George Campbell wrote:
>
> i'm working with an older version of the kernel so the interfaces may have changed,
>but couldn't you just use:
>
> tty->ldisc.recieve_buf( tty, data, NULL, urb->actual_length );
I had a look at this.
This would work assuming low_latency is set (which is it in most of the
serial drivers now I think). If low_latency was not set then the
behaviour would be incorrect.
I did a quick test doing this (spelling receive_buf properly :-) and
there was no significant speed difference doing loopbacks at 115k
(doing a short approx 4 second test sending and receiving around
50kbytes) between the two versions.
The downside is when the drivers start handling things like Framing
errors, parity errors etc (which I am doing for the ftdi device at the
moment)we would have to create an array of error flags to go with the
data bytes which would be a pain.
>
> > -----Original Message-----
> > From: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED]]On Behalf Of Greg KH
> > Sent: Tuesday, May 15, 2001 9:46 PM
> > To: Johannes Erdfelt
> > Cc: [EMAIL PROTECTED]
> > Subject: [linux-usb-devel] [PATCH] USB acm driver fix
> >
> >
> > Hi,
> >
> > Here's a small patch to the acm driver that fixes a potential buffer
> > overrun problem that can occur. It is against 2.4.5-pre2.
> >
> > thanks,
> >
> > greg k-h
> >
> >
>
> _______________________________________________
> [EMAIL PROTECTED]
> To unsubscribe, use the last form field at:
> http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
--
Bill Ryder SGI New Zealand
Systems Engineer Ph: (+64 4) 494 6326
[EMAIL PROTECTED] icq: 16285091 Mobile: (+64 21 67 9507)
DISCLAIMER: Unless explicitly stated all opinions are mine not SGI's
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
