[heap] Uncommit unused large object page memory.

[heap] Uncommit unused large object page memory.

As a first step I uncommit the memory on the main thread. Also to measure impact and stability of that optimization. In a follow-up CL, the uncommitting should be moved on the concurrent thread.

BUG=

Committed: https://crrev.com/d61a5c376ba51145dc4684e39d5d3a9ce75bcfa6
Cr-Commit-Position: refs/heads/master@{#36763}

[Hannes Payer (out of office), 2016-06-03 12:45:53]

Description was changed from

==========
[heap] Uncommit unused large object page space.

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

to

==========
[heap] Uncommit unused large object page space.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

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

[Hannes Payer (out of office), 2016-06-06 13:21:34]

Description was changed from

==========
[heap] Uncommit unused large object page space.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

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

to

==========
[heap] Uncommit unused large object page memory.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

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

[Hannes Payer (out of office), 2016-06-06 13:33:10]

hpayer@chromium.org changed reviewers:
+ ulan@chromium.org

[ulan, 2016-06-06 13:44:22]

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
File src/base/platform/platform-macos.cc (right):

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
src/base/platform/platform-macos.cc:244: return munmap(free_start, free_size) ==
0;
Don't we need to handle lsan like in linux port?

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
File src/base/platform/platform.h (right):

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
src/base/platform/platform.h:342: void ReleasePartial(void* free_start, size_t
size) {
As discussed, we can make a stricter version of this function that releases the
tail of the region.

https://codereview.chromium.org/2032393002/diff/120001/src/heap/spaces.cc
File src/heap/spaces.cc (right):

https://codereview.chromium.org/2032393002/diff/120001/src/heap/spaces.cc#new...
src/heap/spaces.cc:423: Executability executable,
Maybe completely disallow the function for executable memory?
Since we will not partially uncommit code.

[Hannes Payer (out of office), 2016-06-06 14:42:44]

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
File src/base/platform/platform-macos.cc (right):

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
src/base/platform/platform-macos.cc:244: return munmap(free_start, free_size) ==
0;
On 2016/06/06 13:44:22, ulan wrote:
> Don't we need to handle lsan like in linux port?

We only have it on linux.

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
File src/base/platform/platform.h (right):

https://codereview.chromium.org/2032393002/diff/120001/src/base/platform/plat...
src/base/platform/platform.h:342: void ReleasePartial(void* free_start, size_t
size) {
On 2016/06/06 13:44:22, ulan wrote:
> As discussed, we can make a stricter version of this function that releases
the
> tail of the region.

Done.

https://codereview.chromium.org/2032393002/diff/120001/src/heap/spaces.cc
File src/heap/spaces.cc (right):

https://codereview.chromium.org/2032393002/diff/120001/src/heap/spaces.cc#new...
src/heap/spaces.cc:423: Executability executable,
On 2016/06/06 13:44:22, ulan wrote:
> Maybe completely disallow the function for executable memory?
> Since we will not partially uncommit code.

I added checks to the lower level functions.

[ulan, 2016-06-06 15:53:16]

lgtm

https://codereview.chromium.org/2032393002/diff/140001/src/base/platform/plat...
File src/base/platform/platform.h (right):

https://codereview.chromium.org/2032393002/diff/140001/src/base/platform/plat...
src/base/platform/platform.h:340: // Releases the tail [free_start, free_start +
size] of a previously allocated
Should be [free_start, size_]

https://codereview.chromium.org/2032393002/diff/140001/src/base/platform/plat...
src/base/platform/platform.h:349: DCHECK(free_start != address_);
DCHECK_LT(address_, free_start);
DCHECK_LT(free_start, address_ + size_);

[Hannes Payer (out of office), 2016-06-06 15:56:38]

https://codereview.chromium.org/2032393002/diff/140001/src/base/platform/plat...
File src/base/platform/platform.h (right):

https://codereview.chromium.org/2032393002/diff/140001/src/base/platform/plat...
src/base/platform/platform.h:340: // Releases the tail [free_start, free_start +
size] of a previously allocated
On 2016/06/06 15:53:15, ulan wrote:
> Should be [free_start, size_]

Done.

https://codereview.chromium.org/2032393002/diff/140001/src/base/platform/plat...
src/base/platform/platform.h:349: DCHECK(free_start != address_);
On 2016/06/06 15:53:15, ulan wrote:
> DCHECK_LT(address_, free_start);
> DCHECK_LT(free_start, address_ + size_);

Done.

[Hannes Payer (out of office), 2016-06-06 15:57:24]

The CQ bit was checked by hpayer@chromium.org

[Hannes Payer (out of office), 2016-06-06 15:57:25]

The patchset sent to the CQ was uploaded after l-g-t-m from ulan@chromium.org
Link to the patchset: https://codereview.chromium.org/2032393002/#ps190001
(title: " ")

[commit-bot: I haz the power, 2016-06-06 15:57:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2032393002/190001

[commit-bot: I haz the power, 2016-06-06 16:00:17]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-06-06 16:00:18]

Try jobs failed on following builders:
  v8_linux64_avx2_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_avx2_rel_ng/buil...)
  v8_linux_arm64_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel_ng/build...)
  v8_linux_arm_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel_ng/builds/...)
  v8_mac_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel_ng/builds/2708)

[Hannes Payer (out of office), 2016-06-06 16:06:49]

The CQ bit was checked by hpayer@chromium.org

[Hannes Payer (out of office), 2016-06-06 16:06:50]

The patchset sent to the CQ was uploaded after l-g-t-m from ulan@chromium.org
Link to the patchset: https://codereview.chromium.org/2032393002/#ps210001
(title: " ")

[commit-bot: I haz the power, 2016-06-06 16:06:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2032393002/210001

[commit-bot: I haz the power, 2016-06-06 16:30:30]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-06-06 16:30:31]

Try jobs failed on following builders:
  v8_win_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_rel_ng/builds/8251)
  v8_win_rel_ng_triggered on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_rel_ng_triggered/bui...)

[Hannes Payer (out of office), 2016-06-06 16:49:26]

The CQ bit was checked by hpayer@chromium.org

[Hannes Payer (out of office), 2016-06-06 16:49:30]

The patchset sent to the CQ was uploaded after l-g-t-m from ulan@chromium.org
Link to the patchset: https://codereview.chromium.org/2032393002/#ps230001
(title: " ")

[commit-bot: I haz the power, 2016-06-06 16:49:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2032393002/230001

[commit-bot: I haz the power, 2016-06-06 17:17:52]

Description was changed from

==========
[heap] Uncommit unused large object page memory.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

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

to

==========
[heap] Uncommit unused large object page memory.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

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

[commit-bot: I haz the power, 2016-06-06 17:17:53]

Committed patchset #13 (id:230001)

[commit-bot: I haz the power, 2016-06-06 17:19:37]

Description was changed from

==========
[heap] Uncommit unused large object page memory.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

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

to

==========
[heap] Uncommit unused large object page memory.

As a first step I uncommit the memory on the main thread. Also to measure impact
and stability of that optimization. In a follow-up CL, the uncommitting should
be moved on the concurrent thread.

BUG=

Committed: https://crrev.com/d61a5c376ba51145dc4684e39d5d3a9ce75bcfa6
Cr-Commit-Position: refs/heads/master@{#36763}
==========

[commit-bot: I haz the power, 2016-06-06 17:19:39]

Patchset 13 (id:??) landed as
https://crrev.com/d61a5c376ba51145dc4684e39d5d3a9ce75bcfa6
Cr-Commit-Position: refs/heads/master@{#36763}

[Hannes Payer (out of office), 2016-06-08 06:12:39]

A revert of this CL (patchset #13 id:230001) has been created in
https://codereview.chromium.org/2043263002/ by hpayer@chromium.org.

The reason for reverting is: Needs fixing of slot set..