From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paraic Gallagher Subject: Question about backporting pci passthrough fix to 2.6.18 Date: Thu, 1 Jul 2010 10:04:01 +0100 Message-ID: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0268866689==" Return-path: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org --===============0268866689== Content-Type: multipart/alternative; boundary=0015174c3d8866939a048a4fc0c3 --0015174c3d8866939a048a4fc0c3 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi, I have a question about the bug: http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=3D1632 This is marked as fixed in the pv-ops kernel. I am working for a company developing a xen based virtualisation solution. Unfortunately the design is tied to CentOS and there is resistance to upgrade to > 2.6.31 kernel. My question is whether it is possible for me to try to backport the fixes for this pci passthrough issue to a 2.6.18 kernel, or are there too many differences between the kernel versions to make this feasible/possible. thanks all, P=E1raic. ------- Comment #1 from konrad.wilk@oracle.com 2010-06-30 07:54 ------- There is a fix for this in the pv-ops kernel. Specifically: xen-pciback: Allocate IRQ handler for device that is shared with guest. xen-pciback: Add SysFS instrumentation for the fake IRQ handler. xen-pciback: When device transitions to MSI/MSI-X stop ACK-ing on the legacy interrupt. xen-pciback: Enable interrupt handler when device is enabled. xen-pciback: Probe the IRQ line to check if it is not shared Here is a howto on PV-OPS:http://wiki.xensource.com/xenwiki/XenParavirtOps and how to migrate from RHEL5 to it http://wiki.xensource.com/xenwiki/2.6.18-to-2.6.31-and-higher Closing bug as FIXED in upstream. --0015174c3d8866939a048a4fc0c3 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi,

I have a question about the bug:
http://bugz= illa.xensource.com/bugzilla/show_bug.cgi?id=3D1632

This is marke= d as fixed in the pv-ops kernel.

I am working for a company developing a xen based virtualisation soluti= on.
Unfortunately the design is tied to CentOS and there is resistance = to upgrade to > 2.6.31 kernel.

My question is whether it is possi= ble for me to try to backport the fixes for this pci passthrough issue to a= 2.6.18 kernel, or are there too many differences between the kernel versio= ns to make this feasible/possible.

thanks all,
P=E1raic.


------- Comment #1 from konrad.wilk@oracle.com =A02010-06-30 07:54 -------
There is a fix for this in the pv-ops kernel.

Specifically:
=A0xen-pciback: Allocate IRQ handler for device that is shared with=20 guest.
=A0 =A0 =A0xen-pciback: Add SysFS instrumentation for the fake IRQ handler= .
=A0 =A0 =A0xen-pciback: When device transitions to MSI/MSI-X stop ACK-ing = on=20 the
legacy interrupt.
=A0 =A0 =A0xen-pciback: Enable interrupt handler when device is enabled. =A0 =A0 =A0xen-pciback: Probe the IRQ line to check if it is not shared
Here is a howto on PV-OPS:http://wiki.xensource.com/xenwiki/XenParavirt= Ops
and how to migrate from RHEL5 to it
http://wiki.xensource.com/xenwiki/2.6.18-to-2.6.31-and-hig= her

Closing bug as FIXED in upstream.
--0015174c3d8866939a048a4fc0c3-- --===============0268866689== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel --===============0268866689==-- From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: Question about backporting pci passthrough fix to 2.6.18 Date: Thu, 1 Jul 2010 11:40:14 -0400 Message-ID: <20100701154014.GF31947@phenom.dumpdata.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Paraic Gallagher Cc: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org On Thu, Jul 01, 2010 at 10:04:01AM +0100, Paraic Gallagher wrote: > Hi, > > I have a question about the bug: > http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=1632 > > This is marked as fixed in the pv-ops kernel. > > I am working for a company developing a xen based virtualisation solution. > Unfortunately the design is tied to CentOS and there is resistance to > upgrade to > 2.6.31 kernel. > > My question is whether it is possible for me to try to backport the fixes > for this pci passthrough issue to a 2.6.18 kernel, or are there too many > differences between the kernel versions to make this feasible/possible. You should be able to do it. The SysFS parts might need some tweaking. > > > ------- Comment #1 from konrad.wilk@oracle.com 2010-06-30 07:54 ------- > There is a fix for this in the pv-ops kernel. > > Specifically: > xen-pciback: Allocate IRQ handler for device that is shared with guest. > xen-pciback: Add SysFS instrumentation for the fake IRQ handler. > xen-pciback: When device transitions to MSI/MSI-X stop ACK-ing on the > legacy interrupt. > xen-pciback: Enable interrupt handler when device is enabled. > xen-pciback: Probe the IRQ line to check if it is not shared > > Here is a howto on PV-OPS:http://wiki.xensource.com/xenwiki/XenParavirtOps > and how to migrate from RHEL5 to it > http://wiki.xensource.com/xenwiki/2.6.18-to-2.6.31-and-higher > > Closing bug as FIXED in upstream. > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel