Tim Renouf 76007138f4 [LLVM] New NoDivergenceSource function attribute (#111832) ...

A call to a function that has this attribute is not a source of
divergence, as used by UniformityAnalysis. That allows a front-end to
use known-name calls as an instruction extension mechanism (e.g.
https://github.com/GPUOpen-Drivers/llvm-dialects ) without such a call
being a source of divergence.

2024-10-12 09:34:45 +01:00

..

deprecated

…

irreducible

…

MIR

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

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

always_uniform.ll

AMDGPU: Add some leaf intrinsics to isAlwaysUniform (#101925)

2024-08-06 21:09:04 +04:00

atomics.ll

…

b42473-r1-crash.ll

…

control-flow-intrinsics.ll

…

hidden_diverge.ll

…

hidden_loopdiverge.ll

…

inline-asm.ll

…

interp_f16.ll

…

intrinsics.ll

AMDGPU: Add back half and bfloat support for global_load_tr16 pats (#99540)

2024-07-18 11:23:35 -07:00

join-at-loop-exit.ll

…

kernel-args.ll

…

lit.local.cfg

…

llvm.amdgcn.buffer.atomic.ll

[RFC][AMDGPU] Remove old llvm.amdgcn.buffer.* and tbuffer intrinsics (#93801)

2024-06-10 12:14:51 +01:00

llvm.amdgcn.image.atomic.ll

[AMDGPU] Overload resource descriptor in image intrinsics. (#107255)

2024-09-27 15:33:52 +02:00

no-return-blocks.ll

…

nodivergencesource.ll

[LLVM] New NoDivergenceSource function attribute (#111832)

2024-10-12 09:34:45 +01:00

phi-undef.ll

…

propagate-loop-live-out.ll

…

temporal_diverge.ll

…

trivial-join-at-loop-exit.ll

…

unreachable-loop-block.ll

…

workitem-intrinsics.ll

…