X87: [crankshaft] Fragmentation-free allocation folding.

X87: [crankshaft] Fragmentation-free allocation folding.

  port 61f5fbbb1926e1ffb672214a043e8c98db0e8af6 (r36133)

  original commit message:
  The new allocation folding implementation avoids fragmentation between folded allocation. As a consequence, our heap will always be iterable i.e. we do not have to perform a garbage collection
  before iterating the heap.

BUG=

Committed: https://crrev.com/ef778970fd754a5019bb9ad6434958f15ae0de25
Cr-Commit-Position: refs/heads/master@{#36158}

[zhengxing.li, 2016-05-11 07:14:12]

zhengxing.li@intel.com changed reviewers:
+ chunyang.dai@intel.com, hpayer@chromium.org, weiliang.lin@intel.com

[zhengxing.li, 2016-05-11 07:14:13]

PTAL, thanks!

[Weiliang, 2016-05-11 07:42:26]

lgtm

[zhengxing.li, 2016-05-11 07:43:20]

The CQ bit was checked by zhengxing.li@intel.com

[commit-bot: I haz the power, 2016-05-11 07:43:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1969553003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1969553003/1

[Hannes Payer (out of office), 2016-05-11 07:44:42]

https://codereview.chromium.org/1969553003/diff/1/src/crankshaft/x87/lithium-...
File src/crankshaft/x87/lithium-codegen-x87.cc (right):

https://codereview.chromium.org/1969553003/diff/1/src/crankshaft/x87/lithium-...
src/crankshaft/x87/lithium-codegen-x87.cc:5459: void
LCodeGen::DoFastAllocate(LFastAllocate* instr) {
Why do you move that piece of code? I would like to keep the order in sync with
the other platforms: DoAllocate, DoDeferredAllocate, DoFastAllocate.

[zhengxing.li, 2016-05-11 07:48:43]

On 2016/05/11 07:44:42, Hannes Payer wrote:
>
https://codereview.chromium.org/1969553003/diff/1/src/crankshaft/x87/lithium-...
> File src/crankshaft/x87/lithium-codegen-x87.cc (right):
> 
>
https://codereview.chromium.org/1969553003/diff/1/src/crankshaft/x87/lithium-...
> src/crankshaft/x87/lithium-codegen-x87.cc:5459: void
> LCodeGen::DoFastAllocate(LFastAllocate* instr) {
> Why do you move that piece of code? I would like to keep the order in sync
with
> the other platforms: DoAllocate, DoDeferredAllocate, DoFastAllocate.

Because I wanted the x87 source code align with ia32 source code.

Thanks!

[Hannes Payer (out of office), 2016-05-11 07:58:19]

lgtm

[commit-bot: I haz the power, 2016-05-11 08:14:53]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-05-11 08:15:16]

Description was changed from

==========
X87: [crankshaft] Fragmentation-free allocation folding.

  port 61f5fbbb1926e1ffb672214a043e8c98db0e8af6 (r36133)

  original commit message:
  The new allocation folding implementation avoids fragmentation between folded
allocation. As a consequence, our heap will always be iterable i.e. we do not
have to perform a garbage collection
  before iterating the heap.

BUG=
==========

to

==========
X87: [crankshaft] Fragmentation-free allocation folding.

  port 61f5fbbb1926e1ffb672214a043e8c98db0e8af6 (r36133)

  original commit message:
  The new allocation folding implementation avoids fragmentation between folded
allocation. As a consequence, our heap will always be iterable i.e. we do not
have to perform a garbage collection
  before iterating the heap.

BUG=

Committed: https://crrev.com/ef778970fd754a5019bb9ad6434958f15ae0de25
Cr-Commit-Position: refs/heads/master@{#36158}
==========

[commit-bot: I haz the power, 2016-05-11 08:15:18]

Patchset 1 (id:??) landed as
https://crrev.com/ef778970fd754a5019bb9ad6434958f15ae0de25
Cr-Commit-Position: refs/heads/master@{#36158}