Received: by oss.sgi.com id ; Tue, 7 Nov 2000 08:03:20 -0800 Received: from hermes.mixx.net ([212.84.196.2]:24339 "HELO hermes.mixx.net") by oss.sgi.com with SMTP id ; Tue, 7 Nov 2000 08:02:59 -0800 Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251]) by hermes.mixx.net (Postfix) with ESMTP id DF624F893 for ; Tue, 7 Nov 2000 17:02:40 +0100 (CET) Received: by mate.bln.innominate.de (Postfix, from userid 9) id 8943D2CA6F; Tue, 7 Nov 2000 17:02:10 +0100 (CET) From: Thomas Graichen Reply-To: Thomas Graichen X-Newsgroups: innominate.list.sgi.xfs Subject: XFS on ppc status Date: 7 Nov 2000 16:02:10 GMT Organization: innominate AG, Berlin, Germany Lines: 34 Distribution: local Message-ID: Reply-To: thomas.graichen@innominate.de X-Trace: mate.bln.innominate.de 973612930 20045 10.0.0.69 (7 Nov 2000 16:02:10 GMT) X-Complaints-To: news@innominate.de User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.4.0-XFS-test10 (i586)) To: linux-xfs@oss.sgi.com Sender: owner-linux-xfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;linux-xfs-outgoing just as a short note: back in berlin at the weekend i found some time to have a look at XFS on the ppc again and here is the result: * the current cvs tree no longer needs any significant changes to build * there were only two small things: CLOCKS_PER_SEC is not defined in asm/param.h for the ppc - this one is not related to XFS at all looks more like a test10 problem (occures only while compiling fs/binfmt_elf.c) and is easily fixed by defining it there * cmd/xfs/dump/dump/content.c cries about O_DIRECT not being defined but as far as i can see it is defined the same way as on i386 in the includes (asm/fcntl.h) - but maybe an indirect include does not get this one (anyone having an idea?) - just commenting out the affected 3 lines gets it going so far (they are only relevant for realtime volumes which are as far as i remember not supported at all so far anyway) the resulting system run just fine so far - tested it with the rootfs XFS and ran two dbench (one 4 and one 8) parallel for some hours and half of a day an dbench 4 parallel to an endless kernel clean build - both times no problem so far ... will try to start some of the stress testing in the next days (have to get more disk space first) ... so far the satus ... my plans are to setup an alpha and a ppc here at work now for doing more and better stress testing - hope to get this done within the next two weeks t -- thomas.graichen@innominate.com technical director innominate AG clustering & security the linux architects tel: +49-30-308806-13 fax: -77 http://www.innominate.com