Bug 503191

Summary: [RHEL5.4 Xen]: Tun patch causing connectathon to fail
Product: Red Hat Enterprise Linux 5 Reporter: Chris Lalancette <clalance>
Component: kernel-xenAssignee: Herbert Xu <herbert.xu>
Status: CLOSED ERRATA QA Contact: Red Hat Kernel QE team <kernel-qe>
Severity: medium Docs Contact:
Priority: low    
Version: 5.4CC: dzickus, herbert.xu, markmc, peterm, sghosh, syeghiay, xen-maint
Target Milestone: rcKeywords: Regression
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-09-02 08:57:52 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:
Attachments:
Description Flags
epoll keyed wakeups: add __wake_up_locked_key() and __wake_up_sync_key()
none
epoll keyed wakeups: introduce new *_poll() wakeup macros
none
tun: Only wake up writers
none
tun: Only wake up writers none

Description Chris Lalancette 2009-05-29 14:36:42 UTC
Description of problem:
This may or may not be a dup of 500865; I don't think it is at the moment, so I'll file it separately.  Since the -147 kernel, running the RHTS connectathon test in a fully virtualized Xen guest completely fails.  It actually doesn't even start at all; it just hangs when trying to start the first test.  I bisected this down to the rhel5/kernel commit fe4248cfc7fe060a7537f5ae3a37142c77d23bcb.  We'll need to look at fixing this or reverting this patch, otherwise FV guests are going to be very unhappy.

Comment 10 Herbert Xu 2009-06-03 03:35:16 UTC
Created attachment 346345 [details]
epoll keyed wakeups: add __wake_up_locked_key() and __wake_up_sync_key()

Comment 11 Herbert Xu 2009-06-03 03:35:47 UTC
Created attachment 346346 [details]
epoll keyed wakeups: introduce new *_poll() wakeup macros

Comment 12 Herbert Xu 2009-06-03 03:36:16 UTC
Created attachment 346347 [details]
tun: Only wake up writers

Comment 14 Herbert Xu 2009-06-03 12:38:21 UTC
Created attachment 346384 [details]
tun: Only wake up writers

tun: Only wake up writers

When I added socket accounting to tun I inadvertently introduced
spurious wake-up events that kills qemu performance.  The problem
occurs when qemu polls on the tun fd for read, and then transmits
packets.  For each packet transmitted, we will wake up qemu even
if it only cares about read events.

Now this affects all sockets, but it is only a new problem for
tun.  So this patch tries to fix it for tun first and we can then
look at the problem in general.

Signed-off-by: Herbert Xu <herbert.org.au>

Comment 17 Juan Quintela 2009-06-04 11:45:24 UTC
It also breaks kvm networking. Going back to kernel -149 fixed the problem.

Comment 18 Juan Quintela 2009-06-04 11:58:13 UTC
Grrr, wrong typing.  I am using RHEL 5.4 beta. (-150 kernel).
Went back to -146 as clalance suggestion, and now I have bridging working back with kvm.  I installed -147, and that one already get broked networking.

Comment 25 Don Zickus 2009-06-11 15:37:20 UTC
in kernel-2.6.18-153.el5
You can download this test kernel from http://people.redhat.com/dzickus/el5

Please do NOT transition this bugzilla state to VERIFIED until our QE team
has sent specific instructions indicating when to do so.  However feel free
to provide a comment indicating that this fix has been verified.

Comment 28 errata-xmlrpc 2009-09-02 08:57:52 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2009-1243.html