Skip to contentSkip to conversation

modesetting: re-set the crtc's mode when link-status goes BAD

PATCH In review Michel Dänzer 3 participants 3 messages Aug 22, 2017 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 modesetting: re-set the crtc's mode when link-status goes BAD

From: Martin Peres <[email protected]>

Despite all the careful planning of the kernel, a link may become insufficient to handle the currently-set mode. At this point, the kernel should mark this particular configuration as being broken and potentially prune the mode before setting the offending connector's link-status to BAD and send the userspace a hotplug event. This may happen right after a modeset or later on.

Upon receiving a hot-plug event, we iterate through the connectors to re-apply the currently-set mode on all the connectors that have a link-status property set to BAD. The kernel may be able to get the link to work by dropping to using a lower link bpp (with the same display bpp). However, the modeset may fail if the kernel has pruned the mode, so to make users aware of this problem a warning is outputed in the logs to warn about having a potentially-black display.

This patch does not modify the current behaviour of always propagating the events to the randr clients. This allows desktop environments to re-probe the connectors and select a new resolution based on the new (currated) mode list if a mode disapeared. This behaviour is expected in order to pass the Display Port compliance tests.

(Ported from xserver commit bcee1b76aa0db8525b491485e90b8740763d7de6)

[ Michel: Bump libdrm dependency to >= 2.4.78 for
DRM_MODE_LINK_STATUS_BAD ]
(Ported from radeon commit 0472a605e0ec8fec1892bbc3a84698b7ef9c5296)

---
 configure.ac          |  2 +-
 src/drmmode_display.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+), 1 deletion(-)
configure.ac +1 -1
index 0ff2bdfaa..6d7cee476 100644
@@ -70,7 +70,7 @@ XORG_DRIVER_CHECK_EXT(XV, videoproto)
 XORG_DRIVER_CHECK_EXT(DPMSExtension, xextproto)
 
 # Checks for libraries.
-PKG_CHECK_MODULES(LIBDRM, [libdrm >= 2.4.72])
+PKG_CHECK_MODULES(LIBDRM, [libdrm >= 2.4.78])
 PKG_CHECK_MODULES(LIBDRM_AMDGPU, [libdrm_amdgpu >= 2.4.72])
 PKG_CHECK_MODULES(GBM, [gbm])
 
src/drmmode_display.c +43 -0

49 lines not on this page · raw message

Signature · 6 lines
2.14.1

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): 20170822094819.8006-1-michel-otUistvHUpPR7s880joybQ@public.gmane.org

#2 lore raw
Quoted 23 lines · Michel Dänzer

From: Martin Peres <[email protected]>

Despite all the careful planning of the kernel, a link may become insufficient to handle the currently-set mode. At this point, the kernel should mark this particular configuration as being broken and potentially prune the mode before setting the offending connector's link-status to BAD and send the userspace a hotplug event. This may happen right after a modeset or later on.

Upon receiving a hot-plug event, we iterate through the connectors to re-apply the currently-set mode on all the connectors that have a link-status property set to BAD. The kernel may be able to get the link to work by dropping to using a lower link bpp (with the same display bpp). However, the modeset may fail if the kernel has pruned the mode, so to make users aware of this problem a warning is outputed in the logs to warn about having a potentially-black display.

This patch does not modify the current behaviour of always propagating the events to the randr clients. This allows desktop environments to re-probe the connectors and select a new resolution based on the new (currated) mode list if a mode disapeared. This behaviour is expected in order to pass the Display Port compliance tests.

Quoted 7 lines

(Ported from xserver commit bcee1b76aa0db8525b491485e90b8740763d7de6)

[ Michel: Bump libdrm dependency to >= 2.4.78 for
DRM_MODE_LINK_STATUS_BAD ]
(Ported from radeon commit 0472a605e0ec8fec1892bbc3a84698b7ef9c5296)

Thanks, and

I'll send an amd/display kernel patch soon to set link-status to bad.

Harry

Quoted 63 lines
---
 configure.ac          |  2 +-
 src/drmmode_display.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+), 1 deletion(-)
configure.ac +1 -1
index 0ff2bdfaa..6d7cee476 100644
@@ -70,7 +70,7 @@ XORG_DRIVER_CHECK_EXT(XV, videoproto)
 XORG_DRIVER_CHECK_EXT(DPMSExtension, xextproto)
 
 # Checks for libraries.
-PKG_CHECK_MODULES(LIBDRM, [libdrm >= 2.4.72])
+PKG_CHECK_MODULES(LIBDRM, [libdrm >= 2.4.78])
 PKG_CHECK_MODULES(LIBDRM_AMDGPU, [libdrm_amdgpu >= 2.4.72])
 PKG_CHECK_MODULES(GBM, [gbm])
 
src/drmmode_display.c +36 -0
index ad3325be6..17efde8e8 100644
@@ -2627,6 +2627,49 @@ amdgpu_mode_hotplug(ScrnInfoPtr scrn, drmmode_ptr drmmode)
 	Bool changed = FALSE;
 	int num_dvi = 0, num_hdmi = 0;
 
+	/* Try to re-set the mode on all the connectors with a BAD link-state:
+	 * This may happen if a link degrades and a new modeset is necessary, using
+	 * different link-training parameters. If the kernel found that the current
+	 * mode is not achievable anymore, it should have pruned the mode before
+	 * sending the hotplug event. Try to re-set the currently-set mode to keep
+	 * the display alive, this will fail if the mode has been pruned.
+	 * In any case, we will send randr events for the Desktop Environment to
+	 * deal with it, if it wants to.
+	 */
+	for (i = 0; i < config->num_output; i++) {
+		xf86OutputPtr output = config->output[i];
+		drmmode_output_private_ptr drmmode_output = output->driver_private;
+		uint32_t con_id = drmmode_output->mode_output->connector_id;
+		drmModeConnectorPtr koutput;
+
+		/* Get an updated view of the properties for the current connector and
+		 * look for the link-status property
+		 */
+		koutput = drmModeGetConnectorCurrent(pAMDGPUEnt->fd, con_id);
+		for (j = 0; koutput && j < koutput->count_props; j++) {
+			drmModePropertyPtr props;
+			props = drmModeGetProperty(pAMDGPUEnt->fd, koutput->props[j]);
+			if (props && props->flags & DRM_MODE_PROP_ENUM &&
+			    !strcmp(props->name, "link-status") &&
+			    koutput->prop_values[j] == DRM_MODE_LINK_STATUS_BAD) {
+				xf86CrtcPtr crtc = output->crtc;
+				if (!crtc)
+					continue;
+
+				/* the connector got a link failure, re-set the current mode */
+				drmmode_set_mode_major(crtc, &crtc->mode, crtc->rotation,
+						       crtc->x, crtc->y);
+
+				xf86DrvMsg(scrn->scrnIndex, X_WARNING,
+					   "hotplug event: connector %u's link-state is BAD, "
+					   "tried resetting the current mode. You may be left "

src/drmmode_display.c amdgpu_mode_hotplug(ScrnInfoPtr scrn, drmmode_ptr drmmode) · line 2675

Quoted 10 lines
src/drmmode_display.c +7 -0
+					   "with a black screen if this fails...\n", con_id);
+			}
+			drmModeFreeProperty(props);
+		}
+		drmModeFreeConnector(koutput);
+	}
+
 	mode_res = drmModeGetResources(pAMDGPUEnt->fd);
 	if (!mode_res)
 		goto out;

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Reviewed-by Alex Deucher
#3 lore raw
In reply to a message not in the archive
Quoted 31 lines

-----Original Message-----
From: amd-gfx [mailto:[email protected]] On Behalf Of Michel Dänzer
Sent: Tuesday, August 22, 2017 5:48 AM
To: [email protected]
Subject: [PATCH xf86-video-amdgpu] modesetting: re-set the crtc's mode when link-status goes BAD

From: Martin Peres <[email protected]>

Despite all the careful planning of the kernel, a link may become insufficient to handle the currently-set mode. At this point, the kernel should mark this particular configuration as being broken and potentially prune the mode before setting the offending connector's link-status to BAD and send the userspace a hotplug event. This may happen right after a modeset or later on.

Upon receiving a hot-plug event, we iterate through the connectors to re-apply the currently-set mode on all the connectors that have a link-status property set to BAD. The kernel may be able to get the link to work by dropping to using a lower link bpp (with the same display bpp). However, the modeset may fail if the kernel has pruned the mode, so to make users aware of this problem a warning is outputed in the logs to warn about having a potentially-black display.

This patch does not modify the current behaviour of always propagating the events to the randr clients. This allows desktop environments to re-probe the connectors and select a new resolution based on the new (currated) mode list if a mode disapeared. This behaviour is expected in order to pass the Display Port compliance tests.

Quoted 8 lines

(Ported from xserver commit
bcee1b76aa0db8525b491485e90b8740763d7de6)

[ Michel: Bump libdrm dependency to >= 2.4.78 for
DRM_MODE_LINK_STATUS_BAD ]
(Ported from radeon commit 0472a605e0ec8fec1892bbc3a84698b7ef9c5296)

Quoted 94 lines
---
 configure.ac          |  2 +-
 src/drmmode_display.c | 43
+++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+), 1 deletion(-)
configure.ac +1 -1
index 0ff2bdfaa..6d7cee476 100644
@@ -70,7 +70,7 @@ XORG_DRIVER_CHECK_EXT(XV, videoproto)
 XORG_DRIVER_CHECK_EXT(DPMSExtension, xextproto)
 
 # Checks for libraries.
-PKG_CHECK_MODULES(LIBDRM, [libdrm >= 2.4.72])
+PKG_CHECK_MODULES(LIBDRM, [libdrm >= 2.4.78])
 PKG_CHECK_MODULES(LIBDRM_AMDGPU, [libdrm_amdgpu >= 2.4.72])
 PKG_CHECK_MODULES(GBM, [gbm])
 
src/drmmode_display.c +2 -0
index ad3325be6..17efde8e8 100644
@@ -2627,6 +2627,49 @@ amdgpu_mode_hotplug(ScrnInfoPtr scrn,
 drmmode_ptr drmmode)
 	Bool changed = FALSE;
 	int num_dvi = 0, num_hdmi = 0;
 
+	/* Try to re-set the mode on all the connectors with a BAD link-state:
+	 * This may happen if a link degrades and a new modeset is

necessary, using

src/drmmode_display.c +1 -0
+	 * different link-training parameters. If the kernel found that the

current

src/drmmode_display.c +1 -0
+	 * mode is not achievable anymore, it should have pruned the mode

before

src/drmmode_display.c +1 -0
+	 * sending the hotplug event. Try to re-set the currently-set mode to

keep

src/drmmode_display.c +2 -0
+	 * the display alive, this will fail if the mode has been pruned.
+	 * In any case, we will send randr events for the Desktop

Environment to

src/drmmode_display.c +5 -0
+	 * deal with it, if it wants to.
+	 */
+	for (i = 0; i < config->num_output; i++) {
+		xf86OutputPtr output = config->output[i];
+		drmmode_output_private_ptr drmmode_output = output-
Quoted 1 line

driver_private;

src/drmmode_display.c +1 -0
+		uint32_t con_id = drmmode_output->mode_output-
Quoted 1 line

connector_id;

src/drmmode_display.c +3 -0
+		drmModeConnectorPtr koutput;
+
+		/* Get an updated view of the properties for the current

connector and

src/drmmode_display.c +3 -0
+		 * look for the link-status property
+		 */
+		koutput = drmModeGetConnectorCurrent(pAMDGPUEnt-
Quoted 1 line

fd, con_id);

src/drmmode_display.c +3 -0
+		for (j = 0; koutput && j < koutput->count_props; j++) {
+			drmModePropertyPtr props;
+			props = drmModeGetProperty(pAMDGPUEnt->fd,

koutput->props[j]);

src/drmmode_display.c +1 -0
+			if (props && props->flags &

DRM_MODE_PROP_ENUM &&

src/drmmode_display.c +2 -0
+			    !strcmp(props->name, "link-status") &&
+			    koutput->prop_values[j] ==

DRM_MODE_LINK_STATUS_BAD) {

src/drmmode_display.c +5 -0
+				xf86CrtcPtr crtc = output->crtc;
+				if (!crtc)
+					continue;
+
+				/* the connector got a link failure, re-set the

current mode */

src/drmmode_display.c +1 -0
+				drmmode_set_mode_major(crtc, &crtc-
Quoted 1 line

mode, crtc->rotation,

src/drmmode_display.c +4 -0
+						       crtc->x, crtc->y);
+
+				xf86DrvMsg(scrn->scrnIndex, X_WARNING,
+					   "hotplug event: connector %u's link-

state is BAD, "

src/drmmode_display.c +1 -0
+					   "tried resetting the current mode.

You may be left "

src/drmmode_display.c +1 -0
+					   "with a black screen if this fails...\n",

con_id);

src/drmmode_display.c +6 -1
+			}
+			drmModeFreeProperty(props);
+		}
+		drmModeFreeConnector(koutput);
+	}
+
 	mode_res = drmModeGetResources(pAMDGPUEnt->fd);
 	if (!mode_res)
 		goto out;
--

2.14.1

Quoted 4 lines

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx