[Webkit-unassigned] [Bug 115138] [QNX][ARM] Use hardware integer division where available

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Apr 24 17:12:57 PDT 2013


https://bugs.webkit.org/show_bug.cgi?id=115138


Benjamin Poulain <benjamin at webkit.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #199537|review?, commit-queue?      |review-, commit-queue-
               Flag|                            |




--- Comment #3 from Benjamin Poulain <benjamin at webkit.org>  2013-04-24 17:11:16 PST ---
(From update of attachment 199537)
View in context: https://bugs.webkit.org/attachment.cgi?id=199537&action=review

Nope.

> Source/JavaScriptCore/dfg/DFGCommon.h:109
>  #if CPU(APPLE_ARMV7S)
>      return true;
> +#elif CPU(ARM_THUMB2) && OS(QNX)
> +    return MacroAssembler::supportsIntegerDiv();

This is terribly misleading.

What you should do is:
    -MacroAssembler::supportsIntegerDiv() is a _compile time_ constant for everything but QNX
    -It is true for ARMv7s, false everywhere else.

> Source/JavaScriptCore/dfg/DFGSpeculativeJIT32_64.cpp:2256
>  #elif CPU(APPLE_ARMV7S)
> -            compileIntegerArithDivForARMv7s(node);
> +            compileIntegerArithDivForARM(node);
> +#elif CPU(ARM_THUMB2) && OS(QNX)
> +            if (MacroAssembler::supportsIntegerDiv())
> +                compileIntegerArithDivForARM(node);

That makes no sense. What if you  are ArithDiv and MacroAssembler::supportsIntegerDiv() evaluate to false, you just silently ignore the operation?

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



More information about the webkit-unassigned mailing list