Re: test13-pre6 -- OOPS at boot time in kmem_cache_create (usb-ohcirelated)

From: David Brownell (david-b@pacbell.net)
Date: Sun Dec 31 2000 - 09:50:44 EST


This BUG() probably means you need to enable DEBUG in mm/slab.c
before you try have OHCI use slab poisoning.

Enable both, or neither, and all should be fine.

- Dave

----- Original Message -----
From: Miles Lane <miles@megapathdsl.net>
To: Linux Kernel <linux-kernel@vger.kernel.org>; David Brownell <david-b@pacbell.net>
Sent: Saturday, December 30, 2000 11:03 PM
Subject: test13-pre6 -- OOPS at boot time in kmem_cache_create (usb-ohcirelated)

> This OOPS occured at boot time when I did not have my Belkin
> BusPort Mobile USB host-controller inserted. /etc/usb/rc.usb
> tried loading usb-ohci. After the boot process completed
> (after the OOPS occured), I was unable to unload the module
> and lsmod showed that usb-ohci was still initializing.
> Presumably, it got wedged when the OOPS occured?
>
> ksymoops 2.3.5 on i686 2.4.0-test13-pre6. Options used
>
> Kernel command line: BOOT_IMAGE=Serial-Debug ro root=305 pci=biosirq
> console=ttyS0,38400 console=tty0 setup_delay=10
> kernel BUG at slab.c:660!
> invalid operand: 0000
> CPU: 0
> EIP: 0010:[<c01284bc>]
> Using defaults from ksymoops -t elf32-i386 -a i386
> EFLAGS: 00010292
> eax: 0000001a ebx: 00002800 ecx: 00000001 edx: 00000001
> esi: 00002800 edi: c5824929 ebp: c58251bc esp: c4af1eb4
> ds: 0018 es: 0018 ss: 0018
> Process modprobe (pid: 84, stackpage=c4af1000)
> Stack: c01ee425 c01ee4a5 00000294 00002800 c58247d0 00000001 c58251bc
> 000041f0
> c021c900 000038d8 000038db 00000286 00000203 00000000 c58251b4
> c5821079
> c5824921 00000020 00000000 00002800 00000000 00000000 c5821000
> c58247d6
> Call Trace: [<c01ee425>] [<c01ee4a5>] [<c58247d0>] [<c58251bc>]
> [<c58251b4>] [<c5821079>] [<c5824921>]
> [<c5821000>] [<c58247d6>] [<c5821000>] [<c0116b78>] [<c5813000>]
> [<c5813000>] [<c58251a8>] [<c5813000>]
> [<c5821060>] [<c0108d57>]
> Code: 0f 0b 83 c4 0c 6a 07 68 80 c6 21 c0 e8 03 07 00 00 89 c5 83
>
> >>EIP; c01284bc <kmem_cache_create+94/3b8> <=====
> Trace; c01ee425 <tvecs+195d/abb8>
> Trace; c01ee4a5 <tvecs+19dd/abb8>
> Trace; c58247d0 <[usb-ohci]init_module+0/0>
> Trace; c58251bc <[usb-ohci].bss.end+15/39>
> Trace; c58251b4 <[usb-ohci].bss.end+d/39>
> Trace; c5821079 <[usb-ohci]ohci_mem_init+19/78>
> Trace; c5824921 <[usb-ohci].rodata.start+101/76c>
> Trace; c5821000 <[usbcore]__kstrtab_usb_devfs_handle+1864/18c4>
> Trace; c58247d6 <[usb-ohci]ohci_hcd_init+6/3c>
> Trace; c5821000 <[usbcore]__kstrtab_usb_devfs_handle+1864/18c4>
> Trace; c0116b78 <sys_init_module+58c/630>
> Trace; c5813000 <_end+557b9b0/557ba10>
> Trace; c5813000 <_end+557b9b0/557ba10>
> Trace; c58251a8 <[usb-ohci].bss.end+1/39>
> Trace; c5813000 <_end+557b9b0/557ba10>
> Trace; c5821060 <[usb-ohci]ohci_mem_init+0/78>
> Trace; c0108d57 <system_call+33/38>
> Code; c01284bc <kmem_cache_create+94/3b8>
> 00000000 <_EIP>:
> Code; c01284bc <kmem_cache_create+94/3b8> <=====
> 0: 0f 0b ud2a <=====
> Code; c01284be <kmem_cache_create+96/3b8>
> 2: 83 c4 0c add $0xc,%esp
> Code; c01284c1 <kmem_cache_create+99/3b8>
> 5: 6a 07 push $0x7
> Code; c01284c3 <kmem_cache_create+9b/3b8>
> 7: 68 80 c6 21 c0 push $0xc021c680
> Code; c01284c8 <kmem_cache_create+a0/3b8>
> c: e8 03 07 00 00 call 714 <_EIP+0x714> c0128bd0
> <kmem_cache_alloc+0/58>
> Code; c01284cd <kmem_cache_create+a5/3b8>
> 11: 89 c5 mov %eax,%ebp
> Code; c01284cf <kmem_cache_create+a7/3b8>
> 13: 83 00 00 addl $0x0,(%eax)
>
>
> 1 warning issued. Results may not be reliable.
>

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



This archive was generated by hypermail 2b29 : Sun Dec 31 2000 - 21:00:15 EST