I'm the only one whom this patch breaks compilation? Is a
cross-compiled qemu, but the make had run smoothly until now, and a
git bisect has driven me here.
My configuration is a bit strange:
configure:
PKG_CONFIG_PATH=/usr/powerpc-linux-gnu/lib/pkgconfig ./configure
--prefix=/mnt/DATA/DARCO/alie
On Tue, Jun 26, 2012 at 5:37 AM, Jia Liu wrote:
> Hi Max,
>
> On Mon, Jun 25, 2012 at 5:00 PM, Max Filippov wrote:
>> On Mon, Jun 25, 2012 at 6:50 AM, Jia Liu wrote:
>>> Hi Max,
>>>
>>> On Thu, Jun 21, 2012 at 6:24 PM, Max Filippov wrote:
On Thu, Jun 21, 2012 at 6:58 AM, Jia Liu wrote:
>>
On 2012-06-26 00:02, Anthony Liguori wrote:
> On 06/25/2012 03:26 PM, Jan Kiszka wrote:
>> On 2012-06-25 18:55, Luiz Capitulino wrote:
>>> Allow for disabling memory merge support (KSM on Linux), which is
>>> enabled by default otherwise.
>>
>> -machine mem_merge=on|off?
>
> Ack.
>
> But any reas
On 06/25/2012 06:14 PM, Michael S. Tsirkin wrote:
On Mon, Jun 25, 2012 at 05:41:17PM +0800, Jason Wang wrote:
Device specific irq optimizations such as irq affinity may be used by virtio
drivers. So this patch introduce a new method to get the irq of a specific
virtqueue.
After this patch, virt
target-ppc: Fix 2nd parameter for tcg_gen_shri_tl (2012-06-24 22:52:11 +0200)
are available in the git repository at:
git://developer.petalogix.com/public/qemu.git third-party/igor-sdhci.next
Igor Mitsyanko (2):
hw: introduce standard SD host controller
exynos4210: Added SD host c
Hi,
VFIO has been kicking around for well over a year now and has been
posted numerous times for review. The pre-requirements are finally
available in linux-next (or will be in the 20120626 build) so I'd like
to request a new branch be included in linux-next with a goal of being
accepted into v3.
Signed-off-by: Peter A. G. Crosthwaite
---
MAINTAINERS | 11 +++
1 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index d544a9c..e19f491 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -482,6 +482,17 @@ S: Supported
F: hw/virtio-serial*
F: hw/vir
Agreed between myself and Alex:
http://lists.nongnu.org/archive/html/qemu-devel/2012-06/msg03561.html
Signed-off-by: Peter A. G. Crosthwaite
Acked-by: Alexander Graf
---
MAINTAINERS |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index e19
Signed-off-by: Peter A. G. Crosthwaite
---
MAINTAINERS |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index b45f075..d544a9c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -311,6 +311,11 @@ M: Edgar E. Iglesias
S: Maintained
F: hw/petalogi
Set some missing maintainer ships. Patch 1 is the Petalogix ML605 machine model
(me). Patch 2 is the Xilinx EDK device suite (me + Edgar). Patch 3 is the
device tree subsystem (me + Alex).
Changed since v1:
Added Alexs ack for p3.
Peter A. G. Crosthwaite (3):
MAINTAINERS: Add Petalogix ml605
On 26.06.2012, at 04:24, 陳韋任 (Wei-Ren Chen) wrote:
> On Mon, Jun 25, 2012 at 07:32:39PM +0200, Alexander Graf wrote:
>> When forwarding a segmentation fault into the guest process, we were passing
>> the host's address directly into the guest process's signal descriptor.
>>
>> That obviously con
On 25.06.2012 16:00, Andreas Färber wrote:
Am 25.06.2012 13:46, schrieb Evgeny Voevodin:
On 25.06.2012 13:24, Andreas Färber wrote:
Am 25.06.2012 09:55, schrieb Oleg Ogurtsov:
Signed-off-by: Oleg Ogurtsov
---
hw/arm/Makefile.objs |1 +
hw/exynos4210.c |8 +
hw/exynos4210_rt
On Mon, Jun 25, 2012 at 07:32:39PM +0200, Alexander Graf wrote:
> When forwarding a segmentation fault into the guest process, we were passing
> the host's address directly into the guest process's signal descriptor.
>
> That obviously confused the guest process, since it didn't know what to make
On 06/25/2012 08:28 PM, Peter Crosthwaite wrote:
On Tue, Jun 26, 2012 at 12:25 AM, Anthony Liguori wrote:
On 06/25/2012 05:02 AM, Andreas Färber wrote:
Hi Peter,
Am 25.06.2012 11:34, schrieb Peter Crosthwaite:
I have a platform (Xilinx Zynq) that has a USB EHCI controller that
attaches dir
Hi Max,
On Mon, Jun 25, 2012 at 5:00 PM, Max Filippov wrote:
> On Mon, Jun 25, 2012 at 6:50 AM, Jia Liu wrote:
>> Hi Max,
>>
>> On Thu, Jun 21, 2012 at 6:24 PM, Max Filippov wrote:
>>> On Thu, Jun 21, 2012 at 6:58 AM, Jia Liu wrote:
Add OpenRISC instruction tanslation routines.
On Tue, Jun 26, 2012 at 12:25 AM, Anthony Liguori wrote:
> On 06/25/2012 05:02 AM, Andreas Färber wrote:
>>
>> Hi Peter,
>>
>> Am 25.06.2012 11:34, schrieb Peter Crosthwaite:
>>>
>>> I have a platform (Xilinx Zynq) that has a USB EHCI controller that
>>> attaches directly to the system bus and not
-Original Message-
From: Eduardo Habkost [mailto:ehabk...@redhat.com]
Sent: Monday, June 25, 2012 1:01 PM
To: Vinod, Chegu
Cc: Hada, Craig M; Hull, Jim; qemu-devel@nongnu.org; k...@vger.kernel.org
Subject: Re: [Qemu-devel] [PATCH v2] Fixes related to processing of qemu's
-numa option
J
On 06/25/2012 05:46 PM, Anthony Liguori wrote:
From: Amit Shah
The Linux kernel already has a virtio-rng driver, this is the device
implementation.
When the guest asks for entropy from the virtio hwrng, it puts a buffer
in the vq. We then put entropy into that buffer, and push it back to
the g
From: Amit Shah
The Linux kernel already has a virtio-rng driver, this is the device
implementation.
When the guest asks for entropy from the virtio hwrng, it puts a buffer
in the vq. We then put entropy into that buffer, and push it back to
the guest.
Signed-off-by: Amit Shah
Signed-off-by:
This is the backend used by devices that need to request entropy.
Signed-off-by: Anthony Liguori
---
Makefile.objs |2 +
backends/Makefile.objs |1 +
backends/rng.c | 93
include/qemu/rng.h | 93 +++
This series depends on my QOM -object series that I just posted.
In Amit's thread on virtio-rng, danpb mentioned that we really ought to have a
proper RNG backend infrastructure and of course he's correct on that.
Now that we have QOM, I wanted to demonstrate how we can use QOM to construct a
com
This backend talks EGD to a CharDriverState. A typical way to invoke this would
be:
qemu -chardev socket,host=localhost,port=1024,id=chr0 \
-object rng-egd,chardev=chr0,id=egd0 \
-device virtio-rng-pci,rng=egd0
Signed-off-by: Anthony Liguori
---
backends/Makefile.objs |2 +-
back
The filename can be overridden but it expects a non-blocking source of entropy.
A typical invocation would be:
qemu -object rng-urandom,id=rng0 -device virtio-rng-pci,rng=rng0
Signed-off-by: Anthony Liguori
---
backends/Makefile.objs |2 +-
backends/rng-urandom.c | 144
This will create a new QOM object in the '/objects' path. Note that properties
are set in order which allows for simple objects to be initialized entirely
with this option and then realized.
This option is roughly equivalent to -device but for things that are not
devices.
Signed-off-by: Anthony
As part of implementing the virtio-rng backend support, I decided it was best
to split off the generic part of the series to make review easier.
This series does a few things. It adds a realized property to DeviceState.
It does so in the least invasive way possible and I think this is the right w
Signed-off-by: Anthony Liguori
---
hw/qdev.c | 36 +++-
1 files changed, 35 insertions(+), 1 deletions(-)
diff --git a/hw/qdev.c b/hw/qdev.c
index a6c4c02..d305128 100644
--- a/hw/qdev.c
+++ b/hw/qdev.c
@@ -101,7 +101,6 @@ static void bus_add_child(BusState *bus
In order to create qdev objects via -late-object, we almost always have to
specify the parent_bus which is usually created during machine init. Until we
properly support two stage init, introduce a -late-object option that allows for
creation of objects post-machine init.
Signed-off-by: Anthony L
Signed-off-by: Anthony Liguori
---
v1 -> v2
- Fix whitespace (Andreas Faerber)
---
include/qemu/object.h | 16 ++
qom/object.c | 56 +
2 files changed, 72 insertions(+), 0 deletions(-)
diff --git a/include/qemu/object.h b/
On 06/25/2012 11:55 AM, Luiz Capitulino wrote:
Allow to enable/disable memory merging during run-time.
This is implemented by extending the qemu_set_mem_merge() function.
To test on HMP:
(qemu) set_memory_merge on
(qemu) set_memory_merge off
Signed-off-by: Luiz Capitulino
Is there a r
On 06/25/2012 03:26 PM, Jan Kiszka wrote:
On 2012-06-25 18:55, Luiz Capitulino wrote:
Allow for disabling memory merge support (KSM on Linux), which is
enabled by default otherwise.
-machine mem_merge=on|off?
Ack.
But any reason not to call it ksm=on|off?
Regards,
Anthony Liguori
Jan
On 06/25/2012 03:47 PM, Andreas Färber wrote:
Am 25.06.2012 20:35, schrieb Anthony Liguori:
On 06/25/2012 11:32 AM, Andreas Färber wrote:
Am 25.06.2012 18:09, schrieb Anthony Liguori:
diff --git a/qom/object.c b/qom/object.c
index 00bb3b0..1357397 100644
--- a/qom/object.c
+++ b/qom/object.c
@
QEMU emulator version 1.1.0, Copyright (c) 2003-2008 Fabrice Bellard
parameters used: qemu-system-arm -M versatilepb -m 256 -net nic -net
user,hostfwd=tcp::40022-:22 -nographic -hda debian-6-armel.img -kernel
vmlinuz-2.6.32-5-versatile -initrd initrd.img-2.6.32-5-versatile -append
"root=/dev/s
Am 25.06.2012 20:35, schrieb Anthony Liguori:
> On 06/25/2012 11:32 AM, Andreas Färber wrote:
>> Am 25.06.2012 18:09, schrieb Anthony Liguori:
>>> diff --git a/qom/object.c b/qom/object.c
>>> index 00bb3b0..1357397 100644
>>> --- a/qom/object.c
>>> +++ b/qom/object.c
>>> @@ -1232,6 +1232,62 @@ void
On Mon, 25 Jun 2012 22:26:58 +0200
Jan Kiszka wrote:
> On 2012-06-25 18:55, Luiz Capitulino wrote:
> > Allow for disabling memory merge support (KSM on Linux), which is
> > enabled by default otherwise.
>
> -machine mem_merge=on|off?
That's possible. But if we do this, then I think that the set
On 25 June 2012 21:26, malc wrote:
> The field32 and field64 are inconsistent w.r.t. grouping, while both are
> correct, one has to go through precedence list mentally in the latter
> case.
Yeah, unintentional and I agree that if you have to think about
operator precedence there weren't enough br
On 2012-06-25 17:40, Alex Williamson wrote:
> Anyone using these functions has to be prepared that irqchip
> support may not be present. It shouldn't be up to the core
> code to determine whether this is a fatal error. Currently
> code written as:
>
> virq = kvm_irqchip_add_msi_route(...)
> if (
On 2012-06-25 18:55, Luiz Capitulino wrote:
> Allow for disabling memory merge support (KSM on Linux), which is
> enabled by default otherwise.
-machine mem_merge=on|off?
Jan
>
> Signed-off-by: Luiz Capitulino
> ---
> cpu-all.h | 1 +
> exec-obsolete.h | 1 +
> exec.c | 25 ++
On Mon, 25 Jun 2012, Peter Maydell wrote:
> Add field32() and field64() functions which extract a particular
> bit field from a word and return it. Based on an idea by Jia Liu.
>
> Suggested-by: Jia Liu
> Signed-off-by: Peter Maydell
> ---
[..snip..]
> +static inline uint64_t field64(uint64_t
Just found another issue:
On Wed, Jun 20, 2012 at 05:33:29PM -0300, Eduardo Habkost wrote:
[...]
> > @@ -970,27 +974,24 @@ static void numa_add(const char *optarg)
> > }
> > node_mem[nodenr] = sval;
> > }
> > -if (get_param_value(option, 128, "cpus", opta
On Thu, 21 Jun 2012 13:42:19 +0100
"Daniel P. Berrange" wrote:
[...]
> > However, we'd change how we use 'desc' and our error classes. 'desc' would
> > become a string which is filled by a printf-like function (see section 2)
> > and
>
> Good, using a printf-like string for desc is the big cha
Add field32() and field64() functions which extract a particular
bit field from a word and return it. Based on an idea by Jia Liu.
Suggested-by: Jia Liu
Signed-off-by: Peter Maydell
---
Jia Liu had a function like this in the OpenRISC support patchset;
this implementation borrows the API but has
Hans de Goede wrote:
Hi,
On 06/25/2012 06:17 PM, Erik Rull wrote:
Hi all,
is there a way to deny passing certain devices to the guest? I want to
deny two devices to get assigned to the guest, all others on all ports
are allowed, ony two each with a defined pid/vid should not get routed.
Provid
Thank you guys for the reply :)
One last question,
my device is like a physical console with buttons and indicators.
I saw the android emulator also draws the keyboard using a modified
version of the qemu.
Is there some kind of "presentation" layer, or the just modified the
window size and add it
On 06/25/2012 11:32 AM, Andreas Färber wrote:
Am 25.06.2012 18:09, schrieb Anthony Liguori:
Signed-off-by: Anthony Liguori
---
include/qemu/object.h | 16 ++
qom/object.c | 56 +
2 files changed, 72 insertions(+), 0 del
On 06/25/2012 10:40 AM, Andreas Färber wrote:
> Am 25.06.2012 17:52, schrieb Peter Maydell:
>> Define a set of TARGET_PRI*PHYS format specifier macros for working
>> with target_phys_addr_t types. These follow the standard pattern
>> for such macros, and are more flexible than TARGET_FMT_plx, which
Hi,
On 06/25/2012 06:17 PM, Erik Rull wrote:
Hi all,
is there a way to deny passing certain devices to the guest? I want to deny two
devices to get assigned to the guest, all others on all ports are allowed, ony
two each with a defined pid/vid should not get routed.
Providing a positive list
Hi
Please send in any agenda items you are interested in covering.
jbaron suggested: q35, what are the mergin dependences?
Later, Juan.
On 25 June 2012 18:32, Alexander Graf wrote:
> When forwarding a segmentation fault into the guest process, we were passing
> the host's address directly into the guest process's signal descriptor.
>
> That obviously confused the guest process, since it didn't know what to make
> of the (usually 3
When forwarding a segmentation fault into the guest process, we were passing
the host's address directly into the guest process's signal descriptor.
That obviously confused the guest process, since it didn't know what to make
of the (usually 32-bit truncated) address. Passing in g2h(address) makes
On 06/25/2012 11:26 AM, Kevin Wolf wrote:
Am 25.06.2012 16:34, schrieb Eric Blake:
Also, I noticed in the fnctl man page that F_SETFL: "On Linux this
command can change only the O_APPEND, O_ASYNC, O_DIRECT, O_NOATIME, and
O_NONBLOCK flags." So I'll only set/unset these flags.
O_NDELAY is t
Today, QEMU enables KSM by default if the (Linux) host supports it. However,
it may be desirable by mngt apps to not enable it by default and also to be
able to enable or disable KSM at run-time.
This series allows this by adding the -disable-mem-merge command-line option
and the set-memory-merge
Allow to enable/disable memory merging during run-time.
This is implemented by extending the qemu_set_mem_merge() function.
To test on HMP:
(qemu) set_memory_merge on
(qemu) set_memory_merge off
Signed-off-by: Luiz Capitulino
---
exec.c | 15 +++
hmp-commands.hx | 1
Will be used by the next commit.
Signed-off-by: Luiz Capitulino
---
cpu-all.h | 1 +
exec.c| 3 +++
2 files changed, 4 insertions(+)
diff --git a/cpu-all.h b/cpu-all.h
index 429c1c3..9a36f7e 100644
--- a/cpu-all.h
+++ b/cpu-all.h
@@ -469,6 +469,7 @@ extern ram_addr_t ram_size;
/* RAM is
Allow for disabling memory merge support (KSM on Linux), which is
enabled by default otherwise.
Signed-off-by: Luiz Capitulino
---
cpu-all.h | 1 +
exec-obsolete.h | 1 +
exec.c | 25 +
memory.c| 5 +
memory.h| 5 +
qemu-options.
Am 25.06.2012 17:52, schrieb Peter Maydell:
> Now we have TARGET_PRI*PHYS for printing target_phys_addr_t values,
> we can use them in monitor.c rather than having duplicate code
> in two arms of a TARGET_PHYS_ADDR_BITS ifdef.
>
> Signed-off-by: Peter Maydell
Reviewed-by: Andreas Färber
/-F
-
Am 25.06.2012 17:52, schrieb Peter Maydell:
> Switch a format string from %x to TARGET_PRIxPHYS so that it will
> continue to work even if target_phys_addr_t is changed
> to 64 bits in the future.
>
> Signed-off-by: Peter Maydell
Reviewed-by: Andreas Färber
/-F
--
SUSE LINUX Products GmbH, M
Signed-off-by: Anthony Liguori
---
backends/Makefile.objs |2 +-
backends/devrng.c | 139
2 files changed, 140 insertions(+), 1 deletions(-)
create mode 100644 backends/devrng.c
diff --git a/backends/Makefile.objs b/backends/Makefile.ob
Am 25.06.2012 17:52, schrieb Peter Maydell:
> Use the new TARGET_PRIxPHYS macro to avoid the need to define an
> OMAP_FMT_plx macro whose expansion depends directly on
> TARGET_PHYS_ADDR_BITS.
>
> Signed-off-by: Peter Maydell
Acked-by: Andreas Färber
/-F
--
SUSE LINUX Products GmbH, Maxfelds
Am 25.06.2012 17:52, schrieb Peter Maydell:
> Define a set of TARGET_PRI*PHYS format specifier macros for working
> with target_phys_addr_t types. These follow the standard pattern
> for such macros, and are more flexible than TARGET_FMT_plx, which
> does not allow specification of field widths.
>
From: Amit Shah
The Linux kernel already has a virtio-rng driver, this is the device
implementation.
When the guest asks for entropy from the virtio hwrng, it puts a buffer
in the vq. We then put entropy into that buffer, and push it back to
the guest.
Signed-off-by: Amit Shah
Signed-off-by:
In Amit's thread on virtio-rng, danpb mentioned that we really ought to have a
proper RNG backend infrastructure and of course he's correct on that.
Now that we have QOM, I wanted to demonstrate how we can use QOM to construct a
complete backend without adding any new infrastructure.
For this RFC
Am 25.06.2012 18:09, schrieb Anthony Liguori:
> Signed-off-by: Anthony Liguori
> ---
> include/qemu/object.h | 16 ++
> qom/object.c | 56
> +
> 2 files changed, 72 insertions(+), 0 deletions(-)
>
> diff --git a/include/qe
On 06/25/2012 09:33 AM, Mike Lovell wrote:
On 06/25/2012 04:40 AM, Stefan Hajnoczi wrote:
Have you looked at QEMU's net/vde.c backend?
Does VDE (http://vde.sourceforge.net/) already do everything that
QDES does?
Stefan
I have looked at VDE and used it for a few things. I think QDES has an
On 06/25/2012 10:34 AM, Eric Blake wrote:
On 06/25/2012 08:24 AM, Corey Bryant wrote:
+if (fcntl_setfl(ret, O_CLOEXEC, (flags & O_CLOEXEC) ? 1
: 0) < 0) {
Broken. O_CLOEXEC _only_ affects open(); to change it on an existing
fd, you have to use fcntl(F_GETFD/F_SETFD) (not F_GETF
Hello,
Am 25.06.2012 12:59, schrieb Alberich de megres:
> Together with some colleagues, we are developing a USB hardware. But
> as the prototype is not ready to be built, and is not cheap,
> or idea is to develop at the same time a virtual device for the qemu.
>
> I searched for documentation ab
Hi all,
is there a way to deny passing certain devices to the guest? I want to deny
two devices to get assigned to the guest, all others on all ports are
allowed, ony two each with a defined pid/vid should not get routed.
Providing a positive list would exceed the max. command line parameter
l
On 06/25/2012 11:34 AM, Kevin Wolf wrote:
Am 25.06.2012 16:51, schrieb Corey Bryant:
Thanks for catching this. I'll fix this in v5. In terms of platforms
that support dup3 vs dup2, I'm assuming the following preprocessor
checks will do what we need:
#if defined(__linux__) || defined(__CYGWI
On 06/25/2012 10:34 AM, Eric Blake wrote:
On 06/25/2012 08:24 AM, Corey Bryant wrote:
+if (fcntl_setfl(ret, O_CLOEXEC, (flags & O_CLOEXEC) ? 1
: 0) < 0) {
Broken. O_CLOEXEC _only_ affects open(); to change it on an existing
fd, you have to use fcntl(F_GETFD/F_SETFD) (not F_GETF
On 06/25/2012 09:34 AM, Kevin Wolf wrote:
> Am 25.06.2012 16:51, schrieb Corey Bryant:
>> Thanks for catching this. I'll fix this in v5. In terms of platforms
>> that support dup3 vs dup2, I'm assuming the following preprocessor
>> checks will do what we need:
>>
>> #if defined(__linux__) || de
Signed-off-by: Anthony Liguori
---
include/qemu/object.h | 16 ++
qom/object.c | 56 +
2 files changed, 72 insertions(+), 0 deletions(-)
diff --git a/include/qemu/object.h b/include/qemu/object.h
index 8b17776..3711054 1006
This is the backend used by devices that need to request entropy.
Signed-off-by: Anthony Liguori
---
Makefile.objs |2 +
backends/Makefile.objs |1 +
backends/rng.c | 84
include/qemu/rng.h | 83 +++
On Mon, Jun 25, 2012 at 05:14:08PM +0200, Kevin Wolf wrote:
> Am 25.06.2012 16:14, schrieb Michael Roth:
> > On Mon, Jun 25, 2012 at 02:58:40PM +0200, Kevin Wolf wrote:
> >> Am 23.06.2012 02:33, schrieb Michael Roth:
> >>> Hi everyone,
> >>>
> >>> I've set up a tree for stable-1.1 at:
> >>>
> >>> g
Define a set of TARGET_PRI*PHYS format specifier macros for working
with target_phys_addr_t types. These follow the standard pattern
for such macros, and are more flexible than TARGET_FMT_plx, which
does not allow specification of field widths.
Suggested-by: Andreas Färber
Signed-off-by: Peter Ma
This patchset defines a set of macros for use in printf format strings
for handling target_phys_addr_t values, as suggested by Andreas Färber
(http://lists.gnu.org/archive/html/qemu-devel/2012-01/msg01547.html)
We can then use them in some places which were previously either
making assumptions abo
Use the new TARGET_PRIxPHYS macro to avoid the need to define an
OMAP_FMT_plx macro whose expansion depends directly on
TARGET_PHYS_ADDR_BITS.
Signed-off-by: Peter Maydell
---
hw/omap.h |8 +---
1 files changed, 1 insertions(+), 7 deletions(-)
diff --git a/hw/omap.h b/hw/omap.h
index 3d
Switch a format string from %x to TARGET_PRIxPHYS so that it will
continue to work even if target_phys_addr_t is changed
to 64 bits in the future.
Signed-off-by: Peter Maydell
---
hw/sh_serial.c |6 --
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/hw/sh_serial.c b/hw/sh_
On Mon, Jun 25, 2012 at 4:10 PM, Kevin Wolf wrote:
> Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
>> +echo
>> +echo "== Read-only access must still work =="
>> +
>> +$QEMU_IO -r -c "read -P 0x5a 0 512" $TEST_IMG | _filter_qemu_io
>
> I would add a check here that the dirty bit is still set.
>
> M
On Mon, Jun 25, 2012 at 3:35 PM, Kevin Wolf wrote:
> Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
>> The lazy refcounts bit indicates that this image can take advantage of
>> the dirty bit and that refcount updates can be postponed.
>>
>> Signed-off-by: Stefan Hajnoczi
>> ---
>> docs/specs/qcow
Now we have TARGET_PRI*PHYS for printing target_phys_addr_t values,
we can use them in monitor.c rather than having duplicate code
in two arms of a TARGET_PHYS_ADDR_BITS ifdef.
Signed-off-by: Peter Maydell
---
monitor.c | 29 -
1 files changed, 4 insertions(+), 25 d
On Mon, Jun 25, 2012 at 3:18 PM, Kevin Wolf wrote:
> Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
>> This patch adds an incompatible feature bit to mark images that have not
>> been closed cleanly. When a dirty image file is opened a consistency
>> check and repair is performed.
>>
>> Signed-off
On Mon, Jun 25, 2012 at 3:06 PM, Zhi Yong Wu wrote:
> On Fri, Jun 22, 2012 at 11:08 PM, Stefan Hajnoczi
> wrote:
>> This series aims to improve qcow2 performance with cache=writethrough and
>> cache=directsync. In particular it reduces the impact of metadata updates
>> for
> What does cache=dir
On Mon, Jun 25, 2012 at 11:59 AM, Alberich de megres
wrote:
> Together with some colleagues, we are developing a USB hardware. But
> as the prototype is not ready to be built, and is not cheap,
> or idea is to develop at the same time a virtual device for the qemu.
>
> I searched for documentation
Anyone using these functions has to be prepared that irqchip
support may not be present. It shouldn't be up to the core
code to determine whether this is a fatal error. Currently
code written as:
virq = kvm_irqchip_add_msi_route(...)
if (virq < 0) {
} else {
}
works on x86 with and wi
Am 25.06.2012 16:51, schrieb Corey Bryant:
> Thanks for catching this. I'll fix this in v5. In terms of platforms
> that support dup3 vs dup2, I'm assuming the following preprocessor
> checks will do what we need:
>
> #if defined(__linux__) || defined(__CYGWIN__)
> dup3(fd, monfd->fd, O_CLOEXE
On 06/25/2012 04:40 AM, Stefan Hajnoczi wrote:
On Mon, Jun 25, 2012 at 6:42 AM, Mike Lovell wrote:
This is what I've been calling QDES or QEMU Distributed Ethernet Switch. I
first had the idea when I was playing with the udp and mcast socket network
backends while exploring how to build a VM in
On 06/22/2012 04:24 PM, Eric Blake wrote:
On 06/22/2012 12:36 PM, Corey Bryant wrote:
This patch adds the pass-fd QMP command using the QAPI framework.
Like the getfd command, it is used to pass a file descriptor via
SCM_RIGHTS and associate it with a name. However, the pass-fd
command also r
Am 25.06.2012 16:34, schrieb Eric Blake:
>> Also, I noticed in the fnctl man page that F_SETFL: "On Linux this
>> command can change only the O_APPEND, O_ASYNC, O_DIRECT, O_NOATIME, and
>> O_NONBLOCK flags." So I'll only set/unset these flags.
>
> O_NDELAY is the obsolete spelling of O_NONBLOCK;
Hello,
Together with some colleagues, we are developing a USB hardware. But
as the prototype is not ready to be built, and is not cheap,
or idea is to develop at the same time a virtual device for the qemu.
I searched for documentation about the USB layer on qemu but I could
not find a good point
Am 25.06.2012 16:14, schrieb Michael Roth:
> On Mon, Jun 25, 2012 at 02:58:40PM +0200, Kevin Wolf wrote:
>> Am 23.06.2012 02:33, schrieb Michael Roth:
>>> Hi everyone,
>>>
>>> I've set up a tree for stable-1.1 at:
>>>
>>> git://github.com/mdroth/qemu.git stable-1.1
>>>
>>> That is simply the initia
Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
> This tests establishes the basic post-conditions of the qcow2 lazy
> refcounts features:
>
> 1. If the image was closed normally, it is marked clean.
>
> 2. If an allocating write was performed and the image was not close
> normally, then i
Am 25.06.2012 16:54, schrieb Peter Maydell:
> On 25 June 2012 15:43, Andreas Färber wrote:
>> Am 25.06.2012 16:10, schrieb Peter Maydell:
>>> On 25 June 2012 14:48, Andreas Färber wrote:
Thus I would suggest TARGET_PRIdPHYSADDR, also addressing Eric's
comment. Judging from patches 2-4 i
Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
> Lazy refcounts is a performance optimization for qcow2 that postpones
> refcount metadata updates and instead marks the image dirty. In the
> case of crash or power failure the image will be left in a dirty state
> and repaired next time it is opened
In our disassembly code, the bfd_vma type is always 64 bits,
even if the target's virtual address width is only 32 bits. This
means that when we print out addresses we need to truncate them
to 32 bits, to avoid odd output which has incorrectly sign-extended
a value to 64 bits, for instance this ARM
On 25 June 2012 15:43, Andreas Färber wrote:
> Am 25.06.2012 16:10, schrieb Peter Maydell:
>> On 25 June 2012 14:48, Andreas Färber wrote:
>>> Thus I would suggest TARGET_PRIdPHYSADDR, also addressing Eric's
>>> comment. Judging from patches 2-4 it's not used to frequently unlike its
>>> user TAR
Am 25.06.2012 16:10, schrieb Peter Maydell:
> On 25 June 2012 14:48, Andreas Färber wrote:
>> Am 25.06.2012 14:45, schrieb Peter Maydell:
>>> This patchset defines a set of macros for use in printf format strings
>>> for handling target_phys_addr_t values, as suggested by Andreas Färber
>>> (http:
Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
> The lazy refcounts bit indicates that this image can take advantage of
> the dirty bit and that refcount updates can be postponed.
>
> Signed-off-by: Stefan Hajnoczi
> ---
> docs/specs/qcow2.txt |6 +-
> 1 file changed, 5 insertions(+), 1 d
On 06/25/2012 08:24 AM, Corey Bryant wrote:
>>> +if (fcntl_setfl(ret, O_CLOEXEC, (flags & O_CLOEXEC) ? 1
>>> : 0) < 0) {
>>
>> Broken. O_CLOEXEC _only_ affects open(); to change it on an existing
>> fd, you have to use fcntl(F_GETFD/F_SETFD) (not F_GETFL/F_SETFL).
>>
>>
>
> I'll fix
On 06/22/2012 03:58 PM, Eric Blake wrote:
On 06/22/2012 12:36 PM, Corey Bryant wrote:
This patch adds support to qemu_open to dup(fd) a pre-opened file
descriptor if the filename is of the format /dev/fd/X.
This can be used when QEMU is restricted from opening files, and
the management applic
On 06/25/2012 05:02 AM, Andreas Färber wrote:
Hi Peter,
Am 25.06.2012 11:34, schrieb Peter Crosthwaite:
I have a platform (Xilinx Zynq) that has a USB EHCI controller that
attaches directly to the system bus and not through PCI. We are
looking for a way to disentangle EHCI from PCI - currently
Am 22.06.2012 17:08, schrieb Stefan Hajnoczi:
> This patch adds an incompatible feature bit to mark images that have not
> been closed cleanly. When a dirty image file is opened a consistency
> check and repair is performed.
>
> Signed-off-by: Stefan Hajnoczi
> ---
> block/qcow2.c | 44 ++
1 - 100 of 189 matches
Mail list logo
