i386/105616: UART PCI device just silent...
Helge Oldach
puc-uart at oldach.net
Tue Nov 28 12:31:32 PST 2006
The following reply was made to PR i386/105616; it has been noted by GNATS.
From: puc-uart at oldach.net (Helge Oldach)
To: xcllnt at mac.com (Marcel Moolenaar)
Cc: FreeBSD-gnats-submit at FreeBSD.org
Subject: Re: i386/105616: UART PCI device just silent...
Date: Tue, 28 Nov 2006 21:14:19 +0100 (CET)
Hi Marcel,
>> The only thing that may be botched is speed, or potentially also
>> parity or stop bits.
>It's definitely an odd failure mode. What happens if you wire one of
>the UART ports on the PCI card to one of the "legacy" on-board serial
>ports? Could you try both the PCI UARTs?
I did. What happens is that single, typed-in characters don't make
to the remote, while files piped to the remote via ~>file appear as
garbage. The garbage contains a lot of 0xff characters. So there *is*
some sort of communication in place. It doesn't matter whether I send
on the puc uart or the on-board uart. It also doesn't make too much
difference if I play with the transmit and receive speeds.
This all points into the direction of speed (parity, stopbits) mismatch.
Going to extremes: When I set the puc UART to 1200 bps and the on-board
to 115200, I constantly receive a 0xff on the on-board for any arbitrary
character typed on the puc uart. But there is silence the other way
'round.
There is no difference between the two puc/uart ports. Garbage either
way, but at least there is communication on both ports. So IO ports and
IRQ are likely correct.
Any further ideas? Is there a simple means to measure the actual speed?
Helge
More information about the freebsd-i386
mailing list