> > Go to <URL: http://www.equinox.com/tech/ftp_sst2.htm#a7>, look for
> > "kernel bug". Anyone who advocates binary-only drivers should be
> > forced to try to get an Equinox SST card working on 2.0.35.
> >
> Its also derived from Ted Tso's serial driver and seems in violation of the GPL
> if you compare their assembler with the output of gcc -S on serial.c
>
> The good side of that btw, is its probably trivial to turn their driver
> back into C since you can paste bits back into serial.c
Can Tso comment on this? The price of having closed drivers is that you
can't derrive from other drivers, we can't let them have their cake and
eat it too. :)
This is actually the first time I've heard of this. Thank you Alan for
bringing it to my attention. I'll have to look into this.
It's actually their loss, since I usually try to make best efforts to
notify tty driver maintainers when a bugfix to the high-level tty code
also requires changings in the low-level tty drivers. This has happened
in the past, and it'll likely happen again.
(It's also likely that sometime during 2.3 I may want to change the tty
interface, to rationalize how hangup processing is done. The current
architecture is pretty ugly, and it really wants to be cleaned up.
There may also be changes required if and when we want to support
finer-grained SMP locking within the tty layer. As per usual,
binary-only distributions will lose if and when these interface changes
happen.)
- Ted
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/