Bug #391

Boot time panic with latest Preview

Added by steve over 7 years ago. Updated over 7 years ago.

Status:ClosedStart date:
Priority:NormalDue date:
Assignee:-% Done:

0%

Category:-
Target version:-

Description

Hi,

Following the note that the Preview tag has been bumped I updated
my sources today, built world and kernel (custom kernel config at
ftp://sohara.org/pub/STEVE) rebooted and got a panic during the boot with
the message below. I've put a full dmesg.boot from the box running the
previous kernel at ftp://sohara.org/pub/dmesg.boot for reference. Hardware
is an Asus A8V with a no name NGE based board installed and a Pinnacle/Miro
TV tuner card.

...
sio1: type 16550A
ppc0: parallel port not found.
panic: bogus ife->ifm_data (10)

Trace beginning at frame 0xd1416c1c
panic(c040bdbd,c048eee0,c04097f0,d1416c4c,2) at panic+0x99
panic(c04097f0,a,850e0000,7000,0) at panic+0x99
mii_phy_set_media(c822dd88,0,0,80207,c822dd88) at mii_phy_set_media+0xa3
nsgphy_service(c822dd88,c1762f70,2,c1725460,20) at nsgphy_service+0x128
mii_mediachg(c1762f70,0,8802,c1762f70,ccedb0b8) at mii_mediachg+0x32
nge_stop(ccedb0b8,ccedb0b8,c01d6e2c,ccedb254,ff800000) at nge_stop+0x149
nge_intr(ccedb0b8,0,0,c824bb68,ccedb254) at bge_intr+0x1d
lwkt_serialize_handler_call(ccedb254,c019bb69,ccedb0b8,0,1) at
lwkt_serialize_ha ndler_call+0x5f
ithread_handler(a,0,0,0,0) at ithread_handler+0x8a
lwkt_exit() at lwkt_exit
Debugger("panic")
Stopped at Debugger+0x44: movb $0,in_Debugger.0

History

#1 Updated by dillon over 7 years ago

: Hi,
:
: Following the note that the Preview tag has been bumped I updated
:my sources today, built world and kernel (custom kernel config at
:ftp://sohara.org/pub/STEVE) rebooted and got a panic during the boot with
:the message below. I've put a full dmesg.boot from the box running the
:previous kernel at ftp://sohara.org/pub/dmesg.boot for reference. Hardware
:is an Asus A8V with a no name NGE based board installed and a Pinnacle/Miro
:TV tuner card.
:
:...
:sio1: type 16550A
:ppc0: parallel port not found.
:panic: bogus ife->ifm_data (10)
:
:Trace beginning at frame 0xd1416c1c
:panic(c040bdbd,c048eee0,c04097f0,d1416c4c,2) at panic+0x99
:panic(c04097f0,a,850e0000,7000,0) at panic+0x99
:mii_phy_set_media(c822dd88,0,0,80207,c822dd88) at mii_phy_set_media+0xa3
:nsgphy_service(c822dd88,c1762f70,2,c1725460,20) at nsgphy_service+0x128
:mii_mediachg(c1762f70,0,8802,c1762f70,ccedb0b8) at mii_mediachg+0x32
:nge_stop(ccedb0b8,ccedb0b8,c01d6e2c,ccedb254,ff800000) at nge_stop+0x149
:nge_intr(ccedb0b8,0,0,c824bb68,ccedb254) at bge_intr+0x1d
:lwkt_serialize_handler_call(ccedb254,c019bb69,ccedb0b8,0,1) at
:lwkt_serialize_ha ndler_call+0x5f
:ithread_handler(a,0,0,0,0) at ithread_handler+0x8a
:lwkt_exit() at lwkt_exit
:Debugger("panic")
:Stopped at Debugger+0x44: movb $0,in_Debugger.0

Maybe Sephe can work out what is going on here. Somehow it is hitting
a phy entry created at dev/netif/mii_layer/mii_physubr.c line 379.
I don't know whether this is correct or not, but if it calls
mii_phy_set_media() with that entry it will panic just as you have
above.

-Matt
Matthew Dillon
<>

#2 Updated by sepherosa over 7 years ago

Oops, I didn't notice that nge(4) uses mii in this way :-(
Please review/test following patch:
http://leaf.dragonflybsd.org/~sephe/nge.diff

Best Regards,
sephe

#3 Updated by steve over 7 years ago

On Wed, 29 Nov 2006 17:20:20 +0800
"Sepherosa Ziehau" <> wrote:

Applies cleanly, compiles fine and works a treat.

Thank you very much.

#4 Updated by dillon over 7 years ago

:On Wed, 29 Nov 2006 17:20:20 +0800
:"Sepherosa Ziehau" <> wrote:
:
:> Oops, I didn't notice that nge(4) uses mii in this way :-(
:> Please review/test following patch:
:> http://leaf.dragonflybsd.org/~sephe/nge.diff

:Steve O'Hara-Smith wrote:
:
: Applies cleanly, compiles fine and works a treat.
:
: Thank you very much.
:
:--

Nice! I'll slip the Preview tag as soon as you commit that.

-Matt
Matthew Dillon
<>

Also available in: Atom PDF