[llvm] [VP][RISCV] Introduce llvm.vp.minimum/maximum intrinsics (PR #74840)

[Craig Topper via llvm-commits]

================
@@ -0,0 +1,1245 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+zvfh,+v,+m -target-abi=ilp32d \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,ZVFH
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfh,+v,+m -target-abi=lp64d \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,ZVFH
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+zvfhmin,+v,+m -target-abi=ilp32d \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,ZVFHMIN
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+zvfhmin,+v,+m -target-abi=lp64d \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,ZVFHMIN
+
+declare <vscale x 1 x half> @llvm.vp.maximum.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x half> @vfmax_vv_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; ZVFH-LABEL: vfmax_vv_nxv1f16:
+; ZVFH:       # %bb.0:
+; ZVFH-NEXT:    vmv1r.v v10, v0
+; ZVFH-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; ZVFH-NEXT:    vmfeq.vv v0, v8, v8, v0.t
+; ZVFH-NEXT:    vmerge.vvm v11, v8, v9, v0
+; ZVFH-NEXT:    vmv1r.v v0, v10
+; ZVFH-NEXT:    vmfeq.vv v0, v9, v9, v0.t
+; ZVFH-NEXT:    vmerge.vvm v8, v9, v8, v0
+; ZVFH-NEXT:    vmv1r.v v0, v10
+; ZVFH-NEXT:    vfmax.vv v8, v8, v11, v0.t
+; ZVFH-NEXT:    ret
+;
+; ZVFHMIN-LABEL: vfmax_vv_nxv1f16:
+; ZVFHMIN:       # %bb.0:
+; ZVFHMIN-NEXT:    vmv1r.v v10, v0
+; ZVFHMIN-NEXT:    vsetvli a1, zero, e16, mf4, ta, ma
+; ZVFHMIN-NEXT:    vfwcvt.f.f.v v11, v8
+; ZVFHMIN-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; ZVFHMIN-NEXT:    vmfeq.vv v0, v11, v11, v0.t
+; ZVFHMIN-NEXT:    vsetvli a1, zero, e16, mf4, ta, ma
+; ZVFHMIN-NEXT:    vfwcvt.f.f.v v8, v9
----------------
topperc wrote:

This is a VLMAX vfwcvt.f.f.v, it should use the VL from the VP intrinsic.

https://github.com/llvm/llvm-project/pull/74840