* [PATCH 1/7] drm/amd/display: Create dc_sink when EDID fail
2021-08-13 6:15 [PATCH 0/7] DC Patches Aug 13, 2021 Wayne Lin
@ 2021-08-13 6:15 ` Wayne Lin
2021-08-13 6:15 ` [PATCH 2/7] drm/amd/display: Fix multi-display support for idle opt workqueue Wayne Lin
` (3 subsequent siblings)
4 siblings, 0 replies; 7+ messages in thread
From: Wayne Lin @ 2021-08-13 6:15 UTC (permalink / raw)
To: amd-gfx
Cc: Harry.Wentland, Sunpeng.Li, Bhawanpreet.Lakha, Rodrigo.Siqueira,
Aurabindo.Pillai, Qingqing.Zhuo, bindu.r, Anson.Jacob, Wayne Lin,
Nicholas Kazlauskas, Wayne Lin
[Why]
While reading remote EDID via Startech 1-to-4 hub, occasionally we
won't get response in time and won't light up corresponding monitor.
Ideally, we can still add generic modes for userspace to choose to try
to light up the monitor and which is done in
drm_helper_probe_single_connector_modes(). So the main problem here is
that we fail .mode_valid since we don't create remote dc_sink for this
case.
[How]
Also add default dc_sink if we can't get the EDID.
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
---
.../display/amdgpu_dm/amdgpu_dm_mst_types.c | 23 +++++++++++++++++++
1 file changed, 23 insertions(+)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 5568d4e518e6..1bcba6943fd7 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -213,6 +213,29 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector)
drm_connector_update_edid_property(
&aconnector->base,
NULL);
+
+ DRM_DEBUG_KMS("Can't get EDID of %s. Add default remote sink.", connector->name);
+ if (!aconnector->dc_sink) {
+ struct dc_sink *dc_sink;
+ struct dc_sink_init_data init_params = {
+ .link = aconnector->dc_link,
+ .sink_signal = SIGNAL_TYPE_DISPLAY_PORT_MST };
+
+ dc_sink = dc_link_add_remote_sink(
+ aconnector->dc_link,
+ NULL,
+ 0,
+ &init_params);
+
+ if (!dc_sink) {
+ DRM_ERROR("Unable to add a remote sink\n");
+ return 0;
+ }
+
+ dc_sink->priv = aconnector;
+ aconnector->dc_sink = dc_sink;
+ }
+
return ret;
}
--
2.25.1
^ permalink raw reply related [flat|nested] 7+ messages in thread* [PATCH 2/7] drm/amd/display: Fix multi-display support for idle opt workqueue
2021-08-13 6:15 [PATCH 0/7] DC Patches Aug 13, 2021 Wayne Lin
2021-08-13 6:15 ` [PATCH 1/7] drm/amd/display: Create dc_sink when EDID fail Wayne Lin
@ 2021-08-13 6:15 ` Wayne Lin
2021-08-13 6:15 ` [PATCH 3/7] drm/amd/display: Use vblank control events for PSR enable/disable Wayne Lin
` (2 subsequent siblings)
4 siblings, 0 replies; 7+ messages in thread
From: Wayne Lin @ 2021-08-13 6:15 UTC (permalink / raw)
To: amd-gfx
Cc: Harry.Wentland, Sunpeng.Li, Bhawanpreet.Lakha, Rodrigo.Siqueira,
Aurabindo.Pillai, Qingqing.Zhuo, bindu.r, Anson.Jacob,
Nicholas Kazlauskas, Roman Li, Wayne Lin
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
[Why]
The current implementation for idle optimization support only has a
single work item that gets reshuffled into the system workqueue
whenever we receive an enable or disable event.
We can have mismatched events if the work hasn't been processed or if
we're getting control events from multiple displays at once.
This fixes this issue and also makes the implementation usable for
PSR control - which will be addressed in another patch.
[How]
We need to be able to flush remaining work out on demand for driver stop
and psr disable so create a driver specific workqueue instead of using
the system one. The workqueue will be single threaded to guarantee the
ordering of enable/disable events.
Refactor the queue to allocate the control work and deallocate it
after processing it.
Pass the acrtc directly to make it easier to handle psr enable/disable
in a later patch.
Rename things to indicate that it's not just MALL specific.
Reviewed-by: Roman Li <Roman.Li@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
---
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 62 ++++++++-----------
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 21 ++++---
2 files changed, 36 insertions(+), 47 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 3e28f17c84fa..f88b6c5b83cd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1044,10 +1044,10 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_
}
#endif
#if defined(CONFIG_DRM_AMD_DC_DCN)
-static void event_mall_stutter(struct work_struct *work)
+static void vblank_control_worker(struct work_struct *work)
{
-
- struct vblank_workqueue *vblank_work = container_of(work, struct vblank_workqueue, mall_work);
+ struct vblank_control_work *vblank_work =
+ container_of(work, struct vblank_control_work, work);
struct amdgpu_display_manager *dm = vblank_work->dm;
mutex_lock(&dm->dc_lock);
@@ -1062,22 +1062,9 @@ static void event_mall_stutter(struct work_struct *work)
DRM_DEBUG_KMS("Allow idle optimizations (MALL): %d\n", dm->active_vblank_irq_count == 0);
mutex_unlock(&dm->dc_lock);
+ kfree(vblank_work);
}
-static struct vblank_workqueue *vblank_create_workqueue(struct amdgpu_device *adev, struct dc *dc)
-{
- struct vblank_workqueue *vblank_work;
-
- vblank_work = kzalloc(sizeof(*vblank_work), GFP_KERNEL);
- if (ZERO_OR_NULL_PTR(vblank_work)) {
- kfree(vblank_work);
- return NULL;
- }
-
- INIT_WORK(&vblank_work->mall_work, event_mall_stutter);
-
- return vblank_work;
-}
#endif
static int amdgpu_dm_init(struct amdgpu_device *adev)
{
@@ -1220,12 +1207,10 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
#if defined(CONFIG_DRM_AMD_DC_DCN)
if (adev->dm.dc->caps.max_links > 0) {
- adev->dm.vblank_workqueue = vblank_create_workqueue(adev, adev->dm.dc);
-
- if (!adev->dm.vblank_workqueue)
+ adev->dm.vblank_control_workqueue =
+ create_singlethread_workqueue("dm_vblank_control_workqueue");
+ if (!adev->dm.vblank_control_workqueue)
DRM_ERROR("amdgpu: failed to initialize vblank_workqueue.\n");
- else
- DRM_DEBUG_DRIVER("amdgpu: vblank_workqueue init done %p.\n", adev->dm.vblank_workqueue);
}
#endif
@@ -1298,6 +1283,13 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev)
{
int i;
+#if defined(CONFIG_DRM_AMD_DC_DCN)
+ if (adev->dm.vblank_control_workqueue) {
+ destroy_workqueue(adev->dm.vblank_control_workqueue);
+ adev->dm.vblank_control_workqueue = NULL;
+ }
+#endif
+
for (i = 0; i < adev->dm.display_indexes_num; i++) {
drm_encoder_cleanup(&adev->dm.mst_encoders[i].base);
}
@@ -1321,14 +1313,6 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev)
dc_deinit_callbacks(adev->dm.dc);
#endif
-#if defined(CONFIG_DRM_AMD_DC_DCN)
- if (adev->dm.vblank_workqueue) {
- adev->dm.vblank_workqueue->dm = NULL;
- kfree(adev->dm.vblank_workqueue);
- adev->dm.vblank_workqueue = NULL;
- }
-#endif
-
dc_dmub_srv_destroy(&adev->dm.dc->ctx->dmub_srv);
if (dc_enable_dmub_notifications(adev->dm.dc)) {
@@ -6000,7 +5984,7 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state);
#if defined(CONFIG_DRM_AMD_DC_DCN)
struct amdgpu_display_manager *dm = &adev->dm;
- unsigned long flags;
+ struct vblank_control_work *work;
#endif
int rc = 0;
@@ -6025,12 +6009,16 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
return 0;
#if defined(CONFIG_DRM_AMD_DC_DCN)
- spin_lock_irqsave(&dm->vblank_lock, flags);
- dm->vblank_workqueue->dm = dm;
- dm->vblank_workqueue->otg_inst = acrtc->otg_inst;
- dm->vblank_workqueue->enable = enable;
- spin_unlock_irqrestore(&dm->vblank_lock, flags);
- schedule_work(&dm->vblank_workqueue->mall_work);
+ work = kzalloc(sizeof(*work), GFP_ATOMIC);
+ if (!work)
+ return -ENOMEM;
+
+ INIT_WORK(&work->work, vblank_control_worker);
+ work->dm = dm;
+ work->acrtc = acrtc;
+ work->enable = enable;
+
+ queue_work(dm->vblank_control_workqueue, &work->work);
#endif
return 0;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index ab1670b16b02..c6b8b835b08a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -60,6 +60,7 @@ enum aux_return_code_type;
/* Forward declarations */
struct amdgpu_device;
+struct amdgpu_crtc;
struct drm_device;
struct dc;
struct amdgpu_bo;
@@ -86,16 +87,16 @@ struct dm_compressor_info {
};
/**
- * struct vblank_workqueue - Works to be executed in a separate thread during vblank
- * @mall_work: work for mall stutter
+ * struct vblank_control_work - Work data for vblank control
+ * @work: Kernel work data for the work event
* @dm: amdgpu display manager device
- * @otg_inst: otg instance of which vblank is being set
- * @enable: true if enable vblank
+ * @acrtc: amdgpu CRTC instance for which the event has occurred
+ * @enable: true if enabling vblank
*/
-struct vblank_workqueue {
- struct work_struct mall_work;
+struct vblank_control_work {
+ struct work_struct work;
struct amdgpu_display_manager *dm;
- int otg_inst;
+ struct amdgpu_crtc *acrtc;
bool enable;
};
@@ -380,11 +381,11 @@ struct amdgpu_display_manager {
#if defined(CONFIG_DRM_AMD_DC_DCN)
/**
- * @vblank_workqueue:
+ * @vblank_control_workqueue:
*
- * amdgpu workqueue during vblank
+ * Deferred work for vblank control events.
*/
- struct vblank_workqueue *vblank_workqueue;
+ struct workqueue_struct *vblank_control_workqueue;
#endif
struct drm_atomic_state *cached_state;
--
2.25.1
^ permalink raw reply related [flat|nested] 7+ messages in thread* [PATCH 3/7] drm/amd/display: Use vblank control events for PSR enable/disable
2021-08-13 6:15 [PATCH 0/7] DC Patches Aug 13, 2021 Wayne Lin
2021-08-13 6:15 ` [PATCH 1/7] drm/amd/display: Create dc_sink when EDID fail Wayne Lin
2021-08-13 6:15 ` [PATCH 2/7] drm/amd/display: Fix multi-display support for idle opt workqueue Wayne Lin
@ 2021-08-13 6:15 ` Wayne Lin
2021-08-13 6:15 ` [PATCH 4/7] drm/amd/display: Ensure DCN save after VM setup Wayne Lin
2021-08-16 13:27 ` [PATCH 0/7] DC Patches Aug 13, 2021 Wheeler, Daniel
4 siblings, 0 replies; 7+ messages in thread
From: Wayne Lin @ 2021-08-13 6:15 UTC (permalink / raw)
To: amd-gfx
Cc: Harry.Wentland, Sunpeng.Li, Bhawanpreet.Lakha, Rodrigo.Siqueira,
Aurabindo.Pillai, Qingqing.Zhuo, bindu.r, Anson.Jacob,
Nicholas Kazlauskas, Roman Li, Wayne Lin
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
[Why]
PSR can disable the HUBP along with the OTG when PSR is active.
We'll hit a pageflip timeout when the OTG is disable because we're no
longer updating the CRTC vblank counter and the pflip high IRQ will
not fire on the flip.
In order to flip the page flip timeout occur we should modify the
enter/exit conditions to match DRM requirements.
[How]
Use our deferred handlers for DRM vblank control to notify DMCU(B)
when it can enable or disable PSR based on whether vblank is disabled or
enabled respectively.
We'll need to pass along the stream with the notification now because
we want to access the CRTC state while the CRTC is locked to get the
stream state prior to the commit.
Retain a reference to the stream so it remains safe to continue to
access and release that reference once we're done with it.
Enable/disable logic follows what we were previously doing in
update_planes.
The workqueue has to be flushed before programming streams or planes
to ensure that we exit out of idle optimizations and PSR before
these events occur if necessary.
To keep the skip count logic the same to avoid FBCON PSR enablement
requires copying the allow condition onto the DM IRQ parameters - a
field that we can actually access from the worker.
Reviewed-by: Roman Li <Roman.Li@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
---
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 48 +++++++++++++++----
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 2 +
.../display/amdgpu_dm/amdgpu_dm_irq_params.h | 1 +
3 files changed, 43 insertions(+), 8 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 f88b6c5b83cd..cebd663b6708 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1061,7 +1061,22 @@ static void vblank_control_worker(struct work_struct *work)
DRM_DEBUG_KMS("Allow idle optimizations (MALL): %d\n", dm->active_vblank_irq_count == 0);
+ /* Control PSR based on vblank requirements from OS */
+ if (vblank_work->stream && vblank_work->stream->link) {
+ if (vblank_work->enable) {
+ if (vblank_work->stream->link->psr_settings.psr_allow_active)
+ amdgpu_dm_psr_disable(vblank_work->stream);
+ } else if (vblank_work->stream->link->psr_settings.psr_feature_enabled &&
+ !vblank_work->stream->link->psr_settings.psr_allow_active &&
+ vblank_work->acrtc->dm_irq_params.allow_psr_entry) {
+ amdgpu_dm_psr_enable(vblank_work->stream);
+ }
+ }
+
mutex_unlock(&dm->dc_lock);
+
+ dc_stream_release(vblank_work->stream);
+
kfree(vblank_work);
}
@@ -6018,6 +6033,11 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
work->acrtc = acrtc;
work->enable = enable;
+ if (acrtc_state->stream) {
+ dc_stream_retain(acrtc_state->stream);
+ work->stream = acrtc_state->stream;
+ }
+
queue_work(dm->vblank_control_workqueue, &work->work);
#endif
@@ -8623,6 +8643,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
/* Update the planes if changed or disable if we don't have any. */
if ((planes_count || acrtc_state->active_planes == 0) &&
acrtc_state->stream) {
+ /*
+ * If PSR or idle optimizations are enabled then flush out
+ * any pending work before hardware programming.
+ */
+ flush_workqueue(dm->vblank_control_workqueue);
+
bundle->stream_update.stream = acrtc_state->stream;
if (new_pcrtc_state->mode_changed) {
bundle->stream_update.src = acrtc_state->stream->src;
@@ -8691,16 +8717,20 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
acrtc_state->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED &&
!acrtc_state->stream->link->psr_settings.psr_feature_enabled)
amdgpu_dm_link_setup_psr(acrtc_state->stream);
- else if ((acrtc_state->update_type == UPDATE_TYPE_FAST) &&
- acrtc_state->stream->link->psr_settings.psr_feature_enabled &&
- !acrtc_state->stream->link->psr_settings.psr_allow_active) {
- struct amdgpu_dm_connector *aconn = (struct amdgpu_dm_connector *)
- acrtc_state->stream->dm_stream_context;
+
+ /* Decrement skip count when PSR is enabled and we're doing fast updates. */
+ if (acrtc_state->update_type == UPDATE_TYPE_FAST &&
+ acrtc_state->stream->link->psr_settings.psr_feature_enabled) {
+ struct amdgpu_dm_connector *aconn =
+ (struct amdgpu_dm_connector *)acrtc_state->stream->dm_stream_context;
if (aconn->psr_skip_count > 0)
aconn->psr_skip_count--;
- else
- amdgpu_dm_psr_enable(acrtc_state->stream);
+
+ /* Allow PSR when skip count is 0. */
+ acrtc_attach->dm_irq_params.allow_psr_entry = !aconn->psr_skip_count;
+ } else {
+ acrtc_attach->dm_irq_params.allow_psr_entry = false;
}
mutex_unlock(&dm->dc_lock);
@@ -8949,8 +8979,10 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
if (dc_state) {
/* if there mode set or reset, disable eDP PSR */
- if (mode_set_reset_required)
+ if (mode_set_reset_required) {
+ flush_workqueue(dm->vblank_control_workqueue);
amdgpu_dm_psr_disable_all(dm);
+ }
dm_enable_per_frame_crtc_master_sync(dc_state);
mutex_lock(&dm->dc_lock);
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index c6b8b835b08a..d1d353a7c77d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -91,12 +91,14 @@ struct dm_compressor_info {
* @work: Kernel work data for the work event
* @dm: amdgpu display manager device
* @acrtc: amdgpu CRTC instance for which the event has occurred
+ * @stream: DC stream for which the event has occurred
* @enable: true if enabling vblank
*/
struct vblank_control_work {
struct work_struct work;
struct amdgpu_display_manager *dm;
struct amdgpu_crtc *acrtc;
+ struct dc_stream_state *stream;
bool enable;
};
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
index f3b93ba69a27..79b5f9999fec 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
@@ -33,6 +33,7 @@ struct dm_irq_params {
struct mod_vrr_params vrr_params;
struct dc_stream_state *stream;
int active_planes;
+ bool allow_psr_entry;
struct mod_freesync_config freesync_config;
#ifdef CONFIG_DEBUG_FS
--
2.25.1
^ permalink raw reply related [flat|nested] 7+ messages in thread* [PATCH 4/7] drm/amd/display: Ensure DCN save after VM setup
2021-08-13 6:15 [PATCH 0/7] DC Patches Aug 13, 2021 Wayne Lin
` (2 preceding siblings ...)
2021-08-13 6:15 ` [PATCH 3/7] drm/amd/display: Use vblank control events for PSR enable/disable Wayne Lin
@ 2021-08-13 6:15 ` Wayne Lin
2021-08-16 13:27 ` [PATCH 0/7] DC Patches Aug 13, 2021 Wheeler, Daniel
4 siblings, 0 replies; 7+ messages in thread
From: Wayne Lin @ 2021-08-13 6:15 UTC (permalink / raw)
To: amd-gfx
Cc: Harry.Wentland, Sunpeng.Li, Bhawanpreet.Lakha, Rodrigo.Siqueira,
Aurabindo.Pillai, Qingqing.Zhuo, bindu.r, Anson.Jacob, Jake Wang,
Nicholas Kazlauskas, Wayne Lin
From: Jake Wang <haonan.wang2@amd.com>
[Why]
DM initializes VM context after DMCUB initialization.
This results in loss of DCN_VM_CONTEXT registers after z10.
[How]
Notify DMCUB when VM setup is complete, and have DMCUB
save init registers.
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Jake Wang <haonan.wang2@amd.com>
---
drivers/gpu/drm/amd/display/dc/core/dc.c | 6 ++++++
drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c | 3 +++
drivers/gpu/drm/amd/display/dc/dc.h | 1 +
drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c | 12 ++++++++++++
drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.h | 1 +
drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c | 1 +
drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h | 1 +
drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 5 +++++
8 files changed, 30 insertions(+)
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 7a442fcfa6ac..c798c65d4276 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -1549,6 +1549,12 @@ void dc_z10_restore(struct dc *dc)
if (dc->hwss.z10_restore)
dc->hwss.z10_restore(dc);
}
+
+void dc_z10_save_init(struct dc *dc)
+{
+ if (dc->hwss.z10_save_init)
+ dc->hwss.z10_save_init(dc);
+}
#endif
/*
* Applies given context to HW and copy it into current context.
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c b/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c
index f2b39ec35c89..80c752ca4a5a 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_vm_helper.c
@@ -47,6 +47,9 @@ int dc_setup_system_context(struct dc *dc, struct dc_phy_addr_space_config *pa_c
*/
memcpy(&dc->vm_pa_config, pa_config, sizeof(struct dc_phy_addr_space_config));
dc->vm_pa_config.valid = true;
+#if defined(CONFIG_DRM_AMD_DC_DCN3_1)
+ dc_z10_save_init(dc);
+#endif
}
return num_vmids;
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 03b81e5c5d67..0494e6dcf4dc 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1338,6 +1338,7 @@ void dc_hardware_release(struct dc *dc);
bool dc_set_psr_allow_active(struct dc *dc, bool enable);
#if defined(CONFIG_DRM_AMD_DC_DCN)
void dc_z10_restore(struct dc *dc);
+void dc_z10_save_init(struct dc *dc);
#endif
bool dc_enable_dmub_notifications(struct dc *dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
index 6399e8acd093..3f2333ec67e2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@@ -407,6 +407,18 @@ void dcn31_update_info_frame(struct pipe_ctx *pipe_ctx)
&pipe_ctx->stream_res.encoder_info_frame);
}
}
+void dcn31_z10_save_init(struct dc *dc)
+{
+ union dmub_rb_cmd cmd;
+
+ memset(&cmd, 0, sizeof(cmd));
+ cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT;
+ cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT;
+
+ dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
+ dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
+ dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+}
void dcn31_z10_restore(struct dc *dc)
{
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.h
index 40dfebe78fdd..140435e4f7ff 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.h
@@ -44,6 +44,7 @@ void dcn31_enable_power_gating_plane(
void dcn31_update_info_frame(struct pipe_ctx *pipe_ctx);
void dcn31_z10_restore(struct dc *dc);
+void dcn31_z10_save_init(struct dc *dc);
void dcn31_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on);
int dcn31_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, struct dc_phy_addr_space_config *pa_config);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
index 05954045c332..40011cd3c8ef 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
@@ -97,6 +97,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = {
.set_abm_immediate_disable = dcn21_set_abm_immediate_disable,
.set_pipe = dcn21_set_pipe,
.z10_restore = dcn31_z10_restore,
+ .z10_save_init = dcn31_z10_save_init,
.is_abm_supported = dcn31_is_abm_supported,
.set_disp_pattern_generator = dcn30_set_disp_pattern_generator,
.update_visual_confirm_color = dcn20_update_visual_confirm_color,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index 5ab008e62b82..ad5f2adcc40d 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -237,6 +237,7 @@ struct hw_sequencer_funcs {
int width, int height, int offset);
void (*z10_restore)(struct dc *dc);
+ void (*z10_save_init)(struct dc *dc);
void (*update_visual_confirm_color)(struct dc *dc,
struct pipe_ctx *pipe_ctx,
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 5950da7bf252..a6f3d58f82c6 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -860,6 +860,11 @@ enum dmub_cmd_idle_opt_type {
* DCN hardware restore.
*/
DMUB_CMD__IDLE_OPT_DCN_RESTORE = 0,
+
+ /**
+ * DCN hardware save.
+ */
+ DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT = 1
};
/**
--
2.25.1
^ permalink raw reply related [flat|nested] 7+ messages in thread* RE: [PATCH 0/7] DC Patches Aug 13, 2021
2021-08-13 6:15 [PATCH 0/7] DC Patches Aug 13, 2021 Wayne Lin
` (3 preceding siblings ...)
2021-08-13 6:15 ` [PATCH 4/7] drm/amd/display: Ensure DCN save after VM setup Wayne Lin
@ 2021-08-16 13:27 ` Wheeler, Daniel
4 siblings, 0 replies; 7+ messages in thread
From: Wheeler, Daniel @ 2021-08-16 13:27 UTC (permalink / raw)
To: Lin, Wayne, amd-gfx@lists.freedesktop.org
Cc: Wentland, Harry, Li, Sun peng (Leo), Lakha, Bhawanpreet,
Siqueira, Rodrigo, Pillai, Aurabindo, Zhuo, Qingqing, R, Bindu,
Jacob, Anson, Lin, Wayne
[Public]
Hi all,
This week this patchset was tested on the following systems:
HP Envy 360, with Ryzen 5 4500U, with the following display types: eDP 1080p 60hz, 4k 60hz (via USB-C to DP/HDMI), 1440p 144hz (via USB-C to DP/HDMI), 1680*1050 60hz (via USB-C to DP and then DP to DVI/VGA)
AMD Ryzen 9 5900H, with the following display types: eDP 1080p 60hz, 4k 60hz (via USB-C to DP/HDMI), 1440p 144hz (via USB-C to DP/HDMI), 1680*1050 60hz (via USB-C to DP and then DP to DVI/VGA)
Sapphire Pulse RX5700XT with the following display types:
4k 60hz (via DP/HDMI), 1440p 144hz (via DP/HDMI), 1680*1050 60hz (via DP to DVI/VGA)
Reference AMD RX6800 with the following display types:
4k 60hz (via DP/HDMI and USB-C to DP/HDMI), 1440p 144hz (via USB-C to DP/HDMI and USB-C to DP/HDMI), 1680*1050 60hz (via DP to DVI/VGA)
Included testing using a Startech DP 1.4 MST hub at 2x 4k 60hz, and 3x 1080p 60hz on all systems.
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Thank you,
Dan Wheeler
Technologist | AMD
SW Display
------------------------------------------------------------------------------------------------------------------
1 Commerce Valley Dr E, Thornhill, ON L3T 7X6
Facebook | Twitter | amd.com
-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Wayne Lin
Sent: August 13, 2021 2:16 AM
To: amd-gfx@lists.freedesktop.org
Cc: Wentland, Harry <Harry.Wentland@amd.com>; Li, Sun peng (Leo) <Sunpeng.Li@amd.com>; Lakha, Bhawanpreet <Bhawanpreet.Lakha@amd.com>; Siqueira, Rodrigo <Rodrigo.Siqueira@amd.com>; Pillai, Aurabindo <Aurabindo.Pillai@amd.com>; Zhuo, Qingqing <Qingqing.Zhuo@amd.com>; R, Bindu <Bindu.R@amd.com>; Jacob, Anson <Anson.Jacob@amd.com>; Lin, Wayne <Wayne.Lin@amd.com>
Subject: [PATCH 0/7] DC Patches Aug 13, 2021
This DC patchset brings improvements in multiple areas. In summary, we highlight:
* Ensure DCN save init registers after VM setup
* Fix multi-display support for idle opt workqueue
* Use vblank control events for PSR enable/disable
* Create default dc_sink when fail reading EDID under MST
---
Anthony Koo (1):
drm/amd/display: [FW Promotion] Release 0.0.79
Aric Cyr (1):
drm/amd/display: 3.2.149
Jake Wang (1):
drm/amd/display: Ensure DCN save after VM setup
Nicholas Kazlauskas (3):
drm/amd/display: Fix multi-display support for idle opt workqueue
drm/amd/display: Use vblank control events for PSR enable/disable
drm/amd/display: Guard vblank wq flush with DCN guards
Wayne Lin (1):
drm/amd/display: Create dc_sink when EDID fail
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 112 +++++++++++------- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 23 ++--
.../display/amdgpu_dm/amdgpu_dm_irq_params.h | 1 +
.../display/amdgpu_dm/amdgpu_dm_mst_types.c | 23 ++++
drivers/gpu/drm/amd/display/dc/core/dc.c | 6 +
.../drm/amd/display/dc/core/dc_vm_helper.c | 3 +
drivers/gpu/drm/amd/display/dc/dc.h | 3 +-
.../drm/amd/display/dc/dcn31/dcn31_hwseq.c | 12 ++
.../drm/amd/display/dc/dcn31/dcn31_hwseq.h | 1 +
.../gpu/drm/amd/display/dc/dcn31/dcn31_init.c | 1 +
.../gpu/drm/amd/display/dc/inc/hw_sequencer.h | 1 +
.../gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 19 ++-
12 files changed, 148 insertions(+), 57 deletions(-)
--
2.25.1
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH 2/7] drm/amd/display: Fix multi-display support for idle opt workqueue
2021-08-13 6:34 [PATCH 0/7] DC Patches August " Wayne Lin
@ 2021-08-13 6:34 ` Wayne Lin
0 siblings, 0 replies; 7+ messages in thread
From: Wayne Lin @ 2021-08-13 6:34 UTC (permalink / raw)
To: amd-gfx
Cc: Harry.Wentland, Sunpeng.Li, Bhawanpreet.Lakha, Rodrigo.Siqueira,
Aurabindo.Pillai, Qingqing.Zhuo, bindu.r, Anson.Jacob,
Nicholas Kazlauskas, Roman Li, Wayne Lin
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
[Why]
The current implementation for idle optimization support only has a
single work item that gets reshuffled into the system workqueue
whenever we receive an enable or disable event.
We can have mismatched events if the work hasn't been processed or if
we're getting control events from multiple displays at once.
This fixes this issue and also makes the implementation usable for
PSR control - which will be addressed in another patch.
[How]
We need to be able to flush remaining work out on demand for driver stop
and psr disable so create a driver specific workqueue instead of using
the system one. The workqueue will be single threaded to guarantee the
ordering of enable/disable events.
Refactor the queue to allocate the control work and deallocate it
after processing it.
Pass the acrtc directly to make it easier to handle psr enable/disable
in a later patch.
Rename things to indicate that it's not just MALL specific.
Reviewed-by: Roman Li <Roman.Li@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
---
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 62 ++++++++-----------
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 21 ++++---
2 files changed, 36 insertions(+), 47 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 3e28f17c84fa..f88b6c5b83cd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1044,10 +1044,10 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_
}
#endif
#if defined(CONFIG_DRM_AMD_DC_DCN)
-static void event_mall_stutter(struct work_struct *work)
+static void vblank_control_worker(struct work_struct *work)
{
-
- struct vblank_workqueue *vblank_work = container_of(work, struct vblank_workqueue, mall_work);
+ struct vblank_control_work *vblank_work =
+ container_of(work, struct vblank_control_work, work);
struct amdgpu_display_manager *dm = vblank_work->dm;
mutex_lock(&dm->dc_lock);
@@ -1062,22 +1062,9 @@ static void event_mall_stutter(struct work_struct *work)
DRM_DEBUG_KMS("Allow idle optimizations (MALL): %d\n", dm->active_vblank_irq_count == 0);
mutex_unlock(&dm->dc_lock);
+ kfree(vblank_work);
}
-static struct vblank_workqueue *vblank_create_workqueue(struct amdgpu_device *adev, struct dc *dc)
-{
- struct vblank_workqueue *vblank_work;
-
- vblank_work = kzalloc(sizeof(*vblank_work), GFP_KERNEL);
- if (ZERO_OR_NULL_PTR(vblank_work)) {
- kfree(vblank_work);
- return NULL;
- }
-
- INIT_WORK(&vblank_work->mall_work, event_mall_stutter);
-
- return vblank_work;
-}
#endif
static int amdgpu_dm_init(struct amdgpu_device *adev)
{
@@ -1220,12 +1207,10 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
#if defined(CONFIG_DRM_AMD_DC_DCN)
if (adev->dm.dc->caps.max_links > 0) {
- adev->dm.vblank_workqueue = vblank_create_workqueue(adev, adev->dm.dc);
-
- if (!adev->dm.vblank_workqueue)
+ adev->dm.vblank_control_workqueue =
+ create_singlethread_workqueue("dm_vblank_control_workqueue");
+ if (!adev->dm.vblank_control_workqueue)
DRM_ERROR("amdgpu: failed to initialize vblank_workqueue.\n");
- else
- DRM_DEBUG_DRIVER("amdgpu: vblank_workqueue init done %p.\n", adev->dm.vblank_workqueue);
}
#endif
@@ -1298,6 +1283,13 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev)
{
int i;
+#if defined(CONFIG_DRM_AMD_DC_DCN)
+ if (adev->dm.vblank_control_workqueue) {
+ destroy_workqueue(adev->dm.vblank_control_workqueue);
+ adev->dm.vblank_control_workqueue = NULL;
+ }
+#endif
+
for (i = 0; i < adev->dm.display_indexes_num; i++) {
drm_encoder_cleanup(&adev->dm.mst_encoders[i].base);
}
@@ -1321,14 +1313,6 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev)
dc_deinit_callbacks(adev->dm.dc);
#endif
-#if defined(CONFIG_DRM_AMD_DC_DCN)
- if (adev->dm.vblank_workqueue) {
- adev->dm.vblank_workqueue->dm = NULL;
- kfree(adev->dm.vblank_workqueue);
- adev->dm.vblank_workqueue = NULL;
- }
-#endif
-
dc_dmub_srv_destroy(&adev->dm.dc->ctx->dmub_srv);
if (dc_enable_dmub_notifications(adev->dm.dc)) {
@@ -6000,7 +5984,7 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state);
#if defined(CONFIG_DRM_AMD_DC_DCN)
struct amdgpu_display_manager *dm = &adev->dm;
- unsigned long flags;
+ struct vblank_control_work *work;
#endif
int rc = 0;
@@ -6025,12 +6009,16 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
return 0;
#if defined(CONFIG_DRM_AMD_DC_DCN)
- spin_lock_irqsave(&dm->vblank_lock, flags);
- dm->vblank_workqueue->dm = dm;
- dm->vblank_workqueue->otg_inst = acrtc->otg_inst;
- dm->vblank_workqueue->enable = enable;
- spin_unlock_irqrestore(&dm->vblank_lock, flags);
- schedule_work(&dm->vblank_workqueue->mall_work);
+ work = kzalloc(sizeof(*work), GFP_ATOMIC);
+ if (!work)
+ return -ENOMEM;
+
+ INIT_WORK(&work->work, vblank_control_worker);
+ work->dm = dm;
+ work->acrtc = acrtc;
+ work->enable = enable;
+
+ queue_work(dm->vblank_control_workqueue, &work->work);
#endif
return 0;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index ab1670b16b02..c6b8b835b08a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -60,6 +60,7 @@ enum aux_return_code_type;
/* Forward declarations */
struct amdgpu_device;
+struct amdgpu_crtc;
struct drm_device;
struct dc;
struct amdgpu_bo;
@@ -86,16 +87,16 @@ struct dm_compressor_info {
};
/**
- * struct vblank_workqueue - Works to be executed in a separate thread during vblank
- * @mall_work: work for mall stutter
+ * struct vblank_control_work - Work data for vblank control
+ * @work: Kernel work data for the work event
* @dm: amdgpu display manager device
- * @otg_inst: otg instance of which vblank is being set
- * @enable: true if enable vblank
+ * @acrtc: amdgpu CRTC instance for which the event has occurred
+ * @enable: true if enabling vblank
*/
-struct vblank_workqueue {
- struct work_struct mall_work;
+struct vblank_control_work {
+ struct work_struct work;
struct amdgpu_display_manager *dm;
- int otg_inst;
+ struct amdgpu_crtc *acrtc;
bool enable;
};
@@ -380,11 +381,11 @@ struct amdgpu_display_manager {
#if defined(CONFIG_DRM_AMD_DC_DCN)
/**
- * @vblank_workqueue:
+ * @vblank_control_workqueue:
*
- * amdgpu workqueue during vblank
+ * Deferred work for vblank control events.
*/
- struct vblank_workqueue *vblank_workqueue;
+ struct workqueue_struct *vblank_control_workqueue;
#endif
struct drm_atomic_state *cached_state;
--
2.25.1
^ permalink raw reply related [flat|nested] 7+ messages in thread