Merge tag 'amd-drm-next-5.7-2020-03-26' of git://people.freedesktop.org/~agd5f/linux into drm-next

airlied · airlied · commit c0ca5437c509 · 2020-03-30T15:21:03.000+10:00
amd-drm-next-5.7-2020-03-26: amdgpu: - Remove a dpm quirk that is not necessary - Fix handling of AC/DC mode in newer SMU firmwares on navi - SR-IOV fixes - RAS fixes scheduler: - Fix a race condition radeon: - Remove a dpm quirk that is not necessary Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexdeucher@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200326155310.5486-1-alexander.deucher@amd.com
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2742,6 +2742,9 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
 
 		if (adev->asic_reset_res)
 			goto fail;
+
+		if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count)
+			adev->mmhub.funcs->reset_ras_error_count(adev);
 	} else {
 
 		task_barrier_full(&hive->tb);
@@ -3910,8 +3913,15 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 		}
 	}
 
-	if (!r && amdgpu_ras_intr_triggered())
+	if (!r && amdgpu_ras_intr_triggered()) {
+		list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
+			if (tmp_adev->mmhub.funcs &&
+			    tmp_adev->mmhub.funcs->reset_ras_error_count)
+				tmp_adev->mmhub.funcs->reset_ras_error_count(tmp_adev);
+		}
+
 		amdgpu_ras_intr_cleared();
+	}
 
 	list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
 		if (need_full_reset) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -92,6 +92,9 @@ void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev)
 		if (adev->powerplay.pp_funcs->enable_bapm)
 			amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power);
 		mutex_unlock(&adev->pm.mutex);
+
+		if (is_support_sw_smu(adev))
+			smu_set_ac_dc(&adev->smu);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -159,10 +159,6 @@ static int psp_sw_fini(void *handle)
 	adev->psp.sos_fw = NULL;
 	release_firmware(adev->psp.asd_fw);
 	adev->psp.asd_fw = NULL;
-	if (adev->psp.cap_fw) {
-		release_firmware(adev->psp.cap_fw);
-		adev->psp.cap_fw = NULL;
-	}
 	if (adev->psp.ta_fw) {
 		release_firmware(adev->psp.ta_fw);
 		adev->psp.ta_fw = NULL;
@@ -250,7 +246,7 @@ psp_cmd_submit_buf(struct psp_context *psp,
 		DRM_WARN("psp command (0x%X) failed and response status is (0x%X)\n",
 			 psp->cmd_buf_mem->cmd_id,
 			 psp->cmd_buf_mem->resp.status);
-		if ((ucode->ucode_id == AMDGPU_UCODE_ID_CAP) || !timeout) {
+		if (!timeout) {
 			mutex_unlock(&psp->mutex);
 			return -EINVAL;
 		}
@@ -1192,9 +1188,6 @@ static int psp_get_fw_type(struct amdgpu_firmware_info *ucode,
 			   enum psp_gfx_fw_type *type)
 {
 	switch (ucode->ucode_id) {
-	case AMDGPU_UCODE_ID_CAP:
-		*type = GFX_FW_TYPE_CAP;
-		break;
 	case AMDGPU_UCODE_ID_SDMA0:
 		*type = GFX_FW_TYPE_SDMA0;
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -252,9 +252,6 @@ struct psp_context
 	uint32_t			asd_ucode_size;
 	uint8_t				*asd_start_addr;
 
-	/* cap firmware */
-	const struct firmware		*cap_fw;
-
 	/* fence buffer */
 	struct amdgpu_bo		*fence_buf_bo;
 	uint64_t			fence_buf_mc_addr;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -281,6 +281,11 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *
 	struct ras_debug_if data;
 	int ret = 0;
 
+	if (amdgpu_ras_intr_triggered()) {
+		DRM_WARN("RAS WARN: error injection currently inaccessible\n");
+		return size;
+	}
+
 	ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data);
 	if (ret)
 		return -EINVAL;
@@ -394,6 +399,10 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev,
 		.head = obj->head,
 	};
 
+	if (amdgpu_ras_intr_triggered())
+		return snprintf(buf, PAGE_SIZE,
+				"Query currently inaccessible\n");
+
 	if (amdgpu_ras_error_query(obj->adev, &info))
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1840,9 +1840,11 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	 *The reserved vram for memory training must be pinned to the specified
 	 *place on the VRAM, so reserve it early.
 	 */
-	r = amdgpu_ttm_training_reserve_vram_init(adev);
-	if (r)
-		return r;
+	if (!amdgpu_sriov_vf(adev)) {
+		r = amdgpu_ttm_training_reserve_vram_init(adev);
+		if (r)
+			return r;
+	}
 
 	/* allocate memory as required for VGA
 	 * This is used for VGA emulation and pre-OS scanout buffers to
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
@@ -283,8 +283,7 @@ union amdgpu_firmware_header {
  * fw loading support
  */
 enum AMDGPU_UCODE_ID {
-	AMDGPU_UCODE_ID_CAP = 0, /* CAP must be the 1st fw to be loaded */
-	AMDGPU_UCODE_ID_SDMA0,
+	AMDGPU_UCODE_ID_SDMA0 = 0,
 	AMDGPU_UCODE_ID_SDMA1,
 	AMDGPU_UCODE_ID_SDMA2,
 	AMDGPU_UCODE_ID_SDMA3,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -1940,6 +1940,11 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev)
 		if (!amdgpu_sriov_vf(adev)) /* enable RLC SRM */
 			gfx_v10_0_rlc_enable_srm(adev);
 	} else {
+		if (amdgpu_sriov_vf(adev)) {
+			gfx_v10_0_init_csb(adev);
+			return 0;
+		}
+
 		adev->gfx.rlc.funcs->stop(adev);
 
 		/* disable CG */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4306,7 +4306,7 @@ static const struct soc15_reg_entry vgpr_init_regs_arcturus[] = {
    { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_X), 0x40 },
    { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_Y), 4 },
    { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_NUM_THREAD_Z), 1 },
-   { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC1), 0x81 },
+   { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC1), 0xbf },
    { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_PGM_RSRC2), 0x400000 },  /* 64KB LDS */
    { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE0), 0xffffffff },
    { SOC15_REG_ENTRY(GC, 0, mmCOMPUTE_STATIC_THREAD_MGMT_SE1), 0xffffffff },
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
@@ -246,7 +246,6 @@ enum psp_gfx_fw_type {
 	GFX_FW_TYPE_SDMA6                           = 56,   /* SDMA6                    MI      */
 	GFX_FW_TYPE_SDMA7                           = 57,   /* SDMA7                    MI      */
 	GFX_FW_TYPE_VCN1                            = 58,   /* VCN1                     MI      */
-	GFX_FW_TYPE_CAP                             = 62,   /* CAP_FW                   VG      */
 	GFX_FW_TYPE_MAX
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
@@ -44,7 +44,6 @@
 
 MODULE_FIRMWARE("amdgpu/vega10_sos.bin");
 MODULE_FIRMWARE("amdgpu/vega10_asd.bin");
-MODULE_FIRMWARE("amdgpu/vega10_cap.bin");
 MODULE_FIRMWARE("amdgpu/vega12_sos.bin");
 MODULE_FIRMWARE("amdgpu/vega12_asd.bin");
 
@@ -64,7 +63,6 @@ static int psp_v3_1_init_microcode(struct psp_context *psp)
 	char fw_name[30];
 	int err = 0;
 	const struct psp_firmware_header_v1_0 *hdr;
-	struct amdgpu_firmware_info *info = NULL;
 
 	DRM_DEBUG("\n");
 
@@ -114,26 +112,6 @@ static int psp_v3_1_init_microcode(struct psp_context *psp)
 	adev->psp.asd_start_addr = (uint8_t *)hdr +
 				le32_to_cpu(hdr->header.ucode_array_offset_bytes);
 
-	if (amdgpu_sriov_vf(adev) && adev->asic_type == CHIP_VEGA10) {
-		snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_cap.bin",
-			 chip_name);
-		err = request_firmware(&adev->psp.cap_fw, fw_name, adev->dev);
-		if (err)
-			goto out;
-
-		err = amdgpu_ucode_validate(adev->psp.cap_fw);
-		if (err)
-			goto out;
-
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_CAP];
-		info->ucode_id = AMDGPU_UCODE_ID_CAP;
-		info->fw = adev->psp.cap_fw;
-		hdr = (const struct psp_firmware_header_v1_0 *)
-			      adev->psp.cap_fw->data;
-		adev->firmware.fw_size += ALIGN(
-			le32_to_cpu(hdr->header.ucode_size_bytes), PAGE_SIZE);
-	}
-
 	return 0;
 out:
 	if (err) {
@@ -144,8 +122,6 @@ static int psp_v3_1_init_microcode(struct psp_context *psp)
 		adev->psp.sos_fw = NULL;
 		release_firmware(adev->psp.asd_fw);
 		adev->psp.asd_fw = NULL;
-		release_firmware(adev->psp.cap_fw);
-		adev->psp.cap_fw = NULL;
 	}
 
 	return err;
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c
@@ -3439,7 +3439,6 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 
 	if (adev->asic_type == CHIP_HAINAN) {
 		if ((adev->pdev->revision == 0x81) ||
-		    (adev->pdev->revision == 0x83) ||
 		    (adev->pdev->revision == 0xC3) ||
 		    (adev->pdev->device == 0x6664) ||
 		    (adev->pdev->device == 0x6665) ||
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -524,8 +524,9 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params)
 
 	acrtc_state = to_dm_crtc_state(acrtc->base.state);
 
-	DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id,
-		      amdgpu_dm_vrr_active(acrtc_state));
+	DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d, planes:%d\n", acrtc->crtc_id,
+			 amdgpu_dm_vrr_active(acrtc_state),
+			 acrtc_state->active_planes);
 
 	amdgpu_dm_crtc_handle_crc_irq(&acrtc->base);
 	drm_crtc_handle_vblank(&acrtc->base);
@@ -545,7 +546,18 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params)
 			&acrtc_state->vrr_params.adjust);
 	}
 
-	if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED) {
+	/*
+	 * If there aren't any active_planes then DCH HUBP may be clock-gated.
+	 * In that case, pageflip completion interrupts won't fire and pageflip
+	 * completion events won't get delivered. Prevent this by sending
+	 * pending pageflip events from here if a flip is still pending.
+	 *
+	 * If any planes are enabled, use dm_pflip_high_irq() instead, to
+	 * avoid race conditions between flip programming and completion,
+	 * which could cause too early flip completion events.
+	 */
+	if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED &&
+	    acrtc_state->active_planes == 0) {
 		if (acrtc->event) {
 			drm_crtc_send_vblank_event(&acrtc->base, acrtc->event);
 			acrtc->event = NULL;
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -1154,6 +1154,21 @@ static int smu_smc_table_hw_init(struct smu_context *smu,
 				}
 			}
 		}
+
+		if (smu->ppt_funcs->set_power_source) {
+			/*
+			 * For Navi1X, manually switch it to AC mode as PMFW
+			 * may boot it with DC mode.
+			 */
+			if (adev->pm.ac_power)
+				ret = smu_set_power_source(smu, SMU_POWER_SOURCE_AC);
+			else
+				ret = smu_set_power_source(smu, SMU_POWER_SOURCE_DC);
+			if (ret) {
+				pr_err("Failed to switch to %s mode!\n", adev->pm.ac_power ? "AC" : "DC");
+				return ret;
+			}
+		}
 	}
 	if (adev->asic_type != CHIP_ARCTURUS) {
 		ret = smu_notify_display_change(smu);
@@ -2072,6 +2087,29 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu,
 	return 0;
 }
 
+int smu_set_ac_dc(struct smu_context *smu)
+{
+	int ret = 0;
+
+	/* controlled by firmware */
+	if (smu->dc_controlled_by_gpio)
+		return 0;
+
+	mutex_lock(&smu->mutex);
+	if (smu->ppt_funcs->set_power_source) {
+		if (smu->adev->pm.ac_power)
+			ret = smu_set_power_source(smu, SMU_POWER_SOURCE_AC);
+		else
+			ret = smu_set_power_source(smu, SMU_POWER_SOURCE_DC);
+		if (ret)
+			pr_err("Failed to switch to %s mode!\n",
+			       smu->adev->pm.ac_power ? "AC" : "DC");
+	}
+	mutex_unlock(&smu->mutex);
+
+	return ret;
+}
+
 const struct amd_ip_funcs smu_ip_funcs = {
 	.name = "smu",
 	.early_init = smu_early_init,
diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
@@ -408,6 +408,7 @@ struct smu_context
 	uint32_t smc_if_version;
 
 	bool uploading_custom_pp_table;
+	bool dc_controlled_by_gpio;
 };
 
 struct i2c_adapter;
@@ -570,6 +571,7 @@ struct pptable_funcs {
 	int (*override_pcie_parameters)(struct smu_context *smu);
 	uint32_t (*get_pptable_power_limit)(struct smu_context *smu);
 	int (*disable_umc_cdr_12gbps_workaround)(struct smu_context *smu);
+	int (*set_power_source)(struct smu_context *smu, enum smu_power_src_type power_src);
 };
 
 int smu_load_microcode(struct smu_context *smu);
@@ -718,6 +720,7 @@ int smu_get_dpm_level_range(struct smu_context *smu, enum smu_clk_type clk_type,
 enum amd_dpm_forced_level smu_get_performance_level(struct smu_context *smu);
 int smu_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level);
 int smu_set_display_count(struct smu_context *smu, uint32_t count);
+int smu_set_ac_dc(struct smu_context *smu);
 bool smu_clk_dpm_is_enabled(struct smu_context *smu, enum smu_clk_type clk_type);
 const char *smu_get_message_name(struct smu_context *smu, enum smu_message_type type);
 const char *smu_get_feature_name(struct smu_context *smu, enum smu_feature_mask feature);
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h
@@ -267,4 +267,7 @@ uint32_t smu_v11_0_get_max_power_limit(struct smu_context *smu);
 int smu_v11_0_set_performance_level(struct smu_context *smu,
 				    enum amd_dpm_forced_level level);
 
+int smu_v11_0_set_power_source(struct smu_context *smu,
+			       enum smu_power_src_type power_src);
+
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
@@ -347,7 +347,6 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
 				| FEATURE_MASK(FEATURE_DS_DCEFCLK_BIT)
 				| FEATURE_MASK(FEATURE_FW_DSTATE_BIT)
 				| FEATURE_MASK(FEATURE_BACO_BIT)
-				| FEATURE_MASK(FEATURE_ACDC_BIT)
 				| FEATURE_MASK(FEATURE_GFX_SS_BIT)
 				| FEATURE_MASK(FEATURE_APCC_DFLL_BIT)
 				| FEATURE_MASK(FEATURE_FW_CTF_BIT)
@@ -391,6 +390,9 @@ navi10_get_allowed_feature_mask(struct smu_context *smu,
 	if (smu->adev->pg_flags & AMD_PG_SUPPORT_JPEG)
 		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_JPEG_PG_BIT);
 
+	if (smu->dc_controlled_by_gpio)
+		*(uint64_t *)feature_mask |= FEATURE_MASK(FEATURE_ACDC_BIT);
+
 	/* disable DPM UCLK and DS SOCCLK on navi10 A0 secure board */
 	if (is_asic_secure(smu)) {
 		/* only for navi10 A0 */
@@ -525,6 +527,9 @@ static int navi10_store_powerplay_table(struct smu_context *smu)
 
 	table_context->thermal_controller_type = powerplay_table->thermal_controller_type;
 
+	if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_HARDWAREDC)
+		smu->dc_controlled_by_gpio = true;
+
 	mutex_lock(&smu_baco->mutex);
 	if (powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_BACO ||
 	    powerplay_table->platform_caps & SMU_11_0_PP_PLATFORM_CAP_MACO)
@@ -2369,6 +2374,7 @@ static const struct pptable_funcs navi10_ppt_funcs = {
 	.get_pptable_power_limit = navi10_get_pptable_power_limit,
 	.run_btc = navi10_run_btc,
 	.disable_umc_cdr_12gbps_workaround = navi10_disable_umc_cdr_12gbps_workaround,
+	.set_power_source = smu_v11_0_set_power_source,
 };
 
 void navi10_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/smu_internal.h b/drivers/gpu/drm/amd/powerplay/smu_internal.h
@@ -211,4 +211,7 @@ static inline int smu_send_smc_msg(struct smu_context *smu, enum smu_message_typ
 #define smu_disable_umc_cdr_12gbps_workaround(smu) \
 	((smu)->ppt_funcs->disable_umc_cdr_12gbps_workaround ? (smu)->ppt_funcs->disable_umc_cdr_12gbps_workaround((smu)) : 0)
 
+#define smu_set_power_source(smu, power_src) \
+	((smu)->ppt_funcs->set_power_source ? (smu)->ppt_funcs->set_power_source((smu), (power_src)) : 0)
+
 #endif
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c

Original file line number	Diff line number	Diff line change
`@@ -92,6 +92,9 @@ void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev)`
`92`	`92`	`if (adev->powerplay.pp_funcs->enable_bapm)`
`93`	`93`	`amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power);`
`94`	`94`	`mutex_unlock(&adev->pm.mutex);`
	`95`	`+`
	`96`	`+ if (is_support_sw_smu(adev))`
	`97`	`+ smu_set_ac_dc(&adev->smu);`
`95`	`98`	`}`
`96`	`99`	`}`
`97`	`100`