On Tue, 2004-09-07 at 21:38, Christian T. Steigies wrote: > On Tue, Sep 07, 2004 at 08:26:07PM +0200, Kars de Jong wrote: > > Now it was time to find out which config option caused the problem... > > > > Took me quite some recompile/reboot cycles, but it turned out to be > > CONFIG_SINGLE_MEMORY_CHUNK. If this is NOT defined as "Y", the kernel > > crashes. And this is apparently also true for 2.4.x (x > ??) kernels.
> Found it under advanced options. Do we need this only for mvme16x or for > other systems as well? Any other advance option, or only this? Well, I have CONFIG_RMW_INSNS enabled too, but its help suggests you shouldn't :-P Also, CONFIG_060_WRITETHROUGH specifically mentions the 53C710 driver as a possible source of problems if this option is not enabled... But I don't have a '060 MVME board, so I'm not sure. I'd say leave it disabled unless someone reports a problem with it. > You are saying we need to build 2.4.27 with this too? What about 2.4.26, > does that debian version work? No idea, I'll try it tomorrow. > I don't have a problem to activate this option, if it does not hurt anybody. > But it will be a challenge to get this into the archive for 2.4.27... I don't know... it breaks for subarchitectures with multiple non-contiguous memory regions, like my PME68-22 board. But AFAIK the other VME boards all have a single chunk of memory, so at least the mvme16x kernel could be built with that option enabled. I'll try the MVME147 kernels tomorrow as well. Kind regards, Kars.
