On Fri, Jun 21, 2019 at 4:21 AM Hongtao Liu <crazy...@gmail.com> wrote: > > On Thu, Jun 20, 2019 at 10:58 PM H.J. Lu <hjl.to...@gmail.com> wrote: > > > > On Thu, Jun 20, 2019 at 3:54 AM Hongtao Liu <crazy...@gmail.com> wrote: > > > > > > On Thu, Jun 20, 2019 at 2:13 PM Uros Bizjak <ubiz...@gmail.com> wrote: > > > > > > > > On Thu, Jun 20, 2019 at 7:36 AM Hongtao Liu <crazy...@gmail.com> wrote: > > > > > > > > > > On Sat, Jun 8, 2019 at 4:12 AM Uros Bizjak <ubiz...@gmail.com> wrote: > > > > > > > > > > > > On 6/7/19, H.J. Lu <hjl.to...@gmail.com> wrote: > > > > > > > > > > > > >> > > +/* Register pair. */ > > > > > > >> > > +VECTOR_MODES_WITH_PREFIX (P, INT, 2); /* P2QI */ > > > > > > >> > > +VECTOR_MODES_WITH_PREFIX (P, INT, 4); /* P2HI P4QI */ > > > > > > >> > > > > > > > > >> > > I think > > > > > > >> > > > > > > > > >> > > INT_MODE (P2QI, 16); > > > > > > >> > > INT_MODE (P2HI, 32); > Why P2QI need 16 bytes but not 2 bytes? > Same question with P2HI.
Because we made a mistake. It should be 2 and 4, since these arguments are bytes, not bits. This will also fix gen_highpart issue. Uros. > > > > > > >> > > > > > > > > >> > > with the above subreg approach should work. > > > Yes, it works. > > > > > > But i didn't figure out how did pass_reload correctly handle such subreg, > > > do you have suggestions such as "which function i can dig into first" or > > > "which piece of codes handle subreg"? > > > > You need to define REGMODE_NATURAL_SIZE. > > > > -- > > H.J. > > > > -- > BR, > Hongtao
