)
On Fri, 10 Oct 2025 at 11:05, Richard Biener <[email protected]> wrote:
>
> On Fri, Oct 10, 2025 at 10:56 AM Richard Biener
> <[email protected]> wrote:
> >
> > On Thu, Oct 9, 2025 at 10:58 PM Robin Dapp <[email protected]> wrote:
> > >
> > > Hi,
> > >
> > > In gcc-16-4314-g5e9eecc6686 I meant to remove all uses of TYPE
> > > in support_vector_misalignment but apparently forgot this one.
> > > Fixing by using the inner mode's size.
> > >
> > > I'm a bit short on build machines right now so this is untested.
> > > Is it still OK?
> >
> > I think you want GET_MODE_BITSIZE here?
>
> Ah no, scratch that.
>
At least, that's consistent with the change for risc-v in
gcc-16-4314-g5e9eecc6686 ;-)
Is GET_MODE_UNIT_SIZE (mode) always equal to TYPE_ALIGN_UNIT (type) at
this point?
(Sorry I didn't follow the discussions about gcc-16-4314-g5e9eecc6686)
> I think your patch is OK. Build on arm is broken right now, so it can't
> get worse.
:-) Indeed, it was broken by gcc-16-4314-g5e9eecc6686
Thanks,
Christophe
>
> Richard.
>
> > > Regards
> > > Robin
> > >
> > >
> > > gcc/ChangeLog:
> > >
> > > * config/arm/arm.cc (arm_builtin_support_vector_misalignment):
> > > Remove use of type.
> > > ---
> > > gcc/config/arm/arm.cc | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/gcc/config/arm/arm.cc b/gcc/config/arm/arm.cc
> > > index f074a429200..da28d96298a 100644
> > > --- a/gcc/config/arm/arm.cc
> > > +++ b/gcc/config/arm/arm.cc
> > > @@ -30667,7 +30667,7 @@ arm_builtin_support_vector_misalignment
> > > (machine_mode mode,
> > > {
> > > if (TARGET_NEON && !BYTES_BIG_ENDIAN && unaligned_access)
> > > {
> > > - HOST_WIDE_INT align = TYPE_ALIGN_UNIT (type);
> > > + HOST_WIDE_INT align = GET_MODE_UNIT_SIZE (mode);
> > >
> > > if (is_gather_scatter)
> > > return true;
> > > --
> > > 2.51.0
> > >
