[TEST(/review)] major/minor/devname fix

[Poul-Henning Kamp]

In message <20050310065126.GA28001 at nagual.pp.ru>, Andrey Chernov writes:
>On Thu, Mar 10, 2005 at 07:32:39AM +0100, Poul-Henning Kamp wrote:
>> There is no risk of collision, look at what devfs_random() does.
>
>What happens with whole devfs_random() logic, in case first devfs_seed 
>obtained after masking with 0xf0f becomes 0 again (as it was before)? And 
>second? And third? It may happens especially in some "no good entropy yet" 
>scenario.

Then somebody beats our PRNG maintainers over the head until they get
their act together.

-- 
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.