[ANNOUNCE] kvm-61 release

[ANNOUNCE] kvm-61 release

[Avi Kivity]

Long time since the last release, hence the long changelog.  If you have 
an AMD Barcelona, you should see a very significant performance boost 
from the Nested Page Table feature.

Changes from kvm-60:
- paravirtualized clock (Glauber de Oliveira Costa)
- mmu debug compile fix (Marcelo Tosatti)
- last branch record virtualization on AMD (Joerg Roedel)
- fix Windows x64 installer crash on AMD (Joerg Roedel)
- use new kvm semaphore instead of mmap_sem (Izik Eidus)
   - fixes potential deadlock
- fix type in vmx header (Ryan Harper)
- make the supported cpuid ioctl system-wide instead of per-vm
- fix setting cr0 from userspace (Paul Knowles)
- add group decoding support to the x86 emulator
- nested paging support on AMD (Joerg Roedel)
- allow access to EFER on 32-bit hosts (Joerg Roedel)
   - allows using the nx bit
- fix copy_from_user_inatomic() call non-preemptible hosts
   (Andrea Arcangeli)
- fix smp Linux on AMD (Joerg Roedel)
- make the code more ppc friendly (Hollis Blanchard)
- add virtual processor identifier support (VPID) on Intel
   (Sheng Yang)
- don't expose pio page to host userspace on non-x86
- decouple mmio from shadow page tables
- make function pointer tables constant (Jan Engelhardt)
- merge qemu-cvs
   - reverse vnc connections
   - e1000 nic emulation
   - more audio options
   - usb serial device emulation
   - cdrom/dvd improvements
   - timer tick reduction
- e1000 pxe boot rom (Alexey Eremenko)
- fix 'make clean'
- merge bios from boch-cvs
   - adds dmi tables
   - enables acpi on 32-bit Linux guests
- workaround qemu tcg breakage on non-x86 (Anthony Liguori, Xiantao Zhang)
- adjust external module for removal of fastcall in mainline
   (Andrea Arcangeli)
- improve qemu main loop
- fix live migration with tpr optimization
- fix external module miscompile with newer kernels (Andrea Arcangeli)
- fix external module include directory priority problems (Andrea Arcangeli)
- fix 32-bit host userspace on 64-bit host kernels
- fix preempt notifier emulation clash with host hardware debug registers
   (Andrea Arcangeli)
- adjust external module for 2.6.25
- fix ioapic live migration not transferring some registers
   (Chris Lalancette)
- separate tpr optimization paths for smp and up
- pio emulation on mmio for ia64 (Xiantao Zhang)
- fix qemu timer rearm fallout from last merge (Anders Melchiorsen)
- clean up kvm/qemu interaction (Anthony Liguori)
- improve ppc dcr handling (Jerone Young)
- fix 64-bit guest migration failure due to bad lm_capable_kernel init
   (Uri Lublin)
- fix bios parallel build (Carlo Marcelo Arenas Belon)
- update virtio abi (Anthony Liguori)
- support virtio abi versioning (Anthony Liguori)
- add powerpc kvm/qemu support (Jerone Young)
- support ppc 440 "bamboo" board (Jerone Young)
- improve migration error handling (Chris Lalancette)
- enable cache on reset in bios
- fix up testsuite warnings (Bernhard Kaindl)
- fix aio deadlocks with smp guests


Notes:
      If you use the modules bundled with kvm-61, you can use any version
of Linux from 2.6.17 upwards.
      If you use the modules bundled with Linux 2.6.20, you need to use
kvm-12.
      If you use the modules bundled with Linux 2.6.21, you need to use
kvm-17.
      Modules from Linux 2.6.22 and up will work with any kvm version from
kvm-22.  Some features may only be available in newer releases.
      For best performance, use Linux 2.6.23-rc2 or later as the host.

http://kvm.qumranet.com


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

Visual bug in kvm-60 and 61

[Pelle]

[-- Attachment #1: Type: text/plain, Size: 278 bytes --]

When moving my mouse slowly upwards the lowest black pixels are not erased.
The attached screenshot shows the effect.

I'm using -std-vga and the image has been the same since kvm-9 :). It 
did not have this problem before.
-no-kvm  does not appea to have this.

regards,
Pelle

[-- Attachment #2: Screenshot.png --]
[-- Type: image/png, Size: 65180 bytes --]

[-- Attachment #3: Type: text/plain, Size: 228 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

[-- Attachment #4: Type: text/plain, Size: 158 bytes --]

_______________________________________________
kvm-devel mailing list
kvm-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Re: Visual bug in kvm-60 and 61

[Andreas Winkelbauer]

hi,

> When moving my mouse slowly upwards the lowest black pixels are not erased.
> The attached screenshot shows the effect.

I've also seen this effect on kvm-61 but it is new for me, I didn't see it on
older kvm versions.

I tracked it down a litte bit further:
* I don't see this effect at all when I am not using -std-vga
* I don't see this effect when I am using -no-kvm (no matter if I am using
-std-vga or not)
* in fact I only see this when using -std-vga at a resolution of 1440x900. no
effect at 1440x1050 or at standard resolutions like 1024x786.

maybe someone can check my statements on his/her setup.

bye,
Andi


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

Re: Visual bug in kvm-60 and 61

[Pelle]

I start seeing the effect at 1400x1050 and higher. (including 1440x900)
This happens to be the native resolution for my monitor and the size of 
my desktop.
The effect is notably worse when selecting 1600x1200
It happens at both full-screen and windowed.

btw: kvm crashes when in full screen with a resolution higher than the 
current desktop with a "Could not open SDL display" message.


Andreas Winkelbauer wrote:
> hi,
>
>   
>> When moving my mouse slowly upwards the lowest black pixels are not erased.
>> The attached screenshot shows the effect.
>>     
>
> I've also seen this effect on kvm-61 but it is new for me, I didn't see it on
> older kvm versions.
>
> I tracked it down a litte bit further:
> * I don't see this effect at all when I am not using -std-vga
> * I don't see this effect when I am using -no-kvm (no matter if I am using
> -std-vga or not)
> * in fact I only see this when using -std-vga at a resolution of 1440x900. no
> effect at 1440x1050 or at standard resolutions like 1024x786.
>
> maybe someone can check my statements on his/her setup.
>
> bye,
> Andi
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> kvm-devel mailing list
> kvm-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/kvm-devel
>
>   

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

Re: [ANNOUNCE] kvm-61 release

[Carlo Marcelo Arenas Belon]

On Tue, Feb 19, 2008 at 05:23:58PM +0200, Avi Kivity wrote:
> - e1000 pxe boot rom (Alexey Eremenko)

what specific NIC/ROM type and version was added for kvm-61?, if the ROM came
from ROM-o-matic version 5.4.3 (while all others are from 5.4.2) could that
be spelled as well in qemu/pc-bios/README?

Carlo

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/