Re: [PATCH] Fix for segmentation fault due to outdated list link pointers

2011-10-24 Thread Kristian Høgsberg
On Thu, Oct 20, 2011 at 3:38 PM, Jonas Ådahl wrote: > Hi, > > Attached you will find two patches (one for wayland.git and one for > wayland-demos.git) that fixes some crashes I've seen. > > The reason has been that after a surface destroyed listener gets > called, its link is not reset. This may r

Re: [PATCH 0/2] Fix release of wrong buffer on frame complete

2011-10-24 Thread Kristian Høgsberg
2011/10/21 Kristian Høgsberg : > On Tue, Oct 18, 2011 at 10:02 AM, Tiago Vignatti > wrote: >> From: Ander Conselvan de Oliveira >> >> If scanning out a client buffer, the compositor will send a buffer >> release event for the front buffer as soon as the flip completes. This >> series change the b

Re: [PATCH] compositor: fix drag segfault

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 10:34 AM, Pekka Paalanen wrote: > Without this fix, the dnd demo would make the demo compositor crash in > shell.c:drag_offer() because resource->data is NULL. > > Initialise resource->data in shell_create_drag(). Right, thanks. Kristian > Signed-off-by: Pekka Paalanen

Re: [PATCH] Add missing include for gio.h.

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 11:49 AM, Eugeni Dodonov wrote: > This is needed by GFile, used by view.c. Thanks, committed. > Signed-off-by: Eugeni Dodonov > --- >  clients/view.c |    1 + >  1 files changed, 1 insertions(+), 0 deletions(-) > > diff --git a/clients/view.c b/clients/view.c > index d1b

Re: [PATCH 1/2] compositor: don't schedule idle_repaint from calls to repaint

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 3:22 PM, Ander Conselvan de Oliveira wrote: > Em 24-10-2011 18:57, Kristian Høgsberg escreveu: >> >> On Mon, Oct 24, 2011 at 9:30 AM, Ander Conselvan de Oliveira >>  wrote: >>> >>> wlsc_output_repaint may call wlsc_surface_damage indirectly through >>> wlsc_output_set_curs

Re: [PATCH 1/2] compositor: don't schedule idle_repaint from calls to repaint

2011-10-24 Thread Ander Conselvan de Oliveira
Em 24-10-2011 18:57, Kristian Høgsberg escreveu: On Mon, Oct 24, 2011 at 9:30 AM, Ander Conselvan de Oliveira wrote: wlsc_output_repaint may call wlsc_surface_damage indirectly through wlsc_output_set_cursor. If this happens in the call made from repaint, one ends up with repaint being called

Re: [PATCH 1/2] compositor: don't schedule idle_repaint from calls to repaint

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 9:30 AM, Ander Conselvan de Oliveira wrote: > wlsc_output_repaint may call wlsc_surface_damage indirectly through > wlsc_output_set_cursor. If this happens in the call made from repaint, > one ends up with repaint being called from both idle_repaint and > wlsc_output_finish

[PATCH] Add missing include for gio.h.

2011-10-24 Thread Eugeni Dodonov
This is needed by GFile, used by view.c. Signed-off-by: Eugeni Dodonov --- clients/view.c |1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/clients/view.c b/clients/view.c index d1ba3b9..0eec1f9 100644 --- a/clients/view.c +++ b/clients/view.c @@ -31,6 +31,7 @@ #include

Re: [PATCH 0/9] evdev: events processing after EV_SYN and hotplug support

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 10:42 AM, Tiago Vignatti wrote: > Hi, > > There it goes a bit of refactoring on the evdev code to become more, well, > usable. Nothing exceptional really and the most apparent change is the hotplug > feature added... ahh, one would say I'm a thief maybe because a lot of cod

Re: [PATCH 8/9] evdev: add hotplug support

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 10:42 AM, Tiago Vignatti wrote: > Signed-off-by: Tiago Vignatti Very nice, we just need to make sure we filter hotplugged devices for the seat number as we do in the initial device discovery. Kristian > --- >  compositor/evdev.c |  112 > +++

Re: [PATCH 6/9] evdev: organize main event processing loop

2011-10-24 Thread Kristian Høgsberg
2011/10/24 Kristian Høgsberg : > On Mon, Oct 24, 2011 at 10:42 AM, Tiago Vignatti > wrote: >> No functional changes on this commit, but the general idea was stolen from X >> input evdev with the intents of copying pretty much the way done there which >> will ease next when implementing a smarter m

Re: [PATCH 6/9] evdev: organize main event processing loop

2011-10-24 Thread Kristian Høgsberg
On Mon, Oct 24, 2011 at 10:42 AM, Tiago Vignatti wrote: > No functional changes on this commit, but the general idea was stolen from X > input evdev with the intents of copying pretty much the way done there which > will ease next when implementing a smarter method for computing valuators. > > Sig

Re: [PATCH 0/9] evdev: events processing after EV_SYN and hotplug support

2011-10-24 Thread Eugeni Dodonov
On Mon, Oct 24, 2011 at 12:42, Tiago Vignatti wrote: > Hi, > > There it goes a bit of refactoring on the evdev code to become more, well, > usable. Nothing exceptional really and the most apparent change is the > hotplug > feature added... ahh, one would say I'm a thief maybe because a lot of code

[PATCH 9/9] evdev: simplify error path for device create

2011-10-24 Thread Tiago Vignatti
Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 31 +-- 1 files changed, 13 insertions(+), 18 deletions(-) diff --git a/compositor/evdev.c b/compositor/evdev.c index 18ebe04..7f4b57b 100644 --- a/compositor/evdev.c +++ b/compositor/evdev.c @@ -490,33 +490,28

[PATCH 8/9] evdev: add hotplug support

2011-10-24 Thread Tiago Vignatti
Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 112 +--- 1 files changed, 106 insertions(+), 6 deletions(-) diff --git a/compositor/evdev.c b/compositor/evdev.c index b81e513..18ebe04 100644 --- a/compositor/evdev.c +++ b/compositor/evdev.

[PATCH 7/9] evdev: queue events and process them only at EV_SYN time

2011-10-24 Thread Tiago Vignatti
it fix a bug of buttons events being queuing up before motion ones, which caused the behaviour of taps/clicks not being caught in the needed time. This was tested with direct and indirect input devices. Tested-by: Tiago Vignatti Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 188 ++

[PATCH 6/9] evdev: organize main event processing loop

2011-10-24 Thread Tiago Vignatti
No functional changes on this commit, but the general idea was stolen from X input evdev with the intents of copying pretty much the way done there which will ease next when implementing a smarter method for computing valuators. Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 124 +++

[PATCH 4/9] evdev: record absolute events were queued inside evdev_input_device

2011-10-24 Thread Tiago Vignatti
No functional changes. Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 20 +--- 1 files changed, 9 insertions(+), 11 deletions(-) diff --git a/compositor/evdev.c b/compositor/evdev.c index 3027c7d..49897bf 100644 --- a/compositor/evdev.c +++ b/compositor/evdev.c @@ -37,

[PATCH 5/9] evdev: record time inside evdev_input_device

2011-10-24 Thread Tiago Vignatti
No functional changes. Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 16 1 files changed, 8 insertions(+), 8 deletions(-) diff --git a/compositor/evdev.c b/compositor/evdev.c index 49897bf..3661236 100644 --- a/compositor/evdev.c +++ b/compositor/evdev.c @@ -47,13 +4

[PATCH 3/9] evdev: record valuators coordinates in evdev_input_device

2011-10-24 Thread Tiago Vignatti
No functional changes. It will ease the process of converting to a more smart valuators scheme, similar to the one X evdev driver has. Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 46 +++--- 1 files changed, 23 insertions(+), 23 deletions(-) d

[PATCH 2/9] evdev: delete unused fields from evdev_input_device

2011-10-24 Thread Tiago Vignatti
No functional changes. This all should be gone in commit 96c8be98. Signed-off-by: Tiago Vignatti --- compositor/evdev.c |5 + 1 files changed, 1 insertions(+), 4 deletions(-) diff --git a/compositor/evdev.c b/compositor/evdev.c index 3f8b9ad..a0cd1d5 100644 --- a/compositor/evdev.c +++

[PATCH 1/9] evdev: use struct calibration for those kind of fields

2011-10-24 Thread Tiago Vignatti
No functional changes. Signed-off-by: Tiago Vignatti --- compositor/evdev.c | 32 +++- 1 files changed, 19 insertions(+), 13 deletions(-) diff --git a/compositor/evdev.c b/compositor/evdev.c index f0cdf30..3f8b9ad 100644 --- a/compositor/evdev.c +++ b/compositor/ev

[PATCH 0/9] evdev: events processing after EV_SYN and hotplug support

2011-10-24 Thread Tiago Vignatti
Hi, There it goes a bit of refactoring on the evdev code to become more, well, usable. Nothing exceptional really and the most apparent change is the hotplug feature added... ahh, one would say I'm a thief maybe because a lot of code was based on the X input evdev driver :) The same work is here:

[PATCH] compositor: fix drag segfault

2011-10-24 Thread Pekka Paalanen
Without this fix, the dnd demo would make the demo compositor crash in shell.c:drag_offer() because resource->data is NULL. Initialise resource->data in shell_create_drag(). Signed-off-by: Pekka Paalanen --- compositor/shell.c |1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --g

[PATCH 2/2] compositor: fix whitespace error in wlsc_output_set_cursor

2011-10-24 Thread Ander Conselvan de Oliveira
Signed-off-by: Ander Conselvan de Oliveira --- compositor/compositor.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/compositor/compositor.c b/compositor/compositor.c index b67aa33..587c9ed 100644 --- a/compositor/compositor.c +++ b/compositor/compositor.c @@ -828,7

[PATCH 1/2] compositor: don't schedule idle_repaint from calls to repaint

2011-10-24 Thread Ander Conselvan de Oliveira
wlsc_output_repaint may call wlsc_surface_damage indirectly through wlsc_output_set_cursor. If this happens in the call made from repaint, one ends up with repaint being called from both idle_repaint and wlsc_output_finish_frame. Fix this by setting output->repaint_scheduled to 1 before calling wl

RE: some documents for buiding and understanding wayland

2011-10-24 Thread Zhao, Juan J
Hi Tiago, > -Original Message- > yes, it's pretty much the same. The code for the page is here, so your > patches should be against it: > >http://cgit.freedesktop.org/wayland/wayland-web/ My platform is untunt11.04, there are some differences for the default installed packages and t

Re: some documents for buiding and understanding wayland

2011-10-24 Thread Tiago Vignatti
On 10/24/2011 01:25 PM, Zhao, Juan J wrote: For the wayland building sheet, the actual building steps are pretty similar from the ones on the wiki page. I guess we could emphasize there though, a bit more how to circumvent that eglInitialize error; I have myself forgot to install xcb-dri2 and x

RE: some documents for buiding and understanding wayland

2011-10-24 Thread Zhao, Juan J
Hi Tiago, > For the wayland building sheet, the actual building steps are pretty > similar from the ones on the wiki page. I guess we could emphasize there > though, a bit more how to circumvent that eglInitialize error; I have > myself forgot to install xcb-dri2 and xcb-fixes and seen also some g

Re: some documents for buiding and understanding wayland

2011-10-24 Thread Tiago Vignatti
Hi Juan, On 10/23/2011 01:08 PM, Zhao, Juan J wrote: I wrote some documents about wayland. Just share this here and hope this may help you. Welcome to figure out the errors, comments, adding your thoughts and your suggestions. Thanks for doing this! For the wayland building sheet, the actua