[PATCH 00/10] arm64: move thread_info off of the task stack

[Mark Rutland]

On Mon, Oct 24, 2016 at 11:18:35AM -0700, Kees Cook wrote:
> On Mon, Oct 24, 2016 at 11:15 AM, Mark Rutland <mark.rutland at arm.com> wrote:
> > On Mon, Oct 24, 2016 at 07:09:42PM +0100, Mark Rutland wrote:
> >> It's really crazy how broken a kernel can be yet still "work"; clearly
> >> we better tests are needed. :/
> >
> > Clearly we better grammar need too. :(
> 
> Out of curiosity, what workflow would have tripped over the entry.S bug?

There are two bugs:

The issues in [1] would show up if you were attempting to use
breakpoints or watchpoints -- we'd never disable the single step.

The broken 're-entered irq stack' check [2] would be an issue if we were
close to exhausting the stack -- we'd never switch to the IRQ stack when
we take an IRQ in a kernel context. I'm not sure of a particular
workload.

Thanks,
Mark.

[1] http://lists.infradead.org/pipermail/linux-arm-kernel/2016-October/462932.html
[2] http://lists.infradead.org/pipermail/linux-arm-kernel/2016-October/462891.html