Pretenure ASCII cons string in high promotion mode.

Pretenure ASCII cons string in high promotion mode.


BUG=

Committed: https://code.google.com/p/v8/source/detail?r=14540

[Hannes Payer (out of office), 2013-05-03 09:17:55]

after discussing offline with mstarzinger I added a GetPretenureMode() function

[Michael Starzinger, 2013-05-03 09:33:10]

First round of comments. Second round for the new patch set coming up.

https://codereview.chromium.org/14451003/diff/13001/src/assembler.cc
File src/assembler.cc (right):

https://codereview.chromium.org/14451003/diff/13001/src/assembler.cc#newcode1208
src/assembler.cc:1208: return ExternalReference(isolate->heap()->
nit: Use same line-break style as in the other function. Easier to read.

https://codereview.chromium.org/14451003/diff/13001/src/runtime.cc
File src/runtime.cc (right):

https://codereview.chromium.org/14451003/diff/13001/src/runtime.cc#newcode239
src/runtime.cc:239: boilerplate = isolate->factory()->NewJSObjectFromMap(map,
TENURED);
Can we add a "Heap::GetGlobalPretenureFlag()" helper that returns the correct
PretenureFlag depending on the mode. This simplifies all of the call sites in
this file.

[Hannes Payer (out of office), 2013-05-03 09:35:43]

https://codereview.chromium.org/14451003/diff/13001/src/assembler.cc
File src/assembler.cc (right):

https://codereview.chromium.org/14451003/diff/13001/src/assembler.cc#newcode1208
src/assembler.cc:1208: return ExternalReference(isolate->heap()->
On 2013/05/03 09:33:10, Michael Starzinger wrote:
> nit: Use same line-break style as in the other function. Easier to read.

Done.

https://codereview.chromium.org/14451003/diff/13001/src/runtime.cc
File src/runtime.cc (right):

https://codereview.chromium.org/14451003/diff/13001/src/runtime.cc#newcode239
src/runtime.cc:239: boilerplate = isolate->factory()->NewJSObjectFromMap(map,
TENURED);
On 2013/05/03 09:33:10, Michael Starzinger wrote:
> Can we add a "Heap::GetGlobalPretenureFlag()" helper that returns the correct
> PretenureFlag depending on the mode. This simplifies all of the call sites in
> this file.

Done.

[Michael Starzinger, 2013-05-03 09:58:21]

LGTM with a final round of comments.

https://codereview.chromium.org/14451003/diff/20001/src/heap.h
File src/heap.h (right):

https://codereview.chromium.org/14451003/diff/20001/src/heap.h#newcode1988
src/heap.h:1988: intptr_t new_space_high_promotion_mode_active_;
Can we add a TODO here that this is only "intptr_t" because it is accessed from
generated code. Once we no longer access it from generated code it should be
changed back to "bool".

https://codereview.chromium.org/14451003/diff/20001/src/ia32/code-stubs-ia32.cc
File src/ia32/code-stubs-ia32.cc (right):

https://codereview.chromium.org/14451003/diff/20001/src/ia32/code-stubs-ia32....
src/ia32/code-stubs-ia32.cc:7403: { REG(ecx), REG(edx), REG(ebx),
EMIT_REMEMBERED_SET},
The ecx/edx/ebx stub seems to be duplicated.

[Hannes Payer (out of office), 2013-05-03 10:27:32]

https://codereview.chromium.org/14451003/diff/20001/src/heap.h
File src/heap.h (right):

https://codereview.chromium.org/14451003/diff/20001/src/heap.h#newcode1988
src/heap.h:1988: intptr_t new_space_high_promotion_mode_active_;
On 2013/05/03 09:58:21, Michael Starzinger wrote:
> Can we add a TODO here that this is only "intptr_t" because it is accessed
from
> generated code. Once we no longer access it from generated code it should be
> changed back to "bool".

Done.

https://codereview.chromium.org/14451003/diff/20001/src/ia32/code-stubs-ia32.cc
File src/ia32/code-stubs-ia32.cc (right):

https://codereview.chromium.org/14451003/diff/20001/src/ia32/code-stubs-ia32....
src/ia32/code-stubs-ia32.cc:7403: { REG(ecx), REG(edx), REG(ebx),
EMIT_REMEMBERED_SET},
On 2013/05/03 09:58:21, Michael Starzinger wrote:
> The ecx/edx/ebx stub seems to be duplicated.

Done.

[Hannes Payer (out of office), 2013-05-03 10:36:28]

Committed patchset #8 manually as r14540 (presubmit successful).