Support for super assignments in for..in.

Support for super assignments in for..in.

R=ishell@chromium.org, arv@chromium.org
BUG=v8:3330
LOG=N

Committed: https://code.google.com/p/v8/source/detail?r=24560

[Dmitry Lomov (no reviews), 2014-10-09 14:50:44]

One missing case for assignment.
PTAL, no platform ports yet.

[Dmitry Lomov (no reviews), 2014-10-09 15:22:55]

Now with platform ports.

[Igor Sheludko, 2014-10-13 11:39:08]

lgtm with notes:

https://codereview.chromium.org/639243003/diff/50001/src/arm/full-codegen-arm.cc
File src/arm/full-codegen-arm.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/arm/full-codegen-arm...
src/arm/full-codegen-arm.cc:2545: __ ldr(scratch2, MemOperand(sp,
kPointerSize));  // value
I think you could load value directly to r0 and use scratch2 below instead of
r0. (minus one instruction)

https://codereview.chromium.org/639243003/diff/50001/src/arm64/full-codegen-a...
File src/arm64/full-codegen-arm64.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/arm64/full-codegen-a...
src/arm64/full-codegen-arm64.cc:2205: __ Peek(scratch2, kPointerSize);     //
value
Same note here as for arm.

https://codereview.chromium.org/639243003/diff/50001/src/ia32/full-codegen-ia...
File src/ia32/full-codegen-ia32.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/ia32/full-codegen-ia...
src/ia32/full-codegen-ia32.cc:2461: __ mov(scratch2, MemOperand(esp,
kPointerSize));  // value
Same note here.

https://codereview.chromium.org/639243003/diff/50001/src/x64/full-codegen-x64.cc
File src/x64/full-codegen-x64.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/x64/full-codegen-x64...
src/x64/full-codegen-x64.cc:2460: __ movp(scratch2, MemOperand(rsp,
kPointerSize));  // value
And here.

https://codereview.chromium.org/639243003/diff/50001/test/mjsunit/harmony/sup...
File test/mjsunit/harmony/super.js (right):

https://codereview.chromium.org/639243003/diff/50001/test/mjsunit/harmony/sup...
test/mjsunit/harmony/super.js:710: 
Does it make sense to test keyed access via object with toString() method like
you did for ordinary keyed stores to super?

[Dmitry Lomov (no reviews), 2014-10-13 12:39:51]

Comments addressed, landing.

https://codereview.chromium.org/639243003/diff/50001/src/arm/full-codegen-arm.cc
File src/arm/full-codegen-arm.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/arm/full-codegen-arm...
src/arm/full-codegen-arm.cc:2545: __ ldr(scratch2, MemOperand(sp,
kPointerSize));  // value
On 2014/10/13 11:39:07, Igor Sheludko wrote:
> I think you could load value directly to r0 and use scratch2 below instead of
> r0. (minus one instruction)

Great suggestion, done

https://codereview.chromium.org/639243003/diff/50001/src/arm64/full-codegen-a...
File src/arm64/full-codegen-arm64.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/arm64/full-codegen-a...
src/arm64/full-codegen-arm64.cc:2205: __ Peek(scratch2, kPointerSize);     //
value
On 2014/10/13 11:39:07, Igor Sheludko wrote:
> Same note here as for arm.

Done.

https://codereview.chromium.org/639243003/diff/50001/src/ia32/full-codegen-ia...
File src/ia32/full-codegen-ia32.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/ia32/full-codegen-ia...
src/ia32/full-codegen-ia32.cc:2461: __ mov(scratch2, MemOperand(esp,
kPointerSize));  // value
On 2014/10/13 11:39:07, Igor Sheludko wrote:
> Same note here.

Done.

https://codereview.chromium.org/639243003/diff/50001/src/x64/full-codegen-x64.cc
File src/x64/full-codegen-x64.cc (right):

https://codereview.chromium.org/639243003/diff/50001/src/x64/full-codegen-x64...
src/x64/full-codegen-x64.cc:2460: __ movp(scratch2, MemOperand(rsp,
kPointerSize));  // value
On 2014/10/13 11:39:07, Igor Sheludko wrote:
> And here.

Done.

https://codereview.chromium.org/639243003/diff/50001/test/mjsunit/harmony/sup...
File test/mjsunit/harmony/super.js (right):

https://codereview.chromium.org/639243003/diff/50001/test/mjsunit/harmony/sup...
test/mjsunit/harmony/super.js:710: 
On 2014/10/13 11:39:07, Igor Sheludko wrote:
> Does it make sense to test keyed access via object with toString() method like
> you did for ordinary keyed stores to super?

Done.

[Dmitry Lomov (no reviews), 2014-10-13 12:46:57]

Committed patchset #4 (id:140001) manually as 24560 (presubmit successful).

[arv (Not doing code reviews), 2014-10-14 15:33:53]

LGTM