From: sashiko-bot@kernel.org
To: "Vidya Srinivas" <vidya.srinivas@intel.com>,
intel-gfx@lists.freedesktop.org
Cc: intel-xe@lists.freedesktop.org
Subject: Re: [PATCH] drm/i915/dp: Extend HDR mode to HLG
Date: Tue, 17 Nov 2026 11:29:12 +0000 [thread overview]
Message-ID: <sashiko-outbox-218711@kernel.org> (raw)
In-Reply-To: <20261117104808.2023577-1-vidya.srinivas@intel.com>
Thank you for your contribution! Sashiko AI review found 2 potential
issues to consider:
- [Medium] drm/i915/dp: PQ/HLG fastset skips TCON update in
intel_backlight_update()
- [Medium] drm/i915/dp: stale HDR TCON bits in
intel_dp_aux_fill_hdr_tcon_params()
--
commit 12c8fa843bb08be12fb5ea03fd4f7818e8682022
Author: Vidya Srinivas <vidya.srinivas@intel.com>
drm/i915/dp: Extend HDR mode to HLG
This commit extends intel_dp_in_hdr_mode() to recognize BT.2100 HLG in
addition to SMPTE ST 2084 (PQ), while restricting
INTEL_EDP_HDR_TCON_2084_DECODE_ENABLE to PQ outputs in the eDP AUX HDR
backlight path.
> diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
> index dc2f0f3088a94..fa68cc6803a6a 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
> @@ -284,8 +284,16 @@ intel_dp_aux_fill_hdr_tcon_params(const struct drm_connector_state *conn_state,
> * HDR mode.
> */
> if (intel_dp_in_hdr_mode(conn_state)) {
> + const struct hdr_output_metadata *hdr_metadata =
> + conn_state->hdr_output_metadata->data;
> +
> *ctrl |= INTEL_EDP_HDR_TCON_SEGMENTED_BACKLIGHT_ENABLE;
> - *ctrl |= INTEL_EDP_HDR_TCON_2084_DECODE_ENABLE;
> +
> + /* The TCON can only decode SMPTE ST 2084 (PQ), not HLG. */
> + if (hdr_metadata->hdmi_metadata_type1.eotf == HDMI_EOTF_SMPTE_ST2084)
> + *ctrl |= INTEL_EDP_HDR_TCON_2084_DECODE_ENABLE;
> + else
> + *ctrl &= ~INTEL_EDP_HDR_TCON_2084_DECODE_ENABLE;
> }
[Severity: Medium]
Will switching hdr_output_metadata between SMPTE ST 2084 (PQ) and BT.2100
HLG on an active panel skip updating INTEL_EDP_HDR_TCON_2084_DECODE_ENABLE?
Because intel_dp_in_hdr_mode() now returns true for both PQ and HLG, both
compute the same pipe_bpp and infoframes.enable mask, and
intel_pipe_config_compare() skips PIPE_CONF_CHECK_INFOFRAME(drm) when
checking for a fastset. Switching between PQ and HLG therefore takes the
fastset path, where intel_ddi_update_pipe_dp() -> intel_backlight_update()
skips __intel_backlight_enable() -> intel_dp_aux_hdr_enable_backlight()
because panel->backlight.enabled is already true:
drivers/gpu/drm/i915/display/intel_backlight.c:intel_backlight_update() {
...
mutex_lock(&display->backlight.lock);
if (!panel->backlight.enabled)
__intel_backlight_enable(crtc_state, conn_state);
mutex_unlock(&display->backlight.lock);
...
}
As a result, INTEL_EDP_HDR_GETSET_CTRL_PARAMS is not updated in DPCD during
a PQ <-> HLG transition, leaving the panel TCON decoding HLG as PQ or
failing to decode PQ after HLG.
[Severity: Medium]
Does intel_dp_aux_fill_hdr_tcon_params() leave
INTEL_EDP_HDR_TCON_2084_DECODE_ENABLE and
INTEL_EDP_HDR_TCON_SEGMENTED_BACKLIGHT_ENABLE set when
intel_dp_in_hdr_mode(conn_state) is false?
intel_dp_aux_hdr_enable_backlight() initializes ctrl from the current DPCD
INTEL_EDP_HDR_GETSET_CTRL_PARAMS value (old_ctrl) before calling
intel_dp_aux_fill_hdr_tcon_params(). When enabling the backlight in SDR
mode, the if (intel_dp_in_hdr_mode(conn_state)) block is not entered, so
neither bit is cleared if it was previously set in DPCD.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261117104808.2023577-1-vidya.srinivas@intel.com?part=1
next prev parent reply other threads:[~2026-11-17 11:29 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-17 10:48 [PATCH] drm/i915/dp: Extend HDR mode to HLG Vidya Srinivas
2026-11-17 11:07 ` ✓ CI.KUnit: success for " Patchwork
2026-11-17 11:29 ` sashiko-bot [this message]
2026-11-17 11:39 ` ✓ Xe.CI.BAT: " 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=sashiko-outbox-218711@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=sashiko-reviews@lists.linux.dev \
--cc=vidya.srinivas@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