[prev in list] [next in list] [prev in thread] [next in thread] 

List:       fuse-devel
Subject:    Re: [fuse-devel] Horrible mmap write performance (kernel writeback issue?)
From:       Antonio SJ Musumeci <trapexit () spawn ! link>
Date:       2016-03-31 16:58:47
Message-ID: CAB3-ZyQQY=4rRn8Lw7E4EyuJsrunLXDTz3orXmiaPW8cxF6syQ () mail ! gmail ! com
[Download RAW message or body]

[Attachment #2 (multipart/alternative)]


I had a user of mergerfs[0] use mmapwrite[1] and reproduce it immediately.

$ mkdir /tmp/test
$ mergerfs -odefaults /var/tmp /tmp/test
$ mmapwrite /tmp/test/foo

Perhaps of interest is that it affected the user's physical machines
running 4.2.0-34 (Ubuntu 15.10) but neither they or I could reproduce it in
a VM (Virtualbox).

[0] https://github.com/trapexit/mergerfs/releases
[1] https://github.com/rfjakob/mmapwrite

On Wed, Mar 30, 2016 at 2:47 PM, Tejun Heo <tj@kernel.org> wrote:

> Hello,
>
> On Mon, Mar 28, 2016 at 09:45:53PM +0200, Miklos Szeredi wrote:
> > >> commit 947e9762a8ddefda38aa21e249e6a4fec215cd12
> > >> Author: Tejun Heo <tj@kernel.org>
> > >> Date:   Fri May 22 18:23:32 2015 -0400
> > >>
> > >>     writeback: update wb_over_bg_thresh() to use wb_domain aware
> operations
> > >
> >
> > Tejun,
> >
> > Any idea why this commit stalls fuse mmap writes?
> >
> > Here's the start of the this thread:
> >
> >   http://marc.info/?l=fuse-devel&m=145008058603261&w=2
>
> Hmmm... cgroup writeback support shouldn't affect fuse at all as the
> backing device doesn't enable cgroup support.  I probably made some
> silly mistake.  Is there a simple reproducer I can play with?
>
> Thanks.
>
> --
> tejun
>
>
> ------------------------------------------------------------------------------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785471&iu=/4140
> --
> fuse-devel mailing list
> To unsubscribe or subscribe, visit
> https://lists.sourceforge.net/lists/listinfo/fuse-devel
>

[Attachment #5 (text/html)]

<div dir="ltr">I had a user of mergerfs[0] use mmapwrite[1] and reproduce it \
immediately.<div><br></div><div>$ mkdir /tmp/test</div><div>$ mergerfs -odefaults \
/var/tmp /tmp/test</div><div>$ mmapwrite \
/tmp/test/foo</div><div><br></div><div>Perhaps of interest is that it affected the \
user&#39;s physical machines running 4.2.0-34 (Ubuntu 15.10) but neither they or I \
could reproduce it in a VM (Virtualbox).</div><div><br></div><div><span>[0]  \
</span><span><a class="" \
href="https://github.com/trapexit/mergerfs/releases">https://github.com/trapexit/mergerfs/releases</a></span></div><div><span>[1] \
<a class="linkclass" \
href="https://github.com/rfjakob/mmapwrite">https://github.com/rfjakob/mmapwrite</a></span></div><div \
class="gmail_extra"><br><div class="gmail_quote">On Wed, Mar 30, 2016 at 2:47 PM, \
Tejun Heo <span dir="ltr">&lt;<a href="mailto:tj@kernel.org" \
target="_blank">tj@kernel.org</a>&gt;</span> wrote:<br><blockquote \
class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc \
solid;padding-left:1ex">Hello,<br> <br>
On Mon, Mar 28, 2016 at 09:45:53PM +0200, Miklos Szeredi wrote:<br>
&gt; &gt;&gt; commit 947e9762a8ddefda38aa21e249e6a4fec215cd12<br>
&gt; &gt;&gt; Author: Tejun Heo &lt;<a \
href="mailto:tj@kernel.org">tj@kernel.org</a>&gt;<br> &gt; &gt;&gt; Date:     Fri May \
22 18:23:32 2015 -0400<br> &gt; &gt;&gt;<br>
&gt; &gt;&gt;        writeback: update wb_over_bg_thresh() to use wb_domain aware \
operations<br> &gt; &gt;<br>
&gt;<br>
&gt; Tejun,<br>
&gt;<br>
&gt; Any idea why this commit stalls fuse mmap writes?<br>
&gt;<br>
&gt; Here&#39;s the start of the this thread:<br>
&gt;<br>
&gt;     <a href="http://marc.info/?l=fuse-devel&amp;m=145008058603261&amp;w=2" \
rel="noreferrer" target="_blank">http://marc.info/?l=fuse-devel&amp;m=145008058603261&amp;w=2</a><br>
 <br>
Hmmm... cgroup writeback support shouldn&#39;t affect fuse at all as the<br>
backing device doesn&#39;t enable cgroup support.   I probably made some<br>
silly mistake.   Is there a simple reproducer I can play with?<br>
<br>
Thanks.<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
tejun<br>
<br>
------------------------------------------------------------------------------<br>
Transform Data into Opportunity.<br>
Accelerate data analysis in your applications with<br>
Intel Data Analytics Acceleration Library.<br>
Click to learn more.<br>
<a href="http://pubads.g.doubleclick.net/gampad/clk?id=278785471&amp;iu=/4140" \
rel="noreferrer" target="_blank">http://pubads.g.doubleclick.net/gampad/clk?id=278785471&amp;iu=/4140</a><br>
                
--<br>
fuse-devel mailing list<br>
To unsubscribe or subscribe, visit <a \
href="https://lists.sourceforge.net/lists/listinfo/fuse-devel" rel="noreferrer" \
target="_blank">https://lists.sourceforge.net/lists/listinfo/fuse-devel</a><br> \
</font></span></blockquote></div><br></div></div>



------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785471&iu=/4140

-- 
fuse-devel mailing list
To unsubscribe or subscribe, visit https://lists.sourceforge.net/lists/listinfo/fuse-devel


[prev in list] [next in list] [prev in thread] [next in thread] 

Configure | About | News | Add a list | Sponsored by KoreLogic