Re: [PATCH] vect+aarch64: Fix ldp_stp_* regressions

On Mon, 14 Feb 2022, Richard Sandiford wrote: > ldp_stp_1.c, ldp_stp_4.c and ldp_stp_5.c have been failing since > vectorisation was enabled at -O2. In all three cases SLP is > generating vector code when scalar code would be better. > > The problem is that the target costs do not model whether

Re: [PATCH] vect+aarch64: Fix ldp_stp_* regressions

Richard Biener writes: > On Mon, 14 Feb 2022, Richard Sandiford wrote: > >> ldp_stp_1.c, ldp_stp_4.c and ldp_stp_5.c have been failing since >> vectorisation was enabled at -O2. In all three cases SLP is >> generating vector code when scalar code would be better. >> >> The problem is that the ta

Re: [PATCH] vect+aarch64: Fix ldp_stp_* regressions

On Mon, 14 Feb 2022, Richard Sandiford wrote: > ldp_stp_1.c, ldp_stp_4.c and ldp_stp_5.c have been failing since > vectorisation was enabled at -O2. In all three cases SLP is > generating vector code when scalar code would be better. > > The problem is that the target costs do not model whether

[PATCH] vect+aarch64: Fix ldp_stp_* regressions

ldp_stp_1.c, ldp_stp_4.c and ldp_stp_5.c have been failing since vectorisation was enabled at -O2. In all three cases SLP is generating vector code when scalar code would be better. The problem is that the target costs do not model whether STP could be used for the scalar or vector code, so the n