From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jean Baptiste Favre Subject: Re: PCI passthrough issue Date: Mon, 28 Feb 2011 13:18:13 +0100 Message-ID: <4D6B9285.9090903@jbfavre.org> References: <4D4BBEC6.8070809@jbfavre.org> <4D4BD121.2080505@jbfavre.org> <1296817460.13091.646.camel@zakaz.uk.xensource.com> <4D4BE212.1090400@jbfavre.org> <1296818935.13091.648.camel@zakaz.uk.xensource.com> <4D4BFBE4.6080809@jbfavre.org> <1296827449.13091.670.camel@zakaz.uk.xensource.com> <4D4C06BB.8010907@jbfavre.org> <4D52658E.9060907@jbfavre.org> <4D5EE125.6040208@jbfavre.org> <20110225144035.GA7228@dumpdata.com> Reply-To: xen-devel@lists.xensource.com Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <20110225144035.GA7228@dumpdata.com> 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 Hello, Le 25/02/2011 15:40, Konrad Rzeszutek Wilk a =E9crit : > On Fri, Feb 18, 2011 at 10:14:13PM +0100, Jean Baptiste Favre wrote: >> Hello, >> Back online after my exams :) >> >> Had some time to perform tests with my Debian Squeeze 32bits domU and >> 2.6.37 kernel from experimental. >=20 > I am bit lost now.. Can you refresh my memory whether the 'copy_break' > parameter worked or not? >=20 > Experimental is the stock kernel or was that a new proposed kernel? > Where are the sources for the experimental kernel? >=20 >> DomU config is: >> ***************************************************************** >> kernel =3D '/cluster/kernels/vmlinuz-2.6.37-trunk-686-bigmem' >> ramdisk =3D '/cluster/kernels/initrd.img-2.6.37-trunk-686-bigmem' >> #kernel =3D '/cluster/kernels/vmlinuz-2.6.37-trunk-686-bigmem-sk= y2' >> #ramdisk =3D '/cluster/kernels/initrd.img-2.6.37-trunk-686-bigmem= -sky2' >> builder =3D 'linux' >> memory=3D268 >> vcpus =3D '1' >> cpus =3D '2' >> localtime =3D 0 >> serial =3D 'pty' >> boot =3D 'cdn' >> disk =3D [ 'drbd:xps-106,xvda,w' ] >> on_poweroff =3D 'destroy' >> on_reboot =3D 'restart' >> on_crash =3D 'restart' >> name =3D 'xps-106' >> hostname =3D 'xps-106.clichy.jbfavre.org' >=20 > I had a box with a sky2 adapter that looked to have a similar issue but > found the culprit to be the switch. So at this point I am having no luc= k > reproducing this. Would it be possible for you to stick the kernel + di= st > image somewhere so I can try it out on my box? >=20 >> >> extra =3D "root=3D/dev/mapper/xps--106-root ro iommu=3Dsoft swiotlb=3D= force >> console=3Dhvc0 xencons=3Dtty" >=20 > Try without 'swiotlb=3Dforce' on any kernel that is PVOPS. Only the old= er > ones (lenny) required that. And you don't need 'xencons=3Dtty' either w= ith > PVOPS kernels. Thought I already tried without 'swiotlb=3Dforce', but seems not... Just tried it on my Debian domU, and everything works now, whatever copybreak value can be. But trying the same with OpenWRT gives following results: - With 'swiotlb=3Dforce': fails for size equal or greater than 128 bytes (or ping -s86) - Without 'swiotlb=3Dforce': always fails I can provide OpenWRT disk img and kernel. For Debian domU, I can provide a dd from LVM and kernel. Let me know which ones you want. Regards, JB