On Wednesday, 2017-03-01 17:45:12 +0100, Olivier Fourdan wrote:
> Not all compositors allow for customizing the Xwayland command line,
> gnome-shell/mutter for example have the command line and path to
> Xwayland binary hardcoded, which makes it harder for users to disable
> glamor acceleration in
Not all compositors allow for customizing the Xwayland command line,
gnome-shell/mutter for example have the command line and path to
Xwayland binary hardcoded, which makes it harder for users to disable
glamor acceleration in Xwayland (glamor being used by default).
Add an environment variable XW
Hi all,
I am seeing quite a few Xwayland crashes related to glamor.
Various issues, could be with glamor itself or with the drivers (like the
issues I witness with nv30), whatever.
To investigate those issues (or even unblock affected users) it's often
useful to disable glamor -even temporarily-
At the bottom of weston_output_finish_frame(), code exists to account
for flips which have missed the repaint window, by shifting them to lock
on to the next repaint window rather than repainting immediately.
This code only accounted for flips which missed their target by one
repaint window. If th
Paralleling timespec_to_nsec, converts to milliseconds.
Signed-off-by: Daniel Stone
Reviewed-by: Pekka Paalanen
---
libweston/compositor.c | 2 +-
shared/timespec-util.h | 11 +++
tests/timespec-test.c | 9 +
3 files changed, 21 insertions(+), 1 deletion(-)
v2: No changes.
In preparation for grouping output repaint together where possible,
switch the per-output repaint timer, to a global timer which iterates
across all outputs.
This is implemented by storing the absolute time for the next repaint
for each output locally, and maintaining a global timer which iterates
It is only used as a binary value.
Signed-off-by: Daniel Stone
---
libweston/compositor-drm.c | 2 +-
libweston/compositor-fbdev.c | 2 +-
libweston/compositor.c | 4 ++--
libweston/compositor.h | 5 -
4 files changed, 8 insertions(+), 5 deletions(-)
v2: New in this revision.
Implement new repaint_begin and repaint_flush hooks inside
weston_backend, allowing backends to gang together repaints which
trigger at the same time.
Signed-off-by: Daniel Stone
---
libweston/compositor-drm.c | 5 +++--
libweston/compositor-fbdev.c| 3 ++-
libweston/compositor-headle
repaint_scheduled is actually cleverly a tristate. There are three
possible conditions for the repaint loop to be in at any time:
- loop idle; no repaint will occur until specifically requested, which
may be never (repaint_scheduled == 0)
- repaint scheduled: the compositor has definitively
Add a (timespec) = (timespec) + (msec) helper, to save intermediate
conversions in its users.
Signed-off-by: Daniel Stone
Reviewed-by: Pekka Paalanen
---
shared/timespec-util.h | 12
tests/timespec-test.c | 11 +++
2 files changed, 23 insertions(+)
v2: Trivially reimpleme
On startup, we cannot lock on to the repaint timer because it is unknown
to us. We deal with this by claiming that the moment of entry into the
repaint loop is the moment a frame returned, causing finish_frame to
delay our initial repaint to (refresh_time - repaint_delay), typically
around 9ms of u
Rather than determining the time until next-frame repaint in relative
space (time until repaint), determine it first in absolute space, and
then later convert this to relative.
This will later allow us to store these per-output, so we can have a
single idle timer which will allow us to aggregate m
Add helpers to subtract two timespecs, then return the difference in
either milliseconds or nanoseconds. These will be used to compare
timestamps during the repaint cycle.
Signed-off-by: Daniel Stone
Suggested-by: Pekka Paalanen
---
shared/timespec-util.h | 26 ++
tests/
Hi,
I'm submitting v2 of this series after Pekka's review. At this point,
we now keep all time values in absolute timespec structs, only using
normalisation to nsec for relative comparisons. This required the
addition of subtraction helpers, suggested by Pekka, but reimplemented
to work with signed
Add a (timespec) = (timespec) + (nsec) helper, to save intermediate
conversions to nanoseconds in its users.
Signed-off-by: Daniel Stone
Reviewed-by: Pekka Paalanen
---
Makefile.am| 10
shared/timespec-util.h | 21 +
tests/timespec-test.c | 124 ++
Hi,
On 1 March 2017 at 10:29, Pekka Paalanen wrote:
> On Tue, 28 Feb 2017 14:46:21 + Daniel Stone wrote:
>> On 28 February 2017 at 10:59, Pekka Paalanen wrote:
>> > The important thing to document about this change is that is exchanges
>> > a (almost never hit) double-unref into a (always h
On Tue, 28 Feb 2017 14:46:21 +
Daniel Stone wrote:
> Hi,
>
> On 28 February 2017 at 10:59, Pekka Paalanen wrote:
> > On Mon, 27 Feb 2017 22:58:48 + Daniel Stone
> > wrote:
> >> It's indeed almost certainly a bugfix: if you have one client buffer
> >> with multiple views, both of whi
HI,
On 01-03-17 02:48, Peter Hutterer wrote:
If a touch started hovering in the main area, the button state would start
with AREA and never move to the real button state, despite the finger
triggering the pressure thresholds correctly in one of the areas.
This could even happen across touch seq
18 matches
Mail list logo
