Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • hpvm-release-epochs
  • hpvm-release-epochs0
  • main default
  • v2.0
  • v2.0rc
  • v1.0
  • v0.5
7 results
Created with Raphaël 2.2.026Apr23521Mar201918171615131211987543227Feb1615983Dec19Nov1897427Oct16Sep1211109876542131Aug231815542130Jul26212016149230Jun22201814131153229May201312432129Apr10927Mar12987642127Feb26231413121110964130Jan282717Dec165430Nov2927262524211918171614119654331Oct30191612711Sep413Aug12111065432131Jul3029262315108727Jun322May24Apr17143Oct20Sep1Aug4Jul3230Jun28262513Fixed pipeline benchmark to not specify return types of internal nodes. That is automatically inferred from bind out intrinsicsChanged GenVISC pass to automatically infer return type of internal nodes from bind output intrinsicsFixed linera-svm conflictlinear-svm example committed to compute memory branchNo internal nodes for linear-svm. Using just vector opsFixed pipeline code. Allows multiple runs within the code. Fixed visc.mk to support SPIR and PTX bothRemoved bugs in hint generation. Earlier non-gpu hints were all considered sequential cpu.Made visc-rt threadsafe by implementing locks across all ocl functions. Very crude. Fine grained locking possible if performance deterioteslinear-svm example committed to compute memory branchmoved spmv to c++. Required for dummy visc callsfixed lbm hint in viscfixed hint in stencil viscCommiting the pipeline version of pldiModified visc-rt to not create a thread for launch. Improves results for vector casePorted bfs to cpu baselinePorted cutcp for cpu_baseline versionMerge branch 'master' of bitbucket.org:psrivas2/viscPorted histo for opencl cpuStencil vector baselineMerge branch 'master' of bitbucket.org:psrivas2/viscvisc.mk changes missing in previous commitNow we do not need to edit the visc hint for the 5 new benchmarks, to compile them for avx or gpuPorted opencl_base for avx platformSgemm opencl cpu nvidia version (in opencn_cpu_sm)opencl_base version of tpacf with dynamic shared memory allocationbfs working for avx backendcutcp working for avx backendFixed a bug in SPIR backend to generate get_num_groups correctlyMerge branch 'master' of bitbucket.org:psrivas2/visc(1) sgemm visc_sh working for vector backendAdded unsigned min-max atomicsAdded support for umin and umax visc intrinsicsAdding 2 files to histo. host and ptx to run visc generated host code with opencl driver generated ptxAdded a single kernel file for histogram opencl_nvidia kernels. Improved timing calculations for viscAdded single kernel file to opencl_nvidia version of histogramvisc version similar to nvidia versionHistogram improvedMerge branch 'master' of bitbucket.org:psrivas2/viscRemove llvm intrinsic declarationsFixed bug in histo. Was using get_group_id instead of visc intrinsic
Loading