per-device sysctls

[Poul-Henning Kamp]

In message <xzpu11dk7a2.fsf at dwp.des.no>, Dag-Erling =?iso-8859-1?q?Sm=F8rgrav?=
 writes:
>"M. Warner Losh" <imp at bsdimp.com> writes:
>> des at des.no (Dag-Erling Smørgrav) writes:
>> :  1) it is immensely easier to access
>> From whose point of view.
>
>Programmatically.  The in-kernel implementation is much simpler, and
>the userland code required to access it will also be simpler.
>Accessing individual devices (once you know where they are) should
>also be much simpler since devinfo only gives you "all or nothing".
>
>
>> :  2) it gives drivers a well-defined place to put their per-device
>> :     sysctl variables - devinfo doesn't address that issue at all
>> That is a good reason to transitioning to this, so long as we can come
>> up with a good way to represent detached nodes.
>
>As long as they have a device_t, it should be a piece of cake.

Having a device_t is a property of having hardware, not of being a
device driver.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.