On Wed, Nov 30, 2016 at 10:56:57PM -0800, Guenter Roeck wrote:
Hi Paul,
On Wed, Nov 30, 2016 at 05:19:50PM -0800, Paul E. McKenney wrote:
[ ... ]
[ ... ]
BUG: sleeping function called from invalid context at mm/page_alloc.c:3775
Whew! You had me going for a bit there. ;-)
Bisect results are here ... the culprit is, again, commit 2d66cccd73 ("mm:
Prevent __alloc_pages_nodemask() RCU CPU stall warnings"), and reverting that
patch fixes the problem. Good that you dropped it already :-).
"My work is done." ;-)
And apologies for the hassle. I have no idea what I was thinking when
I put the cond_resched_rcu_qs() there!