MUXer issues

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb Andy Green:
> Somebody in the thread at some point said:
> 
> | Anyway I think there shouldn't be a difference between MUX and in fact
> using 2
> | or 3 of the actually available serial links of MC75i. Maybe hw (=some
> others
> | + me) decides we don't need mux and use some 2nd very new (yet to
> discover)
> | additional serial link of 2442 CPU<->GSM - this mustn't break anything
> | though, things have to be consistent and maximum agnostic of actual
> | hw-platform!
> 
> Don't forget we don't have any spare UART on s3c2442 side.

that's why I talk about "very new (yet to discover) additional serial link"
Probably my speech was a little cryptic.
My point was -just again- sw side should be designed as hw-agnostic as 
possible.
Don't care whether it's vitual MUX channel, a rs232 or USBtty: if it's a 
serial link, it has to have a device in /dev called tty*
So this is a kernel-driver topic, to demux the virtual serial links and 
provide a device for each.

/j 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: This is a digitally signed message part.
Url : http://lists.openmoko.org/pipermail/openmoko-kernel/attachments/20080529/22bf610e/attachment.pgp