[PATCH] D135194: [AArch64][NFC] Precommit test case to show sub-optimal codegen for add(lsl(val1,small-shift), lsl(val2,large-shift)).

Mingming Liu via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Oct 4 13:59:21 PDT 2022


mingmingl created this revision.
Herald added a subscriber: kristof.beyls.
Herald added a project: All.
mingmingl requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

Ideally, add operand with smaller shift should be RHS. In that way,
smaller-shift is folded into ADD.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D135194

Files:
  llvm/test/CodeGen/AArch64/logical_shifted_reg.ll


Index: llvm/test/CodeGen/AArch64/logical_shifted_reg.ll
===================================================================
--- llvm/test/CodeGen/AArch64/logical_shifted_reg.ll
+++ llvm/test/CodeGen/AArch64/logical_shifted_reg.ll
@@ -289,4 +289,16 @@
   ret void
 }
 
+define i64 @add_swap_rhs_lhs(i64 %0, i64 %1) {
+; CHECK-LABEL: add_swap_rhs_lhs:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    lsl x8, x1, #3
+; CHECK-NEXT:    add x0, x8, x0, lsl #8
+; CHECK-NEXT:    ret
+  %3 = shl i64 %0, 8
+  %4 = shl i64 %1, 3
+  %5 = add i64 %4, %3
+  ret i64 %5
+}
+
 !1 = !{!"branch_weights", i32 1, i32 1}


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D135194.465148.patch
Type: text/x-patch
Size: 599 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20221004/ab70fb8e/attachment.bin>


More information about the llvm-commits mailing list