Re: Kernel 2.5.26 panics with Adaptec 2930U

[Mike Anderson <andmike@us.ibm.com>]

Alexandru,
	This error has a signature similiar to what Patrick Mansfield
	reported previously. You can review the previous thread here.
	http://marc.theaimsgroup.com/?t=101494907100009&r=1&w=2

	In the thread a temp workaround was mentioned but not a perm
	solution.

-Mike
Alexandru Deac [andu@shaw.ca] wrote:
> Hello,
> I just compiled kernel 2.5.26 but I can't get it to boot up as it always 
> panics when it gets to the scsi devices.
> I have a Adaptec 2939U SCSI controller and a Sony SDT-9000 DDS-3 tape 
> drive connected to it.
> The aic7xxx is built-in to the kernel along  with the tape support as well as 
> scsi cdrom support and the scsi-emulation driver for my CD-RW.
> 
> Here's the panic output:
> 
> Unable to handle NULL pointer dereference at virtual address 0000000c 
> printing eip:
> c022d7e1
> *pde = 00000000
> Oops: 0000
> CPU: 0
> EIP: 0010:[<c022d7e1>]	Not tainted
> EFLAGS: 00010092
> eax: 00000000	ebx: dfd8d380	ecx: 00000018	edx: 00000000
> esi: dfd4e600	edi: 00000000	ebp: 00000013	esp: dffc3e70
> ds: 0018   es: 0018   ss: 0018
> 
> Process swapper(pid: 1, threadinfo=dffc2000 task=dffc6680)
> Stack: dfd8d380 04000001 00000000 c14d5800 00000292 c08093f1 00000013 dfd4e600
>        dffc3ed0 c03e8040 c03aacc0 00000013 dffc3ec8 c01095e6 00000013 dffc3ed0
>        dfd8d380 dffc2000 00000002 00001c80 00000000 dfd8d380 ff000000 c0107893
> Call Trace: [<c01093f1>] [<c01095e6>] [<c0107893>] [<c022fcba>] [<c01e48ac>]
> [<c01e3c03>] [<c01e4d67>] [<c01af686>] [<c02324ab>] [<c022c3dc>] [<c0218cb9>]
> 
> [<c01e4b2f>] [<c02183a0>] [<c010507b>] [<c0105718>]
> 
> Code: 8b 40 0c f0 fe 08 0f 88 d0 1e 00 00 8b 9e 88 00 00 00 81 e3
> <0>Kernel panic: Aiee, killing interrup handeler!
> In interrupt handeler - not syncing
> 
> I didn't run a ksymoops on it since I cannot boot 2.5.26 so I can't use its 
> /proc/modules and /proc/ksym. Okay well I did a ksymoops against my current 
> kernel 2.4.18 but using 2.5.26 sys map and its modules. I'm not sure if this 
> has any value since it was made using a wrong version but here's it output:
> 
> >>EIP; c022d7e1 <ahc_linux_isr+21/2e0>   <=====
> 
> >>ebx; dfd8d380 <_end+1f95dac4/20453744>
> >>esi; dfd4e600 <_end+1f91ed44/20453744>
> >>esp; dffc3e70 <_end+1fb945b4/20453744>
> 
> Trace; c01093f1 <handle_IRQ_event+51/80>
> Trace; c01095e6 <do_IRQ+a6/f0>
> Trace; c0107893 <common_interrupt+1f/24>
> Trace; c022fcba <aic7770_linux_probe+6a/150>
> Trace; c01e48ac <bus_for_each_dev+bc/d0>
> Trace; c01e3c03 <driver_attach+13/20>
> Trace; c01e4d67 <driver_register+87/a0>
> Trace; c01af686 <pci_register_driver+36/50>
> Trace; c02324ab <ahc_linux_pci_probe+b/40>
> Trace; c022c3dc <ahc_linux_detect+2c/70>
> Trace; c0218cb9 <scsi_register_host+49/300>
> Trace; c01e4b2f <bus_register+5f/70>
> Trace; c02183a0 <scsi_softirq+0/e0>
> Trace; c010507b <init+2b/170>
> Trace; c0105718 <kernel_thread+28/40>
> 
> Code;  c022d7e1 <ahc_linux_isr+21/2e0>
> 00000000 <_EIP>:
> Code;  c022d7e1 <ahc_linux_isr+21/2e0>   <=====
>    0:   8b 40 0c                  mov    0xc(%eax),%eax   <=====
> Code;  c022d7e4 <ahc_linux_isr+24/2e0>
>    3:   f0 fe 08                  lock decb (%eax)
> Code;  c022d7e7 <ahc_linux_isr+27/2e0>
>    6:   0f 88 d0 1e 00 00         js     1edc <_EIP+0x1edc> c022f6bd 
> <.text.lock.aic7xxx_linux+18/ab>
> Code;  c022d7ed <ahc_linux_isr+2d/2e0>
>    c:   8b 9e 88 00 00 00         mov    0x88(%esi),%ebx
> Code;  c022d7f3 <ahc_linux_isr+33/2e0>
>   12:   81 e3 00 00 00 00         and    $0x0,%ebx
> 
> I don't have much experience submiting kernel bugs as you  have probably 
> noticed from this post, but if you need anything else to trace this thing now 
> let me know and I'll provide it.
> 
> As a side note I just tested 2.5.25 and it gives the same error.
> 
> Cheers,
> Alex.
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Michael Anderson
andmike@us.ibm.com