Okay, to sort this mess out: I am pretty sure there are actually two
different bugs, or at least two different manifestations of the same
bug:

*** Bug A
- Reported by GI on May 7th

  "When my display is rotated, I find that the drawing of regular
  strokes is delayed. That is, if I move the pen to the right by one
  inch, it takes a second for me to see the stroke appear"

  I'll add to that: If I draw a continuous line, I see every second or
  split second a small part of the line appearing.

- Happens only when display is rotated
- --sync independent

- Enabled INPUT_DEBUG on "visible" terminal make it disappear for some
  (Jonas May 10th)
  Not for all though, apparently (GI May 10th)

- Is not experienced with xf86-video-intel<=2.14.0 or current
  upstream git (Jonas May 10th)

*** Bug B
- Reported by dmg on May 7th
  "This might or might not be related ... in some cases the
  strokes are not displayed when created (but they are saved)
  Only when the page is refreshed."

  I'll add to that: if I draw a continuous long line, the drawing
  stops at some random point and from there on only small stripes or
  blocks show the new lines. All is shown as soon as some drawing
  update is forced, as by hovering over the scrollbar or scrolling
  with cursor keys. (my description)

  "they look like bands across the entire page --rectangles to be more
  precise, where nothing is redrawn" (dmg 10th)

- Affects rotated as well as non-rotated (dmg 9th)
- --sync independent
- Doesn't affect X201T, but X61T (dmg 10th)
- Enabled INPUT_DEBUG on visible terminal make it disappear (dmg 10th)
  I can confirm this now here as well

- Perhaps connected to kernel changes between 2.6.32 and 2.6.38
  ("I rebooted, and upgraded to kernel 2.6.38-2.  Now I have a whole
  bunch of weird drawing problems even in unrotated mode" GI and
  another mail "but not on the 2.6.32 kernel" GI)

- on upstream xf86-video-intel and kernel 2.6.38, this bug still
  occurs, but ironically only in non-rotated mode (Jonas now)

I tried to be as thorough as I could, please complain if you think it
should be split another way, or some info is missing or whatever.

Good day,

Jonas


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Xournal-devel mailing list
Xournal-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xournal-devel

Reply via email to