Messages in this thread | | | From | "Jonathan H. Pickard" <> | Subject | Re: /dev/kbd | Date | Tue, 26 Mar 1996 20:44:02 -0800 (PST) |
| |
Quoth Albert Cahalan: > > > : For a REAL consistent solution IMHO some kernel support is needed as it > > : is only the kernel that knows the current, global, keyboard state. > > > > One might introduce another ioctl that returns the key_down array. > > But I don't like this proliferation of ioctls. And I think svgalib > > can survive just as well as X can without it. > > Forget the ioctls, they are a pain to use. It would be better to > be able to mmap the keyboard.
Yes, please.
> Maybe even be able to request that > some signal be sent whenever a change occurs, but I suppose select > would do. (but getting SIGUSR1 would be very nice)
The user signals should remain user signals. SIGIO exists, it's in the header files, and it happens to be standard. I don't suppose it'd be as bright an idea to use it for a sensible purpose, though. =:^)
-- Jon Pickard * 149 Olive St #45 * Paso Robles, CA 93446 * +1 805 2399518 [.sig under construction -- please excuse our dust]
| |