diff options
| author | Tony Tye <Tony.Tye@amd.com> | 2018-03-23 18:45:18 +0000 |
|---|---|---|
| committer | Tony Tye <Tony.Tye@amd.com> | 2018-03-23 18:45:18 +0000 |
| commit | 7a893d4e34e7261368c58ad583dfaa600108b7e1 (patch) | |
| tree | fb50022d2ccd9637c6baf7cb492e03b6f0e7648b /llvm/docs | |
| parent | a6fb536e5b5c98d3ca46fbd7ea21893f633d21d8 (diff) | |
| download | bcm5719-llvm-7a893d4e34e7261368c58ad583dfaa600108b7e1.tar.gz bcm5719-llvm-7a893d4e34e7261368c58ad583dfaa600108b7e1.zip | |
[AMDGPU] Remove use of OpenCL triple environment and replace with function attribute for AMDGPU
- Remove use of the opencl and amdopencl environment member of the target triple for the AMDGPU target.
- Use function attribute to communicate to the AMDGPU backend to add implicit arguments for OpenCL kernels for the AMDHSA OS.
Differential Revision: https://reviews.llvm.org/D43736
llvm-svn: 328349
Diffstat (limited to 'llvm/docs')
| -rw-r--r-- | llvm/docs/AMDGPUUsage.rst | 36 |
1 files changed, 18 insertions, 18 deletions
diff --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst index d0671b0a74e..6bb64c9e061 100644 --- a/llvm/docs/AMDGPUUsage.rst +++ b/llvm/docs/AMDGPUUsage.rst @@ -64,9 +64,7 @@ specify the target triple: ============ ============================================================== Environment Description ============ ============================================================== - *<empty>* Defaults to ``opencl``. - ``opencl`` OpenCL compute kernel (see :ref:`amdgpu-opencl`). - ``hcc`` AMD HC language compute kernel (see :ref:`amdgpu-hcc`). + *<empty>* Default. ============ ============================================================== .. _amdgpu-processors: @@ -3787,35 +3785,37 @@ Source Languages OpenCL ------ -When generating code for the OpenCL language the target triple environment -should be ``opencl`` or ``amdgizcl`` (see :ref:`amdgpu-target-triples`). - When the language is OpenCL the following differences occur: 1. The OpenCL memory model is used (see :ref:`amdgpu-amdhsa-memory-model`). -2. The AMDGPU backend adds additional arguments to the kernel. +2. The AMDGPU backend appends additional arguments to the kernel's explicit + arguments for the AMDHSA OS (see + :ref:`opencl-kernel-implicit-arguments-appended-for-amdhsa-os-table`). 3. Additional metadata is generated - (:ref:`amdgpu-amdhsa-hsa-code-object-metadata`). + (see :ref:`amdgpu-amdhsa-hsa-code-object-metadata`). -.. TODO - Specify what affect this has. Hidden arguments added. Additional metadata - generated. + .. table:: OpenCL kernel implicit arguments appended for AMDHSA OS + :name: opencl-kernel-implicit-arguments-appended-for-amdhsa-os-table + + ======== ==== ========= =========================================== + Position Byte Byte Description + Size Alignment + ======== ==== ========= =========================================== + 0 8 8 OpenCL Global Offset X + 1 8 8 OpenCL Global Offset Y + 2 8 8 OpenCL Global Offset Z + 3 8 8 OpenCL printf buffer + ======== ==== ========= =========================================== .. _amdgpu-hcc: HCC --- -When generating code for the OpenCL language the target triple environment -should be ``hcc`` (see :ref:`amdgpu-target-triples`). - -When the language is OpenCL the following differences occur: +When the language is HCC the following differences occur: 1. The HSA memory model is used (see :ref:`amdgpu-amdhsa-memory-model`). -.. TODO - Specify what affect this has. - Assembler --------- |
