Re: [PATCH] [RFC] mm: add kmem_cache_create_rcu()

[Vlastimil Babka <vbabka@suse.cz>]

On 8/27/24 17:45, Jens Axboe wrote:
> On 8/27/24 9:33 AM, Jann Horn wrote:
> 
> Yep should work.
> 
> I like this approach as a way of unifying the various ways of creating a
> slab cache. Particularly as I'd want to use a free ptr myself for an
> in-kernel use case that currently creates a cache with
> kmem_cache_create_usercopy().

Ah right so you need effectively a combination of the current
kmem_cache_create_usercopy() and kmem_cache_create_rcu().

>> But I guess if we want to allow leaving out the freeptr_offset
>> parameter, we'd have to have a flag to say whether the freeptr_offset
>> parameter value should be used... Maybe my proposal is too overly
>> fancy...

I think it's a good way. The extra flag to denote the custom freepointer
offset is valid is a bit annoying, but there shouldn't be too many users of
that, and implicit zero-initializing is a nice thing otherwise.

> 
> I think we'll end up going down that route eventually anyway to cover
> all cases, so may as well do a prep patch that does that first. At least
> that's my opinion...
>