On Fri, Jul 30, 2010 at 01:15:14PM +1000, Michael Neuling wrote:
> (adding kexec list to CC)
>
> In message <4c521fd2.4050...@ozlabs.org> you wrote:
> > Michael Neuling wrote:
> > > In message <4c511216.30...@ozlabs.org> you wrote:
> > >> When CPU hotplug is used, some CPUs may be offline at the t
(adding kexec list to CC)
In message <4c521fd2.4050...@ozlabs.org> you wrote:
> Michael Neuling wrote:
> > In message <4c511216.30...@ozlabs.org> you wrote:
> >> When CPU hotplug is used, some CPUs may be offline at the time a kexec is
> >> performed. The subsequent kernel may expect these CPUs t
Michael Neuling wrote:
> In message <4c511216.30...@ozlabs.org> you wrote:
>> When CPU hotplug is used, some CPUs may be offline at the time a kexec is
>> performed. The subsequent kernel may expect these CPUs to be already running
> ,
>> and will declare them stuck. On pseries, there's also a so
In message <4c511216.30...@ozlabs.org> you wrote:
> When CPU hotplug is used, some CPUs may be offline at the time a kexec is
> performed. The subsequent kernel may expect these CPUs to be already running
,
> and will declare them stuck. On pseries, there's also a soft-offline (cede)
> state th
When CPU hotplug is used, some CPUs may be offline at the time a kexec is
performed. The subsequent kernel may expect these CPUs to be already running,
and will declare them stuck. On pseries, there's also a soft-offline (cede)
state that CPUs may be in; this can also cause problems as the kexece
