[deoptimizer] float32 stack on s390 and ppc is on lower 32 bit

[deoptimizer] float32 stack on s390 and ppc is on lower 32 bit

Also add more local variables to regress-v8-6077 to force
register spill on platform with 32 float registers.

BUG=

Review-Url: https://codereview.chromium.org/2822073003
Cr-Commit-Position: refs/heads/master@{#44865}
Committed: https://chromium.googlesource.com/v8/v8/+/3be834e876bc0ecb98dbe6d4957790d1613dc494

[john.yan, 2017-04-18 20:00:17]

Description was changed from

==========
[deoptimizer] float32 stack on s390 and ppc is on lower 32 bit

Also add more local variables to regress-v8-6077 to force
register spill on platform with 32 float registers.

BUG=
==========

to

==========
[deoptimizer] float32 stack on s390 and ppc is on lower 32 bit

Also add more local variables to regress-v8-6077 to force
register spill on platform with 32 float registers.

BUG=
==========

[john.yan, 2017-04-18 20:00:18]

jyan@ca.ibm.com changed reviewers:
+ bbudge@chromium.org, bjaideep@ca.ibm.com, bmeurer@chromium.org,
jarin@chromium.org, joransiu@ca.ibm.com, michael_dawson@ca.ibm.com,
mstarzinger@chromium.org

[john.yan, 2017-04-18 20:00:40]

The CQ bit was checked by jyan@ca.ibm.com to run a CQ dry run

[john.yan, 2017-04-18 20:00:47]

ptal

[commit-bot: I haz the power, 2017-04-18 20:00:48]

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, 2017-04-18 20:29:38]

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

[commit-bot: I haz the power, 2017-04-18 20:29:38]

Dry run: This issue passed the CQ dry run.

[bbudge, 2017-04-24 20:49:24]

https://codereview.chromium.org/2822073003/diff/1/src/deoptimizer.cc
File src/deoptimizer.cc (right):

https://codereview.chromium.org/2822073003/diff/1/src/deoptimizer.cc#newcode3135
src/deoptimizer.cc:3135: #if !V8_TARGET_ARCH_S390X && !V8_TARGET_ARCH_PPC64
Can you just modify the GetUInt32Slot function (combining the preprocessor
directives with the ones there)? Or is there something special about float
slots?

[john.yan, 2017-04-24 21:00:53]

https://codereview.chromium.org/2822073003/diff/1/src/deoptimizer.cc
File src/deoptimizer.cc (right):

https://codereview.chromium.org/2822073003/diff/1/src/deoptimizer.cc#newcode3135
src/deoptimizer.cc:3135: #if !V8_TARGET_ARCH_S390X && !V8_TARGET_ARCH_PPC64
On 2017/04/24 20:49:24, bbudge wrote:
> Can you just modify the GetUInt32Slot function (combining the preprocessor
> directives with the ones there)? Or is there something special about float
> slots?

Yes, float/double is treated differently than int32/int64. So we can't use
GetUInt32Slot.

[bbudge, 2017-04-24 22:14:30]

lgtm, but I'm not that familiar with the deoptimizer if you have any concerns
about correctness.

[Jarin, 2017-04-25 08:31:05]

lgtm

[john.yan, 2017-04-25 16:24:17]

On 2017/04/24 22:14:30, bbudge wrote:
> lgtm, but I'm not that familiar with the deoptimizer if you have any concerns
> about correctness.

thank you!

[john.yan, 2017-04-25 16:24:32]

On 2017/04/25 08:31:05, Jarin wrote:
> lgtm

thank you!

[john.yan, 2017-04-25 16:24:37]

The CQ bit was checked by jyan@ca.ibm.com

[commit-bot: I haz the power, 2017-04-25 16:24:44]

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, 2017-04-25 16:49:34]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1493137477290070, "parent_rev":
"8887945e90dcae78df7ead48ecb965f36d57081d", "commit_rev":
"3be834e876bc0ecb98dbe6d4957790d1613dc494"}

[commit-bot: I haz the power, 2017-04-25 16:49:40]

Description was changed from

==========
[deoptimizer] float32 stack on s390 and ppc is on lower 32 bit

Also add more local variables to regress-v8-6077 to force
register spill on platform with 32 float registers.

BUG=
==========

to

==========
[deoptimizer] float32 stack on s390 and ppc is on lower 32 bit

Also add more local variables to regress-v8-6077 to force
register spill on platform with 32 float registers.

BUG=

Review-Url: https://codereview.chromium.org/2822073003
Cr-Commit-Position: refs/heads/master@{#44865}
Committed:
https://chromium.googlesource.com/v8/v8/+/3be834e876bc0ecb98dbe6d4957790d1613...
==========

[commit-bot: I haz the power, 2017-04-25 16:49:41]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/v8/v8/+/3be834e876bc0ecb98dbe6d4957790d1613...

[john.yan, 2017-04-26 16:55:07]

On 2017/04/25 08:31:05, Jarin wrote:
> lgtm

Hello, could you also help me backport this one please?
https://chromium-review.googlesource.com/c/486274/

Thanks a lot!