5 messages in org.freebsd.freebsd-mobileRe: sio interrupt problems with PAO-9...| From | Sent On | Attachments |
|---|---|---|
| Mark Diekhans | Jun 25, 1996 11:26 pm | |
| Michael Smith | Jun 26, 1996 12:37 am | |
| Mark Diekhans | Jun 26, 1996 8:11 am | |
| Michael Smith | Jun 26, 1996 8:51 pm | |
| Mark Diekhans | Jun 27, 1996 10:21 am |
| Subject: | Re: sio interrupt problems with PAO-960616 [pccard-test] | |
|---|---|---|
| From: | Mark Diekhans (mar...@grizzly.com) | |
| Date: | Jun 26, 1996 8:11:31 am | |
| List: | org.freebsd.freebsd-mobile | |
Hi Mike, Thanks for the response....
Mark Diekhans stands accused of saying:
The same results occur with two difference modems, one of which is on the supported car list (Megahertz XJ2288). I suspect this is either a configuration mistake or a problem with pcic or sio drivers:
/kernel: PC-Card Cirrus Logic PD6710 (5 mem & 2 I/O windows) /kernel: pcic: controller irq 10 /kernel: sio0 at 0x3f8-0x3ff irq 4 on isa /kernel: sio0: type 16550A /kernel: sio1 not found at 0x2f8
WHen the kernel is booting I get the message: /kernel: sio0: 64 events for device with no tp
although there is not device attached to sio0.
This sounds like the card doesn't have a 'real' UART in it, but rather a micro pretending to be one. This is pretty tragic.
Don't know for sure, but these are both two well know cards: Cardinal MPV288C2 Megahertz XJ2288
You could look the places in sioprobe() where the EXTRA DELAY comments are, and try adding 'DELAY(500)' in these places.
Didn't help. I got the passed the probe error on the cardinal but then go the unfielded interrupt later.
/kernel: sio1: probe test 3 failed
The card failed to generate an interrupt when it should have.
/kernel: sio1: type 16550A /kernel: Slot 0, unfielded interrupt (3)
... but it's generated it later.
Actually one or the other occurs depending on the reset delay: probe 3 fails or the fielded interrupt. An then there is the 64 stray strange interrups on sio0..
Thanks for the suggestions, Mark