Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Govindapillai, Vinod" <vinod.govindapillai@intel.com>
To: "ville.syrjala@linux.intel.com" <ville.syrjala@linux.intel.com>,
	"intel-gfx@lists.freedesktop.org"
	<intel-gfx@lists.freedesktop.org>
Subject: Re: [PATCH 03/19] drm/i915: Don't clobber crtc_state->cpu_transcoder for inactive crtcs
Date: Sun, 2 Mar 2025 17:28:41 +0000	[thread overview]
Message-ID: <fca75928ea91759ddba4650c4a7ed7ec77398583.camel@intel.com> (raw)
In-Reply-To: <20250218211913.27867-4-ville.syrjala@linux.intel.com>

On Tue, 2025-02-18 at 23:18 +0200, Ville Syrjala wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> 
> Inactive crtcs are supposed to have their crtc_state completely
> cleared. Currently we are clobbering crtc_state->cpu_transcoder
> before determining whether it's actually enabled or not. Don't
> do that.
> 
> I want to rework the inherited flag handling for inactive crtcs
> a bit, and having a bogus cpu_transcoder in the crtc state can
> then cause confusing fastset mismatches even when the crtc never
> changes state during the commit.
> 
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> ---
>  drivers/gpu/drm/i915/display/intel_display.c | 28 +++++++++-----------
>  1 file changed, 13 insertions(+), 15 deletions(-)
> 

Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>

> diff --git a/drivers/gpu/drm/i915/display/intel_display.c
> b/drivers/gpu/drm/i915/display/intel_display.c
> index 8ee8c1df075a..75b184c933e7 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -3109,25 +3109,24 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
>  	struct intel_display *display = to_intel_display(crtc);
>  	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
>  	enum intel_display_power_domain power_domain;
> +	enum transcoder cpu_transcoder = (enum transcoder)crtc->pipe;
>  	intel_wakeref_t wakeref;
> +	bool ret = false;
>  	u32 tmp;
> -	bool ret;
>  
>  	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
>  	wakeref = intel_display_power_get_if_enabled(display, power_domain);
>  	if (!wakeref)
>  		return false;
>  
> +	tmp = intel_de_read(dev_priv, TRANSCONF(dev_priv, cpu_transcoder));
> +	if (!(tmp & TRANSCONF_ENABLE))
> +		goto out;
> +
> +	pipe_config->cpu_transcoder = cpu_transcoder;
> +
>  	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
>  	pipe_config->sink_format = pipe_config->output_format;
> -	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
> -
> -	ret = false;
> -
> -	tmp = intel_de_read(dev_priv,
> -			    TRANSCONF(dev_priv, pipe_config->cpu_transcoder));
> -	if (!(tmp & TRANSCONF_ENABLE))
> -		goto out;
>  
>  	if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) ||
>  	    IS_CHERRYVIEW(dev_priv)) {
> @@ -3492,23 +3491,22 @@ static bool ilk_get_pipe_config(struct intel_crtc *crtc,
>  	struct drm_device *dev = crtc->base.dev;
>  	struct drm_i915_private *dev_priv = to_i915(dev);
>  	enum intel_display_power_domain power_domain;
> +	enum transcoder cpu_transcoder = (enum transcoder)crtc->pipe;
>  	intel_wakeref_t wakeref;
> +	bool ret = false;
>  	u32 tmp;
> -	bool ret;
>  
>  	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
>  	wakeref = intel_display_power_get_if_enabled(display, power_domain);
>  	if (!wakeref)
>  		return false;
>  
> -	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
> -
> -	ret = false;
> -	tmp = intel_de_read(dev_priv,
> -			    TRANSCONF(dev_priv, pipe_config->cpu_transcoder));
> +	tmp = intel_de_read(dev_priv, TRANSCONF(dev_priv, cpu_transcoder));
>  	if (!(tmp & TRANSCONF_ENABLE))
>  		goto out;
>  
> +	pipe_config->cpu_transcoder = cpu_transcoder;
> +
>  	switch (tmp & TRANSCONF_BPC_MASK) {
>  	case TRANSCONF_BPC_6:
>  		pipe_config->pipe_bpp = 18;


  reply	other threads:[~2025-03-02 17:28 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-18 21:18 [PATCH 00/19] drm/i915: cdclk/bw/dbuf readout/sanitation cleanup Ville Syrjala
2025-02-18 21:18 ` [PATCH 01/19] drm/i915/cdclk: Do cdclk post plane programming later Ville Syrjala
2025-02-25 21:19   ` Govindapillai, Vinod
2025-02-18 21:18 ` [PATCH 02/19] drm/i915: Drop redundant shared_dpll=NULL assignments Ville Syrjala
2025-02-25 21:32   ` Govindapillai, Vinod
2025-02-18 21:18 ` [PATCH 03/19] drm/i915: Don't clobber crtc_state->cpu_transcoder for inactive crtcs Ville Syrjala
2025-03-02 17:28   ` Govindapillai, Vinod [this message]
2025-02-18 21:18 ` [PATCH 04/19] drm/i915: Use intel_plane_set_invisible() in intel_plane_disable_noatomic() Ville Syrjala
2025-03-02 17:57   ` Govindapillai, Vinod
2025-02-18 21:18 ` [PATCH 05/19] drm/i915: Extract intel_cdclk_crtc_disable_noatomic() Ville Syrjala
2025-03-02 18:00   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 06/19] drm/i915: Extract skl_wm_crtc_disable_noatomic() Ville Syrjala
2025-03-02 18:06   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 07/19] drm/i915: clean up pipe's ddb usage in intel_crtc_disable_noatomic() Ville Syrjala
2025-03-02 18:19   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 08/19] drm/i915: Add skl_wm_plane_disable_noatomic() Ville Syrjala
2025-03-04 13:47   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 09/19] drm/i915: Extract intel_bw_crtc_disable_noatomic() Ville Syrjala
2025-03-04 13:49   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 10/19] drm/i915: Extract intel_cdclk_update_hw_state() Ville Syrjala
2025-03-04 14:04   ` Govindapillai, Vinod
2025-03-06 15:46     ` Ville Syrjälä
2025-02-18 21:19 ` [PATCH 11/19] drm/i915: Extract intel_bw_update_hw_state() Ville Syrjala
2025-03-04 14:08   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 12/19] drm/i915: Update bw_state->active_pipes during readout Ville Syrjala
2025-03-04 14:10   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 13/19] drm/i915: Skip some bw_state readout on pre-icl Ville Syrjala
2025-03-04 14:20   ` Govindapillai, Vinod
2025-03-06 16:01     ` Ville Syrjälä
2025-02-18 21:19 ` [PATCH 14/19] sem/i915: Simplify intel_cdclk_update_hw_state() Ville Syrjala
2025-03-04 14:25   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 15/19] drm/i915: Simplify cdclk_disable_noatomic() Ville Syrjala
2025-03-04 15:04   ` Govindapillai, Vinod
2025-03-04 15:22     ` Govindapillai, Vinod
2025-03-06 16:10     ` Ville Syrjälä
2025-02-18 21:19 ` [PATCH 16/19] drm/i915: Split wm sanitize from readout Ville Syrjala
2025-03-04 15:10   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 17/19] drm/i915: Do wm readout ealier for skl+ Ville Syrjala
2025-03-04 15:10   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 18/19] drm/i915: Move dbuf_state->active_piepes into skl_wm_get_hw_state() Ville Syrjala
2025-03-04 15:12   ` Govindapillai, Vinod
2025-02-18 21:19 ` [PATCH 19/19] drm/i915: Relocate intel_bw_crtc_update() Ville Syrjala
2025-03-04 15:12   ` Govindapillai, Vinod
2025-02-18 22:51 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: cdclk/bw/dbuf readout/sanitation cleanup Patchwork
2025-02-18 22:51 ` ✗ Fi.CI.SPARSE: " Patchwork
2025-02-18 23:04 ` ✓ i915.CI.BAT: success " Patchwork
2025-02-19  1:24 ` ✗ i915.CI.Full: 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=fca75928ea91759ddba4650c4a7ed7ec77398583.camel@intel.com \
    --to=vinod.govindapillai@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox