From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-201667-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 42985 invoked by alias); 17 Jan 2020 08:19:46 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 42974 invoked by uid 89); 17 Jan 2020 08:19:46 -0000
Authentication-Results: sourceware.org; auth=none
X-Spam-SWARE-Status: No, score=-0.7 required=5.0 tests=AWL,BAYES_00,BODY_8BITS,FREEMAIL_FROM,GARBLED_BODY,RCVD_IN_DNSWL_NONE,SPF_PASS autolearn=no version=3.3.1 spammy=H*f:sk:GcaLIHZ, H*f:sk:CCED278, H*f:sk:hd-6QLi, H*f:sk:1bP43sI
X-HELO: mail-lf1-f67.google.com
Received: from mail-lf1-f67.google.com (HELO mail-lf1-f67.google.com) (209.85.167.67) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Fri, 17 Jan 2020 08:19:35 +0000
Received: by mail-lf1-f67.google.com with SMTP id m30so17635194lfp.8        for <gcc@gcc.gnu.org>; Fri, 17 Jan 2020 00:19:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;        d=gmail.com; s=20161025;        h=mime-version:references:in-reply-to:from:date:message-id:subject:to         :cc:content-transfer-encoding;        bh=4Vw9gpJWYrYcJG2gcBbM0WIxXGZAbqm/iU4ESrYIo80=;        b=SFxvLVvzwgPRb7XITBcjlbnosSajvcrp6RTSJHac+VtkhqczATCat1NYfPaBLGxUlg         0Ptub6vTEFSQOlrK+LhO9rVmA8KngHNjINI1r+P+RyAfshg8BmuYiTvLaIo7yzdcBFiL         us67YEOliLtlD7ewsf2ZPz/ALQSCdFCF0glRaIjfMRXcgOMbYSJkuQHiDqoO5ihCuJbf         /zOFfc+sw4tsLdlpfJAADpTwyL0Qj6B8BvMAk0LsLsnv+gz/b1z3BZP43qOfxOKYzb7G         WHcg2DiW7lJyA0Ix6ixH3/GppV6D99RLpvvaDMGlBSCSQrEAOH2mr7kSRwyK39JdIgSH         Qp1w==
MIME-Version: 1.0
References: <GcaLIHZaxSm2PVhunNuxEwEJB-BC3PeXIBlmYb6eMw54vZQ_KF6uBjlTgWmWtPNIMMeXu7UyYg797u1vV1_GS2Qv3ThxrUeORQD0wcddwwo=@protonmail.com> <20191206111540.GC10088@tucnak> <CAFiYyc2WJrh2BW5891XhjxM0SsJkxbSOoOh58AeTBmz+Bed9Tw@mail.gmail.com> <hd-6QLiJS0PVrxLBOkT_w0FiGyHomRJkmlR_hRrIGCHZJXhITGVMZ6fzbOFOOoqt6fZ3E1m5tDhtzwDfbeeusk1IdGwsONZFF2nIA4WvxMw=@protonmail.com> <CCED278B-4638-4238-A798-9FD4C98EF2BE@gmail.com> <lyW7bEDQcGe8hV1r_r-252urMynbbu4bJHT-mEJOb4rljDo4-H-BpPiZcKO3EVm2p2EkkvF0Xl_CLCFRael9C5TiBdoNLJ4TCEXbNTPe8iU=@protonmail.com> <CAFiYyc0Pr58VFoR+tKrgi_pvO3OGcHnm4JH=itYF+NdvaqmJUQ@mail.gmail.com> <Ob53m7cSAsYBYj1DHIj9VH8DF5Hihjngpocha0Ctn5zbjEYG8ixA7ZoKAzR4Hm_2XT74neDGiZukyW_c95ppV_dG4XKwRDrybQAfiOUGhSo=@protonmail.com> <CAFiYyc0sSJdCOeqdKtSi+7+u8y_aZ-P6mCgp_nkq5ybfEghb9w@mail.gmail.com> <WmCoNrPoL6wrMXaqKofP1mAcNkuqDYShTab5uZln-mNhVgnv62OLWvqpKeYQyGbhh__GYoyt058t86W2zIXbdtSjJo9aSZsZ4H_HGRJUKL4=@protonmail.com> <1bP43sIDtgGcokozEcGeINTETW5pPYQ93fU4Eo0b856l8R-Nm6eYIaOTA9T6TRVou2CJTWkya2uRIP4SJHX2vuXhyOaxH-IKMrw8aFht2-Q=@protonmail.com>
In-Reply-To: <1bP43sIDtgGcokozEcGeINTETW5pPYQ93fU4Eo0b856l8R-Nm6eYIaOTA9T6TRVou2CJTWkya2uRIP4SJHX2vuXhyOaxH-IKMrw8aFht2-Q=@protonmail.com>
From: Richard Biener <richard.guenther@gmail.com>
Date: Fri, 17 Jan 2020 16:37:00 -0000
Message-ID: <CAFiYyc1QNhacUPMesnrFqnBOGa0j9RhW3O-m=Drb1KZ5-6gk_w@mail.gmail.com>
Subject: Re: PPC64 libmvec implementation of sincos
To: GT <tnggil@protonmail.com>
Cc: Jakub Jelinek <jakub@redhat.com>, "gcc@gcc.gnu.org" <gcc@gcc.gnu.org>, 	Szabolcs Nagy <Szabolcs.Nagy@arm.com>, nd <nd@arm.com>, 	Bill Schmidt <wschmidt@linux.ibm.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-IsSubscribed: yes
X-SW-Source: 2020-01/txt/msg00337.txt.bz2

On Thu, Jan 16, 2020 at 4:54 PM GT <tnggil@protonmail.com> wrote:
>
>
> =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90 Original =
Message =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90
> On Wednesday, January 15, 2020 3:20 PM, GT <tnggil@protonmail.com> wrote:
>
> > =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90 Origina=
l Message =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90
> > On Thursday, January 9, 2020 8:42 AM, Richard Biener richard.guenther@g=
mail.com wrote:
> >
> > > As for the other question for testing you probably want to provide a
> > > OMP simd declaration
> > > of a function like
> > > _Complex double mycexpi (double);
> > > and make a testcase like
> > > void foo (_Complex double * __restrict out, double *in)
> > > {
> > > for (int i =3D 0; i < 1024; ++i)
> > > {
> > > out[i] =3D mycexpi (in[i]);
> > > }
> > > }
> > > or eventually with two output arrays and explicit __real/__imag
> > > processing. The real
> > > and main question is how is the OMP SIMD declaration of mycexpi looki=
ng like?
> > > So I'd completely side-step sincos() and GCCs sincos() ->
> > > __builtin_cepxi transform
> > > and concentrate on OMP SIMD of a function with the signature we need =
to handle.
> > > Richard.
> >
> > I think what is required here is to attach either #pragma omp declare s=
imd orattribute ((simd))
> > to the declaration of builtin cexpi. In gcc/builtins.def, some attribut=
es are provided during
> > creation of cexpi (line 656, call containing BUILT_IN_CEXPI). Attaching=
 the simd attributes to
> > function declarations is how sin, cos, and the other math functions wer=
e handled in math-vector.h
> > glibc header file.
>
> You probably intended that we first teach GCC how to vectorize any functi=
on which returns a
> _Complex double and has a single parameter of type double. When that's do=
ne, move on to solving
> the specific vectorization of __builtin_cexpi. Right?

Yes.

Richard.

> Bert.