Re: kmemleak panic

[Robin Murphy]

On 2019-01-18 3:36 pm, Marc Gonzalez wrote:
> Unable to handle kernel paging request at virtual address ffffffc021e00000

I can't help but notice that you seem to get the same address in 4 
different logs - if it really is that deterministic then that's quite 
the boon for debugging (FWIW my first thought is that it looks a lot 
like a phys_to_virt() of something bogus). I've never used kmemleak 
myself, but looking at where it's crashing it appears to think that that 
address is a valid object - if I'm reading the docs correctly then I 
guess the "dump" command ought to be able to tell you where it thinks 
that object came from.

Robin.