Wenju He 89ac672f5b [libclc] Delete wrong implementation nvptx clc_isinf (#205699) ...

The file calls __nv_isinf which return 1 for true on vector input, while
the generic clc_isinf which return -1 for true on vector input. Using
nvptx clc_isinf in OpenCL isinf violates OpenCL spec.

Found the issue in https://github.com/intel/llvm/pull/22413

2026-06-25 08:47:23 +08:00

..

amdgpu

[libclc] Use unchecked division for f64 AMDGPU reciprocal (#203809)

2026-06-15 07:26:32 -05:00

generic

[libclc] Use __CLC_SCALAR instead of nonexistant __CLC_SCALAR1 for sin (#203807)

2026-06-14 20:33:36 -05:00

nvptx

[libclc] Delete wrong implementation nvptx clc_isinf (#205699)

2026-06-25 08:47:23 +08:00

spirv

[libclc] Use float sw_fma for spirv*-mesa-mesa3d targets (#199626)

2026-05-29 13:28:17 +08:00