[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <df2597f6-af21-5547-d39c-94c02ad17adb@linux.alibaba.com>
Date: Thu, 9 Jul 2020 23:11:11 +0800
From: Alex Shi <alex.shi@...ux.alibaba.com>
To: Johannes Weiner <hannes@...xchg.org>,
"Kirill A. Shutemov" <kirill@...temov.name>,
Matthew Wilcox <willy@...radead.org>,
Linux-MM <linux-mm@...ck.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Hugh Dickins <hughd@...gle.com>
Subject: a question of split_huge_page
Hi Kirill & Matthew,
In the func call chain, from split_huge_page() to lru_add_page_tail(),
Seems tail pages are added to lru list at line 963, but in this scenario
the head page has no lru bit and isn't set the bit later. Why we do this?
or do I miss sth?
Many Thanks
Alex
938 void lru_add_page_tail(struct page *page, struct page *page_tail,
939 struct lruvec *lruvec, struct list_head *list)
940 {
941 VM_BUG_ON_PAGE(!PageHead(page), page);
942 VM_BUG_ON_PAGE(PageCompound(page_tail), page);
943 VM_BUG_ON_PAGE(PageLRU(page_tail), page);
944 lockdep_assert_held(&lruvec_pgdat(lruvec)->lru_lock);
945
946 if (!list)
947 SetPageLRU(page_tail);
948
949 if (likely(PageLRU(page)))
950 list_add_tail(&page_tail->lru, &page->lru);
951 else if (list) {
952 /* page reclaim is reclaiming a huge page */
953 get_page(page_tail);
954 list_add_tail(&page_tail->lru, list);
955 } else {
956 /*
957 * Head page has not yet been counted, as an hpage,
958 * so we must account for each subpage individually.
959 *
960 * Put page_tail on the list at the correct position
961 * so they all end up in order.
962 */
963 add_page_to_lru_list_tail(page_tail, lruvec,
964 page_lru(page_tail));
965 }
966 }
Powered by blists - more mailing lists