[cfe-commits] [llvm-commits] [patch][pr14113] Use inreg in combination with x86_fastcallcc

Eli Friedman eli.friedman at gmail.com
Fri Oct 19 13:53:30 PDT 2012


On Fri, Oct 19, 2012 at 8:01 AM, Rafael EspĂ­ndola
<rafael.espindola at gmail.com> wrote:
>> (Note to anyone else reviewing: don't compare against gcc 4.2; gcc's
>> behavior changed in 4.3.)
>>
>> Please add a testcase for something like "void bar5(struct S2 y, int
>> z) {}", with S2 from your testcase.  Please add a testcase with an
>> argument which is a C++ struct with a non-trivial copy constructor.
>> Please add testcases involving sret.
>
> Thanks for the suggestions. This found a bug in regparm too!
>
>> This change is going to break any other sources of IR which expect the
>> current behavior of x86_fastcall; at the very least, please start a
>> thread on llvmdev so this doesn't surprise anyone.
>
> A new patch is attached, I will start a thread on llvmdev about the
> representation.

The changes involving padding types are relatively difficult to
follow; can you split them out into a followup patch?

-Eli




More information about the cfe-commits mailing list