Re: [PATCH 2/8] drm/i915/de: Pimp the intel_de_{read, write}_fw() docs

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Fri, Nov 13, 2026 at 04:12:51PM +0200, Jani Nikula wrote:
> On Fri, 13 Nov 2026, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> >
> > Include a bit more guidance on when to use or not use
> > intel_de_{read,write}_fw(), and what the extra caveats
> > are.
> >
> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > ---
> >  drivers/gpu/drm/i915/display/intel_de.h | 13 ++++++++++---
> >  1 file changed, 10 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_de.h b/drivers/gpu/drm/i915/display/intel_de.h
> > index a6dcb91b36f2..2c469a2d6a20 100644
> > --- a/drivers/gpu/drm/i915/display/intel_de.h
> > +++ b/drivers/gpu/drm/i915/display/intel_de.h
> > @@ -87,12 +87,19 @@ int intel_de_wait_for_clear_ms(struct intel_display *display, intel_reg_t reg,
> >  			       u32 mask, unsigned int timeout_ms);
> >  
> >  /*
> > - * Unlocked mmio-accessors, think carefully before using these.
> > + * Unlocked mmio-accessors, think carefully before using these. Generally only
> > + * highly performance sensitive code should use these (eg. vblank evade critical
> > + * section, or interrupt handlers).
> >   *
> >   * Certain architectures will die if the same cacheline is concurrently accessed
> > - * by different clients (e.g. on Ivybridge). Access to registers should
> > - * therefore generally be serialised, by either the dev_priv->uncore.lock or
> > + * by different clients (e.g. on Ivybridge/Haswell). Access to registers should
> > + * therefore generally be serialised, by either the display->mmio.lock or
> >   * a more localised lock guarding all access to that bank of registers.
> > + *
> > + * DMC wakelock will not be taken by these. It must be done manually,
> > + * if needed.
> 
> Is there anything we could mention here about what "if needed" really
> means?

"See intel_dmc_wl.c which registers need the wakelock."
?

> 
> Regardless,
> 
> Reviewed-by: Jani Nikula <jani.nikula@intel.com>
> 
> 
> > + *
> > + * These will also skip the mmio_debug unclaimed register detection.
> >   */
> >  u32 intel_de_read_fw(struct intel_display *display, intel_reg_t reg);
> >  void intel_de_write_fw(struct intel_display *display, intel_reg_t reg, u32 val);
> 
> -- 
> Jani Nikula, Intel

-- 
Ville Syrjälä
Intel