Bug#278724: Info received (was Bug#278724: [nv] system hang when using gv, xmms, or lopster on NV5 [Aladdin TNT2] rev 0x20)
Hi!
in my last mesage relating to the captioned bug I wrote (Nov. 24, 2004):
----------
I'm glad to inform you that changing the Section "Screen" DefaultDepth
from 24 to 16 the problem seems to have disappeared: at least I could
run my famous test (i.e. gv foo.txt) many times without locking the
system :-)
Although I think you could reasonably close the bug, one question
remains: I had not set that value to 24 and, if I recall well, when I
configured X in old Woody and tried to run it with the wrong value, some
'goblin' would come out from within the X Window System and tell me
that depth 24 was not supported by my video card!
----------
Despite that I did experiment many other hanging-up, but still I had no
clue as to what was happening, so I didn't report them.
To-day, after my PC froze again, I took different steps which may
possibly throw some light on the subject.
I had all six consoles opened as follows:
F1 - root (idle)
F2 - user (idle)
F3 - user (idle)
F6 - user -->start x --> running lopster (on F7)
F5 - user -->startx -- :1 --> running speedy (on F8)
then I opened
F4 - user -->startx -- :2 --> running mplayer -vo xv -autosync 1 file.mpg (on F9)
after half minute watching on F9, the screen (and the PC) froze.
I ssh-ed in with my laptop and run top, which showed an XFree86
abosrbing from 60 to 94,6% of CPU!
Instead of rebooting (as in past occasions) I killed the above high
absorbing XFree86 process and nothing changed; killed .xinitrc running
on tty4 (running mplayer on F9, the screen the PC had frozen on) and
nothing happened; killed .xinitrc on tty5 (running speedy on F8) and saw
a black line across my F9 screen: something had happened!
While trying all the CTRL-ALT+F1 through F6 I could see part of the F9
screen progressively full of garbage blacking-out the window ...
finally, CTRL-ALT+F7 showed me the window where lopster was running and
after that I could re-gain control of my PC!
While still having the laptop connected via ssh I tried to reproduce the
hanging-up doing again the same steps and watching from top running on
the laptop what went on on the PC: the speedy-preloader took almost 45%
CPU and about 49% was taken by XFree86 when I launched mplayer ...;
after a while (i.e. whan F9 window froze), XFree86 was absorbing up to
96,5% CPU.
What I can conclude from my un-technical point of view is that it's the
CPU's lack of resources to cause the freezing and wonder whether there
could be a sort of pre-warning as to avoid it.
Thanks for your attention and for your wonderful work.
Regards,
Ennio
--
[Perche' usare Win$ozz (dico io) se ..."anche uno sciocco sa farlo. \\?//
Fa' qualche cosa di cui non sei capace!" (diceva Henry Miller) ] (°|°)
[Why use Win$ozz (I say) if ... "even a fool can do that. )=(
Do something you aren't good at!" (as Henry Miller used to say) ]
Reply to: