[PATCH] D43735: [AMDGPU] Remove use of OpenCL triple environment and replace with function attribute for AMDGPU

[Konstantin Zhuravlyov via Phabricator via llvm-commits]

kzhuravl requested changes to this revision.
kzhuravl added inline comments.
This revision now requires changes to proceed.


================
Comment at: docs/UsersManual.rst:2182
      $ clang -target nvptx64-unknown-unknown test.cl
-     $ clang -target amdgcn-amd-amdhsa-opencl test.cl
+     $ clang -target amdgcn-amd-amdhsa -mcpu gfx900 test.cl
 
----------------
-mcpu=gfx900


================
Comment at: docs/UsersManual.rst:2290
 
-     $ clang -target amdgcn-amd-amdhsa-opencl test.cl
+     $ clang -target amdgcn-amd-amdhsa- -mcpu gfx900 test.cl
 
----------------
-mcpu=gfx900


================
Comment at: lib/CodeGen/TargetInfo.cpp:7660
+
+  if (M.getLangOpts().OpenCL && FD->hasAttr<OpenCLKernelAttr>() && (M.getTriple().getOS() == llvm::Triple::AMDHSA))
+    F->addFnAttr("amdgpu-implicitarg-num-bytes", "32");
----------------
Make <= 80 chars long.


https://reviews.llvm.org/D43735