| Issue |
209237
|
| Summary |
RISC-V performance regression in telecomm-gsm benchmark after a00f7be392d9
|
| Labels |
backend:RISC-V,
llvm:SLPVectorizer
|
| Assignees |
|
| Reporter |
lukel97
|
telecomm-gsm (and toast, which contains the same function) are 8% slower after a00f7be392d9: https://cc-perf.igalia.com/db_default/v4/nts/703
It's caused by unprofitable SLP vectorization.
I think we're overcosting the scalar llvm.sadd.sat.i16: https://godbolt.org/z/ffadqhEW1
We're also probably undercosting the roots, we end up with both a `<1 x i16>` and `i16` store somehow:
```asm
vmv.x.s a0, v8
vslidedown.vi v8, v8, 1
vsetivli zero, 1, e16, mf4, ta, ma
vse16.v v8, (a3)
sh a0, 0(a4)
```
![]()
_______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
