Re: [kvm-devel] file-backed guest memory

2007-04-14 Thread Avi Kivity
Christian Borntraeger wrote: > On Thursday 12 April 2007 17:38, Avi Kivity wrote: > >>> But wouldn't it also be possible to create a file in >>> userspace and pass its descriptor to kvm? >>> > [...] > >> swap, hugetlbfs, and maybe other nifty stuff. I think I know how to do >> this f

Re: [kvm-devel] irqdevice INTR example

2007-04-14 Thread Avi Kivity
Gregory Haskins wrote: >> It's really subtle. >> >> With respect to interrupts, VT^H^Hthe hardware provides an override over >> IF. If an interrupt happens while this override is enabled, we exit >> guest mode regardless of the state of IF. In effect interrupts are >> enabled but _delivery_ is

Re: [kvm-devel] emulate 64 bit mmx Instructionen

2007-04-14 Thread Avi Kivity
Michael Westermann wrote: > Hello, > > when I boot a debian-sarge with kvm and qemu, I get a exeption with the > instruction 06 6f . This Instruction is am "movq". > My System is a Dual Opteron 2214. > > The debian-installer install's a K7 kernel with a mmx optimising. > > I think it is a good

Re: [kvm-devel] Is KVM source code available through http or https?

2007-04-14 Thread Avi Kivity
Li, Xin B wrote: >> Li, Xin B wrote: >> If the git: and svn: protocols are blocked for you, I can >> try to make them available through http. >> >>> That would be helpful, thanks! >>> -Xin >>> >>> >> http://www.kernel.org/pub/scm/linux/kernel/git/avi/kvm.git,

Re: [kvm-devel] Is KVM source code available through http or https?

2007-04-14 Thread Thomas Glanzmann
Hello Avi, > git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm-userspace.git is this now the main repository? I love to see that, if it is the case. Thomas - This SF.net email is sponsored by DB2 Express Downl

Re: [kvm-devel] Swapping out kvm pages

2007-04-14 Thread Chen Haogang
I encountered another problem. Because guest uses its own PTEs, host cannot track page usage of the guest, including dirty flag in guest PTEs. So host will consider those pages clean and reclaim them without writing back. What I currently think to walk around this is set the PG_dirty flag manually