Fundamental issues for the ASU

[Carsten Haitzler (The Rasterman)]

On Fri, 16 May 2008 16:25:47 +0200 Holger Freyther <zecke at openmoko.org> babbled:

> Hey,
> 
> for the ASU I only have two bugs that worry me:
> 
> 	- Xglamo eating all of my CPU after a while (94% now). I will try to
> get a backtrace. Anyone with a spare cycle to look into this? This issue only
> comes after the device is on for a long time(in my case over 20 hours). And
> we already have a open bug for this. Anyone interested in fixing it?
> 
> 	The backtace looks innocent:
> 		#0  0x401f88ac in gettimeofday () from /lib/libc.so.6
> 		#1  0x00014eb4 in GLAMOEngineWaitReal ()
> 		#2  0x00015270 in GLAMOFlushCMDQCache ()
> 		#3  0x00015e70 in GLAMOBlockHandler ()
> 		#4  0x0001f704 in BlockHandler ()
> 		#5  0x000ec274 in WaitForSomething ()
> 		#6  0x00038a2c in Dispatch ()
> 		#7  0x00022c60 in main ()
> 		#8  0x401909fc in __libc_start_main () from /lib/libc.so.6
> 		#9  0x00012614 in _start ()

gah! waiting for the cmdq to flush and it never does. thats bad. too much
context switching - lets keep this filed for now?

> 	- modem resume. The modem is waking up the host but no data is
> actually transferred on the serial. This might be a bug in the modem or the
> infamous flow control issue (first mentioned by Uli Luckas in 2005, and we at
> Openmoko have it as well, reported and investigated by mwester).

yup. i can confirm this is always happening to me too. i need to resolve
qt/qtopia vs illume behaviro issues right now - after that i can dig up the
glamo code and have a hack.

> 	I will investigate after fixing some kernel suspend/resume issues and
> file a bug and try to reach sean? Or who could help me finding out what the
> modem is thinking?
> 
> 
> the remaining issues not scaring me:
> 	- kernel suspend/resume
> 	- Qtopia/Illume bugs
> 	- Finding out if our touchscreen controller is really that "bad"
> 
> 
> 
> 
> z.
> 


-- 
Carsten Haitzler (The Rasterman) <raster at openmoko.org>