Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Imre Deak <imre.deak@intel.com>
To: Jani Nikula <jani.nikula@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH 6/7] drm/i915: Factor out function to get/put AUX_IO power for main link
Date: Thu, 3 Nov 2022 13:54:22 +0200	[thread overview]
Message-ID: <Y2Or7h3iqB0iM//I@ideak-desk.fi.intel.com> (raw)
In-Reply-To: <875yfxch7y.fsf@intel.com>

On Wed, Nov 02, 2022 at 09:02:41PM +0200, Jani Nikula wrote:
> On Wed, 02 Nov 2022, Imre Deak <imre.deak@intel.com> wrote:
> > Factor out functions to get/put the AUX_IO power domain for the main
> > link on DDI ports.
> >
> > While at it clarify the corresponding code comment.
> >
> > No functional change.
> >
> > Signed-off-by: Imre Deak <imre.deak@intel.com>
> > ---
> >  drivers/gpu/drm/i915/display/intel_ddi.c | 84 ++++++++++++++----------
> >  1 file changed, 51 insertions(+), 33 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
> > index 00b577a5b9a76..7453772d2073d 100644
> > --- a/drivers/gpu/drm/i915/display/intel_ddi.c
> > +++ b/drivers/gpu/drm/i915/display/intel_ddi.c
> > @@ -846,26 +846,63 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
> >  }
> >  
> >  static enum intel_display_power_domain
> > -intel_ddi_main_link_aux_domain(struct intel_digital_port *dig_port)
> > +intel_ddi_main_link_aux_domain(struct intel_digital_port *dig_port,
> > +			       const struct intel_crtc_state *crtc_state)
> >  {
> >  	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
> > +	enum phy phy = intel_port_to_phy(i915, dig_port->base.port);
> >  
> > -	/* ICL+ HW requires corresponding AUX IOs to be powered up for PSR with
> > +	/*
> > +	 * ICL+ HW requires corresponding AUX IOs to be powered up for PSR with
> >  	 * DC states enabled at the same time, while for driver initiated AUX
> >  	 * transfers we need the same AUX IOs to be powered but with DC states
> > -	 * disabled. Accordingly use the AUX power domain here which leaves DC
> > -	 * states enabled.
> > -	 * However, for non-A AUX ports the corresponding non-EDP transcoders
> > -	 * would have already enabled power well 2 and DC_OFF. This means we can
> > -	 * acquire a wider POWER_DOMAIN_AUX_{B,C,D,F} reference instead of a
> > -	 * specific AUX_IO reference without powering up any extra wells.
> > -	 * Note that PSR is enabled only on Port A even though this function
> > -	 * returns the correct domain for other ports too.
> > +	 * disabled. Accordingly use the AUX_IO_<port> power domain here which
> > +	 * leaves DC states enabled.
> > +	 *
> > +	 * Before MTL all DP ports and HDMI ports on TypeC PHYs also require
> > +	 * AUX IO to be enabled, but all these require DC_OFF to be enabled as
> > +	 * well, so we can acquire a wider AUX_<port> power domain reference
> > +	 * instead of a specific AUX_IO_<port> reference without powering up any
> > +	 * extra wells.
> >  	 */
> >  	if (intel_dp_is_edp(&dig_port->dp))
> >  		return intel_display_power_aux_io_domain(i915, dig_port->aux_ch);
> > -	else
> > +	else if (intel_crtc_has_dp_encoder(crtc_state) ||
> > +		 intel_phy_is_tc(i915, phy))
> >  		return intel_aux_power_domain(dig_port);
> > +	else
> > +		return POWER_DOMAIN_INVALID;
> > +}
> > +
> > +static void
> > +get_aux_power_for_main_link(struct intel_digital_port *dig_port,
> > +			    const struct intel_crtc_state *crtc_state)
> > +{
> > +	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
> > +	enum intel_display_power_domain domain =
> > +		intel_ddi_main_link_aux_domain(dig_port, crtc_state);
> > +
> > +	drm_WARN_ON(&i915->drm, dig_port->aux_wakeref);
> > +
> > +	if (domain == POWER_DOMAIN_INVALID)
> > +		return;
> > +
> > +	dig_port->aux_wakeref = intel_display_power_get(i915, domain);
> > +}
> > +
> > +static void
> > +put_aux_power_for_main_link(struct intel_digital_port *dig_port,
> > +			    const struct intel_crtc_state *crtc_state)
> > +{
> > +	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
> > +	intel_wakeref_t wf = fetch_and_zero(&dig_port->aux_wakeref);
> > +	enum intel_display_power_domain domain =
> > +		intel_ddi_main_link_aux_domain(dig_port, crtc_state);
> > +
> > +	if (!wf)
> > +		return;
> > +
> > +	intel_display_power_put(i915, domain, wf);
> >  }
> 
> Nitpick, I think foo_get()/foo_put() are much more common in the driver
> than get_foo()/put_foo().

Ok, can rename these.

> >  static void intel_ddi_get_power_domains(struct intel_encoder *encoder,
> > @@ -873,7 +910,6 @@ static void intel_ddi_get_power_domains(struct intel_encoder *encoder,
> >  {
> >  	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
> >  	struct intel_digital_port *dig_port;
> > -	enum phy phy = intel_port_to_phy(dev_priv, encoder->port);
> >  
> >  	/*
> >  	 * TODO: Add support for MST encoders. Atm, the following should never
> > @@ -892,17 +928,7 @@ static void intel_ddi_get_power_domains(struct intel_encoder *encoder,
> >  								   dig_port->ddi_io_power_domain);
> >  	}
> >  
> > -	/*
> > -	 * AUX power is only needed for (e)DP mode, and for HDMI mode on TC
> > -	 * ports.
> > -	 */
> > -	if (intel_crtc_has_dp_encoder(crtc_state) ||
> > -	    intel_phy_is_tc(dev_priv, phy)) {
> > -		drm_WARN_ON(&dev_priv->drm, dig_port->aux_wakeref);
> > -		dig_port->aux_wakeref =
> > -			intel_display_power_get(dev_priv,
> > -						intel_ddi_main_link_aux_domain(dig_port));
> > -	}
> > +	get_aux_power_for_main_link(dig_port, crtc_state);
> >  }
> >  
> >  void intel_ddi_enable_pipe_clock(struct intel_encoder *encoder,
> > @@ -2741,10 +2767,7 @@ static void intel_ddi_post_disable(struct intel_atomic_state *state,
> >  		intel_ddi_post_disable_dp(state, encoder, old_crtc_state,
> >  					  old_conn_state);
> >  
> > -	if (intel_crtc_has_dp_encoder(old_crtc_state) || is_tc_port)
> > -		intel_display_power_put(dev_priv,
> > -					intel_ddi_main_link_aux_domain(dig_port),
> > -					fetch_and_zero(&dig_port->aux_wakeref));
> > +	put_aux_power_for_main_link(dig_port, old_crtc_state);
> >  
> >  	if (is_tc_port)
> >  		intel_tc_port_put_link(dig_port);
> > @@ -3065,12 +3088,7 @@ intel_ddi_pre_pll_enable(struct intel_atomic_state *state,
> >  	if (is_tc_port)
> >  		intel_tc_port_get_link(dig_port, crtc_state->lane_count);
> >  
> > -	if (intel_crtc_has_dp_encoder(crtc_state) || is_tc_port) {
> > -		drm_WARN_ON(&dev_priv->drm, dig_port->aux_wakeref);
> > -		dig_port->aux_wakeref =
> > -			intel_display_power_get(dev_priv,
> > -						intel_ddi_main_link_aux_domain(dig_port));
> > -	}
> > +	get_aux_power_for_main_link(dig_port, crtc_state);
> >  
> >  	if (is_tc_port && !intel_tc_port_in_tbt_alt_mode(dig_port))
> >  		/*
> 
> -- 
> Jani Nikula, Intel Open Source Graphics Center

  reply	other threads:[~2022-11-03 11:54 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-02 17:15 [Intel-gfx] [PATCH 0/7] drm/i915/tgl+: Enable DC power states on all eDP ports Imre Deak
2022-11-02 17:15 ` [Intel-gfx] [PATCH 1/7] drm/i915: Allocate power domain set wakerefs dynamically Imre Deak
2022-11-02 18:57   ` Jani Nikula
2022-11-03 11:53     ` Imre Deak
2022-11-02 17:15 ` [Intel-gfx] [PATCH 2/7] drm/i915: Move the POWER_DOMAIN_AUX_IO_A definition to its logical place Imre Deak
2022-11-02 17:15 ` [Intel-gfx] [PATCH 3/7] drm/i915/tgl+: Enable display DC power states on all eDP ports Imre Deak
2022-11-02 17:35   ` Ville Syrjälä
2022-11-02 18:34     ` Imre Deak
2022-11-02 19:07       ` Ville Syrjälä
2022-11-03  8:08         ` Imre Deak
2022-11-02 17:15 ` [Intel-gfx] [PATCH 4/7] drm/i915: Add missing AUX_IO_A power domain->well mappings Imre Deak
2022-11-02 17:15 ` [Intel-gfx] [PATCH 5/7] drm/i915: Add missing DC_OFF " Imre Deak
2022-11-02 17:15 ` [Intel-gfx] [PATCH 6/7] drm/i915: Factor out function to get/put AUX_IO power for main link Imre Deak
2022-11-02 19:02   ` Jani Nikula
2022-11-03 11:54     ` Imre Deak [this message]
2022-11-02 17:15 ` [Intel-gfx] [PATCH 7/7] drm/i915/mtl+: Don't enable the AUX_IO power for non-eDP port main links Imre Deak
2022-11-02 19:32 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/tgl+: Enable DC power states on all eDP ports Patchwork
2022-11-02 19:33 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2022-11-02 19:52 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2022-11-03  2:02 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Y2Or7h3iqB0iM//I@ideak-desk.fi.intel.com \
    --to=imre.deak@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=jani.nikula@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox