Commit 8789d19
committed
[AArch64] Use i32 extract from UADDV in popcount lowering.
We need the top bits to be zeroes, but an v8i8->i32 EXTRACT_VECTOR_ELT will
anyext into the top bits. The instruction we create (UADDV) is known to be
zeroes in the upper bits, so we can convert to a larger v2i32 vector and
extract from there, similar to the operation currently performed for i64 types.1 parent e233002 commit 8789d19
File tree
2 files changed
+6
-9
lines changed- llvm
- lib/Target/AArch64
- test/CodeGen/AArch64
2 files changed
+6
-9
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
10852 | 10852 | | |
10853 | 10853 | | |
10854 | 10854 | | |
10855 | | - | |
10856 | | - | |
10857 | | - | |
10858 | | - | |
10859 | | - | |
10860 | | - | |
10861 | | - | |
| 10855 | + | |
| 10856 | + | |
| 10857 | + | |
| 10858 | + | |
10862 | 10859 | | |
10863 | 10860 | | |
10864 | 10861 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
686 | 686 | | |
687 | 687 | | |
688 | 688 | | |
689 | | - | |
| 689 | + | |
690 | 690 | | |
691 | 691 | | |
692 | 692 | | |
| |||
701 | 701 | | |
702 | 702 | | |
703 | 703 | | |
704 | | - | |
| 704 | + | |
705 | 705 | | |
706 | 706 | | |
707 | 707 | | |
| |||
0 commit comments