summaryrefslogtreecommitdiffstats
path: root/llvm/test/Transforms/LoopUnroll/AMDGPU
Commit message (Expand)AuthorAgeFilesLines
* [AMDGPU] Tune inlining parameters for AMDGPU target (part 2)dfukalov2019-11-191-1/+1
* [AMDGPU] Improve code size cost model (part 2)dfukalov2019-11-061-1/+1
* [lit] Delete empty lines at the end of lit.local.cfg NFCFangrui Song2019-06-171-1/+0
* TTI: Improve default costs for addrspacecastMatt Arsenault2019-06-031-0/+77
* Revert "Temporarily Revert "Add basic loop fusion pass.""Eric Christopher2019-04-174-0/+267
* Temporarily Revert "Add basic loop fusion pass."Eric Christopher2019-04-174-267/+0
* AMDGPU: Fix tests using old number for constant address spaceMatt Arsenault2018-09-101-2/+2
* LoopUnroll: Allow analyzing intrinsic call costsMatt Arsenault2018-06-261-0/+77
* [AMDGPU] Switch to the new addr space mapping by defaultYaxun Liu2018-02-021-21/+21
* AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernelMatt Arsenault2017-03-212-5/+5
* AMDGPU: Don't unroll for private with dynamic allocasMatt Arsenault2017-02-031-0/+34
* [AMDGPU] Unroll preferences improvementsStanislav Mekhanoshin2017-02-031-0/+120
* AMDGPU: Remove some old intrinsic uses from testsMatt Arsenault2016-02-111-6/+6
* AMDGPU: Switch barrier intrinsics to using convergentMatt Arsenault2015-12-192-0/+36
OpenPOWER on IntegriCloud