[llvm-branch-commits] [llvm] PeepholeOpt: Fix looking for def of current copy to coalesce (PR #125533)
Pengcheng Wang via llvm-branch-commits
llvm-branch-commits at lists.llvm.org
Tue Feb 4 22:55:36 PST 2025
================
@@ -1891,31 +1886,44 @@ define <32 x i64> @vp_cttz_v32i64(<32 x i64> %va, <32 x i1> %m, i32 zeroext %evl
; RV32-NEXT: addi a0, a0, 48
; RV32-NEXT: vl8r.v v16, (a0) # Unknown-size Folded Reload
; RV32-NEXT: vand.vv v16, v24, v16, v0.t
-; RV32-NEXT: vsub.vv v24, v8, v16, v0.t
+; RV32-NEXT: vsub.vv v8, v8, v16, v0.t
; RV32-NEXT: csrr a0, vlenb
-; RV32-NEXT: slli a0, a0, 5
+; RV32-NEXT: li a1, 40
+; RV32-NEXT: mul a0, a0, a1
; RV32-NEXT: add a0, sp, a0
; RV32-NEXT: addi a0, a0, 48
-; RV32-NEXT: vl8r.v v8, (a0) # Unknown-size Folded Reload
-; RV32-NEXT: vand.vv v16, v24, v8, v0.t
+; RV32-NEXT: vs8r.v v8, (a0) # Unknown-size Folded Spill
----------------
wangpc-pp wrote:
For RISCV, this is a regerssion since it requires more spill/reload.
https://github.com/llvm/llvm-project/pull/125533
More information about the llvm-branch-commits
mailing list