[PATCH] D50222: [CodeGen] [SelectionDAG] More efficient code for X % C == 0
Dmytro Shynkevych via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Aug 2 21:07:16 PDT 2018
hermord added inline comments.
================
Comment at: test/CodeGen/Hexagon/swp-const-tc2.ll:9
; CHECK-NOT: loop0
-; CHECK: = mpy
-; CHECK-NOT: = mpy
+; CHECK: = memw
+; CHECK-NOT: = memw
----------------
To the best of my understanding, it is valid to not perform the multiplication at all, since its can never be used because of the infinite loop at `b3`. This is the output I get:
```
// %bb.0: // %b0
{
r0 = ##-1431655765
}
{
r0 = #0
}
{
r0 = memw(r0+#0)
}
.p2align 4
.LBB0_1: // %b3
// =>This Inner Loop Header: Depth=1
{
jump .LBB0_1
}
```
Repository:
rL LLVM
https://reviews.llvm.org/D50222
More information about the llvm-commits
mailing list