[PATCH] D76357: [AMDGPU] Implement wave32 DWARF register mapping

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Mar 19 10:17:57 PDT 2020


arsenm added inline comments.


================
Comment at: llvm/unittests/MC/AMDGPU/DwarfRegMappings.cpp:44
+
+  return std::unique_ptr<const MCRegisterInfo>(TM->getMCRegisterInfo());
+}
----------------
This isn't creating an owning reference? Why is this unique ptr?


================
Comment at: llvm/unittests/MC/AMDGPU/DwarfRegMappings.cpp:49
+  // Test triple OS strings
+  std::vector<std::string> OSTestValues = {"", "amdhsa", "amdpal"};
+
----------------
The multiple OS tests aren't needed, my point was it should just be set to something sensible other than the defaulted host OS


================
Comment at: llvm/unittests/MC/AMDGPU/DwarfRegMappings.cpp:55
+  // A0 => 3072, A255 => 3327
+  std::vector<int> wave64DwarfTestValues = {16,   17,   32,   95,   1088,
+                                            1129, 2560, 2815, 3072, 3327};
----------------
This can be a regular array


================
Comment at: llvm/unittests/MC/AMDGPU/DwarfRegMappings.cpp:79
+  // A0 => 2048, A255 => 2303
+  std::vector<int> wave32DwarfTestValues = {0,    1,    32,   95,   1088,
+                                            1129, 1536, 1791, 2048, 2303};
----------------
Regular array


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76357/new/

https://reviews.llvm.org/D76357





More information about the llvm-commits mailing list