From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <libc-alpha-return-52754-listarch-libc-alpha=sources.redhat.com@sourceware.org>
Received: (qmail 7782 invoked by alias); 12 Sep 2014 00:06:28 -0000
Mailing-List: contact libc-alpha-help@sourceware.org; run by ezmlm
Precedence: bulk
List-Id: <libc-alpha.sourceware.org>
List-Subscribe: <mailto:libc-alpha-subscribe@sourceware.org>
List-Archive: <http://sourceware.org/ml/libc-alpha/>
List-Post: <mailto:libc-alpha@sourceware.org>
List-Help: <mailto:libc-alpha-help@sourceware.org>, <http://sourceware.org/ml/#faqs>
Sender: libc-alpha-owner@sourceware.org
Received: (qmail 7772 invoked by uid 89); 12 Sep 2014 00:06:27 -0000
Authentication-Results: sourceware.org; auth=none
X-Virus-Found: No
X-Spam-SWARE-Status: No, score=-3.9 required=5.0 tests=AWL,BAYES_00,RP_MATCHES_RCVD,SPF_HELO_PASS,SPF_PASS autolearn=ham version=3.3.2
X-HELO: mx1.redhat.com
Message-ID: <541238F4.80103@redhat.com>
Date: Fri, 12 Sep 2014 00:06:00 -0000
From: "Carlos O'Donell" <carlos@redhat.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.7.0
MIME-Version: 1.0
To: Rich Felker <dalias@libc.org>,
        Matthew Fortune <Matthew.Fortune@imgtec.com>
CC: "Joseph S. Myers" <joseph@codesourcery.com>,
        Andrew Senkevich <andrew.n.senkevich@gmail.com>,
        libc-alpha <libc-alpha@sourceware.org>,
        "igor.zamyatin@intel.com" <igor.zamyatin@intel.com>,
        "Melik-Adamyan, Areg" <areg.melik-adamyan@intel.com>,
        "jakub@redhat.com" <jakub@redhat.com>
Subject: Re: [PATCH 1/N] x86_64 vectorization support: vectorized math functions
 addition to Glibc
References: <CAMXFM3t=ppndDUBzHzSus7xyuF5hTaLFZ5b273jD39NtddSvsw@mail.gmail.com> <Pine.LNX.4.64.1409101549490.12853@digraph.polyomino.org.uk> <6D39441BF12EF246A7ABCE6654B0235320F09D65@LEMAIL01.le.imgtec.org> <20140911210246.GN23797@brightrain.aerifal.cx>
In-Reply-To: <20140911210246.GN23797@brightrain.aerifal.cx>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-SW-Source: 2014-09/txt/msg00251.txt.bz2

On 09/11/2014 05:02 PM, Rich Felker wrote:
> This really seems like something the compiler should be doing --
> translating parallelizable calls to the standard math functions into
> calls to special simd versions (or better yet, LTO'ing in an
> on-the-fly simd version based on the non-simd-specific code in libm.a)
> rather than having applications written to a klunky API that's
> designed around particular hardware features.

Nothing is stopping the compiler from doing exactly what you say and
I know people who are working on it.

However, I think we should also support the klunky APIs that have been
in use at companies like Intel and IBM in something like libmvec.

So it's not one or the other, but both.

Cheers,
Carlos.