Takuya Yoshikawa takuya.yoshik...@gmail.com wrote:
@@ -1265,22 +1263,19 @@ int emulate_int_real(struct x86_emulate_ctxt *ctxt,
/* TODO: Add limit checks */
c-src.val = ctxt-eflags;
- emulate_push(ctxt, ops);
- rc = writeback(ctxt, ops);
+ rc = emulate_push(ctxt,
Hi,
I'm trying to write a virtio-blk driver for Solaris. I've gotten it to the point
where Solaris can see the device and create a ZFS file system on it.
However when I try and create a UFS filesystem on the device, the VM crashed
with the error
*** glibc detected *** /usr/bin/qemu-kvm: double
On 03/30/2011 10:41 AM, Alexander Graf wrote:
On 30.03.2011, at 10:15, Conor Murphy wrote:
Hi,
I'm trying to write a virtio-blk driver for Solaris. I've gotten it to the
point
where Solaris can see the device and create a ZFS file system on it.
However when I try and create a UFS
On Wed, Mar 30, 2011 at 9:15 AM, Conor Murphy
conor_murphy_v...@hotmail.com wrote:
I'm trying to write a virtio-blk driver for Solaris. I've gotten it to the
point
where Solaris can see the device and create a ZFS file system on it.
However when I try and create a UFS filesystem on the
On 03/29/2011 02:08 PM, Gleb Natapov wrote:
Currently we sync registers back and forth before/after exiting
to userspace for IO, but during IO device model shouldn't need to
read/write the registers, so we can as well skip those sync points. The
only exaception is broken vmware backdor
On 03/29/2011 02:56 AM, Sasha Levin wrote:
It is possible to create CPU-less NUMA nodes, node amount shouldn't be
limited by amount of CPUs.
Patch seems fine; but please send to qemu-de...@nongnu.org.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this
On Wed, Mar 30, 2011 at 12:38:53PM +0200, Avi Kivity wrote:
On 03/29/2011 02:08 PM, Gleb Natapov wrote:
Currently we sync registers back and forth before/after exiting
to userspace for IO, but during IO device model shouldn't need to
read/write the registers, so we can as well skip those sync
On 03/30/2011 12:47 PM, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 12:38:53PM +0200, Avi Kivity wrote:
On 03/29/2011 02:08 PM, Gleb Natapov wrote:
Currently we sync registers back and forth before/after exiting
to userspace for IO, but during IO device model shouldn't need to
read/write
On Wed, Mar 30, 2011 at 12:50:53PM +0200, Avi Kivity wrote:
On 03/30/2011 12:47 PM, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 12:38:53PM +0200, Avi Kivity wrote:
On 03/29/2011 02:08 PM, Gleb Natapov wrote:
Currently we sync registers back and forth before/after exiting
to userspace
On Wed, Mar 30, 2011 at 01:22:43PM +0200, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 12:50:53PM +0200, Avi Kivity wrote:
On 03/30/2011 12:47 PM, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 12:38:53PM +0200, Avi Kivity wrote:
On 03/29/2011 02:08 PM, Gleb Natapov wrote:
Currently we
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
The patch saves copying of 256 bytes on each MMIO/PIO read. It may not
save a lot comparing to time it takes to do one MMIO to userspace, but
do 1 million of those and you saved a lot of CPU cycles. I do not think
we should abandon the
On 03/30/2011 02:12 PM, Avi Kivity wrote:
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
The patch saves copying of 256 bytes on each MMIO/PIO read. It may
not
save a lot comparing to time it takes to do one MMIO to userspace,
but
do 1 million of those and you saved a lot of CPU cycles. I
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
After reboot perf started to work. I ran modified emulator.flat unit
test. It was modified to run test_cmps() in an endless loop.
Without patch:
1.71% qemu-system-x86 [kvm] [k] x86_emulate_instruction
1.51% qemu-system-x86 [kvm]
On Wed, Mar 30, 2011 at 02:17:55PM +0200, Avi Kivity wrote:
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
After reboot perf started to work. I ran modified emulator.flat unit
test. It was modified to run test_cmps() in an endless loop.
Without patch:
1.71% qemu-system-x86 [kvm]
When KVM scans the 0xD CPUID leaf for propagating the XSAVE save area
leaves, it assumes that the leaves are contigious and stops at the
first zero one. On AMD hardware there is a gap, though, as LWP uses
leaf 62 to announce it's state save area.
So lets iterate through all 64 possible leaves and
If KVM cannot find an exact match for a requested CPUID leaf, the
code will try to find the closest match instead of simply confessing
it's failure. The heuristic is on one hand wrong nowadays,
since it does not take the KVM CPUID leaves (0x40xx) into
account. On the other hand the callers of
On 03/30/2011 03:01 PM, Andre Przywara wrote:
If KVM cannot find an exact match for a requested CPUID leaf, the
code will try to find the closest match instead of simply confessing
it's failure. The heuristic is on one hand wrong nowadays,
since it does not take the KVM CPUID leaves (0x40xx)
On Wed, Mar 30, 2011 at 02:48:28PM +0200, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 02:17:55PM +0200, Avi Kivity wrote:
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
After reboot perf started to work. I ran modified emulator.flat unit
test. It was modified to run test_cmps() in an endless
On 03/30/2011 03:26 PM, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 02:48:28PM +0200, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 02:17:55PM +0200, Avi Kivity wrote:
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
After reboot perf started to work. I ran modified emulator.flat unit
On 03/30/2011 03:26 PM, Avi Kivity wrote:
This behaviour is mandated by the spec (looking at the Intel one),
though it is implemented incorrectly - should always return largest
basic leaf, and ignore the kvm leaves.
I think the correct behaviour is:
if (e-function 1 (!best ||
On Wed, Mar 30, 2011 at 03:29:02PM +0200, Avi Kivity wrote:
On 03/30/2011 03:26 PM, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 02:48:28PM +0200, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 02:17:55PM +0200, Avi Kivity wrote:
On 03/30/2011 01:43 PM, Gleb Natapov wrote:
After reboot
On 03/30/2011 03:36 PM, Gleb Natapov wrote:
It's wierd. Do you get perf hits in the copying?
How can I check. The memcpy is inlined.
perf annotate x86_emulate_instruction
(newer perf allows you to get there interactively from 'perf report')
Copying a couple of hot cache lines
On Wed, Mar 30, 2011 at 03:41:09PM +0200, Avi Kivity wrote:
On 03/30/2011 03:36 PM, Gleb Natapov wrote:
It's wierd. Do you get perf hits in the copying?
How can I check. The memcpy is inlined.
perf annotate x86_emulate_instruction
(newer perf allows you to get there interactively
On 03/30/2011 03:43 PM, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 03:41:09PM +0200, Avi Kivity wrote:
On 03/30/2011 03:36 PM, Gleb Natapov wrote:
It's wierd. Do you get perf hits in the copying?
How can I check. The memcpy is inlined.
perf annotate x86_emulate_instruction
I apologize if this is the wrong list. I have just installed Fedora
14 and gotten KVM up and running. I installed a Windows 7 guest
without issue using NAT for networking. The guest can ping the
default gateway, but can't reach the internet or the rest of the
network. Here's the really odd
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows up,
grepping for base_address gets me lots of output, telling me that
this is a typo, but could be wrong.
Signed-off-by: Justin P. Mattock justinmatt...@gmail.com
---
On 03/30/2011 06:19 PM, Justin P. Mattock wrote:
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows up,
grepping for base_address gets me lots of output, telling me that
this is a typo, but could be wrong.
Signed-off-by: Justin P.
Based on Gleb's idea, fix race between nmi injection and enabling
nmi window in a simpler way.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index a6a129f..9a7cc1be 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -5152,6
On 03/30/2011 09:26 AM, Avi Kivity wrote:
On 03/30/2011 06:19 PM, Justin P. Mattock wrote:
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows up,
grepping for base_address gets me lots of output, telling me that
this is a typo, but
On Wed, Mar 30, 2011 at 01:30:28PM -0300, Marcelo Tosatti wrote:
Based on Gleb's idea, fix race between nmi injection and enabling
nmi window in a simpler way.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
But we need to revert the patch that introduced use of request for NMI
first.
Hi Marc,
On 03/30/2011 08:46 AM, Marc Boorshtein wrote:
I apologize if this is the wrong list. I have just installed Fedora
14 and gotten KVM up and running. I installed a Windows 7 guest
without issue using NAT for networking. The guest can ping the
default gateway, but can't reach the
The patch below removes unsigned long base_addresss; in i8254.h
since it is unused.
Signed-off-by: Justin P. Mattock justinmatt...@gmail.com
---
arch/x86/kvm/i8254.h |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/i8254.h b/arch/x86/kvm/i8254.h
index
On Wed, Mar 30, 2011 at 06:33:22PM +0200, Gleb Natapov wrote:
On Wed, Mar 30, 2011 at 01:30:28PM -0300, Marcelo Tosatti wrote:
Based on Gleb's idea, fix race between nmi injection and enabling
nmi window in a simpler way.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
But we
On 03/30/2011 06:30 PM, Marcelo Tosatti wrote:
Based on Gleb's idea, fix race between nmi injection and enabling
nmi window in a simpler way.
Signed-off-by: Marcelo Tosattimtosa...@redhat.com
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index a6a129f..9a7cc1be 100644
---
OK, I''ll go ask them.
Thanks!
Marc
On Wed, Mar 30, 2011 at 12:44 PM, David Mair dm...@mair-family.org wrote:
Hi Marc,
On 03/30/2011 08:46 AM, Marc Boorshtein wrote:
I apologize if this is the wrong list. I have just installed Fedora
14 and gotten KVM up and running. I installed a
On 03/30/2011 06:30 PM, Justin P. Mattock wrote:
On 03/30/2011 09:26 AM, Avi Kivity wrote:
On 03/30/2011 06:19 PM, Justin P. Mattock wrote:
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows up,
grepping for base_address gets me lots
On 03/30/2011 10:17 AM, Avi Kivity wrote:
On 03/30/2011 06:30 PM, Justin P. Mattock wrote:
On 03/30/2011 09:26 AM, Avi Kivity wrote:
On 03/30/2011 06:19 PM, Justin P. Mattock wrote:
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows
On Wed, Mar 30, 2011 at 07:16:34PM +0200, Avi Kivity wrote:
On 03/30/2011 06:30 PM, Marcelo Tosatti wrote:
Based on Gleb's idea, fix race between nmi injection and enabling
nmi window in a simpler way.
Signed-off-by: Marcelo Tosattimtosa...@redhat.com
diff --git a/arch/x86/kvm/x86.c
On Wed, 30 Mar 2011, Avi Kivity wrote:
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows up,
grepping for base_address gets me lots of output, telling me that
this is a typo, but could be wrong.
Signed-off-by:
On 03/30/2011 03:21 PM, Jiri Kosina wrote:
On Wed, 30 Mar 2011, Avi Kivity wrote:
The below patch changes base_addresss to base_address.
Note: I have grepped for base_addresss and nothing shows up,
grepping for base_address gets me lots of output, telling me that
this is a typo, but could be
On Wed, 30 Mar 2011 11:00:26 +0530
Balbir Singh bal...@linux.vnet.ibm.com wrote:
Data from the previous patchsets can be found at
https://lkml.org/lkml/2010/11/30/79
It would be nice if the data for the current patchset was present in
the current patchset's changelog!
--
To unsubscribe from
On Wed, 30 Mar 2011 11:02:38 +0530
Balbir Singh bal...@linux.vnet.ibm.com wrote:
Changelog v4
1. Added documentation for max_unmapped_pages
2. Better #ifdef'ing of max_unmapped_pages and min_unmapped_pages
Changelog v2
1. Use a config option to enable the code (Andrew Morton)
2. Explain
* Andrew Morton a...@linux-foundation.org [2011-03-30 16:36:07]:
On Wed, 30 Mar 2011 11:00:26 +0530
Balbir Singh bal...@linux.vnet.ibm.com wrote:
Data from the previous patchsets can be found at
https://lkml.org/lkml/2010/11/30/79
It would be nice if the data for the current patchset
On Thu, 31 Mar 2011 10:57:03 +0530 Balbir Singh bal...@linux.vnet.ibm.com
wrote:
* Andrew Morton a...@linux-foundation.org [2011-03-30 16:36:07]:
On Wed, 30 Mar 2011 11:00:26 +0530
Balbir Singh bal...@linux.vnet.ibm.com wrote:
Data from the previous patchsets can be found at
The following series implements page cache control,
this is a split out version of patch 1 of version 3 of the
page cache optimization patches posted earlier at
Previous posting http://lwn.net/Articles/425851/ and analysis
at http://lwn.net/Articles/419713/
Detailed Description
* Andrew Morton a...@linux-foundation.org [2011-03-30 16:35:45]:
On Wed, 30 Mar 2011 11:02:38 +0530
Balbir Singh bal...@linux.vnet.ibm.com wrote:
Changelog v4
1. Added documentation for max_unmapped_pages
2. Better #ifdef'ing of max_unmapped_pages and min_unmapped_pages
Changelog
On 29.03.2011, at 23:49, Scott Wood wrote:
Return the actual host SVR for now, as we already do for PVR. Eventually
we may support Qemu overriding PVR/SVR if the situation is appropriate,
once we implement KVM_SET_SREGS on e500.
Haha - maybe I should have read your v2 before replying to the
On 03/30/2011 09:28 AM, Alexander Graf wrote:
On 29.03.2011, at 23:49, Scott Wood wrote:
Return the actual host SVR for now, as we already do for PVR. Eventually
we may support Qemu overriding PVR/SVR if the situation is appropriate,
once we implement KVM_SET_SREGS on e500.
Haha - maybe
On 30.03.2011, at 09:56, Avi Kivity wrote:
On 03/30/2011 09:28 AM, Alexander Graf wrote:
On 29.03.2011, at 23:49, Scott Wood wrote:
Return the actual host SVR for now, as we already do for PVR. Eventually
we may support Qemu overriding PVR/SVR if the situation is appropriate,
once
On Wed, 30 Mar 2011 10:17:55 +0200
Alexander Graf ag...@suse.de wrote:
On 30.03.2011, at 01:43, Scott Wood wrote:
+ case BOOKE_INTERRUPT_SPE_UNAVAIL: {
+ extern void kvmppc_vcpu_spe_load(struct kvm_vcpu *vcpu);
+
+ /* reload the SPE env if guest first use SPE
Keep the guest MSR and the guest-mode true MSR separate, rather than
modifying the guest MSR on each guest entry to produce a true MSR.
Any bits which should be modified based on guest MSR must be explicitly
propagated from vcpu-arch.shared-msr to vcpu-arch.shadow_msr in
kvmppc_set_msr().
While
This allows reuse for saving/restoring KVM SPE state.
Signed-off-by: Scott Wood scottw...@freescale.com
---
v4 of patchset, first version of this patch
Kumar, please ack (or comment).
arch/powerpc/include/asm/ppc_asm.h | 28
arch/powerpc/kernel/head_fsl_booke.S
giveup_spe() saves the SPE state which is protected by MSR[SPE].
However, modifying SPEFSCR does not trap when MSR[SPE]=0.
And since SPEFSCR is already saved/restored in _switch(),
not all the callers want to save SPEFSCR again.
Thus, saving SPEFSCR should not belong to giveup_spe().
This patch
On Thu, 31 Mar 2011 00:43:27 +0200
Alexander Graf ag...@suse.de wrote:
@@ -78,6 +79,43 @@ void kvmppc_dump_vcpu(struct kvm_vcpu *vcpu)
}
}
+#ifdef CONFIG_SPE
+static void kvmppc_vcpu_enable_spe(struct kvm_vcpu *vcpu)
+{
+ enable_kernel_spe();
+
On 31.03.2011, at 00:56, Scott Wood wrote:
On Thu, 31 Mar 2011 00:43:27 +0200
Alexander Graf ag...@suse.de wrote:
@@ -78,6 +79,43 @@ void kvmppc_dump_vcpu(struct kvm_vcpu *vcpu)
}
}
+#ifdef CONFIG_SPE
+static void kvmppc_vcpu_enable_spe(struct kvm_vcpu *vcpu)
+{
+
This is done lazily. The SPE save will be done only if the guest has
used SPE since the last preemption or heavyweight exit. Restore will be
done only on demand, when enabling MSR_SPE in the shadow MSR, in response
to an SPE fault or mtmsr emulation.
For SPEFSCR, Linux already switches it on
-Original Message-
From: kvm-ppc-ow...@vger.kernel.org
[mailto:kvm-ppc-ow...@vger.kernel.org] On Behalf Of Scott Wood
Sent: Thursday, March 31, 2011 7:35 AM
To: ag...@suse.de
Cc: kvm-ppc@vger.kernel.org
Subject: [PATCH v5 4/4] KVM: PPC: e500: Save/restore SPE state
This is
57 matches
Mail list logo