Re: 5.3-RELEASE TODO

On Thu, Jul 15, 2004 at 09:18:29PM -0600, Scott Long wrote:
> Scott Long wrote:
> >Tim Robbins wrote:
> >
> >>On Thu, Jul 15, 2004 at 03:20:19PM -0700, Colin Percival wrote:
> >>
> >>>At 15:04 15/07/2004, Kris Kennaway wrote:
> >>>
> >>>>* linprocfs [...]
> >>>>     if (pvd->pvd_pid != NO_PID) {
> >>>>              if ((proc = pfind(pvd->pvd_pid)) == NULL)
> >>>>                      PFS_RETURN (ENOENT);
> >>>>-->             vap->va_uid = proc->p_ucred->cr_ruid;
> >>>>
> >>>>rwatson has a patch that works around this particular null pointer
> >>>>deref, but the underlying cause is not addressed.
> [...]

> Ok, I've officially been in front of the computer for too long today.
> This response was actually meant for a different message.  Sorry for the
> confusion.
> 
> Scott

Hmm.. on a hunch I'll assume this was meant for the plip thread on
-current ;)

> >Please don't take this the wrong way, but would you consider using a
> >null-modem serial cable and SLIP or PPP instead?  PLIP has been
> >neglected nice it was first introduced, and I doubt that it has much
> >life left.  Setting up SLIP and PPP is trivially easy and has a much
> >better chance of working between OS's.
> >
> >Scott

I can set that up yeah, I used plip since I assumed it would be
faster than ppp over serial line.  I've been getting at most
~60 KBps.

My only reservation in removing this would be, it came in handy
and still primarily works.  It's useful absent any other link but
imposes a large overhead.  Should this driver still use splhigh()?

I'll also note that even NetBSD seems to omit plip support at the
moment.   So I'll take that as a prophetic nod from the code-reaper.

Funny thing is now I'm having trouble reproducing the IP header
problem.. could it be cable/hardware that was the culprit?

> Scott

Thanks..
  Allan