Re: [PATCH 4.4 07/40] arm64: ensure __dump_instr() checks addr_limit

From: Mark Rutland
Date: Mon Nov 20 2017 - 06:26:32 EST


On Sun, Nov 19, 2017 at 11:29:35AM +0100, Greg Kroah-Hartman wrote:
> On Tue, Nov 14, 2017 at 04:18:23PM +0000, Mark Rutland wrote:
> > I agree that we need both patches. On its own (in the absence of PAN/UAO),
> > 57f4959bad0a154a wouldn't have any effect, but it is critical to ensure that we
> > can dump kernel instructions when we intend to, with 7a7003b1da010d2b
> > backported.
> >
> > Backporting 57f4959bad0a154a as a prerequisite makes sense to me.
>
> But that is a "new feature", are you sure about this?

Sorry, I copied the wrong commit ID when replying. I meant to say that
we should backport commit:

c5cea06be060f38e ("arm64: fix dump_instr when PAN and UAO are in use")

... which on its own does nothing, but will prevent 7a7003b1da010d2b
from breaking kernel dumps.

We should *not* backport commit:

57f4959bad0a154a ("arm64: kernel: Add support for User Access Override")

> So, just apply that one patch and we are all ok? Nothing else?

Just apply c5cea06be060f38e as a prerequisite for 7a7003b1da010d2b, and
I think we're good. I'm not aware of anything else we need.

> confused,

Sorry about that. Hopefully the above clarified things?

Thanks,
Mark.