From owner-freebsd-amd64@FreeBSD.ORG Sat Feb 4 13:36:55 2012 Return-Path: Delivered-To: freebsd-amd64@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 68946106566C for ; Sat, 4 Feb 2012 13:36:55 +0000 (UTC) (envelope-from andrej@antiszoc.hu) Received: from mail.deployis.eu (mail.deployis.eu [217.20.135.253]) by mx1.freebsd.org (Postfix) with ESMTP id DCA358FC18 for ; Sat, 4 Feb 2012 13:36:54 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=antiszoc.hu; s=default; h=Message-ID:References:In-Reply-To:Subject:To:From:Date:Content-Transfer-Encoding:Content-Type:MIME-Version; bh=S/E9QkXj+0wFmp6gFuYPd7VZyDIqej04QnGIeD+/xDc=; b=blw0gzZXkzNwA73GK6XCgQV5ekiL7WRbvy1lH7vlsEbw4Vjy0rBlo5STmIE4IUbO3OEfVmd/f56PeeD8kWPz7QOI6t5fF6f+14cIAcCgyzVIqLWB9FOrvRBz2vuQ84j8; Authentication-Results: mail.deployis.eu dkim=none Received: from localhost ([127.0.0.1]:40464 helo=mail.deployis.eu) by mail.deployis.eu with esmtp (Exim 4.71 #1 (Debian)) id 1RtfNp-0006Hm-N8 from for ; Sat, 04 Feb 2012 14:10:37 +0100 Received: from pool-232-162.ippark.hu ([31.223.232.162]) by mail.deployis.eu with HTTP (HTTP/1.1 POST); Sat, 04 Feb 2012 14:10:25 +0100 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Date: Sat, 04 Feb 2012 14:10:25 +0100 From: =?UTF-8?Q?G=C3=B3t_Andr=C3=A1s?= To: In-Reply-To: <20120204124952.GK3283@deviant.kiev.zoral.com.ua> References: <4F2CFFC3.30909@gurgelmail.com> <20120204124952.GK3283@deviant.kiev.zoral.com.ua> Message-ID: <7e0e01cc2b957c417a3f69232b33ebe0@antiszoc.hu> X-Sender: andrej@antiszoc.hu User-Agent: RoundCube Webmail/0.2.1 X-Mail-Status-postahivatal: trustedmail (from 127.0.0.1) X-Spam-Score-postahivatal: -43 Subject: Re: aesni module with AMD CPUs? X-BeenThere: freebsd-amd64@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the AMD64 platform List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 04 Feb 2012 13:36:55 -0000 Hi, This is from a Linux machine: processor : 15 vendor_id : AuthenticAMD cpu family : 21 model : 1 model name : AMD Opteron(tm) Processor 4280 stepping : 2 microcode : 0x6000624 cpu MHz : 2800.277 cache size : 2048 KB physical id : 1 siblings : 8 core id : 7 cpu cores : 4 apicid : 39 initial apicid : 23 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 cx16 sse4_1 sse4_2 popcnt aes xsave avx lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 nodeid_msr topoext perfctr_core arat cpb npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold AFAIK Bulldozer only has AVX, but unfortunately not AESNI. Andras On Sat, 4 Feb 2012 14:49:52 +0200, Konstantin Belousov wrote: > On Sat, Feb 04, 2012 at 10:52:03AM +0100, Mario Fleischmann wrote: >> Hi Konstantin, >> I hope you don't mind me writing to you directly, but I'm having a >> hard >> time finding information on Google. > You would get much higher chances of getting response if you post on > the > public list Cc:ing corresponding developer. Since the question > seemingly > not contain any sensitive information, and discussion may be useful > to > the wider audience, I added appropriate ml to Cc:. > >> >> I'm currently in the market for a new system, and I need AES-NI >> acceleration with FreeBSD 9.0. I'd like to buy an AMD-FX 4100 CPU, >> which >> supports the AES-NI instruction set. > Do you have a reference to the CPU datasheet or programming manual ? > >> >> What I'd like to know is: Does the aesni driver in FreeBSD also >> support >> AMD CPUs? The man page explicitly states the driver is for Intel >> CPUs, >> but since the AES-NI instruction set is licensed by AMD, I think it >> should also work with those CPUs? > The aesni(4) driver attaches on any CPU that reports AESNI support > through > the CPUID instruction. If amd implemented bit-to-bit compatible AESNI > instruction set and report the capability in CPUID, driver shall just > work. > > I never saw any bulldozer live, so this is the most I can answer to > you. > >> >> >> A reply would be greatly appreciated, >> >> Best Regards, >> Mario