[PATCH 00/32] DC Patches July 29 2022

[PATCH 00/32] DC Patches July 29 2022

[Tom Chung]

This DC patchset brings along following fixes:

- Move fpu file to dml folder
- Fix CAB allocation for multiple displays
- Add a variable to update FCLK
- Fix TMDS 4K@60Hz YCbCr420 corruption issue
- Fix MPC tree infinite loop in some case
- Fix SubVP calculations
- Fix HDMI VSIF V3 blank screen issue
- Add some documentation to the code
- Allow alternate prefetch modes in DML for DCN32
- Fix VPG instancing for dcn314 HPO
- Check correct bounds for stream encoder instances for DCN303
- Enable SubVP by default on DCN32 & DCN321
- Fix DTBCLK not correct for dcn314
- Fix Compile-time warning
- Fix the stereo mode hang issue
- Fix display light up issue

Alvin Lee (8):
  drm/amd/display: Revert "fix 32 bit compilation errors in
    dc_dmub_srv.c"
  drm/amd/display: Add a variable to update FCLK latency
  drm/amd/display: Refactor SubVP calculation to remove FPU
  drm/amd/display: Add 16 lines margin for SubVP
  drm/amd/display: Fix subvp calculations
  drm/amd/display: Revert "attempt to fix the logic in
    commit_planes_for_stream()"
  drm/amd/display: For stereo keep "FLIP_ANY_FRAME"
  drm/amd/display: Don't try to enter MALL SS if stereo3d

Aric Cyr (1):
  drm/amd/display: 3.2.197

Aurabindo Pillai (3):
  drm/amd/display: fix CAB allocation for multiple displays
  drm/amd/display: Check correct bounds for stream encoder instances for
    DCN303
  drm/amd/display: Enable SubVP by default on DCN32 & DCN321

Chiawen Huang (1):
  drm/amd/display: Device flash garbage before get in OS

Daniel Miess (2):
  drm/amd/display: Use pixels per container logic for DCN314 DCCG
    dividers
  drm/amd/display: Fix TMDS 4K@60Hz YCbCr420 corruption issue

David Galiffi (2):
  drm/amd/display: Allow alternate prefetch modes in DML for DCN32
  drm/amd/display: Fix Compile-time Warning

Duncan Ma (2):
  drm/amd/display: Fix VPG instancing for dcn314 HPO
  drm/amd/display: Correct DTBCLK for dcn314

Fudong Wang (1):
  drm/amd/display: clear optc underflow before turn off odm clock

Josip Pavic (1):
  drm/amd/display: Avoid MPC infinite loop

Leo Ma (1):
  drm/amd/display: Fix HDMI VSIF V3 incorrect issue

Meenakshikumar Somasundaram (1):
  drm/amd/display: Fix TDR eDP and USB4 display light up issue

Rodrigo Siqueira (9):
  drm/amd/display: Create FPU files for DCN314
  drm/amd/display: Move populate dml pipes from DCN314 to dml
  drm/amd/display: Drop FPU flags from Makefile
  drm/amd/display: Add documentation to some of the cursor struct
  drm/amd/display: Add basic kernel doc to CRC code under DC
  drm/amd/display: Add some extra kernel doc to amdgpu_dm
  drm/amd/display: Document pipe split policy
  drm/amd/display: Expand documentation for timing
  drm/amd/display: Document some of the DML structs

 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |  16 +-
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  46 ++-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.c        |  12 +-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.h        |   2 +-
 drivers/gpu/drm/amd/display/dc/core/dc.c      |  45 ++-
 drivers/gpu/drm/amd/display/dc/dc.h           |  40 +-
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c  |  93 +++--
 drivers/gpu/drm/amd/display/dc/dc_hw_types.h  | 148 ++++++-
 drivers/gpu/drm/amd/display/dc/dc_link.h      |   1 +
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.c |   1 +
 .../gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c  |   6 +
 .../gpu/drm/amd/display/dc/dcn10/dcn10_optc.c |  44 +-
 .../gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c  |   6 +
 .../gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c |   2 +-
 .../amd/display/dc/dcn303/dcn303_resource.c   |   2 +-
 .../gpu/drm/amd/display/dc/dcn314/Makefile    |  25 --
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  |  42 +-
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.h  |   4 +
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |   2 +
 .../amd/display/dc/dcn314/dcn314_resource.c   | 341 +---------------
 .../amd/display/dc/dcn314/dcn314_resource.h   |   3 +
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.c    |  39 +-
 .../gpu/drm/amd/display/dc/dcn32/dcn32_optc.c |   2 +-
 .../drm/amd/display/dc/dcn32/dcn32_resource.c |   3 +-
 .../amd/display/dc/dcn321/dcn321_resource.c   |   3 +-
 drivers/gpu/drm/amd/display/dc/dml/Makefile   |   2 +
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 376 ++++++++++++++++++
 .../amd/display/dc/dml/dcn314/dcn314_fpu.h    |  40 ++
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  |  46 ++-
 .../amd/display/dc/dml/dcn321/dcn321_fpu.c    |   7 +
 .../amd/display/dc/dml/display_mode_structs.h |  34 ++
 .../gpu/drm/amd/display/dc/inc/core_types.h   |   4 +
 .../amd/display/dc/inc/hw/timing_generator.h  |  14 +-
 .../amd/display/modules/freesync/freesync.c   |  15 +-
 34 files changed, 968 insertions(+), 498 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
 create mode 100644 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h

-- 
2.25.1

[PATCH 01/32] drm/amd/display: Use pixels per container logic for DCN314 DCCG dividers

[Tom Chung]

From: Daniel Miess <Daniel.Miess@amd.com>

[Why]
DP DSC YCbCr422 not working because dcn314_calculate_dccg_k1_k2_values
does not account for two pixels per container

[How]
Replace the contents of dcn314_calculate_dccg_k1_k2_values with the
code from the function dcn32_calculate_dccg_k1_k2_values

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Daniel Miess <Daniel.Miess@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c   | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index 755c715ad8dc..452f4a9bf251 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -343,7 +343,10 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	unsigned int odm_combine_factor = 0;
+	struct dc *dc = pipe_ctx->stream->ctx->dc;
+	bool two_pix_per_container = false;
 
+	two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing);
 	odm_combine_factor = get_odm_config(pipe_ctx, NULL);
 
 	if (is_dp_128b_132b_signal(pipe_ctx)) {
@@ -355,16 +358,13 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 		else
 			*k2_div = PIXEL_RATE_DIV_BY_4;
 	} else if (dc_is_dp_signal(pipe_ctx->stream->signal)) {
-		if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) {
+		if (two_pix_per_container) {
 			*k1_div = PIXEL_RATE_DIV_BY_1;
 			*k2_div = PIXEL_RATE_DIV_BY_2;
-		} else if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422) {
-			*k1_div = PIXEL_RATE_DIV_BY_2;
-			*k2_div = PIXEL_RATE_DIV_BY_2;
 		} else {
-			if (odm_combine_factor == 1)
-				*k2_div = PIXEL_RATE_DIV_BY_4;
-			else if (odm_combine_factor == 2)
+			*k1_div = PIXEL_RATE_DIV_BY_1;
+			*k2_div = PIXEL_RATE_DIV_BY_4;
+			if ((odm_combine_factor == 2) || dc->debug.enable_dp_dig_pixel_rate_div_policy)
 				*k2_div = PIXEL_RATE_DIV_BY_2;
 		}
 	}
-- 
2.25.1

[PATCH 02/32] drm/amd/display: Create FPU files for DCN314

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

DCN314 has multiple references to FPU operations inside the resource
files, and we need to move those codes to the DML folder. This commit
creates the dcn314_fpu files and moves the bounding box operation to
this file.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../amd/display/dc/dcn314/dcn314_resource.c   | 235 +---------------
 .../amd/display/dc/dcn314/dcn314_resource.h   |   3 +
 drivers/gpu/drm/amd/display/dc/dml/Makefile   |   2 +
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 266 ++++++++++++++++++
 .../amd/display/dc/dml/dcn314/dcn314_fpu.h    |  37 +++
 5 files changed, 312 insertions(+), 231 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
 create mode 100644 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 63861cdfb09f..376488198363 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -70,6 +70,7 @@
 #include "dce110/dce110_resource.h"
 #include "dml/display_mode_vba.h"
 #include "dml/dcn31/dcn31_fpu.h"
+#include "dml/dcn314/dcn314_fpu.h"
 #include "dcn314/dcn314_dccg.h"
 #include "dcn10/dcn10_resource.h"
 #include "dcn31/dcn31_panel_cntl.h"
@@ -132,155 +133,6 @@ static const struct IP_BASE DCN_BASE = { { { { 0x00000012, 0x000000C0, 0x000034C
 
 #define DC_LOGGER_INIT(logger)
 
-#define DCN3_14_DEFAULT_DET_SIZE 384
-#define DCN3_14_MAX_DET_SIZE 384
-#define DCN3_14_MIN_COMPBUF_SIZE_KB 128
-#define DCN3_14_CRB_SEGMENT_SIZE_KB 64
-struct _vcs_dpi_ip_params_st dcn3_14_ip = {
-	.VBlankNomDefaultUS = 668,
-	.gpuvm_enable = 1,
-	.gpuvm_max_page_table_levels = 1,
-	.hostvm_enable = 1,
-	.hostvm_max_page_table_levels = 2,
-	.rob_buffer_size_kbytes = 64,
-	.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE,
-	.config_return_buffer_size_in_kbytes = 1792,
-	.compressed_buffer_segment_size_in_kbytes = 64,
-	.meta_fifo_size_in_kentries = 32,
-	.zero_size_buffer_entries = 512,
-	.compbuf_reserved_space_64b = 256,
-	.compbuf_reserved_space_zs = 64,
-	.dpp_output_buffer_pixels = 2560,
-	.opp_output_buffer_lines = 1,
-	.pixel_chunk_size_kbytes = 8,
-	.meta_chunk_size_kbytes = 2,
-	.min_meta_chunk_size_bytes = 256,
-	.writeback_chunk_size_kbytes = 8,
-	.ptoi_supported = false,
-	.num_dsc = 4,
-	.maximum_dsc_bits_per_component = 10,
-	.dsc422_native_support = false,
-	.is_line_buffer_bpp_fixed = true,
-	.line_buffer_fixed_bpp = 48,
-	.line_buffer_size_bits = 789504,
-	.max_line_buffer_lines = 12,
-	.writeback_interface_buffer_size_kbytes = 90,
-	.max_num_dpp = 4,
-	.max_num_otg = 4,
-	.max_num_hdmi_frl_outputs = 1,
-	.max_num_wb = 1,
-	.max_dchub_pscl_bw_pix_per_clk = 4,
-	.max_pscl_lb_bw_pix_per_clk = 2,
-	.max_lb_vscl_bw_pix_per_clk = 4,
-	.max_vscl_hscl_bw_pix_per_clk = 4,
-	.max_hscl_ratio = 6,
-	.max_vscl_ratio = 6,
-	.max_hscl_taps = 8,
-	.max_vscl_taps = 8,
-	.dpte_buffer_size_in_pte_reqs_luma = 64,
-	.dpte_buffer_size_in_pte_reqs_chroma = 34,
-	.dispclk_ramp_margin_percent = 1,
-	.max_inter_dcn_tile_repeaters = 8,
-	.cursor_buffer_size = 16,
-	.cursor_chunk_size = 2,
-	.writeback_line_buffer_buffer_size = 0,
-	.writeback_min_hscl_ratio = 1,
-	.writeback_min_vscl_ratio = 1,
-	.writeback_max_hscl_ratio = 1,
-	.writeback_max_vscl_ratio = 1,
-	.writeback_max_hscl_taps = 1,
-	.writeback_max_vscl_taps = 1,
-	.dppclk_delay_subtotal = 46,
-	.dppclk_delay_scl = 50,
-	.dppclk_delay_scl_lb_only = 16,
-	.dppclk_delay_cnvc_formatter = 27,
-	.dppclk_delay_cnvc_cursor = 6,
-	.dispclk_delay_subtotal = 119,
-	.dynamic_metadata_vm_enabled = false,
-	.odm_combine_4to1_supported = false,
-	.dcc_supported = true,
-};
-
-struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
-		/*TODO: correct dispclk/dppclk voltage level determination*/
-	.clock_limits = {
-		{
-			.state = 0,
-			.dispclk_mhz = 1200.0,
-			.dppclk_mhz = 1200.0,
-			.phyclk_mhz = 600.0,
-			.phyclk_d18_mhz = 667.0,
-			.dscclk_mhz = 186.0,
-			.dtbclk_mhz = 625.0,
-		},
-		{
-			.state = 1,
-			.dispclk_mhz = 1200.0,
-			.dppclk_mhz = 1200.0,
-			.phyclk_mhz = 810.0,
-			.phyclk_d18_mhz = 667.0,
-			.dscclk_mhz = 209.0,
-			.dtbclk_mhz = 625.0,
-		},
-		{
-			.state = 2,
-			.dispclk_mhz = 1200.0,
-			.dppclk_mhz = 1200.0,
-			.phyclk_mhz = 810.0,
-			.phyclk_d18_mhz = 667.0,
-			.dscclk_mhz = 209.0,
-			.dtbclk_mhz = 625.0,
-		},
-		{
-			.state = 3,
-			.dispclk_mhz = 1200.0,
-			.dppclk_mhz = 1200.0,
-			.phyclk_mhz = 810.0,
-			.phyclk_d18_mhz = 667.0,
-			.dscclk_mhz = 371.0,
-			.dtbclk_mhz = 625.0,
-		},
-		{
-			.state = 4,
-			.dispclk_mhz = 1200.0,
-			.dppclk_mhz = 1200.0,
-			.phyclk_mhz = 810.0,
-			.phyclk_d18_mhz = 667.0,
-			.dscclk_mhz = 417.0,
-			.dtbclk_mhz = 625.0,
-		},
-	},
-	.num_states = 5,
-	.sr_exit_time_us = 9.0,
-	.sr_enter_plus_exit_time_us = 11.0,
-	.sr_exit_z8_time_us = 442.0,
-	.sr_enter_plus_exit_z8_time_us = 560.0,
-	.writeback_latency_us = 12.0,
-	.dram_channel_width_bytes = 4,
-	.round_trip_ping_latency_dcfclk_cycles = 106,
-	.urgent_latency_pixel_data_only_us = 4.0,
-	.urgent_latency_pixel_mixed_with_vm_data_us = 4.0,
-	.urgent_latency_vm_data_only_us = 4.0,
-	.urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096,
-	.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096,
-	.urgent_out_of_order_return_per_channel_vm_only_bytes = 4096,
-	.pct_ideal_sdp_bw_after_urgent = 80.0,
-	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
-	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
-	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
-	.max_avg_sdp_bw_use_normal_percent = 60.0,
-	.max_avg_dram_bw_use_normal_percent = 60.0,
-	.fabric_datapath_to_dcn_data_return_bytes = 32,
-	.return_bus_width_bytes = 64,
-	.downspread_percent = 0.38,
-	.dcn_downspread_percent = 0.5,
-	.gpuvm_min_page_size_bytes = 4096,
-	.hostvm_min_page_size_bytes = 4096,
-	.do_urgent_latency_adjustment = false,
-	.urgent_latency_adjustment_fabric_clock_component_us = 0,
-	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
-};
-
 enum dcn31_clk_src_array_id {
 	DCN31_CLK_SRC_PLL0,
 	DCN31_CLK_SRC_PLL1,
@@ -1906,88 +1758,9 @@ static struct dc_cap_funcs cap_funcs = {
 
 static void dcn314_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
 {
-	struct clk_limit_table *clk_table = &bw_params->clk_table;
-	struct _vcs_dpi_voltage_scaling_st *clock_tmp = dcn3_14_soc._clock_tmp;
-	unsigned int i, closest_clk_lvl;
-	int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
-	int j;
-
-	// Default clock levels are used for diags, which may lead to overclocking.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-
-		dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
-		dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count;
-
-		if (bw_params->num_channels > 0)
-			dcn3_14_soc.num_chans = bw_params->num_channels;
-
-		ASSERT(dcn3_14_soc.num_chans);
-		ASSERT(clk_table->num_entries);
-
-		/* Prepass to find max clocks independent of voltage level. */
-		for (i = 0; i < clk_table->num_entries; ++i) {
-			if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
-				max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
-			if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
-				max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
-		}
-
-		for (i = 0; i < clk_table->num_entries; i++) {
-			/* loop backwards*/
-			for (closest_clk_lvl = 0, j = dcn3_14_soc.num_states - 1; j >= 0; j--) {
-				if ((unsigned int) dcn3_14_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
-					closest_clk_lvl = j;
-					break;
-				}
-			}
-			if (clk_table->num_entries == 1) {
-				/*smu gives one DPM level, let's take the highest one*/
-				closest_clk_lvl = dcn3_14_soc.num_states - 1;
-			}
-
-			clock_tmp[i].state = i;
-
-			/* Clocks dependent on voltage level. */
-			clock_tmp[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
-			if (clk_table->num_entries == 1 &&
-				clock_tmp[i].dcfclk_mhz < dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
-				/*SMU fix not released yet*/
-				clock_tmp[i].dcfclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
-			}
-			clock_tmp[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
-			clock_tmp[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
-
-			if (clk_table->entries[i].memclk_mhz && clk_table->entries[i].wck_ratio)
-				clock_tmp[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2 * clk_table->entries[i].wck_ratio;
-
-			/* Clocks independent of voltage level. */
-			clock_tmp[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
-				dcn3_14_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
-
-			clock_tmp[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
-				dcn3_14_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
-
-			clock_tmp[i].dram_bw_per_chan_gbps = dcn3_14_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
-			clock_tmp[i].dscclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
-			clock_tmp[i].dtbclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
-			clock_tmp[i].phyclk_d18_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
-			clock_tmp[i].phyclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
-		}
-		for (i = 0; i < clk_table->num_entries; i++)
-			dcn3_14_soc.clock_limits[i] = clock_tmp[i];
-		if (clk_table->num_entries)
-			dcn3_14_soc.num_states = clk_table->num_entries;
-	}
-
-	if (max_dispclk_mhz) {
-		dcn3_14_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
-		dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
-	}
-
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31);
-	else
-		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA);
+	DC_FP_START();
+	dcn314_update_bw_bounding_box_fpu(dc, bw_params);
+	DC_FP_END();
 }
 
 static struct resource_funcs dcn314_res_pool_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h
index c41108847ce0..0dd3153aa5c1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h
@@ -29,6 +29,9 @@
 
 #include "core_types.h"
 
+extern struct _vcs_dpi_ip_params_st dcn3_14_ip;
+extern struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc;
+
 #define TO_DCN314_RES_POOL(pool)\
 	container_of(pool, struct dcn314_resource_pool, base)
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile
index 359f6e9a1da0..3b813f011569 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
@@ -71,6 +71,7 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_ccflags) $(fram
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_mode_vba_31.o := $(dml_ccflags) $(frame_warn_flag)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_rq_dlg_calc_31.o := $(dml_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn314/dcn314_fpu.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/dcn32_fpu.o := $(dml_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_32.o := $(dml_ccflags) $(frame_warn_flag)
@@ -131,6 +132,7 @@ DML += dcn321/dcn321_fpu.o
 DML += dcn301/dcn301_fpu.o
 DML += dcn302/dcn302_fpu.o
 DML += dcn303/dcn303_fpu.o
+DML += dcn314/dcn314_fpu.o
 DML += dsc/rc_calc_fpu.o
 DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o
 endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
new file mode 100644
index 000000000000..340c3ca22345
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -0,0 +1,266 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#include "clk_mgr.h"
+#include "resource.h"
+#include "dcn314_fpu.h"
+#include "dml/display_mode_vba.h"
+
+struct _vcs_dpi_ip_params_st dcn3_14_ip = {
+	.VBlankNomDefaultUS = 668,
+	.gpuvm_enable = 1,
+	.gpuvm_max_page_table_levels = 1,
+	.hostvm_enable = 1,
+	.hostvm_max_page_table_levels = 2,
+	.rob_buffer_size_kbytes = 64,
+	.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE,
+	.config_return_buffer_size_in_kbytes = 1792,
+	.compressed_buffer_segment_size_in_kbytes = 64,
+	.meta_fifo_size_in_kentries = 32,
+	.zero_size_buffer_entries = 512,
+	.compbuf_reserved_space_64b = 256,
+	.compbuf_reserved_space_zs = 64,
+	.dpp_output_buffer_pixels = 2560,
+	.opp_output_buffer_lines = 1,
+	.pixel_chunk_size_kbytes = 8,
+	.meta_chunk_size_kbytes = 2,
+	.min_meta_chunk_size_bytes = 256,
+	.writeback_chunk_size_kbytes = 8,
+	.ptoi_supported = false,
+	.num_dsc = 4,
+	.maximum_dsc_bits_per_component = 10,
+	.dsc422_native_support = false,
+	.is_line_buffer_bpp_fixed = true,
+	.line_buffer_fixed_bpp = 48,
+	.line_buffer_size_bits = 789504,
+	.max_line_buffer_lines = 12,
+	.writeback_interface_buffer_size_kbytes = 90,
+	.max_num_dpp = 4,
+	.max_num_otg = 4,
+	.max_num_hdmi_frl_outputs = 1,
+	.max_num_wb = 1,
+	.max_dchub_pscl_bw_pix_per_clk = 4,
+	.max_pscl_lb_bw_pix_per_clk = 2,
+	.max_lb_vscl_bw_pix_per_clk = 4,
+	.max_vscl_hscl_bw_pix_per_clk = 4,
+	.max_hscl_ratio = 6,
+	.max_vscl_ratio = 6,
+	.max_hscl_taps = 8,
+	.max_vscl_taps = 8,
+	.dpte_buffer_size_in_pte_reqs_luma = 64,
+	.dpte_buffer_size_in_pte_reqs_chroma = 34,
+	.dispclk_ramp_margin_percent = 1,
+	.max_inter_dcn_tile_repeaters = 8,
+	.cursor_buffer_size = 16,
+	.cursor_chunk_size = 2,
+	.writeback_line_buffer_buffer_size = 0,
+	.writeback_min_hscl_ratio = 1,
+	.writeback_min_vscl_ratio = 1,
+	.writeback_max_hscl_ratio = 1,
+	.writeback_max_vscl_ratio = 1,
+	.writeback_max_hscl_taps = 1,
+	.writeback_max_vscl_taps = 1,
+	.dppclk_delay_subtotal = 46,
+	.dppclk_delay_scl = 50,
+	.dppclk_delay_scl_lb_only = 16,
+	.dppclk_delay_cnvc_formatter = 27,
+	.dppclk_delay_cnvc_cursor = 6,
+	.dispclk_delay_subtotal = 119,
+	.dynamic_metadata_vm_enabled = false,
+	.odm_combine_4to1_supported = false,
+	.dcc_supported = true,
+};
+
+struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
+		/*TODO: correct dispclk/dppclk voltage level determination*/
+	.clock_limits = {
+		{
+			.state = 0,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 600.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 186.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 1,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 2,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 209.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 3,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 371.0,
+			.dtbclk_mhz = 625.0,
+		},
+		{
+			.state = 4,
+			.dispclk_mhz = 1200.0,
+			.dppclk_mhz = 1200.0,
+			.phyclk_mhz = 810.0,
+			.phyclk_d18_mhz = 667.0,
+			.dscclk_mhz = 417.0,
+			.dtbclk_mhz = 625.0,
+		},
+	},
+	.num_states = 5,
+	.sr_exit_time_us = 9.0,
+	.sr_enter_plus_exit_time_us = 11.0,
+	.sr_exit_z8_time_us = 442.0,
+	.sr_enter_plus_exit_z8_time_us = 560.0,
+	.writeback_latency_us = 12.0,
+	.dram_channel_width_bytes = 4,
+	.round_trip_ping_latency_dcfclk_cycles = 106,
+	.urgent_latency_pixel_data_only_us = 4.0,
+	.urgent_latency_pixel_mixed_with_vm_data_us = 4.0,
+	.urgent_latency_vm_data_only_us = 4.0,
+	.urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096,
+	.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096,
+	.urgent_out_of_order_return_per_channel_vm_only_bytes = 4096,
+	.pct_ideal_sdp_bw_after_urgent = 80.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0,
+	.pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0,
+	.max_avg_sdp_bw_use_normal_percent = 60.0,
+	.max_avg_dram_bw_use_normal_percent = 60.0,
+	.fabric_datapath_to_dcn_data_return_bytes = 32,
+	.return_bus_width_bytes = 64,
+	.downspread_percent = 0.38,
+	.dcn_downspread_percent = 0.5,
+	.gpuvm_min_page_size_bytes = 4096,
+	.hostvm_min_page_size_bytes = 4096,
+	.do_urgent_latency_adjustment = false,
+	.urgent_latency_adjustment_fabric_clock_component_us = 0,
+	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
+};
+
+
+void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
+{
+	struct clk_limit_table *clk_table = &bw_params->clk_table;
+	struct _vcs_dpi_voltage_scaling_st *clock_limits =
+		dcn3_14_soc.clock_limits;
+	unsigned int i, closest_clk_lvl;
+	int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+	int j;
+
+	dc_assert_fp_enabled();
+
+	// Default clock levels are used for diags, which may lead to overclocking.
+	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
+
+		dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+		dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count;
+
+		if (bw_params->num_channels > 0)
+			dcn3_14_soc.num_chans = bw_params->num_channels;
+
+		ASSERT(dcn3_14_soc.num_chans);
+		ASSERT(clk_table->num_entries);
+
+		/* Prepass to find max clocks independent of voltage level. */
+		for (i = 0; i < clk_table->num_entries; ++i) {
+			if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+				max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+			if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+				max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+		}
+
+		for (i = 0; i < clk_table->num_entries; i++) {
+			/* loop backwards*/
+			for (closest_clk_lvl = 0, j = dcn3_14_soc.num_states - 1; j >= 0; j--) {
+				if ((unsigned int) dcn3_14_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
+					closest_clk_lvl = j;
+					break;
+				}
+			}
+			if (clk_table->num_entries == 1) {
+				/*smu gives one DPM level, let's take the highest one*/
+				closest_clk_lvl = dcn3_14_soc.num_states - 1;
+			}
+
+			clock_limits[i].state = i;
+
+			/* Clocks dependent on voltage level. */
+			clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+			if (clk_table->num_entries == 1 &&
+				clock_limits[i].dcfclk_mhz < dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
+				/*SMU fix not released yet*/
+				clock_limits[i].dcfclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
+			}
+			clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+			clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+
+			if (clk_table->entries[i].memclk_mhz && clk_table->entries[i].wck_ratio)
+				clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2 * clk_table->entries[i].wck_ratio;
+
+			/* Clocks independent of voltage level. */
+			clock_limits[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+				dcn3_14_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+			clock_limits[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+				dcn3_14_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+			clock_limits[i].dram_bw_per_chan_gbps = dcn3_14_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+			clock_limits[i].dscclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+			clock_limits[i].dtbclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+			clock_limits[i].phyclk_d18_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+			clock_limits[i].phyclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+		}
+		for (i = 0; i < clk_table->num_entries; i++)
+			dcn3_14_soc.clock_limits[i] = clock_limits[i];
+		if (clk_table->num_entries) {
+			dcn3_14_soc.num_states = clk_table->num_entries;
+		}
+	}
+
+	if (max_dispclk_mhz) {
+		dcn3_14_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+		dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2;
+	}
+
+	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
+		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31);
+	else
+		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA);
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h
new file mode 100644
index 000000000000..79466d4f4185
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#ifndef __DCN314_FPU_H__
+#define __DCN314_FPU_H__
+
+#define DCN3_14_DEFAULT_DET_SIZE 384
+#define DCN3_14_MAX_DET_SIZE 384
+#define DCN3_14_MIN_COMPBUF_SIZE_KB 128
+#define DCN3_14_CRB_SEGMENT_SIZE_KB 64
+
+void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+
+#endif
-- 
2.25.1

[PATCH 03/32] drm/amd/display: Move populate dml pipes from DCN314 to dml

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

The function responsible for populating DML pipes has some FPU
operations, and for this reason, it must be moved to the dml folder.
This commit moves such function from resource to the fpu file under the
dml folder.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../amd/display/dc/dcn314/dcn314_resource.c   | 101 +---------------
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 110 ++++++++++++++++++
 .../amd/display/dc/dml/dcn314/dcn314_fpu.h    |   3 +
 3 files changed, 117 insertions(+), 97 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 376488198363..1f095f05d986 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -1645,109 +1645,16 @@ static struct clock_source *dcn31_clock_source_create(
 	return NULL;
 }
 
-static bool is_dual_plane(enum surface_pixel_format format)
-{
-	return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA;
-}
-
 static int dcn314_populate_dml_pipes_from_context(
 	struct dc *dc, struct dc_state *context,
 	display_e2e_pipe_params_st *pipes,
 	bool fast_validate)
 {
-	int i, pipe_cnt;
-	struct resource_context *res_ctx = &context->res_ctx;
-	struct pipe_ctx *pipe;
-	bool upscaled = false;
-
-	dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate);
-
-	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
-		struct dc_crtc_timing *timing;
-
-		if (!res_ctx->pipe_ctx[i].stream)
-			continue;
-		pipe = &res_ctx->pipe_ctx[i];
-		timing = &pipe->stream->timing;
-
-		if (dc_extended_blank_supported(dc) && pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min
-			&& pipe->stream->adjust.v_total_min > timing->v_total)
-			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
-
-		if (pipe->plane_state &&
-				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
-				pipe->plane_state->src_rect.width < pipe->plane_state->dst_rect.width))
-			upscaled = true;
-
-		/*
-		 * Immediate flip can be set dynamically after enabling the plane.
-		 * We need to require support for immediate flip or underflow can be
-		 * intermittently experienced depending on peak b/w requirements.
-		 */
-		pipes[pipe_cnt].pipe.src.immediate_flip = true;
-
-		pipes[pipe_cnt].pipe.src.unbounded_req_mode = false;
-		pipes[pipe_cnt].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active;
-		pipes[pipe_cnt].pipe.src.gpuvm = true;
-		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0;
-		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0;
-		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
-		pipes[pipe_cnt].pipe.src.dcc_rate = 3;
-		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
-
-		if (pipes[pipe_cnt].dout.dsc_enable) {
-			switch (timing->display_color_depth) {
-			case COLOR_DEPTH_888:
-				pipes[pipe_cnt].dout.dsc_input_bpc = 8;
-				break;
-			case COLOR_DEPTH_101010:
-				pipes[pipe_cnt].dout.dsc_input_bpc = 10;
-				break;
-			case COLOR_DEPTH_121212:
-				pipes[pipe_cnt].dout.dsc_input_bpc = 12;
-				break;
-			default:
-				ASSERT(0);
-				break;
-			}
-		}
-
-		pipe_cnt++;
-	}
-	context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE;
-
-	dc->config.enable_4to1MPC = false;
-	if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) {
-		if (is_dual_plane(pipe->plane_state->format)
-				&& pipe->plane_state->src_rect.width <= 1920 && pipe->plane_state->src_rect.height <= 1080) {
-			dc->config.enable_4to1MPC = true;
-		} else if (!is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 5120) {
-			/* Limit to 5k max to avoid forced pipe split when there is not enough detile for swath */
-			context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
-			pipes[0].pipe.src.unbounded_req_mode = true;
-		}
-	} else if (context->stream_count >= dc->debug.crb_alloc_policy_min_disp_count
-			&& dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) {
-		context->bw_ctx.dml.ip.det_buffer_size_kbytes = dc->debug.crb_alloc_policy * 64;
-	} else if (context->stream_count >= 3 && upscaled) {
-		context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
-	}
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
-
-		if (!pipe->stream)
-			continue;
+	int pipe_cnt;
 
-		if (pipe->stream->signal == SIGNAL_TYPE_EDP && dc->debug.seamless_boot_odm_combine &&
-				pipe->stream->apply_seamless_boot_optimization) {
-
-			if (pipe->stream->apply_boot_odm_mode == dm_odm_combine_policy_2to1) {
-				context->bw_ctx.dml.vba.ODMCombinePolicy = dm_odm_combine_policy_2to1;
-				break;
-			}
-		}
-	}
+	DC_FP_START();
+	pipe_cnt = dcn314_populate_dml_pipes_from_context_fpu(dc, context, pipes, fast_validate);
+	DC_FP_END();
 
 	return pipe_cnt;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 340c3ca22345..d52a16d280e9 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -26,7 +26,9 @@
 
 #include "clk_mgr.h"
 #include "resource.h"
+#include "dcn31/dcn31_hubbub.h"
 #include "dcn314_fpu.h"
+#include "dml/dcn20/dcn20_fpu.h"
 #include "dml/display_mode_vba.h"
 
 struct _vcs_dpi_ip_params_st dcn3_14_ip = {
@@ -264,3 +266,111 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 	else
 		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA);
 }
+
+static bool is_dual_plane(enum surface_pixel_format format)
+{
+	return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA;
+}
+
+int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context,
+					       display_e2e_pipe_params_st *pipes,
+					       bool fast_validate)
+{
+	int i, pipe_cnt;
+	struct resource_context *res_ctx = &context->res_ctx;
+	struct pipe_ctx *pipe;
+	bool upscaled = false;
+
+	dc_assert_fp_enabled();
+
+	dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate);
+
+	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
+		struct dc_crtc_timing *timing;
+
+		if (!res_ctx->pipe_ctx[i].stream)
+			continue;
+		pipe = &res_ctx->pipe_ctx[i];
+		timing = &pipe->stream->timing;
+
+		if (dc_extended_blank_supported(dc) && pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min
+			&& pipe->stream->adjust.v_total_min > timing->v_total)
+			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
+
+		if (pipe->plane_state &&
+				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
+				pipe->plane_state->src_rect.width < pipe->plane_state->dst_rect.width))
+			upscaled = true;
+
+		/*
+		 * Immediate flip can be set dynamically after enabling the plane.
+		 * We need to require support for immediate flip or underflow can be
+		 * intermittently experienced depending on peak b/w requirements.
+		 */
+		pipes[pipe_cnt].pipe.src.immediate_flip = true;
+
+		pipes[pipe_cnt].pipe.src.unbounded_req_mode = false;
+		pipes[pipe_cnt].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active;
+		pipes[pipe_cnt].pipe.src.gpuvm = true;
+		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0;
+		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0;
+		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
+		pipes[pipe_cnt].pipe.src.dcc_rate = 3;
+		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
+
+		if (pipes[pipe_cnt].dout.dsc_enable) {
+			switch (timing->display_color_depth) {
+			case COLOR_DEPTH_888:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 8;
+				break;
+			case COLOR_DEPTH_101010:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 10;
+				break;
+			case COLOR_DEPTH_121212:
+				pipes[pipe_cnt].dout.dsc_input_bpc = 12;
+				break;
+			default:
+				ASSERT(0);
+				break;
+			}
+		}
+
+		pipe_cnt++;
+	}
+	context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE;
+
+	dc->config.enable_4to1MPC = false;
+	if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) {
+		if (is_dual_plane(pipe->plane_state->format)
+				&& pipe->plane_state->src_rect.width <= 1920 && pipe->plane_state->src_rect.height <= 1080) {
+			dc->config.enable_4to1MPC = true;
+		} else if (!is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 5120) {
+			/* Limit to 5k max to avoid forced pipe split when there is not enough detile for swath */
+			context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+			pipes[0].pipe.src.unbounded_req_mode = true;
+		}
+	} else if (context->stream_count >= dc->debug.crb_alloc_policy_min_disp_count
+			&& dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes = dc->debug.crb_alloc_policy * 64;
+	} else if (context->stream_count >= 3 && upscaled) {
+		context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192;
+	}
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->stream)
+			continue;
+
+		if (pipe->stream->signal == SIGNAL_TYPE_EDP && dc->debug.seamless_boot_odm_combine &&
+				pipe->stream->apply_seamless_boot_optimization) {
+
+			if (pipe->stream->apply_boot_odm_mode == dm_odm_combine_policy_2to1) {
+				context->bw_ctx.dml.vba.ODMCombinePolicy = dm_odm_combine_policy_2to1;
+				break;
+			}
+		}
+	}
+
+	return pipe_cnt;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h
index 79466d4f4185..d32c5bb99f4c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h
@@ -33,5 +33,8 @@
 #define DCN3_14_CRB_SEGMENT_SIZE_KB 64
 
 void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params);
+int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context,
+					       display_e2e_pipe_params_st *pipes,
+					       bool fast_validate);
 
 #endif
-- 
2.25.1

[PATCH 04/32] drm/amd/display: Drop FPU flags from Makefile

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

At this point, we isolate the FPU code associated with DCN314 under the
DML folder. This commit drops the FPU flags from the Makefile.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn314/Makefile    | 25 -------------------
 1 file changed, 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/Makefile b/drivers/gpu/drm/amd/display/dc/dcn314/Makefile
index e3b5a95e03b1..702c28c2560e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/Makefile
@@ -13,31 +13,6 @@
 DCN314 = dcn314_resource.o dcn314_hwseq.o dcn314_init.o \
 		dcn314_dio_stream_encoder.o dcn314_dccg.o dcn314_optc.o
 
-ifdef CONFIG_X86
-CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o := -mhard-float -msse
-endif
-
-ifdef CONFIG_PPC64
-CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o := -mhard-float -maltivec
-endif
-
-ifdef CONFIG_CC_IS_GCC
-ifeq ($(call cc-ifversion, -lt, 0701, y), y)
-IS_OLD_GCC = 1
-endif
-endif
-
-ifdef CONFIG_X86
-ifdef IS_OLD_GCC
-# Stack alignment mismatch, proceed with caution.
-# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
-# (8B stack alignment).
-CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o += -mpreferred-stack-boundary=4
-else
-CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o += -msse2
-endif
-endif
-
 AMD_DAL_DCN314 = $(addprefix $(AMDDALPATH)/dc/dcn314/,$(DCN314))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_DCN314)
-- 
2.25.1

[PATCH 05/32] drm/amd/display: fix CAB allocation for multiple displays

[Tom Chung]

From: Aurabindo Pillai <aurabindo.pillai@amd.com>

[Why & How]
When multiple displays are used, the underlying framebuffers could be
two separate framebuffers, or a single large framebuffer. Fix the
calculation logic for CAB to account for large framebuffer. Current
logic assumes that any FB that the plane points to are independent. When
a single FB is used on the system, this does 2 times allocation.

Add a check to prevent duplicate allocation by checking if the base
addresses are the same, and then ensuring that the if we allocate using
the pitch, whole of the other fbs will be accounted for in the first
allocation.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index d38341f68b17..5ba503d80728 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -250,6 +250,7 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c
 	uint32_t total_lines = 0;
 	uint32_t lines_per_way = 0;
 	uint32_t num_ways = 0;
+	uint32_t prev_addr_low = 0;
 
 	for (i = 0; i < ctx->stream_count; i++) {
 		stream = ctx->streams[i];
@@ -267,10 +268,20 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c
 			plane = ctx->stream_status[i].plane_states[j];
 
 			// Calculate total surface size
-			surface_size = plane->plane_size.surface_pitch *
+			if (prev_addr_low != plane->address.grph.addr.u.low_part) {
+				/* if plane address are different from prev FB, then userspace allocated separate FBs*/
+				surface_size += plane->plane_size.surface_pitch *
 					plane->plane_size.surface_size.height *
 					(plane->format >= SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616 ? 8 : 4);
 
+				prev_addr_low = plane->address.grph.addr.u.low_part;
+			} else {
+				/* We have the same fb for all the planes.
+				 * Xorg always creates one giant fb that holds all surfaces,
+				 * so allocating it once is sufficient.
+				 * */
+				continue;
+			}
 			// Convert surface size + starting address to number of cache lines required
 			// (alignment accounted for)
 			cache_lines_used += dcn32_cache_lines_for_surface(dc, surface_size,
-- 
2.25.1

[PATCH 06/32] drm/amd/display: Revert "fix 32 bit compilation errors in dc_dmub_srv.c"

[Tom Chung]

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

[Description]
Revert commit that refactors SubVP calculation incorrectly.

Reviewed-by: Chaitanya Dhere <chaitanya.dhere@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 67 ++++++++++----------
 1 file changed, 33 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 2d61c2a91cee..c7582f834a43 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -433,28 +433,27 @@ static void populate_subvp_cmd_drr_info(struct dc *dc,
 	pipe_data->pipe_config.vblank_data.drr_info.use_ramping = false; // for now don't use ramping
 	pipe_data->pipe_config.vblank_data.drr_info.drr_window_size_ms = 4; // hardcode 4ms DRR window for now
 
-	drr_frame_us = div64_s64(drr_timing->v_total * drr_timing->h_total,
-				 (int64_t)(drr_timing->pix_clk_100hz * 100) * 1000000);
+	drr_frame_us = drr_timing->v_total * drr_timing->h_total /
+			(double)(drr_timing->pix_clk_100hz * 100) * 1000000;
 	// P-State allow width and FW delays already included phantom_timing->v_addressable
-	mall_region_us = div64_s64(phantom_timing->v_addressable * phantom_timing->h_total,
-				   (int64_t)(phantom_timing->pix_clk_100hz * 100) * 1000000);
+	mall_region_us = phantom_timing->v_addressable * phantom_timing->h_total /
+			(double)(phantom_timing->pix_clk_100hz * 100) * 1000000;
 	min_drr_supported_us = drr_frame_us + mall_region_us + SUBVP_DRR_MARGIN_US;
-	min_vtotal_supported = div64_s64(drr_timing->pix_clk_100hz * 100 *
-					 (div64_s64((int64_t)min_drr_supported_us, 1000000)),
-					 (int64_t)drr_timing->h_total);
-
-	prefetch_us = div64_s64((phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total,
-				(int64_t)(phantom_timing->pix_clk_100hz * 100) * 1000000 +
-				dc->caps.subvp_prefetch_end_to_mall_start_us);
-	subvp_active_us = div64_s64(main_timing->v_addressable * main_timing->h_total,
-				    (int64_t)(main_timing->pix_clk_100hz * 100) * 1000000);
-	drr_active_us = div64_s64(drr_timing->v_addressable * drr_timing->h_total,
-				  (int64_t)(drr_timing->pix_clk_100hz * 100) * 1000000);
-	max_drr_vblank_us = div64_s64((int64_t)(subvp_active_us - prefetch_us - drr_active_us), 2) + drr_active_us;
+	min_vtotal_supported = drr_timing->pix_clk_100hz * 100 * ((double)min_drr_supported_us / 1000000) /
+			(double)drr_timing->h_total;
+
+	prefetch_us = (phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total /
+			(double)(phantom_timing->pix_clk_100hz * 100) * 1000000 +
+			dc->caps.subvp_prefetch_end_to_mall_start_us;
+	subvp_active_us = main_timing->v_addressable * main_timing->h_total /
+			(double)(main_timing->pix_clk_100hz * 100) * 1000000;
+	drr_active_us = drr_timing->v_addressable * drr_timing->h_total /
+			(double)(drr_timing->pix_clk_100hz * 100) * 1000000;
+	max_drr_vblank_us = (double)(subvp_active_us - prefetch_us - drr_active_us) / 2 + drr_active_us;
 	max_drr_mallregion_us = subvp_active_us - prefetch_us - mall_region_us;
 	max_drr_supported_us = max_drr_vblank_us > max_drr_mallregion_us ? max_drr_vblank_us : max_drr_mallregion_us;
-	max_vtotal_supported = div64_s64(drr_timing->pix_clk_100hz * 100 * (div64_s64((int64_t)max_drr_supported_us, 1000000)),
-					 (int64_t)drr_timing->h_total);
+	max_vtotal_supported = drr_timing->pix_clk_100hz * 100 * ((double)max_drr_supported_us / 1000000) /
+			(double)drr_timing->h_total;
 
 	pipe_data->pipe_config.vblank_data.drr_info.min_vtotal_supported = min_vtotal_supported;
 	pipe_data->pipe_config.vblank_data.drr_info.max_vtotal_supported = max_vtotal_supported;
@@ -548,10 +547,10 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 	struct dc_crtc_timing *phantom_timing1 = &subvp_pipes[1]->stream->mall_stream_config.paired_stream->timing;
 	struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 *pipe_data = NULL;
 
-	subvp0_prefetch_us = div64_s64((phantom_timing0->v_total - phantom_timing0->v_front_porch) * phantom_timing0->h_total,
-				       (int64_t)(phantom_timing0->pix_clk_100hz * 100) * 1000000 + dc->caps.subvp_prefetch_end_to_mall_start_us);
-	subvp1_prefetch_us = div64_s64((phantom_timing1->v_total - phantom_timing1->v_front_porch) * phantom_timing1->h_total,
-				       (int64_t)(phantom_timing1->pix_clk_100hz * 100) * 1000000 + dc->caps.subvp_prefetch_end_to_mall_start_us);
+	subvp0_prefetch_us = (phantom_timing0->v_total - phantom_timing0->v_front_porch) * phantom_timing0->h_total /
+				(double)(phantom_timing0->pix_clk_100hz * 100) * 1000000 + dc->caps.subvp_prefetch_end_to_mall_start_us;
+	subvp1_prefetch_us = (phantom_timing1->v_total - phantom_timing1->v_front_porch) * phantom_timing1->h_total /
+					(double)(phantom_timing1->pix_clk_100hz * 100) * 1000000 + dc->caps.subvp_prefetch_end_to_mall_start_us;
 
 	// Whichever SubVP PIPE has the smaller prefetch (including the prefetch end to mall start time)
 	// should increase it's prefetch time to match the other
@@ -559,16 +558,16 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 		pipe_data = &cmd->fw_assisted_mclk_switch_v2.config_data.pipe_data[1];
 		prefetch_delta_us = subvp0_prefetch_us - subvp1_prefetch_us;
 		pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-			div64_s64(((div64_s64((int64_t)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us), 1000000)) *
-				   (phantom_timing1->pix_clk_100hz * 100) + phantom_timing1->h_total - 1),
-				  (int64_t)phantom_timing1->h_total);
+					(((double)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) / 1000000) *
+					(phantom_timing1->pix_clk_100hz * 100) + phantom_timing1->h_total - 1) /
+					(double)phantom_timing1->h_total;
 	} else if (subvp1_prefetch_us >  subvp0_prefetch_us) {
 		pipe_data = &cmd->fw_assisted_mclk_switch_v2.config_data.pipe_data[0];
 		prefetch_delta_us = subvp1_prefetch_us - subvp0_prefetch_us;
 		pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-			div64_s64(((div64_s64((int64_t)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us), 1000000)) *
-				   (phantom_timing0->pix_clk_100hz * 100) + phantom_timing0->h_total - 1),
-				  (int64_t)phantom_timing0->h_total);
+					(((double)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) / 1000000) *
+					(phantom_timing0->pix_clk_100hz * 100) + phantom_timing0->h_total - 1) /
+					(double)phantom_timing0->h_total;
 	}
 }
 
@@ -619,13 +618,13 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc,
 
 	// Round up
 	pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-		div64_s64(((div64_s64((int64_t)dc->caps.subvp_prefetch_end_to_mall_start_us, 1000000)) *
-			   (phantom_timing->pix_clk_100hz * 100) + phantom_timing->h_total - 1),
-			  (int64_t)phantom_timing->h_total);
+			(((double)dc->caps.subvp_prefetch_end_to_mall_start_us / 1000000) *
+			(phantom_timing->pix_clk_100hz * 100) + phantom_timing->h_total - 1) /
+			(double)phantom_timing->h_total;
 	pipe_data->pipe_config.subvp_data.processing_delay_lines =
-		div64_s64(((div64_s64((int64_t)dc->caps.subvp_fw_processing_delay_us, 1000000)) *
-			   (phantom_timing->pix_clk_100hz * 100) + phantom_timing->h_total - 1),
-			  (int64_t)phantom_timing->h_total);
+			(((double)dc->caps.subvp_fw_processing_delay_us / 1000000) *
+			(phantom_timing->pix_clk_100hz * 100) + phantom_timing->h_total - 1) /
+			(double)phantom_timing->h_total;
 	// Find phantom pipe index based on phantom stream
 	for (j = 0; j < dc->res_pool->pipe_count; j++) {
 		struct pipe_ctx *phantom_pipe = &context->res_ctx.pipe_ctx[j];
-- 
2.25.1

[PATCH 07/32] drm/amd/display: Add a variable to update FCLK latency

[Tom Chung]

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

[Description]
Add a variable to update FCLK latency

Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                    | 1 +
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c   | 7 +++++++
 drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c | 7 +++++++
 3 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 8e1e40083ec8..d73e68c2fd51 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -609,6 +609,7 @@ struct dc_bounding_box_overrides {
 	int percent_of_ideal_drambw;
 	int dram_clock_change_latency_ns;
 	int dummy_clock_change_latency_ns;
+	int fclk_clock_change_latency_ns;
 	/* This forces a hard min on the DCFCLK we use
 	 * for DML.  Unlike the debug option for forcing
 	 * DCFCLK, this override affects watermark calculations
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 66453546e24f..f8023cfe79f8 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2098,6 +2098,13 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa
 				dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
 		}
 
+		if ((int)(dcn3_2_soc.fclk_change_latency_us * 1000)
+				!= dc->bb_overrides.fclk_clock_change_latency_ns
+				&& dc->bb_overrides.fclk_clock_change_latency_ns) {
+			dcn3_2_soc.fclk_change_latency_us =
+				dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
+		}
+
 		if ((int)(dcn3_2_soc.dummy_pstate_latency_us * 1000)
 				!= dc->bb_overrides.dummy_clock_change_latency_ns
 				&& dc->bb_overrides.dummy_clock_change_latency_ns) {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
index 84b4b00f29cb..c87091683b5d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
@@ -498,6 +498,13 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 				dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
 		}
 
+		if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000)
+				!= dc->bb_overrides.fclk_clock_change_latency_ns
+				&& dc->bb_overrides.fclk_clock_change_latency_ns) {
+			dcn3_21_soc.fclk_change_latency_us =
+				dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
+		}
+
 		if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000)
 				!= dc->bb_overrides.dummy_clock_change_latency_ns
 				&& dc->bb_overrides.dummy_clock_change_latency_ns) {
-- 
2.25.1

[PATCH 08/32] drm/amd/display: Refactor SubVP calculation to remove FPU

[Tom Chung]

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

[Description]
Refactor calculation to remove floating point operations

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 86 ++++++++++----------
 1 file changed, 42 insertions(+), 44 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index c7582f834a43..ffa1c0148330 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -417,43 +417,42 @@ static void populate_subvp_cmd_drr_info(struct dc *dc,
 	struct dc_crtc_timing *main_timing = &subvp_pipe->stream->timing;
 	struct dc_crtc_timing *phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing;
 	struct dc_crtc_timing *drr_timing = &vblank_pipe->stream->timing;
-	int16_t drr_frame_us = 0;
-	int16_t min_drr_supported_us = 0;
-	int16_t max_drr_supported_us = 0;
-	int16_t max_drr_vblank_us = 0;
-	int16_t max_drr_mallregion_us = 0;
-	int16_t mall_region_us = 0;
-	int16_t prefetch_us = 0;
-	int16_t subvp_active_us = 0;
-	int16_t drr_active_us = 0;
-	int16_t min_vtotal_supported = 0;
-	int16_t max_vtotal_supported = 0;
+	uint16_t drr_frame_us = 0;
+	uint16_t min_drr_supported_us = 0;
+	uint16_t max_drr_supported_us = 0;
+	uint16_t max_drr_vblank_us = 0;
+	uint16_t max_drr_mallregion_us = 0;
+	uint16_t mall_region_us = 0;
+	uint16_t prefetch_us = 0;
+	uint16_t subvp_active_us = 0;
+	uint16_t drr_active_us = 0;
+	uint16_t min_vtotal_supported = 0;
+	uint16_t max_vtotal_supported = 0;
 
 	pipe_data->pipe_config.vblank_data.drr_info.drr_in_use = true;
 	pipe_data->pipe_config.vblank_data.drr_info.use_ramping = false; // for now don't use ramping
 	pipe_data->pipe_config.vblank_data.drr_info.drr_window_size_ms = 4; // hardcode 4ms DRR window for now
 
-	drr_frame_us = drr_timing->v_total * drr_timing->h_total /
-			(double)(drr_timing->pix_clk_100hz * 100) * 1000000;
+	drr_frame_us = (uint64_t)drr_timing->v_total * drr_timing->h_total * 1000000 /
+			(uint64_t)(drr_timing->pix_clk_100hz * 100);
 	// P-State allow width and FW delays already included phantom_timing->v_addressable
-	mall_region_us = phantom_timing->v_addressable * phantom_timing->h_total /
-			(double)(phantom_timing->pix_clk_100hz * 100) * 1000000;
+	mall_region_us = (uint64_t)phantom_timing->v_addressable * phantom_timing->h_total * 1000000 /
+			(uint64_t)(phantom_timing->pix_clk_100hz * 100);
 	min_drr_supported_us = drr_frame_us + mall_region_us + SUBVP_DRR_MARGIN_US;
-	min_vtotal_supported = drr_timing->pix_clk_100hz * 100 * ((double)min_drr_supported_us / 1000000) /
-			(double)drr_timing->h_total;
-
-	prefetch_us = (phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total /
-			(double)(phantom_timing->pix_clk_100hz * 100) * 1000000 +
-			dc->caps.subvp_prefetch_end_to_mall_start_us;
-	subvp_active_us = main_timing->v_addressable * main_timing->h_total /
-			(double)(main_timing->pix_clk_100hz * 100) * 1000000;
-	drr_active_us = drr_timing->v_addressable * drr_timing->h_total /
-			(double)(drr_timing->pix_clk_100hz * 100) * 1000000;
-	max_drr_vblank_us = (double)(subvp_active_us - prefetch_us - drr_active_us) / 2 + drr_active_us;
+	min_vtotal_supported = (uint64_t)drr_timing->pix_clk_100hz * 100 * min_drr_supported_us /
+			(uint64_t)(drr_timing->h_total * 1000000);
+
+	prefetch_us = (uint64_t)(phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total * 1000000
+			/ (uint64_t)(phantom_timing->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us;
+	subvp_active_us = (uint64_t)main_timing->v_addressable * main_timing->h_total * 1000000 /
+			(uint64_t)(main_timing->pix_clk_100hz * 100);
+	drr_active_us = (uint64_t)drr_timing->v_addressable * drr_timing->h_total * 1000000 /
+			(uint64_t)(drr_timing->pix_clk_100hz * 100);
+	max_drr_vblank_us = (subvp_active_us - prefetch_us - drr_active_us) / 2 + drr_active_us;
 	max_drr_mallregion_us = subvp_active_us - prefetch_us - mall_region_us;
 	max_drr_supported_us = max_drr_vblank_us > max_drr_mallregion_us ? max_drr_vblank_us : max_drr_mallregion_us;
-	max_vtotal_supported = drr_timing->pix_clk_100hz * 100 * ((double)max_drr_supported_us / 1000000) /
-			(double)drr_timing->h_total;
+	max_vtotal_supported = (uint64_t)drr_timing->pix_clk_100hz * 100 * max_drr_supported_us /
+			(uint64_t)(1000000 * drr_timing->h_total);
 
 	pipe_data->pipe_config.vblank_data.drr_info.min_vtotal_supported = min_vtotal_supported;
 	pipe_data->pipe_config.vblank_data.drr_info.max_vtotal_supported = max_vtotal_supported;
@@ -547,10 +546,10 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 	struct dc_crtc_timing *phantom_timing1 = &subvp_pipes[1]->stream->mall_stream_config.paired_stream->timing;
 	struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 *pipe_data = NULL;
 
-	subvp0_prefetch_us = (phantom_timing0->v_total - phantom_timing0->v_front_porch) * phantom_timing0->h_total /
-				(double)(phantom_timing0->pix_clk_100hz * 100) * 1000000 + dc->caps.subvp_prefetch_end_to_mall_start_us;
-	subvp1_prefetch_us = (phantom_timing1->v_total - phantom_timing1->v_front_porch) * phantom_timing1->h_total /
-					(double)(phantom_timing1->pix_clk_100hz * 100) * 1000000 + dc->caps.subvp_prefetch_end_to_mall_start_us;
+	subvp0_prefetch_us = (uint64_t)(phantom_timing0->v_total - phantom_timing0->v_front_porch) * phantom_timing0->h_total * 1000000 /
+			(uint64_t)(phantom_timing0->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us;
+	subvp1_prefetch_us = (uint64_t)(phantom_timing1->v_total - phantom_timing1->v_front_porch) * phantom_timing1->h_total * 1000000 /
+			(uint64_t)(phantom_timing1->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us;
 
 	// Whichever SubVP PIPE has the smaller prefetch (including the prefetch end to mall start time)
 	// should increase it's prefetch time to match the other
@@ -558,16 +557,17 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 		pipe_data = &cmd->fw_assisted_mclk_switch_v2.config_data.pipe_data[1];
 		prefetch_delta_us = subvp0_prefetch_us - subvp1_prefetch_us;
 		pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-					(((double)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) / 1000000) *
-					(phantom_timing1->pix_clk_100hz * 100) + phantom_timing1->h_total - 1) /
-					(double)phantom_timing1->h_total;
+				((uint64_t)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) *
+					(phantom_timing1->pix_clk_100hz * 100) + ((uint64_t)phantom_timing1->h_total * 1000000 - 1)) /
+					((uint64_t)phantom_timing1->h_total * 1000000);
+
 	} else if (subvp1_prefetch_us >  subvp0_prefetch_us) {
 		pipe_data = &cmd->fw_assisted_mclk_switch_v2.config_data.pipe_data[0];
 		prefetch_delta_us = subvp1_prefetch_us - subvp0_prefetch_us;
 		pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-					(((double)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) / 1000000) *
-					(phantom_timing0->pix_clk_100hz * 100) + phantom_timing0->h_total - 1) /
-					(double)phantom_timing0->h_total;
+				((uint64_t)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) *
+					(phantom_timing0->pix_clk_100hz * 100) + ((uint64_t)phantom_timing0->h_total * 1000000 - 1)) /
+					((uint64_t)phantom_timing0->h_total * 1000000);
 	}
 }
 
@@ -618,13 +618,11 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc,
 
 	// Round up
 	pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-			(((double)dc->caps.subvp_prefetch_end_to_mall_start_us / 1000000) *
-			(phantom_timing->pix_clk_100hz * 100) + phantom_timing->h_total - 1) /
-			(double)phantom_timing->h_total;
+			((uint64_t)dc->caps.subvp_prefetch_end_to_mall_start_us * (phantom_timing->pix_clk_100hz * 100) +
+					((uint64_t)phantom_timing->h_total * 1000000 - 1)) / ((uint64_t)phantom_timing->h_total * 1000000);
 	pipe_data->pipe_config.subvp_data.processing_delay_lines =
-			(((double)dc->caps.subvp_fw_processing_delay_us / 1000000) *
-			(phantom_timing->pix_clk_100hz * 100) + phantom_timing->h_total - 1) /
-			(double)phantom_timing->h_total;
+			((uint64_t)(dc->caps.subvp_fw_processing_delay_us) * (phantom_timing->pix_clk_100hz * 100) +
+					((uint64_t)phantom_timing->h_total * 1000000 - 1)) / ((uint64_t)phantom_timing->h_total * 1000000);
 	// Find phantom pipe index based on phantom stream
 	for (j = 0; j < dc->res_pool->pipe_count; j++) {
 		struct pipe_ctx *phantom_pipe = &context->res_ctx.pipe_ctx[j];
-- 
2.25.1

[PATCH 09/32] drm/amd/display: Fix TMDS 4K@60Hz YCbCr420 corruption issue

[Tom Chung]

From: Daniel Miess <Daniel.Miess@amd.com>

[Why]
DIG_FIFO_OUTPUT_PIXEL_MODE not being set for dcn314
resulting in incorrect timing for YCbCr4:2:0

[How]
Copy the implementation of set_pixels_per_cycle from dcn32
over to dcn314

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Daniel Miess <Daniel.Miess@amd.com>
---
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  | 28 +++++++++++++++++++
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.h  |  4 +++
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |  2 ++
 3 files changed, 34 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index 452f4a9bf251..39931d48f385 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -374,3 +374,31 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 
 	return odm_combine_factor;
 }
+
+void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
+{
+	uint32_t pix_per_cycle = 1;
+	uint32_t odm_combine_factor = 1;
+
+	if (!pipe_ctx || !pipe_ctx->stream || !pipe_ctx->stream_res.stream_enc)
+		return;
+
+	odm_combine_factor = get_odm_config(pipe_ctx, NULL);
+	if (optc2_is_two_pixels_per_containter(&pipe_ctx->stream->timing) || odm_combine_factor > 1
+		|| dcn314_is_dp_dig_pixel_rate_div_policy(pipe_ctx))
+		pix_per_cycle = 2;
+
+	if (pipe_ctx->stream_res.stream_enc->funcs->set_input_mode)
+		pipe_ctx->stream_res.stream_enc->funcs->set_input_mode(pipe_ctx->stream_res.stream_enc,
+				pix_per_cycle);
+}
+
+bool dcn314_is_dp_dig_pixel_rate_div_policy(struct pipe_ctx *pipe_ctx)
+{
+	struct dc *dc = pipe_ctx->stream->ctx->dc;
+
+	if (dc_is_dp_signal(pipe_ctx->stream->signal) && !is_dp_128b_132b_signal(pipe_ctx) &&
+		dc->debug.enable_dp_dig_pixel_rate_div_policy)
+		return true;
+	return false;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
index be0f5e4d48e1..d014580592ac 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
@@ -39,4 +39,8 @@ void dcn314_enable_power_gating_plane(struct dce_hwseq *hws, bool enable);
 
 unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div);
 
+void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);
+
+bool dcn314_is_dp_dig_pixel_rate_div_policy(struct pipe_ctx *pipe_ctx);
+
 #endif /* __DC_HWSS_DCN314_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
index b9debeb081fd..fcf67eb3478f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@@ -145,6 +145,8 @@ static const struct hwseq_private_funcs dcn314_private_funcs = {
 	.set_shaper_3dlut = dcn20_set_shaper_3dlut,
 	.setup_hpo_hw_control = dcn31_setup_hpo_hw_control,
 	.calculate_dccg_k1_k2_values = dcn314_calculate_dccg_k1_k2_values,
+	.set_pixels_per_cycle = dcn314_set_pixels_per_cycle,
+	.is_dp_dig_pixel_rate_div_policy = dcn314_is_dp_dig_pixel_rate_div_policy,
 };
 
 void dcn314_hw_sequencer_construct(struct dc *dc)
-- 
2.25.1

[PATCH 10/32] drm/amd/display: Device flash garbage before get in OS

[Tom Chung]

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

[Why]
Enabling stream with tg lock makes config settings
pending causing the garbage until tg unlock.

[How]
Keep the original lock mechanism
The driver doesn't lock tg if plane_state is null.

Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Chiawen Huang <chiawen.huang@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c | 1 +
 1 file changed, 1 insertion(+)

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 bed783747f16..5b5d952b2b8c 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
@@ -110,6 +110,7 @@ void dcn10_lock_all_pipes(struct dc *dc,
 		 */
 		if (pipe_ctx->top_pipe ||
 		    !pipe_ctx->stream ||
+		    !pipe_ctx->plane_state ||
 		    !tg->funcs->is_tg_enabled(tg))
 			continue;
 
-- 
2.25.1

[PATCH 11/32] drm/amd/display: Add 16 lines margin for SubVP

[Tom Chung]

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

[Description]
SUBVP_START_LINE must be aligned to 2 swaths, so add
16 lines of margin so the start line can be adjusted by
up to 16 lines for alignment purposes in FW.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                     | 1 +
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c            | 4 ++--
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c       | 2 +-
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c   | 1 +
 drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c | 1 +
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c    | 5 ++++-
 6 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index d73e68c2fd51..f2accacae85b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -213,6 +213,7 @@ struct dc_caps {
 	uint32_t cache_num_ways;
 	uint16_t subvp_fw_processing_delay_us;
 	uint16_t subvp_prefetch_end_to_mall_start_us;
+	uint8_t subvp_swath_height_margin_lines; // subvp start line must be aligned to 2 x swath height	
 	uint16_t subvp_pstate_allow_width_us;
 	uint16_t subvp_vertical_int_margin_us;
 	bool seamless_odm;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index ffa1c0148330..759d54843eff 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -275,8 +275,7 @@ void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst)
 	union dmub_rb_cmd cmd = { 0 };
 
 	cmd.drr_update.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
-	// TODO: Uncomment once FW headers are promoted
-	//cmd.drr_update.header.sub_type = DMUB_CMD__FAMS_SET_MANUAL_TRIGGER;
+	cmd.drr_update.header.sub_type = DMUB_CMD__FAMS_SET_MANUAL_TRIGGER;
 	cmd.drr_update.dmub_optc_state_req.tg_inst = tg_inst;
 
 	cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header);
@@ -611,6 +610,7 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc,
 			main_timing->v_total - main_timing->v_front_porch - main_timing->v_addressable;
 	pipe_data->pipe_config.subvp_data.mall_region_lines = phantom_timing->v_addressable;
 	pipe_data->pipe_config.subvp_data.main_pipe_index = subvp_pipe->pipe_idx;
+	pipe_data->pipe_config.subvp_data.is_drr = subvp_pipe->stream->ignore_msa_timing_param;	
 
 	// Prefetch lines is equal to VACTIVE + BP + VSYNC
 	pipe_data->pipe_config.subvp_data.prefetch_lines =
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
index 992e56c6907e..9861be1dc063 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
@@ -282,7 +282,7 @@ static struct timing_generator_funcs dcn32_tg_funcs = {
 		.lock_doublebuffer_disable = optc3_lock_doublebuffer_disable,
 		.enable_optc_clock = optc1_enable_optc_clock,
 		.set_vrr_m_const = optc3_set_vrr_m_const,
-		.set_drr = optc31_set_drr, // TODO: Update to optc32_set_drr once FW headers are promoted
+		.set_drr = optc32_set_drr,
 		.get_last_used_drr_vtotal = optc2_get_last_used_drr_vtotal,
 		.set_vtotal_min_max = optc3_set_vtotal_min_max,
 		.set_static_screen_control = optc1_set_static_screen_control,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 9a26d24b579f..96076d633273 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -2051,6 +2051,7 @@ static bool dcn32_resource_construct(
 	dc->caps.max_cab_allocation_bytes = 67108864; // 64MB = 1024 * 1024 * 64
 	dc->caps.subvp_fw_processing_delay_us = 15;
 	dc->caps.subvp_prefetch_end_to_mall_start_us = 15;
+	dc->caps.subvp_swath_height_margin_lines = 16;
 	dc->caps.subvp_pstate_allow_width_us = 20;
 	dc->caps.subvp_vertical_int_margin_us = 30;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index 8157e40d2c7e..f627b494737a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -1662,6 +1662,7 @@ static bool dcn321_resource_construct(
 	dc->caps.max_cab_allocation_bytes = 33554432; // 32MB = 1024 * 1024 * 32
 	dc->caps.subvp_fw_processing_delay_us = 15;
 	dc->caps.subvp_prefetch_end_to_mall_start_us = 15;
+	dc->caps.subvp_swath_height_margin_lines = 16;
 	dc->caps.subvp_pstate_allow_width_us = 20;
 
 	dc->caps.max_slave_planes = 1;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index f8023cfe79f8..988e28a8e90e 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -473,8 +473,11 @@ void dcn32_set_phantom_stream_timing(struct dc *dc,
 
 	// DML calculation for MALL region doesn't take into account FW delay
 	// and required pstate allow width for multi-display cases
+	/* Add 16 lines margin to the MALL REGION because SUB_VP_START_LINE must be aligned
+	 * to 2 swaths (i.e. 16 lines)
+	 */
 	phantom_vactive = get_subviewport_lines_needed_in_mall(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx) +
-				pstate_width_fw_delay_lines;
+				pstate_width_fw_delay_lines + dc->caps.subvp_swath_height_margin_lines;
 
 	// For backporch of phantom pipe, use vstartup of the main pipe
 	phantom_bp = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
-- 
2.25.1

[PATCH 12/32] drm/amd/display: Avoid MPC infinite loop

[Tom Chung]

From: Josip Pavic <Josip.Pavic@amd.com>

[Why]
In some cases MPC tree bottom pipe ends up point to itself.  This causes
iterating from top to bottom to hang the system in an infinite loop.

[How]
When looping to next MPC bottom pipe, check that the pointer is not same
as current to avoid infinite loop.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Josip Pavic <Josip.Pavic@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c | 6 ++++++
 2 files changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c
index 769974375b4b..8e9384094f6d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c
@@ -131,6 +131,12 @@ struct mpcc *mpc1_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id)
 	while (tmp_mpcc != NULL) {
 		if (tmp_mpcc->dpp_id == dpp_id)
 			return tmp_mpcc;
+
+		/* avoid circular linked list */
+		ASSERT(tmp_mpcc != tmp_mpcc->mpcc_bot);
+		if (tmp_mpcc == tmp_mpcc->mpcc_bot)
+			break;
+
 		tmp_mpcc = tmp_mpcc->mpcc_bot;
 	}
 	return NULL;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c
index 3d307dd58e9a..116f67a0b989 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c
@@ -531,6 +531,12 @@ static struct mpcc *mpc2_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id)
 	while (tmp_mpcc != NULL) {
 		if (tmp_mpcc->dpp_id == 0xf || tmp_mpcc->dpp_id == dpp_id)
 			return tmp_mpcc;
+
+		/* avoid circular linked list */
+		ASSERT(tmp_mpcc != tmp_mpcc->mpcc_bot);
+		if (tmp_mpcc == tmp_mpcc->mpcc_bot)
+			break;
+
 		tmp_mpcc = tmp_mpcc->mpcc_bot;
 	}
 	return NULL;
-- 
2.25.1

[PATCH 13/32] drm/amd/display: Fix subvp calculations

[Tom Chung]

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

[Descriptions]
Missed some brackets in order of operations

Reviewed-by: Samson Tam <Samson.Tam@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 48 ++++++++++----------
 1 file changed, 25 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 759d54843eff..62f7dd94f498 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -432,26 +432,26 @@ static void populate_subvp_cmd_drr_info(struct dc *dc,
 	pipe_data->pipe_config.vblank_data.drr_info.use_ramping = false; // for now don't use ramping
 	pipe_data->pipe_config.vblank_data.drr_info.drr_window_size_ms = 4; // hardcode 4ms DRR window for now
 
-	drr_frame_us = (uint64_t)drr_timing->v_total * drr_timing->h_total * 1000000 /
-			(uint64_t)(drr_timing->pix_clk_100hz * 100);
+	drr_frame_us = ((uint64_t)drr_timing->v_total * drr_timing->h_total * 1000000) /
+			(((uint64_t)drr_timing->pix_clk_100hz * 100));
 	// P-State allow width and FW delays already included phantom_timing->v_addressable
-	mall_region_us = (uint64_t)phantom_timing->v_addressable * phantom_timing->h_total * 1000000 /
-			(uint64_t)(phantom_timing->pix_clk_100hz * 100);
+	mall_region_us = ((uint64_t)phantom_timing->v_addressable * phantom_timing->h_total * 1000000) /
+			(((uint64_t)phantom_timing->pix_clk_100hz * 100));
 	min_drr_supported_us = drr_frame_us + mall_region_us + SUBVP_DRR_MARGIN_US;
-	min_vtotal_supported = (uint64_t)drr_timing->pix_clk_100hz * 100 * min_drr_supported_us /
-			(uint64_t)(drr_timing->h_total * 1000000);
-
-	prefetch_us = (uint64_t)(phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total * 1000000
-			/ (uint64_t)(phantom_timing->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us;
-	subvp_active_us = (uint64_t)main_timing->v_addressable * main_timing->h_total * 1000000 /
-			(uint64_t)(main_timing->pix_clk_100hz * 100);
-	drr_active_us = (uint64_t)drr_timing->v_addressable * drr_timing->h_total * 1000000 /
-			(uint64_t)(drr_timing->pix_clk_100hz * 100);
+	min_vtotal_supported = ((uint64_t)drr_timing->pix_clk_100hz * 100 * min_drr_supported_us) /
+			(((uint64_t)drr_timing->h_total * 1000000));
+
+	prefetch_us = ((uint64_t)(phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total * 1000000)
+			/ (((uint64_t)phantom_timing->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us);
+	subvp_active_us = ((uint64_t)main_timing->v_addressable * main_timing->h_total * 1000000) /
+			(((uint64_t)main_timing->pix_clk_100hz * 100));
+	drr_active_us = ((uint64_t)drr_timing->v_addressable * drr_timing->h_total * 1000000) /
+			(((uint64_t)drr_timing->pix_clk_100hz * 100));
 	max_drr_vblank_us = (subvp_active_us - prefetch_us - drr_active_us) / 2 + drr_active_us;
 	max_drr_mallregion_us = subvp_active_us - prefetch_us - mall_region_us;
 	max_drr_supported_us = max_drr_vblank_us > max_drr_mallregion_us ? max_drr_vblank_us : max_drr_mallregion_us;
-	max_vtotal_supported = (uint64_t)drr_timing->pix_clk_100hz * 100 * max_drr_supported_us /
-			(uint64_t)(1000000 * drr_timing->h_total);
+	max_vtotal_supported = ((uint64_t)drr_timing->pix_clk_100hz * 100 * max_drr_supported_us) /
+			(((uint64_t)drr_timing->h_total * 1000000));
 
 	pipe_data->pipe_config.vblank_data.drr_info.min_vtotal_supported = min_vtotal_supported;
 	pipe_data->pipe_config.vblank_data.drr_info.max_vtotal_supported = max_vtotal_supported;
@@ -545,10 +545,12 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 	struct dc_crtc_timing *phantom_timing1 = &subvp_pipes[1]->stream->mall_stream_config.paired_stream->timing;
 	struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 *pipe_data = NULL;
 
-	subvp0_prefetch_us = (uint64_t)(phantom_timing0->v_total - phantom_timing0->v_front_porch) * phantom_timing0->h_total * 1000000 /
-			(uint64_t)(phantom_timing0->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us;
-	subvp1_prefetch_us = (uint64_t)(phantom_timing1->v_total - phantom_timing1->v_front_porch) * phantom_timing1->h_total * 1000000 /
-			(uint64_t)(phantom_timing1->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us;
+	subvp0_prefetch_us = ((uint64_t)(phantom_timing0->v_total - phantom_timing0->v_front_porch) *
+			(uint64_t)phantom_timing0->h_total * 1000000) /
+			(((uint64_t)phantom_timing0->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us);
+	subvp1_prefetch_us = ((uint64_t)(phantom_timing1->v_total - phantom_timing1->v_front_porch) *
+			(uint64_t)phantom_timing1->h_total * 1000000) /
+			(((uint64_t)phantom_timing1->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us);
 
 	// Whichever SubVP PIPE has the smaller prefetch (including the prefetch end to mall start time)
 	// should increase it's prefetch time to match the other
@@ -557,7 +559,7 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 		prefetch_delta_us = subvp0_prefetch_us - subvp1_prefetch_us;
 		pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
 				((uint64_t)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) *
-					(phantom_timing1->pix_clk_100hz * 100) + ((uint64_t)phantom_timing1->h_total * 1000000 - 1)) /
+					((uint64_t)phantom_timing1->pix_clk_100hz * 100) + ((uint64_t)phantom_timing1->h_total * 1000000 - 1)) /
 					((uint64_t)phantom_timing1->h_total * 1000000);
 
 	} else if (subvp1_prefetch_us >  subvp0_prefetch_us) {
@@ -565,7 +567,7 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc,
 		prefetch_delta_us = subvp1_prefetch_us - subvp0_prefetch_us;
 		pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
 				((uint64_t)(dc->caps.subvp_prefetch_end_to_mall_start_us + prefetch_delta_us) *
-					(phantom_timing0->pix_clk_100hz * 100) + ((uint64_t)phantom_timing0->h_total * 1000000 - 1)) /
+					((uint64_t)phantom_timing0->pix_clk_100hz * 100) + ((uint64_t)phantom_timing0->h_total * 1000000 - 1)) /
 					((uint64_t)phantom_timing0->h_total * 1000000);
 	}
 }
@@ -618,10 +620,10 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc,
 
 	// Round up
 	pipe_data->pipe_config.subvp_data.prefetch_to_mall_start_lines =
-			((uint64_t)dc->caps.subvp_prefetch_end_to_mall_start_us * (phantom_timing->pix_clk_100hz * 100) +
+			((uint64_t)dc->caps.subvp_prefetch_end_to_mall_start_us * ((uint64_t)phantom_timing->pix_clk_100hz * 100) +
 					((uint64_t)phantom_timing->h_total * 1000000 - 1)) / ((uint64_t)phantom_timing->h_total * 1000000);
 	pipe_data->pipe_config.subvp_data.processing_delay_lines =
-			((uint64_t)(dc->caps.subvp_fw_processing_delay_us) * (phantom_timing->pix_clk_100hz * 100) +
+			((uint64_t)(dc->caps.subvp_fw_processing_delay_us) * ((uint64_t)phantom_timing->pix_clk_100hz * 100) +
 					((uint64_t)phantom_timing->h_total * 1000000 - 1)) / ((uint64_t)phantom_timing->h_total * 1000000);
 	// Find phantom pipe index based on phantom stream
 	for (j = 0; j < dc->res_pool->pipe_count; j++) {
-- 
2.25.1

[PATCH 14/32] drm/amd/display: Fix HDMI VSIF V3 incorrect issue

[Tom Chung]

From: Leo Ma <hanghong.ma@amd.com>

[Why]
Reported from customer the checksum in AMD VSIF V3 is incorrect and
causing blank screen issue.

[How]
Fix the packet length issue on AMD HDMI VSIF V3.

Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Leo Ma <hanghong.ma@amd.com>
---
 .../drm/amd/display/modules/freesync/freesync.c   | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index da09ba7589f7..0f39ab9dc5b4 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -613,10 +613,6 @@ static void build_vrr_infopacket_data_v1(const struct mod_vrr_params *vrr,
 	 * Note: We should never go above the field rate of the mode timing set.
 	 */
 	infopacket->sb[8] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000);
-
-	/* FreeSync HDR */
-	infopacket->sb[9] = 0;
-	infopacket->sb[10] = 0;
 }
 
 static void build_vrr_infopacket_data_v3(const struct mod_vrr_params *vrr,
@@ -684,10 +680,6 @@ static void build_vrr_infopacket_data_v3(const struct mod_vrr_params *vrr,
 
 	/* PB16 : Reserved bits 7:1, FixedRate bit 0 */
 	infopacket->sb[16] = (vrr->state == VRR_STATE_ACTIVE_FIXED) ? 1 : 0;
-
-	//FreeSync HDR
-	infopacket->sb[9] = 0;
-	infopacket->sb[10] = 0;
 }
 
 static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf,
@@ -772,8 +764,7 @@ static void build_vrr_infopacket_header_v2(enum signal_type signal,
 		/* HB2  = [Bits 7:5 = 0] [Bits 4:0 = Length = 0x09] */
 		infopacket->hb2 = 0x09;
 
-		*payload_size = 0x0A;
-
+		*payload_size = 0x09;
 	} else if (dc_is_dp_signal(signal)) {
 
 		/* HEADER */
@@ -822,9 +813,9 @@ static void build_vrr_infopacket_header_v3(enum signal_type signal,
 		infopacket->hb1 = version;
 
 		/* HB2  = [Bits 7:5 = 0] [Bits 4:0 = Length] */
-		*payload_size = 0x10;
-		infopacket->hb2 = *payload_size - 1; //-1 for checksum
+		infopacket->hb2 = 0x10;
 
+		*payload_size = 0x10;
 	} else if (dc_is_dp_signal(signal)) {
 
 		/* HEADER */
-- 
2.25.1

[PATCH 15/32] drm/amd/display: Add documentation to some of the cursor struct

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

Reviewed-by: Harry Wentland <Harry.Wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_hw_types.h | 62 +++++++++++++++++---
 1 file changed, 55 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
index 584aaf6967fd..d158aa4985f8 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
@@ -417,19 +417,43 @@ enum dc_scan_direction {
 	SCAN_DIRECTION_VERTICAL = 2,    /* 90, 270 rotation */
 };
 
+/**
+ * struct dc_cursor_position: Hardware cursor data.
+ *
+ * This struct keeps the action information related to the cursor that will be
+ * sent and received from our DC core.
+ */
 struct dc_cursor_position {
+	/**
+	 * @x: It represents the top left abscissa coordinate of the cursor.
+	 */
 	uint32_t x;
+
+	/**
+	 * @y: It is the top ordinate of the cursor coordinate.
+	 */
 	uint32_t y;
 
+	/**
+	 * @x_hotspot: Define the abscissa point where mouse click happens.
+	 */
 	uint32_t x_hotspot;
+
+	/**
+	 * @y_hotspot: Define the ordinate point where mouse click happens.
+	 */
 	uint32_t y_hotspot;
 
-	/*
-	 * This parameter indicates whether HW cursor should be enabled
+	/**
+	 * @enable: This parameter indicates whether hardware cursor should be
+	 * enabled.
 	 */
 	bool enable;
 
-	/* Translate cursor x/y by the source rectangle for each plane. */
+	/**
+	 * @translate_by_source: Translate cursor x/y by the source rectangle
+	 * for each plane.
+	 */
 	bool translate_by_source;
 };
 
@@ -494,7 +518,9 @@ struct dc_gamma {
 /* Used by both ipp amd opp functions*/
 /* TODO: to be consolidated with enum color_space */
 
-/*
+/**
+ * enum dc_cursor_color_format - DC cursor programming mode
+ *
  * This enum is for programming CURSOR_MODE register field. What this register
  * should be programmed to depends on OS requested cursor shape flags and what
  * we stored in the cursor surface.
@@ -530,17 +556,39 @@ union dc_cursor_attribute_flags {
 };
 
 struct dc_cursor_attributes {
+	/**
+	 * @address: This field represents the framebuffer address associated
+	 * with the cursor. It is important to highlight that this address is
+	 * divided into a high and low parts.
+	 */
 	PHYSICAL_ADDRESS_LOC address;
+
+	/**
+	 * @pitch: Cursor line stride.
+	 */
 	uint32_t pitch;
 
-	/* Width and height should correspond to cursor surface width x heigh */
+	/**
+	 * @width: Width should correspond to cursor surface width.
+	 */
 	uint32_t width;
+	/**
+	 * @heigh: Height should correspond to cursor surface heigh.
+	 */
 	uint32_t height;
 
+	/**
+	 * @color_format: DC cursor programming mode.
+	 */
 	enum dc_cursor_color_format color_format;
-	uint32_t sdr_white_level; // for boosting (SDR) cursor in HDR mode
+	/**
+	 * @sdr_white_level: Boosting (SDR) cursor in HDR mode.
+	 */
+	uint32_t sdr_white_level;
 
-	/* In case we support HW Cursor rotation in the future */
+	/**
+	 * @rotation_angle: In case we support HW Cursor rotation in the future
+	 */
 	enum dc_rotation_angle rotation_angle;
 
 	union dc_cursor_attribute_flags attribute_flags;
-- 
2.25.1

[PATCH 16/32] drm/amd/display: Add basic kernel doc to CRC code under DC

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

Reviewed-by: Harry Wentland <Harry.Wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 17 +++++++------
 .../gpu/drm/amd/display/dc/dcn10/dcn10_optc.c | 25 ++++++++++++++++---
 .../gpu/drm/amd/display/dc/inc/core_types.h   |  4 +++
 .../amd/display/dc/inc/hw/timing_generator.h  | 14 ++++++++---
 4 files changed, 46 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 e42f44fc1c08..6f5a7e4378a7 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -638,14 +638,17 @@ bool dc_stream_configure_crc(struct dc *dc, struct dc_stream_state *stream,
 
 /**
  * dc_stream_get_crc() - Get CRC values for the given stream.
- * @dc: DC object
+ *
+ * @dc: DC object.
  * @stream: The DC stream state of the stream to get CRCs from.
- * @r_cr: CRC value for the first of the 3 channels stored here.
- * @g_y:  CRC value for the second of the 3 channels stored here.
- * @b_cb: CRC value for the third of the 3 channels stored here.
+ * @r_cr: CRC value for the red component.
+ * @g_y:  CRC value for the green component.
+ * @b_cb: CRC value for the blue component.
  *
  * dc_stream_configure_crc needs to be called beforehand to enable CRCs.
- * Return false if stream is not found, or if CRCs are not enabled.
+ *
+ * Return:
+ * false if stream is not found, or if CRCs are not enabled.
  */
 bool dc_stream_get_crc(struct dc *dc, struct dc_stream_state *stream,
 		       uint32_t *r_cr, uint32_t *g_y, uint32_t *b_cb)
@@ -4271,8 +4274,8 @@ void dc_mclk_switch_using_fw_based_vblank_stretch_shut_down(struct dc *dc)
 /*
  *****************************************************************************
  * Function: dc_is_dmub_outbox_supported -
- * 
- * @brief 
+ *
+ * @brief
  *      Checks whether DMUB FW supports outbox notifications, if supported
  *		DM should register outbox interrupt prior to actually enabling interrupts
  *		via dc_enable_dmub_outbox
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index e1a9a45b03b6..1b14250d1b33 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -1493,8 +1493,23 @@ bool optc1_configure_crc(struct timing_generator *optc,
 	return true;
 }
 
+/**
+ * optc1_get_crc - Capture CRC result per component
+ *
+ * @optc: timing_generator instance.
+ * @r_cr: 16-bit primary CRC signature for red data.
+ * @g_y: 16-bit primary CRC signature for green data.
+ * @b_cb: 16-bit primary CRC signature for blue data.
+ *
+ * This function reads the CRC signature from the OPTC registers. Notice that
+ * we have three registers to keep the CRC result per color component (RGB).
+ *
+ * Returns:
+ * If CRC is disabled, return false; otherwise, return true, and the CRC
+ * results in the parameters.
+ */
 bool optc1_get_crc(struct timing_generator *optc,
-		    uint32_t *r_cr, uint32_t *g_y, uint32_t *b_cb)
+		   uint32_t *r_cr, uint32_t *g_y, uint32_t *b_cb)
 {
 	uint32_t field = 0;
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);
@@ -1505,12 +1520,14 @@ bool optc1_get_crc(struct timing_generator *optc,
 	if (!field)
 		return false;
 
+	/* OTG_CRC0_DATA_RG has the CRC16 results for the red and green component */
 	REG_GET_2(OTG_CRC0_DATA_RG,
-			CRC0_R_CR, r_cr,
-			CRC0_G_Y, g_y);
+		  CRC0_R_CR, r_cr,
+		  CRC0_G_Y, g_y);
 
+	/* OTG_CRC0_DATA_B has the CRC16 results for the blue component */
 	REG_GET(OTG_CRC0_DATA_B,
-			CRC0_B_CB, b_cb);
+		CRC0_B_CB, b_cb);
 
 	return true;
 }
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 b3d0a4ea2446..8919a2092ac5 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -399,6 +399,10 @@ struct pipe_ctx {
 	struct dc_stream_state *stream;
 
 	struct plane_resource plane_res;
+
+	/**
+	 * @stream_res: Reference to DCN resource components such OPP and DSC.
+	 */
 	struct stream_resource stream_res;
 	struct link_resource link_res;
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
index 62d4683f17a2..828e5c6ad1cf 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
@@ -137,7 +137,13 @@ struct crc_params {
 	bool enable;
 };
 
+/**
+ * struct timing_generator - Entry point to Output Timing Generator feature.
+ */
 struct timing_generator {
+	/**
+	 * @funcs: Timing generator control functions
+	 */
 	const struct timing_generator_funcs *funcs;
 	struct dc_bios *bp;
 	struct dc_context *ctx;
@@ -148,7 +154,9 @@ struct dc_crtc_timing;
 
 struct drr_params;
 
-
+/**
+ * struct timing_generator_funcs - Control timing generator on a given device.
+ */
 struct timing_generator_funcs {
 	bool (*validate_timing)(struct timing_generator *tg,
 							const struct dc_crtc_timing *timing);
@@ -273,8 +281,8 @@ struct timing_generator_funcs {
 			       const struct crc_params *params);
 
 	/**
-	 * Get CRCs for the given timing generator. Return false if CRCs are
-	 * not enabled (via configure_crc).
+	 * @get_crc: Get CRCs for the given timing generator. Return false if
+	 * CRCs are not enabled (via configure_crc).
 	 */
 	bool (*get_crc)(struct timing_generator *tg,
 			uint32_t *r_cr, uint32_t *g_y, uint32_t *b_cb);
-- 
2.25.1

[PATCH 17/32] drm/amd/display: Add some extra kernel doc to amdgpu_dm

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

Reviewed-by: Harry Wentland <Harry.Wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 16 +++++--
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 46 +++++++++++++++++--
 2 files changed, 54 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 0d54c1486739..ddcb7845f642 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -9322,6 +9322,7 @@ static int add_affected_mst_dsc_crtcs(struct drm_atomic_state *state, struct drm
 
 /**
  * amdgpu_dm_atomic_check() - Atomic check implementation for AMDgpu DM.
+ *
  * @dev: The DRM device
  * @state: The atomic state to commit
  *
@@ -9936,8 +9937,18 @@ static int parse_hdmi_amd_vsdb(struct amdgpu_dm_connector *aconnector,
 	return valid_vsdb_found ? i : -ENODEV;
 }
 
+/**
+ * amdgpu_dm_update_freesync_caps - Update Freesync capabilities
+ *
+ * @aconnector: Connector to query.
+ *
+ * Amdgpu supports Freesync in DP and HDMI displays, and it is required to keep
+ * track of some of the display information in the internal data struct used by
+ * amdgpu_dm. This function checks which type of connector we need to set the
+ * FreeSync parameters.
+ */
 void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
-					struct edid *edid)
+				    struct edid *edid)
 {
 	int i = 0;
 	struct detailed_timing *timing;
@@ -9950,8 +9961,8 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 
 	struct drm_device *dev = connector->dev;
 	struct amdgpu_device *adev = drm_to_adev(dev);
-	bool freesync_capable = false;
 	struct amdgpu_hdmi_vsdb_info vsdb_info = {0};
+	bool freesync_capable = false;
 
 	if (!connector->state) {
 		DRM_ERROR("%s - Connector has no state", __func__);
@@ -9980,7 +9991,6 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 	if (!adev->dm.freesync_module)
 		goto update;
 
-
 	if (sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT
 		|| sink->sink_signal == SIGNAL_TYPE_EDP) {
 		bool edid_check_required = false;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 90b306a1dd68..b44faaad9b0b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -598,6 +598,10 @@ struct amdgpu_dm_connector {
 	 * The 'current' sink is in dc_link->sink. */
 	struct dc_sink *dc_sink;
 	struct dc_link *dc_link;
+
+	/**
+	 * @dc_em_sink: Reference to the emulated (virtual) sink.
+	 */
 	struct dc_sink *dc_em_sink;
 
 	/* DM only */
@@ -610,7 +614,16 @@ struct amdgpu_dm_connector {
 	struct amdgpu_i2c_adapter *i2c;
 
 	/* Monitor range limits */
-	int min_vfreq ;
+	/**
+	 * @min_vfreq: Minimal frequency supported by the display in Hz. This
+	 * value is set to zero when there is no FreeSync support.
+	 */
+	int min_vfreq;
+
+	/**
+	 * @max_vfreq: Maximum frequency supported by the display in Hz. This
+	 * value is set to zero when there is no FreeSync support.
+	 */
 	int max_vfreq ;
 	int pixel_clock_mhz;
 
@@ -705,11 +718,34 @@ struct dm_connector_state {
 	uint64_t pbn;
 };
 
+/**
+ * struct amdgpu_hdmi_vsdb_info - Keep track of the VSDB info
+ *
+ * AMDGPU supports FreeSync over HDMI by using the VSDB section, and this
+ * struct is useful to keep track of the display-specific information about
+ * FreeSync.
+ */
 struct amdgpu_hdmi_vsdb_info {
-	unsigned int amd_vsdb_version;		/* VSDB version, should be used to determine which VSIF to send */
-	bool freesync_supported;		/* FreeSync Supported */
-	unsigned int min_refresh_rate_hz;	/* FreeSync Minimum Refresh Rate in Hz */
-	unsigned int max_refresh_rate_hz;	/* FreeSync Maximum Refresh Rate in Hz */
+	/**
+	 * @amd_vsdb_version: Vendor Specific Data Block Version, should be
+	 * used to determine which Vendor Specific InfoFrame (VSIF) to send.
+	 */
+	unsigned int amd_vsdb_version;
+
+	/**
+	 * @freesync_supported: FreeSync Supported.
+	 */
+	bool freesync_supported;
+
+	/**
+	 * @min_refresh_rate_hz: FreeSync Minimum Refresh Rate in Hz.
+	 */
+	unsigned int min_refresh_rate_hz;
+
+	/**
+	 * @max_refresh_rate_hz: FreeSync Maximum Refresh Rate in Hz
+	 */
+	unsigned int max_refresh_rate_hz;
 };
 
 
-- 
2.25.1

[PATCH 18/32] drm/amd/display: Document pipe split policy

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

In the DCN code, we can find some references to three different pipe
split policies but no clear description. This commit adds some
kernel-doc that explains more about these options and a little bit of
their impact.

Reviewed-by: Harry Wentland <Harry.Wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 35 +++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index f2accacae85b..07feb8551436 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -385,9 +385,31 @@ enum dcc_option {
 	DCC_HALF_REQ_DISALBE = 2,
 };
 
+/**
+ * enum pipe_split_policy - Pipe split strategy supported by DCN
+ *
+ * This enum is used to define the pipe split policy supported by DCN. By
+ * default, DC favors MPC_SPLIT_DYNAMIC.
+ */
 enum pipe_split_policy {
+	/**
+	 * @MPC_SPLIT_DYNAMIC: DC will automatically decide how to split the
+	 * pipe in order to bring the best trade-off between performance and
+	 * power consumption. This is the recommended option.
+	 */
 	MPC_SPLIT_DYNAMIC = 0,
+
+	/**
+	 * @MPC_SPLIT_DYNAMIC: Avoid pipe split, which means that DC will not
+	 * try any sort of split optimization.
+	 */
 	MPC_SPLIT_AVOID = 1,
+
+	/**
+	 * @MPC_SPLIT_DYNAMIC: With this option, DC will only try to optimize
+	 * the pipe utilization when using a single display; if the user
+	 * connects to a second display, DC will avoid pipe split.
+	 */
 	MPC_SPLIT_AVOID_MULT_DISP = 2,
 };
 
@@ -622,6 +644,14 @@ struct dc_state;
 struct resource_pool;
 struct dce_hwseq;
 
+/**
+ * struct dc_debug_options - DC debug struct
+ *
+ * This struct provides a simple mechanism for developers to change some
+ * configurations, enable/disable features, and activate extra debug options.
+ * This can be very handy to narrow down whether some specific feature is
+ * causing an issue or not.
+ */
 struct dc_debug_options {
 	bool native422_support;
 	bool disable_dsc;
@@ -641,6 +671,11 @@ struct dc_debug_options {
 	bool disable_stutter;
 	bool use_max_lb;
 	enum dcc_option disable_dcc;
+
+	/**
+	 * @pipe_split_policy: Define which pipe split policy is used by the
+	 * display core.
+	 */
 	enum pipe_split_policy pipe_split_policy;
 	bool force_single_disp_pipe_split;
 	bool voltage_align_fclk;
-- 
2.25.1

[PATCH 19/32] drm/amd/display: Expand documentation for timing

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

The timing programming inside DCN is far from trivial, it has multiple
parameters associated with that, and the lack of documentation does not
help comprehend this already complicated topic. This commit tries to
improve this situation by expanding the documentation of dc_crtc_timing
and the VTG program function.

Reviewed-by: Harry Wentland <Harry.Wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_hw_types.h  | 86 +++++++++++++++++++
 .../gpu/drm/amd/display/dc/dcn10/dcn10_optc.c | 14 +++
 2 files changed, 100 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
index d158aa4985f8..848db8676adf 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h
@@ -812,22 +812,108 @@ struct dc_dsc_config {
 	bool is_dp; /* indicate if DSC is applied based on DP's capability */
 	uint32_t mst_pbn; /* pbn of display on dsc mst hub */
 };
+
+/**
+ * struct dc_crtc_timing - Timing parameters used to configure DCN blocks
+ *
+ * DCN provides multiple signals and parameters that can be used to adjust
+ * timing parameters, this struct aggregate multiple of these values for easy
+ * access. In this struct, fields prefixed with h_* are related to horizontal
+ * timing, and v_* to vertical timing. Keep in mind that when we talk about
+ * vertical timings, the values, in general, are described in the number of
+ * lines; on the other hand, the horizontal values are in pixels.
+ */
 struct dc_crtc_timing {
+	/**
+	 * @h_total: The total number of pixels from the rising edge of HSync
+	 * until the rising edge of the current HSync.
+	 */
 	uint32_t h_total;
+
+	/**
+	 * @h_border_left: The black pixels related to the left border
+	 */
 	uint32_t h_border_left;
+
+	/**
+	 * @h_addressable: It is the range of pixels displayed horizontally.
+	 * For example, if the display resolution is 3840@2160, the horizontal
+	 * addressable area is 3840.
+	 */
 	uint32_t h_addressable;
+
+	/**
+	 * @h_border_right: The black pixels related to the right border
+	 */
 	uint32_t h_border_right;
+
+	/**
+	 * @h_front_porch: Period (in pixels) between HBlank start and the
+	 * rising edge of HSync.
+	 */
 	uint32_t h_front_porch;
+
+	/**
+	 * @h_sync_width: HSync duration in pixels.
+	 */
 	uint32_t h_sync_width;
 
+	/**
+	 * @v_total: It is the total number of lines from the rising edge of
+	 * the previous VSync until the rising edge of the current VSync.
+	 *
+	 *          |--------------------------|
+	 *          +-+        V_TOTAL         +-+
+	 *          | |                        | |
+	 * VSync ---+ +--------- // -----------+ +---
+	 */
 	uint32_t v_total;
+
+	/**
+	 * @v_border_top: The black border on the top.
+	 */
 	uint32_t v_border_top;
+
+	/**
+	 * @v_addressable: It is the range of the scanout at which the
+	 * framebuffer is displayed. For example, if the display resolution is
+	 * 3840@2160, the addressable area is 2160 lines, or if the resolution
+	 * is 1920x1080, the addressable area is 1080 lines.
+	 */
 	uint32_t v_addressable;
+
+	/**
+	 * @v_border_bottom: The black border on the bottom.
+	 */
 	uint32_t v_border_bottom;
+
+	/**
+	 * @v_front_porch: Period (in lines) between VBlank start and rising
+	 * edge of VSync.
+	 *                  +-+
+	 * VSync            | |
+	 *        ----------+ +--------...
+	 *          +------------------...
+	 * VBlank   |
+	 *        --+
+	 *          |-------|
+	 *        v_front_porch
+	 */
 	uint32_t v_front_porch;
+
+	/**
+	 * @v_sync_width: VSync signal width in lines.
+	 */
 	uint32_t v_sync_width;
 
+	/**
+	 * @pix_clk_100hz: Pipe pixel precision
+	 *
+	 * This field is used to communicate pixel clocks with 100 Hz accuracy
+	 * from dc_crtc_timing to BIOS command table.
+	 */
 	uint32_t pix_clk_100hz;
+
 	uint32_t min_refresh_in_uhz;
 
 	uint32_t vic;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index 1b14250d1b33..bcca4847590d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -312,6 +312,20 @@ void optc1_program_timing(
 	}
 }
 
+/**
+ * optc1_set_vtg_params - Set Vertical Timing Generator (VTG) parameters
+ *
+ * @optc: timing_generator struct used to extract the optc parameters
+ * @dc_crtc_timing: Timing parameters configured
+ * @program_fp2: Boolean value indicating if FP2 will be programmed or not
+ *
+ * OTG is responsible for generating the global sync signals, including
+ * vertical timing information for each HUBP in the dcfclk domain. Each VTG is
+ * associated with one OTG that provides HUBP with vertical timing information
+ * (i.e., there is 1:1 correspondence between OTG and VTG). This function is
+ * responsible for setting the OTG parameters to the VTG during the pipe
+ * programming.
+ */
 void optc1_set_vtg_params(struct timing_generator *optc,
 		const struct dc_crtc_timing *dc_crtc_timing, bool program_fp2)
 {
-- 
2.25.1

[PATCH 20/32] drm/amd/display: Document some of the DML structs

[Tom Chung]

From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>

Reviewed-by: Harry Wentland <Harry.Wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 .../amd/display/dc/dml/display_mode_structs.h | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
index e8b094006d95..8538588e7754 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
@@ -26,6 +26,16 @@
 #include "dc_features.h"
 #include "display_mode_enums.h"
 
+/**
+ * DOC: overview
+ *
+ * Most of the DML code is automatically generated and tested via hardware
+ * description language. Usually, we use the reference _vcs_dpi in the code
+ * where VCS means "Verilog Compiled Simulator" and DPI stands for "Direct
+ * Programmer Interface". In other words, those structs can be used to
+ * interface with Verilog with other languages such as C.
+ */
+
 #ifndef __DISPLAY_MODE_STRUCTS_H__
 #define __DISPLAY_MODE_STRUCTS_H__
 
@@ -159,6 +169,14 @@ struct _vcs_dpi_voltage_scaling_st {
 	double dtbclk_mhz;
 };
 
+/**
+ * _vcs_dpi_soc_bounding_box_st: SOC definitions
+ *
+ * This struct maintains the SOC Bounding Box information for the ASIC; it
+ * defines things such as clock, voltage, performance, etc. Usually, we load
+ * these values from VBIOS; if something goes wrong, we use some hard-coded
+ * values, which will enable the ASIC to light up with limitations.
+ */
 struct _vcs_dpi_soc_bounding_box_st {
 	struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES];
 	/*
@@ -166,6 +184,11 @@ struct _vcs_dpi_soc_bounding_box_st {
 	 * clock table. Do not use outside of *update_bw_boudning_box functions.
 	 */
 	struct _vcs_dpi_voltage_scaling_st _clock_tmp[DC__VOLTAGE_STATES];
+
+	/**
+	 * @num_states: It represents the total of Display Power Management
+	 * (DPM) supported by the specific ASIC.
+	 */
 	unsigned int num_states;
 	double sr_exit_time_us;
 	double sr_enter_plus_exit_time_us;
@@ -231,6 +254,14 @@ struct _vcs_dpi_soc_bounding_box_st {
 	enum self_refresh_affinity allow_dram_self_refresh_or_dram_clock_change_in_vblank;
 };
 
+/**
+ * @_vcs_dpi_ip_params_st: IP configuraion for DCN blocks
+ *
+ * In this struct you can find the DCN configuration associated to the specific
+ * ASIC. For example, here we can save how many DPPs the ASIC is using and it
+ * is available.
+ *
+ */
 struct _vcs_dpi_ip_params_st {
 	bool use_min_dcfclk;
 	bool clamp_min_dcfclk;
@@ -283,6 +314,9 @@ struct _vcs_dpi_ip_params_st {
 	unsigned int writeback_line_buffer_chroma_buffer_size;
 
 	unsigned int max_page_table_levels;
+	/**
+	 * @max_num_dpp: Maximum number of DPP supported in the target ASIC.
+	 */
 	unsigned int max_num_dpp;
 	unsigned int max_num_otg;
 	unsigned int cursor_chunk_size;
-- 
2.25.1

[PATCH 21/32] drm/amd/display: Allow alternate prefetch modes in DML for DCN32

[Tom Chung]

From: David Galiffi <David.Galiffi@amd.com>

[Why]
Driver is restricting voltage levels if system cannot switch
in vblank.

[How]
Change allow_for_pstate_or_stutter_in_vblank_final from
dm_prefetch_support_uclk_fclk_and_stutter to
dm_prefetch_support_uclk_fclk_and_stutter_if_possible.
Add support for a new registry property,
DalDMLDisallowAlternatePrefetchModes, for easier debugging.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h           |  1 +
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 34 +++++++++++++------
 2 files changed, 25 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 07feb8551436..86fe4b9f10c2 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -788,6 +788,7 @@ struct dc_debug_options {
 	uint32_t mst_start_top_delay;
 	uint8_t psr_power_use_phy_fsm;
 	enum dml_hostvm_override_opts dml_hostvm_override;
+	bool dml_disallow_alternate_prefetch_modes;
 	bool use_legacy_soc_bb_mechanism;
 	bool exit_idle_opt_for_cursor_updates;
 	bool enable_single_display_2to1_odm_policy;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 988e28a8e90e..85ecd94e2a4a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -986,9 +986,15 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 	 * DML favors voltage over p-state, but we're more interested in
 	 * supporting p-state over voltage. We can't support p-state in
 	 * prefetch mode > 0 so try capping the prefetch mode to start.
+	 * Override present for testing.
 	 */
-	context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
+	if (dc->debug.dml_disallow_alternate_prefetch_modes)
+		context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
 			dm_prefetch_support_uclk_fclk_and_stutter;
+	else
+		context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
+			dm_prefetch_support_uclk_fclk_and_stutter_if_possible;
+
 	*vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, *pipe_cnt);
 	/* This may adjust vlevel and maxMpcComb */
 	if (*vlevel < context->bw_ctx.dml.soc.num_states)
@@ -1017,7 +1023,9 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 			 * will not allow for switch in VBLANK. The DRR display must have it's VBLANK stretched
 			 * enough to support MCLK switching.
 			 */
-			if (*vlevel == context->bw_ctx.dml.soc.num_states) {
+			if (*vlevel == context->bw_ctx.dml.soc.num_states &&
+				context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final ==
+					dm_prefetch_support_uclk_fclk_and_stutter) {
 				context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
 								dm_prefetch_support_stutter;
 				/* There are params (such as FabricClock) that need to be recalculated
@@ -1347,7 +1355,8 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 	int split[MAX_PIPES] = { 0 };
 	bool merge[MAX_PIPES] = { false };
 	bool newly_split[MAX_PIPES] = { false };
-	int pipe_cnt, i, pipe_idx, vlevel;
+	int pipe_cnt, i, pipe_idx;
+	int vlevel = context->bw_ctx.dml.soc.num_states;
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 
 	dc_assert_fp_enabled();
@@ -1376,17 +1385,22 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 		DC_FP_END();
 	}
 
-	if (fast_validate || vlevel == context->bw_ctx.dml.soc.num_states ||
-			vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported) {
+	if (fast_validate ||
+			dc->debug.dml_disallow_alternate_prefetch_modes &&
+			(vlevel == context->bw_ctx.dml.soc.num_states ||
+				vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported)) {
 		/*
-		 * If mode is unsupported or there's still no p-state support then
-		 * fall back to favoring voltage.
+		 * If dml_disallow_alternate_prefetch_modes is false, then we have already
+		 * tried alternate prefetch modes during full validation.
+		 *
+		 * If mode is unsupported or there is no p-state support, then
+		 * fall back to favouring voltage.
 		 *
-		 * If Prefetch mode 0 failed for this config, or passed with Max UCLK, try if
-		 * supported with Prefetch mode 1 (dm_prefetch_support_fclk_and_stutter == 2)
+		 * If Prefetch mode 0 failed for this config, or passed with Max UCLK, then try
+		 * to support with Prefetch mode 1 (dm_prefetch_support_fclk_and_stutter == 2)
 		 */
 		context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
-				dm_prefetch_support_fclk_and_stutter;
+			dm_prefetch_support_fclk_and_stutter;
 
 		vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt);
 
-- 
2.25.1

[PATCH 22/32] drm/amd/display: Fix Compile-time Warning

[Tom Chung]

From: David Galiffi <David.Galiffi@amd.com>

Fixed bracketing around condition statement.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 85ecd94e2a4a..3316c4a64901 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -1386,9 +1386,9 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 	}
 
 	if (fast_validate ||
-			dc->debug.dml_disallow_alternate_prefetch_modes &&
+			(dc->debug.dml_disallow_alternate_prefetch_modes &&
 			(vlevel == context->bw_ctx.dml.soc.num_states ||
-				vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported)) {
+				vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported))) {
 		/*
 		 * If dml_disallow_alternate_prefetch_modes is false, then we have already
 		 * tried alternate prefetch modes during full validation.
-- 
2.25.1

[PATCH 23/32] drm/amd/display: Fix VPG instancing for dcn314 HPO

[Tom Chung]

From: Duncan Ma <duncan.ma@amd.com>

[Why]
An issue during VPG indexing offset generation causing
to use the incorrect VPG. HW team placed VPG instances
5 at end of list, making it VPG 9 in register headers.

[How]
Correct VPG instance for HPO encoders.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Duncan Ma <duncan.ma@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 1f095f05d986..85f32206a766 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -1254,7 +1254,7 @@ static struct stream_encoder *dcn314_stream_encoder_create(
 	int afmt_inst;
 
 	/* Mapping of VPG, AFMT, DME register blocks to DIO block instance */
-	if (eng_id <= ENGINE_ID_DIGF) {
+	if (eng_id < ENGINE_ID_DIGF) {
 		vpg_inst = eng_id;
 		afmt_inst = eng_id;
 	} else
@@ -1299,7 +1299,8 @@ static struct hpo_dp_stream_encoder *dcn31_hpo_dp_stream_encoder_create(
 	 * VPG[8] -> HPO_DP[2]
 	 * VPG[9] -> HPO_DP[3]
 	 */
-	vpg_inst = hpo_dp_inst + 6;
+	//Uses offset index 5-8, but actually maps to vpg_inst 6-9
+	vpg_inst = hpo_dp_inst + 5;
 
 	/* Mapping of APG register blocks to HPO DP block instance:
 	 * APG[0] -> HPO_DP[0]
-- 
2.25.1

[PATCH 24/32] drm/amd/display: Check correct bounds for stream encoder instances for DCN303

[Tom Chung]

From: Aurabindo Pillai <aurabindo.pillai@amd.com>

[Why & How]
eng_id for DCN303 cannot be more than 1, since we have only two
instances of stream encoders.

Check the correct boundary condition for engine ID for DCN303 prevent
the potential out of bounds access.

Fixes: cd6d421e3d1a ("drm/amd/display: Initial DC support for Beige Goby")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: <Stable@vger.kernel.org>

Reviewed-by: Chris Park <Chris.Park@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
index 0a67f8a5656d..d97076648acb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
@@ -372,7 +372,7 @@ static struct stream_encoder *dcn303_stream_encoder_create(enum engine_id eng_id
 	int afmt_inst;
 
 	/* Mapping of VPG, AFMT, DME register blocks to DIO block instance */
-	if (eng_id <= ENGINE_ID_DIGE) {
+	if (eng_id <= ENGINE_ID_DIGB) {
 		vpg_inst = eng_id;
 		afmt_inst = eng_id;
 	} else
-- 
2.25.1

[PATCH 24/32] drm/amd/display: Check correct bounds for stream encoder instances for DCN303

[Tom Chung]

From: Aurabindo Pillai <aurabindo.pillai@amd.com>

[Why & How]
eng_id for DCN303 cannot be more than 1, since we have only two
instances of stream encoders.

Check the correct boundary condition for engine ID for DCN303 prevent
the potential out of bounds access.

Fixes: cd6d421e3d1a ("drm/amd/display: Initial DC support for Beige Goby")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: <Stable@vger.kernel.org>

Reviewed-by: Chris Park <Chris.Park@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
index 0a67f8a5656d..d97076648acb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
@@ -372,7 +372,7 @@ static struct stream_encoder *dcn303_stream_encoder_create(enum engine_id eng_id
 	int afmt_inst;
 
 	/* Mapping of VPG, AFMT, DME register blocks to DIO block instance */
-	if (eng_id <= ENGINE_ID_DIGE) {
+	if (eng_id <= ENGINE_ID_DIGB) {
 		vpg_inst = eng_id;
 		afmt_inst = eng_id;
 	} else
-- 
2.25.1

[PATCH 25/32] drm/amd/display: Enable SubVP by default on DCN32 & DCN321

[Tom Chung]

From: Aurabindo Pillai <aurabindo.pillai@amd.com>

[Why&How]
Set the debug options to enable SubVP feature

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 96076d633273..8b887b552f2c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -867,7 +867,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		}
 	},
 	.use_max_lb = true,
-	.force_disable_subvp = true,
+	.force_disable_subvp = false,
 	.exit_idle_opt_for_cursor_updates = true,
 	.enable_single_display_2to1_odm_policy = true,
 	.enable_dp_dig_pixel_rate_div_policy = 1,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index f627b494737a..e9db5f8b6fdc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -868,7 +868,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		}
 	},
 	.use_max_lb = true,
-	.force_disable_subvp = true,
+	.force_disable_subvp = false,
 	.exit_idle_opt_for_cursor_updates = true,
 	.enable_single_display_2to1_odm_policy = true,
 	.enable_dp_dig_pixel_rate_div_policy = 1,
-- 
2.25.1

[PATCH 26/32] drm/amd/display: Correct DTBCLK for dcn314

[Tom Chung]

From: Duncan Ma <duncan.ma@amd.com>

[Why]
DTBCLK clocks reset after clocks are initialized
and bounding box values are also incorrect.

[How]
Use dcn31 init clock function programming sequence
and correct bounding box values for dcn314

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Duncan Ma <duncan.ma@amd.com>
---
 .../amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c   | 12 +-----------
 .../amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h   |  2 +-
 .../gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c   | 10 +++++-----
 3 files changed, 7 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index ee99974b3b62..7af19823a29d 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -307,16 +307,6 @@ static void dcn314_enable_pme_wa(struct clk_mgr *clk_mgr_base)
 	dcn314_smu_enable_pme_wa(clk_mgr);
 }
 
-void dcn314_init_clocks(struct clk_mgr *clk_mgr)
-{
-	memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks));
-	// Assumption is that boot state always supports pstate
-	clk_mgr->clks.p_state_change_support = true;
-	clk_mgr->clks.prev_p_state_change_support = true;
-	clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN;
-	clk_mgr->clks.zstate_support = DCN_ZSTATE_SUPPORT_UNKNOWN;
-}
-
 bool dcn314_are_clock_states_equal(struct dc_clocks *a,
 		struct dc_clocks *b)
 {
@@ -641,7 +631,7 @@ static struct clk_mgr_funcs dcn314_funcs = {
 	.get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
 	.get_dtb_ref_clk_frequency = dcn31_get_dtb_ref_freq_khz,
 	.update_clocks = dcn314_update_clocks,
-	.init_clocks = dcn314_init_clocks,
+	.init_clocks = dcn31_init_clocks,
 	.enable_pme_wa = dcn314_enable_pme_wa,
 	.are_clock_states_equal = dcn314_are_clock_states_equal,
 	.notify_wm_ranges = dcn314_notify_wm_ranges
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
index c695a4498c50..171f84340eb2 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
@@ -42,7 +42,7 @@ struct clk_mgr_dcn314 {
 
 bool dcn314_are_clock_states_equal(struct dc_clocks *a,
 		struct dc_clocks *b);
-void dcn314_init_clocks(struct clk_mgr *clk_mgr);
+
 void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
 			struct dc_state *context,
 			bool safe_to_lower);
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index d52a16d280e9..c80307a6af1b 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -106,7 +106,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 			.phyclk_mhz = 600.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 186.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 1,
@@ -115,7 +115,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 209.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 2,
@@ -124,7 +124,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 209.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 3,
@@ -133,7 +133,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 371.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 4,
@@ -142,7 +142,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 417.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 	},
 	.num_states = 5,
-- 
2.25.1

[PATCH 27/32] drm/amd/display: Revert "attempt to fix the logic in commit_planes_for_stream()"

[Tom Chung]

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

[Description]
Reverts commit "attempt to fix the logic in commit_planes_for_stream()"
since it caused a regression.

Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 27 +++++++++++-------------
 1 file changed, 12 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 6f5a7e4378a7..08f533f86520 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3232,7 +3232,7 @@ static void commit_planes_for_stream(struct dc *dc,
 				odm_pipe->ttu_regs.min_ttu_vblank = MAX_TTU;
 	}
 
-	if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) {
+	if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed)
 		if (top_pipe_to_program &&
 			top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) {
 			if (should_use_dmub_lock(stream->link)) {
@@ -3250,7 +3250,6 @@ static void commit_planes_for_stream(struct dc *dc,
 				top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable(
 						top_pipe_to_program->stream_res.tg);
 		}
-	}
 
 	if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) {
 		if (dc->hwss.subvp_pipe_control_lock)
@@ -3469,7 +3468,7 @@ static void commit_planes_for_stream(struct dc *dc,
 		dc->hwss.pipe_control_lock(dc, top_pipe_to_program, false);
 	}
 
-	if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) {
+	if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed)
 		if (top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) {
 			top_pipe_to_program->stream_res.tg->funcs->wait_for_state(
 				top_pipe_to_program->stream_res.tg,
@@ -3496,21 +3495,19 @@ static void commit_planes_for_stream(struct dc *dc,
 				top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_disable(
 					top_pipe_to_program->stream_res.tg);
 		}
-	}
 
-	if (update_type != UPDATE_TYPE_FAST) {
+	if (update_type != UPDATE_TYPE_FAST)
 		dc->hwss.post_unlock_program_front_end(dc, context);
 
-		/* Since phantom pipe programming is moved to post_unlock_program_front_end,
-		 * move the SubVP lock to after the phantom pipes have been setup
-		 */
-		if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) {
-			if (dc->hwss.subvp_pipe_control_lock)
-				dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, NULL, subvp_prev_use);
-		} else {
-			if (dc->hwss.subvp_pipe_control_lock)
-				dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, top_pipe_to_program, subvp_prev_use);
-		}
+	/* Since phantom pipe programming is moved to post_unlock_program_front_end,
+	 * move the SubVP lock to after the phantom pipes have been setup
+	 */
+	if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) {
+		if (dc->hwss.subvp_pipe_control_lock)
+			dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, NULL, subvp_prev_use);
+	} else {
+		if (dc->hwss.subvp_pipe_control_lock)
+			dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, top_pipe_to_program, subvp_prev_use);
 	}
 
 	// Fire manual trigger only when bottom plane is flipped
-- 
2.25.1

[PATCH 28/32] drm/amd/display: For stereo keep "FLIP_ANY_FRAME"

[Tom Chung]

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

[Description]
Observed in stereomode that programming FLIP_LEFT_EYE
can cause hangs. Keep FLIP_ANY_FRAME in stereo mode so
the surface flip can take place before left or right eye

Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c
index 6a4dcafb9bba..dc3e8df706b3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c
@@ -86,7 +86,7 @@ bool hubp3_program_surface_flip_and_addr(
 			VMID, address->vmid);
 
 	if (address->type == PLN_ADDR_TYPE_GRPH_STEREO) {
-		REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_MODE_FOR_STEREOSYNC, 0x1);
+		REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_MODE_FOR_STEREOSYNC, 0);
 		REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_IN_STEREOSYNC, 0x1);
 
 	} else {
-- 
2.25.1

[PATCH 29/32] drm/amd/display: Don't try to enter MALL SS if stereo3d

[Tom Chung]

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

[Description]
MALL not supported with stereo3D according
to DCN doc.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.c    | 26 ++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 5ba503d80728..4aecbf230446 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -331,7 +331,9 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c
 bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable)
 {
 	union dmub_rb_cmd cmd;
-	uint8_t ways;
+	uint8_t ways, i, j;
+	bool stereo_in_use = false;
+	struct dc_plane_state *plane = NULL;
 
 	if (!dc->ctx->dmub_srv)
 		return false;
@@ -360,7 +362,23 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable)
 			 * and configure HUBP's to fetch from MALL
 			 */
 			ways = dcn32_calculate_cab_allocation(dc, dc->current_state);
-			if (ways <= dc->caps.cache_num_ways) {
+
+			/* MALL not supported with Stereo3D. If any plane is using stereo,
+			 * don't try to enter MALL.
+			 */
+			for (i = 0; i < dc->current_state->stream_count; i++) {
+				for (j = 0; j < dc->current_state->stream_status[i].plane_count; j++) {
+					plane = dc->current_state->stream_status[i].plane_states[j];
+
+					if (plane->address.type == PLN_ADDR_TYPE_GRPH_STEREO) {
+						stereo_in_use = true;
+						break;
+					}
+				}
+				if (stereo_in_use)
+					break;
+			}
+			if (ways <= dc->caps.cache_num_ways && !stereo_in_use) {
 				memset(&cmd, 0, sizeof(cmd));
 				cmd.cab.header.type = DMUB_CMD__CAB_FOR_SS;
 				cmd.cab.header.sub_type = DMUB_CMD__CAB_DCN_SS_FIT_IN_CAB;
@@ -694,9 +712,11 @@ void dcn32_update_mall_sel(struct dc *dc, struct dc_state *context)
 			if (pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) {
 					hubp->funcs->hubp_update_mall_sel(hubp, 1, false);
 			} else {
+				// MALL not supported with Stereo3D
 				hubp->funcs->hubp_update_mall_sel(hubp,
 					num_ways <= dc->caps.cache_num_ways &&
-					pipe->stream->link->psr_settings.psr_version == DC_PSR_VERSION_UNSUPPORTED ? 2 : 0,
+					pipe->stream->link->psr_settings.psr_version == DC_PSR_VERSION_UNSUPPORTED &&
+					pipe->plane_state->address.type !=  PLN_ADDR_TYPE_GRPH_STEREO ? 2 : 0,
 							cache_cursor);
 			}
 		}
-- 
2.25.1

[PATCH 30/32] drm/amd/display: clear optc underflow before turn off odm clock

[Tom Chung]

From: Fudong Wang <Fudong.Wang@amd.com>

[Why]
After ODM clock off, optc underflow bit will be kept there always and clear not work.
We need to clear that before clock off.

[How]
Clear that if have when clock off.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Fudong Wang <Fudong.Wang@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index bcca4847590d..cdb6e2f220f4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -479,6 +479,11 @@ void optc1_enable_optc_clock(struct timing_generator *optc, bool enable)
 				OTG_CLOCK_ON, 1,
 				1, 1000);
 	} else  {
+
+		//last chance to clear underflow, otherwise, it will always there due to clock is off.
+		if (optc->funcs->is_optc_underflow_occurred(optc) == true)
+			optc->funcs->clear_optc_underflow(optc);
+
 		REG_UPDATE_2(OTG_CLOCK_CONTROL,
 				OTG_CLOCK_GATE_DIS, 0,
 				OTG_CLOCK_EN, 0);
-- 
2.25.1

[PATCH 31/32] drm/amd/display: Fix TDR eDP and USB4 display light up issue

[Tom Chung]

From: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>

[Why]
After TDR recovery, eDP and USB4 display does not light up. Because
dmub outbox notifications are not enabled after dmub reload and link
encoder assignments for the streams are not cleared before dc state
reset.

[How]
- Dmub outbox notification is enabled after tdr recovery by issuing
  inbox command to dmub.
- Link encoders for the streams are unassigned before dc state reset.
- Added WindowsDM::forceDpiaUnplugForTdr() API to remove all the enabled
  DPIA links. Because after fw reload during TDR, current DPIA pairing is
  meaningless and there would be new pairing process.

Reviewed-by: Jimmy Kizito <Jimmy.Kizito@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 1 +
 drivers/gpu/drm/amd/display/dc/dc_link.h | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 08f533f86520..0bc6174aae61 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -4340,6 +4340,7 @@ void dc_enable_dmub_outbox(struct dc *dc)
 	struct dc_context *dc_ctx = dc->ctx;
 
 	dmub_enable_outbox_notification(dc_ctx->dmub_srv);
+	DC_LOG_DC("%s: dmub outbox notifications enabled\n", __func__);
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index a0af0f6afeef..9544abf75e84 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -344,6 +344,7 @@ enum dc_detect_reason {
 	DETECT_REASON_HPDRX,
 	DETECT_REASON_FALLBACK,
 	DETECT_REASON_RETRAIN,
+	DETECT_REASON_TDR,
 };
 
 bool dc_link_detect(struct dc_link *dc_link, enum dc_detect_reason reason);
-- 
2.25.1

[PATCH 32/32] drm/amd/display: 3.2.197

[Tom Chung]

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

This version brings along following fixes:

- Move fpu file to dml folder
- Fix CAB allocation for multiple displays
- Add a variable to update FCLK
- Fix TMDS 4K@60Hz YCbCr420 corruption issue
- Fix MPC tree infinite loop in some case
- Fix SubVP calculations
- Fix HDMI VSIF V3 blank screen issue
- Add some documentation to the code
- Allow alternate prefetch modes in DML for DCN32
- Fix VPG instancing for dcn314 HPO
- Check correct bounds for stream encoder instances for DCN303
- Enable SubVP by default on DCN32 & DCN321
- Fix DTBCLK not correct for dcn314
- Fix Compile-time warning
- Fix the stereo mode hang issue
- Fix display light up issue

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 86fe4b9f10c2..792fd57c1615 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -47,7 +47,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.196"
+#define DC_VER "3.2.197"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
2.25.1

RE: [PATCH 00/32] DC Patches July 29 2022

[Wheeler, Daniel]

[AMD Official Use Only - General]

Hi all,
 
This week this patchset was tested on the following systems:
 
HP Envy 360, with Ryzen 5 4500U
Lenovo Thinkpad T14s Gen2, with AMD Ryzen 5 5650U 
Sapphire Pulse RX5700XT 
Reference AMD RX6800
Engineering board with Ryzen 9 5900H
 
These systems were tested on the following display types: 
eDP, (1080p 60hz [4500U, 5650U, 5900H])
VGA and DVI (1680x1050 60HZ [DP to VGA/DVI, USB-C to DVI/VGA])
DP/HDMI/USB-C (1440p 170hz, 4k 60hz, 4k 144hz [Includes USB-C to DP/HDMI adapters])
 
MST tested with Startech MST14DP123DP and 2x 4k 60Hz displays
DSC tested with Cable Matters 101075 (DP to 3x DP), and 201375 (USB-C to 3x DP) with 3x 4k60 displays
 
The testing is a mix of automated and manual tests. Manual testing includes (but is not limited to):
Changing display configurations and settings
Benchmark testing
Feature testing (Freesync, etc.)
 
Automated testing includes (but is not limited to):
Script testing (scripts to automate some of the manual checks)
IGT testing
 
The patchset consists of the amd-staging-drm-next branch (Head commit - 83ace616212127dcf707897e39196f45f5bede29 -> drm/amd/display: 3.2.196) with new patches added on top of it. This branch is used for both Ubuntu and Chrome OS testing (ChromeOS on a bi-weekly basis).

 
Tested on Ubuntu 22.04 and Chrome OS
 
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
 
 
Thank you,
 
Dan Wheeler
Sr. Technologist | AMD
SW Display
------------------------------------------------------------------------------------------------------------------
1 Commerce Valley Dr E, Thornhill, ON L3T 7X6
amd.com

-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Tom Chung
Sent: August 5, 2022 1:58 PM
To: amd-gfx@lists.freedesktop.org
Cc: Wang, Chao-kai (Stylon) <Stylon.Wang@amd.com>; Chung, ChiaHsuan (Tom) <ChiaHsuan.Chung@amd.com>; Li, Sun peng (Leo) <Sunpeng.Li@amd.com>; Wentland, Harry <Harry.Wentland@amd.com>; Zhuo, Qingqing (Lillian) <Qingqing.Zhuo@amd.com>; Siqueira, Rodrigo <Rodrigo.Siqueira@amd.com>; Li, Roman <Roman.Li@amd.com>; Chiu, Solomon <Solomon.Chiu@amd.com>; Pillai, Aurabindo <Aurabindo.Pillai@amd.com>; Lin, Wayne <Wayne.Lin@amd.com>; Lakha, Bhawanpreet <Bhawanpreet.Lakha@amd.com>; Gutierrez, Agustin <Agustin.Gutierrez@amd.com>; Kotarac, Pavle <Pavle.Kotarac@amd.com>
Subject: [PATCH 00/32] DC Patches July 29 2022

This DC patchset brings along following fixes:

- Move fpu file to dml folder
- Fix CAB allocation for multiple displays
- Add a variable to update FCLK
- Fix TMDS 4K@60Hz YCbCr420 corruption issue
- Fix MPC tree infinite loop in some case
- Fix SubVP calculations
- Fix HDMI VSIF V3 blank screen issue
- Add some documentation to the code
- Allow alternate prefetch modes in DML for DCN32
- Fix VPG instancing for dcn314 HPO
- Check correct bounds for stream encoder instances for DCN303
- Enable SubVP by default on DCN32 & DCN321
- Fix DTBCLK not correct for dcn314
- Fix Compile-time warning
- Fix the stereo mode hang issue
- Fix display light up issue

Alvin Lee (8):
  drm/amd/display: Revert "fix 32 bit compilation errors in
    dc_dmub_srv.c"
  drm/amd/display: Add a variable to update FCLK latency
  drm/amd/display: Refactor SubVP calculation to remove FPU
  drm/amd/display: Add 16 lines margin for SubVP
  drm/amd/display: Fix subvp calculations
  drm/amd/display: Revert "attempt to fix the logic in
    commit_planes_for_stream()"
  drm/amd/display: For stereo keep "FLIP_ANY_FRAME"
  drm/amd/display: Don't try to enter MALL SS if stereo3d

Aric Cyr (1):
  drm/amd/display: 3.2.197

Aurabindo Pillai (3):
  drm/amd/display: fix CAB allocation for multiple displays
  drm/amd/display: Check correct bounds for stream encoder instances for
    DCN303
  drm/amd/display: Enable SubVP by default on DCN32 & DCN321

Chiawen Huang (1):
  drm/amd/display: Device flash garbage before get in OS

Daniel Miess (2):
  drm/amd/display: Use pixels per container logic for DCN314 DCCG
    dividers
  drm/amd/display: Fix TMDS 4K@60Hz YCbCr420 corruption issue

David Galiffi (2):
  drm/amd/display: Allow alternate prefetch modes in DML for DCN32
  drm/amd/display: Fix Compile-time Warning

Duncan Ma (2):
  drm/amd/display: Fix VPG instancing for dcn314 HPO
  drm/amd/display: Correct DTBCLK for dcn314

Fudong Wang (1):
  drm/amd/display: clear optc underflow before turn off odm clock

Josip Pavic (1):
  drm/amd/display: Avoid MPC infinite loop

Leo Ma (1):
  drm/amd/display: Fix HDMI VSIF V3 incorrect issue

Meenakshikumar Somasundaram (1):
  drm/amd/display: Fix TDR eDP and USB4 display light up issue

Rodrigo Siqueira (9):
  drm/amd/display: Create FPU files for DCN314
  drm/amd/display: Move populate dml pipes from DCN314 to dml
  drm/amd/display: Drop FPU flags from Makefile
  drm/amd/display: Add documentation to some of the cursor struct
  drm/amd/display: Add basic kernel doc to CRC code under DC
  drm/amd/display: Add some extra kernel doc to amdgpu_dm
  drm/amd/display: Document pipe split policy
  drm/amd/display: Expand documentation for timing
  drm/amd/display: Document some of the DML structs

 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |  16 +-  .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  46 ++-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.c        |  12 +-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.h        |   2 +-
 drivers/gpu/drm/amd/display/dc/core/dc.c      |  45 ++-
 drivers/gpu/drm/amd/display/dc/dc.h           |  40 +-
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c  |  93 +++--  drivers/gpu/drm/amd/display/dc/dc_hw_types.h  | 148 ++++++-
 drivers/gpu/drm/amd/display/dc/dc_link.h      |   1 +
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.c |   1 +
 .../gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c  |   6 +
 .../gpu/drm/amd/display/dc/dcn10/dcn10_optc.c |  44 +-
 .../gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c  |   6 +
 .../gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c |   2 +-
 .../amd/display/dc/dcn303/dcn303_resource.c   |   2 +-
 .../gpu/drm/amd/display/dc/dcn314/Makefile    |  25 --
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  |  42 +-
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.h  |   4 +
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |   2 +
 .../amd/display/dc/dcn314/dcn314_resource.c   | 341 +---------------
 .../amd/display/dc/dcn314/dcn314_resource.h   |   3 +
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.c    |  39 +-
 .../gpu/drm/amd/display/dc/dcn32/dcn32_optc.c |   2 +-
 .../drm/amd/display/dc/dcn32/dcn32_resource.c |   3 +-
 .../amd/display/dc/dcn321/dcn321_resource.c   |   3 +-
 drivers/gpu/drm/amd/display/dc/dml/Makefile   |   2 +
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 376 ++++++++++++++++++
 .../amd/display/dc/dml/dcn314/dcn314_fpu.h    |  40 ++
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  |  46 ++-
 .../amd/display/dc/dml/dcn321/dcn321_fpu.c    |   7 +
 .../amd/display/dc/dml/display_mode_structs.h |  34 ++
 .../gpu/drm/amd/display/dc/inc/core_types.h   |   4 +
 .../amd/display/dc/inc/hw/timing_generator.h  |  14 +-
 .../amd/display/modules/freesync/freesync.c   |  15 +-
 34 files changed, 968 insertions(+), 498 deletions(-)  create mode 100644 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
 create mode 100644 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h

--
2.25.1