[ARM] Refactor NEON binary ops in assembler.

[ARM] Refactor NEON binary ops in assembler.
- Refactors many FP, integer, and signed integer instructions where
possible.

LOG=N
BUG=v8:4124
Review-Url: https://codereview.chromium.org/2639443002
Cr-Commit-Position: refs/heads/master@{#42463}
Committed: https://chromium.googlesource.com/v8/v8/+/ac620154eeb042f17e9e35abb6b93635f89bf51f

[bbudge, 2017-01-16 22:50:03]

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

[commit-bot: I haz the power, 2017-01-16 22:50:14]

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, 2017-01-16 23:15:12]

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

[commit-bot: I haz the power, 2017-01-16 23:15:12]

Dry run: This issue passed the CQ dry run.

[bbudge, 2017-01-17 00:15:26]

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

[commit-bot: I haz the power, 2017-01-17 00:15:30]

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

[bbudge, 2017-01-17 00:31:09]

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

[commit-bot: I haz the power, 2017-01-17 00:41:43]

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

[commit-bot: I haz the power, 2017-01-17 00:41:43]

Dry run: This issue passed the CQ dry run.

[martyn.capewell, 2017-01-17 13:41:34]

This looks good, but I wonder if the encoding of NeonDataType should be changed.

I think NeonDataType has been encoded to match the requirement of VMOVL, but
that seems like an exception to me.

If it looked more like the values in NeonSize, it should be easier to encode the
enum directly in the instruction, and then combine the signed and sign-agnostic
binary ops into a single emit encoder.

It would tidy up the binops, but would need modification of VMOVL.

What do you think?

[bbudge, 2017-01-17 14:36:21]

On 2017/01/17 13:41:34, martyn.capewell wrote:
> This looks good, but I wonder if the encoding of NeonDataType should be
changed.
> 
> I think NeonDataType has been encoded to match the requirement of VMOVL, but
> that seems like an exception to me.
> 
> If it looked more like the values in NeonSize, it should be easier to encode
the
> enum directly in the instruction, and then combine the signed and
sign-agnostic
> binary ops into a single emit encoder.
> 
> It would tidy up the binops, but would need modification of VMOVL.
> 
> What do you think?

It sounds like a good idea.

Would it be OK if I include that change in this CL?

[martyn.capewell, 2017-01-17 15:02:45]

On 2017/01/17 14:36:21, bbudge_ooo_until_mid_jan wrote:

> Would it be OK if I include that change in this CL?

If this works in merging the integer ops, you'll change the same code anyway, so
doing it in this patch makes sense.

[bbudge, 2017-01-17 18:05:55]

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

[commit-bot: I haz the power, 2017-01-17 18:05:59]

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, 2017-01-17 18:35:07]

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

[commit-bot: I haz the power, 2017-01-17 18:35:08]

Dry run: This issue passed the CQ dry run.

[bbudge, 2017-01-17 19:01:14]

Your suggestion to make NeonDataType match up with NeonSize seems OK. PTAL.

[martyn.capewell, 2017-01-18 13:47:35]

lgtm

[bbudge, 2017-01-18 15:09:55]

The CQ bit was checked by bbudge@chromium.org

[commit-bot: I haz the power, 2017-01-18 15:09:59]

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 15:11:36]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1484752195706670,
"parent_rev": "761f373b520d201848c0ef6b209039bf12d4eded", "commit_rev":
"ac620154eeb042f17e9e35abb6b93635f89bf51f"}

[commit-bot: I haz the power, 2017-01-18 15:11:43]

Description was changed from

==========
[ARM] Refactor NEON binary ops in assembler.
- Refactors many FP, integer, and signed integer instructions where
possible.

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

to

==========
[ARM] Refactor NEON binary ops in assembler.
- Refactors many FP, integer, and signed integer instructions where
possible.

LOG=N
BUG=v8:4124

Review-Url: https://codereview.chromium.org/2639443002
Cr-Commit-Position: refs/heads/master@{#42463}
Committed:
https://chromium.googlesource.com/v8/v8/+/ac620154eeb042f17e9e35abb6b93635f89...
==========

[commit-bot: I haz the power, 2017-01-18 15:11:44]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/v8/v8/+/ac620154eeb042f17e9e35abb6b93635f89...