Bug 163734 - GC should scan stack at the end
Summary: GC should scan stack at the end
Status: RESOLVED DUPLICATE of bug 164454
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: WebKit Nightly Build
Hardware: All All
: P2 Normal
Assignee: Filip Pizlo
URL:
Keywords:
Depends on:
Blocks: 149432
  Show dependency tree / graph
 
Reported: 2016-10-20 08:58 PDT by Filip Pizlo
Modified: 2016-11-09 20:27 PST (History)
5 users (show)

See Also:

Attachments
the patch (5.56 KB, patch)
2016-11-04 13:05 PDT, Filip Pizlo 		mark.lam: review+
Details | Formatted Diff | Diff
the patch (7.81 KB, application/octet-stream)
2016-11-04 13:39 PDT, Filip Pizlo 		no flags Details
possibly better patch (11.84 KB, patch)
2016-11-04 13:40 PDT, Filip Pizlo 		no flags Details | Formatted Diff | Diff
the patch (12.01 KB, patch)
2016-11-04 14:20 PDT, Filip Pizlo 		no flags Details | Formatted Diff | Diff
possibly better patch (12.20 KB, patch)
2016-11-04 14:24 PDT, Filip Pizlo 		mark.lam: review+
Details | Formatted Diff | Diff
patch for landing (12.78 KB, patch)
2016-11-04 16:07 PDT, Filip Pizlo 		no flags Details | Formatted Diff | Diff
Show Obsolete (5) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Filip Pizlo 2016-10-20 08:58:22 PDT 
Because we will have a grey stack GC.
Comment 1 Filip Pizlo 2016-11-04 13:05:50 PDT 
Created attachment 293917 [details]
the patch
Comment 2 Mark Lam 2016-11-04 13:31:28 PDT 
Comment on attachment 293917 [details]
the patch

r=me.  Looks reasonable to me.
Comment 3 Filip Pizlo 2016-11-04 13:39:52 PDT 
Created attachment 293920 [details]
the patch
Comment 4 Filip Pizlo 2016-11-04 13:40:17 PDT 
Created attachment 293921 [details]
possibly better patch
Comment 5 Filip Pizlo 2016-11-04 13:40:41 PDT 
(In reply to comment #2)
> Comment on attachment 293917 [details]
> the patch
> 
> r=me.  Looks reasonable to me.

Sorry, it was broken. :-(  There's something subtle about CodeBlock marking, I'm still investigating the best approach.
Comment 6 Filip Pizlo 2016-11-04 14:20:40 PDT 
Created attachment 293927 [details]
the patch
Comment 7 Filip Pizlo 2016-11-04 14:24:25 PDT 
Created attachment 293928 [details]
possibly better patch

Still experimenting.
Comment 8 Mark Lam 2016-11-04 15:30:41 PDT 
Comment on attachment 293928 [details]
possibly better patch

View in context: https://bugs.webkit.org/attachment.cgi?id=293928&action=review

r=me.  Looks good to me.

> Source/JavaScriptCore/ChangeLog:18
> +        This change should not change performance or behavior. I'm still testing it.

Please replace "I'm still testing it." with test results when done if all is well.
Comment 9 Filip Pizlo 2016-11-04 15:31:31 PDT 
(In reply to comment #8)
> Comment on attachment 293928 [details]
> possibly better patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=293928&action=review
> 
> r=me.  Looks good to me.
> 
> > Source/JavaScriptCore/ChangeLog:18
> > +        This change should not change performance or behavior. I'm still testing it.
> 
> Please replace "I'm still testing it." with test results when done if all is
> well.

Right!  Will do!
Comment 10 Filip Pizlo 2016-11-04 16:07:55 PDT 
Created attachment 293949 [details]
patch for landing

Still testing it.
Comment 11 Filip Pizlo 2016-11-04 20:51:00 PDT 
OK, this is perf-neutral.
Comment 12 Filip Pizlo 2016-11-09 20:27:52 PST 

*** This bug has been marked as a duplicate of bug 164454 ***