[llvm] [RISCV] Re-model RVV comparison instructions (PR #88868)

[Pengcheng Wang via llvm-commits]

================
@@ -585,13 +569,25 @@ define <16 x double> @vfmax_vv_v16f64(<16 x double> %va, <16 x double> %vb, <16
 define <16 x double> @vfmax_vv_v16f64_unmasked(<16 x double> %va, <16 x double> %vb, i32 zeroext %evl) {
 ; CHECK-LABEL: vfmax_vv_v16f64_unmasked:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 1
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x02, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 2 * vlenb
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; CHECK-NEXT:    vmfeq.vv v0, v8, v8
-; CHECK-NEXT:    vmfeq.vv v7, v16, v16
+; CHECK-NEXT:    vmfeq.vv v24, v16, v16
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vs1r.v v24, (a0) # Unknown-size Folded Spill
 ; CHECK-NEXT:    vmerge.vvm v24, v8, v16, v0
-; CHECK-NEXT:    vmv1r.v v0, v7
+; CHECK-NEXT:    vl1r.v v0, (a0) # Unknown-size Folded Reload
----------------
wangpc-pp wrote:

An example of regression here.

https://github.com/llvm/llvm-project/pull/88868