[llvm] d48f6df - [RISCV] Create the correct mask type when lowering EXTRACT_VECTOR_ELT

[Fraser Cormack via llvm-commits]

Author: Fraser Cormack
Date: 2021-09-27T09:43:40+01:00
New Revision: d48f6df1f8ef84d97856697a6366f440f0d9bb54

URL: https://github.com/llvm/llvm-project/commit/d48f6df1f8ef84d97856697a6366f440f0d9bb54
DIFF: https://github.com/llvm/llvm-project/commit/d48f6df1f8ef84d97856697a6366f440f0d9bb54.diff

LOG: [RISCV] Create the correct mask type when lowering EXTRACT_VECTOR_ELT

This particular case was creating a `VMSET_VL` using the old
fixed-length type in order to pass a mask to other custom nodes
operating on the scalable container type. This kind of thing wasn't
caught for us; I only noticed when experimenting with odd-length
vectors, where it was trying to generate an invalid `v3i1` MVT.

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D110420

Added: 
    

Modified: 
    llvm/lib/Target/RISCV/RISCVISelLowering.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index 646346d4acfa..ff8d2d948c5c 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -5791,7 +5791,7 @@ void RISCVTargetLowering::ReplaceNodeResults(SDNode *N,
     MVT XLenVT = Subtarget.getXLenVT();
 
     // Use a VL of 1 to avoid processing more elements than we need.
-    MVT MaskVT = MVT::getVectorVT(MVT::i1, VecVT.getVectorElementCount());
+    MVT MaskVT = MVT::getVectorVT(MVT::i1, ContainerVT.getVectorElementCount());
     SDValue VL = DAG.getConstant(1, DL, XLenVT);
     SDValue Mask = DAG.getNode(RISCVISD::VMSET_VL, DL, MaskVT, VL);