Re: [Spice-devel] [PATCH spice-server] reds: fix not sending the mm-time after migration when there is no audio playback

2013-04-30 Thread Marc-André Lureau
ack On Tue, Apr 30, 2013 at 8:50 PM, Yonit Halperin wrote: > This bug results in the client dropping all the video frames after > migration in case that (1) the hosts involved in migration have different > mm-time; and that (2) there is no audio playback. > This is relvant only for the client t

Re: [Spice-devel] [PATCH spice-server] red_channel: fix not handling self pipe items in red_channel_client_release_item

2013-04-30 Thread Marc-André Lureau
ack On Tue, Apr 30, 2013 at 9:07 PM, Yonit Halperin wrote: > When a client disconnects, red_channel_client_pipe_clear is called. > Releasing pipe items of type == MIGRATE||EMPTY_MSG||PING > wasn't handled, and was passed to channel_cbs.release_item. > There, an error occured since the pipe item

Re: [Spice-devel] [PATCH spice-gtk v2 2/2] channel-display: protect video streaming from temporarily unsynced mm_time (migration related)

2013-04-30 Thread Marc-André Lureau
looks good, ack On Tue, Apr 30, 2013 at 8:24 PM, Yonit Halperin wrote: > rhbz#951664 > > When the src and dst servers have different mm-times, we can > hit a case when for a short period of time the session mm-time and > the video mm-time are not synced. If the video mm-time is much > bigger th

Re: [Spice-devel] [PATCH spice-gtk v2 1/2] spice-session: new signal for notifying on a significant change in mm-time

2013-04-30 Thread Marc-André Lureau
On Tue, Apr 30, 2013 at 8:24 PM, Yonit Halperin wrote: > mm-time can change after migration. This can cause video synchronization > problems after migration if not handled appropriately (see rhbz#951664). > --- > gtk/spice-session.c | 29 + > gtk/spice-session.h | 3

[Spice-devel] [PATCH spice-server] red_channel: fix not handling self pipe items in red_channel_client_release_item

2013-04-30 Thread Yonit Halperin
When a client disconnects, red_channel_client_pipe_clear is called. Releasing pipe items of type == MIGRATE||EMPTY_MSG||PING wasn't handled, and was passed to channel_cbs.release_item. There, an error occured since the pipe items were not recognized. --- server/red_channel.c | 3 +++ 1 file change

[Spice-devel] [PATCH spice-server] reds: fix not sending the mm-time after migration when there is no audio playback

2013-04-30 Thread Yonit Halperin
This bug results in the client dropping all the video frames after migration in case that (1) the hosts involved in migration have different mm-time; and that (2) there is no audio playback. This is relvant only for the client that was connected during the migration. rhbz#958276 --- server/reds.c

[Spice-devel] [PATCH spice-gtk v2 2/2] channel-display: protect video streaming from temporarily unsynced mm_time (migration related)

2013-04-30 Thread Yonit Halperin
rhbz#951664 When the src and dst servers have different mm-times, we can hit a case when for a short period of time the session mm-time and the video mm-time are not synced. If the video mm-time is much bigger than the session mm-time, the next stream rendering will be scheduled to (video-mm-time

[Spice-devel] [PATCH spice-gtk v2 1/2] spice-session: new signal for notifying on a significant change in mm-time

2013-04-30 Thread Yonit Halperin
mm-time can change after migration. This can cause video synchronization problems after migration if not handled appropriately (see rhbz#951664). --- gtk/spice-session.c | 29 + gtk/spice-session.h | 3 ++- 2 files changed, 31 insertions(+), 1 deletion(-) diff --git a

Re: [Spice-devel] [RFC PATCH spice-gtk] channel-display: protect video streaming from temporally unsynced mm_time (migration related)

2013-04-30 Thread Yonit Halperin
On 04/30/2013 07:59 AM, Marc-André Lureau wrote: Hi, Thanks for tracking this down. First, the patch doesn't compile. Sorry, forgot to add channel-display-priv.h to the patch This is indeed what I imagined, that the mm time is not monotonic over migrations on different machines (iirc, I was t

Re: [Spice-devel] [RFC PATCH spice-gtk] channel-display: protect video streaming from temporally unsynced mm_time (migration related)

2013-04-30 Thread Marc-André Lureau
Hi, Thanks for tracking this down. First, the patch doesn't compile. This is indeed what I imagined, that the mm time is not monotonic over migrations on different machines (iirc, I was told the stream were stopped before migration) Imho, there should be also a base mm time after migration compl

Re: [Spice-devel] [win-vdagent PATCH] vdagent: protect against NULL entry in _displays

2013-04-30 Thread agilboa
ack On 04/30/2013 01:10 PM, Uri Lublin wrote: rhbz#958051 It may be that a _displays entry will be NULL. I encountered it when running with multiple QXL devices, for one of which the driver failed to load since it was "out of resources" Iterations over _displays should handle that case. I foun

[Spice-devel] [win-vdagent PATCH] vdagent: protect against NULL entry in _displays

2013-04-30 Thread Uri Lublin
rhbz#958051 It may be that a _displays entry will be NULL. I encountered it when running with multiple QXL devices, for one of which the driver failed to load since it was "out of resources" Iterations over _displays should handle that case. I found three such iterations, and fixed them in this p

Re: [Spice-devel] [PATCH spice-gtk] widget: don't grab mouse when switching to server mode

2013-04-30 Thread Christophe Fergeau
On Tue, Apr 30, 2013 at 01:44:25AM +0200, Marc-André Lureau wrote: > spice-gtk used to try taking the grab when switching to server mode, so > that the current mouse interaction, ex a drag, isn't interrupted if the s/ex/for example/ > agent dies. However, shuting down a VM will automatically try