On Thu, May 04, 2000, Alan Cox <[EMAIL PROTECTED]> wrote:
> > > devfs has some unsolved
> > > race conditions at least the last time Al Viro reviewed it.
> >
> > So, limit the functionality because of some minor bugs?
>
> Actually we are more concerned with 'bugtraq specials'
This is a valid concern, however I fail to see why this is such a concern
to avoid deploying it.
Why would usbdevfs (with the custom VFS) have any less race conditions than
devfs?
In fact, with the extra eyes, statistics say there should be less.
> > > So USB needs to work without devfs..
> >
> > I don't buy this. I am not willing to hold back USB and more importantly
> > Linux because of FUD like this.
>
> Ok then I'll just have to fix USB so that it works without devfs.
So, following that logic, we should avoid depending on anything which is
new.
That would include usbdevfs as well as much of 2.3 since many things have
changed recently.
To be blunt, this is plain and simple FUD Alan.
However, I'm intested to see how you intend to make this work without
unnecessarily reinventing an already existing system and without reducing
functionality.
JE
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
