Matt Arsenault 8520061281 AMDGPU: Support local atomicrmw fmin/fmax for float/double (#95590) ...

This has always been supported. Somehow, we ended up with 2
copies of clang builtins for this case, and the newer one
erroneously requires gfx8-insts.

2024-06-18 18:34:34 +02:00

..

irreducible

…

always-uniform-gmir.mir

…

always-uniform.mir

…

atomics-gmir.mir

AMDGPU: Support local atomicrmw fmin/fmax for float/double (#95590)

2024-06-18 18:34:34 +02:00

atomics.mir

…

control-flow-intrinsics.mir

…

hidden-diverge-gmir.mir

…

hidden-loop-diverge.mir

…

join-loopexit-gmir.mir

…

loads-gmir.mir

…

never-uniform.mir

…

temporal-divergence.mir

…

uses-value-from-cycle.mir

…