Merge tag 'amd-drm-next-5.8-2020-04-30' of git://people.freedesktop.org/~agd5f/linux into drm-next

amd-drm-next-5.8-2020-04-30:

amdgpu:
- SR-IOV fixes
- SDMA fix for Navi
- VCN 2.5 DPG fixes
- Display fixes
- Display stuttering fixes for pageflip and cursor
- Add support for handling encrypted GPU memory
- Add UAPI for encrypted GPU memory
- Rework IB pool handling

amdkfd:
- Expose asic revision in topology
- Add UAPI for GWS (Global Wave Sync) resource management

UAPI:
- Add amdgpu UAPI for encrypted GPU memory
  Used by: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4401
- Add amdkfd UAPI for GWS (Global Wave Sync) resource management
  Thunk usage of KFD ioctl: https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/blob/roc-2.8.0/src/queues.c#L840
  ROCr usage of Thunk API: https://github.com/RadeonOpenCompute/ROCR-Runtime/blob/roc-3.1.0/src/core/runtime/amd_gpu_agent.cpp#L597
  HCC code using ROCr API: https://github.com/RadeonOpenCompute/hcc/blob/98ee9f34945d3b5f572d7a4c15cbffa506487734/lib/hsa/mcwamp_hsa.cpp#L2161
  HIP code using HCC API: https://github.com/ROCm-Developer-Tools/HIP/blob/cf8589b8c8a40ddcc55fa3a51e23390a49824130/src/hip_module.cpp#L567

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200430212951.3902-1-alexander.deucher@amd.com

4 files changed, 63 insertions, 27 deletions

diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index fdff3e1c5e95..fc31499c2e5c 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -1438,7 +1438,8 @@ static int pp_get_asic_baco_capability(void *handle, bool *cap)
 	if (!hwmgr)
 		return -EINVAL;
 
-	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_capability)
+	if (!(hwmgr->not_vf && amdgpu_dpm) ||
+		!hwmgr->hwmgr_func->get_asic_baco_capability)
 		return 0;
 
 	mutex_lock(&hwmgr->smu_lock);
@@ -1455,8 +1456,7 @@ static int pp_get_asic_baco_state(void *handle, int *state)
 	if (!hwmgr)
 		return -EINVAL;
 
-	if (!(hwmgr->not_vf && amdgpu_dpm) ||
-		!hwmgr->hwmgr_func->get_asic_baco_state)
+	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_state)
 		return 0;
 
 	mutex_lock(&hwmgr->smu_lock);
@@ -1473,7 +1473,8 @@ static int pp_set_asic_baco_state(void *handle, int state)
 	if (!hwmgr)
 		return -EINVAL;
 
-	if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_asic_baco_state)
+	if (!(hwmgr->not_vf && amdgpu_dpm) ||
+		!hwmgr->hwmgr_func->set_asic_baco_state)
 		return 0;
 
 	mutex_lock(&hwmgr->smu_lock);
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
index 88b4e5642302..5964d6323a13 100644
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -347,13 +347,13 @@ int smu_get_dpm_freq_by_index(struct smu_context *smu, enum smu_clk_type clk_typ
 	param = (uint32_t)(((clk_id & 0xffff) << 16) | (level & 0xffff));
 
 	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex,
-					  param, &param);
+					  param, value);
 	if (ret)
 		return ret;
 
 	/* BIT31:  0 - Fine grained DPM, 1 - Dicrete DPM
 	 * now, we un-support it */
-	*value = param & 0x7fffffff;
+	*value = *value & 0x7fffffff;
 
 	return ret;
 }
@@ -535,7 +535,6 @@ int smu_update_table(struct smu_context *smu, enum smu_table_id table_index, int
 	int table_id = smu_table_get_index(smu, table_index);
 	uint32_t table_size;
 	int ret = 0;
-
 	if (!table_data || table_id >= SMU_TABLE_COUNT || table_id < 0)
 		return -EINVAL;
 
@@ -571,15 +570,10 @@ bool is_support_sw_smu(struct amdgpu_device *adev)
 	if (adev->asic_type == CHIP_VEGA20)
 		return (amdgpu_dpm == 2) ? true : false;
 	else if (adev->asic_type >= CHIP_ARCTURUS) {
-		if (amdgpu_sriov_vf(adev) &&
-		    !(adev->asic_type == CHIP_ARCTURUS &&
-		      amdgpu_sriov_is_pp_one_vf(adev)))
-
-			return false;
-		else
+	      if (amdgpu_sriov_is_pp_one_vf(adev) || !amdgpu_sriov_vf(adev))
 			return true;
-	} else
-		return false;
+	}
+	return false;
 }
 
 bool is_support_sw_smu_xgmi(struct amdgpu_device *adev)
@@ -696,7 +690,6 @@ int smu_feature_is_enabled(struct smu_context *smu, enum smu_feature_mask mask)
 
 	if (smu->is_apu)
 		return 1;
-
 	feature_id = smu_feature_get_index(smu, mask);
 	if (feature_id < 0)
 		return 0;
@@ -1344,6 +1337,9 @@ static int smu_hw_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct smu_context *smu = &adev->smu;
 
+	if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
+		return 0;
+
 	ret = smu_start_smc_engine(smu);
 	if (ret) {
 		pr_err("SMU is not ready yet!\n");
@@ -1357,9 +1353,6 @@ static int smu_hw_init(void *handle)
 		smu_set_gfx_cgpg(&adev->smu, true);
 	}
 
-	if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
-		return 0;
-
 	if (!smu->pm_enabled)
 		return 0;
 
@@ -1403,6 +1396,9 @@ failed:
 
 static int smu_stop_dpms(struct smu_context *smu)
 {
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	return smu_system_features_control(smu, false);
 }
 
diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
index c94270f7c198..2184d247a9f7 100644
--- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
@@ -1817,7 +1817,8 @@ static int navi10_get_power_limit(struct smu_context *smu,
 	int power_src;
 
 	if (!smu->power_limit) {
-		if (smu_feature_is_enabled(smu, SMU_FEATURE_PPT_BIT)) {
+		if (smu_feature_is_enabled(smu, SMU_FEATURE_PPT_BIT) &&
+			!amdgpu_sriov_vf(smu->adev)) {
 			power_src = smu_power_get_index(smu, SMU_POWER_SOURCE_AC);
 			if (power_src < 0)
 				return -EINVAL;
@@ -1960,6 +1961,9 @@ static int navi10_set_default_od_settings(struct smu_context *smu, bool initiali
 	OverDriveTable_t *od_table, *boot_od_table;
 	int ret = 0;
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	ret = smu_v11_0_set_default_od_settings(smu, initialize, sizeof(OverDriveTable_t));
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
index 0045b54b19ed..d5a3487ccfac 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
@@ -57,7 +57,7 @@ static int smu_v11_0_send_msg_without_waiting(struct smu_context *smu,
 					      uint16_t msg)
 {
 	struct amdgpu_device *adev = smu->adev;
-	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_66, msg);
+	WREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_66, msg);
 	return 0;
 }
 
@@ -65,7 +65,7 @@ static int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg)
 {
 	struct amdgpu_device *adev = smu->adev;
 
-	*arg = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82);
+	*arg = RREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_82);
 	return 0;
 }
 
@@ -75,7 +75,7 @@ static int smu_v11_0_wait_for_response(struct smu_context *smu)
 	uint32_t cur_value, i, timeout = adev->usec_timeout * 10;
 
 	for (i = 0; i < timeout; i++) {
-		cur_value = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90);
+		cur_value = RREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_90);
 		if ((cur_value & MP1_C2PMSG_90__CONTENT_MASK) != 0)
 			return cur_value == 0x1 ? 0 : -EIO;
 
@@ -83,7 +83,10 @@ static int smu_v11_0_wait_for_response(struct smu_context *smu)
 	}
 
 	/* timeout means wrong logic */
-	return -ETIME;
+	if (i == timeout)
+		return -ETIME;
+
+	return RREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_90) == 0x1 ? 0 : -EIO;
 }
 
 int
@@ -107,9 +110,9 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu,
 		goto out;
 	}
 
-	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0);
+	WREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_90, 0);
 
-	WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82, param);
+	WREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_82, param);
 
 	smu_v11_0_send_msg_without_waiting(smu, (uint16_t)index);
 
@@ -119,6 +122,7 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu,
 		       smu_get_message_name(smu, msg), index, param, ret);
 		goto out;
 	}
+
 	if (read_arg) {
 		ret = smu_v11_0_read_arg(smu, read_arg);
 		if (ret) {
@@ -728,8 +732,9 @@ int smu_v11_0_parse_pptable(struct smu_context *smu)
 	struct smu_table_context *table_context = &smu->smu_table;
 	struct smu_table *table = &table_context->tables[SMU_TABLE_PPTABLE];
 
+	/* during TDR we need to free and alloc the pptable */
 	if (table_context->driver_pptable)
-		return -EINVAL;
+		kfree(table_context->driver_pptable);
 
 	table_context->driver_pptable = kzalloc(table->size, GFP_KERNEL);
 
@@ -769,6 +774,9 @@ int smu_v11_0_set_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk)
 {
 	int ret;
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	ret = smu_send_smc_msg_with_param(smu,
 					  SMU_MSG_SetMinDeepSleepDcefclk, clk, NULL);
 	if (ret)
@@ -812,6 +820,9 @@ int smu_v11_0_set_tool_table_location(struct smu_context *smu)
 	int ret = 0;
 	struct smu_table *tool_table = &smu->smu_table.tables[SMU_TABLE_PMSTATUSLOG];
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	if (tool_table->mc_address) {
 		ret = smu_send_smc_msg_with_param(smu,
 				SMU_MSG_SetToolsDramAddrHigh,
@@ -831,6 +842,12 @@ int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count)
 {
 	int ret = 0;
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
+	if (!smu->pm_enabled)
+		return ret;
+
 	ret = smu_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count, NULL);
 	return ret;
 }
@@ -842,6 +859,9 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu)
 	int ret = 0;
 	uint32_t feature_mask[2];
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	mutex_lock(&feature->mutex);
 	if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || feature->feature_num < 64)
 		goto failed;
@@ -870,6 +890,9 @@ int smu_v11_0_get_enabled_mask(struct smu_context *smu,
 	struct smu_feature *feature = &smu->smu_feature;
 	int ret = 0;
 
+	if (amdgpu_sriov_vf(smu->adev) && !amdgpu_sriov_is_pp_one_vf(smu->adev))
+		return 0;
+
 	if (!feature_mask || num < 2)
 		return -EINVAL;
 
@@ -925,6 +948,12 @@ int smu_v11_0_notify_display_change(struct smu_context *smu)
 {
 	int ret = 0;
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
+	if (!smu->pm_enabled)
+		return ret;
+
 	if (smu_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT) &&
 	    smu->adev->gmc.vram_type == AMDGPU_VRAM_TYPE_HBM)
 		ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1, NULL);
@@ -1084,6 +1113,9 @@ int smu_v11_0_set_power_limit(struct smu_context *smu, uint32_t n)
 	int ret = 0;
 	uint32_t max_power_limit;
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	max_power_limit = smu_v11_0_get_max_power_limit(smu);
 
 	if (n > max_power_limit) {
@@ -1815,6 +1847,9 @@ int smu_v11_0_override_pcie_parameters(struct smu_context *smu)
 	uint32_t pcie_gen = 0, pcie_width = 0;
 	int ret;
 
+	if (amdgpu_sriov_vf(smu->adev))
+		return 0;
+
 	if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
 		pcie_gen = 3;
 	else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)