On Fri, 23 Jul 2021, Christophe JAILLET wrote: > This function logs a warning if the workqueue gets too big. > In order to save a few cycles, use 'atomic_inc_return()' instead of an > 'atomic_inc()/atomic_read()' sequence. > > This axes a line of code and saves a 'atomic_read()' call. > > Signed-off-by: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx> > --- > drivers/hid/hid-logitech-hidpp.c | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/drivers/hid/hid-logitech-hidpp.c b/drivers/hid/hid-logitech-hidpp.c > index 61635e629469..a7fa35245c2e 100644 > --- a/drivers/hid/hid-logitech-hidpp.c > +++ b/drivers/hid/hid-logitech-hidpp.c > @@ -2240,11 +2240,10 @@ static int hidpp_ff_queue_work(struct hidpp_ff_private_data *data, int effect_id > wd->size = size; > memcpy(wd->params, params, size); > > - atomic_inc(&data->workqueue_size); > + s = atomic_inc_return(&data->workqueue_size); > queue_work(data->wq, &wd->work); > > /* warn about excessive queue size */ > - s = atomic_read(&data->workqueue_size); > if (s >= 20 && s % 20 == 0) > hid_warn(data->hidpp->hid_dev, "Force feedback command queue contains %d commands, causing substantial delays!", s); Applied, thank you. -- Jiri Kosina SUSE Labs