On 05/06/13 22:49, Andreas Färber wrote:
> Am 06.05.2013 19:14, schrieb akoskov...@gmx.com:
>> From: Ákos Kovács
>>
>> Cc: Laszlo Ersek
>> Signed-off-by: Ákos Kovács
>> ---
>> Changes in v2:
>> * Fixed input release/grab translations
>> * Fixed inconsistency with the "leállítva"/"megál
Hi all,
When we use the latest version of QEMU to build ovirt,
we get this error reported from libvirt.
qemu-system-ppc64: -device
virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel0,id=channel0,name=com.redhat.rhevm.vdsm:
Bus 'virtio-serial0.0' is full
qemu-system-ppc64: -device
virts
Hello All,
Is there any new regarding this patch?
Thanks,
Dmitry
On Thu, May 2, 2013 at 8:20 AM, Michael S. Tsirkin wrote:
> On Mon, Apr 22, 2013 at 12:58:26PM -0500, Anthony Liguori wrote:
> > "Michael S. Tsirkin" writes:
> >
> > > On Sun, Apr 07, 2013 at 09:34:08AM +0300, Dmitry Fleytman w
This series of patches add the support for the i.MX25 processor through the
Freescale 3DS evaluation board.
For now a limited set of devices are supported.
* GPT timers (from i.MX31)
* EPIT timers (from i.MX31)
* Serial ports (from i.MX31)
* Ethernet FEC port
* I2C controller
This is based on the mcf_fec.c FEC implementation for ColdFire.
* a generic phy was added (borrowed from lan9118).
* The buffer management is also modified as buffers are
slightly different between coldfire and i.MX.
Signed-off-by: Jean-Christophe DUBOIS
---
Changes since V1:
This is using a ds1338 RTC chip on the i2c bus. This RTC
chip is not present on the real board.
Signed-off-by: Jean-Christophe DUBOIS
---
Changes since v1:
* not present on v1
Changes since v2:
* use a common header file for I2C regs definition
Changes since v3:
* rework GPL header
For now we support:
* timers (GPT and EPIT)
* serial ports (only 2 out of 5 possible)
* ethernet (through the newly added FEC driver)
* I2C (through the newly added I2C driver)
A ds1338 I2C RTC chip was added on the first i2c bus to allow
automatic test through qtest. This RTC is n
The slave mode is not implemented.
Signed-off-by: Jean-Christophe DUBOIS
---
Changes since v1:
* use QOM cast
* run checkpatch on code
* added restrictin on MemoryRegionOps
* use DeviceClass::realise as init function
Changes since v2:
* use CamelCase for state type
* use
Dong Xu Wang writes:
> On 2013/5/6 20:20, Markus Armbruster wrote:
>> Dong Xu Wang writes:
>>
>>> These functions will be used in next commit. qemu_opt_get_(*)_del functions
>>> are used to make sure we have the same behaviors as before: after get an
>>> option value, options++.
>>
>> I don't un
Add API to create header, sub header of vmcore in kdump-compressed format.
The data is store in struct DumpState.
The following patch will use this function to gather data of header, then
write them into vmcore.
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
---
dump.c|
On 2013/5/7 15:38, Markus Armbruster wrote:
Dong Xu Wang writes:
On 2013/5/6 20:20, Markus Armbruster wrote:
Dong Xu Wang writes:
These functions will be used in next commit. qemu_opt_get_(*)_del functions
are used to make sure we have the same behaviors as before: after get an
option valu
Add API to get data of page desc and page data and save them into tmp files.
The following patch will use these functions to gather data of page desc and
page data, then write them into vmcore
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
---
dump.c| 254
Hi, all
Command 'dump-guest-memory' was introduced to dump guest's memory. But the
vmcore's format is only elf32 or elf64. The message is here:
http://lists.gnu.org/archive/html/qemu-devel/2012-04/msg03379.html
These patches are used to make 'dump-guest-memory' be able to dump guest's
memory in k
Make monitor command 'dump-guest-memory' dump in kdump-compressed format.
The command's usage:
dump [-p] protocol [flags] [begin] [length]
With 'flags' set, the core file will be in kdump-compress format, and without
it, the format is ELF. 'flags' can be:
1. FLAG_DUMP_COMPRESS_ZLIB(0x1), compress
The following patch will use these functions to write cached data into vmcore.
Header is cached in DumpState, and bitmap and page are cached in tmp files.
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
---
dump.c| 247 +
inc
When calling create_header, create_dump_bitmap and create_pages, some memory
spaces
are allocated. The following patch will use this function to free these spaces.
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
---
dump.c | 17 +
1 files changed, 17 insertions(+), 0 del
Il 07/05/2013 09:22, Laszlo Ersek ha scritto:
>>> diff --git a/po/hu.po b/po/hu.po
>>> new file mode 100644
>>> index 000..340709f
>>> --- /dev/null
>>> +++ b/po/hu.po
>>> @@ -0,0 +1,63 @@
>>> +# Hungarian translation for QEMU.
>>> +# This file is put in the public domain.
>>
>> Same issue as w
On Tue, May 07, 2013 at 01:42:49PM +0800, Jason Wang wrote:
> There are several several issues in the current checking:
>
> - The check was based on the minus of unsigned values which can overflow
> - It was done after .{set|get}_config() which can lead crash when config_len
> is zero since vdev
Move includes and definition of struct DumpState into include/sysemu/dump.h.
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
---
dump.c| 30 --
include/sysemu/dump.h | 31 +++
2 files changed, 31 insertions(+), 30
Struct dump_bitmap is associated with a tmp file, and the tmp file can be used
to save data of page desc and page data in kdump-compressed format temporarily.
The following patch will use these function to get the data of page desc and
page data and cache them in tmp files.
Signed-off-by: Qiao Nuo
Struct dump_bitmap is associated with a tmp file, and the tmp file can be used
to save data of bitmap in kdump-compressed format temporarily.
The following patch will use these functions to get the data of bitmap and cache
them into tmp files.
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
Add API to get data of the 1st and 2nd dump bitmap and save them into tmp files.
The following patch will use these functions to gather data of dump bitmap,
then write them into vmcore.
Signed-off-by: Qiao Nuohan
Reviewed-by: Zhang Xiaohe
---
dump.c| 98 +++
On 05/07/13 10:01, Paolo Bonzini wrote:
> Il 07/05/2013 09:22, Laszlo Ersek ha scritto:
diff --git a/po/hu.po b/po/hu.po
new file mode 100644
index 000..340709f
--- /dev/null
+++ b/po/hu.po
@@ -0,0 +1,63 @@
+# Hungarian translation for QEMU.
+# This file
On Tue, May 07, 2013 at 01:42:49PM +0800, Jason Wang wrote:
> There are several several issues in the current checking:
>
> - The check was based on the minus of unsigned values which can overflow
> - It was done after .{set|get}_config() which can lead crash when config_len
> is zero since vdev
Il 07/05/2013 10:26, Laszlo Ersek ha scritto:
> I suspected that something like this was in the background, but what I
> didn't understand was: why single out the public domain, as the GPL
> itself is in the exact same bucket, generally speaking.
I think because public domain is a concept that _do
Hello,
On 17/01/13 21:21, Stefan Weil wrote:
> Am 17.01.2013 18:10, schrieb Wolfgang Mauerer:
>> For slow targets and fast hosts, the emulation may be faster
>> than the actual hardware, which can be undesirable for various
>> reasons. Add a run-time option to slow down the emulation
>> by sleepin
On 7 May 2013 01:39, Peter Crosthwaite wrote:
> There may also be cases where machine model want to create a NIC
> regardless of whether its used or not. Relevant for sysbus NICs as we
> don't have the luxury of a PCI probe process so a generic guest (e.g.
> a kernel and its pre-canned dtb) may as
On 06/05/2013 22:51, Anthony Liguori wrote:
"Michael S. Tsirkin" writes:
mask notifiers are never called without msix,
so devices with backend masking like vhost don't work.
Call mask notifiers explicitly at
startup/cleanup to make it work.
Signed-off-by: Michael S. Tsirkin
Tested-by: Alexan
i.MX25 PDK only has 64 MB DDR
Signed-off-by: Jean-Christophe DUBOIS
---
arch/arm/board/imx25/dts/3dstack/3dstack.dtsi| 3 ++-
arch/arm/board/imx25/dts/3dstack/one_guest_versatile.dts | 5 +
2 files changed, 3 insertions(+), 5 deletions(-)
diff --git a/arch/arm/board/imx25/dts/3d
Sorry for the 14/14 in the subject.
JC
On 05/07/2013 11:29 AM, Jean-Christophe DUBOIS wrote:
i.MX25 PDK only has 64 MB DDR
Signed-off-by: Jean-Christophe DUBOIS
---
arch/arm/board/imx25/dts/3dstack/3dstack.dtsi| 3 ++-
arch/arm/board/imx25/dts/3dstack/one_guest_versatile.dts |
Il 06/05/2013 22:46, Peter Maydell ha scritto:
> On 6 May 2013 15:26, Jan Kiszka wrote:
>> Simplify the sub-page handling by implementing it directly in the
>> dispatcher instead of using a redirection memory region. We extend the
>> phys_sections entries to optionally hold a pointer to the sub-se
Signed-off-by: Gerd Hoffmann
---
hw/usb/hcd-xhci.c |5 +
1 file changed, 5 insertions(+)
diff --git a/hw/usb/hcd-xhci.c b/hw/usb/hcd-xhci.c
index 2c90e56..8813bdf 100644
--- a/hw/usb/hcd-xhci.c
+++ b/hw/usb/hcd-xhci.c
@@ -1716,6 +1716,11 @@ static int xhci_complete_packet(XHCITransfer *x
Signed-off-by: Gerd Hoffmann
---
hw/usb/host-libusb.c | 54 +-
1 file changed, 49 insertions(+), 5 deletions(-)
diff --git a/hw/usb/host-libusb.c b/hw/usb/host-libusb.c
index d1186b8..ee67c4c 100644
--- a/hw/usb/host-libusb.c
+++ b/hw/usb/host-li
On 05/07/2013 09:35 AM, Jean-Christophe DUBOIS wrote:
The slave mode is not implemented.
Signed-off-by: Jean-Christophe DUBOIS
---
Changes since v1:
* use QOM cast
* run checkpatch on code
* added restrictin on MemoryRegionOps
* use DeviceClass::realise as init function
Ch
From: Hans de Goede
Due to various unfortunate reasons we cannot reliable detect a guest
cancelling a packet as soon as it happens, instead we detect cancels
with some delay.
When packets are handled async, and we directly pass the guest memory for
the packet to the usb-device as iovec, this mea
Il 07/05/2013 09:20, Li Zhang ha scritto:
> Hi all,
>
> When we use the latest version of QEMU to build ovirt,
> we get this error reported from libvirt.
What QEMU commit is this?
It might have been fixed already.
Paolo
>
> qemu-system-ppc64: -device
> virtserialport,bus=virtio-serial0.0,nr=1
On Mon, May 06, 2013 at 09:56:57PM +0200, Andrea Arcangeli wrote:
> Hello everyone,
>
> this is a patchset to implement two new kernel features:
> MADV_USERFAULT and remap_anon_pages.
>
> The combination of the two features are what I would propose to
> implement postcopy live migration, and in g
mask notifiers are never called without msix,
so devices with backend masking like vhost don't work.
Call mask notifiers explicitly at
startup/cleanup to make it work.
Signed-off-by: Michael S. Tsirkin
Tested-by: Alexander Graf
---
Changes from v1:
- rebase to master
hw/virtio/virtio-pci.
All buses do this, and if they don't they get subtle
bugs related to cross version migration.
Let's add an assert to catch these bugs early.
Signed-off-by: Michael S. Tsirkin
---
Just a cleanup so not 1.5 material.
Seems to work fine here - any opinions?
hw/net/virtio-net.c | 7 ---
1 file
Hi,
A few small usb fixes for 1.5. Two host controller bugfixes, and two
small enhancements for the libusb based usb-host implementation. They
bring the libusb on par with the old linux usbfs implementation and are
needed to make sure usb passthrough doesn't lose features in 1.5.
please pull,
Alloes to pass through usb2 devices on usb1 host controllers if possible.
Brings the libusb implementation to feature-parity with the linux usbfs
code, so the usb-host implementation in 1.5 (libusb) doesn't regress
compared to 1.4 (usbfs).
Signed-off-by: Gerd Hoffmann
---
hw/usb/host-libusb.c |
Alexander Graf writes:
There's a certain chance that happens, yes. We don't have instruction
test suites for the PPC target.
There certainly are more bugs. GMP still crashes all over the place.
I have semi-isolated one more.
Extracted stand-alone sources:
bug-qemu-ppc-again.c
Descript
On 7 May 2013 11:27, Torbjorn Granlund wrote:
> Alexander Graf writes:
>
> There's a certain chance that happens, yes. We don't have instruction
> test suites for the PPC target.
>
> There certainly are more bugs. GMP still crashes all over the place.
If you want to more seriously test the
Wolfgang Mauerer writes:
> Hello,
>
> On 17/01/13 21:21, Stefan Weil wrote:
>> Am 17.01.2013 18:10, schrieb Wolfgang Mauerer:
>>> For slow targets and fast hosts, the emulation may be faster
>>> than the actual hardware, which can be undesirable for various
>>> reasons. Add a run-time option to s
When testing vmdk, qemu-iotests fails on 017, 018, 042 and 043 for trivial
reasons. These patches try to fix those cases to pass tests on all four
subformats of vmdk {monolithicSparse monolithicFlat twoGbMaxExtentSparse
twoGbMaxExtentFlat}.
Fam Zheng (3):
qemu-iotests: exclude vmdk for test 042
017 and 018 use /bin/mv to move base img from t.IMGFMG to t.IMGFMT.base
after filling data, this is not enough for vmdk, when t.IMGFMT is only a
description text file who points to t-{flat,s001,f001,...}.IMGFMT as
data extent, so testing such subformats alway fails on them.
This patch use the tric
Zero sized disk is not supported by qemu vmdk driver, exclude vmdk from
the test script.
As tested on vmware-vdiskmanager and vmware workstation, zero sized disk
is not supported by vmware, either.
Signed-off-by: Fam Zheng
---
tests/qemu-iotests/042 | 2 +-
1 file changed, 1 insertion(+), 1 del
043 tests recursive backing file by changing backing file. VMDK has not
implemented this yet.
Signed-off-by: Fam Zheng
---
tests/qemu-iotests/043 | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/qemu-iotests/043 b/tests/qemu-iotests/043
index 3ba08dc..dd06c37 100755
--- a
Il 07/05/2013 07:00, Chegu Vinod ha scritto:
> Busy enterprise workloads hosted on large sized VM's tend to dirty
> memory faster than the transfer rate achieved via live guest migration.
> Despite some good recent improvements (& using dedicated 10Gig NICs
> between hosts) the live migration does
On Mon, May 06, 2013 at 09:56:58PM +0200, Andrea Arcangeli wrote:
> +++ b/mm/madvise.c
> @@ -93,6 +93,21 @@ static long madvise_behavior(struct vm_area_struct * vma,
> if (error)
> goto out;
> break;
> + case MADV_USERFAULT:
> + if (
On Fri, Apr 26, 2013 at 06:27:40PM +0800, Jason Wang wrote:
> Commit 32993698 (vhost: disable on tap link down) tries to disable the vhost
> also when the peer's link is down. But the check was not done properly, the
> vhost were only started when:
>
> 1) peer's link is not down
> 2) virtio-net ha
On 07/05/13 12:56, Markus Armbruster wrote:
(...)
>> +if (slowdown_option && (kvm_enabled() || xen_enabled())) {
>> +fprintf(stderr, "-slowdown is not allowed with kvm or xen\n");
>> +exit(1);
>
> Suggest to mention this in the commit message, and perhaps the manual
> (STEXI.
Hi Andrew,
On Tue, May 07, 2013 at 01:16:30PM +0200, Andrew Jones wrote:
> On Mon, May 06, 2013 at 09:56:58PM +0200, Andrea Arcangeli wrote:
> > @@ -405,6 +420,7 @@ madvise_behavior_valid(int behavior)
> > case MADV_HUGEPAGE:
> > case MADV_NOHUGEPAGE:
> > #endif
> > + case MADV_USERFAUL
On Mon, May 06, 2013 at 09:56:57PM +0200, Andrea Arcangeli wrote:
>
> The current behavior of remap_anon_pages is very strict to avoid any
> chance of memory corruption going unnoticed, and it will return
> -EFAULT at the first sign of something unexpected (like a page already
> mapped in the dest
I am trying to use qemu to
1. cover more of the assembly code in GMP
2. check configuration logic of GMP
but I am not as successful as I would like to be.
The 2nd table of http://gmplib.org/devel/testsystems.html shows all
emulated systems I am using, most of which are qemu-based.
Unfortunately
From: Laszlo Ersek
The qemu guest agent creates a bunch of files with insecure permissions
when started in daemon mode. For example:
-rw-rw-rw- 1 root root /var/log/qemu-ga.log
-rw-rw-rw- 1 root root /var/run/qga.state
-rw-rw-rw- 1 root root /var/log/qga-fsfreeze-hook.log
In addition, at
Peter Maydell writes:
On 7 May 2013 11:27, Torbjorn Granlund wrote:
> Alexander Graf writes:
>
> There's a certain chance that happens, yes. We don't have instruction
> test suites for the PPC target.
>
> There certainly are more bugs. GMP still crashes all over the place.
On 7 May 2013 12:48, Torbjorn Granlund wrote:
> Peter Maydell writes:
> On 7 May 2013 11:27, Torbjorn Granlund wrote:
> > Alexander Graf writes:
> > There's a certain chance that happens, yes. We don't have instruction
> > test suites for the PPC target.
> >
> > There certainly
Hi Isaku,
On Tue, May 07, 2013 at 07:07:40PM +0900, Isaku Yamahata wrote:
> On Mon, May 06, 2013 at 09:56:57PM +0200, Andrea Arcangeli wrote:
> > Hello everyone,
> >
> > this is a patchset to implement two new kernel features:
> > MADV_USERFAULT and remap_anon_pages.
> >
> > The combination of t
On Tue, May 07, 2013 at 01:38:10PM +0200, Andrew Jones wrote:
> What about instead of adding a new syscall (remap_anon_pages) to
> instead extend mremap with new flags giving it a strict mode?
I actually thought about this and it's a very interesting argument.
When I thought about it, I felt the
On 07/05/2013 12:20, Michael S. Tsirkin wrote:
mask notifiers are never called without msix,
so devices with backend masking like vhost don't work.
Call mask notifiers explicitly at
startup/cleanup to make it work.
Signed-off-by: Michael S. Tsirkin
Tested-by: Alexander Graf
---
Changes from v
On Tue, May 07, 2013 at 02:20:25PM +0200, KONRAD Frédéric wrote:
> On 07/05/2013 12:20, Michael S. Tsirkin wrote:
> >mask notifiers are never called without msix,
> >so devices with backend masking like vhost don't work.
> >Call mask notifiers explicitly at
> >startup/cleanup to make it work.
> >
>
Hi,
I think it is not a good idea, as we wanted to make VirtIODevice
hot-pluggable for virtio-mmio.
Maybe we can use a callback which is called by virtio-bus before
plugging, to ensure that this
config size is computed?
On 07/05/2013 12:22, Michael S. Tsirkin wrote:
All buses do this, and
On Tue, May 07, 2013 at 02:29:38PM +0200, KONRAD Frédéric wrote:
> Hi,
>
> I think it is not a good idea, as we wanted to make VirtIODevice
> hot-pluggable for virtio-mmio.
And then this hack will break cross version migration.
> Maybe we can use a callback which is called by virtio-bus before
>
Il 06/05/2013 22:46, Peter Maydell ha scritto:
> On 6 May 2013 15:26, Jan Kiszka wrote:
>> Simplify the sub-page handling by implementing it directly in the
>> dispatcher instead of using a redirection memory region. We extend the
>> phys_sections entries to optionally hold a pointer to the sub-se
Upstream dtc.git introduced a change in libfdt_env.h, which breaks
compilation with QEMU's version of it:
CC arm-softmmu/device_tree.o
In file included from /usr/include/libfdt.h:55:0,
from /src/qemu.git/device_tree.c:28:
/usr/include/fdt.h:7:2: error: unknown type name 'fdt32_t
On 7 May 2013 13:36, Andre Przywara wrote:
> Upstream dtc.git introduced a change in libfdt_env.h, which breaks
> compilation with QEMU's version of it:
>
> CC arm-softmmu/device_tree.o
> In file included from /usr/include/libfdt.h:55:0,
> from /src/qemu.git/device_tree.c:28:
>
On 05/07/2013 02:44 PM, Peter Maydell wrote:
On 7 May 2013 13:36, Andre Przywara wrote:
Upstream dtc.git introduced a change in libfdt_env.h, which breaks
compilation with QEMU's version of it:
CC arm-softmmu/device_tree.o
In file included from /usr/include/libfdt.h:55:0,
On 07/05/2013 14:33, Michael S. Tsirkin wrote:
On Tue, May 07, 2013 at 02:29:38PM +0200, KONRAD Frédéric wrote:
Hi,
I think it is not a good idea, as we wanted to make VirtIODevice
hot-pluggable for virtio-mmio.
And then this hack will break cross version migration.
Why?
virtio_net_set_conf
A little delayed as we had some last minute pull requests but we are now
bug fixes only from this point on. We'll have the first -rc1 on
Wednesday. I'll post an -rc0 tarball shortly.
Regards,
Anthony Liguori
On 10 March 2013 14:21, Igor Mitsyanko wrote:
> First two patches fix issues in exynos4210 code which were blocking proper
> memory
> migration.
>
> Third patch makes memory_region_init_ram_ptr assert if memory region size is
> not a
> multiple of TARGET_PAGE_SIZE.
>
> Igor Mitsyanko (3):
> hw
Il 07/05/2013 15:09, Peter Maydell ha scritto:
> On 10 March 2013 14:21, Igor Mitsyanko wrote:
>> First two patches fix issues in exynos4210 code which were blocking proper
>> memory
>> migration.
>>
>> Third patch makes memory_region_init_ram_ptr assert if memory region size is
>> not a
>> mult
On 7 May 2013 13:52, Andre Przywara wrote:
> On 05/07/2013 02:44 PM, Peter Maydell wrote:
>> I'm not entirely sure I understand why we need change.
>> Have upstream really introduced a breaking change for
>> everybody who uses libfdt, or are we using it wrongly?
>
>
> Everybody who copies and chan
On 07/05/13 12:38, Torbjorn Granlund wrote:
Hi Torbjorn,
I am trying to use qemu to
1. cover more of the assembly code in GMP
2. check configuration logic of GMP
but I am not as successful as I would like to be.
The 2nd table of http://gmplib.org/devel/testsystems.html shows all
emulated sys
Signed-off-by: Kevin Wolf
---
po/de_DE.po | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/po/de_DE.po b/po/de_DE.po
index 92c5df5..e35aaf4 100644
--- a/po/de_DE.po
+++ b/po/de_DE.po
@@ -18,11 +18,11 @@ msgstr ""
#: ../ui/gtk.c:213
msgid " - Press Ctrl+Alt+G to release
Factor out endpoint context initialization to a separate function.
xhci live migration will need that too, in post_load.
Signed-off-by: Gerd Hoffmann
---
hw/usb/hcd-xhci.c | 43 +--
1 file changed, 25 insertions(+), 18 deletions(-)
diff --git a/hw/usb/h
Using a trick cut+pasted from vmstate_scsi_device
to wind up msix_save and msix_load.
---
hw/pci/msix.c | 33 +
include/hw/pci/msix.h | 11 +++
2 files changed, 44 insertions(+)
diff --git a/hw/pci/msix.c b/hw/pci/msix.c
index e231a0d..6da75ec 1
Factor out endpoint context allocation to a separate function.
xhci live migration will need that too, in post_load.
Signed-off-by: Gerd Hoffmann
---
hw/usb/hcd-xhci.c | 32 +---
1 file changed, 21 insertions(+), 11 deletions(-)
diff --git a/hw/usb/hcd-xhci.c b/hw/
Hi,
This patch series adds live migration support to the xhci usb host
adapter. Very first cut, survived light testing, not stressed yet.
/me looks for review comments, especially on patch #1 for better
msix live migration.
cheers,
Gerd
Gerd Hoffmann (6):
pci: add VMSTATE_MSIX
xhci: ad
Preparing for live-migration support, post_load will need that.
Signed-off-by: Gerd Hoffmann
---
hw/usb/hcd-xhci.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/hw/usb/hcd-xhci.c b/hw/usb/hcd-xhci.c
index 8813bdf..ac683ce 100644
--- a/hw/usb/hcd-xhci.c
+++ b/hw/usb/hcd-xhci.c
@@ -405
Signed-off-by: Gerd Hoffmann
---
hw/usb/hcd-xhci.c | 145 -
1 file changed, 144 insertions(+), 1 deletion(-)
diff --git a/hw/usb/hcd-xhci.c b/hw/usb/hcd-xhci.c
index 9b90067..426478c 100644
--- a/hw/usb/hcd-xhci.c
+++ b/hw/usb/hcd-xhci.c
@@ -3
On 05/06/13 10:48, Hans de Goede wrote:
> Due to various unfortunate reasons we cannot reliable detect a guest
> cancelling a packet as soon as it happens, instead we detect cancels
> with some delay.
>
> When packets are handled async, and we directly pass the guest memory for
> the packet to the
For testing only, migration support isn't finished yet.
Signed-off-by: Gerd Hoffmann
---
hw/usb/hcd-xhci.c |1 -
1 file changed, 1 deletion(-)
diff --git a/hw/usb/hcd-xhci.c b/hw/usb/hcd-xhci.c
index 426478c..4ab138a 100644
--- a/hw/usb/hcd-xhci.c
+++ b/hw/usb/hcd-xhci.c
@@ -3501,7 +3501,6
On 2013年05月07日 17:50, Paolo Bonzini wrote:
Il 07/05/2013 09:20, Li Zhang ha scritto:
Hi all,
When we use the latest version of QEMU to build ovirt,
we get this error reported from libvirt.
What QEMU commit is this?
It might have been fixed already.
You mean this is one bug of QEMU, right?
mask notifiers are never called without msix,
so devices with backend masking like vhost don't work.
Call mask notifiers explicitly at
startup/cleanup to make it work.
Signed-off-by: Michael S. Tsirkin
Tested-by: Alexander Graf
---
changes from v2:
fix bug spotted by Konrad
changes from v1:
Juan Quintela wrote:
> Hi
>
> Please send in any agenda topics you are interested in.
As there are no topics, call gets cancelled.
Have a nice day.
Later, Juan.
On 2013年05月07日 21:05, Alexey Kardashevskiy wrote:
On 05/07/2013 07:50 PM, Paolo Bonzini wrote:
Il 07/05/2013 09:20, Li Zhang ha scritto:
Hi all,
When we use the latest version of QEMU to build ovirt,
we get this error reported from libvirt.
What QEMU commit is this?
b3e6d591b05538056d665572
On 05/07/2013 03:24 PM, Peter Maydell wrote:
On 7 May 2013 13:52, Andre Przywara wrote:
On 05/07/2013 02:44 PM, Peter Maydell wrote:
I'm not entirely sure I understand why we need change.
Have upstream really introduced a breaking change for
everybody who uses libfdt, or are we using it wrongl
On Tue, May 07, 2013 at 02:54:38PM +0200, KONRAD Frédéric wrote:
> On 07/05/2013 14:33, Michael S. Tsirkin wrote:
> >On Tue, May 07, 2013 at 02:29:38PM +0200, KONRAD Frédéric wrote:
> >>Hi,
> >>
> >>I think it is not a good idea, as we wanted to make VirtIODevice
> >>hot-pluggable for virtio-mmio.
On Tue, May 07, 2013 at 03:34:31PM +0200, Gerd Hoffmann wrote:
> Using a trick cut+pasted from vmstate_scsi_device
> to wind up msix_save and msix_load.
Any reason this is not signed-off by you?
> ---
> hw/pci/msix.c | 33 +
> include/hw/pci/msix.h | 1
On Tue, May 07, 2013 at 03:34:35PM +0200, Gerd Hoffmann wrote:
> Signed-off-by: Gerd Hoffmann
> ---
> hw/usb/hcd-xhci.c | 145
> -
> 1 file changed, 144 insertions(+), 1 deletion(-)
>
> diff --git a/hw/usb/hcd-xhci.c b/hw/usb/hcd-xhci.c
> ind
USB_RET_ASYNC is -6, so inflight was always false.
Signed-off-by: Ed Maste
---
hw/usb/host-libusb.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/usb/host-libusb.c b/hw/usb/host-libusb.c
index d1186b8..0af6ce3 100644
--- a/hw/usb/host-libusb.c
+++ b/hw/usb/host-libusb.c
On 05/07/13 16:05, Michael S. Tsirkin wrote:
> On Tue, May 07, 2013 at 03:34:31PM +0200, Gerd Hoffmann wrote:
>> > Using a trick cut+pasted from vmstate_scsi_device
>> > to wind up msix_save and msix_load.
> Any reason this is not signed-off by you?
>
Just forgot it, here we go:
Signed-off-by: Ge
On Tue, May 07, 2013 at 04:08:46PM +0200, Gerd Hoffmann wrote:
> On 05/07/13 16:05, Michael S. Tsirkin wrote:
> > On Tue, May 07, 2013 at 03:34:31PM +0200, Gerd Hoffmann wrote:
> >> > Using a trick cut+pasted from vmstate_scsi_device
> >> > to wind up msix_save and msix_load.
> > Any reason this is
On 05/07/13 16:06, Michael S. Tsirkin wrote:
>> +#if 0
> should have a comment explaining why is this
> commented out.
> Or just drop this section.
>
>> > +/* event queue (used if ring is full) */
>> > +VMSTATE_BOOL(er_full, XHCIInterrupter),
>> > +VMSTATE_UINT32(ev
From: Avi Kivity
Add a new memory region type that translates addresses it is given,
then forwards them to a target address space. This is similar to
an alias, except that the mapping is more flexible than a linear
translation and trucation, and also less efficient since the
translation happens
Signed-off-by: Paolo Bonzini
---
hw/core/sysbus.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/hw/core/sysbus.c b/hw/core/sysbus.c
index 9004d8c..788696b 100644
--- a/hw/core/sysbus.c
+++ b/hw/core/sysbus.c
@@ -115,6 +115,8 @@ void sysbus_init_mmio(SysBusDevice *dev,
From: Avi Kivity
Use the new iommu support in the memory core for iommu support. The only
user, spapr, is also converted, but it still provides a DMAContext
interface until the non-PCI bits switch to AddressSpace.
Cc: Michael S. Tsirkin
Signed-off-by: Avi Kivity
Signed-off-by: Paolo Bonzini
Fetch the root region from the sPAPRTCETable, and use it to build
an AddressSpace and DMAContext.
Now, everywhere we have a DMAContext we also have access to the
corresponding AddressSpace (either because we create it just before
the DMAContext, or because dma_context_memory's AddressSpace is
triv
The TCE table is currently returned as a DMAContext, and non-type-safe
APIs are called later passing back the DMAContext. Since we want to move
away from DMAContext, use an opaque type instead, and add an accessor
to retrieve the DMAContext from it.
Acked-by: David Gibson
Signed-off-by: Paolo Bo
1 - 100 of 250 matches
Mail list logo
