Bug 163766

Summary: bmalloc api should crash on failure to allocate when !isBmallocEnabled.
Product: WebKit Reporter: Mark Lam <mark.lam>
Component: bmallocAssignee: Mark Lam <mark.lam>
Status: RESOLVED FIXED    
Severity: Normal CC: ggaren
Priority: P2    
Version: WebKit Local Build   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
proposed patch. keith_miller: review+

Mark Lam
Reported 2016-10-20 16:43:51 PDT
Patch coming.
Attachments
proposed patch. (2.11 KB, patch)
2016-10-20 16:55 PDT, Mark Lam 		keith_miller: review+
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Mark Lam
Comment 1 2016-10-20 16:55:03 PDT
Created attachment 292285 [details] proposed patch.
Keith Miller
Comment 2 2016-10-20 17:00:23 PDT
Comment on attachment 292285 [details] proposed patch. View in context: https://bugs.webkit.org/attachment.cgi?id=292285&action=review r=me with comments > Source/bmalloc/bmalloc/Allocator.cpp:107 > + auto result = realloc(object, newSize); nit: auto seems a little unnecessary here > Source/bmalloc/bmalloc/Allocator.cpp:197 > + auto result = malloc(size); nit: ditto
Filip Pizlo
Comment 3 2016-10-20 17:00:41 PDT
Comment on attachment 292285 [details] proposed patch. View in context: https://bugs.webkit.org/attachment.cgi?id=292285&action=review R=me. I'm assuming that there's a reason why we need to crash. > Source/bmalloc/ChangeLog:3 > + bmalloc api should crash on failure to allocate when GuardMalloc is used. I recommend changing the title or adding a clarification that the crash-on-failure is for the !isBmallocEnabled case, which is broader than the GuardMalloc case.
Mark Lam
Comment 4 2016-10-20 17:19:37 PDT
Thanks for the review. I've changed the autos to void *s. I've also changed the title, and added a comment in the ChangeLog to explain why crashing in bmalloc is what we want when we fail to allocate. Landed in r207646: <http://trac.webkit.org/r207646>.
Note You need to log in before you can comment on or make changes to this bug.