Thread (45 messages) 45 messages, 4 authors, 2021-09-20

Re: [PATCH 14/15] drm/amd: cleanup: drm_modeset_lock_all() --> DRM_MODESET_LOCK_ALL_BEGIN()

From: Fernando Ramos <hidden>
Date: 2021-09-17 23:17:12
Also in: amd-gfx, dri-devel, intel-gfx, linux-arm-msm, linux-renesas-soc, linux-tegra, lkml, nouveau

quoted
+	struct drm_modeset_acquire_ctx ctx;
 	int r;
+	int ret;
Relocate ret with r please
Done!
quoted
-	drm_modeset_unlock_all(dev);
+	DRM_MODESET_LOCK_ALL_END(dev, ctx, ret);
You should check ret here
Done!

quoted
 	int r;
+	int ret;
Relocate ret with r
Done!

quoted
-	drm_modeset_unlock_all(dev);
+	DRM_MODESET_LOCK_ALL_END(dev, ctx, ret);
 
 	return 0;
Return ret
Done!

quoted
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 9b1fc54555ee..5196c1d26f87 100644
@@ -2661,13 +2657,18 @@ static void handle_hpd_irq(void *param)
 
 		amdgpu_dm_update_connector_after_detect(aconnector);
 
-		drm_modeset_lock_all(dev);
-		dm_restore_drm_connector_state(dev, connector);
-		drm_modeset_unlock_all(dev);
-
-		if (aconnector->base.force == DRM_FORCE_UNSPECIFIED)
-			drm_kms_helper_hotplug_event(dev);
+	} else {
+		goto out;
 	}
+
+	DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx, 0, ret);
+	dm_restore_drm_connector_state(dev, connector);
+	DRM_MODESET_LOCK_ALL_END(dev, ctx, ret);
Check ret here please
This function ("handle_hpd_irq") returns void. What would the appropriate way of
checking the error be?

quoted
@@ -2841,14 +2838,17 @@ static void handle_hpd_rx_irq(void *param)
 
 			amdgpu_dm_update_connector_after_detect(aconnector);
 
+		} else {
+			goto finish;
You used 'out' above and 'finish' here. It would be nice to be consistent with
naming, I see 'out' a lot more than 'finish', so my vote would be to change this
label to 'out'.
I originally used "out", but turns out there is already an "out" label in this
function :)

I then searched for other "go to end" labels and found "finish" being used in 
this same file.

But I can rename it to somehitng else ("out2" maybe?) to make it less confusing.

 
quoted
+		}
 
-			drm_modeset_lock_all(dev);
-			dm_restore_drm_connector_state(dev, connector);
-			drm_modeset_unlock_all(dev);
+		DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx, 0, ret);
+		dm_restore_drm_connector_state(dev, connector);
+		DRM_MODESET_LOCK_ALL_END(dev, ctx, ret);
Check ret here?
This is another irq-like function returning void.

What can we do here after having checked the error?

quoted
+#include <drm/drm_drv.h>
Top-level headers generally come above the driver headers. Also, now that I think
about this a bit more, all of the new includes in this set should probably be
for 'drm_modeset_lock.h' instead of 'drm_drv.h'.
Ok. Let me try that.

quoted
@@ -1259,13 +1257,16 @@ static ssize_t trigger_hotplug(struct file *f, const char __user *buf,
+	DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx, 0, ret);
+	dm_restore_drm_connector_state(dev, connector);
+	DRM_MODESET_LOCK_ALL_END(dev, ctx, ret);
Check ret here?
This is a .write file_operations function expected to return a "size". Is it ok
for it to return an error? I guess so, right?
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help