[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Xen-devel][PV-ops][PATCH] Netback: Fix PV network issue fornetback multiple threads patchset



I can confirm that this patch solves the problem for me.  Tested in Win7
x64 with GPLPV 0.11.0.213, and with a PV linux domain running fedora 12
64bit.

Doug Magee

-----Original Message-----
From: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
[mailto:xen-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Xu, Dongxiao
Sent: Thursday, June 10, 2010 7:49 AM
To: Jeremy Fitzhardinge
Cc: xen-devel@xxxxxxxxxxxxxxxxxxx; djmagee@xxxxxxxxxxxx;
Pasi@xxxxxxxxxxxxxxxxxxxx; Fantu
Subject: [Xen-devel][PV-ops][PATCH] Netback: Fix PV network issue
fornetback multiple threads patchset

Hi Jeremy,

The attached patch should fix the PV network issue after applying the
netback multiple threads patchset.

This bug comes from the version upgrade when resending my patchset last
time. 

In patch v1 & v2, I used kzalloc() to allocate memory, so there is no
problem.
In patch v3, I changed it to "__get_free_pages(GFP_KERNEL | __GFP_ZERO,
order)", so there is also no problem.
However in patch v4, when modified it to vmalloc according to the
comments, memset(0) is missing. Sorry for that.

*ONE POINT IS*, the phenomenon on my side is different from the bug
reports from mailing list. They often saw GPLPV network long ping
latency, and on my side, system would crash when enabling guest network.


After applying the patch, network in the following cases are working
fine in my side. 

WinXP 32bit with gplpv_XP_0.11.0.213.msi 64bit Linux PV guest.
64bit RHEL5u3 HVM guest with PV driver

Can somebody help to try this fixing for long ping latency phenomenon?
Appreciate for that!

Thanks,
Dongxiao

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.