From: Christophe Fergeau
Acked-by: Jonathon Jongsma
---
server/spicevmc.c | 80 +--
1 file changed, 66 insertions(+), 14 deletions(-)
diff --git a/server/spicevmc.c b/server/spicevmc.c
index d1e8222..241d87d 100644
--- a/server/spicevmc.c
+++
make the function names match the type names. So
spice_char_device_state_* becomes red_char_device_* and
spice_char_device_* also becomes red_char_device_*.
---
server/char-device.c | 320 +--
server/char-device.h | 108 -
server/reds
From: Christophe Fergeau
This structure holding virtual function pointers was kept until now as a
RedCharDevice member in order to make the GObject conversion easier.
Now that all RedCharDevice children are converted to GObject, it can be
moved into RedCharDeviceClass.
---
server/char-device.c |
From: Christophe Fergeau
Reset pointer after freeing the structure pointing to it.
Acked-by: Jonathon Jongsma
---
server/char-device.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/server/char-device.c b/server/char-device.c
index 3056f51..2957a1b 100644
--- a/server/char-device.c
+++ b/s
From: Christophe Fergeau
Reviewed-by: Jonathon Jongsma
---
server/char-device.c | 353 ++-
server/char-device.h | 43 ++-
2 files changed, 304 insertions(+), 92 deletions(-)
diff --git a/server/char-device.c b/server/char-device.c
index 2957
From: Christophe Fergeau
This inherits from RedCharDevice.
Signed-off-by: Christophe Fergeau
---
server/reds-private.h | 2 +-
server/reds.c | 246 --
2 files changed, 159 insertions(+), 89 deletions(-)
diff --git a/server/reds-private
From: Christophe Fergeau
VDIPortState and VDIReadBuf don't really need to be defined in a header
file as they are only used in reds.c
---
server/reds-private.h | 56 +
server/reds.c | 169 +-
2 files changed, 114 insertions
From: Christophe Fergeau
Nothing is using it anymore now that CharDevice classes are
gobjectified.
---
server/char-device.c | 42 --
server/char-device.h | 9 +
2 files changed, 1 insertion(+), 50 deletions(-)
diff --git a/server/char-device.c b/
From: Christophe Fergeau
This pipe item belongs to the char device, not to the spicevmc channel.
---
server/spicevmc.c | 35 ---
1 file changed, 28 insertions(+), 7 deletions(-)
diff --git a/server/spicevmc.c b/server/spicevmc.c
index 241d87d..9969cf3 100644
---
Internal types should use Red namespace for consistency
---
server/char-device.c | 36 ++--
server/char-device.h | 12 ++--
server/reds.c| 12 ++--
server/smartcard.c | 12 ++--
server/spicevmc.c| 16
5 files ch
Make it consistent with the renamed RedCharDevice and the convention of
using 'Red' namespaces for internal types
---
server/char-device.c | 108 +--
1 file changed, 54 insertions(+), 54 deletions(-)
diff --git a/server/char-device.c b/server/char-d
Internal types should use 'Red' namespace for consistency
---
server/char-device.c | 72 +--
server/char-device.h | 16 ++--
server/reds-private.h | 2 +-
server/reds.c | 4 +--
server/smartcard.c| 12 -
server/spicevm
This is more consistent with internal type naming convention, and it
paves the way for a new char device GObject heirarchy
---
server/char-device.c | 107 +-
server/char-device.h | 58 ++-
server/reds-private.h | 2 +-
se
Here's an updated version of the series Frediano posted yesterday, but includes
some introductory commits that rename some types for improved consistency
within the codebase.
Christophe Fergeau (10):
char-device: Avoid use-after-free
char-device: Make SpiceCharDeviceState a GObject
spicevmc:
From: Christophe Fergeau
This inherits from RedCharDevice. Once all char device states are
converted, we can turn the associated vfuncs into RedCharDeviceClass
vfuncs.
Signed-off-by: Christophe Fergeau
---
server/smartcard.c | 124 +++--
server/s
From: Christophe Fergeau
This inherits from RedCharDevice.
Signed-off-by: Christophe Fergeau
Signed-off-by: Frediano Ziglio
---
server/reds.c | 395 +-
1 file changed, 200 insertions(+), 195 deletions(-)
diff --git a/server/reds.c b/ser
On Wed, 2016-03-30 at 18:21 +0100, Frediano Ziglio wrote:
> From: Christophe Fergeau
>
> This inherits from RedCharDevice. Once all char device states are
> converted, we can turn the associated vfuncs into RedCharDeviceClass
> vfuncs.
>
> Signed-off-by: Christophe Fergeau
> ---
> server/smart
Hi,
On Thu, Mar 31, 2016 at 12:37:31PM +0200, Fabio Fantoni wrote:
> Il 31/03/2016 12:21, Victor Toso ha scritto:
> > From: Jeremy White
> >
> > This patch requires the patch series provided by Francois Gouget
> > which add gstreamer support to Spice.
> >
> > It also should finally include the fi
On Thu, 2016-03-31 at 05:01 -0400, Frediano Ziglio wrote:
> > Here's a patch that I think could be squashed into this one:
> >
> > diff --git a/server/smartcard.c b/server/smartcard.c
> > index f592a9f..3d40f59 100644
> > --- a/server/smartcard.c
> > +++ b/server/smartcard.c
> > @@ -67,7 +67,6 @@
On Thu, 2016-03-31 at 03:12 -0400, Frediano Ziglio wrote:
> >
> > Hi Jonathon,
> >
> > On Wed, Mar 30, 2016 at 05:17:47PM -0500, Jonathon Jongsma wrote:
> > > These two type names are typedefs of eachother, but it's confusing to
> > > have two different typenames for the same type throughout the
>
> From: Takao Fujiwara
>
> Some of the keys in JP keyboard do no send WM_KEYUP and it causes
> unlimited key events on Linux desktop. This sends the virtual
> key release events to avoid the desktop hangup.
> ---
> src/spice-widget.c | 9 +
> 1 file changed, 9 insertions(+)
>
> diff
On Thu, Mar 31, 2016 at 08:36:34PM +0900, Takao Fujiwara wrote:
> I noticed Linux GNOME desktop hangs up when Windows virt-viewer connects to
> the desktop and type some specific keys on JP keyboard.
>
> Server: GNOME desktop and IBus in RHEL 7
> Client virt-viewer in Windows 7 with JP 106 keyboa
>
> From: Takao Fujiwara
>
> Some of the keys in JP keyboard do no send WM_KEYUP and it causes
> unlimited key events on Linux desktop. This sends the virtual
> key release events to avoid the desktop hangup.
> ---
> src/spice-widget.c | 18 ++
> 1 file changed, 18 insertions(+)
>
> From: Takao Fujiwara
>
> MapVirtualKey() returns the scancode of the released Zenkaku_Hankaku key
> but does not pressed one and modifiered one. Any cases should be sent
> to use input method engines.
> ---
> src/spice-widget-priv.h | 1 +
> src/spice-widget.c | 37 +++
On 03/26/2016 10:53 AM, Brad Campbell wrote:
> Checking the git logs, in Oct the required glib version was bumped to
> 2.22. Since that point logging has been added using g_test_* functions
> which were apparently added in 2.38. Trying to build on 2.33 (which I
> was) therefore passes the configure
On 03/30/2016 09:15 AM, Eduardo Lima (Etrunko) wrote:
> On 03/29/2016 12:08 PM, Mahmood Naderan wrote:
>> O that is good :thumsup:
>> I applied the patch and the compilation passed that file. However, I get
>> a new undefined reference in another file
>>
>>
>> root@cluster:spice-src# patch < patch
Ack,
Pavel
On Thu, 2016-03-31 at 13:35 +0200, Christophe Fergeau wrote:
> The glib version checks were using #ifdef GLIB_CHECK_VERSION() rather
> than #if GLIB_CHECK_VERSION()
> ---
> tests/test-logging.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/tests/test-lo
Sorry, the subject and description is not correct.
I will send this patch again with the correct subject.
On 03/31/16 19:43, Takao Fujiwara-san wrote:
> Some of the keys in JP keyboard do no send WM_KEYUP and it causes
> unlimited key events on Linux desktop. This sends the virtual
> key release e
On Thu, Mar 31, 2016 at 1:43 PM, Frediano Ziglio wrote:
>>
>> I noticed Linux GNOME desktop hangs up when Windows virt-viewer connects to
>> the desktop and type some specific keys on JP keyboard.
>>
>> Server: GNOME desktop and IBus in RHEL 7
>> Client virt-viewer in Windows 7 with JP 106 keyboar
From: Takao Fujiwara
Some of the keys in JP keyboard do no send WM_KEYUP and it causes
unlimited key events on Linux desktop. This sends the virtual
key release events to avoid the desktop hangup.
---
src/spice-widget.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/src/
From: Takao Fujiwara
Some of the keys in JP keyboard do no send WM_KEYUP and it causes
unlimited key events on Linux desktop. This sends the virtual
key release events to avoid the desktop hangup.
---
src/spice-widget.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/src/spice-widge
From: Takao Fujiwara
MapVirtualKey() returns the scancode of the released Zenkaku_Hankaku key
but does not pressed one and modifiered one. Any cases should be sent
to use input method engines.
---
src/spice-widget-priv.h | 1 +
src/spice-widget.c | 37 +++--
In order to speed up the process I took the freedom to apply Takao's
patches locally, regenerate the patches (and cover letter) and submit
them upstream using git send-email.
Please, for more info, take a look on:
https://lists.freedesktop.org/archives/spice-devel/2016-March/027878.html
Takao Fuj
>
> I noticed Linux GNOME desktop hangs up when Windows virt-viewer connects to
> the desktop and type some specific keys on JP keyboard.
>
> Server: GNOME desktop and IBus in RHEL 7
> Client virt-viewer in Windows 7 with JP 106 keyboard
>
> When I see the key events with 'xev' command in GNOME
I noticed Linux GNOME desktop hangs up when Windows virt-viewer connects to the
desktop and type some specific keys on JP keyboard.
Server: GNOME desktop and IBus in RHEL 7
Client virt-viewer in Windows 7 with JP 106 keyboard
When I see the key events with 'xev' command in GNOME desktop, I obser
The glib version checks were using #ifdef GLIB_CHECK_VERSION() rather
than #if GLIB_CHECK_VERSION()
---
tests/test-logging.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/test-logging.c b/tests/test-logging.c
index c02905f..cb929f9 100644
--- a/tests/test-logging.c
>
> > Here's a patch that I think could be squashed into this one:
> >
> > diff --git a/server/smartcard.c b/server/smartcard.c
> > index f592a9f..3d40f59 100644
> > --- a/server/smartcard.c
> > +++ b/server/smartcard.c
> > @@ -67,7 +67,6 @@ G_DEFINE_TYPE(RedCharDeviceSmartcard,
> > red_char_devi
Il 31/03/2016 12:21, Victor Toso ha scritto:
> From: Jeremy White
>
> This patch requires the patch series provided by Francois Gouget
> which add gstreamer support to Spice.
>
> It also should finally include the fix that Fabio keeps pointing out.
If you mean me (I not remember other Fabio in th
From: Jeremy White
This patch requires the patch series provided by Francois Gouget
which add gstreamer support to Spice.
It also should finally include the fix that Fabio keeps pointing out.
Signed-off-by: Jeremy White
---
qemu-options.hx | 8
ui/spice-core.c | 17 +
Sending the rebased version of this patch... changing the server
version to the unstable as well
___
Spice-devel mailing list
Spice-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/spice-devel
On Thu, Mar 31, 2016 at 06:02:39AM -0400, Frediano Ziglio wrote:
> libvirt IMHO is not enforcing this. I would be surprised if libvirt based
> its naming on the internal naming conventions of spice-server, it's just
> using public names so it has to be SpiceVmc.
>
Yes it's using a public name, wh
> On Thu, Mar 31, 2016 at 04:23:47AM -0400, Frediano Ziglio wrote:
> > >
> > > On Wed, 2016-03-30 at 18:21 +0100, Frediano Ziglio wrote:
> > > > From: Christophe Fergeau
> > > >
> > > > ---
> > > > server/spicevmc.c | 80
> > > > +-
> > > > -
On Tue, Mar 29, 2016 at 10:24:53AM +0200, Christophe Fergeau wrote:
> On Mon, Mar 28, 2016 at 06:00:28PM -0300, Eduardo Lima (Etrunko) wrote:
> > From: "Eduardo Lima (Etrunko)"
> >
> > It is required to compile ssl-verify.[ch].
>
> Makes sense. Any idea why it could be built without issues so fa
Acked-by: Christophe Fergeau
I'll push this in a minute.
On Wed, Mar 30, 2016 at 03:39:06PM -0300, Eduardo Lima (Etrunko) wrote:
> Tests now include functions only available in glib 2.38. To avoid
> setting the dependency bar too high, we simply put them between #ifdef
> guards.
>
> Signed-off-
On Thu, Mar 31, 2016 at 03:12:44AM -0400, Frediano Ziglio wrote:
> >
> > Hi Jonathon,
> >
> > On Wed, Mar 30, 2016 at 05:17:47PM -0500, Jonathon Jongsma wrote:
> > > These two type names are typedefs of eachother, but it's confusing to
> > > have two different typenames for the same type througho
On Thu, Mar 31, 2016 at 04:23:47AM -0400, Frediano Ziglio wrote:
> >
> > On Wed, 2016-03-30 at 18:21 +0100, Frediano Ziglio wrote:
> > > From: Christophe Fergeau
> > >
> > > ---
> > > server/spicevmc.c | 80
> > > +-
> > > -
> > > 1 file chan
> Hi,
>
> I agree that there should be the only one typedef. But I don't
> understand why not to include the needed header directly. We should try
> to "untagle" these header dependencies.
>
> Pavel
>
Recently I read an article (I think this
http://developers.redhat.com/blog/2016/03/03/curing
> Here's a patch that I think could be squashed into this one:
>
> diff --git a/server/smartcard.c b/server/smartcard.c
> index f592a9f..3d40f59 100644
> --- a/server/smartcard.c
> +++ b/server/smartcard.c
> @@ -67,7 +67,6 @@ G_DEFINE_TYPE(RedCharDeviceSmartcard,
> red_char_device_smartcard, RED_
>
> On Wed, 2016-03-30 at 18:21 +0100, Frediano Ziglio wrote:
> > From: Christophe Fergeau
> >
> > ---
> > server/spicevmc.c | 80
> > +-
> > -
> > 1 file changed, 66 insertions(+), 14 deletions(-)
> >
> > diff --git a/server/spicevmc.c b/se
Hi,
On Thu, Mar 31, 2016 at 03:12:44AM -0400, Frediano Ziglio wrote:
> >
> > Hi Jonathon,
> >
> > On Wed, Mar 30, 2016 at 05:17:47PM -0500, Jonathon Jongsma wrote:
> > > These two type names are typedefs of eachother, but it's confusing to
> > > have two different typenames for the same type thro
>
> These two type names are typedefs of eachother, but it's confusing to
> have two different typenames for the same type throughout the code.
> Convert everythign to use RedCharDevice.
> ---
>
> This patch should apply to the end of this patch series. Comments?
>
Patch is good but should appl
This is causing compile errors with older gcc versions.
red-qxl.h includes spice-qxl.h through red-channel.h and then spice.h,
so it will get the QXLState typedef this way.
Signed-off-by: Christophe Fergeau
Signed-off-by: Frediano Ziglio
---
server/main-dispatcher.h | 2 +-
server/red-qxl.h
>
> Hi Jonathon,
>
> On Wed, Mar 30, 2016 at 05:17:47PM -0500, Jonathon Jongsma wrote:
> > These two type names are typedefs of eachother, but it's confusing to
> > have two different typenames for the same type throughout the code.
> > Convert everythign to use RedCharDevice.
>
> I agree with y
53 matches
Mail list logo
