[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20160208155534.GD2343@linux.intel.com>
Date: Mon, 8 Feb 2016 08:55:34 -0700
From: Ross Zwisler <ross.zwisler@...ux.intel.com>
To: Christoph Hellwig <hch@...radead.org>
Cc: Ross Zwisler <ross.zwisler@...ux.intel.com>,
Jan Kara <jack@...e.cz>,
Dan Williams <dan.j.williams@...el.com>,
Dave Chinner <david@...morbit.com>,
Matthew Wilcox <willy@...ux.intel.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Alexander Viro <viro@...iv.linux.org.uk>,
Andrew Morton <akpm@...ux-foundation.org>,
Jan Kara <jack@...e.com>,
linux-fsdevel <linux-fsdevel@...r.kernel.org>,
linux-nvdimm <linux-nvdimm@...1.01.org>
Subject: Re: [PATCH 2/2] dax: fix bdev NULL pointer dereferences
On Sun, Feb 07, 2016 at 12:38:14AM -0800, Christoph Hellwig wrote:
> On Fri, Feb 05, 2016 at 03:25:00PM -0700, Ross Zwisler wrote:
> > According to the standard specification (e.g., POSIX.1-2001), sync()
> > schedules the writes, but may return before the actual writing
> > is done. However, since version 1.3.20 Linux does actually wait.
> > (This still does not guarantee data integrity: modern disks have large
> > caches.)
> >
> > Based on this I don't believe that it is a requirement that sync and syncfs
> > actually flush the data durably to media before they return - they just need
> > to make sure it has been sent to the device, which is always true for all
> > writes PMEM via DAX, even without any calls to dax_writeback_mapping_range().
>
> For Linux there is a requirement to not return before the data is on disk,
> and our users rely on it. The man page is rather confusing, and I'll see
> if I can fix it up.
Okay, thank you for the clarification. :)
Powered by blists - more mailing lists