linux-ext4 - Re: unexpected sync delays in dpkg for small pre-allocated files on ext4

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <57505D8E.6080306@siemens.com>
Date:	Thu, 2 Jun 2016 18:23:42 +0200
From:	Gernot Hillier <gernot.hillier@...mens.com>
To:	"Theodore Ts'o" <tytso@....edu>,
	Dave Chinner <david@...morbit.com>, linux-scsi@...r.kernel.org,
	MPT-FusionLinux.pdl@...adcom.com, linux-ext4@...r.kernel.org,
	sathya.prakash@...adcom.com, chaitra.basappa@...adcom.com,
	suganath-prabu.subramani@...adcom.com
Subject: Re: unexpected sync delays in dpkg for small pre-allocated files on
 ext4

On 01.06.2016 16:12, Theodore Ts'o wrote:
> On Wed, Jun 01, 2016 at 03:17:14PM +0200, Gernot Hillier wrote:
>> I repeated the discussed tests and found comparable results on this machine:
>>
>> - 3 seconds dpkg install time on ext3 vs. 80 seconds for ext4
>>   on same partition for same package
>> - 40 ms for fallocate+write+sync_file_range for writing a few bytes
>> - 15 ms for write+fdatasync vs. 45 ms for BLKZEROOUT on raw device
>>
>> So this seems to be not bound to one specific disk+controller setup, but
>> it still can't be a common problem affecting many people as then we
>> would see more reports about it...
> 
> OK, so let's try to get common baseline, shall we?  

Sure!

> I'm using as my test package:
> 
> http://debug.mirrors.debian.org/debian-debug/pool/main/e/e2fsprogs/e2fsprogs-dbgsym_1.43-3_amd64.deb
[...]
> # /usr/bin/time dpkg --force-depends -i /tmp/e2fsprogs-dbgsym_1.43-3_amd64.deb 
[...]
> 0.21user 0.19system 0:00.54elapsed 74%CPU (0avgtext+0avgdata 73188maxresident)k
> 0inputs+10168outputs (0major+24935minor)pagefaults 0swaps

On machine with Intel SCSI controller and on ext4 root, I see:

0.15user 0.00system 0:00.92elapsed 16%CPU (0avgtext+0avgdata
14144maxresident)k
0inputs+0outputs (0major+5447minor)pagefaults 0swaps

Installing to some ext3 partition:

0.14user 0.01system 0:00.56elapsed 26%CPU (0avgtext+0avgdata
14092maxresident)k
0inputs+0outputs (0major+5431minor)pagefaults 0swaps

This doesn't really advertise the issue as it only installs a small
number of files - yet, I still see a reproducible factor of 1.5-2 for
total elapsed time between ext3 and ext4...

> Something else would be useful is to make sure /tmp is a tmpfs mounted
> file system (so you don't get any I/O's caused by /tmp), and then do:

Ok.

> # dpkg --purge e2fsprogs-dbgsym
> # btrace /dev/sda3 > /tmp/btrace.out &
> # /usr/bin/time dpkg --force-depends -i /tmp/e2fsprogs-dbgsym_1.43-3_amd64.deb
> # kill %btrace

Please find btrace.out.bz2 attached. HTH...

Please note that I'm offline for the next couple of days, so there will
be some delay in my answers!

-- 
Gernot Hillier

Siemens AG, Corporate Technology
Corporate Competence Center Embedded Linux

Download attachment "btrace.out.bz2" of type "application/x-bzip" (14203 bytes)