xfs
[Top] [All Lists]

Re: [RFC 00/11] DAX fsynx/msync support

To: Dave Chinner <david@xxxxxxxxxxxxx>, Jeff Moyer <jmoyer@xxxxxxxxxx>
Subject: Re: [RFC 00/11] DAX fsynx/msync support
From: Jens Axboe <axboe@xxxxxxxxx>
Date: Thu, 5 Nov 2015 13:54:19 -0700
Cc: Ross Zwisler <ross.zwisler@xxxxxxxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, "H. Peter Anvin" <hpa@xxxxxxxxx>, "J. Bruce Fields" <bfields@xxxxxxxxxxxx>, Theodore Ts'o <tytso@xxxxxxx>, Alexander Viro <viro@xxxxxxxxxxxxxxxxxx>, Andreas Dilger <adilger.kernel@xxxxxxxxx>, Dan Williams <dan.j.williams@xxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, Jan Kara <jack@xxxxxxxx>, Jeff Layton <jlayton@xxxxxxxxxxxxxxx>, Matthew Wilcox <willy@xxxxxxxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, linux-ext4@xxxxxxxxxxxxxxx, linux-fsdevel@xxxxxxxxxxxxxxx, linux-mm@xxxxxxxxx, linux-nvdimm@xxxxxxxxxxx, x86@xxxxxxxxxx, xfs@xxxxxxxxxxx, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>, Matthew Wilcox <matthew.r.wilcox@xxxxxxxxx>
Delivered-to: xfs@xxxxxxxxxxx
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel_dk.20150623.gappssmtp.com; s=20150623; h=subject:to:references:cc:from:message-id:date:user-agent :mime-version:in-reply-to:content-type:content-transfer-encoding; bh=MBzeWYsf6lngQ+jiU950oq/SOS9HaWr70k/TpYIVqiA=; b=OZo7Me1hQLazE0C7QdzB5+wc+qnYuuW/dREr2NLqxMCrk5wRCttcDCXaeyw79HuJHv LCeuaD/SrVrK23y83bJinaUfpjk1gSK3VXYJ0bfm5AdBWhuYrhpAzm83+pvJ5EAzlRIk 5hR+IvQqP78rWDMmWdORBA+l9Fl+uBzw0TBYEV/bjAxbnne4ONRpCmu0AplNy8vMZrQG lakWUN2gp+S/8mD/SHsfNEsiSDTr2rY8wNVpxPbJFX7s4cM2rZzrPzFGdbDYD/9FYlHK ypa4bFdjisRYq0UEbfgGn6fX73pA/Y2DWWljiIQ5JNayj+QeyF/hHYlubVwpxH3at+rJ Ga/g==
In-reply-to: <20151105083309.GJ19199@dastard>
References: <1446149535-16200-1-git-send-email-ross.zwisler@xxxxxxxxxxxxxxx> <20151030035533.GU19199@dastard> <20151030183938.GC24643@xxxxxxxxxxxxxxx> <20151101232948.GF10656@dastard> <x49vb9kqy5k.fsf@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> <20151102201029.GI10656@dastard> <x49twp4p11j.fsf@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> <20151105083309.GJ19199@dastard>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.3.0
On 11/05/2015 01:33 AM, Dave Chinner wrote:
Des xfs rely on this model for correctness?  If so, I'd say we've got a
problem

No, it doesn't. The XFS integrity model doesn't trust the IO layers
to tell the truth about IO ordering and completion or for it's
developers to fully understand how IO layer ordering works. :P

That's good, because the storage developers simplified the model so that fs developers would be able to get and use it.

i.e. we wait for full completions of all dependent IO before issuing
flushes or log writes that use REQ_FLUSH|REQ_FUA semantics to ensure
the dependent IOs are fully caught by the cache flushes...

... which is what you are supposed to do, that's how it works.

--
Jens Axboe

<Prev in Thread] Current Thread [Next in Thread>