[kvm-devel] [ kvm-Bugs-1848393 ] crash in vmx; most likely linked to suspend/resume

2007-12-10 Thread SourceForge.net
Bugs item #1848393, was opened at 2007-12-11 11:21 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=893831&aid=1848393&group_id=180599 Please note that this message will contain a full copy

[kvm-devel] 供应：剪板机、折弯机、卷板机、冲床、校平机

2007-12-10 Thread 安徽省三力机床制造有限公司
供应：剪板机、折弯机、卷板机、冲床、校平机 安徽省三力机床制造有限公司，属安徽省名牌、免检产品生产企业，安徽省高新技术企业，马鞍山市“专、精、特、新”企业，马鞍山市下岗失业职工再就业基地，介于南京、芜湖之间，坐落在著名的钢城马鞍山市，丰富的钢铁资源为公司的发展提供了得天独厚的优势。公司占地面积108000平方米，主要生产“大盛”牌系列剪板机、折弯机、卷板机、快速压力机、冲床、开卷校平机、型材弯曲机刀模具等产品，专业为航空、轻工、冶金、化工、建筑汽车、电力、装潢等行业提供所需要的专用机械和成套设备，产品销往全国各地和欧盟及东南亚地区并远销南非。 　　　

Re: [kvm-devel] [PATCH] Fix SMP shadow instantiation race

2007-12-10 Thread Marcelo Tosatti
On Mon, Dec 10, 2007 at 11:27:45PM +0200, Avi Kivity wrote: > Marcelo Tosatti wrote: > > On Mon, Dec 10, 2007 at 07:07:54PM +0200, Avi Kivity wrote: > > > >> Marcelo Tosatti wrote: > >> > >>> There is a race where VCPU0 is shadowing a pagetable entry while VCPU1 > >>> is updating it, which

Re: [kvm-devel] [PATCH] emulated cmpxchg8b should be atomic on i386

2007-12-10 Thread Avi Kivity
Marcelo Tosatti wrote: > Emulate cmpxchg8b atomically on i386. This is required to avoid a guest > pte walker from seeing a splitted write. > > Signed-off-by: Marcelo Tosatti <[EMAIL PROTECTED]> > > diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c > index 9db4e32..d58d795 100644 > --- a/drivers/k

Re: [kvm-devel] [PATCH] Fix SMP shadow instantiation race

2007-12-10 Thread Avi Kivity
Marcelo Tosatti wrote: > On Mon, Dec 10, 2007 at 07:07:54PM +0200, Avi Kivity wrote: > >> Marcelo Tosatti wrote: >> >>> There is a race where VCPU0 is shadowing a pagetable entry while VCPU1 >>> is updating it, which results in a stale shadow copy. >>> >>> Fix that by comparing the contents

[kvm-devel] [PATCH] emulated cmpxchg8b should be atomic on i386

2007-12-10 Thread Marcelo Tosatti
Emulate cmpxchg8b atomically on i386. This is required to avoid a guest pte walker from seeing a splitted write. Signed-off-by: Marcelo Tosatti <[EMAIL PROTECTED]> diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c index 9db4e32..d58d795 100644 --- a/drivers/kvm/x86.c +++ b/drivers/kvm/x86.c @@

Re: [kvm-devel] [PATCH] Fix SMP shadow instantiation race

2007-12-10 Thread Marcelo Tosatti
On Mon, Dec 10, 2007 at 07:07:54PM +0200, Avi Kivity wrote: > Marcelo Tosatti wrote: > >There is a race where VCPU0 is shadowing a pagetable entry while VCPU1 > >is updating it, which results in a stale shadow copy. > > > >Fix that by comparing the contents of the cached guest pte with the > >curre

Re: [kvm-devel] [PATCH] fix -kernel option

2007-12-10 Thread Dor Laor
Glauber de Oliveira Costa wrote: Dor Laor wrote: > Glauber de Oliveira Costa wrote: >> >> Currently, the -kernel option is not working. >> >> Reason is, because we're registering chunks for regions 0-0xa and >> 0x10-ram_size, the phys_ram_addr + PA is broken. >> The real fix should be to

Re: [kvm-devel] [PATCH] Fix SMP shadow instantiation race

2007-12-10 Thread Avi Kivity
Marcelo Tosatti wrote: > There is a race where VCPU0 is shadowing a pagetable entry while VCPU1 > is updating it, which results in a stale shadow copy. > > Fix that by comparing the contents of the cached guest pte with the > current guest pte after write-protecting the guest pagetable. > > Attache

[kvm-devel] [PATCH] Fix SMP shadow instantiation race

2007-12-10 Thread Marcelo Tosatti
There is a race where VCPU0 is shadowing a pagetable entry while VCPU1 is updating it, which results in a stale shadow copy. Fix that by comparing the contents of the cached guest pte with the current guest pte after write-protecting the guest pagetable. Attached program kvm_shadow_race.c demons

Re: [kvm-devel] [PATCH] fix -kernel option

2007-12-10 Thread Glauber de Oliveira Costa
Dor Laor wrote: > Glauber de Oliveira Costa wrote: >> >> Currently, the -kernel option is not working. >> >> Reason is, because we're registering chunks for regions 0-0xa and >> 0x10-ram_size, the phys_ram_addr + PA is broken. >> The real fix should be to rewrite all the load_linux() code t

Re: [kvm-devel] kvm-55 and ne2k-pci

2007-12-10 Thread Avi Kivity
Jerry Geis wrote: > The extra -net user did the trick for redhat 9. THanks, I'll try redhat > 7 now. > > The network "seemed" slow. ethtool eth0 reports 100m full. when copying all > my files over it was slow. THese were source files (text) not big files > at all. > -net user is quite slow.

Re: [kvm-devel] kvm-55 and ne2k-pci

2007-12-10 Thread Jerry Geis
Izik Eidus wrote: > On Mon, 2007-12-10 at 08:44 -0500, Jerry Geis wrote: > >> I installed redhat7.3 and redhat 9 an a qemu-img with 12G each. >> >> When I am running the ne2k-pci the network stalls. >> I tried both redhat 7.3 and redhat 9. both stall. >> >> I then tried to change to use "-net ni

Re: [kvm-devel] kvm-55 and ne2k-pci

2007-12-10 Thread Izik Eidus
On Mon, 2007-12-10 at 08:44 -0500, Jerry Geis wrote: > I installed redhat7.3 and redhat 9 an a qemu-img with 12G each. > > When I am running the ne2k-pci the network stalls. > I tried both redhat 7.3 and redhat 9. both stall. > > I then tried to change to use "-net nic,model=rtl8139" > and I cou

[kvm-devel] kvm-55 and ne2k-pci

2007-12-10 Thread Jerry Geis
I installed redhat7.3 and redhat 9 an a qemu-img with 12G each. When I am running the ne2k-pci the network stalls. I tried both redhat 7.3 and redhat 9. both stall. I then tried to change to use "-net nic,model=rtl8139" and I could not get the network to even start up on redhat 9. I re-ran kudzu

[kvm-devel] frastert

2007-12-10 Thread tarcea tarikci
Don't what to buy as a present for your little "guy"? Well, we have a perfect present. http://www.gorgeousdomination.com/ - SF.Net email is sponsored by: Check out the new SourceForge.net Marketplace. It's the best place to b

Re: [kvm-devel] kvm-55 test

2007-12-10 Thread Shahar Frank
> > -Original Message- > From: [EMAIL PROTECTED] [mailto:kvm-devel- > [EMAIL PROTECTED] On Behalf Of Fabian Deutsch > Sent: Sunday, December 09, 2007 3:24 PM > To: Farkas Levente > Cc: kvm-devel; Avi Kivity > Subject: Re: [kvm-devel] kvm-55 test > > Am Sonntag, den 09.12.2007, 12:26 +0100