[interpreter] Merge {OsrPoll} with {Jump} bytecode.

[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performing the
backwards branch. This reduces the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
Committed: https://crrev.com/c9864173f145c14866bcbf94759c53aa65847291
Cr-Commit-Position: refs/heads/master@{#39384}

[Michael Starzinger, 2016-09-12 16:31:16]

mstarzinger@chromium.org changed reviewers:
+ rmcilroy@chromium.org

[Michael Starzinger, 2016-09-12 16:31:35]

Description was changed from

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performaing the
backwards branch. This redcues the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=
BUG=v8:4764
==========

to

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performaing the
backwards branch. This redcues the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
==========

[Michael Starzinger, 2016-09-12 16:33:02]

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

[commit-bot: I haz the power, 2016-09-12 16:33:06]

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

[Michael Starzinger, 2016-09-12 16:35:40]

Description was changed from

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performaing the
backwards branch. This redcues the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
==========

to

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performing the
backwards branch. This reduces the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
==========

[Michael Starzinger, 2016-09-12 16:36:59]

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

[commit-bot: I haz the power, 2016-09-12 16:37:04]

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, 2016-09-12 16:49:34]

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

[commit-bot: I haz the power, 2016-09-12 16:49:35]

Dry run: Try jobs failed on following builders:
  v8_linux64_gyp_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng/build...)
  v8_linux64_gyp_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng_trigg...)

[Michael Starzinger, 2016-09-12 17:04:02]

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

[commit-bot: I haz the power, 2016-09-12 17:04:12]

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, 2016-09-12 17:31:41]

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

[commit-bot: I haz the power, 2016-09-12 17:31:42]

Dry run: This issue passed the CQ dry run.

[rmcilroy, 2016-09-13 10:47:34]

LGTM. If all backjumps are now JumpLoop, could you add a TODO(rmcilroy) in
InterpreterAssembler::UpdateInterruptBudget to only do the check on
back-branches.

https://codereview.chromium.org/2331033002/diff/60001/src/interpreter/bytecod...
File src/interpreter/bytecode-array-writer.cc (right):

https://codereview.chromium.org/2331033002/diff/60001/src/interpreter/bytecod...
src/interpreter/bytecode-array-writer.cc:361:
node->set_bytecode(node->bytecode(), delta, node->operand(1));
Do we ever have label's bound which don't have Bytecode::kJumpLoop? If not,
could you remove the branch and just make this a DCHECK, otherwise, do you know
where they come from?

https://codereview.chromium.org/2331033002/diff/60001/src/interpreter/bytecod...
src/interpreter/bytecode-array-writer.cc:366: // The label has not yet been
bound so this is a forward reference
Could you add a DCHECK that this is not Bytecode::kJumpLoop

[Michael Starzinger, 2016-09-13 12:01:20]

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

[commit-bot: I haz the power, 2016-09-13 12:01:28]

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

[Michael Starzinger, 2016-09-13 12:01:46]

Thanks. Addressed comments. Landing.

https://codereview.chromium.org/2331033002/diff/60001/src/interpreter/bytecod...
File src/interpreter/bytecode-array-writer.cc (right):

https://codereview.chromium.org/2331033002/diff/60001/src/interpreter/bytecod...
src/interpreter/bytecode-array-writer.cc:361:
node->set_bytecode(node->bytecode(), delta, node->operand(1));
On 2016/09/13 10:47:34, rmcilroy wrote:
> Do we ever have label's bound which don't have Bytecode::kJumpLoop? If not,
> could you remove the branch and just make this a DCHECK, otherwise, do you
know
> where they come from?

Done. Just the unit-tests, adapted the tests.

https://codereview.chromium.org/2331033002/diff/60001/src/interpreter/bytecod...
src/interpreter/bytecode-array-writer.cc:366: // The label has not yet been
bound so this is a forward reference
On 2016/09/13 10:47:34, rmcilroy wrote:
> Could you add a DCHECK that this is not Bytecode::kJumpLoop

Done.

[commit-bot: I haz the power, 2016-09-13 12:03:05]

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

[commit-bot: I haz the power, 2016-09-13 12:03:06]

Dry run: Try jobs failed on following builders:
  v8_linux64_gyp_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng/build...)
  v8_linux_mipsel_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_mipsel_compile_rel...)

[Michael Starzinger, 2016-09-13 12:11:59]

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

[commit-bot: I haz the power, 2016-09-13 12:12:05]

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

[rmcilroy, 2016-09-13 12:26:50]

LGTM, thanks!

[commit-bot: I haz the power, 2016-09-13 12:28:15]

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

[commit-bot: I haz the power, 2016-09-13 12:28:16]

Dry run: Try jobs failed on following builders:
  v8_linux_dbg_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_dbg_ng/builds/12621)
  v8_linux_dbg_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_dbg_ng_triggered/b...)

[Michael Starzinger, 2016-09-13 12:34:59]

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

[commit-bot: I haz the power, 2016-09-13 12:35:07]

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, 2016-09-13 13:04:16]

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

[commit-bot: I haz the power, 2016-09-13 13:04:17]

Dry run: This issue passed the CQ dry run.

[Michael Starzinger, 2016-09-13 13:04:24]

The CQ bit was checked by mstarzinger@chromium.org

[Michael Starzinger, 2016-09-13 13:04:25]

The patchset sent to the CQ was uploaded after l-g-t-m from
rmcilroy@chromium.org
Link to the patchset: https://codereview.chromium.org/2331033002/#ps140001
(title: "One more test.")

[commit-bot: I haz the power, 2016-09-13 13:04:36]

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, 2016-09-13 13:07:19]

Description was changed from

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performing the
backwards branch. This reduces the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
==========

to

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performing the
backwards branch. This reduces the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
==========

[commit-bot: I haz the power, 2016-09-13 13:07:20]

Committed patchset #8 (id:140001)

[commit-bot: I haz the power, 2016-09-13 13:07:41]

Description was changed from

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performing the
backwards branch. This reduces the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764
==========

to

==========
[interpreter] Merge {OsrPoll} with {Jump} bytecode.

This introduces a new {JumpLoop} bytecode to combine the OSR polling
mechanism modeled by {OsrPoll} with the actual {Jump} performing the
backwards branch. This reduces the overall size and also avoids one
additional dispatch. It also makes sure that OSR polling is only done
within real loops.

R=rmcilroy@chromium.org
BUG=v8:4764

Committed: https://crrev.com/c9864173f145c14866bcbf94759c53aa65847291
Cr-Commit-Position: refs/heads/master@{#39384}
==========

[commit-bot: I haz the power, 2016-09-13 13:07:43]

Patchset 8 (id:??) landed as
https://crrev.com/c9864173f145c14866bcbf94759c53aa65847291
Cr-Commit-Position: refs/heads/master@{#39384}