[PATCH] D71365: expand printf when compiling HIP to AMDGPU

[Sameer Sahasrabuddhe via Phabricator via cfe-commits]

sameerds created this revision.
Herald added subscribers: llvm-commits, cfe-commits, hiraditya, t-tye, tpr, dstuttard, yaxunl, mgorny, nhaehnle, wdng, jvesely, kzhuravl.
Herald added projects: clang, LLVM.

This change implements the expansion in two parts:

- Add a utility function emitAMDGPUPrintfCall() in LLVM.
- Invoke the above function from Clang CodeGen, when processing a HIP program for the AMDGPU target.

The printf expansion has undefined behaviour if the format string is
not a compile-time constant. As a sufficient condition, the HIP
ToolChain now emits -Werror=format-nonliteral.

clang format


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D71365

Files:
  clang/lib/CodeGen/CGBuiltin.cpp
  clang/lib/CodeGen/CGGPUBuiltin.cpp
  clang/lib/CodeGen/CodeGenFunction.h
  clang/lib/Driver/ToolChains/HIP.cpp
  clang/test/CodeGenHIP/printf-aggregate.cpp
  clang/test/CodeGenHIP/printf.cpp
  clang/test/Driver/hip-printf.hip
  llvm/include/llvm/Transforms/Utils/AMDGPUEmitPrintf.h
  llvm/lib/Transforms/Utils/AMDGPUEmitPrintf.cpp
  llvm/lib/Transforms/Utils/CMakeLists.txt

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D71365.233399.patch
Type: text/x-patch
Size: 17482 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20191211/be3d042c/attachment-0001.bin>