Better support for AMD multi-GPU on linux (#7212)

* Better support for AMD multi-GPU

This resolves a number of problems related to AMD multi-GPU setups on linux.

The numeric IDs used by rocm are not the same as the numeric IDs exposed in
sysfs although the ordering is consistent.  We have to count up from the first
valid gfx (major/minor/patch with non-zero values) we find starting at zero.

There are 3 different env vars for selecting GPUs, and only ROCR_VISIBLE_DEVICES
supports UUID based identification, so we should favor that one, and try
to use UUIDs if detected to avoid potential ordering bugs with numeric IDs

* ROCR_VISIBLE_DEVICES only works on linux

Use the numeric ID only HIP_VISIBLE_DEVICES on windows
This commit is contained in:
Daniel Hiltgen
2024-10-26 14:04:14 -07:00
committed by GitHub
parent 35ec7f079f
commit d7c94e0ca6
6 changed files with 69 additions and 43 deletions

View File

@@ -265,9 +265,9 @@ func AsMap() map[string]EnvVar {
if runtime.GOOS != "darwin" {
ret["CUDA_VISIBLE_DEVICES"] = EnvVar{"CUDA_VISIBLE_DEVICES", CudaVisibleDevices(), "Set which NVIDIA devices are visible"}
ret["HIP_VISIBLE_DEVICES"] = EnvVar{"HIP_VISIBLE_DEVICES", HipVisibleDevices(), "Set which AMD devices are visible"}
ret["ROCR_VISIBLE_DEVICES"] = EnvVar{"ROCR_VISIBLE_DEVICES", RocrVisibleDevices(), "Set which AMD devices are visible"}
ret["GPU_DEVICE_ORDINAL"] = EnvVar{"GPU_DEVICE_ORDINAL", GpuDeviceOrdinal(), "Set which AMD devices are visible"}
ret["HIP_VISIBLE_DEVICES"] = EnvVar{"HIP_VISIBLE_DEVICES", HipVisibleDevices(), "Set which AMD devices are visible by numeric ID"}
ret["ROCR_VISIBLE_DEVICES"] = EnvVar{"ROCR_VISIBLE_DEVICES", RocrVisibleDevices(), "Set which AMD devices are visible by UUID or numeric ID"}
ret["GPU_DEVICE_ORDINAL"] = EnvVar{"GPU_DEVICE_ORDINAL", GpuDeviceOrdinal(), "Set which AMD devices are visible by numeric ID"}
ret["HSA_OVERRIDE_GFX_VERSION"] = EnvVar{"HSA_OVERRIDE_GFX_VERSION", HsaOverrideGfxVersion(), "Override the gfx used for all detected AMD GPUs"}
ret["OLLAMA_INTEL_GPU"] = EnvVar{"OLLAMA_INTEL_GPU", IntelGPU(), "Enable experimental Intel GPU detection"}
}