Merging upstream version 6.10.3.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

19 files changed, 170 insertions, 160 deletions

diff --git a/drivers/gpu/drm/amd/display/dc/resource/Makefile b/drivers/gpu/drm/amd/display/dc/resource/Makefile
index 184b1f23aa..db9048974d 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/resource/Makefile
@@ -102,10 +102,6 @@ AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCN21)
 
 ###############################################################################
 
-###############################################################################
-
-###############################################################################
-
 RESOURCE_DCN30 = dcn30_resource.o
 
 AMD_DAL_RESOURCE_DCN30 = $(addprefix $(AMDDALPATH)/dc/resource/dcn30/,$(RESOURCE_DCN30))
@@ -202,6 +198,4 @@ AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCN351)
 
 ###############################################################################
 
-###############################################################################
-
 endif
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
index d1edac46c9..88afb2a30e 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
@@ -864,8 +864,6 @@ static struct clock_source *find_matching_pll(
 	default:
 		return NULL;
 	}
-
-	return NULL;
 }
 
 static enum dc_status build_mapped_resource(
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
index 20662edd0a..621825a51f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
@@ -1060,7 +1060,7 @@ static bool dce120_resource_construct(
 	struct irq_service_init_data irq_init_data;
 	static const struct resource_create_funcs *res_funcs;
 	bool is_vg20 = ASICREV_IS_VEGA20_P(ctx->asic_id.hw_internal_rev);
-	uint32_t pipe_fuses;
+	uint32_t pipe_fuses = 0;
 
 	ctx->dc_bios->regs = &bios_regs;
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
index 35a2cce0c2..56ee45e12b 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
@@ -56,7 +56,6 @@
 #include "dce/dce_aux.h"
 #include "dce/dce_abm.h"
 #include "dce/dce_i2c.h"
-/* TODO remove this include */
 
 #ifndef mmMC_HUB_RDREQ_DMIF_LIMIT
 #include "gmc/gmc_7_1_d.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c
index d08d109692..563c5eec83 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c
@@ -513,7 +513,7 @@ static const struct dc_plane_cap plane_cap = {
 			.argb8888 = true,
 			.nv12 = true,
 			.fp16 = true,
-			.p010 = true
+			.p010 = false
 	},
 
 	.max_upscale_factor = {
@@ -569,6 +569,7 @@ static const struct dc_debug_options debug_defaults_diags = {
 		.disable_pplib_clock_request = true,
 		.disable_pplib_wm_range = true,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
+		.enable_legacy_fast_update = true,
 };
 
 static void dcn10_dpp_destroy(struct dpp **dpp)
@@ -1631,6 +1632,7 @@ static bool dcn10_resource_construct(
 	/* valid pipe num */
 	pool->base.pipe_count = j;
 	pool->base.timing_generator_count = j;
+	pool->base.mpcc_count = j;
 
 	/* within dml lib, it is hard code to 4. If ASIC pipe is fused,
 	 * the value may be changed
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
index 622214b365..0a939437e1 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
@@ -62,6 +62,9 @@
 #include "dcn20/dcn20_vmid.h"
 #include "dce/dce_panel_cntl.h"
 
+#include "dcn20/dcn20_dwb.h"
+#include "dcn20/dcn20_mmhubbub.h"
+
 #include "navi10_ip_offset.h"
 
 #include "dcn/dcn_2_0_0_offset.h"
@@ -71,9 +74,6 @@
 
 #include "nbio/nbio_2_3_offset.h"
 
-#include "dcn20/dcn20_dwb.h"
-#include "dcn20/dcn20_mmhubbub.h"
-
 #include "mmhub/mmhub_2_0_0_offset.h"
 #include "mmhub/mmhub_2_0_0_sh_mask.h"
 
@@ -83,11 +83,10 @@
 #include "dce/dce_aux.h"
 #include "dce/dce_i2c.h"
 #include "vm_helper.h"
-#include "link_enc_cfg.h"
-
-#include "amdgpu_socbb.h"
 
+#include "link_enc_cfg.h"
 #include "link.h"
+
 #define DC_LOGGER_INIT(logger)
 
 #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL
@@ -1282,8 +1281,13 @@ void dcn20_build_pipe_pix_clk_params(struct pipe_ctx *pipe_ctx)
 
 static enum dc_status build_pipe_hw_param(struct pipe_ctx *pipe_ctx)
 {
+	struct resource_pool *pool = pipe_ctx->stream->ctx->dc->res_pool;
 
-	dcn20_build_pipe_pix_clk_params(pipe_ctx);
+	if (pool->funcs->build_pipe_pix_clk_params) {
+		pool->funcs->build_pipe_pix_clk_params(pipe_ctx);
+	} else {
+		dcn20_build_pipe_pix_clk_params(pipe_ctx);
+	}
 
 	pipe_ctx->stream->clamping.pixel_encoding = pipe_ctx->stream->timing.pixel_encoding;
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c
index 914b234d7f..070a4efb30 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c
@@ -55,7 +55,6 @@
 #include "dce110/dce110_resource.h"
 #include "dce/dce_aux.h"
 #include "dce/dce_i2c.h"
-#include "dcn201/dcn201_hubbub.h"
 #include "dcn10/dcn10_resource.h"
 
 #include "cyan_skillfish_ip_offset.h"
@@ -182,6 +181,7 @@ static struct _vcs_dpi_soc_bounding_box_st dcn201_soc = {
 				.socclk_mhz = 1254.0,
 				.dram_speed_mts = 14000.0,
 			},
+			/* state4 is not an actual state, just defines unsupported for dml*/
 			{
 				.state = 4,
 				.dscclk_mhz = 400.0,
@@ -566,6 +566,8 @@ static const struct resource_caps res_cap_dnc201 = {
 		.num_audio = 2,
 		.num_stream_encoder = 2,
 		.num_pll = 2,
+		.num_dwb = 0,
+		.num_dsc = 0,
 		.num_ddc = 2,
 };
 
@@ -612,7 +614,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.scl_reset_length10 = true,
 		.sanity_checks = false,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
-		.enable_tri_buf = false,
+		.enable_tri_buf = true,
 		.enable_legacy_fast_update = true,
 		.using_dml2 = false,
 };
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c
index 65d337731f..8663cbc3d1 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c
@@ -581,32 +581,6 @@ static const struct resource_caps res_cap_rn = {
 		.num_dsc = 3,
 };
 
-#ifdef DIAGS_BUILD
-static const struct resource_caps res_cap_rn_FPGA_4pipe = {
-		.num_timing_generator = 4,
-		.num_opp = 4,
-		.num_video_plane = 4,
-		.num_audio = 7,
-		.num_stream_encoder = 4,
-		.num_pll = 4,
-		.num_dwb = 1,
-		.num_ddc = 4,
-		.num_dsc = 0,
-};
-
-static const struct resource_caps res_cap_rn_FPGA_2pipe_dsc = {
-		.num_timing_generator = 2,
-		.num_opp = 2,
-		.num_video_plane = 2,
-		.num_audio = 7,
-		.num_stream_encoder = 2,
-		.num_pll = 4,
-		.num_dwb = 1,
-		.num_ddc = 4,
-		.num_dsc = 2,
-};
-#endif
-
 static const struct dc_plane_cap plane_cap = {
 	.type = DC_PLANE_TYPE_DCN_UNIVERSAL,
 	.per_pixel_alpha = true,
@@ -1415,16 +1389,11 @@ static bool dcn21_resource_construct(
 	struct dc_context *ctx = dc->ctx;
 	struct irq_service_init_data init_data;
 	uint32_t pipe_fuses = read_pipe_fuses(ctx);
-	uint32_t num_pipes;
+	uint32_t num_pipes = 0;
 
 	ctx->dc_bios->regs = &bios_regs;
 
 	pool->base.res_cap = &res_cap_rn;
-#ifdef DIAGS_BUILD
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		//pool->base.res_cap = &res_cap_nv10_FPGA_2pipe_dsc;
-		pool->base.res_cap = &res_cap_rn_FPGA_4pipe;
-#endif
 
 	pool->base.funcs = &dcn21_res_pool_funcs;
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
index b427a98066..f35cc30783 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
@@ -1639,7 +1639,7 @@ noinline bool dcn30_internal_validate_bw(
 	int split[MAX_PIPES] = { 0 };
 	bool merge[MAX_PIPES] = { false };
 	bool newly_split[MAX_PIPES] = { false };
-	int pipe_cnt, i, pipe_idx, vlevel;
+	int pipe_cnt, i, pipe_idx, vlevel = 0;
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 
 	ASSERT(pipes);
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c
index 25cd6236b0..8bc1bcaeaa 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c
@@ -1143,7 +1143,7 @@ static bool dcn303_resource_construct(
 	int i;
 	struct dc_context *ctx = dc->ctx;
 	struct irq_service_init_data init_data;
-	struct ddc_service_init_data ddc_init_data;
+	struct ddc_service_init_data ddc_init_data = {0};
 
 	ctx->dc_bios->regs = &bios_regs;
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
index f38de53911..d4c3e2754f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
@@ -75,7 +75,6 @@
 #include "dcn30/dcn30_dwb.h"
 #include "dcn30/dcn30_mmhubbub.h"
 
-// TODO: change include headers /amd/include/asic_reg after upstream
 #include "yellow_carp_offset.h"
 #include "dcn/dcn_3_1_2_offset.h"
 #include "dcn/dcn_3_1_2_sh_mask.h"
@@ -1647,7 +1646,7 @@ int dcn31_populate_dml_pipes_from_context(
 {
 	int i, pipe_cnt;
 	struct resource_context *res_ctx = &context->res_ctx;
-	struct pipe_ctx *pipe;
+	struct pipe_ctx *pipe = 0;
 	bool upscaled = false;
 
 	DC_FP_START();
@@ -1776,7 +1775,7 @@ bool dcn31_validate_bandwidth(struct dc *dc,
 	out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, true);
 	DC_FP_END();
 
-	// Disable fast_validate to set min dcfclk in alculate_wm_and_dlg
+	// Disable fast_validate to set min dcfclk in calculate_wm_and_dlg
 	if (pipe_cnt == 0)
 		fast_validate = false;
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
index 2791fc45bb..ff50f43e4c 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
@@ -925,27 +925,10 @@ static const struct dc_debug_options debug_defaults_drv = {
 	},
 
 	.seamless_boot_odm_combine = true,
+	.enable_legacy_fast_update = true,
 	.using_dml2 = false,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 	.psr = {
 		.disable_psr = false,
@@ -1943,8 +1926,6 @@ static bool dcn314_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else
-		dc->debug = debug_defaults_diags;
 
 	/* Disable pipe power gating */
 	dc->debug.disable_dpp_power_gate = true;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c
index efa5627b0c..5fd52c5fce 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c
@@ -125,7 +125,6 @@
 #include "link_enc_cfg.h"
 
 #define DCN3_16_MAX_DET_SIZE 384
-#define DCN3_16_MIN_COMPBUF_SIZE_KB 128
 #define DCN3_16_CRB_SEGMENT_SIZE_KB 64
 
 enum dcn31_clk_src_array_id {
@@ -1616,7 +1615,7 @@ static int dcn316_populate_dml_pipes_from_context(
 {
 	int i, pipe_cnt;
 	struct resource_context *res_ctx = &context->res_ctx;
-	struct pipe_ctx *pipe;
+	struct pipe_ctx *pipe = 0;
 	const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_16_MIN_COMPBUF_SIZE_KB;
 
 	DC_FP_START();
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
index 1f5a91b764..d84c8e0e5c 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
@@ -1804,7 +1804,9 @@ bool dcn32_validate_bandwidth(struct dc *dc,
 	bool out = false;
 
 	if (dc->debug.using_dml2)
-		out = dml2_validate(dc, context, fast_validate);
+		out = dml2_validate(dc, context,
+				context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
+				fast_validate);
 	else
 		out = dml1_validate(dc, context, fast_validate);
 	return out;
@@ -1820,9 +1822,48 @@ int dcn32_populate_dml_pipes_from_context(
 	struct pipe_ctx *pipe = NULL;
 	bool subvp_in_use = false;
 	struct dc_crtc_timing *timing;
+	int subvp_main_pipe_index = -1;
+	enum mall_stream_type mall_type;
+	bool single_display_subvp = false;
+	struct dc_stream_state *stream = NULL;
+	int num_subvp_main = 0;
+	int num_subvp_phantom = 0;
+	int num_subvp_none = 0;
+	int odm_slice_count;
 
 	dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate);
 
+	/* For single display subvp, look for subvp main so if we have phantom
+	 *  pipe, we can set odm policy to match main pipe
+	 */
+	for (i = 0; i < context->stream_count; i++) {
+		stream = context->streams[i];
+		mall_type = dc_state_get_stream_subvp_type(context, stream);
+		if (mall_type == SUBVP_MAIN)
+			num_subvp_main++;
+		else if (mall_type == SUBVP_PHANTOM)
+			num_subvp_phantom++;
+		else
+			num_subvp_none++;
+	}
+	if (num_subvp_main == 1 && num_subvp_phantom == 1 && num_subvp_none == 0)
+		single_display_subvp = true;
+
+	if (single_display_subvp) {
+		for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
+			pipe = &res_ctx->pipe_ctx[i];
+			if (!res_ctx->pipe_ctx[i].stream)
+				continue;
+
+			mall_type = dc_state_get_pipe_subvp_type(context, pipe);
+			if (mall_type == SUBVP_MAIN) {
+				if (resource_is_pipe_type(pipe, OTG_MASTER))
+					subvp_main_pipe_index = i;
+			}
+			pipe_cnt++;
+		}
+	}
+
 	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
 
 		if (!res_ctx->pipe_ctx[i].stream)
@@ -1837,7 +1878,21 @@ int dcn32_populate_dml_pipes_from_context(
 		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
 		if (dc->config.enable_windowed_mpo_odm &&
 				dc->debug.enable_single_display_2to1_odm_policy) {
-			switch (resource_get_odm_slice_count(pipe)) {
+			/* For single display subvp, if pipe is phantom pipe,
+			 *  then copy odm policy from subvp main pipe
+			 */
+			mall_type = dc_state_get_pipe_subvp_type(context, pipe);
+			if (single_display_subvp && (mall_type == SUBVP_PHANTOM)) {
+				if (subvp_main_pipe_index < 0) {
+					odm_slice_count = -1;
+					ASSERT(0);
+				} else {
+					odm_slice_count = resource_get_odm_slice_count(&res_ctx->pipe_ctx[subvp_main_pipe_index]);
+				}
+			} else {
+				odm_slice_count = resource_get_odm_slice_count(pipe);
+			}
+			switch (odm_slice_count) {
 			case 2:
 				pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1;
 				break;
@@ -1850,6 +1905,7 @@ int dcn32_populate_dml_pipes_from_context(
 		} else {
 			pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal;
 		}
+
 		pipes[pipe_cnt].pipe.src.gpuvm_min_page_size_kbytes = 256; // according to spreadsheet
 		pipes[pipe_cnt].pipe.src.unbounded_req_mode = false;
 		pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_19;
@@ -1917,6 +1973,22 @@ int dcn32_populate_dml_pipes_from_context(
 	return pipe_cnt;
 }
 
+unsigned int dcn32_calculate_mall_ways_from_bytes(const struct dc *dc, unsigned int total_size_in_mall_bytes)
+{
+	uint32_t cache_lines_used, lines_per_way, total_cache_lines, num_ways;
+
+	/* add 2 lines for worst case alignment */
+	cache_lines_used = total_size_in_mall_bytes / dc->caps.cache_line_size + 2;
+
+	total_cache_lines = dc->caps.max_cab_allocation_bytes / dc->caps.cache_line_size;
+	lines_per_way = total_cache_lines / dc->caps.cache_num_ways;
+	num_ways = cache_lines_used / lines_per_way;
+	if (cache_lines_used % lines_per_way > 0)
+		num_ways++;
+
+	return num_ways;
+}
+
 static struct dc_cap_funcs cap_funcs = {
 	.get_dcc_compression_cap = dcn20_get_dcc_compression_cap,
 	.get_subvp_en = dcn32_subvp_in_use,
@@ -1934,10 +2006,22 @@ void dcn32_calculate_wm_and_dlg(struct dc *dc, struct dc_state *context,
 
 static void dcn32_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
 {
+	struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp;
+
+	memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options));
+
 	DC_FP_START();
+
 	dcn32_update_bw_bounding_box_fpu(dc, bw_params);
+
+	dml2_opt->use_clock_dc_limits = false;
 	if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2)
-		dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2);
+		dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2);
+
+	dml2_opt->use_clock_dc_limits = true;
+	if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source)
+		dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source);
+
 	DC_FP_END();
 }
 
@@ -1965,6 +2049,7 @@ static struct resource_funcs dcn32_res_pool_funcs = {
 	.update_soc_for_wm_a = dcn30_update_soc_for_wm_a,
 	.add_phantom_pipes = dcn32_add_phantom_pipes,
 	.build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params,
+	.calculate_mall_ways_from_bytes = dcn32_calculate_mall_ways_from_bytes,
 };
 
 static uint32_t read_pipe_fuses(struct dc_context *ctx)
@@ -2053,7 +2138,8 @@ static bool dcn32_resource_construct(
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dmdata_alloc_size = 2048;
 	dc->caps.mall_size_per_mem_channel = 4;
-	dc->caps.mall_size_total = 0;
+	/* total size = mall per channel * num channels * 1024 * 1024 */
+	dc->caps.mall_size_total = dc->caps.mall_size_per_mem_channel * dc->ctx->dc_bios->vram_info.num_chans * 1048576;
 	dc->caps.cursor_cache_size = dc->caps.max_cursor_size * dc->caps.max_cursor_size * 8;
 
 	dc->caps.cache_line_size = 64;
@@ -2367,30 +2453,10 @@ static bool dcn32_resource_construct(
 	dc->dml2_options.use_native_soc_bb_construction = true;
 	dc->dml2_options.minimize_dispclk_using_odm = true;
 
-	dc->dml2_options.callbacks.dc = dc;
-	dc->dml2_options.callbacks.build_scaling_params = &resource_build_scaling_params;
+	resource_init_common_dml2_callbacks(dc, &dc->dml2_options);
 	dc->dml2_options.callbacks.can_support_mclk_switch_using_fw_based_vblank_stretch = &dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch;
-	dc->dml2_options.callbacks.acquire_secondary_pipe_for_mpc_odm = &dc_resource_acquire_secondary_pipe_for_mpc_odm_legacy;
-	dc->dml2_options.callbacks.update_pipes_for_stream_with_slice_count = &resource_update_pipes_for_stream_with_slice_count;
-	dc->dml2_options.callbacks.update_pipes_for_plane_with_slice_count = &resource_update_pipes_for_plane_with_slice_count;
-	dc->dml2_options.callbacks.get_mpc_slice_index = &resource_get_mpc_slice_index;
-	dc->dml2_options.callbacks.get_odm_slice_index = &resource_get_odm_slice_index;
-	dc->dml2_options.callbacks.get_opp_head = &resource_get_opp_head;
-
-	dc->dml2_options.svp_pstate.callbacks.dc = dc;
-	dc->dml2_options.svp_pstate.callbacks.add_phantom_plane = &dc_state_add_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.add_phantom_stream = &dc_state_add_phantom_stream;
-	dc->dml2_options.svp_pstate.callbacks.build_scaling_params = &resource_build_scaling_params;
-	dc->dml2_options.svp_pstate.callbacks.create_phantom_plane = &dc_state_create_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.remove_phantom_plane = &dc_state_remove_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.remove_phantom_stream = &dc_state_remove_phantom_stream;
-	dc->dml2_options.svp_pstate.callbacks.create_phantom_stream = &dc_state_create_phantom_stream;
-	dc->dml2_options.svp_pstate.callbacks.release_phantom_plane = &dc_state_release_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.release_phantom_stream = &dc_state_release_phantom_stream;
 	dc->dml2_options.svp_pstate.callbacks.release_dsc = &dcn20_release_dsc;
-	dc->dml2_options.svp_pstate.callbacks.get_pipe_subvp_type = &dc_state_get_pipe_subvp_type;
-	dc->dml2_options.svp_pstate.callbacks.get_stream_subvp_type = &dc_state_get_stream_subvp_type;
-	dc->dml2_options.svp_pstate.callbacks.get_paired_subvp_stream = &dc_state_get_paired_subvp_stream;
+	dc->dml2_options.svp_pstate.callbacks.calculate_mall_ways_from_bytes = pool->base.funcs->calculate_mall_ways_from_bytes;
 
 	dc->dml2_options.svp_pstate.subvp_fw_processing_delay_us = dc->caps.subvp_fw_processing_delay_us;
 	dc->dml2_options.svp_pstate.subvp_prefetch_end_to_mall_start_us = dc->caps.subvp_prefetch_end_to_mall_start_us;
@@ -2488,7 +2554,7 @@ struct resource_pool *dcn32_create_resource_pool(
  * full update which delays the flip for 1 frame. If we use the original pipe
  * we don't have to toggle its power. So we can flip faster.
  */
-static int find_optimal_free_pipe_as_secondary_dpp_pipe(
+int dcn32_find_optimal_free_pipe_as_secondary_dpp_pipe(
 		const struct resource_context *cur_res_ctx,
 		struct resource_context *new_res_ctx,
 		const struct resource_pool *pool,
@@ -2671,7 +2737,7 @@ struct pipe_ctx *dcn32_acquire_free_pipe_as_secondary_dpp_pipe(
 		return dcn32_acquire_idle_pipe_for_head_pipe_in_layer(
 				new_ctx, pool, opp_head_pipe->stream, opp_head_pipe);
 
-	free_pipe_idx = find_optimal_free_pipe_as_secondary_dpp_pipe(
+	free_pipe_idx = dcn32_find_optimal_free_pipe_as_secondary_dpp_pipe(
 					&cur_ctx->res_ctx, &new_ctx->res_ctx,
 					pool, opp_head_pipe);
 	if (free_pipe_idx >= 0) {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
index 2258c5c721..fee67fbab8 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
@@ -113,10 +113,6 @@ void dcn32_calculate_wm_and_dlg(
 		int pipe_cnt,
 		int vlevel);
 
-uint32_t dcn32_helper_mall_bytes_to_ways(
-		struct dc *dc,
-		uint32_t total_size_in_mall_bytes);
-
 uint32_t dcn32_helper_calculate_mall_bytes_for_cursor(
 		struct dc *dc,
 		struct pipe_ctx *pipe_ctx,
@@ -141,6 +137,12 @@ bool dcn32_any_surfaces_rotated(struct dc *dc, struct dc_state *context);
 bool dcn32_is_center_timing(struct pipe_ctx *pipe);
 bool dcn32_is_psr_capable(struct pipe_ctx *pipe);
 
+int dcn32_find_optimal_free_pipe_as_secondary_dpp_pipe(
+		const struct resource_context *cur_res_ctx,
+		struct resource_context *new_res_ctx,
+		const struct resource_pool *pool,
+		const struct pipe_ctx *new_opp_head);
+
 struct pipe_ctx *dcn32_acquire_free_pipe_as_secondary_dpp_pipe(
 		const struct dc_state *cur_ctx,
 		struct dc_state *new_ctx,
@@ -184,6 +186,8 @@ void dcn32_update_dml_pipes_odm_policy_based_on_context(struct dc *dc, struct dc
 
 void dcn32_override_min_req_dcfclk(struct dc *dc, struct dc_state *context);
 
+unsigned int dcn32_calculate_mall_ways_from_bytes(const struct dc *dc, unsigned int total_size_in_mall_bytes);
+
 /* definitions for run time init of reg offsets */
 
 /* CLK SRC */
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
index e83d340ed6..9a3cc0514a 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
@@ -1581,10 +1581,22 @@ static struct dc_cap_funcs cap_funcs = {
 
 static void dcn321_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
 {
+	struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp;
+
+	memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options));
+
 	DC_FP_START();
+
 	dcn321_update_bw_bounding_box_fpu(dc, bw_params);
+
+	dml2_opt->use_clock_dc_limits = false;
 	if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2)
-		dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2);
+		dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2);
+
+	dml2_opt->use_clock_dc_limits = true;
+	if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source)
+		dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source);
+
 	DC_FP_END();
 }
 
@@ -1612,6 +1624,7 @@ static struct resource_funcs dcn321_res_pool_funcs = {
 	.update_soc_for_wm_a = dcn30_update_soc_for_wm_a,
 	.add_phantom_pipes = dcn32_add_phantom_pipes,
 	.build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params,
+	.calculate_mall_ways_from_bytes = dcn32_calculate_mall_ways_from_bytes,
 };
 
 static uint32_t read_pipe_fuses(struct dc_context *ctx)
@@ -1699,7 +1712,9 @@ static bool dcn321_resource_construct(
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dmdata_alloc_size = 2048;
 	dc->caps.mall_size_per_mem_channel = 4;
-	dc->caps.mall_size_total = 0;
+	/* total size = mall per channel * num channels * 1024 * 1024 */
+	dc->caps.mall_size_total = dc->caps.mall_size_per_mem_channel * dc->ctx->dc_bios->vram_info.num_chans * 1048576;
+
 	dc->caps.cursor_cache_size = dc->caps.max_cursor_size * dc->caps.max_cursor_size * 8;
 	dc->caps.cache_line_size = 64;
 	dc->caps.cache_num_ways = 16;
@@ -2000,30 +2015,10 @@ static bool dcn321_resource_construct(
 	dc->dml2_options.use_native_soc_bb_construction = true;
 	dc->dml2_options.minimize_dispclk_using_odm = true;
 
-	dc->dml2_options.callbacks.dc = dc;
-	dc->dml2_options.callbacks.build_scaling_params = &resource_build_scaling_params;
+	resource_init_common_dml2_callbacks(dc, &dc->dml2_options);
 	dc->dml2_options.callbacks.can_support_mclk_switch_using_fw_based_vblank_stretch = &dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch;
-	dc->dml2_options.callbacks.acquire_secondary_pipe_for_mpc_odm = &dc_resource_acquire_secondary_pipe_for_mpc_odm_legacy;
-	dc->dml2_options.callbacks.update_pipes_for_stream_with_slice_count = &resource_update_pipes_for_stream_with_slice_count;
-	dc->dml2_options.callbacks.update_pipes_for_plane_with_slice_count = &resource_update_pipes_for_plane_with_slice_count;
-	dc->dml2_options.callbacks.get_mpc_slice_index = &resource_get_mpc_slice_index;
-	dc->dml2_options.callbacks.get_odm_slice_index = &resource_get_odm_slice_index;
-	dc->dml2_options.callbacks.get_opp_head = &resource_get_opp_head;
-
-	dc->dml2_options.svp_pstate.callbacks.dc = dc;
-	dc->dml2_options.svp_pstate.callbacks.add_phantom_plane = &dc_state_add_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.add_phantom_stream = &dc_state_add_phantom_stream;
-	dc->dml2_options.svp_pstate.callbacks.build_scaling_params = &resource_build_scaling_params;
-	dc->dml2_options.svp_pstate.callbacks.create_phantom_plane = &dc_state_create_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.remove_phantom_plane = &dc_state_remove_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.remove_phantom_stream = &dc_state_remove_phantom_stream;
-	dc->dml2_options.svp_pstate.callbacks.create_phantom_stream = &dc_state_create_phantom_stream;
-	dc->dml2_options.svp_pstate.callbacks.release_phantom_plane = &dc_state_release_phantom_plane;
-	dc->dml2_options.svp_pstate.callbacks.release_phantom_stream = &dc_state_release_phantom_stream;
 	dc->dml2_options.svp_pstate.callbacks.release_dsc = &dcn20_release_dsc;
-	dc->dml2_options.svp_pstate.callbacks.get_pipe_subvp_type = &dc_state_get_pipe_subvp_type;
-	dc->dml2_options.svp_pstate.callbacks.get_stream_subvp_type = &dc_state_get_stream_subvp_type;
-	dc->dml2_options.svp_pstate.callbacks.get_paired_subvp_stream = &dc_state_get_paired_subvp_stream;
+	dc->dml2_options.svp_pstate.callbacks.calculate_mall_ways_from_bytes = pool->base.funcs->calculate_mall_ways_from_bytes;
 
 	dc->dml2_options.svp_pstate.subvp_fw_processing_delay_us = dc->caps.subvp_fw_processing_delay_us;
 	dc->dml2_options.svp_pstate.subvp_prefetch_end_to_mall_start_us = dc->caps.subvp_prefetch_end_to_mall_start_us;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
index cf0cb5cf4b..2df8a74251 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
@@ -721,7 +721,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_dpp_power_gate = true,
 	.disable_hubp_power_gate = true,
 	.disable_optc_power_gate = true, /*should the same as above two*/
-	.disable_hpo_power_gate = true, /*dmubfw force domain25 on*/
+	.disable_hpo_power_gate = false, /*dmubfw force domain25 on*/
 	.disable_clock_gate = false,
 	.disable_dsc_power_gate = true,
 	.vsr_support = true,
@@ -764,12 +764,12 @@ static const struct dc_debug_options debug_defaults_drv = {
 	},
 	.seamless_boot_odm_combine = DML_FAIL_SOURCE_PIXEL_FORMAT,
 	.enable_z9_disable_interface = true, /* Allow support for the PMFW interface for disable Z9*/
-	.minimum_z8_residency_time = 2100,
+	.minimum_z8_residency_time = 1, /* Always allow when other conditions are met */
 	.using_dml2 = true,
 	.support_eDP1_5 = true,
 	.enable_hpo_pg_support = false,
 	.enable_legacy_fast_update = true,
-	.enable_single_display_2to1_odm_policy = false,
+	.enable_single_display_2to1_odm_policy = true,
 	.disable_idle_power_optimizations = false,
 	.dmcub_emulation = false,
 	.disable_boot_optimizations = false,
@@ -783,7 +783,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.psp_disabled_wa = true,
 	.ips2_eval_delay_us = 2000,
 	.ips2_entry_delay_us = 800,
-	.disable_dmub_reallow_idle = true,
+	.disable_dmub_reallow_idle = false,
 	.static_screen_wait_frames = 2,
 };
 
@@ -1736,7 +1736,9 @@ static bool dcn35_validate_bandwidth(struct dc *dc,
 {
 	bool out = false;
 
-	out = dml2_validate(dc, context, fast_validate);
+	out = dml2_validate(dc, context,
+			context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
+			fast_validate);
 
 	if (fast_validate)
 		return out;
@@ -2140,15 +2142,9 @@ static bool dcn35_resource_construct(
 		dc->dml2_options.minimize_dispclk_using_odm = true;
 	dc->dml2_options.enable_windowed_mpo_odm = dc->config.enable_windowed_mpo_odm;
 
-	dc->dml2_options.callbacks.dc = dc;
-	dc->dml2_options.callbacks.build_scaling_params = &resource_build_scaling_params;
+	resource_init_common_dml2_callbacks(dc, &dc->dml2_options);
 	dc->dml2_options.callbacks.can_support_mclk_switch_using_fw_based_vblank_stretch = &dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch;
-	dc->dml2_options.callbacks.acquire_secondary_pipe_for_mpc_odm = &dc_resource_acquire_secondary_pipe_for_mpc_odm_legacy;
-	dc->dml2_options.callbacks.update_pipes_for_stream_with_slice_count = &resource_update_pipes_for_stream_with_slice_count;
-	dc->dml2_options.callbacks.update_pipes_for_plane_with_slice_count = &resource_update_pipes_for_plane_with_slice_count;
-	dc->dml2_options.callbacks.get_mpc_slice_index = &resource_get_mpc_slice_index;
-	dc->dml2_options.callbacks.get_odm_slice_index = &resource_get_odm_slice_index;
-	dc->dml2_options.callbacks.get_opp_head = &resource_get_opp_head;
+
 	dc->dml2_options.max_segments_per_hubp = 24;
 
 	dc->dml2_options.det_segment_size = DCN3_2_DET_SEG_SIZE;/*todo*/
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.h
index a51c4a9eaa..f97bb4cb37 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.h
@@ -240,6 +240,8 @@ struct resource_pool *dcn35_create_resource_pool(
 	SRI_ARR(OTG_V_TOTAL_MAX, OTG, inst),\
 	SRI_ARR(OTG_V_TOTAL_MIN, OTG, inst),\
 	SRI_ARR(OTG_V_TOTAL_CONTROL, OTG, inst),\
+	SRI_ARR(OTG_V_COUNT_STOP_CONTROL, OTG, inst),\
+	SRI_ARR(OTG_V_COUNT_STOP_CONTROL2, OTG, inst),\
 	SRI_ARR(OTG_TRIGA_CNTL, OTG, inst),\
 	SRI_ARR(OTG_FORCE_COUNT_NOW_CNTL, OTG, inst),\
 	SRI_ARR(OTG_STATIC_SCREEN_CONTROL, OTG, inst),\
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
index 116b591231..ddf9560ab7 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
@@ -758,7 +758,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	//must match enable_single_display_2to1_odm_policy to support dynamic ODM transitions
 	.enable_double_buffered_dsc_pg_support = true,
 	.enable_dp_dig_pixel_rate_div_policy = 1,
-	.disable_z10 = true,
+	.disable_z10 = false,
 	.ignore_pg = true,
 	.psp_disabled_wa = true,
 	.ips2_eval_delay_us = 2000,
@@ -1716,19 +1716,20 @@ static bool dcn351_validate_bandwidth(struct dc *dc,
 {
 	bool out = false;
 
-	out = dml2_validate(dc, context, fast_validate);
+	out = dml2_validate(dc, context,
+			context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
+			fast_validate);
 
 	if (fast_validate)
 		return out;
 
 	DC_FP_START();
-	dcn351_decide_zstate_support(dc, context);
+	dcn35_decide_zstate_support(dc, context);
 	DC_FP_END();
 
 	return out;
 }
 
-
 static struct resource_funcs dcn351_res_pool_funcs = {
 	.destroy = dcn351_destroy_resource_pool,
 	.link_enc_create = dcn35_link_encoder_create,
@@ -1871,6 +1872,9 @@ static bool dcn351_resource_construct(
 	/* Use pipe context based otg sync logic */
 	dc->config.use_pipe_ctx_sync_logic = true;
 
+	/* Use psp mailbox to enable assr */
+	dc->config.use_assr_psp_message = true;
+
 	/* read VBIOS LTTPR caps */
 	{
 		if (ctx->dc_bios->funcs->get_lttpr_caps) {
@@ -1890,6 +1894,8 @@ static bool dcn351_resource_construct(
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
 
+	/*HW default is to have all the FGCG enabled, SW no need to program them*/
+	dc->debug.enable_fine_grain_clock_gating.u32All = 0xFFFF;
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -2120,15 +2126,9 @@ static bool dcn351_resource_construct(
 		dc->dml2_options.minimize_dispclk_using_odm = true;
 	dc->dml2_options.enable_windowed_mpo_odm = dc->config.enable_windowed_mpo_odm;
 
-	dc->dml2_options.callbacks.dc = dc;
-	dc->dml2_options.callbacks.build_scaling_params = &resource_build_scaling_params;
+	resource_init_common_dml2_callbacks(dc, &dc->dml2_options);
 	dc->dml2_options.callbacks.can_support_mclk_switch_using_fw_based_vblank_stretch = &dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch;
-	dc->dml2_options.callbacks.acquire_secondary_pipe_for_mpc_odm = &dc_resource_acquire_secondary_pipe_for_mpc_odm_legacy;
-	dc->dml2_options.callbacks.update_pipes_for_stream_with_slice_count = &resource_update_pipes_for_stream_with_slice_count;
-	dc->dml2_options.callbacks.update_pipes_for_plane_with_slice_count = &resource_update_pipes_for_plane_with_slice_count;
-	dc->dml2_options.callbacks.get_mpc_slice_index = &resource_get_mpc_slice_index;
-	dc->dml2_options.callbacks.get_odm_slice_index = &resource_get_odm_slice_index;
-	dc->dml2_options.callbacks.get_opp_head = &resource_get_opp_head;
+
 	dc->dml2_options.max_segments_per_hubp = 24;
 	dc->dml2_options.det_segment_size = DCN3_2_DET_SEG_SIZE;/*todo*/