Bug 51881

Summary: StackBounds initialization in WTFThreadData should be guarded by #if USE(JSC)
Product: WebKit Reporter: James Robinson <jamesr>
Component: New BugsAssignee: James Robinson <jamesr>
Status: RESOLVED FIXED    
Severity: Normal CC: abarth, barraclough, commit-queue, eric, oliver, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Other   
OS: OS X 10.5   
Attachments:
Description Flags
Patch none

Description James Robinson 2011-01-04 10:32:43 PST
StackBounds initialization in WTFThreadData should be guarded by #if USE(JSC)
Comment 1 James Robinson 2011-01-04 10:39:30 PST
Created attachment 77901 [details]
Patch
Comment 2 WebKit Commit Bot 2011-01-04 14:47:59 PST
Comment on attachment 77901 [details]
Patch

Clearing flags on attachment: 77901

Committed r75011: <http://trac.webkit.org/changeset/75011>
Comment 3 WebKit Commit Bot 2011-01-04 14:48:05 PST
All reviewed patches have been landed.  Closing bug.
Comment 4 WebKit Review Bot 2011-01-04 14:54:34 PST
http://trac.webkit.org/changeset/75011 might have broken Qt Linux Release
Comment 5 Gavin Barraclough 2011-01-04 21:26:41 PST
This seems like a slightly odd change to make.  There are a number of features of WTF that are presently only used in JSC, but we don't choose to #ifdef, and there is unsafe recursion in WebCore that could be addressed using this mechanism.  Still, this doesn't do any actual harm - we can always just remove the #ifdefs if and when we do start making use of this from WebCore.
Comment 6 James Robinson 2011-01-04 22:23:17 PST
This particular bit of code was causing problems with the Chromium linux sandbox, which motivated the change.  If we want to use this functionality elsewhere we'll have to figure out how to make those pthread_* calls from within the sandbox (or route the calls some other way for Chromium linux).  It seemed easier to just #ifdef the code out for non-JSC builds since it's not currently being used outside of JSC.
Comment 7 Gavin Barraclough 2011-01-04 22:40:54 PST
(In reply to comment #6)
> This particular bit of code was causing problems with the Chromium linux sandbox, which motivated the change.  If we want to use this functionality elsewhere we'll have to figure out how to make those pthread_* calls from within the sandbox (or route the calls some other way for Chromium linux).  It seemed easier to just #ifdef the code out for non-JSC builds since it's not currently being used outside of JSC.

Ah, makes sense now, & good to know.  If we plan on making use of this from WebCore then I'll try to make sure we give you some heads up, so you have time to look into the sandboxing issues.