Open ScottTodd opened 1 month ago
We have this snippet in our Vulkan docs: https://iree.dev/guides/deployment-configurations/gpu-vulkan/#get-the-iree-runtime
Should add similar snippets to CUDA, ROCm, Metal, etc. pages as well as the common https://iree.dev/guides/deployment-configurations/#runtime-hal-driversdevices page.
Explain devices UUIDs and ordinals, as well as usage from CLI tools, C API bindings, and Python bindings.
Troubleshoot if the device(s) you expect to show up do not (e.g. missing from the build configuration, system GPU driver missing, GPU offline, etc.)
Can probably batch these updates together with multi-device doc updates (see --iree-hal-target-device usage explained on https://github.com/iree-org/iree/pull/17987)
--iree-hal-target-device
We have this snippet in our Vulkan docs: https://iree.dev/guides/deployment-configurations/gpu-vulkan/#get-the-iree-runtime![image](https://github.com/user-attachments/assets/d557f5a9-8c42-4ef7-bb65-97d413b7f510)
Should add similar snippets to CUDA, ROCm, Metal, etc. pages as well as the common https://iree.dev/guides/deployment-configurations/#runtime-hal-driversdevices page.
Explain devices UUIDs and ordinals, as well as usage from CLI tools, C API bindings, and Python bindings.
Troubleshoot if the device(s) you expect to show up do not (e.g. missing from the build configuration, system GPU driver missing, GPU offline, etc.)