Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible

NOPRESUBMIT=true

Committed: https://crrev.com/2cf2eef77bec7d30ccc80cb74950f35bc37404ed
Cr-Commit-Position: refs/heads/master@{#38497}

[Jakob Kummerow, 2016-08-09 12:13:25]

jkummerow@chromium.org changed reviewers:
+ verwaest@chromium.org

[Jakob Kummerow, 2016-08-09 12:13:25]

PTAL.

[Toon Verwaest, 2016-08-09 13:59:01]

lgtm

https://codereview.chromium.org/2227703003/diff/20001/src/code-stub-assembler.cc
File src/code-stub-assembler.cc (right):

https://codereview.chromium.org/2227703003/diff/20001/src/code-stub-assembler...
src/code-stub-assembler.cc:3456: // |is_jsarray| should be non-zero for
JSArrays.
Add comment, fast elements only.

https://codereview.chromium.org/2227703003/diff/20001/src/code-stub-assembler...
src/code-stub-assembler.cc:3688: Node* protector_cell =
LoadRoot(Heap::kArrayProtectorRootIndex);
DCHECK that ArrayProtector is a PropertyCell at this point.

[Jakob Kummerow, 2016-08-09 14:30:45]

Thanks for the review; landing.

https://codereview.chromium.org/2227703003/diff/20001/src/code-stub-assembler.cc
File src/code-stub-assembler.cc (right):

https://codereview.chromium.org/2227703003/diff/20001/src/code-stub-assembler...
src/code-stub-assembler.cc:3456: // |is_jsarray| should be non-zero for
JSArrays.
On 2016/08/09 13:59:01, Toon Verwaest wrote:
> Add comment, fast elements only.

WontFix -- my next CL is adding DICTIONARY_ELEMENTS support, so the comment
would be very short-lived ;-)

https://codereview.chromium.org/2227703003/diff/20001/src/code-stub-assembler...
src/code-stub-assembler.cc:3688: Node* protector_cell =
LoadRoot(Heap::kArrayProtectorRootIndex);
On 2016/08/09 13:59:01, Toon Verwaest wrote:
> DCHECK that ArrayProtector is a PropertyCell at this point.

Done.

[Jakob Kummerow, 2016-08-09 14:31:00]

The CQ bit was checked by jkummerow@chromium.org

[Jakob Kummerow, 2016-08-09 14:31:00]

The patchset sent to the CQ was uploaded after l-g-t-m from
verwaest@chromium.org
Link to the patchset: https://codereview.chromium.org/2227703003/#ps40001
(title: "add DCHECK")

[commit-bot: I haz the power, 2016-08-09 14:31:07]

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, 2016-08-09 14:56:48]

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

[commit-bot: I haz the power, 2016-08-09 14:56:49]

Failed to commit the patch.

[Michael Achenbach, 2016-08-09 15:01:31]

The CQ bit was checked by machenbach@chromium.org

[commit-bot: I haz the power, 2016-08-09 15:01:47]

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

[Michael Achenbach, 2016-08-09 15:01:55]

Description was changed from

==========
Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible
==========

to

==========
Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible

NOPRESUBMIT=true
==========

[Michael Achenbach, 2016-08-09 15:02:06]

The CQ bit was unchecked by machenbach@chromium.org

[Michael Achenbach, 2016-08-09 15:02:10]

The CQ bit was checked by machenbach@chromium.org

[commit-bot: I haz the power, 2016-08-09 15:02:21]

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, 2016-08-09 15:02:42]

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

[commit-bot: I haz the power, 2016-08-09 15:02:43]

Failed to commit the patch.

[tandrii(chromium), 2016-08-09 15:08:08]

The CQ bit was checked by tandrii@chromium.org

[commit-bot: I haz the power, 2016-08-09 15:08:42]

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, 2016-08-09 15:10:20]

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

[commit-bot: I haz the power, 2016-08-09 15:10:21]

Failed to commit the patch.

[tandrii(chromium), 2016-08-09 15:16:00]

The CQ bit was checked by tandrii@chromium.org

[commit-bot: I haz the power, 2016-08-09 15:16:11]

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, 2016-08-09 15:16:31]

Description was changed from

==========
Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible

NOPRESUBMIT=true
==========

to

==========
Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible

NOPRESUBMIT=true
==========

[commit-bot: I haz the power, 2016-08-09 15:16:32]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-08-09 15:17:06]

Description was changed from

==========
Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible

NOPRESUBMIT=true
==========

to

==========
Reland "[KeyedLoadIC] Support Smi "handlers" for element loads"

This reverts commit d9d719e7a89dfcb90365404574f452009f2cb409.
Original review: https://codereview.chromium.org/2180273002/

With additional fixes to repair performance regressions:
- check PropertyCell::kValueOffset, not Cell::kValueOffset
- convert HeapNumbers to int on the fly if possible

NOPRESUBMIT=true

Committed: https://crrev.com/2cf2eef77bec7d30ccc80cb74950f35bc37404ed
Cr-Commit-Position: refs/heads/master@{#38497}
==========

[commit-bot: I haz the power, 2016-08-09 15:17:07]

Patchset 3 (id:??) landed as
https://crrev.com/2cf2eef77bec7d30ccc80cb74950f35bc37404ed
Cr-Commit-Position: refs/heads/master@{#38497}