From: Jani Nikula <jani.nikula@intel.com>
To: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/i915/pps: split intel_pps_reset_all() to vlv and bxt variants
Date: Mon, 23 Sep 2024 11:22:22 +0300 [thread overview]
Message-ID: <87o74e7pa9.fsf@intel.com> (raw)
In-Reply-To: <ZuwJ2Htfhbgk4bGW@intel.com>
On Thu, 19 Sep 2024, Ville Syrjälä <ville.syrjala@linux.intel.com> wrote:
> On Thu, Sep 19, 2024 at 12:04:27PM +0300, Jani Nikula wrote:
>> The intel_pps_reset_all() function does similar but not quite the same
>> things for VLV/CHV and BXT/GLK. Observe that it's called from platform
>> specific code only, and a split to two functions vlv_pps_reset_all() and
>> bxt_pps_reset_all() is natural.
>>
>> Remove the platform checks and warnings from the functions. We don't
>> usually have them, unless we're unsure. To make this easier to reason
>> about for BXT/GLK, change the condition on caller side from "!PCH" to
>> "BXT || GLK".
>>
>> Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
>> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
>
> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Thanks, pushed to din.
BR,
Jani.
>
>> ---
>> .../i915/display/intel_display_power_well.c | 11 +++---
>> drivers/gpu/drm/i915/display/intel_pps.c | 34 +++++++++++--------
>> drivers/gpu/drm/i915/display/intel_pps.h | 3 +-
>> 3 files changed, 27 insertions(+), 21 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
>> index 1898aff50ac4..adaf7cf3a33b 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
>> +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
>> @@ -879,12 +879,11 @@ void bxt_enable_dc9(struct intel_display *display)
>>
>> drm_dbg_kms(display->drm, "Enabling DC9\n");
>> /*
>> - * Power sequencer reset is not needed on
>> - * platforms with South Display Engine on PCH,
>> - * because PPS registers are always on.
>> + * Power sequencer reset is needed on BXT/GLK, because the PPS registers
>> + * aren't always on, unlike with South Display Engine on PCH.
>> */
>> - if (!HAS_PCH_SPLIT(dev_priv))
>> - intel_pps_reset_all(display);
>> + if (IS_BROXTON(dev_priv) || IS_GEMINILAKE(dev_priv))
>> + bxt_pps_reset_all(display);
>> gen9_set_dc_state(display, DC_STATE_EN_DC9);
>> }
>>
>> @@ -1270,7 +1269,7 @@ static void vlv_display_power_well_deinit(struct drm_i915_private *dev_priv)
>> /* make sure we're done processing display irqs */
>> intel_synchronize_irq(dev_priv);
>>
>> - intel_pps_reset_all(display);
>> + vlv_pps_reset_all(display);
>>
>> /* Prevent us from re-enabling polling on accident in late suspend */
>> if (!dev_priv->drm.dev->power.is_suspended)
>> diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
>> index 819b2843946f..88abc4c7cda1 100644
>> --- a/drivers/gpu/drm/i915/display/intel_pps.c
>> +++ b/drivers/gpu/drm/i915/display/intel_pps.c
>> @@ -70,7 +70,7 @@ intel_wakeref_t intel_pps_lock(struct intel_dp *intel_dp)
>> intel_wakeref_t wakeref;
>>
>> /*
>> - * See intel_pps_reset_all() why we need a power domain reference here.
>> + * See vlv_pps_reset_all() why we need a power domain reference here.
>> */
>> wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_DISPLAY_CORE);
>> mutex_lock(&display->pps.mutex);
>> @@ -448,14 +448,10 @@ pps_initial_setup(struct intel_dp *intel_dp)
>> return intel_pps_is_valid(intel_dp);
>> }
>>
>> -void intel_pps_reset_all(struct intel_display *display)
>> +void vlv_pps_reset_all(struct intel_display *display)
>> {
>> - struct drm_i915_private *dev_priv = to_i915(display->drm);
>> struct intel_encoder *encoder;
>>
>> - if (drm_WARN_ON(display->drm, !IS_LP(dev_priv)))
>> - return;
>> -
>> if (!HAS_DISPLAY(display))
>> return;
>>
>> @@ -472,16 +468,26 @@ void intel_pps_reset_all(struct intel_display *display)
>> for_each_intel_dp(display->drm, encoder) {
>> struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
>>
>> - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
>> - drm_WARN_ON(display->drm,
>> - intel_dp->pps.vlv_active_pipe != INVALID_PIPE);
>> -
>> - if (encoder->type != INTEL_OUTPUT_EDP)
>> - continue;
>> + drm_WARN_ON(display->drm, intel_dp->pps.vlv_active_pipe != INVALID_PIPE);
>>
>> - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
>> + if (encoder->type == INTEL_OUTPUT_EDP)
>> intel_dp->pps.vlv_pps_pipe = INVALID_PIPE;
>> - else
>> + }
>> +}
>> +
>> +void bxt_pps_reset_all(struct intel_display *display)
>> +{
>> + struct intel_encoder *encoder;
>> +
>> + if (!HAS_DISPLAY(display))
>> + return;
>> +
>> + /* See vlv_pps_reset_all() for why we can't grab pps_mutex here. */
>> +
>> + for_each_intel_dp(display->drm, encoder) {
>> + struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
>> +
>> + if (encoder->type == INTEL_OUTPUT_EDP)
>> intel_dp->pps.bxt_pps_reset = true;
>> }
>> }
>> diff --git a/drivers/gpu/drm/i915/display/intel_pps.h b/drivers/gpu/drm/i915/display/intel_pps.h
>> index a5339a65485d..bc5046d53626 100644
>> --- a/drivers/gpu/drm/i915/display/intel_pps.h
>> +++ b/drivers/gpu/drm/i915/display/intel_pps.h
>> @@ -43,7 +43,6 @@ void intel_pps_wait_power_cycle(struct intel_dp *intel_dp);
>> bool intel_pps_init(struct intel_dp *intel_dp);
>> void intel_pps_init_late(struct intel_dp *intel_dp);
>> void intel_pps_encoder_reset(struct intel_dp *intel_dp);
>> -void intel_pps_reset_all(struct intel_display *display);
>>
>> void vlv_pps_pipe_init(struct intel_dp *intel_dp);
>> void vlv_pps_pipe_reset(struct intel_dp *intel_dp);
>> @@ -52,6 +51,8 @@ void vlv_pps_port_enable_unlocked(struct intel_encoder *encoder,
>> const struct intel_crtc_state *crtc_state);
>> void vlv_pps_port_disable(struct intel_encoder *encoder,
>> const struct intel_crtc_state *crtc_state);
>> +void vlv_pps_reset_all(struct intel_display *display);
>> +void bxt_pps_reset_all(struct intel_display *display);
>>
>> void intel_pps_unlock_regs_wa(struct intel_display *display);
>> void intel_pps_setup(struct intel_display *display);
>> --
>> 2.39.2
--
Jani Nikula, Intel
next prev parent reply other threads:[~2024-09-23 8:22 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-19 9:04 [PATCH] drm/i915/pps: split intel_pps_reset_all() to vlv and bxt variants Jani Nikula
2024-09-19 11:24 ` Ville Syrjälä
2024-09-23 8:22 ` Jani Nikula [this message]
2024-09-20 3:42 ` ✓ Fi.CI.BAT: success for " Patchwork
2024-09-20 21:48 ` ✗ 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=87o74e7pa9.fsf@intel.com \
--to=jani.nikula@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=ville.syrjala@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.