PPC: [crankshaft] Fragmentation-free allocation folding.

PPC: [crankshaft] Fragmentation-free allocation folding.

Port 61f5fbbb1926e1ffb672214a043e8c98db0e8af6

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.

R=hpayer@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com, mbrandy@us.ibm.com

BUG=chromium:580959
LOG=N

Committed: https://crrev.com/ce7d7761ac513a5e23efef260b74b58b955c5bbb
Cr-Commit-Position: refs/heads/master@{#36184}

[JaideepBajwa, 2016-05-11 01:40:10]

PTAL

[Hannes Payer (out of office), 2016-05-11 07:41:18]

lgtm

[MTBrandyberry, 2016-05-11 15:18:22]

A few comments.  You probably should include the changes for
6e15433db4df53bdb6d7e84fb7eba544ef6097f3 here as well.

https://codereview.chromium.org/1970633002/diff/1/src/crankshaft/ppc/lithium-...
File src/crankshaft/ppc/lithium-codegen-ppc.cc (right):

https://codereview.chromium.org/1970633002/diff/1/src/crankshaft/ppc/lithium-...
src/crankshaft/ppc/lithium-codegen-ppc.cc:5490: CHECK(size <=
Page::kMaxRegularHeapObjectSize);
Why omit this check for 64-bit?

https://codereview.chromium.org/1970633002/diff/1/src/ppc/macro-assembler-ppc.cc
File src/ppc/macro-assembler-ppc.cc (right):

https://codereview.chromium.org/1970633002/diff/1/src/ppc/macro-assembler-ppc...
src/ppc/macro-assembler-ppc.cc:2084: if (is_int16(object_size)) {
 I think you just want Add here.

[JaideepBajwa, 2016-05-11 15:54:05]

On 2016/05/11 15:18:22, MTBrandyberry wrote:
> A few comments.  You probably should include the changes for
> 6e15433db4df53bdb6d7e84fb7eba544ef6097f3 here as well.
> 
>
https://codereview.chromium.org/1970633002/diff/1/src/crankshaft/ppc/lithium-...
> File src/crankshaft/ppc/lithium-codegen-ppc.cc (right):
> 
>
https://codereview.chromium.org/1970633002/diff/1/src/crankshaft/ppc/lithium-...
> src/crankshaft/ppc/lithium-codegen-ppc.cc:5490: CHECK(size <=
> Page::kMaxRegularHeapObjectSize);
> Why omit this check for 64-bit?
> 
>
https://codereview.chromium.org/1970633002/diff/1/src/ppc/macro-assembler-ppc.cc
> File src/ppc/macro-assembler-ppc.cc (right):
> 
>
https://codereview.chromium.org/1970633002/diff/1/src/ppc/macro-assembler-ppc...
> src/ppc/macro-assembler-ppc.cc:2084: if (is_int16(object_size)) {
>  I think you just want Add here.

I have the patch ready for 6e15433db4df53bdb6d7e84fb7eba544ef6097f3, going to
put it up for review.
I've removed the 64-bit guard and updated addi to Add.

[MTBrandyberry, 2016-05-11 16:07:56]

lgtm

[JaideepBajwa, 2016-05-11 16:08:22]

The CQ bit was checked by bjaideep@ca.ibm.com

[JaideepBajwa, 2016-05-11 16:08:22]

The patchset sent to the CQ was uploaded after l-g-t-m from hpayer@chromium.org
Link to the patchset: https://codereview.chromium.org/1970633002/#ps40001
(title: "updated the if-else block")

[commit-bot: I haz the power, 2016-05-11 16:08:31]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1970633002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1970633002/40001

[commit-bot: I haz the power, 2016-05-11 16:57:53]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-05-11 17:00:28]

Description was changed from

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

Port 61f5fbbb1926e1ffb672214a043e8c98db0e8af6

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.

R=hpayer@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com,
michael_dawson@ca.ibm.com, mbrandy@us.ibm.com

BUG=chromium:580959
LOG=N
==========

to

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

Port 61f5fbbb1926e1ffb672214a043e8c98db0e8af6

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.

R=hpayer@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com,
michael_dawson@ca.ibm.com, mbrandy@us.ibm.com

BUG=chromium:580959
LOG=N

Committed: https://crrev.com/ce7d7761ac513a5e23efef260b74b58b955c5bbb
Cr-Commit-Position: refs/heads/master@{#36184}
==========

[commit-bot: I haz the power, 2016-05-11 17:00:29]

Patchset 3 (id:??) landed as
https://crrev.com/ce7d7761ac513a5e23efef260b74b58b955c5bbb
Cr-Commit-Position: refs/heads/master@{#36184}