From owner-freebsd-questions@FreeBSD.ORG Wed Nov 10 23:21:37 2010 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2B0201065670 for ; Wed, 10 Nov 2010 23:21:37 +0000 (UTC) (envelope-from ohartman@mail.zedat.fu-berlin.de) Received: from outpost1.zedat.fu-berlin.de (outpost1.zedat.fu-berlin.de [130.133.4.66]) by mx1.freebsd.org (Postfix) with ESMTP id D5B148FC17 for ; Wed, 10 Nov 2010 23:21:36 +0000 (UTC) Received: from inpost2.zedat.fu-berlin.de ([130.133.4.69]) by outpost1.zedat.fu-berlin.de (Exim 4.69) with esmtp (envelope-from ) id <1PGJyx-00046Y-NG>; Thu, 11 Nov 2010 00:21:35 +0100 Received: from e178026090.adsl.alicedsl.de ([85.178.26.90] helo=thor.walstatt.dyndns.org) by inpost2.zedat.fu-berlin.de (Exim 4.69) with esmtpsa (envelope-from ) id <1PGJyx-00048N-Ci>; Thu, 11 Nov 2010 00:21:35 +0100 Message-ID: <4CDB28FF.9090002@mail.zedat.fu-berlin.de> Date: Thu, 11 Nov 2010 00:21:35 +0100 From: "O. Hartmann" User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.2.12) Gecko/20101029 Lightning/1.0b2 Thunderbird/3.1.6 MIME-Version: 1.0 To: =?ISO-8859-15?Q?=22C=2E_Bergstr=F6m=22?= References: <4CDA8F15.506@zedat.fu-berlin.de> <4CDAED1F.7090909@mail.zedat.fu-berlin.de> <4CDAEF60.1090403@pathscale.com> In-Reply-To: <4CDAEF60.1090403@pathscale.com> Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: quoted-printable X-Originating-IP: 85.178.26.90 Cc: App Deb , "O. Hartmann" , freebsd-questions@freebsd.org Subject: Re: FreeBSD-similar build-from-source Linux? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 10 Nov 2010 23:21:37 -0000 On 11/10/10 20:15, "C. Bergstr=F6m" wrote: > O. Hartmann wrote: >> >>> The NVIDIA FreeBSD driver provides the cuda libraries for linux >>> compatibilty. >>> >>> So 32-bit Linux Cuda applications should work on FreeBSD. >> >> There are some other, very serious questions. AMD calims that they >> made the specs of their 3D chipset internals public. > The docs had big missing chunks which I tried to get clarification on > and got some help and then it just one day stopped >> There is OpenCL as an open standard > Try to find the validation test suite.. ($$$) >> , there is the CLANG/LLVM project even for FreeBSD > Apple licenses it and doesn't publish all opencl work. (or I think may > not even contribute to the opencl side at all anymore.. someone correct= me) >> to become the new standard compiler and, not at last, there is work >> done on drivers for AMD graphics boards but there is no, not even >> rudimentary, support for GPGPU. I preferr a clean open source >> solution, but at the moment, it seems to be the best and easiest path >> to switch to an operating system that is fully supported, even 64 bit.= > Side question - Why care in the least bit about AMD? Their hw sucks and= > their software is a joke.. (if you bought Evergreen I'm sorry..) By the way - way is the hardware crap? Can you please be more specific?=20 AMD claims to have a higher effective double precision throughput than=20 nVidia, they also claim not having cut off most of their double=20 precision capable facilities like nVidia even on their mass market=20 products (but I'm not sure about that since this fact is very often=20 hidden by intention). Their software doesn't really matter since I most=20 FreeBSD users use the open source driver. Well, I never saw a working 64 = bit capable driver on FreeBSD, that's right. Please do not understand this as a tricky question, I'm serious, since I = plan to purchase new hardware for my computer at home (FreeBSD) and even = for my lab's computer(s). And by the way, no, I havn't bought 'Evergreen', I have much more=20 disappointing HD4830 and not properly working HD4770 and HD4670 ... (not = GPGPU capable by design). > > There's no such thing as a complete open source GPGPU solution as of > today (to the best of my knowledge) > > ./C >