add-cow file format core code. It use block-cache.c as cache code.
It lacks of snapshot_blkdev support.
v15->v16):
1) Judge if opts is null in add_cow_create function.
Signed-off-by: Dong Xu Wang
---
block/Makefile.objs |1 +
block/add-cow.c | 690 ++
It will introduce a new file format: add-cow.
The add-cow file format makes it possible to perform copy-on-write on top of
a raw disk image. When we know that no backing file clusters remain visible
(e.g. we have streamed the entire image and copied all data from the backing
file), then it is pos
We will re-use qcow2-cache as block layer common cache code,
so change its name and made some changes, define a struct named
BlockTableType, pass BlockTableType and table size parameters to
block cache initialization function.
Signed-off-by: Dong Xu Wang
---
block/Makefile.objs|3 +-
blo
Signed-off-by: Dong Xu Wang
---
qemu-option.c | 285 -
qemu-option.h | 32 ---
2 files changed, 0 insertions(+), 317 deletions(-)
diff --git a/qemu-option.c b/qemu-option.c
index 63f3474..cda8794 100644
--- a/qemu-option.c
+++ b/qemu
qemu_opts_print has no user now, so can re-write the function safely.
qemu_opts_print will be used while using "qemu-img create", it will
produce the same output as previous code.
The behavior of this function has changed:
1. Print every possible option, whether a value has been set or not.
2. O
> > Yup, it's already not too bad. I haven't looked into it in much
> > detail, but I'd like to reduce it even a bit more. In particular, the
> > backup_info field in the BlockDriverState feels wrong to me. In the
> > long term the generic block layer shouldn't know at all what a backup
> > is, and
While id is NULL, qemu_opts_create can not fail, so ignore
errors is fine.
Signed-off-by: Dong Xu Wang
---
qemu-option.c |9 +
qemu-option.h |1 +
2 files changed, 10 insertions(+), 0 deletions(-)
diff --git a/qemu-option.c b/qemu-option.c
index e0131ce..1303188 100644
--- a/qem
Signed-off-by: Dong Xu Wang
---
qemu-option.c | 22 ++
qemu-option.h |1 +
2 files changed, 23 insertions(+), 0 deletions(-)
diff --git a/qemu-option.c b/qemu-option.c
index 1303188..94557cf 100644
--- a/qemu-option.c
+++ b/qemu-option.c
@@ -695,6 +695,28 @@ int qemu_op
These patches fix problems in the handling of freeing QOM/qdev
objects. Together, they fix hot-unplug of USB mass storage devices,
which crashed with an assertion failure.
I'm not 100% sure, but I think we were always leaking the scsi-disk in
pre-QOM days. Now we're freeing it properly, and the
All conditional deallocation can now be done with object_delete.
Remove the @qom_allocated and @glib_allocated fields; replace the latter
with a direct assignment of the @free function pointer.
Signed-off-by: Paolo Bonzini
---
hw/pci.c | 2 +-
hw/qdev-core.h | 5 -
hw/qdev.c | 1
Add an ObjectClass method that is done at object_unparent time. It
should remove any backlinks to the object in the composition tree,
so that object_delete will be able to drop the last reference and
free the object.
Use it for qdev buses.
Signed-off-by: Paolo Bonzini
---
Ping Fan, with
The reference count for embedded objects is always one too low, because
object_initialize_with_type returns with zero references to the object.
This causes premature finalization of the object (or an assertion failure)
after calling object_ref to add an extra reference and object_unref to
remove it
It is not used anymore, and there is no need to make it public.
Signed-off-by: Paolo Bonzini
---
include/qemu/object.h | 9 -
qom/object.c | 2 +-
2 files changed, 1 insertion(+), 10 deletions(-)
diff --git a/include/qemu/object.h b/include/qemu/object.h
index 5ddcb4a..ed1f47f
There's no need to add a space at the end of line.
Moreover, it can make problems in some projects that
store the help output into a file (and run couple of
tests based on that) and have space at EOL forbidden.
Signed-off-by: Michal Privoznik
---
qemu-options.hx |2 +-
1 files changed, 1 ins
Store in the object the freeing function that will be used at deletion
time. This makes it possible to use object_delete on statically-allocated
(embedded) objects. Dually, it makes it possible to use object_unparent
and object_unref without leaking memory, when the lifetime of object
might exten
> Is there a 1:1 relationship between BackupInfo and BackupBlockJob? Then it
> would be nicer to move BlockupInfo fields into BackupBlockJob (which is
> empty right now). Then you also don't need to add fields to BlockDriverState
> because you know that if your blockjob is running you can access
On 23 November 2012 08:52, Michal Privoznik wrote:
> There's no need to add a space at the end of line.
> Moreover, it can make problems in some projects that
> store the help output into a file (and run couple of
> tests based on that) and have space at EOL forbidden.
I have no problem with this
> > My plan was to have something like bs->job->job_type-
> >{before,after}_write.
> >
> >int coroutine_fn (*before_write)(BlockDriverState *bs,
> > int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
> > void **cookie);
> >int coroutine_fn (*after_write)(BlockDriverStat
> > > My plan was to have something like bs->job->job_type-
> > >{before,after}_write.
> > >
> > >int coroutine_fn (*before_write)(BlockDriverState *bs,
> > > int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
> > > void **cookie);
> > >int coroutine_fn (*after_write)(B
Am 23.11.2012 08:38, schrieb Dietmar Maurer:
>> In short, the idea is that you can stick filters on top of a
>> BlockDriverState, so
>> that any read/writes (and possibly more requests, if necessary) are routed
>> through the filter before they are passed to the block driver of this BDS.
>> Filter
Document for add-cow format, the usage and spec of add-cow are introduced.
Signed-off-by: Dong Xu Wang
---
docs/specs/add-cow.txt | 154
1 files changed, 154 insertions(+), 0 deletions(-)
create mode 100644 docs/specs/add-cow.txt
diff --git a/d
This patch will create 4 functions, count_opts_list, append_opts_list,
free_opts_list and print_opts_list, they will used in following commits.
v5->v6):
1) allocate enough space in append_opts_list function.
Signed-off-by: Dong Xu Wang
---
qemu-option.c | 90 ++
Make qed_read_string function to a common interface, so move it to block.c.
Signed-off-by: Dong Xu Wang
---
block.c | 27 +++
block.h |2 ++
block/qed.c | 34 --
3 files changed, 33 insertions(+), 30 deletions(-)
diff --git
We will use path_has_protocol outside block.c, so just make it public.
Signed-off-by: Dong Xu Wang
Reviewed-by: Michael Roth
---
block.c |2 +-
block.h |1 +
2 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/block.c b/block.c
index 7d5d628..6b57611 100644
--- a/block.c
+++
This patch will use qemu-iotests to test add-cow file format.
v15->v16):
1) Rebased on QEMU upstream source tree.
Signed-off-by: Dong Xu Wang
---
tests/qemu-iotests/017 |2 +-
tests/qemu-iotests/020 |2 +-
tests/qemu-iotests/common|6 ++
tests/qemu-iotests/common
Il 23/11/2012 10:05, Dietmar Maurer ha scritto:
My plan was to have something like bs->job->job_type-
{before,after}_write.
int coroutine_fn (*before_write)(BlockDriverState *bs,
int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
void **cookie);
> > BTW, will such filters work with the new virtio-blk-data-plane?
>
> No, virtio-blk-data-plane is a hack and will be slowly rewritten to support
> all
> fancy features.
Ah, good to know ;-) thanks.
Il 23/11/2012 08:42, Dietmar Maurer ha scritto:
>> > My plan was to have something like bs->job->job_type->{before,after}_write.
>> >
>> >int coroutine_fn (*before_write)(BlockDriverState *bs,
>> > int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
>> > void **cookie);
>>
> >> Filters would be implemented as BlockDrivers, i.e. you could
> >> implement
> >> .bdrv_co_write() in a filter to intercept all writes to an image.
> >
> > I am quite unsure if that make things easier.
>
> At least it would make for a much cleaner design compared to putting code
> for every fe
> Actually this was plan B, as a poor-man implementation of the filter
> infrastructure. Plan A was that the block filters would materialize suddenly
> in
> someone's git tree.
OK, so let us summarize the options:
a.) wait untit it materialize suddenly in someone's git tree.
b.) add BlockFilter
> > >> Filters would be implemented as BlockDrivers, i.e. you could
> > >> implement
> > >> .bdrv_co_write() in a filter to intercept all writes to an image.
> > >
> > > I am quite unsure if that make things easier.
> >
> > At least it would make for a much cleaner design compared to putting
> > co
Am 23.11.2012 um 08:02 schrieb Stefan Hajnoczi:
> On Thu, Nov 22, 2012 at 03:29:52PM +0100, Peter Lieven wrote:
>> is anyone aware of a problem with the linux network bridge that in very rare
>> circumstances stops
>> a bridge from sending pakets to a tap device?
>>
>> My problem occurs in conj
Am 23.11.2012 10:05, schrieb Dietmar Maurer:
My plan was to have something like bs->job->job_type-
{before,after}_write.
int coroutine_fn (*before_write)(BlockDriverState *bs,
int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
void **cookie);
>>
From: Viktor Mihajlovski
This enables qemu -cpu ? to return the list of supported CPU models
on s390. Since only the host model is supported at this point in time
this is pretty straight-forward. Further, a validity check for the
requested CPU model was added.
This change is needed to allow libvi
From: Cornelia Huck
Some gcc versions rightly complain about a possibly unitialized rc,
so let's move setting it before the QTAILQ_FOREACH().
Signed-off-by: Cornelia Huck
Signed-off-by: Christian Borntraeger
Signed-off-by: Jens Freimann
---
hw/s390x/event-facility.c | 3 ++-
1 file changed,
Am 23.11.2012 10:31, schrieb Dietmar Maurer:
> Filters would be implemented as BlockDrivers, i.e. you could
> implement
> .bdrv_co_write() in a filter to intercept all writes to an image.
I am quite unsure if that make things easier.
>>>
>>> At least it would make for a much c
When resetting vcpus on s390/kvm we have to clear registers, psw
and prefix as described in the z/Architecture PoP, otherwise a
reboot won't work. IPL PSW and prefix are set later on by the
s390-ipl device reset code.
Signed-off-by: Jens Freimann
---
target-s390x/kvm.c | 26 +
Hi Alex,
a few more s390 patches:
Patch 1 enables the -cpu ? option for s390 CPU models.
Patch 2 implements vcpu reset
Patch 3 takes care of an possibly uninitialized variable in SCLP code
regards
Jens
Cornelia Huck (1):
sclp: Fix uninitialized var in handle_write_event_buf().
Jens Freimann (
On Wed, 2012-11-21 at 10:15 +0100, Kevin Wolf wrote:
> Am 21.11.2012 09:58, schrieb Christian Borntraeger:
> > From: Heinz Graalfs
> >
> > While testing IPL code (booting) for s390x we faced some problems
> > with cache=none on dasds (4k block size) on bdrv_preads with length
> > values != block
On Fri, Nov 23, 2012 at 03:13:07PM +1100, David Gibson wrote:
> On Thu, Nov 22, 2012 at 09:23:03AM +0200, Michael S. Tsirkin wrote:
> > On Thu, Nov 22, 2012 at 01:27:18PM +1100, David Gibson wrote:
> > > On Wed, Nov 21, 2012 at 05:27:37PM +0200, Michael S. Tsirkin wrote:
> > > > On Wed, Nov 21, 201
Kevin Wolf writes:
> Am 23.11.2012 10:05, schrieb Dietmar Maurer:
> My plan was to have something like bs->job->job_type-
> {before,after}_write.
>
>int coroutine_fn (*before_write)(BlockDriverState *bs,
> int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
>>>
On Fri, Nov 23, 2012 at 10:41:21AM +0100, Peter Lieven wrote:
>
> Am 23.11.2012 um 08:02 schrieb Stefan Hajnoczi:
>
> > On Thu, Nov 22, 2012 at 03:29:52PM +0100, Peter Lieven wrote:
> >> is anyone aware of a problem with the linux network bridge that in very
> >> rare circumstances stops
> >> a
Am 23.11.2012 um 12:01 schrieb Michael S. Tsirkin:
> On Fri, Nov 23, 2012 at 10:41:21AM +0100, Peter Lieven wrote:
>>
>> Am 23.11.2012 um 08:02 schrieb Stefan Hajnoczi:
>>
>>> On Thu, Nov 22, 2012 at 03:29:52PM +0100, Peter Lieven wrote:
is anyone aware of a problem with the linux network
We already depend on working __thread support for coroutines, so this
complication here is no longer needed.
Signed-off-by: Jan Kiszka
---
Post-1.3 material.
qemu-thread-win32.c | 24 ++--
1 files changed, 6 insertions(+), 18 deletions(-)
diff --git a/qemu-thread-win32.c
> We already depend on working __thread support for coroutines, so this
> complication here is no longer needed.
>
> Signed-off-by: Jan Kiszka
Reviewed-by: Paolo Bonzini
> ---
>
> Post-1.3 material.
>
> qemu-thread-win32.c | 24 ++--
> 1 files changed, 6 insertions(+)
On Fri, 23 Nov 2012 07:50:49 +0100
Stefan Hajnoczi wrote:
> On Thu, Nov 22, 2012 at 10:34:11AM -0200, Luiz Capitulino wrote:
> > On Wed, 21 Nov 2012 13:12:53 +0100
> > Jan Kiszka wrote:
> >
> > > On 2012-11-20 16:14, Markus Armbruster wrote:
> > > > Cc'ing Marcelo & Jan to speed up patch applic
To avoid it getting lost in qemu forest.
On Thu, 22 Nov 2012 15:50:50 +0100
fred.kon...@greensocs.com wrote:
> +/* Create a virtio bus. */
> +VirtioBus *virtio_bus_new(DeviceState *host, const VirtioBusInfo *info)
> +{
> +/*
> + * This is needed, as we want to have different names for each
> virtio-bus.
> + * If we don't d
Am 21.11.2012 18:17, schrieb Pavel Hrdina:
> If you have a guest with a media in the optical drive and you change the media
> the windows guest cannot properly recognize this media change.
>
> Windows needs to detect the sense "NOT_READY with ASC_MEDIUM_NOT_PRESENT"
> before we send the sense "UNI
On Thu, 22 Nov 2012 15:50:51 +0100
fred.kon...@greensocs.com wrote:
> +static void virtiopci_class_init(ObjectClass *oc, void *data)
> +{
> +DeviceClass *dc = DEVICE_CLASS(oc);
> +PCIDeviceClass *pc = PCI_DEVICE_CLASS(oc);
> +
> +pc->init = virtiopci_qdev_init;
> +pc->exit = virti
On Thu, Nov 22, 2012 at 03:50:50PM +0100, fred.kon...@greensocs.com wrote:
> +/* Bind the VirtIODevice to the VirtioBus. */
> +void virtio_bus_bind_device(VirtioBus *bus)
> +{
> +BusState *qbus = BUS(bus);
> +assert(bus != NULL);
> +assert(bus->vdev != NULL);
> +virtio_bind_device(b
On Thu, Nov 22, 2012 at 03:50:51PM +0100, fred.kon...@greensocs.com wrote:
> +static const struct VirtioBusInfo virtio_bus_info = {
> +.init_cb = virtio_pci_init_cb,
> +.exit_cb = virtio_pci_exit_cb,
> +
> +.virtio_bindings = {
Eventually VirtIOBindings can probably be inlined into Vir
On Thu, Nov 22, 2012 at 03:50:52PM +0100, fred.kon...@greensocs.com wrote:
> @@ -656,3 +658,83 @@ void virtio_blk_exit(VirtIODevice *vdev)
> blockdev_mark_auto_del(s->bs);
> virtio_cleanup(vdev);
> }
> +
> +static int virtio_blk_qdev_init(DeviceState *qdev)
> +{
> +VirtIOBLKState *s
On 23 November 2012 12:29, Stefan Hajnoczi wrote:
> On Thu, Nov 22, 2012 at 03:50:51PM +0100, fred.kon...@greensocs.com wrote:
>> +static const struct VirtioBusInfo virtio_bus_info = {
>> +.init_cb = virtio_pci_init_cb,
>> +.exit_cb = virtio_pci_exit_cb,
>> +
>> +.virtio_bindings = {
>
From: Viktor Mihajlovski
This enables qemu -cpu ? to return the list of supported CPU models
on s390. Since only the host model is supported at this point in time
this is pretty straight-forward. Further, a validity check for the
requested CPU model was added.
This change is needed to allow libvi
On Thu, Nov 22, 2012 at 03:50:49PM +0100, fred.kon...@greensocs.com wrote:
> From: KONRAD Frederic
> I made the changes you suggest in the last RFC.
>
> There are still two issues with the command line :
>
> * When I use ./qemu* -device virtio-blk -device virtio-pci
> It is said that
The following changes since commit 1ccbc2851282564308f790753d7158487b6af8e2:
qemu-sockets: Fix parsing of the inet option 'to'. (2012-11-21 12:07:59 +0400)
are available in the git repository at:
git://git.kernel.org/pub/scm/virt/kvm/qemu-kvm.git uq/master
Bruce Rogers (1):
Legacy qemu
From: Bruce Rogers
The options no-kvm, no-kvm-pit, no-kvm-pit-reinjection, and no-kvm-irqchip
should be marked as having no argument.
Signed-off-by: Bruce Rogers
Reviewed-by: Jan Kiszka
Signed-off-by: Marcelo Tosatti
---
qemu-options.hx |8
1 files changed, 4 insertions(+), 4 de
On Fri, Nov 23, 2012 at 12:53:23PM +0200, Michael S. Tsirkin wrote:
> On Fri, Nov 23, 2012 at 03:13:07PM +1100, David Gibson wrote:
> > On Thu, Nov 22, 2012 at 09:23:03AM +0200, Michael S. Tsirkin wrote:
> > > On Thu, Nov 22, 2012 at 01:27:18PM +1100, David Gibson wrote:
> > > > On Wed, Nov 21, 201
Hi all,
I have just updated the qemu-xen tree (qemu-upstream-unstable.git) that
we use with xen-unstable for QEMU development to QEMU 1.3.0-rc0 (plus a
couple of commits).
After the updated qemu-xen tree manages to pass the required automated
tests, I'll send a patch to flip the switch in libxl to
On 23.11.2012, at 13:36, Jens Freimann wrote:
> From: Viktor Mihajlovski
>
> This enables qemu -cpu ? to return the list of supported CPU models
> on s390. Since only the host model is supported at this point in time
> this is pretty straight-forward. Further, a validity check for the
> request
On 23.11.2012, at 11:18, Jens Freimann wrote:
> When resetting vcpus on s390/kvm we have to clear registers, psw
> and prefix as described in the z/Architecture PoP, otherwise a
> reboot won't work. IPL PSW and prefix are set later on by the
> s390-ipl device reset code.
>
> Signed-off-by: Jens
On Fri, Nov 23, 2012 at 11:59:51PM +1100, David Gibson wrote:
> > Look, even if solution using a required property is less elegant for CLI
> > use, it will work, won't it?
> > So how about we merge it so that things work, and then we can discuss a
> > patch on top that auto-generates this property?
On 23.11.2012, at 11:18, Jens Freimann wrote:
> From: Cornelia Huck
>
> Some gcc versions rightly complain about a possibly unitialized rc,
> so let's move setting it before the QTAILQ_FOREACH().
>
> Signed-off-by: Cornelia Huck
> Signed-off-by: Christian Borntraeger
> Signed-off-by: Jens Fr
On 23.11.2012, at 14:44, Michael S. Tsirkin wrote:
> On Fri, Nov 23, 2012 at 11:59:51PM +1100, David Gibson wrote:
>>> Look, even if solution using a required property is less elegant for CLI
>>> use, it will work, won't it?
>>> So how about we merge it so that things work, and then we can discus
On Fri, Nov 23, 2012 at 02:44:15PM +0100, Alexander Graf wrote:
>
> On 23.11.2012, at 14:44, Michael S. Tsirkin wrote:
>
> > On Fri, Nov 23, 2012 at 11:59:51PM +1100, David Gibson wrote:
> >>> Look, even if solution using a required property is less elegant for CLI
> >>> use, it will work, won't
On 23.11.2012, at 15:01, Michael S. Tsirkin wrote:
> On Fri, Nov 23, 2012 at 02:44:15PM +0100, Alexander Graf wrote:
>>
>> On 23.11.2012, at 14:44, Michael S. Tsirkin wrote:
>>
>>> On Fri, Nov 23, 2012 at 11:59:51PM +1100, David Gibson wrote:
> Look, even if solution using a required proper
performance is also reduced when snapshot exist. (like if they are no
preallocated metadatas)
see initial git commit
http://git.qemu.org/?p=qemu.git;a=commit;h=a35e1c177debb01240243bd656caca302410d38c
"qcow2: Metadata preallocation
This introduces a qemu-img create option for qcow2 which allow
Am 23.11.2012 15:03, schrieb Alexandre DERUMIER:
> performance is also reduced when snapshot exist. (like if they are no
> preallocated metadatas)
Preallocation doesn't matter much these days.
> see initial git commit
>
> http://git.qemu.org/?p=qemu.git;a=commit;h=a35e1c177debb01240243bd656caca
On Thu, Nov 22, 2012 at 10:07 AM, Stefan Priebe wrote:
> diff --git a/block/rbd.c b/block/rbd.c
> index 5a0f79f..0384c6c 100644
> --- a/block/rbd.c
> +++ b/block/rbd.c
> @@ -69,7 +69,7 @@ typedef enum {
> typedef struct RBDAIOCB {
> BlockDriverAIOCB common;
> QEMUBH *bh;
> -int ret;
On 23/11/2012 13:08, Cornelia Huck wrote:
On Thu, 22 Nov 2012 15:50:50 +0100
fred.kon...@greensocs.com wrote:
+/* Create a virtio bus. */
+VirtioBus *virtio_bus_new(DeviceState *host, const VirtioBusInfo *info)
+{
+/*
+ * This is needed, as we want to have different names for each vir
On 23 November 2012 14:11, Stefan Hajnoczi wrote:
> On Thu, Nov 22, 2012 at 10:07 AM, Stefan Priebe wrote:
>> diff --git a/block/rbd.c b/block/rbd.c
>> index 5a0f79f..0384c6c 100644
>> --- a/block/rbd.c
>> +++ b/block/rbd.c
>> @@ -69,7 +69,7 @@ typedef enum {
>> typedef struct RBDAIOCB {
>>
On Fri, Nov 23, 2012 at 03:03:24PM +0100, Alexander Graf wrote:
>
> On 23.11.2012, at 15:01, Michael S. Tsirkin wrote:
>
> > On Fri, Nov 23, 2012 at 02:44:15PM +0100, Alexander Graf wrote:
> >>
> >> On 23.11.2012, at 14:44, Michael S. Tsirkin wrote:
> >>
> >>> On Fri, Nov 23, 2012 at 11:59:51PM
>>Preallocation doesn't matter much these days.
I have done benchmark through nfs on a netapp san, (with directio),
I got 300 iops without preallocation, and 6000 iops with preallocation
- Mail original -
De: "Kevin Wolf"
À: "Alexandre DERUMIER"
Cc: "Dietmar Maurer" , qemu-deve
On Thu, Nov 22, 2012 at 9:02 PM, Christian Borntraeger
wrote:
> here are two followup patches.
>
>
> Christian Borntraeger (2):
> block: simply default_drive
> block: clarify comment about IF_IDE = 0
>
> blockdev.h |6 +-
> vl.c | 20 ++--
> 2 files changed, 11
On 23/11/2012 13:23, Stefan Hajnoczi wrote:
On Thu, Nov 22, 2012 at 03:50:50PM +0100, fred.kon...@greensocs.com wrote:
+/* Bind the VirtIODevice to the VirtioBus. */
+void virtio_bus_bind_device(VirtioBus *bus)
+{
+BusState *qbus = BUS(bus);
+assert(bus != NULL);
+assert(bus->vdev !=
On 23/11/2012 13:34, Peter Maydell wrote:
On 23 November 2012 12:29, Stefan Hajnoczi wrote:
On Thu, Nov 22, 2012 at 03:50:51PM +0100, fred.kon...@greensocs.com wrote:
+static const struct VirtioBusInfo virtio_bus_info = {
+.init_cb = virtio_pci_init_cb,
+.exit_cb = virtio_pci_exit_cb,
On 23 November 2012 14:23, Konrad Frederic wrote:
> On 23/11/2012 13:34, Peter Maydell wrote:
>> On 23 November 2012 12:29, Stefan Hajnoczi wrote:
>>> Eventually VirtIOBindings can probably be inlined into VirtioBusInfo. I
>>> don't see a need for separate structs.
>>
>> I agree. It might (or mi
On 23.11.2012, at 15:18, Michael S. Tsirkin wrote:
> On Fri, Nov 23, 2012 at 03:03:24PM +0100, Alexander Graf wrote:
>>
>> On 23.11.2012, at 15:01, Michael S. Tsirkin wrote:
>>
>>> On Fri, Nov 23, 2012 at 02:44:15PM +0100, Alexander Graf wrote:
On 23.11.2012, at 14:44, Michael S. Tsi
On 23/11/2012 13:38, Stefan Hajnoczi wrote:
On Thu, Nov 22, 2012 at 03:50:49PM +0100, fred.kon...@greensocs.com wrote:
From: KONRAD Frederic
I made the changes you suggest in the last RFC.
There are still two issues with the command line :
* When I use ./qemu* -device virtio-blk -device v
On 23/11/12 14:40, Alexander Graf wrote:
>
> On 23.11.2012, at 11:18, Jens Freimann wrote:
>
>> When resetting vcpus on s390/kvm we have to clear registers, psw
>> and prefix as described in the z/Architecture PoP, otherwise a
>> reboot won't work. IPL PSW and prefix are set later on by the
>> s3
On 23/11/2012 15:26, Peter Maydell wrote:
On 23 November 2012 14:23, Konrad Frederic wrote:
On 23/11/2012 13:34, Peter Maydell wrote:
On 23 November 2012 12:29, Stefan Hajnoczi wrote:
Eventually VirtIOBindings can probably be inlined into VirtioBusInfo. I
don't see a need for separate stru
On 23.11.2012, at 15:32, Christian Borntraeger wrote:
> On 23/11/12 14:40, Alexander Graf wrote:
>>
>> On 23.11.2012, at 11:18, Jens Freimann wrote:
>>
>>> When resetting vcpus on s390/kvm we have to clear registers, psw
>>> and prefix as described in the z/Architecture PoP, otherwise a
>>> reb
On Fri, 23 Nov 2012 15:12:45 +0100
Konrad Frederic wrote:
> On 23/11/2012 13:08, Cornelia Huck wrote:
> > On Thu, 22 Nov 2012 15:50:50 +0100
> > fred.kon...@greensocs.com wrote:
> >
> >
> >> +/* Create a virtio bus. */
> >> +VirtioBus *virtio_bus_new(DeviceState *host, const VirtioBusInfo *info)
Hi,
i'm not a ceph or inktank guy. I can't made any decision on what to
change. At least right now you'll see failing I/O's in your guest, when
you discard whole disks.
I could fix this for me with int64 and with ssize_t. So if i should
resend another patch i need a concrete advise how to pa
Am 23.11.2012 15:17, schrieb Alexandre DERUMIER:
>>> Preallocation doesn't matter much these days.
>
> I have done benchmark through nfs on a netapp san, (with directio),
>
> I got 300 iops without preallocation, and 6000 iops with preallocation
This doesn't sound right, even for writethrou
The repeated calls to WaitForMultipleObjects may cause a livelock in aio_poll,
where no progress is made on bottom halves. This patch matches the behavior
of the POSIX code.
Signed-off-by: Paolo Bonzini
---
Found by the new unit tests.
aio-win32.c | 5 -
1 file changed, 4 insertion
On 23 November 2012 10:18, Jens Freimann wrote:
> +/* The initial reset call is needed here to reset in-kernel
> + * vcpu data that we can't access directly from QEMU.
If there's in-kernel vcpu data we can't access from QEMU,
doesn't this cause problems for migration?
> Before
> + *
Christian Borntraeger writes:
> here are two followup patches.
Thanks!
Stefan, PATCH 1/2 could be squashed into "Support default block
interfaces per QEMUMachine". Your choice.
Reviewed-by: Markus Armbruster
Adjust the conditional which guards the implementation of
cpu_get_real_ticks() via RDTSC, so that we don't try to use it
on x86 CPUs which don't implement RDTSC. Instead we will fall
back to the no-cycle-counter-available default implementation.
Reported-by: Yurij Popov
Signed-off-by: Peter Mayd
On 23.11.2012, at 16:02, Peter Maydell wrote:
> On 23 November 2012 10:18, Jens Freimann wrote:
>> +/* The initial reset call is needed here to reset in-kernel
>> + * vcpu data that we can't access directly from QEMU.
>
> If there's in-kernel vcpu data we can't access from QEMU,
> doesn
As requested. :) The tests pass on both Linux and Wine.
I'm not sure if it makes sense to include these in 1.3, but the idea
has some worth.
Paolo Bonzini (2):
tests: add AioContext unit tests
tests: add thread pool unit tests
tests/Makefile | 4 +
tests/test-aio.c | 66
Signed-off-by: Paolo Bonzini
---
tests/Makefile | 2 +
tests/test-thread-pool.c | 213 +++
2 files changed, 215 insertions(+)
create mode 100644 tests/test-thread-pool.c
diff --git a/tests/Makefile b/tests/Makefile
index 61cbe3b..b60f0fb 1
Signed-off-by: Paolo Bonzini
---
tests/Makefile | 2 +
tests/test-aio.c | 667 +++
2 files changed, 669 insertions(+)
create mode 100644 tests/test-aio.c
diff --git a/tests/Makefile b/tests/Makefile
index ca680e5..61cbe3b 100644
--- a/test
Il 23/11/2012 16:12, Peter Maydell ha scritto:
> Adjust the conditional which guards the implementation of
> cpu_get_real_ticks() via RDTSC, so that we don't try to use it
> on x86 CPUs which don't implement RDTSC. Instead we will fall
> back to the no-cycle-counter-available default implementatio
On 23 November 2012 15:15, Paolo Bonzini wrote:
> Il 23/11/2012 16:12, Peter Maydell ha scritto:
>> Adjust the conditional which guards the implementation of
>>
>> -#elif defined(__i386__)
>> +#elif defined(__i586__)
>>
>> static inline int64_t cpu_get_real_ticks(void)
>> {
>>
>
> You should at
Am 23.11.2012 11:18, schrieb Jens Freimann:
> From: Viktor Mihajlovski
>
> This enables qemu -cpu ? to return the list of supported CPU models
> on s390. Since only the host model is supported at this point in time
> this is pretty straight-forward. Further, a validity check for the
> requested C
Paolo Bonzini writes:
> Il 22/11/2012 15:02, Markus Armbruster ha scritto:
>> Watch this:
>>
>> (qemu) drive_add 0 if=scsi
>> Segmentation fault (core dumped)
>>
>> Broken in commit 0d936928 "qdev: Convert busses to QEMU Object Model".
>> Before:
>>
>> (qemu) drive_add 0 if=scsi
>>
Am 23.11.2012 08:26, schrieb Dietmar Maurer:
> qcow2 snapshot on newly created files are fast:
>
> # qemu-img create -f qcow2 test.img 200G
> # time qemu-img snapshot -c snap1 test.img
> real 0m0.014s
>
> but if metadata is allocated it gets very slow:
>
> # qemu-img create -f qcow2 -o "p
Peter Maydell wrote:
> On 23 November 2012 15:15, Paolo Bonzini wrote:
> > Il 23/11/2012 16:12, Peter Maydell ha scritto:
> >> Adjust the conditional which guards the implementation of
> >>
> >> -#elif defined(__i386__)
> >> +#elif defined(__i586__)
> >>
> >> static inline int64_t cpu_get_real_ti
1 - 100 of 161 matches
Mail list logo
