Merge tag 'amd-drm-next-6.15-2025-03-14' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.15-2025-03-14:

amdgpu:
- GC 12.x DCC fixes
- VCN 2.5 fix
- Replay/PSR fixes
- HPD fixes
- DMUB fixes
- Backlight fixes
- DM suspend/resume cleanup
- Misc DC fixes
- HDCP UAF fix
- Misc code cleanups
- VCE 2.x fix
- Wedged event support
- GC 12.x PTE fixes
- Misc multimedia cap fixes
- Enable unique id support for GC 12.x
- XGMI code cleanup
- GC 11.x and 12.x MQD cleanups
- SMU 13.x updates
- SMU 14.x fan speed reporting
- Enable VCN activity reporting for additional chips
- SR-IOV fixes
- RAS fixes
- MES fixes

amdkfd:
- Dequeue wait count API cleanups
- Queue eviction cleanup fixes
- Retry fault fixes
- Dequeue retry timeout adjustments
- GC 12.x trap handler fixes
- GC 9.5.x updates

radeon:
- VCE command parser fix

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314170618.3142042-1-alexander.deucher@amd.com

8 files changed, 103 insertions, 14 deletions

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 1d04f1b79ded..922def51685b 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -1936,7 +1936,7 @@ static int pp_od_clk_voltage_attr_update(struct amdgpu_device *adev, struct amdg
 	if (gc_ver == IP_VERSION(9, 4, 3) ||
 	    gc_ver == IP_VERSION(9, 4, 4) ||
 	    gc_ver == IP_VERSION(9, 5, 0)) {
-		if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
+		if (amdgpu_sriov_multi_vf_mode(adev))
 			*states = ATTR_STATE_UNSUPPORTED;
 		return 0;
 	}
@@ -1971,7 +1971,7 @@ static int pp_dpm_dcefclk_attr_update(struct amdgpu_device *adev, struct amdgpu_
 	 * setting should not be allowed from VF if not in one VF mode.
 	 */
 	if (gc_ver >= IP_VERSION(10, 0, 0) ||
-	    (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))) {
+	    (amdgpu_sriov_multi_vf_mode(adev))) {
 		dev_attr->attr.mode &= ~S_IWUGO;
 		dev_attr->store = NULL;
 	}
@@ -2314,13 +2314,22 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 		    gc_ver == IP_VERSION(9, 0, 1))
 			*states = ATTR_STATE_UNSUPPORTED;
 	} else if (DEVICE_ATTR_IS(vcn_busy_percent)) {
-		if (!(gc_ver == IP_VERSION(10, 3, 1) ||
-			  gc_ver == IP_VERSION(10, 3, 3) ||
-			  gc_ver == IP_VERSION(10, 3, 6) ||
-			  gc_ver == IP_VERSION(10, 3, 7) ||
-			  gc_ver == IP_VERSION(11, 0, 1) ||
-			  gc_ver == IP_VERSION(11, 0, 4) ||
-			  gc_ver == IP_VERSION(11, 5, 0)))
+		if (!(gc_ver == IP_VERSION(9, 3, 0) ||
+		      gc_ver == IP_VERSION(10, 3, 1) ||
+		      gc_ver == IP_VERSION(10, 3, 3) ||
+		      gc_ver == IP_VERSION(10, 3, 6) ||
+		      gc_ver == IP_VERSION(10, 3, 7) ||
+		      gc_ver == IP_VERSION(11, 0, 0) ||
+		      gc_ver == IP_VERSION(11, 0, 1) ||
+		      gc_ver == IP_VERSION(11, 0, 2) ||
+		      gc_ver == IP_VERSION(11, 0, 3) ||
+		      gc_ver == IP_VERSION(11, 0, 4) ||
+		      gc_ver == IP_VERSION(11, 5, 0) ||
+		      gc_ver == IP_VERSION(11, 5, 1) ||
+		      gc_ver == IP_VERSION(11, 5, 2) ||
+		      gc_ver == IP_VERSION(11, 5, 3) ||
+		      gc_ver == IP_VERSION(12, 0, 0) ||
+		      gc_ver == IP_VERSION(12, 0, 1)))
 			*states = ATTR_STATE_UNSUPPORTED;
 	} else if (DEVICE_ATTR_IS(pcie_bw)) {
 		/* PCIe Perf counters won't work on APU nodes */
@@ -2341,6 +2350,8 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 		case IP_VERSION(11, 0, 1):
 		case IP_VERSION(11, 0, 2):
 		case IP_VERSION(11, 0, 3):
+		case IP_VERSION(12, 0, 0):
+		case IP_VERSION(12, 0, 1):
 			*states = ATTR_STATE_SUPPORTED;
 			break;
 		default:
diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 8cfb07549f54..d1aa6075c4f4 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -1814,7 +1814,7 @@ static int smu_hw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	struct smu_context *smu = adev->powerplay.pp_handle;
 
-	if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) {
+	if (amdgpu_sriov_multi_vf_mode(adev)) {
 		smu->pm_enabled = false;
 		return 0;
 	}
@@ -2038,7 +2038,7 @@ static int smu_hw_fini(struct amdgpu_ip_block *ip_block)
 	struct smu_context *smu = adev->powerplay.pp_handle;
 	int i, ret;
 
-	if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
+	if (amdgpu_sriov_multi_vf_mode(adev))
 		return 0;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
@@ -2106,7 +2106,7 @@ static int smu_suspend(struct amdgpu_ip_block *ip_block)
 	int ret;
 	uint64_t count;
 
-	if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
+	if (amdgpu_sriov_multi_vf_mode(adev))
 		return 0;
 
 	if (!smu->pm_enabled)
@@ -2142,7 +2142,7 @@ static int smu_resume(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	struct smu_context *smu = adev->powerplay.pp_handle;
 
-	if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev))
+	if (amdgpu_sriov_multi_vf_mode(adev))
 		return 0;
 
 	if (!smu->pm_enabled)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
index 9c8468fb203a..c9dee09395e3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
@@ -356,6 +356,7 @@ enum smu_clk_type {
        __SMU_DUMMY_MAP(DS_FCLK),                       	\
        __SMU_DUMMY_MAP(DS_MP1CLK),                     	\
        __SMU_DUMMY_MAP(DS_MP0CLK),                     	\
+       __SMU_DUMMY_MAP(DS_MPIOCLK),                     \
        __SMU_DUMMY_MAP(XGMI_PER_LINK_PWR_DWN),          \
        __SMU_DUMMY_MAP(DPM_GFX_PACE),                  	\
        __SMU_DUMMY_MAP(MEM_VDDCI_SCALING),             	\
@@ -452,7 +453,8 @@ enum smu_clk_type {
        __SMU_DUMMY_MAP(APT_PF_DCS),				\
        __SMU_DUMMY_MAP(GFX_EDC_XVMIN),				\
        __SMU_DUMMY_MAP(GFX_DIDT_XVMIN),				\
-       __SMU_DUMMY_MAP(FAN_ABNORMAL),
+       __SMU_DUMMY_MAP(FAN_ABNORMAL),				\
+       __SMU_DUMMY_MAP(PIT),
 
 #undef __SMU_DUMMY_MAP
 #define __SMU_DUMMY_MAP(feature)	SMU_FEATURE_##feature##_BIT
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
index 37d82a71a2d7..9481f897432d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
@@ -1285,6 +1285,12 @@ static int renoir_read_sensor(struct smu_context *smu,
 						  (uint32_t *)data);
 		*size = 4;
 		break;
+	case AMDGPU_PP_SENSOR_VCN_LOAD:
+		ret = renoir_get_smu_metrics_data(smu,
+						  METRICS_AVERAGE_VCNACTIVITY,
+						  (uint32_t *)data);
+		*size = 4;
+		break;
 	case AMDGPU_PP_SENSOR_EDGE_TEMP:
 		ret = renoir_get_smu_metrics_data(smu,
 						  METRICS_TEMPERATURE_EDGE,
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 898487ad6cd2..5a9711e8cf68 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
@@ -836,6 +836,10 @@ static int smu_v13_0_0_get_smu_metrics_data(struct smu_context *smu,
 	case METRICS_AVERAGE_MEMACTIVITY:
 		*value = metrics->AverageUclkActivity;
 		break;
+	case METRICS_AVERAGE_VCNACTIVITY:
+		*value = max(metrics->Vcn0ActivityPercentage,
+			     metrics->Vcn1ActivityPercentage);
+		break;
 	case METRICS_AVERAGE_SOCKETPOWER:
 		*value = metrics->AverageSocketPower << 8;
 		break;
@@ -962,6 +966,12 @@ static int smu_v13_0_0_read_sensor(struct smu_context *smu,
 						       (uint32_t *)data);
 		*size = 4;
 		break;
+	case AMDGPU_PP_SENSOR_VCN_LOAD:
+		ret = smu_v13_0_0_get_smu_metrics_data(smu,
+						       METRICS_AVERAGE_VCNACTIVITY,
+						       (uint32_t *)data);
+		*size = 4;
+		break;
 	case AMDGPU_PP_SENSOR_GPU_AVG_POWER:
 		ret = smu_v13_0_0_get_smu_metrics_data(smu,
 						       METRICS_AVERAGE_SOCKETPOWER,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 285dbfe10303..51078accc0f8 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -76,6 +76,11 @@ const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[SMU_FEATURE_COUNT] =
 	SMU_13_0_12_FEA_MAP(SMU_FEATURE_THERMAL_BIT, 			FEATURE_THERMAL),
 	SMU_13_0_12_FEA_MAP(SMU_FEATURE_SOC_PCC_BIT, 			FEATURE_SOC_PCC),
 	SMU_13_0_12_FEA_MAP(SMU_FEATURE_XGMI_PER_LINK_PWR_DWN_BIT,	FEATURE_XGMI_PER_LINK_PWR_DOWN),
+	SMU_13_0_12_FEA_MAP(SMU_FEATURE_DS_VCN_BIT,			FEATURE_DS_VCN),
+	SMU_13_0_12_FEA_MAP(SMU_FEATURE_DS_MP1CLK_BIT,			FEATURE_DS_MP1CLK),
+	SMU_13_0_12_FEA_MAP(SMU_FEATURE_DS_MPIOCLK_BIT,			FEATURE_DS_MPIOCLK),
+	SMU_13_0_12_FEA_MAP(SMU_FEATURE_DS_MP0CLK_BIT,			FEATURE_DS_MP0CLK),
+	SMU_13_0_12_FEA_MAP(SMU_FEATURE_PIT_BIT,			FEATURE_PIT),
 };
 
 // clang-format off
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 19f47811f6db..c8f4f6fb4083 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -807,6 +807,10 @@ static int smu_v13_0_7_get_smu_metrics_data(struct smu_context *smu,
 		else
 			*value = metrics->AverageMemclkFrequencyPreDs;
 		break;
+	case METRICS_AVERAGE_VCNACTIVITY:
+		*value = max(metrics->Vcn0ActivityPercentage,
+			     metrics->Vcn1ActivityPercentage);
+		break;
 	case METRICS_AVERAGE_VCLK:
 		*value = metrics->AverageVclk0Frequency;
 		break;
@@ -951,6 +955,12 @@ static int smu_v13_0_7_read_sensor(struct smu_context *smu,
 						       (uint32_t *)data);
 		*size = 4;
 		break;
+	case AMDGPU_PP_SENSOR_VCN_LOAD:
+		ret = smu_v13_0_7_get_smu_metrics_data(smu,
+						       METRICS_AVERAGE_VCNACTIVITY,
+						       (uint32_t *)data);
+		*size = 4;
+		break;
 	case AMDGPU_PP_SENSOR_GPU_AVG_POWER:
 		ret = smu_v13_0_7_get_smu_metrics_data(smu,
 						       METRICS_AVERAGE_SOCKETPOWER,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
index 5cad09c5f2ff..21d1b34ada05 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
@@ -756,6 +756,10 @@ static int smu_v14_0_2_get_smu_metrics_data(struct smu_context *smu,
 	case METRICS_AVERAGE_MEMACTIVITY:
 		*value = metrics->AverageUclkActivity;
 		break;
+	case METRICS_AVERAGE_VCNACTIVITY:
+		*value = max(metrics->AverageVcn0ActivityPercentage,
+			     metrics->Vcn1ActivityPercentage);
+		break;
 	case METRICS_AVERAGE_SOCKETPOWER:
 		*value = metrics->AverageSocketPower << 8;
 		break;
@@ -882,6 +886,12 @@ static int smu_v14_0_2_read_sensor(struct smu_context *smu,
 						       (uint32_t *)data);
 		*size = 4;
 		break;
+	case AMDGPU_PP_SENSOR_VCN_LOAD:
+		ret = smu_v14_0_2_get_smu_metrics_data(smu,
+						       METRICS_AVERAGE_VCNACTIVITY,
+						       (uint32_t *)data);
+		*size = 4;
+		break;
 	case AMDGPU_PP_SENSOR_GPU_AVG_POWER:
 		ret = smu_v14_0_2_get_smu_metrics_data(smu,
 						       METRICS_AVERAGE_SOCKETPOWER,
@@ -1627,6 +1637,39 @@ out:
 	adev->unique_id = ((uint64_t)upper32 << 32) | lower32;
 }
 
+static int smu_v14_0_2_get_fan_speed_pwm(struct smu_context *smu,
+					 uint32_t *speed)
+{
+	int ret;
+
+	if (!speed)
+		return -EINVAL;
+
+	ret = smu_v14_0_2_get_smu_metrics_data(smu,
+					       METRICS_CURR_FANPWM,
+					       speed);
+	if (ret) {
+		dev_err(smu->adev->dev, "Failed to get fan speed(PWM)!");
+		return ret;
+	}
+
+	/* Convert the PMFW output which is in percent to pwm(255) based */
+	*speed = min(*speed * 255 / 100, (uint32_t)255);
+
+	return 0;
+}
+
+static int smu_v14_0_2_get_fan_speed_rpm(struct smu_context *smu,
+					 uint32_t *speed)
+{
+	if (!speed)
+		return -EINVAL;
+
+	return smu_v14_0_2_get_smu_metrics_data(smu,
+						METRICS_CURR_FANSPEED,
+						speed);
+}
+
 static int smu_v14_0_2_get_power_limit(struct smu_context *smu,
 				       uint32_t *current_power_limit,
 				       uint32_t *default_power_limit,
@@ -2804,6 +2847,8 @@ static const struct pptable_funcs smu_v14_0_2_ppt_funcs = {
 	.set_performance_level = smu_v14_0_set_performance_level,
 	.gfx_off_control = smu_v14_0_gfx_off_control,
 	.get_unique_id = smu_v14_0_2_get_unique_id,
+	.get_fan_speed_pwm = smu_v14_0_2_get_fan_speed_pwm,
+	.get_fan_speed_rpm = smu_v14_0_2_get_fan_speed_rpm,
 	.get_power_limit = smu_v14_0_2_get_power_limit,
 	.set_power_limit = smu_v14_0_2_set_power_limit,
 	.get_power_profile_mode = smu_v14_0_2_get_power_profile_mode,