Something (obviously) got screwed up with my X windows. I'm running
Mandrake 9.2 with the 2.6.0-test9 kernel. X comes up all right if I
type startx from the command line as superuser. Doing this as a normal
user, freezes the system.
Where can I look? Can somebody help walk me through this, please?

[ Followup-To: set to comp.os.linux.x . That means you should reply to
this message in comp.os.linux.x , manually if necessary, since Google's
NNTP client doesn't seem to grok followups.]

On 1 Dec 2003 09:54:42 -0800, Yitzhak Bar Geva staggered into the Black
Sun and said:
Kernel 2.6.0-test9 is (obviously) not ready for prime time yet. Do not
use it unless you are willing to put up with things breaking. The
advantages it provides over 2.4.22 are, AFAICT, minimal on an Athlon XP
2000 (1G PC2100, parallel IDE, nVidia TNT2)--KDE 3.1 takes almost 5
seconds longer to start up when I'm running 2.6.0-test9. Also, the VESA
framebuffer is br0ken and won't be fixed until 2.6.1 or so, nVidia's
binary-only kernel module requires some fiddling to work properly, and
Lucent's LoseModem module won't work at all (so I can't really test it
on my lower-spec laptop.) It'll be ready for prime time around 2.6.3 or
2.6.5, if it works like the 2.4 kernels did.

/var/log/XFree86*.log is the first place to look. You should also read
http://catb.org/~esr/faqs/smart-questions.html , then post your
/etc/X11/XF86Config file on your webspace and provide a URL. Did you
turn on anything in the "Security options" section of menuconfig? Like
the documentation says, that can break X. What's the make and model of
the video card you're using? Are you trying to use DRI? Are there any
strange messages in /var/log/warn after you try to use X as a user?

--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / mail: TRAP + SPAN don't belong
http://www.brainbench.com / "He is a rhythmic movement of the
-----------------------------/ penguins, is Tux." --MegaHAL

Dances With Crows <danSPANceswitTRAPhcrows@usa.net> wrote in message news:<slrnbsn23n.ipv.danSPANceswitTRAPhcrows@saman tha.crow202.dyndns.org>...
You have definitely given me a lot of good advice, thanks.
Nevertheless, I did have this working at some point, but I haven't
managed to roll it back. I have applied the necessary patches to use
the nvidia driver, which BTW is universal for all their cards.
Following the phenomenon (crash?) I described, /var/log/XFree86.0.log
is corrupt.
After the reboot, it's tail looks as follows:
(II) NVIDIA(0): AGP 8X successfully initialized
(II) NVIDIA(0): Setting mode "1024x768"
(II) NVIDIA(0): Using the NVIDIA 2D acceleration architecture
(==) NVIDIA(0): Backing store disabled
(==) NVIDIA(0): Silken mouse enabled
(**) Option "dpms"
(**) NVIDIA(0): DPMS enabled
(II) Loading extension NV-GLX
(II) Loading extension NV-CONTROL
(==) RandR enabled
(II) Initializing built-in extension MIT-SHM
(II) Initializing built-in extension XInputExtension
(II) Initializing built-in extension XTEST
(II) Initializing built-in extension XKEYBOARD
(II) Initializing built-in extension LBX
(II) Initializing built-in extension XC-APPGROUP
(II) Initializing built-in extension SECURITY
(II) Initializing built-in extension XINERAMA
(II) Initializing built-in extension XFree86-Bigfont
(II) Initializing built-in extension RENDER
(II) Initializing built-in extension RANDR
(II) [GLX]: Initializing GLX extension
(II) Keyboard "Keyboard1" handled by legacy driver
(**) Option "Protocol" "IMPS/2"
(**) Mouse1: Protocol: "IMPS/2"
(**) Option "CorePointer"
(**) Mouse1: Core Pointer
(**) Option "Device" "/dev/psaux"
(**) Mouse1: Emulate3Buttons, Emulate3Timeout: 50
(**) Option "ZAxisMapping" "4 5"
(**) Mouse1: ZAxisMapping: buttons 4 and 5
(**) Mouse1: Buttons: 5
(II) XINPUT: Adding extended input device "Mouse1" (type: MOUSE)
(II) XINPUT: Adding extended input device "NVIDIA Event Handler"
(type: Other)
(II) Mouse1: ps2EnableDataReporting: succeeded
AUDIT: Tue Dec 2 02:12:53 2003: 1374 X: client 4 rejected from local
host
AUDIT: Tue Dec 2 03:05:18 2003: 1374 X: client 19 rejected from local
host

Notice the last two lines. Is that a security problem?

/var/kernel/warnings does have some problem info:
Dec 2 02:12:46 localhost kernel: in_atomic():1, irqs_disabled():0
Dec 2 02:12:46 localhost kernel: Call Trace:
Dec 2 02:12:46 localhost kernel: [__might_sleep+169/207]
__might_sleep+0xa9/0xcf
Dec 2 02:12:46 localhost kernel: [<c011c0e6>]
__might_sleep+0xa9/0xcf
Dec 2 02:12:46 localhost kernel: [s_show+346/611]
kmem_cache_alloc+0x5b/0x6b
Dec 2 02:12:46 localhost kernel: [<c013d97e>]
kmem_cache_alloc+0x5b/0x6b
Dec 2 02:12:46 localhost kernel: [add_to_swap+23/232]
__get_vm_area+0x21/0xfa
Dec 2 02:12:46 localhost kernel: [<c014bf62>]
__get_vm_area+0x21/0xfa
Dec 2 02:12:46 localhost kernel: [delete_from_swap_cache+59/184]
get_vm_area+0x33/0x35
Dec 2 02:12:46 localhost kernel: [<c014c06e>] get_vm_area+0x33/0x35
Dec 2 02:12:46 localhost kernel: [__ioremap+181/263]
__ioremap+0xb5/0x107
Dec 2 02:12:46 localhost kernel: [<c01195a5>] __ioremap+0xb5/0x107
Dec 2 02:12:46 localhost kernel: [ioremap_nocache+41/192]
ioremap_nocache+0x29/0xc0
Dec 2 02:12:46 localhost kernel: [<c0119620>]
ioremap_nocache+0x29/0xc0
Dec 2 02:12:46 localhost kernel: [_end+547028621/1068566168]
os_map_kernel_space+0x2f/0x68 [nvidia]
Dec 2 02:12:46 localhost kernel: [<e0e9d7f5>]
os_map_kernel_space+0x2f/0x68 [nvidia]
Dec 2 02:12:46 localhost kernel: [_end+545802767/1068566168]
__nvsym00568+0x1f/0x2c [nvidia]
Dec 2 02:12:46 localhost kernel: [<e0d72377>] __nvsym00568+0x1f/0x2c
[nvidia]
Dec 2 02:12:46 localhost kernel: [_end+545811246/1068566168]
__nvsym00775+0x6e/0xe0 [nvidia]
Dec 2 02:12:46 localhost kernel: [<e0d74496>] __nvsym00775+0x6e/0xe0
[nvidia]
Dec 2 02:12:46 localhost kernel: [_end+545811390/1068566168]
__nvsym00781+0x1e/0x190 [nvidia]
Dec 2 02:12:46 localhost kernel: [<e0d74526>]
__nvsym00781+0x1e/0x190 [nvidia]
Dec 2 02:12:46 localhost kernel: [_end+545818180/1068566168]
rm_init_adapter+0xc/0x10 [nvidia]
Dec 2 02:12:46 localhost kernel: [<e0d75fac>]
rm_init_adapter+0xc/0x10 [nvidia]
Dec 2 02:12:46 localhost kernel: [_end+547015179/1068566168]
nv_kern_open+0xf4/0x256 [nvidia]
Dec 2 02:12:46 localhost kernel: [<e0e9a373>]
nv_kern_open+0xf4/0x256 [nvidia]
Dec 2 02:12:46 localhost kernel: [inode_sub_bytes+22/162]
chrdev_open+0x16a/0x230
Dec 2 02:12:46 localhost kernel: [<c015a910>]
chrdev_open+0x16a/0x230
Dec 2 02:12:46 localhost kernel: [sys_llseek+183/254]
dentry_open+0x13a/0x208
Dec 2 02:12:46 localhost kernel: [<c0150a6c>]
dentry_open+0x13a/0x208
Dec 2 02:12:46 localhost kernel: [sys_lseek+41/178]
filp_open+0x5a/0x60
Dec 2 02:12:46 localhost kernel: [<c015092c>] filp_open+0x5a/0x60
Dec 2 02:12:46 localhost kernel: [vfs_write+90/270]
sys_open+0x55/0x92
Dec 2 02:12:46 localhost kernel: [<c0150dad>] sys_open+0x55/0x92
Dec 2 02:12:46 localhost kernel: [syscall_call+7/11]
syscall_call+0x7/0xb
Dec 2 02:12:46 localhost kernel: [<c010b287>] syscall_call+0x7/0xb
Dec 2 02:12:46 localhost kernel:

The result of the hang appears in /var/kernel/errors:
Dec 1 00:46:55 localhost kernel: Debug: sleeping function called from
invalid context at include/linux/rwsem.h:43
Dec 1 00:46:55 localhost kernel: bad: scheduling while atomic!
Dec 1 00:49:54 localhost kernel: Debug: sleeping function called from
invalid context at mm/slab.c:1856
Dec 1 00:49:55 localhost kernel: Debug: sleeping function called from
invalid context at mm/slab.c:1856
Dec 1 10:56:58 localhost kernel: Debug: sleeping function called from
invalid context at mm/slab.c:1856
Dec 1 19:29:55 localhost last message repeated 2 times
Dec 1 23:56:29 localhost last message repeated 2 times
Dec 1 23:59:03 localhost last message repeated 2 times

I agree with you to an extent: I have neither time nor interest in
kernel debugging. However, when run as superuser, it's OK and it
already did run before as a normal user using 2.6.0-test9.
Could it be a security problem?:
See this:
#startx
...... X comes up OK. I use Gnome...
root># su - yitzhakbg

yitzhakbg>$ gedit
Xlib: connection to ":0.0" refused by server
Xlib: No protocol specified


(gedit:2051): Gtk-WARNING **: cannot open display:

At this point, I could use some assistance in tracking it down.