Avi Kivity wrote:
On 12/22/2011 05:41 PM, Liu, Jinsong wrote:
Avi Kivity wrote:
On 12/21/2011 12:25 PM, Jan Kiszka wrote:
We must not report the TSC deadline timer feature on our own when
user space provides the APIC as we have no clue about its features.
We must not report the TSC
Avi Kivity wrote:
From: Jan Kiszka jan.kis...@siemens.com
Unlike all of the other cpuid bits, the TSC deadline timer bit is set
unconditionally, regardless of what userspace wants.
This is broken in several ways:
- if userspace doesn't use KVM_CREATE_IRQCHIP, and doesn't emulate
the
Sasha Levin wrote:
On Sun, 2011-12-25 at 21:00 +0200, Sasha Levin wrote:
On Sun, 2011-12-25 at 15:03 +0200, Avi Kivity wrote:
+ if (apic) {
+ if (best-ecx bit(X86_FEATURE_TSC_DEADLINE_TIMER))
+ apic-lapic_timer.timer_mode_mask = 3 17;
+ else
+
Right now we wait for any of the children to stop, which means we might
exit before /bin/sh stopped.
Instead, wait only for the /bin/sh (or the sandboxed) process.
Signed-off-by: Sasha Levin levinsasha...@gmail.com
---
tools/kvm/guest/init_stage2.c |2 +-
1 files changed, 1 insertions(+), 1
On 12/26/2011 10:42 AM, Liu, Jinsong wrote:
}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index df23dff..1171def 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2089,6 +2089,9 @@ int kvm_dev_ioctl_check_extension(long ext)
case KVM_CAP_TSC_CONTROL:
On Mon, Dec 26, 2011 at 12:35, Avi Kivity a...@redhat.com wrote:
On 12/26/2011 10:42 AM, Liu, Jinsong wrote:
}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index df23dff..1171def 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2089,6 +2089,9 @@ int
On 12/26/2011 12:43 PM, Alexey Zaytsev wrote:
On Mon, Dec 26, 2011 at 12:35, Avi Kivity a...@redhat.com wrote:
On 12/26/2011 10:42 AM, Liu, Jinsong wrote:
}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index df23dff..1171def 100644
--- a/arch/x86/kvm/x86.c
+++
On Sat, Dec 17, 2011 at 11:19:35AM +0800, Liu Ping Fan wrote:
From: Liu Ping Fan pingf...@linux.vnet.ibm.com
Currently, vcpu can be destructed only when kvm instance destroyed.
Change this to vcpu's destruction before kvm instance, so vcpu MUST
and CAN be destroyed before kvm's destroy.
On 12/25/2011 08:39 PM, Jan Kiszka wrote:
On 2011-12-25 14:03, Avi Kivity wrote:
From: Jan Kiszka jan.kis...@siemens.com
Unlike all of the other cpuid bits, the TSC deadline timer bit is set
unconditionally, regardless of what userspace wants.
This is broken in several ways:
- if
On 12/26/2011 01:09 PM, Gleb Natapov wrote:
+
+ idx = srcu_read_lock(kvm-srcu_vcpus);
+ kvm_for_each_vcpu(vcpu, kvm) {
+ if (!pass !firststart
+ vcpu != kvm-last_boosted_vcpu
+ kvm-last_boosted_vcpu !=
On Mon, Dec 26, 2011 at 01:17:39PM +0200, Avi Kivity wrote:
On 12/26/2011 01:09 PM, Gleb Natapov wrote:
+
+ idx = srcu_read_lock(kvm-srcu_vcpus);
+ kvm_for_each_vcpu(vcpu, kvm) {
+ if (!pass !firststart
+ vcpu !=
On Mon, Dec 26, 2011 at 12:45, Avi Kivity a...@redhat.com wrote:
On 12/26/2011 12:43 PM, Alexey Zaytsev wrote:
On Mon, Dec 26, 2011 at 12:35, Avi Kivity a...@redhat.com wrote:
On 12/26/2011 10:42 AM, Liu, Jinsong wrote:
}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index
On 12/26/2011 01:25 PM, Alexey Zaytsev wrote:
If you can, please test it:
git://git.kernel.org/pub/scm/virt/kvm/kvm.git kvm-updates/3.2
Works for me, thank you.
Thanks, added your Tested-by:.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from
On 12/26/2011 10:11 AM, Liu, Jinsong wrote:
It breaks live migration: if you start a guest on a TSC-deadline
capable host kernel, and migrate it to a TSC-deadline incapable host
kernel, you end up with a broken guest.
More broadly, kvm never exposes features transparently to the
On 12/15/2011 02:04 PM, Jan Kiszka wrote:
Not all compilers accept anonymous field initialization, e.g. gcc
4.5.1.
Thanks, applied.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to
On 12/20/2011 06:37 AM, Takuya Yoshikawa wrote:
ping
(2011/11/29 14:03), Takuya Yoshikawa wrote:
Signed-off-by: Takuya Yoshikawayoshikawa.tak...@oss.ntt.co.jp
---
arch/x86/kvm/mmu.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kvm/mmu.c
On 11/29/2011 07:04 AM, Takuya Yoshikawa wrote:
There are many places where we drop large spte and we are always doing
much the same as drop_large_spte().
To avoid these duplications, this patch makes drop_large_spte() more
generically usable: it now takes an argument to know if it must flush
On 11/29/2011 07:01 AM, Takuya Yoshikawa wrote:
Made when I was reading mmu code.
Takuya
BTW, is threre any good way to test large page functionality?
Just booting a guest with transparent hugepages enabled is a good test.
The guest kernel will use hugepages itself even if
On 12/20/2011 11:21 AM, Paul Mackerras wrote:
This moves gfn_to_memslot(), and the functions it calls, that is,
search_memslots() and __gfn_to_memslot(), from kvm_main.c to kvm_host.h
so that gfn_to_memslot() can be called from non-modular code even
when KVM is a module. On powerpc, the
On 12/19/2011 07:46 PM, Boris Ostrovsky wrote:
From: Boris Ostrovsky boris.ostrov...@amd.com
In some cases guests should not provide workarounds for errata even when the
physical processor is affected. For example, because of erratum 400 on family
10h processors a Linux guest will read an MSR
Linus, please pull from:
git://git.kernel.org/pub/scm/virt/kvm/kvm.git kvm-updates/3.2
to receive an unfortunately large bunch of fixes for Linux 3.2 KVM, some
fairly critical:
- two DoS issues in the device assignment code
- an oops in the emulated PIT code
- the regression caused by
On 12/19/2011 04:11 PM, Jan Kiszka wrote:
Backwards compatibility
If we want backwards compatibility, we need more than just a simple feature
check, no? Oh, you feed that into CPUID? That's nifty. Ok, so you behave
like VMX/SVM do on real hardware - you always expose the
On 25.12.2011, at 12:01, Jörg Sommer wrote:
Alexander Graf hat am Sat 24. Dec, 18:05 (+0100) geschrieben:
On 24.12.2011, at 12:13, Jörg Sommer wrote:
Jörg Sommer hat am Sat 24. Dec, 11:31 (+0100) geschrieben:
Alexander Graf hat am Fri 23. Dec, 14:50 (+0100) geschrieben:
On 20.12.2011, at
Avi Kivity wrote:
On 12/26/2011 10:11 AM, Liu, Jinsong wrote:
It breaks live migration: if you start a guest on a TSC-deadline
capable host kernel, and migrate it to a TSC-deadline incapable host
kernel, you end up with a broken guest.
More broadly, kvm never exposes features
On 12/26/2011 07:37 PM, Avi Kivity wrote:
On 12/19/2011 04:11 PM, Jan Kiszka wrote:
Backwards compatibility
If we want backwards compatibility, we need more than just a simple feature
check, no? Oh, you feed that into CPUID? That's nifty. Ok, so you behave like
VMX/SVM do on real hardware
From 19caf1db1f93e6f6b736e1dfd5e91a0c7669adec Mon Sep 17 00:00:00 2001
From: Liu Jinsong jinsong@intel.com
Date: Tue, 27 Dec 2011 04:08:27 +0800
Subject: [PATCH] Expose tsc deadline timer cpuid to guest
Depend on several factors:
1. Considering live migration, user enable/disable tsc deadline
Commit 64c8f7b9a24b1de710ad8413da503d61ede6ae45 to Linux 2.6.32.y
broke kvm-kmod. There are preprocessor directives in kvm-kmod that
appear to be intended to remedy this, but there is a logical error in
them. I have attached a patch that corrects that.
I am not on the list, so please CC any
I forgot to include the patch. Sorry about that.
On Mon, Dec 26, 2011 at 11:08 PM, Richard Yao r...@cs.stonybrook.edu wrote:
Commit 64c8f7b9a24b1de710ad8413da503d61ede6ae45 to Linux 2.6.32.y
broke kvm-kmod. There are preprocessor directives in kvm-kmod that
appear to be intended to remedy
On 12/26/11 08:53, Avi Kivity wrote:
On 12/19/2011 07:46 PM, Boris Ostrovsky wrote:
From: Boris Ostrovskyboris.ostrov...@amd.com
In some cases guests should not provide workarounds for errata even when the
physical processor is affected. For example, because of erratum 400 on family
10h
On Mon, Dec 26, 2011 at 7:09 PM, Gleb Natapov g...@redhat.com wrote:
On Sat, Dec 17, 2011 at 11:19:35AM +0800, Liu Ping Fan wrote:
From: Liu Ping Fan pingf...@linux.vnet.ibm.com
Currently, vcpu can be destructed only when kvm instance destroyed.
Change this to vcpu's destruction before kvm
On 12/20/2011 11:21 AM, Paul Mackerras wrote:
This moves gfn_to_memslot(), and the functions it calls, that is,
search_memslots() and __gfn_to_memslot(), from kvm_main.c to kvm_host.h
so that gfn_to_memslot() can be called from non-modular code even
when KVM is a module. On powerpc, the
On 25.12.2011, at 12:01, Jörg Sommer wrote:
Alexander Graf hat am Sat 24. Dec, 18:05 (+0100) geschrieben:
On 24.12.2011, at 12:13, Jörg Sommer wrote:
Jörg Sommer hat am Sat 24. Dec, 11:31 (+0100) geschrieben:
Alexander Graf hat am Fri 23. Dec, 14:50 (+0100) geschrieben:
On 20.12.2011, at
32 matches
Mail list logo