Crash after 'reboot' due to 9be4fd2c7723a

[Russell King - ARM Linux]

On Mon, May 23, 2016 at 06:38:31PM -0300, Fabio Estevam wrote:
> On Mon, May 23, 2016 at 6:07 PM, Russell King - ARM Linux
> <linux at armlinux.org.uk> wrote:
> 
> > Read the code again.  There's three states:
> >
> > 1. CONFIG_SMP=n - it always returns false.
> > 2. CONFIG_SMP=y, CONFIG_SMP_ON_UP=n - it always returns true.  Such a
> >    kernel can _only_ be run on SMP capable systems.
> > 3. CONFIG_SMP=y, CONFIG_SMP_ON_UP=y - it returns depending on whether
> >    we detect that the CPU is SMP capable.
> 
> Yes, I meant that on imx6ul we have option 2 above with imx_v6_v7_defconfig.

I don't think so, unless you specifically turn off CONFIG_SMP_ON_UP.
It defaults to 'y', so imx_v6_v7_defconfig should result in this being
'y'.

> Also tested option 3 and is_smp() returns true as well.

So that means the MPIDR register is telling the kernel...

* The boot CPU has multiprocessor extensions.
* The boot CPU is not part of a uniprocessor system.

> I will try to see why GIC on this SoC cannot work in SMP.

-- 
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line: currently at 9.6Mbps down 400kbps up
according to speedtest.net.