[Webkit-unassigned] [Bug 143754] [W32] weakCompareAndSwap assembler code is not used when building with MinGW GCC

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=143754

Csaba Osztrogonác <ossy at webkit.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #250789|review?                     |review-
              Flags|                            |

--- Comment #2 from Csaba Osztrogonác <ossy at webkit.org> ---
Comment on attachment 250789
  --> https://bugs.webkit.org/attachment.cgi?id=250789
[GTK][W32] Allow MinGW-GCC to use GCC asm code for weakCompareAndSwap

View in context: https://bugs.webkit.org/attachment.cgi?id=250789&action=review

> Source/JavaScriptCore/ChangeLog:8
> +        [W32] Fixup dummy HeapStatistics implementation
> +
> +        GCC warns that exitWithFailure is marked as noreturn, but it does return:
> +        ../webkitgtk-2.4.8/Source/JavaScriptCore/heap/HeapStatistics.cpp:135:1: warning: 'noreturn' function does return
> +
> +        Call exit(-1) to indicate that no, there is no returning from here.

It is unrelated to this patch.

> Source/JavaScriptCore/ChangeLog:16
> +        * heap/HeapStatistics.cpp:
> +        (JSC::HeapStatistics::exitWithFailure):

ditto

> Source/WTF/wtf/Atomics.h:263
> -#if !OS(WINDOWS) && (CPU(X86) || CPU(X86_64))
> +#if (!OS(WINDOWS) || COMPILER(GCC)) && (CPU(X86) || CPU(X86_64))

I think OS(WINDOWS) is very confusing and isn't good at all,
COMPILER(GCC) && (CPU(X86) || CPU(X86_64)) would be better.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20150415/1ed369b0/attachment-0001.html>