---

yaml
---
r: 221724
b: refs/heads/master
c: 203f40a
h: refs/heads/master
v: v3

[refs]

@@ -1,2 +1,2 @@
 ---
-refs/heads/master: 1dce071e18b7264457d17c0dec4c7e430bfaee7d
+refs/heads/master: 203f40a5a030ed4048cd40e3bd9ab5df6c5df589

trunk/Documentation/ABI/obsolete/proc-pid-oom_adj

@@ -0,0 +1,22 @@
+What:	/proc/<pid>/oom_adj
+When:	August 2012
+Why:	/proc/<pid>/oom_adj allows userspace to influence the oom killer's
+	badness heuristic used to determine which task to kill when the kernel
+	is out of memory.
+
+	The badness heuristic has since been rewritten since the introduction of
+	this tunable such that its meaning is deprecated.  The value was
+	implemented as a bitshift on a score generated by the badness()
+	function that did not have any precise units of measure.  With the
+	rewrite, the score is given as a proportion of available memory to the
+	task allocating pages, so using a bitshift which grows the score
+	exponentially is, thus, impossible to tune with fine granularity.
+
+	A much more powerful interface, /proc/<pid>/oom_score_adj, was
+	introduced with the oom killer rewrite that allows users to increase or
+	decrease the badness() score linearly.  This interface will replace
+	/proc/<pid>/oom_adj.
+
+	A warning will be emitted to the kernel log if an application uses this
+	deprecated interface.  After it is printed once, future warnings will be
+	suppressed until the kernel is rebooted.