similar to: AMD iommu , pci passthrough problems assigning devices to HVM

*RESEND* due to exceeding the mailinglists limit for attachment size. Hi, I''m trying to get secondary vga-passthrough on a HVM guest to work with a AMD HD6570 and the native kernel radeon driver and kernel modesetting. So the guest still gets the emulated stdvga or cirrus device(used in my case here) as primary/boot vga adapter. - When i don''t passthrough the radeon card, the

From: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> This patch handle additional cases for IVRS bugs where special->handle is not correctly initialized for IOAPIC and HPETS due to firmware bugs. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Provide logic in "is_ioapic_overidden()" Signed-off-by: Jan Beulich <JBeulich@suse.com> ---

This patch set adds basic supports for amd next generation iommu (IOMMUv2) hardware. IOMMUv2 supports various new features advertised by iommu extended feature register. It introduces guest level IO translation and supports state-of-the-art ATS/ATC devices with demand paging capability. Please refer to AMD IOMMU Architectural Specification [1] for more details. Thanks, Wei [1]

Hello everyone, Reposting with after having applied the (minor) fixes suggested by Wei and Jan. Allen, if you can tell us what you think about this, or suggest someone else to ask some feedback to, if you''re no longer involved with VT-d, that would be great! :-) -- As already discussed here [1], dealing with IOMMU faults in interrupt context may cause nasty things to happen, up to

From: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> This patch add an additional logic to check for the often case when the special->handle is not initialized due to firmware bugs. but the special->usedid is correct. If users overide this using the command line option ivrs_ioapic, then it should use the value instead. --- This patch is supposed to follow the patches:

Hi, All the reviews for the AMD 890fx chipset reported that it supports IOMMU v1.2. But I cannot find any mention of that on AMD site. Does 890fx support IOMMU? Do 890fx motherboards with IOMMU exist? Does Xen support 890fx? Regards, Leonardo _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

The iommu is enabled by default when xen is booting and later disabled in iommu_setup() when no iommu is present. But under some circumstances iommu-code can be called before iommu_setup() is processed. If there is no iommu available xen crashes. This can happen for example when panic(...) is called that got introduced with patch "x86-64: detect processors subject to AMD erratum #121 and

-- Advanced Micro Devices GmbH Sitz: Dornach, Gemeinde Aschheim, Landkreis München Registergericht München, HRB Nr. 43632 WEEE-Reg-Nr: DE 12919551 Geschäftsführer: Alberto Bozzo, Andrew Bowd _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

+cc: ocfs2-devel, sunil Marek, Thanks for the detailed report! Can I trouble you to file a bug at http://oss.oracle.com/bugzilla ... oh, is that your report http://oss.oracle.com/bugzilla/show_bug.cgi?id=1339? This looks like a deadlock of some sort. I don't think your hardware configuration really matters here. We'll keep looking into it. Joel On Tue, Dec 20, 2011 at 08:22:48AM

Hello, I was hoping someone might be able to tell me why IOMMU does not seem to be found on my system. At the very least is there any way I can get more debug information in addition to adding iommu=verbose to the command line? Xen 4.0.0 MSI 890FXA-GD70 (I spoke with an MSI representative who confirmed that this motherboard does support IOMMU) AMD Phenom II 1090T x6 # dmesg __ __

Hi Jan, Attached patch dumps io page fault flags. The flags show the reason of the fault and tell us if this is an unmapped interrupt fault or a DMA fault. Thanks, Wei signed-off-by: Wei Wang <wei.wang2@amd.com> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel

Hi, I migrate Xen4.0 to XEN4.1, and found XEN4.1 performance is not as good as 4.0.2. xm shutdown takes longer time, control thru libvirt take longer time, and since I''m testing PCI/VGA passthru, the "createdevice: pci" also take around 5~8 seconds comparing to 1 second in XEN4.0.2. below is my configuration, do you have any suggestions to make the performance better?

I'm running the following hardware: MSI 890FXA-GD70 motherboard With a AMI v1.7 BIOS AMD Phenom II X6 (1055T) 8 GB RAM (2 x 4 GB DIMM) nVidia Geforce 7300 (G72) video Card ViewSonic VA2431wvm monitor Plus The OS is: CentOS release 6.3 (Final) 2.6.32-279.5.2.el6.centos.plus.x86_64 Actually, on this machine I'm getting the following error during

Hello, I would like to report to the group that I have been successful with VGA passthrough using the following configuration: Gigabyte GA-890FXA-UD5 ASUS EAH5750 1GB GDDR5 video Fedora 13 XEN 4.0.1 Windows 7 64 bit HVM guest Xen was installed per Pasi''s Fedora 13 Xen 4.0 Tutorial found here: http://wiki.xen.org/xenwiki/Fedora13Xen4Tutorial The tutorial was very helpful and many

Hi Jan, After commit 2ca9fbd739b8a72b16dd790d0fff7b75f5488fb8 AMD IOMMU: allocate IRTE entries instead of using a static mapping, booting dom0 stalls several times. Sometimes this results in RCU stall warnings from the dom0 kernel, hitting the "any" key, on normal or serial console, makes the boot continue for a while but it stalls several times. (It also stalls on shutdown BTW) I have

https://bugs.freedesktop.org/show_bug.cgi?id=84424 Priority: medium Bug ID: 84424 Assignee: nouveau at lists.freedesktop.org Summary: nouveau crash log [mesa-10.3.0] Severity: normal Classification: Unclassified OS: All Reporter: reuben.m at gmail.com Hardware: Other Status: NEW

after a clean install of ubuntu 10.4 (Lucid) i copied a completely installed wow game over from another computer, installed wine (1.2 i believe) through the get manager. changed the opengl settings. first issue. anything i run in wine takes over five minutes to load. the cursor changes to the thinking cursor for about thirty seconds and the opening wow.exe (or whatever.exe) bar appears on the

Hi Wei, I ran into the following problem when trying to boot another guest after less than a day of uptime. (the system started 15 guests at boot already which went fine). dom0 is allocated a fixed 1536M. Both host as pv guests run the same kernel, some hvm''s run a slightly older kernel (3.9 f.e.) The are quite some granttable messages in xl dmesg, i also included these and a

Package: xen-utils-common Version: 4.4.0-5 Severity: important Tags: patch Hello, xen-init-list fails to parse output of xen list -l when using XM toolstack. Therefore, /etc/init.d/xendomains stop fails to shutdown any guests still running on shutdown. xen-init-list fails with the following message: /usr/lib/xen-common/bin/xen-init-list Traceback (most recent call last): File

Hi All, Please help me. I am now installing the XEN on my PC and trying to config the pci pass through. My motherboard is intel DQ45CB which is vtd enabled. I downloaded the latest xen4.0 source from xen.org which is the latest release. I downloaded stable-2.6.32-x dom0 kernel source from jeremy''s repository. After compiling the xen and dom0 kernel, I could play it. I could create