From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 31732 invoked by alias); 20 Jan 2004 00:19:57 -0000 Mailing-List: contact libc-hacker-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libc-hacker-owner@sources.redhat.com Received: (qmail 31698 invoked from network); 20 Jan 2004 00:19:56 -0000 Received: from unknown (HELO are.twiddle.net) (64.81.246.98) by sources.redhat.com with SMTP; 20 Jan 2004 00:19:56 -0000 Received: from are.twiddle.net (localhost.localdomain [127.0.0.1]) by are.twiddle.net (8.12.8/8.12.8) with ESMTP id i0K0Jqj1003421; Mon, 19 Jan 2004 16:19:52 -0800 Received: (from rth@localhost) by are.twiddle.net (8.12.8/8.12.8/Submit) id i0K0JpwD003419; Mon, 19 Jan 2004 16:19:51 -0800 X-Authentication-Warning: are.twiddle.net: rth set sender to rth@twiddle.net using -f Date: Tue, 20 Jan 2004 00:19:00 -0000 From: Richard Henderson To: sjmunroe@vnet.ibm.com Cc: libc-hacker@sources.redhat.com, Tom Gall , Andreas Jaeger , Segher Boessenkoof , Hartmut Penner Subject: Re: [RFC] Vector extentions for vprintf/vscanf Message-ID: <20040120001950.GA3410@twiddle.net> Mail-Followup-To: sjmunroe@vnet.ibm.com, libc-hacker@sources.redhat.com, Tom Gall , Andreas Jaeger , Segher Boessenkoof , Hartmut Penner References: <400C5DB8.7010400@us.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <400C5DB8.7010400@us.ibm.com> User-Agent: Mutt/1.4i X-SW-Source: 2004-01/txt/msg00075.txt.bz2 On Mon, Jan 19, 2004 at 04:44:08PM -0600, Steve Munroe wrote: > One useful extention to glibc would be adding vector formating to > printf/scanf. This would add the type modifier prefix v to the type > modifiers hh, h, l. For example to print a vector of 16 signed bytes > "%vhhd", 8 signed shorts "%vhd", 4 signed ints "%vd", or 4 floats > "%vhf". (note the form "%vf" is reserved for a vector of 2 doubles but > most current Vector SIMD implimentations don't support double float). Are you going to hard-code the width of the vector? That's not friendly for x86, where there are two sizes. r~