Bug 95834 - Web Inspector: NMI: extract overloaded instrumentation members for WebCore classes from core NMI code.
Summary: Web Inspector: NMI: extract overloaded instrumentation members for WebCore cl...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Web Inspector (Deprecated) (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Ilya Tikhonovsky
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-09-05 02:52 PDT by Ilya Tikhonovsky
Modified: 2012-09-05 07:52 PDT (History)
11 users (show)

See Also:


Attachments
Patch (31.73 KB, patch)
2012-09-05 02:59 PDT, Ilya Tikhonovsky
no flags Details | Formatted Diff | Diff
patch for mac try bot (31.73 KB, patch)
2012-09-05 04:45 PDT, Ilya Tikhonovsky
no flags Details | Formatted Diff | Diff
another patch for mac try bot (31.72 KB, patch)
2012-09-05 05:28 PDT, Ilya Tikhonovsky
no flags Details | Formatted Diff | Diff
Patch (39.42 KB, patch)
2012-09-05 07:23 PDT, Ilya Tikhonovsky
yurys: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Ilya Tikhonovsky 2012-09-05 02:52:08 PDT
I'd like to upstream core memory instrumentation code to WTF.
This will give us a chance to implement reportMemoryUsage members for containers.
On this way I have to remove WebCore specific members from MemoryInstrumentation class.
Comment 1 Ilya Tikhonovsky 2012-09-05 02:59:29 PDT
Created attachment 162200 [details]
Patch
Comment 2 Build Bot 2012-09-05 03:28:52 PDT
Comment on attachment 162200 [details]
Patch

Attachment 162200 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/13765140
Comment 3 Ilya Tikhonovsky 2012-09-05 04:45:49 PDT
Created attachment 162219 [details]
patch for mac try bot
Comment 4 Ilya Tikhonovsky 2012-09-05 04:48:20 PDT
(In reply to comment #2)
> (From update of attachment 162200 [details])
> Attachment 162200 [details] did not pass mac-ews (mac):
> Output: http://queues.webkit.org/results/13765140

looks like gcc bug http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15867
Comment 5 Build Bot 2012-09-05 05:17:11 PDT
Comment on attachment 162219 [details]
patch for mac try bot

Attachment 162219 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/13741894
Comment 6 Ilya Tikhonovsky 2012-09-05 05:28:26 PDT
Created attachment 162225 [details]
another patch for mac try bot
Comment 7 Yury Semikhatsky 2012-09-05 07:08:53 PDT
Comment on attachment 162225 [details]
another patch for mac try bot

View in context: https://bugs.webkit.org/attachment.cgi?id=162225&action=review

> Source/WebCore/dom/MemoryInstrumentation.cpp:55
> +template <> void MemoryInstrumentationTraits::addInstrumentedObject<String>(MemoryInstrumentation* instrumentation, const String* const& string, MemoryObjectType ownerObjectType, MemoryOwningType owningType)

template <> -> template<> here and in other places.

> Source/WebCore/dom/MemoryInstrumentation.h:54
> +class MemoryInstrumentationTraits {

Please add an explanation(at least in the ChangeLog) why we cannot use template class here.
Comment 8 Ilya Tikhonovsky 2012-09-05 07:23:39 PDT
Created attachment 162241 [details]
Patch
Comment 9 Yury Semikhatsky 2012-09-05 07:49:35 PDT
Comment on attachment 162241 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=162241&action=review

> Source/WebCore/dom/MemoryInstrumentation.h:393
> +template<> void MemoryInstrumentationTraits::addInstrumentedObject<const KURL>(MemoryInstrumentation*, const KURL* const&, MemoryObjectType, MemoryOwningType);

Why do we need overload for T in addition to const T ?
Comment 10 Ilya Tikhonovsky 2012-09-05 07:52:26 PDT
Committed r127592: <http://trac.webkit.org/changeset/127592>