Jan Kiszka wrote:
> Gilles Chanteperdrix wrote:
> > Jan Kiszka wrote:
> > > Hi,
> > >
> > > while my notebook (Thinkpad R40, i82801DB chipset) produces very nice
> > > worst-case numbers under RTAI/fusion with SMI disabled, it
> > unfortunately
> > > fails to reboot when this fix is enabled. Therefore, I quickly patched
> > > rtai_hal to restore the SMI settings on reboot - works great for me. :)
> >
> > Thanks, applied. But I wonder, would not it be possible to call
> > __fusion_sys_exit from nucleus/module.c upon reboot notification ? If
> > you have opportunity to check, I would be interested.
> >
>
> ??? Did not yet really get what you mean. In what way should
> __fusion_sys_exit help to deal with this problem? Note that the issue
> already show up when only rtai_hal is loaded and no nucleus.
I did not think about the hal module only issue, and it looks like I
mixed up xnarch_exit and __rthal_exit. What I meant is that in order to
have a "generic" clean exit we may call each module exit function upon
reboot notification. Obviously, we need one such callback by module,
though.
--
Gilles Chanteperdrix.
