[PATCH 1/2] PCI: use reset_root_port() for bus reset of devices on a root bus

[PATCH 1/2] PCI: use reset_root_port() for bus reset of devices on a root bus

[Hamza Mahfooz]

The "bus" reset method resets a device by asserting Secondary Bus Reset
on its upstream bridge, so it is not available for a device on a root
bus, which has no upstream bridge.

Some platforms have no Root Ports and expose PCI devices directly on a
root bus, e.g., Hyper-V vPCI places each assigned PCI function directly
on the root bus of a virtual host bridge. Such a device has no reset
method at all if it supports neither FLR nor PM reset, even if the
platform is able to reset it.

pci_host_bridge::reset_root_port() already allows a host bridge to reset
a device on a root bus in a platform-specific way, so far only in place
of a Secondary Bus Reset by a Root Port. Use it for the "bus" reset
method of a device that is the only device on a root bus, and wait for
the device to become ready afterwards, as is done for other reset
methods.

Signed-off-by: Hamza Mahfooz <hamzamahfooz@linux.microsoft.com>
---
 drivers/pci/pci.c | 32 ++++++++++++++++++++++++++++++--
 1 file changed, 30 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 8cc6a89130b5..fd4e958d6b99 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -4924,18 +4924,46 @@ int pci_bridge_secondary_bus_reset(struct pci_dev *dev)
 }
 EXPORT_SYMBOL_GPL(pci_bridge_secondary_bus_reset);
 
+static int pci_root_bus_reset(struct pci_dev *dev, bool probe)
+{
+	struct pci_host_bridge *host = pci_find_host_bridge(dev->bus);
+	int rc;
+
+	/*
+	 * Bridges, e.g., Root Ports, are not eligible.  Check the header type
+	 * because dev->subordinate is not set up yet when the reset methods
+	 * are probed during enumeration.
+	 */
+	if (pci_is_bridge(dev) || !host->reset_root_port)
+		return -ENOTTY;
+
+	if (probe)
+		return 0;
+
+	rc = host->reset_root_port(host, dev);
+	if (rc)
+		return rc;
+
+	return pci_dev_wait(dev, "bus reset", PCIE_RESET_READY_POLL_MS);
+}
+
 static int pci_parent_bus_reset(struct pci_dev *dev, bool probe)
 {
 	struct pci_dev *pdev;
 
-	if (pci_is_root_bus(dev->bus) || dev->subordinate ||
-	    !dev->bus->self || dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET)
+	if (dev->subordinate || dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET)
 		return -ENOTTY;
 
 	list_for_each_entry(pdev, &dev->bus->devices, bus_list)
 		if (pdev != dev)
 			return -ENOTTY;
 
+	if (pci_is_root_bus(dev->bus))
+		return pci_root_bus_reset(dev, probe);
+
+	if (!dev->bus->self)
+		return -ENOTTY;
+
 	if (probe)
 		return 0;
 
-- 
2.56.0

[PATCH 2/2] PCI: hv: add vPCI device reset support

[Hamza Mahfooz]

Hyper-V vPCI protocol version 1.5 adds a RESET_DEVICE request for
projected PCI functions. Negotiate the new protocol version and, when it
is in use, implement the pci_host_bridge::reset_root_port() callback by
issuing the request through the vPCI VMBus channel.

Hyper-V has no Root Ports and exposes projected functions directly on
the root bus, so PCI core uses the callback for the "bus" reset method
of a function that is the only one on its vPCI bus.

Use the existing VMBus response path. Return -ENOTTY when the host
reports STATUS_NOT_SUPPORTED so PCI core may try another reset method.

Co-developed-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Hamza Mahfooz <hamzamahfooz@linux.microsoft.com>
---
 Documentation/virt/hyperv/vpci.rst  |  2 +-
 drivers/pci/controller/pci-hyperv.c | 54 +++++++++++++++++++++++++++++
 2 files changed, 55 insertions(+), 1 deletion(-)

diff --git a/Documentation/virt/hyperv/vpci.rst b/Documentation/virt/hyperv/vpci.rst
index b65b2126ede3..6bfee7225c14 100644
--- a/Documentation/virt/hyperv/vpci.rst
+++ b/Documentation/virt/hyperv/vpci.rst
@@ -65,7 +65,7 @@ exchange messages with the vPCI VSP for the purpose of setting
 up and configuring the vPCI device in Linux.  Once the device
 is fully configured in Linux as a PCI device, the VMBus
 channel is used only if Linux changes the vCPU to be interrupted
-in the guest, or if the vPCI device is removed from
+in the guest, or if the vPCI device is reset or removed from
 the VM while the VM is running.  The ongoing operation of the
 device happens directly between the Linux device driver for
 the device and the hardware, with VMBus and the VMBus channel
diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
index 85d826f8f68d..790bf533a7e3 100644
--- a/drivers/pci/controller/pci-hyperv.c
+++ b/drivers/pci/controller/pci-hyperv.c
@@ -68,6 +68,7 @@ enum pci_protocol_version_t {
 	PCI_PROTOCOL_VERSION_1_2 = PCI_MAKE_VERSION(1, 2),	/* RS1 */
 	PCI_PROTOCOL_VERSION_1_3 = PCI_MAKE_VERSION(1, 3),	/* Vibranium */
 	PCI_PROTOCOL_VERSION_1_4 = PCI_MAKE_VERSION(1, 4),	/* WS2022 */
+	PCI_PROTOCOL_VERSION_1_5 = PCI_MAKE_VERSION(1, 5),	/* GE, device reset */
 };
 
 #define CPU_AFFINITY_ALL	-1ULL
@@ -77,6 +78,7 @@ enum pci_protocol_version_t {
  * first.
  */
 static enum pci_protocol_version_t pci_protocol_versions[] = {
+	PCI_PROTOCOL_VERSION_1_5,
 	PCI_PROTOCOL_VERSION_1_4,
 	PCI_PROTOCOL_VERSION_1_3,
 	PCI_PROTOCOL_VERSION_1_2,
@@ -90,6 +92,7 @@ static enum pci_protocol_version_t pci_protocol_versions[] = {
 #define MAX_SUPPORTED_MSI_MESSAGES 0x400
 
 #define STATUS_REVISION_MISMATCH 0xC0000059
+#define STATUS_NOT_SUPPORTED     0xC00000BB
 
 /* space for 32bit serial number as string */
 #define SLOT_NAME_SIZE 11
@@ -136,6 +139,7 @@ enum pci_message_type {
 	PCI_BUS_RELATIONS2		= PCI_MESSAGE_BASE + 0x19,
 	PCI_RESOURCES_ASSIGNED3         = PCI_MESSAGE_BASE + 0x1A,
 	PCI_CREATE_INTERRUPT_MESSAGE3   = PCI_MESSAGE_BASE + 0x1B,
+	PCI_RESET_DEVICE                = PCI_MESSAGE_BASE + 0x1C,
 	PCI_MESSAGE_MAXIMUM
 };
 
@@ -1416,6 +1420,52 @@ static int hv_pcifront_write_config(struct pci_bus *bus, unsigned int devfn,
 	return PCIBIOS_SUCCESSFUL;
 }
 
+static int hv_pci_reset_device(struct pci_host_bridge *bridge,
+			       struct pci_dev *pdev)
+{
+	struct hv_pcibus_device *hbus =
+		container_of(bridge->sysdata, struct hv_pcibus_device, sysdata);
+	struct pci_child_message reset = {};
+	struct hv_pci_compl comp_pkt;
+	struct pci_packet pkt = {
+		.completion_func = hv_pci_generic_compl,
+		.compl_ctxt = &comp_pkt,
+	};
+	enum hv_pcibus_state state;
+	int ret;
+
+	/* Do not take state_lock: eject holds it while removing/locking pdev. */
+	state = READ_ONCE(hbus->state);
+	if (state != hv_pcibus_probed && state != hv_pcibus_installed)
+		return -ENODEV;
+
+	init_completion(&comp_pkt.host_event);
+	reset.message_type.type = PCI_RESET_DEVICE;
+	reset.wslot.slot = devfn_to_wslot(pdev->devfn);
+
+	ret = vmbus_sendpacket(hbus->hdev->channel, &reset, sizeof(reset),
+			       (unsigned long)&pkt, VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (ret)
+		return ret;
+
+	ret = wait_for_response(hbus->hdev, &comp_pkt.host_event,
+				"PCI_RESET_DEVICE");
+	if (ret)
+		return ret;
+
+	if (comp_pkt.completion_status == STATUS_NOT_SUPPORTED)
+		return -ENOTTY;
+
+	if (comp_pkt.completion_status) {
+		pci_err(pdev, "Hyper-V device reset failed: %#x\n",
+			comp_pkt.completion_status);
+		return -EIO;
+	}
+
+	return 0;
+}
+
 /* PCIe operations */
 static struct pci_ops hv_pcifront_ops = {
 	.read  = hv_pcifront_read_config,
@@ -2017,6 +2067,7 @@ static void hv_compose_msi_msg(struct irq_data *data, struct msi_msg *msg)
 		break;
 
 	case PCI_PROTOCOL_VERSION_1_4:
+	case PCI_PROTOCOL_VERSION_1_5:
 		size = hv_compose_msi_req_v3(&ctxt.int_pkts.v3,
 					cpu,
 					hpdev->desc.win_slot.slot,
@@ -2546,6 +2597,9 @@ static int create_root_hv_pci_bus(struct hv_pcibus_device *hbus)
 	bridge->sysdata = &hbus->sysdata;
 	bridge->ops = &hv_pcifront_ops;
 
+	if (hbus->protocol_version >= PCI_PROTOCOL_VERSION_1_5)
+		bridge->reset_root_port = hv_pci_reset_device;
+
 	error = pci_scan_root_bus_bridge(bridge);
 	if (error)
 		return error;
-- 
2.56.0