[llvm-dev] appropriate substitution for nvvm_atomic_load_add_f32?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

[llvm-dev] appropriate substitution for nvvm_atomic_load_add_f32?

Jeremy Morse via llvm-dev

I notice nvvm_atomic_load_add_f32 was removed. Is there a well known substitution for atomic_load_add?



 


_______________________________________________
LLVM Developers mailing list
[hidden email]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
Reply | Threaded
Open this post in threaded view
|

Re: [llvm-dev] appropriate substitution for nvvm_atomic_load_add_f32?

Jeremy Morse via llvm-dev
On Wed, Jan 22, 2020 at 3:28 PM david via llvm-dev
<[hidden email]> wrote:
> I notice nvvm_atomic_load_add_f32 was removed. Is there a well known substitution for atomic_load_add?

I don't know the specifics of this intrinsic and backend, but the
native IR atomicrmw instruction has grown fadd/fsub operation support,
so that seems like it ought to be the canonical replacement.

Cheers,
Nicolai


>
>
>
>
>
> _______________________________________________
> LLVM Developers mailing list
> [hidden email]
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev



--
Lerne, wie die Welt wirklich ist,
aber vergiss niemals, wie sie sein sollte.
_______________________________________________
LLVM Developers mailing list
[hidden email]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev