Bug 126478 - CStack Branch: slowPathFor() should emit an epilogue instead of popping the callFrameRegister
Summary: CStack Branch: slowPathFor() should emit an epilogue instead of popping the c...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Michael Saboff
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-01-03 17:45 PST by Michael Saboff
Modified: 2014-01-04 10:26 PST (History)
0 users

See Also:


Attachments
Patch (1.29 KB, patch)
2014-01-03 17:51 PST, Michael Saboff
mark.lam: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Michael Saboff 2014-01-03 17:45:05 PST
Before ThunkGenerators::slowPathFor() jumps to the target function, it returns the stack to the function entry state.  Currently it does this using a pop(callFrameRegister).  Instead it should emit a function epilogue().  This will work for architectures that have a link register that the callee will want to save in its prologue in addition to the callFrameRegister.
Comment 1 Michael Saboff 2014-01-03 17:51:30 PST
Created attachment 220360 [details]
Patch
Comment 2 Mark Lam 2014-01-04 05:46:31 PST
Comment on attachment 220360 [details]
Patch

r=me
Comment 3 Michael Saboff 2014-01-04 10:26:51 PST
Committed r161312: <http://trac.webkit.org/changeset/161312>