Re: PATCH/RFC: Bring lin-lwp performance back to the real world

[Daniel Jacobowitz <drow at mvista dot com>]

On Fri, Nov 22, 2002 at 10:58:44AM -0800, Michael Snyder wrote:
> Andrew Cagney wrote:
> > 
> > >
> > > Hrm, possibly.  I needed to create linux-nat.c anyway (I'll need it for
> > > some things that are definitely not /proc related) but I could be
> > > persuaded either way on linux_proc_xfer_memory.  It's not focused on
> > > the "proc" bit as much as the "xfer" bit, but it's definitely using
> > > /proc.  If you prefer I'll move it, and save linux-nat.c for another
> > > patch.
> > 
> > Not really my problem (It's a linux / lin-lwp area).  I just figure
> > that, if you put it in linux-proc.c, you've a more compelling argument
> > for getting the change into 5.3 (as if I'm going to stand in its way :-):
> > 
> > - linux-proc.c provides you with the `prior art'.  The other code in
> > that file is pulling an identical trick - using /proc when it should
> > really be using ptrace().
> > 
> > - it trims the change back to something more managable (all the config
> > parts go) so it is easier to be sure it's right.
> > 
> > enjoy,
> > Andrew
> 
> Honest, I made my reply before reading Andrew's!  ;-)
> Daniel, that's two votes for using linux-proc.c.

OK.

I'm going to sit on this for a couple of hours while I pound on it; I
was testing the patch in another environment and saw very strange
performance numbers.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer