[Webkit-unassigned] [Bug 129807] [Win64] Compile error after r165128.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=129807





--- Comment #10 from peavo at outlook.com  2014-03-07 12:41:48 PST ---
(In reply to comment #9)
> (From update of attachment 226133 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=226133&action=review
> 
> >>> Source/JavaScriptCore/bytecode/Opcode.h:82
> >>> +#if ENABLE(COMPUTED_GOTO_OPCODES) || !ENABLE(LLINT_C_LOOP)
> >> 
> >> Why do you need to add the !ENABLE(LLINT_C_LOOP) conditional?  The 2 seems independent to me.  I think this is not needed.
> > 
> > This was done to fix compile errors for Win32 when COMPUTED_GOTO_OPCODES is not enabled, but this is probably not the best way to fix it ... :)
> > 
> > e.g.
> > 
> > 1>..\llint\LLIntData.cpp(59): error C2440: 'static_cast' : cannot convert from 'void *' to 'JSC::Opcode' There is no context in which this conversion is possible
> 
> Sounds like the real issue is that for 32-bit Win, you want the equivalent of COMPUTED_GOTO_OPCODES, but for 64-bit Win, you don’t.  It doesn’t directly have to do with LLINT_C_LOOP, but I guess it is fair to add !ENABLE(LLINT_C_LOOP) as a condition because the ASM LLINT only use the equivalent of computed gotos.  But instead of adding “|| !ENABLE(LLINT_C_LOOP)” all over, lets’s change, in Platform.h, the condition for "#define ENABLE_COMPUTED_GOTO_OPCODES 1” to "#if HAVE(COMPUTED_GOTO) || !ENABLE(LLINT_C_LOOP)”.

Great, thanks, patch coming up :)

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