libxsmm: new build-block added

Compiler/libxsmm/build

@@ -0,0 +1,19 @@
+#!/usr/bin/env modbuild
+
+pbuild::set_download_url "https://github.com/hfp/libxsmm/archive/1.13/libxsmm-1.13.tar.gz"
+pbuild::add_to_group 'Compiler'
+
+pbuild::install_docfiles 'CONTRIBUTING.md'
+pbuild::install_docfiles 'LICENSE.md'
+pbuild::install_docfiles 'README.md'
+
+pbuild::compile_in_sourcetree
+
+pbuild::configure() {
+	:
+}
+
+pbuild::compile() {
+	make PREFIX="${PREFIX}"
+}
+

Compiler/libxsmm/files/variants.rhel6

@@ -0,0 +1 @@
+libxsmm/1.13	unstable	intel/19.4

Compiler/libxsmm/modulefile

@@ -0,0 +1,17 @@
+#%Module1.0
+
+module-whatis		"specialized dense and sparse matrix operations and deep learning primitives"
+module-url		"https://github.com/hfp/libxsmm"
+module-license		"BSD 3-Clause License"
+module-maintainer	"Achim Gsell <achim.gsell@psi.ch>"
+module-help		"
+LIBXSMM is a library for specialized dense and sparse matrix operations
+as well as for deep learning primitives such as small convolutions
+targeting Intel Architecture. Small marix multiplication kernels (SMMs)
+are generated for Intel SSE,  Intel AVX, Intel AVX2, and Intel AVX-512
+as found in the Intel Xeon Phi processor family (KNL, KNM) and Intel
+Xeon processors (SKX). Highly optimized code for small convolutions
+is targeting Intel AVX2 and Intel AVX-512, whereas other targets can
+automatically leverage specialized SMMs to perform convolutions.
+"
+