Glauber Costa wrote:
> As KVM cpus runs on threads, it is possible that
> we call kvm_load_registers() from a cpu thread, while the
> apic has not yet fully initialized. kvm_load_registers() is called
> from ap_main_loop.
>
> This is not a problem when we're starting the whole machine together,
>
As KVM cpus runs on threads, it is possible that
we call kvm_load_registers() from a cpu thread, while the
apic has not yet fully initialized. kvm_load_registers() is called
from ap_main_loop.
This is not a problem when we're starting the whole machine together,
but is a problem for hotplug, since
I don't think it's a cr3 OOS optimisation break, as kvm-72 faild too.
2009/4/17 David S. Ahern :
> Have you seen this thread:
>
> http://article.gmane.org/gmane.comp.emulators.kvm.devel/28649
>
> It's the one that resulted in these patches by Marcelo:
> http://article.gmane.org/gmane.comp.emulator
Have you seen this thread:
http://article.gmane.org/gmane.comp.emulators.kvm.devel/28649
It's the one that resulted in these patches by Marcelo:
http://article.gmane.org/gmane.comp.emulators.kvm.devel/30452
http://article.gmane.org/gmane.comp.emulators.kvm.devel/30453
david
Dongsheng Song wrot
No good news, same as kvm-84 vanilla.
2009/4/17 David S. Ahern :
> I believe this is the patch that should fix the problem:
>
> Index: kvm/arch/x86/kvm/mmu.c
> ===
> --- kvm.orig/arch/x86/kvm/mmu.c
> +++ kvm/arch/x86/kvm/mmu.c
> @@ -1
I believe this is the patch that should fix the problem:
Index: kvm/arch/x86/kvm/mmu.c
===
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -1254,7 +1254,7 @@ static struct kvm_mmu_page *kvm_mmu_get_
pgprintk("%s:
2009/4/16 Brian Jackson :
> it's a module option to the kvm option.
>
> parm: oos_shadow:bool
>
When I set 'oos_shadow=Y' for kvm-84 module, OpenBSD 4.5 stopped
after display 'clock: unknown CMOS layout'.
When I set 'oos_shadow=N' for kvm-84 module, OpenBSD 4.5 go farther,
stopped after
Jim Keniston wrote:
It looks like AT2(Ev,Gv) would yield the same bits as AT2(Gv,Ev). It'd
be nice not to lose the operand-order information. And we'd have to
make clear whether which notation we're using -- src,dest as in the gnu
assembler, or dest,src as in the AMD (and Intel?) manuals.
S
On Thu, 2009-04-16 at 19:31 -0400, Masami Hiramatsu wrote:
...
>
> Hmm, I have an idea about instruction table. Usually, instruction tables
> are encoded with code defined by each decoder/emulator. This method
> will show their internal code directly, and is hard to maintain when
> the opcode map
Masami Hiramatsu wrote:
Hmm, I have an idea about instruction table. Usually, instruction tables
are encoded with code defined by each decoder/emulator. This method
will show their internal code directly, and is hard to maintain when
the opcode map is updated. Instead of that, I'd like to sugges
H. Peter Anvin wrote:
> Jim Keniston wrote:
>> For user-space probing, we've been concentrating on native-built
>> executables. Am I correct in thinking that we'll see 16-bit or V86 mode
>> only on legacy apps built elsewhere? In any case, it only makes sense
>> to build on the kvm folks' work in
Avi Kivity wrote:
We don't emulate guest user mode.
Well, if guest userspace can convince its kernel to give it access to
some memory mapped I/O register, I guess it can execute repeated 'mov
ss, mmio' and starve the guest kernel.
It doesn't need a MMIO register to do that, even.
Any idea when the split will happen?
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Peter Teoh writes:
> I got a mid-way hang while installing OpenSolaris on KVM guest yesterday.
> [...]
>
> How can i debug via kvmtrace (or systemtap) to find out at which point it
> hanged? [...]
In recent kernels, kvm tracing is done by markers, which systemtap can
attach to. Their use of
Peter,
On Sat, Apr 11, 2009 at 12:45:21AM -0400, Peter Teoh wrote:
> In this function, the TLB flushing comes before spin unlock,
>
> void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
> {
>struct kvm_mmu_page *sp;
>
>spin_lock(&kvm->mmu_lock);
>
>kvm_flush
On Thu, Apr 16, 2009 at 09:11:03AM -0400, Peter Teoh wrote:
> I got a mid-way hang while installing OpenSolaris on KVM guest yesterday.
> The hardware is kvm-intel enabled. I tried the same thing on
> Athlon64, but it does not have svm hardware enabled. So qemu startup saying
> /dev/kvm cannot
Nick Piggin wrote:
On Wednesday 15 April 2009 08:09:03 Andrew Morton wrote:
On Thu, 9 Apr 2009 06:58:37 +0300
Izik Eidus wrote:
KSM is a linux driver that allows dynamicly sharing identical memory
pages between one or more processes.
Generally looks OK to me. But that doesn'
On Wednesday 15 April 2009 08:09:03 Andrew Morton wrote:
> On Thu, 9 Apr 2009 06:58:37 +0300
> Izik Eidus wrote:
>
> > KSM is a linux driver that allows dynamicly sharing identical memory
> > pages between one or more processes.
>
> Generally looks OK to me. But that doesn't mean much. We sho
Andrea Arcangeli wrote:
On Wed, Apr 15, 2009 at 05:43:03PM -0700, Jeremy Fitzhardinge wrote:
Shouldn't that be kmap_atomic's job anyway? Otherwise it would be hard to
No because those are full noops in no-highmem kernels. I commented in
other email why I think it's safe thanks to the
On Thu, Apr 16, 2009 at 4:59 AM, Theo de Raadt wrote:
> If they claim to provide a proper i386/amd64 vm environment, and we
> don't run properly ... Look, considering we run very well on 99% of
> PCs, it means they their claims are balony.
I concur,
I had problems with VMware (don't cut my head
BRAUN, Stefanie wrote:
Hello,
I've compiled a new kernel v2.6.27-rc5 with the modified svm.c.
But the behaviour of the vlc process in the guest is still the same.
I've exported additional cpu features to the guest, e.g. mmxext with kvm-84.
But no performance changes.
I was not able to export
Hi List,
I'm a newbie at kvm/linux so hopefully I'm not asking anything too obvious.
Can anyone please recommend the best bits to use for getting VT-d passthru to
work? I've tried kvm-84 with Fedora 10, but the kernel is 2.6.27.5. I'm
trying the Fedora 11 Beta but I've hit other issues. My s
Bugs item #2768533, was opened at 2009-04-16 16:01
Message generated for change (Comment added) made by technologov
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=893831&aid=2768533&group_id=180599
Please note that this message will contain a full copy of the comme
Bugs item #2768533, was opened at 2009-04-16 16:01
Message generated for change (Settings changed) made by technologov
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=893831&aid=2768533&group_id=180599
Please note that this message will contain a full copy of the co
it's a module option to the kvm option.
parm: oos_shadow:bool
On Thursday 16 April 2009 03:43:56 Dongsheng Song wrote:
> kvm-84 has same error.
>
> How to disable oos optimization, is there a switch?
>
> 2009/4/16 Brian Jackson :
> > You might want to try 84 with oos optimization off (
Hello,
I've compiled a new kernel v2.6.27-rc5 with the modified svm.c.
But the behaviour of the vlc process in the guest is still the same.
I've exported additional cpu features to the guest, e.g. mmxext with kvm-84.
But no performance changes.
I was not able to export the cpu flags 3dnow and 3d
I got a mid-way hang while installing OpenSolaris on KVM guest yesterday.
The hardware is kvm-intel enabled. I tried the same thing on
Athlon64, but it does not have svm hardware enabled. So qemu startup saying
/dev/kvm cannot be found, even though i had explicitly insmod kvm, and
dmesg show sv
Zhang, Xiantao wrote:
From dcb73b5db2e91213a2536542945645caaf23bfe3 Mon Sep 17 00:00:00 2001
From: Xiantao Zhang
Date: Thu, 16 Apr 2009 17:59:16 +0800
Subject: [PATCH] KVM: ia64: Flush all TLBs once guest's memory mapping changes.
Flush all vcpu's TLB entries once changes guest's memory mapping
Bugs item #2768533, was opened at 2009-04-16 16:01
Message generated for change (Tracker Item Submitted) made by technologov
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=893831&aid=2768533&group_id=180599
Please note that this message will contain a full copy of
Beth Kon wrote:
These patches resolve the irq0->inti2 override issue, and get the hpet working
on kvm with and without -no-kvm-irqchip (i.e., when hpet takes over, it
disables userspace or in-kernel pit as appropriate).
The irq0->inti2 override will always be used unless the kernel cannot do i
Avi Kivity wrote:
Marcelo Tosatti wrote:
Matt T. Yourst notes that kvm_arch_vcpu_ioctl_set_sregs lacks validity
checking for the new cr3 value:
"Userspace callers of KVM_SET_SREGS can pass a bogus value of
cr3 to
the kernel. This will trigger a NULL pointer access in gfn_to_rmap()
when u
Marcelo Tosatti wrote:
Matt T. Yourst notes that kvm_arch_vcpu_ioctl_set_sregs lacks validity
checking for the new cr3 value:
"Userspace callers of KVM_SET_SREGS can pass a bogus value of cr3 to
the kernel. This will trigger a NULL pointer access in gfn_to_rmap()
when userspace next tries
On Wed, Apr 15, 2009 at 05:43:03PM -0700, Jeremy Fitzhardinge wrote:
> Shouldn't that be kmap_atomic's job anyway? Otherwise it would be hard to
No because those are full noops in no-highmem kernels. I commented in
other email why I think it's safe thanks to the wrprotect + smp tlb
flush of the
Matt T. Yourst notes that kvm_arch_vcpu_ioctl_set_sregs lacks validity
checking for the new cr3 value:
"Userspace callers of KVM_SET_SREGS can pass a bogus value of cr3 to
the kernel. This will trigger a NULL pointer access in gfn_to_rmap()
when userspace next tries to call KVM_RUN on the a
>From dcb73b5db2e91213a2536542945645caaf23bfe3 Mon Sep 17 00:00:00 2001
From: Xiantao Zhang
Date: Thu, 16 Apr 2009 17:59:16 +0800
Subject: [PATCH] KVM: ia64: Flush all TLBs once guest's memory mapping changes.
Flush all vcpu's TLB entries once changes guest's memory mapping.
Signed-off-by: Xiant
Marcelo Tosatti wrote:
Humpf. And something like this? Or GP# instead of triple fault?
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 148cde2..3e63bac 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3986,7 +3986,10 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vc
H. Peter Anvin wrote:
Avi Kivity wrote:
Why do we care? The guest can only harm itself, and if it wants to
disable interrupts, it would be a lot easier for it to run a plain 'cli'.
I guess it would be a problem if we emulated 'mov ss' for ordinary
userspace or vm86 mode, but we don't.
On Thu, Apr 16, 2009 at 11:56:15AM +0300, Avi Kivity wrote:
> Marcelo Tosatti wrote:
>> Matt T. Yourst notes that kvm_arch_vcpu_ioctl_set_sregs lacks validity
>> checking for the new cr3 value:
>>
>> "Userspace callers of KVM_SET_SREGS can pass a bogus value of cr3 to
>> the kernel. This will trigg
Marcelo Tosatti wrote:
Matt T. Yourst notes that kvm_arch_vcpu_ioctl_set_sregs lacks validity
checking for the new cr3 value:
"Userspace callers of KVM_SET_SREGS can pass a bogus value of cr3 to
the kernel. This will trigger a NULL pointer access in gfn_to_rmap()
when userspace next tries to cal
kvm-84 has same error.
How to disable oos optimization, is there a switch?
2009/4/16 Brian Jackson :
> You might want to try 84 with oos optimization off (or better yet 85 when it
> comes out). There was a bug recently (post 84) fixed that affected some BSDs.
>
>
>
> On Wednesday 15 April 2009 22
How about performance comparision with kvm guests?
Xiantao
-Original Message-
From: kvm-ow...@vger.kernel.org [mailto:kvm-ow...@vger.kernel.org] On Behalf Of
Xiaodong Yi
Sent: Thursday, April 16, 2009 9:54 AM
To: kvm@vger.kernel.org
Subject: Luvalley-2 has been released: running KVM be
41 matches
Mail list logo