Subject : [3.0.0-rc6-git7] possible recursive locking at cache_alloc_refill Submitter : Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Date : 2011-07-11 12:37 Message-ID : 201107112137.FAD00545.SHtLOFOJOMFQFV@I-love.SAKURA.ne.jp References : http://marc.info/?l=linux-kernel&m=131038788902151&w=2 This entry is being used for tracking a regression from 2.6.39. Please don't close it until the problem is fixed in the mainline.
merged in v3.1-rc1: commit 30765b92ada267c5395fc788623cb15233276f5c Author: Peter Zijlstra <peterz@infradead.org> Date: Thu Jul 28 23:22:56 2011 +0200 slab, lockdep: Annotate the locks before using them commit 83835b3d9aec8e9f666d8223d8a386814f756266 Author: Peter Zijlstra <peterz@infradead.org> Date: Fri Jul 22 15:26:05 2011 +0200 slab, lockdep: Annotate slab -> rcu -> debug_object -> slab
On Monday, August 15, 2011, Tetsuo Handa wrote: > Pekka Enberg wrote: > > Tetsuio, did Peter's commit 30765b9 ("slab, lockdep: Annotate the locks > > before using them") and commit 83835b3 ("slab, lockdep: Annotate slab -> > > rcu -> debug_object -> slab") in Linus' git fix this for you? > > Yes. Applying these two commits fixes this problem.
(In reply to comment #2) > On Monday, August 15, 2011, Tetsuo Handa wrote: > > Pekka Enberg wrote: > > > Tetsuio, did Peter's commit 30765b9 ("slab, lockdep: Annotate the locks > > > before using them") and commit 83835b3 ("slab, lockdep: Annotate slab -> > > > rcu -> debug_object -> slab") in Linus' git fix this for you? > > > > Yes. Applying these two commits fixes this problem. But probably not in all cases.. I have a machine in which 3.1-rc3 still shows this issue. Attached are the dmesg/.config dheeraj
Created attachment 69742 [details] dmesg of the log
Created attachment 69752 [details] The respective .config