mm/page_owner: record single timestamp value for high order allocations

When allocating a high-order page, separate allocation timestamp is
recorded for each sub-page resulting in different timestamp values between
them.

This behavior is not consistent with the behavior when recording free
timestamp and caused confusion when analyzing memory dumps.  Record single
timestamp for the entire allocation, aligning with the behavior for free
timestamps.

Link: https://lkml.kernel.org/r/20230121165054.520507-1-42.hyeyoo@gmail.com
Signed-off-by: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/page_owner.c b/mm/page_owner.c
index f0553bedb39de38324057c952e05c7a3cdc1b783..80dc8f4050faa6f1a6da2f87e8a4bafbc7266a24 100644 (file)
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -163,6 +163,7 @@ static inline void __set_page_owner_handle(struct page_ext *page_ext,
 {
        struct page_owner *page_owner;
        int i;
+       u64 ts_nsec = local_clock();
 
        for (i = 0; i < (1 << order); i++) {
                page_owner = get_page_owner(page_ext);
@@ -172,7 +173,7 @@ static inline void __set_page_owner_handle(struct page_ext *page_ext,
                page_owner->last_migrate_reason = -1;
                page_owner->pid = current->pid;
                page_owner->tgid = current->tgid;
-               page_owner->ts_nsec = local_clock();
+               page_owner->ts_nsec = ts_nsec;
                strscpy(page_owner->comm, current->comm,
                        sizeof(page_owner->comm));
                __set_bit(PAGE_EXT_OWNER, &page_ext->flags);