2.5.38: oops with kernel LLC support enabled

2.5.38: oops with kernel LLC support enabled

[Bob_Tracy]

Firstly, a big thanks to Ingo for kksymoops...  2.5.38 panics right
out of the chute with no way to save the oops output, so it's REALLY
nice to have it decoded on the screen!!  Manual transcription of the
oops output follows below.

What was I trying to do with LLC support?  I have an old JetDirect
interface in a LJ-III that speaks DLC/LLC only, and wanted to try to
use it with Linux.  (It should go without saying that NT and Win2K
can talk to it just fine :-)).  Both machines in a sample set of two
panic the same way...  One is a desktop with the NIC driver built-in.
The other (that produced the oops output below) is a Dell Latitude CPxJ
with a modular Xircom Tulip cardbus driver.

I guess I should mention that I haven't tried this with a 2.4 kernel.

 Unable to handle kernel NULL pointer dereference at virtual address 00000000
  printing eip:
 c0245da5
 *pde = 00000000
 Oops: 0000
 kernel
 CPU:    0
 EIP:    0060:[<c0245da5>]	Not tainted
 EFLAGS: 00010202
 EIP is at llc_sap_find [kernel] 0x25
 eax: c7fae000   ebx: c02e8e28   ecx: 000000aa   edx: 00000000
 esi: 00000000   edi: 00000000   ebp: 00000000   esp: c7faffa8
 ds: 0068   es: 0068   ss: 0068
 Process swapper (pid: 1, threadinfo=c7fae000 task=c7fac040)
 Stack: c02e8e28 000000aa 00000000 c0242141 000000aa c02e8e28 00000000 c02df8e1
        c01f0ca0 00000000 000000aa c02d0712 c7fae000 c0105093 c0105060 00000000
        00000000 00000000 c01054e5 00000000 00000000 00000000
 Call Trace: [<c0242141>] llc_sap_open [kernel] 0x11
 [<c01f0ca0>] snap_indicate [kernel] 0x0
 [<c0105093>] init [kernel] 0x33
 [<c0105060>] init [kernel] 0x0
 [<c01054e5>] kernel_thread_helper [kernel] 0x5
 
 
 Code: 8b 02 0f 18 00 81 fa 04 69 33 c0 74 08 38 4a f8 8d 72 a0 75
  <0>Kernel panic: Aiee, killing interrupt handler!
 In interrupt handler - not syncing
 
 
-- 
-----------------------------------------------------------------------
Bob Tracy                   WTO + WIPO = DMCA? http://www.anti-dmca.org
rct@frus.com
-----------------------------------------------------------------------

Re: 2.5.38: oops with kernel LLC support enabled

[Arnaldo Carvalho de Melo]

Em Wed, Sep 25, 2002 at 10:49:20AM -0500, Bob_Tracy escreveu:
> Firstly, a big thanks to Ingo for kksymoops...  2.5.38 panics right
> out of the chute with no way to save the oops output, so it's REALLY
> nice to have it decoded on the screen!!  Manual transcription of the
> oops output follows below.

yes, its a great help
 
> What was I trying to do with LLC support?  I have an old JetDirect

Hey, no need to excuses, its there, test and use it, using old protocols
is not something to be ashamed, neither hacking it... 8)

> interface in a LJ-III that speaks DLC/LLC only, and wanted to try to
> use it with Linux.  (It should go without saying that NT and Win2K
> can talk to it just fine :-)).  Both machines in a sample set of two
> panic the same way...  One is a desktop with the NIC driver built-in.
> The other (that produced the oops output below) is a Dell Latitude CPxJ
> with a modular Xircom Tulip cardbus driver.
> 
> I guess I should mention that I haven't tried this with a 2.4 kernel.

you can try, but the codebase there (and I assume you're talking about
the LLC stack in the 2.4-ac tree, as the one in the 2.4 stock is
incomplete) is old and I'm only now getting to a point where LLC in
2.5 is getting pretty stable (now as in today, 4am :) )

>  Unable to handle kernel NULL pointer dereference at virtual address 00000000
>   printing eip:
>  c0245da5
>  *pde = 00000000
>  Oops: 0000
>  kernel
>  CPU:    0
>  EIP:    0060:[<c0245da5>]	Not tainted
>  EFLAGS: 00010202
>  EIP is at llc_sap_find [kernel] 0x25
>  eax: c7fae000   ebx: c02e8e28   ecx: 000000aa   edx: 00000000
>  esi: 00000000   edi: 00000000   ebp: 00000000   esp: c7faffa8
>  ds: 0068   es: 0068   ss: 0068
>  Process swapper (pid: 1, threadinfo=c7fae000 task=c7fac040)
>  Stack: c02e8e28 000000aa 00000000 c0242141 000000aa c02e8e28 00000000 c02df8e1
>         c01f0ca0 00000000 000000aa c02d0712 c7fae000 c0105093 c0105060 00000000
>         00000000 00000000 c01054e5 00000000 00000000 00000000
>  Call Trace: [<c0242141>] llc_sap_open [kernel] 0x11
>  [<c01f0ca0>] snap_indicate [kernel] 0x0
>  [<c0105093>] init [kernel] 0x33
>  [<c0105060>] init [kernel] 0x0
>  [<c01054e5>] kernel_thread_helper [kernel] 0x5

Could you try with the next version Linus releases? I some fixes already
commited into Linus bk tree, some still in DaveM's tree and some waiting
for DaveM to pull that, I think, will solve your problem, and wow, don't
give up, at least I'd have one more person testing my work 8)

Just let me know if you'd like to get a diff from the latest Linus
released kernel to the latest stuff I have, I'll be glad to provide and
to receive test results.

Best Regards,

- Arnaldo