Merge branch 'pci/virtualization'

- Fix erroneous intel-iommu dependency on CONFIG_AMD_IOMMU (Bjorn
    Helgaas)

  - Move pci_prg_resp_pasid_required() to CONFIG_PCI_PRI (Bjorn Helgaas)

  - Allow VFs to use PRI (the PF PRI is shared by the VFs, but the code
    previously didn't recognize that) (Kuppuswamy Sathyanarayanan)

  - Allow VFs to use PASID (the PF PASID capability is shared by the VFs,
    but the code previously didn't recognize that) (Kuppuswamy
    Sathyanarayanan)

  - Disconnect PF and VF ATS enablement, since ATS in PFs and associated
    VFs can be enabled independently (Kuppuswamy Sathyanarayanan)

  - Cache PRI and PASID capability offsets (Kuppuswamy Sathyanarayanan)

  - Cache the PRI PRG Response PASID Required bit (Bjorn Helgaas)

  - Consolidate ATS declarations in linux/pci-ats.h (Krzysztof Wilczynski)

  - Remove unused PRI and PASID stubs (Bjorn Helgaas)

  - Removed unnecessary EXPORT_SYMBOL_GPL() from ATS, PRI, and PASID
    interfaces that are only used by built-in IOMMU drivers (Bjorn Helgaas)

  - Hide PRI and PASID state restoration functions used only inside the PCI
    core (Bjorn Helgaas)

  - Fix the UPDCR register address in the Intel ACS quirk (Steffen
    Liebergeld)

  - Add a DMA alias quirk for the Intel VCA NTB (Slawomir Pawlowski)

  - Serialize sysfs sriov_numvfs reads vs writes (Pierre Crégut)

  - Update Cavium ACS quirk for ThunderX2 and ThunderX3 (George Cherian)

  - Unify ACS quirk implementations (Bjorn Helgaas)

* pci/virtualization:
  PCI: Unify ACS quirk desired vs provided checking
  PCI: Make ACS quirk implementations more uniform
  PCI: Apply Cavium ACS quirk to ThunderX2 and ThunderX3
  PCI/IOV: Serialize sysfs sriov_numvfs reads vs writes
  PCI: Add DMA alias quirk for Intel VCA NTB
  PCI: Fix Intel ACS quirk UPDCR register address
  PCI/ATS: Make pci_restore_pri_state(), pci_restore_pasid_state() private
  PCI/ATS: Remove unnecessary EXPORT_SYMBOL_GPL()
  PCI/ATS: Remove unused PRI and PASID stubs
  PCI/ATS: Consolidate ATS declarations in linux/pci-ats.h
  PCI/ATS: Cache PRI PRG Response PASID Required bit
  PCI/ATS: Cache PASID Capability offset
  PCI/ATS: Cache PRI Capability offset
  PCI/ATS: Disable PF/VF ATS service independently
  PCI/ATS: Handle sharing of PF PASID Capability with all VFs
  PCI/ATS: Handle sharing of PF PRI Capability with all VFs
  PCI/ATS: Move pci_prg_resp_pasid_required() to CONFIG_PCI_PRI
  iommu/vt-d: Select PCI_PRI for INTEL_IOMMU_SVM

drivers/iommu/Kconfig

@@ -207,6 +207,7 @@ config INTEL_IOMMU_SVM
 	bool "Support for Shared Virtual Memory with Intel IOMMU"
 	depends on INTEL_IOMMU && X86
 	select PCI_PASID
+	select PCI_PRI
 	select MMU_NOTIFIER
 	help
 	  Shared Virtual Memory (SVM) provides a facility for devices

drivers/pci/ats.c

@@ -60,8 +60,6 @@ int pci_enable_ats(struct pci_dev *dev, int ps)
 		pdev = pci_physfn(dev);
 		if (pdev->ats_stu != ps)
 			return -EINVAL;
-
-		atomic_inc(&pdev->ats_ref_cnt);  /* count enabled VFs */
 	} else {
 		dev->ats_stu = ps;
 		ctrl |= PCI_ATS_CTRL_STU(dev->ats_stu - PCI_ATS_MIN_STU);
@@ -71,7 +69,6 @@ int pci_enable_ats(struct pci_dev *dev, int ps)
 	dev->ats_enabled = 1;
 	return 0;
 }
-EXPORT_SYMBOL_GPL(pci_enable_ats);
 
 /**
  * pci_disable_ats - disable the ATS capability
@@ -79,27 +76,17 @@ EXPORT_SYMBOL_GPL(pci_enable_ats);
  */
 void pci_disable_ats(struct pci_dev *dev)
 {
-	struct pci_dev *pdev;
 	u16 ctrl;
 
 	if (WARN_ON(!dev->ats_enabled))
 		return;
 
-	if (atomic_read(&dev->ats_ref_cnt))
-		return;		/* VFs still enabled */
-
-	if (dev->is_virtfn) {
-		pdev = pci_physfn(dev);
-		atomic_dec(&pdev->ats_ref_cnt);
-	}
-
 	pci_read_config_word(dev, dev->ats_cap + PCI_ATS_CTRL, &ctrl);
 	ctrl &= ~PCI_ATS_CTRL_ENABLE;
 	pci_write_config_word(dev, dev->ats_cap + PCI_ATS_CTRL, ctrl);
 
 	dev->ats_enabled = 0;
 }
-EXPORT_SYMBOL_GPL(pci_disable_ats);
 
 void pci_restore_ats_state(struct pci_dev *dev)
 {
@@ -113,7 +100,6 @@ void pci_restore_ats_state(struct pci_dev *dev)
 		ctrl |= PCI_ATS_CTRL_STU(dev->ats_stu - PCI_ATS_MIN_STU);
 	pci_write_config_word(dev, dev->ats_cap + PCI_ATS_CTRL, ctrl);
 }
-EXPORT_SYMBOL_GPL(pci_restore_ats_state);
 
 /**
  * pci_ats_queue_depth - query the ATS Invalidate Queue Depth
@@ -140,7 +126,6 @@ int pci_ats_queue_depth(struct pci_dev *dev)
 	pci_read_config_word(dev, dev->ats_cap + PCI_ATS_CAP, &cap);
 	return PCI_ATS_CAP_QDEP(cap) ? PCI_ATS_CAP_QDEP(cap) : PCI_ATS_MAX_QDEP;
 }
-EXPORT_SYMBOL_GPL(pci_ats_queue_depth);
 
 /**
  * pci_ats_page_aligned - Return Page Aligned Request bit status.
@@ -167,9 +152,22 @@ int pci_ats_page_aligned(struct pci_dev *pdev)
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(pci_ats_page_aligned);
 
 #ifdef CONFIG_PCI_PRI
+void pci_pri_init(struct pci_dev *pdev)
+{
+	u16 status;
+
+	pdev->pri_cap = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI);
+
+	if (!pdev->pri_cap)
+		return;
+
+	pci_read_config_word(pdev, pdev->pri_cap + PCI_PRI_STATUS, &status);
+	if (status & PCI_PRI_STATUS_PASID)
+		pdev->pasid_required = 1;
+}
+
 /**
  * pci_enable_pri - Enable PRI capability
  * @ pdev: PCI device structure
@@ -180,32 +178,41 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs)
 {
 	u16 control, status;
 	u32 max_requests;
-	int pos;
+	int pri = pdev->pri_cap;
+
+	/*
+	 * VFs must not implement the PRI Capability.  If their PF
+	 * implements PRI, it is shared by the VFs, so if the PF PRI is
+	 * enabled, it is also enabled for the VF.
+	 */
+	if (pdev->is_virtfn) {
+		if (pci_physfn(pdev)->pri_enabled)
+			return 0;
+		return -EINVAL;
+	}
 
 	if (WARN_ON(pdev->pri_enabled))
 		return -EBUSY;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI);
+	if (!pri)
-	if (!pos)
 		return -EINVAL;
 
-	pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status);
+	pci_read_config_word(pdev, pri + PCI_PRI_STATUS, &status);
 	if (!(status & PCI_PRI_STATUS_STOPPED))
 		return -EBUSY;
 
-	pci_read_config_dword(pdev, pos + PCI_PRI_MAX_REQ, &max_requests);
+	pci_read_config_dword(pdev, pri + PCI_PRI_MAX_REQ, &max_requests);
 	reqs = min(max_requests, reqs);
 	pdev->pri_reqs_alloc = reqs;
-	pci_write_config_dword(pdev, pos + PCI_PRI_ALLOC_REQ, reqs);
+	pci_write_config_dword(pdev, pri + PCI_PRI_ALLOC_REQ, reqs);
 
 	control = PCI_PRI_CTRL_ENABLE;
-	pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control);
+	pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control);
 
 	pdev->pri_enabled = 1;
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(pci_enable_pri);
 
 /**
  * pci_disable_pri - Disable PRI capability
@@ -216,18 +223,21 @@ EXPORT_SYMBOL_GPL(pci_enable_pri);
 void pci_disable_pri(struct pci_dev *pdev)
 {
 	u16 control;
-	int pos;
+	int pri = pdev->pri_cap;
+
+	/* VFs share the PF PRI */
+	if (pdev->is_virtfn)
+		return;
 
 	if (WARN_ON(!pdev->pri_enabled))
 		return;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI);
+	if (!pri)
-	if (!pos)
 		return;
 
-	pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control);
+	pci_read_config_word(pdev, pri + PCI_PRI_CTRL, &control);
 	control &= ~PCI_PRI_CTRL_ENABLE;
-	pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control);
+	pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control);
 
 	pdev->pri_enabled = 0;
 }
@@ -241,19 +251,20 @@ void pci_restore_pri_state(struct pci_dev *pdev)
 {
 	u16 control = PCI_PRI_CTRL_ENABLE;
 	u32 reqs = pdev->pri_reqs_alloc;
-	int pos;
+	int pri = pdev->pri_cap;
+
+	if (pdev->is_virtfn)
+		return;
 
 	if (!pdev->pri_enabled)
 		return;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI);
+	if (!pri)
-	if (!pos)
 		return;
 
-	pci_write_config_dword(pdev, pos + PCI_PRI_ALLOC_REQ, reqs);
+	pci_write_config_dword(pdev, pri + PCI_PRI_ALLOC_REQ, reqs);
-	pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control);
+	pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control);
 }
-EXPORT_SYMBOL_GPL(pci_restore_pri_state);
 
 /**
  * pci_reset_pri - Resets device's PRI state
@@ -265,24 +276,45 @@ EXPORT_SYMBOL_GPL(pci_restore_pri_state);
 int pci_reset_pri(struct pci_dev *pdev)
 {
 	u16 control;
-	int pos;
+	int pri = pdev->pri_cap;
+
+	if (pdev->is_virtfn)
+		return 0;
 
 	if (WARN_ON(pdev->pri_enabled))
 		return -EBUSY;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI);
+	if (!pri)
-	if (!pos)
 		return -EINVAL;
 
 	control = PCI_PRI_CTRL_RESET;
-	pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control);
+	pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control);
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(pci_reset_pri);
+
+/**
+ * pci_prg_resp_pasid_required - Return PRG Response PASID Required bit
+ *				 status.
+ * @pdev: PCI device structure
+ *
+ * Returns 1 if PASID is required in PRG Response Message, 0 otherwise.
+ */
+int pci_prg_resp_pasid_required(struct pci_dev *pdev)
+{
+	if (pdev->is_virtfn)
+		pdev = pci_physfn(pdev);
+
+	return pdev->pasid_required;
+}
 #endif /* CONFIG_PCI_PRI */
 
 #ifdef CONFIG_PCI_PASID
+void pci_pasid_init(struct pci_dev *pdev)
+{
+	pdev->pasid_cap = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID);
+}
+
 /**
  * pci_enable_pasid - Enable the PASID capability
  * @pdev: PCI device structure
@@ -295,7 +327,17 @@ EXPORT_SYMBOL_GPL(pci_reset_pri);
 int pci_enable_pasid(struct pci_dev *pdev, int features)
 {
 	u16 control, supported;
-	int pos;
+	int pasid = pdev->pasid_cap;
+
+	/*
+	 * VFs must not implement the PASID Capability, but if a PF
+	 * supports PASID, its VFs share the PF PASID configuration.
+	 */
+	if (pdev->is_virtfn) {
+		if (pci_physfn(pdev)->pasid_enabled)
+			return 0;
+		return -EINVAL;
+	}
 
 	if (WARN_ON(pdev->pasid_enabled))
 		return -EBUSY;
@@ -303,11 +345,10 @@ int pci_enable_pasid(struct pci_dev *pdev, int features)
 	if (!pdev->eetlp_prefix_path)
 		return -EINVAL;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID);
+	if (!pasid)
-	if (!pos)
 		return -EINVAL;
 
-	pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported);
+	pci_read_config_word(pdev, pasid + PCI_PASID_CAP, &supported);
 	supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV;
 
 	/* User wants to enable anything unsupported? */
@@ -317,13 +358,12 @@ int pci_enable_pasid(struct pci_dev *pdev, int features)
 	control = PCI_PASID_CTRL_ENABLE | features;
 	pdev->pasid_features = features;
 
-	pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control);
+	pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control);
 
 	pdev->pasid_enabled = 1;
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(pci_enable_pasid);
 
 /**
  * pci_disable_pasid - Disable the PASID capability
@@ -332,20 +372,22 @@ EXPORT_SYMBOL_GPL(pci_enable_pasid);
 void pci_disable_pasid(struct pci_dev *pdev)
 {
 	u16 control = 0;
-	int pos;
+	int pasid = pdev->pasid_cap;
+
+	/* VFs share the PF PASID configuration */
+	if (pdev->is_virtfn)
+		return;
 
 	if (WARN_ON(!pdev->pasid_enabled))
 		return;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID);
+	if (!pasid)
-	if (!pos)
 		return;
 
-	pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control);
+	pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control);
 
 	pdev->pasid_enabled = 0;
 }
-EXPORT_SYMBOL_GPL(pci_disable_pasid);
 
 /**
  * pci_restore_pasid_state - Restore PASID capabilities
@@ -354,19 +396,20 @@ EXPORT_SYMBOL_GPL(pci_disable_pasid);
 void pci_restore_pasid_state(struct pci_dev *pdev)
 {
 	u16 control;
-	int pos;
+	int pasid = pdev->pasid_cap;
+
+	if (pdev->is_virtfn)
+		return;
 
 	if (!pdev->pasid_enabled)
 		return;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID);
+	if (!pasid)
-	if (!pos)
 		return;
 
 	control = PCI_PASID_CTRL_ENABLE | pdev->pasid_features;
-	pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control);
+	pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control);
 }
-EXPORT_SYMBOL_GPL(pci_restore_pasid_state);
 
 /**
  * pci_pasid_features - Check which PASID features are supported
@@ -381,49 +424,20 @@ EXPORT_SYMBOL_GPL(pci_restore_pasid_state);
 int pci_pasid_features(struct pci_dev *pdev)
 {
 	u16 supported;
-	int pos;
+	int pasid = pdev->pasid_cap;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID);
+	if (pdev->is_virtfn)
-	if (!pos)
+		pdev = pci_physfn(pdev);
+
+	if (!pasid)
 		return -EINVAL;
 
-	pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported);
+	pci_read_config_word(pdev, pasid + PCI_PASID_CAP, &supported);
 
 	supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV;
 
 	return supported;
 }
-EXPORT_SYMBOL_GPL(pci_pasid_features);
-
-/**
- * pci_prg_resp_pasid_required - Return PRG Response PASID Required bit
- *				 status.
- * @pdev: PCI device structure
- *
- * Returns 1 if PASID is required in PRG Response Message, 0 otherwise.
- *
- * Even though the PRG response PASID status is read from PRI Status
- * Register, since this API will mainly be used by PASID users, this
- * function is defined within #ifdef CONFIG_PCI_PASID instead of
- * CONFIG_PCI_PRI.
- */
-int pci_prg_resp_pasid_required(struct pci_dev *pdev)
-{
-	u16 status;
-	int pos;
-
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI);
-	if (!pos)
-		return 0;
-
-	pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status);
-
-	if (status & PCI_PRI_STATUS_PASID)
-		return 1;
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required);
 
 #define PASID_NUMBER_SHIFT	8
 #define PASID_NUMBER_MASK	(0x1f << PASID_NUMBER_SHIFT)
@@ -437,17 +451,18 @@ EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required);
 int pci_max_pasids(struct pci_dev *pdev)
 {
 	u16 supported;
-	int pos;
+	int pasid = pdev->pasid_cap;
 
-	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID);
+	if (pdev->is_virtfn)
-	if (!pos)
+		pdev = pci_physfn(pdev);
+
+	if (!pasid)
 		return -EINVAL;
 
-	pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported);
+	pci_read_config_word(pdev, pasid + PCI_PASID_CAP, &supported);
 
 	supported = (supported & PASID_NUMBER_MASK) >> PASID_NUMBER_SHIFT;
 
 	return (1 << supported);
 }
-EXPORT_SYMBOL_GPL(pci_max_pasids);
 #endif /* CONFIG_PCI_PASID */

drivers/pci/iov.c

@@ -254,8 +254,14 @@ static ssize_t sriov_numvfs_show(struct device *dev,
 				 char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
+	u16 num_vfs;
 
-	return sprintf(buf, "%u\n", pdev->sriov->num_VFs);
+	/* Serialize vs sriov_numvfs_store() so readers see valid num_VFs */
+	device_lock(&pdev->dev);
+	num_vfs = pdev->sriov->num_VFs;
+	device_unlock(&pdev->dev);
+
+	return sprintf(buf, "%u\n", num_vfs);
 }
 
 /*

drivers/pci/pci.h

@@ -459,6 +459,22 @@ static inline void pci_ats_init(struct pci_dev *d) { }
 static inline void pci_restore_ats_state(struct pci_dev *dev) { }
 #endif /* CONFIG_PCI_ATS */
 
+#ifdef CONFIG_PCI_PRI
+void pci_pri_init(struct pci_dev *dev);
+void pci_restore_pri_state(struct pci_dev *pdev);
+#else
+static inline void pci_pri_init(struct pci_dev *dev) { }
+static inline void pci_restore_pri_state(struct pci_dev *pdev) { }
+#endif
+
+#ifdef CONFIG_PCI_PASID
+void pci_pasid_init(struct pci_dev *dev);
+void pci_restore_pasid_state(struct pci_dev *pdev);
+#else
+static inline void pci_pasid_init(struct pci_dev *dev) { }
+static inline void pci_restore_pasid_state(struct pci_dev *pdev) { }
+#endif
+
 #ifdef CONFIG_PCI_IOV
 int pci_iov_init(struct pci_dev *dev);
 void pci_iov_release(struct pci_dev *dev);

drivers/pci/probe.c

@@ -2335,6 +2335,12 @@ static void pci_init_capabilities(struct pci_dev *dev)
 	/* Address Translation Services */
 	pci_ats_init(dev);
 
+	/* Page Request Interface */
+	pci_pri_init(dev);
+
+	/* Process Address Space ID */
+	pci_pasid_init(dev);
+
 	/* Enable ACS P2P upstream forwarding */
 	pci_enable_acs(dev);
 

drivers/pci/quirks.c

@@ -4080,6 +4080,40 @@ static void quirk_mic_x200_dma_alias(struct pci_dev *pdev)
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2260, quirk_mic_x200_dma_alias);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2264, quirk_mic_x200_dma_alias);
 
+/*
+ * Intel Visual Compute Accelerator (VCA) is a family of PCIe add-in devices
+ * exposing computational units via Non Transparent Bridges (NTB, PEX 87xx).
+ *
+ * Similarly to MIC x200, we need to add DMA aliases to allow buffer access
+ * when IOMMU is enabled.  These aliases allow computational unit access to
+ * host memory.  These aliases mark the whole VCA device as one IOMMU
+ * group.
+ *
+ * All possible slot numbers (0x20) are used, since we are unable to tell
+ * what slot is used on other side.  This quirk is intended for both host
+ * and computational unit sides.  The VCA devices have up to five functions
+ * (four for DMA channels and one additional).
+ */
+static void quirk_pex_vca_alias(struct pci_dev *pdev)
+{
+	const unsigned int num_pci_slots = 0x20;
+	unsigned int slot;
+
+	for (slot = 0; slot < num_pci_slots; slot++) {
+		pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x0));
+		pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x1));
+		pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x2));
+		pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x3));
+		pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x4));
+	}
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2954, quirk_pex_vca_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2955, quirk_pex_vca_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2956, quirk_pex_vca_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2958, quirk_pex_vca_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2959, quirk_pex_vca_alias);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x295A, quirk_pex_vca_alias);
+
 /*
  * The IOMMU and interrupt controller on Broadcom Vulcan/Cavium ThunderX2 are
  * associated not at the root bus, but at a bridge below. This quirk avoids
@@ -4262,6 +4296,24 @@ static void quirk_chelsio_T5_disable_root_port_attributes(struct pci_dev *pdev)
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_CHELSIO, PCI_ANY_ID,
 			 quirk_chelsio_T5_disable_root_port_attributes);
 
+/*
+ * pci_acs_ctrl_enabled - compare desired ACS controls with those provided
+ *			  by a device
+ * @acs_ctrl_req: Bitmask of desired ACS controls
+ * @acs_ctrl_ena: Bitmask of ACS controls enabled or provided implicitly by
+ *		  the hardware design
+ *
+ * Return 1 if all ACS controls in the @acs_ctrl_req bitmask are included
+ * in @acs_ctrl_ena, i.e., the device provides all the access controls the
+ * caller desires.  Return 0 otherwise.
+ */
+static int pci_acs_ctrl_enabled(u16 acs_ctrl_req, u16 acs_ctrl_ena)
+{
+	if ((acs_ctrl_req & acs_ctrl_ena) == acs_ctrl_req)
+		return 1;
+	return 0;
+}
+
 /*
  * AMD has indicated that the devices below do not support peer-to-peer
  * in any system where they are found in the southbridge with an AMD
@@ -4305,7 +4357,7 @@ static int pci_quirk_amd_sb_acs(struct pci_dev *dev, u16 acs_flags)
 	/* Filter out flags not applicable to multifunction */
 	acs_flags &= (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_EC | PCI_ACS_DT);
 
-	return acs_flags & ~(PCI_ACS_RR | PCI_ACS_CR) ? 0 : 1;
+	return pci_acs_ctrl_enabled(acs_flags, PCI_ACS_RR | PCI_ACS_CR);
 #else
 	return -ENODEV;
 #endif
@@ -4313,33 +4365,38 @@ static int pci_quirk_amd_sb_acs(struct pci_dev *dev, u16 acs_flags)
 
 static bool pci_quirk_cavium_acs_match(struct pci_dev *dev)
 {
+	if (!pci_is_pcie(dev) || pci_pcie_type(dev) != PCI_EXP_TYPE_ROOT_PORT)
+		return false;
+
+	switch (dev->device) {
 	/*
-	 * Effectively selects all downstream ports for whole ThunderX 1
+	 * Effectively selects all downstream ports for whole ThunderX1
-	 * family by 0xf800 mask (which represents 8 SoCs), while the lower
+	 * (which represents 8 SoCs).
-	 * bits of device ID are used to indicate which subdevice is used
-	 * within the SoC.
 	 */
-	return (pci_is_pcie(dev) &&
+	case 0xa000 ... 0xa7ff: /* ThunderX1 */
-		(pci_pcie_type(dev) == PCI_EXP_TYPE_ROOT_PORT) &&
+	case 0xaf84:  /* ThunderX2 */
-		((dev->device & 0xf800) == 0xa000));
+	case 0xb884:  /* ThunderX3 */
+		return true;
+	default:
+		return false;
+	}
 }
 
 static int pci_quirk_cavium_acs(struct pci_dev *dev, u16 acs_flags)
 {
+	if (!pci_quirk_cavium_acs_match(dev))
+		return -ENOTTY;
+
 	/*
-	 * Cavium root ports don't advertise an ACS capability.  However,
+	 * Cavium Root Ports don't advertise an ACS capability.  However,
 	 * the RTL internally implements similar protection as if ACS had
-	 * Request Redirection, Completion Redirection, Source Validation,
+	 * Source Validation, Request Redirection, Completion Redirection,
 	 * and Upstream Forwarding features enabled.  Assert that the
 	 * hardware implements and enables equivalent ACS functionality for
 	 * these flags.
 	 */
-	acs_flags &= ~(PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_SV | PCI_ACS_UF);
+	return pci_acs_ctrl_enabled(acs_flags,
-
+		PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
-	if (!pci_quirk_cavium_acs_match(dev))
-		return -ENOTTY;
-
-	return acs_flags ? 0 : 1;
 }
 
 static int pci_quirk_xgene_acs(struct pci_dev *dev, u16 acs_flags)
@@ -4349,13 +4406,12 @@ static int pci_quirk_xgene_acs(struct pci_dev *dev, u16 acs_flags)
 	 * transactions with others, allowing masking out these bits as if they
 	 * were unimplemented in the ACS capability.
 	 */
-	acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
+	return pci_acs_ctrl_enabled(acs_flags,
-
+		PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
-	return acs_flags ? 0 : 1;
 }
 
 /*
- * Many Intel PCH root ports do provide ACS-like features to disable peer
+ * Many Intel PCH Root Ports do provide ACS-like features to disable peer
  * transactions and validate bus numbers in requests, but do not provide an
  * actual PCIe ACS capability.  This is the list of device IDs known to fall
  * into that category as provided by Intel in Red Hat bugzilla 1037684.
@@ -4403,37 +4459,32 @@ static bool pci_quirk_intel_pch_acs_match(struct pci_dev *dev)
 	return false;
 }
 
-#define INTEL_PCH_ACS_FLAGS (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_SV)
-
 static int pci_quirk_intel_pch_acs(struct pci_dev *dev, u16 acs_flags)
 {
-	u16 flags = dev->dev_flags & PCI_DEV_FLAGS_ACS_ENABLED_QUIRK ?
-		    INTEL_PCH_ACS_FLAGS : 0;
-
 	if (!pci_quirk_intel_pch_acs_match(dev))
 		return -ENOTTY;
 
-	return acs_flags & ~flags ? 0 : 1;
+	if (dev->dev_flags & PCI_DEV_FLAGS_ACS_ENABLED_QUIRK)
+		return pci_acs_ctrl_enabled(acs_flags,
+			PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
+
+	return pci_acs_ctrl_enabled(acs_flags, 0);
 }
 
 /*
- * These QCOM root ports do provide ACS-like features to disable peer
+ * These QCOM Root Ports do provide ACS-like features to disable peer
  * transactions and validate bus numbers in requests, but do not provide an
  * actual PCIe ACS capability.  Hardware supports source validation but it
  * will report the issue as Completer Abort instead of ACS Violation.
- * Hardware doesn't support peer-to-peer and each root port is a root
+ * Hardware doesn't support peer-to-peer and each Root Port is a Root
- * complex with unique segment numbers.  It is not possible for one root
+ * Complex with unique segment numbers.  It is not possible for one Root
- * port to pass traffic to another root port.  All PCIe transactions are
+ * Port to pass traffic to another Root Port.  All PCIe transactions are
- * terminated inside the root port.
+ * terminated inside the Root Port.
  */
 static int pci_quirk_qcom_rp_acs(struct pci_dev *dev, u16 acs_flags)
 {
-	u16 flags = (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_SV);
+	return pci_acs_ctrl_enabled(acs_flags,
-	int ret = acs_flags & ~flags ? 0 : 1;
+		PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
-
-	pci_info(dev, "Using QCOM ACS Quirk (%d)\n", ret);
-
-	return ret;
 }
 
 static int pci_quirk_al_acs(struct pci_dev *dev, u16 acs_flags)
@@ -4534,7 +4585,7 @@ static int pci_quirk_intel_spt_pch_acs(struct pci_dev *dev, u16 acs_flags)
 
 	pci_read_config_dword(dev, pos + INTEL_SPT_ACS_CTRL, &ctrl);
 
-	return acs_flags & ~ctrl ? 0 : 1;
+	return pci_acs_ctrl_enabled(acs_flags, ctrl);
 }
 
 static int pci_quirk_mf_endpoint_acs(struct pci_dev *dev, u16 acs_flags)
@@ -4548,10 +4599,9 @@ static int pci_quirk_mf_endpoint_acs(struct pci_dev *dev, u16 acs_flags)
 	 * perform peer-to-peer with other functions, allowing us to mask out
 	 * these bits as if they were unimplemented in the ACS capability.
 	 */
-	acs_flags &= ~(PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR |
+	return pci_acs_ctrl_enabled(acs_flags,
-		       PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_DT);
+		PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR |
-
+		PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_DT);
-	return acs_flags ? 0 : 1;
 }
 
 static int pci_quirk_brcm_acs(struct pci_dev *dev, u16 acs_flags)
@@ -4562,9 +4612,8 @@ static int pci_quirk_brcm_acs(struct pci_dev *dev, u16 acs_flags)
 	 * Allow each Root Port to be in a separate IOMMU group by masking
 	 * SV/RR/CR/UF bits.
 	 */
-	acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
+	return pci_acs_ctrl_enabled(acs_flags,
-
+		PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF);
-	return acs_flags ? 0 : 1;
 }
 
 static const struct pci_dev_acs_enabled {
@@ -4666,6 +4715,17 @@ static const struct pci_dev_acs_enabled {
 	{ 0 }
 };
 
+/*
+ * pci_dev_specific_acs_enabled - check whether device provides ACS controls
+ * @dev:	PCI device
+ * @acs_flags:	Bitmask of desired ACS controls
+ *
+ * Returns:
+ *   -ENOTTY:	No quirk applies to this device; we can't tell whether the
+ *		device provides the desired controls
+ *   0:		Device does not provide all the desired controls
+ *   >0:	Device provides all the controls in @acs_flags
+ */
 int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags)
 {
 	const struct pci_dev_acs_enabled *i;
@@ -4706,7 +4766,7 @@ int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags)
 #define INTEL_BSPR_REG_BPPD  (1 << 9)
 
 /* Upstream Peer Decode Configuration Register */
-#define INTEL_UPDCR_REG 0x1114
+#define INTEL_UPDCR_REG 0x1014
 /* 5:0 Peer Decode Enable bits */
 #define INTEL_UPDCR_REG_MASK 0x3f
 

include/linux/pci-ats.h

@@ -4,74 +4,39 @@
 
 #include <linux/pci.h>
 
-#ifdef CONFIG_PCI_PRI
+#ifdef CONFIG_PCI_ATS
+/* Address Translation Service */
+int pci_enable_ats(struct pci_dev *dev, int ps);
+void pci_disable_ats(struct pci_dev *dev);
+int pci_ats_queue_depth(struct pci_dev *dev);
+int pci_ats_page_aligned(struct pci_dev *dev);
+#else /* CONFIG_PCI_ATS */
+static inline int pci_enable_ats(struct pci_dev *d, int ps)
+{ return -ENODEV; }
+static inline void pci_disable_ats(struct pci_dev *d) { }
+static inline int pci_ats_queue_depth(struct pci_dev *d)
+{ return -ENODEV; }
+static inline int pci_ats_page_aligned(struct pci_dev *dev)
+{ return 0; }
+#endif /* CONFIG_PCI_ATS */
 
+#ifdef CONFIG_PCI_PRI
 int pci_enable_pri(struct pci_dev *pdev, u32 reqs);
 void pci_disable_pri(struct pci_dev *pdev);
-void pci_restore_pri_state(struct pci_dev *pdev);
 int pci_reset_pri(struct pci_dev *pdev);
-
+int pci_prg_resp_pasid_required(struct pci_dev *pdev);
-#else /* CONFIG_PCI_PRI */
-
-static inline int pci_enable_pri(struct pci_dev *pdev, u32 reqs)
-{
-	return -ENODEV;
-}
-
-static inline void pci_disable_pri(struct pci_dev *pdev)
-{
-}
-
-static inline void pci_restore_pri_state(struct pci_dev *pdev)
-{
-}
-
-static inline int pci_reset_pri(struct pci_dev *pdev)
-{
-	return -ENODEV;
-}
-
 #endif /* CONFIG_PCI_PRI */
 
 #ifdef CONFIG_PCI_PASID
-
 int pci_enable_pasid(struct pci_dev *pdev, int features);
 void pci_disable_pasid(struct pci_dev *pdev);
-void pci_restore_pasid_state(struct pci_dev *pdev);
 int pci_pasid_features(struct pci_dev *pdev);
 int pci_max_pasids(struct pci_dev *pdev);
-int pci_prg_resp_pasid_required(struct pci_dev *pdev);
+#else /* CONFIG_PCI_PASID */
-
-#else  /* CONFIG_PCI_PASID */
-
-static inline int pci_enable_pasid(struct pci_dev *pdev, int features)
-{
-	return -EINVAL;
-}
-
-static inline void pci_disable_pasid(struct pci_dev *pdev)
-{
-}
-
-static inline void pci_restore_pasid_state(struct pci_dev *pdev)
-{
-}
-
 static inline int pci_pasid_features(struct pci_dev *pdev)
-{
+{ return -EINVAL; }
-	return -EINVAL;
-}
-
 static inline int pci_max_pasids(struct pci_dev *pdev)
-{
+{ return -EINVAL; }
-	return -EINVAL;
-}
-
-static inline int pci_prg_resp_pasid_required(struct pci_dev *pdev)
-{
-	return 0;
-}
 #endif /* CONFIG_PCI_PASID */
 
-
+#endif /* LINUX_PCI_ATS_H */
-#endif /* LINUX_PCI_ATS_H*/

include/linux/pci.h

@@ -284,7 +284,6 @@ struct irq_affinity;
 struct pcie_link_state;
 struct pci_vpd;
 struct pci_sriov;
-struct pci_ats;
 struct pci_p2pdma;
 
 /* The pci_dev structure describes PCI devices */
@@ -452,12 +451,14 @@ struct pci_dev {
 	};
 	u16		ats_cap;	/* ATS Capability offset */
 	u8		ats_stu;	/* ATS Smallest Translation Unit */
-	atomic_t	ats_ref_cnt;	/* Number of VFs with ATS enabled */
 #endif
 #ifdef CONFIG_PCI_PRI
+	u16		pri_cap;	/* PRI Capability offset */
 	u32		pri_reqs_alloc; /* Number of PRI requests allocated */
+	unsigned int	pasid_required:1; /* PRG Response PASID Required */
 #endif
 #ifdef CONFIG_PCI_PASID
+	u16		pasid_cap;	/* PASID Capability offset */
 	u16		pasid_features;
 #endif
 #ifdef CONFIG_PCI_P2PDMA
@@ -1770,19 +1771,6 @@ pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs,
 					      NULL);
 }
 
-#ifdef CONFIG_PCI_ATS
-/* Address Translation Service */
-int pci_enable_ats(struct pci_dev *dev, int ps);
-void pci_disable_ats(struct pci_dev *dev);
-int pci_ats_queue_depth(struct pci_dev *dev);
-int pci_ats_page_aligned(struct pci_dev *dev);
-#else
-static inline int pci_enable_ats(struct pci_dev *d, int ps) { return -ENODEV; }
-static inline void pci_disable_ats(struct pci_dev *d) { }
-static inline int pci_ats_queue_depth(struct pci_dev *d) { return -ENODEV; }
-static inline int pci_ats_page_aligned(struct pci_dev *dev) { return 0; }
-#endif
-
 /* Include architecture-dependent settings and functions */
 
 #include <asm/pci.h>