summaryrefslogtreecommitdiff
path: root/scripts/generate_rust_analyzer.py
diff options
context:
space:
mode:
authorPaul E. McKenney <paulmck@kernel.org>2024-11-06 15:42:06 -0800
committerUladzislau Rezki (Sony) <urezki@gmail.com>2024-12-14 17:05:27 +0100
commit282e06cc8f595e999f778bd284d6a3608f7d1d62 (patch)
tree36c9257bff40e21a165903bf1d55d72a6eca785a /scripts/generate_rust_analyzer.py
parentcae7f6319e6451dc49e4239652c6c7b6f54008d2 (diff)
rcutorture: Add parameters to control polled/conditional wait interval
This commit adds rcutorture module parameters gp_cond_wi, gp_cond_wi_exp, gp_poll_wi, and gp_poll_wi_exp to control the wait interval for conditional, conditional expedited, polled, and polled expedited grace periods, respectively. When rcu_torture_writer() is testing these types of grace periods, hrtimers are used to randomly wait up to the specified number of microseconds, but with nanosecond granularity. In the case of conditional grace periods (get_state_synchronize_rcu() and cond_synchronize_rcu(), for example) there is just one wait. For polled grace periods (start_poll_synchronize_rcu() and poll_state_synchronize_rcu(), for example), there is a repeated series of waits until the grace period ends. For normal grace periods, the default is 16 jiffies (for example, 16,000 microseconds on a HZ=1000 system) and for expedited grace periods the default is 128 microseconds. Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Diffstat (limited to 'scripts/generate_rust_analyzer.py')
0 files changed, 0 insertions, 0 deletions