Oilpan: Don't use guard pages in NaCl

Oilpan: Don't use guard pages in NaCl

NaCl's system page size is 64 KB. This causes a problem in Oilpan's heap layout
because Oilpan allocates two guard pages for each blink page (whose size is 128 KB).
After a couple of discussions, we decided not to use guard pages in NaCl.

This CL adds a flag to change the guard page size to 0.
It will be used by NaCl in a follow-up CL.

BUG=

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=200620

[haraken, 2015-08-16 13:33:39]

haraken@chromium.org changed reviewers:
+ keishi@chromium.org, nya@chromium.org, oilpan-reviews@chromium.org

[haraken, 2015-08-16 13:33:39]

PTAL

[sof, 2015-08-16 16:41:56]

sigbjornf@opera.com changed reviewers:
+ sigbjornf@opera.com

[sof, 2015-08-16 16:41:56]

lgtm

https://codereview.chromium.org/1297873002/diff/20001/Source/platform/heap/He...
File Source/platform/heap/Heap.h (right):

https://codereview.chromium.org/1297873002/diff/20001/Source/platform/heap/He...
Source/platform/heap/Heap.h:61: #if 0
will be adjusted/fixed before landing, presumably?

https://codereview.chromium.org/1297873002/diff/20001/Source/platform/heap/He...
Source/platform/heap/Heap.h:64: // (whose size is 128 KB). After a couple of
discussions, we decided not
Just saying that the address space expenditure would be too large, would do, I
think.

btw, guard page usage like this isn't entirely novel - has it been disabled for
other codebases that use guard pages & target NaCl also?

https://codereview.chromium.org/1297873002/diff/20001/Source/platform/heap/He...
Source/platform/heap/Heap.h:814: // pages are aligned at blinkPageBase plus an
OS page size.
"plus an OS page size" => plus the size of a guard page ?

https://codereview.chromium.org/1297873002/diff/20001/Source/platform/heap/Pa...
File Source/platform/heap/PageMemory.h (right):

https://codereview.chromium.org/1297873002/diff/20001/Source/platform/heap/Pa...
Source/platform/heap/PageMemory.h:207: // Setup the payload one OS page into the
page memory. The
nit: could you adjust the comment a bit to reflect what will now happen?

[haraken, 2015-08-16 23:59:56]

Thanks for review! Comments addressed.

[haraken, 2015-08-17 00:00:04]

The CQ bit was checked by haraken@chromium.org

[haraken, 2015-08-17 00:00:05]

The patchset sent to the CQ was uploaded after l-g-t-m from sigbjornf@opera.com
Link to the patchset: https://codereview.chromium.org/1297873002/#ps40001
(title: " ")

[commit-bot: I haz the power, 2015-08-17 00:00:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1297873002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1297873002/40001

[commit-bot: I haz the power, 2015-08-17 01:56:16]

Committed patchset #3 (id:40001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=200620