Hello!

This series contains miscellaneous fixes:

1.      Don't invoke try_invoke_on_locked_down_task() with irqs disabled.

2.      Move x86's rcu_cpu_starting() earlier.

3.      Panic after fixed number of stalls.

4.      Update list.h comment to explicitly note circular lists.

5.      Implement rcu_segcblist_is_offloaded() config dependent.

6.      Fix single-CPU check in rcu_blocking_is_gp().

7.      Clarify nocb kthreads naming in RCU_NOCB_CPU config.

8.      Add a warning if CPU being onlined did not report QS already.

9.      Make struct kernel_param_ops definitions const.

10.     Fix ftrace recursion.

11.     nocb: Avoid raising softirq for offloaded ready-to-execute CBs.

12.     Prevent lockdep-RCU splats on lock acquisition/release.

13.     Fix a typo in rcu_blocking_is_gp() header comment.

14.     Do not report strict GPs for outgoing CPUs.

15.     Defer kvfree_rcu() allocation to a clean context.

16.     Take early exit on memory-allocation failure.

                                                Thanx, Paul

------------------------------------------------------------------------

 arch/x86/kernel/cpu/mtrr/mtrr.c |    2 
 arch/x86/kernel/smpboot.c       |    1 
 include/linux/kernel.h          |    1 
 include/linux/list.h            |    2 
 kernel/rcu/Kconfig              |   20 ++--
 kernel/rcu/rcu_segcblist.h      |    2 
 kernel/rcu/srcutree.c           |    4 
 kernel/rcu/tree.c               |  188 +++++++++++++++++++++++++---------------
 kernel/rcu/tree.h               |    2 
 kernel/rcu/tree_plugin.h        |    2 
 kernel/rcu/tree_stall.h         |   28 ++++-
 kernel/sysctl.c                 |   11 ++
 12 files changed, 178 insertions(+), 85 deletions(-)

Reply via email to