Merge tag 'amd-drm-next-6.4-2023-04-14' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.4-2023-04-14:

amdgpu:
- S4 fixes for APUs
- GFX11 fixes
- Misc code cleanups
- DCN 3.2 fixes
- DCN 3.1.4 fixes
- FPO/FAMS work to improve display power savings
- DP fixes
- UMC 8.10 code cleanup
- SDMA v4 fix
- GPU clock counter fixes
- SMU 13 fixes
- Sdma v6 invalidation fix for preemption
- RAS fixes
- S0ix fix
- GC 9.4.3 updates

amdkfd:
- Fix user pointers with IOMMU
- Fix coherency flag handling

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230414204609.7942-1-alexander.deucher@amd.com

35 files changed, 852 insertions, 65 deletions

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index e29655a26dd4..6cacb76f389e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1890,7 +1890,8 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev)
 	if (adev->dm.dc)
 		dc_deinit_callbacks(adev->dm.dc);
 
-	dc_dmub_srv_destroy(&adev->dm.dc->ctx->dmub_srv);
+	if (adev->dm.dc)
+		dc_dmub_srv_destroy(&adev->dm.dc->ctx->dmub_srv);
 
 	if (dc_enable_dmub_notifications(adev->dm.dc)) {
 		kfree(adev->dm.dmub_notify);
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
index f0f948501e9a..cce47d3f1a13 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -515,11 +515,8 @@ static enum bp_result get_gpio_i2c_info(
 	info->i2c_slave_address = record->i2c_slave_addr;
 
 	/* TODO: check how to get register offset for en, Y, etc. */
-	info->gpio_info.clk_a_register_index =
-			le16_to_cpu(
-			header->gpio_pin[table_index].data_a_reg_index);
-	info->gpio_info.clk_a_shift =
-			header->gpio_pin[table_index].gpio_bitshift;
+	info->gpio_info.clk_a_register_index = le16_to_cpu(pin->data_a_reg_index);
+	info->gpio_info.clk_a_shift = pin->gpio_bitshift;
 
 	return BP_RESULT_OK;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index af108f88b112..ea753f8fa175 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -399,6 +399,23 @@ static void dcn32_update_clocks_update_dentist(
 
 }
 
+static int dcn32_get_dispclk_from_dentist(struct clk_mgr *clk_mgr_base)
+{
+	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+	uint32_t dispclk_wdivider;
+	int disp_divider;
+
+	REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, &dispclk_wdivider);
+	disp_divider = dentist_get_divider_from_did(dispclk_wdivider);
+
+	/* Return DISPCLK freq in Khz */
+	if (disp_divider)
+		return (DENTIST_DIVIDER_RANGE_SCALE_FACTOR * clk_mgr->base.dentist_vco_freq_khz) / disp_divider;
+
+	return 0;
+}
+
+
 static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 			struct dc_state *context,
 			bool safe_to_lower)
@@ -852,6 +869,7 @@ static struct clk_mgr_funcs dcn32_funcs = {
 		.are_clock_states_equal = dcn32_are_clock_states_equal,
 		.enable_pme_wa = dcn32_enable_pme_wa,
 		.is_smu_present = dcn32_is_smu_present,
+		.get_dispclk_from_dentist = dcn32_get_dispclk_from_dentist,
 };
 
 void dcn32_clk_mgr_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c
index 41198c729d90..30c0644d4418 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c
@@ -622,7 +622,6 @@ bool link_enc_cfg_validate(struct dc *dc, struct dc_state *state)
 	int i, j;
 	uint8_t valid_count = 0;
 	uint8_t dig_stream_count = 0;
-	int matching_stream_ptrs = 0;
 	int eng_ids_per_ep_id[MAX_PIPES] = {0};
 	int ep_ids_per_eng_id[MAX_PIPES] = {0};
 	int valid_bitmap = 0;
@@ -645,9 +644,7 @@ bool link_enc_cfg_validate(struct dc *dc, struct dc_state *state)
 		struct link_enc_assignment assignment = state->res_ctx.link_enc_cfg_ctx.link_enc_assignments[i];
 
 		if (assignment.valid) {
-			if (assignment.stream == state->streams[i])
-				matching_stream_ptrs++;
-			else
+			if (assignment.stream != state->streams[i])
 				valid_stream_ptrs = false;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index e363a3c88250..23ee63b98dcd 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.229"
+#define DC_VER "3.2.230"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
@@ -405,6 +405,7 @@ struct dc_config {
 	bool force_bios_enable_lttpr;
 	uint8_t force_bios_fixed_vs;
 	int sdpif_request_limit_words_per_umc;
+	bool use_old_fixed_vs_sequence;
 	bool disable_subvp_drr;
 };
 
@@ -875,6 +876,9 @@ struct dc_debug_options {
 	bool override_dispclk_programming;
 	bool disable_fpo_optimizations;
 	bool support_eDP1_5;
+	uint32_t fpo_vactive_margin_us;
+	bool disable_fpo_vactive;
+	bool disable_boot_optimizations;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index dd6f643254fe..a9b9490a532c 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -327,6 +327,7 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru
 	int i = 0, k = 0;
 	int ramp_up_num_steps = 1; // TODO: Ramp is currently disabled. Reenable it.
 	uint8_t visual_confirm_enabled;
+	int pipe_idx = 0;
 
 	if (dc == NULL)
 		return false;
@@ -339,6 +340,25 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru
 	cmd.fw_assisted_mclk_switch.config_data.fams_enabled = should_manage_pstate;
 	cmd.fw_assisted_mclk_switch.config_data.visual_confirm_enabled = visual_confirm_enabled;
 
+	if (should_manage_pstate) {
+		for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
+			struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+			if (!pipe->stream)
+				continue;
+
+			/* If FAMS is being used to support P-State and there is a stream
+			 * that does not use FAMS, we are in an FPO + VActive scenario.
+			 * Assign vactive stretch margin in this case.
+			 */
+			if (!pipe->stream->fpo_in_use) {
+				cmd.fw_assisted_mclk_switch.config_data.vactive_stretch_margin_us = dc->debug.fpo_vactive_margin_us;
+				break;
+			}
+			pipe_idx++;
+		}
+	}
+
 	for (i = 0, k = 0; context && i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
index a142a00bc432..bf399819ca80 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c
@@ -755,8 +755,8 @@ bool hubp1_is_flip_pending(struct hubp *hubp)
 	return false;
 }
 
-uint32_t aperture_default_system = 1;
-uint32_t context0_default_system; /* = 0;*/
+static uint32_t aperture_default_system = 1;
+static uint32_t context0_default_system; /* = 0;*/
 
 static void hubp1_set_vm_system_aperture_settings(struct hubp *hubp,
 		struct vm_system_aperture_param *apt)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
index 893c0809cd4e..7bdc146f7cb5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
@@ -205,6 +205,11 @@
 	type PHYDSYMCLK_GATE_DISABLE; \
 	type PHYESYMCLK_GATE_DISABLE;
 
+#define DCCG314_REG_FIELD_LIST(type) \
+	type DSCCLK3_DTO_PHASE;\
+	type DSCCLK3_DTO_MODULO;\
+	type DSCCLK3_DTO_ENABLE;
+
 #define DCCG32_REG_FIELD_LIST(type) \
 	type DPSTREAMCLK0_EN;\
 	type DPSTREAMCLK1_EN;\
@@ -237,6 +242,7 @@ struct dccg_shift {
 	DCCG_REG_FIELD_LIST(uint8_t)
 	DCCG3_REG_FIELD_LIST(uint8_t)
 	DCCG31_REG_FIELD_LIST(uint8_t)
+	DCCG314_REG_FIELD_LIST(uint8_t)
 	DCCG32_REG_FIELD_LIST(uint8_t)
 };
 
@@ -244,6 +250,7 @@ struct dccg_mask {
 	DCCG_REG_FIELD_LIST(uint32_t)
 	DCCG3_REG_FIELD_LIST(uint32_t)
 	DCCG31_REG_FIELD_LIST(uint32_t)
+	DCCG314_REG_FIELD_LIST(uint32_t)
 	DCCG32_REG_FIELD_LIST(uint32_t)
 };
 
@@ -273,6 +280,7 @@ struct dccg_registers {
 	uint32_t DSCCLK0_DTO_PARAM;
 	uint32_t DSCCLK1_DTO_PARAM;
 	uint32_t DSCCLK2_DTO_PARAM;
+	uint32_t DSCCLK3_DTO_PARAM;
 	uint32_t DPSTREAMCLK_ROOT_GATE_DISABLE;
 	uint32_t DPSTREAMCLK_GATE_DISABLE;
 	uint32_t DCCG_GATE_DISABLE_CNTL;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
index 7d2b982506fd..4c2fdfea162f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
@@ -360,6 +360,15 @@ void dccg31_disable_dscclk(struct dccg *dccg, int inst)
 				DSCCLK2_DTO_PHASE, 0,
 				DSCCLK2_DTO_MODULO, 1);
 		break;
+	case 3:
+		if (REG(DSCCLK3_DTO_PARAM)) {
+			REG_UPDATE(DSCCLK_DTO_CTRL,
+					DSCCLK3_DTO_ENABLE, 1);
+			REG_UPDATE_2(DSCCLK3_DTO_PARAM,
+					DSCCLK3_DTO_PHASE, 0,
+					DSCCLK3_DTO_MODULO, 1);
+		}
+		break;
 	default:
 		BREAK_TO_DEBUGGER();
 		return;
@@ -395,6 +404,15 @@ void dccg31_enable_dscclk(struct dccg *dccg, int inst)
 		REG_UPDATE(DSCCLK_DTO_CTRL,
 				DSCCLK2_DTO_ENABLE, 0);
 		break;
+	case 3:
+		if (REG(DSCCLK3_DTO_PARAM)) {
+			REG_UPDATE(DSCCLK_DTO_CTRL,
+					DSCCLK3_DTO_ENABLE, 0);
+			REG_UPDATE_2(DSCCLK3_DTO_PARAM,
+					DSCCLK3_DTO_PHASE, 0,
+					DSCCLK3_DTO_MODULO, 0);
+		}
+		break;
 	default:
 		BREAK_TO_DEBUGGER();
 		return;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
index eaaa2e01f6d0..ff8cd5076434 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
@@ -1965,6 +1965,8 @@ static bool dcn31_resource_construct(
 	dc->caps.color.mpc.ogam_rom_caps.hlg = 0;
 	dc->caps.color.mpc.ocsc = 1;
 
+	dc->config.use_old_fixed_vs_sequence = true;
+
 	/* Use pipe context based otg sync logic */
 	dc->config.use_pipe_ctx_sync_logic = true;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
index 081ce168f621..6f879265ad9c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
@@ -274,6 +274,32 @@ static void dccg314_set_dpstreamclk(
 	}
 }
 
+void dccg314_init(struct dccg *dccg)
+{
+	int otg_inst;
+
+	/* Set HPO stream encoder to use refclk to avoid case where PHY is
+	 * disabled and SYMCLK32 for HPO SE is sourced from PHYD32CLK which
+	 * will cause DCN to hang.
+	 */
+	for (otg_inst = 0; otg_inst < 4; otg_inst++)
+		dccg31_disable_symclk32_se(dccg, otg_inst);
+
+	if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le)
+		for (otg_inst = 0; otg_inst < 2; otg_inst++)
+			dccg31_disable_symclk32_le(dccg, otg_inst);
+
+	if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpstream)
+		for (otg_inst = 0; otg_inst < 4; otg_inst++)
+			dccg314_set_dpstreamclk(dccg, REFCLK, otg_inst,
+						otg_inst);
+
+	if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
+		for (otg_inst = 0; otg_inst < 5; otg_inst++)
+			dccg31_set_physymclk(dccg, otg_inst,
+					     PHYSYMCLK_FORCE_SRC_SYMCLK, false);
+}
+
 static void dccg314_set_valid_pixel_rate(
 		struct dccg *dccg,
 		int ref_dtbclk_khz,
@@ -315,7 +341,7 @@ static const struct dccg_funcs dccg314_funcs = {
 	.update_dpp_dto = dccg31_update_dpp_dto,
 	.dpp_root_clock_control = dccg314_dpp_root_clock_control,
 	.get_dccg_ref_freq = dccg31_get_dccg_ref_freq,
-	.dccg_init = dccg31_init,
+	.dccg_init = dccg314_init,
 	.set_dpstreamclk = dccg314_set_dpstreamclk,
 	.enable_symclk32_se = dccg31_enable_symclk32_se,
 	.disable_symclk32_se = dccg31_disable_symclk32_se,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h
index 6a35986307af..90687a9e8fdd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h
@@ -68,6 +68,7 @@
 	SR(DSCCLK0_DTO_PARAM),\
 	SR(DSCCLK1_DTO_PARAM),\
 	SR(DSCCLK2_DTO_PARAM),\
+	SR(DSCCLK3_DTO_PARAM),\
 	SR(DSCCLK_DTO_CTRL),\
 	SR(DCCG_GATE_DISABLE_CNTL2),\
 	SR(DCCG_GATE_DISABLE_CNTL3),\
@@ -149,12 +150,20 @@
 	DCCG_SF(DSCCLK1_DTO_PARAM, DSCCLK1_DTO_MODULO, mask_sh),\
 	DCCG_SF(DSCCLK2_DTO_PARAM, DSCCLK2_DTO_PHASE, mask_sh),\
 	DCCG_SF(DSCCLK2_DTO_PARAM, DSCCLK2_DTO_MODULO, mask_sh),\
+	DCCG_SF(DSCCLK3_DTO_PARAM, DSCCLK3_DTO_PHASE, mask_sh),\
+	DCCG_SF(DSCCLK3_DTO_PARAM, DSCCLK3_DTO_MODULO, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_SE0_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_SE1_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_SE2_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_SE3_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_LE0_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_LE1_GATE_DISABLE, mask_sh),\
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_SE0_GATE_DISABLE, mask_sh),\
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_SE1_GATE_DISABLE, mask_sh),\
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_SE2_GATE_DISABLE, mask_sh),\
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_SE3_GATE_DISABLE, mask_sh),\
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_LE0_GATE_DISABLE, mask_sh),\
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_LE1_GATE_DISABLE, mask_sh),\
 	DCCG_SF(HDMISTREAMCLK0_DTO_PARAM, HDMISTREAMCLK0_DTO_PHASE, mask_sh),\
 	DCCG_SF(HDMISTREAMCLK0_DTO_PARAM, HDMISTREAMCLK0_DTO_MODULO, mask_sh)
 
@@ -178,6 +187,7 @@
 	DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK0_DTO_ENABLE, mask_sh),\
 	DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK1_DTO_ENABLE, mask_sh),\
 	DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK2_DTO_ENABLE, mask_sh),\
+	DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK3_DTO_ENABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYASYMCLK_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYBSYMCLK_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYCSYMCLK_GATE_DISABLE, mask_sh),\
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index b3824287c224..db0974fe58ab 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -721,10 +721,19 @@ static void dcn32_initialize_min_clocks(struct dc *dc)
 	clocks->socclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].socclk_mhz * 1000;
 	clocks->dramclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 1000;
 	clocks->dppclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dppclk_mhz * 1000;
-	clocks->dispclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dispclk_mhz * 1000;
-	clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000;
-	clocks->fclk_p_state_change_support = true;
-	clocks->p_state_change_support = true;
+	if (dc->debug.disable_boot_optimizations) {
+		clocks->dispclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dispclk_mhz * 1000;
+	} else {
+		/* Even though DPG_EN = 1 for the connected display, it still requires the
+		 * correct timing so we cannot set DISPCLK to min freq or it could cause
+		 * audio corruption. Read current DISPCLK from DENTIST and request the same
+		 * freq to ensure that the timing is valid and unchanged.
+		 */
+		clocks->dispclk_khz = dc->clk_mgr->funcs->get_dispclk_from_dentist(dc->clk_mgr);
+		clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000;
+		clocks->fclk_p_state_change_support = true;
+		clocks->p_state_change_support = true;
+	}
 
 	dc->clk_mgr->funcs->update_clocks(
 			dc->clk_mgr,
@@ -823,7 +832,14 @@ void dcn32_init_hw(struct dc *dc)
 	 * everything down.
 	 */
 	if (dcb->funcs->is_accelerated_mode(dcb) || !dc->config.seamless_boot_edp_requested) {
-		hws->funcs.init_pipes(dc, dc->current_state);
+		/* Disable boot optimizations means power down everything including PHY, DIG,
+		 * and OTG (i.e. the boot is not optimized because we do a full power down).
+		 */
+		if (dc->hwss.enable_accelerated_mode && dc->debug.disable_boot_optimizations)
+			dc->hwss.enable_accelerated_mode(dc, dc->current_state);
+		else
+			hws->funcs.init_pipes(dc, dc->current_state);
+
 		if (dc->res_pool->hubbub->funcs->allow_self_refresh_control)
 			dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub,
 					!dc->res_pool->hubbub->ctx->dc->debug.disable_stutter);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 502f990346b1..e30d1f60695d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -726,6 +726,9 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_unbounded_requesting = false,
 	.override_dispclk_programming = true,
 	.disable_fpo_optimizations = false,
+	.fpo_vactive_margin_us = 2000, // 2000us
+	.disable_fpo_vactive = true,
+	.disable_boot_optimizations = false,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 10a3350376e4..3937dbc1e552 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -39,6 +39,7 @@
 #define DCN3_2_MBLK_HEIGHT_8BPE 64
 #define DCN3_2_VMIN_DISPCLK_HZ 717000000
 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq
+#define DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US 100 // Only allow FPO + Vactive if active margin >= 100
 
 #define TO_DCN32_RES_POOL(pool)\
 	container_of(pool, struct dcn32_resource_pool, base)
@@ -146,6 +147,8 @@ void dcn32_restore_mall_state(struct dc *dc,
 		struct dc_state *context,
 		struct mall_temp_config *temp_config);
 
+struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, const struct dc_state *context);
+
 bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe);
 
 unsigned int dcn32_calc_num_avail_chans_for_mall(struct dc *dc, int num_chans);
@@ -472,6 +475,7 @@ double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *conte
       SRI_ARR(OTG_H_BLANK, DSCL, id), SRI_ARR(OTG_V_BLANK, DSCL, id),          \
       SRI_ARR(SCL_MODE, DSCL, id), SRI_ARR(LB_DATA_FORMAT, DSCL, id),          \
       SRI_ARR(LB_MEMORY_CTRL, DSCL, id), SRI_ARR(DSCL_AUTOCAL, DSCL, id),      \
+      SRI_ARR(DSCL_CONTROL, DSCL, id),                                         \
       SRI_ARR(SCL_TAP_CONTROL, DSCL, id),                                      \
       SRI_ARR(SCL_COEF_RAM_TAP_SELECT, DSCL, id),                              \
       SRI_ARR(SCL_COEF_RAM_TAP_DATA, DSCL, id),                                \
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index 47fa51c1d3f4..eeca16faf31a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -27,6 +27,7 @@
 #include "dcn32_resource.h"
 #include "dcn20/dcn20_resource.h"
 #include "dml/dcn32/display_mode_vba_util_32.h"
+#include "dml/dcn32/dcn32_fpu.h"
 
 static bool is_dual_plane(enum surface_pixel_format format)
 {
@@ -500,3 +501,158 @@ void dcn32_restore_mall_state(struct dc *dc,
 			pipe->plane_state->is_phantom = temp_config->is_phantom_plane[i];
 	}
 }
+
+#define MAX_STRETCHED_V_BLANK 1000 // in micro-seconds (must ensure to match value in FW)
+/*
+ * Scaling factor for v_blank stretch calculations considering timing in
+ * micro-seconds and pixel clock in 100hz.
+ * Note: the parenthesis are necessary to ensure the correct order of
+ * operation where V_SCALE is used.
+ */
+#define V_SCALE (10000 / MAX_STRETCHED_V_BLANK)
+
+static int get_frame_rate_at_max_stretch_100hz(
+		struct dc_stream_state *fpo_candidate_stream,
+		uint32_t fpo_vactive_margin_us)
+{
+	struct dc_crtc_timing *timing = NULL;
+	uint32_t sec_per_100_lines;
+	uint32_t max_v_blank;
+	uint32_t curr_v_blank;
+	uint32_t v_stretch_max;
+	uint32_t stretched_frame_pix_cnt;
+	uint32_t scaled_stretched_frame_pix_cnt;
+	uint32_t scaled_refresh_rate;
+	uint32_t v_scale;
+
+	if (fpo_candidate_stream == NULL)
+		return 0;
+
+	/* check if refresh rate at least 120hz */
+	timing = &fpo_candidate_stream->timing;
+	if (timing == NULL)
+		return 0;
+
+	v_scale = 10000 / (MAX_STRETCHED_V_BLANK + fpo_vactive_margin_us);
+
+	sec_per_100_lines = timing->pix_clk_100hz / timing->h_total + 1;
+	max_v_blank = sec_per_100_lines / v_scale + 1;
+	curr_v_blank = timing->v_total - timing->v_addressable;
+	v_stretch_max = (max_v_blank > curr_v_blank) ? (max_v_blank - curr_v_blank) : (0);
+	stretched_frame_pix_cnt = (v_stretch_max + timing->v_total) * timing->h_total;
+	scaled_stretched_frame_pix_cnt = stretched_frame_pix_cnt / 10000;
+	scaled_refresh_rate = (timing->pix_clk_100hz) / scaled_stretched_frame_pix_cnt + 1;
+
+	return scaled_refresh_rate;
+
+}
+
+static bool is_refresh_rate_support_mclk_switch_using_fw_based_vblank_stretch(
+		struct dc_stream_state *fpo_candidate_stream, uint32_t fpo_vactive_margin_us)
+{
+	int refresh_rate_max_stretch_100hz;
+	int min_refresh_100hz;
+
+	if (fpo_candidate_stream == NULL)
+		return false;
+
+	refresh_rate_max_stretch_100hz = get_frame_rate_at_max_stretch_100hz(fpo_candidate_stream, fpo_vactive_margin_us);
+	min_refresh_100hz = fpo_candidate_stream->timing.min_refresh_in_uhz / 10000;
+
+	if (refresh_rate_max_stretch_100hz < min_refresh_100hz)
+		return false;
+
+	return true;
+}
+
+static int get_refresh_rate(struct dc_stream_state *fpo_candidate_stream)
+{
+	int refresh_rate = 0;
+	int h_v_total = 0;
+	struct dc_crtc_timing *timing = NULL;
+
+	if (fpo_candidate_stream == NULL)
+		return 0;
+
+	/* check if refresh rate at least 120hz */
+	timing = &fpo_candidate_stream->timing;
+	if (timing == NULL)
+		return 0;
+
+	h_v_total = timing->h_total * timing->v_total;
+	if (h_v_total == 0)
+		return 0;
+
+	refresh_rate = ((timing->pix_clk_100hz * 100) / (h_v_total)) + 1;
+	return refresh_rate;
+}
+
+/**
+ * dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch - Determines if config can support FPO
+ *
+ * @param [in]: dc - current dc state
+ * @param [in]: context - new dc state
+ *
+ * Return: Pointer to FPO stream candidate if config can support FPO, otherwise NULL
+ */
+struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, const struct dc_state *context)
+{
+	int refresh_rate = 0;
+	const int minimum_refreshrate_supported = 120;
+	struct dc_stream_state *fpo_candidate_stream = NULL;
+	bool is_fpo_vactive = false;
+	uint32_t fpo_vactive_margin_us = 0;
+
+	if (context == NULL)
+		return NULL;
+
+	if (dc->debug.disable_fams)
+		return NULL;
+
+	if (!dc->caps.dmub_caps.mclk_sw)
+		return NULL;
+
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching_shut_down)
+		return NULL;
+
+	/* For FPO we can support up to 2 display configs if:
+	 * - first display uses FPO
+	 * - Second display switches in VACTIVE */
+	if (context->stream_count > 2)
+		return NULL;
+	else if (context->stream_count == 2) {
+		DC_FP_START();
+		dcn32_assign_fpo_vactive_candidate(dc, context, &fpo_candidate_stream);
+		DC_FP_END();
+
+		DC_FP_START();
+		is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US);
+		DC_FP_END();
+		if (!is_fpo_vactive || dc->debug.disable_fpo_vactive)
+			return NULL;
+	} else
+		fpo_candidate_stream = context->streams[0];
+
+	if (!fpo_candidate_stream)
+		return NULL;
+
+	if (fpo_candidate_stream->sink->edid_caps.panel_patch.disable_fams)
+		return NULL;
+
+	refresh_rate = get_refresh_rate(fpo_candidate_stream);
+	if (refresh_rate < minimum_refreshrate_supported)
+		return NULL;
+
+	fpo_vactive_margin_us = is_fpo_vactive ? dc->debug.fpo_vactive_margin_us : 0; // For now hardcode the FPO + Vactive stretch margin to be 2000us
+	if (!is_refresh_rate_support_mclk_switch_using_fw_based_vblank_stretch(fpo_candidate_stream, fpo_vactive_margin_us))
+		return NULL;
+
+	// check if freesync enabled
+	if (!fpo_candidate_stream->allow_freesync)
+		return NULL;
+
+	if (fpo_candidate_stream->vrr_active_variable)
+		return NULL;
+
+	return fpo_candidate_stream;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index 79664ba7e7af..e5ab7f3077c4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -724,6 +724,9 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_unbounded_requesting = false,
 	.override_dispclk_programming = true,
 	.disable_fpo_optimizations = false,
+	.fpo_vactive_margin_us = 2000, // 2000us
+	.disable_fpo_vactive = true,
+	.disable_boot_optimizations = false,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
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 d0303173ce80..7d0626e42ea6 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
@@ -4864,7 +4864,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->DETBufferSizeCThisState[k],
 							&v->UrgentBurstFactorCursorPre[k],
 							&v->UrgentBurstFactorLumaPre[k],
-							&v->UrgentBurstFactorChroma[k],
+							&v->UrgentBurstFactorChromaPre[k],
 							&v->NoUrgentLatencyHidingPre[k]);
 				}
 
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 536a63624595..bd674dc30df3 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
@@ -5191,7 +5191,7 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->DETBufferSizeCThisState[k],
 							&v->UrgentBurstFactorCursorPre[k],
 							&v->UrgentBurstFactorLumaPre[k],
-							&v->UrgentBurstFactorChroma[k],
+							&v->UrgentBurstFactorChromaPre[k],
 							&v->NotUrgentLatencyHidingPre[k]);
 				}
 
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 c52b76610bd2..44082f65de1f 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
@@ -97,7 +97,7 @@ struct _vcs_dpi_ip_params_st dcn3_14_ip = {
 	.dcc_supported = true,
 };
 
-struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
+static struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 		/*TODO: correct dispclk/dppclk voltage level determination*/
 	.clock_limits = {
 		{
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 daf319370190..7eb2173b7691 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
@@ -5288,7 +5288,7 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_
 							v->DETBufferSizeCThisState[k],
 							&v->UrgentBurstFactorCursorPre[k],
 							&v->UrgentBurstFactorLumaPre[k],
-							&v->UrgentBurstFactorChroma[k],
+							&v->UrgentBurstFactorChromaPre[k],
 							&v->NotUrgentLatencyHidingPre[k]);
 				}
 
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 1e26adf987cc..4548320217fc 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
@@ -1927,6 +1927,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 	unsigned int min_dram_speed_mts_margin;
 	bool need_fclk_lat_as_dummy = false;
 	bool is_subvp_p_drr = false;
+	struct dc_stream_state *fpo_candidate_stream = NULL;
 
 	dc_assert_fp_enabled();
 
@@ -1968,8 +1969,11 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 	if (!pstate_en || (!dc->debug.disable_fpo_optimizations &&
 			pstate_en && vlevel != 0)) {
 		/* only when the mclk switch can not be natural, is the fw based vblank stretch attempted */
-		context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching =
-			dcn30_can_support_mclk_switch_using_fw_based_vblank_stretch(dc, context);
+		fpo_candidate_stream = dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(dc, context);
+		if (fpo_candidate_stream) {
+			fpo_candidate_stream->fpo_in_use = true;
+			context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = true;
+		}
 
 		if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
 			dummy_latency_index = dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(dc,
@@ -2002,6 +2006,10 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 				 * voltage level)
 				 */
 				context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false;
+				for (i = 0; i < context->stream_count; i++) {
+					if (context->streams[i])
+						context->streams[i]->fpo_in_use = false;
+				}
 				context->bw_ctx.dml.soc.fclk_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us;
 				dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false);
 			}
@@ -2157,7 +2165,13 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 		 * DCFCLK: Min, as reported by PM FW, when available
 		 * UCLK: Min, as reported by PM FW, when available
 		 */
-		dc->res_pool->funcs->update_soc_for_wm_a(dc, context);
+
+		/* For set A set the correct latency values (i.e. non-dummy values) unconditionally
+		 */
+		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
+		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us;
+		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us;
+
 		context->bw_ctx.bw.dcn.watermarks.a.urgent_ns = get_wm_urgent(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 		context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
@@ -2792,3 +2806,68 @@ double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *conte
 	}
 	return max_vratio_pre;
 }
+
+/**
+ * dcn32_assign_fpo_vactive_candidate - Assign the FPO stream candidate for FPO + VActive case
+ *
+ * This function chooses the FPO candidate stream for FPO + VActive cases (2 stream config).
+ * For FPO + VAtive cases, the assumption is that one display has ActiveMargin > 0, and the
+ * other display has ActiveMargin <= 0. This function will choose the pipe/stream that has
+ * ActiveMargin <= 0 to be the FPO stream candidate if found.
+ *
+ *
+ * @param [in]: dc - current dc state
+ * @param [in]: context - new dc state
+ * @param [out]: fpo_candidate_stream - pointer to FPO stream candidate if one is found
+ *
+ * Return: void
+ */
+void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *context, struct dc_stream_state **fpo_candidate_stream)
+{
+	unsigned int i, pipe_idx;
+	const struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+
+	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
+		const struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->stream)
+			continue;
+
+		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0) {
+			*fpo_candidate_stream = pipe->stream;
+			break;
+		}
+		pipe_idx++;
+	}
+}
+
+/**
+ * dcn32_find_vactive_pipe - Determines if the config has a pipe that can switch in VACTIVE
+ *
+ * @param [in]: dc - current dc state
+ * @param [in]: context - new dc state
+ * @param [in]: vactive_margin_req_us - The vactive marign required for a vactive pipe to be
+ *                                      considered "found"
+ *
+ * Return: True if VACTIVE display is found, false otherwise
+ */
+bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req_us)
+{
+	unsigned int i, pipe_idx;
+	const struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+	bool vactive_found = false;
+
+	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
+		const struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->stream)
+			continue;
+
+		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us) {
+			vactive_found = true;
+			break;
+		}
+		pipe_idx++;
+	}
+	return vactive_found;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
index ab010e7e840b..9a0806a0e2ef 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
@@ -76,4 +76,8 @@ void dcn32_patch_dpm_table(struct clk_bw_params *bw_params);
 void dcn32_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
 				  int pipe_cnt);
 
+void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *context, struct dc_stream_state **fpo_candidate_stream);
+
+bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
index f74730c2abbd..13c7e7394b1c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
@@ -3354,7 +3354,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							/* Output */
 							&mode_lib->vba.UrgentBurstFactorCursorPre[k],
 							&mode_lib->vba.UrgentBurstFactorLumaPre[k],
-							&mode_lib->vba.UrgentBurstFactorChroma[k],
+							&mode_lib->vba.UrgentBurstFactorChromaPre[k],
 							&mode_lib->vba.NotUrgentLatencyHidingPre[k]);
 				}
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
index 4125d3d111d1..bdf3ac6cadd5 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
@@ -41,51 +41,51 @@
 #include "dcn32/display_rq_dlg_calc_32.h"
 #include "dml_logger.h"
 
-const struct dml_funcs dml20_funcs = {
+static const struct dml_funcs dml20_funcs = {
 	.validate = dml20_ModeSupportAndSystemConfigurationFull,
 	.recalculate = dml20_recalculate,
 	.rq_dlg_get_dlg_reg = dml20_rq_dlg_get_dlg_reg,
 	.rq_dlg_get_rq_reg = dml20_rq_dlg_get_rq_reg
 };
 
-const struct dml_funcs dml20v2_funcs = {
+static const struct dml_funcs dml20v2_funcs = {
 	.validate = dml20v2_ModeSupportAndSystemConfigurationFull,
 	.recalculate = dml20v2_recalculate,
 	.rq_dlg_get_dlg_reg = dml20v2_rq_dlg_get_dlg_reg,
 	.rq_dlg_get_rq_reg = dml20v2_rq_dlg_get_rq_reg
 };
 
-const struct dml_funcs dml21_funcs = {
-        .validate = dml21_ModeSupportAndSystemConfigurationFull,
-        .recalculate = dml21_recalculate,
-        .rq_dlg_get_dlg_reg = dml21_rq_dlg_get_dlg_reg,
-        .rq_dlg_get_rq_reg = dml21_rq_dlg_get_rq_reg
+static const struct dml_funcs dml21_funcs = {
+	.validate = dml21_ModeSupportAndSystemConfigurationFull,
+	.recalculate = dml21_recalculate,
+	.rq_dlg_get_dlg_reg = dml21_rq_dlg_get_dlg_reg,
+	.rq_dlg_get_rq_reg = dml21_rq_dlg_get_rq_reg
 };
 
-const struct dml_funcs dml30_funcs = {
+static const struct dml_funcs dml30_funcs = {
 	.validate = dml30_ModeSupportAndSystemConfigurationFull,
 	.recalculate = dml30_recalculate,
 	.rq_dlg_get_dlg_reg = dml30_rq_dlg_get_dlg_reg,
 	.rq_dlg_get_rq_reg = dml30_rq_dlg_get_rq_reg
 };
 
-const struct dml_funcs dml31_funcs = {
+static const struct dml_funcs dml31_funcs = {
 	.validate = dml31_ModeSupportAndSystemConfigurationFull,
 	.recalculate = dml31_recalculate,
 	.rq_dlg_get_dlg_reg = dml31_rq_dlg_get_dlg_reg,
 	.rq_dlg_get_rq_reg = dml31_rq_dlg_get_rq_reg
 };
 
-const struct dml_funcs dml314_funcs = {
+static const struct dml_funcs dml314_funcs = {
 	.validate = dml314_ModeSupportAndSystemConfigurationFull,
 	.recalculate = dml314_recalculate,
 	.rq_dlg_get_dlg_reg = dml314_rq_dlg_get_dlg_reg,
 	.rq_dlg_get_rq_reg = dml314_rq_dlg_get_rq_reg
 };
 
-const struct dml_funcs dml32_funcs = {
+static const struct dml_funcs dml32_funcs = {
 	.validate = dml32_ModeSupportAndSystemConfigurationFull,
-    .recalculate = dml32_recalculate,
+	.recalculate = dml32_recalculate,
 	.rq_dlg_get_dlg_reg_v2 = dml32_rq_dlg_get_dlg_reg,
 	.rq_dlg_get_rq_reg_v2 = dml32_rq_dlg_get_rq_reg
 };
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
index 591ab1389e3b..bef843cc32a1 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
@@ -293,6 +293,9 @@ struct clk_mgr_funcs {
 
 	/* Get SMU present */
 	bool (*is_smu_present)(struct clk_mgr *clk_mgr);
+
+	int (*get_dispclk_from_dentist)(struct clk_mgr *clk_mgr_base);
+
 };
 
 struct clk_mgr {
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index f6c5ee2d639b..027ad1f0144d 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -2035,6 +2035,12 @@ static enum dc_status enable_link_dp(struct dc_state *state,
 	uint32_t post_oui_delay = 30; // 30ms
 	/* Reduce link bandwidth between failed link training attempts. */
 	bool do_fallback = false;
+	int lt_attempts = LINK_TRAINING_ATTEMPTS;
+
+	// Increase retry count if attempting DP1.x on FIXED_VS link
+	if ((link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) &&
+			link_dp_get_encoding_format(link_settings) == DP_8b_10b_ENCODING)
+		lt_attempts = 10;
 
 	// check for seamless boot
 	for (i = 0; i < state->stream_count; i++) {
@@ -2099,7 +2105,7 @@ static enum dc_status enable_link_dp(struct dc_state *state,
 
 	if (perform_link_training_with_retries(link_settings,
 					       skip_video_pattern,
-					       LINK_TRAINING_ATTEMPTS,
+					       lt_attempts,
 					       pipe_ctx,
 					       pipe_ctx->stream->signal,
 					       do_fallback)) {
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index c840ef17802e..ba98013fecd0 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1043,6 +1043,9 @@ static enum dc_status wake_up_aux_channel(struct dc_link *link)
 				DP_SET_POWER,
 				&dpcd_power_state,
 				sizeof(dpcd_power_state));
+		if (status < 0)
+			DC_LOG_DC("%s: Failed to power up sink: %s\n", __func__,
+				  dpcd_power_state == DP_SET_POWER_D0 ? "D0" : "D3");
 		return DC_ERROR_UNEXPECTED;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
index f301c9eaf2f9..170f33835930 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
@@ -212,27 +212,36 @@ enum dpcd_training_patterns
 
 	switch (pattern) {
 	case DP_TRAINING_PATTERN_SEQUENCE_1:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP training pattern TPS1\n", __func__);
 		dpcd_tr_pattern = DPCD_TRAINING_PATTERN_1;
 		break;
 	case DP_TRAINING_PATTERN_SEQUENCE_2:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP training pattern TPS2\n", __func__);
 		dpcd_tr_pattern = DPCD_TRAINING_PATTERN_2;
 		break;
 	case DP_TRAINING_PATTERN_SEQUENCE_3:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP training pattern TPS3\n", __func__);
 		dpcd_tr_pattern = DPCD_TRAINING_PATTERN_3;
 		break;
 	case DP_TRAINING_PATTERN_SEQUENCE_4:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP training pattern TPS4\n", __func__);
 		dpcd_tr_pattern = DPCD_TRAINING_PATTERN_4;
 		break;
 	case DP_128b_132b_TPS1:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP 128b/132b training pattern TPS1\n", __func__);
 		dpcd_tr_pattern = DPCD_128b_132b_TPS1;
 		break;
 	case DP_128b_132b_TPS2:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP 128b/132b training pattern TPS2\n", __func__);
 		dpcd_tr_pattern = DPCD_128b_132b_TPS2;
 		break;
 	case DP_128b_132b_TPS2_CDS:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP 128b/132b training pattern TPS2 CDS\n",
+					__func__);
 		dpcd_tr_pattern = DPCD_128b_132b_TPS2_CDS;
 		break;
 	case DP_TRAINING_PATTERN_VIDEOIDLE:
+		DC_LOG_HW_LINK_TRAINING("%s: Using DP training pattern videoidle\n", __func__);
 		dpcd_tr_pattern = DPCD_TRAINING_PATTERN_VIDEOIDLE;
 		break;
 	default:
@@ -1496,7 +1505,10 @@ enum link_training_result dp_perform_link_training(
 	 * Non-LT AUX transactions inside training mode.
 	 */
 	if ((link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && encoding == DP_8b_10b_ENCODING)
-		status = dp_perform_fixed_vs_pe_training_sequence(link, link_res, &lt_settings);
+		if (link->dc->config.use_old_fixed_vs_sequence)
+			status = dp_perform_fixed_vs_pe_training_sequence_legacy(link, link_res, &lt_settings);
+		else
+			status = dp_perform_fixed_vs_pe_training_sequence(link, link_res, &lt_settings);
 	else if (encoding == DP_8b_10b_ENCODING)
 		status = dp_perform_8b_10b_link_training(link, link_res, &lt_settings);
 	else if (encoding == DP_128b_132b_ENCODING)
@@ -1557,9 +1569,10 @@ bool perform_link_training_with_retries(
 	j = 0;
 	while (j < attempts && fail_count < (attempts * 10)) {
 
-		DC_LOG_HW_LINK_TRAINING("%s: Beginning link(%d) training attempt %u of %d @ rate(%d) x lane(%d)\n",
-			__func__, link->link_index, (unsigned int)j + 1, attempts, cur_link_settings.link_rate,
-			cur_link_settings.lane_count);
+		DC_LOG_HW_LINK_TRAINING("%s: Beginning link(%d) training attempt %u of %d @ rate(%d) x lane(%d) @ spread = %x\n",
+					__func__, link->link_index, (unsigned int)j + 1, attempts,
+				       cur_link_settings.link_rate, cur_link_settings.lane_count,
+				       cur_link_settings.link_spread);
 
 		dp_enable_link_phy(
 			link,
@@ -1637,9 +1650,10 @@ bool perform_link_training_with_retries(
 				break;
 		}
 
-		DC_LOG_WARNING("%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) : fail reason:(%d)\n",
-			__func__, link->link_index, (unsigned int)j + 1, attempts, cur_link_settings.link_rate,
-			cur_link_settings.lane_count, status);
+		DC_LOG_WARNING("%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n",
+			       __func__, link->link_index, (unsigned int)j + 1, attempts,
+			      cur_link_settings.link_rate, cur_link_settings.lane_count,
+			      cur_link_settings.link_spread, status);
 
 		dp_disable_link_phy(link, &pipe_ctx->link_res, signal);
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
index 14b98e096d39..3889ebb2256b 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
@@ -225,8 +225,10 @@ enum link_training_result perform_8b_10b_clock_recovery_sequence(
 				offset);
 
 		/* 5. check CR done*/
-		if (dp_is_cr_done(lane_count, dpcd_lane_status))
+		if (dp_is_cr_done(lane_count, dpcd_lane_status)) {
+			DC_LOG_HW_LINK_TRAINING("%s: Clock recovery OK\n", __func__);
 			return LINK_TRAINING_SUCCESS;
+		}
 
 		/* 6. max VS reached*/
 		if ((link_dp_get_encoding_format(&lt_settings->link_settings) ==
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_dpia.c
index ab4aafdb5e5c..4f4e899e5c46 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_dpia.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_dpia.c
@@ -401,6 +401,7 @@ static enum link_training_result dpia_training_cr_non_transparent(
 
 		/* Check if clock recovery successful. */
 		if (dp_is_cr_done(lane_count, dpcd_lane_status)) {
+			DC_LOG_HW_LINK_TRAINING("%s: Clock recovery OK\n", __func__);
 			result = LINK_TRAINING_SUCCESS;
 			break;
 		}
@@ -508,6 +509,7 @@ static enum link_training_result dpia_training_cr_transparent(
 
 		/* Check if clock recovery successful. */
 		if (dp_is_cr_done(lane_count, dpcd_lane_status)) {
+			DC_LOG_HW_LINK_TRAINING("%s: Clock recovery OK\n", __func__);
 			result = LINK_TRAINING_SUCCESS;
 			break;
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
index a4071d2959a0..5731c4b61f9f 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
@@ -223,7 +223,7 @@ static enum link_training_result perform_fixed_vs_pe_nontransparent_training_seq
 }
 
 
-enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
+enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
 	struct dc_link *link,
 	const struct link_resource *link_res,
 	struct link_training_settings *lt_settings)
@@ -577,3 +577,379 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 
 	return status;
 }
+
+enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
+	struct dc_link *link,
+	const struct link_resource *link_res,
+	struct link_training_settings *lt_settings)
+{
+	const uint8_t vendor_lttpr_write_data_reset[4] = {0x1, 0x50, 0x63, 0xFF};
+	const uint8_t offset = dp_parse_lttpr_repeater_count(
+			link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
+	const uint8_t vendor_lttpr_write_data_intercept_en[4] = {0x1, 0x55, 0x63, 0x0};
+	const uint8_t vendor_lttpr_write_data_intercept_dis[4] = {0x1, 0x55, 0x63, 0x6E};
+	const uint8_t vendor_lttpr_write_data_adicora_eq1[4] = {0x1, 0x55, 0x63, 0x2E};
+	const uint8_t vendor_lttpr_write_data_adicora_eq2[4] = {0x1, 0x55, 0x63, 0x01};
+	const uint8_t vendor_lttpr_write_data_adicora_eq3[4] = {0x1, 0x55, 0x63, 0x68};
+	uint32_t pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa;
+	uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0};
+	uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0};
+
+	uint32_t vendor_lttpr_write_address = 0xF004F;
+	enum link_training_result status = LINK_TRAINING_SUCCESS;
+	uint8_t lane = 0;
+	union down_spread_ctrl downspread = {0};
+	union lane_count_set lane_count_set = {0};
+	uint8_t toggle_rate;
+	uint8_t rate;
+
+	/* Only 8b/10b is supported */
+	ASSERT(link_dp_get_encoding_format(&lt_settings->link_settings) ==
+			DP_8b_10b_ENCODING);
+
+	if (lt_settings->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) {
+		status = perform_fixed_vs_pe_nontransparent_training_sequence(link, link_res, lt_settings);
+		return status;
+	}
+
+	if (offset != 0xFF) {
+		vendor_lttpr_write_address +=
+				((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1));
+
+		/* Certain display and cable configuration require extra delay */
+		if (offset > 2)
+			pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2;
+	}
+
+	/* Vendor specific: Reset lane settings */
+	core_link_write_dpcd(
+			link,
+			vendor_lttpr_write_address,
+			&vendor_lttpr_write_data_reset[0],
+			sizeof(vendor_lttpr_write_data_reset));
+	core_link_write_dpcd(
+			link,
+			vendor_lttpr_write_address,
+			&vendor_lttpr_write_data_vs[0],
+			sizeof(vendor_lttpr_write_data_vs));
+	core_link_write_dpcd(
+			link,
+			vendor_lttpr_write_address,
+			&vendor_lttpr_write_data_pe[0],
+			sizeof(vendor_lttpr_write_data_pe));
+
+	/* Vendor specific: Enable intercept */
+	core_link_write_dpcd(
+			link,
+			vendor_lttpr_write_address,
+			&vendor_lttpr_write_data_intercept_en[0],
+			sizeof(vendor_lttpr_write_data_intercept_en));
+
+	/* 1. set link rate, lane count and spread. */
+
+	downspread.raw = (uint8_t)(lt_settings->link_settings.link_spread);
+
+	lane_count_set.bits.LANE_COUNT_SET =
+	lt_settings->link_settings.lane_count;
+
+	lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing;
+	lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0;
+
+
+	if (lt_settings->pattern_for_eq < DP_TRAINING_PATTERN_SEQUENCE_4) {
+		lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED =
+				link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED;
+	}
+
+	core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL,
+		&downspread.raw, sizeof(downspread));
+
+	core_link_write_dpcd(link, DP_LANE_COUNT_SET,
+		&lane_count_set.raw, 1);
+
+	rate = get_dpcd_link_rate(&lt_settings->link_settings);
+
+	/* Vendor specific: Toggle link rate */
+	toggle_rate = (rate == 0x6) ? 0xA : 0x6;
+
+	if (link->vendor_specific_lttpr_link_rate_wa == rate) {
+		core_link_write_dpcd(
+				link,
+				DP_LINK_BW_SET,
+				&toggle_rate,
+				1);
+	}
+
+	link->vendor_specific_lttpr_link_rate_wa = rate;
+
+	core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1);
+
+	DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x framing = %x\n %x spread = %x\n",
+		__func__,
+		DP_LINK_BW_SET,
+		lt_settings->link_settings.link_rate,
+		DP_LANE_COUNT_SET,
+		lt_settings->link_settings.lane_count,
+		lt_settings->enhanced_framing,
+		DP_DOWNSPREAD_CTRL,
+		lt_settings->link_settings.link_spread);
+
+	/* 2. Perform link training */
+
+	/* Perform Clock Recovery Sequence */
+	if (status == LINK_TRAINING_SUCCESS) {
+		const uint8_t max_vendor_dpcd_retries = 10;
+		uint32_t retries_cr;
+		uint32_t retry_count;
+		uint32_t wait_time_microsec;
+		enum dc_lane_count lane_count = lt_settings->link_settings.lane_count;
+		union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX];
+		union lane_align_status_updated dpcd_lane_status_updated;
+		union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0};
+		enum dc_status dpcd_status = DC_OK;
+		uint8_t i = 0;
+
+		retries_cr = 0;
+		retry_count = 0;
+
+		memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status));
+		memset(&dpcd_lane_status_updated, '\0',
+		sizeof(dpcd_lane_status_updated));
+
+		while ((retries_cr < LINK_TRAINING_MAX_RETRY_COUNT) &&
+			(retry_count < LINK_TRAINING_MAX_CR_RETRY)) {
+
+
+			/* 1. call HWSS to set lane settings */
+			dp_set_hw_lane_settings(
+					link,
+					link_res,
+					lt_settings,
+					0);
+
+			/* 2. update DPCD of the receiver */
+			if (!retry_count) {
+				/* EPR #361076 - write as a 5-byte burst,
+				 * but only for the 1-st iteration.
+				 */
+				dpcd_set_lt_pattern_and_lane_settings(
+						link,
+						lt_settings,
+						lt_settings->pattern_for_cr,
+						0);
+				/* Vendor specific: Disable intercept */
+				for (i = 0; i < max_vendor_dpcd_retries; i++) {
+					msleep(pre_disable_intercept_delay_ms);
+					dpcd_status = core_link_write_dpcd(
+							link,
+							vendor_lttpr_write_address,
+							&vendor_lttpr_write_data_intercept_dis[0],
+							sizeof(vendor_lttpr_write_data_intercept_dis));
+
+					if (dpcd_status == DC_OK)
+						break;
+
+					core_link_write_dpcd(
+							link,
+							vendor_lttpr_write_address,
+							&vendor_lttpr_write_data_intercept_en[0],
+							sizeof(vendor_lttpr_write_data_intercept_en));
+				}
+			} else {
+				vendor_lttpr_write_data_vs[3] = 0;
+				vendor_lttpr_write_data_pe[3] = 0;
+
+				for (lane = 0; lane < lane_count; lane++) {
+					vendor_lttpr_write_data_vs[3] |=
+							lt_settings->dpcd_lane_settings[lane].bits.VOLTAGE_SWING_SET << (2 * lane);
+					vendor_lttpr_write_data_pe[3] |=
+							lt_settings->dpcd_lane_settings[lane].bits.PRE_EMPHASIS_SET << (2 * lane);
+				}
+
+				/* Vendor specific: Update VS and PE to DPRX requested value */
+				core_link_write_dpcd(
+						link,
+						vendor_lttpr_write_address,
+						&vendor_lttpr_write_data_vs[0],
+						sizeof(vendor_lttpr_write_data_vs));
+				core_link_write_dpcd(
+						link,
+						vendor_lttpr_write_address,
+						&vendor_lttpr_write_data_pe[0],
+						sizeof(vendor_lttpr_write_data_pe));
+
+				dpcd_set_lane_settings(
+						link,
+						lt_settings,
+						0);
+			}
+
+			/* 3. wait receiver to lock-on*/
+			wait_time_microsec = lt_settings->cr_pattern_time;
+
+			dp_wait_for_training_aux_rd_interval(
+					link,
+					wait_time_microsec);
+
+			/* 4. Read lane status and requested drive
+			 * settings as set by the sink
+			 */
+			dp_get_lane_status_and_lane_adjust(
+					link,
+					lt_settings,
+					dpcd_lane_status,
+					&dpcd_lane_status_updated,
+					dpcd_lane_adjust,
+					0);
+
+			/* 5. check CR done*/
+			if (dp_is_cr_done(lane_count, dpcd_lane_status)) {
+				status = LINK_TRAINING_SUCCESS;
+				break;
+			}
+
+			/* 6. max VS reached*/
+			if (dp_is_max_vs_reached(lt_settings))
+				break;
+
+			/* 7. same lane settings */
+			/* Note: settings are the same for all lanes,
+			 * so comparing first lane is sufficient
+			 */
+			if (lt_settings->dpcd_lane_settings[0].bits.VOLTAGE_SWING_SET ==
+					dpcd_lane_adjust[0].bits.VOLTAGE_SWING_LANE)
+				retries_cr++;
+			else
+				retries_cr = 0;
+
+			/* 8. update VS/PE/PC2 in lt_settings*/
+			dp_decide_lane_settings(lt_settings, dpcd_lane_adjust,
+					lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings);
+			retry_count++;
+		}
+
+		if (retry_count >= LINK_TRAINING_MAX_CR_RETRY) {
+			ASSERT(0);
+			DC_LOG_ERROR("%s: Link Training Error, could not get CR after %d tries. Possibly voltage swing issue",
+				__func__,
+				LINK_TRAINING_MAX_CR_RETRY);
+
+		}
+
+		status = dp_get_cr_failure(lane_count, dpcd_lane_status);
+	}
+
+	/* Perform Channel EQ Sequence */
+	if (status == LINK_TRAINING_SUCCESS) {
+		enum dc_dp_training_pattern tr_pattern;
+		uint32_t retries_ch_eq;
+		uint32_t wait_time_microsec;
+		enum dc_lane_count lane_count = lt_settings->link_settings.lane_count;
+		union lane_align_status_updated dpcd_lane_status_updated = {0};
+		union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = {0};
+		union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0};
+
+		core_link_write_dpcd(
+				link,
+				vendor_lttpr_write_address,
+				&vendor_lttpr_write_data_adicora_eq1[0],
+				sizeof(vendor_lttpr_write_data_adicora_eq1));
+		core_link_write_dpcd(
+				link,
+				vendor_lttpr_write_address,
+				&vendor_lttpr_write_data_adicora_eq2[0],
+				sizeof(vendor_lttpr_write_data_adicora_eq2));
+
+		/* Note: also check that TPS4 is a supported feature*/
+		tr_pattern = lt_settings->pattern_for_eq;
+
+		dp_set_hw_training_pattern(link, link_res, tr_pattern, 0);
+
+		status = LINK_TRAINING_EQ_FAIL_EQ;
+
+		for (retries_ch_eq = 0; retries_ch_eq <= LINK_TRAINING_MAX_RETRY_COUNT;
+			retries_ch_eq++) {
+
+			dp_set_hw_lane_settings(link, link_res, lt_settings, 0);
+
+			vendor_lttpr_write_data_vs[3] = 0;
+			vendor_lttpr_write_data_pe[3] = 0;
+
+			for (lane = 0; lane < lane_count; lane++) {
+				vendor_lttpr_write_data_vs[3] |=
+						lt_settings->dpcd_lane_settings[lane].bits.VOLTAGE_SWING_SET << (2 * lane);
+				vendor_lttpr_write_data_pe[3] |=
+						lt_settings->dpcd_lane_settings[lane].bits.PRE_EMPHASIS_SET << (2 * lane);
+			}
+
+			/* Vendor specific: Update VS and PE to DPRX requested value */
+			core_link_write_dpcd(
+					link,
+					vendor_lttpr_write_address,
+					&vendor_lttpr_write_data_vs[0],
+					sizeof(vendor_lttpr_write_data_vs));
+			core_link_write_dpcd(
+					link,
+					vendor_lttpr_write_address,
+					&vendor_lttpr_write_data_pe[0],
+					sizeof(vendor_lttpr_write_data_pe));
+
+			/* 2. update DPCD*/
+			if (!retries_ch_eq) {
+				/* EPR #361076 - write as a 5-byte burst,
+				 * but only for the 1-st iteration
+				 */
+
+				dpcd_set_lt_pattern_and_lane_settings(
+					link,
+					lt_settings,
+					tr_pattern, 0);
+
+				core_link_write_dpcd(
+					link,
+					vendor_lttpr_write_address,
+					&vendor_lttpr_write_data_adicora_eq3[0],
+					sizeof(vendor_lttpr_write_data_adicora_eq3));
+			} else
+				dpcd_set_lane_settings(link, lt_settings, 0);
+
+			/* 3. wait for receiver to lock-on*/
+			wait_time_microsec = lt_settings->eq_pattern_time;
+
+			dp_wait_for_training_aux_rd_interval(
+					link,
+					wait_time_microsec);
+
+			/* 4. Read lane status and requested
+			 * drive settings as set by the sink
+			 */
+			dp_get_lane_status_and_lane_adjust(
+				link,
+				lt_settings,
+				dpcd_lane_status,
+				&dpcd_lane_status_updated,
+				dpcd_lane_adjust,
+				0);
+
+			/* 5. check CR done*/
+			if (!dp_is_cr_done(lane_count, dpcd_lane_status)) {
+				status = LINK_TRAINING_EQ_FAIL_CR;
+				break;
+			}
+
+			/* 6. check CHEQ done*/
+			if (dp_is_ch_eq_done(lane_count, dpcd_lane_status) &&
+					dp_is_symbol_locked(lane_count, dpcd_lane_status) &&
+					dp_is_interlane_aligned(dpcd_lane_status_updated)) {
+				status = LINK_TRAINING_SUCCESS;
+				break;
+			}
+
+			/* 7. update VS/PE/PC2 in lt_settings*/
+			dp_decide_lane_settings(lt_settings, dpcd_lane_adjust,
+					lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings);
+		}
+	}
+
+	return status;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h
index e61970e27661..c0d6ea329504 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h
@@ -28,6 +28,11 @@
 #define __DC_LINK_DP_FIXED_VS_PE_RETIMER_H__
 #include "link_dp_training.h"
 
+enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
+	struct dc_link *link,
+	const struct link_resource *link_res,
+	struct link_training_settings *lt_settings);
+
 enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 	struct dc_link *link,
 	const struct link_resource *link_res,
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 15d26222597a..598fa1de54ce 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -362,7 +362,7 @@ union dmub_fw_boot_status {
 		uint32_t defer_load : 1; /**< 1 if VBIOS data is deferred programmed */
 		uint32_t reserved : 1;
 		uint32_t detection_required: 1; /**<  if detection need to be triggered by driver */
-
+		uint32_t hw_power_init_done: 1; /**< 1 if hw power init is completed */
 	} bits; /**< status bits */
 	uint32_t all; /**< 32-bit access to status bits */
 };
@@ -377,6 +377,7 @@ enum dmub_fw_boot_status_bit {
 	DMUB_FW_BOOT_STATUS_BIT_RESTORE_REQUIRED = (1 << 3), /**< 1 if driver should call restore */
 	DMUB_FW_BOOT_STATUS_BIT_DEFERRED_LOADED = (1 << 4), /**< 1 if VBIOS data is deferred programmed */
 	DMUB_FW_BOOT_STATUS_BIT_DETECTION_REQUIRED = (1 << 6), /**< 1 if detection need to be triggered by driver*/
+	DMUB_FW_BOOT_STATUS_BIT_HW_POWER_INIT_DONE = (1 << 7), /**< 1 if hw power init is completed */
 };
 
 /* Register bit definition for SCRATCH5 */
@@ -1104,7 +1105,12 @@ enum dmub_cmd_idle_opt_type {
 	/**
 	 * DCN hardware save.
 	 */
-	DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT = 1
+	DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT = 1,
+
+	/**
+	 * DCN hardware notify idle.
+	 */
+	DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE = 2
 };
 
 /**
@@ -1115,6 +1121,24 @@ struct dmub_rb_cmd_idle_opt_dcn_restore {
 };
 
 /**
+ * struct dmub_dcn_notify_idle_cntl_data - Data passed to FW in a DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE command.
+ */
+struct dmub_dcn_notify_idle_cntl_data {
+	uint8_t driver_idle;
+	uint8_t d3_entry;
+	uint8_t trigger;
+	uint8_t pad[1];
+};
+
+/**
+ * struct dmub_rb_cmd_idle_opt_dcn_notify_idle - Data passed to FW in a DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE command.
+ */
+struct dmub_rb_cmd_idle_opt_dcn_notify_idle {
+	struct dmub_cmd_header header; /**< header */
+	struct dmub_dcn_notify_idle_cntl_data cntl_data;
+};
+
+/**
  * struct dmub_clocks - Clock update notification.
  */
 struct dmub_clocks {
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 315da61ee897..5c41a4751db4 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -1129,7 +1129,6 @@ void mod_freesync_handle_preflip(struct mod_freesync *mod_freesync,
 {
 	struct core_freesync *core_freesync = NULL;
 	unsigned int last_render_time_in_us = 0;
-	unsigned int average_render_time_in_us = 0;
 
 	if (mod_freesync == NULL)
 		return;
@@ -1138,7 +1137,6 @@ void mod_freesync_handle_preflip(struct mod_freesync *mod_freesync,
 
 	if (in_out_vrr->supported &&
 			in_out_vrr->state == VRR_STATE_ACTIVE_VARIABLE) {
-		unsigned int i = 0;
 		unsigned int oldest_index = plane->time.index + 1;
 
 		if (oldest_index >= DC_PLANE_UPDATE_TIMES_MAX)
@@ -1147,18 +1145,6 @@ void mod_freesync_handle_preflip(struct mod_freesync *mod_freesync,
 		last_render_time_in_us = curr_time_stamp_in_us -
 				plane->time.prev_update_time_in_us;
 
-		/* Sum off all entries except oldest one */
-		for (i = 0; i < DC_PLANE_UPDATE_TIMES_MAX; i++) {
-			average_render_time_in_us +=
-					plane->time.time_elapsed_in_us[i];
-		}
-		average_render_time_in_us -=
-				plane->time.time_elapsed_in_us[oldest_index];
-
-		/* Add render time for current flip */
-		average_render_time_in_us += last_render_time_in_us;
-		average_render_time_in_us /= DC_PLANE_UPDATE_TIMES_MAX;
-
 		if (in_out_vrr->btr.btr_enabled) {
 			apply_below_the_range(core_freesync,
 					stream,