[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

MIPS port contributed by Balazs Kilvady <balazs.kilvady@imgtec.com>;.

BUG=v8:4280
LOG=N

Committed: https://crrev.com/84f8a506e204b1e70df5b9a8036cc97170e661fd
Cr-Commit-Position: refs/heads/master@{#33334}

[rmcilroy, 2016-01-12 16:03:33]

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

[commit-bot: I haz the power, 2016-01-12 16:03:42]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1576093004/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1576093004/1

[commit-bot: I haz the power, 2016-01-12 16:34:24]

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

[commit-bot: I haz the power, 2016-01-12 16:34:24]

Dry run: Try jobs failed on following builders:
  v8_android_arm_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_android_arm_compile_rel/...)

[rmcilroy, 2016-01-13 10:29:20]

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

[commit-bot: I haz the power, 2016-01-13 10:29:32]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1576093004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1576093004/20001

[commit-bot: I haz the power, 2016-01-13 10:31:18]

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

[commit-bot: I haz the power, 2016-01-13 10:31:19]

Dry run: Try jobs failed on following builders:
  v8_linux_arm64_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel/builds/1...)
  v8_linux_mipsel_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_mipsel_compile_rel...)

[rmcilroy, 2016-01-13 12:34:07]

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

[commit-bot: I haz the power, 2016-01-13 12:34:11]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1576093004/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1576093004/40001

[commit-bot: I haz the power, 2016-01-13 12:37:24]

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

[commit-bot: I haz the power, 2016-01-13 12:37:24]

Dry run: Try jobs failed on following builders:
  v8_linux_mips64el_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_mips64el_compile_r...)

[rmcilroy, 2016-01-13 13:19:57]

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

[commit-bot: I haz the power, 2016-01-13 13:20:03]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1576093004/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1576093004/60001

[commit-bot: I haz the power, 2016-01-13 13:23:12]

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

[commit-bot: I haz the power, 2016-01-13 13:23:13]

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

[rmcilroy, 2016-01-13 13:43:18]

rmcilroy@chromium.org changed reviewers:
+ bmeurer@chromium.org, oth@chromium.org

[rmcilroy, 2016-01-13 13:43:19]

Benedikt, Orion, please take a look.

MIPS-ports, could you please port to MIPS

PPC/x87-ports, FYI - let me know if there is anything in this CL which might
cause you problems porting to PPC/x87.

[rmcilroy, 2016-01-13 13:43:24]

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

[commit-bot: I haz the power, 2016-01-13 13:43:29]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1576093004/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1576093004/80001

[commit-bot: I haz the power, 2016-01-13 14:45:00]

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

[commit-bot: I haz the power, 2016-01-13 14:45:00]

Dry run: Try jobs failed on following builders:
  v8_linux_mips64el_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_mips64el_compile_r...)

[Benedikt Meurer, 2016-01-13 14:48:53]

Awesome. I like it.
Can you also change JSGenericLowering to lower JSForInPrepare to the runtime
function?

https://codereview.chromium.org/1576093004/diff/80001/src/runtime/runtime-for...
File src/runtime/runtime-forin.cc (right):

https://codereview.chromium.org/1576093004/diff/80001/src/runtime/runtime-for...
src/runtime/runtime-forin.cc:55: cache_type = Handle<Object>(Smi::FromInt(0),
isolate);
This no longer matches the current implementatin fullcodegen and TurboFan.
cache_type is always Smi 1 independent of whether it's a proxy or not.

[oth, 2016-01-13 16:06:36]

lgtm.

[rmcilroy, 2016-01-14 10:04:12]

> Can you also change JSGenericLowering to lower JSForInPrepare to the runtime
? function?

If I do this, won't I have to change full-codegen too (at least change the
bailout point) so that TF->FCG still works, and if I do that, won't I have to
change Crankshaft too? I'd rather not do all this in this CL, WDYT?

[Benedikt Meurer, 2016-01-14 11:18:09]

Yeah maybe. Let's just land your CL. I'll look into the FCG interaction later.

[Benedikt Meurer, 2016-01-14 11:18:19]

Ah yeah, LGTM BTW :-)

[rmcilroy, 2016-01-14 12:13:52]

https://codereview.chromium.org/1576093004/diff/80001/src/runtime/runtime-for...
File src/runtime/runtime-forin.cc (right):

https://codereview.chromium.org/1576093004/diff/80001/src/runtime/runtime-for...
src/runtime/runtime-forin.cc:55: cache_type = Handle<Object>(Smi::FromInt(0),
isolate);
On 2016/01/13 14:48:53, Benedikt Meurer wrote:
> This no longer matches the current implementatin fullcodegen and TurboFan.
> cache_type is always Smi 1 independent of whether it's a proxy or not.

Done.

[rmcilroy, 2016-01-14 13:11:51]

> Yeah maybe. Let's just land your CL. I'll look into the FCG interaction later.

Great, thanks for looking into it! I'll land this once we have MIPS ports. For
Ignition+TF, will I land something in BytecodeGraphBuilder which directly
performs the runtime call, then replace this with the modified ForInPrepare node
once you've done the FGC / Crankshaft work, or would you prefer I create a
seperate ForInPrepareForIgnition, which lowers to that runtime call, and will
eventually replace ForInPrepare?

[balazs.kilvady, 2016-01-14 16:41:00]

We are working on the MIPS port: https://codereview.chromium.org/1589913003
It needs to be tested although.

[balazs.kilvady, 2016-01-14 20:07:54]

On 2016/01/14 16:41:00, balazs.kilvady wrote:
> We are working on the MIPS port: https://codereview.chromium.org/1589913003
> It needs to be tested although.

After testing on boards an a0 - v0 fix patchset is uploaded.

[rmcilroy, 2016-01-15 13:54:31]

Description was changed from

==========
[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

BUG=v8:4280
LOG=N
==========

to

==========
[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

MIPS port contributed by Balazs Kilvady <balazs.kilvady@imgtec.com>.

BUG=v8:4280
LOG=N
==========

[rmcilroy, 2016-01-15 14:07:51]

On 2016/01/14 20:07:54, balazs.kilvady wrote:
> On 2016/01/14 16:41:00, balazs.kilvady wrote:
> > We are working on the MIPS port: https://codereview.chromium.org/1589913003
> > It needs to be tested although.
> 
> After testing on boards an a0 - v0 fix patchset is uploaded.

Thanks for the MIPS port! I've patched it into this CL.

[rmcilroy, 2016-01-15 14:07:56]

The CQ bit was checked by rmcilroy@chromium.org

[rmcilroy, 2016-01-15 14:07:56]

The patchset sent to the CQ was uploaded after l-g-t-m from
bmeurer@chromium.org, oth@chromium.org
Link to the patchset: https://codereview.chromium.org/1576093004/#ps120001
(title: "Add MIPS port")

[commit-bot: I haz the power, 2016-01-15 14:08:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1576093004/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1576093004/120001

[commit-bot: I haz the power, 2016-01-15 14:34:41]

Description was changed from

==========
[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

MIPS port contributed by Balazs Kilvady <balazs.kilvady@imgtec.com>.

BUG=v8:4280
LOG=N
==========

to

==========
[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

MIPS port contributed by Balazs Kilvady <balazs.kilvady@imgtec.com>.

BUG=v8:4280
LOG=N
==========

[commit-bot: I haz the power, 2016-01-15 14:34:42]

Committed patchset #7 (id:120001)

[commit-bot: I haz the power, 2016-01-15 14:35:40]

Description was changed from

==========
[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

MIPS port contributed by Balazs Kilvady <balazs.kilvady@imgtec.com>.

BUG=v8:4280
LOG=N
==========

to

==========
[Interpreter] Add ForInPrepare runtime function which returns a ObjectTriple.

Adds a ForInPrepare Runtime function which returns a triple of
cache_type, cache_array and cache_length.

This requires adding support to CEntryStub to call runtime functions
which return a ObjectTriple - a struct containing three Object*
pointers. Also did some cleanup of the x64 CEntryStub to avoid
replicated code.

Replaces the interpreter's use of the ad-hock InterpreterForInPrepare
Runtime function with ForInPrepare in preparation for fixing deopt in
BytecodeGraphBuilder for ForIn (which will be done in a followup CL).

MIPS port contributed by Balazs Kilvady <balazs.kilvady@imgtec.com>.

BUG=v8:4280
LOG=N

Committed: https://crrev.com/84f8a506e204b1e70df5b9a8036cc97170e661fd
Cr-Commit-Position: refs/heads/master@{#33334}
==========

[commit-bot: I haz the power, 2016-01-15 14:35:41]

Patchset 7 (id:??) landed as
https://crrev.com/84f8a506e204b1e70df5b9a8036cc97170e661fd
Cr-Commit-Position: refs/heads/master@{#33334}