diff options
| author |   Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2015-10-01 10:26:24 -0700 |
|---|---|---|
| committer | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2015-12-04 12:26:52 -0800 |
| commit | df5bd5144a80a9f6c3807383b11f735dae9caf9d (patch) | |
| tree | fce44d0970a70446c5fd6b50f2e764db0efb8e56 /ipc | |
| parent | rcu: Invert sync_rcu_exp_select_cpus() "if" statement (diff) | |
| download | linux-dev-df5bd5144a80a9f6c3807383b11f735dae9caf9d.tar.xz linux-dev-df5bd5144a80a9f6c3807383b11f735dae9caf9d.zip | |
rcu: Reduce expedited GP memory contention via per-CPU variables
Currently, the piggybacked-work checks carried out by sync_exp_work_done()
atomically increment a small set of variables (the ->expedited_workdone0,
->expedited_workdone1, ->expedited_workdone2, ->expedited_workdone3
fields in the rcu_state structure), which will form a memory-contention
bottleneck given a sufficiently large number of CPUs concurrently invoking
either synchronize_rcu_expedited() or synchronize_sched_expedited().
This commit therefore moves these for fields to the per-CPU rcu_data
structure, eliminating the memory contention. The show_rcuexp() function
also changes to sum up each field in the rcu_data structures.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Diffstat (limited to 'ipc')
0 files changed, 0 insertions, 0 deletions