Re: USB device allocation

Zack Weinberg (zack@bitmover.com)
Wed, 06 Oct 1999 11:12:11 -0700


david parsons wrote:
> In article <linux.kernel.37FA7D3A.F66BFB27@pobox.com>,
> Jeff Garzik <jgarzik@pobox.com> wrote:
> >"H. Peter Anvin" wrote:
> >> Actually, the need is for a decent-sized dev_t.
> >
> >We're still pre-code-freeze... it would be great to change this before
> >2.4.0.
>
> 3.0
>
> That's a pretty major kernel interface change that will break
> close to everything in userland. If version numbers mean
> anything, this sort of change is suited for a major version
> number revision.

It should suffice to find all of the system calls and ioctls that
take dev_t arguments, and add new versions. Old code will keep using
the old system calls.

Whether this can be done without running out of system call numbers is
another question entirely :)

Also, the encoding of a user visible dev_t will have to be rather icky
in order to avoid breaking stuff:

| high 24 bits major | high 24 bits minor | low 8 maj | low 8 min |

This is the only definition that will allow the macros in glibc's
<sys/sysmacros.h> to be transparently updated.

kdev_t, of course, can be completely different. Didn't someone have
patches that made it a pointer?

zw

-
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/