From owner-freebsd-arm@freebsd.org Sun Mar 5 11:13:51 2017 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 7F8BDCF904D for ; Sun, 5 Mar 2017 11:13:51 +0000 (UTC) (envelope-from markmi@dsl-only.net) Received: from asp.reflexion.net (outbound-mail-211-192.reflexion.net [208.70.211.192]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 27D9F17F0 for ; Sun, 5 Mar 2017 11:13:50 +0000 (UTC) (envelope-from markmi@dsl-only.net) Received: (qmail 7226 invoked from network); 5 Mar 2017 11:16:01 -0000 Received: from unknown (HELO mail-cs-02.app.dca.reflexion.local) (10.81.19.2) by 0 (rfx-qmail) with SMTP; 5 Mar 2017 11:16:01 -0000 Received: by mail-cs-02.app.dca.reflexion.local (Reflexion email security v8.30.2) with SMTP; Sun, 05 Mar 2017 06:13:43 -0500 (EST) Received: (qmail 2158 invoked from network); 5 Mar 2017 11:02:25 -0000 Received: from unknown (HELO iron2.pdx.net) (69.64.224.71) by 0 (rfx-qmail) with (AES256-SHA encrypted) SMTP; 5 Mar 2017 11:02:25 -0000 Received: from [192.168.1.111] (c-67-170-167-181.hsd1.or.comcast.net [67.170.167.181]) by iron2.pdx.net (Postfix) with ESMTPSA id AE363EC8807; Sun, 5 Mar 2017 03:02:24 -0800 (PST) Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 10.2 \(3259\)) Subject: Re: Is CPUTYPE=cortex-A7 supposed to work? From: Mark Millard In-Reply-To: <87h93814rb.fsf@news-spur.riddles.org.uk> Date: Sun, 5 Mar 2017 03:02:24 -0800 Cc: "freebsd-arm@freebsd.org" Content-Transfer-Encoding: 7bit Message-Id: <70AE704A-C9FF-4742-88E9-147CD5B77BE8@dsl-only.net> References: <871suc3nv8.fsf@news-spur.riddles.org.uk> <87tw7820fc.fsf@news-spur.riddles.org.uk> <644D1F49-BF5D-409D-BFC4-4F7E6E73085B@dsl-only.net> <87lgsk1udz.fsf@news-spur.riddles.org.uk> <9677298B-5A5E-44BF-928E-28DDDADB310A@dsl-only.net> <87h93814rb.fsf@news-spur.riddles.org.uk> To: Andrew Gierth X-Mailer: Apple Mail (2.3259) X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 05 Mar 2017 11:13:51 -0000 On 2017-Mar-5, at 2:01 AM, Andrew Gierth wrote: >>>>>> "Mark" == Mark Millard writes: > >>> 3. The errors aren't consistent at all. For example, sometimes I run >>> openssl speed rsa512 and it succeeds without error. When testing >>> with git, the failures were not always at the same place. > > Mark> Interesting. My context had MALLOC_PRODUCTION. I wonder if > Mark> without that and having junk filled in systematically might > Mark> produce more stable results. (I've no specific evidence that this > Mark> would make a difference.) > > I don't think this can explain my results with git; openssl's sha1 > doesn't call malloc (or any library function other than memcpy, memset, > OPENSSL_cleanse), and I have logs of the data being hashed that show the > wrong results returned. Also, I've checked that while these hashes are > being done, there are no other threads in the process. > > -- > Andrew. FYI: I've updated to -r314687 (so now clang 4.0 based) and openssl speed still gets the same failures. (By contrast a pine64+ 2 GiByte [an arm64] worked fine.) I have not tried a normal build without the -mcpu=cortex-a7 use (or equivalent). Have you? === Mark Millard markmi at dsl-only.net