summaryrefslogtreecommitdiffstats
path: root/libclc/amdgcn
Commit message (Expand)AuthorAgeFilesLines
* amdgcn: Consolidate atomic minmax helpersJan Vesely2018-11-2711-57/+4
* amdgcn: Move __clc_amdgcn_s_waitcnt definition to clc fileJan Vesely2018-11-044-15/+1
* amdgcn: Convert get_num_groups to clcJan Vesely2018-11-0413-75/+16
* amdgcn: Convert get_global_size to clcJan Vesely2018-11-0413-75/+16
* amdgcn: Convert get_local_size to clcJan Vesely2018-11-0413-75/+16
* amdgcn: Use __constant AS for amdgcn builtins.Jan Vesely2018-08-032-2/+6
* Add initial support for half precision builtinsJan Vesely2018-05-172-0/+30
* amdgcn/fmin: Fix typos that reduced precisionJan Vesely2018-04-171-3/+3
* amdgcn: Update datalayout after LLVM r328656Jan Vesely2018-04-054-4/+4
* amdgcn/fmax: fcanonicalize operandsJan Vesely2018-03-082-0/+32
* amdgcn/fmin: fcanonicalize operandsJan Vesely2018-03-082-0/+32
* amdgcn,popcount: Workaround broken llvm.ctpop intrinsic on some GCN ASICsJan Vesely2018-03-083-0/+24
* amdgcn: Fix build after GDS/const AS swap in r325030Jan Vesely2018-02-236-10/+20
* amdgcn: Fix datalayout after addition of 32bit const AS in r324747Jan Vesely2018-02-234-4/+4
* amdgcn: Fix datalayout after clang r324101Jan Vesely2018-02-2317-5/+150
* amdgcn: Add missing datalayout info to .ll filesJan Vesely2017-10-207-0/+14
* Let get_work_dim take exactly 0 argumentsJeroen Ketema2017-10-011-1/+1
* Restore support for llvm-3.9Jan Vesely2017-09-295-0/+60
* Implement cl_khr_int64_extended_atomics builtinsJan Vesely2017-09-202-0/+48
* amdgcn,waitcnt: Add datalayout infoJan Vesely2017-09-041-0/+2
* amdgcn: rewrite barrier() using fence and clang __builtin_amdgcn_s_barrierJan Vesely2017-08-163-33/+8
* amdgcn: Implement {read_,write_,}mem_fence builtinJan Vesely2017-08-163-0/+52
* amdgcn: Fix return type of get_num_groupsMatt Arsenault2016-08-252-0/+22
* amdgcn: Fix return type for get_global_sizeMatt Arsenault2016-08-242-0/+22
* amdgpu: Fix default case value for get_local_sizeMatt Arsenault2016-08-201-1/+1
* amdgcn: Fix get_local_size IR return typeMatt Arsenault2016-08-202-0/+22
* amdgcn: Correct return types to be size_tMatt Arsenault2016-08-193-3/+3
* AMDGPU: Implement get_global_offset builtinJan Vesely2016-07-222-0/+12
* AMDGPU: Use clang intrinsics for workitem builtinsJan Vesely2016-07-226-62/+34
* Replace llvm.AMDGPU.ldexp with llvm.amdgcn.ldexpMatt Arsenault2016-07-182-0/+48
* amdgcn: Use new workitem intrinsicsMatt Arsenault2016-02-173-0/+62
* Split sources for amdgcn and r600Matt Arsenault2016-02-133-0/+33
OpenPOWER on IntegriCloud