Bug 78423 - [DFG] Misuse of WeakJSConstants in silentFillGPR code.
Summary: [DFG] Misuse of WeakJSConstants in silentFillGPR code.
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Nobody
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2012-02-11 14:36 PST by Filip Pizlo
Modified: 2012-02-13 11:34 PST (History)
4 users (show)

See Also:


Attachments
the patch (1.57 KB, patch)
2012-02-11 14:41 PST, Filip Pizlo
sam: review+
Details | Formatted Diff | Diff
the patch with test case (60.41 KB, patch)
2012-02-11 16:03 PST, Filip Pizlo
ggaren: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Filip Pizlo 2012-02-11 14:36:11 PST
Patch forthcoming.
Comment 1 Filip Pizlo 2012-02-11 14:36:19 PST
<rdar://problem/10849353>
Comment 2 Filip Pizlo 2012-02-11 14:41:48 PST
Created attachment 126647 [details]
the patch
Comment 3 Filip Pizlo 2012-02-11 14:50:01 PST
Going to attempt to write a test that repros this.  If I can't after some amount of trying then I'll land without a test.
Comment 4 Filip Pizlo 2012-02-11 16:03:59 PST
Created attachment 126655 [details]
the patch with test case
Comment 5 WebKit Review Bot 2012-02-11 16:06:45 PST
Attachment 126655 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'LayoutTests/ChangeLog', u'LayoutTests/fast..." exit_code: 1

Source/JavaScriptCore/ChangeLog:31:  Line contains tab character.  [whitespace/tab] [5]
Source/JavaScriptCore/ChangeLog:32:  Line contains tab character.  [whitespace/tab] [5]
Total errors found: 2 in 6 files


If any of these errors are false positives, please file a bug against check-webkit-style.
Comment 6 Filip Pizlo 2012-02-11 16:09:37 PST
http://trac.webkit.org/changeset/107492
Comment 7 Eric Seidel (no email) 2012-02-13 11:20:46 PST
Comment on attachment 126655 [details]
the patch with test case

View in context: https://bugs.webkit.org/attachment.cgi?id=126655&action=review

Why does this test repeat 1000 times?

> Source/JavaScriptCore/ChangeLog:20
> +	https://bugs.webkit.org/show_bug.cgi?id=78062
> +	<rdar://problem/10326439> <rdar://problem/10824839>

Are these tabs?
Comment 8 Geoffrey Garen 2012-02-13 11:29:11 PST
r=me

Please replace your tabs with spaces.

> Why does this test repeat 1000 times?

It's the best way we have to test the optimizing compiler. :(
Comment 9 Filip Pizlo 2012-02-13 11:32:34 PST
(In reply to comment #7)
> (From update of attachment 126655 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=126655&action=review
> 
> Why does this test repeat 1000 times?
> 
> > Source/JavaScriptCore/ChangeLog:20
> > +	https://bugs.webkit.org/show_bug.cgi?id=78062
> > +	<rdar://problem/10326439> <rdar://problem/10824839>
> 
> Are these tabs?

Yes. :-(  Stupid emacs!  Somehow I missed this, but fixed it in a later commit.
Comment 10 Filip Pizlo 2012-02-13 11:34:00 PST
(In reply to comment #8)
> r=me
> 
> Please replace your tabs with spaces.

Already fixed!

By the way, I landed the code with the test case (thanks to Sam's rubber stamp) in http://trac.webkit.org/changeset/107492

> 
> > Why does this test repeat 1000 times?
> 
> It's the best way we have to test the optimizing compiler. :(