summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll
Commit message (Collapse)AuthorAgeFilesLines
* AMDGPU: Add intrinsics for address space identificationMatt Arsenault2019-09-051-0/+19
| | | | | | | The library currently uses ptrtoint and directly checks the queue ptr for this, which counts as a pointer capture. llvm-svn: 371009
* [AMDGPU] Change constant addr space to 4Yaxun Liu2018-02-131-17/+17
| | | | | | Differential Revision: https://reviews.llvm.org/D43170 llvm-svn: 325030
* [AMDGPU] Switch to the new addr space mapping by defaultYaxun Liu2018-02-021-23/+23
| | | | | | | | This requires corresponding clang change. Differential Revision: https://reviews.llvm.org/D40955 llvm-svn: 324101
* AMDGPU: Detect kernarg segment pointerMatt Arsenault2017-07-141-0/+11
| | | | | | | | This is necessary to pass the kernarg segment pointer to callee functions. Also don't unconditionally enable for kernels. llvm-svn: 307978
* AMDGPU: Make intrinsics speculatableMatt Arsenault2017-05-021-2/+2
| | | | llvm-svn: 301937
* AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernelMatt Arsenault2017-03-211-50/+50
| | | | | | | | | | | | Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444
* AMDGPU: Implement addrspacecastMatt Arsenault2016-04-251-0/+57
| | | | llvm-svn: 267452
* AMDGPU: Add queue ptr intrinsicMatt Arsenault2016-04-251-0/+11
| | | | llvm-svn: 267451
* AMDGPU: Stop checking intrinsics not used by HSA for dispatch-ptrMatt Arsenault2016-01-301-0/+170
Only the dispatch.ptr intrinsic is supposed to be used now to get the workgroup size, and the read.local.size intrinsics do not work correctly. llvm-svn: 259296
OpenPOWER on IntegriCloud