Merging upstream version 6.10.3.

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

13 files changed, 183 insertions, 180 deletions

diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile
index c4a5efd2dd..a94b6d546c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
@@ -25,40 +25,8 @@
 # It provides the general basic services required by other DAL
 # subcomponents.
 
-ifdef CONFIG_X86
-dml_ccflags-$(CONFIG_CC_IS_GCC) := -mhard-float
-dml_ccflags := $(dml_ccflags-y) -msse
-endif
-
-ifdef CONFIG_PPC64
-dml_ccflags := -mhard-float -maltivec
-endif
-
-ifdef CONFIG_ARM64
-dml_rcflags := -mgeneral-regs-only
-endif
-
-ifdef CONFIG_LOONGARCH
-dml_ccflags := -mfpu=64
-dml_rcflags := -msoft-float
-endif
-
-ifdef CONFIG_CC_IS_GCC
-ifneq ($(call gcc-min-version, 70100),y)
-IS_OLD_GCC = 1
-endif
-endif
-
-ifdef CONFIG_X86
-ifdef IS_OLD_GCC
-# Stack alignment mismatch, proceed with caution.
-# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3
-# (8B stack alignment).
-dml_ccflags += -mpreferred-stack-boundary=4
-else
-dml_ccflags += -msse2
-endif
-endif
+dml_ccflags := $(CC_FLAGS_FPU)
+dml_rcflags := $(CC_FLAGS_NO_FPU)
 
 ifneq ($(CONFIG_FRAME_WARN),0)
 ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
index 38ab9ad60e..74da9ebda0 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
@@ -1085,6 +1085,9 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 		int minmum_z8_residency = dc->debug.minimum_z8_residency_time > 0 ? dc->debug.minimum_z8_residency_time : 1000;
 		bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > (double)minmum_z8_residency;
 		bool is_pwrseq0 = link->link_index == 0;
+		bool is_psr = (link && (link->psr_settings.psr_version == DC_PSR_VERSION_1 ||
+						link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) && !link->panel_config.psr.disable_psr);
+		bool is_replay = link && link->replay_settings.replay_feature_enabled;
 
 		/* Don't support multi-plane configurations */
 		if (stream_status->plane_count > 1)
@@ -1092,8 +1095,8 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 
 		if (is_pwrseq0 && context->bw_ctx.dml.vba.StutterPeriod > 5000.0)
 			return DCN_ZSTATE_SUPPORT_ALLOW;
-		else if (is_pwrseq0 && link->psr_settings.psr_version == DC_PSR_VERSION_1 && !link->panel_config.psr.disable_psr)
-			return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY : DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY;
+		else if (is_pwrseq0 && (is_psr || is_replay))
+			return DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY;
 		else
 			return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY : DCN_ZSTATE_SUPPORT_DISALLOW;
 	} else {
@@ -2369,7 +2372,7 @@ validate_out:
 
 static struct _vcs_dpi_voltage_scaling_st construct_low_pstate_lvl(struct clk_limit_table *clk_table, unsigned int high_voltage_lvl)
 {
-	struct _vcs_dpi_voltage_scaling_st low_pstate_lvl;
+	struct _vcs_dpi_voltage_scaling_st low_pstate_lvl = {0};
 	int i;
 
 	low_pstate_lvl.state = 1;
@@ -2474,7 +2477,7 @@ void dcn201_populate_dml_writeback_from_context_fpu(struct dc *dc,
 	int pipe_cnt, i, j;
 	double max_calc_writeback_dispclk;
 	double writeback_dispclk;
-	struct writeback_st dout_wb;
+	struct writeback_st dout_wb = {0};
 
 	dc_assert_fp_enabled();
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
index ccb4ad78f6..81f7b90849 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
@@ -260,7 +260,7 @@ void dcn30_fpu_populate_dml_writeback_from_context(
 	int pipe_cnt, i, j;
 	double max_calc_writeback_dispclk;
 	double writeback_dispclk;
-	struct writeback_st dout_wb;
+	struct writeback_st dout_wb = {0};
 
 	dc_assert_fp_enabled();
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
index e7f4a2d491..e0b52db2c2 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
@@ -3535,7 +3535,6 @@ static double TruncToValidBPP(
 			return DesiredBPP;
 		}
 	}
-	return BPP_INVALID;
 }
 
 void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
index 7307b7b8d8..94317b2e4a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
@@ -487,6 +487,7 @@ void dcn31_calculate_wm_and_dlg_fp(
 {
 	int i, pipe_idx, total_det = 0, active_hubp_count = 0;
 	double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
+	uint32_t cstate_enter_plus_exit_z8_ns;
 
 	dc_assert_fp_enabled();
 
@@ -506,6 +507,13 @@ void dcn31_calculate_wm_and_dlg_fp(
 	pipes[0].clks_cfg.dcfclk_mhz = dcfclk;
 	pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz;
 
+	cstate_enter_plus_exit_z8_ns =
+		get_wm_z8_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+
+	if (get_stutter_period(&context->bw_ctx.dml, pipes, pipe_cnt) < dc->debug.minimum_z8_residency_time &&
+			cstate_enter_plus_exit_z8_ns < dc->debug.minimum_z8_residency_time * 1000)
+		cstate_enter_plus_exit_z8_ns = dc->debug.minimum_z8_residency_time * 1000;
+
 	/* Set A:
 	 * All clocks min required
 	 *
@@ -516,7 +524,7 @@ void dcn31_calculate_wm_and_dlg_fp(
 	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
-	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_z8_ns = get_wm_z8_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_z8_ns = cstate_enter_plus_exit_z8_ns;
 	context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_z8_ns = get_wm_z8_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h
index 8f9c8faed2..d2ae43a82b 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h
@@ -30,6 +30,7 @@
 #define DCN3_15_DEFAULT_DET_SIZE 192
 #define DCN3_15_MIN_COMPBUF_SIZE_KB 128
 #define DCN3_16_DEFAULT_DET_SIZE 192
+#define DCN3_16_MIN_COMPBUF_SIZE_KB 128
 
 void dcn31_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
 				  int pipe_cnt);
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
index adea459e7d..33cf824c5d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
@@ -3679,7 +3679,6 @@ static double TruncToValidBPP(
 			return DesiredBPP;
 		}
 	}
-	return BPP_INVALID;
 }
 
 static noinline void CalculatePrefetchSchedulePerPlane(
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index fb21572750..21f637ae4a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -310,7 +310,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 {
 	int i, pipe_cnt;
 	struct resource_context *res_ctx = &context->res_ctx;
-	struct pipe_ctx *pipe;
+	struct pipe_ctx *pipe = 0;
 	bool upscaled = false;
 	const unsigned int max_allowed_vblank_nom = 1023;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
index 88e56889a6..3242957d00 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
@@ -3788,7 +3788,6 @@ static double TruncToValidBPP(
 			return DesiredBPP;
 		}
 	}
-	return BPP_INVALID;
 }
 
 static noinline void CalculatePrefetchSchedulePerPlane(
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index a0a65e0991..f6fe0a64be 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -180,6 +180,9 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc = {
 	.urgent_latency_adjustment_fabric_clock_reference_mhz = 3000,
 };
 
+static bool dcn32_apply_merge_split_flags_helper(struct dc *dc, struct dc_state *context,
+	bool *repopulate_pipes, int *split, bool *merge);
+
 void dcn32_build_wm_range_table_fpu(struct clk_mgr_internal *clk_mgr)
 {
 	/* defaults */
@@ -622,7 +625,7 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 		 *   to combine this with SubVP can cause issues with the scheduling).
 		 * - Not TMZ surface
 		 */
-		if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) &&
+		if (pipe->plane_state && !pipe->top_pipe && !pipe->prev_odm_pipe && !dcn32_is_center_timing(pipe) &&
 				!(pipe->stream->timing.pix_clk_100hz / 10000 > DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ) &&
 				(!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) &&
 				dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_NONE &&
@@ -720,7 +723,7 @@ static bool dcn32_enough_pipes_for_subvp(struct dc *dc, struct dc_state *context
  */
 static bool subvp_subvp_schedulable(struct dc *dc, struct dc_state *context)
 {
-	struct pipe_ctx *subvp_pipes[2];
+	struct pipe_ctx *subvp_pipes[2] = {0};
 	struct dc_stream_state *phantom = NULL;
 	uint32_t microschedule_lines = 0;
 	uint32_t index = 0;
@@ -1425,13 +1428,14 @@ static bool is_test_pattern_enabled(
 	return false;
 }
 
-static void dcn32_full_validate_bw_helper(struct dc *dc,
+static bool dcn32_full_validate_bw_helper(struct dc *dc,
 				   struct dc_state *context,
 				   display_e2e_pipe_params_st *pipes,
 				   int *vlevel,
 				   int *split,
 				   bool *merge,
-				   int *pipe_cnt)
+				   int *pipe_cnt,
+				   bool *repopulate_pipes)
 {
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 	unsigned int dc_pipe_idx = 0;
@@ -1461,6 +1465,12 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 		vba->VoltageLevel = *vlevel;
 	}
 
+	/* Apply split and merge flags before checking for subvp */
+	if (!dcn32_apply_merge_split_flags_helper(dc, context, repopulate_pipes, split, merge))
+		return false;
+	memset(split, 0, MAX_PIPES * sizeof(int));
+	memset(merge, 0, MAX_PIPES * sizeof(bool));
+
 	/* Conditions for setting up phantom pipes for SubVP:
 	 * 1. Not force disable SubVP
 	 * 2. Full update (i.e. !fast_validate)
@@ -1475,19 +1485,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 	    vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported ||
 	    dc->debug.force_subvp_mclk_switch)) {
 
-		dcn32_merge_pipes_for_subvp(dc, context);
-		memset(merge, 0, MAX_PIPES * sizeof(bool));
-
 		vlevel_temp = *vlevel;
-		/* to re-initialize viewport after the pipe merge */
-		for (i = 0; i < dc->res_pool->pipe_count; i++) {
-			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-
-			if (!pipe_ctx->plane_state || !pipe_ctx->stream)
-				continue;
-
-			resource_build_scaling_params(pipe_ctx);
-		}
 
 		while (!found_supported_config && dcn32_enough_pipes_for_subvp(dc, context) &&
 			dcn32_assign_subvp_pipe(dc, context, &dc_pipe_idx)) {
@@ -1576,8 +1574,6 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 			 * add phantom pipes. If pipe split (ODM / MPC) is required, both the main
 			 * and phantom pipes will be split in the regular pipe splitting sequence.
 			 */
-			memset(split, 0, MAX_PIPES * sizeof(int));
-			memset(merge, 0, MAX_PIPES * sizeof(bool));
 			*vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, *vlevel, split, merge);
 			vba->VoltageLevel = *vlevel;
 			// Note: We can't apply the phantom pipes to hardware at this time. We have to wait
@@ -1590,6 +1586,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 		try_odm_power_optimization_and_revalidate(
 				dc, context, pipes, split, merge, vlevel, *pipe_cnt);
 
+	return true;
 }
 
 static bool is_dtbclk_required(struct dc *dc, struct dc_state *context)
@@ -1929,106 +1926,23 @@ static bool dcn32_split_stream_for_mpc_or_odm(
 	return true;
 }
 
-bool dcn32_internal_validate_bw(struct dc *dc,
-				struct dc_state *context,
-				display_e2e_pipe_params_st *pipes,
-				int *pipe_cnt_out,
-				int *vlevel_out,
-				bool fast_validate)
+static bool dcn32_apply_merge_split_flags_helper(
+		struct dc *dc,
+		struct dc_state *context,
+		bool *repopulate_pipes,
+		int *split,
+		bool *merge)
 {
-	bool out = false;
-	bool repopulate_pipes = false;
-	int split[MAX_PIPES] = { 0 };
-	bool merge[MAX_PIPES] = { false };
+	int i, pipe_idx;
 	bool newly_split[MAX_PIPES] = { false };
-	int pipe_cnt, i, pipe_idx;
-	int vlevel = context->bw_ctx.dml.soc.num_states;
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 
-	dc_assert_fp_enabled();
-
-	ASSERT(pipes);
-	if (!pipes)
-		return false;
-
-	// For each full update, remove all existing phantom pipes first
-	dc_state_remove_phantom_streams_and_planes(dc, context);
-	dc_state_release_phantom_streams_and_planes(dc, context);
-
-	dc->res_pool->funcs->update_soc_for_wm_a(dc, context);
-
-	pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate);
-
-	if (!pipe_cnt) {
-		out = true;
-		goto validate_out;
-	}
-
-	dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt);
-	context->bw_ctx.dml.soc.max_vratio_pre = dcn32_determine_max_vratio_prefetch(dc, context);
-
-	if (!fast_validate)
-		dcn32_full_validate_bw_helper(dc, context, pipes, &vlevel, split, merge, &pipe_cnt);
-
-	if (fast_validate ||
-			(dc->debug.dml_disallow_alternate_prefetch_modes &&
-			(vlevel == context->bw_ctx.dml.soc.num_states ||
-				vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported))) {
-		/*
-		 * If dml_disallow_alternate_prefetch_modes is false, then we have already
-		 * tried alternate prefetch modes during full validation.
-		 *
-		 * If mode is unsupported or there is no p-state support, then
-		 * fall back to favouring voltage.
-		 *
-		 * If Prefetch mode 0 failed for this config, or passed with Max UCLK, then try
-		 * to support with Prefetch mode 1 (dm_prefetch_support_fclk_and_stutter == 2)
-		 */
-		context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
-			dm_prefetch_support_none;
-
-		context->bw_ctx.dml.validate_max_state = fast_validate;
-		vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt);
-
-		context->bw_ctx.dml.validate_max_state = false;
-
-		if (vlevel < context->bw_ctx.dml.soc.num_states) {
-			memset(split, 0, sizeof(split));
-			memset(merge, 0, sizeof(merge));
-			vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split, merge);
-			// dcn20_validate_apply_pipe_split_flags can modify voltage level outside of DML
-			vba->VoltageLevel = vlevel;
-		}
-	}
-
-	dml_log_mode_support_params(&context->bw_ctx.dml);
-
-	if (vlevel == context->bw_ctx.dml.soc.num_states)
-		goto validate_fail;
-
-	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
-		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
-		struct pipe_ctx *mpo_pipe = pipe->bottom_pipe;
-
-		if (!pipe->stream)
-			continue;
-
-		if (vba->ODMCombineEnabled[vba->pipe_plane[pipe_idx]] != dm_odm_combine_mode_disabled
-				&& !dc->config.enable_windowed_mpo_odm
-				&& pipe->plane_state && mpo_pipe
-				&& memcmp(&mpo_pipe->plane_state->clip_rect,
-						&pipe->stream->src,
-						sizeof(struct rect)) != 0) {
-			ASSERT(mpo_pipe->plane_state != pipe->plane_state);
-			goto validate_fail;
-		}
-		pipe_idx++;
-	}
-
 	if (dc->config.enable_windowed_mpo_odm) {
-		repopulate_pipes = update_pipes_with_split_flags(
-				dc, context, vba, split, merge);
+		if (update_pipes_with_split_flags(
+			dc, context, vba, split, merge))
+			*repopulate_pipes = true;
 	} else {
+
 		/* the code below will be removed once windowed mpo odm is fully
 		 * enabled.
 		 */
@@ -2085,7 +1999,7 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 				memset(&pipe->plane_res, 0, sizeof(pipe->plane_res));
 				memset(&pipe->stream_res, 0, sizeof(pipe->stream_res));
 				memset(&pipe->link_res, 0, sizeof(pipe->link_res));
-				repopulate_pipes = true;
+				*repopulate_pipes = true;
 			} else if (pipe->top_pipe && pipe->top_pipe->plane_state == pipe->plane_state) {
 				struct pipe_ctx *top_pipe = pipe->top_pipe;
 				struct pipe_ctx *bottom_pipe = pipe->bottom_pipe;
@@ -2101,7 +2015,7 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 				memset(&pipe->plane_res, 0, sizeof(pipe->plane_res));
 				memset(&pipe->stream_res, 0, sizeof(pipe->stream_res));
 				memset(&pipe->link_res, 0, sizeof(pipe->link_res));
-				repopulate_pipes = true;
+				*repopulate_pipes = true;
 			} else
 				ASSERT(0); /* Should never try to merge master pipe */
 
@@ -2140,15 +2054,15 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 				hsplit_pipe = dcn32_find_split_pipe(dc, context, old_index);
 				ASSERT(hsplit_pipe);
 				if (!hsplit_pipe)
-					goto validate_fail;
+					return false;
 
 				if (!dcn32_split_stream_for_mpc_or_odm(
 						dc, &context->res_ctx,
 						pipe, hsplit_pipe, odm))
-					goto validate_fail;
+					return false;
 
 				newly_split[hsplit_pipe->pipe_idx] = true;
-				repopulate_pipes = true;
+				*repopulate_pipes = true;
 			}
 			if (split[i] == 4) {
 				struct pipe_ctx *pipe_4to1;
@@ -2163,11 +2077,11 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 				pipe_4to1 = dcn32_find_split_pipe(dc, context, old_index);
 				ASSERT(pipe_4to1);
 				if (!pipe_4to1)
-					goto validate_fail;
+					return false;
 				if (!dcn32_split_stream_for_mpc_or_odm(
 						dc, &context->res_ctx,
 						pipe, pipe_4to1, odm))
-					goto validate_fail;
+					return false;
 				newly_split[pipe_4to1->pipe_idx] = true;
 
 				if (odm && old_pipe->next_odm_pipe && old_pipe->next_odm_pipe->next_odm_pipe
@@ -2182,11 +2096,11 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 				pipe_4to1 = dcn32_find_split_pipe(dc, context, old_index);
 				ASSERT(pipe_4to1);
 				if (!pipe_4to1)
-					goto validate_fail;
+					return false;
 				if (!dcn32_split_stream_for_mpc_or_odm(
 						dc, &context->res_ctx,
 						hsplit_pipe, pipe_4to1, odm))
-					goto validate_fail;
+					return false;
 				newly_split[pipe_4to1->pipe_idx] = true;
 			}
 			if (odm)
@@ -2198,11 +2112,122 @@ bool dcn32_internal_validate_bw(struct dc *dc,
 
 			if (pipe->plane_state) {
 				if (!resource_build_scaling_params(pipe))
-					goto validate_fail;
+					return false;
 			}
 		}
+
+		for (i = 0; i < context->stream_count; i++) {
+			struct pipe_ctx *otg_master = resource_get_otg_master_for_stream(&context->res_ctx,
+					context->streams[i]);
+
+			if (otg_master)
+				resource_build_test_pattern_params(&context->res_ctx, otg_master);
+		}
+	}
+	return true;
+}
+
+bool dcn32_internal_validate_bw(struct dc *dc,
+				struct dc_state *context,
+				display_e2e_pipe_params_st *pipes,
+				int *pipe_cnt_out,
+				int *vlevel_out,
+				bool fast_validate)
+{
+	bool out = false;
+	bool repopulate_pipes = false;
+	int split[MAX_PIPES] = { 0 };
+	bool merge[MAX_PIPES] = { false };
+	int pipe_cnt, i, pipe_idx;
+	int vlevel = context->bw_ctx.dml.soc.num_states;
+	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+
+	dc_assert_fp_enabled();
+
+	ASSERT(pipes);
+	if (!pipes)
+		return false;
+
+	/* For each full update, remove all existing phantom pipes first */
+	dc_state_remove_phantom_streams_and_planes(dc, context);
+	dc_state_release_phantom_streams_and_planes(dc, context);
+
+	dc->res_pool->funcs->update_soc_for_wm_a(dc, context);
+
+	pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate);
+
+	if (!pipe_cnt) {
+		out = true;
+		goto validate_out;
+	}
+
+	dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt);
+	context->bw_ctx.dml.soc.max_vratio_pre = dcn32_determine_max_vratio_prefetch(dc, context);
+
+	if (!fast_validate) {
+		if (!dcn32_full_validate_bw_helper(dc, context, pipes, &vlevel, split, merge,
+			&pipe_cnt, &repopulate_pipes))
+			goto validate_fail;
+	}
+
+	if (fast_validate ||
+			(dc->debug.dml_disallow_alternate_prefetch_modes &&
+			(vlevel == context->bw_ctx.dml.soc.num_states ||
+				vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported))) {
+		/*
+		 * If dml_disallow_alternate_prefetch_modes is false, then we have already
+		 * tried alternate prefetch modes during full validation.
+		 *
+		 * If mode is unsupported or there is no p-state support, then
+		 * fall back to favouring voltage.
+		 *
+		 * If Prefetch mode 0 failed for this config, or passed with Max UCLK, then try
+		 * to support with Prefetch mode 1 (dm_prefetch_support_fclk_and_stutter == 2)
+		 */
+		context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final =
+			dm_prefetch_support_none;
+
+		context->bw_ctx.dml.validate_max_state = fast_validate;
+		vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt);
+
+		context->bw_ctx.dml.validate_max_state = false;
+
+		if (vlevel < context->bw_ctx.dml.soc.num_states) {
+			memset(split, 0, sizeof(split));
+			memset(merge, 0, sizeof(merge));
+			vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split, merge);
+			/* dcn20_validate_apply_pipe_split_flags can modify voltage level outside of DML */
+			vba->VoltageLevel = vlevel;
+		}
 	}
 
+	dml_log_mode_support_params(&context->bw_ctx.dml);
+
+	if (vlevel == context->bw_ctx.dml.soc.num_states)
+		goto validate_fail;
+
+	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+		struct pipe_ctx *mpo_pipe = pipe->bottom_pipe;
+
+		if (!pipe->stream)
+			continue;
+
+		if (vba->ODMCombineEnabled[vba->pipe_plane[pipe_idx]] != dm_odm_combine_mode_disabled
+				&& !dc->config.enable_windowed_mpo_odm
+				&& pipe->plane_state && mpo_pipe
+				&& memcmp(&mpo_pipe->plane_state->clip_rect,
+						&pipe->stream->src,
+						sizeof(struct rect)) != 0) {
+			ASSERT(mpo_pipe->plane_state != pipe->plane_state);
+			goto validate_fail;
+		}
+		pipe_idx++;
+	}
+
+	if (!dcn32_apply_merge_split_flags_helper(dc, context, &repopulate_pipes, split, merge))
+		goto validate_fail;
+
 	/* Actual dsc count per stream dsc validation*/
 	if (!dcn20_validate_dsc(dc, context)) {
 		vba->ValidationStatus[vba->soc.num_states] = DML_FAIL_DSC_VALIDATION_FAILURE;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
index 80fccd4999..ba1310c8fd 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
@@ -1650,6 +1650,8 @@ double dml32_TruncToValidBPP(
 			MaxLinkBPP = 2 * MaxLinkBPP;
 	}
 
+	*RequiredSlots = dml_ceil(DesiredBPP / MaxLinkBPP * 64, 1);
+
 	if (DesiredBPP == 0) {
 		if (DSCEnable) {
 			if (MaxLinkBPP < MinDSCBPP)
@@ -1676,10 +1678,6 @@ double dml32_TruncToValidBPP(
 		else
 			return DesiredBPP;
 	}
-
-	*RequiredSlots = dml_ceil(DesiredBPP / MaxLinkBPP * 64, 1);
-
-	return BPP_INVALID;
 } // TruncToValidBPP
 
 double dml32_RequiredDTBCLK(
@@ -1975,8 +1973,8 @@ void dml32_CalculateVMRowAndSwath(
 	unsigned int PTEBufferSizeInRequestsForChroma[DC__NUM_DPP__MAX];
 	unsigned int PDEAndMetaPTEBytesFrameY;
 	unsigned int PDEAndMetaPTEBytesFrameC;
-	unsigned int MetaRowByteY[DC__NUM_DPP__MAX];
-	unsigned int MetaRowByteC[DC__NUM_DPP__MAX];
+	unsigned int MetaRowByteY[DC__NUM_DPP__MAX] = {0};
+	unsigned int MetaRowByteC[DC__NUM_DPP__MAX] = {0};
 	unsigned int PixelPTEBytesPerRowY[DC__NUM_DPP__MAX];
 	unsigned int PixelPTEBytesPerRowC[DC__NUM_DPP__MAX];
 	unsigned int PixelPTEBytesPerRowY_one_row_per_frame[DC__NUM_DPP__MAX];
@@ -4291,7 +4289,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 	unsigned int i, j, k;
 	unsigned int SurfaceWithMinActiveFCLKChangeMargin = 0;
 	unsigned int DRAMClockChangeSupportNumber = 0;
-	unsigned int LastSurfaceWithoutMargin;
+	unsigned int LastSurfaceWithoutMargin = 0;
 	unsigned int DRAMClockChangeMethod = 0;
 	bool FoundFirstSurfaceWithMinActiveFCLKChangeMargin = false;
 	double MinActiveFCLKChangeMargin = 0.;
@@ -5656,9 +5654,9 @@ void dml32_CalculateStutterEfficiency(
 	double LastZ8StutterPeriod = 0.0;
 	double LastStutterPeriod = 0.0;
 	unsigned int TotalNumberOfActiveOTG = 0;
-	double doublePixelClock;
-	unsigned int doubleHTotal;
-	unsigned int doubleVTotal;
+	double doublePixelClock = 0;
+	unsigned int doubleHTotal = 0;
+	unsigned int doubleVTotal = 0;
 	bool SameTiming = true;
 	double DETBufferingTimeY;
 	double SwathWidthYCriticalSurface = 0.0;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c
index 6716696df7..beed7adbbd 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c
@@ -439,7 +439,7 @@ int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc,
 {
 	int i, pipe_cnt;
 	struct resource_context *res_ctx = &context->res_ctx;
-	struct pipe_ctx *pipe;
+	struct pipe_ctx *pipe = 0;
 	bool upscaled = false;
 	const unsigned int max_allowed_vblank_nom = 1023;
 
@@ -577,6 +577,7 @@ void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context)
 {
 	enum dcn_zstate_support_state support = DCN_ZSTATE_SUPPORT_DISALLOW;
 	unsigned int i, plane_count = 0;
+	DC_LOGGER_INIT(dc->ctx->logger);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		if (context->res_ctx.pipe_ctx[i].plane_state)
@@ -602,11 +603,14 @@ void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context)
 		if (is_pwrseq0 && allow_z10)
 			support = DCN_ZSTATE_SUPPORT_ALLOW;
 		else if (is_pwrseq0 && (is_psr || is_replay))
-			support = allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY : DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY;
+			support = DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY;
 		else if (allow_z8)
 			support = DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY;
 
 	}
 
+	DC_LOG_SMU("zstate_support: %d, StutterPeriod: %d\n", support,
+		   (int)context->bw_ctx.dml.vba.StutterPeriod);
+
 	context->bw_ctx.bw.dcn.clk.zstate_support = support;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c
index 40ca38dd1b..a201dbb743 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c
@@ -237,7 +237,6 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_51_soc = {
 	.urgent_latency_adjustment_fabric_clock_component_us = 0,
 	.urgent_latency_adjustment_fabric_clock_reference_mhz = 0,
 	.num_chans = 4,
-	.dram_clock_change_latency_us = 11.72,
 	.dispclk_dppclk_vco_speed_mhz = 2400.0,
 };
 
@@ -474,7 +473,7 @@ int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc,
 {
 	int i, pipe_cnt;
 	struct resource_context *res_ctx = &context->res_ctx;
-	struct pipe_ctx *pipe;
+	struct pipe_ctx *pipe = 0;
 	bool upscaled = false;
 	const unsigned int max_allowed_vblank_nom = 1023;