drm/amd/display: Add misc DC changes for DCN401

Add miscellaneous changes to enable DCN401 init

Signed-off-by: Aurabindo Pillai <[e-mail]>
Acked-by: Rodrigo Siqueira <[e-mail]>
Signed-off-by: Alex Deucher <[e-mail]>

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

···
 #define FIRMWARE_DCN_351_DMUB "amdgpu/dcn_3_5_1_dmcub.bin"
 MODULE_FIRMWARE(FIRMWARE_DCN_351_DMUB);
 
+#define FIRMWARE_DCN_401_DMUB "amdgpu/dcn_4_0_1_dmcub.bin"
+MODULE_FIRMWARE(FIRMWARE_DCN_401_DMUB);
+
 /* Number of bytes in PSP header for firmware. */
 #define PSP_HEADER_BYTES 0x100
 
···
 	case IP_VERSION(3, 1, 4):
 	case IP_VERSION(3, 5, 0):
 	case IP_VERSION(3, 5, 1):
+	case IP_VERSION(4, 0, 1):
 		hw_params.dpia_supported = true;
 		hw_params.disable_dpia = adev->dm.dc->debug.dpia_debug.bits.disable_dpia;
 		break;
···
 	if (amdgpu_dc_debug_mask & DC_FORCE_SUBVP_MCLK_SWITCH)
 		adev->dm.dc->debug.force_subvp_mclk_switch = true;
 
-	if (amdgpu_dc_debug_mask & DC_ENABLE_DML2)
+	if (amdgpu_dc_debug_mask & DC_ENABLE_DML2) {
 		adev->dm.dc->debug.using_dml2 = true;
+		adev->dm.dc->debug.using_dml21 = true;
+	}
 
 	adev->dm.dc->debug.visual_confirm = amdgpu_dc_visual_confirm;
 
···
 		case IP_VERSION(3, 2, 1):
 		case IP_VERSION(3, 5, 0):
 		case IP_VERSION(3, 5, 1):
+		case IP_VERSION(4, 0, 1):
 			return 0;
 		default:
 			break;
···
 	case IP_VERSION(3, 5, 1):
 		dmub_asic = DMUB_ASIC_DCN35;
 		break;
+	case IP_VERSION(4, 0, 1):
+		dmub_asic = DMUB_ASIC_DCN401;
+		break;
+
 	default:
 		/* ASIC doesn't support DMUB. */
 		return 0;
···
 	case IP_VERSION(2, 1, 0):
 	case IP_VERSION(3, 5, 0):
 	case IP_VERSION(3, 5, 1):
+	case IP_VERSION(4, 0, 1):
 		if (register_outbox_irq_handlers(dm->adev)) {
 			DRM_ERROR("DM: Failed to initialize IRQ\n");
 			goto fail;
···
 		case IP_VERSION(3, 2, 1):
 		case IP_VERSION(3, 5, 0):
 		case IP_VERSION(3, 5, 1):
+		case IP_VERSION(4, 0, 1):
 			psr_feature_enabled = true;
 			break;
 		default:
···
 		case IP_VERSION(3, 2, 1):
 		case IP_VERSION(3, 5, 0):
 		case IP_VERSION(3, 5, 1):
+		case IP_VERSION(4, 0, 1):
 			if (dcn10_register_irq_handlers(dm->adev)) {
 				DRM_ERROR("DM: Failed to initialize IRQ\n");
 				goto fail;
···
 	case IP_VERSION(3, 5, 1):
 		fw_name_dmub = FIRMWARE_DCN_351_DMUB;
 		break;
+	case IP_VERSION(4, 0, 1):
+		fw_name_dmub = FIRMWARE_DCN_401_DMUB;
+		break;
 	default:
 		/* ASIC doesn't support DMUB. */
 		return 0;
···
 		case IP_VERSION(3, 2, 1):
 		case IP_VERSION(3, 5, 0):
 		case IP_VERSION(3, 5, 1):
+		case IP_VERSION(4, 0, 1):
 			adev->mode_info.num_crtc = 4;
 			adev->mode_info.num_hpd = 4;
 			adev->mode_info.num_dig = 4;
···
 		return true;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 static void update_dsc_caps(struct amdgpu_dm_connector *aconnector,
 			    struct dc_sink *sink, struct dc_stream_state *stream,
 			    struct dsc_dec_dpcd_caps *dsc_caps)
···
 				dsc_caps);
 	}
 }
-
 
 static void apply_dsc_policy_for_edp(struct amdgpu_dm_connector *aconnector,
 				    struct dc_sink *sink, struct dc_stream_state *stream,
···
 		stream->timing.flags.DSC = 1;
 	}
 }
-
 
 static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector,
 					struct dc_sink *sink, struct dc_stream_state *stream,
···
 	if (stream->timing.flags.DSC && aconnector->dsc_settings.dsc_bits_per_pixel)
 		stream->timing.dsc_cfg.bits_per_pixel = aconnector->dsc_settings.dsc_bits_per_pixel;
 }
+#endif
 
 static struct dc_stream_state *
 create_stream_for_sink(struct drm_connector *connector,
···
 	int mode_refresh;
 	int preferred_refresh = 0;
 	enum color_transfer_func tf = TRANSFER_FUNC_UNKNOWN;
+#if defined(CONFIG_DRM_AMD_DC_FP)
 	struct dsc_dec_dpcd_caps dsc_caps;
-
+#endif
 	struct dc_link *link = NULL;
 	struct dc_sink *sink = NULL;
 
···
 		stream->timing = *aconnector->timing_requested;
 	}
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 	/* SST DSC determination policy */
 	update_dsc_caps(aconnector, sink, stream, &dsc_caps);
 	if (aconnector->dsc_settings.dsc_force_enable != DSC_CLK_FORCE_DISABLE && dsc_caps.is_dsc_supported)
 		apply_dsc_policy_for_stream(aconnector, sink, stream, &dsc_caps);
+#endif
 
 	update_stream_scaling_settings(&mode, dm_state, stream);
 
···
 		}
 	}
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 	if (dc_resource_is_dsc_encoding_supported(dc)) {
 		ret = pre_validate_dsc(state, &dm_state, vars);
 		if (ret != 0)
 			goto fail;
 	}
+#endif
 
 	/* Run this here since we want to validate the streams we created */
 	ret = drm_atomic_helper_check_planes(dev, state);
···
 			goto fail;
 		}
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 		if (dc_resource_is_dsc_encoding_supported(dc)) {
 			ret = compute_mst_dsc_configs_for_state(state, dm_state->context, vars);
 			if (ret) {
···
 				goto fail;
 			}
 		}
+#endif
 
 		ret = dm_update_mst_vcpi_slots_for_dsc(state, dm_state->context, vars);
 		if (ret) {

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c

···
 	return false;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 static bool is_synaptics_cascaded_panamera(struct dc_link *link, struct drm_dp_mst_port *port)
 {
 	u8 branch_vendor_data[4] = { 0 }; // Vendor data 0x50C ~ 0x50F
···
 
 	return true;
 }
+#endif
 
 static bool retrieve_downstream_port_device(struct amdgpu_dm_connector *aconnector)
 {
···
 			amdgpu_dm_update_freesync_caps(
 					connector, aconnector->edid);
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 			if (!validate_dsc_caps_on_connector(aconnector))
 				memset(&aconnector->dc_sink->dsc_caps,
 				       0, sizeof(aconnector->dc_sink->dsc_caps));
+#endif
 
 			if (!retrieve_downstream_port_device(aconnector))
 				memset(&aconnector->mst_downstream_port_present,
···
 	struct amdgpu_dm_connector *aconnector;
 };
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 static int kbps_to_peak_pbn(int kbps)
 {
 	u64 peak_kbps = kbps;
···
 
 	return bw_range->max_target_bpp_x16 && bw_range->min_target_bpp_x16;
 }
+#endif
 
 enum dc_status dm_dp_mst_is_port_support_mode(
 	struct amdgpu_dm_connector *aconnector,
 	struct dc_stream_state *stream)
 {
-	int pbn, branch_max_throughput_mps = 0;
+	int branch_max_throughput_mps = 0;
+#if defined(CONFIG_DRM_AMD_DC_FP)
 	struct dc_link_settings cur_link_settings;
+	int pbn;
 	unsigned int end_to_end_bw_in_kbps = 0;
 	unsigned int upper_link_bw_in_kbps = 0, down_link_bw_in_kbps = 0;
 	struct dc_dsc_bw_range bw_range = {0};
···
 			return DC_FAIL_BANDWIDTH_VALIDATE;
 		}
 	}
-
 	/* check is mst dsc output bandwidth branch_overall_throughput_0_mps */
 	switch (stream->timing.pixel_encoding) {
 	case PIXEL_ENCODING_RGB:
···
 	default:
 		break;
 	}
+#endif
 
 	if (branch_max_throughput_mps != 0 &&
 		((stream->timing.pix_clk_100hz / 10) >  branch_max_throughput_mps * 1000))

drivers/gpu/drm/amd/display/dc/Makefile

···
 DC_LIBS += dcn32
 DC_LIBS += dcn321
 DC_LIBS += dcn35
+DC_LIBS += dcn401
 DC_LIBS += dml
 DC_LIBS += dml2
 endif
···
 
 DC_LIBS += hdcp
 
+ifdef CONFIG_DRM_AMD_DC_FP
+DC_LIBS += spl
+DC_SPL_TRANS += dc_spl_translate.o
+endif
+
 AMD_DC = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/dc/,$(DC_LIBS)))
 
 include $(AMD_DC)
···
 
 AMD_DM_REG_UPDATE = $(addprefix $(AMDDALPATH)/dc/,dc_helper.o)
 
+AMD_DC_SPL_TRANS = $(addprefix $(AMDDALPATH)/dc/,$(DC_SPL_TRANS))
+
 AMD_DISPLAY_FILES += $(AMD_DISPLAY_CORE)
 AMD_DISPLAY_FILES += $(AMD_DM_REG_UPDATE)
 
···
 AMD_DISPLAY_EDID = $(addprefix $(AMDDALPATH)/dc/,$(DC_EDID))
 AMD_DISPLAY_FILES += $(AMD_DISPLAY_DMUB) $(AMD_DISPLAY_EDID)
 
+AMD_DISPLAY_FILES += $(AMD_DC_SPL_TRANS)

drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c

···
 	struct bios_parser *bp,
 	struct dc_firmware_info *info);
 
+static enum bp_result get_firmware_info_v3_5(
+	struct bios_parser *bp,
+	struct dc_firmware_info *info);
+
 static struct atom_hpd_int_record *get_hpd_record(struct bios_parser *bp,
 		struct atom_display_object_path_v2 *object);
 
···
 			case 4:
 				result = get_firmware_info_v3_4(bp, info);
 				break;
+			case 5:
+				result = get_firmware_info_v3_5(bp, info);
+				break;
 			default:
 				break;
 			}
···
 	} else {
 		info->oem_i2c_present = false;
 	}
+
+	return BP_RESULT_OK;
+}
+
+static enum bp_result get_firmware_info_v3_5(
+	struct bios_parser *bp,
+	struct dc_firmware_info *info)
+{
+	struct atom_firmware_info_v3_5 *firmware_info;
+	struct atom_common_table_header *header;
+	struct atom_data_revision revision;
+	struct atom_display_controller_info_v4_5 *dce_info_v4_5 = NULL;
+
+	if (!info)
+		return BP_RESULT_BADINPUT;
+
+	firmware_info = GET_IMAGE(struct atom_firmware_info_v3_5,
+			DATA_TABLES(firmwareinfo));
+
+	if (!firmware_info)
+		return BP_RESULT_BADBIOSTABLE;
+
+	memset(info, 0, sizeof(*info));
+
+	if (firmware_info->board_i2c_feature_id == 0x2) {
+		info->oem_i2c_present = true;
+		info->oem_i2c_obj_id = firmware_info->board_i2c_feature_gpio_id;
+	} else {
+		info->oem_i2c_present = false;
+	}
+
+	header = GET_IMAGE(struct atom_common_table_header,
+					DATA_TABLES(dce_info));
+
+	get_atom_data_table_revision(header, &revision);
+
+	switch (revision.major) {
+	case 4:
+		switch (revision.minor) {
+		case 5:
+			dce_info_v4_5 = GET_IMAGE(struct atom_display_controller_info_v4_5,
+							DATA_TABLES(dce_info));
+
+			if (!dce_info_v4_5)
+				return BP_RESULT_BADBIOSTABLE;
+
+			 /* 100MHz expected */
+			info->pll_info.crystal_frequency = dce_info_v4_5->dce_refclk_10khz * 10;
+			break;
+		default:
+			break;
+		}
+		break;
+	default:
+		break;
+	}
+
 
 	return BP_RESULT_OK;
 }
···
 
 	info->num_chans = info_v30->channel_num;
 	info->dram_channel_width_bytes = (1 << info_v30->channel_width) / 8;
+
+	return result;
+}
+
+static enum bp_result get_vram_info_from_umc_info_v40(
+		struct bios_parser *bp,
+		struct dc_vram_info *info)
+{
+	struct atom_umc_info_v4_0 *info_v40;
+	enum bp_result result = BP_RESULT_OK;
+
+	info_v40 = GET_IMAGE(struct atom_umc_info_v4_0,
+						DATA_TABLES(umc_info));
+
+	if (info_v40 == NULL)
+		return BP_RESULT_BADBIOSTABLE;
+
+	info->num_chans = info_v40->channel_num;
+	info->dram_channel_width_bytes = (1 << info_v40->channel_width) / 8;
 
 	return result;
 }
···
 	struct atom_common_table_header *header;
 	struct atom_data_revision revision;
 
-	if (info && DATA_TABLES(vram_info)) {
+	// vram info moved to umc_info for DCN4x
+	if (info && DATA_TABLES(umc_info)) {
+		header = GET_IMAGE(struct atom_common_table_header,
+					DATA_TABLES(umc_info));
+
+		get_atom_data_table_revision(header, &revision);
+
+		switch (revision.major) {
+		case 4:
+			switch (revision.minor) {
+			case 0:
+				result = get_vram_info_from_umc_info_v40(bp, info);
+				break;
+			default:
+				break;
+			}
+			break;
+		default:
+			break;
+		}
+	}
+
+	if (result != BP_RESULT_OK && info && DATA_TABLES(vram_info)) {
 		header = GET_IMAGE(struct atom_common_table_header,
 					DATA_TABLES(vram_info));
 
···
 	bp->base.integrated_info = bios_parser_create_integrated_info(&bp->base);
 	bp->base.fw_info_valid = bios_parser_get_firmware_info(&bp->base, &bp->base.fw_info) == BP_RESULT_OK;
 	bios_parser_get_vram_info(&bp->base, &bp->base.vram_info);
-
+	bios_parser_get_soc_bb_info(&bp->base, &bp->base.bb_info);
 	return true;
 }
 

drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c

···
 	case DCN_VERSION_3_21:
 	case DCN_VERSION_3_5:
 	case DCN_VERSION_3_51:
+	case DCN_VERSION_4_01:
 		*h = dal_cmd_tbl_helper_dce112_get_table2();
 		return true;
 

drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile

···
 AMD_DAL_CLK_MGR_DCN35 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn35/,$(CLK_MGR_DCN35))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN35)
+
+###############################################################################
+# DCN401
+###############################################################################
+CLK_MGR_DCN401 = dcn401_clk_mgr.o dcn401_clk_mgr_smu_msg.o
+
+AMD_DAL_CLK_MGR_DCN401 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn401/,$(CLK_MGR_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN401)
 endif

drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c

···
 #include "dcn316/dcn316_clk_mgr.h"
 #include "dcn32/dcn32_clk_mgr.h"
 #include "dcn35/dcn35_clk_mgr.h"
+#include "dcn401/dcn401_clk_mgr.h"
 
 int clk_mgr_helper_get_active_display_cnt(
 		struct dc *dc,
···
 	}
 	break;
 
+	case AMDGPU_FAMILY_GC_12_0_0: {
+		struct clk_mgr_internal *clk_mgr = kzalloc(sizeof(*clk_mgr), GFP_KERNEL);
+
+		if (clk_mgr == NULL) {
+			BREAK_TO_DEBUGGER();
+			return NULL;
+		}
+
+		dcn401_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg);
+		return &clk_mgr->base;
+	}
+	break;
 #endif	/* CONFIG_DRM_AMD_DC_FP */
 	default:
 		ASSERT(0); /* Unknown Asic */
···
 
 	case AMDGPU_FAMILY_GC_11_5_0:
 		dcn35_clk_mgr_destroy(clk_mgr);
+		break;
+	case AMDGPU_FAMILY_GC_12_0_0:
+		dcn401_clk_mgr_destroy(clk_mgr);
 		break;
 
 	default:

drivers/gpu/drm/amd/display/dc/core/dc.c

···
 
 #include "hw_sequencer_private.h"
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 #include "dml2/dml2_internal_types.h"
+#endif
 
 #include "dce/dmub_outbox.h"
 
···
 				get_subvp_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color));
 			else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MCLK_SWITCH)
 				get_mclk_switch_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color));
+			else if (dc->debug.visual_confirm == VISUAL_CONFIRM_FAMS2)
+				get_fams2_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color));
 		}
 	}
 }
···
 	dc->build_id = DC_BUILD_ID;
 
 	DC_LOG_DC("Display Core initialized\n");
-
-
 
 	return dc;
 
···
 	 */
 	if (dc->hwss.subvp_pipe_control_lock)
 		dc->hwss.subvp_pipe_control_lock(dc, context, true, true, NULL, subvp_prev_use);
+	if (dc->hwss.fams2_global_control_lock)
+		dc->hwss.fams2_global_control_lock(dc, context, true);
 
 	if (dc->hwss.update_dsc_pg)
 		dc->hwss.update_dsc_pg(dc, context, false);
···
 		dc->hwss.commit_subvp_config(dc, context);
 	if (dc->hwss.subvp_pipe_control_lock)
 		dc->hwss.subvp_pipe_control_lock(dc, context, false, true, NULL, subvp_prev_use);
+	if (dc->hwss.fams2_global_control_lock)
+		dc->hwss.fams2_global_control_lock(dc, context, false);
 
 	for (i = 0; i < context->stream_count; i++) {
 		const struct dc_link *link = context->streams[i]->link;
···
 			elevate_update_type(&overall_type, UPDATE_TYPE_MED);
 		}
 
+	if (u->cm2_params) {
+		if ((u->cm2_params->component_settings.shaper_3dlut_setting
+					!= u->surface->mcm_shaper_3dlut_setting)
+				|| (u->cm2_params->component_settings.lut1d_enable
+					!= u->surface->mcm_lut1d_enable))
+			update_flags->bits.mcm_transfer_function_enable_change = 1;
+		if (u->cm2_params->cm2_luts.lut3d_data.lut3d_src
+				!= u->surface->mcm_luts.lut3d_data.lut3d_src)
+			update_flags->bits.mcm_transfer_function_enable_change = 1;
+	}
 	if (update_flags->bits.in_transfer_func_change) {
 		type = UPDATE_TYPE_MED;
 		elevate_update_type(&overall_type, type);
 	}
 
 	if (update_flags->bits.lut_3d) {
+		type = UPDATE_TYPE_FULL;
+		elevate_update_type(&overall_type, type);
+	}
+	if (update_flags->bits.mcm_transfer_function_enable_change) {
 		type = UPDATE_TYPE_FULL;
 		elevate_update_type(&overall_type, type);
 	}
···
 	if (srf_update->gamut_remap_matrix)
 		surface->gamut_remap_matrix =
 			*srf_update->gamut_remap_matrix;
+	if (srf_update->cm2_params) {
+		surface->mcm_shaper_3dlut_setting = srf_update->cm2_params->component_settings.shaper_3dlut_setting;
+		surface->mcm_lut1d_enable = srf_update->cm2_params->component_settings.lut1d_enable;
+		surface->mcm_luts = srf_update->cm2_params->cm2_luts;
+	}
+	if (srf_update->cursor_csc_color_matrix)
+		surface->cursor_csc_color_matrix =
+			*srf_update->cursor_csc_color_matrix;
 }
 
 static void copy_stream_update_to_stream(struct dc *dc,
···
 	}
 }
 
+static bool check_address_only_update(union surface_update_flags update_flags)
+{
+	union surface_update_flags addr_only_update_flags;
+	addr_only_update_flags.raw = 0;
+	addr_only_update_flags.bits.addr_update = 1;
+
+	return update_flags.bits.addr_update &&
+			!(update_flags.raw & ~addr_only_update_flags.raw);
+}
 
 /**
  * build_dmub_cmd_list() - Build an array of DMCUB commands to be sent to DMCUB
···
 	build_dmub_update_dirty_rect(dc, surface_count, stream, srf_updates, context, dc_dmub_cmd, dmub_cmd_count);
 }
 
+static void commit_plane_for_stream_offload_fams2_flip(struct dc *dc,
+		struct dc_surface_update *srf_updates,
+		int surface_count,
+		struct dc_stream_state *stream,
+		struct dc_state *context)
+{
+	int i, j;
+
+	/* update dirty rect for PSR */
+	dc_dmub_update_dirty_rect(dc, surface_count, stream,
+			srf_updates, context);
+
+	/* Perform requested Updates */
+	for (i = 0; i < surface_count; i++) {
+		struct dc_plane_state *plane_state = srf_updates[i].surface;
+
+		/* set offload flag so driver does not program address */
+		plane_state->address.offload_flip = true;
+
+		for (j = 0; j < dc->res_pool->pipe_count; j++) {
+			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+
+			if (!should_update_pipe_for_stream(context, pipe_ctx, stream))
+				continue;
+
+			if (!should_update_pipe_for_plane(context, pipe_ctx, plane_state))
+				continue;
+
+			/* update pipe context for plane */
+			if (pipe_ctx->plane_state->update_flags.bits.addr_update)
+				dc->hwss.update_plane_addr(dc, pipe_ctx);
+		}
+	}
+
+	/* Send commands to DMCUB */
+	dc_dmub_srv_fams2_passthrough_flip(dc,
+				context,
+				stream,
+				srf_updates,
+				surface_count);
+
+	/* reset offload flip flag */
+	for (i = 0; i < surface_count; i++) {
+		struct dc_plane_state *plane_state = srf_updates[i].surface;
+		plane_state->address.offload_flip = false;
+	}
+}
+
 static void commit_planes_for_stream_fast(struct dc *dc,
 		struct dc_surface_update *srf_updates,
 		int surface_count,
···
 	int i, j;
 	struct pipe_ctx *top_pipe_to_program = NULL;
 	struct dc_stream_status *stream_status = NULL;
+	bool should_offload_fams2_flip = false;
+
+	if (dc->debug.fams2_config.bits.enable &&
+			dc->debug.fams2_config.bits.enable_offload_flip &&
+			dc_state_is_fams2_in_use(dc, context)) {
+		/* if not offloading to HWFQ, offload to FAMS2 if needed */
+		should_offload_fams2_flip = true;
+		for (i = 0; i < surface_count; i++) {
+			if (srf_updates[i].surface &&
+					srf_updates[i].surface->update_flags.raw &&
+					!check_address_only_update(srf_updates[i].surface->update_flags)) {
+				/* more than address update, need to acquire FAMS2 lock */
+				should_offload_fams2_flip = false;
+				break;
+			}
+		}
+	}
 
 	dc_exit_ips_for_hw_access(dc);
 
···
 				continue;
 			pipe_ctx->plane_state->triplebuffer_flips = false;
 			if (update_type == UPDATE_TYPE_FAST &&
-			    dc->hwss.program_triplebuffer &&
+			    dc->hwss.program_triplebuffer != NULL &&
 			    !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) {
 				/*triple buffer for VUpdate  only*/
 				pipe_ctx->plane_state->triplebuffer_flips = true;
···
 	}
 
 	stream_status = dc_state_get_stream_status(context, stream);
+
+	if (should_offload_fams2_flip) {
+		commit_plane_for_stream_offload_fams2_flip(dc,
+				srf_updates,
+				surface_count,
+				stream,
+				context);
+	} else {
+		build_dmub_cmd_list(dc,
+				srf_updates,
+				surface_count,
+				stream,
+				context,
+				context->dc_dmub_cmd,
+				&(context->dmub_cmd_count));
+		hwss_build_fast_sequence(dc,
+				context->dc_dmub_cmd,
+				context->dmub_cmd_count,
+				context->block_sequence,
+				&(context->block_sequence_steps),
+				top_pipe_to_program,
+				stream_status,
+				context);
+		hwss_execute_sequence(dc,
+				context->block_sequence,
+				context->block_sequence_steps);
+	}
 
 	build_dmub_cmd_list(dc,
 			srf_updates,
···
 
 	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, true, should_lock_all_pipes, NULL, subvp_prev_use);
-		dc->hwss.interdependent_update_lock(dc, context, true);
+			dc->hwss.subvp_pipe_control_lock(dc, context, true, should_lock_all_pipes, NULL, subvp_prev_use);
 
+		if (dc->hwss.fams2_global_control_lock)
+			dc->hwss.fams2_global_control_lock(dc, context, true);
+
+		dc->hwss.interdependent_update_lock(dc, context, true);
 	} else {
 		if (dc->hwss.subvp_pipe_control_lock)
 			dc->hwss.subvp_pipe_control_lock(dc, context, true, should_lock_all_pipes, top_pipe_to_program, subvp_prev_use);
+
+		if (dc->hwss.fams2_global_control_lock)
+			dc->hwss.fams2_global_control_lock(dc, context, true);
+
 		/* Lock the top pipe while updating plane addrs, since freesync requires
 		 *  plane addr update event triggers to be synchronized.
 		 *  top_pipe_to_program is expected to never be NULL
···
 		if (dc->hwss.subvp_pipe_control_lock)
 			dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes,
 							 NULL, subvp_prev_use);
+
+		if (dc->hwss.fams2_global_control_lock)
+			dc->hwss.fams2_global_control_lock(dc, context, false);
+
 		return;
 	}
 
···
 	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);
+		if (dc->hwss.fams2_global_control_lock)
+			dc->hwss.fams2_global_control_lock(dc, context, false);
 	} 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);
+		if (dc->hwss.fams2_global_control_lock)
+			dc->hwss.fams2_global_control_lock(dc, context, false);
 	}
 
 	// Fire manual trigger only when bottom plane is flipped
···
 		fast_update[i].gamut_remap_matrix = srf_updates[i].gamut_remap_matrix;
 		fast_update[i].input_csc_color_matrix = srf_updates[i].input_csc_color_matrix;
 		fast_update[i].coeff_reduction_factor = srf_updates[i].coeff_reduction_factor;
+		fast_update[i].cursor_csc_color_matrix = srf_updates[i].cursor_csc_color_matrix;
 	}
 }
 
···
 				fast_update[i].gamma ||
 				fast_update[i].gamut_remap_matrix ||
 				fast_update[i].input_csc_color_matrix ||
+				fast_update[i].cursor_csc_color_matrix ||
 				fast_update[i].coeff_reduction_factor)
 			return true;
 	}
···
 				srf_updates[i].surface->force_full_update ||
 				(srf_updates[i].flip_addr &&
 				srf_updates[i].flip_addr->address.tmz_surface != srf_updates[i].surface->address.tmz_surface) ||
+				(srf_updates[i].cm2_params &&
+				 (srf_updates[i].cm2_params->component_settings.shaper_3dlut_setting != srf_updates[i].surface->mcm_shaper_3dlut_setting ||
+				  srf_updates[i].cm2_params->component_settings.lut1d_enable != srf_updates[i].surface->mcm_lut1d_enable)) ||
 				!is_surface_in_context(context, srf_updates[i].surface)))
 			return true;
 	}
···
 	 * specially handle compatibility problems with transitions among those
 	 * features as they are now transparent to the new sequence.
 	 */
-	if (dc->ctx->dce_version > DCN_VERSION_3_51)
+	if (dc->ctx->dce_version > DCN_VERSION_4_01)
 		return update_planes_and_stream_v3(dc, srf_updates,
 				surface_count, stream, stream_update);
 	return update_planes_and_stream_v2(dc, srf_updates,
···
 	 * we get more confident about this change we'll need to enable
 	 * the new sequence for all ASICs.
 	 */
-	if (dc->ctx->dce_version > DCN_VERSION_3_51) {
+	if (dc->ctx->dce_version > DCN_VERSION_4_01) {
 		update_planes_and_stream_v3(dc, srf_updates, surface_count,
 				stream, stream_update);
 		return;
···
 	return profile;
 }
 
+/* Need to account for padding due to pixel-to-symbol packing
+ * for uncompressed 128b/132b streams.
+ */
+static uint32_t apply_128b_132b_stream_overhead(
+	const struct dc_crtc_timing *timing, const uint32_t kbps)
+{
+	uint32_t total_kbps = kbps;
+#if defined(CONFIG_DRM_AMD_DC_FP)
+	if (dc_get_disable_128b_132b_stream_overhead())
+		return kbps;
+#endif
+
+	if (!timing->flags.DSC) {
+		struct fixed31_32 bpp;
+		struct fixed31_32 overhead_factor;
+
+		bpp = dc_fixpt_from_int(kbps);
+		bpp = dc_fixpt_div_int(bpp, timing->pix_clk_100hz / 10);
+
+		/* Symbols_per_HActive = HActive * bpp / (4 lanes * 32-bit symbol size)
+		 * Overhead_factor = ceil(Symbols_per_HActive) / Symbols_per_HActive
+		 */
+		overhead_factor = dc_fixpt_from_int(timing->h_addressable);
+		overhead_factor = dc_fixpt_mul(overhead_factor, bpp);
+		overhead_factor = dc_fixpt_div_int(overhead_factor, 128);
+		overhead_factor = dc_fixpt_div(
+			dc_fixpt_from_int(dc_fixpt_ceil(overhead_factor)),
+			overhead_factor);
+
+		total_kbps = dc_fixpt_ceil(
+			dc_fixpt_mul_int(overhead_factor, total_kbps));
+	}
+
+	return total_kbps;
+}
+
+uint32_t dc_bandwidth_in_kbps_from_timing(
+	const struct dc_crtc_timing *timing,
+	const enum dc_link_encoding_format link_encoding)
+{
+	uint32_t bits_per_channel = 0;
+	uint32_t kbps;
+
+#if defined(CONFIG_DRM_AMD_DC_FP)
+	if (timing->flags.DSC)
+		return dc_dsc_stream_bandwidth_in_kbps(timing,
+				timing->dsc_cfg.bits_per_pixel,
+				timing->dsc_cfg.num_slices_h,
+				timing->dsc_cfg.is_dp);
+#endif
+
+	switch (timing->display_color_depth) {
+	case COLOR_DEPTH_666:
+		bits_per_channel = 6;
+		break;
+	case COLOR_DEPTH_888:
+		bits_per_channel = 8;
+		break;
+	case COLOR_DEPTH_101010:
+		bits_per_channel = 10;
+		break;
+	case COLOR_DEPTH_121212:
+		bits_per_channel = 12;
+		break;
+	case COLOR_DEPTH_141414:
+		bits_per_channel = 14;
+		break;
+	case COLOR_DEPTH_161616:
+		bits_per_channel = 16;
+		break;
+	default:
+		ASSERT(bits_per_channel != 0);
+		bits_per_channel = 8;
+		break;
+	}
+
+	kbps = timing->pix_clk_100hz / 10;
+	kbps *= bits_per_channel;
+
+	if (timing->flags.Y_ONLY != 1) {
+		/*Only YOnly make reduce bandwidth by 1/3 compares to RGB*/
+		kbps *= 3;
+		if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
+			kbps /= 2;
+		else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
+			kbps = kbps * 2 / 3;
+	}
+
+	if (link_encoding == DC_LINK_ENCODING_DP_128b_132b)
+		kbps = apply_128b_132b_stream_overhead(timing, kbps);
+
+	if (link_encoding == DC_LINK_ENCODING_HDMI_FRL &&
+			timing->vic == 0 && timing->hdmi_vic == 0 &&
+			timing->frl_uncompressed_video_bandwidth_in_kbps != 0)
+		kbps = timing->frl_uncompressed_video_bandwidth_in_kbps;
+
+	return kbps;
+}

drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c

···
 	}
 }
 
+void get_fams2_visual_confirm_color(
+		struct dc *dc,
+		struct dc_state *context,
+		struct pipe_ctx *pipe_ctx,
+		struct tg_color *color)
+{
+	uint32_t color_value = MAX_TG_COLOR_VALUE;
+
+	if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !context || !dc->debug.fams2_config.bits.enable)
+		return;
+
+	/* driver only handles visual confirm when FAMS2 is disabled */
+	if (!dc_state_is_fams2_in_use(dc, context)) {
+		/* when FAMS2 is disabled, all pipes are grey */
+		color->color_g_y = color_value / 2;
+		color->color_b_cb = color_value / 2;
+		color->color_r_cr = color_value / 2;
+	}
+}
+
 void hwss_build_fast_sequence(struct dc *dc,
 		struct dc_dmub_cmd *dc_dmub_cmd,
 		unsigned int dmub_cmd_count,
···
 		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.subvp_immediate_flip =
 				plane->flip_immediate && stream_status->mall_stream_config.type == SUBVP_MAIN;
 		block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST;
+		(*num_steps)++;
+	}
+	if (dc->hwss.fams2_global_control_lock_fast) {
+		block_sequence[*num_steps].params.fams2_global_control_lock_fast_params.dc = dc;
+		block_sequence[*num_steps].params.fams2_global_control_lock_fast_params.lock = true;
+		block_sequence[*num_steps].params.fams2_global_control_lock_fast_params.is_required = dc_state_is_fams2_in_use(dc, context);
+		block_sequence[*num_steps].func = DMUB_FAMS2_GLOBAL_CONTROL_LOCK_FAST;
 		(*num_steps)++;
 	}
 	if (dc->hwss.pipe_control_lock) {
···
 		block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST;
 		(*num_steps)++;
 	}
+	if (dc->hwss.fams2_global_control_lock_fast) {
+		block_sequence[*num_steps].params.fams2_global_control_lock_fast_params.dc = dc;
+		block_sequence[*num_steps].params.fams2_global_control_lock_fast_params.lock = false;
+		block_sequence[*num_steps].params.fams2_global_control_lock_fast_params.is_required = dc_state_is_fams2_in_use(dc, context);
+		block_sequence[*num_steps].func = DMUB_FAMS2_GLOBAL_CONTROL_LOCK_FAST;
+		(*num_steps)++;
+	}
 
 	current_pipe = pipe_ctx;
 	while (current_pipe) {
···
 		case DMUB_SUBVP_SAVE_SURF_ADDR:
 			hwss_subvp_save_surf_addr(params);
 			break;
+		case DMUB_FAMS2_GLOBAL_CONTROL_LOCK_FAST:
+			dc->hwss.fams2_global_control_lock_fast(params);
+			break;
 		default:
 			ASSERT(false);
 			break;
···
 				plane_state->input_csc_color_matrix,
 				plane_state->color_space,
 				NULL);
+	}
+
+	if (dpp && dpp->funcs->set_cursor_matrix) {
+		dpp->funcs->set_cursor_matrix(dpp,
+			plane_state->color_space,
+			plane_state->cursor_csc_color_matrix);
 	}
 }
 

drivers/gpu/drm/amd/display/dc/core/dc_resource.c

···
 #include "dcn321/dcn321_resource.h"
 #include "dcn35/dcn35_resource.h"
 #include "dcn351/dcn351_resource.h"
+#include "dcn401/dcn401_resource.h"
+#if defined(CONFIG_DRM_AMD_DC_FP)
+#include "dc_spl_translate.h"
+#endif
 
 #define VISUAL_CONFIRM_BASE_DEFAULT 3
 #define VISUAL_CONFIRM_BASE_MIN 1
···
 		if (ASICREV_IS_GC_11_0_4(asic_id.hw_internal_rev))
 			dc_version = DCN_VERSION_3_51;
 		break;
+	case AMDGPU_FAMILY_GC_12_0_0:
+		if (ASICREV_IS_DCN401(asic_id.hw_internal_rev))
+			dc_version = DCN_VERSION_4_01;
+		break;
 	default:
 		dc_version = DCE_VERSION_UNKNOWN;
 		break;
···
 		break;
 	case DCN_VERSION_3_51:
 		res_pool = dcn351_create_resource_pool(init_data, dc);
+		break;
+	case DCN_VERSION_4_01:
+		res_pool = dcn401_create_resource_pool(init_data, dc);
 		break;
 #endif /* CONFIG_DRM_AMD_DC_FP */
 	default:
···
 	pipe_ctx->plane_res.scl_data.format = convert_pixel_format_to_dalsurface(
 			pipe_ctx->plane_state->format);
 
+	if (pipe_ctx->stream->ctx->dc->config.use_spl)	{
+#if defined(CONFIG_DRM_AMD_DC_FP)
+		struct spl_in *spl_in = &pipe_ctx->plane_res.spl_in;
+		struct spl_out *spl_out = &pipe_ctx->plane_res.spl_out;
+
+		if (plane_state->ctx->dce_version > DCE_VERSION_MAX)
+			pipe_ctx->plane_res.scl_data.lb_params.depth = LB_PIXEL_DEPTH_36BPP;
+		else
+			pipe_ctx->plane_res.scl_data.lb_params.depth = LB_PIXEL_DEPTH_30BPP;
+
+		pipe_ctx->plane_res.scl_data.lb_params.alpha_en = plane_state->per_pixel_alpha;
+		spl_out->scl_data.h_active = pipe_ctx->plane_res.scl_data.h_active;
+		spl_out->scl_data.v_active = pipe_ctx->plane_res.scl_data.v_active;
+
+		// Convert pipe_ctx to respective input params for SPL
+		translate_SPL_in_params_from_pipe_ctx(pipe_ctx, spl_in);
+		// Set SPL output parameters to dscl_prog_data to be used for hw registers
+		spl_out->dscl_prog_data = resource_get_dscl_prog_data(pipe_ctx);
+		// Calculate scaler parameters from SPL
+		res = spl_calculate_scaler_params(spl_in, spl_out);
+		// Convert respective out params from SPL to scaler data
+		translate_SPL_out_params_to_pipe_ctx(pipe_ctx, spl_out);
+#endif
+	} else {
+
 	/* depends on h_active */
 	calculate_recout(pipe_ctx);
 	/* depends on pixel format */
···
 		pipe_ctx->plane_res.scl_data.viewport.height = MIN_VIEWPORT_SIZE;
 	if (pipe_ctx->plane_res.scl_data.viewport.width < MIN_VIEWPORT_SIZE)
 		pipe_ctx->plane_res.scl_data.viewport.width = MIN_VIEWPORT_SIZE;
-
-
+	}
 	DC_LOG_SCALER("%s pipe %d:\nViewport: height:%d width:%d x:%d y:%d  Recout: height:%d width:%d x:%d y:%d  HACTIVE:%d VACTIVE:%d\n"
 			"src_rect: height:%d width:%d x:%d y:%d  dst_rect: height:%d width:%d x:%d y:%d  clip_rect: height:%d width:%d x:%d y:%d\n",
 			__func__,
···
 		return true;
 
 	return false;
+}
+
+struct dscl_prog_data *resource_get_dscl_prog_data(struct pipe_ctx *pipe_ctx)
+{
+	return &pipe_ctx->plane_res.scl_data.dscl_prog_data;
 }
 
 void resource_init_common_dml2_callbacks(struct dc *dc, struct dml2_configuration_options *dml2_options)

drivers/gpu/drm/amd/display/dc/core/dc_state.c

···
 #include "resource.h"
 #include "link_enc_cfg.h"
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
 #include "dml2/dml2_wrapper.h"
 #include "dml2/dml2_internal_types.h"
+#endif
 
 #define DC_LOGGER \
 	dc->ctx->logger
···
 	return stream;
 }
 
+bool dc_state_is_fams2_in_use(
+		const struct dc *dc,
+		const struct dc_state *state)
+{
+	bool is_fams2_in_use = false;
+
+	if (state)
+		is_fams2_in_use |= state->bw_ctx.bw.dcn.fams2_stream_count > 0;
+
+	if (dc->current_state)
+		is_fams2_in_use |= dc->current_state->bw_ctx.bw.dcn.fams2_stream_count > 0;
+
+	return is_fams2_in_use;
+}

drivers/gpu/drm/amd/display/dc/dc.h

···
 	bool zstate_support;
 	bool ips_support;
 	uint32_t num_of_internal_disp;
+	uint32_t max_dwb_htap;
+	uint32_t max_dwb_vtap;
 	enum dp_protocol_version max_dp_protocol_version;
+	bool spdif_aud;
 	unsigned int mall_size_per_mem_channel;
 	unsigned int mall_size_total;
 	unsigned int cursor_cache_size;
···
 	uint32_t max_v_total;
 	uint32_t max_disp_clock_khz_at_vmin;
 	uint8_t subvp_drr_vblank_start_margin_us;
+	bool cursor_not_scaled;
 };
 
 struct dc_bug_wa {
···
 		uint8_t dcfclk : 1;
 		uint8_t dcfclk_ds: 1;
 	} clock_update_disable_mask;
+	//Customer Specific WAs
+	uint32_t force_backlight_start_level;
 };
 struct dc_dcc_surface_param {
 	struct dc_size surface_size;
 	enum surface_pixel_format format;
-	enum swizzle_mode_values swizzle_mode;
+	unsigned int plane0_pitch;
+	struct dc_size plane1_size;
+	unsigned int plane1_pitch;
+	union {
+		enum swizzle_mode_values swizzle_mode;
+		enum swizzle_mode_addr3_values swizzle_mode_addr3;
+	};
 	enum dc_scan_direction scan;
 };
 
···
 struct dc_plane_state;
 struct dc_state;
 
-
 struct dc_cap_funcs {
 	bool (*get_dcc_compression_cap)(const struct dc *dc,
 			const struct dc_dcc_surface_param *input,
···
 	bool is_asymmetric_memory;
 	bool is_single_rank_dimm;
 	bool is_vmin_only_asic;
+	bool use_spl;
+	bool prefer_easf;
 	bool use_pipe_ctx_sync_logic;
 	bool ignore_dpref_ss;
 	bool enable_mipi_converter_optimization;
···
 	VISUAL_CONFIRM_REPLAY = 12,
 	VISUAL_CONFIRM_SUBVP = 14,
 	VISUAL_CONFIRM_MCLK_SWITCH = 16,
+	VISUAL_CONFIRM_FAMS2 = 19,
 };
 
 enum dc_psr_power_opts {
···
 	bool enable_single_display_2to1_odm_policy;
 	bool enable_double_buffered_dsc_pg_support;
 	bool enable_dp_dig_pixel_rate_div_policy;
+	bool using_dml21;
 	enum lttpr_mode lttpr_mode_override;
 	unsigned int dsc_delay_factor_wa_x1000;
 	unsigned int min_prefetch_in_strobe_ns;
···
 	unsigned int static_screen_wait_frames;
 	bool force_chroma_subsampling_1tap;
 	bool disable_422_left_edge_pixel;
+	bool dml21_force_pstate_method;
+	uint32_t dml21_force_pstate_method_value;
+	uint32_t dml21_disable_pstate_method_mask;
+	union dmub_fams2_global_feature_config fams2_config;
 	unsigned int force_cositing;
 };
 
···
 		uint32_t stereo_format_change:1;
 		uint32_t lut_3d:1;
 		uint32_t tmz_changed:1;
+		uint32_t mcm_transfer_function_enable_change:1; /* disable or enable MCM transfer func */
 		uint32_t full_update:1;
 	} bits;
 
···
 
 	bool is_statically_allocated;
 	enum chroma_cositing cositing;
+	enum dc_cm2_shaper_3dlut_setting mcm_shaper_3dlut_setting;
+	bool mcm_lut1d_enable;
+	struct dc_cm2_func_luts mcm_luts;
+	bool lut_bank_a;
+	enum mpcc_movable_cm_location mcm_location;
+	struct dc_csc_transform cursor_csc_color_matrix;
+	bool adaptive_sharpness_en;
+	unsigned int sharpnessX1000;
+	enum linear_light_scaling linear_light_scaling;
 };
 
 struct dc_plane_info {
···
 	int  global_alpha_value;
 	bool input_csc_enabled;
 	int layer_index;
+	bool front_buffer_rendering_active;
 	enum chroma_cositing cositing;
 };
 
···
 	const struct fixed31_32 *coeff_reduction_factor;
 	struct dc_transfer_func *out_transfer_func;
 	struct dc_csc_transform *output_csc_transform;
+	const struct dc_csc_transform *cursor_csc_color_matrix;
 };
 
 struct dc_surface_update {
···
 	const struct dc_3dlut *lut3d_func;
 	const struct dc_transfer_func *blend_tf;
 	const struct colorspace_transform *gamut_remap_matrix;
+	/*
+	 * Color Transformations for pre-blend MCM (Shaper, 3DLUT, 1DLUT)
+	 *
+	 * change cm2_params.component_settings: Full update
+	 * change cm2_params.cm2_luts: Fast update
+	 */
+	struct dc_cm2_parameters *cm2_params;
+	const struct dc_csc_transform *cursor_csc_color_matrix;
 };
 
 /*
···
 uint32_t dc_get_opp_for_plane(struct dc *dc, struct dc_plane_state *plane);
 
 void dc_set_disable_128b_132b_stream_overhead(bool disable);
+bool dc_get_disable_128b_132b_stream_overhead(void);
 
 /* The function returns minimum bandwidth required to drive a given timing
  * return - minimum required timing bandwidth in kbps.
···
 	union dpcd_sink_ext_caps dpcd_sink_ext_caps;
 
 	struct psr_settings psr_settings;
-
 	struct replay_settings replay_settings;
 
 	/* Drive settings read from integrated info table */

drivers/gpu/drm/amd/display/dc/dc_bios_types.h

···
 	struct dc_firmware_info fw_info;
 	bool fw_info_valid;
 	struct dc_vram_info vram_info;
+	struct bp_soc_bb_info bb_info;
 	struct dc_golden_table golden_table;
 };
 

drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c

···
 	return result;
 }
 
+void dc_dmub_srv_fams2_update_config(struct dc *dc,
+		struct dc_state *context,
+		bool enable)
+{
+	uint8_t num_cmds = 1;
+	uint32_t i;
+	union dmub_rb_cmd cmd[MAX_STREAMS + 1];
+	struct dmub_rb_cmd_fams2 *global_cmd = &cmd[0].fams2_config;
+
+	memset(cmd, 0, sizeof(union dmub_rb_cmd) * (MAX_STREAMS + 1));
+	/* fill in generic command header */
+	global_cmd->header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
+	global_cmd->header.sub_type = DMUB_CMD__FAMS2_CONFIG;
+	global_cmd->header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
+
+	/* send global configuration parameters */
+	global_cmd->config.global.max_allow_delay_us = 100 * 1000; //100ms
+	global_cmd->config.global.lock_wait_time_us = 5000; //5ms
+
+	/* copy static feature configuration */
+	global_cmd->config.global.features.all = dc->debug.fams2_config.all;
+
+	/* apply feature configuration based on current driver state */
+	global_cmd->config.global.features.bits.enable_visual_confirm = dc->debug.visual_confirm == VISUAL_CONFIRM_FAMS2;
+	global_cmd->config.global.features.bits.enable = enable;
+
+	/* construct per-stream configs */
+	if (enable) {
+		for (i = 0; i < context->bw_ctx.bw.dcn.fams2_stream_count; i++) {
+			struct dmub_rb_cmd_fams2 *stream_cmd = &cmd[i+1].fams2_config;
+
+			/* configure command header */
+			stream_cmd->header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
+			stream_cmd->header.sub_type = DMUB_CMD__FAMS2_CONFIG;
+			stream_cmd->header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
+			stream_cmd->header.multi_cmd_pending = 1;
+			/* copy stream static state */
+			memcpy(&stream_cmd->config.stream,
+					&context->bw_ctx.bw.dcn.fams2_stream_params[i],
+					sizeof(struct dmub_fams2_stream_static_state));
+		}
+	}
+
+	if (enable && context->bw_ctx.bw.dcn.fams2_stream_count) {
+		/* set multi pending for global, and unset for last stream cmd */
+		global_cmd->config.global.num_streams = context->bw_ctx.bw.dcn.fams2_stream_count;
+		global_cmd->header.multi_cmd_pending = 1;
+		cmd[context->bw_ctx.bw.dcn.fams2_stream_count].fams2_config.header.multi_cmd_pending = 0;
+		num_cmds += context->bw_ctx.bw.dcn.fams2_stream_count;
+	}
+
+	dm_execute_dmub_cmd_list(dc->ctx, num_cmds, cmd, DM_DMUB_WAIT_TYPE_WAIT);
+}
+
+void dc_dmub_srv_fams2_drr_update(struct dc *dc,
+		uint32_t tg_inst,
+		uint32_t vtotal_min,
+		uint32_t vtotal_max,
+		uint32_t vtotal_mid,
+		uint32_t vtotal_mid_frame_num,
+		bool program_manual_trigger)
+{
+	union dmub_rb_cmd cmd = { 0 };
+
+	cmd.fams2_drr_update.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
+	cmd.fams2_drr_update.header.sub_type = DMUB_CMD__FAMS2_DRR_UPDATE;
+	cmd.fams2_drr_update.dmub_optc_state_req.tg_inst = tg_inst;
+	cmd.fams2_drr_update.dmub_optc_state_req.v_total_max = vtotal_max;
+	cmd.fams2_drr_update.dmub_optc_state_req.v_total_min = vtotal_min;
+	cmd.fams2_drr_update.dmub_optc_state_req.v_total_mid = vtotal_mid;
+	cmd.fams2_drr_update.dmub_optc_state_req.v_total_mid_frame_num = vtotal_mid_frame_num;
+	cmd.fams2_drr_update.dmub_optc_state_req.program_manual_trigger = program_manual_trigger;
+
+	cmd.fams2_drr_update.header.payload_bytes = sizeof(cmd.fams2_drr_update) - sizeof(cmd.fams2_drr_update.header);
+
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
+}
+
+void dc_dmub_srv_fams2_passthrough_flip(
+		struct dc *dc,
+		struct dc_state *state,
+		struct dc_stream_state *stream,
+		struct dc_surface_update *srf_updates,
+		int surface_count)
+{
+	int plane_index;
+	union dmub_rb_cmd cmds[MAX_PLANES];
+	struct dc_plane_address *address;
+	struct dc_plane_state *plane_state;
+	int num_cmds = 0;
+	struct dc_stream_status *stream_status = dc_stream_get_status(stream);
+
+	if (surface_count <= 0 || stream_status == NULL)
+		return;
+
+	memset(cmds, 0, sizeof(union dmub_rb_cmd) * MAX_PLANES);
+
+	/* build command for each surface update */
+	for (plane_index = 0; plane_index < surface_count; plane_index++) {
+		plane_state = srf_updates[plane_index].surface;
+		address = &plane_state->address;
+
+		/* skip if there is no address update for plane */
+		if (!srf_updates[plane_index].flip_addr)
+			continue;
+
+		/* build command header */
+		cmds[num_cmds].fams2_flip.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
+		cmds[num_cmds].fams2_flip.header.sub_type = DMUB_CMD__FAMS2_FLIP;
+		cmds[num_cmds].fams2_flip.header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2_flip);
+
+		/* for chaining multiple commands, all but last command should set to 1 */
+		cmds[num_cmds].fams2_flip.header.multi_cmd_pending = 1;
+
+		/* set topology info */
+		cmds[num_cmds].fams2_flip.flip_info.pipe_mask = dc_plane_get_pipe_mask(state, plane_state);
+		if (stream_status)
+			cmds[num_cmds].fams2_flip.flip_info.otg_inst = stream_status->primary_otg_inst;
+
+		cmds[num_cmds].fams2_flip.flip_info.config.bits.is_immediate = plane_state->flip_immediate;
+
+		/* build address info for command */
+		switch (address->type) {
+		case PLN_ADDR_TYPE_GRAPHICS:
+			if (address->grph.addr.quad_part == 0) {
+				BREAK_TO_DEBUGGER();
+				break;
+			}
+
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.meta_addr_lo =
+					address->grph.meta_addr.low_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.meta_addr_hi =
+					(uint16_t)address->grph.meta_addr.high_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.surf_addr_lo =
+					address->grph.addr.low_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.surf_addr_hi =
+					(uint16_t)address->grph.addr.high_part;
+			break;
+		case PLN_ADDR_TYPE_VIDEO_PROGRESSIVE:
+			if (address->video_progressive.luma_addr.quad_part == 0 ||
+				address->video_progressive.chroma_addr.quad_part == 0) {
+				BREAK_TO_DEBUGGER();
+				break;
+			}
+
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.meta_addr_lo =
+					address->video_progressive.luma_meta_addr.low_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.meta_addr_hi =
+					(uint16_t)address->video_progressive.luma_meta_addr.high_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.meta_addr_c_lo =
+					address->video_progressive.chroma_meta_addr.low_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.meta_addr_c_hi =
+					(uint16_t)address->video_progressive.chroma_meta_addr.high_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.surf_addr_lo =
+					address->video_progressive.luma_addr.low_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.surf_addr_hi =
+					(uint16_t)address->video_progressive.luma_addr.high_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.surf_addr_c_lo =
+					address->video_progressive.chroma_addr.low_part;
+			cmds[num_cmds].fams2_flip.flip_info.addr_info.surf_addr_c_hi =
+					(uint16_t)address->video_progressive.chroma_addr.high_part;
+			break;
+		default:
+			// Should never be hit
+			BREAK_TO_DEBUGGER();
+			break;
+		}
+
+		num_cmds++;
+	}
+
+	if (num_cmds > 0)  {
+		cmds[num_cmds - 1].fams2_flip.header.multi_cmd_pending = 0;
+		dm_execute_dmub_cmd_list(dc->ctx, num_cmds, cmds, DM_DMUB_WAIT_TYPE_WAIT);
+	}
+}

drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h

···
 bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type);
 
 bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv,
-				    unsigned int stream_mask);
+				   unsigned int stream_mask);
 
 bool dc_dmub_srv_is_restore_required(struct dc_dmub_srv *dc_dmub_srv);
 
···
 bool dc_wake_and_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_command command_code,
 			       uint16_t param, uint32_t *response, enum dm_dmub_wait_type wait_type);
 
+void dc_dmub_srv_fams2_update_config(struct dc *dc,
+		struct dc_state *context,
+		bool enable);
+void dc_dmub_srv_fams2_drr_update(struct dc *dc,
+		uint32_t tg_inst,
+		uint32_t vtotal_min,
+		uint32_t vtotal_max,
+		uint32_t vtotal_mid,
+		uint32_t vtotal_mid_frame_num,
+		bool program_manual_trigger);
+void dc_dmub_srv_fams2_passthrough_flip(
+		struct dc *dc,
+		struct dc_state *state,
+		struct dc_stream_state *stream,
+		struct dc_surface_update *srf_updates,
+		int surface_count);
 #endif /* _DMUB_DC_SRV_H_ */

drivers/gpu/drm/amd/display/dc/dc_helper.c

···
 
 	va_end(ap);
 
-
 	/* mmio write directly */
 	reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value;
 
···
 		return "DCN 3.5";
 	case DCN_VERSION_3_51:
 		return "DCN 3.5.1";
+	case DCN_VERSION_4_01:
+		return "DCN 4.0.1";
 	default:
 		return "Unknown";
 	}

drivers/gpu/drm/amd/display/dc/dc_hw_types.h

···
 
 enum dc_plane_addr_type {
 	PLN_ADDR_TYPE_GRAPHICS = 0,
+	PLN_ADDR_TYPE_3DLUT,
 	PLN_ADDR_TYPE_GRPH_STEREO,
 	PLN_ADDR_TYPE_VIDEO_PROGRESSIVE,
 	PLN_ADDR_TYPE_RGBEA
···
 			PHYSICAL_ADDRESS_LOC meta_addr;
 			union large_integer dcc_const_color;
 		} grph;
+
+		struct {
+			PHYSICAL_ADDRESS_LOC addr;
+		} lut3d;
 
 		/*stereo*/
 		struct {
···
 			PHYSICAL_ADDRESS_LOC right_alpha_addr;
 			PHYSICAL_ADDRESS_LOC right_alpha_meta_addr;
 			union large_integer right_alpha_dcc_const_color;
-
 		} grph_stereo;
 
 		/*video  progressive*/
···
 	union large_integer page_table_base;
 
 	uint8_t vmid;
+	/* dc should use hw flip queue rather than directly programming the surface address.
+	 * Value is determined on each flip. */
+	bool offload_flip;
 };
 
 struct dc_size {
···
 	DC_TRIPLEBUFFER_DISABLE = 0x0,
 	DC_TRIPLEBUFFER_ENABLE = 0x1,
 };
+enum tile_split_values_new {
+	DC_SURF_TILE_SPLIT_1KB = 0x4,
+};
 
 /* TODO: These values come from hardware spec. We need to readdress this
  * if they ever change.
···
 	DC_SW_VAR_R_X = 31,
 	DC_SW_MAX = 32,
 	DC_SW_UNKNOWN = DC_SW_MAX
+};
+
+// Definition of swizzle modes with addr3 ASICs
+enum swizzle_mode_addr3_values {
+	DC_ADDR3_SW_LINEAR = 0,
+	DC_ADDR3_SW_256B_2D = 1,
+	DC_ADDR3_SW_4KB_2D = 2,
+	DC_ADDR3_SW_64KB_2D = 3,
+	DC_ADDR3_SW_256KB_2D = 4,
+	DC_ADDR3_SW_4KB_3D = 5,
+	DC_ADDR3_SW_64KB_3D = 6,
+	DC_ADDR3_SW_256KB_3D = 7,
+	DC_ADDR3_SW_MAX = 8,
+	DC_ADDR3_SW_UNKNOWN = DC_ADDR3_SW_MAX
 };
 
 union dc_tiling_info {
···
 		bool rb_aligned;
 		bool pipe_aligned;
 		unsigned int num_pkrs;
-	} gfx9;
+	} gfx9;/*gfx9, gfx10 and above*/
+	struct {
+		enum swizzle_mode_addr3_values swizzle;
+	} gfx_addr3;/*gfx with addr3 and above*/
 };
 
 /* Rotation angle */
···
 	unsigned int pixel_clk_khz;
 	unsigned int ref_clk_khz;
 	struct rect viewport;
+	struct rect recout;
 	struct fixed31_32 h_scale_ratio;
 	struct fixed31_32 v_scale_ratio;
 	enum dc_rotation_angle rotation;
···
 enum cm_gamut_coef_format {
 	CM_GAMUT_REMAP_COEF_FORMAT_S2_13 = 0,
 	CM_GAMUT_REMAP_COEF_FORMAT_S3_12 = 1
+};
+
+enum mpcc_gamut_remap_mode_select {
+	MPCC_GAMUT_REMAP_MODE_SELECT_0 = 0,
+	MPCC_GAMUT_REMAP_MODE_SELECT_1,
+	MPCC_GAMUT_REMAP_MODE_SELECT_2
+};
+
+enum mpcc_gamut_remap_id {
+	MPCC_OGAM_GAMUT_REMAP,
+	MPCC_MCM_FIRST_GAMUT_REMAP,
+	MPCC_MCM_SECOND_GAMUT_REMAP
+};
+
+enum cursor_matrix_mode {
+	CUR_MATRIX_BYPASS = 0,
+	CUR_MATRIX_SET_A,
+	CUR_MATRIX_SET_B
 };
 
 struct mcif_warmup_params {

drivers/gpu/drm/amd/display/dc/dc_state_priv.h

···
 		const struct dc *dc,
 		struct dc_state *state);
 
+bool dc_state_is_fams2_in_use(
+		const struct dc *dc,
+		const struct dc_state *state);
+
 #endif /* _DC_STATE_PRIV_H_ */

drivers/gpu/drm/amd/display/dc/dc_types.h

···
 	HPD_EN_FOR_SECONDARY_EDP_ONLY,
 };
 
+enum dc_cm2_shaper_3dlut_setting {
+	DC_CM2_SHAPER_3DLUT_SETTING_BYPASS_ALL,
+	DC_CM2_SHAPER_3DLUT_SETTING_ENABLE_SHAPER,
+	/* Bypassing Shaper will always bypass 3DLUT */
+	DC_CM2_SHAPER_3DLUT_SETTING_ENABLE_SHAPER_3DLUT
+};
+
+enum dc_cm2_gpu_mem_layout {
+	DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_RGB,
+	DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_BGR,
+	DC_CM2_GPU_MEM_LAYOUT_1D_PACKED_LINEAR
+};
+
+enum dc_cm2_gpu_mem_pixel_component_order {
+	DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA,
+};
+
+enum dc_cm2_gpu_mem_format {
+	DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12MSB,
+	DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12LSB,
+	DC_CM2_GPU_MEM_FORMAT_16161616_FLOAT_FP1_5_10
+};
+
+struct dc_cm2_gpu_mem_format_parameters {
+	enum dc_cm2_gpu_mem_format format;
+	union {
+		struct {
+			/* bias & scale for float only */
+			uint16_t bias;
+			uint16_t scale;
+		} float_params;
+	};
+};
+
+enum dc_cm2_gpu_mem_size {
+	DC_CM2_GPU_MEM_SIZE_171717,
+	DC_CM2_GPU_MEM_SIZE_TRANSFORMED
+};
+
+struct dc_cm2_gpu_mem_parameters {
+	struct dc_plane_address addr;
+	enum dc_cm2_gpu_mem_layout layout;
+	struct dc_cm2_gpu_mem_format_parameters format_params;
+	enum dc_cm2_gpu_mem_pixel_component_order component_order;
+	enum dc_cm2_gpu_mem_size  size;
+};
+
+enum dc_cm2_transfer_func_source {
+	DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM,
+	DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM
+};
+
+struct dc_cm2_component_settings {
+	enum dc_cm2_shaper_3dlut_setting shaper_3dlut_setting;
+	bool lut1d_enable;
+};
+
+/*
+ * All pointers in this struct must remain valid for as long as the 3DLUTs are used
+ */
+struct dc_cm2_func_luts {
+	const struct dc_transfer_func *shaper;
+	struct {
+		enum dc_cm2_transfer_func_source lut3d_src;
+		union {
+			const struct dc_3dlut *lut3d_func;
+			struct dc_cm2_gpu_mem_parameters gpu_mem_params;
+		};
+	} lut3d_data;
+	const struct dc_transfer_func *lut1d_func;
+};
+
+struct dc_cm2_parameters {
+	struct dc_cm2_component_settings component_settings;
+	struct dc_cm2_func_luts cm2_luts;
+};
+
 enum mall_stream_type {
 	SUBVP_NONE, // subvp not in use
 	SUBVP_MAIN, // subvp in use, this stream is main stream

drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c

···
 	return true;
 }
 
+static bool dcn401_program_pix_clk(
+		struct clock_source *clock_source,
+		struct pixel_clk_params *pix_clk_params,
+		enum dp_link_encoding encoding,
+		struct pll_settings *pll_settings)
+{
+	struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source);
+	unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0;
+	const struct pixel_rate_range_table_entry *e =
+			look_up_in_video_optimized_rate_tlb(pix_clk_params->requested_pix_clk_100hz / 10);
+	struct bp_pixel_clock_parameters bp_pc_params = {0};
+	enum transmitter_color_depth bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24;
+	struct dp_dto_params dto_params = { 0 };
+
+	dto_params.otg_inst = inst;
+	dto_params.signal = pix_clk_params->signal_type;
+
+	// all but TMDS gets Driver to program DP_DTO without calling VBIOS Command table
+	if (!dc_is_hdmi_tmds_signal(pix_clk_params->signal_type)) {
+		long long ref_dtbclk_khz = clock_source->ctx->dc->clk_mgr->funcs->get_dtb_ref_clk_frequency(clock_source->ctx->dc->clk_mgr);
+		long long dprefclk_khz = clock_source->ctx->dc->clk_mgr->dprefclk_khz;
+		long long dtbclk_p_src_clk_khz;
+		/* if signal is DP132B128B dtbclk_p_src is DTBCLK else DPREFCLK */
+		dtbclk_p_src_clk_khz = encoding == DP_128b_132b_ENCODING ? ref_dtbclk_khz : dprefclk_khz;
+		if (e) {
+			dto_params.pixclk_hz = e->target_pixel_rate_khz * e->mult_factor;
+			dto_params.refclk_hz = dtbclk_p_src_clk_khz * e->div_factor;
+		} else {
+			dto_params.pixclk_hz = pix_clk_params->requested_pix_clk_100hz * 100;
+			dto_params.refclk_hz = dtbclk_p_src_clk_khz * 1000;
+		}
+
+		/* enable DP DTO */
+		clock_source->ctx->dc->res_pool->dccg->funcs->set_dp_dto(
+				clock_source->ctx->dc->res_pool->dccg,
+				&dto_params);
+
+	} else {
+		/* disables DP DTO when provided with TMDS signal type */
+		clock_source->ctx->dc->res_pool->dccg->funcs->set_dp_dto(
+				clock_source->ctx->dc->res_pool->dccg,
+				&dto_params);
+
+		/*ATOMBIOS expects pixel rate adjusted by deep color ratio)*/
+		bp_pc_params.controller_id = pix_clk_params->controller_id;
+		bp_pc_params.pll_id = clock_source->id;
+		bp_pc_params.target_pixel_clock_100hz = pll_settings->actual_pix_clk_100hz;
+		bp_pc_params.encoder_object_id = pix_clk_params->encoder_object_id;
+		bp_pc_params.signal_type = pix_clk_params->signal_type;
+
+		// Make sure we send the correct color depth to DMUB for HDMI
+		if (pix_clk_params->signal_type == SIGNAL_TYPE_HDMI_TYPE_A) {
+			switch (pix_clk_params->color_depth) {
+			case COLOR_DEPTH_888:
+				bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24;
+				break;
+			case COLOR_DEPTH_101010:
+				bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_30;
+				break;
+			case COLOR_DEPTH_121212:
+				bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_36;
+				break;
+			case COLOR_DEPTH_161616:
+				bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_48;
+				break;
+			default:
+				bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24;
+				break;
+			}
+			bp_pc_params.color_depth = bp_pc_colour_depth;
+		}
+
+		if (clock_source->id != CLOCK_SOURCE_ID_DP_DTO) {
+			bp_pc_params.flags.SET_GENLOCK_REF_DIV_SRC =
+							pll_settings->use_external_clk;
+			bp_pc_params.flags.SET_XTALIN_REF_SRC =
+							!pll_settings->use_external_clk;
+			if (pix_clk_params->flags.SUPPORT_YCBCR420) {
+				bp_pc_params.flags.SUPPORT_YUV_420 = 1;
+			}
+		}
+		if (clk_src->bios->funcs->set_pixel_clock(
+				clk_src->bios, &bp_pc_params) != BP_RESULT_OK)
+			return false;
+		/* Resync deep color DTO */
+		if (clock_source->id != CLOCK_SOURCE_ID_DP_DTO)
+			dce112_program_pixel_clk_resync(clk_src,
+						pix_clk_params->signal_type,
+						pix_clk_params->color_depth,
+						pix_clk_params->flags.SUPPORT_YCBCR420);
+	}
+
+	return true;
+}
+
 static bool dce110_clock_source_power_down(
 		struct clock_source *clk_src)
 {
···
 static const struct clock_source_funcs dcn31_clk_src_funcs = {
 	.cs_power_down = dce110_clock_source_power_down,
 	.program_pix_clk = dcn31_program_pix_clk,
+	.get_pix_clk_dividers = dcn3_get_pix_clk_dividers,
+	.get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz
+};
+
+static const struct clock_source_funcs dcn401_clk_src_funcs = {
+	.cs_power_down = dce110_clock_source_power_down,
+	.program_pix_clk = dcn401_program_pix_clk,
 	.get_pix_clk_dividers = dcn3_get_pix_clk_dividers,
 	.get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz
 };
···
 	return ret;
 }
 
+bool dcn401_clk_src_construct(
+	struct dce110_clk_src *clk_src,
+	struct dc_context *ctx,
+	struct dc_bios *bios,
+	enum clock_source_id id,
+	const struct dce110_clk_src_regs *regs,
+	const struct dce110_clk_src_shift *cs_shift,
+	const struct dce110_clk_src_mask *cs_mask)
+{
+	bool ret = dce112_clk_src_construct(clk_src, ctx, bios, id, regs, cs_shift, cs_mask);
+
+	clk_src->base.funcs = &dcn401_clk_src_funcs;
+
+	return ret;
+}
 bool dcn301_clk_src_construct(
 	struct dce110_clk_src *clk_src,
 	struct dc_context *ctx,

drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h

···
 	const struct dce110_clk_src_shift *cs_shift,
 	const struct dce110_clk_src_mask *cs_mask);
 
+bool dcn401_clk_src_construct(
+	struct dce110_clk_src *clk_src,
+	struct dc_context *ctx,
+	struct dc_bios *bios,
+	enum clock_source_id id,
+	const struct dce110_clk_src_regs *regs,
+	const struct dce110_clk_src_shift *cs_shift,
+	const struct dce110_clk_src_mask *cs_mask);
 /* this table is use to find *1.001 and /1.001 pixel rates from non-precise pixel rate */
 struct pixel_rate_range_table_entry {
 	unsigned int range_min_khz;

drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h

···
 	I2C_COMMON_MASK_SH_LIST_DCN30(mask_sh),\
 	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_CLK_EN, mask_sh)
 
+#define I2C_COMMON_MASK_SH_LIST_DCN401(mask_sh)\
+	I2C_COMMON_MASK_SH_LIST_DCN30(mask_sh),\
+	I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_CLK_EN, mask_sh)
+
 struct dce_i2c_registers {
 	uint32_t SETUP;
 	uint32_t SPEED;

drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h

···
 #define SE_REG_FIELD_LIST_DCN4_01_COMMON(type) \
 	type COMPRESSED_PIXEL_FORMAT;\
 	type DP_VID_N_INTERVAL;\
-	type DIG_FIFO_OUTPUT_PIXEL_PER_CYCLE
-
+	type DIG_FIFO_OUTPUT_PIXEL_PER_CYCLE;\
+	type DP_STEER_FIFO_ENABLE
 struct dcn10_stream_encoder_shift {
 	SE_REG_FIELD_LIST_DCN1_0(uint8_t);
 	uint8_t HDMI_ACP_SEND;

drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c

···
 			MPC_RMU_3DLUT_SIZE, &s->lut3d_size);
 	}
 
-        REG_GET_2(MPCC_OGAM_CONTROL[mpcc_inst],
+	REG_GET_2(MPCC_OGAM_CONTROL[mpcc_inst],
 		  MPCC_OGAM_MODE_CURRENT, &s->rgam_mode,
 		  MPCC_OGAM_SELECT_CURRENT, &s->rgam_lut);
 }

drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.h

···
 
 void mpc3_init_mpcc(struct mpcc *mpcc, int mpcc_inst);
 
+void mpc3_mpc_init_single_inst(
+	struct mpc *mpc,
+	unsigned int mpcc_id);
+
 enum dc_lut_mode mpc3_get_ogam_current(
 	struct mpc *mpc,
 	int mpcc_id);

drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c

···
 	}
 }
 
-static void dcn32_program_compbuf_size(struct hubbub *hubbub, unsigned int compbuf_size_kb, bool safe_to_increase)
+void dcn32_program_compbuf_size(struct hubbub *hubbub, unsigned int compbuf_size_kb, bool safe_to_increase)
 {
 	struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub);
 	unsigned int compbuf_size_segments = (compbuf_size_kb + DCN32_CRB_SEGMENT_SIZE_KB - 1) / DCN32_CRB_SEGMENT_SIZE_KB;

drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.h

···
 
 void hubbub32_get_mall_en(struct hubbub *hubbub, unsigned int *mall_in_use);
 
+void dcn32_program_compbuf_size(struct hubbub *hubbub, unsigned int compbuf_size_kb, bool safe_to_increase);
+
 #endif

drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.c

···
 #include "basics/conversion.h"
 #include "dcn10/dcn10_cm_common.h"
 #include "dc.h"
+#include "dcn401/dcn401_mpc.h"
 
 #define REG(reg)\
 	mpc30->mpc_regs->reg
···
 	.power_on_mpc_mem_pwr = mpc3_power_on_ogam_lut,
 	.get_mpc_out_mux = mpc1_get_mpc_out_mux,
 	.set_bg_color = mpc1_set_bg_color,
+	.set_movable_cm_location = mpc401_set_movable_cm_location,
+	.populate_lut = mpc401_populate_lut,
 };
 
 

drivers/gpu/drm/amd/display/dc/dml/Makefile

···
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_rcflags)
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags)
 
+CFLAGS_$(AMDDALPATH)/dc/dml/dcn401/dcn401_fpu.o := $(dml_ccflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn401/dcn401_fpu.o := $(dml_rcflags)
+
 ifdef CONFIG_DRM_AMD_DC_FP
 DML += display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o
 DML += dcn10/dcn10_fpu.o
···
 DML += dcn314/dcn314_fpu.o
 DML += dcn35/dcn35_fpu.o
 DML += dcn351/dcn351_fpu.o
+DML += dcn401/dcn401_fpu.o
 DML += dsc/rc_calc_fpu.o
 DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o
 endif

drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c

···
  *
  */
 
-#define UNIT_TEST 0
-#if !UNIT_TEST
 #include "dc.h"
-#endif
 #include "../display_mode_lib.h"
 #include "display_mode_vba_314.h"
 #include "../dml_inline_defs.h"

drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h

···
 	unsigned int ref_freq_to_pix_freq;
 	unsigned int vratio_prefetch;
 	unsigned int vratio_prefetch_c;
+	unsigned int refcyc_per_tdlut_group;
 	unsigned int refcyc_per_pte_group_vblank_l;
 	unsigned int refcyc_per_pte_group_vblank_c;
 	unsigned int refcyc_per_meta_chunk_vblank_l;

drivers/gpu/drm/amd/display/dc/dml2/Makefile

···
 endif
 endif
 
+# DRIVER_BUILD is mostly used in DML2.1 source
+subdir-ccflags-y += -DDRIVER_BUILD=1
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/src/dml2_core
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/src/dml2_mcg/
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/src/dml2_dpmm/
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/src/dml2_pmo/
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/src/dml2_standalone_libraries/
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/src/inc
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/inc
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2/dml21/
+
 CFLAGS_$(AMDDALPATH)/dc/dml2/display_mode_core.o := $(dml2_ccflags) $(frame_warn_flag)
 CFLAGS_$(AMDDALPATH)/dc/dml2/display_mode_util.o := $(dml2_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml2/dml2_wrapper.o := $(dml2_ccflags)
···
 AMD_DAL_DML2 = $(addprefix $(AMDDALPATH)/dc/dml2/,$(DML2))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_DML2)
+
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml_top.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml_top_mcache.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_optimization := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.o := $(dml2_ccflags) -Wframe-larger-than=2048
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_factory.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_shared.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_standalone_libraries/lib_float_math.o := $(dml2_ccflags)
+
+
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml21_wrapper.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_translation_helper.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_ccflags)
+CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/inc/dml2_debug.o := $(dml2_ccflags)
+
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml_top.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml_top_mcache.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_optimization.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.o := $(dml2_rcflags) -Wframe-larger-than=2048
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_factory.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_shared.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_standalone_libraries/lib_float_math.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml21_wrapper.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/dml21_translation_helper.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_rcflags)
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/inc/dml2_debug.o := $(dml2_rcflags)
+
+DML21 := src/dml2_top/dml_top.o
+DML21 += src/dml2_top/dml_top_mcache.o
+DML21 += src/dml2_top/dml2_top_optimization.o
+DML21 += src/inc/dml2_debug.o
+DML21 += src/dml2_core/dml2_core_dcn4.o
+DML21 += src/dml2_core/dml2_core_factory.o
+DML21 += src/dml2_core/dml2_core_dcn4_calcs.o
+DML21 += src/dml2_core/dml2_core_shared.o
+DML21 += src/dml2_dpmm/dml2_dpmm_dcn4.o
+DML21 += src/dml2_dpmm/dml2_dpmm_factory.o
+DML21 += src/dml2_mcg/dml2_mcg_dcn4.o
+DML21 += src/dml2_mcg/dml2_mcg_factory.o
+DML21 += src/dml2_pmo/dml2_pmo_dcn3.o
+DML21 += src/dml2_pmo/dml2_pmo_dcn4.o
+DML21 += src/dml2_pmo/dml2_pmo_factory.o
+DML21 += src/dml2_pmo/dml2_pmo_dcn4_fams2.o
+DML21 += src/dml2_standalone_libraries/lib_float_math.o
+DML21 += dml21_translation_helper.o
+DML21 += dml21_wrapper.o
+DML21 += dml21_utils.o
+
+AMD_DAL_DML21 = $(addprefix $(AMDDALPATH)/dc/dml2/dml21/,$(DML21))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_DML21)
 

drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h

···
 	dml_project_dcn321 = 2,
 	dml_project_dcn35 = 3,
 	dml_project_dcn351 = 4,
+	dml_project_dcn401 = 5,
 };
 enum dml_prefetch_modes {
 	dml_prefetch_support_uclk_fclk_and_stutter_if_possible = 0,

drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c

···
 				stream->stream_id, plane_idx, &plane_id);
 		cfg_idx = find_disp_cfg_idx_by_plane_id(mapping, plane_id);
 		mpc_factor = (unsigned int)disp_cfg->hw.DPPPerSurface[cfg_idx];
+	} else if (ctx->architecture == dml2_architecture_21) {
+		if (ctx->config.svp_pstate.callbacks.get_stream_subvp_type(state, stream) == SUBVP_PHANTOM) {
+			struct dc_stream_state *main_stream;
+			struct dc_stream_status *main_stream_status;
+
+			/* get stream id of main stream */
+			main_stream = ctx->config.svp_pstate.callbacks.get_paired_subvp_stream(state, stream);
+			main_stream_status = ctx->config.callbacks.get_stream_status(state, main_stream);
+
+			/* get plane id for associated main plane */
+			get_plane_id(ctx, state, main_stream_status->plane_states[plane_idx],
+					main_stream->stream_id, plane_idx, &plane_id);
+		} else {
+			get_plane_id(ctx, state, status->plane_states[plane_idx],
+					stream->stream_id, plane_idx, &plane_id);
+		}
+
+		cfg_idx = find_disp_cfg_idx_by_plane_id(mapping, plane_id);
+		mpc_factor = ctx->v21.mode_programming.programming->plane_programming[cfg_idx].num_dpps_required;
 	} else {
 		mpc_factor = 1;
 		ASSERT(false);
···
 			break;
 		}
 	}
+	else if (ctx->architecture == dml2_architecture_21) {
+		if (ctx->config.svp_pstate.callbacks.get_stream_subvp_type(state, stream) == SUBVP_PHANTOM) {
+			struct dc_stream_state *main_stream;
+
+			/* get stream id of main stream */
+			main_stream = ctx->config.svp_pstate.callbacks.get_paired_subvp_stream(state, stream);
+
+			/* get cfg idx for associated main stream */
+			cfg_idx = find_disp_cfg_idx_by_stream_id(
+					mapping, main_stream->stream_id);
+		} else {
+			cfg_idx = find_disp_cfg_idx_by_stream_id(
+					mapping, stream->stream_id);
+		}
+
+		return ctx->v21.mode_programming.programming->stream_programming[cfg_idx].num_odms_required;
+	}
+
 	ASSERT(false);
 	return 1;
 }
···
 	unsigned int stream_id;
 
 	const unsigned int *ODMMode, *DPPPerSurface;
+	unsigned int odm_mode_array[__DML2_WRAPPER_MAX_STREAMS_PLANES__] = {0}, dpp_per_surface_array[__DML2_WRAPPER_MAX_STREAMS_PLANES__] = {0};
 	struct dc_pipe_mapping_scratch scratch;
 
 	if (ctx->config.map_dc_pipes_with_callbacks)
 		return map_dc_pipes_with_callbacks(
 				ctx, state, disp_cfg, mapping, existing_state);
 
-	ODMMode = (unsigned int *)disp_cfg->hw.ODMMode;
-	DPPPerSurface = disp_cfg->hw.DPPPerSurface;
+	if (ctx->architecture == dml2_architecture_21) {
+		/*
+		 * Extract ODM and DPP outputs from DML2.1 and map them in an array as required for pipe mapping in dml2_map_dc_pipes.
+		 * As data cannot be directly extracted in const pointers, assign these arrays to const pointers before proceeding to
+		 * maximize the reuse of existing code. Const pointers are required because dml2.0 dml_display_cfg_st is const.
+		 *
+		 */
+		for (i = 0; i < __DML2_WRAPPER_MAX_STREAMS_PLANES__; i++) {
+			odm_mode_array[i] = ctx->v21.mode_programming.programming->stream_programming[i].num_odms_required;
+			dpp_per_surface_array[i] = ctx->v21.mode_programming.programming->plane_programming[i].num_dpps_required;
+		}
+
+		ODMMode = (const unsigned int *)odm_mode_array;
+		DPPPerSurface = (const unsigned int *)dpp_per_surface_array;
+	} else {
+		ODMMode = (unsigned int *)disp_cfg->hw.ODMMode;
+		DPPPerSurface = disp_cfg->hw.DPPPerSurface;
+	}
 
 	for (stream_index = 0; stream_index < state->stream_count; stream_index++) {
 		memset(&scratch, 0, sizeof(struct dc_pipe_mapping_scratch));
···
 			scratch.odm_info.odm_factor = 1;
 		}
 
+		/* After DML2.1 update, ODM interpretation needs to change and is no longer same as for DML2.0.
+		 * This is not an issue with new resource management logic. This block ensure backcompat
+		 * with legacy pipe management with updated DML.
+		 * */
+		if (ctx->architecture == dml2_architecture_21) {
+			if (ODMMode[stream_disp_cfg_index] == 1) {
+				scratch.odm_info.odm_factor = 1;
+			} else if (ODMMode[stream_disp_cfg_index] == 2) {
+				scratch.odm_info.odm_factor = 2;
+			} else if (ODMMode[stream_disp_cfg_index] == 4) {
+				scratch.odm_info.odm_factor = 4;
+			} else {
+				ASSERT(false);
+				scratch.odm_info.odm_factor = 1;
+			}
+		}
 		calculate_odm_slices(state->streams[stream_index], scratch.odm_info.odm_factor, scratch.odm_info.odm_slice_end_x);
 
 		// If there are no planes, you still want to setup ODM...

drivers/gpu/drm/amd/display/dc/dml2/dml2_internal_types.h

···
 #include "dml2_wrapper.h"
 #include "dml2_policy.h"
 
+#include "dml_top.h"
+#include "dml21_wrapper.h"
 
 struct dml2_wrapper_optimize_configuration_params {
 	struct display_mode_lib_st *dml_core_ctx;
···
 
 enum dml2_architecture {
 	dml2_architecture_20,
+	dml2_architecture_21
+};
+
+struct prepare_mcache_programming_locals {
+	struct dml2_build_mcache_programming_in_out build_mcache_programming_params;
+};
+
+struct dml21_wrapper_scratch {
+	struct prepare_mcache_programming_locals prepare_mcache_locals;
+	struct pipe_ctx temp_pipe;
 };
 
 struct dml2_pipe_combine_factor {
···
 			struct dml2_wrapper_scratch scratch;
 			struct dcn_watermarks g6_temp_read_watermark_set;
 		} v20;
+		struct {
+			struct dml21_wrapper_scratch scratch;
+			struct dml2_initialize_instance_in_out dml_init;
+			struct dml2_display_cfg display_config;
+			struct dml2_check_mode_supported_in_out mode_support;
+			struct dml2_build_mode_programming_in_out mode_programming;
+			struct dml2_dml_to_dc_pipe_mapping dml_to_dc_pipe_mapping;
+		} v21;
 	};
 };
 

drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c

···
 void dml2_init_ip_params(struct dml2_context *dml2, const struct dc *in_dc, struct ip_params_st *out)
 {
 	switch (dml2->v20.dml_core_ctx.project) {
-
 	case dml_project_dcn32:
 	case dml_project_dcn321:
 	default:
···
 		out->config_return_buffer_segment_size_in_kbytes = 64; /*required, but not exist,, hard coded in dml2_translate_ip_params*/
 		break;
 
+	case dml_project_dcn401:
+		// Hardcoded values for DCN4m
+		out->vblank_nom_default_us = 668;	//600;
+		out->rob_buffer_size_kbytes = 192;	//128;
+		out->config_return_buffer_size_in_kbytes = 1344;	//1280;
+		out->config_return_buffer_segment_size_in_kbytes = 64;
+		out->compressed_buffer_segment_size_in_kbytes = 64;
+		out->meta_fifo_size_in_kentries = 22;
+		out->dpte_buffer_size_in_pte_reqs_luma = 68;
+		out->dpte_buffer_size_in_pte_reqs_chroma = 36;
+		out->gpuvm_max_page_table_levels = 4;
+		out->pixel_chunk_size_kbytes = 8;
+		out->alpha_pixel_chunk_size_kbytes = 4;
+		out->min_pixel_chunk_size_bytes = 1024;
+		out->writeback_chunk_size_kbytes = 8;
+		out->line_buffer_size_bits = 1171920;
+		out->max_line_buffer_lines = 32;
+		out->writeback_interface_buffer_size_kbytes = 90;
+		//Number of pipes after DCN Pipe harvesting
+		out->max_num_dpp = dml2->config.dcn_pipe_count;
+		out->max_num_otg = dml2->config.dcn_pipe_count;
+		out->max_num_wb = 1;
+		out->max_dchub_pscl_bw_pix_per_clk = 4;
+		out->max_pscl_lb_bw_pix_per_clk = 2;
+		out->max_lb_vscl_bw_pix_per_clk = 4;
+		out->max_vscl_hscl_bw_pix_per_clk = 4;
+		out->max_hscl_ratio = 6;
+		out->max_vscl_ratio = 6;
+		out->max_hscl_taps = 8;
+		out->max_vscl_taps = 8;
+		out->dispclk_ramp_margin_percent = 1;
+		out->dppclk_delay_subtotal = 47;
+		out->dppclk_delay_scl = 50;
+		out->dppclk_delay_scl_lb_only = 16;
+		out->dppclk_delay_cnvc_formatter = 28;
+		out->dppclk_delay_cnvc_cursor = 6;
+		out->dispclk_delay_subtotal = 125;
+		out->cursor_buffer_size = 24;	//16
+		out->cursor_chunk_size = 2;
+		out->max_inter_dcn_tile_repeaters = 8;
+		out->writeback_max_hscl_ratio = 1;
+		out->writeback_max_vscl_ratio = 1;
+		out->writeback_min_hscl_ratio = 1;
+		out->writeback_min_vscl_ratio = 1;
+		out->writeback_max_hscl_taps = 1;
+		out->writeback_max_vscl_taps = 1;
+		out->writeback_line_buffer_buffer_size = 0;
+		out->num_dsc = 4;
+		out->maximum_dsc_bits_per_component = 12;
+		out->maximum_pixels_per_line_per_dsc_unit = 5760;
+		out->dsc422_native_support = true;
+		out->dcc_supported = true;
+		out->ptoi_supported = false;
+
+		out->gpuvm_enable = false;
+		out->hostvm_enable = false;
+		out->cursor_64bpp_support = true;	//false;
+		out->dynamic_metadata_vm_enabled = false;
+
+		out->max_num_hdmi_frl_outputs = 1;
+		out->max_num_dp2p0_outputs = 4;		//2;
+		out->max_num_dp2p0_streams = 4;
+		break;
 	}
 }
 
···
 		out->dispclk_dppclk_vco_speed_mhz = 3600;
 		break;
 
+	case dml_project_dcn401:
+		out->pct_ideal_fabric_bw_after_urgent = 76;			//67;
+		out->max_avg_sdp_bw_use_normal_percent = 75;		//80;
+		out->max_avg_fabric_bw_use_normal_percent = 57;		//60;
+
+		out->urgent_out_of_order_return_per_channel_pixel_only_bytes = 0;	//4096;
+		out->urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 0;	//4096;
+		out->urgent_out_of_order_return_per_channel_vm_only_bytes = 0;		//4096;
+
+		out->num_chans = 16;
+		out->round_trip_ping_latency_dcfclk_cycles = 1000;	//263;
+		out->smn_latency_us = 0;							//2 us
+		out->mall_allocated_for_dcn_mbytes = dml2->config.mall_cfg.max_cab_allocation_bytes / 1048576; // 64;
+		break;
 	}
 	/* ---Overrides if available--- */
 	if (dml2->config.bbox_overrides.dram_num_chan)
···
 		p->in_states->state_array[1].fabricclk_mhz = 2250.0;
 		p->in_states->state_array[1].dcfclk_mhz = 1434.0;
 		p->in_states->state_array[1].dram_speed_mts = 1000 * transactions_per_mem_clock;
+		break;
+	case dml_project_dcn401:
+		p->in_states->num_states = 2;
+		transactions_per_mem_clock = 16;
+		p->in_states->state_array[0].socclk_mhz = 300;		//620.0;
+		p->in_states->state_array[0].dscclk_mhz = 666.667;	//716.667;
+		p->in_states->state_array[0].phyclk_mhz = 810;
+		p->in_states->state_array[0].phyclk_d18_mhz = 667;
+		p->in_states->state_array[0].phyclk_d32_mhz = 625;
+		p->in_states->state_array[0].dtbclk_mhz = 2000;		//1564.0;
+		p->in_states->state_array[0].fabricclk_mhz = 300;	//450.0;
+		p->in_states->state_array[0].dcfclk_mhz = 200;		//300.0;
+		p->in_states->state_array[0].dispclk_mhz = 2000;	//2150.0;
+		p->in_states->state_array[0].dppclk_mhz = 2000;		//2150.0;
+		p->in_states->state_array[0].dram_speed_mts = 97 * transactions_per_mem_clock; //100 *
+
+		p->in_states->state_array[0].urgent_latency_pixel_data_only_us = 4;
+		p->in_states->state_array[0].urgent_latency_pixel_mixed_with_vm_data_us = 0;
+		p->in_states->state_array[0].urgent_latency_vm_data_only_us = 0;
+		p->in_states->state_array[0].writeback_latency_us = 12;
+		p->in_states->state_array[0].urgent_latency_adjustment_fabric_clock_component_us = 1;
+		p->in_states->state_array[0].urgent_latency_adjustment_fabric_clock_reference_mhz = 1000;	//3000;
+		p->in_states->state_array[0].sr_exit_z8_time_us = 0;
+		p->in_states->state_array[0].sr_enter_plus_exit_z8_time_us = 0;
+		p->in_states->state_array[0].dram_clock_change_latency_us = 400;
+		p->in_states->state_array[0].use_ideal_dram_bw_strobe = true;
+		p->in_states->state_array[0].sr_exit_time_us = 15.70;	//42.97;
+		p->in_states->state_array[0].sr_enter_plus_exit_time_us = 20.20;	//49.94;
+		p->in_states->state_array[0].fclk_change_latency_us = 0;	//20;
+		p->in_states->state_array[0].usr_retraining_latency_us = 0;	//2;
+
+		p->in_states->state_array[1].socclk_mhz = 1600;		//1200.0;
+		p->in_states->state_array[1].fabricclk_mhz = 2500;	//2500.0;
+		p->in_states->state_array[1].dcfclk_mhz = 1800;		//1564.0;
+		p->in_states->state_array[1].dram_speed_mts = 1125 * transactions_per_mem_clock;
 		break;
 	}
 
···
 	default:
 		out->SurfaceTiling[location] = (enum dml_swizzle_mode)in->tiling_info.gfx9.swizzle;
 		break;
+	case dml_project_dcn401:
+		// Temporary use gfx11 swizzle in dml, until proper dml for DCN4x is integrated/implemented
+		switch (in->tiling_info.gfx_addr3.swizzle) {
+		case DC_ADDR3_SW_4KB_2D:
+		case DC_ADDR3_SW_64KB_2D:
+		case DC_ADDR3_SW_256KB_2D:
+		default:
+			out->SurfaceTiling[location] = dml_sw_64kb_r_x;
+			break;
+		case DC_ADDR3_SW_LINEAR:
+			out->SurfaceTiling[location] = dml_sw_linear;
+			break;
+		}
 	}
 
 	switch (in->format) {

drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c

···
 #include "dml2_translation_helper.h"
 #include "dml2_mall_phantom.h"
 #include "dml2_dc_resource_mgmt.h"
+#include "dml21_wrapper.h"
 
 
 static void initialize_dml2_ip_params(struct dml2_context *dml2, const struct dc *in_dc, struct ip_params_st *out)
···
 		return false;
 	dml2_apply_debug_options(in_dc, dml2);
 
+	/* DML2.1 validation path */
+	if (dml2->architecture == dml2_architecture_21) {
+		out = dml21_validate(in_dc, context, dml2, fast_validate);
+		return out;
+	}
 
 	/* Use dml_validate_only for fast_validate path */
 	if (fast_validate)
···
 
 static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2)
 {
+	// TODO : Temporarily add DCN_VERSION_3_2 for N-1 validation. Remove DCN_VERSION_3_2 after N-1 validation phase is complete.
+        if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01 || in_dc->ctx->dce_version == DCN_VERSION_3_2)) {
+                dml21_reinit(in_dc, dml2, config);
+        }
 
 	// Store config options
 	(*dml2)->config = *config;
···
 	case DCN_VERSION_3_21:
 		(*dml2)->v20.dml_core_ctx.project = dml_project_dcn321;
 		break;
+	case DCN_VERSION_4_01:
+		(*dml2)->v20.dml_core_ctx.project = dml_project_dcn401;
+		break;
 	default:
 		(*dml2)->v20.dml_core_ctx.project = dml_project_default;
 		break;
···
 
 bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2)
 {
+	DC_FP_START();
+	// TODO : Temporarily add DCN_VERSION_3_2 for N-1 validation. Remove DCN_VERSION_3_2 after N-1 validation phase is complete.
+	if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01 || in_dc->ctx->dce_version == DCN_VERSION_3_2)) {
+		return dml21_create(in_dc, dml2, config);
+	}
+
 	// Allocate Mode Lib Ctx
 	*dml2 = dml2_allocate_memory();
 
···
 
 	dml2_init(in_dc, config, dml2);
 
+	DC_FP_END();
 	return true;
 }
 
···
 void dml2_copy(struct dml2_context *dst_dml2,
 	struct dml2_context *src_dml2)
 {
+	if (src_dml2->architecture == dml2_architecture_21) {
+		dml21_copy(dst_dml2, src_dml2);
+		return;
+	}
 	/* copy Mode Lib Ctx */
 	memcpy(dst_dml2, src_dml2, sizeof(struct dml2_context));
 }
···
 bool dml2_create_copy(struct dml2_context **dst_dml2,
 	struct dml2_context *src_dml2)
 {
+	if (src_dml2->architecture == dml2_architecture_21)
+		return dml21_create_copy(dst_dml2, src_dml2);
 	/* Allocate Mode Lib Ctx */
 	*dst_dml2 = dml2_allocate_memory();
 
···
 				 const struct dml2_configuration_options *config,
 				 struct dml2_context **dml2)
 {
+	// TODO : Temporarily add DCN_VERSION_3_2 for N-1 validation. Remove DCN_VERSION_3_2 after N-1 validation phase is complete.
+        if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01 || in_dc->ctx->dce_version == DCN_VERSION_3_2)) {
+                dml21_reinit(in_dc, dml2, config);
+        }
 
 	dml2_init(in_dc, config, dml2);
 }

drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h

···
 	struct dml2_clks_limit_table clks_table;
 };
 
+enum dml2_force_pstate_methods {
+	dml2_force_pstate_method_auto = 0,
+	dml2_force_pstate_method_vactive,
+	dml2_force_pstate_method_vblank,
+	dml2_force_pstate_method_drr,
+	dml2_force_pstate_method_subvp,
+};
+
 struct dml2_configuration_options {
 	int dcn_pipe_count;
 	bool use_native_pstate_optimization;
···
 	struct dml2_soc_bbox_overrides bbox_overrides;
 	unsigned int max_segments_per_hubp;
 	unsigned int det_segment_size;
+	/* Only for debugging purposes when initializing SOCBB params via tool for DML21. */
+	struct socbb_ip_params_external *external_socbb_ip_params;
+	struct {
+		bool force_pstate_method_enable;
+		enum dml2_force_pstate_methods force_pstate_method_value;
+	} pmo;
 	bool map_dc_pipes_with_callbacks;
 
 	bool use_clock_dc_limits;

drivers/gpu/drm/amd/display/dc/dpp/Makefile

···
 
 ###############################################################################
 
+DPP_DCN401 = dcn401_dpp.o dcn401_dpp_cm.o dcn401_dpp_dscl.o
+
+AMD_DAL_DPP_DCN401 = $(addprefix $(AMDDALPATH)/dc/dpp/dcn401/,$(DPP_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_DPP_DCN401)
+
 endif

drivers/gpu/drm/amd/display/dc/dpp/dcn20/dcn20_dpp.c

···
 
 	return true;
 }
+/*compute the maximum number of lines that we can fit in the line buffer*/
+void dscl2_spl_calc_lb_num_partitions(
+		bool alpha_en,
+		const struct spl_scaler_data *scl_data,
+		enum lb_memory_config lb_config,
+		int *num_part_y,
+		int *num_part_c)
+{
+	int memory_line_size_y, memory_line_size_c, memory_line_size_a,
+	lb_memory_size, lb_memory_size_c, lb_memory_size_a, num_partitions_a;
 
+	int line_size = scl_data->viewport.width < scl_data->recout.width ?
+			scl_data->viewport.width : scl_data->recout.width;
+	int line_size_c = scl_data->viewport_c.width < scl_data->recout.width ?
+			scl_data->viewport_c.width : scl_data->recout.width;
+
+	if (line_size == 0)
+		line_size = 1;
+
+	if (line_size_c == 0)
+		line_size_c = 1;
+
+	memory_line_size_y = (line_size + 5) / 6; /* +5 to ceil */
+	memory_line_size_c = (line_size_c + 5) / 6; /* +5 to ceil */
+	memory_line_size_a = (line_size + 5) / 6; /* +5 to ceil */
+
+	if (lb_config == LB_MEMORY_CONFIG_1) {
+		lb_memory_size = 970;
+		lb_memory_size_c = 970;
+		lb_memory_size_a = 970;
+	} else if (lb_config == LB_MEMORY_CONFIG_2) {
+		lb_memory_size = 1290;
+		lb_memory_size_c = 1290;
+		lb_memory_size_a = 1290;
+	} else if (lb_config == LB_MEMORY_CONFIG_3) {
+		/* 420 mode: using 3rd mem from Y, Cr and Cb */
+		lb_memory_size = 970 + 1290 + 484 + 484 + 484;
+		lb_memory_size_c = 970 + 1290;
+		lb_memory_size_a = 970 + 1290 + 484;
+	} else {
+		lb_memory_size = 970 + 1290 + 484;
+		lb_memory_size_c = 970 + 1290 + 484;
+		lb_memory_size_a = 970 + 1290 + 484;
+	}
+	*num_part_y = lb_memory_size / memory_line_size_y;
+	*num_part_c = lb_memory_size_c / memory_line_size_c;
+	num_partitions_a = lb_memory_size_a / memory_line_size_a;
+
+	if (alpha_en
+			&& (num_partitions_a < *num_part_y))
+		*num_part_y = num_partitions_a;
+
+	if (*num_part_y > 64)
+		*num_part_y = 64;
+	if (*num_part_c > 64)
+		*num_part_c = 64;
+}

drivers/gpu/drm/amd/display/dc/dpp/dcn20/dcn20_dpp.h

···
 #define __DCN20_DPP_H__
 
 #include "dcn10/dcn10_dpp.h"
-
+#include "spl/dc_spl_types.h"
 #define TO_DCN20_DPP(dpp)\
 	container_of(dpp, struct dcn20_dpp, base)
 
···
 			enum lb_memory_config lb_config,
 			int *num_part_y,
 			int *num_part_c);
+
+void dscl2_spl_calc_lb_num_partitions(
+	bool alpha_en,
+	const struct spl_scaler_data *scl_data,
+	enum lb_memory_config lb_config,
+	int *num_part_y,
+	int *num_part_c);
 
 void dpp2_set_cursor_attributes(
 		struct dpp *dpp_base,

drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.h

···
 	SRI(CM_BLNDGAM_RAMB_START_SLOPE_CNTL_R, CM, id),\
 	SRI(CM_BLNDGAM_LUT_CONTROL, CM, id)
 
-
-
 #define DPP_REG_LIST_SH_MASK_DCN30_COMMON(mask_sh)\
 	TF_SF(CM0_CM_MEM_PWR_STATUS, GAMCOR_MEM_PWR_STATE, mask_sh),\
 	TF_SF(CM0_CM_DEALPHA, CM_DEALPHA_EN, mask_sh),\
···
 		struct scaler_data *scl_data,
 		const struct scaling_taps *in_taps);
 
-void dpp3_cnv_setup (
+void dpp3_cnv_setup(
 		struct dpp *dpp_base,
 		enum surface_pixel_format format,
 		enum expansion_mode mode,

drivers/gpu/drm/amd/display/dc/dpp/dcn32/dcn32_dpp.c

···
 
 	return true;
 }
+void dscl32_spl_calc_lb_num_partitions(
+		bool alpha_en,
+		const struct spl_scaler_data *scl_data,
+		enum lb_memory_config lb_config,
+		int *num_part_y,
+		int *num_part_c)
+{
+	int memory_line_size_y, memory_line_size_c, memory_line_size_a,
+	lb_memory_size, lb_memory_size_c, lb_memory_size_a, num_partitions_a;
+
+	int line_size = scl_data->viewport.width < scl_data->recout.width ?
+			scl_data->viewport.width : scl_data->recout.width;
+	int line_size_c = scl_data->viewport_c.width < scl_data->recout.width ?
+			scl_data->viewport_c.width : scl_data->recout.width;
+
+	if (line_size == 0)
+		line_size = 1;
+
+	if (line_size_c == 0)
+		line_size_c = 1;
+
+	memory_line_size_y = (line_size + 5) / 6; /* +5 to ceil */
+	memory_line_size_c = (line_size_c + 5) / 6; /* +5 to ceil */
+	memory_line_size_a = (line_size + 5) / 6; /* +5 to ceil */
+
+	if (lb_config == LB_MEMORY_CONFIG_1) {
+		lb_memory_size = 970;
+		lb_memory_size_c = 970;
+		lb_memory_size_a = 970;
+	} else if (lb_config == LB_MEMORY_CONFIG_2) {
+		lb_memory_size = 1290;
+		lb_memory_size_c = 1290;
+		lb_memory_size_a = 1290;
+	} else if (lb_config == LB_MEMORY_CONFIG_3) {
+		if (scl_data->viewport.width  == scl_data->h_active &&
+			scl_data->viewport.height == scl_data->v_active) {
+			/* 420 mode: luma using all 3 mem from Y, plus 3rd mem from Cr and Cb */
+			/* use increased LB size for calculation only if Scaler not enabled */
+			lb_memory_size = 970 + 1290 + 1170 + 1170 + 1170;
+			lb_memory_size_c = 970 + 1290;
+			lb_memory_size_a = 970 + 1290 + 1170;
+		} else {
+			/* 420 mode: luma using all 3 mem from Y, plus 3rd mem from Cr and Cb */
+			lb_memory_size = 970 + 1290 + 484 + 484 + 484;
+			lb_memory_size_c = 970 + 1290;
+			lb_memory_size_a = 970 + 1290 + 484;
+		}
+	} else {
+		if (scl_data->viewport.width  == scl_data->h_active &&
+			scl_data->viewport.height == scl_data->v_active) {
+			/* use increased LB size for calculation only if Scaler not enabled */
+			lb_memory_size = 970 + 1290 + 1170;
+			lb_memory_size_c = 970 + 1290 + 1170;
+			lb_memory_size_a = 970 + 1290 + 1170;
+		} else {
+			lb_memory_size = 970 + 1290 + 484;
+			lb_memory_size_c = 970 + 1290 + 484;
+			lb_memory_size_a = 970 + 1290 + 484;
+		}
+	}
+	*num_part_y = lb_memory_size / memory_line_size_y;
+	*num_part_c = lb_memory_size_c / memory_line_size_c;
+	num_partitions_a = lb_memory_size_a / memory_line_size_a;
+
+	if (alpha_en
+			&& (num_partitions_a < *num_part_y))
+		*num_part_y = num_partitions_a;
+
+	if (*num_part_y > 32)
+		*num_part_y = 32;
+	if (*num_part_c > 32)
+		*num_part_c = 32;
+}

drivers/gpu/drm/amd/display/dc/dpp/dcn32/dcn32_dpp.h

···
 
 #include "dcn20/dcn20_dpp.h"
 #include "dcn30/dcn30_dpp.h"
+#include "spl/dc_spl_types.h"
 
 bool dpp32_construct(struct dcn3_dpp *dpp3,
 	struct dc_context *ctx,
···
 	const struct dcn3_dpp_registers *tf_regs,
 	const struct dcn3_dpp_shift *tf_shift,
 	const struct dcn3_dpp_mask *tf_mask);
+
+void dscl32_spl_calc_lb_num_partitions(
+		bool alpha_en,
+		const struct spl_scaler_data *scl_data,
+		enum lb_memory_config lb_config,
+		int *num_part_y,
+		int *num_part_c);
 
 #endif /* __DCN32_DPP_H__ */

drivers/gpu/drm/amd/display/dc/dsc/Makefile

···
 
 AMD_DISPLAY_FILES += $(addprefix $(AMDDALPATH)/dc/dsc/dcn35/,$(DSC_DCN35))
 
+###############################################################################
+# DCN401
+###############################################################################
 
+DSC_DCN401 += dcn401_dsc.o
 
-endif
+AMD_DISPLAY_FILES += $(addprefix $(AMDDALPATH)/dc/dsc/dcn401/,$(DSC_DCN401))
+
 
 DSC = dc_dsc.o rc_calc.o rc_calc_dpi.o
 
···
 
 AMD_DISPLAY_FILES += $(AMD_DAL_DSC)
 
+endif

drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c

···
 #define MIN(X, Y) ((X) < (Y) ? (X) : (Y))
 #endif
 
-/* Need to account for padding due to pixel-to-symbol packing
- * for uncompressed 128b/132b streams.
- */
-static uint32_t apply_128b_132b_stream_overhead(
-	const struct dc_crtc_timing *timing, const uint32_t kbps)
-{
-	uint32_t total_kbps = kbps;
-
-	if (disable_128b_132b_stream_overhead)
-		return kbps;
-
-	if (!timing->flags.DSC) {
-		struct fixed31_32 bpp;
-		struct fixed31_32 overhead_factor;
-
-		bpp = dc_fixpt_from_int(kbps);
-		bpp = dc_fixpt_div_int(bpp, timing->pix_clk_100hz / 10);
-
-		/* Symbols_per_HActive = HActive * bpp / (4 lanes * 32-bit symbol size)
-		 * Overhead_factor = ceil(Symbols_per_HActive) / Symbols_per_HActive
-		 */
-		overhead_factor = dc_fixpt_from_int(timing->h_addressable);
-		overhead_factor = dc_fixpt_mul(overhead_factor, bpp);
-		overhead_factor = dc_fixpt_div_int(overhead_factor, 128);
-		overhead_factor = dc_fixpt_div(
-			dc_fixpt_from_int(dc_fixpt_ceil(overhead_factor)),
-			overhead_factor);
-
-		total_kbps = dc_fixpt_ceil(
-			dc_fixpt_mul_int(overhead_factor, total_kbps));
-	}
-
-	return total_kbps;
-}
-
-uint32_t dc_bandwidth_in_kbps_from_timing(
-	const struct dc_crtc_timing *timing,
-	const enum dc_link_encoding_format link_encoding)
-{
-	uint32_t bits_per_channel = 0;
-	uint32_t kbps;
-
-	if (timing->flags.DSC)
-		return dc_dsc_stream_bandwidth_in_kbps(timing,
-				timing->dsc_cfg.bits_per_pixel,
-				timing->dsc_cfg.num_slices_h,
-				timing->dsc_cfg.is_dp);
-
-	switch (timing->display_color_depth) {
-	case COLOR_DEPTH_666:
-		bits_per_channel = 6;
-		break;
-	case COLOR_DEPTH_888:
-		bits_per_channel = 8;
-		break;
-	case COLOR_DEPTH_101010:
-		bits_per_channel = 10;
-		break;
-	case COLOR_DEPTH_121212:
-		bits_per_channel = 12;
-		break;
-	case COLOR_DEPTH_141414:
-		bits_per_channel = 14;
-		break;
-	case COLOR_DEPTH_161616:
-		bits_per_channel = 16;
-		break;
-	default:
-		ASSERT(bits_per_channel != 0);
-		bits_per_channel = 8;
-		break;
-	}
-
-	kbps = timing->pix_clk_100hz / 10;
-	kbps *= bits_per_channel;
-
-	if (timing->flags.Y_ONLY != 1) {
-		/*Only YOnly make reduce bandwidth by 1/3 compares to RGB*/
-		kbps *= 3;
-		if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
-			kbps /= 2;
-		else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422)
-			kbps = kbps * 2 / 3;
-	}
-
-	if (link_encoding == DC_LINK_ENCODING_DP_128b_132b)
-		kbps = apply_128b_132b_stream_overhead(timing, kbps);
-
-	if (link_encoding == DC_LINK_ENCODING_HDMI_FRL &&
-			timing->vic == 0 && timing->hdmi_vic == 0 &&
-			timing->frl_uncompressed_video_bandwidth_in_kbps != 0)
-		kbps = timing->frl_uncompressed_video_bandwidth_in_kbps;
-
-	return kbps;
-}
-
-
 /* Forward Declerations */
 static bool decide_dsc_bandwidth_range(
 		const uint32_t min_bpp_x16,
···
 void dc_set_disable_128b_132b_stream_overhead(bool disable)
 {
 	disable_128b_132b_stream_overhead = disable;
+}
+
+bool dc_get_disable_128b_132b_stream_overhead(void)
+{
+	return disable_128b_132b_stream_overhead;
 }
 
 void dc_dsc_get_default_config_option(const struct dc *dc, struct dc_dsc_config_options *options)

drivers/gpu/drm/amd/display/dc/gpio/Makefile

···
 AMD_DAL_GPIO_DCN32 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn32/,$(GPIO_DCN32))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN32)
+
+###############################################################################
+# DCN 4.01
+###############################################################################
+GPIO_DCN401 = hw_translate_dcn401.o hw_factory_dcn401.o
+
+AMD_DAL_GPIO_DCN401 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn401/,$(GPIO_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN401)
+

drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c

···
 #include "dcn30/hw_factory_dcn30.h"
 #include "dcn315/hw_factory_dcn315.h"
 #include "dcn32/hw_factory_dcn32.h"
+#include "dcn401/hw_factory_dcn401.h"
 
 bool dal_hw_factory_init(
 	struct hw_factory *factory,
···
 	case DCN_VERSION_3_5:
 	case DCN_VERSION_3_51:
 		dal_hw_factory_dcn32_init(factory);
+		return true;
+	case DCN_VERSION_4_01:
+		dal_hw_factory_dcn401_init(factory);
 		return true;
 	default:
 		ASSERT_CRITICAL(false);

drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c

···
 #include "dcn30/hw_translate_dcn30.h"
 #include "dcn315/hw_translate_dcn315.h"
 #include "dcn32/hw_translate_dcn32.h"
+#include "dcn401/hw_translate_dcn401.h"
 
 /*
  * This unit
···
 	case DCN_VERSION_3_5:
 	case DCN_VERSION_3_51:
 		dal_hw_translate_dcn32_init(translate);
+		return true;
+	case DCN_VERSION_4_01:
+		dal_hw_translate_dcn401_init(translate);
 		return true;
 	default:
 		BREAK_TO_DEBUGGER();

drivers/gpu/drm/amd/display/dc/hwss/Makefile

···
 
 ###############################################################################
 
+HWSS_DCN401 = dcn401_hwseq.o dcn401_init.o
+
+AMD_DAL_HWSS_DCN401 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn401/,$(HWSS_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN401)
 endif

drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c

···
 			"  rc_pg_flc  rc_mc_fll  rc_mc_flc  pr_nom_l   pr_nom_c   rc_pg_nl   rc_pg_nc "
 			"  mr_nom_l   mr_nom_c   rc_mc_nl   rc_mc_nc   rc_ld_pl   rc_ld_pc   rc_ld_l  "
 			"  rc_ld_c    cha_cur0   ofst_cur1  cha_cur1   vr_af_vc0  ddrq_limt  x_rt_dlay"
-			"  x_rp_dlay  x_rr_sfl\n");
+			"  x_rp_dlay  x_rr_sfl  rc_td_grp\n");
+
 	for (i = 0; i < pool->pipe_count; i++) {
 		struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state);
 		struct _vcs_dpi_display_dlg_regs_st *dlg_regs = &s->dlg_attr;
···
 		if (!s->blank_en)
 			DTN_INFO("[%2d]:  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh"
 				"  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh"
-				"  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh\n",
+				"  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh  %8xh %xh\n",
 				pool->hubps[i]->inst, dlg_regs->refcyc_h_blank_end, dlg_regs->dlg_vblank_end, dlg_regs->min_dst_y_next_start,
 				dlg_regs->refcyc_per_htotal, dlg_regs->refcyc_x_after_scaler, dlg_regs->dst_y_after_scaler,
 				dlg_regs->dst_y_prefetch, dlg_regs->dst_y_per_vm_vblank, dlg_regs->dst_y_per_row_vblank,
···
 				dlg_regs->refcyc_per_line_delivery_c, dlg_regs->chunk_hdl_adjust_cur0, dlg_regs->dst_y_offset_cur1,
 				dlg_regs->chunk_hdl_adjust_cur1, dlg_regs->vready_after_vcount0, dlg_regs->dst_y_delta_drq_limit,
 				dlg_regs->xfc_reg_transfer_delay, dlg_regs->xfc_reg_precharge_delay,
-				dlg_regs->xfc_reg_remote_surface_flip_latency);
+				dlg_regs->xfc_reg_remote_surface_flip_latency, dlg_regs->refcyc_per_tdlut_group);
 	}
 
 	DTN_INFO("========TTU========\n");

drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c

···
  * gsl_0 <=> pipe_ctx->stream_res.gsl_group == 1
  * Using a magic value like -1 would require tracking all inits/resets
  */
- void dcn20_setup_gsl_group_as_lock(
+void dcn20_setup_gsl_group_as_lock(
 		const struct dc *dc,
 		struct pipe_ctx *pipe_ctx,
 		bool enable)
···
 				plane_state->color_space,
 				NULL);
 
+		if (dpp->funcs->set_cursor_matrix) {
+			dpp->funcs->set_cursor_matrix(dpp,
+				plane_state->color_space,
+				plane_state->cursor_csc_color_matrix);
+		}
 		if (dpp->funcs->dpp_program_bias_and_scale) {
 			//TODO :for CNVC set scale and bias registers if necessary
 			build_prescale_params(&bns_params, plane_state);
···
 		if (dc->res_pool->hubbub->funcs->program_det_size)
 			dc->res_pool->hubbub->funcs->program_det_size(
 				dc->res_pool->hubbub, pipe_ctx->plane_res.hubp->inst, pipe_ctx->det_buffer_size_kb);
+
+		if (dc->res_pool->hubbub->funcs->program_det_segments)
+			dc->res_pool->hubbub->funcs->program_det_segments(
+				dc->res_pool->hubbub, pipe_ctx->plane_res.hubp->inst, pipe_ctx->hubp_regs.det_size);
 	}
 
 	if (pipe_ctx->update_flags.raw || pipe_ctx->plane_state->update_flags.raw || pipe_ctx->stream->update_flags.raw)
···
 			|| pipe_ctx->plane_state->update_flags.bits.hdr_mult)
 		hws->funcs.set_hdr_multiplier(pipe_ctx);
 
+	if (hws->funcs.populate_mcm_luts) {
+		hws->funcs.populate_mcm_luts(dc, pipe_ctx, pipe_ctx->plane_state->mcm_luts,
+				pipe_ctx->plane_state->lut_bank_a);
+		pipe_ctx->plane_state->lut_bank_a = !pipe_ctx->plane_state->lut_bank_a;
+	}
 	if (pipe_ctx->update_flags.bits.enable ||
 	    pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change ||
 	    pipe_ctx->plane_state->update_flags.bits.gamma_change ||
···
 					(context->res_ctx.pipe_ctx[i].plane_state && dc_state_get_pipe_subvp_type(context, &context->res_ctx.pipe_ctx[i]) == SUBVP_PHANTOM))) {
 				if (hubbub->funcs->program_det_size)
 					hubbub->funcs->program_det_size(hubbub, dc->current_state->res_ctx.pipe_ctx[i].plane_res.hubp->inst, 0);
+				if (dc->res_pool->hubbub->funcs->program_det_segments)
+					dc->res_pool->hubbub->funcs->program_det_segments(hubbub, dc->current_state->res_ctx.pipe_ctx[i].plane_res.hubp->inst, 0);
 			}
 			hws->funcs.plane_atomic_disconnect(dc, dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]);
 			DC_LOG_DC("Reset mpcc for pipe %d\n", dc->current_state->res_ctx.pipe_ctx[i].pipe_idx);

drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c

···
 #include "dcn30/dcn30_hwseq.h"
 #include "dcn31/dcn31_hwseq.h"
 #include "dcn32/dcn32_hwseq.h"
+#include "dcn401/dcn401_hwseq.h"
 #include "dcn32_init.h"
 
 static const struct hw_sequencer_funcs dcn32_funcs = {
···
 	.is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy,
 	.apply_single_controller_ctx_to_hw = dce110_apply_single_controller_ctx_to_hw,
 	.reset_back_end_for_pipe = dcn20_reset_back_end_for_pipe,
+	.populate_mcm_luts = dcn401_populate_mcm_luts,
 };
 
 void dcn32_hw_sequencer_init_functions(struct dc *dc)

drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h

···
 	uint8_t subvp_index;
 };
 
+struct fams2_global_control_lock_fast_params {
+	struct dc *dc;
+	bool is_required;
+	bool lock;
+};
+
 union block_sequence_params {
 	struct update_plane_addr_params update_plane_addr_params;
 	struct subvp_pipe_control_lock_fast_params subvp_pipe_control_lock_fast_params;
···
 	struct set_output_csc_params set_output_csc_params;
 	struct set_ocsc_default_params set_ocsc_default_params;
 	struct subvp_save_surf_addr subvp_save_surf_addr;
+	struct fams2_global_control_lock_fast_params fams2_global_control_lock_fast_params;
 };
 
 enum block_sequence_func {
···
 	MPC_SET_OUTPUT_CSC,
 	MPC_SET_OCSC_DEFAULT,
 	DMUB_SUBVP_SAVE_SURF_ADDR,
+	DMUB_FAMS2_GLOBAL_CONTROL_LOCK_FAST,
+
 };
 
 struct block_sequence {
···
 	bool (*is_pipe_topology_transition_seamless)(struct dc *dc,
 			const struct dc_state *cur_ctx,
 			const struct dc_state *new_ctx);
+	void (*fams2_global_control_lock)(struct dc *dc,
+			struct dc_state *context,
+			bool lock);
+	void (*fams2_update_config)(struct dc *dc,
+			struct dc_state *context,
+			bool enable);
+	void (*fams2_global_control_lock_fast)(union block_sequence_params *params);
 	void (*set_long_vtotal)(struct pipe_ctx **pipe_ctx, int num_pipes, uint32_t v_total_min, uint32_t v_total_max);
 };
 
···
 		struct tg_color *color);
 
 void get_subvp_visual_confirm_color(
+	struct pipe_ctx *pipe_ctx,
+	struct tg_color *color);
+
+void get_fams2_visual_confirm_color(
+	struct dc *dc,
+	struct dc_state *context,
 	struct pipe_ctx *pipe_ctx,
 	struct tg_color *color);
 

drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h

···
 	void (*reset_back_end_for_pipe)(struct dc *dc,
 			struct pipe_ctx *pipe_ctx,
 			struct dc_state *context);
+	void (*populate_mcm_luts)(struct dc *dc,
+			struct pipe_ctx *pipe_ctx,
+			struct dc_cm2_func_luts mcm_luts,
+			bool lut_bank_a);
 };
 
 struct dce_hwseq {

drivers/gpu/drm/amd/display/dc/inc/core_types.h

···
 #include "panel_cntl.h"
 #include "dmub/inc/dmub_cmd.h"
 #include "pg_cntl.h"
+#include "spl/dc_spl.h"
 
 #define MAX_CLOCK_SOURCES 7
 #define MAX_SVP_PHANTOM_STREAMS 2
···
 /********** DAL Core*********************/
 #include "transform.h"
 #include "dpp.h"
+
+#include "dml2/dml21/inc/dml_top_dchub_registers.h"
+#include "dml2/dml21/inc/dml_top_types.h"
 
 struct resource_pool;
 struct dc_state;
···
 				struct dc *dc,
 				struct dc_state *new_ctx,
 				struct dc_stream_state *stream);
+
 	enum dc_status (*patch_unknown_plane_state)(
 			struct dc_plane_state *plane_state);
 
···
 			struct resource_context *res_ctx,
 			const struct resource_pool *pool,
 			struct dc_stream_state *stream);
+
 	void (*populate_dml_writeback_from_context)(
 			struct dc *dc,
 			struct resource_context *res_ctx,
···
 			struct dc_state *context,
 			display_e2e_pipe_params_st *pipes,
 			int pipe_cnt);
+
 	void (*update_bw_bounding_box)(
 			struct dc *dc,
 			struct clk_bw_params *bw_params);
···
 	struct abm *abm;
 	struct dmcu *dmcu;
 	struct dmub_psr *psr;
-
 	struct dmub_replay *replay;
 
 	struct abm *multiple_abms[MAX_PIPES];
···
 struct plane_resource {
 	/* scl_data is scratch space required to program a plane */
 	struct scaler_data scl_data;
+	/* Below pointers to hw objects are required to enable the plane */
+	/* spl_in and spl_out are the input and output structures for SPL
+	 * which are required when using Scaler Programming Library
+	 * these are scratch spaces needed when programming a plane
+	 */
+	struct spl_in spl_in;
+	struct spl_out spl_out;
 	/* Below pointers to hw objects are required to enable the plane */
 	struct hubp *hubp;
 	struct mem_input *mi;
···
 	int det_buffer_size_kb;
 	bool unbounded_req;
 	unsigned int surface_size_in_mall_bytes;
+	struct dml2_dchub_per_pipe_register_set hubp_regs;
+	struct dml2_hubp_pipe_mcache_regs mcache_regs;
 
 	struct dwbc *dwbc;
 	struct mcif_wb *mcif_wb;
···
 	unsigned int mall_subvp_size_bytes;
 	unsigned int legacy_svp_drr_stream_index;
 	bool legacy_svp_drr_stream_index_valid;
+	struct dml2_mcache_surface_allocation mcache_allocations[DML2_MAX_PLANES];
+	struct dmub_fams2_stream_static_state fams2_stream_params[DML2_MAX_PLANES];
+	unsigned fams2_stream_count;
+	struct dml2_display_arb_regs arb_regs;
 };
 
 union bw_output {

drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h

···
 #define WM_C 2
 #define WM_D 3
 #define WM_SET_COUNT 4
+#define WM_1A 2
+#define WM_1B 3
 
 #define DCN_MINIMUM_DISPCLK_Khz 100000
 #define DCN_MINIMUM_DPPCLK_Khz 100000
···
 
 struct dummy_pstate_entry {
 	unsigned int dram_speed_mts;
-	double dummy_pstate_latency_us;
+	unsigned int dummy_pstate_latency_us;
 };
 
 struct clk_bw_params {
 	unsigned int vram_type;
 	unsigned int num_channels;
 	unsigned int dram_channel_width_bytes;
- 	unsigned int dispclk_vco_khz;
+	unsigned int dispclk_vco_khz;
 	unsigned int dc_mode_softmax_memclk;
 	unsigned int max_memclk_mhz;
 	struct clk_limit_table clk_table;

drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h

···
 #define CLK_COMMON_REG_LIST_DCN_BASE() \
 	SR(DENTIST_DISPCLK_CNTL)
 
-#define VBIOS_SMU_MSG_BOX_REG_LIST_RV() \
-	.MP1_SMN_C2PMSG_91 = mmMP1_SMN_C2PMSG_91, \
-	.MP1_SMN_C2PMSG_83 = mmMP1_SMN_C2PMSG_83, \
-	.MP1_SMN_C2PMSG_67 = mmMP1_SMN_C2PMSG_67
-
 #define CLK_COMMON_REG_LIST_DCN_201() \
 	SR(DENTIST_DISPCLK_CNTL), \
 	CLK_SRI(CLK4_CLK_PLL_REQ, CLK4, 0), \
···
 	CLK_SRI(CLK3_CLK2_DFS_CNTL, CLK3, 0)
 
 #define CLK_REG_LIST_DCN3()	  \
-	CLK_COMMON_REG_LIST_DCN_BASE(), \
+	SR(DENTIST_DISPCLK_CNTL), \
 	CLK_SRI(CLK0_CLK_PLL_REQ,   CLK02, 0), \
 	CLK_SRI(CLK0_CLK2_DFS_CNTL, CLK02, 0)
 
···
 	type DENTIST_DISPCLK_WDIVIDER; \
 	type DENTIST_DISPCLK_CHG_DONE;
 
-/*
- ***************************************************************************************
- ****************** Clock Manager Private Structures ***********************************
- ***************************************************************************************
- */
 #define CLK20_REG_FIELD_LIST(type) \
 	type DENTIST_DPPCLK_WDIVIDER; \
 	type DENTIST_DPPCLK_CHG_DONE; \
 	type FbMult_int; \
 	type FbMult_frac;
 
-#define VBIOS_SMU_REG_FIELD_LIST(type) \
-	type CONTENT;
-
-struct clk_mgr_shift {
-	CLK_REG_FIELD_LIST(uint8_t)
-	CLK20_REG_FIELD_LIST(uint8_t)
-	VBIOS_SMU_REG_FIELD_LIST(uint32_t)
-};
-
-struct clk_mgr_mask {
-	CLK_REG_FIELD_LIST(uint32_t)
-	CLK20_REG_FIELD_LIST(uint32_t)
-	VBIOS_SMU_REG_FIELD_LIST(uint32_t)
-};
+/*
+ ***************************************************************************************
+ ****************** Clock Manager Private Structures ***********************************
+ ***************************************************************************************
+ */
 
 struct clk_mgr_registers {
 	uint32_t DPREFCLK_CNTL;
 	uint32_t DENTIST_DISPCLK_CNTL;
+
 	uint32_t CLK4_CLK2_CURRENT_CNT;
 	uint32_t CLK4_CLK_PLL_REQ;
 
···
 	uint32_t CLK0_CLK1_DFS_CNTL;
 	uint32_t CLK0_CLK3_DFS_CNTL;
 	uint32_t CLK0_CLK4_DFS_CNTL;
+};
 
-	uint32_t MP1_SMN_C2PMSG_67;
-	uint32_t MP1_SMN_C2PMSG_83;
-	uint32_t MP1_SMN_C2PMSG_91;
+struct clk_mgr_shift {
+	CLK_REG_FIELD_LIST(uint8_t)
+	CLK20_REG_FIELD_LIST(uint8_t)
+};
+
+struct clk_mgr_mask {
+	CLK_REG_FIELD_LIST(uint32_t)
+	CLK20_REG_FIELD_LIST(uint32_t)
 };
 
 enum clock_type {

drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h

···
  * blocks for the Data Fabric Interface that are not clock/power gated.
  */
 
+#include "dc/dc_hw_types.h"
+
 enum dcc_control {
 	dcc_control__256_256_xxx,
 	dcc_control__128_128_xxx,
···
 			enum segment_order *segment_order_horz,
 			enum segment_order *segment_order_vert);
 
+	bool (*dcc_support_swizzle_addr3)(
+			enum swizzle_mode_addr3_values swizzle,
+			unsigned int plane_pitch,
+			unsigned int bytes_per_element,
+			enum segment_order *segment_order_horz,
+			enum segment_order *segment_order_vert);
+
+	bool (*dcc_support_pixel_format_plane0_plane1)(
+			enum surface_pixel_format format,
+			unsigned int *plane0_bpe,
+			unsigned int *plane1_bpe);
 	bool (*dcc_support_pixel_format)(
 			enum surface_pixel_format format,
 			unsigned int *bytes_per_element);
···
 	void (*set_request_limit)(struct hubbub *hubbub, int memory_channel_count, int words_per_channel);
 	void (*dchubbub_init)(struct hubbub *hubbub);
 	void (*get_mall_en)(struct hubbub *hubbub, unsigned int *mall_in_use);
+	void (*program_det_segments)(struct hubbub *hubbub, int hubp_inst, unsigned det_buffer_size_seg);
+	void (*program_compbuf_segments)(struct hubbub *hubbub, unsigned compbuf_size_seg, bool safe_to_increase);
 };
 
 struct hubbub {

drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h

···
 
 	void (*dpp_get_gamut_remap)(struct dpp *dpp_base,
 				    struct dpp_grph_csc_adjustment *adjust);
+	void (*set_cursor_matrix)(
+		struct dpp *dpp_base,
+		enum dc_color_space color_space,
+		struct dc_csc_transform cursor_csc_color_matrix);
 };
 
 

drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h

···
 #include "mem_input.h"
 #include "cursor_reg_cache.h"
 
+#include "dml2/dml21/inc/dml_top_dchub_registers.h"
+
 #define OPP_ID_INVALID 0xf
 #define MAX_TTU 0xffffff
 
···
 	hubp_ind_block_64b,
 	hubp_ind_block_128b,
 	hubp_ind_block_64b_no_128bcl,
+};
+
+enum hubp_3dlut_fl_mode {
+	hubp_3dlut_fl_mode_disable = 0,
+	hubp_3dlut_fl_mode_native_1 = 1,
+	hubp_3dlut_fl_mode_native_2 = 2,
+	hubp_3dlut_fl_mode_transform = 3
+};
+
+enum hubp_3dlut_fl_format {
+	hubp_3dlut_fl_format_unorm_12msb_bitslice = 0,
+	hubp_3dlut_fl_format_unorm_12lsb_bitslice = 1,
+	hubp_3dlut_fl_format_float_fp1_5_10 = 2
+};
+
+enum hubp_3dlut_fl_addressing_mode {
+	hubp_3dlut_fl_addressing_mode_sw_linear = 0,
+	hubp_3dlut_fl_addressing_mode_simple_linear = 1
+};
+
+enum hubp_3dlut_fl_width {
+	hubp_3dlut_fl_width_17 = 17,
+	hubp_3dlut_fl_width_33 = 33,
+	hubp_3dlut_fl_width_transformed = 4916
+};
+
+enum hubp_3dlut_fl_crossbar_bit_slice {
+	hubp_3dlut_fl_crossbar_bit_slice_0_15 = 0,
+	hubp_3dlut_fl_crossbar_bit_slice_16_31 = 1,
+	hubp_3dlut_fl_crossbar_bit_slice_32_47 = 2,
+	hubp_3dlut_fl_crossbar_bit_slice_48_63 = 3
 };
 
 struct hubp {
···
 	bool (*hubp_in_blank)(struct hubp *hubp);
 	void (*hubp_soft_reset)(struct hubp *hubp, bool reset);
 
+	void (*hubp_set_flip_int)(struct hubp *hubp);
+
 	void (*hubp_update_force_pstate_disallow)(struct hubp *hubp, bool allow);
 	void (*hubp_update_force_cursor_pstate_disallow)(struct hubp *hubp, bool allow);
 	void (*hubp_update_mall_sel)(struct hubp *hubp, uint32_t mall_sel, bool c_cursor);
 	void (*hubp_prepare_subvp_buffering)(struct hubp *hubp, bool enable);
-
-	void (*hubp_set_flip_int)(struct hubp *hubp);
+	void (*hubp_surface_update_lock)(struct hubp *hubp,
+			bool lock);
 
 	void (*program_extended_blank)(struct hubp *hubp,
 			unsigned int min_dst_y_next_start_optimized);
 
 	void (*hubp_wait_pipe_read_start)(struct hubp *hubp);
+	void (*hubp_update_3dlut_fl_bias_scale)(struct hubp *hubp, uint16_t bias, uint16_t scale);
+	void (*hubp_program_3dlut_fl_mode)(struct hubp *hubp,
+			enum hubp_3dlut_fl_mode mode);
+	void (*hubp_program_3dlut_fl_format)(struct hubp *hubp,
+			enum hubp_3dlut_fl_format format);
+	void (*hubp_program_3dlut_fl_addr)(struct hubp *hubp,
+		const struct dc_plane_address address);
+	void (*hubp_program_3dlut_fl_dlg_param)(struct hubp *hubp, int refcyc_per_3dlut_group);
+	void (*hubp_enable_3dlut_fl)(struct hubp *hubp, bool enable);
+	void (*hubp_program_3dlut_fl_addressing_mode)(struct hubp *hubp, enum hubp_3dlut_fl_addressing_mode addr_mode);
+	void (*hubp_program_3dlut_fl_width)(struct hubp *hubp, enum hubp_3dlut_fl_width width);
+	void (*hubp_program_3dlut_fl_tmz_protected)(struct hubp *hubp, bool protection_enabled);
+	void (*hubp_program_3dlut_fl_crossbar)(struct hubp *hubp,
+			enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_y_g,
+			enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_cb_b,
+			enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_cr_r);
+	int (*hubp_get_3dlut_fl_done)(struct hubp *hubp);
 };
 
 #endif

drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h

···
 #include "include/grph_object_id.h"
 
 #include "dml/display_mode_structs.h"
+#include "dml2/dml21/inc/dml_top_dchub_registers.h"
 
 struct dchub_init_data;
 struct cstate_pstate_watermarks_st {
···
 	uint32_t urgent_ns;
 	uint32_t frac_urg_bw_nom;
 	uint32_t frac_urg_bw_flip;
-	int32_t urgent_latency_ns;
+	uint32_t urgent_latency_ns;
 	struct cstate_pstate_watermarks_st cstate_pstate;
 	uint32_t usr_retraining_ns;
 };
···
 		struct dcn_watermarks c;
 		struct dcn_watermarks d;
 	}; // legacy
+	struct {
+		struct dml2_dchub_watermark_regs a;
+		struct dml2_dchub_watermark_regs b;
+		struct dml2_dchub_watermark_regs c;
+		struct dml2_dchub_watermark_regs d;
+	} dcn4; //dcn4+
 };
 
 struct dce_watermarks {

drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h

···
 	MPCC_ALPHA_BLEND_MODE_GLOBAL_ALPHA
 };
 
+enum mpcc_movable_cm_location {
+	MPCC_MOVABLE_CM_LOCATION_BEFORE,
+	MPCC_MOVABLE_CM_LOCATION_AFTER,
+};
+
+enum MCM_LUT_XABLE {
+	MCM_LUT_DISABLE,
+	MCM_LUT_DISABLED = MCM_LUT_DISABLE,
+	MCM_LUT_ENABLE,
+	MCM_LUT_ENABLED = MCM_LUT_ENABLE,
+};
+
+enum MCM_LUT_ID {
+	MCM_LUT_3DLUT,
+	MCM_LUT_1DLUT,
+	MCM_LUT_SHAPER
+};
+
+union mcm_lut_params {
+	const struct pwl_params *pwl;
+	const struct tetrahedral_params *lut3d;
+};
+
 /**
  * struct mpcc_blnd_cfg - MPCC blending configuration
  */
···
 struct mpc_grph_gamut_adjustment {
 	struct fixed31_32 temperature_matrix[CSC_TEMPERATURE_MATRIX_SIZE];
 	enum graphics_gamut_adjust_type gamut_adjust_type;
+	enum mpcc_gamut_remap_id mpcc_gamut_remap_block_id;
 };
 
 struct mpcc_sm_cfg {
···
 	int (*release_rmu)(struct mpc *mpc, int mpcc_id);
 
 	unsigned int (*get_mpc_out_mux)(
-			struct mpc *mpc,
-			int opp_id);
+				struct mpc *mpc,
+				int opp_id);
 
 	void (*set_bg_color)(struct mpc *mpc,
 			struct tg_color *bg_color,
 			int mpcc_id);
 	void (*set_mpc_mem_lp_mode)(struct mpc *mpc);
+	void (*set_movable_cm_location)(struct mpc *mpc, enum mpcc_movable_cm_location location, int mpcc_id);
+	void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx);
+	void (*get_3dlut_fast_load_status)(struct mpc *mpc, int mpcc_id, uint32_t *done, uint32_t *soft_underflow, uint32_t *hard_underflow);
+	void (*populate_lut)(struct mpc *mpc, const enum MCM_LUT_ID id, const union mcm_lut_params params,
+			bool lut_bank_a, int mpcc_id);
+	void (*program_lut_read_write_control)(struct mpc *mpc, const enum MCM_LUT_ID id, bool lut_bank_a, int mpcc_id);
+	void (*program_lut_mode)(struct mpc *mpc, const enum MCM_LUT_ID id, const enum MCM_LUT_XABLE xable,
+			bool lut_bank_a, int mpcc_id);
 };
 
 #endif

drivers/gpu/drm/amd/display/dc/inc/hw/transform.h

···
 #include "hw_shared.h"
 #include "dc_hw_types.h"
 #include "fixed31_32.h"
+#include "spl/dc_spl_types.h"
 
 #include "spl/dc_spl_types.h"
 
···
 	struct sharpness_adj sharpness;
 	enum pixel_format format;
 	struct line_buffer_params lb_params;
+	// Below struct holds the scaler values to program hw registers
+	struct dscl_prog_data dscl_prog_data;
 };
 
 struct transform_funcs {
···
 			struct transform *xfm_base,
 			const struct dc_cursor_attributes *attr);
 
+	bool (*transform_program_blnd_lut)(
+			struct transform *xfm,
+			const struct pwl_params *params);
+	bool (*transform_program_shaper_lut)(
+			struct transform *xfm,
+			const struct pwl_params *params);
+	bool (*transform_program_3dlut)(
+			struct transform *xfm,
+			struct tetrahedral_params *params);
 };
 
 const uint16_t *get_filter_2tap_16p(void);

drivers/gpu/drm/amd/display/dc/inc/resource.h

···
 #include "core_status.h"
 #include "dal_asic_id.h"
 #include "dm_pp_smu.h"
+#include "spl/dc_spl.h"
 
 #define MEMORY_TYPE_MULTIPLIER_CZ 4
 #define MEMORY_TYPE_HBM 2
···
 
 	struct hpo_dp_stream_encoder *(*create_hpo_dp_stream_encoder)(
 			enum engine_id eng_id, struct dc_context *ctx);
-
 	struct hpo_dp_link_encoder *(*create_hpo_dp_link_encoder)(
 			uint8_t inst,
 			struct dc_context *ctx);
-
 	struct dce_hwseq *(*create_hwseq)(
 			struct dc_context *ctx);
 };
···
 
 bool check_subvp_sw_cursor_fallback_req(const struct dc *dc, struct dc_stream_state *stream);
 
+/* Get hw programming parameters container from pipe context
+ * @pipe_ctx: pipe context
+ * @dscl_prog_data: struct to hold programmable hw reg values
+ */
+struct dscl_prog_data *resource_get_dscl_prog_data(struct pipe_ctx *pipe_ctx);
 /* Setup dc callbacks for dml2
  * @dc: the display core structure
  * @dml2_options: struct to hold callbacks

drivers/gpu/drm/amd/display/dc/irq/Makefile

···
 AMD_DAL_IRQ_DCN351= $(addprefix $(AMDDALPATH)/dc/irq/dcn351/,$(IRQ_DCN351))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN351)
+
+###############################################################################
+# DCN 401
+###############################################################################
+IRQ_DCN401 = irq_service_dcn401.o
+
+AMD_DAL_IRQ_DCN401= $(addprefix $(AMDDALPATH)/dc/irq/dcn401/,$(IRQ_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN401)

drivers/gpu/drm/amd/display/dc/optc/Makefile

···
 ###############################################################################
 
 ###############################################################################
+OPTC_DCN401 = dcn401_optc.o
+
+AMD_DAL_OPTC_DCN401 = $(addprefix $(AMDDALPATH)/dc/optc/dcn401/,$(OPTC_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_OPTC_DCN401)
 endif
+

drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h

···
 	SF(GSL_SOURCE_SELECT, GSL2_READY_SOURCE_SEL, mask_sh),\
 	SF(OTG0_OTG_GLOBAL_CONTROL2, MANUAL_FLOW_CONTROL_SEL, mask_sh)
 
-
-
 #define TG_COMMON_MASK_SH_LIST_DCN1_0(mask_sh)\
 	TG_COMMON_MASK_SH_LIST_DCN(mask_sh),\
 	SF(OTG0_OTG_TEST_PATTERN_PARAMETERS, OTG_TEST_PATTERN_INC0, mask_sh),\
···
 
 #define TG_REG_FIELD_LIST_DCN3_2(type) \
 	type OTG_H_TIMING_DIV_MODE_MANUAL;
-
 
 #define TG_REG_FIELD_LIST_DCN3_5(type) \
 	type OTG_CRC0_WINDOWA_X_START_READBACK;\

drivers/gpu/drm/amd/display/dc/resource/Makefile

···
 
 ###############################################################################
 
+###############################################################################
+
+RESOURCE_DCN401 = dcn401_resource.o
+
+AMD_DAL_RESOURCE_DCN401 = $(addprefix $(AMDDALPATH)/dc/resource/dcn401/,$(RESOURCE_DCN401))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCN401)
+
 endif

drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c

···
 	.force_disable_subvp = false,
 	.exit_idle_opt_for_cursor_updates = true,
 	.using_dml2 = false,
+	.using_dml21 = false, // TODO : Temporary for N-1 validation. Remove after N-1 is done.
 	.enable_single_display_2to1_odm_policy = true,
 
 	/* Must match enable_single_display_2to1_odm_policy to support dynamic ODM transitions*/

drivers/gpu/drm/amd/display/include/dal_asic_id.h

···
 #define ASICREV_IS_GC_11_0_3(eChipRev) (eChipRev >= GC_11_0_3_A0 && eChipRev < GC_11_UNKNOWN)
 #define ASICREV_IS_GC_11_0_4(eChipRev) (eChipRev >= GC_11_0_4_A0 && eChipRev < GC_11_UNKNOWN)
 
+#define AMDGPU_FAMILY_GC_12_0_0         152 /* GC 12.0.0 */
+
+enum {
+	GC_12_0_0_A0 = 0x50,
+	GC_12_0_1_A0 = 0x40,
+	GC_12_UNKNOWN = 0xFF,
+};
+
+#define ASICREV_IS_DCN4(eChipRev)	(eChipRev >= GC_12_0_1_A0 && eChipRev < GC_12_0_0_A0)
+#define ASICREV_IS_DCN401(eChipRev)	(eChipRev >= GC_12_0_0_A0 && eChipRev < GC_12_UNKNOWN)
+
 /*
  * ASIC chip ID
  */

drivers/gpu/drm/amd/display/include/dal_types.h

···
 	DCN_VERSION_3_21,
 	DCN_VERSION_3_5,
 	DCN_VERSION_3_51,
+	DCN_VERSION_4_01,
 	DCN_VERSION_MAX
 };