aic7xxx still flaky

Martin Maciaszek (fastjack@i-s-o.net)
Sun, 17 Jan 1999 19:00:12 +0100

The problem occurs since 2.1.131 or even earlier (I just tested 2.1.13[12], 2.2.0-pre7, 2.2.0-pre7-ac[245]) I'm using an Adaptec AHA-2940UW (BIOS v1.21) on a Gigabyte GA-586ATE (Triton) board. I built aic7xxx, sd_mod, sr_mod, sg and st as modules and loaded them at boot time from a initrd. After some minutes of scsi-activity (e2fsck or dd) I get an oops. It happened to me the first time on a forces fsck (maximal mount count). I can reproduce this oops easily with "dd if=/dev/scd2 of=/dev/null"

Here is the output of ksymoops:
Jan 17 17:10:42 nexus kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000400
Jan 17 17:10:42 nexus kernel: current->tss.cr3 = 0310b000, `r3 = 0310b000
Jan 17 17:10:42 nexus kernel: *pde = 00000000
Jan 17 17:10:42 nexus kernel: Oops: 0000
Jan 17 17:10:42 nexus kernel: CPU: 0
Jan 17 17:10:42 nexus kernel: EIP: 0010:[<c012483e>]
Jan 17 17:10:42 nexus kernel: EFLAGS: 00010206
Jan 17 17:10:42 nexus kernel: eax: 00000400 ebx: 000144fa ecx: 00000b02 edx: 00000400
Jan 17 17:10:42 nexus kernel: esi: 00000800 edi: 00000800 ebp: 000144fa esp: c311fcec
Jan 17 17:10:42 nexus kernel: ds: 0018 es: 0018 ss: 0018
Jan 17 17:10:42 nexus kernel: Process dd (pid: 601, process nr: 23, stackpage=c311f000)
Jan 17 17:10:42 nexus kernel: Stack: c0124877 00000b02 000144fa 00000800 00000001 00000800 000144fa c3110b02
Jan 17 17:10:42 nexus kernel: c0124b0b 00000b02 000144fa 00000800 00000001 00000000 c311fe90 c311fe94
Jan 17 17:10:42 nexus kernel: 00000002 c0127cd1 00000b02 000144fa 00000800 c30fc080 0804c3da 0804c3da
Jan 17 17:10:42 nexus kernel: Call Trace: [<c0124877>] [<c0124b0b>] [<c0127cd1>] [<c4810d34>] [<c4810d6b>] [<c4810002>] [<c4810cf3>]
Jan 17 17:10:42 nexus kernel: [<c4810d6b>] [<c4822b70>] [<c017ecf9>] [<c481017a>] [<c017ecf9>] [<c481017a>] [<c481084d>] [<c4810865>]
Jan 17 17:10:42 nexus kernel: [<c011053f>] [<c011053f>] [<c010f713>] [<c011053f>] [<c011060a>] [<c012341a>] [<c01089e0>]
Jan 17 17:10:42 nexus kernel: Code: 8b 12 39 58 04 75 f3 39 70 08 75 ee 66 39 48 0c 75 e8 89 c2

>>EIP: c012483e <find_buffer+2a/44>
Trace: c0124877 <get_hash_table+1f/60>
Trace: c0124b0b <getblk+1f/224>
Trace: c0127cd1 <block_read+30d/524>
Trace: c4810d34 <do_sd_request+1b8/1c4>
Trace: c4810d6b <requeue_sd_request+2b/f00>
Trace: c4810002 <scsi_mlqueue_remove_lock+1012/105c>
Trace: c4810cf3 <do_sd_request+177/1c4>
Trace: c4810d6b <requeue_sd_request+2b/f00>
Trace: c011053f <update_process_times+5b/64>
Code: c012483e <find_buffer+2a/44> 00000000 <_EIP>: <===
Code: c012483e <find_buffer+2a/44> 0: 8b 12 movl (%edx),%edx <===
Code: c0124840 <find_buffer+2c/44> 2: 39 58 04 cmpl %ebx,0x4(%eax)
Code: c0124843 <find_buffer+2f/44> 5: 75 f3 jne c0124838 <find_buffer+24/44>
Code: c0124845 <find_buffer+31/44> 7: 39 70 08 cmpl %esi,0x8(%eax)
Code: c0124848 <find_buffer+34/44> a: 75 ee jne c0124838 <find_buffer+24/44>
Code: c012484a <find_buffer+36/44> c: 66 39 48 0c cmpw %cx,0xc(%eax)
Code: c012484e <find_buffer+3a/44> 10: 75 e8 jne c0124838 <find_buffer+24/44>
Code: c0124850 <find_buffer+3c/44> 12: 89 c2 movl %eax,%edx

Didn't anyone else encounter similar problems? My programming skills aren enought to fix this problem myself.

Regards
Martin

PS: I you guys need some extra information on my system configuration just tell me what you need.

-- 
This dungeon is owned and operated by Frobozz Magic Co., Ltd.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/