[llvm-dev] NEON FP flags

[Hal Finkel via llvm-dev]

Hi Renato,

As I understand it, the fundamental property being addresses here is: Are the semantics of scalar FP math the same as vector FP math? TTI seems like a good place to expose that information. If the semantics are indeed different, then the vectorizer would require fast-math flags in order to vectorize FP operations (similarly, gcc's man page says it requires -funsafe-math-optimizations for vectorization unless -mfpu=neon or similar is specified). In this context, this different-semantics query would return true if:

  !(isDarwin OR ARMISA >= v8 OR fpMath == NEON)

and then we need to teach people to use -mfpu=neon ;)

I think this more-or-less matches what you've proposed. Is that right?

 -Hal

P.S. Looking at gcc's man page, gcc seems to use -mfpu for ARM and -mfpmath for x86. Do we use -mfpmath for both?

----- Original Message -----
> From: "Renato Golin" <renato.golin at linaro.org>
> To: "James Molloy" <James.Molloy at arm.com>
> Cc: "Hal Finkel" <hfinkel at anl.gov>, "LLVM Dev" <llvm-dev at lists.llvm.org>
> Sent: Tuesday, March 22, 2016 6:50:40 AM
> Subject: Re: NEON FP flags
> 
> On 22 March 2016 at 11:34, James Molloy <James.Molloy at arm.com> wrote:
> > I don’t think this part is right. The denormal flag would have to
> > be set by
> > whatever code generates the FP instruction, which would be Clang’s
> > codegen
> > layer. So the if (Darwin) would be there, not in TTI.
> 
> Right, I meant the information to set/not set would be in TTI, not
> the
> actual setting.
> 
> I don't want Clang to have that kind of information independent of
> LLVM.
> 
> cheers,
> --renato
> 

-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory