[cfe-commits] Bug 11709 Fix: va_list on ARM is not following AAPCS 7.1.4

[Sean Silva]

> - AST can't depend on Sema. Is there a reason for the include of Sema.h in
> ASTContext.cpp?

Is this documented anywhere? If not you should mention it somewhere.

-- Sean Silva

On Thu, Oct 4, 2012 at 9:19 PM, Jordan Rose <jordan_rose at apple.com> wrote:
> A couple comments (sorry for the long long delay):
>
> - AST can't depend on Sema. Is there a reason for the include of Sema.h in
> ASTContext.cpp?
>
> - It would be a good idea to leave a comment in ExprClassification.cpp about
> why __va_list needs an exception.
>
> - For ARMTargetInfo::getBuiltinVaListKind, maybe you can just decide this at
> construction time and stick it in a local variable? 'unsigned usesAAPCS :
> 1'?
>
> Thanks for working on this, Weiming, Logan.
> Jordan
>
>
> On Oct 2, 2012, at 22:34 , Logan Chien <tzuhsiang.chien at gmail.com> wrote:
>
> LGTM.
>
> However, I think we can enhance the test cases a little.
> 1. Keep the void* test in builtins-arm.c for apcs-gnu ABI.
> 2. Add the unit test for va_list name mangling.
>
> Sincerely,
> Logan
>
> ps. The attached patch is the revised patch with the updated
> test cases.
>
> On Thu, Sep 20, 2012 at 6:13 AM, Weiming Zhao <weimingz at codeaurora.org>
> wrote:
>>
>> Ping...
>>
>> Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted
>> by
>> The Linux Foundation
>
> <mangle-valist.cpp><builtins-arm.c><0001-Bug-11709-Change-the-definition-of-va_list-to-meet-A.patch>
>
>
>
> _______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>