if_dc and trap on shutdown

[Panagiotis Astithas]

Hi all,

on my supposedly rock-solid (as of late) HP Omnibook XE3, I still seem 
to get some interrupt storms rarely when booting. For the first time 
though, I got a trap as well when shutting down. The culprit seems to be 
the dc driver which gets an interrupt a bit late in the shutdown 
sequence. I usually get the storms when I reboot from Windows into 
FreeBSD *with* the netowrk cable plugged in. Before the last upgrade, I 
used to get storms with or without the cable plugged in, on occasion. 
This time though I had the laptop hibernated (ACPI S4BIOS) from FreeBSD 
when I was not connected and awaken with the network cable plugged in. 
This is what happenned when I tried to shut down.


The following is a hand-copied output:
--------------------------------------

[...]
Waiting (max 60 seconds) for system process `bufdaemon' to 
stop...Interrupt storm detected on "irq: acpi0"; throttling interrupt source
stopped
[...]
Shutting down ACPI


Fatal trap 12: page fault while in kernel mode
fault virtual address	= 0x18
fault code		= supervisor write, page not present
instruction pointer	= 0x8:0xc05b8054
stack pointer		= 0x10:0xcd316cb0
frame pointer		= 0x10:0xcd316ccc
code segment		= base 0x0, limit 0xfffff, type 0x1b
			= DPL 0, pres 1, def32 1, gran 1
processor eflags	= interrupt enabled, resume, IOPL = 0
current process		= 22 (irq11: cbb0 cbb1+)
kernel: type 12 trap, code=0
Stopped at	dc_rxeof+0x168: movl	%edx,0x18(%eax)
db> tr
dc_rxeof(c2b99000) at dc_rxeof+0x168
dc_intr(c2b99000) at dc_intr+0xef
ithread_loop(c14f8800,cd316d48,c06b9980,0,c065d0f4) at ithread_loop+0x174
fork_exit(c04d92ec,c14f8800,cd316d48) at fork_exit+0x60
fork_trampoline() at fork_trampoline+0x8
--- trap 0x1, eip = 0, esp = 0xcd316d7c, ebp = 0 ---
db> sh intrcnt
irq0: clk		571043
irq1: atkbd0		6659
irq3: sio1		1
irq4: sio0 		1
irq5: pcm0 uhci0	37227
irq6: fdc0		5
irq7: ppc0		1
irq8: rtc		730958
irq9: acpi0		17550
irq11: cbb0 cbb1+	56
irq12: psm0		50115
irq13: npxo		1
irq14: ata0		17764
irq15: ata1		25900

--------------------------------------
The machine is a May 11 -current and unfortunately I don't have a debug 
kernel handy. You can find a standard dmesg in:
http://noc.ntua.gr/~past/dmesg.HPXE3

Cheers,
-- 
Panagiotis Astithas
Electrical & Computer Engineer, PhD
Network Management Center
National Technical University of Athens, Greece