Skip to content

Conversation

@jacquesguan
Copy link
Contributor

This patch supports STRICT_FP_ROUND and STRICT_FP_EXTEND when we only have Zvfhmin but no Zvfh.

…vfhmin

This patch supports STRICT_FP_ROUND and STRICT_FP_EXTEND when we only have Zvfhmin but no Zvfh
@llvmbot
Copy link
Member

llvmbot commented Oct 9, 2023

@llvm/pr-subscribers-backend-risc-v

Changes

This patch supports STRICT_FP_ROUND and STRICT_FP_EXTEND when we only have Zvfhmin but no Zvfh.


Full diff: https://github.com/llvm/llvm-project/pull/68559.diff

3 Files Affected:

  • (modified) llvm/lib/Target/RISCV/RISCVISelLowering.cpp (+4)
  • (modified) llvm/test/CodeGen/RISCV/rvv/fixed-vectors-vfptrunc-constrained-sdnode.ll (+4)
  • (modified) llvm/test/CodeGen/RISCV/rvv/vfptrunc-constrained-sdnode.ll (+4)
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index b66a596f1cbe692..241d2f3de4d76cf 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -941,6 +941,8 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
         if (!isTypeLegal(VT))
           continue;
         setOperationAction({ISD::FP_ROUND, ISD::FP_EXTEND}, VT, Custom);
+        setOperationAction({ISD::STRICT_FP_ROUND, ISD::STRICT_FP_EXTEND}, VT,
+                           Custom);
         setOperationAction({ISD::VP_FP_ROUND, ISD::VP_FP_EXTEND}, VT, Custom);
         setOperationAction({ISD::VP_MERGE, ISD::VP_SELECT, ISD::SELECT}, VT,
                            Custom);
@@ -1153,6 +1155,8 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
         if (VT.getVectorElementType() == MVT::f16 &&
             !Subtarget.hasVInstructionsF16()) {
           setOperationAction({ISD::FP_ROUND, ISD::FP_EXTEND}, VT, Custom);
+          setOperationAction({ISD::STRICT_FP_ROUND, ISD::STRICT_FP_EXTEND}, VT,
+                             Custom);
           setOperationAction({ISD::VP_FP_ROUND, ISD::VP_FP_EXTEND}, VT, Custom);
           setOperationAction(
               {ISD::VP_MERGE, ISD::VP_SELECT, ISD::VSELECT, ISD::SELECT}, VT,
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-vfptrunc-constrained-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-vfptrunc-constrained-sdnode.ll
index 32a050800b9792b..fd53113741de00c 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-vfptrunc-constrained-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-vfptrunc-constrained-sdnode.ll
@@ -3,6 +3,10 @@
 ; RUN:     -verify-machineinstrs < %s | FileCheck %s
 ; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfh,+v -target-abi=lp64d \
 ; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+zvfhmin,+v -target-abi=ilp32d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfhmin,+v -target-abi=lp64d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
 
 declare <2 x float> @llvm.experimental.constrained.fptrunc.v2f32.v2f64(<2 x double>, metadata, metadata)
 define <2 x float> @vfptrunc_v2f64_v2f32(<2 x double> %va) strictfp {
diff --git a/llvm/test/CodeGen/RISCV/rvv/vfptrunc-constrained-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vfptrunc-constrained-sdnode.ll
index 72bf2b94e6f9f02..4404a275858f23a 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vfptrunc-constrained-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vfptrunc-constrained-sdnode.ll
@@ -3,6 +3,10 @@
 ; RUN:     -verify-machineinstrs < %s | FileCheck %s
 ; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfh,+v -target-abi=lp64d \
 ; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+zvfhmin,+v -target-abi=ilp32d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfhmin,+v -target-abi=lp64d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
 
 declare <vscale x 1 x float> @llvm.experimental.constrained.fptrunc.nxv1f32.nxv1f64(<vscale x 1 x double>, metadata, metadata)
 define <vscale x 1 x float> @vfptrunc_nxv1f64_nxv1f32(<vscale x 1 x double> %va) strictfp {

Copy link
Contributor

@michaelmaitland michaelmaitland left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants