On Fri, Aug 29, 2025 at 12:39 PM Robin Dapp <rdapp....@gmail.com> wrote: > > > This is v2, changed to estimated_poly_value / 2. Regtested on > > rv64gcv_zvl512b > > and aarch64 via qemu. > > I looked at the test suite results more closely now. While those apply_scale > ICEs vanish with the patch there are still a few execution failures with the > VLA peeling patch remaining: > > One is gcc.c-torture/execute/memset-1.c and that one doesn't look as > straightforward. Will have a look as soon as I get to it.
Yeah, it's not a strict bound, so the function needs to return -1 aka UNKNOWN. But how this -1 should be interpreted differs on context. But for sure -1 cannot be interpreted as actual bound. For frequency scaling I'd use the same logic as for costing - use estimated_poly_value / 2 Richard. > -- > Regards > Robin >
