Re: [PATCH v4 00/12] mm: make userland page table freeing RCU-safe

[Vlastimil Babka (SUSE)]

On 9/23/26 11:15, David Hildenbrand (Arm) wrote:
> On 9/23/26 11:04, Lorenzo Stoakes (ARM) wrote:
>> On Wed, Sep 23, 2026 at 10:45:33AM +0200, David Hildenbrand (Arm) wrote:
>>> On 9/23/26 01:00, Andrew Morton wrote:
>>>>
>>>>
>>>> Do we have reason to believe that anyone is seeing any effects from
>>>> their arch's failure to rcu-free the page tables?
>>>>
>>>> This sounds like a rather long-standing wart, yes?
>>>>
>>>> The patchset is thinly reviewed and mm.git is bursting at the seams
>>>> again.  I'll play wait-and-see on this series.
>>> Do we want to nudge arch maintainers to review?
>> 
>> (Just to note that I replied separately, 'thinly reviewed' is incorrect, it's
>> been in -next for weeks and the whole thing was reviewed by Kiryl.)
> 
> Ack.
> 
>> 
>> The only places where that would be useful is m68k motorola, sparc32.
>> 
>> But in any case it has been heavily tested including on every single arch
>> listed with stress testing performed (I think I cover that in the cover
>> letter).
>> 
>> The changes are not controversial so I don't think that's a blocker, but
>> it'd be nice?
> 
> I doubt we'll ever get sparc32 reviews :)
> 
> I'd definitely appreciate some review on the arch bits from arch maintainers,
> but if they don't reply within the next weeks I wouldn't wait on that.

Indeed, for minor arches like this the lack of review (it would be a whole
cycle by the merge window) shouldn't be a blocker for important tree-wide
progress.