Richard Biener 209ed06c3a Fix architecture support in OMP_OFFLOAD_init_device for gcn ...

The following makes the existing architecture support check work
instead of being optimized away (enum vs. -1).  This avoids
later asserts when we assume such devices are never actually
used.

libgomp/
	* plugin/plugin-gcn.c
	(EF_AMDGPU_MACH::EF_AMDGPU_MACH_UNSUPPORTED): Add.
	(isa_code): Return that instead of -1.
	(GOMP_OFFLOAD_init_device): Adjust.

2024-01-26 15:36:35 +01:00

..

configfrag.ac

Update copyright years.

2024-01-03 12:19:35 +01:00

cuda-lib.def

OpenMP: Call cuMemcpy2D/cuMemcpy3D for nvptx for omp_target_memcpy_rect

2023-07-26 16:22:35 +02:00

Makefrag.am

Update copyright years.

2024-01-03 12:19:35 +01:00

plugin-gcn.c

Fix architecture support in OMP_OFFLOAD_init_device for gcn

2024-01-26 15:36:35 +01:00

plugin-nvptx.c

Update copyright years.

2024-01-03 12:19:35 +01:00