4.4 pccard insanity
Rudolfo Munguia
I have taken the time now to re-install bsd and test using both my NP10T
and an addtron 802.11b card.
I set the "hw.pcic.intr_path="1" for ISA routing, and left the pcic in
polling mode,
The messages now vary slightly from before, but both cards hang the
system.
I tried with -h in boot to monitor over a serial connection, then found
that I cannot get my main freebsd workstation to display the input from
the mobile.
Failing that, I tried booting with "pccard_enable=YES" and
"pccard_flags=-i 11" and then inserting the card after boot, this yields:
card inserted slot 0
after which I sometimes have enough time to enter "pccardc dumpcis"
before it hangs and I have to remove the battery,
Debug does not come up though,
I decided to try and force a panic to get debug up by pulling the NP10T
out after entering the "pccardc dumpcis" command and get the following:
Tuple #1, code 0xff (Terminator, length = 0
1 slots found
No card in database for "(null)"("(null)")
and I am returned to a working prompt.
I thought then that maybe the NP10T was bad and tried it out on a fujitsu
lifebook, and it was recognized and installed by Windows ME without
problems.
pccardc is obviously trying to read the CIS before the system Hangs,
however if I wait longer than 4-5 seconds before pulling the card (hoping
that dumpcis will yield more info) then I get debug:
Fatal trap 12: page fault while in kernel mode
fault virtual address = 0xff90afa1
fault code = supervisor read, page not present
instruction pointer = 0x8:0xc0159883
stack pointer = 0x10:0xc2173ea0
frame pointer = 0x10:0xc2173eb0
code segment = base 0x0, limit 0xfffff, type x1b
= DPL 0, pres 1, def32 1, gran 1
processor ef lags = interrupt enabled, resume, IOPL = 0
current process = 203 (pccardc)
interupt mask =
kernel: type 12 trap, code=0
stopped at lockmgr+0x27: orl 0x4(%ebx),%esi
This message is consistent across reboots (excepting process id)
can anyone interpret this to determine if this is the actual system
problem, or is this just complaining about the card being removed?
ru...@primenet.com
'If you're not living on the edge, then you're taking up too much space'
To Unsubscribe: send mail to majo...@FreeBSD.org
with "unsubscribe freebsd-mobile" in the body of the message
Rudolfo Munguia
Re-read the pccardd man page and ran pccardd -dv from the command line;
The pccardd keeps hanging at ata, which is of course the hard drive which
may explain the fact that I have to manually fsck after each time I start
pccardd.
So, taking the brute-force route, I did:
mv /etc/defaults/pccard.conf /etc/defaults/_pccard.conf
then vi /etc/pccard.conf
io 0x240-0x360
irq 3 5 10 11 15
memory 0xd40000 96k
# Network Everywhere Ethernet 10BaseT PC Card
card "Network Everywhere" "Ethernet 10BaseT PC Card"
config auto "ed" ?
insert /etc/pccard_ether $device start
remove /etc/pccard_ether $device stop
then vi /etc/rc.conf
pccard_conf="/etc/pccard.conf"
The whole upshot of this is that now when I start pccardd I don't have to
fsck the drive, although the system still hangs and I have to remove the
battery to reboot.
still searching.....
Rudolfo Munguia
Just to blatantly display my ignorance, and hopefully in the process reach
some level of enlightenment; I present the following observations:
A) In looking through /etc/ I happened upon rc.pccard, upon inspecting
this it appears that pccardc is used to set mem/beep/conf before pccardd
is called.
B) I find on many posts to the group that pccardc is unable to operate
unless pccardd is started first.
Which is the proper case?
If A then pccardc should be able to pull a dumpcis regardless of the
pccardd state. (not the observed behavior)
If B then could rc.pccard be the point of failure?