[llvm] [RISCV] Add vmv.x.s to RISCVOptWInstrs. (PR #74519)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

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

Author: Craig Topper (topperc)

<details>
<summary>Changes</summary>

This instruction produces a sign extend value if the SEW is less than or equal to 32.

---
Full diff: https://github.com/llvm/llvm-project/pull/74519.diff


2 Files Affected:

- (modified) llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp (+12) 
- (modified) llvm/test/CodeGen/RISCV/sextw-removal.ll (+64-3) 


``````````diff
diff --git a/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp b/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp
index b667a2b7a11ba..174a85d8dee55 100644
--- a/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp
+++ b/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp
@@ -368,6 +368,18 @@ static bool isSignExtendingOpW(const MachineInstr &MI,
     return MI.getOperand(1).getReg() == RISCV::X0;
   case RISCV::PseudoAtomicLoadNand32:
     return true;
+  case RISCV::PseudoVMV_X_S_MF8:
+  case RISCV::PseudoVMV_X_S_MF4:
+  case RISCV::PseudoVMV_X_S_MF2:
+  case RISCV::PseudoVMV_X_S_M1:
+  case RISCV::PseudoVMV_X_S_M2:
+  case RISCV::PseudoVMV_X_S_M4:
+  case RISCV::PseudoVMV_X_S_M8: {
+    // vmv.x.s returns a sign extended value if log2(sew) <= 5.
+    int64_t Log2SEW = MI.getOperand(2).getImm();
+    assert(Log2SEW >= 3 && Log2SEW <= 6 && "Unexpected Log2SEW");
+    return Log2SEW <= 5;
+  }
   }
 
   return false;
diff --git a/llvm/test/CodeGen/RISCV/sextw-removal.ll b/llvm/test/CodeGen/RISCV/sextw-removal.ll
index 3babef93499c8..e730092674e06 100644
--- a/llvm/test/CodeGen/RISCV/sextw-removal.ll
+++ b/llvm/test/CodeGen/RISCV/sextw-removal.ll
@@ -1,9 +1,9 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+f,+zknh -target-abi=lp64f \
+; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+f,+zknh,+v -target-abi=lp64f \
 ; RUN:   | FileCheck %s --check-prefixes=CHECK,RV64I
-; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+zba,+zbb,+f,+zknh -target-abi=lp64f \
+; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+zba,+zbb,+f,+zknh,+v -target-abi=lp64f \
 ; RUN:   | FileCheck %s --check-prefixes=CHECK,RV64ZBB
-; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+zba,+zbb,+f,+zknh -target-abi=lp64f \
+; RUN: llc < %s -mtriple=riscv64 -mattr=+m,+zba,+zbb,+f,+zknh,+v -target-abi=lp64f \
 ; RUN:   -riscv-disable-sextw-removal | FileCheck %s --check-prefix=NOREMOVAL
 
 define void @test1(i32 signext %arg, i32 signext %arg1) nounwind {
@@ -1436,3 +1436,64 @@ bb7:                                              ; preds = %bb2
 }
 
  declare void @side_effect(i64)
+
+define void @test20(<vscale x 1 x i32> %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test20:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi sp, sp, -32
+; CHECK-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
+; CHECK-NEXT:    sd s0, 16(sp) # 8-byte Folded Spill
+; CHECK-NEXT:    sd s1, 8(sp) # 8-byte Folded Spill
+; CHECK-NEXT:    mv s0, a0
+; CHECK-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
+; CHECK-NEXT:    vmv.x.s s1, v8
+; CHECK-NEXT:  .LBB24_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    mv a0, s1
+; CHECK-NEXT:    call bar at plt
+; CHECK-NEXT:    sllw s1, s1, s0
+; CHECK-NEXT:    bnez a0, .LBB24_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
+; CHECK-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
+; CHECK-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
+; CHECK-NEXT:    addi sp, sp, 32
+; CHECK-NEXT:    ret
+;
+; NOREMOVAL-LABEL: test20:
+; NOREMOVAL:       # %bb.0: # %bb
+; NOREMOVAL-NEXT:    addi sp, sp, -32
+; NOREMOVAL-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
+; NOREMOVAL-NEXT:    sd s0, 16(sp) # 8-byte Folded Spill
+; NOREMOVAL-NEXT:    sd s1, 8(sp) # 8-byte Folded Spill
+; NOREMOVAL-NEXT:    mv s0, a0
+; NOREMOVAL-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
+; NOREMOVAL-NEXT:    vmv.x.s s1, v8
+; NOREMOVAL-NEXT:  .LBB24_1: # %bb2
+; NOREMOVAL-NEXT:    # =>This Inner Loop Header: Depth=1
+; NOREMOVAL-NEXT:    sext.w a0, s1
+; NOREMOVAL-NEXT:    call bar at plt
+; NOREMOVAL-NEXT:    sllw s1, s1, s0
+; NOREMOVAL-NEXT:    bnez a0, .LBB24_1
+; NOREMOVAL-NEXT:  # %bb.2: # %bb7
+; NOREMOVAL-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
+; NOREMOVAL-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
+; NOREMOVAL-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
+; NOREMOVAL-NEXT:    addi sp, sp, 32
+; NOREMOVAL-NEXT:    ret
+bb:
+  %i = call i32 @llvm.riscv.vmv.x.s.nxv1i32(<vscale x 1 x i32> %arg)
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call signext i32 @bar(i32 signext %i3)
+  %i5 = shl i32 %i3, %arg1
+  %i6 = icmp eq i32 %i4, 0
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+declare i32 @llvm.riscv.vmv.x.s.nxv1i32( <vscale x 1 x i32>)

``````````

</details>


https://github.com/llvm/llvm-project/pull/74519