[PATCH] D62911: WIP: AMDGPU: Use fixup for local linkage functions

[Matt Arsenault via Phabricator via llvm-commits]

arsenm marked an inline comment as done.
arsenm added inline comments.


================
Comment at: lib/Target/AMDGPU/AMDGPUMCInstLower.cpp:156
+    if (MO.getTargetFlags() == SIInstrInfo::MO_PCREL32_HI)
+      Expr = AMDGPUMCExpr::create(AMDGPUMCExpr::VK_AMDGPU_PCREL_HI32, Expr, Ctx);
+
----------------
hliao wrote:
> Why not use `MCBinaryExpr::createAShr` to shift that high bits into low bits directly? We don't need invent a new target fixup.
That might work. Do we need an explicit truncate from 64-bit to 32-bit operator though?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D62911/new/

https://reviews.llvm.org/D62911