linux/kernel/rcu/tree.c, branch v3.19

Merge branches 'torture.2014.11.03a', 'cpu.2014.11.03a', 'doc.2014.11.13a', 'fixes.2014.11.13a', 'signal.2014.10.29a' and 'rt.2014.10.29a' into HEAD

2014-11-13T18:39:04Z

cpu.2014.11.03a: Changes for per-CPU variables. doc.2014.11.13a: Documentation updates. fixes.2014.11.13a: Miscellaneous fixes. signal.2014.10.29a: Signal changes. rt.2014.10.29a: Real-time changes. torture.2014.11.03a: torture-test changes.

rcutorture: Add early boot self tests

2014-11-04T03:26:37Z

Add early boot self tests for RCU under CONFIG_PROVE_RCU. Currently the only test is adding a dummy callback which increments a counter which we then later verify after calling rcu_barrier*(). Signed-off-by: Pranith Kumar Signed-off-by: Paul E. McKenney

rcu: Remove "cpu" argument to rcu_cleanup_after_idle()

2014-11-04T03:20:56Z

The "cpu" argument to rcu_cleanup_after_idle() is always the current CPU, so drop it. This moves the smp_processor_id() from the caller to rcu_cleanup_after_idle(), saving argument-passing overhead. Again, the anticipated cross-CPU uses of these functions has been replaced by NO_HZ_FULL. Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Remove "cpu" argument to rcu_prepare_for_idle()

2014-11-04T03:20:49Z

The "cpu" argument to rcu_prepare_for_idle() is always the current CPU, so drop it. This in turn allows two of the uses of "cpu" in this function to be replaced with a this_cpu_ptr() and the third by smp_processor_id(), replacing that of the call to rcu_prepare_for_idle(). Again, the anticipated cross-CPU uses of these functions has been replaced by NO_HZ_FULL. Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Remove "cpu" argument to rcu_needs_cpu()

2014-11-04T03:20:43Z

The "cpu" argument to rcu_needs_cpu() is always the current CPU, so drop it. This in turn allows the "cpu" argument to rcu_cpu_has_callbacks() to be removed, which allows the uses of "cpu" in both functions to be replaced with a this_cpu_ptr(). Again, the anticipated cross-CPU uses of these functions has been replaced by NO_HZ_FULL. Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Remove "cpu" argument to rcu_note_context_switch()

2014-11-04T03:20:34Z

The "cpu" argument to rcu_note_context_switch() is always the current CPU, so drop it. This in turn allows the "cpu" argument to rcu_preempt_note_context_switch() to be removed, which allows the sole use of "cpu" in both functions to be replaced with a this_cpu_ptr(). Again, the anticipated cross-CPU uses of these functions has been replaced by NO_HZ_FULL. Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Remove "cpu" argument to rcu_preempt_check_callbacks()

2014-11-04T03:20:26Z

Because rcu_preempt_check_callbacks()'s argument is guaranteed to always be the current CPU, drop the argument and replace per_cpu() with __this_cpu_read(). Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Remove "cpu" argument to rcu_pending()

2014-11-04T03:20:18Z

Because rcu_pending()'s argument is guaranteed to always be the current CPU, drop the argument and replace per_cpu_ptr() with this_cpu_ptr(). Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Remove "cpu" argument to rcu_check_callbacks()

2014-11-04T03:20:11Z

The "cpu" argument was kept around on the off-chance that RCU might offload scheduler-clock interrupts. However, this offload approach has been replaced by NO_HZ_FULL, which offloads -all- RCU processing from qualifying CPUs. It is therefore time to remove the "cpu" argument to rcu_check_callbacks(), which this commit does. Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar

rcu: Use DEFINE_PER_CPU_SHARED_ALIGNED for rcu_data

2014-11-04T03:20:03Z

The rcu_data per-CPU variable has a number of fields that are atomically manipulated, potentially by any CPU. This situation can result in false sharing with per-CPU variables that have the misfortune of being allocated adjacent to rcu_data in memory. This commit therefore changes the DEFINE_PER_CPU() to DEFINE_PER_CPU_SHARED_ALIGNED() in order to avoid this false sharing. Reported-by: Christoph Lameter Signed-off-by: Paul E. McKenney Reviewed-by: Christoph Lameter Reviewed-by: Pranith Kumar