Skip to contentSkip to conversation

new OOPS 2.4.10-pre11, do_generic_file_read [devfs related?]

DISCUSSION Giacomo Catenazzi 2 participants 2 messages Sep 18, 2001

lore
Share & tools
lore
#0 lore raw

Starting in 2.4.10-pre11 I have a new oops.
This time the oops happens also without floppy support, but also this bug happen at boot time, when mounting root rw. Also this time the oops seem reproducible. devfs=nomount solve this bug (like the old bug).

What the status of your devfs rewrite ?

giacomo

Receiver lock-up bug exists -- enabling work-around.

Unable to handle kernel NULL pointer dereference at virtual address
00000000
00000000
*pde = 00000000
Oops: 0000
CPU:    0
EIP:    0010:[<00000000>]

Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010202
eax: c0262e00 ebx: cff61e34 ecx: cfb310c0 edx: cff61e34
esi: c13a9034 edi: cfab6cc0 ebp: 00000000 esp: cfa11ed8
ds: 0018 es: 0018 ss: 0018
Process fsck.ext2 (pid: 37, stackpage=cfa11000)

Stack: c0123064 cfb310c0 c13a9034 00000c00 00000001 00000000 00000400
cfab6c20
         00000000 cfe9a420 cfbaeea0 cfa10000 c010fedc cfe9a420 cfbaeea0
080640d4
         cfb310c0 ffffffea 00000000 00000400 c01236af cfb310c0 cfb310e0
cfa11f4c
Call Trace: [<c0123064>] [<c010fedc>] [<c01236af>] [<c01234b0>]
[<c012f410>]
     [<c0136490>] [<c012f26e>] [<c0106d3b>]
Code:  Bad EIP value.
Quoted 1 line
Quoted 1 line

EIP; 00000000 Before first symbol

Trace; c0123064 <do_generic_file_read+2b4/4f0> Trace; c010fedc <do_page_fault+2fc/4e0>
Trace; c01236ae <generic_file_read+19e/1c0> Trace; c01234b0 <file_read_actor+0/60>
Trace; c012f410 <sys_read+b0/d0>
Trace; c0136490 <block_llseek+0/a0>
Trace; c012f26e <sys_lseek+6e/80>
Trace; c0106d3a <system_call+32/38>

#1 lore raw
Quoted 4 lines · Giacomo Catenazzi

Starting in 2.4.10-pre11 I have a new oops.
This time the oops happens also without floppy support, but also this bug happen at boot time, when mounting root rw. Also this time the oops seem reproducible.

Wow! I should try 2.4.10-pre11. I have't had a chance yet.

Quoted 3 lines

devfs=nomount solve this bug (like the old bug).

What the status of your devfs rewrite ?

Only small progress: I spent last week in Boston setting up some test boxes, which was slower than expected thanks to battling with the non-standardness of Compaq boxen. Grrr.

				Regards,

					Richard....
Permanent: [email protected]
Current:   [email protected]