diff options
| author | Yuri Victorovich <yuri@FreeBSD.org> | 2018-09-20 21:48:56 +0000 |
|---|---|---|
| committer | Yuri Victorovich <yuri@FreeBSD.org> | 2018-09-20 21:48:56 +0000 |
| commit | 57abffa30e291e4864b86e5cb4281ad2f1ccae70 (patch) | |
| tree | 0878676eca00daf88b3de5bc16133cb6dbc0997b /math/py-gimmik | |
| parent | 63c5825f95bed656a20f07bb296e863eb518003b (diff) | |
| download | ports-57abffa30e291e4864b86e5cb4281ad2f1ccae70.tar.gz ports-57abffa30e291e4864b86e5cb4281ad2f1ccae70.zip | |
New port: math/py-gimmik: Generator of Matrix Multiplication Kernels
Notes
Notes:
svn path=/head/; revision=480219
Diffstat (limited to 'math/py-gimmik')
| -rw-r--r-- | math/py-gimmik/Makefile | 21 | ||||
| -rw-r--r-- | math/py-gimmik/distinfo | 3 | ||||
| -rw-r--r-- | math/py-gimmik/pkg-descr | 5 |
3 files changed, 29 insertions, 0 deletions
diff --git a/math/py-gimmik/Makefile b/math/py-gimmik/Makefile new file mode 100644 index 000000000000..9c675fdf3b32 --- /dev/null +++ b/math/py-gimmik/Makefile @@ -0,0 +1,21 @@ +# $FreeBSD$ + +PORTNAME= gimmik +DISTVERSION= 2.1 +CATEGORIES= math python +MASTER_SITES= CHEESESHOP +PKGNAMEPREFIX= ${PYTHON_PKGNAMEPREFIX} + +MAINTAINER= yuri@FreeBSD.org +COMMENT= Generator of Matrix Multiplication Kernels + +LICENSE= BSD3CLAUSE + +RUN_DEPENDS= ${PYNUMPY} \ + ${PYTHON_PKGNAMEPREFIX}mako>0:textproc/py-mako@${PY_FLAVOR} + +USES= python:3.3+ +USE_PYTHON= distutils autoplist +NO_ARCH= yes + +.include <bsd.port.mk> diff --git a/math/py-gimmik/distinfo b/math/py-gimmik/distinfo new file mode 100644 index 000000000000..02e1cc2e7185 --- /dev/null +++ b/math/py-gimmik/distinfo @@ -0,0 +1,3 @@ +TIMESTAMP = 1537478754 +SHA256 (gimmik-2.1.tar.gz) = 36ec13aff705e787b856ae236211d1826ceaa4b22d52d4d98ff72642bb4021a3 +SIZE (gimmik-2.1.tar.gz) = 3503 diff --git a/math/py-gimmik/pkg-descr b/math/py-gimmik/pkg-descr new file mode 100644 index 000000000000..9ef1522318aa --- /dev/null +++ b/math/py-gimmik/pkg-descr @@ -0,0 +1,5 @@ +Generator of Matrix Multiplication Kernels (GiMMiK) is a tool for generation of +high performance matrix multiplication kernel code for various accelerator +platforms. Currently CUDA and OpenCL are the only supported platforms. + +WWW: https://github.com/vincentlab/GiMMiK |