Unify the stack layout for construct frames

Unify the stack layout for construct frames

The stack layout was different for different ports.

BUG=v8:3887
LOG=N
R=dslomov@chromium.org, adamk@chromium.org

Committed: https://crrev.com/876ae425980f67f489323671f13d2314b0ce0a91
Cr-Commit-Position: refs/heads/master@{#29292}

[arv (Not doing code reviews), 2015-06-24 15:38:13]

Fix comments

[arv (Not doing code reviews), 2015-06-24 15:38:58]

PTAL

other ports on the way

[Dmitry Lomov (no reviews), 2015-06-24 15:49:48]

https://codereview.chromium.org/1203103003/diff/1/src/arm/builtins-arm.cc
File src/arm/builtins-arm.cc (right):

https://codereview.chromium.org/1203103003/diff/1/src/arm/builtins-arm.cc#new...
src/arm/builtins-arm.cc:627: __ pop(r1);  // Constructor function.
Use __ Drop(1)

[arv (Not doing code reviews), 2015-06-24 15:54:10]

https://codereview.chromium.org/1203103003/diff/1/src/arm/builtins-arm.cc
File src/arm/builtins-arm.cc (right):

https://codereview.chromium.org/1203103003/diff/1/src/arm/builtins-arm.cc#new...
src/arm/builtins-arm.cc:627: __ pop(r1);  // Constructor function.
On 2015/06/24 15:49:47, Dmitry S. Lomov wrote:
> Use __ Drop(1)

That is not an option here because we need to have the constructor function in
r1. It is used on line 672

[Dmitry Lomov (no reviews), 2015-06-24 15:58:32]

https://codereview.chromium.org/1203103003/diff/1/src/arm/builtins-arm.cc
File src/arm/builtins-arm.cc (right):

https://codereview.chromium.org/1203103003/diff/1/src/arm/builtins-arm.cc#new...
src/arm/builtins-arm.cc:627: __ pop(r1);  // Constructor function.
On 2015/06/24 15:54:10, arv wrote:
> On 2015/06/24 15:49:47, Dmitry S. Lomov wrote:
> > Use __ Drop(1)
> 
> That is not an option here because we need to have the constructor function in
> r1. It is used on line 672

Ah silly me. Thanks

[arv (Not doing code reviews), 2015-06-24 16:14:43]

arm64, mips, mips64

[arv (Not doing code reviews), 2015-06-24 16:18:55]

ports done

[arv (Not doing code reviews), 2015-06-24 16:19:12]

All done. PTAL

[arv (Not doing code reviews), 2015-06-24 16:25:49]

also ppc

[Dmitry Lomov (no reviews), 2015-06-24 16:52:19]

lgtm

[arv (Not doing code reviews), 2015-06-24 16:54:36]

This looks bad:

http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel/builds/547...

[Dmitry Lomov (no reviews), 2015-06-24 16:57:16]

On 2015/06/24 16:54:36, arv wrote:
> This looks bad:
> 
>
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel/builds/547...

Indeed. Seems to have to do with inlining?

[arv (Not doing code reviews), 2015-06-24 17:04:03]

On 2015/06/24 16:57:16, Dmitry S. Lomov wrote:
> On 2015/06/24 16:54:36, arv wrote:
> > This looks bad:
> > 
> >
>
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel/builds/547...
> 
> Indeed. Seems to have to do with inlining?

Yup. I can repro in release mode but all tests pass in debug.

Strange that it doesn't happen in arm64.

[arv (Not doing code reviews), 2015-06-24 17:05:15]

On 2015/06/24 17:04:03, arv wrote:
> On 2015/06/24 16:57:16, Dmitry S. Lomov wrote:
> > On 2015/06/24 16:54:36, arv wrote:
> > > This looks bad:
> > > 
> > >
> >
>
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel/builds/547...
> > 
> > Indeed. Seems to have to do with inlining?
> 
> Yup. I can repro in release mode but all tests pass in debug.
> 
> Strange that it doesn't happen in arm64.

I can repro in both arms in release mode now. Digging...

[arv (Not doing code reviews), 2015-06-24 19:15:57]

Fix deopt case and clean up constants

[arv (Not doing code reviews), 2015-06-24 19:16:14]

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

[arv (Not doing code reviews), 2015-06-24 19:16:15]

The patchset sent to the CQ was uploaded after l-g-t-m from dslomov@chromium.org
Link to the patchset: https://codereview.chromium.org/1203103003/#ps100001
(title: "Fix deopt case and clean up constants")

[commit-bot: I haz the power, 2015-06-24 19:16:36]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1203103003/100001

[commit-bot: I haz the power, 2015-06-24 20:27:37]

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

[commit-bot: I haz the power, 2015-06-24 20:27:38]

Dry run: This issue passed the CQ dry run.

[arv (Not doing code reviews), 2015-06-25 12:42:37]

Dmitry, can you look again or add some other reviewer?

[Dmitry Lomov (no reviews), 2015-06-25 12:43:55]

still lgtm.
Great cleanup!

[arv (Not doing code reviews), 2015-06-25 12:51:00]

The CQ bit was checked by arv@chromium.org

[arv (Not doing code reviews), 2015-06-25 12:51:08]

On 2015/06/25 12:43:55, Dmitry S. Lomov wrote:
> still lgtm.
> Great cleanup!

Thanks

[commit-bot: I haz the power, 2015-06-25 12:51:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1203103003/100001

[commit-bot: I haz the power, 2015-06-25 12:52:29]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-06-25 12:52:44]

Patchset 6 (id:??) landed as
https://crrev.com/876ae425980f67f489323671f13d2314b0ce0a91
Cr-Commit-Position: refs/heads/master@{#29292}