Re: [PATCH] allow enabling/disabling NPT by reloading only the architecture module

2008-07-16 Thread Yang, Sheng
On Tuesday 15 July 2008 18:55:37 Avi Kivity wrote: Yang, Sheng wrote: On Tuesday 15 July 2008 02:36:36 Joerg Roedel wrote: If NPT is enabled after loading both KVM modules on AMD and it should be disabled, both KVM modules must be reloaded. If only the architecture module is reloaded the

[PATCH 1/2] KVM: VMX: Fix bypass_guest_pf enabling when disable EPT in module parameter

2008-07-16 Thread Yang, Sheng
From c4a2cad8b91ac4c0b04a5ccd1f0bfab1d7e6ef37 Mon Sep 17 00:00:00 2001 From: Sheng Yang [EMAIL PROTECTED] Date: Wed, 16 Jul 2008 09:21:22 +0800 Subject: [PATCH] KVM: VMX: Fix bypass_guest_pf enabling when disable EPT in module parameter Signed-off-by: Sheng Yang [EMAIL PROTECTED] ---

[PATCH 2/2] KVM: VMX: Fix undefined beaviour of EPT after reload kvm-intel.ko

2008-07-16 Thread Yang, Sheng
From bcbe1b5c4c6098f122accba4f00f6617baf807f7 Mon Sep 17 00:00:00 2001 From: Sheng Yang [EMAIL PROTECTED] Date: Wed, 16 Jul 2008 09:25:40 +0800 Subject: [PATCH] KVM: VMX: Fix undefined beaviour of EPT after reload kvm-intel.ko As well as move set base/mask ptes to vmx_init(). Signed-off-by:

[PATCH 0/2] configure: add support for audio-{drv,card}-list

2008-07-16 Thread Carlo Marcelo Arenas Belon
The following series adds support for qemu's audio configure option lists that were added in kvm-71 to support selecting which interface will be used to enable audio in the host from the guest (oss, alsa, sdl, esd, fmod, or pulseaudio) and which audio devices emulation to enable for the guest

[PATCH 2/2] configure: passthrough for audio-{drv,card}-list and logic cleanup

2008-07-16 Thread Carlo Marcelo Arenas Belon
Extending the cleanup logic used in a patch from Jindrich Makovicka, changes the default option to pass the full option to qemu's configure and add a passthrough for qemu options that use a space separated list of options like the list for audio drivers enabled or the list for audio devices

RE: [PATCH 01/04]Create x86 directory to hold x86-specific files.

2008-07-16 Thread Zhang, Xiantao
Avi Kivity wrote: Zhang, Xiantao wrote: From 03ac444d1ab4446c587e8180ceaba60b9e75b28d Mon Sep 17 00:00:00 2001 From: Xiantao Zhang [EMAIL PROTECTED] Date: Fri, 11 Jul 2008 10:13:08 +0800 Subject: [PATCH] KVM: external module: Moving x86-speicif files to x86 directory. Create x86

Re: [PATCH] Ignore DEBUGCTL MSRs

2008-07-16 Thread Alexander Graf
Avi Kivity wrote: Alexander Graf wrote: Avi Kivity wrote: Alexander Graf wrote: Netware writes and reads to the DEBUGCTL and LAST*IP MSRs without further checks and is really confused to receive a #GP during that. To make it happy we should just make them stubs, which is exactly what SVM

[PATCH] qemu: re-add definition for qemu_get_launch_info

2008-07-16 Thread Carlo Marcelo Arenas Belon
somehow missing from sysemu.h after a qemu merge and otherwise complaining with the following warning : kvm-71/qemu/migration.c: In function 'migration_init_ssh': kvm-71/qemu/migration.c:629: warning: implicit declaration of function 'qemu_get_launch_info' Signed-off-by: Carlo Marcelo

[PATCH] qemu: remove duplicated inclusion of signal.h in qemu-kvm.h

2008-07-16 Thread Carlo Marcelo Arenas Belon
added by mistake as part of 4820cce75999b2673a964eb87601229a4bd78ad9 Signed-off-by: Carlo Marcelo Arenas Belon [EMAIL PROTECTED] --- qemu/qemu-kvm.h |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/qemu/qemu-kvm.h b/qemu/qemu-kvm.h index 8b7dcde..7e28428 100644 ---

Re: kvm: Unknown error 524, Fail to handle apic access vmexit

2008-07-16 Thread Martin Michlmayr
* Yang, Sheng [EMAIL PROTECTED] [2008-07-16 11:26]: Hi Martin, can you show more dmesg here? It doesn't contain any other messages from kvm. If you still want it, let me know. And if it can be reproduce stable? I can reproduce this 100%. Anyway, I just tried 2.6.26 with FlexPriority

PCI passthrough with VT-d - native performance

2008-07-16 Thread Ben-Ami Yassour
In last few tests that we made with PCI-passthrough and VT-d using iperf, we were able to get the same throughput as on native OS with a 1G NIC (with higher CPU utilization). The following patches are the PCI-passthrough patches that Amit sent (re-based on the last kvm tree), followed by a few

[PATCH 5/8] KVM: PCIPT: change order of device release

2008-07-16 Thread Ben-Ami Yassour
Signed-off-by: Ben-Ami Yassour [EMAIL PROTECTED] --- arch/x86/kvm/x86.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 8d25b4a..65b307d 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -343,9 +343,9 @@ static void

[PATCH 4/8] KVM: PCIPT: fix interrupt handling

2008-07-16 Thread Ben-Ami Yassour
This patch fixes a few problems with the interrupt handling for passthrough devices. 1. Pass the interrupt handler the pointer to the device, so we do not need to lock the pcipt lock in the interrupt handler. 2. Remove the pt_irq_handled bitmap - it is no longer needed. 3. Split

[PATCH 2/8] KVM: Introduce a callback routine for PIC ack handling

2008-07-16 Thread Ben-Ami Yassour
From: Amit Shah [EMAIL PROTECTED] This is useful for acking irqs of assigned devices Signed-off-by: Amit Shah [EMAIL PROTECTED] --- arch/x86/kvm/i8259.c |6 +- arch/x86/kvm/irq.c |2 +- arch/x86/kvm/irq.h |3 ++- 3 files changed, 8 insertions(+), 3 deletions(-) diff --git

[PATCH 7/8] KVM: PCIPT: VT-d support

2008-07-16 Thread Ben-Ami Yassour
This patch includes the functions to support VT-d for passthrough devices. [Ben: fixed memory pinning, cleanup] Signed-off-by: Kay, Allen M [EMAIL PROTECTED] Signed-off-by: Weidong Han [EMAIL PROTECTED] Signed-off-by: Ben-Ami Yassour [EMAIL PROTECTED] --- arch/x86/kvm/Makefile |2 +-

[PATCH 8/8] KVM: PCIPT: VT-d: dont map mmio memory slots

2008-07-16 Thread Ben-Ami Yassour
Avoid mapping mmio memory slots. Signed-off-by: Ben-Ami Yassour [EMAIL PROTECTED] --- arch/x86/kvm/vtd.c | 20 +--- include/asm-x86/kvm_host.h |2 ++ virt/kvm/kvm_main.c|2 +- 3 files changed, 16 insertions(+), 8 deletions(-) diff --git

[PATCH 6/8] VT-d: changes to support KVM

2008-07-16 Thread Ben-Ami Yassour
From: Kay, Allen M [EMAIL PROTECTED] This patch extends the VT-d driver to support KVM [Ben: fixed memory pinning] Signed-off-by: Kay, Allen M [EMAIL PROTECTED] Signed-off-by: Weidong Han [EMAIL PROTECTED] Signed-off-by: Ben-Ami Yassour [EMAIL PROTECTED] --- drivers/pci/dmar.c

[PATCH 1/8] KVM: Introduce a callback routine for IOAPIC ack handling

2008-07-16 Thread Ben-Ami Yassour
From: Amit Shah [EMAIL PROTECTED] This will be useful for acking irqs of assigned devices Signed-off-by: Amit Shah [EMAIL PROTECTED] --- virt/kvm/ioapic.c |3 +++ virt/kvm/ioapic.h |1 + 2 files changed, 4 insertions(+), 0 deletions(-) diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c

[PATCH 2/2] PCIPT: direct mmio

2008-07-16 Thread Ben-Ami Yassour
Enable a guest to access a device's memory mapped I/O regions directly. When the guest changes the I/O memory mappings for a passthrough device, send the updated mmio region to the kernel. Signed-off-by: Ben-Ami Yassour [EMAIL PROTECTED] Signed-off-by: Muli Ben-Yehuda [EMAIL PROTECTED] ---

Re: kvm: Unknown error 524, Fail to handle apic access vmexit

2008-07-16 Thread Dor Laor
Yang, Sheng wrote: On Tuesday 15 July 2008 23:19:07 Dor Laor wrote: Martin Michlmayr wrote: I installed a Windows XP SP2 guest on a Debian x86_64 host The installation itself went fine but kvm aborts when when XP starts during Windows XP Setup. XP mentions something with intelppm.sys

Re: networking setup problem

2008-07-16 Thread Uri Lublin
paolo pedaletti wrote: Hi, I hope this is the right ml to submit my problem. Abstract: I can't setup 2 different network inside my VMs, one public and one private. Scheme: eth0 - -| proxy |---eth1 |- | H| | O| eth0 - |

[ kvm-Bugs-2019053 ] tbench fails on guest when AMD NPT enabled

2008-07-16 Thread SourceForge.net
Bugs item #2019053, was opened at 2008-07-16 03:10 Message generated for change (Comment added) made by avik You can respond by visiting: https://sourceforge.net/tracker/?func=detailatid=893831aid=2019053group_id=180599 Please note that this message will contain a full copy of the comment

[PATCH] posix-timers: Do not modify an already queued timer signal

2008-07-16 Thread Mark McLoughlin
When a timer fires, posix_timer_event() zeroes out its pre-allocated siginfo structure, initialises it and then queues up the signal with send_sigqueue(). However, we may have previously queued up this signal, in which case we only want to increment si_overrun and re-initialising the siginfo

Re: [PATCH 4/8] KVM: PCIPT: fix interrupt handling

2008-07-16 Thread Avi Kivity
Ben-Ami Yassour wrote: This patch fixes a few problems with the interrupt handling for passthrough devices. Well, fold it into the patch it fixes. There is no point in sending a buggy patch and a fix in the same patchset. -- error compiling committee.c: too many arguments to function

Re: [PATCH 3/8] KVM: Handle device assignment to guests

2008-07-16 Thread Avi Kivity
Ben-Ami Yassour wrote: From: Han, Weidong [EMAIL PROTECTED] This patch adds support for handling PCI devices that are assigned to the guest (PCI passthrough). + +/* + * Used to find a registered host PCI device (a passthrough device) + * during ioctls, interrupts or EOI + */ +struct

[ kvm-Bugs-2019053 ] tbench fails on guest when AMD NPT enabled

2008-07-16 Thread SourceForge.net
Bugs item #2019053, was opened at 2008-07-15 18:10 Message generated for change (Comment added) made by alex_williamson You can respond by visiting: https://sourceforge.net/tracker/?func=detailatid=893831aid=2019053group_id=180599 Please note that this message will contain a full copy of the

Re: PCI passthrough with VT-d - native performance

2008-07-16 Thread Ben-Ami Yassour
On Wed, 2008-07-16 at 17:36 +0300, Avi Kivity wrote: Ben-Ami Yassour wrote: In last few tests that we made with PCI-passthrough and VT-d using iperf, we were able to get the same throughput as on native OS with a 1G NIC Excellent! (with higher CPU utilization). How much

Re: PCI passthrough with VT-d - native performance

2008-07-16 Thread Avi Kivity
Ben-Ami Yassour wrote: (with higher CPU utilization). How much higher? Here are some numbers for running iperf -l 1M: e1000 NIC (behind a PCI bridge) Bandwidth (Mbit/sec)CPU utilization Native OS 771 18% Native

Re: PCI passthrough with VT-d - native performance

2008-07-16 Thread Anthony Liguori
Ben-Ami Yassour wrote: On Wed, 2008-07-16 at 17:36 +0300, Avi Kivity wrote: Ben-Ami Yassour wrote: In last few tests that we made with PCI-passthrough and VT-d using iperf, we were able to get the same throughput as on native OS with a 1G NIC Excellent! (with higher CPU

[PATCH v2/RFC] libkvm-s390

2008-07-16 Thread Christian Borntraeger
This is an update patch for libkvm to build and work on s390. It should address all comments from Avi as well as some aspects I have found: o implement kvm_show_regs o use s390 instead of s390x in file names. It is commonly used for 31 and 64bit systems o dont define __s390__ and __s390x__ in

Re: [PATCH] posix-timers: Do not modify an already queued timer signal

2008-07-16 Thread Mark McLoughlin
On Wed, 2008-07-16 at 15:50 +0100, Mark McLoughlin wrote: The race was observed with a modified kvm-userspace when running a guest under heavy network load. When it occurs, KVM never sees another SIGALRM signal because although the signal is queued up the appropriate bit is never set in the

[PATCH 3/6] KVM: Handle device assignment to guests

2008-07-16 Thread Ben-Ami Yassour
From: Amit Shah [EMAIL PROTECTED] This patch adds support for handling PCI devices that are assigned to the guest (PCI passthrough). The device to be assigned to the guest is registered in the host kernel and interrupt delivery is handled. If a device is already assigned, or the device driver

[PATCH 3/6] KVM: Handle device assignment to guests

2008-07-16 Thread Ben-Ami Yassour
From: Amit Shah [EMAIL PROTECTED] This patch adds support for handling PCI devices that are assigned to the guest (PCI passthrough). The device to be assigned to the guest is registered in the host kernel and interrupt delivery is handled. If a device is already assigned, or the device driver

[PATCH 1/6] KVM: Introduce a callback routine for IOAPIC ack handling

2008-07-16 Thread Ben-Ami Yassour
From: Amit Shah [EMAIL PROTECTED] This will be useful for acking irqs of assigned devices Signed-off-by: Amit Shah [EMAIL PROTECTED] --- virt/kvm/ioapic.c |3 +++ virt/kvm/ioapic.h |1 + 2 files changed, 4 insertions(+), 0 deletions(-) diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c

Re: [PATCH 3/6] KVM: Handle device assignment to guests

2008-07-16 Thread Ben-Ami Yassour
Please ignore this repeated patch -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH] posix-timers: Do not modify an already queued timer signal

2008-07-16 Thread Oleg Nesterov
On 07/16, Mark McLoughlin wrote: When a timer fires, posix_timer_event() zeroes out its pre-allocated siginfo structure, initialises it and then queues up the signal with send_sigqueue(). However, we may have previously queued up this signal, in which case we only want to increment

Re: PCI passthrough with VT-d - native performance

2008-07-16 Thread Avi Kivity
Ben-Ami Yassour wrote: That CPU utilization is extremely high and somewhat illogical if native w/vt-d has almost no CPU impact. Have you run oprofile yet or have any insight into where CPU is being burnt? What does kvm_stat look like? I wonder if there are a large number of PIO exits.

kvm_queue_exception

2008-07-16 Thread Thomas Mueller
hi there i was using kvm-70 and kernel 2.6.25 on debian lenny for two weeks without problems. processor is an AMD Opteron 2350. now out of the nowwhere there is a problem with 1 of 4 guests (1x debian etch amd64, 1x debian etch i386, 1x ubuntu 7.10 amd64, 1x ubuntu 7.10 i386) On the host i

Re: kvm causing memory corruption? now 2.6.26

2008-07-16 Thread Dave Hansen
On a suggestion of Anthony's, I tried a defconfig kernel. It is now bombing out on an assertion in the lapic code: http://sr71.net/~dave/linux/2.6.26-oops1.txt -- Dave -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to [EMAIL PROTECTED] More

[patch 3/3] KVM: VMX: handle segment limit granularity special case in software

2008-07-16 Thread Marcelo Tosatti
As the comment in the diff mentions, VMX does not accept any bit in the range 11:0 of ES,CS,FS,GS,SS segment registers limit field to be zero with the granulity bit set to one. So clear granularity and adjust the limit accordingly. Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED] Index:

KVM overflows the stack

2008-07-16 Thread Dave Hansen
On Wed, 2008-07-16 at 14:44 -0700, Dave Hansen wrote: On a suggestion of Anthony's, I tried a defconfig kernel. It is now bombing out on an assertion in the lapic code: http://sr71.net/~dave/linux/2.6.26-oops1.txt I think I found it!!! $ (objdump -d kvm.ko ; objdump -d kvm-intel.ko

Re: networking setup problem

2008-07-16 Thread David Mair
paolo pedaletti wrote: Hi, I hope this is the right ml to submit my problem. Abstract: I can't setup 2 different network inside my VMs, one public and one private. Scheme: eth0 - -| proxy |---eth1 |- | H| | O| eth0 - |

Re: [ANNOUNCE] kvm-autotest

2008-07-16 Thread Ryan Harper
* Uri Lublin [EMAIL PROTECTED] [2008-07-16 18:15]: Client side, for installation, we already have a solution that works for all types of guests: http://kvm.qumranet.com/kvmwiki/KVMTest which is already integrated as a client test in autotest. Once you record your installation via

Re: [PATCH 3/6] KVM: Handle device assignment to guests

2008-07-16 Thread Yang, Sheng
Some comments below. :) On Wednesday 16 July 2008 23:56:50 Ben-Ami Yassour wrote: From: Amit Shah [EMAIL PROTECTED] This patch adds support for handling PCI devices that are assigned to the guest (PCI passthrough). The device to be assigned to the guest is registered in the host kernel and

RE: [PATCH 3/8] KVM: Handle device assignment to guests

2008-07-16 Thread Han, Weidong
Avi Kivity wrote: +static void kvm_pci_pt_work_fn(struct work_struct *work) +{ +struct kvm_pci_pt_dev_list *match; +struct kvm_pci_pt_work *int_work; +int source; +unsigned long flags; +int guest_irq; +int host_irq; + +int_work = container_of(work, struct

Re: [PATCH 3/8] KVM: Handle device assignment to guests

2008-07-16 Thread Yang, Sheng
On Thursday 17 July 2008 10:09:57 Han, Weidong wrote: Avi Kivity wrote: +static void kvm_pci_pt_work_fn(struct work_struct *work) +{ + struct kvm_pci_pt_dev_list *match; + struct kvm_pci_pt_work *int_work; + int source; + unsigned long flags; + int guest_irq; + int host_irq;

RE: PCI passthrough with VT-d - native performance

2008-07-16 Thread Han, Weidong
Anthony Liguori wrote: Ben-Ami Yassour wrote: On Wed, 2008-07-16 at 17:36 +0300, Avi Kivity wrote: Ben-Ami Yassour wrote: In last few tests that we made with PCI-passthrough and VT-d using iperf, we were able to get the same throughput as on native OS with a 1G NIC Excellent!

Re: kvm causing memory corruption? now 2.6.26

2008-07-16 Thread Avi Kivity
Dave Hansen wrote: On a suggestion of Anthony's, I tried a defconfig kernel. It is now bombing out on an assertion in the lapic code: http://sr71.net/~dave/linux/2.6.26-oops1.txt Well that assert is plain wrong: static int apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic

Re: KVM overflows the stack

2008-07-16 Thread Avi Kivity
Dave Hansen wrote: On Wed, 2008-07-16 at 14:44 -0700, Dave Hansen wrote: On a suggestion of Anthony's, I tried a defconfig kernel. It is now bombing out on an assertion in the lapic code: http://sr71.net/~dave/linux/2.6.26-oops1.txt I think I found it!!! $ (objdump -d

Re: missing git tags

2008-07-16 Thread Avi Kivity
Hollis Blanchard wrote: Hi Avi, I noticed that the kvm.git tree seems to be missing every Linus tag after v2.6.24-rc8. (Of course, it does have all the kvm-NN tags.) Can this be fixed? Sure, done. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to