Re: [PATCH v5 02/11] powerpc/mm: Adds counting method to monitor lockless pgtable walks

[Christopher Lameter]

On Tue, 8 Oct 2019, Leonardo Bras wrote:

> So you say that the performance impact of using my approach is the same
> as using locks? (supposing that lock never waits)
>
> So, there are 'lockless pagetable walks' only for the sake of better
> performance?

I thought that was the major motivation here.

> I thought they existed to enable doing pagetable walks in states where
> locking was not safe.

That sounds profoundly dangerous. Usually locking makes things safe to
access. Accesses without locking require a lot of care.