VectorICs: VisitClassLiteral needs adjustment for IC slot usage.

VectorICs: VisitClassLiteral needs adjustment for IC slot usage.

Also, generic lowering for keyed stores needs to handle the case when
there is no IC slot available (it can use the generic keyed store).

BUG=
R=mstarzinger@chromium.org

Committed: https://crrev.com/878071af3e4ecf403f3a4ba7833245e0f5ddf579
Cr-Commit-Position: refs/heads/master@{#29847}

[mvstanton, 2015-07-24 10:35:00]

Hi Michael, here is the thing I mentioned at lunch, thx!
--Michael

[Michael Starzinger, 2015-07-24 11:51:30]

https://codereview.chromium.org/1252303002/diff/20001/src/compiler/js-generic...
File src/compiler/js-generic-lowering.cc (right):

https://codereview.chromium.org/1252303002/diff/20001/src/compiler/js-generic...
src/compiler/js-generic-lowering.cc:362: InlineCacheState state =
FLAG_vector_stores && p.feedback().index() == -1
Can we restructure this a little to be more readable. I had a hard time
understanding it and would have come to the wrong conclusion without your
offline help. How about this ...

// We have a special case where we do keyed stores but don't have a type
// feedback vector slot allocated to support it. In this case, install
// the megamorphic keyed store stub which needs neither vector nor slot.
bool use_vector_slot = FLAG_vector_stores && p.feedback().index() != -1;
Callable callable = CodeFactory::KeyedStoreICInOptimizedCode(isolate(),
language_mode, use_vector_slot ? UNINITIALIZED : MEGAMORPHIC);
if (use_vector_slot) {
  node->InsertInput(zone(), 3, jsgraph()->SmiConstant(p.feedback().index()));
} else {
  node->RemoveInput(3);
}

[mvstanton, 2015-07-24 12:24:58]

Thx for the comments, note my comment over in ast-graph-builder.cc,
VisitArrayLiteral().

https://codereview.chromium.org/1252303002/diff/20001/src/compiler/js-generic...
File src/compiler/js-generic-lowering.cc (right):

https://codereview.chromium.org/1252303002/diff/20001/src/compiler/js-generic...
src/compiler/js-generic-lowering.cc:362: InlineCacheState state =
FLAG_vector_stores && p.feedback().index() == -1
On 2015/07/24 11:51:30, Michael Starzinger wrote:
> Can we restructure this a little to be more readable. I had a hard time
> understanding it and would have come to the wrong conclusion without your
> offline help. How about this ...
> 
> // We have a special case where we do keyed stores but don't have a type
> // feedback vector slot allocated to support it. In this case, install
> // the megamorphic keyed store stub which needs neither vector nor slot.
> bool use_vector_slot = FLAG_vector_stores && p.feedback().index() != -1;
> Callable callable = CodeFactory::KeyedStoreICInOptimizedCode(isolate(),
> language_mode, use_vector_slot ? UNINITIALIZED : MEGAMORPHIC);
> if (use_vector_slot) {
>   node->InsertInput(zone(), 3, jsgraph()->SmiConstant(p.feedback().index()));
> } else {
>   node->RemoveInput(3);
> }

Most wise, thanks!

[Michael Starzinger, 2015-07-24 12:26:35]

LGTM.

[mvstanton, 2015-07-24 12:28:40]

The CQ bit was checked by mvstanton@chromium.org

[commit-bot: I haz the power, 2015-07-24 12:28:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1252303002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1252303002/40001

[commit-bot: I haz the power, 2015-07-24 12:36:34]

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

[commit-bot: I haz the power, 2015-07-24 12:36:34]

Try jobs failed on following builders:
  v8_mac_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel/builds/8064)

[mvstanton, 2015-07-24 12:55:20]

The CQ bit was checked by mvstanton@chromium.org

[mvstanton, 2015-07-24 12:55:20]

The patchset sent to the CQ was uploaded after l-g-t-m from
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1252303002/#ps60001
(title: "doh! typo.")

[commit-bot: I haz the power, 2015-07-24 12:55:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1252303002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1252303002/60001

[commit-bot: I haz the power, 2015-07-24 13:03:08]

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

[commit-bot: I haz the power, 2015-07-24 13:03:09]

Try jobs failed on following builders:
  v8_mac_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel/builds/8067)

[mvstanton, 2015-07-24 14:07:20]

The CQ bit was checked by mvstanton@chromium.org

[mvstanton, 2015-07-24 14:07:21]

The patchset sent to the CQ was uploaded after l-g-t-m from
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1252303002/#ps80001
(title: "The wrong IC was chosen without vector stores on.")

[commit-bot: I haz the power, 2015-07-24 14:07:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1252303002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1252303002/80001

[commit-bot: I haz the power, 2015-07-24 14:08:35]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-07-24 14:09:02]

Patchset 5 (id:??) landed as
https://crrev.com/878071af3e4ecf403f3a4ba7833245e0f5ddf579
Cr-Commit-Position: refs/heads/master@{#29847}