On 22/05/12 15:52, Alexander Graf wrote:
On 22.05.2012, at 05:44, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 13:21, Alexander Graf wrote:
On 22.05.2012, at 04:02, Benjamin Herrenschmidt b...@kernel.crashing.org
wrote:
On Fri, 2012-05-18 at 15:12 +1000, Alexey
On 22.05.2012, at 08:11, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 15:52, Alexander Graf wrote:
On 22.05.2012, at 05:44, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 13:21, Alexander Graf wrote:
On 22.05.2012, at 04:02, Benjamin Herrenschmidt
VFIO PCI support will make use of these for user initiated
PCI config accesses.
Signed-off-by: Alex Williamson alex.william...@redhat.com
Acked-by: Bjorn Helgaas bhelg...@google.com
---
drivers/pci/access.c |6 --
drivers/pci/pci.h|7 ---
include/linux/pci.h |8
On 22.05.2012, at 08:11, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 15:52, Alexander Graf wrote:
On 22.05.2012, at 05:44, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 13:21, Alexander Graf wrote:
On 22.05.2012, at 04:02, Benjamin Herrenschmidt
VFIO is a secure user level driver for use with both virtual machines
and user level drivers. VFIO makes use of IOMMU groups to ensure the
isolation of devices in use, allowing unprivileged user access. It's
intended that VFIO will replace KVM device assignment and UIO drivers
(in cases where
On 22/05/12 16:31, Alexander Graf wrote:
On 22.05.2012, at 08:11, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 15:52, Alexander Graf wrote:
On 22.05.2012, at 05:44, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 13:21, Alexander Graf wrote:
On 22.05.2012, at
DMA transactions are tagged with the source ID of the device making
the request. Occasionally hardware screws this up and uses the
source ID of a different device (often the wrong function number of
a multifunction device). A specific Ricoh multifunction device is
a prime example of this problem
x86 is probably the wrong name for this VFIO IOMMU driver, but x86
is the primary target for it. This driver support a very simple
usage model using the existing IOMMU API. The IOMMU is expected to
support the full host address space with no special IOVA windows,
number of mappings restrictions,
On 22.05.2012, at 09:01, Alexey Kardashevskiy wrote:
On 22/05/12 16:31, Alexander Graf wrote:
On 22.05.2012, at 08:11, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 22/05/12 15:52, Alexander Graf wrote:
On 22.05.2012, at 05:44, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On
On Tue, 2012-05-22 at 14:34 +1000, Benjamin Herrenschmidt wrote:
The emulated devices can run simultaneously with the guest, so
we need to be careful with ordering of load and stores done by
them to the guest system memory, which need to be observed in
the right order by the guest operating
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
Documentation/vfio.txt | 315
1 files changed, 315 insertions(+), 0 deletions(-)
create mode 100644 Documentation/vfio.txt
diff --git a/Documentation/vfio.txt
Add back group support for AMD Intel. amd_iommu already tracks
devices and has init and uninit routines to manage groups.
intel-iommu does this on the fly, so we make use of the notifier
support built into iommu groups to create and remove groups.
Signed-off-by: Alex Williamson
In a PCI environment, transactions aren't always required to reach
the root bus before being re-routed. Intermediate switches between
an endpoint and the root bus can redirect DMA back downstream before
things like IOMMUs have a chance to intervene. Legacy PCI is always
susceptible to this as it
IOMMU groups allow IOMMU drivers to represent DMA visibility
and isolation of devices. Multiple devices may be grouped
together for the purposes of DMA. Placing a pointer on
struct device enable easy access for things like streaming
DMA programming and drivers like VFIO.
Signed-off-by: Alex
On Tue, 2012-05-22 at 09:13 +0200, Alexander Graf wrote:
On 22.05.2012, at 09:01, Alexey Kardashevskiy wrote:
This is internal kitchen of PCIDevice which I do not want to touch
from anywhere but pci.c. And
there is no fixup_capability or something.
Hrm. Maybe we should have one? :) Or
For returning errors out to non-PCI code. Re-name xen's version.
Signed-off-by: Alex Williamson alex.william...@redhat.com
Acked-by: Konrad Rzeszutek Wilk konrad.w...@oracle.com
---
drivers/xen/xen-pciback/conf_space.c |6 +++---
include/linux/pci.h | 26
Fill in many missing definitions and add sizeof fields for many
sections allowing for more extensive config parsing.
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
include/linux/pci_regs.h | 112 +-
1 files changed, 100 insertions(+),
Version 2 incorporating acks and feedback from v1. The PCI DMA quirk
and ACS check are reworked, sysfs iommu groups ABI Documentation
added as well as numerous other fixes, including patches from Alexey
Kardashevskiy towards supporting POWER usage of VFIO and IOMMU groups.
This series can be
Incorporate DMA quirking and ACS checking into amd_iommu and
intel-iommu. Note that IOMMU groups are not yet used for
streaming DMA, so this doesn't immediately solve the problems
with broken Ricoh devices. This a very strict implementation of
ACS checking, which will often result in
On 05/22/2012 02:34 AM, Andreas Färber wrote:
Am 21.05.2012 20:20, schrieb Blue Swirl:
On Mon, May 21, 2012 at 9:09 AM, Andreas Färber afaer...@suse.de wrote:
Am 14.05.2012 23:22, schrieb Blue Swirl:
On Mon, May 14, 2012 at 8:59 PM, Andreas Färber afaer...@suse.de wrote:
Am 14.05.2012 21:54,
IOMMU device groups are currently a rather vague associative notion
with assembly required by the user or user level driver provider to
do anything useful. This patch intends to grow the IOMMU group concept
into something a bit more consumable.
To do this, we first create an object representing
Signed-off-by: Fabien Chouteau chout...@adacore.com
---
MAINTAINERS |6 ++
1 file changed, 6 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index b45f075..3d773d3 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -387,6 +387,12 @@ M: Blue Swirl blauwir...@gmail.com
S: Maintained
F:
Am 21.05.2012 22:19, schrieb Corey Bryant:
libvirt's sVirt security driver provides SELinux MAC isolation for
Qemu guest processes and their corresponding image files. In other
words, sVirt uses SELinux to prevent a QEMU process from opening
files that do not belong to it.
sVirt provides
Am 21.05.2012 19:56, schrieb Blue Swirl:
On Mon, May 21, 2012 at 10:25 AM, Jim Meyering j...@meyering.net wrote:
Blue Swirl wrote:
On Tue, May 15, 2012 at 1:04 PM, j...@meyering.net wrote:
From: Jim Meyering meyer...@redhat.com
Without this, envlist_to_environ may silently fail to copy all
Am 17.05.2012 22:42, schrieb Josh Durgin:
Writeback caching was added in Ceph 0.46, and writethrough will be in
0.47. These are controlled by general config options, so there's no
need to check for librbd version.
Signed-off-by: Josh Durgin josh.dur...@inktank.com
Thanks, applied to the
Am 16.05.2012 12:16, schrieb Pavel Dovgaluk:
I use qemu under Windows and it has two windows when executes - console and
SDL ones.
When I close SDL window main loop function terminates correctly, and when I
close
console window to terminate qemu then the code after main loop is not
From: Jim Meyering meyer...@redhat.com
Per discussion, let's switch envlist.c to indent with spaces,
and then make the fix:
Jim Meyering (2):
envlist.c: convert many leading TABs to spaces via expand -i
envlist.c: handle strdup failure
envlist.c | 272
From: Jim Meyering meyer...@redhat.com
Signed-off-by: Jim Meyering meyer...@redhat.com
---
envlist.c | 256 +++---
1 file changed, 128 insertions(+), 128 deletions(-)
diff --git a/envlist.c b/envlist.c
index f2303cd..1d98108 100644
---
From: Jim Meyering meyer...@redhat.com
Without this, envlist_to_environ may silently fail to copy all
strings into the destination buffer, and both callers would leak
any env strings allocated after a failing strdup, because the
freeing code stops at the first NULL pointer.
Signed-off-by: Jim
Kevin Wolf wrote:
A patch replacing tabs by spaces isn't really the kind of patches that
we would want to avoid during freeze. It's easy enough to check with git
diff -w that it doesn't change anything semantically.
That makes sense, so I've posted two patches:
1) two patches: one
Hi Stefan
On Mon, May 21, 2012 at 11:20 PM, Stefan Hajnoczi stefa...@gmail.com wrote:
On Mon, May 21, 2012 at 12:00 PM, ronnie sahlberg
ronniesahlb...@gmail.com wrote:
Yes, I use IDE since I boot from this LUN.
I just managed to track it down to the IDE changes.
It looks like basically
Hi,
Now that I see what happens, I can easily workaround this in block/iscsi.c
by the patch below, but I dont know if this is the right thing to do.
It does appear that here, when I use qemu_set_fd_handler() and add a
handler for writeble it takes 55ms before the event system notices
this and
Il 17/05/2012 22:42, Josh Durgin ha scritto:
+ * Fallback to more conservative semantics if setting cache
+ * options fails. Ignore errors from setting rbd_cache because the
+ * only possible error is that the option does not exist, and
+ * librbd defaults to no caching. If
Il 22/05/2012 10:51, Kevin Wolf ha scritto:
Am 16.05.2012 12:16, schrieb Pavel Dovgaluk:
I use qemu under Windows and it has two windows when executes - console and
SDL ones.
When I close SDL window main loop function terminates correctly, and when I
close
console window to terminate qemu
On Sat, May 19, 2012 at 12:14 AM, Shirley Ma mashi...@us.ibm.com wrote:
On Thu, 2012-05-17 at 17:20 +0800, Liu Ping Fan wrote:
Currently, the guest can not know the NUMA info of the vcpu, which
will
result in performance drawback.
This is the discovered and experiment by
Shirley Ma
Il 22/05/2012 11:15, ronnie sahlberg ha scritto:
Hi,
Now that I see what happens, I can easily workaround this in block/iscsi.c
by the patch below, but I dont know if this is the right thing to do.
It does appear that here, when I use qemu_set_fd_handler() and add a
handler for writeble
Am 22.05.2012 11:05, schrieb Jim Meyering:
Kevin Wolf wrote:
A patch replacing tabs by spaces isn't really the kind of patches that
we would want to avoid during freeze. It's easy enough to check with git
diff -w that it doesn't change anything semantically.
That makes sense, so I've posted
On Tue, May 22, 2012 at 7:29 PM, Paolo Bonzini pbonz...@redhat.com wrote:
Il 22/05/2012 11:15, ronnie sahlberg ha scritto:
Hi,
Now that I see what happens, I can easily workaround this in block/iscsi.c
by the patch below, but I dont know if this is the right thing to do.
It does appear that
From: Jim Meyering meyer...@redhat.com
This is the same as v1, except that two lines of non-leading TABs
in envlist.c (indenting comments after code) have also been converted
to use equivalent spaces instead of TABs.
Jim Meyering (2):
envlist.c: convert all TABs to equivalent spaces
From: Jim Meyering meyer...@redhat.com
Without this, envlist_to_environ may silently fail to copy all
strings into the destination buffer, and both callers would leak
any env strings allocated after a failing strdup, because the
freeing code stops at the first NULL pointer.
Signed-off-by: Jim
From: Jim Meyering meyer...@redhat.com
Signed-off-by: Jim Meyering meyer...@redhat.com
---
envlist.c | 256 +++---
1 file changed, 128 insertions(+), 128 deletions(-)
diff --git a/envlist.c b/envlist.c
index f2303cd..be0addb 100644
---
Kevin Wolf wrote:
Am 22.05.2012 11:05, schrieb Jim Meyering:
Kevin Wolf wrote:
A patch replacing tabs by spaces isn't really the kind of patches that
we would want to avoid during freeze. It's easy enough to check with git
diff -w that it doesn't change anything semantically.
That makes
Hi Peter
On 05/22/2012 07:45 AM, Peter Crosthwaite wrote:
Hi Michael,
The microblaze linux kernel is available at:
http://wiki.xilinx.com/
If you wish to do your own kernel config from scratch. There is some
brief documentation there on how to do a bringup.
The qemu port for mb
On 22 May 2012 10:50, Jim Meyering j...@meyering.net wrote:
From: Jim Meyering meyer...@redhat.com
If we're going to go to the effort of a complete reindent
patch we should actually reindent to the QEMU coding style
standard, which is four-space, not eight.
-- PMM
Otherwise, If we add an event for -is-writeable but the socket is already
writeable there may be a short delay before the event callback is actually
triggered.
Those delays would in particular hurt performance during BIOS boot and when the
GRUB bootloader reads the kernel and initrd.
None of the other drivers in block/*.c call qemu_notify_event()
Do you you think those should be audited and have this call added to
where required ?
regards
ronnie sahlberg
On Tue, May 22, 2012 at 7:48 PM, ronnie sahlberg
ronniesahlb...@gmail.com wrote:
On Tue, May 22, 2012 at 7:29 PM, Paolo
On Mon, May 21, 2012 at 5:54 PM, Erik Rull webmas...@rdsoftware.de wrote:
is there a summary existing that shows up the rough or actual differences
between qemu --enable-kvm and qemu-kvm? I tested both versions with the same
compile and start options, the CPU performance results are identical,
This allows using LUNs bigger than 2TB.
Signed-off-by: Ronnie Sahlberg ronniesahlb...@gmail.com
---
block/iscsi.c | 101
trace-events |4 +-
2 files changed, 74 insertions(+), 31 deletions(-)
diff --git a/block/iscsi.c
On 11.05.2012, at 16:42, Jan Kiszka wrote:
v3 is v2 - xhci changes as MSI is instable there anyway. So, only
patches 1 and 2 are for 1.1/stable now.
Just saw these in my inbox as unanswered, so mst: ping? :)
Alex
From: Jim Meyering meyer...@redhat.com
Same as v2, but now with TABs converted using expand --tabs=4.
Jim Meyering (2):
envlist.c: convert each TAB(width-4) to equivalent spaces
envlist.c: handle strdup failure
envlist.c | 272 --
From: Jim Meyering meyer...@redhat.com
Signed-off-by: Jim Meyering meyer...@redhat.com
---
envlist.c | 256 +++---
1 file changed, 128 insertions(+), 128 deletions(-)
diff --git a/envlist.c b/envlist.c
index f2303cd..e44889b 100644
---
On 11.05.2012, at 16:42, Jan Kiszka wrote:
Properly register reset functions via the device class.
CC: Alexander Graf ag...@suse.de
Signed-off-by: Jan Kiszka jan.kis...@siemens.com
Looks good to me, but I'm no infrastructure expert :)
Reviewed-by: Alexander Graf ag...@suse.de
Alex
From: Jim Meyering meyer...@redhat.com
Add comments so no one else will be tempted to reduce the scope
of this global variable.
Signed-off-by: Jim Meyering meyer...@redhat.com
---
tcg/tcg.c | 10 +++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/tcg/tcg.c b/tcg/tcg.c
Peter Maydell wrote:
On 21 May 2012 21:10, Jim Meyering j...@meyering.net wrote:
Peter Maydell wrote:
On 21 May 2012 20:51, Jim Meyering j...@meyering.net wrote:
From: Jim Meyering meyer...@redhat.com
Signed-off-by: Jim Meyering meyer...@redhat.com
---
tcg/tcg.c | 2 +-
1 file changed,
Peter Maydell wrote:
On 22 May 2012 10:50, Jim Meyering j...@meyering.net wrote:
From: Jim Meyering meyer...@redhat.com
If we're going to go to the effort of a complete reindent
patch we should actually reindent to the QEMU coding style
standard, which is four-space, not eight.
Good point.
On 11.05.2012, at 16:42, Jan Kiszka wrote:
Call msi_reset on device reset as still required by the core.
CC: Alexander Graf ag...@suse.de
CC: qemu-sta...@nongnu.org
Signed-off-by: Jan Kiszka jan.kis...@siemens.com
Acked-by: Alexander Graf ag...@suse.de
Alex
On 2012-05-22 07:04, Stefan Hajnoczi wrote:
On Mon, May 21, 2012 at 5:54 PM, Erik Rull webmas...@rdsoftware.de wrote:
is there a summary existing that shows up the rough or actual differences
between qemu --enable-kvm and qemu-kvm? I tested both versions with the same
compile and start
Moving code related to CPU creation and initialization internal parts
from board level into apic and cpu objects will allow X86CPU to better
model QOM object life-cycle.
It will allow to create X86CPU as any other object by creating it with
object_new() then setting properties and then calling
it's probably intermidiate step till cpu modeled as
sub-classes. After then we probably could drop it.
However it still could be used for overiding default
cpu subclasses definition, and probably renamed to
something like 'features'.
Signed-off-by: Igor Mammedov imamm...@redhat.com
---
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
qapi/qapi-visit-core.h |4
qapi/qapi-visit-core.c |7 +++
scripts/qapi.py|2 +-
3 files changed, 12 insertions(+), 1 deletions(-)
diff --git a/qapi/qapi-visit-core.h b/qapi/qapi-visit-core.h
index e850746..bab2fec
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net/tap.h |2 +-
net/tap-aix.c |2 +-
net/tap-bsd.c |2 +-
net/tap-haiku.c |2 +-
net/tap-linux.c |9 +++-
net/tap-solaris.c |2 +-
net/tap-win32.c | 11 +++--
net/tap.c | 111
On 2012-05-22 07:35, Igor Mammedov wrote:
Move from apic_init in pc.c the code that belongs to apic_init_common
and create/init apic in pc_new_cpu instead of separate func.
Signed-off-by: Igor Mammedov imamm...@redhat.com
---
hw/apic_common.c | 16
hw/msi.h |
On 2012-05-22 07:35, Igor Mammedov wrote:
Move from apic_init in pc.c the code that belongs to apic_init_common
and create/init apic in pc_new_cpu instead of separate func.
Signed-off-by: Igor Mammedov imamm...@redhat.com
---
hw/apic_common.c | 16
hw/msi.h |
On 2012-05-22 07:35, Igor Mammedov wrote:
Make CPU creation/initialization consistent with QOM object
behavior in this, by moving tcg and apic initialization from board
level into CPU's initfn/realize calls and cpu_model property setter.
Which makes CPU object self-sufficient in respect of
Move from apic_init in pc.c the code that belongs to apic_init_common
and create/init apic in pc_new_cpu instead of separate func.
Signed-off-by: Igor Mammedov imamm...@redhat.com
---
hw/apic_common.c | 16
hw/msi.h |2 ++
hw/pc.c | 47
NetdevTapOptions::sndbuf and NetdevDumpOptions::len use the new size
type.
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
qapi-schema.json | 275 ++
1 files changed, 275 insertions(+), 0 deletions(-)
diff --git a/qapi-schema.json
On 22 May 2012 11:35, Igor Mammedov imamm...@redhat.com wrote:
From: Igor Mammedov niall...@gmail.com
MP initialization protocol differs between cpu families, and for P6 and
onward models it is up to CPU to decide if it will be BSP using this
protocol, so try to model this. However there is
If you start guest with floppy drive but without media inserted,
the guest now can see floppy drive.
Signed-off-by: Pavel Hrdina phrd...@redhat.com
---
hw/pc.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index 4d34a33..967c17a 100644
--- a/hw/pc.c
This patch series fixes handling of FDC when you don't have media inserted.
Guest should see floppy drive if you start guest without media and should
detect that there is no media in drive.
Signed-off-by: Pavel Hrdina phrd...@redhat.com
Pavel Hrdina (2):
fdc: floppy drive should be visible
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net/tap.c | 23 ---
1 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/net/tap.c b/net/tap.c
index 7501eba..fdaab2b 100644
--- a/net/tap.c
+++ b/net/tap.c
@@ -512,21 +512,22 @@ static int
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net/slirp.c | 93 ---
1 files changed, 25 insertions(+), 68 deletions(-)
diff --git a/net/slirp.c b/net/slirp.c
index 9b925b7..166304c 100644
--- a/net/slirp.c
+++ b/net/slirp.c
@@
The net_client_init() prototype is kept intact.
Based on is_netdev, the QemuOpts-rooted QemuOpt-list is parsed as a
Netdev or a NetLegacy. The original meat of net_client_init() is moved to
and simplified in net_client_init1():
Fields not common between -net and -netdev are clearly separated:
We have to set up 'media_changed' after guest start so floppy driver
could detect that there is no media in drive. For this purpose we call
'fdctrl_change_cb' instead of 'fd_revalidate' in 'fdctrl_connect_drives'.
'fd_revalidate' is called inside 'fdctrl_change_cb'.
In 'fdctrl_handle_seek' we
This visitor supports parsing
-option [type=]discriminator[,optarg1=val1][,optarg2=val2][,...]
style QemuOpts objects into native C structures. After defining the type
tree in the qapi schema (see below), a root type traversal with this
visitor linked to the underlying QemuOpts object will
Inspired by [1], the first half of this series attempts to implement a new
visitor that should clean up defining and processing command line options.
For a more detailed description, please see [PATCH 04/16] qapi: introduce
OptsVisitor.
The second half converts -net/-netdev parsing to the new
Make CPU creation/initialization consistent with QOM object
behavior in this, by moving tcg and apic initialization from board
level into CPU's initfn/realize calls and cpu_model property setter.
Which makes CPU object self-sufficient in respect of creation/initialization
and matches a typical
On Tue, May 22, 2012 at 05:17:39PM +1000, Benjamin Herrenschmidt wrote:
On Tue, 2012-05-22 at 14:34 +1000, Benjamin Herrenschmidt wrote:
The emulated devices can run simultaneously with the guest, so
we need to be careful with ordering of load and stores done by
them to the guest system
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
qapi-schema.json | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/qapi-schema.json b/qapi-schema.json
index bb1f806..a8b3803 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -315,7 +315,7 @@
# @CPU: the
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net/dump.c | 21 +
1 files changed, 17 insertions(+), 4 deletions(-)
diff --git a/net/dump.c b/net/dump.c
index 27e9528..a1de09e 100644
--- a/net/dump.c
+++ b/net/dump.c
@@ -144,22 +144,35 @@ static int
On 2012-05-21 03:50, Wen Congyang wrote:
When the guest is panicked, it will write 0x1 to the port 0x505. So if
qemu reads 0x1 from this port, we can do the folloing three things
according to the parameter -onpanic:
1. emit QEVENT_GUEST_PANICKED only
2. emit QEVENT_GUEST_PANICKED and pause VM
NET_CLIENT_TYPE_ - NET_CLIENT_OPTIONS_KIND_
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net.h | 16 +-
hw/cadence_gem.c|2 +-
hw/dp8393x.c|2 +-
hw/e1000.c |2 +-
hw/eepro100.c |2 +-
From: Igor Mammedov niall...@gmail.com
MP initialization protocol differs between cpu families, and for P6 and
onward models it is up to CPU to decide if it will be BSP using this
protocol, so try to model this. However there is no point in implementing
MP initialization protocol in qemu. Thus
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net/dump.h |5 ++---
net/slirp.h |5 ++---
net/socket.h|5 ++---
net/tap.h |9 -
net/vde.h |5 ++---
net.c | 14 ++
net/dump.c |4 ++--
net/slirp.c |4
On Tue, 2012-05-22 at 14:14 +0300, Michael S. Tsirkin wrote:
The baseline is the laptop without kvm talking to the server. The
TCP_STREAM test results are:
It's not a good test. The thing most affecting throughput results is
how
much CPU does you guest get. So as a minumum you need to
From: Paolo Bonzini pbonz...@redhat.com
Don't overwrite / leak previously set errors.
Don't try to end a container that could not be started.
Signed-off-by: Paolo Bonzini pbonz...@redhat.com
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
error.h|4 +-
error.c
I reverse engineered the following permissions between the -socket
sub-options:
fd listen connect mcast udp | localaddr
fd x . .. . | .
listen . x .. . | .
connect. . x. . | .
mcast
From: Jim Meyering meyer...@redhat.com
Without this, envlist_to_environ may silently fail to copy all
strings into the destination buffer, and both callers would leak
any env strings allocated after a failing strdup, because the
freeing code stops at the first NULL pointer.
Signed-off-by: Jim
Moving reset callback into cpu object from board level will allow
properly create object during run-time (hotplug).
When reset over QOM hierarchy is implemented, this reset callback
should be removed.
Signed-off-by: Igor Mammedov imamm...@redhat.com
---
hw/pc.c |7 ---
The only clients should be the existent qemu-option.c, and the upcoming
qapi/opts-visitor.c.
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
qemu-option-internal.h | 53
qemu-option.c | 24 +
2 files changed, 54
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net.c | 39 ++-
1 files changed, 22 insertions(+), 17 deletions(-)
diff --git a/net.c b/net.c
index 5ac5cf0..bd2fd23 100644
--- a/net.c
+++ b/net.c
@@ -748,12 +748,15 @@ int net_handle_fd_param(Monitor *mon,
Am 22.05.2012 12:59, schrieb Pavel Hrdina:
We have to set up 'media_changed' after guest start so floppy driver
could detect that there is no media in drive. For this purpose we call
'fdctrl_change_cb' instead of 'fd_revalidate' in 'fdctrl_connect_drives'.
'fd_revalidate' is called inside
On 05/22/2012 02:18 AM, Kevin Wolf wrote:
This patch series adds the -filefd command-line option and the
getfd_file monitor command. This will enable libvirt to open a
file and push the corresponding filename and file descriptor to
QEMU. When QEMU needs to open a file, it will first check
On Tue, May 22, 2012 at 09:41:41PM +1000, Benjamin Herrenschmidt wrote:
On Tue, 2012-05-22 at 14:14 +0300, Michael S. Tsirkin wrote:
The baseline is the laptop without kvm talking to the server. The
TCP_STREAM test results are:
It's not a good test. The thing most affecting throughput
Signed-off-by: Laszlo Ersek ler...@redhat.com
---
net/vde.c | 37 +
1 files changed, 29 insertions(+), 8 deletions(-)
diff --git a/net/vde.c b/net/vde.c
index 8e60f68..35e8113 100644
--- a/net/vde.c
+++ b/net/vde.c
@@ -110,20 +110,41 @@ static int
Am 22.05.2012 14:02, schrieb Eric Blake:
On 05/22/2012 02:18 AM, Kevin Wolf wrote:
This patch series adds the -filefd command-line option and the
getfd_file monitor command. This will enable libvirt to open a
file and push the corresponding filename and file descriptor to
QEMU. When QEMU
Il 22/05/2012 12:10, Ronnie Sahlberg ha scritto:
This allows using LUNs bigger than 2TB.
Applied to scsi-next for 1.2, thanks.
Paolo
Signed-off-by: Ronnie Sahlberg ronniesahlb...@gmail.com
---
block/iscsi.c | 101
trace-events |
Il 22/05/2012 11:56, Ronnie Sahlberg ha scritto:
Otherwise, If we add an event for -is-writeable but the socket is already
writeable there may be a short delay before the event callback is actually
triggered.
Those delays would in particular hurt performance during BIOS boot and when
the
Historically for QEMU we haven't implemented TrustZone support even
though we claim to emulate CPUs that provide it. Instead we provide a
CPU which mostly looks like a variant of the real thing without the
TrustZone feature. We then bolt on a few extra cp15 registers (eg the
SCR) as a pragmatic
From: Jim Meyering meyer...@redhat.com
Signed-off-by: Jim Meyering meyer...@redhat.com
---
hw/scsi-bus.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/scsi-bus.c b/hw/scsi-bus.c
index 8ab9bcd..f10f3ec 100644
--- a/hw/scsi-bus.c
+++ b/hw/scsi-bus.c
@@ -1561,7 +1561,7
The following changes since commit 76ee152a86d5f2533443ce4d2be6fe253cfb3c45:
Update version to 1.1.0-rc2 (2012-05-14 17:56:50 -0500)
are available in the git repository at:
git://github.com/bonzini/qemu.git scsi-next
for you to fetch changes up to e1a2d34f4abd8a117f5c5a25a5bb2e67d597de23:
1 - 100 of 278 matches
Mail list logo