[Webkit-unassigned] [Bug 104807] Adds support for fromCharCode intrinsic

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=104807





--- Comment #25 from Filip Pizlo <fpizlo at apple.com>  2013-04-08 14:57:07 PST ---
(In reply to comment #24)
> (In reply to comment #19)
> > (In reply to comment #17)
> > > (From update of attachment 196949 [details] [details] [details])
> > > View in context: https://bugs.webkit.org/attachment.cgi?id=196949&action=review
> > > 
> > > > Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp:2109
> > > > +    JITCompiler::Jump jmpAfter = m_jit.jump();
> > > > +
> > > > +    isNotASCIICharacter.link(&m_jit);
> > > > +    JITCompiler::Jump jmpSlowCase = m_jit.jump();
> > > > +    addSlowPathGenerator(slowPathCall(jmpSlowCase, this, operationStringFromCharCode, scratchReg, propertyReg));
> > > > +
> > > > +    jmpAfter.link(&m_jit);
> > > 
> > > This is wrong.  You should be passing isNotASCIICharacter as the first argument to slowPathCall(), and omitting the jmpAfter thing.
> > > 
> > > > Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp:2110
> > > > +    speculationCheck(Uncountable, JSValueRegs(), 0, m_jit.branchTest32(MacroAssembler::Zero, scratchReg));
> > > 
> > > Why?
> > 
> It fails on some tests without this check, for example on stringSHA1 from browsemark.

OK, but why?

> > Fixed

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.