Re: [uCsimm] Minor progresses on CS8900 driver for uCsimm

From: Craig Peacock (Craig.Peacock@senet.com.au)
Date: Mon Apr 09 2001 - 05:53:11 EDT


Bernhard,

> I have done some work on the CS8900 driver for uCsimm:
>
> Instead of using uCcs8900, i took the more generic
> driver for cs89x0, because it is already adapted to
> linux 2.4.x. Currently, the device is recognized
> during the boot progress, and can be configured
> using ifattach, but when pinging the uCsimm it
> crashes with a kernel bug :-(
>
> Here is the relevant output:
>
> [...]
> cs89x0:cs89x0_probe()
> cs89x0.c: v2.4.0-test11-pre4 Russell Nelson <nelson@crynwr.com>,
> Andrew Morton <andrewm@uow.edu.au>
> eth0: cs8900 rev J found at 0x10000300 media RJ-45,,
> programmed I/O, MAC 00:50:c2:03:6b:15
> cs89x0_probe1() successful
> [...]
> # /sbin/ifattach \
> --addr 192.168.1.2 \
> --mask 255.255.255.0 \
> --net 192.168.1.0 \
> --gw 192.168.1.1 eth0
> eth0: using half-duplex 10Base-T (RJ-45)
> # skput:over: 0063643e:196 put:98 dev:eth0kernel BUG!
> skput:over: 0063643e:196 put:98 dev:eth0kernel BUG!
>
> The patch can be downloaded at:
>

Yes, this looks familiar - the error is coming from the net_rx()
routine. I had to fiddle the insw(ioaddr + RX_FRAME_PORT, skb_put(skb,
length), length >> 1); line to get it to work.

The patch I have at ftp://ftp.beyondlogic.org/uClinux/uCcs89x0.diff.gz
is based on the cs89x0 source, I have just renamed it back to
uCcs89(x)0.c and hacked out all the junk we didn't need.

Regards,

Craig Peacock
This message resent by the ucsimm@uclinux.com list server http://www.uClinux.com/



This archive was generated by hypermail 2b30 : Sun Apr 07 2002 - 00:01:42 EST