On Thu, 07 May 2026, James Xiong <[email protected]> wrote:
> During driver probe, DMC firmware is loaded asynchronously via a
> workqueue. There is a race between parse_dmc_fw() setting the payload
> pointer (making has_dmc_id_fw() return true) and intel_dmc_load_program()
> writing the firmware to hardware registers. If the probe thread calls
> intel_dmc_enable_pipe() -> assert_dmc_loaded() in this window via
> intel_modeset_setup_hw_state(), it sees parsed payload but stale HW
> registers, triggering a ~20% intermittent WARNING on ADL-N warm boot.

I wonder if intel_dmc_enable_pipe() should call intel_dmc_wait_fw_load()
itself?

>
> v2: Fix by calling intel_dmc_wait_fw_load() in
>     intel_modeset_setup_hw_state() before iterating the CRTCs (Gustavo
>     Sousa).
>
> Signed-off-by: James Xiong <[email protected]>
> ---
>  drivers/gpu/drm/i915/display/intel_modeset_setup.c | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c 
> b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
> index e88082c8caac..277e56848470 100644
> --- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
> +++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
> @@ -961,6 +961,8 @@ void intel_modeset_setup_hw_state(struct intel_display 
> *display,
>        * intel_sanitize_plane_mapping() may need to do vblank
>        * waits, so we need vblank interrupts restored beforehand.
>        */
> +     intel_dmc_wait_fw_load(display);
> +

No matter what, the comment above now appears to refer to this
intel_dmc_wait_fw_load() call, which makes no sense at all.

BR,
Jani.

>       for_each_intel_crtc(display->drm, crtc) {
>               struct intel_crtc_state *crtc_state =
>                       to_intel_crtc_state(crtc->base.state);

-- 
Jani Nikula, Intel

Reply via email to