MIPS[64]: Port `[deoptimizer] Fill the single precision registers in the deoptimizer entry stub.`

MIPS[64]: Port `[deoptimizer] Fill the single precision registers in the deoptimizer entry stub.`

Port 798ffc9d67e25a0aa9b0be145892c27f2ac5391f

TEST=mjsunit/compiler/regress-v8-6077
BUG=

Review-Url: https://codereview.chromium.org/2773583003
Cr-Commit-Position: refs/heads/master@{#44086}
Committed: https://chromium.googlesource.com/v8/v8/+/2458683f5542aeaa6ff0b32dc4d376c48dde4509

[ivica.bogosavljevic, 2017-03-23 11:01:02]

ivica.bogosavljevic@imgtec.com changed reviewers:
+ bbudge@chromium.org, jarin@chromium.org, mstarzinger@chromium.org

[ivica.bogosavljevic, 2017-03-23 11:01:03]

PTAL

[Jarin, 2017-03-23 12:31:32]

lgtm, thanks!

https://codereview.chromium.org/2773583003/diff/1/src/mips/deoptimizer-mips.cc
File src/mips/deoptimizer-mips.cc (right):

https://codereview.chromium.org/2773583003/diff/1/src/mips/deoptimizer-mips.c...
src/mips/deoptimizer-mips.cc:217: //
float_registers_[FloarRegister::kNumAllocatableRegisters]
Floar -> Float?

[ivica.bogosavljevic, 2017-03-23 14:17:46]

The CQ bit was checked by ivica.bogosavljevic@imgtec.com to run a CQ dry run

[commit-bot: I haz the power, 2017-03-23 14:17:49]

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, 2017-03-23 14:43:41]

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

[commit-bot: I haz the power, 2017-03-23 14:43:42]

Dry run: This issue passed the CQ dry run.

[ivica.bogosavljevic, 2017-03-24 08:20:00]

The CQ bit was checked by ivica.bogosavljevic@imgtec.com

[ivica.bogosavljevic, 2017-03-24 08:20:01]

The patchset sent to the CQ was uploaded after l-g-t-m from jarin@chromium.org
Link to the patchset: https://codereview.chromium.org/2773583003/#ps20001
(title: "Address typo")

[commit-bot: I haz the power, 2017-03-24 08:20:11]

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, 2017-03-24 08:21:26]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1490343600806750,
"parent_rev": "16c38aa820521e2621ffd121977e93d912fbf2aa", "commit_rev":
"2458683f5542aeaa6ff0b32dc4d376c48dde4509"}

[commit-bot: I haz the power, 2017-03-24 08:21:31]

Description was changed from

==========
MIPS[64]: Port `[deoptimizer] Fill the single precision registers in the
deoptimizer entry stub.`

Port 798ffc9d67e25a0aa9b0be145892c27f2ac5391f

TEST=mjsunit/compiler/regress-v8-6077
BUG=
==========

to

==========
MIPS[64]: Port `[deoptimizer] Fill the single precision registers in the
deoptimizer entry stub.`

Port 798ffc9d67e25a0aa9b0be145892c27f2ac5391f

TEST=mjsunit/compiler/regress-v8-6077
BUG=

Review-Url: https://codereview.chromium.org/2773583003
Cr-Commit-Position: refs/heads/master@{#44086}
Committed:
https://chromium.googlesource.com/v8/v8/+/2458683f5542aeaa6ff0b32dc4d376c48dd...
==========

[commit-bot: I haz the power, 2017-03-24 08:21:32]

Committed patchset #2 (id:20001) as
https://chromium.googlesource.com/v8/v8/+/2458683f5542aeaa6ff0b32dc4d376c48dd...

[ivica.bogosavljevic, 2017-03-24 12:03:30]

On 2017/03/23 12:31:32, Jarin wrote:
> lgtm, thanks!
> 
> https://codereview.chromium.org/2773583003/diff/1/src/mips/deoptimizer-mips.cc
> File src/mips/deoptimizer-mips.cc (right):
> 
>
https://codereview.chromium.org/2773583003/diff/1/src/mips/deoptimizer-mips.c...
> src/mips/deoptimizer-mips.cc:217: //
> float_registers_[FloarRegister::kNumAllocatableRegisters]
> Floar -> Float?

Jarin, I just realized I missed something. The single and double registers on
MIPS are overlayed i.e. 
float f0 and f1 are same as double f0, float f2 and f3 are same as double f2.
I don't understand completely the motivation for this change and if it's needed,
but without it the newly introduced test fails. Do you think we could lower the
deoptimization
costs by taking advantage of the fact that single and double registers are
overlayed for MIPS?