From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin K. Petersen" Subject: Re: [PATCH 0 of 3] [RFC] I/O Hints Date: Mon, 09 Jun 2008 11:05:53 -0400 Message-ID: References: <20080605104042.GB20308@shareable.org> <20080606045150.GL10720@disturbed> <20080607205438.GS10720@disturbed> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from rgminet01.oracle.com ([148.87.113.118]:44532 "EHLO rgminet01.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754739AbYFIPLB (ORCPT ); Mon, 9 Jun 2008 11:11:01 -0400 In-Reply-To: <20080607205438.GS10720@disturbed> (Dave Chinner's message of "Sun\, 8 Jun 2008 06\:54\:38 +1000") Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: "Martin K. Petersen" Cc: linux-scsi@vger.kernel.org, linux-fsdevel@vger.kernel.org >>>>> "Dave" == Dave Chinner writes: Dave> Sure it could, but xfs_growfs is currently just a wrapper around Dave> an ioctl that tells the kernel to grow the filesystem by N bytes Dave> and the kernel does all the changes transactionally. So changing Dave> xfs_growfs to provide this information to the kernel will Dave> require a new userspace API for XFS to push this information Dave> straight back down into the kernel.... Dave> Also, it would be good to verify that stored filesystem Dave> parameters are still optimal at mount time by querying the block Dave> device(s) when internal structures are first initialised from Dave> disk. Hence even a simple kernel API to this info would be Dave> handy. Ok, I'll see if I can come up with a reasonable interface for this. -- Martin K. Petersen Oracle Linux Engineering