Re: [PATCH] Re: [PATCH] Multi protocol support (stage #1)

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

 



On Sun, 7 May 2006, Johannes Stezenbach wrote:
> On Sat, May 06, 2006, Trent Piepho wrote:
> > NAME
> >   FE_GET_CAPS - query front-end capabilities
> >
> > SYNOPSIS
> >   int ioctl(int fd, int request = FE_GET_CAPS, struct dvb_frontend_cap *cap);
>
> [snip]
>
> Wow, lot's of work you put into this.
>
> However, IMHO if one needs such a verbose description,
> then the API is bad.

I was trying to write a real spec.  Something that you can use to tell if you
code is correct without having to look at the driver source, and something
that will let you write a driver and know it will work correct software.  So,
I tried to make all the details very clear.

This is probably enough for most programmers:

Set delivery to 0 to get a bitset of supported delivery systems and the name.
Set delivery to a single delivery system to get that system's capabilities.

> See my other mail about the seperate FE_GET_DELSYS ioctl, I think
> this is much friendlier to the poor users' brains ;-)

Basically that takes GET_CAPS with delivery == 0 and makes it a new ioctl.
GET_CAPS is now simpler, but you have another ioctl in the API.  For single
standard tuners, the get_caps ioctl's code stays the same, but a new get_info
or get_delsys ioctl needs to be added.  So the trade off is more work in the
drivers, for a simpler API.  Either way seems reasonable to me.

_______________________________________________

linux-dvb@xxxxxxxxxxx
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

[Index of Archives]     [Linux Media]     [Video 4 Linux]     [Asterisk]     [Samba]     [Xorg]     [Xfree86]     [Linux USB]

  Powered by Linux