Hi, On Sat, Dec 9, 2023 at 7:31 AM Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx> wrote: > > It's the ti_sn65dsi86.pwm auxiliary driver that creates the pwmchip, so > let the auxiliary device be the parent of the pwm device. > > Note that getting a reference to the ti-sn65dsi86's pwm using pwm_get() > isn't affected by this change as ti_sn65dsi86_add_aux_device() sets the > auxiliary device's of_node to that of the main device. > > Also change PM runtime tracking and diagnostic messages to use that one. > After enabling runtime PM operation for the auxiliary device, all works > as expected as parent devices are handled just fine. > > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx> > --- > Changes since v2 > (https://lore.kernel.org/dri-devel/20231209152520.1987483-2-u.kleine-koenig@xxxxxxxxxxxxxx): > > - Make use of devm_pm_runtime_enable as suggested by Douglas Anderson > in reply to v1 already. (Sorry, missed that while preparing v2 :-\) > > Changes since (implicit) v1 > (https://lore.kernel.org/dri-devel/20231127101547.734061-2-u.kleine-koenig@xxxxxxxxxxxxxx): > > - Add a call to pm_runtime_enable() for the aux device > (tested and diagnosed by Nikita Travkin). > - Rebased to yesterday's next, which required some (easy) conflict > resolution for commit c9d99c73940e ("drm/bridge: ti-sn65dsi86: > Simplify using pm_runtime_resume_and_get()"). > > Best regards > Uwe > > drivers/gpu/drm/bridge/ti-sn65dsi86.c | 16 +++++++++------- > 1 file changed, 9 insertions(+), 7 deletions(-) Reviewed-by: Douglas Anderson <dianders@xxxxxxxxxxxx> I'll also note that, via IRC, Steev also confirmed that "on the c630 with the devm_pm.... display works too" Pushed to drm-misc-next: eb3f7cbee294 drm/bridge: ti-sn65dsi86: Associate PWM device to auxiliary device
