From: Imre Deak <imre.deak@intel.com>
To: "Souza, Jose" <jose.souza@intel.com>
Cc: "intel-gfx@lists.freedesktop.org"
<intel-gfx@lists.freedesktop.org>,
"Zanoni, Paulo R" <paulo.r.zanoni@intel.com>
Subject: Re: [PATCH 13/23] drm/i915: Fix the TypeC port mode sanitization during loading/resume
Date: Sat, 8 Jun 2019 20:50:41 +0300 [thread overview]
Message-ID: <20190608175041.GD27722@ideak-desk.fi.intel.com> (raw)
In-Reply-To: <25b091ad328d00ee23159438fd4c7336f826fc10.camel@intel.com>
On Sat, Jun 08, 2019 at 01:39:36AM +0300, Souza, Jose wrote:
> On Tue, 2019-06-04 at 17:58 +0300, Imre Deak wrote:
> > For using the correct AUX power domains we have to sanitize the TypeC
> > port mode early, so move that before encoder sanitization. To do this
> > properly read out the actual port mode instead of just relying on the
> > VBT legacy port flag (which can be incorrect).
> >
> > We also verify that the PHY is connected as expected if the port is
> > active. In case the port is inactive we connect the PHY in case of a
> > legacy port - as we did so far. The PHY will be connected during
> > detection for DP-alt mode - as it was done so far. For TBT-alt mode
> > nothing needs to be done to connect the PHY.
> >
> > Cc: José Roberto de Souza <jose.souza@intel.com>
> > Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
> > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > Signed-off-by: Imre Deak <imre.deak@intel.com>
> > ---
> > drivers/gpu/drm/i915/intel_ddi.c | 16 +-----
> > drivers/gpu/drm/i915/intel_display.c | 10 ++++
> > drivers/gpu/drm/i915/intel_dp_mst.h | 8 ++-
> > drivers/gpu/drm/i915/intel_tc.c | 83
> > ++++++++++++++++++++++++++++
> > drivers/gpu/drm/i915/intel_tc.h | 2 +
> > 5 files changed, 103 insertions(+), 16 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/intel_ddi.c
> > b/drivers/gpu/drm/i915/intel_ddi.c
> > index d236839bee19..2bc3b4f2c9a5 100644
> > --- a/drivers/gpu/drm/i915/intel_ddi.c
> > +++ b/drivers/gpu/drm/i915/intel_ddi.c
> > @@ -3918,17 +3918,6 @@ static void intel_ddi_encoder_suspend(struct
> > intel_encoder *encoder)
> > icl_tc_phy_disconnect(dig_port);
> > }
> >
> > -static void intel_ddi_encoder_reset(struct drm_encoder *drm_encoder)
> > -{
> > - struct intel_digital_port *dig_port =
> > enc_to_dig_port(drm_encoder);
> > - struct drm_i915_private *i915 = to_i915(drm_encoder->dev);
> > -
> > - if (intel_port_is_tc(i915, dig_port->base.port))
> > - intel_digital_port_connected(&dig_port->base);
> > -
> > - intel_dp_encoder_reset(drm_encoder);
> > -}
> > -
> > static void intel_ddi_encoder_destroy(struct drm_encoder *encoder)
> > {
> > struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
> > @@ -3944,7 +3933,7 @@ static void intel_ddi_encoder_destroy(struct
> > drm_encoder *encoder)
> > }
> >
> > static const struct drm_encoder_funcs intel_ddi_funcs = {
> > - .reset = intel_ddi_encoder_reset,
> > + .reset = intel_dp_encoder_reset,
> > .destroy = intel_ddi_encoder_destroy,
> > };
> >
> > @@ -4309,9 +4298,6 @@ void intel_ddi_init(struct drm_i915_private
> > *dev_priv, enum port port)
> >
> > intel_infoframe_init(intel_dig_port);
> >
> > - if (intel_port_is_tc(dev_priv, port))
> > - intel_digital_port_connected(intel_encoder);
> > -
> > return;
> >
> > err:
> > diff --git a/drivers/gpu/drm/i915/intel_display.c
> > b/drivers/gpu/drm/i915/intel_display.c
> > index 0e425a6eebf3..91d18cd0371c 100644
> > --- a/drivers/gpu/drm/i915/intel_display.c
> > +++ b/drivers/gpu/drm/i915/intel_display.c
> > @@ -75,6 +75,7 @@
> > #include "intel_sdvo.h"
> > #include "intel_sideband.h"
> > #include "intel_sprite.h"
> > +#include "intel_tc.h"
> > #include "intel_tv.h"
> > #include "intel_vdsc.h"
> >
> > @@ -16683,6 +16684,15 @@ intel_modeset_setup_hw_state(struct
> > drm_device *dev,
> > intel_modeset_readout_hw_state(dev);
> >
> > /* HW state is read out, now we need to sanitize this mess. */
> > +
> > + /* Sanitize the TypeC port mode upfront, encoders depend on
> > this */
> > + for_each_intel_encoder(dev, encoder) {
> > + /* We need to sanitize only the MST primary port. */
> > + if (encoder->type != INTEL_OUTPUT_DP_MST &&
> > + intel_port_is_tc(dev_priv, encoder->port))
> > + intel_tc_port_sanitize(enc_to_dig_port(&encoder
> > ->base));
> > + }
> > +
> > get_encoder_power_domains(dev_priv);
> >
> > if (HAS_PCH_IBX(dev_priv))
> > diff --git a/drivers/gpu/drm/i915/intel_dp_mst.h
> > b/drivers/gpu/drm/i915/intel_dp_mst.h
> > index 1470c6e0514b..6754c211205a 100644
> > --- a/drivers/gpu/drm/i915/intel_dp_mst.h
> > +++ b/drivers/gpu/drm/i915/intel_dp_mst.h
> > @@ -6,9 +6,15 @@
> > #ifndef __INTEL_DP_MST_H__
> > #define __INTEL_DP_MST_H__
> >
> > -struct intel_digital_port;
> > +#include "intel_drv.h"
> >
> > int intel_dp_mst_encoder_init(struct intel_digital_port
> > *intel_dig_port, int conn_id);
> > void intel_dp_mst_encoder_cleanup(struct intel_digital_port
> > *intel_dig_port);
> > +static inline int
> > +intel_dp_mst_encoder_active_links(struct intel_digital_port
> > *intel_dig_port)
> > +{
> > + return intel_dig_port->dp.active_mst_links;
> > +}
> > +
> >
> > #endif /* __INTEL_DP_MST_H__ */
> > diff --git a/drivers/gpu/drm/i915/intel_tc.c
> > b/drivers/gpu/drm/i915/intel_tc.c
> > index 9832e2ddb92e..9ebf25d7931c 100644
> > --- a/drivers/gpu/drm/i915/intel_tc.c
> > +++ b/drivers/gpu/drm/i915/intel_tc.c
> > @@ -3,6 +3,7 @@
> > * Copyright © 2019 Intel Corporation
> > */
> > #include "intel_display.h"
> > +#include "intel_dp_mst.h"
> > #include "i915_drv.h"
> > #include "intel_tc.h"
> >
> > @@ -167,6 +168,15 @@ static bool icl_tc_phy_set_safe_mode(struct
> > intel_digital_port *dig_port,
> > return true;
> > }
> >
> > +static bool icl_tc_phy_is_in_safe_mode(struct intel_digital_port
> > *dig_port)
> > +{
> > + struct drm_i915_private *dev_priv = to_i915(dig_port-
> > >base.base.dev);
> > + enum tc_port tc_port = intel_port_to_tc(dev_priv, dig_port-
> > >base.port);
> > +
> > + return !(I915_READ(PORT_TX_DFLEXDPCSSS) &
> > + DP_PHY_MODE_STATUS_NOT_SAFE(tc_port));
> > +}
> > +
> > /*
> > * This function implements the first part of the Connect Flow
> > described by our
> > * specification, Gen11 TypeC Programming chapter. The rest of the
> > flow (reading
> > @@ -250,6 +260,49 @@ void icl_tc_phy_disconnect(struct
> > intel_digital_port *dig_port)
> > }
> > }
> >
> > +static bool icl_tc_phy_is_connected(struct intel_digital_port
> > *dig_port)
> > +{
> > + struct drm_i915_private *dev_priv = to_i915(dig_port-
> > >base.base.dev);
> > + enum tc_port tc_port = intel_port_to_tc(dev_priv, dig_port-
> > >base.port);
> > +
> > + if (!icl_tc_phy_status_complete(dig_port)) {
> > + DRM_DEBUG_KMS("Port %s: PHY status not complete\n",
> > + tc_port_name(dev_priv, tc_port));
> > + return dig_port->tc_mode == TC_PORT_TBT_ALT;
> > + }
> > +
> > + if (icl_tc_phy_is_in_safe_mode(dig_port)) {
> > + DRM_DEBUG_KMS("Port %s: PHY still in safe mode\n",
> > + tc_port_name(dev_priv, tc_port));
> > +
> > + return false;
> > + }
> > +
> > + return dig_port->tc_mode == TC_PORT_DP_ALT ||
> > + dig_port->tc_mode == TC_PORT_LEGACY;
> > +}
> > +
> > +static enum tc_port_mode
> > +intel_tc_port_get_current_mode(struct intel_digital_port *dig_port)
> > +{
> > + u32 live_status_mask = tc_port_live_status_mask(dig_port);
> > + bool in_safe_mode = icl_tc_phy_is_in_safe_mode(dig_port);
> > + enum tc_port_mode mode;
> > +
> > + if (in_safe_mode ||
> > WARN_ON(!icl_tc_phy_status_complete(dig_port)))
> > + return TC_PORT_TBT_ALT;
> > +
> > + mode = dig_port->tc_legacy_port ? TC_PORT_LEGACY :
> > TC_PORT_DP_ALT;
> > + if (live_status_mask) {
> > + enum tc_port_mode live_mode = fls(live_status_mask) -
> > 1;
> > +
> > + if (!WARN_ON(live_mode == TC_PORT_TBT_ALT))
> > + mode = live_mode;
> > + }
> > +
> > + return mode;
> > +}
> > +
> > static enum tc_port_mode
> > intel_tc_port_get_target_mode(struct intel_digital_port *dig_port)
> > {
> > @@ -278,6 +331,36 @@ static void intel_tc_port_reset_mode(struct
> > intel_digital_port *dig_port)
> > tc_port_mode_name(dig_port->tc_mode));
> > }
> >
> > +void intel_tc_port_sanitize(struct intel_digital_port *dig_port)
> > +{
> > + struct drm_i915_private *dev_priv = to_i915(dig_port-
> > >base.base.dev);
> > + struct intel_encoder *encoder = &dig_port->base;
> > + enum tc_port tc_port = intel_port_to_tc(dev_priv, encoder-
> > >port);
> > + int active_links = 0;
> > +
> > + dig_port->tc_mode = intel_tc_port_get_current_mode(dig_port);
> > + if (dig_port->dp.is_mst)
> > + active_links =
> > intel_dp_mst_encoder_active_links(dig_port);
> > + else if (encoder->base.crtc)
> > + active_links = to_intel_crtc(encoder->base.crtc)-
> > >active;
> > +
> > + if (active_links) {
> > + if (!icl_tc_phy_is_connected(dig_port))
> > + DRM_DEBUG_DRIVER("Port %s: PHY disconnected
> > with %d active link(s)\n",
> > + tc_port_name(dev_priv,
> > tc_port),
> > + active_links);
> > + goto out;
>
> And then probe sequences will take care of call
> intel_digital_port_connected() to reset TC state.
Right, once the active mode (we detected here) gets disabled, since
before that you can't reset the mode. Btw, we'll also check if a mode
reset is needed before AUX transfers (if the port is not active).
>
> > + }
> > +
> > + if (dig_port->tc_legacy_port)
> > + icl_tc_phy_connect(dig_port);
> > +
> > +out:
> > + DRM_DEBUG_DRIVER("Port %s: sanitize mode (%s)\n",
> > + tc_port_name(dev_priv, tc_port),
> > + tc_port_mode_name(dig_port->tc_mode));
>
> Should be:
>
> DRM_DEBUG_DRIVER()
>
> I guess I missed this on previous patches too.
Ah right, DRM_DEBUG_KMS().
>
>
> With that:
> Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
>
>
> > +}
> > +
> > static bool
> > intel_tc_port_needs_reset(struct intel_digital_port *dig_port)
> > {
> > diff --git a/drivers/gpu/drm/i915/intel_tc.h
> > b/drivers/gpu/drm/i915/intel_tc.h
> > index e937f5326959..10f0f0d81ee4 100644
> > --- a/drivers/gpu/drm/i915/intel_tc.h
> > +++ b/drivers/gpu/drm/i915/intel_tc.h
> > @@ -11,4 +11,6 @@ bool intel_tc_port_connected(struct
> > intel_digital_port *dig_port);
> > u32 intel_tc_port_get_lane_info(struct intel_digital_port
> > *dig_port);
> > int intel_tc_port_fia_max_lane_count(struct intel_digital_port
> > *dig_port);
> >
> > +void intel_tc_port_sanitize(struct intel_digital_port *dig_port);
> > +
> > #endif /* __INTEL_TC_H__ */
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2019-06-08 17:50 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-04 14:58 [PATCH 00/23] drm/i915: Fix TypeC port mode switching Imre Deak
2019-06-04 14:58 ` [PATCH 01/23] drm/i915/icl: Add support to read out the TBT PLL HW state Imre Deak
2019-06-07 17:36 ` Souza, Jose
2019-06-04 14:58 ` [PATCH 02/23] drm/i915: Tune down WARNs about TBT AUX power well enabling Imre Deak
2019-06-07 17:50 ` Souza, Jose
2019-06-08 13:33 ` Imre Deak
2019-06-04 14:58 ` [PATCH 03/23] drm/i915: Move the TypeC port handling code to a separate file Imre Deak
2019-06-06 8:42 ` Jani Nikula
2019-06-06 8:43 ` Jani Nikula
2019-06-06 9:09 ` Imre Deak
2019-06-10 22:57 ` Lucas De Marchi
2019-06-11 10:01 ` Jani Nikula
2019-06-07 17:56 ` Souza, Jose
2019-06-04 14:58 ` [PATCH 04/23] drm/i915: Sanitize the terminology used for TypeC port modes Imre Deak
2019-06-07 19:15 ` Souza, Jose
2019-06-08 13:43 ` Imre Deak
2019-06-10 23:21 ` Lucas De Marchi
2019-06-11 11:19 ` Imre Deak
2019-06-04 14:58 ` [PATCH 05/23] drm/i915: Don't enable the DDI-IO power in the TypeC TBT-alt mode Imre Deak
2019-06-07 19:23 ` Souza, Jose
2019-06-04 14:58 ` [PATCH 06/23] drm/i915: Fix the TBT AUX power well enabling Imre Deak
2019-06-07 19:58 ` Souza, Jose
2019-06-08 13:55 ` Imre Deak
2019-06-04 14:58 ` [PATCH 07/23] drm/i915: Use the correct AUX power domain in TypeC TBT-alt mode Imre Deak
2019-06-07 20:02 ` Souza, Jose
2019-06-04 14:58 ` [PATCH 08/23] drm/i915: Unify the TypeC port notation in debug/error messages Imre Deak
2019-06-07 20:21 ` Souza, Jose
2019-06-07 20:42 ` Imre Deak
2019-06-07 20:45 ` Souza, Jose
2019-06-07 20:46 ` Imre Deak
2019-06-04 14:58 ` [PATCH 09/23] drm/i915: Factor out common parts from TypeC port handling functions Imre Deak
2019-06-06 8:47 ` Jani Nikula
2019-06-06 9:19 ` Imre Deak
2019-06-07 21:22 ` Souza, Jose
2019-06-08 17:23 ` Imre Deak
2019-06-18 16:33 ` Ville Syrjälä
2019-06-18 16:44 ` Imre Deak
2019-06-04 14:58 ` [PATCH 10/23] drm/i915: Wait for TypeC PHY complete flag to clear in safe mode Imre Deak
2019-06-07 21:32 ` Souza, Jose
2019-06-04 14:58 ` [PATCH 11/23] drm/i915: Handle the TCCOLD power-down event Imre Deak
2019-06-07 21:41 ` Souza, Jose
2019-06-08 17:31 ` Imre Deak
2019-06-04 14:58 ` [PATCH 12/23] drm/i915: Sanitize the TypeC connect/detect sequences Imre Deak
2019-06-07 22:15 ` Souza, Jose
2019-06-08 17:40 ` Imre Deak
2019-06-04 14:58 ` [PATCH 13/23] drm/i915: Fix the TypeC port mode sanitization during loading/resume Imre Deak
2019-06-07 22:39 ` Souza, Jose
2019-06-08 17:50 ` Imre Deak [this message]
2019-06-04 14:58 ` [PATCH 14/23] drm/i915: Keep the TypeC port mode fixed for detect/AUX transfers Imre Deak
2019-06-04 14:58 ` [PATCH 15/23] drm/i915: Sanitize the TypeC FIA lane configuration decoding Imre Deak
2019-06-07 22:49 ` Souza, Jose
2019-06-18 16:39 ` Ville Syrjälä
2019-06-18 16:46 ` Imre Deak
2019-06-04 14:58 ` [PATCH 16/23] drm/i915: Sanitize the shared DPLL reserve/release interface Imre Deak
2019-06-04 14:58 ` [PATCH 17/23] drm/i915: Sanitize the shared DPLL find/reference interface Imre Deak
2019-06-04 14:58 ` [PATCH 18/23] drm/i915/icl: Split getting the DPLLs to port type specific functions Imre Deak
2019-06-04 14:58 ` [PATCH 19/23] drm/i915/icl: Reserve all required PLLs for TypeC ports Imre Deak
2019-06-07 17:41 ` [PATCH v2 " Imre Deak
2019-06-18 17:25 ` Ville Syrjälä
2019-06-18 18:02 ` Imre Deak
2019-06-04 14:58 ` [PATCH 20/23] drm/i915: Keep the TypeC port mode fixed when the port is active Imre Deak
2019-06-19 12:58 ` Ville Syrjälä
2019-06-19 16:08 ` Imre Deak
2019-06-04 14:58 ` [PATCH 21/23] drm/i915: Add state verification for the TypeC port mode Imre Deak
2019-06-04 14:58 ` [PATCH 22/23] drm/i915: Remove unneeded disconnect in TypeC legacy " Imre Deak
2019-06-08 0:18 ` Souza, Jose
2019-06-04 14:58 ` [PATCH 23/23] drm/i915: WARN about invalid lane reversal in TBT-alt/DP-alt modes Imre Deak
2019-06-08 0:13 ` Souza, Jose
2019-06-08 0:25 ` Souza, Jose
2019-06-04 17:30 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Fix TypeC port mode switching Patchwork
2019-06-04 17:39 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-06-05 5:24 ` ✓ Fi.CI.BAT: success " Patchwork
2019-06-05 16:36 ` ✓ Fi.CI.IGT: " Patchwork
2019-06-07 17:59 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Fix TypeC port mode switching (rev2) Patchwork
2019-06-07 18:07 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-06-07 18:34 ` ✓ Fi.CI.BAT: success " Patchwork
2019-06-10 12:34 ` ✓ Fi.CI.IGT: " 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=20190608175041.GD27722@ideak-desk.fi.intel.com \
--to=imre.deak@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jose.souza@intel.com \
--cc=paulo.r.zanoni@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