[PATCH v6.1] drm/amd/display: Check dce_hwseq before dereferencing it

[PATCH v6.1] drm/amd/display: Check dce_hwseq before dereferencing it

[Rahul Sharma]

From: Alex Hung <alex.hung@amd.com>

[ Upstream b669507b637eb6b1aaecf347f193efccc65d756e commit ]

[WHAT]

hws was checked for null earlier in dce110_blank_stream, indicating hws
can be null, and should be checked whenever it is used.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 79db43611ff61280b6de58ce1305e0b2ecf675ad)
Cc: stable@vger.kernel.org
[ The context change is due to the commit 8e7b3f5435b3
("drm/amd/display: Add control flag to dc_stream_state to skip eDP BL off/link off")
and the commit a8728dbb4ba2 ("drm/amd/display: Refactor edp power
control") and the proper adoption is done. ]
Signed-off-by: Rahul Sharma <black.hawk@163.com>
---
 drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 508f5fe26848..c542d2ab9160 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -1233,7 +1233,8 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx)
 	struct dce_hwseq *hws = link->dc->hwseq;
 
 	if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
-		hws->funcs.edp_backlight_control(link, false);
+		if (hws)
+			hws->funcs.edp_backlight_control(link, false);
 		link->dc->hwss.set_abm_immediate_disable(pipe_ctx);
 	}
 
-- 
2.34.1

Re: [PATCH v6.1] drm/amd/display: Check dce_hwseq before dereferencing it

[Greg KH]

On Thu, Jan 15, 2026 at 12:19:19PM +0800, Rahul Sharma wrote:
> From: Alex Hung <alex.hung@amd.com>
> 
> [ Upstream b669507b637eb6b1aaecf347f193efccc65d756e commit ]
> 
> [WHAT]
> 
> hws was checked for null earlier in dce110_blank_stream, indicating hws
> can be null, and should be checked whenever it is used.
> 
> Cc: Mario Limonciello <mario.limonciello@amd.com>
> Cc: Alex Deucher <alexander.deucher@amd.com>
> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
> Signed-off-by: Alex Hung <alex.hung@amd.com>
> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
> (cherry picked from commit 79db43611ff61280b6de58ce1305e0b2ecf675ad)
> Cc: stable@vger.kernel.org
> [ The context change is due to the commit 8e7b3f5435b3
> ("drm/amd/display: Add control flag to dc_stream_state to skip eDP BL off/link off")
> and the commit a8728dbb4ba2 ("drm/amd/display: Refactor edp power
> control") and the proper adoption is done. ]
> Signed-off-by: Rahul Sharma <black.hawk@163.com>
> ---
>  drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)

We need 6.6.y backport first, before we can take this one, for obvious
reasons (i.e. you do not want to have a regression).  Can you submit
that one first and then this again?

thanks,

greg k-h

Re: Re: [PATCH v6.1] drm/amd/display: Check dce_hwseq before dereferencing it

[Greg KH]

On Fri, Jan 16, 2026 at 10:57:51AM +0800, Rahul Sharma wrote:
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> At 2026-01-15 19:44:04, "Greg KH" <gregkh@linuxfoundation.org> wrote:
> >On Thu, Jan 15, 2026 at 12:19:19PM +0800, Rahul Sharma wrote:
> >> From: Alex Hung <alex.hung@amd.com>
> >> 
> >> [ Upstream b669507b637eb6b1aaecf347f193efccc65d756e commit ]
> >> 
> >> [WHAT]
> >> 
> >> hws was checked for null earlier in dce110_blank_stream, indicating hws
> >> can be null, and should be checked whenever it is used.
> >> 
> >> Cc: Mario Limonciello <mario.limonciello@amd.com>
> >> Cc: Alex Deucher <alexander.deucher@amd.com>
> >> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
> >> Signed-off-by: Alex Hung <alex.hung@amd.com>
> >> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
> >> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
> >> (cherry picked from commit 79db43611ff61280b6de58ce1305e0b2ecf675ad)
> >> Cc: stable@vger.kernel.org
> >> [ The context change is due to the commit 8e7b3f5435b3
> >> ("drm/amd/display: Add control flag to dc_stream_state to skip eDP BL off/link off")
> >> and the commit a8728dbb4ba2 ("drm/amd/display: Refactor edp power
> >> control") and the proper adoption is done. ]
> >> Signed-off-by: Rahul Sharma <black.hawk@163.com>
> >> ---
> >>  drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 3 ++-
> >>  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> >We need 6.6.y backport first, before we can take this one, for obvious
> >reasons (i.e. you do not want to have a regression).  Can you submit
> >that one first and then this again?
> 
> >
> Thanks Greg, I have sent out the backported patch for 6.6.y.

Great, can you resend this one as well?  It's long-gone from my mail
system.

thanks,

greg k-h

[PATCH v6.1] drm/amd/display: Check dce_hwseq before dereferencing it

[Rahul Sharma]

From: Alex Hung <alex.hung@amd.com>

[ Upstream commit b669507b637eb6b1aaecf347f193efccc65d756e ]

[WHAT]

hws was checked for null earlier in dce110_blank_stream, indicating hws
can be null, and should be checked whenever it is used.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 79db43611ff61280b6de58ce1305e0b2ecf675ad)
Cc: stable@vger.kernel.org
[ The context change is due to the commit 8e7b3f5435b3
("drm/amd/display: Add control flag to dc_stream_state to skip eDP BL off/link off")
and the commit a8728dbb4ba2 ("drm/amd/display: Refactor edp power
control") and the proper adoption is done. ]
Signed-off-by: Rahul Sharma <black.hawk@163.com>
---
 drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 508f5fe26848..c542d2ab9160 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -1233,7 +1233,8 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx)
 	struct dce_hwseq *hws = link->dc->hwseq;
 
 	if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
-		hws->funcs.edp_backlight_control(link, false);
+		if (hws)
+			hws->funcs.edp_backlight_control(link, false);
 		link->dc->hwss.set_abm_immediate_disable(pipe_ctx);
 	}
 
-- 
2.34.1