Re: owner field in `struct fs'

[Ingo Oeser]

On Sat, Jun 24, 2000 at 11:38:11PM -0400, Alexander Viro wrote:
> Random module writer does not have to think about it - the rule is dead
> simple. You have a character device and it's modular - stick
> owner:THIS_MODULE into file_operations and forget about it. Similar rules
> go for other subsystems. Notice that they are replacing the current rules
> regarding the places where you have to put MOD_{INC,DEC}_MOD_COUNT and
> they are way simpler.

Aehm, what should I do in the hardware detection phase? There I
don't know if I have the hardware at all. Or do you propose, that
the module is locked during init() and cleanup of my module?

I[1] am ok with these changes as long as these mechanisms are
_DOCUMENTED_ somewhere _EXCEPT_ in the source itself.

<HINT>Kernelbooks are hip these days</HINT>

Looking at old code (which you proposed for subsystem writers)
also means making old mistakes again.

You[2] should know this from your "I change core semantics and
have to implement it everywhere"-experience ;-)

[1] as writing a upcoming subsystem
[2] And all other people explicit listed in the header of this
   mail
-- 
Feel the power of the penguin - run [e-mail]
<esc>:x