[LLVMdev] NEON vector instructions and the fast math IR flags

[Tobias Grosser]

On 06/07/2013 10:24 AM, Renato Golin wrote:
> On 7 June 2013 18:08, Arnold Schwaighofer <aschwaighofer at apple.com> wrote:
>
>> What I am suggesting is that (if you care about denormals):
>>
>> * the arm backend has to be fixed to scalarize floating point vector
>> operations (behind a flag)
>> * the arm target transform model has to correctly reflect that
>>
>
> Yup. What I had in mind, too. This is why I asked Tobi to create two bugs,
> and we would fix them accordingly. ;)

Done.

1) Fix the ARM target to only introduce NEON if valid

llvm.org/PR16274

2) Fix the vectorizer cost function

llvm.org/PR16275

Thanks for your insights.

Cheers,
Tobi