On Sat, Aug 04, 2012 at 08:30:08AM +1000, Benjamin Herrenschmidt wrote:
> On Fri, 2012-08-03 at 14:41 -0300, Marcelo Tosatti wrote:
>
> > > Hrm, except that doing KVM_RUN with a signal is very cumbersome to do
> > > and I couldn't quite find the logic in qemu to do it ... but I might
> > > just ha
On Sun, 2012-08-05 at 11:55 +0300, Avi Kivity wrote:
>
> I'm afraid I no longer know the details so closely, the code has
> changed
> quite a lot. But the self-signal happens in kvm_cpu_exec(), see also
> env->exit_request.
Right, I think I eventually grasped it :-) It is fairly fragile however,
On Sun, Aug 5, 2012 at 5:58 AM, Gleb Natapov wrote:
> APIC code has a lot of checks for apic presence and apic HW/SW enable
> state. Most common configuration is when each vcpu has in kernel apic
> and it is fully enabled. This path series uses jump labels to turn checks
> to nops in the common c
Am 05.08.2012 17:52, schrieb Stefan Priebe:
Am 05.08.2012 12:29, schrieb Avi Kivity:
On 08/05/2012 01:08 PM, Stefan Priebe wrote:
Am 01.08.2012 11:53, schrieb Avi Kivity:
On 08/01/2012 12:42 PM, Stefan Priebe - Profihost AG wrote:
Am 01.08.2012 11:33, schrieb Avi Kivity:
So here are 3 backtr
Am 05.08.2012 12:29, schrieb Avi Kivity:
On 08/05/2012 01:08 PM, Stefan Priebe wrote:
Am 01.08.2012 11:53, schrieb Avi Kivity:
On 08/01/2012 12:42 PM, Stefan Priebe - Profihost AG wrote:
Am 01.08.2012 11:33, schrieb Avi Kivity:
So here are 3 backtraces from booting the rescue system:
http://p
On 08/05/2012 06:18 PM, Gleb Natapov wrote:
> On Sun, Aug 05, 2012 at 06:03:12PM +0300, Avi Kivity wrote:
>> On 07/30/2012 05:38 PM, Gleb Natapov wrote:
>> > Optimize "rep ins" by allowing emulator to write back more than one
>> > datum at a time. Introduce new operand type OP_MEM_STR which tells
>
On Sun, Aug 05, 2012 at 06:03:12PM +0300, Avi Kivity wrote:
> On 07/30/2012 05:38 PM, Gleb Natapov wrote:
> > Optimize "rep ins" by allowing emulator to write back more than one
> > datum at a time. Introduce new operand type OP_MEM_STR which tells
> > writeback() that dst contains pointer to an ar
On 07/30/2012 05:38 PM, Gleb Natapov wrote:
> Optimize "rep ins" by allowing emulator to write back more than one
> datum at a time. Introduce new operand type OP_MEM_STR which tells
> writeback() that dst contains pointer to an array that should be written
> back as opposite to just one data eleme
On Sun, Aug 05, 2012 at 05:48:42PM +0300, Avi Kivity wrote:
> On 08/05/2012 05:42 PM, Gleb Natapov wrote:
> > On Sun, Aug 05, 2012 at 05:35:21PM +0300, Avi Kivity wrote:
> >> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> >> > Usually all APICs are HW enabled so the check can be optimized out.
> >>
On Thu, Aug 02, 2012 at 04:26:29PM -0300, Marcelo Tosatti wrote:
> On Mon, Jul 30, 2012 at 05:38:18PM +0300, Gleb Natapov wrote:
> > int r;
> > @@ -5554,9 +5568,13 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu,
> > struct kvm_run *kvm_run)
> > }
> > }
> >
> > - r = c
On 08/05/2012 05:42 PM, Gleb Natapov wrote:
> On Sun, Aug 05, 2012 at 05:35:21PM +0300, Avi Kivity wrote:
>> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
>> > Usually all APICs are HW enabled so the check can be optimized out.
>> >
>> > Signed-off-by: Gleb Natapov
>> > ---
>> > arch/x86/kvm/lapic
On Sun, Aug 05, 2012 at 05:39:33PM +0300, Avi Kivity wrote:
> On 08/05/2012 05:17 PM, Gleb Natapov wrote:
> > On Sun, Aug 05, 2012 at 05:14:59PM +0300, Avi Kivity wrote:
> >> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> >> > According to SDM apic is enabled on start up.
> >> >
> >> > Signed-off-b
On Sun, Aug 05, 2012 at 05:35:21PM +0300, Avi Kivity wrote:
> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> > Usually all APICs are HW enabled so the check can be optimized out.
> >
> > Signed-off-by: Gleb Natapov
> > ---
> > arch/x86/kvm/lapic.c | 29 -
> > arch/x8
On 08/05/2012 05:17 PM, Gleb Natapov wrote:
> On Sun, Aug 05, 2012 at 05:14:59PM +0300, Avi Kivity wrote:
>> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
>> > According to SDM apic is enabled on start up.
>> >
>> > Signed-off-by: Gleb Natapov
>> > ---
>> > arch/x86/kvm/lapic.c |3 ++-
>> > 1
On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> Usually all APICs are HW enabled so the check can be optimized out.
>
> Signed-off-by: Gleb Natapov
> ---
> arch/x86/kvm/lapic.c | 29 -
> arch/x86/kvm/lapic.h |1 +
> arch/x86/kvm/x86.c |1 +
> 3 files changed
On Sun, Aug 05, 2012 at 05:14:59PM +0300, Avi Kivity wrote:
> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> > According to SDM apic is enabled on start up.
> >
> > Signed-off-by: Gleb Natapov
> > ---
> > arch/x86/kvm/lapic.c |3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> >
On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> CC: Jason Baron
> CC: Ingo Molnar
> CC: Peter Zijlstra
> Signed-off-by: Gleb Natapov
> ---
> kernel/jump_label.c |1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/kernel/jump_label.c b/kernel/jump_label.c
> index 4304919..60f48fa 100644
On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> According to SDM apic is enabled on start up.
>
> Signed-off-by: Gleb Natapov
> ---
> arch/x86/kvm/lapic.c |3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> index 49f4ac0..c3f
On Sun, Aug 05, 2012 at 05:00:37PM +0300, Avi Kivity wrote:
> On 08/05/2012 04:48 PM, Gleb Natapov wrote:
> >>
> >> >> During guest boot up, some of these jump keys will change, no? Does
> >> >> this mean a stop_machine() or equivalent? I'm worried about real-time
> >> >> response or one guest b
On 08/05/2012 04:48 PM, Gleb Natapov wrote:
>>
>> >> During guest boot up, some of these jump keys will change, no? Does
>> >> this mean a stop_machine() or equivalent? I'm worried about real-time
>> >> response or one guest being affected by another.
>> >>
>> > Yes, SW enable bit changes durin
On Sun, Aug 05, 2012 at 04:42:17PM +0300, Avi Kivity wrote:
> On 08/05/2012 04:35 PM, Gleb Natapov wrote:
> > On Sun, Aug 05, 2012 at 04:33:02PM +0300, Avi Kivity wrote:
> >> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> >> > APIC code has a lot of checks for apic presence and apic HW/SW enable
> >
On 08/05/2012 04:35 PM, Gleb Natapov wrote:
> On Sun, Aug 05, 2012 at 04:33:02PM +0300, Avi Kivity wrote:
>> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
>> > APIC code has a lot of checks for apic presence and apic HW/SW enable
>> > state. Most common configuration is when each vcpu has in kernel
On Sun, Aug 05, 2012 at 04:33:02PM +0300, Avi Kivity wrote:
> On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> > APIC code has a lot of checks for apic presence and apic HW/SW enable
> > state. Most common configuration is when each vcpu has in kernel apic
> > and it is fully enabled. This path serie
On 08/05/2012 03:58 PM, Gleb Natapov wrote:
> APIC code has a lot of checks for apic presence and apic HW/SW enable
> state. Most common configuration is when each vcpu has in kernel apic
> and it is fully enabled. This path series uses jump labels to turn checks
> to nops in the common case.
>
CC: Jason Baron
CC: Ingo Molnar
CC: Peter Zijlstra
Signed-off-by: Gleb Natapov
---
kernel/jump_label.c |1 +
1 file changed, 1 insertion(+)
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
index 4304919..60f48fa 100644
--- a/kernel/jump_label.c
+++ b/kernel/jump_label.c
@@ -118,6 +1
Usually all vcpus have local apic pointer initialized, so the check may
be completely skipped.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/lapic.c | 62 +++---
arch/x86/kvm/x86.c |7 +-
arch/x86/kvm/x86.h |1 +
3 files changed, 41 in
Those functions are used during interrupt injection. When inlined they
become nops on the fast path.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/lapic.c | 143 +++---
arch/x86/kvm/lapic.h | 45 +++-
2 files changed, 96 insertions(+), 92
Usually all APICs are SW enabled so the check can be optimized out.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/lapic.c | 39 +++
1 file changed, 31 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 1aa5528..95277
Usually all APICs are HW enabled so the check can be optimized out.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/lapic.c | 29 -
arch/x86/kvm/lapic.h |1 +
arch/x86/kvm/x86.c |1 +
3 files changed, 30 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm
According to SDM apic is enabled on start up.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/lapic.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 49f4ac0..c3f14fe 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapi
Do not change apic_base directly. Use kvm_lapic_set_base() instead.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/lapic.c |9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 0cd431c..49f4ac0 100644
--- a/arch/x86/kvm/la
kvm_get_apic_base() needlessly checks irqchip_in_kernel although it does
the same no matter what result of the check is. kvm_set_apic_base() also
checks for irqchip_in_kernel, but kvm_lapic_set_base() can handle this
case.
Signed-off-by: Gleb Natapov
---
arch/x86/kvm/x86.c | 10 ++
1 f
APIC code has a lot of checks for apic presence and apic HW/SW enable
state. Most common configuration is when each vcpu has in kernel apic
and it is fully enabled. This path series uses jump labels to turn checks
to nops in the common case.
Gleb Natapov (8):
KVM: clean up kvm_(set|get)_apic_b
On 08/05/2012 01:08 PM, Stefan Priebe wrote:
> Am 01.08.2012 11:53, schrieb Avi Kivity:
>> On 08/01/2012 12:42 PM, Stefan Priebe - Profihost AG wrote:
>>> Am 01.08.2012 11:33, schrieb Avi Kivity:
> So here are 3 backtraces from booting the rescue system:
> http://pastebin.com/raw.php?i=xCy2
Am 01.08.2012 11:53, schrieb Avi Kivity:
On 08/01/2012 12:42 PM, Stefan Priebe - Profihost AG wrote:
Am 01.08.2012 11:33, schrieb Avi Kivity:
So here are 3 backtraces from booting the rescue system:
http://pastebin.com/raw.php?i=xCy2pEcP
To me they all look the same.
They are. What version
From: Julia Lawall
Convert a 0 error return code to a negative one, as returned elsewhere in the
function.
A new label is also added to avoid freeing things that are known to not yet
be allocated.
A simplified version of the semantic match that finds the first problem is as
follows: (http://coc
On Sun, Aug 05, 2012 at 12:47:42PM +0300, Avi Kivity wrote:
> On 08/03/2012 09:08 PM, Marcelo Tosatti wrote:
> > On Wed, Aug 01, 2012 at 05:01:42PM +0300, Gleb Natapov wrote:
> >> When MSR_KVM_PV_EOI_EN was added to msrs_to_save array
> >> KVM_SAVE_MSRS_BEGIN was not updated accordingly.
> >>
> >>
On 08/03/2012 09:08 PM, Marcelo Tosatti wrote:
> On Wed, Aug 01, 2012 at 05:01:42PM +0300, Gleb Natapov wrote:
>> When MSR_KVM_PV_EOI_EN was added to msrs_to_save array
>> KVM_SAVE_MSRS_BEGIN was not updated accordingly.
>>
>> Signed-off-by: Gleb Natapov
>
> Applied, thanks.
This should go into
On 08/05/2012 11:08 AM, Pekka Enberg wrote:
>> Isn't there a capability flag that KVM sets if KVM_SIGNAL_MSI is
>> supported? Just store that in 'struct kvm" and switch between
>> virtio_pci__signal_msi() and kvm__irq_trigger() depending on wheter the
>> flag is set.
On Sun, Aug 5, 2012 at 12:14 P
On 08/03/2012 01:57 PM, Stefan Bader wrote:
>> No, you're backporting the entire feature. All we need is to expose
>> RDPMC intercept to the guest.
>
> Oh well, I thought that was the thing you asked for...
Sorry for being unclear.
>
>> It should be sufficient to backport the bits in
>> nested
On 08/05/2012 11:08 AM, Pekka Enberg wrote:
> On 08/04/2012 01:02 PM, Pekka Enberg wrote:
>>> No, it's not running 3.5. We need to support older *host* kernels,
>>> though.
>
> On Sun, Aug 5, 2012 at 10:02 AM, Sasha Levin wrote:
>> Do we? Don't we need to support just the kernel that the tool was
On 08/04/2012 01:02 PM, Pekka Enberg wrote:
>> No, it's not running 3.5. We need to support older *host* kernels,
>> though.
On Sun, Aug 5, 2012 at 10:02 AM, Sasha Levin wrote:
> Do we? Don't we need to support just the kernel that the tool was
> built with?
We only do that for *guest kernels* i
On 08/04/2012 01:32 AM, Benjamin Herrenschmidt wrote:
> On Fri, 2012-08-03 at 15:05 -0300, Marcelo Tosatti wrote:
>
>> See kvm_arch_process_async_events() call to qemu_system_reset_request()
>> in target-i386/kvm.c.
>>
>> The whole thing is fragile, though: we rely on the order events
>> are proc
On 08/02/2012 11:29 PM, Benjamin Herrenschmidt wrote:
> On Thu, 2012-08-02 at 16:05 +0300, Avi Kivity wrote:
>> > Yeah, we stumbled over this chunk as well. So you're saying we
>> should delay the reset by invoking a self-signal if we're in such an
>> operation?
>>
>> Yes. Qemu of course already
On 08/04/2012 01:02 PM, Pekka Enberg wrote:
> On 08/04/2012 11:14 AM, Pekka Enberg wrote:
>>> This patch makes 'make check' hang for me. Full boot log below:
>
> On Sat, Aug 4, 2012 at 12:30 PM, Sasha Levin wrote:
>> Is your host kernel running 3.5? The new MSI injection ioctl is a new 3.5
>> fe
45 matches
Mail list logo