Re: [PATCH v2] mm: Make ksize() a reporting-only function

[Kees Cook]

On Fri, Nov 18, 2022 at 11:32:36AM +0100, Vlastimil Babka wrote:
> On 11/18/22 04:56, Kees Cook wrote:
> > With all "silently resizing" callers of ksize() refactored, remove the
> 
> At cursory look seems it's true now in -next (but not mainline?) can you
> confirm?

Almost, yes. I realized there is 1 case in the BPF verifier that
remains. (I thought it was picked up, but only a prereq patch was.) I'm
going to resend that one today, but I would expect it to be picked
up soon. (But, yes, definitely not for mainline.)

> That would probably be safe enough to have slab.git expose this to -next now
> and time a PR appropriately in the next merge window?

Possibly. I suspect syzkaller might trip KASAN on any larger BPF tests
until I get the last one landed. And if you don't want to do the timing
of the PR, I can carry this patch in my hardening tree, since I already
have to do a two-part early/late-merge-window PR there.

-- 
Kees Cook