Merge tag 'amd-drm-fixes-6.0-2022-09-29' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.0-2022-09-29:

amdgpu:
- GC 11.x fixes
- SMU 13.x fixes
- DCN 3.1.4 fixes
- DCN 3.2.x fixes
- GC 9.x fix
- Fence fix
- SR-IOV supend/resume fix
- PSR regression fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220929144003.8363-1-alexander.deucher@amd.com

23 files changed, 283 insertions, 86 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 130060834b4e95..48bd660ddb858e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1050,6 +1050,10 @@ bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev)
 {
 	if (adev->flags & AMD_IS_APU)
 		return false;
+
+	if (amdgpu_sriov_vf(adev))
+		return false;
+
 	return pm_suspend_target_state != PM_SUSPEND_TO_IDLE;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index be7aff2d4a57ef..25e1f5ed7ead14 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3152,7 +3152,8 @@ static int amdgpu_device_ip_resume_phase1(struct amdgpu_device *adev)
 			continue;
 		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON ||
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC ||
-		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) {
+		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH ||
+		    (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP && amdgpu_sriov_vf(adev))) {
 
 			r = adev->ip_blocks[i].version->funcs->resume(adev);
 			if (r) {
@@ -4064,12 +4065,20 @@ static void amdgpu_device_evict_resources(struct amdgpu_device *adev)
 int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 {
 	struct amdgpu_device *adev = drm_to_adev(dev);
+	int r = 0;
 
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
 	adev->in_suspend = true;
 
+	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_virt_fini_data_exchange(adev);
+		r = amdgpu_virt_request_full_gpu(adev, false);
+		if (r)
+			return r;
+	}
+
 	if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3))
 		DRM_WARN("smart shift update failed\n");
 
@@ -4093,6 +4102,9 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 
 	amdgpu_device_ip_suspend_phase2(adev);
 
+	if (amdgpu_sriov_vf(adev))
+		amdgpu_virt_release_full_gpu(adev, false);
+
 	return 0;
 }
 
@@ -4111,6 +4123,12 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	int r = 0;
 
+	if (amdgpu_sriov_vf(adev)) {
+		r = amdgpu_virt_request_full_gpu(adev, true);
+		if (r)
+			return r;
+	}
+
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
@@ -4125,6 +4143,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	}
 
 	r = amdgpu_device_ip_resume(adev);
+
+	/* no matter what r is, always need to properly release full GPU */
+	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_virt_init_data_exchange(adev);
+		amdgpu_virt_release_full_gpu(adev, true);
+	}
+
 	if (r) {
 		dev_err(adev->dev, "amdgpu_device_ip_resume failed (%d).\n", r);
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 8adeb7469f1e57..d0d99ed607ddd4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -400,7 +400,6 @@ unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring)
 	/* We are not protected by ring lock when reading the last sequence
 	 * but it's ok to report slightly wrong fence count here.
 	 */
-	amdgpu_fence_process(ring);
 	emitted = 0x100000000ull;
 	emitted -= atomic_read(&ring->fence_drv.last_seq);
 	emitted += READ_ONCE(ring->fence_drv.sync_seq);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index 7b46f6bf41875f..ad980f4b66e191 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -222,6 +222,8 @@ struct mes_add_queue_input {
 	uint64_t	tba_addr;
 	uint64_t	tma_addr;
 	uint32_t	is_kfd_process;
+	uint32_t	is_aql_queue;
+	uint32_t	queue_size;
 };
 
 struct mes_remove_queue_input {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index f6b1bb40e5036e..fbb13c9147f8c1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -5260,6 +5260,8 @@ static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid)
 {
 	u32 reg, data;
 
+	amdgpu_gfx_off_ctrl(adev, false);
+
 	reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL);
 	if (amdgpu_sriov_is_pp_one_vf(adev))
 		data = RREG32_NO_KIQ(reg);
@@ -5273,6 +5275,8 @@ static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid)
 		WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
 	else
 		WREG32_SOC15(GC, 0, regRLC_SPM_MC_CNTL, data);
+
+	amdgpu_gfx_off_ctrl(adev, true);
 }
 
 static const struct amdgpu_rlc_funcs gfx_v11_0_rlc_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index fc9c1043244cb3..037af835267733 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -5597,7 +5597,7 @@ static void gfx_v9_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigne
 	BUG_ON(offset > ring->buf_mask);
 	BUG_ON(ring->ring[offset] != 0x55aa55aa);
 
-	cur = (ring->wptr & ring->buf_mask) - 1;
+	cur = (ring->wptr - 1) & ring->buf_mask;
 	if (likely(cur > offset))
 		ring->ring[offset] = cur - offset;
 	else
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index cc3fdbbcd3140a..f92744b8d79dfc 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -185,6 +185,10 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes,
 	mes_add_queue_pkt.is_kfd_process = input->is_kfd_process;
 	mes_add_queue_pkt.trap_en = 1;
 
+	/* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */
+	mes_add_queue_pkt.is_aql_queue = input->is_aql_queue;
+	mes_add_queue_pkt.gds_size = input->queue_size;
+
 	return mes_v11_0_submit_pkt_and_poll_completion(mes,
 			&mes_add_queue_pkt, sizeof(mes_add_queue_pkt),
 			offsetof(union MESAPI__ADD_QUEUE, api_status));
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index e83725a281068b..007a3db69df13a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -205,6 +205,8 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q,
 	}
 
 	queue_input.is_kfd_process = 1;
+	queue_input.is_aql_queue = (q->properties.format == KFD_QUEUE_FORMAT_AQL);
+	queue_input.queue_size = q->properties.queue_size >> 2;
 
 	queue_input.paging = false;
 	queue_input.tba_addr = qpd->tba_addr;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index a6fcbeeb742859..0d53f60674226b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -350,11 +350,11 @@ static void event_interrupt_wq_v11(struct kfd_dev *dev,
 				print_sq_intr_info_inst(context_id0, context_id1);
 				sq_int_priv = REG_GET_FIELD(context_id0,
 						SQ_INTERRUPT_WORD_WAVE_CTXID0, PRIV);
-				if (sq_int_priv /*&& (kfd_set_dbg_ev_from_interrupt(dev, pasid,
+				/*if (sq_int_priv && (kfd_set_dbg_ev_from_interrupt(dev, pasid,
 						KFD_CTXID0_DOORBELL_ID(context_id0),
 						KFD_CTXID0_TRAP_CODE(context_id0),
-						NULL, 0))*/)
-					return;
+						NULL, 0)))
+					return;*/
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
 				print_sq_intr_info_error(context_id0, context_id1);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index b8e14c2cc2956e..3ae350220d42ed 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -126,6 +126,10 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 	m->compute_static_thread_mgmt_se1 = 0xFFFFFFFF;
 	m->compute_static_thread_mgmt_se2 = 0xFFFFFFFF;
 	m->compute_static_thread_mgmt_se3 = 0xFFFFFFFF;
+	m->compute_static_thread_mgmt_se4 = 0xFFFFFFFF;
+	m->compute_static_thread_mgmt_se5 = 0xFFFFFFFF;
+	m->compute_static_thread_mgmt_se6 = 0xFFFFFFFF;
+	m->compute_static_thread_mgmt_se7 = 0xFFFFFFFF;
 
 	m->cp_hqd_persistent_state = CP_HQD_PERSISTENT_STATE__PRELOAD_REQ_MASK |
 			0x55 << CP_HQD_PERSISTENT_STATE__PRELOAD_SIZE__SHIFT;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
index c8da18e45b0e86..8ca10ab3dfc126 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
@@ -170,7 +170,13 @@ bool amdgpu_dm_psr_enable(struct dc_stream_state *stream)
 					   &stream, 1,
 					   &params);
 
-	power_opt |= psr_power_opt_z10_static_screen;
+	/*
+	 * Only enable static-screen optimizations for PSR1. For PSR SU, this
+	 * causes vstartup interrupt issues, used by amdgpu_dm to send vblank
+	 * events.
+	 */
+	if (link->psr_settings.psr_version < DC_PSR_VERSION_SU_1)
+		power_opt |= psr_power_opt_z10_static_screen;
 
 	return dc_link_set_psr_allow_active(link, &psr_enable, false, false, &power_opt);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 8559dcd80af041..4a15aa7a375fe0 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -130,11 +130,20 @@ static void dcn314_disable_otg_wa(struct clk_mgr *clk_mgr_base, struct dc_state
 		if (pipe->top_pipe || pipe->prev_odm_pipe)
 			continue;
 		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) {
+			struct stream_encoder *stream_enc = pipe->stream_res.stream_enc;
+
 			if (disable) {
+				if (stream_enc && stream_enc->funcs->disable_fifo)
+					pipe->stream_res.stream_enc->funcs->disable_fifo(stream_enc);
+
 				pipe->stream_res.tg->funcs->immediate_disable_crtc(pipe->stream_res.tg);
 				reset_sync_context_for_pipe(dc, context, i);
-			} else
+			} else {
 				pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg);
+
+				if (stream_enc && stream_enc->funcs->enable_fifo)
+					pipe->stream_res.stream_enc->funcs->enable_fifo(stream_enc);
+			}
 		}
 	}
 }
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 c6785969eb1aa7..f0f3f66629cc0c 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
@@ -156,12 +156,14 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base)
 {
 	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
 	unsigned int num_levels;
+	unsigned int num_dcfclk_levels, num_dtbclk_levels, num_dispclk_levels;
 
 	memset(&(clk_mgr_base->clks), 0, sizeof(struct dc_clocks));
 	clk_mgr_base->clks.p_state_change_support = true;
 	clk_mgr_base->clks.prev_p_state_change_support = true;
 	clk_mgr_base->clks.fclk_prev_p_state_change_support = true;
 	clk_mgr->smu_present = false;
+	clk_mgr->dpm_present = false;
 
 	if (!clk_mgr_base->bw_params)
 		return;
@@ -179,6 +181,7 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base)
 	dcn32_init_single_clock(clk_mgr, PPCLK_DCFCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dcfclk_mhz,
 			&num_levels);
+	num_dcfclk_levels = num_levels;
 
 	/* SOCCLK */
 	dcn32_init_single_clock(clk_mgr, PPCLK_SOCCLK,
@@ -189,11 +192,16 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base)
 		dcn32_init_single_clock(clk_mgr, PPCLK_DTBCLK,
 				&clk_mgr_base->bw_params->clk_table.entries[0].dtbclk_mhz,
 				&num_levels);
+	num_dtbclk_levels = num_levels;
 
 	/* DISPCLK */
 	dcn32_init_single_clock(clk_mgr, PPCLK_DISPCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dispclk_mhz,
 			&num_levels);
+	num_dispclk_levels = num_levels;
+
+	if (num_dcfclk_levels && num_dtbclk_levels && num_dispclk_levels)
+		clk_mgr->dpm_present = true;
 
 	if (clk_mgr_base->ctx->dc->debug.min_disp_clk_khz) {
 		unsigned int i;
@@ -658,6 +666,12 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 			&num_levels);
 	clk_mgr_base->bw_params->clk_table.num_entries = num_levels ? num_levels : 1;
 
+	if (clk_mgr->dpm_present && !num_levels)
+		clk_mgr->dpm_present = false;
+
+	if (!clk_mgr->dpm_present)
+		dcn32_patch_dpm_table(clk_mgr_base->bw_params);
+
 	DC_FP_START();
 	/* Refresh bounding box */
 	clk_mgr_base->ctx->dc->res_pool->funcs->update_bw_bounding_box(
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index aea49334021c90..38a67051d470f2 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -2164,8 +2164,7 @@ static void dce110_setup_audio_dto(
 			continue;
 		if (pipe_ctx->stream->signal != SIGNAL_TYPE_HDMI_TYPE_A)
 			continue;
-		if (pipe_ctx->stream_res.audio != NULL &&
-			pipe_ctx->stream_res.audio->enabled == false) {
+		if (pipe_ctx->stream_res.audio != NULL) {
 			struct audio_output audio_output;
 
 			build_audio_output(context, pipe_ctx, &audio_output);
@@ -2205,8 +2204,7 @@ static void dce110_setup_audio_dto(
 			if (!dc_is_dp_signal(pipe_ctx->stream->signal))
 				continue;
 
-			if (pipe_ctx->stream_res.audio != NULL &&
-				pipe_ctx->stream_res.audio->enabled == false) {
+			if (pipe_ctx->stream_res.audio != NULL) {
 				struct audio_output audio_output;
 
 				build_audio_output(context, pipe_ctx, &audio_output);
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 232cc15979dda1..fb729674953b22 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
@@ -45,6 +45,48 @@
 #define DC_LOGGER \
 	dccg->ctx->logger
 
+static void dccg314_get_pixel_rate_div(
+		struct dccg *dccg,
+		uint32_t otg_inst,
+		enum pixel_rate_div *k1,
+		enum pixel_rate_div *k2)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+	uint32_t val_k1 = PIXEL_RATE_DIV_NA, val_k2 = PIXEL_RATE_DIV_NA;
+
+	*k1 = PIXEL_RATE_DIV_NA;
+	*k2 = PIXEL_RATE_DIV_NA;
+
+	switch (otg_inst) {
+	case 0:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG0_PIXEL_RATE_DIVK1, &val_k1,
+			OTG0_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	case 1:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG1_PIXEL_RATE_DIVK1, &val_k1,
+			OTG1_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	case 2:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG2_PIXEL_RATE_DIVK1, &val_k1,
+			OTG2_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	case 3:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG3_PIXEL_RATE_DIVK1, &val_k1,
+			OTG3_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	default:
+		BREAK_TO_DEBUGGER();
+		return;
+	}
+
+	*k1 = (enum pixel_rate_div)val_k1;
+	*k2 = (enum pixel_rate_div)val_k2;
+}
+
 static void dccg314_set_pixel_rate_div(
 		struct dccg *dccg,
 		uint32_t otg_inst,
@@ -52,6 +94,11 @@ static void dccg314_set_pixel_rate_div(
 		enum pixel_rate_div k2)
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+	enum pixel_rate_div cur_k1 = PIXEL_RATE_DIV_NA, cur_k2 = PIXEL_RATE_DIV_NA;
+
+	dccg314_get_pixel_rate_div(dccg, otg_inst, &cur_k1, &cur_k2);
+	if (k1 == PIXEL_RATE_DIV_NA || k2 == PIXEL_RATE_DIV_NA || (k1 == cur_k1 && k2 == cur_k2))
+		return;
 
 	switch (otg_inst) {
 	case 0:
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c
index 06d8638db696a8..8c0ab013764e3c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c
@@ -56,7 +56,8 @@ static void enc314_enable_fifo(struct stream_encoder *enc)
 
 	/* TODO: Confirm if we need to wait for DIG_SYMCLK_FE_ON */
 	REG_WAIT(DIG_FE_CNTL, DIG_SYMCLK_FE_ON, 1, 10, 5000);
-	REG_UPDATE_2(DIG_FIFO_CTRL0, DIG_FIFO_RESET, 1, DIG_FIFO_READ_START_LEVEL, 0x7);
+	REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_READ_START_LEVEL, 0x7);
+	REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_RESET, 1);
 	REG_WAIT(DIG_FIFO_CTRL0, DIG_FIFO_RESET_DONE, 1, 10, 5000);
 	REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_RESET, 0);
 	REG_WAIT(DIG_FIFO_CTRL0, DIG_FIFO_RESET_DONE, 0, 10, 5000);
@@ -261,6 +262,16 @@ static bool is_two_pixels_per_containter(const struct dc_crtc_timing *timing)
 	return two_pix;
 }
 
+void enc314_stream_encoder_dp_blank(
+	struct dc_link *link,
+	struct stream_encoder *enc)
+{
+	/* New to DCN314 - disable the FIFO before VID stream disable. */
+	enc314_disable_fifo(enc);
+
+	enc1_stream_encoder_dp_blank(link, enc);
+}
+
 static void enc314_stream_encoder_dp_unblank(
 		struct dc_link *link,
 		struct stream_encoder *enc,
@@ -316,15 +327,11 @@ static void enc314_stream_encoder_dp_unblank(
 	/* switch DP encoder to CRTC data, but reset it the fifo first. It may happen
 	 * that it overflows during mode transition, and sometimes doesn't recover.
 	 */
-	REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_READ_START_LEVEL, 0x7);
 	REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 1);
 	udelay(10);
 
 	REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 0);
 
-	/* DIG Resync FIFO now needs to be explicitly enabled. */
-	enc314_enable_fifo(enc);
-
 	/* wait 100us for DIG/DP logic to prime
 	 * (i.e. a few video lines)
 	 */
@@ -340,6 +347,12 @@ static void enc314_stream_encoder_dp_unblank(
 
 	REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, true);
 
+	/*
+	 * DIG Resync FIFO now needs to be explicitly enabled.
+	 * This should come after DP_VID_STREAM_ENABLE per HW docs.
+	 */
+	enc314_enable_fifo(enc);
+
 	dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_ENABLE_DP_VID_STREAM);
 }
 
@@ -408,7 +421,7 @@ static const struct stream_encoder_funcs dcn314_str_enc_funcs = {
 	.stop_dp_info_packets =
 		enc1_stream_encoder_stop_dp_info_packets,
 	.dp_blank =
-		enc1_stream_encoder_dp_blank,
+		enc314_stream_encoder_dp_blank,
 	.dp_unblank =
 		enc314_stream_encoder_dp_unblank,
 	.audio_mute_control = enc3_audio_mute_control,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
index 0d5e8a44151210..6640d0ac43040a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
@@ -42,6 +42,48 @@
 #define DC_LOGGER \
 	dccg->ctx->logger
 
+static void dccg32_get_pixel_rate_div(
+		struct dccg *dccg,
+		uint32_t otg_inst,
+		enum pixel_rate_div *k1,
+		enum pixel_rate_div *k2)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+	uint32_t val_k1 = PIXEL_RATE_DIV_NA, val_k2 = PIXEL_RATE_DIV_NA;
+
+	*k1 = PIXEL_RATE_DIV_NA;
+	*k2 = PIXEL_RATE_DIV_NA;
+
+	switch (otg_inst) {
+	case 0:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG0_PIXEL_RATE_DIVK1, &val_k1,
+			OTG0_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	case 1:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG1_PIXEL_RATE_DIVK1, &val_k1,
+			OTG1_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	case 2:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG2_PIXEL_RATE_DIVK1, &val_k1,
+			OTG2_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	case 3:
+		REG_GET_2(OTG_PIXEL_RATE_DIV,
+			OTG3_PIXEL_RATE_DIVK1, &val_k1,
+			OTG3_PIXEL_RATE_DIVK2, &val_k2);
+		break;
+	default:
+		BREAK_TO_DEBUGGER();
+		return;
+	}
+
+	*k1 = (enum pixel_rate_div)val_k1;
+	*k2 = (enum pixel_rate_div)val_k2;
+}
+
 static void dccg32_set_pixel_rate_div(
 		struct dccg *dccg,
 		uint32_t otg_inst,
@@ -50,6 +92,17 @@ static void dccg32_set_pixel_rate_div(
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+	enum pixel_rate_div cur_k1 = PIXEL_RATE_DIV_NA, cur_k2 = PIXEL_RATE_DIV_NA;
+
+	// Don't program 0xF into the register field. Not valid since
+	// K1 / K2 field is only 1 / 2 bits wide
+	if (k1 == PIXEL_RATE_DIV_NA || k2 == PIXEL_RATE_DIV_NA)
+		return;
+
+	dccg32_get_pixel_rate_div(dccg, otg_inst, &cur_k1, &cur_k2);
+	if (k1 == cur_k1 && k2 == cur_k2)
+		return;
+
 	switch (otg_inst) {
 	case 0:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
index 99eb239bbc7bba..9aebc1be2f596c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
@@ -98,9 +98,13 @@ static void dcn32_program_det_size(struct hubbub *hubbub, int hubp_inst, unsigne
 	default:
 		break;
 	}
-	/* Should never be hit, if it is we have an erroneous hw config*/
-	ASSERT(hubbub2->det0_size + hubbub2->det1_size + hubbub2->det2_size
-			+ hubbub2->det3_size + hubbub2->compbuf_size_segments <= hubbub2->crb_size_segs);
+	if (hubbub2->det0_size + hubbub2->det1_size + hubbub2->det2_size
+			+ hubbub2->det3_size + hubbub2->compbuf_size_segments > hubbub2->crb_size_segs) {
+		/* This may happen during seamless transition from ODM 2:1 to ODM4:1 */
+		DC_LOG_WARNING("CRB Config Warning: DET size (%d,%d,%d,%d) + Compbuf size (%d) >  CRB segments (%d)\n",
+						hubbub2->det0_size, hubbub2->det1_size, hubbub2->det2_size, hubbub2->det3_size,
+						hubbub2->compbuf_size_segments, hubbub2->crb_size_segs);
+	}
 }
 
 static void dcn32_program_compbuf_size(struct hubbub *hubbub, unsigned int compbuf_size_kb, bool safe_to_increase)
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 f43686997917f2..e573e706430df8 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
@@ -121,8 +121,8 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc = {
 		},
 	},
 	.num_states = 1,
-	.sr_exit_time_us = 20.16,
-	.sr_enter_plus_exit_time_us = 27.13,
+	.sr_exit_time_us = 42.97,
+	.sr_enter_plus_exit_time_us = 49.94,
 	.sr_exit_z8_time_us = 285.0,
 	.sr_enter_plus_exit_z8_time_us = 320,
 	.writeback_latency_us = 12.0,
@@ -1926,6 +1926,45 @@ static void remove_entry_from_table_at_index(struct _vcs_dpi_voltage_scaling_st
 	memset(&table[--(*num_entries)], 0, sizeof(struct _vcs_dpi_voltage_scaling_st));
 }
 
+void dcn32_patch_dpm_table(struct clk_bw_params *bw_params)
+{
+	int i;
+	unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0,
+			max_phyclk_mhz = 0, max_dtbclk_mhz = 0, max_fclk_mhz = 0, max_uclk_mhz = 0;
+
+	for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz)
+			max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
+		if (bw_params->clk_table.entries[i].fclk_mhz > max_fclk_mhz)
+			max_fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz;
+		if (bw_params->clk_table.entries[i].memclk_mhz > max_uclk_mhz)
+			max_uclk_mhz = bw_params->clk_table.entries[i].memclk_mhz;
+		if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
+		if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
+		if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz)
+			max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
+		if (bw_params->clk_table.entries[i].dtbclk_mhz > max_dtbclk_mhz)
+			max_dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
+	}
+
+	/* Scan through clock values we currently have and if they are 0,
+	 *  then populate it with dcn3_2_soc.clock_limits[] value.
+	 *
+	 * Do it for DCFCLK, DISPCLK, DTBCLK and UCLK as any of those being
+	 *  0, will cause it to skip building the clock table.
+	 */
+	if (max_dcfclk_mhz == 0)
+		bw_params->clk_table.entries[0].dcfclk_mhz = dcn3_2_soc.clock_limits[0].dcfclk_mhz;
+	if (max_dispclk_mhz == 0)
+		bw_params->clk_table.entries[0].dispclk_mhz = dcn3_2_soc.clock_limits[0].dispclk_mhz;
+	if (max_dtbclk_mhz == 0)
+		bw_params->clk_table.entries[0].dtbclk_mhz = dcn3_2_soc.clock_limits[0].dtbclk_mhz;
+	if (max_uclk_mhz == 0)
+		bw_params->clk_table.entries[0].memclk_mhz = dcn3_2_soc.clock_limits[0].dram_speed_mts / 16;
+}
+
 static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 		struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries)
 {
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 6ce2210989796c..e1b79e2aab8c6d 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
@@ -77,4 +77,6 @@ int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc,
 							    int pipe_cnt,
 							    int vlevel);
 
+void dcn32_patch_dpm_table(struct clk_bw_params *bw_params);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
index 68c2ed434d2c4f..cff5fd55a0ad73 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
@@ -340,6 +340,8 @@ struct clk_mgr_internal {
 	bool smu_present;
 	void *wm_range_table;
 	long long wm_range_table_addr;
+
+	bool dpm_present;
 };
 
 struct clk_mgr_internal_funcs {
diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
index 50bfa513cb35a6..7e85cdc5bd34ea 100644
--- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h
+++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
@@ -269,7 +269,8 @@ union MESAPI__ADD_QUEUE {
 			uint32_t map_kiq_utility_queue  : 1;
 			uint32_t is_kfd_process		: 1;
 			uint32_t trap_en		: 1;
-			uint32_t reserved		: 21;
+			uint32_t is_aql_queue		: 1;
+			uint32_t reserved		: 20;
 		};
 		struct MES_API_STATUS		api_status;
 		uint64_t                        tma_addr;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index 096327513dd09d..1d454485e0d91f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -239,82 +239,47 @@ smu_v13_0_0_get_allowed_feature_mask(struct smu_context *smu,
 				  uint32_t *feature_mask, uint32_t num)
 {
 	struct amdgpu_device *adev = smu->adev;
+	u32 smu_version;
 
 	if (num > 2)
 		return -EINVAL;
 
-	memset(feature_mask, 0, sizeof(uint32_t) * num);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FW_DATA_READ_BIT);
+	memset(feature_mask, 0xff, sizeof(uint32_t) * num);
 
-	if (adev->pm.pp_feature & PP_SCLK_DPM_MASK) {
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT);
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_IMU_BIT);
+	if (!(adev->pm.pp_feature & PP_SCLK_DPM_MASK)) {
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_GFXCLK_BIT);
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFX_IMU_BIT);
 	}
 
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MM_DPM_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_VCN_BIT);
-
-	if ((adev->pg_flags & AMD_PG_SUPPORT_ATHUB) &&
-	    (adev->pg_flags & AMD_PG_SUPPORT_MMHUB))
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ATHUB_MMHUB_PG_BIT);
+	if (!(adev->pg_flags & AMD_PG_SUPPORT_ATHUB) ||
+	    !(adev->pg_flags & AMD_PG_SUPPORT_MMHUB))
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_ATHUB_MMHUB_PG_BIT);
 
-	if (adev->pm.pp_feature & PP_SOCCLK_DPM_MASK)
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT);
-
-#if 0
-	if (adev->pm.pp_feature & PP_GFXOFF_MASK)
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFXOFF_BIT);
-#endif
+	if (!(adev->pm.pp_feature & PP_SOCCLK_DPM_MASK))
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_SOCCLK_BIT);
 
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_THROTTLERS_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FAN_CONTROL_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DF_CSTATE_BIT);
-
-	if (adev->pm.pp_feature & PP_MCLK_DPM_MASK) {
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_UCLK_BIT);
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VMEMP_SCALING_BIT);
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VDDIO_MEM_SCALING_BIT);
+	/* PMFW 78.58 contains a critical fix for gfxoff feature */
+	smu_cmn_get_smc_version(smu, NULL, &smu_version);
+	if ((smu_version < 0x004e3a00) ||
+	     !(adev->pm.pp_feature & PP_GFXOFF_MASK))
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFXOFF_BIT);
+
+	if (!(adev->pm.pp_feature & PP_MCLK_DPM_MASK)) {
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_UCLK_BIT);
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_VMEMP_SCALING_BIT);
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_VDDIO_MEM_SCALING_BIT);
 	}
 
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_MEM_TEMP_READ_BIT);
-
-	if (adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK)
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_GFXCLK_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_SOCCLK_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_SOC_MPCLK_DS_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_BACO_MPCLK_DS_BIT);
+	if (!(adev->pm.pp_feature & PP_SCLK_DEEP_SLEEP_MASK))
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DS_GFXCLK_BIT);
 
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_FCLK_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_DCN_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_DCFCLK_BIT);
-
-	if (adev->pm.pp_feature & PP_PCIE_DPM_MASK) {
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_LINK_BIT);
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_LCLK_BIT);
+	if (!(adev->pm.pp_feature & PP_PCIE_DPM_MASK)) {
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DPM_LINK_BIT);
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_DS_LCLK_BIT);
 	}
 
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_BACO_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_MP0CLK_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FW_DSTATE_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_OUT_OF_BAND_MONITOR_BIT);
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_SOC_CG_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_FCLK_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_FW_CTF_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DS_UCLK_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_VR0HOT_BIT);
-
-	*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_DPM_GFX_POWER_OPTIMIZER_BIT);
-
-	if (adev->pm.pp_feature & PP_ULV_MASK)
-		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_GFX_ULV_BIT);
+	if (!(adev->pm.pp_feature & PP_ULV_MASK))
+		*(uint64_t *)feature_mask &= ~FEATURE_MASK(FEATURE_GFX_ULV_BIT);
 
 	return 0;
 }