From: Chenyu Chen <chen-yu.chen@amd.com>
To: <amd-gfx@lists.freedesktop.org>
Cc: Harry Wentland <harry.wentland@amd.com>,
Leo Li <sunpeng.li@amd.com>,
Aurabindo Pillai <aurabindo.pillai@amd.com>,
Roman Li <roman.li@amd.com>, Wayne Lin <wayne.lin@amd.com>,
Tom Chung <chiahsuan.chung@amd.com>,
"Fangzhi Zuo" <jerry.zuo@amd.com>,
Dan Wheeler <daniel.wheeler@amd.com>, Ray Wu <Ray.Wu@amd.com>,
Ivan Lipski <ivan.lipski@amd.com>, Alex Hung <alex.hung@amd.com>,
Chuanyu Tseng <Chuanyu.Tseng@amd.com>,
Allen Li <Allen.Li@amd.com>, ChunTao Tso <chuntao.tso@amd.com>,
Allen Li <allen.li@amd.com>, Chenyu Chen <chen-yu.chen@amd.com>
Subject: [PATCH 11/19] drm/amd/display: Add Replay/PSR active check in link loss status check
Date: Wed, 15 Apr 2026 15:39:50 +0800 [thread overview]
Message-ID: <20260415074223.34848-12-chen-yu.chen@amd.com> (raw)
In-Reply-To: <20260415074223.34848-1-chen-yu.chen@amd.com>
From: Allen Li <Allen.Li@amd.com>
[Why&How]
To avoid unnecessary link retraining when the panel is in Replay/PSR mode,
we need to check if it's in active state and ESD information before we
decide to retrain the link.
Reviewed-by: ChunTao Tso <chuntao.tso@amd.com>
Signed-off-by: Allen Li <allen.li@amd.com>
Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com>
---
.../dc/link/protocols/link_dp_irq_handler.c | 57 +++++++++++--------
1 file changed, 34 insertions(+), 23 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
index 1860d44f63c1..dd19b912c48c 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
@@ -223,7 +223,7 @@ static void handle_hpd_irq_vesa_replay_sink(struct dc_link *link)
}
}
-static void handle_hpd_irq_replay_sink(struct dc_link *link, bool *need_re_enable)
+static void handle_hpd_irq_replay_sink(struct dc_link *link, bool *need_re_enable, bool *replay_esd_detection_needed)
{
union dpcd_replay_configuration replay_configuration = {0};
union dpcd_replay_configuration replay_sink_status = {0};
@@ -311,6 +311,14 @@ static void handle_hpd_irq_replay_sink(struct dc_link *link, bool *need_re_enabl
*need_re_enable = true;
}
}
+
+ if (!link->replay_settings.replay_allow_active &&
+ replay_sink_status.bits.SINK_DEVICE_REPLAY_STATUS == 0x7) {
+ /* If sink device replay status is 0x7 and replay is disabled,
+ * it means sink is in a bad state and link retraining is needed to recover
+ */
+ *replay_esd_detection_needed = true;
+ }
}
void dp_handle_link_loss(struct dc_link *link)
@@ -469,6 +477,7 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link,
enum dc_status result;
bool status = false;
bool replay_re_enable_needed = false;
+ bool replay_esd_detection_needed = false;
if (out_link_loss)
*out_link_loss = false;
@@ -482,6 +491,7 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link,
DC_LOG_HW_HPD_IRQ("%s: Got short pulse HPD on link %d\n",
__func__, link->link_index);
+ handle_hpd_irq_replay_sink(link, &replay_re_enable_needed, &replay_esd_detection_needed);
/* All the "handle_hpd_irq_xxx()" methods
* should be called only after
@@ -528,8 +538,6 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link,
/* PSR-related error was detected and handled */
return true;
- handle_hpd_irq_replay_sink(link, &replay_re_enable_needed);
-
/* If PSR-related error handled, Main link may be off,
* so do not handle as a normal sink status change interrupt.
*/
@@ -552,27 +560,30 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link,
* Downstream port status changed,
* then DM should call DC to do the detection.
* NOTE: Now includes eDP link loss detection and retraining
+ * Link will be retrained if panel is not EDP or
+ * Replay ESD recovery is needed.
*/
-
- if (dp_parse_link_loss_status(
- link,
- &hpd_irq_dpcd_data)) {
- /* Connectivity log: link loss */
- CONN_DATA_LINK_LOSS(link,
- hpd_irq_dpcd_data.raw,
- sizeof(hpd_irq_dpcd_data),
- "Status: ");
-
- if (defer_handling && has_left_work)
- *has_left_work = true;
- else
- dp_handle_link_loss(link);
-
- status = false;
- if (out_link_loss)
- *out_link_loss = true;
-
- dp_trace_link_loss_increment(link);
+ if (link->connector_signal != SIGNAL_TYPE_EDP || replay_esd_detection_needed) {
+ if (dp_parse_link_loss_status(
+ link,
+ &hpd_irq_dpcd_data)) {
+ /* Connectivity log: link loss */
+ CONN_DATA_LINK_LOSS(link,
+ hpd_irq_dpcd_data.raw,
+ sizeof(hpd_irq_dpcd_data),
+ "Status: ");
+
+ if (defer_handling && has_left_work)
+ *has_left_work = true;
+ else
+ dp_handle_link_loss(link);
+
+ status = false;
+ if (out_link_loss)
+ *out_link_loss = true;
+
+ dp_trace_link_loss_increment(link);
+ }
}
if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling) {
--
2.43.0
next prev parent reply other threads:[~2026-04-15 7:44 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-15 7:39 [PATCH 00/19] DC Patches Apr 20 2026 Chenyu Chen
2026-04-15 7:39 ` [PATCH 01/19] drm/amd/display: Add allow_clock_gating to dcn42 dccg Chenyu Chen
2026-04-15 7:39 ` [PATCH 02/19] drm/amd/display: bypass post csc for additional color spaces in dcn42 Chenyu Chen
2026-04-15 7:39 ` [PATCH 03/19] drm/amd/display: Remove unused dml2_project Chenyu Chen
2026-04-15 7:39 ` [PATCH 04/19] drm/amd/display: Unset Replay desync error verification by default Chenyu Chen
2026-04-15 7:39 ` [PATCH 05/19] drm/amd/display: Align HWSS fast commit path with legacy path Chenyu Chen
2026-04-15 7:39 ` [PATCH 06/19] drm/amd/display: Fix implicit narrowing conversion warnings Chenyu Chen
2026-04-15 7:39 ` [PATCH 07/19] drm/amd/display: Fix double free Chenyu Chen
2026-04-15 7:39 ` [PATCH 08/19] drm/amd/display: Introduce power module on Linux Chenyu Chen
2026-04-15 7:39 ` [PATCH 09/19] drm/amd/display: Add " Chenyu Chen
2026-04-15 7:39 ` [PATCH 10/19] drm/amd/display: Fix fpu guard warning Chenyu Chen
2026-04-17 8:07 ` mikhail.v.gavrilov
2026-04-15 7:39 ` Chenyu Chen [this message]
2026-04-15 7:39 ` [PATCH 12/19] drm/amd/display: Remove SYMCLK F and G values from link encoder and MANUAL_FLOW_CONTROL from optc Chenyu Chen
2026-04-15 7:39 ` [PATCH 13/19] drm/amd/display: Add minimum vfp requirement Chenyu Chen
2026-04-15 7:39 ` [PATCH 14/19] drm/amd/display: Fix narrowing boundaries and eDP parser assignment Chenyu Chen
2026-04-15 7:39 ` [PATCH 15/19] drm/amd/display: Fix dml2_0 narrowing boundaries Chenyu Chen
2026-04-15 7:39 ` [PATCH 16/19] drm/amd/display: Add README.md file to DML2_0 repository Chenyu Chen
2026-04-15 7:39 ` [PATCH 17/19] drm/amd/display: Fix DPMS using partially updated pipe context Chenyu Chen
2026-04-15 7:39 ` [PATCH 18/19] drm/amd/display: Move dml2_destroy to non-FPU compilation unit Chenyu Chen
2026-04-15 7:39 ` [PATCH 19/19] drm/amd/display: Promote DC to 3.2.379 Chenyu Chen
2026-04-20 12:54 ` [PATCH 00/19] DC Patches Apr 20 2026 Wheeler, Daniel
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=20260415074223.34848-12-chen-yu.chen@amd.com \
--to=chen-yu.chen@amd.com \
--cc=Allen.Li@amd.com \
--cc=Chuanyu.Tseng@amd.com \
--cc=Ray.Wu@amd.com \
--cc=alex.hung@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=aurabindo.pillai@amd.com \
--cc=chiahsuan.chung@amd.com \
--cc=chuntao.tso@amd.com \
--cc=daniel.wheeler@amd.com \
--cc=harry.wentland@amd.com \
--cc=ivan.lipski@amd.com \
--cc=jerry.zuo@amd.com \
--cc=roman.li@amd.com \
--cc=sunpeng.li@amd.com \
--cc=wayne.lin@amd.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