Return-Path: linux-nfs-owner@vger.kernel.org Received: from bombadil.infradead.org ([198.137.202.9]:60365 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750749Ab3KBPHi (ORCPT ); Sat, 2 Nov 2013 11:07:38 -0400 Date: Sat, 2 Nov 2013 08:07:38 -0700 From: Christoph Hellwig To: "Myklebust, Trond" Cc: Christoph Hellwig , Bruce Fields , "Schumaker, Bryan" , Linux NFS Mailing List Subject: Re: [PATCH 3/4] NFSD: Add WRITE_PLUS support for hole punches Message-ID: <20131102150738.GA21133@infradead.org> References: <1382972247-1108-1-git-send-email-bjschuma@netapp.com> <1382972247-1108-4-git-send-email-bjschuma@netapp.com> <20131028214030.GO31322@fieldses.org> <526FAF30.3060502@netapp.com> <20131029130649.GA29606@fieldses.org> <20131102135409.GC18961@infradead.org> <20131102144431.GB26983@fieldses.org> <20131102145105.GA4450@infradead.org> <38D14516-BAA4-45BB-8054-ED05AEC22C8F@netapp.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <38D14516-BAA4-45BB-8054-ED05AEC22C8F@netapp.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Sat, Nov 02, 2013 at 03:02:29PM +0000, Myklebust, Trond wrote: > COMMIT has always had a metadata part to it, though. It guarantees stability of the ctime/mime and change attributes in addition to the file size and data. > > IOW: it really is more akin to fsync() than to fdatasync(). That's fine, but so far it wasn't used for any purely metadata operation. E.g. if we allow it for a preallocation we should also allow it for updating the size put the current file size. And diverting from the spec to your implementation: we'd need a lot more generic infrastructure to deal with unstable random non-data operations.