Bug 841622

Summary: add virtio-scsi unlocked kick patches
Product: Red Hat Enterprise Linux 6 Reporter: Paolo Bonzini <pbonzini>
Component: kernelAssignee: Paolo Bonzini <pbonzini>
Status: CLOSED ERRATA QA Contact: Virtualization Bugs <virt-bugs>
Severity: medium Docs Contact:
Priority: medium    
Version: 6.3CC: areis, juzhang, knoel, pbonzini, perfbz, shu, sluo, wdai
Target Milestone: rc   
Target Release: ---   
Hardware: x86_64   
OS: All   
Whiteboard:
Fixed In Version: kernel-2.6.32-328.el6 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-02-21 06:40:53 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Paolo Bonzini 2012-07-19 15:38:50 UTC
Here are some results with varying number of disks on the same controller (data in MB/s, using fio, ioengine=libaio, iodepth=4):

#    6.3          unlocked kick       unlocked kick
#                 1 target            1 LUN/target
# --------------------------------------------------------------
1    303.679      343.454             << same configuration
2    343.979      393.545             460.247
4    409.953      447.683             455.197 (probably a fluke)
8    438.526      449.390             474.092
16   427.363      448.377             478.958

Comment 2 RHEL Program Management 2012-07-20 16:00:04 UTC
This request was evaluated by Red Hat Product Management for
inclusion in a Red Hat Enterprise Linux release.  Product
Management has requested further review of this request by
Red Hat Engineering, for potential inclusion in a Red Hat
Enterprise Linux release for currently deployed products.
This request is not yet committed for inclusion in a release.

Comment 4 Jarod Wilson 2012-10-10 19:43:53 UTC
Patch(es) available on kernel-2.6.32-328.el6

Comment 8 Shaolong Hu 2013-01-05 09:57:34 UTC
Hi Paolo,

Do you think QE need to repeat your test in comment 0? if so, could you provide qemu-kvm command line and fio command line in guest? In command 0, the "1 2 4 8 16" means boot guest with 1, 2, 4, 8, 16 scsi disks and in guest run fio with -filename=/dev/sda:/dev/sdb:/dev/sdc... ?

If not, we have tested virtio scsi and blk in a lot of test runs since kernel 328, could we simply verify this one based on there is no regression?

Comment 9 Shaolong Hu 2013-01-06 02:53:46 UTC
We have

Comment 10 Shaolong Hu 2013-01-06 02:59:17 UTC
Since kernel 328, we have tested virtio scsi:
Test run for qemu-kvm-0.12.1.2-2.331.el6 - virtio scsi - RHEL6.4-64

And lots of test runs use virtio scsi as block device, no regression so far, for sanity test, i'd like to set this one verified.

Comment 11 Paolo Bonzini 2013-01-09 13:49:53 UTC
I believe sanity testing is enough.

Comment 13 errata-xmlrpc 2013-02-21 06:40:53 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-2013-0496.html