Gentoo Packages

Search

Manage

• Information: euse -i rocm
• Enable: euse -E rocm
• Disable: euse -D rocm

Local and Metadata Descriptions

Metadata Descriptions (metadata.xml)

• cross-x86_64-unknown-linux-musl/gdb: Enable support for AMD GPU debugging via <pkg>dev-libs/rocdbgapi</pkg>
• dev-debug/gdb: Enable support for AMD GPU debugging via <pkg>dev-libs/rocdbgapi</pkg>
• dev-python/cupy: Enable AMD ROCm
• dev-python/vllm: Build for AMD ROCm inference (VLLM_TARGET_DEVICE=rocm); pull HIP libs + torch{audio,vision}
• llvm-runtimes/openmp: Build <pkg>dev-libs/rocr-runtime</pkg> offloading support for AMD GPUs.
• media-gfx/comfyui: Build the torch backend (<pkg>sci-ml/caffe2</pkg>) with AMD ROCm/HIP acceleration; installs the non-CUDA wheels of <pkg>dev-python/comfy-aimdo-bin</pkg> and <pkg>dev-python/comfy-kitchen-bin</pkg> and pulls <pkg>dev-python/triton-bin</pkg>, whose Triton backend runs comfy_kitchen.apply_rope on AMD. Mutually exclusive with cuda
• sci-libs/torchaudio: Enable ROCm gpu computing support
• sci-misc/llama-cpp: Build a HIP (ROCm) backend for AMD GPUs
• sci-ml/caffe2: Enable ROCm gpu computing support
• sci-ml/gloo: Enable ROCm gpu computing support
• sci-ml/llama-cpp: Build a HIP (ROCm) backend
• sci-ml/ollama: Build the ROCm/HIP (rocm_v7_2) llama-server GPU backend
• sci-ml/ollama-bin: Support for AMD ROCm compatible cards
• sci-ml/torchvision: Enable ROCm gpu computing support
• sys-apps/hwloc: Enable AMD RSMI device discovery
• sys-block/libfabric: Enable Radeon Open Compute provider

Local Descriptions (use.local.desc)

• dev-debug/gdb: Enable support for AMD GPU debugging via dev-libs/rocdbgapi
• llvm-runtimes/openmp: Build dev-libs/rocr-runtime offloading support for AMD GPUs.
• sci-misc/ik_llama-cpp: Build a HIP (ROCm) backend
• sci-misc/llama-cpp: Build a HIP (ROCm) backend
• sci-misc/stable-diffusion-cpp: Build a HIP (ROCm) backend
• sci-ml/caffe2: Enable ROCm gpu computing support
• sci-ml/flash-attn: Enable ROCm support.
• sci-ml/gloo: Enable ROCm gpu computing support
• sci-ml/ollama: Enable ROCm gpu computing support
• sci-ml/ollama-bin: Fetch standalone Ollama ROCm libraries
• sci-ml/torchvision: Enable ROCm gpu computing support
• sys-apps/hwloc: Enable AMD RSMI device discovery
• sys-block/libfabric: Enable Radeon Open Compute provider

Warnings

• Warning: USE flag 'rocm' is used in ebuilds/metadata but is not defined in profiles/use.desc
• Warning: USE flag 'rocm' used by app-misc/ollama-bin but has no description in metadata.xml, use.local.desc or use.desc
• Warning: USE flag 'rocm' used by dev-python/jax but has no description in metadata.xml, use.local.desc or use.desc
• Warning: USE flag 'rocm' used by dev-python/rembg but has no description in metadata.xml, use.local.desc or use.desc
• Warning: USE flag 'rocm' used by app-misc/ik_llama-cpp but has no description in metadata.xml, use.local.desc or use.desc
• Warning: USE flag 'rocm' used by app-misc/llama-cpp but has no description in metadata.xml, use.local.desc or use.desc
• Warning: USE flag 'rocm' used by app-misc/stable-diffusion-cpp but has no description in metadata.xml, use.local.desc or use.desc

Packages Using This Flag