On Tue, Mar 26, 2024 at 11:35:43AM -0700, Chris Li wrote: > /* > - * When reading into the swapcache, invalidate our entry. The > + * When reading into the swapcache, erase our entry. The > * swapcache can be the authoritative owner of the page and > * its mappings, and the pressure that results from having two > * in-memory copies outweighs any benefits of caching the Not sure why you editorialized this? Your patch doesn't change those semantics, for which "invalidation" of the backing copy is a much better description, given we're also marking the folio dirty again etc. Can you please send a delta fix to undo that? > @@ -1649,8 +1581,12 @@ bool zswap_load(struct folio *folio) > * the fault fails. We remain the primary owner of the entry.) > */ > if (swapcache) > - zswap_rb_erase(&tree->rbroot, entry); > - spin_unlock(&tree->lock); > + entry = xa_erase(tree, offset); > + else > + entry = xa_load(tree, offset); > + > + if (!entry) > + return false; > > if (entry->length) > zswap_decompress(entry, page); Otherwise, looks good to me Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
