On Wed, Dec 11, 2019 at 01:08:32PM -0700, Jens Axboe wrote: > > The fact that you see that xas_create() so prominently would imply > > perhaps add_to_swap_cache(), which certainly implies that the page > > shrinking isn't hitting the file pages... > > That's presumably misleading, as it's just lookups. But yes, > confusing... While xas_create() could be called directly, it's more often called through xas_store() ... which would be called if we're storing a shadow entry to replace a page, which this workload would presumably be doing.
