Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Nautiyal, Ankit K" <ankit.k.nautiyal@intel.com>
To: <intel-gfx@lists.freedesktop.org>, <intel-xe@lists.freedesktop.org>
Cc: <ville.syrjala@linux.intel.com>
Subject: Re: [PATCH 00/10] Introduce set_context_latency and refactor VRR/DSB timing logic
Date: Thu, 25 Sep 2025 19:06:40 +0530	[thread overview]
Message-ID: <99f4dbe4-e840-4073-ac51-85b3091d4cf9@intel.com> (raw)
In-Reply-To: <20250924141542.3122126-1-ankit.k.nautiyal@intel.com>


On 9/24/2025 7:45 PM, Ankit Nautiyal wrote:
> This patch series refactors VRR and DSB timing logic by introducing
> explicit handling of Set Context Latency (SCL) the number of lines required
> before double buffering to safely program display registers, aka W2 Window.
>
> Previously, SCL was handled implicitly via vblank delay calculations. This
> was a hinderance for later work to optimize guardband length. This series
> formalizes SCL as `set_context_latency` in `intel_crtc_state`, enabling
> consistent tracking and timing/delay computation across platforms.
>
> - Introduces `set_context_latency` to track SCL explicitly.
> - Refactors VRR evasion and DSB wait logic to use SCL directly.
> - Clamps guardband values based on hardware limits and timing constraints.
> - Renames helpers for clarity and removes legacy delay logic.
>
> Rev2: Address comments from Ville:
> - Handle SCL for TGL better and meld patch with previous patch.
> - Drop patch to use set context latency in evasion logic.
> - Add patch to introduce REG_FIELD_MAX based on FIELD_MAX.
> - Added new helper to wait for SCL start and end lines.
> - Other minor refactoring suggested in comments.
>
> Rev3:
> - Drop patch to rename vrr_vblank_delay and instead add a patch to
>    replace it directly with crtc_state->set_context_latency. (Ville)
> - Fix few places where adjusted_mode->crtc_vdisplay was missed. (Ville)
> - Minor refactoring to make the helpers consistent with other parts.
>
> Rev4:
> - Add new Patch#10 to drop the single-use dsb_vblank_delay(). (Ville)
> - Minor changes as suggested in comments in Patch#8.
>
> Ankit Nautiyal (10):
>    drm/i915/psr:
>      s/intel_psr_min_vblank_delay/intel_psr_min_set_context_latency
>    drm/i915/display: Add set_context_latency to crtc_state
>    drm/i915/vrr: Use set_context_latency instead of
>      intel_vrr_real_vblank_delay()
>    drm/i915/vrr: Use SCL for computing guardband
>    drm/i915/dsb:
>      s/intel_dsb_wait_vblank_delay/intel_dsb_wait_for_delayed_vblank
>    drm/i915/display: Wait for scl start instead of dsb_wait_vblanks
>    drm/i915/reg_defs: Add REG_FIELD_MAX wrapper for FIELD_MAX()
>    drm/i915/vrr: Clamp guardband as per hardware and timing constraints
>    drm/i915/display: Drop intel_vrr_vblank_delay and use
>      set_context_latency
>    drm/i915/dsb: Inline dsb_vblank_delay() into
>      intel_dsb_wait_for_delayed_vblank()


Thanks for the reviews, pushed to drm-intel-next.

Regards,

Ankit

>
>   drivers/gpu/drm/i915/display/intel_color.c    |  2 +-
>   .../drm/i915/display/intel_crtc_state_dump.c  |  5 +-
>   drivers/gpu/drm/i915/display/intel_display.c  | 58 ++++++++-----
>   .../drm/i915/display/intel_display_types.h    |  3 +
>   drivers/gpu/drm/i915/display/intel_dsb.c      | 54 +++++++-----
>   drivers/gpu/drm/i915/display/intel_dsb.h      |  4 +-
>   drivers/gpu/drm/i915/display/intel_psr.c      |  6 +-
>   drivers/gpu/drm/i915/display/intel_psr.h      |  2 +-
>   drivers/gpu/drm/i915/display/intel_vblank.c   |  2 +-
>   drivers/gpu/drm/i915/display/intel_vrr.c      | 84 ++++++++++++-------
>   drivers/gpu/drm/i915/display/intel_vrr.h      |  3 +-
>   drivers/gpu/drm/i915/i915_reg_defs.h          | 10 +++
>   12 files changed, 152 insertions(+), 81 deletions(-)
>

      parent reply	other threads:[~2025-09-25 13:36 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-24 14:15 [PATCH 00/10] Introduce set_context_latency and refactor VRR/DSB timing logic Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 01/10] drm/i915/psr: s/intel_psr_min_vblank_delay/intel_psr_min_set_context_latency Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 02/10] drm/i915/display: Add set_context_latency to crtc_state Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 03/10] drm/i915/vrr: Use set_context_latency instead of intel_vrr_real_vblank_delay() Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 04/10] drm/i915/vrr: Use SCL for computing guardband Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 05/10] drm/i915/dsb: s/intel_dsb_wait_vblank_delay/intel_dsb_wait_for_delayed_vblank Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 06/10] drm/i915/display: Wait for scl start instead of dsb_wait_vblanks Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 07/10] drm/i915/reg_defs: Add REG_FIELD_MAX wrapper for FIELD_MAX() Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 08/10] drm/i915/vrr: Clamp guardband as per hardware and timing constraints Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 09/10] drm/i915/display: Drop intel_vrr_vblank_delay and use set_context_latency Ankit Nautiyal
2025-09-24 14:15 ` [PATCH 10/10] drm/i915/dsb: Inline dsb_vblank_delay() into intel_dsb_wait_for_delayed_vblank() Ankit Nautiyal
2025-09-24 15:20   ` Ville Syrjälä
2025-09-25  2:23   ` Ankit Nautiyal
2025-09-25  7:42     ` Ville Syrjälä
2025-09-24 17:13 ` ✓ i915.CI.BAT: success for Introduce set_context_latency and refactor VRR/DSB timing logic (rev4) Patchwork
2025-09-25  0:20 ` ✗ i915.CI.Full: failure " Patchwork
2025-09-25  3:38 ` ✓ i915.CI.BAT: success for Introduce set_context_latency and refactor VRR/DSB timing logic (rev5) Patchwork
2025-09-25  9:16 ` ✗ i915.CI.Full: failure " Patchwork
2025-09-25 12:33 ` ✓ i915.CI.Full: success " Patchwork
2025-09-25 13:36 ` Nautiyal, Ankit K [this message]

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=99f4dbe4-e840-4073-ac51-85b3091d4cf9@intel.com \
    --to=ankit.k.nautiyal@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@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