[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 2.4 kernel performance?

Johan Adolfsson wrote:

> An alternative solution could be to implement char-by-char interrupt driven
> input så you don't have to flush the fifos.
> Another way could be to create a raw serial driver that skips the tty layer
> so that a read(buf,count) would wait for count characters and then return.
> (Either by setting up the DMA to that length or by interrupt driven I/O
> counting the received bytes. )
> I believe Mikael has worked on a raw driver that support up to 6.5Mbps
> without loosing characters, perhaps that could be used as a base
> once it is tested and released.

Is it possible to share this driver with us. I would like to take a look 
at it.
Do you have any example code snippets for implementing 'simple' 
interrupt driven serial IO. (to implement RS485 driver)