This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] [AArch64, NEON] Improve vpmaxX & vpminX intrinsics
- From: "Yangfei (Felix)" <felix dot yang at huawei dot com>
- To: Tejas Belagod <tejas dot belagod at arm dot com>, "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, Marcus Shawcroft <Marcus dot Shawcroft at arm dot com>
- Date: Tue, 9 Dec 2014 08:20:57 +0000
- Subject: Re: [PATCH] [AArch64, NEON] Improve vpmaxX & vpminX intrinsics
- Authentication-results: sourceware.org; auth=none
- References: <DA41BE1DDCA941489001C7FBD7A8820E837A2D0A at szxema507-mbx dot china dot huawei dot com> <5481F71E dot 3000309 at arm dot com>
> You'll need to rebase over Alan Lawrance's patch.
> https://gcc.gnu.org/ml/gcc-patches/2014-12/msg00279.html
Yes, see my new patch: https://gcc.gnu.org/ml/gcc-patches/2014-12/msg00750.html
> > +;; Pairwise Integer Max/Min operations.
> > +(define_insn "aarch64_<maxmin_uns>p<mode>"
> > + [(set (match_operand:VQ_S 0 "register_operand" "=w")
> > + (unspec:VQ_S [(match_operand:VQ_S 1 "register_operand" "w")
> > + (match_operand:VQ_S 2 "register_operand" "w")]
> > + MAXMINV))]
> > + "TARGET_SIMD"
> > + "<maxmin_uns_op>p\t%0.<Vtype>, %1.<Vtype>, %2.<Vtype>"
> > + [(set_attr "type" "neon_minmax<q>")]
> > +)
> > +
>
> Could you roll aarch64_reduc_<maxmin_uns>_internalv2si into this pattern?
Will come up with another patch to fix this. Thanks for pointing this out.
>
> Thanks,
> Tejas.