# Copyright 1999-2026 Gentoo Authors # Distributed under the terms of the GNU General Public License v2 EAPI=8 DISTUTILS_USE_PEP517=setuptools PYTHON_COMPAT=( python3_{12..14} ) DISTUTILS_SINGLE_IMPL=1 inherit distutils-r1 pypi DESCRIPTION="Collection of CuTe-based CUDA kernels" HOMEPAGE=" https://github.com/Dao-AILab/quack https://pypi.org/project/quack-kernels/ " LICENSE="Apache-2.0" SLOT="0" KEYWORDS="~amd64" # Tests compile and execute CUDA kernels. RESTRICT="test" # Pure Python with kernels JIT-compiled through apache-tvm-ffi and CUTLASS DSL. # Pin CUTLASS 4.6.0 exactly: generated kernels are ABI-bound to it. # verified 2026-08-04 RDEPEND=" dev-python/torch-c-dlpack-ext[${PYTHON_SINGLE_USEDEP}] sci-ml/pytorch[${PYTHON_SINGLE_USEDEP}] $(python_gen_cond_dep ' ~dev-python/nvidia-cutlass-dsl-4.6.0[${PYTHON_USEDEP}] >=dev-python/apache-tvm-ffi-0.1.6[${PYTHON_USEDEP}]