5.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Joel Savitz <jsavitz@xxxxxxxxxx>
commit 8d98e42fb20c25e8efdab4cc1ac46d52ba964aca upstream.
commit 5ef64cc8987a ("mm: allow a controlled amount of unfairness in the
page lock") introduced a new systctl but no accompanying documentation.
Add a simple entry to the documentation.
Link: https://lkml.kernel.org/r/20220325164437.120246-1-jsavitz@xxxxxxxxxx
Signed-off-by: Joel Savitz <jsavitz@xxxxxxxxxx>
Cc: Jonathan Corbet <corbet@xxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Suren Baghdasaryan <surenb@xxxxxxxxxx>
Cc: Mike Rapoport <rppt@xxxxxxxxxx>
Cc: "zhangyi (F)" <yi.zhang@xxxxxxxxxx>
Cc: Charan Teja Reddy <charante@xxxxxxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
Documentation/admin-guide/sysctl/vm.rst | 9 +++++++++
1 file changed, 9 insertions(+)
--- a/Documentation/admin-guide/sysctl/vm.rst
+++ b/Documentation/admin-guide/sysctl/vm.rst
@@ -61,6 +61,7 @@ Currently, these files are in /proc/sys/
- overcommit_memory
- overcommit_ratio
- page-cluster
+- page_lock_unfairness
- panic_on_oom
- percpu_pagelist_fraction
- stat_interval
@@ -741,6 +742,14 @@ extra faults and I/O delays for followin
that consecutive pages readahead would have brought in.
+page_lock_unfairness
+====================
+
+This value determines the number of times that the page lock can be
+stolen from under a waiter. After the lock is stolen the number of times
+specified in this file (default is 5), the "fair lock handoff" semantics
+will apply, and the waiter will only be awakened if the lock can be taken.
+
panic_on_oom
============
