On Fri, May 29, 2026 at 02:04:01PM +0300, Jani Nikula wrote:
> Replace direct intel_display_driver_pm_suspend() and additional calls
> with intel_display_driver_pm_shutdown(). This switches to use
^^^
Reviewed-by: Ville Syrjälä <[email protected]>
> drm_atomic_helper_shutdown() instead of drm_atomic_helper_suspend(),
> which is the more appropriate thing to do anyway.
>
> Not calling intel_display_driver_pm_suspend() from the xe shutdown path
> unblocks further follow-up changes.
>
> There's a slight functional change in that
> intel_display_driver_shutdown() returns early for !HAS_DISPLAY(). Assume
> this is what we want, and there are no cases where display engine is
> present but all pipes have been fused off.
>
> Signed-off-by: Jani Nikula <[email protected]>
> ---
> drivers/gpu/drm/xe/display/xe_display.c | 11 +----------
> 1 file changed, 1 insertion(+), 10 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/display/xe_display.c
> b/drivers/gpu/drm/xe/display/xe_display.c
> index fdc999c62bc9..99bcaa49d11a 100644
> --- a/drivers/gpu/drm/xe/display/xe_display.c
> +++ b/drivers/gpu/drm/xe/display/xe_display.c
> @@ -196,16 +196,7 @@ void xe_display_shutdown(struct xe_device *xe)
> if (!xe->info.probe_display)
> return;
>
> - intel_display_driver_pm_suspend(display);
> -
> - intel_encoder_block_all_hpds(display);
> - intel_hpd_cancel_work(display);
> -
> - if (intel_display_device_present(display))
> - intel_display_driver_suspend_access(display);
> -
> - intel_encoder_suspend_all(display);
> - intel_encoder_shutdown_all(display);
> + intel_display_driver_shutdown(display);
>
> intel_opregion_suspend(display, PCI_D3cold);
>
> --
> 2.47.3
--
Ville Syrjälä
Intel
