Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Fri, Apr 08, 2016 at 10:17:03PM +0200, Tomasz Golinski wrote: > >> > Can you > >> > guys please test if the attached patch fixes the issues for you as it > >> > did for me? > > Tested, works as expected. Thanks. Attaching dmesg from patched kernel > in case you need it. Thanks for verifying th

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

>> > Can you >> > guys please test if the attached patch fixes the issues for you as it >> > did for me? Tested, works as expected. Thanks. Attaching dmesg from patched kernel in case you need it. dmesg-patch.log.gz Description: application/gunzip ___

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Fri, Apr 08, 2016 at 05:39:41PM +0200, Tomasz Golinski wrote: > > Okay, with this card I was able to reproduce the problem myself. Can you > > guys please test if the attached patch fixes the issues for you as it > > did for me? > > Sorry for asking (maybe) stupid question. This patch doesn't a

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

> Okay, with this card I was able to reproduce the problem myself. Can you > guys please test if the attached patch fixes the issues for you as it > did for me? Sorry for asking (maybe) stupid question. This patch doesn't apply against 4.5 kernel. Should I apply it against current git head or mayb

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Wed, Apr 06, 2016 at 05:52:57PM +0200, Tomasz Golinski wrote: > I'm using Asus Xonar DX card: https://www.asus.com/Sound-Cards/Xonar_DX/ Okay, with this card I was able to reproduce the problem myself. Can you guys please test if the attached patch fixes the issues for you as it did for me? T

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

Le mercredi 06 avril 2016 à 17:47 +0200, Joerg Roedel a écrit : > Vendor and exact product name would be helpful. I use the Asus Essence STX II (the one withOUT the additional 7.1 surround card): http://www.asus.com/Sound-Cards/Essence_STX_II/ I use it in its original configuration, without chan

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

> > > Martin/Tomasz, could we get the amd_iommu_dump dmesg output for > > > an earlier kernel? Sorry, I don't have access to that system at the moment and supply the info as early as I get back home in 10 days. Please keep me in cc. As for the other questions: - no proprietary drivers installed

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

Joerg Roedel wrote: > Maybe I can get one of those cards myself to reproduce it. Can you guys > send me some more detailed information about the sound cards that cause > the problems? Most of the PCIe Xonar cards use the same architecture: a PCI chip and a PCIe/PCI bridge. Consider the Asus Xonar

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

> Can you guys send me some more detailed information about > the sound cards that cause the problems? Vendor and exact product name > would be helpful. I'm using Asus Xonar DX card: https://www.asus.com/Sound-Cards/Xonar_DX/ Let me add that I also did no BIOS upgrade in the meantime and I use no

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Wed, Apr 06, 2016 at 09:06:31AM +0200, Martin Haiden wrote: > Sorry, I don't have access to that system at the moment and supply the > info as early as I get back home in 10 days. Please keep me in cc. > > As for the other questions: > > - no proprietary drivers installed > - no BIOS update H

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On 05.04.2016 14:26, Clemens Ladisch wrote: > Martin/Tomasz, could we get the amd_iommu_dump dmesg output for an > earlier kernel? > It's for 4.1.15. Both kernels were built with Gentoo patches but I don't think they added anything in related areas. I can build vanilla kernels if it would be bett

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On 05.04.2016 14:49, Joerg Roedel wrote: >> [1.722866] nvidia: module license 'NVIDIA' taints kernel. > > Hmm, is this reproducible without the propietary nvidia-driver loaded? Yes, I get errors also w/o nvidia module (attaching dmesg). On 05.04.2016 15:15, Clemens Ladisch wrote: > Apparently

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Tue, Apr 05, 2016 at 03:15:10PM +0200, Clemens Ladisch wrote: > Tomasz Golinski wrote: > > On 05.04.2016 14:26, Clemens Ladisch wrote: > >> Martin/Tomasz, could we get the amd_iommu_dump dmesg output for an > >> earlier kernel? > > > > It's for 4.1.15. > > ... > > [0.486934] AMD-Vi: DEV_SP

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

Tomasz Golinski wrote: > On 05.04.2016 14:26, Clemens Ladisch wrote: >> Martin/Tomasz, could we get the amd_iommu_dump dmesg output for an >> earlier kernel? > > It's for 4.1.15. > ... > [0.486934] AMD-Vi: DEV_SPECIAL(IOAPIC[5]) devid: 00:14.0 > [0.569054] AMD-Vi: Found IOMM

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Tue, Apr 05, 2016 at 02:39:15PM +0200, Tomasz Golinski wrote: > [1.722866] nvidia: module license 'NVIDIA' taints kernel. Hmm, is this reproducible without the propietary nvidia-driver loaded? Joerg ___ iommu mailing list iommu@lists.l

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

Joerg Roedel wrote: > On Tue, Apr 05, 2016 at 12:02:07PM +0200, Clemens Ladisch wrote: >> Joerg Roedel wrote: >>> I have seen that before, can you please boot with 'amd_iommu_dump' on >>> the kernel command line and send me dmesg after boot? >> >> Full dmesg:

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

On Tue, Apr 05, 2016 at 12:02:07PM +0200, Clemens Ladisch wrote: > Joerg Roedel wrote: > > I have seen that before, can you please boot with 'amd_iommu_dump' on > > the kernel command line and send me dmesg after boot? > > Full dmesg: : > ... >

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

Joerg Roedel wrote: > I have seen that before, can you please boot with 'amd_iommu_dump' on > the kernel command line and send me dmesg after boot? Full dmesg: : ... [0.179504] pci :04:00.0: [10b5:8112] type 01 class 0x060400 [0.179

Re: AMD-Vi regression with PCIe/PCI bridges in 4.4

Hi Clemens, On Sun, Apr 03, 2016 at 08:59:53AM +0200, Clemens Ladisch wrote: > there are several reports of Xonar PCIe sound cards failing to work in > kernel 4.4 (including 4.4.5): > > https://bugzilla.redhat.com/show_bug.cgi?id=1313790 > https://bugzilla.kernel.org/show_bug.cgi?id=111711 > http

AMD-Vi regression with PCIe/PCI bridges in 4.4

Hi, there are several reports of Xonar PCIe sound cards failing to work in kernel 4.4 (including 4.4.5): https://bugzilla.redhat.com/show_bug.cgi?id=1313790 https://bugzilla.kernel.org/show_bug.cgi?id=111711 https://bbs.archlinux.org/viewtopic.php?pid=1601338#p1601338 They show error messages li