From: Eric Sandeen Subject: Re: [PATCH, RFC] mke2fs: get device topology values from blkid Date: Fri, 18 Sep 2009 09:04:37 -0500 Message-ID: <4AB39375.6040506@redhat.com> References: <4AB2B6B9.7010506@redhat.com> <20090918055547.GO2537@webber.adilger.int> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: ext4 development To: Andreas Dilger Return-path: Received: from mx1.redhat.com ([209.132.183.28]:25622 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753688AbZIROEj (ORCPT ); Fri, 18 Sep 2009 10:04:39 -0400 In-Reply-To: <20090918055547.GO2537@webber.adilger.int> Sender: linux-ext4-owner@vger.kernel.org List-ID: Andreas Dilger wrote: > On Sep 17, 2009 17:22 -0500, Eric Sandeen wrote: >> This is just a rough cut, due to the blkid header selection >> issues I mentioned earlier on the list. It'll also need >> some config-fu to be sure we've got a blkid which has these >> calls, but with it in place, we'll finally have automatic >> selection of stride/stripe: >> >> # misc/mke2fs -b 4096 /dev/md0 >> mke2fs 1.41.9 (22-Aug-2009) >> Filesystem label= >> OS type: Linux >> Block size=4096 (log=2) >> Fragment size=4096 (log=2) >> Stride=16 blocks, Stripe width=32 blocks >> ... > > Cool. > >> Signed-off-by: Eric Sandeen >> printf(_("Fragment size=%u (log=%u)\n"), fs->fragsize, >> s->s_log_frag_size); >> + printf(_("Stride=%u blocks, Stripe width=%u blocks\n"), >> + s->s_raid_stride, s->s_raid_stripe_width); > > I was going to say we should only print these if non-zero, but for > Pete's sake we print the "Fragment size", which has never been useful, > so I don't see any harm in this. Maybe "RAID Stride ..."? > >> +static errcode_t ext2fs_get_device_geometry(const char *file, >> + struct ext2_super_block *fs_param) >> +{ >> + rc = blkid_probe_set_device(pr, fd, 0, 0); > > Is this in the e2fsprogs blkid code yet? I'm guessing not, since you right, not yet. I don't know what the long-term plan is for e2fsprogs blkid; keeping 2 trees in sync seems like a lot of work w/o much gain... > are getting build problems, which means anyone building with something > other than latest & greatest util-linux will also get failures. > > Either this needs to be configured in, or you need to upgrade the > blkid included with e2fsprogs to handle this. Yep, that's why I said "It'll also need some config-fu to be sure we've got a blkid which has these calls...." :) and why I was asking about moving the in-tree headers slightly out of the way. -Eric > Cheers, Andreas > -- > Andreas Dilger > Sr. Staff Engineer, Lustre Group > Sun Microsystems of Canada, Inc. >