Kernel oops on XXS1500 in au1000eth.c

Kernel oops on XXS1500 in au1000eth.c

[Goswin von Brederlow]

Hi,

when compiling a kernel for my XXS1500 I allways ended up with a
kernel oops in the network driver (au1000eth.c).

Finaly I checked the actual kernel source the running kernel was build
from and found "CONFIG_BCM5222_DUAL_PHY" was set. Setting that solves
the oops.

Maybe that could be caught in some way and prevented.

MfG
        Goswin

----------------------------------------------------------------------
Start = 0x80274040, range = (0x80100000,0x802bbfff), format = SREC

Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
pty: 256 Unix98 ptys configured
Serial driver version 1.01 (2001-02-08) with no serial options enabled
ttyS00 at 0xb1100000 (irq = 0) is a 16550
ttyS01 at 0xb1200000 (irq = 1) is a 16550
ttyS02 at 0xb1300000 (irq = 2) is a 16550
ttyS03 at 0xb1400000 (irq = 3) is a 16550
Generic MIPS RTC Driver v1.0
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
loop: loaded (max 8 devices)
au1000eth.c:1.4 ppopov@mvista.com
eth0: Au1x Ethernet found at 0xb1500000, irq 28
eth0: Broadcom BCM5222 10/100 BaseT PHY at phy address 0
eth0: Using Broadcom BCM5222 10/100 BaseT PHY as default
eth1: Au1x Ethernet found at 0xb1510000, irq 29
Unable to handle kernel paging request at virtual address 00000000, epc == 801c0
Oops in fault.c::do_page_fault, line 206:
$0 : 00000000 1000fc00 00000000 001e3000 802597c8 0000001f 00000001 00000013
$8 : 810cc800 b1510018 000011e0 802a1434 00000004 ba2e8ba3 1000fc01 00000002
$16: 8029f940 802db12c 00000001 00000020 810cc800 810cc9e4 810cc960 802b4cf4
$24: ffffffff 00000001                   802e4000 802e5ed8 0000ffff 801c5b60
Hi : 000304cc
Lo : ecaf8000
epc   : 801c5c10    Not tainted
Status: 1000fc03
epc   : 00800008
PrId  : 01030200
Process swapper (pid: 1, stackpage=802e4000)
Stack:    b1510000 801196ac 802598b8 810cc800 8029f940 802db12c 00000001
 810cc960 810cc800 810cc9e4 810cc9f4 b1510000 0000001d 801c63a0 80259890
 810cc800 b1510000 0000001d 87000266 00001123 00000001 802db12c 00000001
 04000000 00000000 00000000 00000000 00000000 8008aa54 80287dac 80287c80
 80287c6c 00000000 00000000 8028e530 8028e55c 00010f00 802746ec 80122ed8
 8028100c ...
Call Trace:   [<801196ac>] [<802598b8>] [<801c63a0>] [<80259890>] [<80122ed8>]
 [<8025b9cc>] [<801007c4>] [<801007d4>] [<801007c4>] [<801047d4>] [<80111e94>]
 [<8016125c>] [<80161220>] [<801047c4>]

Code: 8c420004  3c048026  248497c8 <8c460000> 0c044dc1  02802821  0807171f  000
Kernel panic: Attempted to kill init!

Re: Kernel oops on XXS1500 in au1000eth.c

[Pete Popov]

On Tue, 2004-01-13 at 10:07, Goswin von Brederlow wrote:
> Hi,
> 
> when compiling a kernel for my XXS1500 I allways ended up with a
> kernel oops in the network driver (au1000eth.c).
> 
> Finaly I checked the actual kernel source the running kernel was build
> from and found "CONFIG_BCM5222_DUAL_PHY" was set. Setting that solves
> the oops.
> 
> Maybe that could be caught in some way and prevented.

Well, the kernel shouldn't be crashing but as far as the BCM dual phy
thing -- I'm not sure how to detect it at run time without knowing ahead
of time that we've got one.  I admittedly haven't spent too much time
thinking about this problem but I didn't see an easy way to handle it.

Pete

> MfG
>         Goswin
> 
> ----------------------------------------------------------------------
> Start = 0x80274040, range = (0x80100000,0x802bbfff), format = SREC
> 
> Linux NET4.0 for Linux 2.4
> Based upon Swansea University Computer Society NET3.039
> Initializing RT netlink socket
> Starting kswapd
> pty: 256 Unix98 ptys configured
> Serial driver version 1.01 (2001-02-08) with no serial options enabled
> ttyS00 at 0xb1100000 (irq = 0) is a 16550
> ttyS01 at 0xb1200000 (irq = 1) is a 16550
> ttyS02 at 0xb1300000 (irq = 2) is a 16550
> ttyS03 at 0xb1400000 (irq = 3) is a 16550
> Generic MIPS RTC Driver v1.0
> RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
> loop: loaded (max 8 devices)
> au1000eth.c:1.4 ppopov@mvista.com
> eth0: Au1x Ethernet found at 0xb1500000, irq 28
> eth0: Broadcom BCM5222 10/100 BaseT PHY at phy address 0
> eth0: Using Broadcom BCM5222 10/100 BaseT PHY as default
> eth1: Au1x Ethernet found at 0xb1510000, irq 29
> Unable to handle kernel paging request at virtual address 00000000, epc == 801c0
> Oops in fault.c::do_page_fault, line 206:
> $0 : 00000000 1000fc00 00000000 001e3000 802597c8 0000001f 00000001 00000013
> $8 : 810cc800 b1510018 000011e0 802a1434 00000004 ba2e8ba3 1000fc01 00000002
> $16: 8029f940 802db12c 00000001 00000020 810cc800 810cc9e4 810cc960 802b4cf4
> $24: ffffffff 00000001                   802e4000 802e5ed8 0000ffff 801c5b60
> Hi : 000304cc
> Lo : ecaf8000
> epc   : 801c5c10    Not tainted
> Status: 1000fc03
> epc   : 00800008
> PrId  : 01030200
> Process swapper (pid: 1, stackpage=802e4000)
> Stack:    b1510000 801196ac 802598b8 810cc800 8029f940 802db12c 00000001
>  810cc960 810cc800 810cc9e4 810cc9f4 b1510000 0000001d 801c63a0 80259890
>  810cc800 b1510000 0000001d 87000266 00001123 00000001 802db12c 00000001
>  04000000 00000000 00000000 00000000 00000000 8008aa54 80287dac 80287c80
>  80287c6c 00000000 00000000 8028e530 8028e55c 00010f00 802746ec 80122ed8
>  8028100c ...
> Call Trace:   [<801196ac>] [<802598b8>] [<801c63a0>] [<80259890>] [<80122ed8>]
>  [<8025b9cc>] [<801007c4>] [<801007d4>] [<801007c4>] [<801047d4>] [<80111e94>]
>  [<8016125c>] [<80161220>] [<801047c4>]
> 
> Code: 8c420004  3c048026  248497c8 <8c460000> 0c044dc1  02802821  0807171f  000
> Kernel panic: Attempted to kill init!
> 
> 

Re: Kernel oops on XXS1500 in au1000eth.c

[Goswin von Brederlow]

Pete Popov <ppopov@mvista.com> writes:

> On Tue, 2004-01-13 at 10:07, Goswin von Brederlow wrote:
> > Hi,
> > 
> > when compiling a kernel for my XXS1500 I allways ended up with a
> > kernel oops in the network driver (au1000eth.c).
> > 
> > Finaly I checked the actual kernel source the running kernel was build
> > from and found "CONFIG_BCM5222_DUAL_PHY" was set. Setting that solves
> > the oops.
> > 
> > Maybe that could be caught in some way and prevented.
> 
> Well, the kernel shouldn't be crashing but as far as the BCM dual phy
> thing -- I'm not sure how to detect it at run time without knowing ahead
> of time that we've got one.  I admittedly haven't spent too much time
> thinking about this problem but I didn't see an easy way to handle it.
> 
> Pete

Maybe some

if (mach == MACH_XXS1500) ...

construct? The MACH_XXS1500 must be good for something.

Hopefully MyCable will change the name when they design a new board
with a different network thing.

MfG
        Goswin

Re: Kernel oops on XXS1500 in au1000eth.c

[Joerg Ritter]

> 
> Maybe some
> 
> if (mach == MACH_XXS1500) ...
> 
> construct? The MACH_XXS1500 must be good for something.
> 
> Hopefully MyCable will change the name when they design a new board
> with a different network thing.
> 

We will...

Greetings
Joerg


-- 

-------------------------------------------------------
Joerg Ritter        Tel: +49 48 73 90 10 866
mycable GmbH        Fax: +49 48 73 90 19 76
Boeker Stieg 43
D-24613 Aukrug      eMail: jr@mycable.de
-------------------------------------------------------