Daniel Hiltgen 1618700c5a

Workaround buggy P2P ROCm copy on windows (#7466) ...

This enables the workaround code only for windows which should help windows users with muliple AMD GPUs

2024-11-07 14:26:31 -08:00

..

common-defs.make

Soften windows clang requirement (#7428)

2024-10-30 12:28:36 -07:00

cuda.make

Improve dependency gathering logic (#7345)

2024-10-24 09:51:53 -07:00

gpu.make

Be explicit for gpu library link dir (#7560)

2024-11-07 09:20:40 -08:00

Makefile.cuda_v11

Re-introduce the llama package (#5034)

2024-10-08 08:53:54 -07:00

Makefile.cuda_v12

Re-introduce the llama package (#5034)

2024-10-08 08:53:54 -07:00

Makefile.default

Improve dependency gathering logic (#7345)

2024-10-24 09:51:53 -07:00

Makefile.rocm

Workaround buggy P2P ROCm copy on windows (#7466)

2024-11-07 14:26:31 -08:00

Makefile.sync

Remove submodule and shift to Go server - 0.4.0 (#7157)

2024-10-30 10:34:28 -07:00