Bug 708000

Summary: cifs: asynchronous writepages support
Product: Red Hat Enterprise Linux 6 Reporter: Jeff Layton <jlayton>
Component: kernelAssignee: Jeff Layton <jlayton>
Status: CLOSED ERRATA QA Contact: Jian Li <jiali>
Severity: medium Docs Contact:
Priority: medium    
Version: 6.1CC: bfields, dhowells, eguan, jiali, kzhang, nmurray, rwheeler, sprabhu, ssaha, steved, yanwang
Target Milestone: rcKeywords: FutureFeature
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: kernel-2.6.32-168.el6 Doc Type: Enhancement
Doc Text:
Story Points: ---
Clone Of:
: 719308 (view as bug list) Environment:
Last Closed: 2011-12-06 13:08:17 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 695824, 719308    

Description Jeff Layton 2011-05-26 14:12:32 UTC
The patchset to add asynchronous write support for cifs was just merged into Steve F.'s tree and should be going into mainline soon. It speeds up writes fairly significantly and is something we will probably want to backport for 6.2 or 6.3.

Comment 1 RHEL Program Management 2011-05-26 14:30:09 UTC
This request was evaluated by Red Hat Product Management for inclusion
in a Red Hat Enterprise Linux maintenance release. Product Management has 
requested further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed 
products. This request is not yet committed for inclusion in an Update release.

Comment 7 Jeff Layton 2011-07-06 12:50:02 UTC
Basically the best way to test this is to do a lot of buffered I/O and measure the performance. A simple way:

$ dd if=/dev/zero of=/file/on/cifs/share bs=1M count=XXXXX conv=fsync

...or something like that. It should be faster with this patchset in place, but it's hard to know how much faster as that's dependent on many different factors. While you're at it, you can also try mounting with different wsize= options and verifying that they work as expected.

Comment 8 Kyle McMartin 2011-07-13 23:01:23 UTC
Patch(es) available on kernel-2.6.32-168.el6

Comment 11 Jian Li 2011-10-21 05:42:38 UTC
patchs are checked and regression test are done.

[root@dell-pet610-01 lease]# dd if=/dev/zero of=/mnt/test/test bs=1M count=10K conv=fsync
10240+0 records in
10240+0 records out
10737418240 bytes (11 GB) copied, 165.412 s, 64.9 MB/s
[root@dell-pet610-01 lease]# uname -a
Linux dell-pet610-01.lab.bos.redhat.com 2.6.32-209.el6.x86_64 #1 SMP Wed Oct 12 03:54:10 EDT 2011 x86_64 x86_64 x86_64 GNU/Linux
[root@dell-pet610-01 lease]# mount | grep cifs
//dell-pe860-01.rhts.eng.bos.redhat.com/test on /mnt/test type cifs (rw)


[root@dell-pet610-01 ~]# dd if=/dev/zero of=/mnt/test/test2 bs=1M count=10K conv=fsync
10240+0 records in
10240+0 records out
10737418240 bytes (11 GB) copied, 188.81 s, 56.9 MB/s
[root@dell-pet610-01 ~]# uname -a
Linux dell-pet610-01.lab.bos.redhat.com 2.6.32-167.el6.x86_64 #1 SMP Fri Jul 8 11:24:13 EDT 2011 x86_64 x86_64 x86_64 GNU/Linux

Comment 12 errata-xmlrpc 2011-12-06 13:08:17 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHSA-2011-1530.html