From: "Nautiyal, Ankit K" <ankit.k.nautiyal@intel.com>
To: "Hogander, Jouni" <jouni.hogander@intel.com>,
"intel-xe@lists.freedesktop.org" <intel-xe@lists.freedesktop.org>,
"intel-gfx@lists.freedesktop.org"
<intel-gfx@lists.freedesktop.org>
Cc: "ville.syrjala@linux.intel.com" <ville.syrjala@linux.intel.com>,
"Manna, Animesh" <animesh.manna@intel.com>
Subject: Re: [PATCH 4/5] drm/i915/alpm: Simplify and align LOBF checks in pre/post plane update
Date: Thu, 18 Dec 2025 15:07:55 +0530 [thread overview]
Message-ID: <2b5abd99-69fe-4b40-af9b-dffbd633e645@intel.com> (raw)
In-Reply-To: <d8e65d354c42eb4241c75b3c0f6b0125b5f48b73.camel@intel.com>
On 12/17/2025 7:19 PM, Hogander, Jouni wrote:
> On Wed, 2025-11-19 at 19:21 +0530, Ankit Nautiyal wrote:
>> The pre_plane_update and post_plane_update hooks essentially
>> disable/enable lobf feature. Use the existing
>> _is_enabling/is_disabling
>> logic for this in the pre_plane_update and post_plane_update paths.
>>
>> Also rename the helpers to intel_alpm_lobf_{en,dis}able().
>>
>> Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
>> ---
>> drivers/gpu/drm/i915/display/intel_alpm.c | 45 ++++++------------
>> --
>> drivers/gpu/drm/i915/display/intel_alpm.h | 6 +--
>> drivers/gpu/drm/i915/display/intel_display.c | 28 +++++++++++-
>> 3 files changed, 41 insertions(+), 38 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c
>> b/drivers/gpu/drm/i915/display/intel_alpm.c
>> index 5cfb9345776a..abeb70fcf92c 100644
>> --- a/drivers/gpu/drm/i915/display/intel_alpm.c
>> +++ b/drivers/gpu/drm/i915/display/intel_alpm.c
>> @@ -447,25 +447,17 @@ void intel_alpm_port_configure(struct intel_dp
>> *intel_dp,
>> intel_de_write(display, PORT_ALPM_LFPS_CTL(port),
>> lfps_ctl_val);
>> }
>>
>> -void intel_alpm_pre_plane_update(struct intel_atomic_state *state,
>> - struct intel_crtc *crtc)
>> +void intel_alpm_lobf_disable(const struct intel_crtc_state
>> *new_crtc_state)
>> {
>> - struct intel_display *display = to_intel_display(state);
>> - const struct intel_crtc_state *crtc_state =
>> - intel_atomic_get_new_crtc_state(state, crtc);
>> - const struct intel_crtc_state *old_crtc_state =
>> - intel_atomic_get_old_crtc_state(state, crtc);
>> - enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
>> + struct intel_display *display =
>> to_intel_display(new_crtc_state);
>> + enum transcoder cpu_transcoder = new_crtc_state-
>>> cpu_transcoder;
>> struct intel_encoder *encoder;
>>
>> if (DISPLAY_VER(display) < 20)
>> return;
> You can remove this as well. intel_crtc_lobf_enabling is supposed to be
> true only on display version >= 20.
Ok sure will remove this check.
>
>>
>> - if (crtc_state->has_lobf || crtc_state->has_lobf ==
>> old_crtc_state->has_lobf)
>> - return;
>> -
>> for_each_intel_encoder_mask(display->drm, encoder,
>> - crtc_state->uapi.encoder_mask) {
>> + new_crtc_state-
>>> uapi.encoder_mask) {
>> struct intel_dp *intel_dp;
>>
>> if (!intel_encoder_is_dp(encoder))
>> @@ -476,12 +468,10 @@ void intel_alpm_pre_plane_update(struct
>> intel_atomic_state *state,
>> if (!intel_dp_is_edp(intel_dp))
>> continue;
>>
>> - if (old_crtc_state->has_lobf) {
>> - mutex_lock(&intel_dp->alpm.lock);
>> - intel_de_write(display, ALPM_CTL(display,
>> cpu_transcoder), 0);
>> - drm_dbg_kms(display->drm, "Link off between
>> frames (LOBF) disabled\n");
>> - mutex_unlock(&intel_dp->alpm.lock);
>> - }
>> + mutex_lock(&intel_dp->alpm.lock);
>> + intel_de_write(display, ALPM_CTL(display,
>> cpu_transcoder), 0);
>> + drm_dbg_kms(display->drm, "Link off between frames
>> (LOBF) disabled\n");
>> + mutex_unlock(&intel_dp->alpm.lock);
>> }
>> }
>>
>> @@ -502,22 +492,13 @@ void intel_alpm_enable_sink(struct intel_dp
>> *intel_dp,
>> drm_dp_dpcd_writeb(&intel_dp->aux, DP_RECEIVER_ALPM_CONFIG,
>> val);
>> }
>>
>> -void intel_alpm_post_plane_update(struct intel_atomic_state *state,
>> - struct intel_crtc *crtc)
>> +void intel_alpm_lobf_enable(const struct intel_crtc_state
>> *new_crtc_state)
>> {
>> - struct intel_display *display = to_intel_display(state);
>> - const struct intel_crtc_state *crtc_state =
>> - intel_atomic_get_new_crtc_state(state, crtc);
>> - const struct intel_crtc_state *old_crtc_state =
>> - intel_atomic_get_old_crtc_state(state, crtc);
>> + struct intel_display *display =
>> to_intel_display(new_crtc_state);
>> struct intel_encoder *encoder;
>>
>> - if (crtc_state->has_psr || !crtc_state->has_lobf ||
>> - crtc_state->has_lobf == old_crtc_state->has_lobf)
>> - return;
>> -
>> for_each_intel_encoder_mask(display->drm, encoder,
>> - crtc_state->uapi.encoder_mask) {
>> + new_crtc_state-
>>> uapi.encoder_mask) {
>> struct intel_dp *intel_dp;
>>
>> if (!intel_encoder_is_dp(encoder))
>> @@ -526,8 +507,8 @@ void intel_alpm_post_plane_update(struct
>> intel_atomic_state *state,
>> intel_dp = enc_to_intel_dp(encoder);
>>
>> if (intel_dp_is_edp(intel_dp)) {
>> - intel_alpm_enable_sink(intel_dp,
>> crtc_state);
>> - intel_alpm_configure(intel_dp, crtc_state);
>> + intel_alpm_enable_sink(intel_dp,
>> new_crtc_state);
>> + intel_alpm_configure(intel_dp,
>> new_crtc_state);
>> }
>> }
>> }
>> diff --git a/drivers/gpu/drm/i915/display/intel_alpm.h
>> b/drivers/gpu/drm/i915/display/intel_alpm.h
>> index 14dc49fee4c3..0b2460b90465 100644
>> --- a/drivers/gpu/drm/i915/display/intel_alpm.h
>> +++ b/drivers/gpu/drm/i915/display/intel_alpm.h
>> @@ -25,12 +25,10 @@ void intel_alpm_configure(struct intel_dp
>> *intel_dp,
>> const struct intel_crtc_state
>> *crtc_state);
>> void intel_alpm_enable_sink(struct intel_dp *intel_dp,
>> const struct intel_crtc_state
>> *crtc_state);
>> -void intel_alpm_pre_plane_update(struct intel_atomic_state *state,
>> - struct intel_crtc *crtc);
>> +void intel_alpm_lobf_disable(const struct intel_crtc_state
>> *new_crtc_state);
>> void intel_alpm_port_configure(struct intel_dp *intel_dp,
>> const struct intel_crtc_state
>> *crtc_state);
>> -void intel_alpm_post_plane_update(struct intel_atomic_state *state,
>> - struct intel_crtc *crtc);
>> +void intel_alpm_lobf_enable(const struct intel_crtc_state
>> *new_crtc_state);
>> void intel_alpm_lobf_debugfs_add(struct intel_connector *connector);
>> bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp);
>> bool intel_alpm_aux_less_wake_supported(struct intel_dp *intel_dp);
>> diff --git a/drivers/gpu/drm/i915/display/intel_display.c
>> b/drivers/gpu/drm/i915/display/intel_display.c
>> index 069967114bd9..ecbdc506a1fc 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display.c
>> +++ b/drivers/gpu/drm/i915/display/intel_display.c
>> @@ -1009,6 +1009,27 @@ static bool intel_casf_disabling(const struct
>> intel_crtc_state *old_crtc_state,
>> return is_disabling(hw.casf_params.casf_enable,
>> old_crtc_state, new_crtc_state);
>> }
>>
>> +static bool intel_crtc_lobf_enabling(const struct intel_crtc_state
>> *old_crtc_state,
>> + const struct intel_crtc_state
>> *new_crtc_state)
>> +{
>> + if (!new_crtc_state->hw.active)
>> + return false;
> Is it possible that new_crtc_state->hw.active == false &&
> new_crtc_state->has_lobf == true
I dont think its possible.
>> +
>> + if (!new_crtc_state->has_psr)
>> + return false;
> I don't understand why you are checking psr here? I should have been
> checked already on compute phase?
hmm yes right this can be dropped here.
>
>> +
>> + return is_enabling(has_lobf, old_crtc_state,
>> new_crtc_state);
>> +}
>> +
>> +static bool intel_crtc_lobf_disabling(const struct intel_crtc_state
>> *old_crtc_state,
>> + const struct intel_crtc_state
>> *new_crtc_state)
>> +{
>> + if (!new_crtc_state->hw.active)
>> + return false;
> No need to disable before everything else is disabled?
Sorry this should be :
if (!old_crtc_state->hw.active)
return false;
If pipe was already inactive we don't need to do anything.
Thanks for catching this.
I will update the patch with suggested changes.
Thanks & Regards,
Ankit
>
> BR,
>
> Jouni Högander
>
>> +
>> + return is_disabling(has_lobf, old_crtc_state,
>> new_crtc_state);
>> +}
>> +
>> #undef is_disabling
>> #undef is_enabling
>>
>> @@ -1056,7 +1077,8 @@ static void intel_post_plane_update(struct
>> intel_atomic_state *state,
>> adl_scaler_ecc_unmask(new_crtc_state);
>> }
>>
>> - intel_alpm_post_plane_update(state, crtc);
>> + if (intel_crtc_lobf_enabling(old_crtc_state,
>> new_crtc_state))
>> + intel_alpm_lobf_enable(new_crtc_state);
>>
>> intel_psr_post_plane_update(state, crtc);
>> }
>> @@ -1153,7 +1175,9 @@ static void intel_pre_plane_update(struct
>> intel_atomic_state *state,
>> intel_atomic_get_new_crtc_state(state, crtc);
>> enum pipe pipe = crtc->pipe;
>>
>> - intel_alpm_pre_plane_update(state, crtc);
>> + if (intel_crtc_lobf_disabling(old_crtc_state,
>> new_crtc_state))
>> + intel_alpm_lobf_disable(new_crtc_state);
>> +
>> intel_psr_pre_plane_update(state, crtc);
>>
>> if (intel_crtc_vrr_disabling(state, crtc)) {
next prev parent reply other threads:[~2025-12-18 9:38 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-19 13:51 [PATCH 0/5] LOBF fixes Ankit Nautiyal
2025-11-19 13:51 ` [PATCH 1/5] drm/i915/alpm: Compute LOBF late after guardband is already determined Ankit Nautiyal
2025-11-20 8:51 ` Michał Grzelak
2025-11-20 12:21 ` Nautiyal, Ankit K
2025-11-24 3:49 ` Ankit Nautiyal
2025-12-17 13:11 ` Hogander, Jouni
2025-12-24 8:46 ` Nautiyal, Ankit K
2026-02-04 6:23 ` Hogander, Jouni
2025-11-19 13:51 ` [PATCH 2/5] drm/i915/alpm: Allow LOBF only if window1 > alpm check_entry lines Ankit Nautiyal
2025-11-20 8:52 ` Michał Grzelak
2025-11-19 13:51 ` [PATCH 3/5] drm/i915/alpm: Allow LOBF only for platform that have Always on VRR TG Ankit Nautiyal
2025-11-20 8:52 ` Michał Grzelak
2025-11-19 13:51 ` [PATCH 4/5] drm/i915/alpm: Simplify and align LOBF checks in pre/post plane update Ankit Nautiyal
2025-11-20 8:52 ` Michał Grzelak
2025-12-17 13:49 ` Hogander, Jouni
2025-12-18 9:37 ` Nautiyal, Ankit K [this message]
2025-11-19 13:51 ` [PATCH 5/5] drm/i915/alpm: Disable LOBF around transitioning for LRR/seamless MN Ankit Nautiyal
2025-11-20 8:53 ` Michał Grzelak
2025-11-19 14:09 ` ✗ CI.checkpatch: warning for LOBF fixes (rev2) Patchwork
2025-11-19 14:11 ` ✓ CI.KUnit: success " Patchwork
2025-11-19 14:26 ` ✗ CI.checksparse: warning " Patchwork
2025-11-19 15:11 ` ✓ Xe.CI.BAT: success " Patchwork
2025-11-19 17:26 ` ✗ Xe.CI.Full: failure " Patchwork
2025-11-25 1:14 ` ✓ CI.KUnit: success for LOBF fixes (rev3) Patchwork
2025-11-25 1:29 ` ✗ CI.checksparse: warning " Patchwork
2025-11-25 2:07 ` ✓ Xe.CI.BAT: success " Patchwork
2025-11-25 4:21 ` ✗ Xe.CI.Full: failure " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2025-12-24 8:48 [PATCH 0/5] LOBF fixes Ankit Nautiyal
2025-12-24 8:48 ` [PATCH 4/5] drm/i915/alpm: Simplify and align LOBF checks in pre/post plane update Ankit Nautiyal
2025-11-14 5:27 [PATCH 0/5] LOBF fixes Ankit Nautiyal
2025-11-14 5:27 ` [PATCH 4/5] drm/i915/alpm: Simplify and align LOBF checks in pre/post plane update Ankit Nautiyal
2025-11-14 15:15 ` Ville Syrjälä
2025-11-17 7:36 ` Nautiyal, Ankit K
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=2b5abd99-69fe-4b40-af9b-dffbd633e645@intel.com \
--to=ankit.k.nautiyal@intel.com \
--cc=animesh.manna@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=jouni.hogander@intel.com \
--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