llvm-project/mlir/test/Conversion/AMDGPUToROCDL
Krzysztof Drewniak 6329562249 [mlir][AMDGPU] Explicitly truncate memory addresses in buffer ops
As a percaution, truncate memory addresses passed to kernels to 48 bits,
since bits 48-63 of the buffer descriptor are used for the stride field
and, on gfx10, to control swizzling.

Reviewed By: ThomasRaoux

Differential Revision: https://reviews.llvm.org/D131016
2022-08-04 19:42:33 +00:00
..
amdgpu-to-rocdl.mlir [mlir][AMDGPU] Explicitly truncate memory addresses in buffer ops 2022-08-04 19:42:33 +00:00