Re: GEM and memory protection

[Konrad Kokoszkiewicz <draco@mi.com.pl>]

Hi

> > The reason is: because the memory protection as it works now with GEM has
> > something to do with system stability. Explanation: you never know what a
> 
> It can cause the desktop [eg Thing] to be killed, but this is a recoverable
> error, surely? 

Of course. A reboot recovers from it.

> > new program you download from the net will kill your desktop and crash the
> > whole system. The result: people don't use memory protection, because the
> > memory protection makes the system less stable. A conclusion: the memory
> > protection has no real use. Proposition: optionally make the GEM
> > applications running in global memory and protect the rest of the system.
> 
> This is a prejudice against all GEM programs, IMO.

I don't think they would get that personally. <g>.

Seriously:

1) if the desktop can be protected agains being killed by a faulty AV
   client, I think it should be protected and no additional patches
   should be made.

2) if the desktop can't be protected, I think it is not reasonable to
   keep things as they are. Perhaps a fix like that (global mem for all
   the gem) is a "prejudice" what "punishes" well written applications,
   but a lack of it "punishes" the whole system bringing it down if a
   faulty AV client is being executed.

> A better solution, IMO, is for a patch to the AV servers such as the 
> one mentioned elsewhere in this thread, so that the offending application 
> is killed rather than the desktop.

Yes.

Gtx,

--
Konrad M.Kokoszkiewicz
|mail: draco@mi.com.pl                  | Atari Falcon030/TT030/65XE |
|http://www.orient.uw.edu.pl/~conradus/ | ** FreeMiNT development ** |

** Ea natura multitudinis est,
** aut servit humiliter, aut superbe dominatur (Liv. XXIV,25)
*************************************************************
** U pospolstwa normalne jest, ze albo sluzy ono unizenie,
** albo bezczelnie sie panoszy.