Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.
RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.

Bug 1140580

Summary: [performance] virtio-blk performance degradation happened with virito-serial
Product: Red Hat Enterprise Linux 7 Reporter: Sibiao Luo <sluo>
Component: qemu-kvmAssignee: Fam Zheng <famz>
Status: CLOSED WONTFIX QA Contact: Virtualization Bugs <virt-bugs>
Severity: high Docs Contact:
Priority: high    
Version: 7.1CC: amit.shah, chayang, famz, hhuang, juzhang, michen, pbonzini, rbalakri, sharpwiner, stefanha, virt-bugs, virt-maint, wquan, xfu, xigao
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: 1140521 Environment:
Last Closed: 2014-11-19 10:44:03 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 1140521    
Bug Blocks:    

Comment 1 Sibiao Luo 2014-09-11 09:49:43 UTC
Tried the qemu-kvm-1.5.3-69.el7.x86_64 also hit this issue, so clone it to qemu-kvm component to fix it.

host info:
# uname -r && rpm -q qemu-kvm
3.10.0-152.el7.x86_64
qemu-kvm-1.5.3-69.el7.x86_64
guest info:
rhel6.5 64bit GA

Results:
1.virtio_console module installed:
  64K-write-sequence: 516 MBPS, 8261 IOPS
2.virtio_console module uninstalled(modprobe -r virtio_console):
  64K-write-sequence: 565 MBPS, 9040 IOPS

Best Regards,
sluo

Comment 2 Fam Zheng 2014-11-19 10:44:03 UTC
Close this one to focus on qemu-kvm-rhev (Bug 1140521) for below reasons:

1) dataplane in qemu-kvm-rhev should already solve this. The root cause is slow "ppoll" implementation of host kernel, which is O(n) to # of internal iohandlers in QEMU. The # of iohandlers increases greatly when virtio-serial is enabled.

2) To fix this, we'd need to introduce Linux "epoll" as a new event loop backend of QEMU. This is relatively an invasive change to RHEL7 qemu-kvm, also as a strict performance enhancement instead of a bug fix.

3) "epoll", as the long term solution for upstream and qemu-kvm-rhev, will firstly be applied on dataplane (iothread), or even just skip the main loop in the end.

If we want to fix this on RHEL 7, feel free to justify and reopen the bug - in which case we can consider the other fix: modify virtio-serial so it doesn't use all 64 virtqueues unconditionally.

Fam