Merge tag 'amd-drm-next-6.9-2024-02-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.9-2024-02-09:

amdgpu:
- Validate DMABuf imports in compute VMs
- Add RAS ACA framework
- PSP 13 fixes
- Misc code cleanups
- Replay fixes
- Atom interpretor PS, WS bounds checking
- DML2 fixes
- Audio fixes
- DCN 3.5 Z state fixes
- Remove deprecated ida_simple usage
- UBSAN fixes
- RAS fixes
- Enable seq64 infrastructure
- DC color block enablement
- Documentation updates
- DC documentation updates
- DMCUB updates
- S3 fixes
- VCN 4.0.5 fixes
- DP MST fixes
- SR-IOV fixes

amdkfd:
- Validate DMABuf imports in compute VMs
- SVM fixes
- Trap handler updates

radeon:
- Atom interpretor PS, WS bounds checking
- Misc code cleanups

UAPI:
- Bump KFD version so UMDs know that the fixes that enable the management of
  VA mappings in compute VMs using the GEM_VA ioctl for DMABufs exported from KFD are present
- Add INFO query for input power.  This matches the existing INFO query for average
  power.  Used in gaming HUDs, etc.
  Example userspace: https://github.com/Umio-Yasuno/libdrm-amdgpu-sys-rs/tree/input_power

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240209221459.5453-1-alexander.deucher@amd.com

9 files changed, 215 insertions, 52 deletions

diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
index f503e61faa60..b1b4c09c3467 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
@@ -226,7 +226,7 @@ int atomctrl_set_engine_dram_timings_rv770(
 
 	return amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, DynamicMemorySettings),
-			(uint32_t *)&engine_clock_parameters);
+			(uint32_t *)&engine_clock_parameters, sizeof(engine_clock_parameters));
 }
 
 /*
@@ -297,7 +297,7 @@ int atomctrl_get_memory_pll_dividers_si(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 		 GetIndexIntoMasterTable(COMMAND, ComputeMemoryClockParam),
-		(uint32_t *)&mpll_parameters);
+		(uint32_t *)&mpll_parameters, sizeof(mpll_parameters));
 
 	if (0 == result) {
 		mpll_param->mpll_fb_divider.clk_frac =
@@ -345,7 +345,7 @@ int atomctrl_get_memory_pll_dividers_vi(struct pp_hwmgr *hwmgr,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ComputeMemoryClockParam),
-			(uint32_t *)&mpll_parameters);
+			(uint32_t *)&mpll_parameters, sizeof(mpll_parameters));
 
 	if (!result)
 		mpll_param->mpll_post_divider =
@@ -366,7 +366,7 @@ int atomctrl_get_memory_pll_dividers_ai(struct pp_hwmgr *hwmgr,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ComputeMemoryClockParam),
-			(uint32_t *)&mpll_parameters);
+			(uint32_t *)&mpll_parameters, sizeof(mpll_parameters));
 
 	/* VEGAM's mpll takes sometime to finish computing */
 	udelay(10);
@@ -396,7 +396,7 @@ int atomctrl_get_engine_pll_dividers_kong(struct pp_hwmgr *hwmgr,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 		 GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL),
-		(uint32_t *)&pll_parameters);
+		(uint32_t *)&pll_parameters, sizeof(pll_parameters));
 
 	if (0 == result) {
 		dividers->pll_post_divider = pll_parameters.ucPostDiv;
@@ -420,7 +420,7 @@ int atomctrl_get_engine_pll_dividers_vi(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 		 GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL),
-		(uint32_t *)&pll_patameters);
+		(uint32_t *)&pll_patameters, sizeof(pll_patameters));
 
 	if (0 == result) {
 		dividers->pll_post_divider =
@@ -457,7 +457,7 @@ int atomctrl_get_engine_pll_dividers_ai(struct pp_hwmgr *hwmgr,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 		 GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL),
-		(uint32_t *)&pll_patameters);
+		(uint32_t *)&pll_patameters, sizeof(pll_patameters));
 
 	if (0 == result) {
 		dividers->usSclk_fcw_frac     = le16_to_cpu(pll_patameters.usSclk_fcw_frac);
@@ -490,7 +490,7 @@ int atomctrl_get_dfs_pll_dividers_vi(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 		 GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL),
-		(uint32_t *)&pll_patameters);
+		(uint32_t *)&pll_patameters, sizeof(pll_patameters));
 
 	if (0 == result) {
 		dividers->pll_post_divider =
@@ -773,7 +773,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 
 	if (result)
 		return result;
@@ -794,7 +794,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 
 	if (result)
 		return result;
@@ -814,7 +814,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 
 	if (result)
 		return result;
@@ -835,7 +835,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 
 	if (result)
 		return result;
@@ -857,7 +857,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 	if (result)
 		return result;
 
@@ -878,7 +878,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 
 	if (result)
 		return result;
@@ -909,7 +909,7 @@ int atomctrl_calculate_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&sOutput_FuseValues);
+			(uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues));
 
 	if (result)
 		return result;
@@ -1134,7 +1134,7 @@ int atomctrl_get_voltage_evv_on_sclk(
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, GetVoltageInfo),
-			(uint32_t *)&get_voltage_info_param_space);
+			(uint32_t *)&get_voltage_info_param_space, sizeof(get_voltage_info_param_space));
 
 	*voltage = result ? 0 :
 			le16_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2 *)
@@ -1179,7 +1179,7 @@ int atomctrl_get_voltage_evv(struct pp_hwmgr *hwmgr,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, GetVoltageInfo),
-			(uint32_t *)&get_voltage_info_param_space);
+			(uint32_t *)&get_voltage_info_param_space, sizeof(get_voltage_info_param_space));
 
 	if (0 != result)
 		return result;
@@ -1359,7 +1359,7 @@ int atomctrl_read_efuse(struct pp_hwmgr *hwmgr, uint16_t start_index,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, ReadEfuseValue),
-			(uint32_t *)&efuse_param);
+			(uint32_t *)&efuse_param, sizeof(efuse_param));
 	*efuse = result ? 0 : le32_to_cpu(efuse_param.ulEfuseValue) & mask;
 
 	return result;
@@ -1380,7 +1380,7 @@ int atomctrl_set_ac_timing_ai(struct pp_hwmgr *hwmgr, uint32_t memory_clock,
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 		 GetIndexIntoMasterTable(COMMAND, DynamicMemorySettings),
-		(uint32_t *)&memory_clock_parameters);
+		(uint32_t *)&memory_clock_parameters, sizeof(memory_clock_parameters));
 
 	return result;
 }
@@ -1399,7 +1399,7 @@ int atomctrl_get_voltage_evv_on_sclk_ai(struct pp_hwmgr *hwmgr, uint8_t voltage_
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, GetVoltageInfo),
-			(uint32_t *)&get_voltage_info_param_space);
+			(uint32_t *)&get_voltage_info_param_space, sizeof(get_voltage_info_param_space));
 
 	*voltage = result ? 0 :
 		le32_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_3 *)(&get_voltage_info_param_space))->ulVoltageLevel);
@@ -1526,7 +1526,7 @@ int atomctrl_get_leakage_id_from_efuse(struct pp_hwmgr *hwmgr, uint16_t *virtual
 
 	result = amdgpu_atom_execute_table(adev->mode_info.atom_context,
 			GetIndexIntoMasterTable(COMMAND, SetVoltage),
-			(uint32_t *)voltage_parameters);
+			(uint32_t *)voltage_parameters, sizeof(*voltage_parameters));
 
 	*virtual_voltage_id = voltage_parameters->usVoltageLevel;
 
diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c
index a47a47238e2b..82d540334318 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c
@@ -258,7 +258,7 @@ int pp_atomfwctrl_get_gpu_pll_dividers_vega10(struct pp_hwmgr *hwmgr,
 	idx = GetIndexIntoMasterCmdTable(computegpuclockparam);
 
 	if (amdgpu_atom_execute_table(
-		adev->mode_info.atom_context, idx, (uint32_t *)&pll_parameters))
+		adev->mode_info.atom_context, idx, (uint32_t *)&pll_parameters, sizeof(pll_parameters)))
 		return -EINVAL;
 
 	pll_output = (struct compute_gpu_clock_output_parameter_v1_8 *)
@@ -505,7 +505,7 @@ int pp_atomfwctrl_get_clk_information_by_clkid(struct pp_hwmgr *hwmgr,
 	ix = GetIndexIntoMasterCmdTable(getsmuclockinfo);
 
 	if (amdgpu_atom_execute_table(
-		adev->mode_info.atom_context, ix, (uint32_t *)&parameters))
+		adev->mode_info.atom_context, ix, (uint32_t *)&parameters, sizeof(parameters)))
 		return -EINVAL;
 
 	output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)&parameters;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
index c7bfa68bf00f..f6545093bfc1 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
@@ -514,7 +514,7 @@ static int smu_v11_0_atom_get_smu_clockinfo(struct amdgpu_device *adev,
 					    getsmuclockinfo);
 
 	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
-					(uint32_t *)&input);
+					(uint32_t *)&input, sizeof(input));
 	if (ret)
 		return -EINVAL;
 
@@ -1432,24 +1432,24 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev,
 		dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n");
 		orderly_poweroff(true);
 	} else if (client_id == SOC15_IH_CLIENTID_MP1) {
-		if (src_id == 0xfe) {
+		if (src_id == SMU_IH_INTERRUPT_ID_TO_DRIVER) {
 			/* ACK SMUToHost interrupt */
 			data = RREG32_SOC15(MP1, 0, mmMP1_SMN_IH_SW_INT_CTRL);
 			data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1);
 			WREG32_SOC15(MP1, 0, mmMP1_SMN_IH_SW_INT_CTRL, data);
 
 			switch (ctxid) {
-			case 0x3:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_AC:
 				dev_dbg(adev->dev, "Switched to AC mode!\n");
 				schedule_work(&smu->interrupt_work);
 				adev->pm.ac_power = true;
 				break;
-			case 0x4:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_DC:
 				dev_dbg(adev->dev, "Switched to DC mode!\n");
 				schedule_work(&smu->interrupt_work);
 				adev->pm.ac_power = false;
 				break;
-			case 0x7:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING:
 				/*
 				 * Increment the throttle interrupt counter
 				 */
@@ -1462,6 +1462,10 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev,
 					schedule_work(&smu->throttling_logging_work);
 
 				break;
+			default:
+				dev_dbg(adev->dev, "Unhandled context id %d from client:%d!\n",
+									ctxid, client_id);
+				break;
 			}
 		}
 	}
@@ -1504,7 +1508,7 @@ int smu_v11_0_register_irq_handler(struct smu_context *smu)
 		return ret;
 
 	ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1,
-				0xfe,
+				SMU_IH_INTERRUPT_ID_TO_DRIVER,
 				irq_src);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c
index 5e408a195860..ed15f5a0fd11 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c
@@ -301,7 +301,7 @@ static int smu_v12_0_atom_get_smu_clockinfo(struct amdgpu_device *adev,
 					    getsmuclockinfo);
 
 	ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index,
-					(uint32_t *)&input);
+					(uint32_t *)&input, sizeof(input));
 	if (ret)
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index c486182ff275..48170bb5112e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -1369,24 +1369,24 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev,
 		dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n");
 		orderly_poweroff(true);
 	} else if (client_id == SOC15_IH_CLIENTID_MP1) {
-		if (src_id == 0xfe) {
+		if (src_id == SMU_IH_INTERRUPT_ID_TO_DRIVER) {
 			/* ACK SMUToHost interrupt */
 			data = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL);
 			data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1);
 			WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, data);
 
 			switch (ctxid) {
-			case 0x3:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_AC:
 				dev_dbg(adev->dev, "Switched to AC mode!\n");
 				smu_v13_0_ack_ac_dc_interrupt(smu);
 				adev->pm.ac_power = true;
 				break;
-			case 0x4:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_DC:
 				dev_dbg(adev->dev, "Switched to DC mode!\n");
 				smu_v13_0_ack_ac_dc_interrupt(smu);
 				adev->pm.ac_power = false;
 				break;
-			case 0x7:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING:
 				/*
 				 * Increment the throttle interrupt counter
 				 */
@@ -1399,7 +1399,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev,
 					schedule_work(&smu->throttling_logging_work);
 
 				break;
-			case 0x8:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL:
 				high = smu->thermal_range.software_shutdown_temp +
 					smu->thermal_range.software_shutdown_temp_offset;
 				high = min_t(typeof(high),
@@ -1416,7 +1416,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev,
 				data = data & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK);
 				WREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL, data);
 				break;
-			case 0x9:
+			case SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY:
 				high = min_t(typeof(high),
 					     SMU_THERMAL_MAXIMUM_ALERT_TEMP,
 					     smu->thermal_range.software_shutdown_temp);
@@ -1429,6 +1429,10 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev,
 				data = data & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK);
 				WREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL, data);
 				break;
+			default:
+				dev_dbg(adev->dev, "Unhandled context id %d from client:%d!\n",
+									ctxid, client_id);
+				break;
 			}
 		}
 	}
@@ -1473,7 +1477,7 @@ int smu_v13_0_register_irq_handler(struct smu_context *smu)
 		return ret;
 
 	ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1,
-				0xfe,
+				SMU_IH_INTERRUPT_ID_TO_DRIVER,
 				irq_src);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 7e1941cf1796..1b96158b63bf 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -45,6 +45,7 @@
 #include <linux/pci.h>
 #include "amdgpu_ras.h"
 #include "amdgpu_mca.h"
+#include "amdgpu_aca.h"
 #include "smu_cmn.h"
 #include "mp/mp_13_0_6_offset.h"
 #include "mp/mp_13_0_6_sh_mask.h"
@@ -1438,7 +1439,10 @@ static int smu_v13_0_6_irq_process(struct amdgpu_device *adev,
 							entry->src_data[1]);
 					schedule_work(&smu->throttling_logging_work);
 				}
-
+				break;
+			default:
+				dev_dbg(adev->dev, "Unhandled context id %d from client:%d!\n",
+									ctxid, client_id);
 				break;
 			}
 		}
@@ -2547,18 +2551,22 @@ static int mca_umc_mca_get_err_count(const struct mca_ras_info *mca_ras, struct
 				     enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count)
 {
 	uint64_t status0;
+	uint32_t ext_error_code;
+	uint32_t odecc_err_cnt;
 
 	status0 = entry->regs[MCA_REG_IDX_STATUS];
+	ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(status0);
+	odecc_err_cnt = MCA_REG__MISC0__ERRCNT(entry->regs[MCA_REG_IDX_MISC0]);
 
 	if (!REG_GET_FIELD(status0, MCMP1_STATUST0, Val)) {
 		*count = 0;
 		return 0;
 	}
 
-	if (type == AMDGPU_MCA_ERROR_TYPE_UE && umc_v12_0_is_uncorrectable_error(adev, status0))
-		*count = 1;
-	else if (type == AMDGPU_MCA_ERROR_TYPE_CE && umc_v12_0_is_correctable_error(adev, status0))
-		*count = 1;
+	if (umc_v12_0_is_deferred_error(adev, status0) ||
+	    umc_v12_0_is_uncorrectable_error(adev, status0) ||
+	    umc_v12_0_is_correctable_error(adev, status0))
+		*count = (ext_error_code == 0) ? odecc_err_cnt : 1;
 
 	return 0;
 }
@@ -2857,6 +2865,143 @@ static const struct amdgpu_mca_smu_funcs smu_v13_0_6_mca_smu_funcs = {
 	.mca_get_valid_mca_count = mca_smu_get_valid_mca_count,
 };
 
+static int aca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+
+	return smu_v13_0_6_mca_set_debug_mode(smu, enable);
+}
+
+static int smu_v13_0_6_get_valid_aca_count(struct smu_context *smu, enum aca_error_type type, u32 *count)
+{
+	uint32_t msg;
+	int ret;
+
+	if (!count)
+		return -EINVAL;
+
+	switch (type) {
+	case ACA_ERROR_TYPE_UE:
+		msg = SMU_MSG_QueryValidMcaCount;
+		break;
+	case ACA_ERROR_TYPE_CE:
+		msg = SMU_MSG_QueryValidMcaCeCount;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ret = smu_cmn_send_smc_msg(smu, msg, count);
+	if (ret) {
+		*count = 0;
+		return ret;
+	}
+
+	return 0;
+}
+
+static int aca_smu_get_valid_aca_count(struct amdgpu_device *adev,
+				       enum aca_error_type type, u32 *count)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	int ret;
+
+	switch (type) {
+	case ACA_ERROR_TYPE_UE:
+	case ACA_ERROR_TYPE_CE:
+		ret = smu_v13_0_6_get_valid_aca_count(smu, type, count);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
+static int __smu_v13_0_6_aca_bank_dump(struct smu_context *smu, enum aca_error_type type,
+				       int idx, int offset, u32 *val)
+{
+	uint32_t msg, param;
+
+	switch (type) {
+	case ACA_ERROR_TYPE_UE:
+		msg = SMU_MSG_McaBankDumpDW;
+		break;
+	case ACA_ERROR_TYPE_CE:
+		msg = SMU_MSG_McaBankCeDumpDW;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	param = ((idx & 0xffff) << 16) | (offset & 0xfffc);
+
+	return smu_cmn_send_smc_msg_with_param(smu, msg, param, (uint32_t *)val);
+}
+
+static int smu_v13_0_6_aca_bank_dump(struct smu_context *smu, enum aca_error_type type,
+				     int idx, int offset, u32 *val, int count)
+{
+	int ret, i;
+
+	if (!val)
+		return -EINVAL;
+
+	for (i = 0; i < count; i++) {
+		ret = __smu_v13_0_6_aca_bank_dump(smu, type, idx, offset + (i << 2), &val[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int aca_bank_read_reg(struct amdgpu_device *adev, enum aca_error_type type,
+			     int idx, int reg_idx, u64 *val)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	u32 data[2] = {0, 0};
+	int ret;
+
+	if (!val || reg_idx >= ACA_REG_IDX_COUNT)
+		return -EINVAL;
+
+	ret = smu_v13_0_6_aca_bank_dump(smu, type, idx, reg_idx * 8, data, ARRAY_SIZE(data));
+	if (ret)
+		return ret;
+
+	*val = (u64)data[1] << 32 | data[0];
+
+	dev_dbg(adev->dev, "mca read bank reg: type:%s, index: %d, reg_idx: %d, val: 0x%016llx\n",
+		type == ACA_ERROR_TYPE_UE ? "UE" : "CE", idx, reg_idx, *val);
+
+	return 0;
+}
+
+static int aca_smu_get_valid_aca_bank(struct amdgpu_device *adev,
+				      enum aca_error_type type, int idx, struct aca_bank *bank)
+{
+	int i, ret, count;
+
+	count = min_t(int, 16, ARRAY_SIZE(bank->regs));
+	for (i = 0; i < count; i++) {
+		ret = aca_bank_read_reg(adev, type, idx, i, &bank->regs[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static const struct aca_smu_funcs smu_v13_0_6_aca_smu_funcs = {
+	.max_ue_bank_count = 12,
+	.max_ce_bank_count = 12,
+	.set_debug_mode = aca_smu_set_debug_mode,
+	.get_valid_aca_count = aca_smu_get_valid_aca_count,
+	.get_valid_aca_bank = aca_smu_get_valid_aca_bank,
+};
+
 static int smu_v13_0_6_select_xgmi_plpd_policy(struct smu_context *smu,
 					       enum pp_xgmi_plpd_mode mode)
 {
@@ -2895,13 +3040,6 @@ static int smu_v13_0_6_select_xgmi_plpd_policy(struct smu_context *smu,
 	return ret;
 }
 
-static ssize_t smu_v13_0_6_get_ecc_info(struct smu_context *smu,
-			void *table)
-{
-	/* Support ecc info by default */
-	return 0;
-}
-
 static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	/* init dpm */
 	.get_allowed_feature_mask = smu_v13_0_6_get_allowed_feature_mask,
@@ -2956,7 +3094,6 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.i2c_init = smu_v13_0_6_i2c_control_init,
 	.i2c_fini = smu_v13_0_6_i2c_control_fini,
 	.send_hbm_bad_pages_num = smu_v13_0_6_smu_send_hbm_bad_page_num,
-	.get_ecc_info = smu_v13_0_6_get_ecc_info,
 };
 
 void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
@@ -2969,4 +3106,5 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
 	smu->smc_driver_if_version = SMU13_0_6_DRIVER_IF_VERSION;
 	smu_v13_0_set_smu_mailbox_registers(smu);
 	amdgpu_mca_smu_init_funcs(smu->adev, &smu_v13_0_6_mca_smu_funcs);
+	amdgpu_aca_set_smu_funcs(smu->adev, &smu_v13_0_6_aca_smu_funcs);
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c
index 4894f7ee737b..2aa7e9945a0b 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c
@@ -892,7 +892,7 @@ int smu_v14_0_register_irq_handler(struct smu_context *smu)
 	// TODO: THM related
 
 	ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1,
-				0xfe,
+				SMU_IH_INTERRUPT_ID_TO_DRIVER,
 				irq_src);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
index 00cd615bbcdc..b8dbd4e25348 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
@@ -378,8 +378,15 @@ int smu_cmn_send_smc_msg_with_param(struct smu_context *smu,
 	res = __smu_cmn_reg2errno(smu, reg);
 	if (res != 0)
 		__smu_cmn_reg_print_error(smu, reg, index, param, msg);
-	if (read_arg)
+	if (read_arg) {
 		smu_cmn_read_arg(smu, read_arg);
+		dev_dbg(adev->dev, "smu send message: %s(%d) param: 0x%08x, resp: 0x%08x,\
+			readval: 0x%08x\n",
+			smu_get_message_name(smu, msg), index, param, reg, *read_arg);
+	} else {
+		dev_dbg(adev->dev, "smu send message: %s(%d) param: 0x%08x, resp: 0x%08x\n",
+			smu_get_message_name(smu, msg), index, param, reg);
+	}
 Out:
 	if (unlikely(adev->pm.smu_debug_mask & SMU_DEBUG_HALT_ON_ERROR) && res) {
 		amdgpu_device_halt(adev);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
index cc590e27d88a..81bfce1406e5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
@@ -30,6 +30,16 @@
 #define FDO_PWM_MODE_STATIC  1
 #define FDO_PWM_MODE_STATIC_RPM 5
 
+#define SMU_IH_INTERRUPT_ID_TO_DRIVER                   0xFE
+#define SMU_IH_INTERRUPT_CONTEXT_ID_BACO                0x2
+#define SMU_IH_INTERRUPT_CONTEXT_ID_AC                  0x3
+#define SMU_IH_INTERRUPT_CONTEXT_ID_DC                  0x4
+#define SMU_IH_INTERRUPT_CONTEXT_ID_AUDIO_D0            0x5
+#define SMU_IH_INTERRUPT_CONTEXT_ID_AUDIO_D3            0x6
+#define SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING  0x7
+#define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL        0x8
+#define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY        0x9
+
 extern const int link_speed[];
 
 /* Helper to Convert from PCIE Gen 1/2/3/4/5/6 to 0.1 GT/s speed units */