On Mon, Nov 16, 2026 at 11:30:19AM +0100, Maxime Ripard wrote:
> The hdlcd_probe() error paths for pm_runtime_set_active() and
> drm_vblank_init() fall through to the err_pm_active label, which calls
> drm_atomic_helper_shutdown().
>
> However, both calls happen before drm_mode_config_reset(), so the CRTC
> state is still NULL at that point. drm_atomic_helper_shutdown() will
> then go through drm_atomic_helper_disable_all() and
> drm_atomic_get_crtc_state() to drm_atomic_helper_crtc_duplicate_state(),
> which trips its WARN_ON(!crtc->state) and returns an error.
>
> Move drm_atomic_helper_shutdown() to the err_register label, the only
> error path where drm_mode_config_reset() has run and thus where
> shutting the device down makes sense. This also brings the error path
> closer to hdlcd_remove(), which calls drm_atomic_helper_shutdown()
> while runtime PM is still enabled.
>
> The pm_runtime_set_active() error path now jumps directly to
> err_unload: runtime PM hasn't been enabled at that point, so calling
> pm_runtime_disable() there wasn't needed either.
>
> It will also be needed when drm_mode_config_reset() gets replaced by
> drm_mode_config_create_initial_state(), whose state allocation can
> fail.
>
> Fixes: d664b851eb2bc ("drm/arm/hdlcd: Reject atomic commits that disable only
> the plane")
> Reported-by: [email protected]
> Link: https://lore.kernel.org/r/[email protected]
> Signed-off-by: Maxime Ripard <[email protected]>
Something seems to have broken down in your setup as I've only been
Cc-ed on patch 04/64, discovered this one during a search.
Reviewed-by: Liviu Dudau <[email protected]>
Best regards,
Liviu
> ---
> drivers/gpu/drm/arm/hdlcd_drv.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/arm/hdlcd_drv.c b/drivers/gpu/drm/arm/hdlcd_drv.c
> index 340b1bb38bf9..f2bf9a61979d 100644
> --- a/drivers/gpu/drm/arm/hdlcd_drv.c
> +++ b/drivers/gpu/drm/arm/hdlcd_drv.c
> @@ -297,11 +297,11 @@ static int hdlcd_probe(struct platform_device *pdev)
> goto err_unload;
> }
>
> ret = pm_runtime_set_active(dev);
> if (ret)
> - goto err_pm_active;
> + goto err_unload;
>
> pm_runtime_enable(dev);
>
> ret = drm_vblank_init(drm, drm->mode_config.num_crtc);
> if (ret < 0) {
> @@ -333,14 +333,13 @@ static int hdlcd_probe(struct platform_device *pdev)
>
> return 0;
>
> err_register:
> drm_kms_helper_poll_fini(drm);
> + drm_atomic_helper_shutdown(drm);
> err_vblank:
> pm_runtime_disable(drm->dev);
> -err_pm_active:
> - drm_atomic_helper_shutdown(drm);
> err_unload:
> hdlcd_irq_uninstall(hdlcd);
> of_reserved_mem_device_release(drm->dev);
> err_free:
> dev_set_drvdata(dev, NULL);
>
> --
> 2.56.1
>
--
====================
| I would like to |
| fix the world, |
| but they're not |
| giving me the |
\ source code! /
---------------
¯\_(ツ)_/¯