Save PC speaker state to remove differences between system
states after saving the snapshot and after loading it again.
This patch is needed for deterministic replay of the execution.
v2: Do not save the data that cannot affect the guest code behavior.
Signed-off-by: Pavel Dovgalyuk
---
hw/pcspk
Add MIPS ASE DSP Accumulator and DSPControl Access instructions.
Signed-off-by: Jia Liu
---
target-mips/dsp_helper.c | 1044 ++
target-mips/helper.h | 52 +++
target-mips/translate.c | 425 +++
3 files changed, 1521 insertions(+
It's been a long time.
I've tested the performances of one jump difference when fast qemu_ld/st
(TLB hit).
The result shows 3.6% CoreMark enhancement when reducing one jump where slow
paths are generated at the end of block as same for the both cases.
That means reducing one jump dominates the
Delete DSP r1 & DSP r2 from TODO file.
Signed-off-by: Jia Liu
---
target-mips/TODO |2 --
1 file changed, 2 deletions(-)
diff --git a/target-mips/TODO b/target-mips/TODO
index 2a3546f..15d67cd 100644
--- a/target-mips/TODO
+++ b/target-mips/TODO
@@ -6,8 +6,6 @@ General
- Unimplemented ASEs
Add MIPS ASE DSP Branch instructions.
Signed-off-by: Jia Liu
---
target-mips/translate.c | 50 +++
1 file changed, 46 insertions(+), 4 deletions(-)
diff --git a/target-mips/translate.c b/target-mips/translate.c
index e1ea9c1..18d827d 100644
--- a/ta
Use correct acc value to index cpu_HI/cpu_LO rather than using a fix number.
Signed-off-by: Jia Liu
---
target-mips/translate.c | 134 +--
1 file changed, 107 insertions(+), 27 deletions(-)
diff --git a/target-mips/translate.c b/target-mips/translate
Add internal functions using by MIPS ASE DSP instructions.
Signed-off-by: Jia Liu
---
target-mips/Makefile.objs |2 +-
target-mips/dsp_helper.c | 1277 +
2 files changed, 1278 insertions(+), 1 deletion(-)
create mode 100644 target-mips/dsp_helper
On 08/27/12 18:21, Søren Sandmann wrote:
> From: Søren Sandmann Pedersen
>
> It's not uncommon for an X workload to have more than 1024 pixmaps
> live at the same time. Ideally, there wouldn't be any fixed limit like
> this, but since we have one, increase it to 4096.
> ---
> ui/spice-display.h
Signed-off-by: Bo Yang
---
Makefile |9 -
kvm_stat.texi | 55 +++
2 files changed, 63 insertions(+), 1 deletions(-)
create mode 100644 kvm_stat.texi
diff --git a/Makefile b/Makefile
index 1cd5bc8..ee524b0 100644
--- a/Makefi
The guest may enter into state of no receive descriptors,
and if there is no interrupt, the descriptor filling function
has no chance to run again,which causes network stall. According
to liunux driver's implementation, the descriptor with EL bit set
must not be touched by hardware, usually, the bu
On 08/27/12 19:20, Søren Sandmann Pedersen wrote:
> From: Søren Sandmann Pedersen
>
> This new interface lets spice server inform the guest whether
>
> (a) a client is connected
> (b) what capabilities the client has
>
> There is a fixed number (464) of bits reserved for capabilities, and
> whe
On 08/27/12 19:20, Søren Sandmann Pedersen wrote:
> From: Søren Sandmann Pedersen
>
> The client_present field is a byte that is set of non-zero when a
> client is connected and to zero when no client is connected.
>
> The client_capabilities[58] array contains 464 bits that indicate the
> capab
This was fixed a long time ago. I just verified it with latest qemu-kvm.git and
kvm.git.
A RHEL6.3 guest can boot up with 10 VFs, and each VF has good network.
So, mark this bug as "fixed" and "verified".
** Changed in: qemu
Status: New => Fix Released
--
You received this bug notificati
I added a special opcode, which is not used by existing x86. When the
process in the guest issues this opcode, the qemu starts to log its
mmu access.
On Mon, Aug 27, 2012 at 11:14 PM, 陳韋任 (Wei-Ren Chen)
wrote:
>> My final goal is to obtain the memory access trace for a particular
>> process in
On Tue, Aug 28, 2012 at 11:09 AM, liu ping fan wrote:
> On Tue, Aug 28, 2012 at 3:38 AM, Jan Kiszka wrote:
>> On 2012-08-27 20:52, Avi Kivity wrote:
>>> On 08/27/2012 11:39 AM, Jan Kiszka wrote:
On 2012-08-27 20:20, Avi Kivity wrote:
> On 08/27/2012 11:17 AM, Jan Kiszka wrote:
>> On
> My final goal is to obtain the memory access trace for a particular
> process in the guest, so your patch really helps, except for too many
> kernel _mmu events.
How do you know guest is running which process, and log it's memory
access trace?
Regards,
chenwj
--
Wei-Ren Chen (陳韋任)
Computer
On Tue, Aug 28, 2012 at 3:38 AM, Jan Kiszka wrote:
> On 2012-08-27 20:52, Avi Kivity wrote:
>> On 08/27/2012 11:39 AM, Jan Kiszka wrote:
>>> On 2012-08-27 20:20, Avi Kivity wrote:
On 08/27/2012 11:17 AM, Jan Kiszka wrote:
> On 2012-08-27 20:09, Avi Kivity wrote:
>> On 08/27/2012 10:14
On 2012-08-27 22:53, Avi Kivity wrote:
> On 08/27/2012 12:38 PM, Jan Kiszka wrote:
Even worse, apply
restrictions on how the dispatched objects, the regions, have to be
treated because of this.
>>>
>>> Please elaborate.
>>
>> The fact that you can't manipulate a memory region object
On 27.08.2012, at 14:13, Aurelien Jarno wrote:
> Hi,
>
> As you probably know, the PPC machines with a heathrow controller is
> broken following commit 9e56edcf ("vga: raise default vgamem size").
> The PCI hole space is not big enough for such a new default size.
>
> Alexander has fixed it i
Hi Blue,
thanks for the review. I addressed most of them, the others a commented
below.
On 2012-08-27 20:56, Blue Swirl wrote:
>> +typedef struct AssignedDevice {
>> +PCIDevice dev;
>> +PCIHostDeviceAddress host;
>> +uint32_t dev_id;
>> +uint32_t features;
>> +int intpin;
>> +
When setting up the NVIC memory regions the memory range
0x100..0xcff is aliased to an IO memory region that belongs
to the ARM GIC. This aliased region should be added to the
NVIC memory container, but the actual GIC IO memory region
was being added instead. This mixup was causing the wrong
IO m
The CONFIG_TCG_PASS_AREG0 code for calling ld/st helpers was
broken in that it did not respect the ABI requirement that 64
bit values were passed in even-odd register pairs. The simplest
way to fix this is to implement some new utility functions
for marshalling function arguments into the correct r
Hi
Please send in any agenda items you are interested in covering.
Thanks, Juan.
Peter Maydell wrote:
> Yes, qemu's linux-user emulation layer doesn't currently support any of
> the posix timer syscalls.
Any idea how much work is involved to implement this?
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https:/
Il 27/08/2012 22:58, Avi Kivity ha scritto:
>> > It's best to start this conversion using very coarse locking. There's
>> > no need to start with ultra fine grain locking.
> How does it work? You have to drop this main loop lock to dispatch the
> callback, so the data structure you use for dispat
Hi,
As you probably know, the PPC machines with a heathrow controller is
broken following commit 9e56edcf ("vga: raise default vgamem size").
The PCI hole space is not big enough for such a new default size.
Alexander has fixed it in OpenBIOS r1063, while the current version in
QEMU is r1062. It
Yes, qemu's linux-user emulation layer doesn't currently support any of
the posix timer syscalls.
** Summary changed:
- qemu: Unsupported syscall: 257
+ qemu: Unsupported syscall: 257 (timer_create)
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subs
On 08/27/2012 12:17 PM, Anthony Liguori wrote:
> Avi Kivity writes:
>
> > On 08/27/2012 09:24 AM, Anthony Liguori wrote:
> >> >
> >> > I'm sure we should leave existing code alone wherever possible, focusing
> >> > on providing alternative versions for those paths that matter. Example:
> >> > Most
On 08/27/2012 12:38 PM, Jan Kiszka wrote:
> >> Even worse, apply
> >> restrictions on how the dispatched objects, the regions, have to be
> >> treated because of this.
> >
> > Please elaborate.
>
> The fact that you can't manipulate a memory region object arbitrarily
> after removing it from the m
Public bug reported:
Running qemu-arm-static for git HEAD. When I try to install ghc from
debian into my arm chroot I get:
Setting up ghc (7.4.1-4) ...
qemu: Unsupported syscall: 257
ghc: timer_create: Function not implemented
qemu: Unsupported syscall: 257
ghc-pkg: timer_create: Function not imp
On 27 August 2012 21:37, Meador Inge wrote:
> This changes the driver behavior to choose the default machine
> model based on the CPU being used. Defaulting the machine this
> way makes it easier to use QEMU as an ISS by just specifying
> the -cpu option since a default machine that is suitable f
This patch adds support for a "dummy" ARMv7-M board so that
QEMU can be used as an ISS for ARMv7-M processors. For example,
running an image compiled for the Cortex-M3 with -cpu cortex-m3
should just work.
Signed-off-by: Meador Inge
---
hw/arm/Makefile.objs |1 +
hw/dummy_armv7m.c| 40
The minimal amount of arm_boot_info has been setup to allow
for machines based off of ARMv7-M processors to be loaded via the
-kernel option.
Signed-off-by: Meador Inge
---
hw/armv7m.c | 13 +
1 files changed, 13 insertions(+), 0 deletions(-)
diff --git a/hw/armv7m.c b/hw/armv7m.c
This changes the driver behavior to choose the default machine
model based on the CPU being used. Defaulting the machine this
way makes it easier to use QEMU as an ISS by just specifying
the -cpu option since a default machine that is suitable for
emulating the full ISA can be chosen.
For example
The current implementation of the ARM semi-hosting SYS_HEAPINFO
system call assumes that the base address of RAM for all ARM devices
is 0x0. This isn't true for ARMv7-M devices, which uses a base of
0x2000 for SRAM.
Signed-off-by: Meador Inge
---
target-arm/arm-semi.c |8 +++-
1 fil
Hi All,
This patch series is an attempt to improve the current ARMv7-M support
by making it easier to run applications that only require architecture
level support from the emulation (basically an ISS). We are mostly there
already, but there are some cases that we don't handle well. For example,
On Mon, Aug 27, 2012 at 7:24 PM, Michael S. Tsirkin wrote:
> On Mon, Aug 27, 2012 at 07:12:27PM +, Blue Swirl wrote:
>> On Mon, Aug 27, 2012 at 7:06 PM, Michael S. Tsirkin wrote:
>> > On Mon, Aug 27, 2012 at 06:58:29PM +, Blue Swirl wrote:
>> >> On Mon, Aug 27, 2012 at 12:20 PM, Michael S
On 2012-08-27 20:52, Avi Kivity wrote:
> On 08/27/2012 11:39 AM, Jan Kiszka wrote:
>> On 2012-08-27 20:20, Avi Kivity wrote:
>>> On 08/27/2012 11:17 AM, Jan Kiszka wrote:
On 2012-08-27 20:09, Avi Kivity wrote:
> On 08/27/2012 10:14 AM, Jan Kiszka wrote:
>>>
>>> Deregistration is fi
Hi,
The following patches add the ability for spice-server to advertise
the capabilities of connected clients to guests. They do this through
adding some new fields to QXLRom:
- whether a client is present
- a bit field to indicate which SPICE_DISPLAY_CAP_* capabilities the
client has.
Th
From: Søren Sandmann Pedersen
A new interface
set_client_capabilities (QXLInstance *qin,
uint8_t client_present,
uint8_t caps[58]);
is added to QXLInstance, and spice server is changed to call it
whenever a client connects or disconnects.
From: Søren Sandmann Pedersen
This new interface lets spice server inform the guest whether
(a) a client is connected
(b) what capabilities the client has
There is a fixed number (464) of bits reserved for capabilities, and
when the capabilities bits change, the QXL_INTERRUPT_CLIENT interrupt
i
Le Monday 27 Aug 2012 à 12:03:48 (-0600), Eric Blake a écrit :
> On 08/27/2012 01:30 AM, Benoît Canet wrote:
> > Signed-off-by: Benoit Canet
> > ---
> > block/quorum.c | 24
> > 1 file changed, 24 insertions(+)
>
> Say I'm using a 2/3 quorum. What happens if:
>
> ima
On Mon, Aug 27, 2012 at 07:12:27PM +, Blue Swirl wrote:
> On Mon, Aug 27, 2012 at 7:06 PM, Michael S. Tsirkin wrote:
> > On Mon, Aug 27, 2012 at 06:58:29PM +, Blue Swirl wrote:
> >> On Mon, Aug 27, 2012 at 12:20 PM, Michael S. Tsirkin
> >> wrote:
> >> > In preparation for adding PV EOI s
On 2012-08-27 21:17, Anthony Liguori wrote:
> Avi Kivity writes:
>
>> On 08/27/2012 09:24 AM, Anthony Liguori wrote:
I'm sure we should leave existing code alone wherever possible, focusing
on providing alternative versions for those paths that matter. Example:
Most timers are
Hi,
The following patches add the ability for spice-server to advertise
the capabilities of connected clients to guests. They do this through
adding some new fields to QXLRom:
- whether a client is present
- a bit field to indicate which SPICE_DISPLAY_CAP_* capabilities the
client has.
Th
Le Monday 27 Aug 2012 à 11:59:34 (-0600), Eric Blake a écrit :
> On 08/27/2012 01:30 AM, Benoît Canet wrote:
> > Valid quorum resources look like
> > quorum:threshold/total:path/to/image_1, ... ,path/to/image_total
> >
> > ',' is used as a separator to allow to use networked path
>
> Isn't this a
Avi Kivity writes:
> On 08/27/2012 09:24 AM, Anthony Liguori wrote:
>> >
>> > I'm sure we should leave existing code alone wherever possible, focusing
>> > on providing alternative versions for those paths that matter. Example:
>> > Most timers are fine under BQL. But some sensitive devices (RTC
On Mon, Aug 27, 2012 at 7:06 PM, Michael S. Tsirkin wrote:
> On Mon, Aug 27, 2012 at 06:58:29PM +, Blue Swirl wrote:
>> On Mon, Aug 27, 2012 at 12:20 PM, Michael S. Tsirkin wrote:
>> > In preparation for adding PV EOI support, disable PV EOI by default for
>> > 1.1 and older machine types, to
From: Søren Sandmann Pedersen
The client_present field is a byte that is set of non-zero when a
client is connected and to zero when no client is connected.
The client_capabilities[58] array contains 464 bits that indicate the
capabilities of the client. Each bit corresponds to a
SPICE_DISPLAY_C
On Mon, Aug 27, 2012 at 7:01 PM, Michael S. Tsirkin wrote:
> On Mon, Aug 27, 2012 at 06:56:38PM +, Blue Swirl wrote:
>> > +static uint32_t slow_bar_readb(void *opaque, target_phys_addr_t addr)
>> > +{
>> > +AssignedDevRegion *d = opaque;
>> > +uint8_t *in = d->u.r_virtbase + addr;
>>
>
On Mon, Aug 27, 2012 at 06:58:29PM +, Blue Swirl wrote:
> On Mon, Aug 27, 2012 at 12:20 PM, Michael S. Tsirkin wrote:
> > In preparation for adding PV EOI support, disable PV EOI by default for
> > 1.1 and older machine types, to avoid CPUID changing during migration.
> >
> > PV EOI can still
On Mon, Aug 27, 2012 at 06:56:38PM +, Blue Swirl wrote:
> > +static uint32_t slow_bar_readb(void *opaque, target_phys_addr_t addr)
> > +{
> > +AssignedDevRegion *d = opaque;
> > +uint8_t *in = d->u.r_virtbase + addr;
>
> Don't perform arithmetic with void pointers.
Why not?
We require
From: Søren Sandmann Pedersen
It's not uncommon for an X workload to have more than 1024 pixmaps
live at the same time. Ideally, there wouldn't be any fixed limit like
this, but since we have one, increase it to 4096.
---
ui/spice-display.h |2 +-
1 files changed, 1 insertions(+), 1 deletion
On Mon, Aug 27, 2012 at 12:20 PM, Michael S. Tsirkin wrote:
> In preparation for adding PV EOI support, disable PV EOI by default for
> 1.1 and older machine types, to avoid CPUID changing during migration.
>
> PV EOI can still be enabled/disabled by specifying it explicitly.
> Enable for 1.1
On 08/27/2012 11:39 AM, Jan Kiszka wrote:
> On 2012-08-27 20:20, Avi Kivity wrote:
> > On 08/27/2012 11:17 AM, Jan Kiszka wrote:
> >> On 2012-08-27 20:09, Avi Kivity wrote:
> >>> On 08/27/2012 10:14 AM, Jan Kiszka wrote:
> >
> > Deregistration is fine, the problem is destruction.
> >
>
On Mon, Aug 27, 2012 at 04:59:40PM +0200, Jan Kiszka wrote:
> On 2012-08-27 16:53, Michael S. Tsirkin wrote:
> > On Mon, Aug 27, 2012 at 02:48:57PM +0200, Jan Kiszka wrote:
> >> On 2012-08-27 14:42, Peter Maydell wrote:
> >>> On 27 August 2012 13:20, Michael S. Tsirkin wrote:
> Update linux-h
On 2012-08-27 20:20, Avi Kivity wrote:
> On 08/27/2012 11:17 AM, Jan Kiszka wrote:
>> On 2012-08-27 20:09, Avi Kivity wrote:
>>> On 08/27/2012 10:14 AM, Jan Kiszka wrote:
>
> Deregistration is fine, the problem is destruction.
>
It isn't as you access memory region states that
On 08/27/2012 09:24 AM, Anthony Liguori wrote:
> >
> > I'm sure we should leave existing code alone wherever possible, focusing
> > on providing alternative versions for those paths that matter. Example:
> > Most timers are fine under BQL. But some sensitive devices (RTC or HPET
> > as clock source
On 27 August 2012 08:23, Yeongkyoon Lee wrote:
> BTW, who will finally confirm my patches?
> I have sent four version of my patches in which I have applied all the
> reasonable feedbacks from this community.
If you'd like your patches committed you should not use the "[RFC]" tag
in the Subject, b
On 08/27/2012 06:19 AM, Anthony Liguori wrote:
> Liu Ping Fan writes:
>
> > From: Liu Ping Fan
> >
> > Scene:
> > obja lies in objA, when objA's ref->0, it will be freed,
> > but at that time obja can still be in use.
> >
> > The real example is:
> > typedef struct PCIIDEState {
> > PCIDevi
On Mon, Aug 27, 2012 at 7:23 AM, Yeongkyoon Lee
wrote:
> On 2012년 07월 29일 00:39, Yeongkyoon Lee wrote:
>>
>> On 2012년 07월 25일 23:00, Richard Henderson wrote:
>>>
>>> On 07/25/2012 12:35 AM, Yeongkyoon Lee wrote:
+#if defined(CONFIG_QEMU_LDST_OPTIMIZATION) && defined(CONFIG_SOFTMMU)
From: Peter Maydell
Now that we default to reserving nearly 4GB of RAM for the guest
address space when running a 32 bit linux-user guest on 64 bit
hosts, users are much more likely to run into it. Reword the
message to be more informative about what failed and provide
suggestions for how to fix
From: Jim Meyering
Declare arg_table to be "static const", and adjust the two users
to also be const.
Signed-off-by: Jim Meyering
Signed-off-by: Riku Voipio
---
linux-user/main.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/linux-user/main.c b/linux-user/main.c
From: Riku Voipio
Hi,
The following changes since commit d03c98d80ffb7c561d9e6874580f52fe7ecc8c6c:
tcg/ia64: fix and optimize ld/st slow path (2012-08-26 21:10:37 +0200)
are available in the git repository at:
git://git.linaro.org/people/rikuvoipio/qemu.git linux-user-for-upstream
for yo
From: "Dmitry V. Levin"
In case when TARGET_ABI_BITS == 32 && HOST_LONG_BITS == 64, the last
byte of the target dirent structure (aka d_type byte) was never copied
from the host dirent structure, thus breaking everything that relies
on valid d_type value, e.g. glob(3).
Reviewed-by: Peter Maydell
On 08/27/2012 11:17 AM, Jan Kiszka wrote:
> On 2012-08-27 20:09, Avi Kivity wrote:
> > On 08/27/2012 10:14 AM, Jan Kiszka wrote:
> >>>
> >>> Deregistration is fine, the problem is destruction.
> >>>
> >>
> >> It isn't as you access memory region states that can change after
> >> deregistration. Dev
On 2012-08-27 20:09, Avi Kivity wrote:
> On 08/27/2012 10:14 AM, Jan Kiszka wrote:
>>>
>>> Deregistration is fine, the problem is destruction.
>>>
>>
>> It isn't as you access memory region states that can change after
>> deregistration. Devices can remove memory regions from the mapping,
>> alter
setup_ioeventfds() is unnecessary and actually causes a segfault when used
ioeventfd=on is used on the command-line. Since ioeventfds are handled within
the memory API, it can be removed.
Signed-off-by: Cam Macdonell
---
hw/ivshmem.c | 15 ---
1 files changed, 0 insertions(+), 15
On 08/27/2012 10:14 AM, Jan Kiszka wrote:
> >
> > Deregistration is fine, the problem is destruction.
> >
>
> It isn't as you access memory region states that can change after
> deregistration. Devices can remove memory regions from the mapping,
> alter and then reinsert them. The last to steps m
On Mon, Aug 27, 2012 at 1:59 AM, Alexander Graf wrote:
>
>
> On 26.08.2012, at 10:34, Blue Swirl wrote:
>
>> On Sat, Aug 25, 2012 at 2:27 PM, Alexander Graf wrote:
>>>
>>>
>>> On 25.08.2012, at 00:43, Blue Swirl wrote:
>>>
On Wed, Aug 22, 2012 at 10:31 AM, Li Zhang wrote:
> When -usb
On 08/27/2012 01:30 AM, Benoît Canet wrote:
> Signed-off-by: Benoit Canet
> ---
> block/quorum.c | 24
> 1 file changed, 24 insertions(+)
Say I'm using a 2/3 quorum. What happens if:
image A and B agree on initial content, but not length
image B and C agree on length
On 08/27/2012 01:30 AM, Benoît Canet wrote:
> Valid quorum resources look like
> quorum:threshold/total:path/to/image_1, ... ,path/to/image_total
>
> ',' is used as a separator to allow to use networked path
Isn't this a step backwards? After all, on the command line, we would
have something lik
On 08/27/2012 01:15 AM, Benoît Canet wrote:
> This patchset add a JSON output mode to the qemu-img info command.
> It's a rewrite from scratch of the original patchset by Wenchao Xia
> following Anthony Liguori advices on JSON formating.
>
> the --output=(json|human) option is now mandatory on the
On Sun, Aug 26, 2012 at 7:08 PM, Aurelien Jarno wrote:
> op_helper.c has been removed in commit 0184e266. Remove the corresponding
> flags from Makefile.objs.
>
> Cc: Blue Swirl
Acked-by: Blue Swirl
As this is cleanup only, it could be postponed to 1.3.
> Signed-off-by: Aurelien Jarno
> ---
On 2012-08-27 14:21, Pavel Dovgaluk wrote:
> Save PC speaker state to remove differences between system
> states after saving the snapshot and after loading it again.
> This patch is needed for deterministic replay of the execution.
>
> Signed-off-by: Pavel Dovgalyuk
> ---
> hw/pcspk.c | 18 +++
On 2012-08-27 19:09, Avi Kivity wrote:
> On 08/27/2012 12:47 AM, Jan Kiszka wrote:
>> On 2012-08-27 09:01, Paolo Bonzini wrote:
>>> Il 25/08/2012 09:42, liu ping fan ha scritto:
>>
>> I don't see why MMIO dispatch should hold the IDEBus ref rather than the
>> PCIIDEState.
>>
Wh
On 08/27/2012 12:47 AM, Jan Kiszka wrote:
> On 2012-08-27 09:01, Paolo Bonzini wrote:
> > Il 25/08/2012 09:42, liu ping fan ha scritto:
>
> I don't see why MMIO dispatch should hold the IDEBus ref rather than the
> PCIIDEState.
>
> >> When transfer memory_region_init_io() 3rd p
On 2012-08-27 18:24, Anthony Liguori wrote:
> Jan Kiszka writes:
>
>> On 2012-08-27 17:14, Anthony Liguori wrote:
>>> Jan Kiszka writes:
>>>
On 2012-08-27 15:19, Anthony Liguori wrote:
> Liu Ping Fan writes:
>
>> From: Liu Ping Fan
>>
>> Scene:
>> obja lies in ob
Am 27.08.2012 14:21, schrieb Pavel Dovgaluk:
> Save PC speaker state to remove differences between system
> states after saving the snapshot and after loading it again.
> This patch is needed for deterministic replay of the execution.
>
> Signed-off-by: Pavel Dovgalyuk
> ---
> hw/pcspk.c | 18 +
Ping, no response so far ...
Thanks,
Alex
Alex ZUEPKE wrote:
> Hi,
>
> x86 software emulation (non-KVM mode) does not check privilege levels on
> task gate switches ... so one can invoke a kernel's double fault handler
> from user space -- very bad.
>
> Expected behaviour (testcase works with
Jan Kiszka writes:
> On 2012-08-27 17:14, Anthony Liguori wrote:
>> Jan Kiszka writes:
>>
>>> On 2012-08-27 15:19, Anthony Liguori wrote:
Liu Ping Fan writes:
> From: Liu Ping Fan
>
> Scene:
> obja lies in objA, when objA's ref->0, it will be freed,
> but at th
On Sat, Aug 25, 2012 at 4:41 PM, Max Filippov wrote:
> On Sat, Aug 25, 2012 at 9:20 PM, Steven wrote:
>> On Tue, Aug 21, 2012 at 3:18 AM, Max Filippov wrote:
>>> On Tue, Aug 21, 2012 at 9:40 AM, Steven wrote:
Hi, Max,
I wrote a small program to verify your patch could catch all the lo
Am 27.08.2012 11:04, schrieb Stefan Hajnoczi:
> On Sun, Aug 26, 2012 at 10:56 AM, Alexandre DERUMIER
> wrote:
>> It is possible to achieve the same behaviour with external snapshot ? (I
>> would like to do it online)
>> I don't see how I can rollback to the point of time of the snapshot.
>
> The
Hi all,
this is an RFC series implementing an NBD server embedded inside QEMU.
This can be used in various cases, including migration with non-shared
storage.
Three new commands are introduced at the QMP level
{ 'command': 'nbd-server-start', 'data': { 'addr': 'IPSocketAddress' } }
{ 'comma
Andreas Färber writes:
> Am 26.08.2012 17:51, schrieb Anthony Liguori:
>> Callers should just use object_unref
>>
>> Signed-off-by: Anthony Liguori
>> ---
>> hw/qdev.c |4
>> include/qemu/object.h |9 -
>> qom/object.c |2 +-
>> 3 files changed, 1
Paolo Bonzini writes:
> Il 27/08/2012 15:55, Anthony Liguori ha scritto:
>>> > This patch provides a way to optionally suppress spurious interrupts,
>>> > as a workaround for systems described below:
>>> >
>>> > Some old operating systems do not handle spurious interrupts well,
>>> > and qemu ten
This is needed so that we can set up a QemuOpts instance from QMP
parameters. The way to go here is to and move qemu-sockets.c away from
QemuOpts and use Laszlo's QemuOptsVisitor whenever *_opts functions are
called now. This can be done later, however.
Signed-off-by: Paolo Bonzini
---
qemu-so
Okay, looks like the performance issue started with introduction of
pc-0.15 machine profile in version 1.0.1. I'll narrow the problem down
further.
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/10363
Signed-off-by: Paolo Bonzini
---
qemu-sockets.c | 2 +-
qemu_socket.h | 1 +
2 file modificati, 2 inserzioni(+). 1 rimozione(-)
diff --git a/qemu-sockets.c b/qemu-sockets.c
index b292311..7a28715 100644
--- a/qemu-sockets.c
+++ b/qemu-sockets.c
@@ -407,7 +407,7 @@ err:
}
/* compatibility wr
Ping^2 since rc2 is just around the corner.
thanks
-- PMM
On 23 August 2012 17:15, Peter Maydell wrote:
> Ping! This didn't get into rc1.
>
> thanks
> -- PMM
>
> On 20 August 2012 11:36, Peter Maydell wrote:
>> Now that we default to reserving nearly 4GB of RAM for the guest
>> address space wh
Because nbd_client_close removes the I/O handlers for the client
socket, there is no way that any suspended coroutines are restarted.
This will be a problem with the QEMU embedded NBD server, because
we will have a QMP command to forcibly close all connections with
the clients.
Instead, we can exp
Signed-off-by: Paolo Bonzini
---
Makefile.objs | 3 ++-
block.c | 19 ++-
block.h | 1 +
block_int.h | 2 ++
4 file modificati, 19 inserzioni(+), 6 rimozioni(-)
diff --git a/Makefile.objs b/Makefile.objs
index c42affc..8321f81 100644
--- a/Makefile.objs
+++ b/Mak
Add an API to register and find named exports.
Signed-off-by: Paolo Bonzini
---
nbd.c | 38 ++
nbd.h | 3 +++
2 file modificati, 41 inserzioni(+)
diff --git a/nbd.c b/nbd.c
index f6eaea3..1249548 100644
--- a/nbd.c
+++ b/nbd.c
@@ -90,13 +90,17 @@ struct NBDR
We will need the NBDClient in nbd_send_negotiate to store the
export requested by the client.
Signed-off-by: Paolo Bonzini
---
nbd.c | 76 ---
1 file modificato, 41 inserzioni(+), 35 rimozioni(-)
diff --git a/nbd.c b/nbd.c
index 82
Clients have a pointer to the NBDExport that they serve. Do not
let a dangling pointer escape. Also flush all pending I/O so that
coroutines are forced to exit.
Signed-off-by: Paolo Bonzini
---
nbd.c | 14 ++
1 file modificato, 14 inserzioni(+)
diff --git a/nbd.c b/nbd.c
index cb4
Signed-off-by: Paolo Bonzini
---
blockdev-nbd.c | 38 ++
qapi/opts-visitor.c | 48
2 file modificati, 58 inserzioni(+), 28 rimozioni(-)
diff --git a/blockdev-nbd.c b/blockdev-nbd.c
index 5a415be..c190caa 10
On 2012-08-27 17:14, Anthony Liguori wrote:
> Jan Kiszka writes:
>
>> On 2012-08-27 15:19, Anthony Liguori wrote:
>>> Liu Ping Fan writes:
>>>
From: Liu Ping Fan
Scene:
obja lies in objA, when objA's ref->0, it will be freed,
but at that time obja can still be in use.
Am 26.08.2012 17:51, schrieb Anthony Liguori:
> Callers should just use object_unref
>
> Signed-off-by: Anthony Liguori
> ---
> hw/qdev.c |4
> include/qemu/object.h |9 -
> qom/object.c |2 +-
> 3 files changed, 1 insertions(+), 14 deletions(-)
>
>
Jan Kiszka writes:
> On 2012-08-27 15:19, Anthony Liguori wrote:
>> Liu Ping Fan writes:
>>
>>> From: Liu Ping Fan
>>>
>>> Scene:
>>> obja lies in objA, when objA's ref->0, it will be freed,
>>> but at that time obja can still be in use.
>>>
>>> The real example is:
>>> typedef struct PCIIDE
malc writes:
> On Mon, 27 Aug 2012, Anthony Liguori wrote:
>
>> malc writes:
>>
>
> [..snip..]
>
>> >
>> > Number 2 was, and should stay, as the emulation wasn't correct before it,
>> > don't really care about the rest.
>>
>> Okay, please revert the rest then.
>>
>
> Done.
Thank you!
Regard
1 - 100 of 186 matches
Mail list logo
