From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Wilcox Subject: Re: [PATCH v6 06/22] Replace XIP read and write with DAX I/O Date: Tue, 11 Mar 2014 08:53:57 -0400 Message-ID: <20140311125357.GA7580@linux.intel.com> References: <1393337918-28265-1-git-send-email-matthew.r.wilcox@intel.com> <1393337918-28265-7-git-send-email-matthew.r.wilcox@intel.com> <1394497958.6784.204.camel@misato.fc.hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Matthew Wilcox , linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org To: Toshi Kani Return-path: Content-Disposition: inline In-Reply-To: <1394497958.6784.204.camel@misato.fc.hp.com> Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Mar 10, 2014 at 06:32:38PM -0600, Toshi Kani wrote: > On Tue, 2014-02-25 at 09:18 -0500, Matthew Wilcox wrote: > > Use the generic AIO infrastructure instead of custom read and write > > methods. In addition to giving us support for AIO, this adds the missing > > locking between read() and truncate(). > > > : > > +static void dax_new_buf(void *addr, unsigned size, unsigned first, > > + loff_t offset, loff_t end, int rw) > > +{ > > + loff_t final = end - offset; /* The final byte in this buffer */ > > I may be missing something, but shouldn't it take first into account? > > loff_t final = end - offset + first; Yes it should. Thanks! (Fortunately, this is only a performance problem as we'll end up zeroing more than we ought to, which is fine as it will be overwritten by the copy_from_user later) -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org