Re: [Spice-devel] [PATCH spice-server 0/1] prevent crash on video stream reattach

2019-03-08 Thread Frediano Ziglio
> > I experienced some crashes with qemu 3.1.0 compiled with libspice-server > 0.14.0 on Gentoo. > > The problem reproduced reliably with a guest running Ubuntu 18.04.2 LTS. > If I connect a viewer at system startup, I would get a crash just after > the fade-in of the login prompt in GDM. > > In

Re: [Spice-devel] [PATCH spice-common 6/9] Generate automatically most C message declarations

2019-03-08 Thread Frediano Ziglio
> > Acked-by: Christophe Fergeau > > only one caveat is that some of the structs are getting renamed to > equivalent ones (SpiceMsgDisplayDrawBlend/SpiceMsgDisplayDrawCopy, > SpiceMsgcMainAgentStart/SpiceMsgcMainAgentTokens and a few more), which > is fine because they are typedef'ed to the othe

Re: [Spice-devel] [PATCH spice-common 5/9] Allow to generate C declarations for spice.proto

2019-03-08 Thread Frediano Ziglio
> > On Sun, Mar 03, 2019 at 07:10:26PM +, Frediano Ziglio wrote: > > Generate and include C declarations. > > Next patch will use this facility. > > Since none of the spice.proto types are decorated with @declare, adding > the #include in messages.h won't have any bad consequences > Are you

Re: [Spice-devel] [PATCH spice-common 4/9] codegen: Allows to generate C declarations automatically

2019-03-08 Thread Frediano Ziglio
> > On Sun, Mar 03, 2019 at 07:10:25PM +, Frediano Ziglio wrote: > > Allows to specify a @declare attribute for messages and structure > > that can generate the needed C structures. > > > > Signed-off-by: Frediano Ziglio > > --- > > python_modules/ptypes.py | 64

Re: [Spice-devel] [PATCH spice-common 3/9] codegen: Generate headers while generating code

2019-03-08 Thread Frediano Ziglio
> > On Sun, Mar 03, 2019 at 07:10:24PM +, Frediano Ziglio wrote: > > Python script generates code and header together however allowed > > to save only one of them. > > Allows to save both of them together to reduce number of time > > we call Python script. > > > > Signed-off-by: Frediano Zigl

Re: [Spice-devel] [spice v1 2/2] display-channel: monitors config debug: add head number

2019-03-08 Thread Frediano Ziglio
> > From: Victor Toso > > The difference is subtle but compared to what client receives, this > could help identify values set to the wrong head, e.g: > > First we received: > | display-channel.c:180:monitors_config_debug: monitors config count:2 max:4 > | display-channel.c:184:monitors_confi

Re: [Spice-devel] [spice v1 1/2] display-channel: monitors_config: add 200ms delay

2019-03-08 Thread Frediano Ziglio
> > From: Victor Toso > > The device driver might take a few interactions to reconfigure all > displays but in each change it can send it down to spice-server. The > client is not interested in temporary states, only the final monitor > config can be helpful. > > This patch adds a 200ms delay t

[Spice-devel] [PATCH spice-server 1/1] video-stream: prevent crash on stream reattach

2019-03-08 Thread Douglas Paul
Check that the video encoder has been created before sending frame drop notification through it. Signed-off-by: Douglas Paul --- server/video-stream.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/server/video-stream.c b/server/video-stream.c index b624093e..19795098 100

[Spice-devel] [PATCH spice-server 0/1] prevent crash on video stream reattach

2019-03-08 Thread Douglas Paul
I experienced some crashes with qemu 3.1.0 compiled with libspice-server 0.14.0 on Gentoo. The problem reproduced reliably with a guest running Ubuntu 18.04.2 LTS. If I connect a viewer at system startup, I would get a crash just after the fade-in of the login prompt in GDM. Interestingly, I us

[Spice-devel] [spice v1 1/2] display-channel: monitors_config: add 200ms delay

2019-03-08 Thread Victor Toso
From: Victor Toso The device driver might take a few interactions to reconfigure all displays but in each change it can send it down to spice-server. The client is not interested in temporary states, only the final monitor config can be helpful. This patch adds a 200ms delay to wait for more cha

[Spice-devel] [spice v1 2/2] display-channel: monitors config debug: add head number

2019-03-08 Thread Victor Toso
From: Victor Toso The difference is subtle but compared to what client receives, this could help identify values set to the wrong head, e.g: First we received: | display-channel.c:180:monitors_config_debug: monitors config count:2 max:4 | display-channel.c:184:monitors_config_debug: +0+0 1015x

Re: [Spice-devel] [PATCH spice-common 9/9] codegen: Rename --prefix parameter to --suffix

2019-03-08 Thread Christophe Fergeau
On Thu, Mar 07, 2019 at 04:05:27PM -0500, Frediano Ziglio wrote: > > > > On Sun, Mar 03, 2019 at 07:10:30PM +, Frediano Ziglio wrote: > > > The option is used to add a suffix to public functions, not a > > > prefix. > > > > Sometimes it's a suffix to a prefix, sometimes it's indeed just a > >

Re: [Spice-devel] [PATCH spice-gtk 1/3] clipboard: accept grab only from the side with keyboard focus

2019-03-08 Thread Marc-André Lureau
Hi On Fri, Mar 8, 2019 at 1:41 PM Jakub Janku wrote: > > Hi, > > On Fri, Mar 8, 2019 at 1:15 PM Marc-André Lureau > wrote: > > > > Hi > > > > On Thu, Mar 7, 2019 at 10:36 PM Jakub Janku wrote: > > > > > > Hi, > > > > > > thanks for having a look! > > > > > > On Wed, Mar 6, 2019 at 6:42 PM Marc-

[Spice-devel] [PATCH spice-server 1/1] video-stream: prevent crash on stream reattach

2019-03-08 Thread Douglas Paul
Check that the video encoder has been created before sending frame drop notification through it. Signed-off-by: Douglas Paul --- server/video-stream.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/server/video-stream.c b/server/video-stream.c index b624093e..19795098 100

[Spice-devel] [PATCH spice-server 0/1] prevent crash on video stream reattach

2019-03-08 Thread Douglas Paul
I experienced some crashes with qemu 3.1.0 compiled with libspice-server 0.14.0 on Gentoo. The problem reproduced reliably with a guest running Ubuntu 18.04.2 LTS. If I connect a viewer at system startup, I would get a crash just after the fade-in of the login prompt in GDM. Interestingly, I us

Re: [Spice-devel] [PATCH spice-gtk 1/3] clipboard: accept grab only from the side with keyboard focus

2019-03-08 Thread Jakub Janku
Hi, On Fri, Mar 8, 2019 at 1:15 PM Marc-André Lureau wrote: > > Hi > > On Thu, Mar 7, 2019 at 10:36 PM Jakub Janku wrote: > > > > Hi, > > > > thanks for having a look! > > > > On Wed, Mar 6, 2019 at 6:42 PM Marc-André Lureau > > wrote: > > > > > > Hi > > > > > > On Thu, Feb 28, 2019 at 8:12 PM

[Spice-devel] [PATCH spice-server 1/2] Update spice-common submodule

2019-03-08 Thread Frediano Ziglio
This brings in the following changes: Frediano Ziglio (11): codegen: Document ptr_array attribute codegen: Use a better type for pointer converted to integer codegen: Reduce indentation codegen: Fix c_type result for TypeAlias codegen: Check wrong attribute code

[Spice-devel] [PATCH spice-server 2/2] Remove support for 64 bit pointers on protocol

2019-03-08 Thread Frediano Ziglio
Import "codegen: Remove support for --ptrsize" change from spice-common and update code accordingly. Signed-off-by: Frediano Ziglio --- server/char-device.c | 2 +- server/dcc-send.c| 2 +- server/reds.c| 2 +- server/smartcard.c | 2 +- subprojects/spice-common | 2

Re: [Spice-devel] [PATCH spice-gtk 1/3] clipboard: accept grab only from the side with keyboard focus

2019-03-08 Thread Marc-André Lureau
Hi On Thu, Mar 7, 2019 at 10:36 PM Jakub Janku wrote: > > Hi, > > thanks for having a look! > > On Wed, Mar 6, 2019 at 6:42 PM Marc-André Lureau > wrote: > > > > Hi > > > > On Thu, Feb 28, 2019 at 8:12 PM Jakub Janků wrote: > > > > > > If two grab messages in opposite directions "meet" on their

Re: [Spice-devel] [PATCH spice-common 7/9] codegen: Remove support for --ptrsize

2019-03-08 Thread Frediano Ziglio
> > On Thu, Mar 07, 2019 at 04:01:02PM -0500, Frediano Ziglio wrote: > > > > > > Acked-by: Christophe Fergeau > > > > > > > Thanks, looking at reply to 9/9 however I think one thing is not clear from > > what I wrote in the commit message. > > The reason is not only that was used in protocol 1

Re: [Spice-devel] [win32/vd_agent v2 2/3] introduce turn_monitor_off method of WDDM interface

2019-03-08 Thread Yuri Benditovich
On Fri, Mar 8, 2019 at 1:34 PM Frediano Ziglio wrote: > > > > > Adding method 'turn_monitor_off' to WDDM interface. > > It sends QXL_ESCAPE_MONITOR_CONFIG escape with zeroed > > display size to qxl-wddm-dod driver. > > > > Signed-off-by: Yuri Benditovich > > --- > > vdagent/display_configuration

Re: [Spice-devel] [win32/vd_agent v2 2/3] introduce turn_monitor_off method of WDDM interface

2019-03-08 Thread Frediano Ziglio
> > Adding method 'turn_monitor_off' to WDDM interface. > It sends QXL_ESCAPE_MONITOR_CONFIG escape with zeroed > display size to qxl-wddm-dod driver. > > Signed-off-by: Yuri Benditovich > --- > vdagent/display_configuration.cpp | 18 ++ > vdagent/display_configuration.h | 2

Re: [Spice-devel] [win32/vd_agent v2 3/3] make display disconnection persistent

2019-03-08 Thread Frediano Ziglio
> > https://bugzilla.redhat.com/show_bug.cgi?id=1477191 > Currently on Win10 when one of multiple displays is > disconnected by the client, the qxl-wddm-dod driver > does not receive any update, the system just stops using > this adapter for rendering. As a result, the 'disconnected' > state of th

Re: [Spice-devel] [PATCH spice-common 7/9] codegen: Remove support for --ptrsize

2019-03-08 Thread Christophe Fergeau
On Thu, Mar 07, 2019 at 04:01:02PM -0500, Frediano Ziglio wrote: > > > > Acked-by: Christophe Fergeau > > > > Thanks, looking at reply to 9/9 however I think one thing is not clear from > what I wrote in the commit message. > The reason is not only that was used in protocol 1 only and now is no

[Spice-devel] [win32/vd_agent v2 0/3] Fix for BZ#1477191

2019-03-08 Thread Yuri Benditovich
https://bugzilla.redhat.com/show_bug.cgi?id=1477191 When one of multiple qxl-wddm-dod displays is turned off by the client, on Windows 10 the display driver does not receive any update, so on next connection the adapter is recognized as active. This series fixes the problem by sending QXL_ESCAPE_MO

[Spice-devel] [win32/vd_agent v2 1/3] construct WDDMMonitorConfigEscape also with NULL DisplayMode

2019-03-08 Thread Yuri Benditovich
Adding possibility to construct WDDMMonitorConfigEscape object with NULL parameter of DisplayMode. In this case all the fields depending on DisplayMode are set to zero. Signed-off-by: Yuri Benditovich --- vdagent/display_configuration.cpp | 8 1 file changed, 4 insertions(+), 4 deletion

[Spice-devel] [win32/vd_agent v2 2/3] introduce turn_monitor_off method of WDDM interface

2019-03-08 Thread Yuri Benditovich
Adding method 'turn_monitor_off' to WDDM interface. It sends QXL_ESCAPE_MONITOR_CONFIG escape with zeroed display size to qxl-wddm-dod driver. Signed-off-by: Yuri Benditovich --- vdagent/display_configuration.cpp | 18 ++ vdagent/display_configuration.h | 2 +- 2 files changed

[Spice-devel] [win32/vd_agent v2 3/3] make display disconnection persistent

2019-03-08 Thread Yuri Benditovich
https://bugzilla.redhat.com/show_bug.cgi?id=1477191 Currently on Win10 when one of multiple displays is disconnected by the client, the qxl-wddm-dod driver does not receive any update, the system just stops using this adapter for rendering. As a result, the 'disconnected' state of the adapter is no