rcu/kvfree: Do not run a page work if a cache is disabled
Uladzislau Rezki (Sony)
maintainerPATCH 1/1 rcu/kvfree: Do not run a page work if a cache is disabled
By default the cache size is 5 pages per-cpu. But it can be disabled at boot time by setting the rcu_min_cached_objs to zero.
--- kernel/rcu/tree.c | 4 ++++ 1 file changed, 4 insertions(+)
kernel/rcu/tree.c +4 -0
@@ -3246,6 +3246,10 @@ static void fill_page_cache_func(struct work_struct *work)static void run_page_cache_worker (struct kfree_rcu_cpu *krcp) { +// If cache disabled, bail out. +if (!rcu_min_cached_objs) +return ; +if (rcu_scheduler_active ==RCU_SCHEDULER_RUNNING && !atomic_xchg (&krcp->work_in_progress,1 )) {if (atomic_read (&krcp->backoff_page_cache_fill)) {
Signature · 1 line
2.30.2
Paul E. McKenney
maintainerQuoted 5 lines · Uladzislau Rezki (Sony)
By default the cache size is 5 pages per-cpu. But it can be disabled at boot time by setting the
rcu_min_cached_objsto zero.Signed-off-by Uladzislau Rezki (Sony) <[email protected]>
That does get rid of a needless hrtimer &c in that case, good!
I have queued this with the usual wordsmithing below, so please check it.
Thanx, Paul
------------------------------------------------------------------------
commit 5e433764beec0134a9a677f399a6e4539eb8870d
Author: Uladzislau Rezki (Sony) <[email protected]> Date: Tue Apr 11 15:13:41 2023 +0200
rcu/kvfree: Do not run a page work if a cache is disabled
By default the cache size is 5 pages per CPU, but it can be disabled at boot time by setting the rcu_min_cached_objs to zero. When that happens, the current code will uselessly set an hrtimer to schedule refilling this cache with zero pages. This commit therefore streamlines this process by simply refusing the set the hrtimer when rcu_min_cached_objs is zero.
kernel/rcu/tree.c +4 -0
@@ -3247,6 +3247,10 @@ static void fill_page_cache_func(struct work_struct *work)static void run_page_cache_worker (struct kfree_rcu_cpu *krcp) { +// If cache disabled, bail out. +if (!rcu_min_cached_objs) +return ; +if (rcu_scheduler_active ==RCU_SCHEDULER_RUNNING && !atomic_xchg (&krcp->work_in_progress,1 )) {if (atomic_read (&krcp->backoff_page_cache_fill)) {