From: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@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>
Subject: Re: [RFC PATCH] drm/i915/psr: Prevent DC entry during active vblank for Panel Replay
Date: Tue, 28 Apr 2026 22:09:15 +0530 [thread overview]
Message-ID: <15f3c557-602f-4fd9-a2fd-d77290e255fc@intel.com> (raw)
In-Reply-To: <70d7c60be91c179b8d47997cd00ac5d84e824c8c.camel@intel.com>
On 28-04-2026 18:26, Hogander, Jouni wrote:
> On Tue, 2026-04-28 at 18:09 +0530, Dibin Moolakadan Subrahmanian wrote:
>> On PTL+ (display ver >= 30), hold the DMC wakelock while vblank is
>> enabled for Panel Replay. Older platforms rely on
>> intel_display_power_set_target_dc_state() for this.
>>
>> The wakelock is acquired/released at two places:
>> 1) Dynamically when vblank is enabled/disabled via
>> intel_psr_notify_vblank_enable_disable().
>> 2) In the PSR enable/disable path.
>>
>> This handles the following ordering scenarios:
>> 1) Panel Replay is enabled before vblank enable arrives.
>> 2) Vblank enable arrives before Panel Replay is updated in
>> intel_psr_post_plane_update().
> As discussed offline I would check if we could just take the wakelock
> when VBI is getting enabled.
Thanks for the review. As discussed offline,
I’ll rework this and post a new series which unifies the handling
by taking the DMC wakelock when VBI/vblank is enabled.
>
> BR,
> Jouni Högander
>
>> ---
>> .../drm/i915/display/intel_display_types.h | 2 +
>> drivers/gpu/drm/i915/display/intel_psr.c | 55
>> +++++++++++++++++--
>> 2 files changed, 53 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h
>> b/drivers/gpu/drm/i915/display/intel_display_types.h
>> index c81916761850..f386d6dba9e5 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display_types.h
>> +++ b/drivers/gpu/drm/i915/display/intel_display_types.h
>> @@ -1775,6 +1775,8 @@ struct intel_psr {
>> bool source_panel_replay_support;
>> bool sink_panel_replay_support;
>> bool panel_replay_enabled;
>> + bool panel_replay_wakelock;
>> + bool vblank_enabled;
>> u32 dc3co_exitline;
>> u32 dc3co_exit_delay;
>> struct delayed_work dc3co_work;
>> diff --git a/drivers/gpu/drm/i915/display/intel_psr.c
>> b/drivers/gpu/drm/i915/display/intel_psr.c
>> index 63c19958a9e3..82bc63054906 100644
>> --- a/drivers/gpu/drm/i915/display/intel_psr.c
>> +++ b/drivers/gpu/drm/i915/display/intel_psr.c
>> @@ -2192,6 +2192,34 @@ static bool psr_interrupt_error_check(struct
>> intel_dp *intel_dp)
>> return true;
>> }
>>
>> +static void intel_panel_replay_get_wakelock_locked(struct intel_dp
>> *intel_dp)
>> +{
>> + struct intel_display *display = to_intel_display(intel_dp);
>> +
>> + if (DISPLAY_VER(display) < 30)
>> + return;
>> +
>> + if (intel_dp->psr.panel_replay_wakelock)
>> + return;
>> +
>> + intel_dmc_wl_get_noreg(display);
>> + intel_dp->psr.panel_replay_wakelock = true;
>> +}
>> +
>> +static void intel_panel_replay_put_wakelock_locked(struct intel_dp
>> *intel_dp)
>> +{
>> + struct intel_display *display = to_intel_display(intel_dp);
>> +
>> + if (DISPLAY_VER(display) < 30)
>> + return;
>> +
>> + if (!intel_dp->psr.panel_replay_wakelock)
>> + return;
>> +
>> + intel_dmc_wl_put_noreg(display);
>> + intel_dp->psr.panel_replay_wakelock = false;
>> +}
>> +
>> static void intel_psr_enable_locked(struct intel_dp *intel_dp,
>> const struct intel_crtc_state
>> *crtc_state)
>> {
>> @@ -2224,8 +2252,11 @@ static void intel_psr_enable_locked(struct
>> intel_dp *intel_dp,
>> if (!psr_interrupt_error_check(intel_dp))
>> return;
>>
>> - if (intel_dp->psr.panel_replay_enabled)
>> + if (intel_dp->psr.panel_replay_enabled) {
>> + if (intel_dp->psr.vblank_enabled)
>> + intel_panel_replay_get_wakelock_locked(intel
>> _dp);
>> drm_dbg_kms(display->drm, "Enabling Panel
>> Replay\n");
>> + }
>> else
>> drm_dbg_kms(display->drm, "Enabling PSR%s\n",
>> intel_dp->psr.sel_update_enabled ? "2" :
>> "1");
>> @@ -2344,8 +2375,10 @@ static void intel_psr_disable_locked(struct
>> intel_dp *intel_dp)
>> if (!intel_dp->psr.enabled)
>> return;
>>
>> - if (intel_dp->psr.panel_replay_enabled)
>> + if (intel_dp->psr.panel_replay_enabled) {
>> + intel_panel_replay_put_wakelock_locked(intel_dp);
>> drm_dbg_kms(display->drm, "Disabling Panel
>> Replay\n");
>> + }
>> else
>> drm_dbg_kms(display->drm, "Disabling PSR%s\n",
>> intel_dp->psr.sel_update_enabled ? "2" :
>> "1");
>> @@ -4143,9 +4176,24 @@ void
>> intel_psr_notify_vblank_enable_disable(struct intel_display *display,
>> struct intel_dp *intel_dp =
>> enc_to_intel_dp(encoder);
>>
>> mutex_lock(&intel_dp->psr.lock);
>> + intel_dp->psr.vblank_enabled = enable;
>> if (intel_dp->psr.panel_replay_enabled) {
>> + /*
>> + * wakelock handling for panel replay
>> + * for older platform rely on
>> intel_display_power_set_target_dc_state().
>> + */
>> + if (DISPLAY_VER(display) < 30) {
>> + mutex_unlock(&intel_dp->psr.lock);
>> + break;
>> + }
>> +
>> + if (enable)
>> + intel_panel_replay_get_wakelock_lock
>> ed(intel_dp);
>> + else
>> + intel_panel_replay_put_wakelock_lock
>> ed(intel_dp);
>> +
>> mutex_unlock(&intel_dp->psr.lock);
>> - break;
>> + return;
>> }
>>
>> if (intel_dp->psr.enabled && intel_dp-
>>> psr.pkg_c_latency_used)
>> @@ -4154,7 +4202,6 @@ void
>> intel_psr_notify_vblank_enable_disable(struct intel_display *display,
>> mutex_unlock(&intel_dp->psr.lock);
>> return;
>> }
>> -
>> /*
>> * NOTE: intel_display_power_set_target_dc_state is used
>> * only by PSR * code for DC3CO handling. DC3CO target
next prev parent reply other threads:[~2026-04-28 16:39 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-28 12:39 [RFC PATCH] drm/i915/psr: Prevent DC entry during active vblank for Panel Replay Dibin Moolakadan Subrahmanian
2026-04-28 12:56 ` Hogander, Jouni
2026-04-28 16:39 ` Dibin Moolakadan Subrahmanian [this message]
2026-04-28 14:43 ` ✓ i915.CI.BAT: success for " Patchwork
2026-04-28 23:56 ` ✓ i915.CI.Full: " 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=15f3c557-602f-4fd9-a2fd-d77290e255fc@intel.com \
--to=dibin.moolakadan.subrahmanian@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=jouni.hogander@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