Re: [RFC PATCH v2] UML: add support for KASAN under x86_64

[Johannes Berg]

On Thu, 2020-02-13 at 09:44 +0100, Dmitry Vyukov wrote:

> > Right, but again like below - that's just mapped, not actually used. But
> > as far as I can tell, once you actually start running and potentially
> > use all of your mem=1024 (MB), you'll actually also use another 128MB on
> > the KASAN shadow, right?
> > 
> > Unlike, say, a real x86_64 machine where if you just have 1024 MB
> > physical memory, the KASAN shadow will have to fit into that as well.
> 
> Depends on what you mean by "real" :)

:)

> Real user-space ASAN will also reserve 1/8th of 47-bit VA on start
> (16TB).

Ah, but I was thinking of actual memory *used*, not just VA.

And of KASAN, not user-space, but yeah, good point.

> This implementation seems to be much closer to user-space ASAN
> rather than to x86_64 KASAN (in particular it seems to be mostly
> portable across archs and is not really x86-specific, which is good).

Indeed.

> I think it's reasonable and good, but the implementation difference
> with other kernel arches may be worth noting somewhere in comments.

Right, I guess that's the broader point. I was thinking mostly of the
memory consumption: if you run with UML KASAN, your UML virtual machine
will use around 12.5% more memory than before, unlike if you say have a
KVM virtual machine - whatever you reserve outside will be what it can
use inside, regardless of KASAN being enabled or not.

This is totally fine, I just thought it should be documented somewhere,
perhaps in the Kconfig option, though I guess there isn't a UML specific
one for this... Not sure where then.

johannes