Andrew Edem wrote:
> I always keep USB a module, and I've set MPS to 1.1, and disabled power
> management. The system works for a while, but locks up whenever a usb
> event occurs, which is bad, because i've got a usb mouse. The lock ups
> start at about 1 second, and increase in length exponentially, until the
> system locks indefinatly.
My laptop is running fine with a usb mouse attached, but it is not a SMP system.
If I get a chance I'll slap a usb mouse on one of the smp systems, see what
happens.
> Has anyone tried GCC 2.97?
2.96.x wouldn't even compile the kernel.
I doubt 2.97 will have much better luck.
Since most recommendations for kernel compilation seem to suggest using 2.91.66
I doubt we will spend much time trying to beat the code in submission.
Especially since other parts of the kernel may not work.
>
>
> On Mon, 23 Oct 2000, kris buggenhout wrote:
>
> > Keith Owens wrote:
> >
> > >
> > > The USB hang is probably SMP related, there is still a lot of work
> > > going on the USB source. Can you compile the kernel without USB or
> > > make all of USB a module instead of builtin?
> >
> > strange... I have done this on a Dell Poweredge 6400 with 4CPU's 4GB ram and
> > had no problem whatsoever...
> > I used the rpm's for the beta as well as the kernel recompile.
> >
> > It even had no "normal" ide or scsi disks, only a hardware raid controler...
> > And that works too ... :)
> >
> > I compiled with the most recent binutils and the gcc that comes with RH 6.2
> > (
> > on of the condemned versions)
> > and have a clean compile 99 out of 100.
> >
> >
> > I had similar trouble on an single cpu system... due to a setting in
> > powermanagement wake on USB... that messed up the boot procedure ...
> >
> > ...
> >
> >
> >
--
Russell Cattelan
cattelan@xxxxxxxxxxx
|