

LOG=N
BUG=v8:4124
==========

to

==========
[Turbofan] Add other integer SIMD types, add more integer ops.
- Adds Int16x8, Int8x16 types.
- Adds neg, abs unary ops.
- Adds add, sub, mul, and signed / unsigned min / max, comparison ops.

LOG=N
BUG=v8:4124

Review-Url: https://codereview.chromium.org/2638133002
Cr-Commit-Position: refs/heads/master@{#42674}
Committed:
https://chromium.googlesource.com/v8/v8/+/c5bdbbbebe12ba41e69f5e4cddc443ede2f...
==========

[commit-bot: I haz the power, 2017-01-26 02:18:07]

Committed patchset #12 (id:240001) as
https://chromium.googlesource.com/v8/v8/+/c5bdbbbebe12ba41e69f5e4cddc443ede2f...