-next boot failures during KVM setup

[Mark Brown <broonie@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 1413 bytes --]

I'm seeing boot failures on a range of physical arm64 platforms in
today's -next.  Turning on earlycon it looks like we're getting bad
pointer dereferences during KVM initialisation:

[    0.728923] kvm [1]: nv: 570 coarse grained trap handlers
[    0.735138] kvm [1]: nv: 710 fine grained trap handlers
[    0.741326] kvm [1]: IPA Size Limit: 40 bits
[    0.748840] Unable to handle kernel paging request at virtual address ffff00000478e000
[    0.757027] Mem abort info:
[    0.759917]   ESR = 0x0000000096000147
[    0.763772]   EC = 0x25: DABT (current EL), IL = 32 bits
[    0.851526] pc : dcache_clean_inval_poc+0x24/0x48
[    0.856367] lr : kvm_arm_init+0xbb0/0x13f0
...

[    0.937120] Call trace:
[    0.939628]  dcache_clean_inval_poc+0x24/0x48 (P)
[    0.944457]  do_one_initcall+0x60/0x1d4
[    0.948393]  kernel_init_freeable+0x250/0x2d8

   https://lava.sirena.org.uk/scheduler/job/2849583#L848

(with other platforms I've got earlycon logs showing basically the same
thing).  I have some bisects but they seem to have been confused by
earlier driver core issues, I've tweaked to try to avoid that and am
retrying.  FVP and qemu seem unaffected:

  https://lava.sirena.org.uk/scheduler/job/2848374#L888
  https://lava.sirena.org.uk/scheduler/job/2848966#L447

The affected platforms thus far are all SMP Cortex A53/5 systems, but
that's the vast majority of my lab.  They have both GICv3 and GICv2.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]