Re: khttpd rotten?

[Martin Dalecki]

Uz.ytkownik David S. Miller napisa?:
>    From: Andrew Morton <[e-mail]>
>    Date: Thu, 09 May 2002 12:30:59 -0700
> 
>    The concern with moving one (major) application into the
>    kernel is that this will weaken the testing/motivation to get
>    zerocopy, aio and sophisticated notifications working well
>    for userspace.
>    
> Actually, to the contrary, TUX was in fact an impetus for the
> userlevel zerocopy and AIO bits :-)
> 
> I personally don't see anything wrong with something like the
> TUX engine being in there.  At the same time I want to reiterate what
> Ingo said which is what we can do in userspace catches up to what
> TUX can do then we pull it out and move on to the next thing :-)

It's far easiet to add then to remove. Trust me ;-).

I vote against both of them: tux and khttpd are should have
no place in the kernel of a General Pupose OS kernel.