diff options
| author | Rafael J. Wysocki <rafael.j.wysocki@intel.com> | 2025-10-23 19:12:57 +0200 |
|---|---|---|
| committer | Rafael J. Wysocki <rafael.j.wysocki@intel.com> | 2025-10-27 14:41:27 +0100 |
| commit | db86f55bf81a3a297be05ee8775ae9a8c6e3a599 (patch) | |
| tree | 6ff8a382e3b603b3b1ae6b9f31b30d599744c111 | |
| parent | dcb6fa37fd7bc9c3d2b066329b0d27dedf8becaa (diff) | |
cpuidle: governors: menu: Select polling state in some more cases
A throughput regression of 11% introduced by commit 779b1a1cb13a ("cpuidle:
governors: menu: Avoid selecting states with too much latency") has been
reported and it is related to the case when the menu governor checks if
selecting a proper idle state instead of a polling one makes sense.
In particular, it is questionable to do so if the exit latency of the
idle state in question exceeds the predicted idle duration, so add a
check for that, which is sufficient to make the reported regression go
away, and update the related code comment accordingly.
Fixes: 779b1a1cb13a ("cpuidle: governors: menu: Avoid selecting states with too much latency")
Closes: https://lore.kernel.org/linux-pm/004501dc43c9$ec8aa930$c59ffb90$@telus.net/
Reported-by: Doug Smythies <dsmythies@telus.net>
Tested-by: Doug Smythies <dsmythies@telus.net>
Cc: All applicable <stable@vger.kernel.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Link: https://patch.msgid.link/12786727.O9o76ZdvQC@rafael.j.wysocki
| -rw-r--r-- | drivers/cpuidle/governors/menu.c | 7 |
1 files changed, 5 insertions, 2 deletions
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index 7d21fb5a72f4..23239b0c04f9 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -318,10 +318,13 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev, /* * Use a physical idle state, not busy polling, unless a timer - * is going to trigger soon enough. + * is going to trigger soon enough or the exit latency of the + * idle state in question is greater than the predicted idle + * duration. */ if ((drv->states[idx].flags & CPUIDLE_FLAG_POLLING) && - s->target_residency_ns <= data->next_timer_ns) { + s->target_residency_ns <= data->next_timer_ns && + s->exit_latency_ns <= predicted_ns) { predicted_ns = s->target_residency_ns; idx = i; break; |