packages: kernel-desktop/bfs357-worker_fix.patch (NEW) - worker fix

[shadzik]

Author: shadzik                      Date: Sun Oct 24 13:30:47 2010 GMT
Module: packages                      Tag: HEAD
---- Log message:
- worker fix

---- Files affected:
packages/kernel-desktop:
   bfs357-worker_fix.patch (NONE -> 1.1)  (NEW)

---- Diffs:

================================================================
Index: packages/kernel-desktop/bfs357-worker_fix.patch
diff -u /dev/null packages/kernel-desktop/bfs357-worker_fix.patch:1.1
--- /dev/null	Sun Oct 24 15:30:47 2010
+++ packages/kernel-desktop/bfs357-worker_fix.patch	Sun Oct 24 15:30:42 2010
@@ -0,0 +1,23 @@
+Seems like some worker function in the wifi code in 2.6.36 hits this bug_on,
+which it shouldn't. It's hard to know if this is caused by BFS per se, or is
+in iwlagn itself. It is, however, harmless to wake up a task on another RQ
+with BFS, so just disable this BUG_ON for now.
+
+-ck
+
+---
+ kernel/sched_bfs.c |    1 -
+ 1 file changed, 1 deletion(-)
+
+Index: linux-2.6.36-bfs/kernel/sched_bfs.c
+===================================================================
+--- linux-2.6.36-bfs.orig/kernel/sched_bfs.c	2010-10-24 21:30:39.164151587 +1100
++++ linux-2.6.36-bfs/kernel/sched_bfs.c	2010-10-24 21:30:54.851780968 +1100
+@@ -1460,7 +1460,6 @@ static void try_to_wake_up_local(struct
+ 	struct rq *rq = task_rq(p);
+ 	bool success = false;
+ 
+-	BUG_ON(rq != this_rq());
+ 	BUG_ON(p == current);
+ 	lockdep_assert_held(&grq.lock);
+ 
================================================================