cvs commit: src/sys/fs/nullfs null.h null_subr.c null_vnops.c

[Poul-Henning Kamp]

In message <20030618122054.GA55870 at fling-wing.demos.su>, Dmitry Sivachenko writes
:
>On Wed, Jun 18, 2003 at 01:53:29PM +0200, Poul-Henning Kamp wrote:
>> All of this have tangled the simple component formerly known as the
>> buffer cache up in so many ways, that it is very hard for anybody
>> to make heads and tails of it any more.
>> 
>> So I am tempted to answer you question with:  "Because it is all a
>> mess"
>> 
>
>Are there any more-or-less correct FS implementation in the system one
>could learn how things should be done from?

The only arguably correct FS we have is by fiat of most use UFS.
Unfortunately that is also the most complex FS we have.

This is also a situation we should try to fix.

-- 
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.