Use PartitionAlloc for fastMalloc on Android

Use PartitionAlloc for fastMalloc on Android

This switches fastMalloc to PartitionAlloc for Android. Previously, we only
used PartitionAlloc for fastMalloc on Mac OS X.

According to primiano's measurements, this CL should improve PLT performance
on Android by 7%.  See referenced bug for details.

alexa_us on N10 with default alloc (without
#     ms/page  PSS   Dirty
#1    227.6   60410 44904
#2    226.1   61547 46008
#3    222.4   55595 40072
#4    224.2   57293 41728
AVG   225.1   58711 43178

alexa_us on N10 with partalloc
#     ms/page  PSS   Dirty
#1    204.68  54862  39456
#2    208.73  59240  43844
#3    208.57  55164  39788
#4    217.28  58280  42904
AVG   209.81  56886  41498

BUG=312725
R=tonyg@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=166271

[abarth-chromium, 2014-01-29 06:05:18]

tonyg suggests that we try landing this CL and seeing how it impacts the perf
bots.  @tonyg: Would you be willing to approve the CL and send it to the commit
queue?

[tonyg, 2014-01-29 06:08:42]

lgtm

We can check in on this tomorrow to see how things are looking.

[commit-bot: I haz the power, 2014-01-29 06:09:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/abarth@chromium.org/148013010/20001

[commit-bot: I haz the power, 2014-01-29 12:30:42]

Retried try job too often on linux_blink for step(s) webkit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_blin...

[commit-bot: I haz the power, 2014-01-29 12:30:53]

CQ bit was unchecked on CL. Ignoring.

[commit-bot: I haz the power, 2014-01-29 16:15:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/abarth@chromium.org/148013010/20001

[abarth-chromium, 2014-01-29 18:19:26]

Committed patchset #2 manually as r166040 (presubmit successful).

[commit-bot: I haz the power, 2014-01-29 18:22:23]

CQ bit was unchecked on CL. Ignoring.

[tonyg, 2014-01-29 18:51:15]

[+andrewhayden, current perf sheriff]

Andrew, we expect a variety of performance and memory changes from this patch on
Android. Please try to group them all in one bug and we'll examine in a day or
two to decide whether this change is a keeper or not.

[dglazkov, 2014-01-29 20:51:14]

I think this may have broken android?

ASSERTION FAILED: isPointerTypeAlignmentOkay(reinterpret_cast<TypePtr>(ptr))
../../third_party/WebKit/Source/wtf/StdLibExtras.h(88) : TypePtr
reinterpret_cast_ptr(void*) [with TypePtr = WebCore::CSSParserValue*]

http://build.chromium.org/p/chromium.webkit/builders/Android%20Tests%20%28dbg...

[abarth-chromium, 2014-02-02 18:39:06]

The CQ bit was checked by abarth@chromium.org

[commit-bot: I haz the power, 2014-02-02 18:39:19]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/abarth@chromium.org/148013010/20001

[abarth-chromium, 2014-02-02 18:42:08]

The CQ bit was unchecked by abarth@chromium.org

[commit-bot: I haz the power, 2014-02-02 18:42:14]

CQ bit was unchecked on CL. Ignoring.

[commit-bot: I haz the power, 2014-02-02 18:42:16]

CQ bit was unchecked on CL. Ignoring.

[commit-bot: I haz the power, 2014-02-02 18:42:18]

CQ bit was unchecked on CL. Ignoring.

[abarth-chromium, 2014-02-02 18:46:32]

Committed patchset #2 manually as r166271 (presubmit successful).

[Chris Evans, 2014-02-03 03:43:46]

On 2014/02/02 18:46:32, abarth wrote:
> Committed patchset #2 manually as r166271 (presubmit successful).

Ah thanks for landing it again! I was about to look into it myself :)