[PATCH 0/4] drm/amd/display: Follow-up patches to analog support

[PATCH 0/4] drm/amd/display: Follow-up patches to analog support

[Timur Kristóf]

This series addresses some feedback that I received after the DC analog
series support landed. Harry already gave a review to the patches.
Please feel free to let me know if you have any other feedback that I
should address regarding the analog connector support.

Timur Kristóf (4):
  drm/amd/display: Fix warning for analog stream encoders
  drm/amd/display: Cleanup uses of the analog flag
  drm/amd/display: Cleanup early return in construct_phy
  drm/amd/display: Move analog check to dce110_hwseq

 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15 ++++++++++++---
 .../drm/amd/display/dc/dce/dce_stream_encoder.c   |  2 +-
 .../drm/amd/display/dc/hwss/dce110/dce110_hwseq.c |  6 ++++--
 .../gpu/drm/amd/display/dc/link/link_detection.c  |  2 --
 drivers/gpu/drm/amd/display/dc/link/link_dpms.c   |  3 +--
 .../gpu/drm/amd/display/dc/link/link_factory.c    | 13 +++++++------
 6 files changed, 25 insertions(+), 16 deletions(-)

-- 
2.51.1

[PATCH 1/4] drm/amd/display: Fix warning for analog stream encoders

[Timur Kristóf]

Fixes the following warning that some users are reporting
with some kernel configurations:

"positional initialization of field in 'struct' declared
with 'designated_init' attribute"

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
index f8996ee2856b..574618d5d4a4 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c
@@ -1568,7 +1568,7 @@ void dce110_stream_encoder_construct(
 	enc110->se_mask = se_mask;
 }
 
-static const struct stream_encoder_funcs dce110_an_str_enc_funcs = {0};
+static const struct stream_encoder_funcs dce110_an_str_enc_funcs = {};
 
 void dce110_analog_stream_encoder_construct(
 	struct dce110_stream_encoder *enc110,
-- 
2.51.1

[PATCH 2/4] drm/amd/display: Cleanup uses of the analog flag

[Timur Kristóf]

In the detect_link_and_local_sink() function, do not modify the
EDID capabilities of the display based on the connector. Instead,
respect the analog flag better and when the analog flag is set,
check that the connector indeed supports analog displays.

Suggested-by: Ray Wu <Ray.Wu@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15 ++++++++++++---
 .../gpu/drm/amd/display/dc/link/link_detection.c  |  2 --
 2 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index aa31e969b39d..9703f7fc957d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8665,8 +8665,14 @@ static void amdgpu_dm_connector_add_freesync_modes(struct drm_connector *connect
 	if (!(amdgpu_freesync_vid_mode && drm_edid))
 		return;
 
-	if (!amdgpu_dm_connector->dc_sink || amdgpu_dm_connector->dc_sink->edid_caps.analog ||
-		!dc_supports_vrr(amdgpu_dm_connector->dc_sink->ctx->dce_version))
+	if (!amdgpu_dm_connector->dc_sink || !amdgpu_dm_connector->dc_link)
+		return;
+
+	if (!dc_supports_vrr(amdgpu_dm_connector->dc_sink->ctx->dce_version))
+		return;
+
+	if (dc_connector_supports_analog(amdgpu_dm_connector->dc_link->link_id.id) &&
+	    amdgpu_dm_connector->dc_sink->edid_caps.analog)
 		return;
 
 	if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10)
@@ -8693,7 +8699,10 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector)
 			amdgpu_dm_connector->num_modes +=
 				drm_add_modes_noedid(connector, 1920, 1080);
 
-		if (amdgpu_dm_connector->dc_sink && amdgpu_dm_connector->dc_sink->edid_caps.analog) {
+		if (amdgpu_dm_connector->dc_sink &&
+		    amdgpu_dm_connector->dc_link &&
+		    amdgpu_dm_connector->dc_sink->edid_caps.analog &&
+		    dc_connector_supports_analog(amdgpu_dm_connector->dc_link->link_id.id)) {
 			/* Analog monitor connected by DAC load detection.
 			 * Add common modes. It will be up to the user to select one that works.
 			 */
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
index d163360a2bf6..6d31f4967f1a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
@@ -1224,8 +1224,6 @@ static bool detect_link_and_local_sink(struct dc_link *link,
 			break;
 		}
 
-		sink->edid_caps.analog &= dc_connector_supports_analog(link->link_id.id);
-
 		// Check if edid is the same
 		if ((prev_sink) &&
 		    (edid_status == EDID_THE_SAME || edid_status == EDID_OK))
-- 
2.51.1

[PATCH 3/4] drm/amd/display: Cleanup early return in construct_phy

[Timur Kristóf]

Match pre-existing patterns in the DC code base.
Instead of returning early from the construct_phy() function,
add a label at the end and use goto to jump there.
Additionally, respect the DC logger and let it log the function
even when it returns early.

Suggested-by: Ray Wu <Ray.Wu@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
---
 drivers/gpu/drm/amd/display/dc/link/link_factory.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
index 7989baf3843c..b177069e4807 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
@@ -529,16 +529,16 @@ static bool construct_phy(struct dc_link *link,
 	enc_init_data.transmitter = translate_encoder_to_transmitter(enc_init_data.encoder);
 	enc_init_data.analog_engine = find_analog_engine(link);
 
-	if (!transmitter_supported(enc_init_data.transmitter) &&
-		!analog_engine_supported(enc_init_data.analog_engine)) {
-		DC_LOG_WARNING("link_id %d has unsupported encoder\n", link->link_id.id);
-		return false;
-	}
-
 	link->ep_type = DISPLAY_ENDPOINT_PHY;
 
 	DC_LOG_DC("BIOS object table - link_id: %d", link->link_id.id);
 
+	if (!transmitter_supported(enc_init_data.transmitter) &&
+	    !analog_engine_supported(enc_init_data.analog_engine)) {
+		DC_LOG_WARNING("link_id %d has unsupported encoder\n", link->link_id.id);
+		goto unsupported_fail;
+	}
+
 	if (bios->funcs->get_disp_connector_caps_info) {
 		bios->funcs->get_disp_connector_caps_info(bios, link->link_id, &disp_connect_caps_info);
 		link->is_internal_display = disp_connect_caps_info.INTERNAL_DISPLAY;
@@ -805,6 +805,7 @@ static bool construct_phy(struct dc_link *link,
 		link->hpd_gpio = NULL;
 	}
 
+unsupported_fail:
 	DC_LOG_DC("BIOS object table - %s failed.\n", __func__);
 	return false;
 }
-- 
2.51.1

[PATCH 4/4] drm/amd/display: Move analog check to dce110_hwseq

[Timur Kristóf]

Instead of checking that the signal is analog before calling the
HWSS disable_audio_stream() function to disable audio, move
the check inside the HWSS function.

Suggested-by: Ray Wu <Ray.Wu@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c | 6 ++++--
 drivers/gpu/drm/amd/display/dc/link/link_dpms.c           | 3 +--
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index 3005115c8505..94e66d96c403 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -1139,6 +1139,9 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx)
 	if (!pipe_ctx || !pipe_ctx->stream)
 		return;
 
+	if (dc_is_rgb_signal(pipe_ctx->stream->signal))
+		return;
+
 	dc = pipe_ctx->stream->ctx->dc;
 	clk_mgr = dc->clk_mgr;
 	link_hwss = get_link_hwss(pipe_ctx->stream->link, &pipe_ctx->link_res);
@@ -1193,8 +1196,7 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx)
 		pipe_ctx->stream_res.stream_enc->funcs->stop_dp_info_packets(
 			pipe_ctx->stream_res.stream_enc);
 
-	if (!dc_is_rgb_signal(pipe_ctx->stream->signal))
-		dc->hwss.disable_audio_stream(pipe_ctx);
+	dc->hwss.disable_audio_stream(pipe_ctx);
 
 	link_hwss->reset_stream_encoder(pipe_ctx);
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 4ddcdc222913..f7ac26598c39 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -2374,8 +2374,7 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx)
 			set_avmute(pipe_ctx, true);
 	}
 
-	if (!dc_is_rgb_signal(pipe_ctx->stream->signal))
-		dc->hwss.disable_audio_stream(pipe_ctx);
+	dc->hwss.disable_audio_stream(pipe_ctx);
 
 	update_psp_stream_config(pipe_ctx, true);
 	dc->hwss.blank_stream(pipe_ctx);
-- 
2.51.1

Re: [PATCH 1/4] drm/amd/display: Fix warning for analog stream encoders

[Kees Cook]

On Thu, Nov 13, 2025 at 05:33:45PM +0100, Timur Kristóf wrote:
> Fixes the following warning that some users are reporting
> with some kernel configurations:
> 
> "positional initialization of field in 'struct' declared
> with 'designated_init' attribute"

Yes, seen with CONFIG_RANDSTRUCT=y

Thanks for the fix!

> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
> Reviewed-by: Harry Wentland <harry.wentland@amd.com>


Acked-by: Kees Cook <kees@kernel.org>

-- 
Kees Cook

Re: [PATCH 0/4] drm/amd/display: Follow-up patches to analog support

[Mario Limonciello]

On 11/13/2025 10:33 AM, Timur Kristóf wrote:
> This series addresses some feedback that I received after the DC analog
> series support landed. Harry already gave a review to the patches.
> Please feel free to let me know if you have any other feedback that I
> should address regarding the analog connector support.
> 
> Timur Kristóf (4):
>    drm/amd/display: Fix warning for analog stream encoders
>    drm/amd/display: Cleanup uses of the analog flag
>    drm/amd/display: Cleanup early return in construct_phy
>    drm/amd/display: Move analog check to dce110_hwseq
> 
>   drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15 ++++++++++++---
>   .../drm/amd/display/dc/dce/dce_stream_encoder.c   |  2 +-
>   .../drm/amd/display/dc/hwss/dce110/dce110_hwseq.c |  6 ++++--
>   .../gpu/drm/amd/display/dc/link/link_detection.c  |  2 --
>   drivers/gpu/drm/amd/display/dc/link/link_dpms.c   |  3 +--
>   .../gpu/drm/amd/display/dc/link/link_factory.c    | 13 +++++++------
>   6 files changed, 25 insertions(+), 16 deletions(-)
> 

As these are reviewed by Harry and display promotion testing wouldn't 
catch issues with analog I've queued them up for the internal CI to run 
on and merge.

Will let you know if anything crops up in that, and if any future 
promotion tests fail and it bisects to these the promoter will respond 
accordingly.

Re: [PATCH 0/4] drm/amd/display: Follow-up patches to analog support

[timur.kristof]

On Mon, 2025-11-17 at 16:18 -0600, Mario Limonciello wrote:
> On 11/13/2025 10:33 AM, Timur Kristóf wrote:
> > This series addresses some feedback that I received after the DC
> > analog
> > series support landed. Harry already gave a review to the patches.
> > Please feel free to let me know if you have any other feedback that
> > I
> > should address regarding the analog connector support.
> > 
> > Timur Kristóf (4):
> >    drm/amd/display: Fix warning for analog stream encoders
> >    drm/amd/display: Cleanup uses of the analog flag
> >    drm/amd/display: Cleanup early return in construct_phy
> >    drm/amd/display: Move analog check to dce110_hwseq
> > 
> >   drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15
> > ++++++++++++---
> >   .../drm/amd/display/dc/dce/dce_stream_encoder.c   |  2 +-
> >   .../drm/amd/display/dc/hwss/dce110/dce110_hwseq.c |  6 ++++--
> >   .../gpu/drm/amd/display/dc/link/link_detection.c  |  2 --
> >   drivers/gpu/drm/amd/display/dc/link/link_dpms.c   |  3 +--
> >   .../gpu/drm/amd/display/dc/link/link_factory.c    | 13 +++++++---
> > ---
> >   6 files changed, 25 insertions(+), 16 deletions(-)
> > 
> 
> As these are reviewed by Harry and display promotion testing wouldn't
> catch issues with analog I've queued them up for the internal CI to
> run 
> on and merge.
> 
> Will let you know if anything crops up in that, and if any future 
> promotion tests fail and it bisects to these the promoter will
> respond 
> accordingly.

That sounds reasonable. Thank you!

Timur

Re: [PATCH 1/4] drm/amd/display: Fix warning for analog stream encoders

[Alex Hung]

On 11/17/25 13:16, Kees Cook wrote:
> On Thu, Nov 13, 2025 at 05:33:45PM +0100, Timur Kristóf wrote:
>> Fixes the following warning that some users are reporting
>> with some kernel configurations:
>>
>> "positional initialization of field in 'struct' declared
>> with 'designated_init' attribute"
> 
> Yes, seen with CONFIG_RANDSTRUCT=y
> 
> Thanks for the fix!
> 
>> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
>> Reviewed-by: Harry Wentland <harry.wentland@amd.com>
> 
> 
> Acked-by: Kees Cook <kees@kernel.org>
> 

Thanks and this patch is now merged to amd-staging-drm-next, but I found 
it causes *some* compiler to fail.

I'd like to test whether the below change is able to fix both issues.

-static const struct stream_encoder_funcs dce110_an_str_enc_funcs = {};
+static const struct stream_encoder_funcs dce110_an_str_enc_funcs = {
+       .dp_set_stream_attribute = NULL,
+};

However, I could not reproduce ""positional initialization of field in 
'struct' declared 'designated_init' attribute" in my local setup, with 
either gcc or clang and "CONFIG_RANDSTRUCT=y"

Can someone try the above changes for me? really appreciate it. I will 
create a patch if it works.

Alex Hung