On 02/21/2014 05:44 PM, Dr. David Alan Gilbert wrote:
It's not clear to me how much of this (or any) of this control loop should
be in QEMU or in the management software, but I would definitely agree
that a minimum of at least the ability to detect the situation and remedy
the situation should be
On Fri, Feb 28, 2014 at 01:29:59PM -0500, Luiz Capitulino wrote:
> On Fri, 21 Feb 2014 18:01:40 +0800
> Amos Kong wrote:
>
> > On Thu, Feb 20, 2014 at 12:46:14PM -0500, Vlad Yasevich wrote:
> > > On 02/20/2014 11:38 AM, Amos Kong wrote:
> > > > Stefan Fritsch just fixed a virtio-net driver bug [1
>From 3a7783cd9a0556787809d3d5ecb5f2b85dd9fc02 Mon Sep 17 00:00:00 2001
From: Liu Jinsong
Date: Mon, 3 Mar 2014 18:56:39 +0800
Subject: [PATCH] target-i386: bugfix of Intel MPX
The correct size of cpuid 0x0d sub-leaf 4 is 0x40, not 0x10.
This is confirmed by Anvin H Peter and Mallick Asit K.
Sig
On Tue, Feb 25, 2014 at 06:09:20PM +0800, Hu Tao wrote:
> On Wed, Feb 19, 2014 at 10:03:13AM +0100, Paolo Bonzini wrote:
>
> <...>
>
> > > static int
> > > ram_backend_memory_init(HostMemoryBackend *backend, Error **errp)
> > > {
> > >+HostMemoryBackendRam *ram_backend = MEMORY_BACKEND_RAM(ba
On 03/02/2014 08:31 AM, Paolo Bonzini wrote:
> Il 01/03/2014 13:30, Alexey Kardashevskiy ha scritto:
>>> >
>>> > Corrupted DMA buffer is 0x e0 -- 0x7f15c313f000.
>>> > The e1000 packet is at 0x12ffac2 -- 0x7f15c313eac2.
>>> >
>>> > (0x7f15c313f000 - 0x7f15c313eac2) = 0x53e which is less than 0
Signed-off-by: Xuebing Wang
Reviewed-By: Alex Bligh
---
qemu-timer.c | 97 ++
1 file changed, 50 insertions(+), 47 deletions(-)
diff --git a/qemu-timer.c b/qemu-timer.c
index c9801da..e592c14 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
Signed-off-by: Xuebing Wang
Reviewed-By: Alex Bligh
---
qemu-timer.c | 86 ++
1 file changed, 45 insertions(+), 41 deletions(-)
diff --git a/qemu-timer.c b/qemu-timer.c
index 471150c..6f13a76 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
Signed-off-by: Xuebing Wang
Reviewed-By: Alex Bligh
---
qemu-timer.c | 160 ++
1 file changed, 82 insertions(+), 78 deletions(-)
diff --git a/qemu-timer.c b/qemu-timer.c
index 2db87ba..471150c 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
Signed-off-by: Xuebing Wang
Reviewed-By: Alex Bligh
---
qemu-timer.c | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/qemu-timer.c b/qemu-timer.c
index 6f13a76..c9801da 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
@@ -548,6 +548,9 @@ int qemu_poll_ns(GPollFD *f
Signed-off-by: Xuebing Wang
Reviewed-By: Alex Bligh
---
qemu-timer.c |9 +++--
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/qemu-timer.c b/qemu-timer.c
index e592c14..21aff82 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
@@ -22,13 +22,10 @@
* THE SOFTWARE.
*/
-#incl
This patch does below:
- Move function timerlist_has_timers() to be first QEMUTimerList function
because there is a chance that we will make it private
- Move QEMUTimerList related functions together
- Remove below comments for function timerlist_deadline_ns(), because:
-- There is a
This patch does below:
- Make QEMUTimerList functions private (remove from APIs)
- Comment out function QEMUTimerList timerlist_get_clock() in order
to fix compile error
Signed-off-by: Xuebing Wang
Reviewed-By: Alex Bligh
---
include/qemu/timer.h | 94 -
Hi Paolo, Stephen,
The following changes since commit 9fbee91a131a05e443d7108d7fbdf3ca91020290:
Merge remote-tracking branch 'remotes/kvm/uq/master' into staging (2014-02-27
16:00:31 +)
are available in the git repository at:
https://github.com/xbing6/qemu tags/xbing.refactor-timer
f
Hi, All
I found some discussion about VM live-snapshot, but haven't seen any progress.
https://lists.gnu.org/archive/html/qemu-devel/2013-08/msg02125.html
http://markmail.org/thread/shneezha7kmtosvb#query:+page:1+mid:shneezha7kmtosvb+state:results
Here I have another proposal, based on the live-m
Am 27.02.2014 00:51, schrieb Peter Maydell:
> The function qobject_from_json() doesn't actually allow its
> argument to be a format string -- it passes a NULL va_list*
> to qobject_from_jsonv(), and the parser code will then never
> actually intepret %-escape sequences (it tests whether the
"inter
Am 02.03.2014 20:36, schrieb Peter Maydell:
> The size of the UID/GID types depends on whether USE_UID16 is
> defined. Define a new put_user_id() which writes a uid/gid
> type to guest memory. This fixes getresuid and getresgid, which
> were always storing 16 bits even if the uid type was 32 bits.
Hello Stefan,
We have a file pc-bios/ppc_rom.bin in qemu.git, used by PReP emulation,
to which we do not have available the full GPL source code. I would like
to replace the current file with a new binary built from original
OpenHack'Ware sources plus our pc-bios/ohw.diff plus build fixes plus
wor
On 28 February 2014 00:57, Rob Herring wrote:
> From: Rob Herring
>
> Intermittent issues have been seen where no serial input occurs. It
> appears the pl011 gets in a state where the rx interrupt never fires
> because the rx interrupt only asserts when crossing the fifo trigger
> level. The fifo
This is a set of miscellaneous patches for linux-user which fix
problems seen with AArch64 guests. Mostly they're from the SuSE
1.6 tree (though I've tweaked one or two of them a bit).
The sendmmsg syscall patch is a significantly revised version
of the patch Alex sent to the list a while back; I'
Fix the loop restoring the FP registers from the signal frame to match
the one used when setting up the signal frame, so that it handles
TARGET_WORDS_BIGENDIAN being set.
Signed-off-by: Peter Maydell
---
linux-user/signal.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff -
From: Alexander Graf
Glibc when built for newer kernels assumes that the sendmmsg syscall is
available. Without it, dns resolution simply fails to work.
Wrap the syscall with existing infrastructure so that we don't have a host
dependency on sendmmsg.
To avoid locking the same area of guest mem
From: Andreas Schwab
The AArch64 kernel defines its __kernel_uid_t type as 32 bits, unlike
32 bit ARM, so don't enable our 16-bit UID wrapper handling.
Signed-off-by: Andreas Schwab
Signed-off-by: Peter Maydell
---
linux-user/syscall_defs.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion
The size of the UID/GID types depends on whether USE_UID16 is
defined. Define a new put_user_id() which writes a uid/gid
type to guest memory. This fixes getresuid and getresgid, which
were always storing 16 bits even if the uid type was 32 bits.
Reported-by: Michael Matz
Signed-off-by: Peter May
From: Michael Matz
Implement support for signal handlers with the SA_RESTORER
flag set.
Signed-off-by: Michael Matz
[PMM: minor tweaks to make patch apply to current master]
Signed-off-by: Peter Maydell
---
linux-user/signal.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletio
Hi Fabien,
Am 14.02.2014 18:27, schrieb Fabien Chouteau:
> On 02/14/2014 04:33 PM, Andreas Färber wrote:
>> As for the other one you'll need to sort our who sends a pull if Blue
>> doesn't resurface -
>
> I didn't see any message about this. Does anyone know why he's not around?
>
>> I note that
On Sun, Mar 02, 2014 at 11:15:15AM -0500, Gabriel L. Somlo wrote:
> On Sun, Mar 02, 2014 at 04:52:22PM +0200, Michael S. Tsirkin wrote:
> > On Fri, Feb 28, 2014 at 01:57:19PM -0500, Gabriel L. Somlo wrote:
> > > Both QEMU and KVM have already accumulated a significant number of
> > > optimizations
On Sun, Mar 02, 2014 at 04:52:22PM +0200, Michael S. Tsirkin wrote:
> On Fri, Feb 28, 2014 at 01:57:19PM -0500, Gabriel L. Somlo wrote:
> > Both QEMU and KVM have already accumulated a significant number of
> > optimizations based on the hard-coded assumption that ioapic polarity
> > will always us
On Sun, Mar 02, 2014 at 04:31:12PM +0200, Michael S. Tsirkin wrote:
> On Sun, Mar 02, 2014 at 09:56:50AM +0100, Paolo Bonzini wrote:
> > Il 02/03/2014 01:17, Gabriel L. Somlo ha scritto:
> > >Although, on KVM, it's simply hardcoded to 0x14 rather than exposing to
> > >the guest whatever the host CP
On 03/02/2014 02:17 PM, Michael Tokarev wrote:
26.02.2014 11:53, Sebastian Huber wrote:
Ping.
There was another comment on this, about the first hunk,
see https://lists.gnu.org/archive/html/qemu-trivial/2014-02/msg00067.html
Note that the comment below this line mentions the +1 which is
being
I am glad to hear that you are also interested in migration. We will be pleasure
if you find some bugs or suggestion and discuss them together with us.
As I know, the qemu will print some log about the migration failure. It is
same to your patch.
> This adds more traces in the migration code.
>
On Thu, Feb 27, 2014 at 11:06:17PM -0500, Gabriel L. Somlo wrote:
> Both QEMU and KVM have already accumulated a significant number of
> optimizations based on the hard-coded assumption that ioapic polarity
> will always use the ActiveHigh convention, where the logical and
> physical states of leve
On Fri, Feb 28, 2014 at 01:57:19PM -0500, Gabriel L. Somlo wrote:
> Both QEMU and KVM have already accumulated a significant number of
> optimizations based on the hard-coded assumption that ioapic polarity
> will always use the ActiveHigh convention, where the logical and
> physical states of leve
On Sun, Mar 02, 2014 at 09:56:50AM +0100, Paolo Bonzini wrote:
> Il 02/03/2014 01:17, Gabriel L. Somlo ha scritto:
> >Although, on KVM, it's simply hardcoded to 0x14 rather than exposing to
> >the guest whatever the host CPU's apic version happens to be, or
> >trying to match it to the CPU model:
>
The model was generating interrupts for all enabled timers after the
expiration of one of them. Avoid this by passing to the timer callback
function a structure containing the index of the expired timer.
Signed-off-by: Beniamino Galvani
---
hw/timer/allwinner-a10-pit.c | 30 +++
Signed-off-by: Beniamino Galvani
---
hw/net/allwinner_emac.c |4 ++--
include/hw/net/allwinner_emac.h |1 +
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/hw/net/allwinner_emac.c b/hw/net/allwinner_emac.c
index 469f2f0..91931ac 100644
--- a/hw/net/allwinner_emac.c
The value of pending register was updated only when an irq was raised
from a device; it should also be updated when an interrupt is cleared.
Signed-off-by: Beniamino Galvani
---
hw/intc/allwinner-a10-pic.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/hw/intc/allwinner-a10-pic.c b/hw/
The irq line status must be updated after writes to the INT_CTL and
INT_STA registers.
Signed-off-by: Beniamino Galvani
---
hw/net/allwinner_emac.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/hw/net/allwinner_emac.c b/hw/net/allwinner_emac.c
index 91931ac..d780ba0 100644
--- a/hw/ne
Convert the interrupt generation logic to the use of level triggered
interrupts.
Signed-off-by: Beniamino Galvani
---
hw/timer/allwinner-a10-pit.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/hw/timer/allwinner-a10-pit.c b/hw/timer/allwinner-a10-pit.c
index
This patch implements proper updating of the vector register which
should hold, according to the A10 user manual, the vector address for
the interrupt currently active on the CPU IRQ input.
Interrupt priority is not implemented at the moment and thus the first
pending interrupt is returned.
Signe
This implements the prescaler and source fields of the timer control
register as described in the A10 user manual.
Signed-off-by: Beniamino Galvani
---
hw/timer/allwinner-a10-pit.c | 19 ++-
1 file changed, 18 insertions(+), 1 deletion(-)
diff --git a/hw/timer/allwinner-a10-pi
This series introduces some fixes and missing features found while
trying to run mainline Linux kernel on emulated Allwinner A10.
Most of the changes concern interrupt handling, but there are also
improvements in the timer and the ethernet MAC.
With this applied I'm able to boot Linux 3.14-rc2 us
This adds more traces in the migration code.
Signed-off-by: Alexey Kardashevskiy
---
Debugging migration lately a lot and this patch helps quite a lot.
Would it make any use for anyone else? Thanks.
---
migration.c | 34 +++---
savevm.c | 14 --
tr
23.02.2014 21:02, Stefan Weil wrote:
> Today, most .o files depend on windows.h when QEMU is built with MinGW or
> MinGW-w64 (32 and 64 bit builds). These patches reduce the number of such
> files from more than 1800 to less than 180. They also allow removing some
> hacks which were needed because
27.02.2014 03:51, Peter Maydell пишет:
> The function qobject_from_json() doesn't actually allow its
> argument to be a format string -- it passes a NULL va_list*
> to qobject_from_jsonv(), and the parser code will then never
> actually intepret %-escape sequences (it tests whether the
> va_list po
27.02.2014 03:39, Peter Maydell wrote:
> clang doesn't support the __noclone__ attribute and emits a warning about
> it. Fortunately clang also implements a mechanism for asking if a particular
> attribute is implemented; use it. We assume that if the compiler doesn't
> support __has_attribute() th
26.02.2014 11:53, Sebastian Huber wrote:
> Ping.
There was another comment on this, about the first hunk,
see https://lists.gnu.org/archive/html/qemu-trivial/2014-02/msg00067.html
Note that the comment below this line mentions the +1 which is
being removed.
Thanks,
/mjt
20.02.2014 23:08, Stefan Weil wrote:
> It does not need qemu-common.h. Including exec/gdbstub.h fixes a warning
> from static code analyzers and avoids mismatching declarations for
> xml_builtin.
Applied to -trivial, thanks!
/mjt
Applied to -trivial, thanks!
/mjt
The QOM machine has both field per QemuOpt and an instance of
QEMUMachineInitArgs. Removed dupplications.
Signed-off-by: Marcel Apfelbaum
---
hw/core/machine.c | 18 +-
include/hw/boards.h | 9 +++--
vl.c| 6 +++---
3 files changed, 15 insertions(+), 18 de
The machine registration flow is refactored to use the QOM functionality.
Instead of linking the machines into a list, each machine has a type
and the types can be traversed in the QOM way.
Signed-off-by: Marcel Apfelbaum
---
include/hw/boards.h | 1 +
vl.c| 75 +
A NULL value is not added to visitor's stack, but there
is no check for that when the visitor tries to return
that value, leading to Qemu crash.
Reviewed-by: Eric Blake
Signed-off-by: Marcel Apfelbaum
---
qapi/qmp-output-visitor.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/qapi/qm
Sometimes is not enough to get property's value,
but it is needed to know if the value was actually set.
This is especially useful when querying bool properties
and having different defaults on different scenarios.
Signed-off-by: Marcel Apfelbaum
---
include/qom/object.h | 11 +++
qom/o
Most of the "Cc" list is due to patch 8: (Should I send each patch to a
different list?)
machine-opts: replace qemu_opt_get by QOM QemuMachine queries.
Status:
- machine_opts are mapped into QemuMachineState's properties,
which can be queried as regular QOM properties.
- Subclas
In order to allow attaching machine options to a machine instance,
current_machine is converted into QemuMachineState.
As a first step of deprecating QEMUMachine, some of the functions
were modified to return QemuMachineCLass.
Signed-off-by: Marcel Apfelbaum
---
device-hotplug.c| 4 +++-
in
It is cleaner to query current_machine's properties than
accessing QemuOpts from the code.
Signed-off-by: Marcel Apfelbaum
---
device_tree.c | 9 +
exec.c| 21 ++---
hw/arm/boot.c | 3 ++-
hw/microblaze/boot.c | 13 +++--
hw/ppc/
Filter out also 'type' property when setting
object's properties
Signed-off-by: Marcel Apfelbaum
---
vl.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/vl.c b/vl.c
index c4939ef..dc206e1 100644
--- a/vl.c
+++ b/vl.c
@@ -2766,7 +2766,8 @@ static int object_set_property(con
It is cleaner to query current_machine's properties than
accessing QemuOpts from the code.
Signed-off-by: Marcel Apfelbaum
---
hw/core/machine.c | 251 ++
1 file changed, 251 insertions(+)
diff --git a/hw/core/machine.c b/hw/core/machine.c
ind
The main functionality change is to convert QEMUMachine into QemuMachineClass
and QEMUMachineInitArgs into QemuMachineState, instance of QemuMachineClass.
As a first step, in order to make possible an incremental developement,
both QEMUMachine and QEMUMachineInitArgs are being embeded into the
new
Hello,
I'm very much interested in working on this idea for the upcoming GSoC.
Which part of the current qemu codebase does this idea correspond to??
and any links and documents that can further aid me??
Regards,
Prashanth
On Fri, 2014-02-28 at 11:37 -0500, Luiz Capitulino wrote:
> On Mon, 17 Feb 2014 20:01:42 +0200
> Marcel Apfelbaum wrote:
>
> > On Mon, 2014-02-17 at 10:38 -0700, Eric Blake wrote:
> > > On 02/17/2014 04:52 AM, Marcel Apfelbaum wrote:
> > > > A NULL value is not added to visitor's stack, but there
Il 02/03/2014 01:17, Gabriel L. Somlo ha scritto:
Although, on KVM, it's simply hardcoded to 0x14 rather than exposing to
the guest whatever the host CPU's apic version happens to be, or
trying to match it to the CPU model:
[somlo@foober kvm]$ grep -i version arch/x86/kvm/lapic.c
...
/* 14 is t
61 matches
Mail list logo
