On Thu, Sep 23, 2021 at 02:10:13PM -0700, Hugh Dickins wrote: > But you have a different point in mind when you refer to superfluous > CoW and GUP: I don't know the score there (and I think we are still in > that halfway zone, since pte CoW was changed to depend on page_count, > but THP CoW still depending on mapcount). I didn't pay enough attention to the topic when the change to depend on page_count was made. I need to catch up. I look at what direction Andrea went in his patchset and so far I *feel* he has a point[1]. I have not read the whole thing yet and I don't have a firm position here, but maybe we need to get to the bottom of the topic before considering ditching per-subpage mapcount. [1] https://git.kernel.org/pub/scm/linux/kernel/git/andrea/aa.git/tree/mm/memory.c?h=mapcount_deshare&id=e1cb3108d4131c2a7da03fbd37c3230cf082bfd9#n3153 -- Kirill A. Shutemov
