Weird Xorg behaviour under recent -current amd64

[Chavdar Ivanov <ci4ic4%gmail.com@localhost>]

Hi,

Trying to start Enlightenment (which used to work fine a few weeks ago
on the same system), now I get:

(random selections from top follows):

load averages:  8.21,  3.03,  1.31;               up 0+00:25:59
                                   12:17:55
75 processes: 11 runnable, 62 sleeping, 2 on CPU
CPU states:  0.0% user,  0.0% nice, 50.2% system, 49.8% interrupt,  0.0% idle
Memory: 717M Act, 2440K Inact, 104K Wired, 73M Exec, 106M File, 342M Free
Swap: 10G Total, 10G Free

  PID USERNAME PRI NICE   SIZE   RES STATE      TIME   WCPU    CPU COMMAND
 7258 root      26   -1   452M   37M CPU/1      3:24   215%   215% Xorg
 4072 root      78    0    38M 3716K RUN/0      0:43 54.00% 54.00% snmpd
 2440 xci       28    0    17M 1916K CPU/0      1:12 31.98% 31.98% [system]
 7252 root      77    0    32M 5288K ttyraw/1   0:18 11.92% 11.91% zsh
 5366 xci       79    0    75M 4572K select/0   0:06 11.25% 10.50% sshd


oad averages:  11.4,  4.51,  1.93;               up 0+00:27:13
                                  12:19:09
75 processes: 3 runnable, 70 sleeping, 2 on CPU
CPU states:  0.0% user,  0.0% nice, 68.0% system, 23.6% interrupt,  8.4% idle
Memory: 718M Act, 2440K Inact, 104K Wired, 73M Exec, 106M File, 342M Free
Swap: 10G Total, 10G Free

  PID USERNAME PRI NICE   SIZE   RES STATE      TIME   WCPU    CPU COMMAND
 7258 root      27   -1   452M   37M RUN/1      4:36   179%   179% Xorg
 4072 root      77    0    38M 3716K select/1   0:52 30.57% 30.57% snmpd
 2440 xci       36    0    17M 1916K CPU/0      0:21 20.74% 20.12% top
    0 root     221    0     0K   37M raidio/0   1:24 19.87% 19.87% [system]
 5309 postfix   78    0    48M 4192K kqueue/0   0:04  8.98%  8.98% pickup
 5366 xci       81    0    75M 4572K select/1   0:13  8.58%  8.50% sshd
 3381 root      82    0    39M 2660K select/1   0:05  8.30%  8.30% snmptrapd
  192 root      78    0    59M 6212K kqueue/1   0:03  5.66%  5.66% cupsd
 7837 pgsql     82    0    64M 4652K RUN/1      0:01  3.71%  3.71% postgres
 8298 pgsql     82    0    27M 3364K select/1   0:06  3.37%  3.37% postgres


(Xorg CPU load shoots up to 240%, whatever that means on a 2 CPU
system...). Remote ssh session to the host stopped responding, I came
back, found that the host is pingable, tried to ssh again, which
eventually succeeded).

This is unmodified GENERIC from overnight update; the card is a FireGL
f2. If I just run startx with twm, it seems to work fine (but I cannot
use the window manager menus, they disappear as soon as I move the
mouse to point the selection). KDE4 appears to be working fine, but if
I run glxgears under kde4, it hard-resets the system (whilst working
fine under twm).

DRMKMS on the same hardware boots and generally works fine, I can even
start Xorg (without anything else - I get a cursor in the middle and a
white square in the top left corner, but at least it doesn't reboot).
It hard-resets if I try startx with twm, though.

Chavdar

(I've got other issues with a recent setup of the same system, I'll
ask about them later).



-- 
----