[PATCH 0/9] DC Patches September 25, 2020

[PATCH 0/9] DC Patches September 25, 2020

[Rodrigo Siqueira]

This DC patchset brings improvements in multiple areas. In summary, we
highlight:
 
* P-state refinements;
* Improvements on DSC
* Updates on NV1x

Best Regards

Alvin Lee (1):
  drm/amd/display: Update NV1x SR latency values

Anthony Koo (1):
  drm/amd/display: [FW Promotion] Release 0.0.35

Aric Cyr (1):
  drm/amd/display: Revert check for flip pending before locking pipes

Chiawen Huang (2):
  drm/amd/display: disable stream if pixel clock changed with link
    active
  drm/amd/display: disable stream if pixel clock changed with link
    active

Eric Bernstein (1):
  drm/amd/display: Add dp_set_dsc_pps_info_packet to virtual stream
    encoder

Joshua Aberback (1):
  drm/amd/display: Calc DLG from dummy p-state if full p-state
    unsupported

Wesley Chalmers (1):
  drm/amd/display: Add debug param to force dio disable

Wyatt Wood (1):
  drm/amd/display: Ensure all debug bits are passed to fw

 drivers/gpu/drm/amd/display/dc/core/dc.c      |  74 +++++++++-
 .../drm/amd/display/dc/core/dc_link_hwss.c    |  21 +--
 drivers/gpu/drm/amd/display/dc/dc.h           |   2 +
 drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c |   5 +-
 .../display/dc/dce110/dce110_hw_sequencer.c   |   2 +-
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.c |  12 +-
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.h |   2 +-
 .../drm/amd/display/dc/dcn20/dcn20_resource.c |   4 +-
 .../drm/amd/display/dc/dcn30/dcn30_resource.c | 114 +++++++++------
 .../drm/amd/display/dc/dcn30/dcn30_resource.h |   5 +
 .../gpu/drm/amd/display/dc/inc/core_types.h   |   2 +-
 .../gpu/drm/amd/display/dc/inc/hw_sequencer.h |   2 +-
 .../dc/virtual/virtual_stream_encoder.c       |   7 +
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 132 +++++++++++++++++-
 14 files changed, 305 insertions(+), 79 deletions(-)

-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 1/9] drm/amd/display: Update NV1x SR latency values

[Rodrigo Siqueira]

From: Alvin Lee <alvin.lee2@amd.com>

[Why]
HW team measurement requires updating values

[How]
Update bounding box values

Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
index 18b9465057ff..bf37a229a342 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@@ -298,8 +298,8 @@ static struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = {
 			},
 		},
 	.num_states = 5,
-	.sr_exit_time_us = 8.6,
-	.sr_enter_plus_exit_time_us = 10.9,
+	.sr_exit_time_us = 11.6,
+	.sr_enter_plus_exit_time_us = 13.9,
 	.urgent_latency_us = 4.0,
 	.urgent_latency_pixel_data_only_us = 4.0,
 	.urgent_latency_pixel_mixed_with_vm_data_us = 4.0,
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 2/9] drm/amd/display: Add dp_set_dsc_pps_info_packet to virtual stream encoder

[Rodrigo Siqueira]

From: Eric Bernstein <eric.bernstein@amd.com>

Signed-off-by: Eric Bernstein <eric.bernstein@amd.com>
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../drm/amd/display/dc/core/dc_link_hwss.c    | 21 ++++++++++---------
 .../dc/virtual/virtual_stream_encoder.c       |  7 +++++++
 2 files changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
index dba338c88256..11a619befb42 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c
@@ -493,13 +493,15 @@ void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 				OPTC_DSC_DISABLED, 0, 0);
 
 		/* disable DSC in stream encoder */
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(
-					pipe_ctx->stream_res.stream_enc,
-					OPTC_DSC_DISABLED, 0, 0);
-
-			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
-					pipe_ctx->stream_res.stream_enc, false, NULL);
+		if (dc_is_dp_signal(stream->signal)) {
+
+			if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+				pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(
+						pipe_ctx->stream_res.stream_enc,
+						OPTC_DSC_DISABLED, 0, 0);
+				pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
+							pipe_ctx->stream_res.stream_enc, false, NULL);
+			}
 		}
 
 		/* disable DSC block */
@@ -536,7 +538,6 @@ bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable)
 bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable)
 {
 	struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc;
-	struct dc *dc = pipe_ctx->stream->ctx->dc;
 	struct dc_stream_state *stream = pipe_ctx->stream;
 
 	if (!pipe_ctx->stream->timing.flags.DSC || !dsc)
@@ -559,7 +560,7 @@ bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable)
 
 		DC_LOG_DSC(" ");
 		dsc->funcs->dsc_get_packed_pps(dsc, &dsc_cfg, &dsc_packed_pps[0]);
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+		if (dc_is_dp_signal(stream->signal)) {
 			DC_LOG_DSC("Setting stream encoder DSC PPS SDP for engine %d\n", (int)pipe_ctx->stream_res.stream_enc->id);
 			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
 									pipe_ctx->stream_res.stream_enc,
@@ -568,7 +569,7 @@ bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable)
 		}
 	} else {
 		/* disable DSC PPS in stream encoder */
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+		if (dc_is_dp_signal(stream->signal)) {
 			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet(
 						pipe_ctx->stream_res.stream_enc, false, NULL);
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
index f0a0d419e555..1053b165c139 100644
--- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
@@ -99,6 +99,12 @@ static void virtual_setup_stereo_sync(
 			bool enable)
 {}
 
+static void virtual_stream_encoder_set_dsc_pps_info_packet(
+		struct stream_encoder *enc,
+		bool enable,
+		uint8_t *dsc_packed_pps)
+{}
+
 static const struct stream_encoder_funcs virtual_str_enc_funcs = {
 	.dp_set_odm_combine =
 		virtual_enc_dp_set_odm_combine,
@@ -128,6 +134,7 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = {
 	.hdmi_reset_stream_attribute = virtual_stream_encoder_reset_hdmi_stream_attribute,
 	.dig_connect_to_otg = virtual_dig_connect_to_otg,
 	.setup_stereo_sync = virtual_setup_stereo_sync,
+	.dp_set_dsc_pps_info_packet = virtual_stream_encoder_set_dsc_pps_info_packet,
 };
 
 bool virtual_stream_encoder_construct(
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 3/9] drm/amd/display: Ensure all debug bits are passed to fw

[Rodrigo Siqueira]

From: Wyatt Wood <wyatt.wood@amd.com>

[Why]
Some debug bits are not being copied from driver to fw.

[How]
Copy debug bits properly.

Signed-off-by: Wyatt Wood <wyatt.wood@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
index bf24f1029547..67af67ef2865 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
@@ -256,10 +256,11 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub,
 	copy_settings_data->smu_optimizations_en		= psr_context->allow_smu_optimizations;
 	copy_settings_data->frame_delay				= psr_context->frame_delay;
 	copy_settings_data->frame_cap_ind			= psr_context->psrFrameCaptureIndicationReq;
+	copy_settings_data->init_sdp_deadline			= psr_context->sdpTransmitLineNumDeadline;
+	copy_settings_data->debug.u32All = 0;
 	copy_settings_data->debug.bitfields.visual_confirm	= dc->dc->debug.visual_confirm == VISUAL_CONFIRM_PSR ?
 									true : false;
-	copy_settings_data->debug.bitfields.use_hw_lock_mgr	= 1;
-	copy_settings_data->init_sdp_deadline			= psr_context->sdpTransmitLineNumDeadline;
+	copy_settings_data->debug.bitfields.use_hw_lock_mgr		= 1;
 
 	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
 	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 4/9] drm/amd/display: disable stream if pixel clock changed with link active

[Rodrigo Siqueira]

From: Chiawen Huang <chiawen.huang@amd.com>

[Why]
Vbios uses preferred timing to turn on edp but OS could use other
timing. If change pixel clock when link active, there is unexpected
garbage on monitor.

[How]
Once pixel clock changed, the driver needs to disable stream.

Signed-off-by: Chiawen Huang <chiawen.huang@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 62 ++++++++++++++++++-
 .../display/dc/dce110/dce110_hw_sequencer.c   |  2 +-
 2 files changed, 60 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 7e74ddc1c708..9fabe264cdea 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -842,6 +842,61 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context)
 	dc_release_state(current_ctx);
 }
 
+static void disable_vbios_mode_if_required(
+		struct dc *dc,
+		struct dc_state *context)
+{
+	unsigned int i;
+
+	/* check if timing_changed, disable stream*/
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct dc_stream_state *stream = NULL;
+		struct dc_link *link = NULL;
+		struct pipe_ctx *pipe = NULL;
+
+		pipe = &context->res_ctx.pipe_ctx[i];
+		stream = pipe->stream;
+		if (stream == NULL)
+			continue;
+
+		if (stream->link->local_sink &&
+			stream->link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
+			link = stream->link;
+			break;
+		}
+
+		if (link != NULL) {
+			unsigned int enc_inst, tg_inst = 0;
+			unsigned int pix_clk_100hz;
+
+			enc_inst = link->link_enc->funcs->get_dig_frontend(link->link_enc);
+			if (enc_inst != ENGINE_ID_UNKNOWN) {
+				for (i = 0; i < dc->res_pool->stream_enc_count; i++) {
+					if (dc->res_pool->stream_enc[i]->id == enc_inst) {
+						tg_inst = dc->res_pool->stream_enc[i]->funcs->dig_source_otg(
+							dc->res_pool->stream_enc[i]);
+						break;
+					}
+				}
+
+				dc->res_pool->dp_clock_source->funcs->get_pixel_clk_frequency_100hz(
+					dc->res_pool->dp_clock_source,
+					tg_inst, &pix_clk_100hz);
+
+				if (link->link_status.link_active) {
+					uint32_t requested_pix_clk_100hz =
+						pipe->stream_res.pix_clk_params.requested_pix_clk_100hz;
+
+					if (pix_clk_100hz != requested_pix_clk_100hz) {
+						core_link_disable_stream(pipe);
+						pipe->stream->dpms_off = false;
+					}
+				}
+			}
+		}
+	}
+}
+
 static void wait_for_no_pipes_pending(struct dc *dc, struct dc_state *context)
 {
 	int i;
@@ -1278,13 +1333,14 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 	for (i = 0; i < context->stream_count; i++)
 		dc_streams[i] =  context->streams[i];
 
-	if (!dcb->funcs->is_accelerated_mode(dcb))
+	if (!dcb->funcs->is_accelerated_mode(dcb)) {
+		disable_vbios_mode_if_required(dc, context);
 		dc->hwss.enable_accelerated_mode(dc, context);
+	}
 
-	for (i = 0; i < context->stream_count; i++) {
+	for (i = 0; i < context->stream_count; i++)
 		if (context->streams[i]->apply_seamless_boot_optimization)
 			dc->optimize_seamless_boot_streams++;
-	}
 
 	if (context->stream_count > dc->optimize_seamless_boot_streams ||
 		context->stream_count == 0)
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 27a1262a20f6..c73768ed250e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -1654,7 +1654,7 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 		// enable fastboot if backend is enabled on eDP
 		if (edp_link->link_enc->funcs->is_dig_enabled(edp_link->link_enc)) {
 			/* Set optimization flag on eDP stream*/
-			if (edp_stream) {
+			if (edp_stream && edp_link->link_status.link_active) {
 				edp_stream->apply_edp_fast_boot_optimization = true;
 				can_apply_edp_fast_boot = true;
 			}
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 5/9] drm/amd/display: Calc DLG from dummy p-state if full p-state unsupported

[Rodrigo Siqueira]

From: Joshua Aberback <joshua.aberback@amd.com>

[Why]
Currently, when full p-state changes are not supported, DLG parameters
are calculated for no p-state support at all. However, we are required
to always support dummy p-state changes, so we should instead calculate
DLG based on dummy p-state latency when full p-state is unsupported.
This behaviour already exists for DCN2.

[How]
 - move DLG calculation inside WM calculation
 - if p-state unsupported, do not recalculate for set A, instead copy from
set C, and perform DLG calculation with dummy p-state latency

Signed-off-by: Joshua Aberback <joshua.aberback@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../drm/amd/display/dc/dcn30/dcn30_resource.c | 114 ++++++++++++------
 .../drm/amd/display/dc/dcn30/dcn30_resource.h |   5 +
 .../gpu/drm/amd/display/dc/inc/core_types.h   |   2 +-
 3 files changed, 80 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
index dde87baf1370..7f3354b3512d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@@ -2127,7 +2127,7 @@ static bool dcn30_internal_validate_bw(
 	return out;
 }
 
-static void dcn30_calculate_wm(
+void dcn30_calculate_wm_and_dlg(
 		struct dc *dc, struct dc_state *context,
 		display_e2e_pipe_params_st *pipes,
 		int pipe_cnt,
@@ -2135,6 +2135,8 @@ static void dcn30_calculate_wm(
 {
 	int i, pipe_idx;
 	double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
+	bool pstate_en = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] !=
+			dm_dram_clock_change_unsupported;
 
 	if (context->bw_ctx.dml.soc.min_dcfclk > dcfclk)
 		dcfclk = context->bw_ctx.dml.soc.min_dcfclk;
@@ -2168,30 +2170,12 @@ static void dcn30_calculate_wm(
 	pipes[0].clks_cfg.voltage = vlevel;
 	pipes[0].clks_cfg.dcfclk_mhz = dcfclk;
 
-	/* Set C:
-	 * DCFCLK: Min Required
-	 * FCLK(proportional to UCLK): 1GHz or Max
-	 * pstate latency overriden to 5us
-	 */
-	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid) {
-		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us;
-		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us;
-		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us;
-	}
-	context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.c.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-
 	/* Set D:
 	 * DCFCLK: Min Required
 	 * FCLK(proportional to UCLK): 1GHz or Max
 	 * sr_enter_exit = 4, sr_exit = 2us
 	 */
+	/*
 	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].valid) {
 		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us;
 		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us;
@@ -2205,29 +2189,72 @@ static void dcn30_calculate_wm(
 	context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.d.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	*/
 
-	/* Set A:
+	/* Set C:
 	 * DCFCLK: Min Required
 	 * FCLK(proportional to UCLK): 1GHz or Max
-	 *
-	 * Set A calculated last so that following calculations are based on Set A
+	 * pstate latency overridden to 5us
 	 */
-	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) {
-		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
-		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us;
-		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us;
+	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid) {
+		unsigned int min_dram_speed_mts = context->bw_ctx.dml.vba.DRAMSpeed;
+		unsigned int min_dram_speed_mts_margin = 160;
+
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[0].dummy_pstate_latency_us;
+
+		if (context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] == dm_dram_clock_change_unsupported)
+			min_dram_speed_mts = dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz * 16;
+
+		for (i = 3; i > 0; i--) {
+			if ((min_dram_speed_mts + min_dram_speed_mts_margin > dc->clk_mgr->bw_params->dummy_pstate_table[i].dram_speed_mts) &&
+					(min_dram_speed_mts - min_dram_speed_mts_margin < dc->clk_mgr->bw_params->dummy_pstate_table[i].dram_speed_mts))
+				context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[i].dummy_pstate_latency_us;
+		}
+
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us;
+		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us;
 	}
-	context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-
-	context->perf_params.stutter_period_us =
-		context->bw_ctx.dml.vba.StutterPeriod;
+	context->bw_ctx.bw.dcn.watermarks.c.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.c.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+
+	if (!pstate_en) {
+		/* The only difference between A and C is p-state latency, if p-state is not supported we want to
+		 * calculate DLG based on dummy p-state latency, and max out the set A p-state watermark
+		 */
+		context->bw_ctx.bw.dcn.watermarks.a = context->bw_ctx.bw.dcn.watermarks.c;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = 0x13FFFF;
+	} else {
+		/* Set A:
+		 * DCFCLK: Min Required
+		 * FCLK(proportional to UCLK): 1GHz or Max
+		 *
+		 * Set A calculated last so that following calculations are based on Set A
+		 */
+		if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) {
+			context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
+			context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us;
+			context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us;
+		}
+		context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+		context->bw_ctx.bw.dcn.watermarks.a.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	}
+
+	context->perf_params.stutter_period_us = context->bw_ctx.dml.vba.StutterPeriod;
+
+	/* Make set D = set A until set D is enabled */
+	context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a;
 
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
 		if (!context->res_ctx.pipe_ctx[i].stream)
@@ -2247,6 +2274,13 @@ static void dcn30_calculate_wm(
 
 		pipe_idx++;
 	}
+
+	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+
+	if (!pstate_en)
+		/* Restore full p-state latency */
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us =
+				dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
 }
 
 bool dcn30_validate_bandwidth(struct dc *dc,
@@ -2279,8 +2313,7 @@ bool dcn30_validate_bandwidth(struct dc *dc,
 		goto validate_out;
 	}
 
-	dcn30_calculate_wm(dc, context, pipes, pipe_cnt, vlevel);
-	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
+	dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel);
 
 	BW_VAL_TRACE_END_WATERMARKS();
 
@@ -2448,6 +2481,7 @@ static const struct resource_funcs dcn30_res_pool_funcs = {
 	.link_enc_create = dcn30_link_encoder_create,
 	.panel_cntl_create = dcn30_panel_cntl_create,
 	.validate_bandwidth = dcn30_validate_bandwidth,
+	.calculate_wm_and_dlg = dcn30_calculate_wm_and_dlg,
 	.populate_dml_pipes = dcn30_populate_dml_pipes_from_context,
 	.acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer,
 	.add_stream_to_ctx = dcn30_add_stream_to_ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h
index c9d5f94092a0..d163812af858 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.h
@@ -55,6 +55,11 @@ unsigned int dcn30_calc_max_scaled_time(
 
 bool dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context,
 		bool fast_validate);
+void dcn30_calculate_wm_and_dlg(
+		struct dc *dc, struct dc_state *context,
+		display_e2e_pipe_params_st *pipes,
+		int pipe_cnt,
+		int vlevel);
 void dcn30_populate_dml_writeback_from_context(
 		struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes);
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index 1daa563c8ff4..6e6bc66e49f0 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -101,7 +101,7 @@ struct resource_funcs {
 					struct dc *dc,
 					struct dc_state *context,
 					bool fast_validate);
-	void (*calculate_wm)(
+	void (*calculate_wm_and_dlg)(
 				struct dc *dc, struct dc_state *context,
 				display_e2e_pipe_params_st *pipes,
 				int pipe_cnt,
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 6/9] drm/amd/display: Add debug param to force dio disable

[Rodrigo Siqueira]

From: Wesley Chalmers <Wesley.Chalmers@amd.com>

[WHY]
At the moment, some tests are failing because cur_link_settings is
invalid. As a workaround, add an option to force dio disable.

Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 2 ++
 drivers/gpu/drm/amd/display/dc/dc.h      | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 9fabe264cdea..dc90d26e2663 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -735,6 +735,8 @@ static bool dc_construct(struct dc *dc,
 	dc->clk_mgr->force_smu_not_present = init_params->force_smu_not_present;
 #endif
 
+	dc->debug.force_ignore_link_settings = init_params->force_ignore_link_settings;
+
 	if (dc->res_pool->funcs->update_bw_bounding_box)
 		dc->res_pool->funcs->update_bw_bounding_box(dc, dc->clk_mgr->bw_params);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 061ca15660ac..82fe0ab56e3a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -503,6 +503,7 @@ struct dc_debug_options {
 	bool usbc_combo_phy_reset_wa;
 	bool disable_dsc;
 	bool enable_dram_clock_change_one_display_vactive;
+	bool force_ignore_link_settings;
 };
 
 struct dc_debug_data {
@@ -660,6 +661,7 @@ struct dc_init_data {
 #if defined(CONFIG_DRM_AMD_DC_DCN3_0)
 	bool force_smu_not_present;
 #endif
+	bool force_ignore_link_settings;
 };
 
 struct dc_callback_init {
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 7/9] drm/amd/display: Revert check for flip pending before locking pipes

[Rodrigo Siqueira]

From: Aric Cyr <aric.cyr@amd.com>

This reverts commit e82d8eae9e81af243256f70bec593baed50b0bdb.

Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c             | 11 +++++++++--
 .../drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c    | 12 ++----------
 .../drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h    |  2 +-
 drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h    |  2 +-
 4 files changed, 13 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index dc90d26e2663..76946a7d47ca 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2384,6 +2384,7 @@ static void commit_planes_for_stream(struct dc *dc,
 		enum surface_update_type update_type,
 		struct dc_state *context)
 {
+	bool mpcc_disconnected = false;
 	int i, j;
 	struct pipe_ctx *top_pipe_to_program = NULL;
 
@@ -2414,8 +2415,14 @@ static void commit_planes_for_stream(struct dc *dc,
 		context_clock_trace(dc, context);
 	}
 
-	if (update_type != UPDATE_TYPE_FAST && dc->hwss.interdependent_update_lock && dc->hwss.wait_for_pending_cleared)
-		dc->hwss.disconnect_pipes(dc, context);
+	if (update_type != UPDATE_TYPE_FAST && dc->hwss.interdependent_update_lock &&
+		dc->hwss.disconnect_pipes && dc->hwss.wait_for_pending_cleared){
+		dc->hwss.interdependent_update_lock(dc, context, true);
+		mpcc_disconnected = dc->hwss.disconnect_pipes(dc, context);
+		dc->hwss.interdependent_update_lock(dc, context, false);
+		if (mpcc_disconnected)
+			dc->hwss.wait_for_pending_cleared(dc, context);
+	}
 
 	for (j = 0; j < dc->res_pool->pipe_count; j++) {
 		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index 79fe9571cf5d..d0f3bf953d02 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -2761,7 +2761,7 @@ static struct pipe_ctx *dcn10_find_top_pipe_for_stream(
 	return NULL;
 }
 
-void dcn10_disconnect_pipes(
+bool dcn10_disconnect_pipes(
 		struct dc *dc,
 		struct dc_state *context)
 {
@@ -2772,10 +2772,6 @@ void dcn10_disconnect_pipes(
 		bool mpcc_disconnected = false;
 		struct pipe_ctx *old_pipe;
 		struct pipe_ctx *new_pipe;
-
-		dc->hwss.wait_for_pending_cleared(dc, context);
-		dc->hwss.interdependent_update_lock(dc, context, true);
-
 		DC_LOGGER_INIT(dc->ctx->logger);
 
 		/* Set pipe update flags and lock pipes */
@@ -2878,11 +2874,7 @@ void dcn10_disconnect_pipes(
 				}
 			}
 		}
-
-		dc->hwss.interdependent_update_lock(dc, context, false);
-
-		if (mpcc_disconnected)
-			dc->hwss.wait_for_pending_cleared(dc, context);
+	return mpcc_disconnected;
 }
 
 void dcn10_wait_for_pending_cleared(struct dc *dc,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
index 9a0f7a8a85cd..e5691e499023 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
@@ -194,7 +194,7 @@ void dcn10_get_surface_visual_confirm_color(
 void dcn10_get_hdr_visual_confirm_color(
 		struct pipe_ctx *pipe_ctx,
 		struct tg_color *color);
-void dcn10_disconnect_pipes(
+bool dcn10_disconnect_pipes(
 		struct dc *dc,
 		struct dc_state *context);
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index f48ee24d42f9..64c1be818b0e 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -67,7 +67,7 @@ struct hw_sequencer_funcs {
 			int num_planes, struct dc_state *context);
 	void (*program_front_end_for_ctx)(struct dc *dc,
 			struct dc_state *context);
-	void (*disconnect_pipes)(struct dc *dc,
+	bool (*disconnect_pipes)(struct dc *dc,
 			struct dc_state *context);
 	void (*wait_for_pending_cleared)(struct dc *dc,
 			struct dc_state *context);
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 8/9] drm/amd/display: [FW Promotion] Release 0.0.35

[Rodrigo Siqueira]

From: Anthony Koo <Anthony.Koo@amd.com>

[Header Changes]
   - Definition for retaining ABM settings during disable
   - Addition of some new AUX interface definitions
   - Addition of some outbox definitions

Signed-off-by: Anthony Koo <Anthony.Koo@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 132 +++++++++++++++++-
 1 file changed, 126 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 0051b1000264..d103ec1eaa73 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -36,10 +36,10 @@
 
 /* Firmware versioning. */
 #ifdef DMUB_EXPOSE_VERSION
-#define DMUB_FW_VERSION_GIT_HASH 0xf547f0b9d
+#define DMUB_FW_VERSION_GIT_HASH 0x9cf8f05fe
 #define DMUB_FW_VERSION_MAJOR 0
 #define DMUB_FW_VERSION_MINOR 0
-#define DMUB_FW_VERSION_REVISION 34
+#define DMUB_FW_VERSION_REVISION 35
 #define DMUB_FW_VERSION_TEST 0
 #define DMUB_FW_VERSION_VBIOS 0
 #define DMUB_FW_VERSION_HOTFIX 0
@@ -57,6 +57,7 @@
 
 #define SET_ABM_PIPE_GRADUALLY_DISABLE           0
 #define SET_ABM_PIPE_IMMEDIATELY_DISABLE         255
+#define SET_ABM_PIPE_IMMEDIATE_KEEP_GAIN_DISABLE 254
 #define SET_ABM_PIPE_NORMAL                      1
 
 /* Maximum number of streams on any ASIC. */
@@ -69,10 +70,6 @@
 #define PHYSICAL_ADDRESS_LOC union large_integer
 #endif
 
-#if defined(__cplusplus)
-extern "C" {
-#endif
-
 #ifndef dmub_memcpy
 #define dmub_memcpy(dest, source, bytes) memcpy((dest), (source), (bytes))
 #endif
@@ -81,6 +78,10 @@ extern "C" {
 #define dmub_memset(dest, val, bytes) memset((dest), (val), (bytes))
 #endif
 
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
 #ifndef dmub_udelay
 #define dmub_udelay(microseconds) udelay(microseconds)
 #endif
@@ -299,11 +300,15 @@ enum dmub_cmd_type {
 	DMUB_CMD__PSR = 64,
 	DMUB_CMD__ABM = 66,
 	DMUB_CMD__HW_LOCK = 69,
+	DMUB_CMD__DP_AUX_ACCESS = 70,
+	DMUB_CMD__OUTBOX1_ENABLE = 71,
 	DMUB_CMD__VBIOS = 128,
 };
 
 enum dmub_out_cmd_type {
 	DMUB_OUT_CMD__NULL = 0,
+	DMUB_OUT_CMD__DP_AUX_REPLY = 1,
+	DMUB_OUT_CMD__DP_HPD_NOTIFY = 2,
 };
 
 #pragma pack(push, 1)
@@ -461,6 +466,78 @@ struct dmub_rb_cmd_dpphy_init {
 	uint8_t reserved[60];
 };
 
+enum dp_aux_request_action {
+	DP_AUX_REQ_ACTION_I2C_WRITE		= 0x00,
+	DP_AUX_REQ_ACTION_I2C_READ		= 0x10,
+	DP_AUX_REQ_ACTION_I2C_STATUS_REQ	= 0x20,
+	DP_AUX_REQ_ACTION_I2C_WRITE_MOT		= 0x40,
+	DP_AUX_REQ_ACTION_I2C_READ_MOT		= 0x50,
+	DP_AUX_REQ_ACTION_I2C_STATUS_REQ_MOT	= 0x60,
+	DP_AUX_REQ_ACTION_DPCD_WRITE		= 0x80,
+	DP_AUX_REQ_ACTION_DPCD_READ		= 0x90
+};
+
+/* DP AUX command */
+struct aux_transaction_parameters {
+	uint8_t is_i2c_over_aux;
+	uint8_t action;
+	uint8_t length;
+	uint8_t pad;
+	uint32_t address;
+	uint8_t data[16];
+};
+
+struct dmub_cmd_dp_aux_control_data {
+	uint32_t handle;
+	uint8_t port_index;
+	uint8_t sw_crc_enabled;
+	uint16_t timeout;
+	struct aux_transaction_parameters dpaux;
+};
+
+struct dmub_rb_cmd_dp_aux_access {
+	struct dmub_cmd_header header;
+	struct dmub_cmd_dp_aux_control_data aux_control;
+};
+
+struct dmub_rb_cmd_outbox1_enable {
+	struct dmub_cmd_header header;
+	uint32_t enable;
+};
+
+/* DP AUX Reply command - OutBox Cmd */
+struct aux_reply_data {
+	uint8_t command;
+	uint8_t length;
+	uint8_t pad[2];
+	uint8_t data[16];
+};
+
+struct aux_reply_control_data {
+	uint32_t handle;
+	uint8_t phy_port_index;
+	uint8_t result;
+	uint16_t pad;
+};
+
+struct dmub_rb_cmd_dp_aux_reply {
+	struct dmub_cmd_header header;
+	struct aux_reply_control_data control;
+	struct aux_reply_data reply_data;
+};
+
+struct dp_hpd_data {
+	uint8_t phy_port_index;
+	uint8_t hpd_type;
+	uint8_t hpd_status;
+	uint8_t pad;
+};
+
+struct dmub_rb_cmd_dp_hpd_notify {
+	struct dmub_cmd_header header;
+	struct dp_hpd_data hpd_data;
+};
+
 /*
  * Command IDs should be treated as stable ABI.
  * Do not reuse or modify IDs.
@@ -690,8 +767,15 @@ union dmub_rb_cmd {
 	struct dmub_rb_cmd_abm_set_ambient_level abm_set_ambient_level;
 	struct dmub_rb_cmd_abm_set_pwm_frac abm_set_pwm_frac;
 	struct dmub_rb_cmd_abm_init_config abm_init_config;
+	struct dmub_rb_cmd_dp_aux_access dp_aux_access;
+	struct dmub_rb_cmd_outbox1_enable outbox1_enable;
 };
 
+union dmub_rb_out_cmd {
+	struct dmub_rb_cmd_common cmd_common;
+	struct dmub_rb_cmd_dp_aux_reply dp_aux_reply;
+	struct dmub_rb_cmd_dp_hpd_notify dp_hpd_notify;
+};
 #pragma pack(pop)
 
 
@@ -764,6 +848,25 @@ static inline bool dmub_rb_push_front(struct dmub_rb *rb,
 	return true;
 }
 
+static inline bool dmub_rb_out_push_front(struct dmub_rb *rb,
+				      const union dmub_rb_out_cmd *cmd)
+{
+	uint8_t *dst = (uint8_t *)(rb->base_address) + rb->wrpt;
+	const uint8_t *src = (uint8_t *)cmd;
+
+	if (dmub_rb_full(rb))
+		return false;
+
+	dmub_memcpy(dst, src, DMUB_RB_CMD_SIZE);
+
+	rb->wrpt += DMUB_RB_CMD_SIZE;
+
+	if (rb->wrpt >= rb->capacity)
+		rb->wrpt %= rb->capacity;
+
+	return true;
+}
+
 static inline bool dmub_rb_front(struct dmub_rb *rb,
 				 union dmub_rb_cmd  *cmd)
 {
@@ -777,6 +880,23 @@ static inline bool dmub_rb_front(struct dmub_rb *rb,
 	return true;
 }
 
+static inline bool dmub_rb_out_front(struct dmub_rb *rb,
+				 union dmub_rb_out_cmd  *cmd)
+{
+	const uint64_t volatile *src = (const uint64_t volatile *)(rb->base_address) + rb->rptr / sizeof(uint64_t);
+	uint64_t *dst = (uint64_t *)cmd;
+	int i;
+
+	if (dmub_rb_empty(rb))
+		return false;
+
+	// copying data
+	for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++)
+		*dst++ = *src++;
+
+	return true;
+}
+
 static inline bool dmub_rb_pop_front(struct dmub_rb *rb)
 {
 	if (dmub_rb_empty(rb))
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 9/9] drm/amd/display: disable stream if pixel clock changed with link active

[Rodrigo Siqueira]

From: Chiawen Huang <chiawen.huang@amd.com>

[Why]
Vbios uses preferred timing to turn on edp but OS could use other
timing.  If change pixel clock when link active, there is unexpected
garbage on monitor.

[How]
Once pixel clock changed, the driver needs to disable stream.

Signed-off-by: Chiawen Huang <chiawen.huang@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 76946a7d47ca..2a725a5fba40 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -864,7 +864,6 @@ static void disable_vbios_mode_if_required(
 		if (stream->link->local_sink &&
 			stream->link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
 			link = stream->link;
-			break;
 		}
 
 		if (link != NULL) {
-- 
2.28.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH 7/9] drm/amd/display: Revert check for flip pending before locking pipes

[Michel Dänzer]

On 2020-09-25 4:54 p.m., Rodrigo Siqueira wrote:
> From: Aric Cyr <aric.cyr@amd.com>
> 
> This reverts commit e82d8eae9e81af243256f70bec593baed50b0bdb.

Every revert should explain in the commit log why the change is being 
reverted. (Your future selves might thank you for that someday :)


-- 
Earthling Michel Dänzer               |               https://redhat.com
Libre software enthusiast             |             Mesa and X developer
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

RE: [PATCH 7/9] drm/amd/display: Revert check for flip pending before locking pipes

[Cyr, Aric]

[AMD Official Use Only - Internal Distribution Only]

It's already un-reverted in latest.  They could be squashed IMO.

-----Original Message-----
From: Michel Dänzer <michel@daenzer.net> 
Sent: Friday, September 25, 2020 11:53
To: Siqueira, Rodrigo <Rodrigo.Siqueira@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Cyr, Aric <Aric.Cyr@amd.com>; Brol, Eryk <Eryk.Brol@amd.com>; Li, Sun peng (Leo) <Sunpeng.Li@amd.com>; Wentland, Harry <Harry.Wentland@amd.com>; Zhuo, Qingqing <Qingqing.Zhuo@amd.com>; Pillai, Aurabindo <Aurabindo.Pillai@amd.com>; Lakha, Bhawanpreet <Bhawanpreet.Lakha@amd.com>
Subject: Re: [PATCH 7/9] drm/amd/display: Revert check for flip pending before locking pipes

On 2020-09-25 4:54 p.m., Rodrigo Siqueira wrote:
> From: Aric Cyr <aric.cyr@amd.com>
> 
> This reverts commit e82d8eae9e81af243256f70bec593baed50b0bdb.

Every revert should explain in the commit log why the change is being 
reverted. (Your future selves might thank you for that someday :)


-- 
Earthling Michel Dänzer               |               https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fredhat.com%2F&amp;data=02%7C01%7CAric.Cyr%40amd.com%7Cff8318f7505f48ac4e6008d8616b1050%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637366459764467283&amp;sdata=eeKdB8NNb3tanvw5GToG7YbETY%2FYKnEJQXRsfuE6u5k%3D&amp;reserved=0
Libre software enthusiast             |             Mesa and X developer
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx