Merge branch 'drm-next-5.3' of git://people.freedesktop.org/~agd5f/linux into drm-next

New stuff for 5.3:
- Add new thermal sensors for vega asics
- Various RAS fixes
- Add sysfs interface for memory interface utilization
- Use HMM rather than mmu notifier for user pages
- Expose xgmi topology via kfd
- SR-IOV fixes
- Fixes for manual driver reload
- Add unique identifier for vega asics
- Clean up user fence handling with UVD/VCE/VCN blocks
- Convert DC to use core bpc attribute rather than a custom one
- Add GWS support for KFD
- Vega powerplay improvements
- Add CRC support for DCE 12
- SR-IOV support for new security policy
- Various cleanups

From: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

1 files changed, 75 insertions, 58 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 9f282e971197..0ffa6733f2b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -97,6 +97,28 @@ static const char *amdgpu_asic_name[] = {
 	"LAST",
 };
 
+/**
+ * DOC: pcie_replay_count
+ *
+ * The amdgpu driver provides a sysfs API for reporting the total number
+ * of PCIe replays (NAKs)
+ * The file pcie_replay_count is used for this and returns the total
+ * number of replays as a sum of the NAKs generated and NAKs received
+ */
+
+static ssize_t amdgpu_device_get_pcie_replay_count(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = ddev->dev_private;
+	uint64_t cnt = amdgpu_asic_get_pcie_replay_count(adev);
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n", cnt);
+}
+
+static DEVICE_ATTR(pcie_replay_count, S_IRUGO,
+		amdgpu_device_get_pcie_replay_count, NULL);
+
 static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev);
 
 /**
@@ -910,8 +932,10 @@ def_value:
  * Validates certain module parameters and updates
  * the associated values used by the driver (all asics).
  */
-static void amdgpu_device_check_arguments(struct amdgpu_device *adev)
+static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
 {
+	int ret = 0;
+
 	if (amdgpu_sched_jobs < 4) {
 		dev_warn(adev->dev, "sched jobs (%d) must be at least 4\n",
 			 amdgpu_sched_jobs);
@@ -956,12 +980,15 @@ static void amdgpu_device_check_arguments(struct amdgpu_device *adev)
 		amdgpu_vram_page_split = 1024;
 	}
 
-	if (amdgpu_lockup_timeout == 0) {
-		dev_warn(adev->dev, "lockup_timeout msut be > 0, adjusting to 10000\n");
-		amdgpu_lockup_timeout = 10000;
+	ret = amdgpu_device_get_job_timeout_settings(adev);
+	if (ret) {
+		dev_err(adev->dev, "invalid lockup_timeout parameter syntax\n");
+		return ret;
 	}
 
 	adev->firmware.load_type = amdgpu_ucode_get_load_type(adev, amdgpu_fw_load_type);
+
+	return ret;
 }
 
 /**
@@ -1505,12 +1532,26 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 		r = amdgpu_virt_request_full_gpu(adev, true);
 		if (r)
 			return -EAGAIN;
+
+		/* query the reg access mode at the very beginning */
+		amdgpu_virt_init_reg_access_mode(adev);
 	}
 
 	adev->pm.pp_feature = amdgpu_pp_feature_mask;
 	if (amdgpu_sriov_vf(adev))
 		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
 
+	/* Read BIOS */
+	if (!amdgpu_get_bios(adev))
+		return -EINVAL;
+
+	r = amdgpu_atombios_init(adev);
+	if (r) {
+		dev_err(adev->dev, "amdgpu_atombios_init failed\n");
+		amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0);
+		return r;
+	}
+
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		if ((amdgpu_ip_block_mask & (1 << i)) == 0) {
 			DRM_ERROR("disabled ip block: %d <%s>\n",
@@ -1550,6 +1591,7 @@ static int amdgpu_device_ip_hw_init_phase1(struct amdgpu_device *adev)
 		if (adev->ip_blocks[i].status.hw)
 			continue;
 		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON ||
+		    (amdgpu_sriov_vf(adev) && (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP)) ||
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) {
 			r = adev->ip_blocks[i].version->funcs->hw_init(adev);
 			if (r) {
@@ -2473,7 +2515,9 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->lock_reset);
 	mutex_init(&adev->virt.dpm_mutex);
 
-	amdgpu_device_check_arguments(adev);
+	r = amdgpu_device_check_arguments(adev);
+	if (r)
+		return r;
 
 	spin_lock_init(&adev->mmio_idx_lock);
 	spin_lock_init(&adev->smc_idx_lock);
@@ -2558,19 +2602,6 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		goto fence_driver_init;
 	}
 
-	/* Read BIOS */
-	if (!amdgpu_get_bios(adev)) {
-		r = -EINVAL;
-		goto failed;
-	}
-
-	r = amdgpu_atombios_init(adev);
-	if (r) {
-		dev_err(adev->dev, "amdgpu_atombios_init failed\n");
-		amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0);
-		goto failed;
-	}
-
 	/* detect if we are with an SRIOV vbios */
 	amdgpu_device_detect_sriov_bios(adev);
 
@@ -2672,6 +2703,10 @@ fence_driver_init:
 	if (r)
 		DRM_ERROR("registering pm debugfs failed (%d).\n", r);
 
+	r = amdgpu_ucode_sysfs_init(adev);
+	if (r)
+		DRM_ERROR("Creating firmware sysfs failed (%d).\n", r);
+
 	r = amdgpu_debugfs_gem_init(adev);
 	if (r)
 		DRM_ERROR("registering gem debugfs failed (%d).\n", r);
@@ -2712,7 +2747,13 @@ fence_driver_init:
 	}
 
 	/* must succeed. */
-	amdgpu_ras_post_init(adev);
+	amdgpu_ras_resume(adev);
+
+	r = device_create_file(adev->dev, &dev_attr_pcie_replay_count);
+	if (r) {
+		dev_err(adev->dev, "Could not create pcie_replay_count");
+		return r;
+	}
 
 	return 0;
 
@@ -2777,6 +2818,8 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	adev->rmmio = NULL;
 	amdgpu_device_doorbell_fini(adev);
 	amdgpu_debugfs_regs_cleanup(adev);
+	device_remove_file(adev->dev, &dev_attr_pcie_replay_count);
+	amdgpu_ucode_sysfs_fini(adev);
 }
 
 
@@ -2857,6 +2900,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool suspend, bool fbcon)
 
 	amdgpu_amdkfd_suspend(adev);
 
+	amdgpu_ras_suspend(adev);
+
 	r = amdgpu_device_ip_suspend_phase1(adev);
 
 	/* evict vram memory */
@@ -2977,6 +3022,8 @@ int amdgpu_device_resume(struct drm_device *dev, bool resume, bool fbcon)
 
 	drm_kms_helper_poll_enable(dev);
 
+	amdgpu_ras_resume(adev);
+
 	/*
 	 * Most of the connector probing functions try to acquire runtime pm
 	 * refs to ensure that the GPU is powered on when connector polling is
@@ -3455,6 +3502,13 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 				if (vram_lost)
 					amdgpu_device_fill_reset_magic(tmp_adev);
 
+				r = amdgpu_device_ip_late_init(tmp_adev);
+				if (r)
+					goto out;
+
+				/* must succeed. */
+				amdgpu_ras_resume(tmp_adev);
+
 				/* Update PSP FW topology after reset */
 				if (hive && tmp_adev->gmc.xgmi.num_physical_nodes > 1)
 					r = amdgpu_xgmi_update_topology(hive, tmp_adev);
@@ -3695,43 +3749,6 @@ skip_hw_reset:
 	return r;
 }
 
-static void amdgpu_device_get_min_pci_speed_width(struct amdgpu_device *adev,
-						  enum pci_bus_speed *speed,
-						  enum pcie_link_width *width)
-{
-	struct pci_dev *pdev = adev->pdev;
-	enum pci_bus_speed cur_speed;
-	enum pcie_link_width cur_width;
-	u32 ret = 1;
-
-	*speed = PCI_SPEED_UNKNOWN;
-	*width = PCIE_LNK_WIDTH_UNKNOWN;
-
-	while (pdev) {
-		cur_speed = pcie_get_speed_cap(pdev);
-		cur_width = pcie_get_width_cap(pdev);
-		ret = pcie_bandwidth_available(adev->pdev, NULL,
-						       NULL, &cur_width);
-		if (!ret)
-			cur_width = PCIE_LNK_WIDTH_RESRV;
-
-		if (cur_speed != PCI_SPEED_UNKNOWN) {
-			if (*speed == PCI_SPEED_UNKNOWN)
-				*speed = cur_speed;
-			else if (cur_speed < *speed)
-				*speed = cur_speed;
-		}
-
-		if (cur_width != PCIE_LNK_WIDTH_UNKNOWN) {
-			if (*width == PCIE_LNK_WIDTH_UNKNOWN)
-				*width = cur_width;
-			else if (cur_width < *width)
-				*width = cur_width;
-		}
-		pdev = pci_upstream_bridge(pdev);
-	}
-}
-
 /**
  * amdgpu_device_get_pcie_info - fence pcie info about the PCIE slot
  *
@@ -3765,8 +3782,8 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 	if (adev->pm.pcie_gen_mask && adev->pm.pcie_mlw_mask)
 		return;
 
-	amdgpu_device_get_min_pci_speed_width(adev, &platform_speed_cap,
-					      &platform_link_width);
+	pcie_bandwidth_available(adev->pdev, NULL,
+				 &platform_speed_cap, &platform_link_width);
 
 	if (adev->pm.pcie_gen_mask == 0) {
 		/* asic caps */