Oops in 2.5 kernels - SCSI related issues
Hi,
I've been trying to compile 2.5 kernels in my all SCSI machine, however failing miserably. It compiles but gives me a nasty oops upon reboot. Since I've not had such show stopper with any 2.4 kernel, I tried to compile an early 2.5, eg. 2.5.1-pre1 (not avoid 2.5.0 to avoid umount breakage) and it compiled, as one would expect. However, I start to get oopses right from 2.5.1-pre2 onwards. I tried up to 2.5.3-final and the same result is obtained.
Upon boot, I see it oopses when detecting SCSI devices (namely the two hard drives) related to the aic7xxx driver. This doesn't surprise me at all knowing that since 2.5.1-pre2 I get compile warnings precisely with this driver. Here's the stderr from the 'make bzImage' process:
In file included from aic7xxx_linux.c:122:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7xxx_proc.c:42:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7770_linux.c:42:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7xxx_linux_pci.c:42:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7xxx.c:45:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7xxx_93cx6.c:70:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7770.c:45:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host' In file included from aic7xxx_pci.c:47:aic7xxx_osm.h: In function `ahc_done_unlock':
aic7xxx_osm.h:750: warning: unused variable `host'
I'm not sure how can I dump the oops output here but if it is really needed, I'll try to provide you with it. My .config is goes attached. This is a Pentium-Classic 100MHz, 32MB RAM.
I don't know if someone arose this problem before. If so, my apologies. I don't know if this is somehow connected with the oopses but it's just too much coincidence. Sorry if I'm talking nonsense here, but what you see here is well...a newbie :|
Thanks in advance,
// Paulo Andre'
/~\ The ASCII
\ / Ribbon Campaign
X Against HTML
/ \ Email!
Ok, here's the ksymoops output for the oops I told you about in the previous email..
Unable to handle kernel NULL pointer dereference at virtual address 000000040 c01c0263 *pde = 00000000 Oops: 0000 CPU: 0 EIP: 0010:[<c01c0263>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010246
eax: 00000000 ebx: c1f4b000 ecx: c1f8a360 edx: 00000000
esi: c1f4b018 edi: c1f4b134 ebp: c1f8a360 esp: c1091e3c
ds: 0018 es: 0018 ss: 0018
Process swapper (pid: 1, stackpage=c1091000)
Stack: c1091fbc c1091fbc c01c059e c1f4b000 c1091fbc c1091fbc c1091fd0
c1f8a360
c1091e78 ffffffff c0249100 c1091e78 00000000 00000000 c1f4b000
00000212
00000001 00000019 00000000 00000000 c00ba300 c017ad2b c1081000
00000000
Call Trace: [<c01c059e>] [<c017ad2b>] [<c01f10c7>] [<c017b866>]
[<c017f61c>]
[<c017ec02>] [<c01131c6>] [<c011322b>] [<c0113301>] [<c0113523>]
[<c0113474>]
[<c01bb9f7>] [<c0105023>] [<c0107004>]
Code: 8b 50 40 8b 40 3c eb 12 90 8d 75 26 00 f6 41 7a 02 74 07 b8Quoted 1 line
Quoted 1 line
EIP;
c01c0262<scsi_initialize_merge_fn+2a/58> <=====
Trace; c01c059e <scan_scsis+c6/444>
Trace; c017ad2a <scrup+6a/104>
Trace; c01f10c6 <vgacon_cursor+17e/188>
Trace; c017b866 <set_cursor+6e/88>
Trace; c017f61c <poke_blanked_console+60/64>
Trace; c017ec02 <vt_console_print+2d2/2e4>
Trace; c01131c6 <__call_console_drivers+3a/4c>
Trace; c011322a <_call_console_drivers+52/58>
Trace; c0113300 <call_console_drivers+d0/d8>
Trace; c0113522 <release_console_sem+72/78>
Trace; c0113474 <printk+104/110>
Trace; c01bb9f6 <scsi_register_host+1ba/2b0>
Trace; c0105022 <init+6/114>
Trace; c0107004 <kernel_thread+28/38>
Code; c01c0262 <scsi_initialize_merge_fn+2a/58>
00000000 <_EIP>:
Code; c01c0262 <scsi_initialize_merge_fn+2a/58> <=====
0: 8b 50 40 mov 0x40(%eax),%edx <=====
Code; c01c0264 <scsi_initialize_merge_fn+2c/58>
3: 8b 40 3c mov 0x3c(%eax),%eax
Code; c01c0268 <scsi_initialize_merge_fn+30/58>
6: eb 12 jmp 1a <_EIP+0x1a> c01c027c <scsi_initialize_merge_fn+44/58>
Code; c01c026a <scsi_initialize_merge_fn+32/58>
8: 90 nop Code; c01c026a
<scsi_initialize_merge_fn+32/58>
9: 8d 75 26 lea 0x26(%ebp),%esi
Code; c01c026e <scsi_initialize_merge_fn+36/58>
c: 00 f6 add %dh,%dh
Code; c01c0270 <scsi_initialize_merge_fn+38/58>
e: 41 inc %ecx
Code; c01c0270 <scsi_initialize_merge_fn+38/58>
f: 7a 02 jp 13 <_EIP+0x13> c01c0274
<scsi_initialize_merge_fn+3c/58>
Code; c01c0272 <scsi_initialize_merge_fn+3a/58>
11: 74 07 je 1a <_EIP+0x1a> c01c027c
<scsi_initialize_merge_fn+44/58>
Code; c01c0274 <scsi_initialize_merge_fn+3c/58>
13: b8 00 00 00 00 mov $0x0,%eax
<0>Kernel panic: Attempted to kill init!
Hope this helps