i915 fixes to handle hotplug cases on 8K tiled monitor
Subsystems (3)
cover i915 fixes to handle hotplug cases on 8K tiled monitor
This has some fixes in the transcoder port sync teardown sequence which ensures now that the hotplug/unplug is handled gracefully
Manasi Navare (3):
drm/i915/dp: Make sure all tiled connectors get added to the state
with full modeset
drm/i915/dp: Make port sync mode assignments only if all tiles present drm/i915/dp: Disable Port sync mode correctly on teardown
drivers/gpu/drm/i915/display/intel_display.c | 109 ++++++++++++++++++- 1 file changed, 106 insertions(+), 3 deletions(-)
Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 1/3 drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in intel_atomic_check() we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)
drivers/gpu/drm/i915/display/intel_display.c +78 -0
@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device
@@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail; +for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {
Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 1 line · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other
Minor typo on "linked to" here.
Quoted 8 lines
for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>
I think we're moving to "Closes:" as the annotation here now that it's not actually a bugzilla bug database anymore.
Quoted 67 lines
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +54
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile)
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14132
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id; + }else
Minor kernel coding style violation; if we use {} on one branch of an if, we need to use them on all.
Quoted 14 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0@@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;
Should this happen before the drm_atomic_helper_check_modeset() just above (or should we re-call that function if we flag the other tile as needing a modeset)? The kerneldoc on that function says:
"""
Drivers which set &drm_crtc_state.mode_changed [...] _must_ call this function afterwards after that change. It is permitted to call this function multiple times for the same update ...
"""
Matt
Quoted 11 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 9 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 4 lines · Matt Roper
Quoted 1 line · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other
Minor typo on "linked to" here.
I will fix it
Quoted 5 lines
Quoted 5 lines
for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.
Quoted 7 lines
Quoted 4 lines
Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>I think we're moving to "Closes:" as the annotation here now that it's not actually a bugzilla bug database anymore.
Ok cool, will change that to Closes
Quoted 73 lines
Quoted 73 lines
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +59
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc);
Quoted 9 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14132
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id; + }else Minor kernel coding style violation; if we use {} on one branch of an if, we need to use them on all.
Yes i got a checkpatch check warning, will fix it
Quoted 25 lines
Quoted 14 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0@@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;Should this happen before the
drm_atomic_helper_check_modeset()just above (or should we re-call that function if we flag the other tile as needing a modeset)? The kerneldoc on that function says:
Quoted 6 lines
"""
Drivers which set &drm_crtc_state.mode_changed [...] _must_ call this function afterwards after that change. It is permitted to call this function multiple times for the same update ...
"""
IMO, here infact it makes sense to call my function after the drm_atomic_helper_check_modeset() because it directly sets the new_crtc_state->mode_changed to true for all tiles if 1 of them needs a full modeset.
And whether that one tile needs a full modeset or not will be decided based on mode changed for that set in drm_atomic_helper_check_modeset.
Manasi
Quoted 10 lines
Matt
Quoted 7 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1
Quoted 11 lines
Quoted 4 lines
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 5 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 140 lines · Manasi Navare
Quoted 4 lines · Matt Roper
Quoted 1 line · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other
Minor typo on "linked to" here.
I will fix it
Quoted 5 lines
Quoted 5 lines
for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Quoted 7 lines
Quoted 4 lines
Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>I think we're moving to "Closes:" as the annotation here now that it's not actually a bugzilla bug database anymore.
Ok cool, will change that to Closes
Quoted 73 lines
Quoted 73 lines
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +59
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc);Quoted 9 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14132
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id; + }else Minor kernel coding style violation; if we use {} on one branch of an if, we need to use them on all.
Yes i got a checkpatch check warning, will fix it
Quoted 25 lines
Quoted 14 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0@@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;Should this happen before the
drm_atomic_helper_check_modeset()just above (or should we re-call that function if we flag the other tile as needing a modeset)? The kerneldoc on that function says:Quoted 6 lines
"""
Drivers which set &drm_crtc_state.mode_changed [...] _must_ call this function afterwards after that change. It is permitted to call this function multiple times for the same update ...
"""
Quoted 5 lines
IMO, here infact it makes sense to call my function after the
drm_atomic_helper_check_modeset()because it directly sets thenew_crtc_state->mode_changedto true for all tiles if 1 of them needs a full modeset.
And whether that one tile needs a full modeset or not will be decided based on mode changed for that set indrm_atomic_helper_check_modeset.
Okay, makes sense. But based on the comment, I believe we still need to call the helper a second time during/after your function in the event that we've switched any additional crtcs to mode_changed=true.
I'm also wondering whether we should eventually move most of your logic here out of i915 and into a general drm helper that other drivers can utilize too. But I'm okay with doing that as a followup after we've landed it and verified it as an i915-specific change first.
Matt
Quoted 24 lines
Manasi
Quoted 21 lines
Matt
Quoted 11 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 5 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795
Signature · 9 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
In reply to Manasi NavareQuoted 19 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;
'connector'
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_state has needlessly wide scope.
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;
'conn_state' is the most popular name.
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter);
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;
The tile group check should be part of the same if with the has_tile check.
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,
Pointless variable. Can be extracted from the atomic state.
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_id is rather pointless. crtc_state and ret can move into tighter scope. And the next suggestion allows you to kill crtc_state entirely...
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; +
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14129
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ;
This should to be able to be shortened to just intel_connector_needs_modeset().
Quoted 17 lines
drivers/gpu/drm/i915/display/intel_display.c +13
-0+ + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0@@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;
We should probably do this from the connector .atomic_check() hook if Jose is going to do the MST thing that way. But no real problem doing here I think.
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 23 lines · Ville Syrjälä
Quoted 22 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; }
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
Ok
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.
Ok will add it inside the loop
Quoted 18 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } +
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.
Ok yes. So not pass the tile grp id at all just compare with current connector's tile_grp_id ?
Quoted 19 lines
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...
tile_grp_id is useless because I should just use it from current connector obtained from state?
Quoted 17 lines
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; +drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14129
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ;This should to be able to be shortened to just
intel_connector_needs_modeset().
Ok will use that instead.
Quoted 21 lines
Quoted 21 lines
drivers/gpu/drm/i915/display/intel_display.c +13
-0+ + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device @@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail;
Quoted 7 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 3 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0+ ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;We should probably do this from the connector .atomic_check() hook if Jose is going to do the MST thing that way. But no real problem doing here I think.
If you dont have a strong preference I will leave it here since Jose's code is not in yet.
Manasi
Quoted 11 lines
Quoted 7 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1Signature · 2 lines
Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 67 lines · Manasi Navare
Quoted 23 lines · Ville Syrjälä
Quoted 22 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; }Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
Ok
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.Ok will add it inside the loop
Quoted 18 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } +Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.
Quoted 2 lines
Ok yes. So not pass the tile grp id at all just compare with current connector's
tile_grp_id?
You still need to pass in the tile_grp_id from the caller's current connector. What I meant is that you should do
if (!connector->has_tile || connector->tile_grp != tile_grp)
continue;
before you do the drm_atomic_get_connector_state(). We don't want to needlessly add connectors belonging to some other tile group to our atomic state.
Quoted 31 lines
Quoted 20 lines
Quoted 7 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...
Quoted 1 line
tile_grp_idis useless because I should just use it from current connector obtained from state?
I just meant that you use that variable exactly once. So instead you can just directly do:
ret = intel_dp_modeset_all_tiles(state, connector->tile_grp_id);
...
Quoted 65 lines
Quoted 18 lines
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; +drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14129
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ;This should to be able to be shortened to just
intel_connector_needs_modeset().Ok will use that instead.
Quoted 21 lines
Quoted 21 lines
drivers/gpu/drm/i915/display/intel_display.c +13
-0+ + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device @@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail;Quoted 7 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 3 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0+ ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;We should probably do this from the connector .atomic_check() hook if Jose is going to do the MST thing that way. But no real problem doing here I think.
If you dont have a strong preference I will leave it here since Jose's code is not in yet.
Manasi
Quoted 11 lines
Quoted 7 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1Signature · 2 lines
Ville Syrjälä Intel
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
In reply to Ville SyrjäläQuoted 78 lines · Ville Syrjälä
Quoted 19 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...
Its not clear why tile_grp_id is pointless, I am using tile_grp_id for the first connector with has_tile and I make sure that I dont enter into the loop to check modeset again for the connector with same tile_grp_id because we have already set its mode changed to true in intel_dp_modeset_all_tiles()
How can I achieve this instead?
Manasi
Quoted 46 lines
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; +drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14129
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ;This should to be able to be shortened to just
intel_connector_needs_modeset().Quoted 17 lines
drivers/gpu/drm/i915/display/intel_display.c +13
-0+ + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +} +/** * intel_atomic_check - validate state object * @dev: drm device drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check(struct drm_device *dev, · line 14172
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +3
-0@@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail;We should probably do this from the connector .atomic_check() hook if Jose is going to do the MST thing that way. But no real problem doing here I think.
Quoted 10 lines
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1Signature · 2 lines
Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 90 lines · Manasi Navare
Quoted 78 lines · Ville Syrjälä
Quoted 19 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...
Quoted 5 lines
Its not clear why
tile_grp_idis pointless, I am usingtile_grp_idfor the first connector withhas_tileand I make sure that I dont enter into the loop to check modeset again for the connector with sametile_grp_idbecause we have already set its mode changed to true inintel_dp_modeset_all_tiles()How can I achieve this instead?
Instead offoo()
{
tile_grp_id = 0;
for_each() {
tile_grp_id = conn->tile_grp_id;
intel_dp_modeset_all_tiles(tile_grp_id);
}
}you just dofoo()
{
for_each() {
intel_dp_modeset_all_tiles(conn->tile_grp_id);
}
}Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 110 lines · Ville Syrjälä
Quoted 90 lines · Manasi Navare
Quoted 78 lines · Ville Syrjälä
Quoted 19 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...Quoted 5 lines
Its not clear why
tile_grp_idis pointless, I am usingtile_grp_idfor the first connector withhas_tileand I make sure that I dont enter into the loop to check modeset again for the connector with sametile_grp_idbecause we have already set its mode changed to true inintel_dp_modeset_all_tiles()How can I achieve this instead?
Instead of
foo(){ tile_grp_id = 0; for_each() { tile_grp_id = conn->tile_grp_id; intel_dp_modeset_all_tiles(tile_grp_id); } }
Quoted 7 lines
you just do
foo(){ for_each() { intel_dp_modeset_all_tiles(conn->tile_grp_id); } }
Yes I understand that we can pass the conn->tile_grp_id directly. But currently I am using tile_grp_id in for_each(), to call intel_dp_modeset_all_tiles() only if that cnnector is from a differnt tile grp id else continue.
foo()
{
for_each() {
if(tile_grp_id != conn->tile_grp_id)
intel_dpmodeset_all_tiles();
else
continue;
}
}calling intel_dp_modeset_all_tiles() for all tiles is kind of redundant. How can I do this with your suggestion, or do you think we should call intel_dp_modeset_all_tiles for each connector anyway and it shouldnt matter?
Regards
Manasi
Quoted 4 lines
Signature · 3 lines
-- Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 132 lines · Manasi Navare
Quoted 110 lines · Ville Syrjälä
Quoted 90 lines · Manasi Navare
Quoted 78 lines · Ville Syrjälä
Quoted 19 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...Quoted 5 lines
Its not clear why
tile_grp_idis pointless, I am usingtile_grp_idfor the first connector withhas_tileand I make sure that I dont enter into the loop to check modeset again for the connector with sametile_grp_idbecause we have already set its mode changed to true inintel_dp_modeset_all_tiles()How can I achieve this instead?
Instead of
foo(){ tile_grp_id = 0; for_each() { tile_grp_id = conn->tile_grp_id; intel_dp_modeset_all_tiles(tile_grp_id); } }Quoted 7 lines
you just do
foo(){ for_each() { intel_dp_modeset_all_tiles(conn->tile_grp_id); } }Yes I understand that we can pass the
conn->tile_grp_iddirectly. But currently I am usingtile_grp_idinfor_each(), to callintel_dp_modeset_all_tiles()only if that cnnector is from a differnt tile grp id else continue.
foo(){ for_each() { if(tile_grp_id != conn->tile_grp_id) intel_dpmodeset_all_tiles(); else continue; } }
Quoted 2 lines
calling
intel_dp_modeset_all_tiles()for all tiles is kind of redundant. How can I do this with your suggestion, or do you think we should callintel_dp_modeset_all_tilesfor each connector anyway and it shouldnt matter?
Yes if that connector is tiled.
Eg. if you have something like
connector A: tile group 1
connector B: tile group 2
connector C: tile group 3
connector D: tile group 2
then your single tile_grp_id variable doesn't work anyway. You'd need track which tile groups you've already handled and skip those. But much easier to just blindly plow ahead.
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 144 lines · Ville Syrjälä
Quoted 132 lines · Manasi Navare
Quoted 110 lines · Ville Syrjälä
Quoted 90 lines · Manasi Navare
Quoted 78 lines · Ville Syrjälä
Quoted 19 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +0
-0drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14073
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter;'connector'
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14075
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state;
crtc_statehas needlessly wide scope.drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14079
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state;'
conn_state' is the most popular name.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14094
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ;The tile group check should be part of the same if with the
has_tilecheck.Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+ + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state);drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14109
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check(struct drm_i915_private *dev_priv,Pointless variable. Can be extracted from the atomic state.
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14115
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0+ struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ;
tile_grp_idis rather pointless.crtc_stateand ret can move into tighter scope. And the next suggestion allows you to killcrtc_stateentirely...Quoted 5 lines
Its not clear why
tile_grp_idis pointless, I am usingtile_grp_idfor the first connector withhas_tileand I make sure that I dont enter into the loop to check modeset again for the connector with sametile_grp_idbecause we have already set its mode changed to true inintel_dp_modeset_all_tiles()How can I achieve this instead?
Instead of
foo(){ tile_grp_id = 0; for_each() { tile_grp_id = conn->tile_grp_id; intel_dp_modeset_all_tiles(tile_grp_id); } }Quoted 7 lines
you just do
foo(){ for_each() { intel_dp_modeset_all_tiles(conn->tile_grp_id); } }Yes I understand that we can pass the
conn->tile_grp_iddirectly. But currently I am usingtile_grp_idinfor_each(), to callintel_dp_modeset_all_tiles()only if that cnnector is from a differnt tile grp id else continue.
foo(){ for_each() { if(tile_grp_id != conn->tile_grp_id) intel_dpmodeset_all_tiles(); else continue; } }Quoted 2 lines
calling
intel_dp_modeset_all_tiles()for all tiles is kind of redundant. How can I do this with your suggestion, or do you think we should callintel_dp_modeset_all_tilesfor each connector anyway and it shouldnt matter?Yes if that connector is tiled.
Eg. if you have something like
connector A: tile group 1
connector B: tile group 2
connector C: tile group 3
connector D: tile group 2
Quoted 3 lines
then your single
tile_grp_idvariable doesn't work anyway. You'd need track which tile groups you've already handled and skip those. But much easier to just blindly plow ahead.
Okay so in our case :
Conn A: tile gro 1
Conn B: tile grp 1
Now foo {
for_each_conn {
//connA
intel_modeset_all_tiles() // this will set mode to true for conn B in iteration 1
}
}
Now in the second iteration
for_each_conn {
// conn B iter 2
intel_modeset_all_tiles() // this will set mode to true for conn A
}redundant second iteration since both are in same tile grp
Manasi
Quoted 4 lines
Signature · 3 lines
-- Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
In reply to Manasi NavareQuoted 85 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +63
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id;
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +}
BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 17 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+/** * intel_atomic_check - validate state object * @dev: drm device @@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail; + ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail; +for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 94 lines · Ville Syrjälä
Quoted 93 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +71
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } +
Quoted 9 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
The crtcs in the current state get synced and master/slave assignments happen in icl_add_sync_mode_crtcs before compute_config call So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?
And why would the tile information be disappeared?
Manasi
Quoted 8 lines
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0+/** * intel_atomic_check - validate state object * @dev: drm device @@ -14093,6 +14167,10 @@ static int intel_atomic_check(struct drm_device *dev,if (ret) goto fail;
Quoted 13 lines
Quoted 9 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-0+ ret =intel_dp_atomic_trans_port_sync_check (dev_priv, state); +if (ret) + goto fail; +for_each_oldnew_intel_crtc_in_state (state, crtc, old_crtc_state, new_crtc_state, i) {if (!needs_modeset (new_crtc_state)) {Signature · 1 line
2.19.1Signature · 2 lines
Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 105 lines · Manasi Navare
Quoted 94 lines · Ville Syrjälä
Quoted 93 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +71
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id; + }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } +Quoted 9 lines
drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 3 lines
The crtcs in the current state get synced and master/slave assignments happen in
icl_add_sync_mode_crtcsbeforecompute_configcall So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?
I think it should look something like:
modeset_tiled_things();modeset_synced_things();
for_each() {
modeset_pipes_config();
if (can_fastset()) {
modeset=false;
fastset=true;
}
}modeset_synced_things();
for_each() {
if (!modeset && fastset)
copy_state();
}Quoted 2 lines
And why would the tile information be disappeared?
It'll get updated whenever someone does a getconnector() or whatever.
Example:
1. sync pipe A and B, pipe A is master
2. swap pipe B display for something else
3. getconnector() -> tile info goes poof
4. do something on pipe A that needs a modeset
no tile info so we miss that pipe B also needs a modeset
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 106 lines · Ville Syrjälä
Quoted 105 lines · Manasi Navare
Quoted 103 lines · Ville Syrjälä
Quoted 85 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +63
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id;drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 8 lines
Quoted 3 lines
The crtcs in the current state get synced and master/slave assignments happen in
icl_add_sync_mode_crtcsbeforecompute_configcall So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?I think it should look something like:
modeset_tiled_things();modeset_synced_things();
but modeset_synced_things() should be called after icl_add_sync_crtcs() which as per your review should be called from within modeset_pipe_config just before compute_config() call.
Quoted 2 lines
for_each(){
This is the for_each_crtc loop in intel_atomic_check() right?
Quoted 1 line
modeset_pipes_config();
So have icl_add_sync_crtcs outside of modeset_pipe_config()?
Quoted 6 lines
if (can_fastset()) {
modeset=false;
fastset=true;
}
}Quoted 6 lines
modeset_synced_things();for_each() { if (!modeset && fastset) copy_state(); }
We already do this in the code right?
Manasi
Quoted 4 lines
Quoted 3 lines
And why would the tile information be disappeared?
Quoted 12 lines
It'll get updated whenever someone does a
getconnector()or whatever.Example:
1. sync pipe A and B, pipe A is master
2. swap pipe B display for something else
3.getconnector()-> tile info goes poof
4. do something on pipe A that needs a modeset
no tile info so we miss that pipe B also needs a modesetSignature · 2 lines
Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 136 lines · Manasi Navare
Quoted 106 lines · Ville Syrjälä
Quoted 105 lines · Manasi Navare
Quoted 103 lines · Ville Syrjälä
Quoted 85 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +63
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id;drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 8 lines
Quoted 3 lines
The crtcs in the current state get synced and master/slave assignments happen in
icl_add_sync_mode_crtcsbeforecompute_configcall So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?I think it should look something like:
modeset_tiled_things();modeset_synced_things();but
modeset_synced_things()should be called aftericl_add_sync_crtcs()which as per your review should be called from withinmodeset_pipe_configjust beforecompute_config()call.Quoted 2 lines
for_each(){This is the
for_each_crtcloop inintel_atomic_check()right?Quoted 1 line
modeset_pipes_config();So have
icl_add_sync_crtcsoutside ofmodeset_pipe_config()?Quoted 8 lines
if (can_fastset()) { modeset=false; fastset=true; } }
modeset_synced_things();
Quoted 17 lines
Quoted 4 lines
for_each() { if (!modeset && fastset) copy_state(); }We already do this in the code right?
Manasi
Quoted 9 lines
Quoted 3 lines
And why would the tile information be disappeared?
It'll get updated whenever someone does a
getconnector()or whatever.Example:
1. sync pipe A and B, pipe A is master
2. swap pipe B display for something else
If we disconnect and connect other display for pipe B, port sync mode is off and
Pipe A no longer a master and we would reset the master_slave assignments and conn on pipe B would not have tile
so why we wold need to add both pipes to modeset in this case at all
Manasi
Quoted 11 lines
Quoted 7 lines
3.
getconnector()-> tile info goes poof
4. do something on pipe A that needs a modeset
no tile info so we miss that pipe B also needs a modesetSignature · 2 lines
Ville Syrjälä Intel_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 149 lines · Manasi Navare
Quoted 148 lines · Manasi Navare
Quoted 106 lines · Ville Syrjälä
Quoted 105 lines · Manasi Navare
Quoted 103 lines · Ville Syrjälä
Quoted 85 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +63
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id;drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 8 lines
Quoted 3 lines
The crtcs in the current state get synced and master/slave assignments happen in
icl_add_sync_mode_crtcsbeforecompute_configcall So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?I think it should look something like:
modeset_tiled_things();modeset_synced_things();but
modeset_synced_things()should be called aftericl_add_sync_crtcs()which as per your review should be called from withinmodeset_pipe_configjust beforecompute_config()call.Quoted 2 lines
for_each(){This is the
for_each_crtcloop inintel_atomic_check()right?Quoted 1 line
modeset_pipes_config();So have
icl_add_sync_crtcsoutside ofmodeset_pipe_config()?Quoted 6 lines
if (can_fastset()) { modeset=false; fastset=true; } }Quoted 6 lines
modeset_synced_things();for_each() { if (!modeset && fastset) copy_state(); }We already do this in the code right?
Manasi
Quoted 4 lines
Quoted 3 lines
And why would the tile information be disappeared?
Quoted 8 lines
Quoted 5 lines
Quoted 5 lines
It'll get updated whenever someone does a
getconnector()or whatever.Example:
1. sync pipe A and B, pipe A is master
2. swap pipe B display for something elseIf we disconnect and connect other display for pipe B, port sync mode is off and
Pipe A no longer a master and we would reset themaster_slaveassignments and conn on pipe B would not have tile
Port sync will stay enabled until we do a modeset to disable it. In this example there is never any modeset on pipe B until we add soemthing to force one in step 4.
Quoted 16 lines
so why we wold need to add both pipes to modeset in this case at all
Manasi
Quoted 11 lines
Quoted 7 lines
3.
getconnector()-> tile info goes poof
4. do something on pipe A that needs a modeset
no tile info so we miss that pipe B also needs a modesetSignature · 2 lines
Ville Syrjälä Intel_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Quoted 156 lines · Ville Syrjälä
Quoted 155 lines · Manasi Navare
Quoted 153 lines · Manasi Navare
Quoted 106 lines · Ville Syrjälä
Quoted 105 lines · Manasi Navare
Quoted 103 lines · Ville Syrjälä
Quoted 85 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +63
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id;drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 8 lines
Quoted 3 lines
The crtcs in the current state get synced and master/slave assignments happen in
icl_add_sync_mode_crtcsbeforecompute_configcall So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?I think it should look something like:
modeset_tiled_things();modeset_synced_things();but
modeset_synced_things()should be called aftericl_add_sync_crtcs()which as per your review should be called from withinmodeset_pipe_configjust beforecompute_config()call.Quoted 2 lines
for_each(){This is the
for_each_crtcloop inintel_atomic_check()right?Quoted 1 line
modeset_pipes_config();So have
icl_add_sync_crtcsoutside ofmodeset_pipe_config()?Quoted 6 lines
if (can_fastset()) { modeset=false; fastset=true; } }Quoted 6 lines
modeset_synced_things();for_each() { if (!modeset && fastset) copy_state(); }We already do this in the code right?
Manasi
Quoted 9 lines
Quoted 3 lines
And why would the tile information be disappeared?
It'll get updated whenever someone does a
getconnector()or whatever.Example:
1. sync pipe A and B, pipe A is master
2. swap pipe B display for something else
Quoted 6 lines
Quoted 2 lines
If we disconnect and connect other display for pipe B, port sync mode is off and
Pipe A no longer a master and we would reset themaster_slaveassignments and conn on pipe B would not have tilePort sync will stay enabled until we do a modeset to disable it. In this example there is never any modeset on pipe B until we add soemthing to force one in step 4.
seems to be working with the current intel_dp_add_modeset_tiles() since the moment
we unplug, it goes through the disable sequence where all the associated tiles get disabled and port sync mode gets disabled. This happens because the Pipe A which is still connected now indicates a mode change since it fallsback to a lower non tiled mode.
But to be on safer side, i can check if say Pipe A is a master or slave (in port sync mode) , check its crtc_state which is still showing the old master slave links since we havent cleared those yet and then add all other synced crtcs to the modeset?
So the order of calls can still be :intel_atomic_check() {
intel_dp_atomic_tiled_check() { modeset_all_tiles modeset_synced_crtcs} intel_modeset_pipe_Config()icl_add_sync_crtcscompute_configfastsetcheck()modeset_synced_things (here it looks at the new master slave assignments)
Does this look correct, I want send this patch out today
Manasi
Quoted 10 lines
Quoted 10 lines
so why we wold need to add both pipes to modeset in this case at all
Manasi
Quoted 4 lines
Quoted 4 lines
3.
getconnector()-> tile info goes poof
4. do something on pipe A that needs a modeset
no tile info so we miss that pipe B also needs a modeset
Quoted 11 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Signature · 2 lines
Ville Syrjälä Intel_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 2 lines
Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 1/3] drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
Hi Ville,
I double checke that I can use new_crtc_state in modeset_synced_crtcs to add all synced crtcs to modeset That is working as expected so that the same function can be reused after fastset check to override mode_changed to true.
Here's what I have:
static intintel_dp_modeset_synced_crtcs(struct intel_atomic_state *state)
{
struct drm_i915_private *dev_priv = to_i915(state->base.dev);
struct intel_crtc_state *old_crtc_state, *new_crtc_state;
struct intel_crtc *crtc;
int i;
for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
new_crtc_state, i) {
if(is_trans_port_sync_mode(new_crtc_state))
DRM_DEBUG_KMS("\n Adding CRTC %d:%s to modeset since old crtc state master trans = %s, slave mask = %d, new crtc state master trans = %s slave mask = %d",
crtc->base.base.id,
crtc->base.name,
transcoder_name(old_crtc_state->master_transcoder),
old_crtc_state->sync_mode_slaves_mask,
transcoder_name(new_crtc_state->master_transcoder),
new_crtc_state->sync_mode_slaves_mask);
new_crtc_state->uapi.mode_changed = true;
}
return 0;
}static intintel_dp_atomic_check_synced_crtcs(struct intel_atomic_state *state)
{
struct drm_i915_private *dev_priv = to_i915(state->base.dev);
struct intel_crtc_state *old_crtc_state, *new_crtc_state;
struct intel_crtc *crtc;
int i;
for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
new_crtc_state, i) {
if(!is_trans_port_sync_mode(new_crtc_state) ||
!needs_modeset(new_crtc_state))
continue;
intel_dp_modeset_synced_crtcs(state);
}
return 0;
}I will add this to this patch after thye call to modeset_all_tiles() and submit a new revision.
Regards
Manasi
Quoted 205 lines · Manasi Navare
Quoted 156 lines · Ville Syrjälä
Quoted 155 lines · Manasi Navare
Quoted 153 lines · Manasi Navare
Quoted 106 lines · Ville Syrjälä
Quoted 105 lines · Manasi Navare
Quoted 103 lines · Ville Syrjälä
Quoted 85 lines · Manasi Navare
In case of tiled displays, all the tiles are linke dto each other for transcoder port sync. So in
intel_atomic_check()we need to make sure that we add all the tiles to the modeset and if one of the tiles needs a full modeset then mark all other tiles for a full modeset.Suggested-by Ville Syrjälä <[email protected]>Cc Ville Syrjälä <[email protected]>Cc José Roberto de Souza <[email protected]>Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 78 ++++++++++++++++++++ 1 file changed, 78 insertions(+)drivers/gpu/drm/i915/display/intel_display.c +63
-0@@ -14066,6 +14066,80 @@ static int intel_atomic_check_crtcs(struct intel_atomic_state *state)return 0 ; } +static int +intel_dp_modeset_all_tiles (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state,int tile_grp_id) +{ +struct drm_connector *conn_iter; +struct drm_connector_list_iter conn_list_iter; +struct drm_crtc_state *crtc_state; + +drm_connector_list_iter_begin (&dev_priv->drm, &conn_list_iter); +drm_for_each_connector_iter (conn_iter, &conn_list_iter) { +struct drm_connector_state *conn_iter_state; + +if (!conn_iter->has_tile) +continue ; + conn_iter_state =drm_atomic_get_connector_state (&state->base, + conn_iter); +if (IS_ERR (conn_iter_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + +if (!conn_iter_state->crtc) +continue ; + +if (conn_iter->tile_group->id != tile_grp_id) +continue ; + + crtc_state =drm_atomic_get_crtc_state (&state->base, conn_iter_state->crtc); +if (IS_ERR (crtc_state)) { +drm_connector_list_iter_end (&conn_list_iter); +return PTR_ERR (conn_iter_state); + } + crtc_state->mode_changed = true; + } +drm_connector_list_iter_end (&conn_list_iter); + +return 0 ; +} + +static int +intel_dp_atomic_trans_port_sync_check (struct drm_i915_private *dev_priv, +struct intel_atomic_state *state) +{ +struct drm_connector *connector; +struct drm_crtc_state *crtc_state; +struct drm_connector_state *connector_state; +int i, ret, tile_grp_id =0 ; + +if (INTEL_GEN (dev_priv) <11 ) +return 0 ; + +/* Is tiled, mark all other tiled CRTCs as needing a modeset */ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (!connector->has_tile) +continue ; +if (connector_state->crtc && + tile_grp_id != connector->tile_group->id) { + crtc_state =drm_atomic_get_new_crtc_state (&state->base, + connector_state->crtc); +if (!drm_atomic_crtc_needs_modeset (crtc_state)) +continue ; + + tile_grp_id = connector->tile_group->id;drivers/gpu/drm/i915/display/intel_display.c static int intel_atomic_check_crtcs(struct intel_atomic_state *state) · line 14141
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-0+ }else +continue ; + + ret =intel_dp_modeset_all_tiles (dev_priv, state, tile_grp_id); +if (ret) +return ret; + } + +return 0 ; +}BTW after some more pondering I don't think this alone is sufficient. The tile information may have already disppeared so I believe we also need to make sure we mark all currently synced crtcs as needing a modeset if any of them need a modeset. And I guess that's pretty much the same function we'll need to handle fastset correctly.
Quoted 8 lines
Quoted 3 lines
The crtcs in the current state get synced and master/slave assignments happen in
icl_add_sync_mode_crtcsbeforecompute_configcall So are you suggesting basically moving this function after the crtcs are synced and then just setting modeset to true for all synced crtcs if one of them needs modeset?I think it should look something like:
modeset_tiled_things();modeset_synced_things();but
modeset_synced_things()should be called aftericl_add_sync_crtcs()which as per your review should be called from withinmodeset_pipe_configjust beforecompute_config()call.Quoted 2 lines
for_each(){This is the
for_each_crtcloop inintel_atomic_check()right?Quoted 1 line
modeset_pipes_config();So have
icl_add_sync_crtcsoutside ofmodeset_pipe_config()?Quoted 6 lines
if (can_fastset()) { modeset=false; fastset=true; } }Quoted 6 lines
modeset_synced_things();for_each() { if (!modeset && fastset) copy_state(); }We already do this in the code right?
Manasi
Quoted 9 lines
Quoted 3 lines
And why would the tile information be disappeared?
It'll get updated whenever someone does a
getconnector()or whatever.Example:
1. sync pipe A and B, pipe A is master
2. swap pipe B display for something elseQuoted 6 lines
Quoted 2 lines
If we disconnect and connect other display for pipe B, port sync mode is off and
Pipe A no longer a master and we would reset themaster_slaveassignments and conn on pipe B would not have tilePort sync will stay enabled until we do a modeset to disable it. In this example there is never any modeset on pipe B until we add soemthing to force one in step 4.
seems to be working with the current
intel_dp_add_modeset_tiles()since the moment
we unplug, it goes through the disable sequence where all the associated tiles get disabled and port sync mode gets disabled. This happens because the Pipe A which is still connected now indicates a mode change since it fallsback to a lower non tiled mode.But to be on safer side, i can check if say Pipe A is a master or slave (in port sync mode) , check its
crtc_statewhich is still showing the old master slave links since we havent cleared those yet and then add all other synced crtcs to the modeset?So the order of calls can still be :
intel_atomic_check(){
intel_dp_atomic_tiled_check(){modeset_all_tilesmodeset_synced_crtcs}intel_modeset_pipe_Config()icl_add_sync_crtcscompute_configfastsetcheck()modeset_synced_things(here it looks at the new master slave assignments)Does this look correct, I want send this patch out today
Manasi
Quoted 18 lines
Quoted 17 lines
so why we wold need to add both pipes to modeset in this case at all
Manasi
Quoted 11 lines
Quoted 7 lines
3.
getconnector()-> tile info goes poof
4. do something on pipe A that needs a modeset
no tile info so we miss that pipe B also needs a modesetSignature · 2 lines
Ville Syrjälä Intel_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 7 lines
Quoted 3 lines
Signature · 2 lines
Ville Syrjälä Intel_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 2/3 drm/i915/dp: Make port sync mode assignments only if all tiles present
In reply to Manasi NavareAdd an extra check before making master slave assignments for tiled displays to make sure we make these assignments only if all tiled connectors are present. If not then initialize the state to defaults so it does a normal non tiled modeset without transcoder port sync.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
--- drivers/gpu/drm/i915/display/intel_display.c | 28 ++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/display/intel_display.c +26 -2
@@ -12048,6 +12048,12 @@ static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state)return !old_crtc_state->c8_planes != !new_crtc_state->c8_planes; } +static void initialize_trans_port_sync_mode_state (struct intel_crtc_state *crtc_state) +{ + crtc_state->master_transcoder =INVALID_TRANSCODER ; + crtc_state->sync_mode_slaves_mask =0 ; +} +static int icl_add_sync_mode_crtcs (struct intel_crtc_state *crtc_state) {struct drm_crtc *crtc = crtc_state->uapi.crtc;
@@ -12059,11 +12065,22 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)struct drm_crtc *master_crtc =NULL ;struct drm_crtc_state *master_crtc_state;struct intel_crtc_state *master_pipe_config; -int i, tile_group_id; +int i, tile_group_id =0 , num_tiled_conns =0 ;if (INTEL_GEN (dev_priv) <11 )return 0 ; +/* If all tiles not present do not make master slave assignments +* Here we assume all tiles belong to the same tile group for now. +*/ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (connector->has_tile) { +if (!tile_group_id) + tile_group_id = connector->tile_group->id; + num_tiled_conns++; + } + } +/* * In case of tiled displays there could be one or more slaves but there is * only one master. Lets make the CRTC used by the connector corresponding
@@ -12077,8 +12094,15 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)if (!connector->has_tile)continue ;if (crtc_state->hw.mode.hdisplay != connector->tile_h_size || - crtc_state->hw.mode.vdisplay != connector->tile_v_size) + crtc_state->hw.mode.vdisplay != connector->tile_v_size) { +initialize_trans_port_sync_mode_state (crtc_state);return 0 ; + } +if (connector->tile_group->id == tile_group_id && + num_tiled_conns < connector->num_h_tile * connector->num_v_tile) { +initialize_trans_port_sync_mode_state (crtc_state); +return 0 ; + }if (connector->tile_h_loc == connector->num_h_tile -1 && connector->tile_v_loc == connector->num_v_tile -1 )continue ;
Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 2/3] drm/i915/dp: Make port sync mode assignments only if all tiles present
Quoted 17 lines · Manasi Navare
Add an extra check before making master slave assignments for tiled displays to make sure we make these assignments only if all tiled connectors are present. If not then initialize the state to defaults so it does a normal non tiled modeset without transcoder port sync.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 28 ++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/display/intel_display.c +0
-0
drivers/gpu/drm/i915/display/intel_display.c static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state) · line 12051
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +1
-0@@ -12048,6 +12048,12 @@ static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state)return !old_crtc_state->c8_planes != !new_crtc_state->c8_planes; } +static void initialize_trans_port_sync_mode_state (struct intel_crtc_state *crtc_state)
"reset" rather than "initialize" might be more consistent with similar things elsewhere in the driver?
Quoted 9 lines
drivers/gpu/drm/i915/display/intel_display.c +5
-0+{ + crtc_state->master_transcoder =INVALID_TRANSCODER ; + crtc_state->sync_mode_slaves_mask =0 ; +} +static int icl_add_sync_mode_crtcs (struct intel_crtc_state *crtc_state) {struct drm_crtc *crtc = crtc_state->uapi.crtc;
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12073
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-1@@ -12059,11 +12065,22 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)struct drm_crtc *master_crtc =NULL ;struct drm_crtc_state *master_crtc_state;struct intel_crtc_state *master_pipe_config; -int i, tile_group_id; +int i, tile_group_id =0 , num_tiled_conns =0 ;if (INTEL_GEN (dev_priv) <11 )return 0 ; +/* If all tiles not present do not make master slave assignments
This comment seems like it should go farther down the function; this block isn't directly responsible for master/slave assignments, so it's a bit confusing here. Also, you might want to clarify what "present" means in this case. E.g., explain that since you've already added all tiles of the same monitor to the transaction (earlier in intel_atomic_check), then if the number of tiles in the tile group is smaller than the size of the tile group it means that at least one of the tiles isn't active.
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12074
Quoted 1 line
drivers/gpu/drm/i915/display/intel_display.c +1
-0+* Here we assume all tiles belong to the same tile group for now.
Should this sentence be a FIXME:? If we plug in 2x 2-tile monitors on TGL, this would be problematic, right?
The logic changes seem correct (if we assume that only a single tiled monitor is in use), so
Matt
Quoted 36 lines
drivers/gpu/drm/i915/display/intel_display.c +17
-1+*/ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (connector->has_tile) { +if (!tile_group_id) + tile_group_id = connector->tile_group->id; + num_tiled_conns++; + } + } +/* * In case of tiled displays there could be one or more slaves but there is * only one master. Lets make the CRTC used by the connector corresponding @@ -12077,8 +12094,15 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)if (!connector->has_tile)continue ;if (crtc_state->hw.mode.hdisplay != connector->tile_h_size || - crtc_state->hw.mode.vdisplay != connector->tile_v_size) + crtc_state->hw.mode.vdisplay != connector->tile_v_size) { +initialize_trans_port_sync_mode_state (crtc_state);return 0 ; + } +if (connector->tile_group->id == tile_group_id && + num_tiled_conns < connector->num_h_tile * connector->num_v_tile) { +initialize_trans_port_sync_mode_state (crtc_state); +return 0 ; + }if (connector->tile_h_loc == connector->num_h_tile -1 && connector->tile_v_loc == connector->num_v_tile -1 )continue ;Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 9 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 2/3] drm/i915/dp: Make port sync mode assignments only if all tiles present
Quoted 21 lines · Matt Roper
Quoted 20 lines · Manasi Navare
Add an extra check before making master slave assignments for tiled displays to make sure we make these assignments only if all tiled connectors are present. If not then initialize the state to defaults so it does a normal non tiled modeset without transcoder port sync.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 28 ++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/display/intel_display.c +0
-0@@ -12048,6 +12048,12 @@ static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state)return !old_crtc_state->c8_planes != !new_crtc_state->c8_planes; }
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state) · line 12051
Quoted 1 line
drivers/gpu/drm/i915/display/intel_display.c +1
-0+static void initialize_trans_port_sync_mode_state (struct intel_crtc_state *crtc_state)"reset" rather than "initialize" might be more consistent with similar things elsewhere in the driver?
Yes I was considering other names like default/reset/initialize, but yes I think reset sounds better, will change that
Quoted 19 lines
Quoted 19 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-1+{ + crtc_state->master_transcoder =INVALID_TRANSCODER ; + crtc_state->sync_mode_slaves_mask =0 ; +} +static int icl_add_sync_mode_crtcs (struct intel_crtc_state *crtc_state) {struct drm_crtc *crtc = crtc_state->uapi.crtc;@@ -12059,11 +12065,22 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)struct drm_crtc *master_crtc =NULL ;struct drm_crtc_state *master_crtc_state;struct intel_crtc_state *master_pipe_config; -int i, tile_group_id; +int i, tile_group_id =0 , num_tiled_conns =0 ;if (INTEL_GEN (dev_priv) <11 )return 0 ;
Quoted 11 lines
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12073
Quoted 1 line
drivers/gpu/drm/i915/display/intel_display.c +1
-0+/* If all tiles not present do not make master slave assignments This comment seems like it should go farther down the function; this block isn't directly responsible for master/slave assignments, so it's a bit confusing here. Also, you might want to clarify what "present" means in this case. E.g., explain that since you've already added all tiles of the same monitor to the transaction (earlier in
intel_atomic_check), then if the number of tiles in the tile group is smaller than the size of the tile group it means that at least one of the tiles isn't active.
Yes will move this comment down
Quoted 1 line
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12074
Quoted 1 line
drivers/gpu/drm/i915/display/intel_display.c +1
-0+* Here we assume all tiles belong to the same tile group for now.
Will add a FIXME
I will make these changes and add your r-b, thanks Matt
Manasi
Quoted 44 lines
Should this sentence be a FIXME:? If we plug in 2x 2-tile monitors on TGL, this would be problematic, right?
The logic changes seem correct (if we assume that only a single tiled monitor is in use), so
Reviewed-by Matt Roper <[email protected]>Matt
Quoted 32 lines
drivers/gpu/drm/i915/display/intel_display.c +17
-1+*/ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (connector->has_tile) { +if (!tile_group_id) + tile_group_id = connector->tile_group->id; + num_tiled_conns++; + } + } +/* * In case of tiled displays there could be one or more slaves but there is * only one master. Lets make the CRTC used by the connector corresponding @@ -12077,8 +12094,15 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)if (!connector->has_tile)continue ;if (crtc_state->hw.mode.hdisplay != connector->tile_h_size || - crtc_state->hw.mode.vdisplay != connector->tile_v_size) + crtc_state->hw.mode.vdisplay != connector->tile_v_size) { +initialize_trans_port_sync_mode_state (crtc_state);return 0 ; + } +if (connector->tile_group->id == tile_group_id && + num_tiled_conns < connector->num_h_tile * connector->num_v_tile) { +initialize_trans_port_sync_mode_state (crtc_state); +return 0 ; + }if (connector->tile_h_loc == connector->num_h_tile -1 && connector->tile_v_loc == connector->num_v_tile -1 )continue ;Signature · 1 line
2.19.1
Quoted 11 lines
Quoted 4 lines
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 5 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 2/3] drm/i915/dp: Make port sync mode assignments only if all tiles present
In reply to Manasi NavareQuoted 36 lines · Manasi Navare
Add an extra check before making master slave assignments for tiled displays to make sure we make these assignments only if all tiled connectors are present. If not then initialize the state to defaults so it does a normal non tiled modeset without transcoder port sync.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 28 ++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/display/intel_display.c +7
-1@@ -12048,6 +12048,12 @@ static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state)return !old_crtc_state->c8_planes != !new_crtc_state->c8_planes; } +static void initialize_trans_port_sync_mode_state (struct intel_crtc_state *crtc_state) +{ + crtc_state->master_transcoder =INVALID_TRANSCODER ; + crtc_state->sync_mode_slaves_mask =0 ; +} +static int icl_add_sync_mode_crtcs (struct intel_crtc_state *crtc_state) {struct drm_crtc *crtc = crtc_state->uapi.crtc;@@ -12059,11 +12065,22 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)struct drm_crtc *master_crtc =NULL ;struct drm_crtc_state *master_crtc_state;struct intel_crtc_state *master_pipe_config; -int i, tile_group_id; +int i, tile_group_id =0 , num_tiled_conns =0 ;
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12079
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +7
-0if (INTEL_GEN (dev_priv) <11 )return 0 ; +/* If all tiles not present do not make master slave assignments +* Here we assume all tiles belong to the same tile group for now. +*/ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (connector->has_tile) { +if (!tile_group_id) + tile_group_id = connector->tile_group->id;
Isn't 0 a valid tile group id?
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12081
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+ num_tiled_conns++; + }
This whole thing looks confused. Should it not just look for the same tile group as what the current connector belongs to?
Quoted 12 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-1+ } +/* * In case of tiled displays there could be one or more slaves but there is * only one master. Lets make the CRTC used by the connector corresponding @@ -12077,8 +12094,15 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)if (!connector->has_tile)continue ;if (crtc_state->hw.mode.hdisplay != connector->tile_h_size || - crtc_state->hw.mode.vdisplay != connector->tile_v_size) + crtc_state->hw.mode.vdisplay != connector->tile_v_size) { +initialize_trans_port_sync_mode_state (crtc_state);
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12108
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0return 0 ; + } +if (connector->tile_group->id == tile_group_id && + num_tiled_conns < connector->num_h_tile * connector->num_v_tile) { +initialize_trans_port_sync_mode_state (crtc_state); +return 0 ; + }if (connector->tile_h_loc == connector->num_h_tile -1 && connector->tile_v_loc == connector->num_v_tile -1 )continue ;
This whole thing seems kinda overly complicated. I suggest it should just blindly go through all connectors of the same tile group and pick the lowest transcoder as the master, which is the logic Jose is using for MST. Except I guess we have to special case the EDP transcoder for port sync since it can't be a slave. So a simple numeric comparison won't quite do like used for MST.
And then we should probably move this thing to the encoder .compute_config(). I suppose it should look in the end something like:
compute_config() {
...
crtc_state->master = compute_master_transcoder();
crtc_state->slaves = 0;
if (master_transcoder == cpu_transcoder)
crtc_state->master = INVALID;
crtc_state->slave = compute_slave_transcoders();
}
}That keeps it very readable and avoids the confusing stuff of comptue_config() for one pipe randomly mutating the states of the other pipes.
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 2/3] drm/i915/dp: Make port sync mode assignments only if all tiles present
Quoted 77 lines · Ville Syrjälä
Quoted 36 lines · Manasi Navare
Add an extra check before making master slave assignments for tiled displays to make sure we make these assignments only if all tiled connectors are present. If not then initialize the state to defaults so it does a normal non tiled modeset without transcoder port sync.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 28 ++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/display/intel_display.c +7
-1@@ -12048,6 +12048,12 @@ static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state)return !old_crtc_state->c8_planes != !new_crtc_state->c8_planes; } +static void initialize_trans_port_sync_mode_state (struct intel_crtc_state *crtc_state) +{ + crtc_state->master_transcoder =INVALID_TRANSCODER ; + crtc_state->sync_mode_slaves_mask =0 ; +} +static int icl_add_sync_mode_crtcs (struct intel_crtc_state *crtc_state) {struct drm_crtc *crtc = crtc_state->uapi.crtc;@@ -12059,11 +12065,22 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)struct drm_crtc *master_crtc =NULL ;struct drm_crtc_state *master_crtc_state;struct intel_crtc_state *master_pipe_config; -int i, tile_group_id; +int i, tile_group_id =0 , num_tiled_conns =0 ;drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12079
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +7
-0if (INTEL_GEN (dev_priv) <11 )return 0 ; +/* If all tiles not present do not make master slave assignments +* Here we assume all tiles belong to the same tile group for now. +*/ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (connector->has_tile) { +if (!tile_group_id) + tile_group_id = connector->tile_group->id;Isn't 0 a valid tile group id?
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12081
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+ num_tiled_conns++; + }This whole thing looks confused. Should it not just look for the same tile group as what the current connector belongs to?
Quoted 21 lines
drivers/gpu/drm/i915/display/intel_display.c +10
-1+ } +/* * In case of tiled displays there could be one or more slaves but there is * only one master. Lets make the CRTC used by the connector corresponding @@ -12077,8 +12094,15 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)if (!connector->has_tile)continue ;if (crtc_state->hw.mode.hdisplay != connector->tile_h_size || - crtc_state->hw.mode.vdisplay != connector->tile_v_size) + crtc_state->hw.mode.vdisplay != connector->tile_v_size) { +initialize_trans_port_sync_mode_state (crtc_state);return 0 ; + } +if (connector->tile_group->id == tile_group_id && + num_tiled_conns < connector->num_h_tile * connector->num_v_tile) { +initialize_trans_port_sync_mode_state (crtc_state); +return 0 ; + }if (connector->tile_h_loc == connector->num_h_tile -1 && connector->tile_v_loc == connector->num_v_tile -1 )
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12108
Quoted 1 line
drivers/gpu/drm/i915/display/intel_display.c +0
-0continue ;This whole thing seems kinda overly complicated. I suggest it should just blindly go through all connectors of the same tile group and pick the lowest transcoder as the master, which is the logic Jose is using for MST. Except I guess we have to special case the EDP transcoder for port sync since it can't be a slave. So a simple numeric comparison won't quite do like used for MST.
Hmm. Except that won't actually work since .cpu_transcoder won't have been populated yet for the later pipes. So we can't use that in .compute_config(). So either we do this after .compute_config() is done for everyone or we just calculate the cpu_transcoder on the spot (I mean it's just a trivial port==A->EDP else->pipe so no big deal).
Quoted 21 lines
And then we should probably move this thing to the encoder .compute_config(). I suppose it should look in the end something like:
compute_config() { ... crtc_state->master = compute_master_transcoder(); crtc_state->slaves = 0; if (master_transcoder == cpu_transcoder) crtc_state->master = INVALID; crtc_state->slave = compute_slave_transcoders(); } }That keeps it very readable and avoids the confusing stuff of
comptue_config()for one pipe randomly mutating the states of the other pipes.Signature · 2 lines
Ville Syrjälä Intel
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 2/3] drm/i915/dp: Make port sync mode assignments only if all tiles present
In reply to Ville SyrjäläQuoted 50 lines · Ville Syrjälä
Quoted 36 lines · Manasi Navare
Add an extra check before making master slave assignments for tiled displays to make sure we make these assignments only if all tiled connectors are present. If not then initialize the state to defaults so it does a normal non tiled modeset without transcoder port sync.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 28 ++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/display/intel_display.c +7
-1@@ -12048,6 +12048,12 @@ static bool c8_planes_changed(const struct intel_crtc_state *new_crtc_state)return !old_crtc_state->c8_planes != !new_crtc_state->c8_planes; } +static void initialize_trans_port_sync_mode_state (struct intel_crtc_state *crtc_state) +{ + crtc_state->master_transcoder =INVALID_TRANSCODER ; + crtc_state->sync_mode_slaves_mask =0 ; +} +static int icl_add_sync_mode_crtcs (struct intel_crtc_state *crtc_state) {struct drm_crtc *crtc = crtc_state->uapi.crtc;@@ -12059,11 +12065,22 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)struct drm_crtc *master_crtc =NULL ;struct drm_crtc_state *master_crtc_state;struct intel_crtc_state *master_pipe_config; -int i, tile_group_id; +int i, tile_group_id =0 , num_tiled_conns =0 ;drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12079
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +7
-0if (INTEL_GEN (dev_priv) <11 )return 0 ; +/* If all tiles not present do not make master slave assignments +* Here we assume all tiles belong to the same tile group for now. +*/ +for_each_new_connector_in_state (&state->base, connector, connector_state, i) { +if (connector->has_tile) { +if (!tile_group_id) + tile_group_id = connector->tile_group->id;Isn't 0 a valid tile group id?
Quoted 5 lines
drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12081
Quoted 2 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-0+ num_tiled_conns++; + }This whole thing looks confused. Should it not just look for the same tile group as what the current connector belongs to?
I was making sure that we dont count the connectors that dont belong to the same tile grp id. But yes I agree that the tile grp id can be assigned by current connector's tile grp id and add num_conns only if they are of the same tile grp id.
Quoted 44 lines
Quoted 13 lines
drivers/gpu/drm/i915/display/intel_display.c +4
-1+ } +/* * In case of tiled displays there could be one or more slaves but there is * only one master. Lets make the CRTC used by the connector corresponding @@ -12077,8 +12094,15 @@ static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state)if (!connector->has_tile)continue ;if (crtc_state->hw.mode.hdisplay != connector->tile_h_size || - crtc_state->hw.mode.vdisplay != connector->tile_v_size) + crtc_state->hw.mode.vdisplay != connector->tile_v_size) { +initialize_trans_port_sync_mode_state (crtc_state);drivers/gpu/drm/i915/display/intel_display.c static int icl_add_sync_mode_crtcs(struct intel_crtc_state *crtc_state) · line 12108
Quoted 10 lines
drivers/gpu/drm/i915/display/intel_display.c +6
-0return 0 ; + } +if (connector->tile_group->id == tile_group_id && + num_tiled_conns < connector->num_h_tile * connector->num_v_tile) { +initialize_trans_port_sync_mode_state (crtc_state); +return 0 ; + }if (connector->tile_h_loc == connector->num_h_tile -1 && connector->tile_v_loc == connector->num_v_tile -1 )continue ;This whole thing seems kinda overly complicated. I suggest it should just blindly go through all connectors of the same tile group and pick the lowest transcoder as the master, which is the logic Jose is using for MST. Except I guess we have to special case the EDP transcoder for port sync since it can't be a slave. So a simple numeric comparison won't quite do like used for MST.
And then we should probably move this thing to the encoder .compute_config(). I suppose it should look in the end something like:
compute_config() { ... crtc_state->master = compute_master_transcoder(); crtc_state->slaves = 0; if (master_transcoder == cpu_transcoder) crtc_state->master = INVALID; crtc_state->slave = compute_slave_transcoders(); } }
Quoted 3 lines
That keeps it very readable and avoids the confusing stuff of
comptue_config()for one pipe randomly mutating the states of the other pipes.
But the logic that we are using is always make the connector corresponding to the last tile the master and all other connectors as slaves since only 1 master (last tile) and multiple slaves.
This was agreed upon with discussions with you and Danvet and Maarten almost 6 months back and the entire enable/disable sequence is based on this.
I would rather keep the logic same for now since the master slave asisgnments which was designed based on consensus during the initial patches and since its working well now.
The only thing that I wanted to add in this patch is reset the master_trans and slave_bitmask properly if all tiles not present in order to handle the hotplug case correctly.
I will make changes for the tile grp id that you suggested for now and we can simplify the logic later as enhncements/optimizations.
Sounds good?
Manasi
Quoted 4 lines
Signature · 3 lines
-- Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 3/3 drm/i915/dp: Disable Port sync mode correctly on teardown
In reply to Manasi NavareWhile clearing the Ports ync mode enable and master select bits we need to make sure that we perform a RMW for disable else it sets the other bits casuing unwanted sideeffects.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
--- drivers/gpu/drm/i915/display/intel_display.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/display/intel_display.c +2 -1
@@ -4599,7 +4599,8 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ transcoder_name(old_crtc_state->cpu_transcoder)); reg =TRANS_DDI_FUNC_CTL2 (old_crtc_state->cpu_transcoder); - trans_ddi_func_ctl2_val = ~(PORT_SYNC_MODE_ENABLE | + trans_ddi_func_ctl2_val =I915_READ (reg); + trans_ddi_func_ctl2_val &= ~(PORT_SYNC_MODE_ENABLE |PORT_SYNC_MODE_MASTER_SELECT_MASK );I915_WRITE (reg, trans_ddi_func_ctl2_val); }
Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 3/3] drm/i915/dp: Disable Port sync mode correctly on teardown
Quoted 9 lines · Manasi Navare
While clearing the Ports ync mode enable and master select bits we need to make sure that we perform a RMW for disable else it sets the other bits casuing unwanted sideeffects.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Cc Jani Nikula <[email protected]>Fixes 51528afe7c5e ("drm/i915/display/icl: Disable transcoder port sync as part of crtc_disable() sequence")Signed-off-by Manasi Navare <[email protected]>
Quoted 25 lines
--- drivers/gpu/drm/i915/display/intel_display.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/display/intel_display.c +2
-1@@ -4599,7 +4599,8 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ transcoder_name(old_crtc_state->cpu_transcoder)); reg =TRANS_DDI_FUNC_CTL2 (old_crtc_state->cpu_transcoder); - trans_ddi_func_ctl2_val = ~(PORT_SYNC_MODE_ENABLE | + trans_ddi_func_ctl2_val =I915_READ (reg); + trans_ddi_func_ctl2_val &= ~(PORT_SYNC_MODE_ENABLE |PORT_SYNC_MODE_MASTER_SELECT_MASK );I915_WRITE (reg, trans_ddi_func_ctl2_val); }Signature · 6 lines
2.19.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 9 lines
Matt Roper Graphics Software Engineer VTT-OSGC Platform Enablement Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 3/3] drm/i915/dp: Disable Port sync mode correctly on teardown
In reply to Manasi NavareQuoted 18 lines · Manasi Navare
While clearing the Ports ync mode enable and master select bits we need to make sure that we perform a RMW for disable else it sets the other bits casuing unwanted sideeffects.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Cc Jani Nikula <[email protected]>Fixes 51528afe7c5e ("drm/i915/display/icl: Disable transcoder port sync as part of crtc_disable() sequence")Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/display/intel_display.c +0
-0
drivers/gpu/drm/i915/display/intel_display.c static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ · line 4605
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-1@@ -4599,7 +4599,8 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ transcoder_name(old_crtc_state->cpu_transcoder)); reg =TRANS_DDI_FUNC_CTL2 (old_crtc_state->cpu_transcoder); - trans_ddi_func_ctl2_val = ~(PORT_SYNC_MODE_ENABLE | + trans_ddi_func_ctl2_val =I915_READ (reg); + trans_ddi_func_ctl2_val &= ~(PORT_SYNC_MODE_ENABLE |PORT_SYNC_MODE_MASTER_SELECT_MASK );I915_WRITE (reg, trans_ddi_func_ctl2_val);
I915_WRITE(TRANS_DDI_FUNC_CTL2, 0);
Quoted 3 lines
drivers/gpu/drm/i915/display/intel_display.c +0
-0}Signature · 1 line
2.19.1
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 3/3] drm/i915/dp: Disable Port sync mode correctly on teardown
Quoted 21 lines · Ville Syrjälä
Quoted 20 lines · Manasi Navare
While clearing the Ports ync mode enable and master select bits we need to make sure that we perform a RMW for disable else it sets the other bits casuing unwanted sideeffects.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Cc Jani Nikula <[email protected]>Fixes 51528afe7c5e ("drm/i915/display/icl: Disable transcoder port sync as part of crtc_disable() sequence")Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/display/intel_display.c +0
-0@@ -4599,7 +4599,8 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ transcoder_name(old_crtc_state->cpu_transcoder));
Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ · line 4605
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-1reg =TRANS_DDI_FUNC_CTL2 (old_crtc_state->cpu_transcoder); - trans_ddi_func_ctl2_val = ~(PORT_SYNC_MODE_ENABLE | + trans_ddi_func_ctl2_val =I915_READ (reg); + trans_ddi_func_ctl2_val &= ~(PORT_SYNC_MODE_ENABLE |PORT_SYNC_MODE_MASTER_SELECT_MASK );I915_WRITE (reg, trans_ddi_func_ctl2_val);
I915_WRITE(TRANS_DDI_FUNC_CTL2, 0);
So not even consider the other values that might have been set in this reg? You would prefer setting this to 0 directly?
Right now i do see that no other bits are set, but things can change when we start using DSI port sync mode or genlock mode etc.
Manasi
Quoted 8 lines
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +0
-0}Signature · 1 line
2.19.1Signature · 2 lines
Ville Syrjälä Intel
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 3/3] drm/i915/dp: Disable Port sync mode correctly on teardown
Quoted 31 lines · Manasi Navare
Quoted 21 lines · Ville Syrjälä
Quoted 20 lines · Manasi Navare
While clearing the Ports ync mode enable and master select bits we need to make sure that we perform a RMW for disable else it sets the other bits casuing unwanted sideeffects.
Bugzilla: https://gitlab.freedesktop.org/drm/intel/issues/5
Cc Ville Syrjälä <[email protected]>Cc Jani Nikula <[email protected]>Fixes 51528afe7c5e ("drm/i915/display/icl: Disable transcoder port sync as part of crtc_disable() sequence")Signed-off-by Manasi Navare <[email protected]>--- drivers/gpu/drm/i915/display/intel_display.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/display/intel_display.c +0
-0@@ -4599,7 +4599,8 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ transcoder_name(old_crtc_state->cpu_transcoder));Quoted 8 lines
drivers/gpu/drm/i915/display/intel_display.c static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ · line 4605
Quoted 6 lines
drivers/gpu/drm/i915/display/intel_display.c +2
-1reg =TRANS_DDI_FUNC_CTL2 (old_crtc_state->cpu_transcoder); - trans_ddi_func_ctl2_val = ~(PORT_SYNC_MODE_ENABLE | + trans_ddi_func_ctl2_val =I915_READ (reg); + trans_ddi_func_ctl2_val &= ~(PORT_SYNC_MODE_ENABLE |PORT_SYNC_MODE_MASTER_SELECT_MASK );I915_WRITE (reg, trans_ddi_func_ctl2_val);
I915_WRITE(TRANS_DDI_FUNC_CTL2, 0);
Quoted 4 lines
So not even consider the other values that might have been set in this reg? You would prefer setting this to 0 directly?
Right now i do see that no other bits are set, but things can change when we start using DSI port sync mode or genlock mode etc.
No point in making it more complicated than it has to be.
In fact after the series I just posted we should probably just suck this whole thing into intel_ddi_{enable,disable}_transcoder_func() since on older platforms these bits live in TRANS_DDI_FUNC_CTL itself.
Quoted 11 lines
Manasi
Quoted 8 lines
Quoted 4 lines
drivers/gpu/drm/i915/display/intel_display.c +0
-0}Signature · 1 line
2.19.1Signature · 2 lines
Ville Syrjälä Intel
Signature · 6 lines
Ville Syrjälä Intel _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for i915 fixes to handle hotplug cases on 8K tiled monitor
In reply to Manasi Navare== Series Details ==
== Series Details ==
Series: i915 fixes to handle hotplug cases on 8K tiled monitor
URL : https://patchwork.freedesktop.org/series/70788/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
8279f374e563 drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
-:84: CHECK:BRACES: braces {} should be used on all arms of this statement
#84: FILE: drivers/gpu/drm/i915/display/intel_display.c:14124:
+ if (connector_state->crtc &&
[...]
+ } else
[...]-:92: CHECK:BRACES: Unbalanced braces around else statement #92: FILE: drivers/gpu/drm/i915/display/intel_display.c:14132:
+ } else
total: 0 errors, 0 warnings, 2 checks, 90 lines checked4fe2b74a1644 drm/i915/dp: Make port sync mode assignments only if all tiles present 58e4b57a23e0 drm/i915/dp: Disable Port sync mode correctly on teardown
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI[Intel-gfx] ✓ Fi.CI.BAT: success for i915 fixes to handle hotplug cases on 8K tiled monitor
In reply to Manasi Navare== Series Details ==
== Series Details ==
Series: i915 fixes to handle hotplug cases on 8K tiled monitor
URL : https://patchwork.freedesktop.org/series/70788/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_7545 -> Patchwork_15700
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/index.html
Known issues
------------
Here are the changes found in Patchwork_15700 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@i915_selftest@live_gem_contexts:
- fi-hsw-peppy: [PASS][1] -> [INCOMPLETE][2] ([i915#694])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-hsw-peppy/igt@i915_selftest@live_gem_contexts.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-hsw-peppy/igt@i915_selftest@live_gem_contexts.html
* igt@i915_selftest@live_requests:
- fi-hsw-4770: [PASS][3] -> [INCOMPLETE][4] ([i915#773])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-hsw-4770/igt@i915_selftest@live_requests.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-hsw-4770/igt@i915_selftest@live_requests.html
#### Possible fixes ####
* igt@kms_chamelium@hdmi-hpd-fast:
- fi-kbl-7500u: [FAIL][5] ([fdo#111096] / [i915#323]) -> [PASS][6]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-kbl-7500u/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-kbl-7500u/igt@[email protected]
#### Warnings ####
* igt@i915_module_load@reload:
- fi-icl-u2: [DMESG-WARN][7] ([i915#109] / [i915#289]) -> [DMESG-WARN][8] ([i915#289])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-icl-u2/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-icl-u2/igt@[email protected]
* igt@i915_pm_rpm@basic-rte:
- fi-kbl-guc: [SKIP][9] ([fdo#109271]) -> [FAIL][10] ([i915#704])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-kbl-guc/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-kbl-guc/igt@[email protected]
* igt@kms_flip@basic-flip-vs-modeset:
- fi-kbl-x1275: [DMESG-WARN][11] ([i915#62] / [i915#92] / [i915#95]) -> [DMESG-WARN][12] ([i915#62] / [i915#92]) +4 similar issues
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-kbl-x1275/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-kbl-x1275/igt@[email protected]
* igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a:
- fi-kbl-x1275: [DMESG-WARN][13] ([i915#62] / [i915#92]) -> [DMESG-WARN][14] ([i915#62] / [i915#92] / [i915#95]) +5 similar issues
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/fi-kbl-x1275/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/fi-kbl-x1275/igt@[email protected]
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#111096]: https://bugs.freedesktop.org/show_bug.cgi?id=111096
[i915#109]: https://gitlab.freedesktop.org/drm/intel/issues/109
[i915#289]: https://gitlab.freedesktop.org/drm/intel/issues/289
[i915#323]: https://gitlab.freedesktop.org/drm/intel/issues/323
[i915#472]: https://gitlab.freedesktop.org/drm/intel/issues/472
[i915#62]: https://gitlab.freedesktop.org/drm/intel/issues/62
[i915#694]: https://gitlab.freedesktop.org/drm/intel/issues/694
[i915#704]: https://gitlab.freedesktop.org/drm/intel/issues/704
[i915#707]: https://gitlab.freedesktop.org/drm/intel/issues/707
[i915#773]: https://gitlab.freedesktop.org/drm/intel/issues/773
[i915#92]: https://gitlab.freedesktop.org/drm/intel/issues/92
[i915#95]: https://gitlab.freedesktop.org/drm/intel/issues/95
Participating hosts (52 -> 45)
------------------------------
Additional (1): fi-kbl-7560u
Missing (8): fi-hsw-4770r fi-tgl-guc fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-ctg-p8600 fi-byt-clapper fi-bdw-samus
Build changes
-------------
* CI: CI-20190529 -> None
* Linux: CI_DRM_7545 -> Patchwork_15700
CI-20190529: 20190529
CI_DRM_7545: b1b808dff985c3c2050b20771050453589a60ca3 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_5346: 466b0e6cbcbaccff012b484d1fd7676364b37b93 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_15700: 58e4b57a23e011523571977f21a2d476e7f349fb @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
58e4b57a23e0 drm/i915/dp: Disable Port sync mode correctly on teardown 4fe2b74a1644 drm/i915/dp: Make port sync mode assignments only if all tiles present 8279f374e563 drm/i915/dp: Make sure all tiled connectors get added to the state with full modeset
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/index.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI[Intel-gfx] ✓ Fi.CI.IGT: success for i915 fixes to handle hotplug cases on 8K tiled monitor
In reply to Manasi Navare== Series Details ==
== Series Details ==
Series: i915 fixes to handle hotplug cases on 8K tiled monitor
URL : https://patchwork.freedesktop.org/series/70788/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_7545_full -> Patchwork_15700_full
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Known issues
------------
Here are the changes found in Patchwork_15700_full that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@gem_busy@extended-parallel-vcs1:
- shard-iclb: [PASS][1] -> [SKIP][2] ([fdo#112080]) +3 similar issues
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb2/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb8/igt@[email protected]
* igt@gem_ctx_shared@exec-single-timeline-bsd:
- shard-iclb: [PASS][3] -> [SKIP][4] ([fdo#110841])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb5/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb4/igt@[email protected]
* igt@gem_ctx_shared@q-smoketest-bsd:
- shard-tglb: [PASS][5] -> [INCOMPLETE][6] ([i915#461])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb3/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb4/igt@[email protected]
* igt@gem_eio@kms:
- shard-kbl: [PASS][7] -> [DMESG-WARN][8] ([i915#180] / [i915#56])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-kbl7/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-kbl2/igt@[email protected]
* igt@gem_exec_balancer@nop:
- shard-tglb: [PASS][9] -> [INCOMPLETE][10] ([fdo#111736])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb3/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb5/igt@[email protected]
* igt@gem_exec_parallel@fds:
- shard-tglb: [PASS][11] -> [INCOMPLETE][12] ([i915#470])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb2/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb6/igt@[email protected]
* igt@gem_exec_schedule@preemptive-hang-bsd:
- shard-iclb: [PASS][13] -> [SKIP][14] ([fdo#112146]) +2 similar issues
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb6/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb1/igt@[email protected]
* igt@gem_softpin@noreloc-s3:
- shard-apl: [PASS][15] -> [DMESG-WARN][16] ([i915#180]) +2 similar issues
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-apl8/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-apl6/igt@[email protected]
* igt@i915_pm_rpm@drm-resources-equal:
- shard-tglb: [PASS][17] -> [DMESG-WARN][18] ([i915#766])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb6/igt@[email protected]
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb8/igt@[email protected]
* igt@i915_pm_rpm@system-suspend-devices:
- shard-tglb: [PASS][19] -> [INCOMPLETE][20] ([i915#456] / [i915#460]) +1 similar issue
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb1/igt@[email protected]
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb5/igt@[email protected]
* igt@i915_pm_rps@waitboost:
- shard-tglb: [PASS][21] -> [FAIL][22] ([i915#413])
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb3/igt@[email protected]
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb3/igt@[email protected]
* igt@kms_color@pipe-b-ctm-0-5:
- shard-skl: [PASS][23] -> [DMESG-WARN][24] ([i915#109])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl9/igt@[email protected]
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl4/igt@[email protected]
* igt@kms_cursor_crc@pipe-b-cursor-256x85-random:
- shard-skl: [PASS][25] -> [FAIL][26] ([i915#54]) +1 similar issue
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl1/igt@[email protected]
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl1/igt@[email protected]
* igt@kms_draw_crc@draw-method-xrgb2101010-mmap-gtt-ytiled:
- shard-skl: [PASS][27] -> [FAIL][28] ([i915#52] / [i915#54])
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl6/igt@[email protected]
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl10/igt@[email protected]
* igt@kms_draw_crc@draw-method-xrgb2101010-render-ytiled:
- shard-tglb: [PASS][29] -> [INCOMPLETE][30] ([i915#667]) +1 similar issue
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb7/igt@[email protected]
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb8/igt@[email protected]
* igt@kms_flip@flip-vs-expired-vblank-interruptible:
- shard-skl: [PASS][31] -> [FAIL][32] ([i915#79])
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl3/igt@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl8/igt@[email protected]
* igt@kms_flip@flip-vs-suspend:
- shard-skl: [PASS][33] -> [INCOMPLETE][34] ([i915#221])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl1/igt@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl4/igt@[email protected]
* igt@kms_flip@plain-flip-fb-recreate:
- shard-skl: [PASS][35] -> [FAIL][36] ([i915#34])
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl4/igt@[email protected]
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl3/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-blt:
- shard-tglb: [PASS][37] -> [FAIL][38] ([i915#49]) +2 similar issues
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb4/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb5/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbc-1p-rte:
- shard-tglb: [PASS][39] -> [INCOMPLETE][40] ([i915#474] / [i915#667])
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb6/igt@[email protected]
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb2/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbc-suspend:
- shard-tglb: [PASS][41] -> [INCOMPLETE][42] ([i915#456] / [i915#460] / [i915#474])
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb8/igt@[email protected]
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb2/igt@[email protected]
* igt@kms_plane@pixel-format-pipe-a-planes:
- shard-kbl: [PASS][43] -> [INCOMPLETE][44] ([fdo#103665] / [i915#648] / [i915#667])
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-kbl4/igt@[email protected]
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-kbl6/igt@[email protected]
* igt@kms_psr@psr2_cursor_render:
- shard-iclb: [PASS][45] -> [SKIP][46] ([fdo#109441]) +1 similar issue
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb2/igt@kms_psr@psr2_cursor_render.html
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb8/igt@kms_psr@psr2_cursor_render.html
* igt@kms_setmode@basic:
- shard-hsw: [PASS][47] -> [FAIL][48] ([i915#31])
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-hsw4/igt@[email protected]
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-hsw7/igt@[email protected]
* igt@kms_vblank@pipe-a-ts-continuation-suspend:
- shard-kbl: [PASS][49] -> [DMESG-WARN][50] ([i915#180]) +6 similar issues
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-kbl6/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-kbl1/igt@[email protected]
* igt@kms_vblank@pipe-d-ts-continuation-suspend:
- shard-tglb: [PASS][51] -> [INCOMPLETE][52] ([i915#460])
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb6/igt@[email protected]
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb3/igt@[email protected]
* igt@prime_vgem@fence-wait-bsd2:
- shard-iclb: [PASS][53] -> [SKIP][54] ([fdo#109276]) +5 similar issues
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb2/igt@[email protected]
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb7/igt@[email protected]
#### Possible fixes ####
* igt@gem_ctx_isolation@vcs0-s3:
- shard-skl: [INCOMPLETE][55] ([i915#69]) -> [PASS][56]
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl5/igt@[email protected]
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl8/igt@[email protected]
* igt@gem_ctx_persistence@vcs1-queued:
- shard-iclb: [SKIP][57] ([fdo#109276] / [fdo#112080]) -> [PASS][58] +2 similar issues
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb7/igt@[email protected]
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb2/igt@[email protected]
* igt@gem_eio@unwedge-stress:
- shard-snb: [FAIL][59] ([i915#232]) -> [PASS][60]
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-snb6/igt@[email protected]
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-snb2/igt@[email protected]
* igt@gem_exec_balancer@bonded-slice:
- shard-tglb: [FAIL][61] ([i915#800]) -> [PASS][62]
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb4/igt@[email protected]
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb5/igt@[email protected]
* igt@gem_exec_parallel@basic:
- shard-tglb: [INCOMPLETE][63] ([i915#476]) -> [PASS][64]
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb6/igt@[email protected]
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb8/igt@[email protected]
* igt@gem_exec_schedule@fifo-bsd:
- shard-iclb: [SKIP][65] ([fdo#112146]) -> [PASS][66] +1 similar issue
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb2/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb7/igt@[email protected]
* {igt@gem_exec_schedule@pi-shared-iova-bsd2}:
- shard-iclb: [SKIP][67] ([fdo#109276]) -> [PASS][68] +8 similar issues
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb5/igt@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb1/igt@[email protected]
* igt@gem_exec_schedule@preempt-queue-contexts-blt:
- shard-tglb: [INCOMPLETE][69] ([fdo#111606] / [fdo#111677]) -> [PASS][70]
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb6/igt@[email protected]
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb1/igt@[email protected]
* igt@gem_exec_suspend@basic-s0:
- shard-iclb: [DMESG-WARN][71] ([fdo#111764]) -> [PASS][72]
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb1/igt@[email protected]
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb2/igt@[email protected]
* igt@gem_userptr_blits@map-fixed-invalidate-busy:
- shard-snb: [DMESG-WARN][73] ([fdo#111870]) -> [PASS][74]
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-snb6/igt@[email protected]
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-snb2/igt@[email protected]
* igt@i915_selftest@live_requests:
- shard-tglb: [INCOMPLETE][75] ([fdo#112057]) -> [PASS][76]
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb7/igt@i915_selftest@live_requests.html
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb5/igt@i915_selftest@live_requests.html
* igt@i915_selftest@mock_sanitycheck:
- shard-hsw: [DMESG-WARN][77] ([i915#747]) -> [PASS][78]
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-hsw4/igt@i915_selftest@mock_sanitycheck.html
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-hsw8/igt@i915_selftest@mock_sanitycheck.html
* igt@i915_suspend@debugfs-reader:
- shard-tglb: [INCOMPLETE][79] ([i915#456] / [i915#460]) -> [PASS][80]
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb4/igt@[email protected]
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb8/igt@[email protected]
* igt@i915_suspend@fence-restore-tiled2untiled:
- shard-apl: [DMESG-WARN][81] ([i915#180]) -> [PASS][82] +1 similar issue
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-apl1/igt@[email protected]
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-apl3/igt@[email protected]
* igt@kms_cursor_crc@pipe-a-cursor-128x42-sliding:
- shard-skl: [FAIL][83] ([i915#54]) -> [PASS][84] +2 similar issues
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl5/igt@[email protected]
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl6/igt@[email protected]
* igt@kms_cursor_crc@pipe-c-cursor-256x256-offscreen:
- shard-hsw: [DMESG-WARN][85] ([IGT#6]) -> [PASS][86]
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-hsw6/igt@[email protected]
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-hsw2/igt@[email protected]
* igt@kms_draw_crc@draw-method-xrgb8888-mmap-cpu-untiled:
- shard-skl: [FAIL][87] ([i915#177] / [i915#52] / [i915#54]) -> [PASS][88]
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl3/igt@[email protected]
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl8/igt@[email protected]
* igt@kms_flip@2x-flip-vs-expired-vblank-interruptible:
- shard-glk: [FAIL][89] ([i915#79]) -> [PASS][90]
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-glk7/igt@[email protected]
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-glk4/igt@[email protected]
* igt@kms_flip@flip-vs-suspend:
- shard-snb: [INCOMPLETE][91] ([i915#82]) -> [PASS][92]
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-snb1/igt@[email protected]
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-snb5/igt@[email protected]
* igt@kms_flip@plain-flip-interruptible:
- shard-hsw: [INCOMPLETE][93] ([i915#61]) -> [PASS][94]
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-hsw2/igt@[email protected]
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-hsw5/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbc-suspend:
- shard-kbl: [DMESG-WARN][95] ([i915#180]) -> [PASS][96] +4 similar issues
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-kbl4/igt@[email protected]
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-kbl6/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbcpsr-rgb565-draw-pwrite:
- shard-tglb: [FAIL][97] ([i915#49]) -> [PASS][98]
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-tglb3/igt@[email protected]
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-tglb5/igt@[email protected]
* igt@kms_plane_alpha_blend@pipe-a-coverage-7efc:
- shard-skl: [FAIL][99] ([fdo#108145]) -> [PASS][100] +1 similar issue
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl8/igt@[email protected]
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl2/igt@[email protected]
* igt@kms_psr@psr2_cursor_blt:
- shard-iclb: [SKIP][101] ([fdo#109441]) -> [PASS][102]
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb1/igt@kms_psr@psr2_cursor_blt.html
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb2/igt@kms_psr@psr2_cursor_blt.html
* igt@kms_sequence@get-forked-busy:
- shard-snb: [SKIP][103] ([fdo#109271]) -> [PASS][104] +1 similar issue
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-snb2/igt@[email protected]
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-snb7/igt@[email protected]
* igt@perf_pmu@init-busy-vcs1:
- shard-iclb: [SKIP][105] ([fdo#112080]) -> [PASS][106] +6 similar issues
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb5/igt@[email protected]
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb4/igt@[email protected]
#### Warnings ####
* igt@gem_ctx_isolation@vcs1-nonpriv:
- shard-iclb: [FAIL][107] ([IGT#28]) -> [SKIP][108] ([fdo#109276] / [fdo#112080])
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-iclb2/igt@[email protected]
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-iclb7/igt@[email protected]
* igt@gem_eio@kms:
- shard-snb: [DMESG-WARN][109] ([i915#444]) -> [INCOMPLETE][110] ([i915#82])
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-snb6/igt@[email protected]
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-snb4/igt@[email protected]
* igt@kms_plane@pixel-format-pipe-a-planes:
- shard-skl: [INCOMPLETE][111] ([fdo#112347] / [i915#648] / [i915#667]) -> [INCOMPLETE][112] ([fdo#112391] / [i915#648] / [i915#667])
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl7/igt@[email protected]
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl2/igt@[email protected]
* igt@kms_plane@pixel-format-pipe-b-planes:
- shard-skl: [INCOMPLETE][113] ([fdo#112391] / [i915#648] / [i915#667]) -> [INCOMPLETE][114] ([fdo#112347] / [fdo#112391] / [i915#648] / [i915#667])
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7545/shard-skl10/igt@[email protected]
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/shard-skl1/igt@[email protected]
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[IGT#28]: https://gitlab.freedesktop.org/drm/igt-gpu-tools/issues/28
[IGT#6]: https://gitlab.freedesktop.org/drm/igt-gpu-tools/issues/6
[fdo#103665]: https://bugs.freedesktop.org/show_bug.cgi?id=103665
[fdo#108145]: https://bugs.freedesktop.org/show_bug.cgi?id=108145
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109276]: https://bugs.freedesktop.org/show_bug.cgi?id=109276
[fdo#109441]: https://bugs.freedesktop.org/show_bug.cgi?id=109441
[fdo#110841]: https://bugs.freedesktop.org/show_bug.cgi?id=110841
[fdo#111606]: https://bugs.freedesktop.org/show_bug.cgi?id=111606
[fdo#111677]: https://bugs.freedesktop.org/show_bug.cgi?id=111677
[fdo#111736]: https://bugs.freedesktop.org/show_bug.cgi?id=111736
[fdo#111764]: https://bugs.freedesktop.org/show_bug.cgi?id=111764
[fdo#111870]: https://bugs.freedesktop.org/show_bug.cgi?id=111870
[fdo#112057]: https://bugs.freedesktop.org/show_bug.cgi?id=112057
[fdo#112080]: https://bugs.freedesktop.org/show_bug.cgi?id=112080
[fdo#112146]: https://bugs.freedesktop.org/show_bug.cgi?id=112146
[fdo#112347]: https://bugs.freedesktop.org/show_bug.cgi?id=112347
[fdo#112391]: https://bugs.freedesktop.org/show_bug.cgi?id=112391
[i915#109]: https://gitlab.freedesktop.org/drm/intel/issues/109
[i915#177]: https://gitlab.freedesktop.org/drm/intel/issues/177
[i915#180]: https://gitlab.freedesktop.org/drm/intel/issues/180
[i915#221]: https://gitlab.freedesktop.org/drm/intel/issues/221
[i915#232]: https://gitlab.freedesktop.org/drm/intel/issues/232
[i915#31]: https://gitlab.freedesktop.org/drm/intel/issues/31
[i915#34]: https://gitlab.freedesktop.org/drm/intel/issues/34
[i915#413]: https://gitlab.freedesktop.org/drm/intel/issues/413
[i915#444]: https://gitlab.freedesktop.org/drm/intel/issues/444
[i915#456]: https://gitlab.freedesktop.org/drm/intel/issues/456
[i915#460]: https://gitlab.freedesktop.org/drm/intel/issues/460
[i915#461]: https://gitlab.freedesktop.org/drm/intel/issues/461
[i915#470]: https://gitlab.freedesktop.org/drm/intel/issues/470
[i915#474]: https://gitlab.freedesktop.org/drm/intel/issues/474
[i915#476]: https://gitlab.freedesktop.org/drm/intel/issues/476
[i915#49]: https://gitlab.freedesktop.org/drm/intel/issues/49
[i915#52]: https://gitlab.freedesktop.org/drm/intel/issues/52
[i915#54]: https://gitlab.freedesktop.org/drm/intel/issues/54
[i915#56]: https://gitlab.freedesktop.org/drm/intel/issues/56
[i915#61]: https://gitlab.freedesktop.org/drm/intel/issues/61
[i915#648]: https://gitlab.freedesktop.org/drm/intel/issues/648
[i915#667]: https://gitlab.freedesktop.org/drm/intel/issues/667
[i915#69]: https://gitlab.freedesktop.org/drm/intel/issues/69
[i915#747]: https://gitlab.freedesktop.org/drm/intel/issues/747
[i915#766]: https://gitlab.freedesktop.org/drm/intel/issues/766
[i915#79]: https://gitlab.freedesktop.org/drm/intel/issues/79
[i915#800]: https://gitlab.freedesktop.org/drm/intel/issues/800
[i915#82]: https://gitlab.freedesktop.org/drm/intel/issues/82
Participating hosts (11 -> 11)
------------------------------
No changes in participating hosts
Build changes
-------------
* CI: CI-20190529 -> None
* Linux: CI_DRM_7545 -> Patchwork_15700
CI-20190529: 20190529
CI_DRM_7545: b1b808dff985c3c2050b20771050453589a60ca3 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_5346: 466b0e6cbcbaccff012b484d1fd7676364b37b93 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_15700: 58e4b57a23e011523571977f21a2d476e7f349fb @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_15700/index.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx