[Openmp-commits] [PATCH] D140783: [OpenMP][AMDGPU] Extract code object version from the ELF

[Jon Chesterfield via Phabricator via Openmp-commits]

JonChesterfield added inline comments.


================
Comment at: openmp/libomptarget/plugins/amdgpu/src/rtl.cpp:128
   std::queue<int> FreeKernargSegments;
+  uint16_t CodeObjectVersion;
 
----------------
I don't think the memory pool needs to know what format the elf is using


================
Comment at: openmp/libomptarget/plugins/amdgpu/src/rtl.cpp:481
   std::vector<std::string> TargetID;
+  uint16_t CodeObjectVersion;
 
----------------
This is per code object, how does a global scalar capture that?


================
Comment at: openmp/libomptarget/plugins/amdgpu/src/rtl.cpp:1382
+
+  if (const auto *ELFObj = dyn_cast<ELF64LEObjectFile>(ElfOrErr->get())) {
+    auto Header = ELFObj->getELFFile().getHeader();
----------------
This seems likely to be something already available in the elf library 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D140783/new/

https://reviews.llvm.org/D140783