diff options
| author | Liu Shixin <liushixin2@huawei.com> | 2022-09-09 04:16:53 +0200 |
|---|---|---|
| committer | Andrew Morton <akpm@linux-foundation.org> | 2022-10-03 23:03:08 +0200 |
| commit | f4981502088f8ea704beeedf3470e1d53bc2e46c (patch) | |
| tree | 01f5ad21737aabffcb9948a10fdf9aa5f4123c76 /mm/huge_memory.c | |
| parent | writeback: remove unused macro DIRTY_FULL_SCOPE (diff) | |
| download | linux-f4981502088f8ea704beeedf3470e1d53bc2e46c.tar.xz linux-f4981502088f8ea704beeedf3470e1d53bc2e46c.zip | |
mm/huge_memory: prevent THP_ZERO_PAGE_ALLOC increased twice
A user who reads THP_ZERO_PAGE_ALLOC may be more concerned about the huge
zero pages that are really allocated for thp. It is misleading to
increase THP_ZERO_PAGE_ALLOC twice if two threads call get_huge_zero_page
concurrently. Don't increase the value if the huge page is not really
used.
Update Documentation/admin-guide/mm/transhuge.rst to suit.
Link: https://lkml.kernel.org/r/20220909021653.3371879-1-liushixin2@huawei.com
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Diffstat (limited to 'mm/huge_memory.c')
| -rw-r--r-- | mm/huge_memory.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 36ef79b85195..4938defe4e73 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -163,7 +163,6 @@ retry: count_vm_event(THP_ZERO_PAGE_ALLOC_FAILED); return false; } - count_vm_event(THP_ZERO_PAGE_ALLOC); preempt_disable(); if (cmpxchg(&huge_zero_page, NULL, zero_page)) { preempt_enable(); @@ -175,6 +174,7 @@ retry: /* We take additional reference here. It will be put back by shrinker */ atomic_set(&huge_zero_refcount, 2); preempt_enable(); + count_vm_event(THP_ZERO_PAGE_ALLOC); return true; } |