summaryrefslogtreecommitdiffstats
path: root/polly/lib/CodeGen/PPCGCodeGeneration.cpp
Commit message (Expand)AuthorAgeFilesLines
...
* GPGPU: Add basic support for kernel launchesTobias Grosser2016-07-271-0/+171
* GPGPU: Load GPU kernelsTobias Grosser2016-07-251-3/+60
* GPGPU: Emit data-transfer codeTobias Grosser2016-07-251-25/+139
* GPGPU: Complete code to allocate and free device arraysTobias Grosser2016-07-251-4/+45
* GPGPU: initialize GPU context and simplify the corresponding GPURuntime inter...Tobias Grosser2016-07-251-0/+116
* IslNodeBuilder: Make finalize() virtualTobias Grosser2016-07-251-1/+1
* GPGPU: Optimize kernel IR before generating assembly codeTobias Grosser2016-07-241-0/+9
* GPGPU: Verify kernel IR before generating assemblyTobias Grosser2016-07-241-0/+5
* GPGPU: Generate PTX assembly code for the kernel modulesTobias Grosser2016-07-221-0/+123
* GPGPU: generate code for ScopStatementsTobias Grosser2016-07-211-15/+202
* GPGPU: Bail out of scops with hoisted invariant loadsTobias Grosser2016-07-191-0/+4
* GPGPU: Emit in-kernel synchronization statementsTobias Grosser2016-07-191-0/+49
* GPGPU: generate control flow within the kernelTobias Grosser2016-07-191-0/+6
* GPGPU: add scop parameters to kernel argumentsTobias Grosser2016-07-191-1/+14
* GPGPU: add host iterators to kernel argumentsTobias Grosser2016-07-191-1/+14
* GPGPU: add intrinsic functions to obtain a kernels thread and block idsTobias Grosser2016-07-191-0/+50
* GPGPU: create kernel function skeletonTobias Grosser2016-07-191-7/+153
* GPGPU: collect array referencesTobias Grosser2016-07-181-0/+2
* GPGPU: Pull implementation out of class definitionTobias Grosser2016-07-181-4/+7
* GPGPU: Create host control flowTobias Grosser2016-07-181-0/+82
* GPGPU: Format statements scheduled on the host ourselvesTobias Grosser2016-07-151-0/+14
* GPGPU: Use schedule whole components for schedulerTobias Grosser2016-07-151-9/+1
* GPGPU: Drop domain constraints from flow dependencesTobias Grosser2016-07-151-0/+9
* GPGPU: Add memory reference tag ids to tagged accessesTobias Grosser2016-07-151-0/+2
* GPGPU: Make sure scops with more than one array workTobias Grosser2016-07-151-0/+1
* GPGPU: Free options to avoid memory leakTobias Grosser2016-07-151-0/+14
* GPGPU: Shorten ppcg include paths to avoid conflict with cuda.hTobias Grosser2016-07-151-5/+5
* GPGPU: Model array access informationTobias Grosser2016-07-151-10/+150
* GPGPU: Generate an AST for the GPU-mapped scheduleTobias Grosser2016-07-141-4/+127
* GPGPU: Add dummy implementation for ast expression constructionTobias Grosser2016-07-141-0/+18
* GPGPU: Use a tile size of 32 by defaultTobias Grosser2016-07-141-0/+2
* Fix formattingTobias Grosser2016-07-141-1/+2
* GPGPU: Map initial schedule to GPU scheduleTobias Grosser2016-07-141-2/+41
* GPGPU: Do not dump schedule by defaultTobias Grosser2016-07-141-1/+1
* GPGPU: compute new schedule from polly scopTobias Grosser2016-07-141-12/+147
* GPGPU: create default initialized PPCG scop and gpu programTobias Grosser2016-07-141-1/+137
* Add accelerator code generation pass skeletonTobias Grosser2016-07-131-0/+82
OpenPOWER on IntegriCloud