Re: [PATCH 0/5] Another set of Nested SVM fixes and cleanups

On Wed, Sep 16, 2009 at 03:24:14PM +0200, Joerg Roedel wrote: Hi, this series of patches contain another set of cleanups and an important fix to the Nested SVM code. These patches make the TSC handling code for SVM aware of a nested guest. This fixes the TSC running backwards on guest and

[PATCH -v3] QEMU-KVM: MCE: Relay UCR MCE to guest

UCR (uncorrected recovery) MCE is supported in recent Intel CPUs, where some hardware error such as some memory error can be reported without PCC (processor context corrupted). To recover from such MCE, the corresponding memory will be unmapped, and all processes accessing the memory will be

Re: Possible problem with -no-kvm-irqchip -no-kvm-pit qemu options detected during git daily testing

On Thu, Sep 17, 2009 at 09:57:05PM -0300, Lucas Meneghel Rodrigues wrote: Hi folks, after looking at today's git testing results, it seems like we have a problem with the command line options: -no-kvm-irqchip -no-kvm-pit This patch series should fix it. Not yet committed:

[KVM-AUTOTEST PATCH 1/2] Adding Cerberus Test Control System 2 (ctcs2) as client test

- Using a newer version compared to the existed cerberus test. - User can specifiy cerberus testcases by providing command line options in control file. - Added a patch to fix the makefile to make cerberus run on x86_64 system. NOTE: this test may need a longer autotest timeout than the default

Re: [Autotest] [KVM-AUTOTEST PATCH 0/7] KVM test: support for the new remote shell server for Windows

On Thu, Sep 17, 2009 at 11:40:46AM -0400, Michael Goldish wrote: - Michael Goldish mgold...@redhat.com wrote: - Yolkfull Chow yz...@redhat.com wrote: On Tue, Aug 18, 2009 at 06:30:14PM -0400, Michael Goldish wrote: - Lucas Meneghel Rodrigues l...@redhat.com wrote:

[PATCH 00/24] configure and Makefile cleanup

This series: - werror is not disabled unconditoanally, you can disable it with --disable-werror. - clean configure and Makefile* of mis-merges - sync such files to qemu upstream - kvm options are detected and configured _after_ detecting kvm, not the other way around - add command line options

[PATCH 01/24] Don't disable werror unconditionally

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/configure b/configure index 97292d9..e6252a2 100755 --- a/configure +++ b/configure @@ -1758,8 +1758,6 @@ if test -z $werror ; then else

[PATCH 02/24] extboot.bin is a generated file

Signed-off-by: Juan Quintela quint...@redhat.com --- .gitignore |1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/.gitignore b/.gitignore index 13c967e..59a0244 100644 --- a/.gitignore +++ b/.gitignore @@ -44,4 +44,5 @@ pc-bios/bios-pq/status pc-bios/vgabios-pq/status

[PATCH 03/24] Remove merge artifacts

CFLAGS and LDFLAGS are defined on config.mak LDFLAGS_BASE is not used anywhere DEPLIBS are not defined anywhere Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target |8 1 files changed, 0 insertions(+), 8 deletions(-) diff --git a/Makefile.target b/Makefile.target

[PATCH 05/24] compatfd is included before, and it is compiled unconditionally

Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile |4 1 files changed, 0 insertions(+), 4 deletions(-) diff --git a/Makefile b/Makefile index 6a9ca7c..88b39bd 100644 --- a/Makefile +++ b/Makefile @@ -78,10 +78,6 @@ block-nested-$(CONFIG_CURL) += curl.o block-obj-y +=

[PATCH 04/24] piix4_dev is not used for pc's

And mips still don't have pci assignment support :) Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target |2 +- hw/piix_pci.c |3 --- 2 files changed, 1 insertions(+), 4 deletions(-) diff --git a/Makefile.target b/Makefile.target index 5fad812..58f8b9e 100644 ---

[PATCH 06/24] Use common style for signalfd

s/CONFIG_signalfd/CONFIG_SIGNALFD/ Signed-off-by: Juan Quintela quint...@redhat.com --- compatfd.c |2 +- configure |4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/compatfd.c b/compatfd.c index 36e37e5..8c78597 100644 --- a/compatfd.c +++ b/compatfd.c @@ -100,7

[PATCH 07/24] One CONFIG_EVENTFD should be enough

(there was a missmerge with need_offset) Signed-off-by: Juan Quintela quint...@redhat.com --- compatfd.c |2 +- configure |4 qemu-kvm.c |2 +- 3 files changed, 2 insertions(+), 6 deletions(-) diff --git a/compatfd.c b/compatfd.c index 8c78597..594e4af 100644 --- a/compatfd.c

[PATCH 08/24] Bring ia64 to current arch selection code

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |3 --- 1 files changed, 0 insertions(+), 3 deletions(-) diff --git a/configure b/configure index bbb0c9d..7806456 100755 --- a/configure +++ b/configure @@ -2262,9 +2262,6 @@ case $target_arch2 in target_phys_bits=64 ;;

[PATCH 11/24] KVM temp hack not needed anymore

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |5 - 1 files changed, 0 insertions(+), 5 deletions(-) diff --git a/configure b/configure index 22d393b..c4ee415 100755 --- a/configure +++ b/configure @@ -2074,11 +2074,6 @@ bsd) ;; esac -# this is a temp hack needed for

[PATCH 10/24] Prin kvm options values as everything else

Signed-off-by: Juan Quintela quint...@redhat.com --- configure | 32 +--- 1 files changed, 9 insertions(+), 23 deletions(-) diff --git a/configure b/configure index 4512686..22d393b 100755 --- a/configure +++ b/configure @@ -2050,6 +2050,9 @@ fi if test

[PATCH 13/24] Use compile_prog function in thee missing compilations

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/configure b/configure index 930e450..8c69894 100755 --- a/configure +++ b/configure @@ -936,7 +936,7 @@ cat $TMPC EOF #endif int main(void) { return 0;

[PATCH 15/24] Rename USE_KVM_* to CONFIG_KVM_*

Once there, simplify Makefile.target with new syntax Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target | 13 +++-- configure|4 ++-- create_config|3 --- hw/ipf.c |4 ++-- hw/pc.c |6 +++--- hw/pci-hotplug.c | 14

[PATCH 19/24] Fix compilation without device assignment

Signed-off-by: Juan Quintela quint...@redhat.com --- qemu-kvm-x86.c |2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/qemu-kvm-x86.c b/qemu-kvm-x86.c index f80d82b..cdf4ebe 100644 --- a/qemu-kvm-x86.c +++ b/qemu-kvm-x86.c @@ -1632,11 +1632,13 @@ void

[PATCH 22/24] fix mis-merge (we define pkgversion 3 lines above

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/configure b/configure index 23a3dfb..289459a 100755 --- a/configure +++ b/configure @@ -236,7 +236,6 @@ blobs=yes pkgversion= ($(kvm_version))

[PATCH 18/24] Add kvm-cap-device-assignment command line flags

Signed-off-by: Juan Quintela quint...@redhat.com --- configure | 20 ++-- 1 files changed, 18 insertions(+), 2 deletions(-) diff --git a/configure b/configure index df45cd4..9598331 100755 --- a/configure +++ b/configure @@ -229,7 +229,7 @@ io_thread=no mixemu=no

[PATCH 21/24] __user is not used anywhere in qemu sources

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/configure b/configure index 91246a1..23a3dfb 100755 --- a/configure +++ b/configure @@ -1785,8 +1785,6 @@ if test $werror = yes ; then QEMU_CFLAGS=-Werror

[PATCH 23/24] Get CONFIG_CPU_EMULATION back to life

Code compile with --disable-cpu-emulation, and that don't compile tcg. This is the minimal set of changes to get it working Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target |6 -- configure |6 -- exec.c |2 ++

[PATCH 20/24] Only print kvm options when kvm is enabled

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |8 +--- 1 files changed, 5 insertions(+), 3 deletions(-) diff --git a/configure b/configure index 9598331..91246a1 100755 --- a/configure +++ b/configure @@ -1881,9 +1881,11 @@ echo IO thread $io_thread echo Linux

[PATCH 17/24] Add kvm-cap-pit command line flags

Signed-off-by: Juan Quintela quint...@redhat.com --- configure | 19 +-- 1 files changed, 17 insertions(+), 2 deletions(-) diff --git a/configure b/configure index 73ba095..df45cd4 100755 --- a/configure +++ b/configure @@ -228,7 +228,7 @@ uname_release= io_thread=no

[PATCH 09/24] Use configure way of enabling kvm

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |7 +-- 1 files changed, 1 insertions(+), 6 deletions(-) diff --git a/configure b/configure index 7806456..4512686 100755 --- a/configure +++ b/configure @@ -362,19 +362,14 @@ AIX) usb=linux if [ $cpu = i386 -o $cpu =

[PATCH 14/24] Test for libpci, not only for header

Signed-off-by: Juan Quintela quint...@redhat.com --- configure |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/configure b/configure index 8c69894..fc067a4 100755 --- a/configure +++ b/configure @@ -961,9 +961,9 @@ cat $TMPC EOF #ifndef PCI_VENDOR_ID #error NO

[PATCH 24/24] Remove build-targets-* rules that are always empty

Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile |7 --- 1 files changed, 0 insertions(+), 7 deletions(-) diff --git a/Makefile b/Makefile index 88b39bd..1e09953 100644 --- a/Makefile +++ b/Makefile @@ -386,10 +386,3 @@ tarbin: # Include automatically generated dependency

[PATCH 12/24] Introduce libs_softmmu to device assignment code

Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target |2 -- configure |2 +- 2 files changed, 1 insertions(+), 3 deletions(-) diff --git a/Makefile.target b/Makefile.target index 58f8b9e..525a8c2 100644 --- a/Makefile.target +++ b/Makefile.target @@ -217,7 +217,6

[PATCH 16/24] Move kvm specific tests after main kvm test

Signed-off-by: Juan Quintela quint...@redhat.com --- configure | 140 ++--- 1 files changed, 69 insertions(+), 71 deletions(-) diff --git a/configure b/configure index ef7d03d..73ba095 100755 --- a/configure +++ b/configure @@ -905,76

Minimum Ethernet packet size problems in qemu-kvm

Hi! I noticed a curious little problem with my virtualized DSL router which seems to indicate that qemu-kvm (kvm-88-1081-g561d3ed) will deliver packets to the guest over the emulated network card (tested with rtl8139 and ne2k_pci) which wouldn't appear on real hardware because they are shorter

[PATCH 0/5] Nested SVM Interrupt Fixes

Looking through the interrupt handling paths I stumbled across several inconsitencies in the interrupt handling paths. Using this patchset, running the most recent KVM version in nested KVM works again. Alexander Graf (5): Implement #NMI exiting for nested SVM Don't call

[PATCH 3/5] Don't #VMEXIT(INTR) if we still have event_inj waiting

Real hardware would first process the event_inj field and then notify the host that an interrupt is waiting. Let's do the same and just not EXIT_INTR if we have an event pending for the L2 guest. Signed-off-by: Alexander Graf ag...@suse.de --- arch/x86/kvm/svm.c |4 1 files changed, 4

[PATCH 5/5] Notify nested hypervisor of lost event injections

Normally when event_inj is valid the host CPU would write the contents to exit_int_info, so the hypervisor knows that the event wasn't injected. We failed to do so so far, so let's model closer to the CPU. Signed-off-by: Alexander Graf ag...@suse.de --- arch/x86/kvm/svm.c | 16

[PATCH 1/5] Implement #NMI exiting for nested SVM

When injecting an NMI to the l1 guest while it was running the l2 guest, we didn't #VMEXIT but just injected the NMI to the l2 guest. Let's be closer to real hardware and #VMEXIT if we're supposed to do so. Signed-off-by: Alexander Graf ag...@suse.de --- arch/x86/kvm/svm.c | 38

[PATCH 2/5] Don't call svm_complete_interrupts for nested guests

SVM has some cleanup code, that tries to reinject interrupts and exceptions when the guest didn't manage to deal with them yet. It basically transfers them to KVM internal state. Unfortunately, the internal state is reserved for the L1 guest state, so we shouldn't try to go through that logic

[PATCH 4/5] Don't bail when injecting an event in nested SVM

There's a warning to alarm the user when the guest has a valid exit_int_info, but really shouldn't have. We don't want that warning when running an L2 guest, because we're in control of the state then anyways, so let's not warn when running nested. Signed-off-by: Alexander Graf ag...@suse.de ---

Re: [Autotest] [KVM-AUTOTEST PATCH 1/2] Adding Cerberus Test Control System 2 (ctcs2) as client test

Oh Chen, great idea! I am going to make some tests here to see what would be the appropriate timeout and update the cerberus wrapper (no need to keep the old version in there). On Fri, Sep 18, 2009 at 7:25 AM, Cao, Chen k...@redhat.com wrote: - Using a newer version compared to the existed

Re: [PATCH 1/5] Implement #NMI exiting for nested SVM

Alexander Graf wrote: When injecting an NMI to the l1 guest while it was running the l2 guest, we didn't #VMEXIT but just injected the NMI to the l2 guest. Let's be closer to real hardware and #VMEXIT if we're supposed to do so. Signed-off-by: Alexander Graf ag...@suse.de ---

Re: [PATCH 2/5] Don't call svm_complete_interrupts for nested guests

Alexander Graf wrote: SVM has some cleanup code, that tries to reinject interrupts and exceptions when the guest didn't manage to deal with them yet. It basically transfers them to KVM internal state. Unfortunately, the internal state is reserved for the L1 guest state, so we shouldn't try

Re: [PATCH] Fix display of package version for KVM

Frank Arnold frank.arn...@amd.com wrote: The configure script is constructing the pkgversion correctly. Due to some merge the same variable is getting purged again some lines later. Removing this line makes things work again. Signed-off-by: Frank Arnold frank.arn...@amd.com Acked-by: Andre

Re: [PATCH 1/5] Implement #NMI exiting for nested SVM

Am 18.09.2009 um 15:33 schrieb Jan Kiszka jan.kis...@siemens.com: Alexander Graf wrote: When injecting an NMI to the l1 guest while it was running the l2 guest, we didn't #VMEXIT but just injected the NMI to the l2 guest. Let's be closer to real hardware and #VMEXIT if we're supposed to

Re: [PATCH 1/5] Implement #NMI exiting for nested SVM

Alexander Graf wrote: Am 18.09.2009 um 15:33 schrieb Jan Kiszka jan.kis...@siemens.com: Alexander Graf wrote: When injecting an NMI to the l1 guest while it was running the l2 guest, we didn't #VMEXIT but just injected the NMI to the l2 guest. Let's be closer to real hardware and

Re: [PATCH 4/4] Make cpu runnable after sipi

On Sun, Sep 13, 2009 at 03:55:21PM +0300, Gleb Natapov wrote: On Sun, Sep 13, 2009 at 03:05:21PM +0300, Avi Kivity wrote: On 09/13/2009 11:30 AM, Gleb Natapov wrote: Signed-off-by: Gleb Natapovg...@redhat.com --- hw/apic.c |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)

Re: [PATCH] update qemu-ifup for modern ip route output

Applied, thanks. On Thu, Sep 17, 2009 at 03:38:38PM -0500, Dustin Kirkland wrote: update qemu-ifup for modern ip route output The output from the /sbin/ip utility has changed, adding two more columns, which breaks the qemu-ifup script, as it was depending on the last column being the

Re: [PATCH] qemu-kvm: Fix guest single-stepping

Applied, thanks. On Thu, Sep 17, 2009 at 07:53:57PM +0200, Jan Kiszka wrote: Hopefully the last regression of 4c0960c0: KVM_SET_GUEST_DEBUG requires properly synchronized guest registers (on x86: eflags) on entry. Signed-off-by: Jan Kiszka jan.kis...@siemens.com --- qemu-kvm.c |4

Re: [PATCH 4/4] Make cpu runnable after sipi

On Fri, Sep 18, 2009 at 07:22:22PM +0300, Gleb Natapov wrote: On Fri, Sep 18, 2009 at 01:18:18PM -0300, Marcelo Tosatti wrote: On Sun, Sep 13, 2009 at 03:55:21PM +0300, Gleb Natapov wrote: On Sun, Sep 13, 2009 at 03:05:21PM +0300, Avi Kivity wrote: On 09/13/2009 11:30 AM, Gleb Natapov

Re: whats the advantage of enable-io-thread

On Fri, Sep 18, 2009 at 07:23:46AM +0200, Dietmar Maurer wrote: Hi all, what is the advantage of the --enable-io-thread options. It is disabled by default - why? - Dietmar qemu-kvm uses its own io thread by default. This option only makes sense in qemu upstream (at the moment). Its

Re: [RFC] defer skb allocation in virtio_net -- mergable buff part

Hello Michael, I am working on the patch to address the question you raised below. I am adding one more function -- destroy_buf in virtqueue_ops, so we don't need to maintain the list of pending buffers in upper layer (like virtio_net), when the device is shutdown or removed, this buffer free

Re: [PATCH -v3] QEMU-KVM: MCE: Relay UCR MCE to guest

On Fri, Sep 18, 2009 at 03:58:59PM +0800, Huang Ying wrote: UCR (uncorrected recovery) MCE is supported in recent Intel CPUs, where some hardware error such as some memory error can be reported without PCC (processor context corrupted). To recover from such MCE, the corresponding memory will

[PATCH] qemu-kvm: Reindent pc_new_cpu

...for better match with upstream. Signed-off-by: Jan Kiszka jan.kis...@siemens.com --- PS: regs_modified = 1 is another sign that the initialization order in qemu-kvm needs a rework. hw/pc.c | 28 +++- 1 files changed, 15 insertions(+), 13 deletions(-) diff --git

[PATCH] qemu-kvm: Cleanup compatfd makefile rule

Signed-off-by: Jan Kiszka jan.kis...@siemens.com --- Makefile |7 +-- 1 files changed, 1 insertions(+), 6 deletions(-) diff --git a/Makefile b/Makefile index 6a9ca7c..6855fa2 100644 --- a/Makefile +++ b/Makefile @@ -65,9 +65,8 @@ recurse-all: $(SUBDIR_RULES) $(ROMSUBDIR_RULES)

Re: [PATCH 1/4] core, x86: Add user return notifiers

On 09/16/2009 05:45 AM, Avi Kivity wrote: Add a general per-cpu notifier that is called whenever the kernel is about to return to userspace. The notifier uses a thread_info flag and existing checks, so there is no impact on user return or context switch fast paths. Signed-off-by: Avi

Re: Possible problem with -no-kvm-irqchip -no-kvm-pit qemu options detected during git daily testing

On Fri, Sep 18, 2009 at 6:51 AM, Gleb Natapov g...@redhat.com wrote: On Thu, Sep 17, 2009 at 09:57:05PM -0300, Lucas Meneghel Rodrigues wrote: Hi folks, after looking at today's git testing results, it seems like we have a problem with the command line options: -no-kvm-irqchip -no-kvm-pit

[PATCH][KVM][2/2] Support Pause Filter in AMD processors

This patch depends on [PATCH] Prevent immediate process rescheduling that I just submitted. New AMD processors (Family 0x10 models 8+) support the Pause Filter Feature. This feature creates a new field in the VMCB called Pause Filter Count. If Pause Filter Count is greater than 0 and

Re: [PATCH 00/24] configure and Makefile cleanup

On Fri, Sep 18, 2009 at 01:41:08PM +0200, Juan Quintela wrote: This series: - werror is not disabled unconditoanally, you can disable it with --disable-werror. - clean configure and Makefile* of mis-merges - sync such files to qemu upstream - kvm options are detected and configured _after_

Re: [PATCH 04/24] piix4_dev is not used for pc's

On Fri, Sep 18, 2009 at 01:41:13PM +0200, Juan Quintela wrote: And mips still don't have pci assignment support :) Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target |2 +- hw/piix_pci.c |3 --- 2 files changed, 1 insertions(+), 4 deletions(-) diff --git

Re: [PATCH 23/24] Get CONFIG_CPU_EMULATION back to life

On Fri, Sep 18, 2009 at 01:41:34PM +0200, Juan Quintela wrote: Code compile with --disable-cpu-emulation, and that don't compile tcg. This is the minimal set of changes to get it working Better see this upstreamed (and perhaps remove support from qemu-kvm, unless someone disagrees).

Re: [PATCH] qemu-kvm: Reindent pc_new_cpu

Applied, thanks. On Fri, Sep 18, 2009 at 07:28:36PM +0200, Jan Kiszka wrote: ...for better match with upstream. Signed-off-by: Jan Kiszka jan.kis...@siemens.com --- PS: regs_modified = 1 is another sign that the initialization order in qemu-kvm needs a rework. hw/pc.c | 28

Re: [PATCH 04/24] piix4_dev is not used for pc's

Marcelo Tosatti mtosa...@redhat.com wrote: On Fri, Sep 18, 2009 at 01:41:13PM +0200, Juan Quintela wrote: And mips still don't have pci assignment support :) Signed-off-by: Juan Quintela quint...@redhat.com --- Makefile.target |2 +- hw/piix_pci.c |3 --- 2 files changed, 1

Re: [PATCH 20/24] Only print kvm options when kvm is enabled

Marcelo Tosatti mtosa...@redhat.com wrote: On Fri, Sep 18, 2009 at 01:41:31PM +0200, Juan Quintela wrote: Signed-off-by: Juan Quintela quint...@redhat.com --- configure |8 +--- 1 files changed, 5 insertions(+), 3 deletions(-) diff --git a/configure b/configure index

Re: [PATCH 23/24] Get CONFIG_CPU_EMULATION back to life

Marcelo Tosatti mtosa...@redhat.com wrote: On Fri, Sep 18, 2009 at 01:41:34PM +0200, Juan Quintela wrote: Code compile with --disable-cpu-emulation, and that don't compile tcg. This is the minimal set of changes to get it working Better see this upstreamed (and perhaps remove support from

kvm-kmod 88 or git not working with 2.6.31

Hello I am trying to compile kvm-kmod-devel-88.tar.gz with 2.6.31. Seems there is a missing header file. vger kvm-kmod-devel-88 # make make -C /lib/modules/2.6.31-gentoo/build M=`pwd` \ LINUXINCLUDE=-I`pwd`/include -Iinclude \ -Iarch/x86/include