Re: raid5d, page_launder and scheduling latency

[Andrew Morton]

Alan Cox wrote:
> 
> > - exit with 1,000 files open
> > - exit with half a million pages to be zapped
> >
> > And "fixing" copy_*_user is outright dumb.  Just fix the four
> > or five places where it matters.
> 
> Depends if you want to use Linux as a Windows 3.1 replacement or a real OS.
> In the latter case we need to fix the stupid cases too.

umm..  That's what I said :)

Hunt them down and fix them, but don't add hundreds of schedules
all over the tree.  gettimeofday(), for heavens sake.