Skip to contentSkip to conversation

drm/dp_mst: Fix return code on sideband message failure

PATCH In review Kuogee Hsieh 4 participants 14 messages Jul 6, 2021 – Sep 7, 2021

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 drm/dp_mst: Fix return code on sideband message failure

From: Rajkumar Subbiah <[email protected]>

Commit 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests") added some debug code for sideband message tracing. But it seems to have unintentionally changed the behavior on sideband message failure. It catches and returns failure only if DRM_UT_DP is enabled. Otherwise it ignores the error code and returns success. So on an MST unplug, the caller is unaware that the clear payload message failed and ends up waiting for 4 seconds for the response. Fixes the issue by returning the proper error code.

Changes in V2:
-- Revise commit text as review comment
-- add Fixes text

Changes in V3:
-- remove "unlikely" optimization

Fixes 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests")
---
 drivers/gpu/drm/drm_dp_mst_topology.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/gpu/drm/drm_dp_mst_topology.c +6 -4
index 1590144..df91110 100644
@@ -2887,11 +2887,13 @@ static int process_single_tx_qlock(struct drm_dp_mst_topology_mgr *mgr,
 	idx += tosend + 1;
 
 	ret = drm_dp_send_sideband_msg(mgr, up, chunk, idx);
-	if (unlikely(ret) && drm_debug_enabled(DRM_UT_DP)) {
-		struct drm_printer p = drm_debug_printer(DBG_PREFIX);
+	if (ret) {
+		if (drm_debug_enabled(DRM_UT_DP)) {
+			struct drm_printer p = drm_debug_printer(DBG_PREFIX);
 
-		drm_printf(&p, "sideband msg failed to send\n");
-		drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+			drm_printf(&p, "sideband msg failed to send\n");
+			drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+		}
 		return ret;
 	}
 
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
Reviewed-by on v3 1/1
#1 lore raw
Quoted 18 lines · Kuogee Hsieh

From: Rajkumar Subbiah <[email protected]>

Commit 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests") added some debug code for sideband message tracing. But it seems to have unintentionally changed the behavior on sideband message failure. It catches and returns failure only if DRM_UT_DP is enabled. Otherwise it ignores the error code and returns success. So on an MST unplug, the caller is unaware that the clear payload message failed and ends up waiting for 4 seconds for the response. Fixes the issue by returning the proper error code.

Changes in V2:
-- Revise commit text as review comment
-- add Fixes text

Changes in V3:
-- remove "unlikely" optimization

Quoted 6 lines
Fixes 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests")
Quoted 25 lines
---
 drivers/gpu/drm/drm_dp_mst_topology.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/gpu/drm/drm_dp_mst_topology.c +6 -4
index 1590144..df91110 100644
@@ -2887,11 +2887,13 @@ static int process_single_tx_qlock(struct drm_dp_mst_topology_mgr *mgr,
 	idx += tosend + 1;
 
 	ret = drm_dp_send_sideband_msg(mgr, up, chunk, idx);
-	if (unlikely(ret) && drm_debug_enabled(DRM_UT_DP)) {
-		struct drm_printer p = drm_debug_printer(DBG_PREFIX);
+	if (ret) {
+		if (drm_debug_enabled(DRM_UT_DP)) {
+			struct drm_printer p = drm_debug_printer(DBG_PREFIX);
 
-		drm_printf(&p, "sideband msg failed to send\n");
-		drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+			drm_printf(&p, "sideband msg failed to send\n");
+			drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+		}
 		return ret;
 	}
Signature · 1 line
Jani Nikula, Intel Open Source Graphics Center
#2 lore raw
Quoted 19 lines · Jani Nikula
Quoted 18 lines · Kuogee Hsieh

From: Rajkumar Subbiah <[email protected]>

Commit 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests") added some debug code for sideband message tracing. But it seems to have unintentionally changed the behavior on sideband message failure. It catches and returns failure only if DRM_UT_DP is enabled. Otherwise it ignores the error code and returns success. So on an MST unplug, the caller is unaware that the clear payload message failed and ends up waiting for 4 seconds for the response. Fixes the issue by returning the proper error code.

Changes in V2:
-- Revise commit text as review comment
-- add Fixes text

Changes in V3:
-- remove "unlikely" optimization

Quoted 11 lines
Quoted 6 lines
Fixes 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests")
Quoted 1 line

---

Lyude,
Any comments from you?
Thanks,

Quoted 25 lines
Quoted 25 lines
 drivers/gpu/drm/drm_dp_mst_topology.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/gpu/drm/drm_dp_mst_topology.c +6 -4
index 1590144..df91110 100644
@@ -2887,11 +2887,13 @@ static int process_single_tx_qlock(struct 
 drm_dp_mst_topology_mgr *mgr,
 	idx += tosend + 1;
 
 	ret = drm_dp_send_sideband_msg(mgr, up, chunk, idx);
-	if (unlikely(ret) && drm_debug_enabled(DRM_UT_DP)) {
-		struct drm_printer p = drm_debug_printer(DBG_PREFIX);
+	if (ret) {
+		if (drm_debug_enabled(DRM_UT_DP)) {
+			struct drm_printer p = drm_debug_printer(DBG_PREFIX);
 
-		drm_printf(&p, "sideband msg failed to send\n");
-		drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+			drm_printf(&p, "sideband msg failed to send\n");
+			drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+		}
 		return ret;
 	}
Reviewed-by
#3 lore raw
Quoted 30 lines · On Tue, 2021-07-13 at 15:24 -0700, [email protected] wrote:
Quoted 29 lines · Jani Nikula
Quoted 27 lines · Kuogee Hsieh

From: Rajkumar Subbiah <[email protected]>

Commit 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests") added some debug code for sideband message tracing. But it seems to have unintentionally changed the behavior on sideband message
failure. It catches and returns failure only if DRM_UT_DP is enabled. Otherwise it ignores the error code and returns success. So on an MST unplug, the caller is unaware that the clear payload message failed and
ends up waiting for 4 seconds for the response. Fixes the issue by returning the proper error code.

Changes in V2:
-- Revise commit text as review comment
-- add Fixes text

Changes in V3:
-- remove "unlikely" optimization

Fixes 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing +

selftests")

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

---

Lyude,
Any comments from you?
Thanks,

Hey! Sorry did I forget to respond to this?

Quoted 29 lines
Quoted 28 lines
Quoted 28 lines

 drivers/gpu/drm/drm_dp_mst_topology.c | 10 ++++++----

 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/gpu/drm/drm_dp_mst_topology.c +6 -4
index 1590144..df91110 100644
@@ -2887,11 +2887,13 @@ static int process_single_tx_qlock(struct 
 drm_dp_mst_topology_mgr *mgr,
         idx += tosend + 1;
 
         ret = drm_dp_send_sideband_msg(mgr, up, chunk, idx);
-       if (unlikely(ret) && drm_debug_enabled(DRM_UT_DP)) {
-               struct drm_printer p = drm_debug_printer(DBG_PREFIX);
+       if (ret) {
+               if (drm_debug_enabled(DRM_UT_DP)) {
+                       struct drm_printer p =
 drm_debug_printer(DBG_PREFIX);
 
-               drm_printf(&p, "sideband msg failed to send\n");
-               drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+                       drm_printf(&p, "sideband msg failed to send\n");
+                       drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+               }
                 return ret;

        }

Signature · 3 lines
Cheers,
 Lyude Paul (she/her)
 Software Engineer at Red Hat
#4 lore raw
Quoted 34 lines · Lyude Paul
Quoted 33 lines · On Tue, 2021-07-13 at 15:24 -0700, [email protected] wrote:
Quoted 32 lines · Jani Nikula
Quoted 20 lines · Kuogee Hsieh

From: Rajkumar Subbiah <[email protected]>

Commit 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests") added some debug code for sideband message tracing. But it seems to have unintentionally changed the behavior on sideband message
failure. It catches and returns failure only if DRM_UT_DP is enabled. Otherwise it ignores the error code and returns success. So on an MST unplug, the caller is unaware that the clear payload message failed and
ends up waiting for 4 seconds for the response. Fixes the issue by returning the proper error code.

Changes in V2:
-- Revise commit text as review comment
-- add Fixes text

Changes in V3:
-- remove "unlikely" optimization

Quoted 7 lines
Fixes 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing +

selftests")

Quoted 9 lines
Quoted 4 lines
Quoted 1 line
Quoted 1 line

---

Lyude,
Any comments from you?
Thanks,

Hey! Sorry did I forget to respond to this?

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Quoted 29 lines
Quoted 29 lines
Quoted 28 lines
Quoted 28 lines

 drivers/gpu/drm/drm_dp_mst_topology.c | 10 ++++++----

 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/gpu/drm/drm_dp_mst_topology.c +6 -4
index 1590144..df91110 100644
@@ -2887,11 +2887,13 @@ static int process_single_tx_qlock(struct
 drm_dp_mst_topology_mgr *mgr,
         idx += tosend + 1;
 
         ret = drm_dp_send_sideband_msg(mgr, up, chunk, idx);
-       if (unlikely(ret) && drm_debug_enabled(DRM_UT_DP)) {
-               struct drm_printer p = drm_debug_printer(DBG_PREFIX);
+       if (ret) {
+               if (drm_debug_enabled(DRM_UT_DP)) {
+                       struct drm_printer p =
 drm_debug_printer(DBG_PREFIX);
 
-               drm_printf(&p, "sideband msg failed to send\n");
-               drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+                       drm_printf(&p, "sideband msg failed to send\n");
+                       drm_dp_mst_dump_sideband_msg_tx(&p, txmsg);
+               }
                 return ret;

        }

#5 lore raw
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I can go ahead and push it to drm-misc-next for you.

Signature · 3 lines
Cheers,
 Lyude Paul (she/her)
 Software Engineer at Red Hat
#6 lore raw
Quoted 8 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I can go ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.

#7 lore raw

Nice timing, you literally got me as I was 2 minutes away from leaving work for the day :P. I will go ahead and push it now.

BTW - in the future I recommend using dim to add Fixes: tags as it'll add Cc:
to stable as appropriate (this patch in particular should be Cc:
[email protected] # v5.3+). will add these tags when I push it

Quoted 14 lines · On Tue, 2021-07-27 at 15:41 -0700, [email protected] wrote:
Quoted 9 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I can go
ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.

Signature · 3 lines
Cheers,
 Lyude Paul (she/her)
 Software Engineer at Red Hat
#8 lore raw
Quoted 3 lines · Lyude Paul

Nice timing, you literally got me as I was 2 minutes away from leaving work for the day :P. I will go ahead and push it now.

Hi Lyude,

Had you pushed this patch yet?
We still did not see this patch at msm-nex and v5.10 branch. Thanks,

Quoted 19 lines

BTW - in the future I recommend using dim to add Fixes: tags as it'll add Cc:
to stable as appropriate (this patch in particular should be Cc:
[email protected] # v5.3+). will add these tags when I push it

Quoted 14 lines · On Tue, 2021-07-27 at 15:41 -0700, [email protected] wrote:
Quoted 9 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I can go
ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.

#9 lore raw

The patch was pushed yes (was part of drm-misc-next-2021-07-29), seems like it just hasn't trickled down to linus's branch quite yet.

Quoted 34 lines · On Wed, 2021-08-25 at 09:06 -0700, [email protected] wrote:
Quoted 4 lines · Lyude Paul

Nice timing, you literally got me as I was 2 minutes away from leaving work
for the day :P. I will go ahead and push it now.

Hi Lyude,

Had you pushed this patch yet?
We still did not see this patch at msm-nex and v5.10 branch. Thanks,

Quoted 21 lines

BTW - in the future I recommend using dim to add Fixes: tags as it'll add Cc:
to stable as appropriate (this patch in particular should be Cc:

[email protected] # v5.3+). will add these tags when I push it
Quoted 15 lines · On Tue, 2021-07-27 at 15:41 -0700, [email protected] wrote:
Quoted 10 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I can
go
ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.

Signature · 3 lines
Cheers,
 Lyude Paul (she/her)
 Software Engineer at Red Hat
#10 lore raw
Quoted 2 lines · Lyude Paul

The patch was pushed yes (was part of drm-misc-next-2021-07-29), seems like it just hasn't trickled down to linus's branch quite yet.

Hi Stephen B,

Would you mind back porting this patch to V5.10 branch? It will have lots of helps for us to support display port MST case. Thanks,

Quoted 36 lines · On Wed, 2021-08-25 at 09:06 -0700, [email protected] wrote:
Quoted 34 lines · Lyude Paul
Quoted 4 lines

Nice timing, you literally got me as I was 2 minutes away from leaving work
for the day :P. I will go ahead and push it now.

Hi Lyude,

Had you pushed this patch yet?
We still did not see this patch at msm-nex and v5.10 branch. Thanks,

Quoted 21 lines

BTW - in the future I recommend using dim to add Fixes: tags as it'll add Cc:
to stable as appropriate (this patch in particular should be Cc:

[email protected] # v5.3+). will add these tags when I push it
Quoted 15 lines · On Tue, 2021-07-27 at 15:41 -0700, [email protected] wrote:
Quoted 10 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I can
go
ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.

#11 lore raw
Quoted 10 lines · On Mon, 2021-08-30 at 08:56 -0700, [email protected] wrote:
Quoted 3 lines · Lyude Paul

The patch was pushed yes (was part of drm-misc-next-2021-07-29), seems like it
just hasn't trickled down to linus's branch quite yet.

Hi Stephen B,

Would you mind back porting this patch to V5.10 branch? It will have lots of helps for us to support display port MST case. Thanks,

I'm assuming you're talking to someone else? A little confused because I don't see a Stephen B in this thread

Quoted 41 lines
Quoted 40 lines · On Wed, 2021-08-25 at 09:06 -0700, [email protected] wrote:
Quoted 35 lines · Lyude Paul
Quoted 4 lines

Nice timing, you literally got me as I was 2 minutes away from leaving work
for the day :P. I will go ahead and push it now.

Hi Lyude,

Had you pushed this patch yet?
We still did not see this patch at msm-nex and v5.10 branch. Thanks,

Quoted 22 lines

BTW - in the future I recommend using dim to add Fixes: tags as it'll add Cc:
to stable as appropriate (this patch in particular should be Cc:

[email protected] # v5.3+). will add these tags when I push it
Quoted 16 lines · On Tue, 2021-07-27 at 15:41 -0700, [email protected] wrote:
Quoted 11 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I
can
go
ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.

Signature · 3 lines
Cheers,
 Lyude Paul (she/her)
 Software Engineer at Red Hat
#12 lore raw

Quoting Lyude Paul (2021-08-30 09:58:01)

Quoted 11 lines · On Mon, 2021-08-30 at 08:56 -0700, [email protected] wrote:
Quoted 10 lines · Lyude Paul
Quoted 3 lines

The patch was pushed yes (was part of drm-misc-next-2021-07-29), seems like it
just hasn't trickled down to linus's branch quite yet.

Hi Stephen B,

Would you mind back porting this patch to V5.10 branch? It will have lots of helps for us to support display port MST case. Thanks,

If the patch is tagged for stable then it will automatically be backported to the 5.10 stable release, or at least attempted. If you don't see it in the stable tree but it's in Linus' tree then you can submit the patch directly to stable. See Documentation/process/stable-kernel-rules.rst for more info.

Quoted 3 lines

I'm assuming you're talking to someone else? A little confused because I don't see a Stephen B in this thread

I assume it is me.

#13 lore raw
In reply to Lyude Paul
Quoted 6 lines · Lyude Paul
Quoted 5 lines · On Mon, 2021-08-30 at 08:56 -0700, [email protected] wrote:
Quoted 3 lines · Lyude Paul

The patch was pushed yes (was part of drm-misc-next-2021-07-29), seems like it
just hasn't trickled down to linus's branch quite yet.

Quoted 8 lines
Quoted 5 lines

Hi Stephen B,

Would you mind back porting this patch to V5.10 branch? It will have lots of helps for us to support display port MST case. Thanks,

I'm assuming you're talking to someone else? A little confused because I don't see a Stephen B in this thread

Yes,
I am asking Stephen B ([email protected]) helps to back port this patch to v5.10.

Quoted 42 lines
Quoted 42 lines
Quoted 41 lines · On Wed, 2021-08-25 at 09:06 -0700, [email protected] wrote:
Quoted 35 lines · Lyude Paul
Quoted 4 lines

Nice timing, you literally got me as I was 2 minutes away from leaving work
for the day :P. I will go ahead and push it now.

Hi Lyude,

Had you pushed this patch yet?
We still did not see this patch at msm-nex and v5.10 branch. Thanks,

Quoted 22 lines

BTW - in the future I recommend using dim to add Fixes: tags as it'll add Cc:
to stable as appropriate (this patch in particular should be Cc:

[email protected] # v5.3+). will add these tags when I push it
Quoted 16 lines · On Tue, 2021-07-27 at 15:41 -0700, [email protected] wrote:
Quoted 11 lines · Lyude Paul
Quoted 4 lines · On Thu, 2021-07-22 at 15:28 -0700, [email protected] wrote:

It looks like this patch is good to go (mainlined). Anything needed from me to do?
Thanks,

Do you have access for pushing this patch? If not let me know and I
can
go
ahead and push it to drm-misc-next for you.

no, I do not have access to drm-misc-next.
Please push it for me.
Thanks a lots.