-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 24/09/2015 01:54, David Gibson wrote:
> On Wed, Sep 23, 2015 at 08:55:01PM +0200, Laurent Vivier wrote:
>>
>>
>> On 17/09/2015 15:09, David Gibson wrote:
>>> At present the PCI host bridge (PHB) for the pseries machine
>>> type has a fixed DMA w
Ping? (also ui, usb, qxl)
Markus Armbruster writes:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> This
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 24/09/2015 03:02, David Gibson wrote:
> On Wed, Sep 23, 2015 at 06:46:17PM +0200, Laurent Vivier wrote:
>>
>>
>> On 17/09/2015 15:09, David Gibson wrote:
>>> Currently the pseries machine type uses two types of PCI Host
>>> Bridge (PHB) devices:
Ping?
Markus Armbruster writes:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> This commit only touches
Ping?
Markus Armbruster writes:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> This commit only touches
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 24/09/2015 01:50, David Gibson wrote:
> On Wed, Sep 23, 2015 at 07:04:55PM +0200, Laurent Vivier wrote:
>>
>>
>> On 17/09/2015 15:09, David Gibson wrote:
>>> When we have guest visible IOMMUs, we allow notifiers to be
>>> registered which will b
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Am 24.09.2015 um 09:06 schrieb Markus Armbruster:
> Ping?
>
> Markus Armbruster writes:
>
>> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also
>> safer, for two reasons. One, it catches multiplication
>> overflowing size_t. Two, it
Move the creation of the dump client from net_dump_init() into
net_init_dump(), so we can later use the former function for
dump via netfilter, too. Also rename net_dump_init() to
net_dump_state_init() to make it easier distinguishable from
net_init_dump().
Signed-off-by: Thomas Huth
---
net/dum
Add a short description for the filter-dump command line options.
Signed-off-by: Thomas Huth
---
qemu-options.hx | 8
1 file changed, 8 insertions(+)
diff --git a/qemu-options.hx b/qemu-options.hx
index b09f97f..347ee28 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -2011,6 +201
Adding a proper receive_iov function to the net dump module.
This will make it easier to support the dump filter feature for
the -netdev option in later patches.
Signed-off-by: Thomas Huth
---
net/dump.c | 24 +---
1 file changed, 21 insertions(+), 3 deletions(-)
diff --git
The "-net dump" option only works with the "-net" option. So far, it
is not possible to dump network traffic with the "-netdev" option yet.
This patch series now fixes this ugliness by enabling dumping for the
netdev devices via the new netfilter infrastructure from Yang Hongyang
(integration into
With the upcoming dumping-via-netfilter patch, the DumpState
should not be related to NetClientState anymore, so move the
related information to a new struct called DumpNetClient.
Signed-off-by: Thomas Huth
---
net/dump.c | 74 +-
1 fil
Use the net filter infrastructure to provide the dumping
functions for netdev devices, too.
Signed-off-by: Thomas Huth
---
net/dump.c | 129 -
vl.c | 8 +++-
2 files changed, 134 insertions(+), 3 deletions(-)
diff --git a/net/d
This has finally reached the front of my review queue. I apologize for
the lng delay.
Copying Paolo for another pair of eyeballs (he wrote this code).
Yang Hongyang writes:
> When delete an object, we need to delete the associated qemu opts,
> otherwise, we can not add another object with
On Mon, 21 Sep 2015 10:09:34 +0200
Greg Kurz wrote:
> On Sat, 19 Sep 2015 15:34:02 +0530
> "Aneesh Kumar K.V" wrote:
>
> > Greg Kurz writes:
> >
> > > In a cross-endian setup, the virtio-9p device has state in
> > > @device_endian. It
> > > must be migrated.
> > >
> > > Signed-off-by: Greg K
Hi Kevin,
I tried the patch you provide, and I haven't seen that problem yet. If the disk
space is full, an error will be reported with the message "Invalid argument"
and the program will stop.
Will you merge the patch to the master?
Thanks.
Guangmu Zhu
diff --git a/block/raw-win32.c b/bl
The filter interface part looks good to me, thanks!
On 09/24/2015 03:22 PM, Thomas Huth wrote:
Use the net filter infrastructure to provide the dumping
functions for netdev devices, too.
Signed-off-by: Thomas Huth
---
net/dump.c | 129 +
Introduce a set of new runtime helpers do handle exclusive instructions.
This helpers are used as hooks to call the respective LL/SC helpers in
softmmu_llsc_template.h from TCG code.
Suggested-by: Jani Kokkonen
Suggested-by: Claudio Fontana
Signed-off-by: Alvise Rigo
---
target-arm/helper.h
The new helpers rely on the legacy ones to perform the actual read/write.
The LoadLink helper (helper_ldlink_name) prepares the way for the
following SC operation. It sets the linked address and the size of the
access.
These helper also update the TLB entry of the page involved in the
LL/SC for th
This is the fifth iteration of the patch series which applies to the
upstream branch of QEMU (v2.4.0).
Changes versus previous versions are at the bottom of this cover letter.
The code is also available at following repository:
https://git.virtualopensystems.com/dev/qemu-mt.git
branch:
slowpath-f
Use the new LL/SC runtime helpers to handle the ARM atomic
instructions in softmmu_llsc_template.h.
In general, the helper generator
gen_helper_{ldlink,stcond}_aa32_i{8,16,32,64}() calls the function
helper_{le,be}_{ldlink,stcond}{ub,uw,ulq}_mmu() implemented in
softmmu_llsc_template.h.
Suggested
The purpose of this new bitmap is to flag the memory pages that are in
the middle of LL/SC operations (after a LL, before a SC) on a per-vCPU
basis.
For all these pages, the corresponding TLB entries will be generated
in such a way to force the slow-path if at least one vCPU has the bit
not set.
Wh
On Tue, 09/22 15:09, John Snow wrote:
>
>
> On 09/21/2015 10:46 PM, Fam Zheng wrote:
> > Sometimes block jobs must execute as a transaction group. Finishing
> > jobs wait until all other jobs are ready to complete successfully.
> > Failure or cancellation of one job cancels the other jobs in the
Add a new TLB flag to force all the accesses made to a page to follow
the slow-path.
In the case we remove a TLB entry marked as EXCL, we unset the
corresponding exclusive bit in the bitmap.
Suggested-by: Jani Kokkonen
Suggested-by: Claudio Fontana
Signed-off-by: Alvise Rigo
---
cputlb.c
Use the new slow path for atomic instruction translation when the
softmmu is enabled.
Suggested-by: Jani Kokkonen
Suggested-by: Claudio Fontana
Signed-off-by: Alvise Rigo
---
configure | 11 +++
1 file changed, 11 insertions(+)
diff --git a/configure b/configure
index cd219d8..5f72977
On 09/24/2015 03:43 PM, Markus Armbruster wrote:
This has finally reached the front of my review queue. I apologize for
the lng delay.
Copying Paolo for another pair of eyeballs (he wrote this code).
[...]
+
+opts = qemu_opts_find(qemu_find_opts_err("object", NULL), id);
+qemu_op
Hello!
Thank you very much for your support and cooperation. I am back from my
vacation and continuing my
work on live migration. Actually i already have working code, but need to
settle down kernel API
first.
Kind regards,
Pavel Fedin
Expert Engineer
Samsung Electronics Research center Russi
Yang Hongyang writes:
> Add a netfilter object based on QOM.
>
> A netfilter is attached to a netdev, captures all network packets
> that pass through the netdev. When we delete the netdev, we also
> delete the netfilter object attached to it, because if the netdev is
> removed, the filter which
From: Peter Lieven
This makes sure the number of reallocs is in O(log N).
Signed-off-by: Peter Lieven
[ rebased to io/buffer.c ]
Signed-off-by: Gerd Hoffmann
---
io/buffer.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/io/buffer.c b/io/buffer.c
index 68ae68d..6aa0
Signed-off-by: Gerd Hoffmann
---
include/io/buffer.h | 12
io/buffer.c | 11 +++
2 files changed, 23 insertions(+)
diff --git a/include/io/buffer.h b/include/io/buffer.h
index 2b1b261..cffad19 100644
--- a/include/io/buffer.h
+++ b/include/io/buffer.h
@@ -34,12 +34,2
Signed-off-by: Gerd Hoffmann
---
include/io/buffer.h | 10 ++
io/buffer.c | 16
2 files changed, 26 insertions(+)
diff --git a/include/io/buffer.h b/include/io/buffer.h
index 1dddc73..5676aff 100644
--- a/include/io/buffer.h
+++ b/include/io/buffer.h
@@ -137,4 +1
Signed-off-by: Gerd Hoffmann
---
ui/vnc-jobs.c | 1 +
ui/vnc.c | 16
2 files changed, 17 insertions(+)
diff --git a/ui/vnc-jobs.c b/ui/vnc-jobs.c
index 9824c34..7a234da 100644
--- a/ui/vnc-jobs.c
+++ b/ui/vnc-jobs.c
@@ -303,6 +303,7 @@ static VncJobQueue *vnc_queue_init(vo
Hi,
Here is a patch series to improve the vnc buffer handling. It picks up
the qio_buffer patches from Daniel, adds move calls (move data from one
buffer to another) and tracing, makes vnc use the new features. Net
effect should be that (a) vnc copies less data around and (b) buffers
don't gro
Signed-off-by: Gerd Hoffmann
---
io/buffer.c | 10 ++
trace-events | 6 ++
2 files changed, 16 insertions(+)
diff --git a/io/buffer.c b/io/buffer.c
index 96077d3..05425c2 100644
--- a/io/buffer.c
+++ b/io/buffer.c
@@ -19,6 +19,7 @@
*/
#include "io/buffer.h"
+#include "trace.h"
Signed-off-by: Gerd Hoffmann
---
Makefile| 2 ++
Makefile.objs | 5 +
Makefile.target | 2 ++
3 files changed, 9 insertions(+)
diff --git a/Makefile b/Makefile
index 8ec9b69..ca18209 100644
--- a/Makefile
+++ b/Makefile
@@ -157,6 +157,7 @@ dummy := $(call unnest-vars,, \
From: "Daniel P. Berrange"
The Buffer code in the VNC server is useful for the IO channel
code, so pull it out into a shared module, QIOBuffer.
Signed-off-by: Daniel P. Berrange
---
include/io/buffer.h | 118
io/Makefile.objs| 1 +
io/
Signed-off-by: Gerd Hoffmann
---
include/io/buffer.h | 10 ++
io/buffer.c | 14 ++
2 files changed, 24 insertions(+)
diff --git a/include/io/buffer.h b/include/io/buffer.h
index cffad19..1dddc73 100644
--- a/include/io/buffer.h
+++ b/include/io/buffer.h
@@ -127,4 +127
Signed-off-by: Gerd Hoffmann
---
ui/vnc-jobs.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/ui/vnc-jobs.c b/ui/vnc-jobs.c
index 7a234da..50e6b37 100644
--- a/ui/vnc-jobs.c
+++ b/ui/vnc-jobs.c
@@ -54,7 +54,6 @@ struct VncJobQueue {
QemuCond cond;
QemuMutex mutex;
QemuThr
On 09/24/2015 04:41 PM, Markus Armbruster wrote:
Yang Hongyang writes:
Add a netfilter object based on QOM.
A netfilter is attached to a netdev, captures all network packets
that pass through the netdev. When we delete the netdev, we also
delete the netfilter object attached to it, because
Signed-off-by: Gerd Hoffmann
---
ui/vnc-jobs.c | 10 +++---
1 file changed, 3 insertions(+), 7 deletions(-)
diff --git a/ui/vnc-jobs.c b/ui/vnc-jobs.c
index 50e6b37..dfc5139 100644
--- a/ui/vnc-jobs.c
+++ b/ui/vnc-jobs.c
@@ -165,8 +165,7 @@ void vnc_jobs_consume_buffer(VncState *vs)
v
On 09/24/2015 04:41 PM, Markus Armbruster wrote:
Yang Hongyang writes:
Add a netfilter object based on QOM.
A netfilter is attached to a netdev, captures all network packets
that pass through the netdev. When we delete the netdev, we also
delete the netfilter object attached to it, because if
On 23/09/15 14:14, Laurent Vivier wrote:
> When DT node names for PCI devices are generated by SLOF,
> they are generated according to the type of the device
> (for instance, ethernet for virtio-net-pci device).
>
> Node name for hotplugged devices is generated by QEMU.
> This patch adds the mecha
Add callbacks for opengl context management and scanout texture
configuration to DisplayChangeListenerOps.
Signed-off-by: Gerd Hoffmann
Reviewed-by: Marc-André Lureau
---
include/ui/console.h | 37 +
ui/console.c | 67 +
Optimizing updates by copying the dirty rectangle
only do not work because of double-buffering.
Signed-off-by: Gerd Hoffmann
Reviewed-by: Marc-André Lureau
Reviewed-by: Max Reitz
---
ui/sdl2-2d.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/ui/sdl2-2d.c b/ui/sdl2-2d.c
ind
Signed-off-by: Gerd Hoffmann
---
hw/display/virtio-gpu.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c
index a67d927..73bd9b6 100644
--- a/hw/display/virtio-gpu.c
+++ b/hw/display/virtio-gpu.c
@@ -563,7 +563,6 @@ int virtio
Hi,
This patch series adds infrastructure for opengl rendering
(context management, define scanout textures), adds support
for this to sdl2 and gtk user interfaces and adds virgl/3d
mode to virtio-gpu by hooking up the virglrenderer library.
v2 fixes a bunch of issues found in patch review, whi
Add helper functions to manage opengl contexts using egl.
Signed-off-by: Gerd Hoffmann
Reviewed-by: Marc-André Lureau
---
include/ui/egl-context.h | 14 ++
ui/Makefile.objs | 1 +
ui/egl-context.c | 34 ++
3 files changed, 49 insertio
This allows virtio-gpu to render in 3d mode.
Uses egl, for gtk versions 3.14 and older.
Signed-off-by: Gerd Hoffmann
Reviewed-by: Marc-André Lureau
---
include/ui/gtk.h | 16 +++
ui/gtk-egl.c | 131 +++
ui/gtk.c | 7 +++
3 f
Create a buffer for the vertex data and place vertexes
there at initialization time. Then just use the buffer
for each texture blit.
Signed-off-by: Gerd Hoffmann
Reviewed-by: Marc-André Lureau
---
include/ui/shader.h | 4 +++-
ui/console-gl.c | 7 ++-
ui/shader.c | 31 +++
This allows virtio-gpu to render in 3d mode.
Signed-off-by: Gerd Hoffmann
---
include/ui/sdl2.h | 22 -
ui/sdl2-gl.c | 135 ++
ui/sdl2.c | 7 +++
3 files changed, 163 insertions(+), 1 deletion(-)
diff --git a/include/ui
---
hw/display/virtio-gpu-pci.c | 4 ++--
hw/display/virtio-gpu.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/hw/display/virtio-gpu-pci.c b/hw/display/virtio-gpu-pci.c
index 5bc62cf..eef137f 100644
--- a/hw/display/virtio-gpu-pci.c
+++ b/hw/display/virtio-gpu-pci.c
@
Sync with linux kernel headers with virgl/3d patches applied.
Signed-off-by: Gerd Hoffmann
Reviewed-by: Marc-André Lureau
---
include/standard-headers/linux/virtio_gpu.h | 112 +++-
1 file changed, 111 insertions(+), 1 deletion(-)
diff --git a/include/standard-headers/l
Add virglrenderer library detection. Add 3d mode to virtio-gpu,
wire up virglrenderer library. When in 3d mode render using the
new context management and texture scanout callbacks.
Signed-off-by: Gerd Hoffmann
---
configure | 32 +++
hw/display/Makefile.objs | 6
This allows virtio-gpu to render in 3d mode.
Uses native opengl support which is present
in gtk versions 3.16 and newer.
Signed-off-by: Gerd Hoffmann
---
configure| 8 ++
include/ui/gtk.h | 23 ++
ui/Makefile.objs | 5 ++
ui/gtk-gl-area.c | 223 ++
Yang Hongyang writes:
> This filter is to buffer/release packets, this feature can be used
> when using MicroCheckpointing, or other Remus like VM FT solutions, you
What's "Remus"?
> can also use it to simulate the network delay.
> It has an interval option, if supplied, this filter will releas
Am 24.09.2015 um 10:01 hat Guangmu Zhu geschrieben:
> Hi Kevin,
>
> I tried the patch you provide, and I haven't seen that problem yet. If the
> disk
> space is full, an error will be reported with the message "Invalid argument"
> and the program will stop.
>
> Will you merge the patch to the ma
aio_worker() wrote the return code to the wrong variable.
Signed-off-by: Kevin Wolf
Tested-by: Guangmu Zhu
---
block/raw-win32.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/block/raw-win32.c b/block/raw-win32.c
index 68f2338..b562c94 100644
--- a/block/raw-win32.c
++
On Wed, Jul 01, 2015 at 09=12=33AM -0400, Andrew Jones wrote:
> cacheflush is an arm-specific syscall that qemu built for arm
> uses. Add it to the whitelist.
>
> Signed-off-by: Andrew Jones
>
> ---
>
> I'm not sure about the priority selection. Maybe cacheflush gets
> used frequently enough th
On Wed, Sep 23, 2015 at 02:14:03PM +0200, Laurent Vivier wrote:
> When DT node names for PCI devices are generated by SLOF,
> they are generated according to the type of the device
> (for instance, ethernet for virtio-net-pci device).
>
> Node name for hotplugged devices is generated by QEMU.
> Th
On Wed 23 Sep 2015 01:15:52 AM CEST, Eric Blake wrote:
> Commit 50b7b000 improved HMP error messages, but forgot to update
> qemu-iotests to match.
>
> Reported-by: Kevin Wolf
> Signed-off-by: Eric Blake
Reviewed-by: Alberto Garcia
Berto
On 24/09/2015 11:34, Michael S. Tsirkin wrote:
> On Wed, Sep 23, 2015 at 02:14:03PM +0200, Laurent Vivier wrote:
>> When DT node names for PCI devices are generated by SLOF,
>> they are generated according to the type of the device
>> (for instance, ethernet for virtio-net-pci device).
>>
>> Node
Yang Hongyang writes:
> On 09/24/2015 03:43 PM, Markus Armbruster wrote:
>> This has finally reached the front of my review queue. I apologize for
>> the lng delay.
>>
>> Copying Paolo for another pair of eyeballs (he wrote this code).
>>
> [...]
>>> +
>>> +opts = qemu_opts_find(qemu_fin
On Wed, 23 Sep 2015 12:04:49 -0300
Eduardo Habkost wrote:
> Version 1 of the pc-*-2.5 machine class series was applied to the PCI
> tree instead of v3 (which was rebased after the broken_reserved_end
> patch by Igor was included).
>
> This patch includes the missing hunks from v3, to make sure
>
Developers who are new to QEMU, or have a background familiarity
with GNU autotools, can have trouble getting their head around the
home-grown QEMU build system. This document attempts to explain
the structure / design of the configure script and the various
Makefile pieces that live across the sou
Am 23.09.2015 um 01:15 hat Eric Blake geschrieben:
> Commit 50b7b000 improved HMP error messages, but forgot to update
> qemu-iotests to match.
>
> Reported-by: Kevin Wolf
> Signed-off-by: Eric Blake
Thanks, applied to the block branch.
> I'm not sure if this should go through Markus' "error"
Hello!
Nice to see things moving on. I am waiting for PULL to be applied to master,
and after this i'll be able to proceed with other things, including live
migration. As part of it, i already rewrote state structure using bitops, so
that it's free of CPU number restrictions. Would be nice if
* Amit Shah (amit.s...@redhat.com) wrote:
> On (Mon) 13 Jul 2015 [20:07:52], Juan Quintela wrote:
> > "Dr. David Alan Gilbert" wrote:
> > > * Juan Quintela (quint...@redhat.com) wrote:
>
> > >> > +void qmp_migrate_start_postcopy(Error **errp)
> > >> > +{
> > >> > +MigrationState *s = migrate_
On Thu, Sep 10, 2015 at 08=54=28PM -0400, namn...@safe-mail.net wrote:
> > The current intention of the seccomp filter in QEMU, is that /all/ existing
> > QEMU features continue to work unchanged. So even if a flag is used in a
> > seemingly uncommon code path, we still need to allow that in a secc
On 09/24/2015 05:42 PM, Markus Armbruster wrote:
Yang Hongyang writes:
On 09/24/2015 03:43 PM, Markus Armbruster wrote:
This has finally reached the front of my review queue. I apologize for
the lng delay.
Copying Paolo for another pair of eyeballs (he wrote this code).
[...]
+
+
On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
Hi,
Here is the updated patch set for enabling vhost-user multiple queue. And I did
proper and formal testing this time.
This patch set introduces 2 more vhost user messages: VHOST_USER_GET_QUEUE_NUM,
for querying how many queues the backend supports,
On 09/24/2015 05:42 PM, Markus Armbruster wrote:
Yang Hongyang writes:
On 09/24/2015 03:43 PM, Markus Armbruster wrote:
This has finally reached the front of my review queue. I apologize for
the lng delay.
Copying Paolo for another pair of eyeballs (he wrote this code).
[...]
+
+
Am 24.09.2015 um 08:45 hat Markus Armbruster geschrieben:
> Max Reitz writes:
>
> > On 23.09.2015 18:11, Alberto Garcia wrote:
> >> Commit 934659c460 disabled the supression of segmentation faults in
> >> bash tests. The new output of test 061, however, assumes that a core
> >> dump will be produ
On Wed, Sep 23, 2015 at 08:43:39AM -0700, Peter Maydell wrote:
> On 23 September 2015 at 07:18, Andrew Jones wrote:
> > ARM/AArch64 KVM guests don't have any way to identify
> > themselves as KVM guests (x86 guests use a CPUID leaf). Now, we
> > could discuss all sorts of reasons why guests should
On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
From: "Michael S. Tsirkin"
Support a separate bitmask for vhost-user protocol features,
and messages to get/set protocol features.
Invoke them at init.
No features are defined yet.
[ leverage vhost_user_call for request handling -- Yuanhan Liu ]
Si
On Thu, Sep 24, 2015 at 11:04:55AM +0200, Gerd Hoffmann wrote:
> Signed-off-by: Gerd Hoffmann
It's easy to see this is what the patch does. But why? Some
explanation in the commit log about why it's done, as opposed to what is
done, would be better.
> ---
> hw/display/virtio-gpu.c | 3 +--
>
On Wed, Sep 23, 2015 at 09:09:11AM -0700, Peter Maydell wrote:
> On 23 September 2015 at 08:43, Peter Maydell wrote:
> > On 23 September 2015 at 07:18, Andrew Jones wrote:
> >> ARM/AArch64 KVM guests don't have any way to identify
> >> themselves as KVM guests (x86 guests use a CPUID leaf). Now,
On Thu, Sep 24, 2015 at 01:05:11PM +0300, Marcel Apfelbaum wrote:
> On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
> >So that we could let vhost_user_call to handle extented requests,
> >such as VHOST_USER_GET/SET_PROTOCOL_FEATURES, instead of invoking
> >vhost_user_read/write and constructing the msg
On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
Quote from Michael:
We really should rename VHOST_RESET_OWNER to VHOST_RESET_DEVICE.
I suggest to change this to a better commit message :)
You can take anything from the mail thread discussion about it.
Suggested-by: Michael S. Tsirkin
Signe
On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
So that we could let vhost_user_call to handle extented requests,
such as VHOST_USER_GET/SET_PROTOCOL_FEATURES, instead of invoking
vhost_user_read/write and constructing the msg again by ourself.
Signed-off-by: Yuanhan Liu
---
hw/virtio/vhost-user.c
On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
This is for querying how many queues the backend supports if it has mq
support(when VHOST_USER_PROTOCOL_F_MQ flag is set from the quried
/s/quried/queried
Only if you plan to send another version, we can fix it on top.
protocol features).
vhost_net
On Mon, Sep 21, 2015 at 02:32:10PM +0200, Jan Kiszka wrote:
> On 2015-09-21 14:13, Michael S. Tsirkin wrote:
> > On Fri, Sep 18, 2015 at 06:29:27PM +0200, Claudio Fontana wrote:
> >> Hello,
> >>
> >> this is a first RFC for virtio-peer 0.1, which is still very much a work
> >> in progress:
> >>
>
When DT node names for PCI devices are generated by SLOF,
they are generated according to the type of the device
(for instance, ethernet for virtio-net-pci device).
Node name for hotplugged devices is generated by QEMU.
This patch adds the mechanic to QEMU to create the node
name according to the
When DT node names for PCI devices are generated by SLOF,
they are generated according to the type of the device
(for instance, ethernet for virtio-net-pci device).
Node name for hotplugged devices is generated by QEMU.
This series adds the mechanic to QEMU to create the node
name according to the
To allow QEMU to add PCI entries in device tree,
we must have a more exhaustive list of PCI class IDs.
This patch synchronizes as much as possible with
pci_ids.h and add some missing IDs from SLOF.
Signed-off-by: Laurent Vivier
Reviewed-by: Michael S. Tsirkin
---
include/hw/pci/pci_ids.h | 112
On 09/24/2015 01:18 PM, Michael S. Tsirkin wrote:
On Thu, Sep 24, 2015 at 01:05:11PM +0300, Marcel Apfelbaum wrote:
On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
So that we could let vhost_user_call to handle extented requests,
such as VHOST_USER_GET/SET_PROTOCOL_FEATURES, instead of invoking
vhos
* Juan Quintela (quint...@redhat.com) wrote:
> "Dr. David Alan Gilbert (git)" wrote:
> > From: "Dr. David Alan Gilbert"
> >
> > Userfault doesn't work with mlock; mlock is designed to nail down pages
> > so they don't move, userfault is designed to tell you when they're not
> > there.
> >
> > mun
On Wed, Sep 23, 2015 at 12:04:49PM -0300, Eduardo Habkost wrote:
> Version 1 of the pc-*-2.5 machine class series was applied to the PCI
> tree instead of v3 (which was rebased after the broken_reserved_end
> patch by Igor was included).
>
> This patch includes the missing hunks from v3, to make s
On 24/09/15 12:27, Laurent Vivier wrote:
> To allow QEMU to add PCI entries in device tree,
> we must have a more exhaustive list of PCI class IDs.
>
> This patch synchronizes as much as possible with
> pci_ids.h and add some missing IDs from SLOF.
>
> Signed-off-by: Laurent Vivier
> Reviewed-by
HTAB size is a factor of maximum memory size that is specified by maxmem=
command line option. In cases where there is shortage of host memory, host
will not be able to allocate contiguous memory for guest HTAB and will
instead allocate a smaller HTAB. This usually is not a problem but when
user st
Quoting Markus Armbruster (2015-09-24 02:02:41)
> Ping?
Thanks, applied to qga tree:
https://github.com/mdroth/qemu/commits/qga
>
> Markus Armbruster writes:
>
> > g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> > for two reasons. One, it catches multiplication over
Allocate HTAB from ppc_spapr_init() so that we can abort the guest
if requested HTAB size is't allocated by the host. However retain the
htab reset call in spapr_reset_htab() so that HTAB gets reset (and
not allocated) during machine reset.
Signed-off-by: Bharata B Rao
---
hw/ppc/spapr.c | 34 ++
Terminate the guest when HTAB of requested size isn't allocated by
the host.
When memory hotplug is attempted on a guest that has booted with
less than requested HTAB size, the guest kernel will not be able
to gracefully fail the hotplug request. This patch will ensure that
we never end up in a si
Hi,
On Thu, 3 Sep 2015 10:19:19 +0300, shmulik.ladk...@ravellosystems.com wrote:
> On Wed, 2 Sep 2015 17:14:52 +0100, stefa...@redhat.com wrote:
> > From: Fam Zheng
> >
> > Commit 6e99c63 ("net/socket: Drop net_socket_can_send") changed the
> > semantics around .can_receive for sockets to now r
On 09/24/15 11:49, Daniel P. Berrange wrote:
> Developers who are new to QEMU, or have a background familiarity
> with GNU autotools, can have trouble getting their head around the
> home-grown QEMU build system. This document attempts to explain
> the structure / design of the configure script and
On Thu, Sep 24, 2015 at 01:13:24PM +0300, Marcel Apfelbaum wrote:
> On 09/23/2015 07:19 AM, Yuanhan Liu wrote:
> >From: "Michael S. Tsirkin"
> >
> >Support a separate bitmask for vhost-user protocol features,
> >and messages to get/set protocol features.
> >
> >Invoke them at init.
> >
> >No featu
On Thu, Sep 24, 2015 at 01:13:24PM +0300, Marcel Apfelbaum wrote:
> >diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
> >index 1d76b94..9d32d76 100644
> >--- a/hw/net/vhost_net.c
> >+++ b/hw/net/vhost_net.c
> >@@ -152,8 +152,10 @@ struct vhost_net *vhost_net_init(VhostNetOptions
> >*options)
>
Yang Hongyang writes:
> On 09/24/2015 05:42 PM, Markus Armbruster wrote:
>> Yang Hongyang writes:
>>
>>> On 09/24/2015 03:43 PM, Markus Armbruster wrote:
This has finally reached the front of my review queue. I apologize for
the lng delay.
Copying Paolo for another pair
Yang Hongyang writes:
> On 09/24/2015 05:42 PM, Markus Armbruster wrote:
>> Yang Hongyang writes:
>>
>>> On 09/24/2015 03:43 PM, Markus Armbruster wrote:
This has finally reached the front of my review queue. I apologize for
the lng delay.
Copying Paolo for another pair
From: Marc-André Lureau
ivshmem is going to use MSIX state conditionally.
Signed-off-by: Marc-André Lureau
---
include/hw/pci/msix.h | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/include/hw/pci/msix.h b/include/hw/pci/msix.h
index 954d82b..72e5f93 100644
From: Marc-André Lureau
Hi,
This series is mostly about adding the client/server code from David
Marchand, code cleanups, and little improvements and fixes for
ivshmem. Finally there is some ivshmem tests (they work fine without
kvm btw).
Git: https://github.com/elmarco/qemu.git ivshmem branch
1 - 100 of 368 matches
Mail list logo
