diff options
| author |   Andrea Righi <arighi@nvidia.com> | 2024-12-28 11:04:11 +0100 |
|---|---|---|
| committer |   Tejun Heo <tj@kernel.org> | 2024-12-29 12:44:15 -1000 |
| commit | 02f034dcbf3dcb0989e638fdc00d10984dc2278b (patch) | |
| tree | 681514f7988f56f9df84a7b1ca8311b139cfea3b | |
| parent | sched_ext: idle: use assign_cpu() to update the idle cpumask (diff) | |
| download | wireguard-linux-02f034dcbf3dcb0989e638fdc00d10984dc2278b.tar.xz wireguard-linux-02f034dcbf3dcb0989e638fdc00d10984dc2278b.zip | |
sched_ext: idle: clarify comments
Add a comments to clarify about the usage of cpumask_intersects().
Moreover, update scx_select_cpu_dfl() description clarifying that the
final step of the idle selection logic involves searching for any idle
CPU in the system that the task can use.
Reviewed-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Diffstat (limited to '')
| -rw-r--r-- | kernel/sched/ext.c | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index eec4716de225..2d701203a3db 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -3180,6 +3180,10 @@ static bool test_and_clear_cpu_idle(int cpu) * scx_pick_idle_cpu() can get caught in an infinite loop as * @cpu is never cleared from idle_masks.smt. Ensure that @cpu * is eventually cleared. + * + * NOTE: Use cpumask_intersects() and cpumask_test_cpu() to + * reduce memory writes, which may help alleviate cache + * coherence pressure. */ if (cpumask_intersects(smt, idle_masks.smt)) cpumask_andnot(idle_masks.smt, idle_masks.smt, smt); @@ -3408,6 +3412,8 @@ static void update_selcpu_topology(void) * 4. Pick a CPU within the same NUMA node, if enabled: * - choose a CPU from the same NUMA node to reduce memory access latency. * + * 5. Pick any idle CPU usable by the task. + * * Step 3 and 4 are performed only if the system has, respectively, multiple * LLC domains / multiple NUMA nodes (see scx_selcpu_topo_llc and * scx_selcpu_topo_numa). |