On Thu, 7 May 2020 at 14:47, Linus Walleij [off-list ref] wrote:

It's time to repost this, now that I got it into better shape.

Main changes:
- Restore the series number to v8 following my previous
  "v1" which was actually v7, following Florians versions.
  What was I thinking. :(
- Doesn't crash randomly in QEMU etc. This was because
  mmu.c was instrumented, and when we were lucky this did
  not poke around in memory that crashed us.
- Handle the undefined symbols better (like suggested by
  Ard)
- Fix up documentation.
- Better patch split in logical chunks.

Remaining issues:
- Does it work for you?
- Crash on ARMv5/v5 in v4wbi_flush_user_tlb_range+0x10/0x4c.
  I wonder why. If you know, please tell me, I naturally
  have to fix this.
- The code in kasan_init.c to first back up TTBR0 and then
  restore it looks *fragile* even if it is working. I think
  we need to do something better in a per-proc file such
  as is done in cpu_v7_switch_mm() and other well organized
  MMU manipulation, I certainly feel this should be done
  like that.

Abbott Liu (1):
  ARM: Define the virtual space of KASan's shadow region

Andrey Ryabinin (4):
  ARM: Disable KASan instrumentation for some code
  ARM: Replace string mem* functions for KASan
  ARM: Initialize the mapping of KASan shadow memory
  ARM: Enable KASan for ARM

For the series,

Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Tested-by: Ard Biesheuvel <ardb@kernel.org> # QEMU/KVM/mach-virt/LPAE/8G

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel