Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Nautiyal, Ankit K" <ankit.k.nautiyal@intel.com>
To: Ville Syrjala <ville.syrjala@linux.intel.com>,
	<intel-gfx@lists.freedesktop.org>
Subject: Re: [PATCH 15/18] drm/i915/vrr: Add extra vblank delay to estimates
Date: Fri, 3 Jan 2025 17:43:22 +0530	[thread overview]
Message-ID: <c01ef3ed-a1ef-4233-a84f-38fd4dc23120@intel.com> (raw)
In-Reply-To: <20241210211007.5976-16-ville.syrjala@linux.intel.com>


On 12/11/2024 2:40 AM, Ville Syrjala wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> On ICL/TGL the VRR hardware injects an extra scanline just after
> vactive. This essentically behaves the same as an extra line of
> vblank delay, except it only appears in this one specific spot.
>
> Considee our DSB interrupt signalling scheme:
typo: Consider
> 1. arm the update
> 2. wait for undelayed vblank (or rather safe window with VRR)
> 3. wait for enough usecs to get past the delayed vblank
> 4. signal interrupt to indicate that arming has latched
>
> If step 2 waits for end of vactive step 3 needs to account for
> the extra one scanline, or else we risk signalling the interrupt
> before the delayed vblank has actuall elapsed. So include the
> extra scanline in our vblank delay estimates.
>
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>


> ---
>   drivers/gpu/drm/i915/display/intel_vrr.c | 33 +++++++++++++++++++-----
>   1 file changed, 27 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c
> index c08dd1e5cffc..98dd92316595 100644
> --- a/drivers/gpu/drm/i915/display/intel_vrr.c
> +++ b/drivers/gpu/drm/i915/display/intel_vrr.c
> @@ -76,12 +76,33 @@ intel_vrr_check_modeset(struct intel_atomic_state *state)
>   	}
>   }
>   
> -int intel_vrr_vblank_delay(const struct intel_crtc_state *crtc_state)
> +static int intel_vrr_real_vblank_delay(const struct intel_crtc_state *crtc_state)
>   {
>   	return crtc_state->hw.adjusted_mode.crtc_vblank_start -
>   		crtc_state->hw.adjusted_mode.crtc_vdisplay;
>   }
>   
> +static int intel_vrr_extra_vblank_delay(struct intel_display *display)
> +{
> +	/*
> +	 * On ICL/TGL VRR hardware inserts one extra scanline
> +	 * just after vactive, which pushes the vmin decision
> +	 * boundary ahead accordingly. We'll include the extra
> +	 * scanline in our vblank delay estimates to make sure
> +	 * that we never underestimate how long we have until
> +	 * the delayed vblank has passed.
> +	 */
> +	return DISPLAY_VER(display) < 13 ? 1 : 0;
> +}
> +
> +int intel_vrr_vblank_delay(const struct intel_crtc_state *crtc_state)
> +{
> +	struct intel_display *display = to_intel_display(crtc_state);
> +
> +	return intel_vrr_real_vblank_delay(crtc_state) +
> +		intel_vrr_extra_vblank_delay(display);
> +}
> +
>   static int intel_vrr_flipline_offset(struct intel_display *display)
>   {
>   	/* ICL/TGL hardware imposes flipline>=vmin+1 */
> @@ -131,7 +152,7 @@ int intel_vrr_vmin_vtotal(const struct intel_crtc_state *crtc_state)
>   		return intel_vrr_vmin_flipline(crtc_state);
>   	else
>   		return intel_vrr_vmin_flipline(crtc_state) +
> -			intel_vrr_vblank_delay(crtc_state);
> +			intel_vrr_real_vblank_delay(crtc_state);
>   }
>   
>   int intel_vrr_vmax_vtotal(const struct intel_crtc_state *crtc_state)
> @@ -142,7 +163,7 @@ int intel_vrr_vmax_vtotal(const struct intel_crtc_state *crtc_state)
>   		return crtc_state->vrr.vmax;
>   	else
>   		return crtc_state->vrr.vmax +
> -			intel_vrr_vblank_delay(crtc_state);
> +			intel_vrr_real_vblank_delay(crtc_state);
>   }
>   
>   int intel_vrr_vmin_vblank_start(const struct intel_crtc_state *crtc_state)
> @@ -310,9 +331,9 @@ void intel_vrr_compute_config_late(struct intel_crtc_state *crtc_state)
>   		 * vmin/vmax/flipline also need to be adjusted by
>   		 * the vblank delay to maintain correct vtotals.
>   		 */
> -		crtc_state->vrr.vmin -= intel_vrr_vblank_delay(crtc_state);
> -		crtc_state->vrr.vmax -= intel_vrr_vblank_delay(crtc_state);
> -		crtc_state->vrr.flipline -= intel_vrr_vblank_delay(crtc_state);
> +		crtc_state->vrr.vmin -= intel_vrr_real_vblank_delay(crtc_state);
> +		crtc_state->vrr.vmax -= intel_vrr_real_vblank_delay(crtc_state);
> +		crtc_state->vrr.flipline -= intel_vrr_real_vblank_delay(crtc_state);
>   	}
>   }
>   

  reply	other threads:[~2025-01-03 12:13 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-10 21:09 [PATCH 00/18] drm/i915: DSB+VRR Ville Syrjala
2024-12-10 21:09 ` [PATCH 01/18] drm/i915: Extract intel_crtc_vblank_delay() Ville Syrjala
2025-01-03 11:24   ` Nautiyal, Ankit K
2025-01-09 15:43     ` Ville Syrjälä
2025-01-10  5:09       ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 02/18] drm/i915: Check vblank delay validity Ville Syrjala
2025-01-03 11:26   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 03/18] drm/i915: Fix include order Ville Syrjala
2025-01-03 11:27   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 04/18] drm/i915: Introduce intel_vrr_{vmin,vmax}_vtotal() Ville Syrjala
2025-01-03 11:28   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 05/18] drm/i915: Move framestart/etc. state dump to a better spot Ville Syrjala
2025-01-03 11:29   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 06/18] drm/i915: Include the vblank delay in the state dump Ville Syrjala
2025-01-03 11:30   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 07/18] drm/i915/vrr: Improve VRR " Ville Syrjala
2025-01-03 11:31   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 08/18] drm/i915: Include the scanline offset in the " Ville Syrjala
2025-01-03 11:32   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 09/18] drm/i915: Extract intel_mode_vblank_delay() Ville Syrjala
2025-01-03 11:34   ` Nautiyal, Ankit K
2024-12-10 21:09 ` [PATCH 10/18] drm/i915: Consolidate intel_pre_commit_crtc_state() Ville Syrjala
2025-01-03 11:38   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 11/18] drm/i915: Extract intel_crtc_active_timings() Ville Syrjala
2025-01-03 11:42   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 12/18] drm/i915/vrr: Introduce intel_vrr_vblank_delay() Ville Syrjala
2025-01-03 11:48   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 13/18] drm/i915/vrr: Drop the extra vmin adjustment for ADL+ Ville Syrjala
2025-01-03 11:56   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 14/18] drm/i915/vrr: Fix vmin/vmax/flipline on TGL when using vblank delay Ville Syrjala
2025-01-03 12:03   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 15/18] drm/i915/vrr: Add extra vblank delay to estimates Ville Syrjala
2025-01-03 12:13   ` Nautiyal, Ankit K [this message]
2024-12-10 21:10 ` [PATCH 16/18] drm/i915/vrr: Plumb the DSB into intel_vrr_send_push() Ville Syrjala
2025-01-03 12:14   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 17/18] drm/i915/dsb: Add support for triggering VRR push with DSB Ville Syrjala
2025-01-03 12:17   ` Nautiyal, Ankit K
2024-12-10 21:10 ` [PATCH 18/18] drm/i915/dsb: Allow DSB to perform commits when VRR is enabled Ville Syrjala
2025-01-03 12:18   ` Nautiyal, Ankit K
2024-12-10 22:17 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: DSB+VRR Patchwork
2024-12-10 22:17 ` ✗ Fi.CI.SPARSE: " Patchwork
2024-12-10 22:30 ` ✓ i915.CI.BAT: success " Patchwork
2024-12-10 23:49 ` ✗ i915.CI.Full: failure " Patchwork
2024-12-13 23:41 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: DSB+VRR (rev2) Patchwork
2024-12-13 23:41 ` ✗ Fi.CI.SPARSE: " Patchwork
2024-12-13 23:55 ` ✗ i915.CI.BAT: failure " Patchwork
2025-01-13 16:48 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: DSB+VRR (rev3) Patchwork
2025-01-13 16:48 ` ✗ Fi.CI.SPARSE: " Patchwork
2025-01-13 17:12 ` ✓ i915.CI.BAT: success " Patchwork
2025-01-15  4:05 ` ✗ i915.CI.Full: failure " Patchwork
2025-01-15 18:11 ` [PATCH 00/18] drm/i915: DSB+VRR Ville Syrjälä

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=c01ef3ed-a1ef-4233-a84f-38fd4dc23120@intel.com \
    --to=ankit.k.nautiyal@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