Re: Bug#750959: linux-image-3.15-rc8-amd64: Add support for VGA VFIO (CONFIG_VFIO_PCI_VGA)

2014-06-09 Thread Andrei POPESCU
Control: reassign -1 src:linux

On Lu, 09 iun 14, 02:55:32, fld wrote:
 Source: linux-image-3.15-rc8-amd64
 Severity: wishlist
 
 Dear Maintainer,
 
 In the past few years with the increased availability of IOMMU/Vt-d supporting
 hardware, there has been an increase in the popularity of experimenting with
 VGA passtrough to virtual machines via IOMMU. Currently the most bleeding edge
 way of doing this is with the combination of kvm, qemu and vfio. However,
 most graphics cards don't play nice with just the plain vfio and need special
 considerations that need to be enabled in the kernel.
 
 I propose that we enable these special features called CONFIG_VFIO_PCI_VGA
 in the kernel so that people can more easily experiment with these new cool
 features. I asked the leading developer of VFIO VGA support (Alex Williamson)
 if he could think of any reason why not to enable this feature by default, and
 he could not think of any. The option should simply add the capability to use
 the x-vga feature in Qemu.
 
 Here some basic information about VFIO:
 https://www.kernel.org/doc/Documentation/vfio.txt
 
 Interesting presentations about VGA Passtroughs with VFIO:
 http://www.linux-kvm.org/wiki/images/b/b4/2012-forum-VFIO.pdf
 http://www.linux-kvm.org/wiki/images/e/ed/Kvm-forum-2013-VFIO-VGA.pdf
 
 There is also quite a bit of mainstream interest about this stuff:
 https://bbs.archlinux.org/viewtopic.php?id=162768
 
 Thank you.
 
 -- System Information:
 Debian Release: jessie/sid
   APT prefers unstable
   APT policy: (500, 'unstable'), (1, 'experimental')
 Architecture: amd64 (x86_64)
 
 Kernel: Linux 3.15.0-rc7 (SMP w/8 CPU cores)
 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
 Shell: /bin/sh linked to /bin/dash

-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic
http://nuvreauspam.ro/gpg-transition.txt


signature.asc
Description: Digital signature


Processed: Re: Bug#750959: linux-image-3.15-rc8-amd64: Add support for VGA VFIO (CONFIG_VFIO_PCI_VGA)

2014-06-09 Thread Debian Bug Tracking System
Processing control commands:

 reassign -1 src:linux
Bug #750959 [src:linux-image-3.15-rc8-amd64] linux-image-3.15-rc8-amd64: Add 
support for VGA VFIO (CONFIG_VFIO_PCI_VGA)
Warning: Unknown package 'src:linux-image-3.15-rc8-amd64'
Bug reassigned from package 'src:linux-image-3.15-rc8-amd64' to 'src:linux'.
Ignoring request to alter found versions of bug #750959 to the same values 
previously set
Ignoring request to alter fixed versions of bug #750959 to the same values 
previously set

-- 
750959: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=750959
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems


--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/handler.s.b750959.14023002009998.transcr...@bugs.debian.org



Bug#750997: linux-image-3.15-rc7-amd64: Kernel message Uhhuh. NMI received for unknown reason 2c on CPU 0. appears after suspend to disk

2014-06-09 Thread Patrick Häcker
Package: src:linux
Version: 3.15~rc7-1~exp1
Severity: normal

Dear Maintainer,

some seconds or minutes after waking up from suspend to disk after the NMI
watchdog has been disabled (e.g. with powertop or with echo 0 
/proc/sys/kernel/nmi_watchdog) before suspend, I get the following message
from the kernel:

 Uhhuh. NMI received for unknown reason 2c on CPU 0.
 Do you have a strange power saving mode enabled?
 Dazed and confused, but trying to continue

The exact timing seems to differ from run to run. The problem seems to be
identical to the problem reported here:
http://linux-kernel.2935.n7.nabble.com/Uhhuh-NMI-received-for-unknown-reason-2c-on-CPU-0-tp591524p628729.html
Please note that the linked thread is interleaved with an independent e1000
networking problem. The NMI problem seems to be unrelated from e1000 there
and it definitely is unrelated here, as no e1000 is loaded or plugged-in.

I only tested 3.15, RC7, but I guess, that the problem occurs with older
kernels, too, and that I didn't see it before, as I just recently switched
the aforementioned watchdog option. I can test different kernels if this
helps, but as the kernels works without any obvious problems, I haven't done
this, yet.

The end of a restore from disk run looks like the following:
[29240.016685] PM: restore of devices complete after 2379.959 msecs
[29240.017030] PM: Image restored successfully.
[29240.017063] PM: Basic memory bitmaps freed
[29240.017064] Restarting tasks ... done.
[29423.719271] Uhhuh. NMI received for unknown reason 2c on CPU 0.
[29423.719281] Do you have a strange power saving mode enabled?
[29423.719286] Dazed and confused, but trying to continue

Another, more detailed run, looks as follows:
[   69.624048] PM: Hibernation mode set to 'platform'
[   69.969074] (NULL device *): firmware: direct-loading firmware 
amd-ucode/microcode_amd.bin
[   69.969114] PM: Syncing filesystems ... done.
[   70.385415] Freezing user space processes ... (elapsed 0.001 seconds) done.
[   70.387587] PM: Marking nosave pages: [mem 0x00098000-0x000f]
[   70.387600] PM: Marking nosave pages: [mem 0xdff9-0x]
[   70.388839] PM: Marking nosave pages: [mem 0xd400-0xd7ff]
[   70.389642] PM: Basic memory bitmaps created
[   70.389699] PM: Preallocating image memory... done (allocated 218996 pages)
[   71.123610] PM: Allocated 875984 kbytes in 0.73 seconds (1199.97 MB/s)
[   71.123612] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) 
done.
[   71.125251] Suspending console(s) (use no_console_suspend to debug)
[   71.175132] PM: freeze of devices complete after 49.868 msecs
[   71.176441] PM: late freeze of devices complete after 1.304 msecs
[   71.177795] PM: noirq freeze of devices complete after 1.350 msecs
[   71.179148] ACPI: Preparing to enter system sleep state S4
[   71.180007] PM: Saving platform NVS memory
[   71.180527] Disabling non-boot CPUs ...
[   71.182262] kvm: disabling virtualization on CPU1
[   71.282478] smpboot: CPU 1 is now offline
[   71.284767] kvm: disabling virtualization on CPU2
[   71.284775] smpboot: CPU 2 is now offline
[   71.286771] kvm: disabling virtualization on CPU3
[   71.286823] smpboot: CPU 3 is now offline
[   71.288525] kvm: disabling virtualization on CPU4
[   71.390451] smpboot: CPU 4 is now offline
[   71.392044] kvm: disabling virtualization on CPU5
[   71.494456] smpboot: CPU 5 is now offline
[   71.494801] PM: Creating hibernation image:
[   71.667480] PM: Need to copy 220062 pages
[   71.667481] PM: Normal pages needed: 220062 + 1024, available pages: 2908835
[   71.495930] PM: Restoring platform NVS memory
[   71.496021] PCI-DMA: Resuming GART IOMMU
[   71.496021] PCI-DMA: Restoring GART aperture settings
[   71.496195] LVT offset 1 assigned for vector 0x400
[   71.496203] IBS: LVT offset 1 assigned
[   71.496274] Enabling non-boot CPUs ...
[   71.496382] x86: Booting SMP configuration:
[   71.496383] smpboot: Booting Node 0 Processor 1 APIC 0x1
[   71.507336] kvm: enabling virtualization on CPU1
[   71.509428] process: Switch to broadcast mode on CPU1
[   71.509647] CPU1 is up
[   71.509701] smpboot: Booting Node 0 Processor 2 APIC 0x2
[   71.520653] kvm: enabling virtualization on CPU2
[   71.522741] process: Switch to broadcast mode on CPU2
[   71.522914] CPU2 is up
[   71.522976] smpboot: Booting Node 0 Processor 3 APIC 0x3
[   71.533929] kvm: enabling virtualization on CPU3
[   71.536031] process: Switch to broadcast mode on CPU3
[   71.536208] CPU3 is up
[   71.536270] smpboot: Booting Node 0 Processor 4 APIC 0x4
[   71.547221] kvm: enabling virtualization on CPU4
[   71.549333] process: Switch to broadcast mode on CPU4
[   71.549510] CPU4 is up
[   71.549574] smpboot: Booting Node 0 Processor 5 APIC 0x5
[   71.560527] kvm: enabling virtualization on CPU5
[   71.562615] process: Switch to broadcast mode on CPU5
[   71.562789] CPU5 is up
[   71.569418] ACPI: Waking up from system sleep state S4
[   71.584196] PM: noirq restore of devices complete 

Bug#750925: make deb-pkg produces a non-installable package on the s390x architecture

2014-06-09 Thread Stephen Powell
On Sun, 08 Jun 2014 09:59:47 -0400 (EDT), maximilian attems wrote:
 
 see scripts/package/builddeb:
 s390*)
 debarch=s390 ;;
 
 this debarch setting very likely predates s390x

Thanks for the tip.  While looking at this file, I discovered
a circumvention.  Use the make variable

   KBUILD_DEBARCH=s390x

This works, even with an unmodified builddeb script.  In my test case,
I set CONFIG_LOCALVERSION to -1custom01-s390x during kernel configuration
(under General Setup).  I also set CONFIG_DEBUG_INFO off.   (This is under
Kernel Hacking.)  Then I issued

   make KDEB_PKGVERSION=3.14.4-1 KBUILD_PKG_ROOTCMD=fakeroot \
   INSTALL_MOD_STRIP=1 KBUILD_DEBARCH=s390x deb-pkg

as a non-root user.  It built a kernel image package just the way I wanted it.

-- 
  .''`. Stephen Powell
 : :'  :
 `. `'`
   `-


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/223528771.113881.1402361863179.javamail.r...@md01.wow.synacor.com