Partitioning a parallel computation into finitely sized chunks for effective mapping onto a parallel machine is a critical concern for source-to-source compilation. In the context of OpenCL and CUDA, this translates to the definition of a uniform hyper-rectangular partitioning of the parallel execution space where each partition is subject to a fine-grained distribution of resources that has a direct yet hard to estimate impact on performance. This paper develops the first compilation scheme for generating parametrically tiled codes for affine loop programs on GPUs which facilitates run-time exploration of partitioning parameters as a fast and portable way of finding the ones that yield maximum performance. Our approach is based on a parametric tiling scheme for producing wavefronts of parallel rectangular partitions of parametric size and a novel runtime system that manages wavefront execution and local memory usage dynamically through an inspector-executor mechanism. Our experimental evaluation demonstrates the effectiveness of our approach for wavefront as well as rectangularly-parallel partitionings.<\/p>\n","protected":false},"excerpt":{"rendered":"
Partitioning a parallel computation into finitely sized chunks for effective mapping onto a parallel machine is a critical concern for source-to-source compilation. In the context of OpenCL and CUDA, this translates to the definition of a uniform hyper-rectangular partitioning of the parallel execution space where each partition is subject to a fine-grained distribution of resources […]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[11,89,90,3],"tags":[215,1782,14,20,1268,974,1793,176,1390,1017],"class_list":["post-10654","post","type-post","status-publish","format-standard","hentry","category-computer-science","category-nvidia-cuda","category-opencl","category-paper","tag-code-generation","tag-computer-science","tag-cuda","tag-nvidia","tag-nvidia-geforce-gt-540-m","tag-nvidia-geforce-gtx-580","tag-opencl","tag-package","tag-tesla-k20","tag-tesla-m2070"],"views":3023,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/10654","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=10654"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/10654\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=10654"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=10654"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=10654"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}