Bug 159688

Summary:

[Cocoa] Simulated memory warning doesn't trigger libcache purge.

Product:

WebKit

Reporter:

Andreas Kling <kling>

Component:

WebCore Misc.

Assignee:

Andreas Kling <kling>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

cdumez, commit-queue, kling

Priority:

Keywords:

InRadar, Performance

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

Attachments:

Description	Flags
Patch	cdumez: review+
Patch for landing	none

Andreas Kling

Reported 2016-07-12 14:15:08 PDT

We're not poking libcache to purge caches in response to fake memory pressure.

Attachments
Patch (2.97 KB, patch) 2016-07-12 14:16 PDT, Andreas Kling	cdumez: review+	Details Formatted Diff Diff
Patch for landing (2.93 KB, patch) 2016-07-12 20:36 PDT, Andreas Kling	no flags	Details Formatted Diff Diff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.

Andreas Kling

Comment 1 2016-07-12 14:16:47 PDT

Created attachment 283446 [details] Patch

Alexey Proskuryakov

Comment 2 2016-07-12 16:58:27 PDT

Comment on attachment 283446 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=283446&action=review > Source/WebCore/platform/MemoryPressureHandler.h:146 > + bool m_isFakeMemoryPressure { false }; What is the official name, "fake" or "simulated"?

Chris Dumez

Comment 3 2016-07-12 16:59:05 PDT

Comment on attachment 283446 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=283446&action=review >> Source/WebCore/platform/MemoryPressureHandler.h:146 >> + bool m_isFakeMemoryPressure { false }; > > What is the official name, "fake" or "simulated"? +1 for simulated

Andreas Kling

Comment 4 2016-07-12 17:04:18 PDT

"simulated" sounds nice, let's call it that instead!

Chris Dumez

Comment 5 2016-07-12 18:27:49 PDT

Comment on attachment 283446 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=283446&action=review R=me with comments. >>> Source/WebCore/platform/MemoryPressureHandler.h:146 >>> + bool m_isFakeMemoryPressure { false }; >> >> What is the official name, "fake" or "simulated"? > > +1 for simulated Let's rename then. > Source/WebCore/platform/cocoa/MemoryPressureHandlerCocoa.mm:142 > + m_isFakeMemoryPressure = true; Can we use WTF TemporaryChange?

Andreas Kling

Comment 6 2016-07-12 19:52:18 PDT

(In reply to comment #5) > Comment on attachment 283446 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=283446&action=review > > R=me with comments. Thanks! > > Source/WebCore/platform/cocoa/MemoryPressureHandlerCocoa.mm:142 > > + m_isFakeMemoryPressure = true; > > Can we use WTF TemporaryChange? Unfortunately not, since the reset to false happens in a block that gets called on next runloop.

Andreas Kling

Comment 7 2016-07-12 20:36:19 PDT

Created attachment 283487 [details] Patch for landing

WebKit Commit Bot

Comment 8 2016-07-12 21:05:52 PDT

Comment on attachment 283487 [details] Patch for landing Clearing flags on attachment: 283487 Committed r203149: <http://trac.webkit.org/changeset/203149>

WebKit Commit Bot

Comment 9 2016-07-12 21:05:56 PDT

All reviewed patches have been landed. Closing bug.

Andreas Kling

Comment 10 2016-07-13 14:41:58 PDT

<rdar://problem/25479969>

Note You need to log in before you can comment on or make changes to this bug.