mirror of
https://github.com/torvalds/linux.git
synced 2026-05-30 18:13:41 +02:00
mm/page_alloc: avoid overcounting bulk alloc in watermark check
alloc_pages_bulk_noprof() only fills NULL slots and already tracks how many entries are pre-populated via nr_populated. The fast watermark check was adding nr_pages unconditionally, which can overestimate the demand. Use (nr_pages - nr_populated) instead, as an upper bound on the remaining pages this call can still allocate without scanning the whole array. Link: https://lkml.kernel.org/r/tencent_F36C5B5FB4DED98C79D9BDEE1210CD338C06@qq.com Signed-off-by: Shengming Hu <hu.shengming@zte.com.cn> Reviewed-by: Vlastimil Babka <vbabka@suse.cz> Cc: Brendan Jackman <jackmanb@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Zi Yan <ziy@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
parent
396f57b572
commit
7e74dd0316
|
|
@ -5136,7 +5136,7 @@ unsigned long alloc_pages_bulk_noprof(gfp_t gfp, int preferred_nid,
|
|||
|
||||
cond_accept_memory(zone, 0, alloc_flags);
|
||||
retry_this_zone:
|
||||
mark = wmark_pages(zone, alloc_flags & ALLOC_WMARK_MASK) + nr_pages;
|
||||
mark = wmark_pages(zone, alloc_flags & ALLOC_WMARK_MASK) + nr_pages - nr_populated;
|
||||
if (zone_watermark_fast(zone, 0, mark,
|
||||
zonelist_zone_idx(ac.preferred_zoneref),
|
||||
alloc_flags, gfp)) {
|
||||
|
|
|
|||
Loading…
Reference in New Issue
Block a user