Re: [Spice-devel] [PATCH spice-server v2 1/2] Introduce some macros to help declaring new GObject

> > On Thu, Sep 07, 2017 at 03:58:21AM -0400, Frediano Ziglio wrote: > > > > > > On Wed, Sep 06, 2017 at 03:42:29PM +0100, Frediano Ziglio wrote: > > > > The macros will implement most of the boilerplate needed > > > > to declare an object. > > > > Their usage are similar to GLib

[Spice-devel] [spice-server] channel: Remove unused red_channel_min_pipe_size

This could have been removed as part of 6e6126e024. Signed-off-by: Christophe Fergeau --- server/red-channel.c | 13 - server/red-channel.h | 2 -- 2 files changed, 15 deletions(-) diff --git a/server/red-channel.c b/server/red-channel.c index

Re: [Spice-devel] [PATCH spice-html5 1/4] spiceconn: Add function to test channel capabilities

> I agree, I will use your first suggestion. > Tomáš As a note, I'm not a Javascript expert as you, I just tested my suggestion with a small test script. Frediano > 2017-09-07 10:31 GMT+02:00 Frediano Ziglio < fzig...@redhat.com > : > > > > > > > This will be used in other commits. > >

Re: [Spice-devel] [PATCH spice-server 1/2] gl: fix client mouse mode

- Original Message - > > > > Hi > > > > - Original Message - > > > Since 2.8, QEMU now longer creates QXL primary surfaces when using GL. > > > This change broke client-side mouse mode, because Spice server relies on > > > primary surface conditions. > > > > > > When GL is

Re: [Spice-devel] [PATCH spice-html5 4/4] Display: Add support for the VP9 codec type

OK, I wil change it. Tomáš 2017-09-07 11:39 GMT+02:00 Frediano Ziglio : > > > Hi, > > I don't think so. Why do you think it would be better? > > Tomáš > > Is easier to extend when new codecs are added > > Frediano > > 2017-09-07 10:50 GMT+02:00 Frediano Ziglio

Re: [Spice-devel] [PATCH spice-server 1/2] gl: fix client mouse mode

> > - Original Message - > > > > > > > Hi > > > > > > - Original Message - > > > > Since 2.8, QEMU now longer creates QXL primary surfaces when using GL. > > > > This change broke client-side mouse mode, because Spice server relies > > > > on > > > > primary surface conditions.

Re: [Spice-devel] [PATCH spice-html5 1/4] spiceconn: Add function to test channel capabilities

I agree, I will use your first suggestion. Tomáš 2017-09-07 10:31 GMT+02:00 Frediano Ziglio : > > > > This will be used in other commits. > > --- > > spiceconn.js | 19 +++ > > 1 file changed, 19 insertions(+) > > > > diff --git a/spiceconn.js

Re: [Spice-devel] [PATCH spice-server] dcc: Make dcc_stop static

On 09/07/2017 12:17 PM, Frediano Ziglio wrote: Just used by dcc_on_disconnect. Signed-off-by: Frediano Ziglio Acked-by: Uri Lublin --- server/dcc.c | 2 +- server/dcc.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git

[Spice-devel] [PATCH spice-html5 v2 4/4] Display: Add support for the VP9 codec type

--- display.js | 27 +-- spiceconn.js | 2 ++ webm.js | 13 +++-- 3 files changed, 34 insertions(+), 8 deletions(-) diff --git a/display.js b/display.js index 0868f91..d711ced 100644 --- a/display.js +++ b/display.js @@ -543,7 +543,8 @@

[Spice-devel] [PATCH spice-server] dcc: Make dcc_stop static

Just used by dcc_on_disconnect. Signed-off-by: Frediano Ziglio --- server/dcc.c | 2 +- server/dcc.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/server/dcc.c b/server/dcc.c index 3fb598ff9..1e0ba790f 100644 --- a/server/dcc.c +++ b/server/dcc.c @@

[Spice-devel] [PATCH spice-server 1/2] reds-stream: Allows to change core interface

When a stream is moved from the main thread to a secondary one the events are potentially registered using a different core interface. This cause memory corruption accessing the watch registered in RedsStream. This patch allows to always use the right interface. Signed-off-by: Frediano Ziglio

[Spice-devel] [PATCH spice-server 2/2] Fix crash attempting to connect duplicate channels

You could easily trigger this issue using multiple monitors and a modified spice-gtk client with this patch: --- a/src/channel-main.c +++ b/src/channel-main.c @@ -1699,6 +1699,7 @@ static gboolean _channel_new(channel_new_t *c) { g_return_val_if_fail(c != NULL, FALSE); +if

[Spice-devel] [PATCH spice-server 0/2] Fix crash attempting to connect duplicate channels

This is a condition a client can potentially trigger connecting to an already connected channel. Changes since v2: - extended commit message. Frediano Ziglio (2): reds-stream: Allows to change core interface Fix crash attempting to connect duplicate channels server/red-channel-client.c |

Re: [Spice-devel] [PATCH spice-html5 4/4] Display: Add support for the VP9 codec type

> Hi, > I don't think so. Why do you think it would be better? > Tomáš Is easier to extend when new codecs are added Frediano > 2017-09-07 10:50 GMT+02:00 Frediano Ziglio < fzig...@redhat.com > : > > > > > > > --- > > > > display.js | 26 -- > > > > spiceconn.js |

[Spice-devel] [PATCH spice-server 6/6] main-dispatcher: Avoid type conversion in dispatcher_handle_read

Pass proper type to callback to avoid having to convert to the right type for each call. Signed-off-by: Frediano Ziglio --- This make dispatcher_handle_read identical to handle_dev_input. --- server/main-dispatcher.c | 6 +++--- 1 file changed, 3 insertions(+), 3

[Spice-devel] [PATCH spice-server 0/6] Minor Dispatcher related changes

Looking at Jonathon work on Dispatcher had some minor changes. Some looks not related to Dispatcher but but are close to it. Frediano Ziglio (6): red-qxl: Unify red_qxl_use_client_monitors_config and red_qxl_client_monitors_config red-qxl: Move QXLInterface wrappers together spice-qxl:

[Spice-devel] [PATCH spice-server 5/6] dispatcher: Remove "opaque" property

Is supposed to be used during initialization but is never used. Signed-off-by: Frediano Ziglio --- server/dispatcher.c | 22 ++ server/dispatcher.h | 2 +- server/red-qxl.c| 2 +- 3 files changed, 4 insertions(+), 22 deletions(-) diff --git

[Spice-devel] [PATCH spice-server 3/6] spice-qxl: Add version information

Signed-off-by: Frediano Ziglio --- server/spice-qxl.h | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/server/spice-qxl.h b/server/spice-qxl.h index b8910bf47..04c657c58 100644 --- a/server/spice-qxl.h +++ b/server/spice-qxl.h @@ -197,12 +197,14 @@

[Spice-devel] [PATCH spice-server 1/6] red-qxl: Unify red_qxl_use_client_monitors_config and red_qxl_client_monitors_config

These 2 functions were doing the same stuff, calling client_monitors_config callback in QXLInterface. The only difference was that red_qxl_use_client_monitors_config used a NULL value. Added the check for proper version, QXLInstance before 3.3 did not have this callback. Signed-off-by: Frediano

[Spice-devel] [PATCH spice-server 2/6] red-qxl: Move QXLInterface wrappers together

Signed-off-by: Frediano Ziglio --- server/red-qxl.c | 24 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/server/red-qxl.c b/server/red-qxl.c index 54d642ad9..bec063c17 100644 --- a/server/red-qxl.c +++ b/server/red-qxl.c @@ -638,18

[Spice-devel] [PATCH spice-server 4/6] red-qxl: Avoid to use AsyncCommand for GL_DRAW_ASYNC message

AsyncCommand is used to handle asynchronous messages from the dispatcher. GL_DRAW_ASYNC is mainly using it to store the cookie. The value of GL_DRAW_COOKIE_INVALID was choosen to allow implementing cookies (which basically are handles) either using indexes (where 0 is valid) or pointers (where 0

[Spice-devel] [PATCH spice-server 1/2] gl: fix client mouse mode

Since 2.8, QEMU now longer creates QXL primary surfaces when using GL. This change broke client-side mouse mode, because Spice server relies on primary surface conditions. When GL is enabled, use GL scanout informations. Mouse mode is always client when GL surfaces are used. This patch and most

[Spice-devel] [PATCH spice-server 2/2] red-qxl: Rename primary_active to display_active

To avoid confusion use the more generic "display" instead of "primary". "primary" is used in the code widely to mean the primary surface. In the past (before 3D support) having a primary surface mean both the availability of something to render (that is the device was enabled and ready) and the

Re: [Spice-devel] [PATCH spice-server 1/2] gl: fix client mouse mode

> Hi > > - Original Message - > > Since 2.8, QEMU now longer creates QXL primary surfaces when using GL. > > This change broke client-side mouse mode, because Spice server relies on > > primary surface conditions. > > > > When GL is enabled, use GL scanout informations. > > Mouse mode

Re: [Spice-devel] [PATCH spice-html5 1/4] spiceconn: Add function to test channel capabilities

> > This will be used in other commits. > --- > spiceconn.js | 19 +++ > 1 file changed, 19 insertions(+) > > diff --git a/spiceconn.js b/spiceconn.js > index 33e7388..78d5820 100644 > --- a/spiceconn.js > +++ b/spiceconn.js > @@ -243,6 +243,9 @@ SpiceConn.prototype = >

Re: [Spice-devel] [PATCH spice-html5 4/4] Display: Add support for the VP9 codec type

> > --- > display.js | 26 -- > spiceconn.js | 2 ++ > webm.js | 12 ++-- > 3 files changed, 32 insertions(+), 8 deletions(-) > > diff --git a/display.js b/display.js > index 0868f91..abd5b1a 100644 > --- a/display.js > +++ b/display.js > @@ -543,7 +543,8

Re: [Spice-devel] [spice-server] channel: Remove unused red_channel_min_pipe_size

> > This could have been removed as part of 6e6126e024. > Symmetric with max but I don't see the purpose, easy to add again if needed. Acked-by: Frediano Ziglio Frediano > Signed-off-by: Christophe Fergeau > --- > server/red-channel.c | 13

Re: [Spice-devel] [PATCH spice-server v2 1/2] Introduce some macros to help declaring new GObject

On Thu, Sep 07, 2017 at 03:58:21AM -0400, Frediano Ziglio wrote: > > > > On Wed, Sep 06, 2017 at 03:42:29PM +0100, Frediano Ziglio wrote: > > > The macros will implement most of the boilerplate needed > > > to declare an object. > > > Their usage are similar to GLib G_DECLARE_*_TYPE macros. > >

Re: [Spice-devel] [PATCH spice-server 1/2] gl: fix client mouse mode

Hi - Original Message - > Since 2.8, QEMU now longer creates QXL primary surfaces when using GL. > This change broke client-side mouse mode, because Spice server relies on > primary surface conditions. > > When GL is enabled, use GL scanout informations. > Mouse mode is always client

Re: [Spice-devel] [PATCH spice-server 2/6] red-qxl: Move QXLInterface wrappers together

Acked-by: Christophe Fergeau On Thu, Sep 07, 2017 at 12:40:22PM +0100, Frediano Ziglio wrote: > Signed-off-by: Frediano Ziglio > --- > server/red-qxl.c | 24 > 1 file changed, 12 insertions(+), 12 deletions(-) > > diff --git

Re: [Spice-devel] [PATCH spice-server 3/6] spice-qxl: Add version information

Acked-by: Christophe Fergeau On Thu, Sep 07, 2017 at 12:40:23PM +0100, Frediano Ziglio wrote: > Signed-off-by: Frediano Ziglio > --- > server/spice-qxl.h | 6 -- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/server/spice-qxl.h

Re: [Spice-devel] [PATCH spice-server 5/6] dispatcher: Remove "opaque" property

On Thu, 2017-09-07 at 16:50 +0200, Christophe Fergeau wrote: > Acked-by: Christophe Fergeau > > On Thu, Sep 07, 2017 at 12:40:25PM +0100, Frediano Ziglio wrote: > > Is supposed to be used during initialization but is never > > used. > > Hmm, I find this commit log to be a

Re: [Spice-devel] [PATCH spice-server v2 3/3] Fix crash attempting to connect duplicate channels

On Thu, 2017-09-07 at 03:50 -0400, Frediano Ziglio wrote: > > > > On Wed, 2017-08-30 at 10:36 +0100, Frediano Ziglio wrote: > > > You could easily trigger this issue using multiple monitors and > > > a modified spice-gtk client with this patch: > > > > > > --- a/src/channel-main.c > > > +++

Re: [Spice-devel] [PATCH spice-server 1/6] red-qxl: Unify red_qxl_use_client_monitors_config and red_qxl_client_monitors_config

On Thu, Sep 07, 2017 at 12:40:21PM +0100, Frediano Ziglio wrote: > These 2 functions were doing the same stuff, calling > client_monitors_config callback in QXLInterface. > The only difference was that red_qxl_use_client_monitors_config > used a NULL value. > Added the check for proper version,

[Spice-devel] [PATCH spice-server v2] gl: fix client mouse mode

Since 2.8, QEMU now longer creates QXL primary surfaces when using GL. This change broke client-side mouse mode, because Spice server relies on primary surface conditions. When GL is enabled, use GL scanout informations. Mouse mode is always client when GL surfaces are used. This patch and most

Re: [Spice-devel] [PATCH spice-server 5/6] dispatcher: Remove "opaque" property

Acked-by: Christophe Fergeau On Thu, Sep 07, 2017 at 12:40:25PM +0100, Frediano Ziglio wrote: > Is supposed to be used during initialization but is never > used. > > Signed-off-by: Frediano Ziglio > --- > server/dispatcher.c | 22

Re: [Spice-devel] [PATCH spice-server 6/6] main-dispatcher: Avoid type conversion in dispatcher_handle_read

Acked-by: Christophe Fergeau On Thu, Sep 07, 2017 at 12:40:26PM +0100, Frediano Ziglio wrote: > Pass proper type to callback to avoid having to convert to > the right type for each call. > > Signed-off-by: Frediano Ziglio > --- > This make

Re: [Spice-devel] [RFC PATCH spice-server v3 10/20] stream-channel: Allows to register callback to get new stream request

On Fri, Aug 25, 2017 at 12:22:41PM -0400, Frediano Ziglio wrote: > > > > On Fri, 2017-08-25 at 05:39 -0400, Frediano Ziglio wrote: > > > > > > > > I had a similar comment in a different patch during the last > > > > version of > > > > the series, but I personally would prefer a signal to handle

Re: [Spice-devel] [PATCH spice-server v4] Add documentation for Dispatcher

> > --- > Changes since v3: > - removed documentation for 'opaque' argument from dispatcher_new() > > server/dispatcher.h | 95 > - > 1 file changed, 80 insertions(+), 15 deletions(-) > > diff --git a/server/dispatcher.h

[Spice-devel] [PATCH spice-server v3 3/3] Add documentation for Dispatcher

--- Changes since v2: - rebased on master - moved handle_async_done removal hunk to first patch server/dispatcher.h | 97 - 1 file changed, 82 insertions(+), 15 deletions(-) diff --git a/server/dispatcher.h b/server/dispatcher.h index

[Spice-devel] [PATCH spice-server v3 1/3] Dispatcher: remove async_done callback

This callback was only executed for message types that were registered with DISPATCHER_ASYNC ack type. However, the async_done handler was called immediately after the message-specific handler and was called in the same thread, so the async_done stuff can just as easily be done from within the

[Spice-devel] [PATCH spice-server v3 2/3] MainDispatcher: use correct argument type

For dispatcher_register_handler(), use 'false' instead of 0 since the last argument is a bool type now. Signed-off-by: Jonathon Jongsma --- Changes since v2: - new patch server/main-dispatcher.c | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git

[Spice-devel] [PATCH spice-server v4] Add documentation for Dispatcher

--- Changes since v3: - removed documentation for 'opaque' argument from dispatcher_new() server/dispatcher.h | 95 - 1 file changed, 80 insertions(+), 15 deletions(-) diff --git a/server/dispatcher.h b/server/dispatcher.h index

Re: [Spice-devel] [PATCH spice-server 4/6] red-qxl: Avoid to use AsyncCommand for GL_DRAW_ASYNC message

OK. The invalid cookie value is a bit odd, but looks safe. Acked-by: Jonathon Jongsma On Thu, 2017-09-07 at 12:40 +0100, Frediano Ziglio wrote: > AsyncCommand is used to handle asynchronous messages from the > dispatcher. > GL_DRAW_ASYNC is mainly using it to store the

[Spice-devel] [PATCH spice-server] RedChannel: Remove obsolete TODO comment

--- server/red-channel.h | 3 --- 1 file changed, 3 deletions(-) diff --git a/server/red-channel.h b/server/red-channel.h index 55cfa7e1d..b5510c301 100644 --- a/server/red-channel.h +++ b/server/red-channel.h @@ -160,9 +160,6 @@ bool red_channel_test_remote_cap(RedChannel *channel, uint32_t

[Spice-devel] [PATCH spice-html5 4/4] Display: Add support for the VP9 codec type

--- display.js | 26 -- spiceconn.js | 2 ++ webm.js | 12 ++-- 3 files changed, 32 insertions(+), 8 deletions(-) diff --git a/display.js b/display.js index 0868f91..abd5b1a 100644 --- a/display.js +++ b/display.js @@ -543,7 +543,8 @@

[Spice-devel] [PATCH spice-html5 3/4] Display: Implement change preferred video codec type message

--- display.js | 16 enums.js| 10 ++ spicemsg.js | 24 3 files changed, 50 insertions(+) diff --git a/display.js b/display.js index 60c79b4..0868f91 100644 --- a/display.js +++ b/display.js @@ -1261,3 +1261,19 @@

[Spice-devel] [PATCH spice-html5 1/4] spiceconn: Add function to test channel capabilities

This will be used in other commits. --- spiceconn.js | 19 +++ 1 file changed, 19 insertions(+) diff --git a/spiceconn.js b/spiceconn.js index 33e7388..78d5820 100644 --- a/spiceconn.js +++ b/spiceconn.js @@ -243,6 +243,9 @@ SpiceConn.prototype = else if (this.state ==

[Spice-devel] [PATCH spice-html5 2/4] Display: Implement change preferred compression message

--- display.js | 16 enums.js| 10 ++ spicemsg.js | 20 3 files changed, 46 insertions(+) diff --git a/display.js b/display.js index 7719b23..60c79b4 100644 --- a/display.js +++ b/display.js @@ -1245,3 +1245,19 @@ function

[Spice-devel] [PATCH spice-html5 0/4] display: Implement preferred compression messages

These patches implement spice messages for changing image and video compression. The last patch adds support for VP9 decoding. Tomáš Bohdálek (4): spiceconn: Add function to test channel capabilities Display: Implement change preferred compression message Display: Implement change preferred

Re: [Spice-devel] [PATCH spice-server v2 10/15] test-display-base: Always compile with AUTOMATED_TESTS enabled

Acked-by: Christophe Fergeau On Wed, Sep 06, 2017 at 05:26:54PM +0100, Frediano Ziglio wrote: > There's no need to not compile this feature, it just enable > a parameters which must be passed in order to change test > behaviour. > > Signed-off-by: Frediano Ziglio

Re: [Spice-devel] [PATCH spice-server v2 14/15] tests: Make test-two-servers work

Acked-by: Christophe Fergeau On Wed, Sep 06, 2017 at 05:26:55PM +0100, Frediano Ziglio wrote: > This test runs 2 spice server in one program. > Use two different tcp port to be able to connect to both servers. > > Signed-off-by: Frediano Ziglio > --- >

Re: [Spice-devel] [PATCH spice-server v2 03/15] test-display-base: Avoid usage after free when the wakeup timer is freed

Acked-by: Christophe Fergeau On Wed, Sep 06, 2017 at 05:26:53PM +0100, Frediano Ziglio wrote: > The wakeup timer is used by the worker thread and by the > main thread. > Destroying the object before destroying the worker thread > can lead to use after free. > Destroying the

Re: [Spice-devel] [PATCH spice-server v2 3/3] Fix crash attempting to connect duplicate channels

> > On Wed, 2017-08-30 at 10:36 +0100, Frediano Ziglio wrote: > > You could easily trigger this issue using multiple monitors and > > a modified spice-gtk client with this patch: > > > > --- a/src/channel-main.c > > +++ b/src/channel-main.c > > @@ -1699,6 +1699,7 @@ static gboolean

Re: [Spice-devel] [PATCH spice-server v2 1/2] Introduce some macros to help declaring new GObject

On Wed, Sep 06, 2017 at 03:42:29PM +0100, Frediano Ziglio wrote: > The macros will implement most of the boilerplate needed > to declare an object. > Their usage are similar to GLib G_DECLARE_*_TYPE macros. Can we/should we use the GLib provided macros when they are available, and copy/paste the

Re: [Spice-devel] [PATCH spice-server v2 1/2] Introduce some macros to help declaring new GObject

> > On Wed, Sep 06, 2017 at 03:42:29PM +0100, Frediano Ziglio wrote: > > The macros will implement most of the boilerplate needed > > to declare an object. > > Their usage are similar to GLib G_DECLARE_*_TYPE macros. > > Can we/should we use the GLib provided macros when they are available, >

Re: [Spice-devel] [PATCH spice-server 1/2] Dispatcher: remove async_done callback

> > This callback was only executed for message types that were registered > with DISPATCHER_ASYNC ack type. However, the async_done handler was > called immediately after the message-specific handler and was called in > the same thread, so the async_done stuff can just as easily be done from >

Re: [Spice-devel] [PATCH spice-server 2/2] Add documentation for Dispatcher

> > --- > server/dispatcher.h | 102 > +--- > 1 file changed, 82 insertions(+), 20 deletions(-) > > diff --git a/server/dispatcher.h b/server/dispatcher.h > index eb93c1358..ed3ecf6bb 100644 > --- a/server/dispatcher.h > +++ b/server/dispatcher.h