[llvm] 681a161 - AMDGPU: Remove outdated comment
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Mon May 18 09:20:24 PDT 2020
Author: Matt Arsenault
Date: 2020-05-18T12:06:16-04:00
New Revision: 681a161ff5419404ac1b3221e9a01ade25802998
URL: https://github.com/llvm/llvm-project/commit/681a161ff5419404ac1b3221e9a01ade25802998
DIFF: https://github.com/llvm/llvm-project/commit/681a161ff5419404ac1b3221e9a01ade25802998.diff
LOG: AMDGPU: Remove outdated comment
Added:
Modified:
llvm/lib/Target/AMDGPU/SIISelLowering.cpp
Removed:
################################################################################
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index 803b188b9488..2eb58841749b 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -4711,13 +4711,6 @@ SDValue SITargetLowering::LowerBRCOND(SDValue BRCOND,
Target = BR->getOperand(1);
}
- // FIXME: This changes the types of the intrinsics instead of introducing new
- // nodes with the correct types.
- // e.g. llvm.amdgcn.loop
-
- // eg: i1,ch = llvm.amdgcn.loop t0, TargetConstant:i32<6271>, t3
- // => t9: ch = llvm.amdgcn.loop t0, TargetConstant:i32<6271>, t3, BasicBlock:ch<bb1 0x7fee5286d088>
-
unsigned CFNode = isCFIntrinsic(Intr);
if (CFNode == 0) {
// This is a uniform branch so we don't need to legalize.
More information about the llvm-commits
mailing list