summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2023-02-25 11:48:02 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2023-02-25 11:48:02 -0800
commit84cc6674b76ba2cdac0df8037b4d8a22a6fc1b77 (patch)
tree6669f09d4ebc3c551e0234abd2f4b82cefd47ff3
parent49d575926890e6ada930bf6f06d62b2fde8fce95 (diff)
parentdeeacf35c922da579637f5db625af20baafc66ed (diff)
Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost
Pull virtio updates from Michael Tsirkin: - device feature provisioning in ifcvf, mlx5 - new SolidNET driver - support for zoned block device in virtio blk - numa support in virtio pmem - VIRTIO_F_RING_RESET support in vhost-net - more debugfs entries in mlx5 - resume support in vdpa - completion batching in virtio blk - cleanup of dma api use in vdpa - now simulating more features in vdpa-sim - documentation, features, fixes all over the place * tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: (64 commits) vdpa/mlx5: support device features provisioning vdpa/mlx5: make MTU/STATUS presence conditional on feature bits vdpa: validate device feature provisioning against supported class vdpa: validate provisioned device features against specified attribute vdpa: conditionally read STATUS in config space vdpa: fix improper error message when adding vdpa dev vdpa/mlx5: Initialize CVQ iotlb spinlock vdpa/mlx5: Don't clear mr struct on destroy MR vdpa/mlx5: Directly assign memory key tools/virtio: enable to build with retpoline vringh: fix a typo in comments for vringh_kiov vhost-vdpa: print warning when vhost_vdpa_alloc_domain fails scsi: virtio_scsi: fix handling of kmalloc failure vdpa: Fix a couple of spelling mistakes in some messages vhost-net: support VIRTIO_F_RING_RESET vhost-scsi: convert sysfs snprintf and sprintf to sysfs_emit vdpa: mlx5: support per virtqueue dma device vdpa: set dma mask for vDPA device virtio-vdpa: support per vq dma device vdpa: introduce get_vq_dma_device() ...
-rw-r--r--Documentation/driver-api/index.rst1
-rw-r--r--Documentation/driver-api/virtio/index.rst11
-rw-r--r--Documentation/driver-api/virtio/virtio.rst145
-rw-r--r--Documentation/driver-api/virtio/writing_virtio_drivers.rst197
-rw-r--r--MAINTAINERS5
-rw-r--r--drivers/block/virtio_blk.c468
-rw-r--r--drivers/nvdimm/virtio_pmem.c11
-rw-r--r--drivers/pci/quirks.c8
-rw-r--r--drivers/scsi/virtio_scsi.c14
-rw-r--r--drivers/vdpa/Kconfig30
-rw-r--r--drivers/vdpa/Makefile1
-rw-r--r--drivers/vdpa/ifcvf/ifcvf_base.c32
-rw-r--r--drivers/vdpa/ifcvf/ifcvf_base.h10
-rw-r--r--drivers/vdpa/ifcvf/ifcvf_main.c162
-rw-r--r--drivers/vdpa/mlx5/Makefile2
-rw-r--r--drivers/vdpa/mlx5/core/mr.c1
-rw-r--r--drivers/vdpa/mlx5/core/resources.c3
-rw-r--r--drivers/vdpa/mlx5/net/debug.c152
-rw-r--r--drivers/vdpa/mlx5/net/mlx5_vnet.c259
-rw-r--r--drivers/vdpa/mlx5/net/mlx5_vnet.h94
-rw-r--r--drivers/vdpa/solidrun/Makefile6
-rw-r--r--drivers/vdpa/solidrun/snet_hwmon.c188
-rw-r--r--drivers/vdpa/solidrun/snet_main.c1111
-rw-r--r--drivers/vdpa/solidrun/snet_vdpa.h194
-rw-r--r--drivers/vdpa/vdpa.c110
-rw-r--r--drivers/vdpa/vdpa_sim/vdpa_sim.c233
-rw-r--r--drivers/vdpa/vdpa_sim/vdpa_sim.h7
-rw-r--r--drivers/vdpa/vdpa_sim/vdpa_sim_blk.c1
-rw-r--r--drivers/vdpa/vdpa_sim/vdpa_sim_net.c219
-rw-r--r--drivers/vhost/net.c5
-rw-r--r--drivers/vhost/scsi.c6
-rw-r--r--drivers/vhost/test.c3
-rw-r--r--drivers/vhost/vdpa.c39
-rw-r--r--drivers/vhost/vhost.c2
-rw-r--r--drivers/vhost/vhost.h2
-rw-r--r--drivers/vhost/vsock.c2
-rw-r--r--drivers/virtio/virtio_ring.c133
-rw-r--r--drivers/virtio/virtio_vdpa.c13
-rw-r--r--include/linux/pci_ids.h2
-rw-r--r--include/linux/vdpa.h12
-rw-r--r--include/linux/virtio_config.h8
-rw-r--r--include/linux/virtio_ring.h16
-rw-r--r--include/linux/vringh.h2
-rw-r--r--include/uapi/linux/vhost.h8
-rw-r--r--include/uapi/linux/vhost_types.h2
-rw-r--r--include/uapi/linux/virtio_blk.h105
-rw-r--r--tools/virtio/Makefile2
47 files changed, 3535 insertions, 502 deletions
diff --git a/Documentation/driver-api/index.rst b/Documentation/driver-api/index.rst
index 7a2584ab63c4..ff9aa1afdc62 100644
--- a/Documentation/driver-api/index.rst
+++ b/Documentation/driver-api/index.rst
@@ -108,6 +108,7 @@ available subsections can be seen below.
vfio-mediated-device
vfio
vfio-pci-device-specific-driver-acceptance
+ virtio/index
xilinx/index
xillybus
zorro
diff --git a/Documentation/driver-api/virtio/index.rst b/Documentation/driver-api/virtio/index.rst
new file mode 100644
index 000000000000..528b14b291e3
--- /dev/null
+++ b/Documentation/driver-api/virtio/index.rst
@@ -0,0 +1,11 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+======
+Virtio
+======
+
+.. toctree::
+ :maxdepth: 1
+
+ virtio
+ writing_virtio_drivers
diff --git a/Documentation/driver-api/virtio/virtio.rst b/Documentation/driver-api/virtio/virtio.rst
new file mode 100644
index 000000000000..7947b4ca690e
--- /dev/null
+++ b/Documentation/driver-api/virtio/virtio.rst
@@ -0,0 +1,145 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+.. _virtio:
+
+===============
+Virtio on Linux
+===============
+
+Introduction
+============
+
+Virtio is an open standard that defines a protocol for communication
+between drivers and devices of different types, see Chapter 5 ("Device
+Types") of the virtio spec (`[1]`_). Originally developed as a standard
+for paravirtualized devices implemented by a hypervisor, it can be used
+to interface any compliant device (real or emulated) with a driver.
+
+For illustrative purposes, this document will focus on the common case
+of a Linux kernel running in a virtual machine and using paravirtualized
+devices provided by the hypervisor, which exposes them as virtio devices
+via standard mechanisms such as PCI.
+
+
+Device - Driver communication: virtqueues
+=========================================
+
+Although the virtio devices are really an abstraction layer in the
+hypervisor, they're exposed to the guest as if they are physical devices
+using a specific transport method -- PCI, MMIO or CCW -- that is
+orthogonal to the device itself. The virtio spec defines these transport
+methods in detail, including device discovery, capabilities and
+interrupt handling.
+
+The communication between the driver in the guest OS and the device in
+the hypervisor is done through shared memory (that's what makes virtio
+devices so efficient) using specialized data structures called
+virtqueues, which are actually ring buffers [#f1]_ of buffer descriptors
+similar to the ones used in a network device:
+
+.. kernel-doc:: include/uapi/linux/virtio_ring.h
+ :identifiers: struct vring_desc
+
+All the buffers the descriptors point to are allocated by the guest and
+used by the host either for reading or for writing but not for both.
+
+Refer to Chapter 2.5 ("Virtqueues") of the virtio spec (`[1]`_) for the
+reference definitions of virtqueues and "Virtqueues and virtio ring: How
+the data travels" blog post (`[2]`_) for an illustrated overview of how
+the host device and the guest driver communicate.
+
+The :c:type:`vring_virtqueue` struct models a virtqueue, including the
+ring buffers and management data. Embedded in this struct is the
+:c:type:`virtqueue` struct, which is the data structure that's
+ultimately used by virtio drivers:
+
+.. kernel-doc:: include/linux/virtio.h
+ :identifiers: struct virtqueue
+
+The callback function pointed by this struct is triggered when the
+device has consumed the buffers provided by the driver. More
+specifically, the trigger will be an interrupt issued by the hypervisor
+(see vring_interrupt()). Interrupt request handlers are registered for
+a virtqueue during the virtqueue setup process (transport-specific).
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: vring_interrupt
+
+
+Device discovery and probing
+============================
+
+In the kernel, the virtio core contains the virtio bus driver and
+transport-specific drivers like `virtio-pci` and `virtio-mmio`. Then
+there are individual virtio drivers for specific device types that are
+registered to the virtio bus driver.
+
+How a virtio device is found and configured by the kernel depends on how
+the hypervisor defines it. Taking the `QEMU virtio-console
+<https://gitlab.com/qemu-project/qemu/-/blob/master/hw/char/virtio-console.c>`__
+device as an example. When using PCI as a transport method, the device
+will present itself on the PCI bus with vendor 0x1af4 (Red Hat, Inc.)
+and device id 0x1003 (virtio console), as defined in the spec, so the
+kernel will detect it as it would do with any other PCI device.
+
+During the PCI enumeration process, if a device is found to match the
+virtio-pci driver (according to the virtio-pci device table, any PCI
+device with vendor id = 0x1af4)::
+
+ /* Qumranet donated their vendor ID for devices 0x1000 thru 0x10FF. */
+ static const struct pci_device_id virtio_pci_id_table[] = {
+ { PCI_DEVICE(PCI_VENDOR_ID_REDHAT_QUMRANET, PCI_ANY_ID) },
+ { 0 }
+ };
+
+then the virtio-pci driver is probed and, if the probing goes well, the
+device is registered to the virtio bus::
+
+ static int virtio_pci_probe(struct pci_dev *pci_dev,
+ const struct pci_device_id *id)
+ {
+ ...
+
+ if (force_legacy) {
+ rc = virtio_pci_legacy_probe(vp_dev);
+ /* Also try modern mode if we can't map BAR0 (no IO space). */
+ if (rc == -ENODEV || rc == -ENOMEM)
+ rc = virtio_pci_modern_probe(vp_dev);
+ if (rc)
+ goto err_probe;
+ } else {
+ rc = virtio_pci_modern_probe(vp_dev);
+ if (rc == -ENODEV)
+ rc = virtio_pci_legacy_probe(vp_dev);
+ if (rc)
+ goto err_probe;
+ }
+
+ ...
+
+ rc = register_virtio_device(&vp_dev->vdev);
+
+When the device is registered to the virtio bus the kernel will look
+for a driver in the bus that can handle the device and call that
+driver's ``probe`` method.
+
+At this point, the virtqueues will be allocated and configured by
+calling the appropriate ``virtio_find`` helper function, such as
+virtio_find_single_vq() or virtio_find_vqs(), which will end up calling
+a transport-specific ``find_vqs`` method.
+
+
+References
+==========
+
+_`[1]` Virtio Spec v1.2:
+https://docs.oasis-open.org/virtio/virtio/v1.2/virtio-v1.2.html
+
+.. Check for later versions of the spec as well.
+
+_`[2]` Virtqueues and virtio ring: How the data travels
+https://www.redhat.com/en/blog/virtqueues-and-virtio-ring-how-data-travels
+
+.. rubric:: Footnotes
+
+.. [#f1] that's why they may be also referred to as virtrings.
diff --git a/Documentation/driver-api/virtio/writing_virtio_drivers.rst b/Documentation/driver-api/virtio/writing_virtio_drivers.rst
new file mode 100644
index 000000000000..e14c58796d25
--- /dev/null
+++ b/Documentation/driver-api/virtio/writing_virtio_drivers.rst
@@ -0,0 +1,197 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+.. _writing_virtio_drivers:
+
+======================
+Writing Virtio Drivers
+======================
+
+Introduction
+============
+
+This document serves as a basic guideline for driver programmers that
+need to hack a new virtio driver or understand the essentials of the
+existing ones. See :ref:`Virtio on Linux <virtio>` for a general
+overview of virtio.
+
+
+Driver boilerplate
+==================
+
+As a bare minimum, a virtio driver needs to register in the virtio bus
+and configure the virtqueues for the device according to its spec, the
+configuration of the virtqueues in the driver side must match the
+virtqueue definitions in the device. A basic driver skeleton could look
+like this::
+
+ #include <linux/virtio.h>
+ #include <linux/virtio_ids.h>
+ #include <linux/virtio_config.h>
+ #include <linux/module.h>
+
+ /* device private data (one per device) */
+ struct virtio_dummy_dev {
+ struct virtqueue *vq;
+ };
+
+ static void virtio_dummy_recv_cb(struct virtqueue *vq)
+ {
+ struct virtio_dummy_dev *dev = vq->vdev->priv;
+ char *buf;
+ unsigned int len;
+
+ while ((buf = virtqueue_get_buf(dev->vq, &len)) != NULL) {
+ /* process the received data */
+ }
+ }
+
+ static int virtio_dummy_probe(struct virtio_device *vdev)
+ {
+ struct virtio_dummy_dev *dev = NULL;
+
+ /* initialize device data */
+ dev = kzalloc(sizeof(struct virtio_dummy_dev), GFP_KERNEL);
+ if (!dev)
+ return -ENOMEM;
+
+ /* the device has a single virtqueue */
+ dev->vq = virtio_find_single_vq(vdev, virtio_dummy_recv_cb, "input");
+ if (IS_ERR(dev->vq)) {
+ kfree(dev);
+ return PTR_ERR(dev->vq);
+
+ }
+ vdev->priv = dev;
+
+ /* from this point on, the device can notify and get callbacks */
+ virtio_device_ready(vdev);
+
+ return 0;
+ }
+
+ static void virtio_dummy_remove(struct virtio_device *vdev)
+ {
+ struct virtio_dummy_dev *dev = vdev->priv;
+
+ /*
+ * disable vq interrupts: equivalent to
+ * vdev->config->reset(vdev)
+ */
+ virtio_reset_device(vdev);
+
+ /* detach unused buffers */
+ while ((buf = virtqueue_detach_unused_buf(dev->vq)) != NULL) {
+ kfree(buf);
+ }
+
+ /* remove virtqueues */
+ vdev->config->del_vqs(vdev);
+
+ kfree(dev);
+ }
+
+ static const struct virtio_device_id id_table[] = {
+ { VIRTIO_ID_DUMMY, VIRTIO_DEV_ANY_ID },
+ { 0 },
+ };
+
+ static struct virtio_driver virtio_dummy_driver = {
+ .driver.name = KBUILD_MODNAME,
+ .driver.owner = THIS_MODULE,
+ .id_table = id_table,
+ .probe = virtio_dummy_probe,
+ .remove = virtio_dummy_remove,
+ };
+
+ module_virtio_driver(virtio_dummy_driver);
+ MODULE_DEVICE_TABLE(virtio, id_table);
+ MODULE_DESCRIPTION("Dummy virtio driver");
+ MODULE_LICENSE("GPL");
+
+The device id ``VIRTIO_ID_DUMMY`` here is a placeholder, virtio drivers
+should be added only for devices that are defined in the spec, see
+include/uapi/linux/virtio_ids.h. Device ids need to be at least reserved
+in the virtio spec before being added to that file.
+
+If your driver doesn't have to do anything special in its ``init`` and
+``exit`` methods, you can use the module_virtio_driver() helper to
+reduce the amount of boilerplate code.
+
+The ``probe`` method does the minimum driver setup in this case
+(memory allocation for the device data) and initializes the
+virtqueue. virtio_device_ready() is used to enable the virtqueue and to
+notify the device that the driver is ready to manage the device
+("DRIVER_OK"). The virtqueues are anyway enabled automatically by the
+core after ``probe`` returns.
+
+.. kernel-doc:: include/linux/virtio_config.h
+ :identifiers: virtio_device_ready
+
+In any case, the virtqueues need to be enabled before adding buffers to
+them.
+
+Sending and receiving data
+==========================
+
+The virtio_dummy_recv_cb() callback in the code above will be triggered
+when the device notifies the driver after it finishes processing a
+descriptor or descriptor chain, either for reading or writing. However,
+that's only the second half of the virtio device-driver communication
+process, as the communication is always started by the driver regardless
+of the direction of the data transfer.
+
+To configure a buffer transfer from the driver to the device, first you
+have to add the buffers -- packed as `scatterlists` -- to the
+appropriate virtqueue using any of the virtqueue_add_inbuf(),
+virtqueue_add_outbuf() or virtqueue_add_sgs(), depending on whether you
+need to add one input `scatterlist` (for the device to fill in), one
+output `scatterlist` (for the device to consume) or multiple
+`scatterlists`, respectively. Then, once the virtqueue is set up, a call
+to virtqueue_kick() sends a notification that will be serviced by the
+hypervisor that implements the device::
+
+ struct scatterlist sg[1];
+ sg_init_one(sg, buffer, BUFLEN);
+ virtqueue_add_inbuf(dev->vq, sg, 1, buffer, GFP_ATOMIC);
+ virtqueue_kick(dev->vq);
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: virtqueue_add_inbuf
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: virtqueue_add_outbuf
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: virtqueue_add_sgs
+
+Then, after the device has read or written the buffers prepared by the
+driver and notifies it back, the driver can call virtqueue_get_buf() to
+read the data produced by the device (if the virtqueue was set up with
+input buffers) or simply to reclaim the buffers if they were already
+consumed by the device:
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: virtqueue_get_buf_ctx
+
+The virtqueue callbacks can be disabled and re-enabled using the
+virtqueue_disable_cb() and the family of virtqueue_enable_cb() functions
+respectively. See drivers/virtio/virtio_ring.c for more details:
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: virtqueue_disable_cb
+
+.. kernel-doc:: drivers/virtio/virtio_ring.c
+ :identifiers: virtqueue_enable_cb
+
+But note that some spurious callbacks can still be triggered under
+certain scenarios. The way to disable callbacks reliably is to reset the
+device or the virtqueue (virtio_reset_device()).
+
+
+References
+==========
+
+_`[1]` Virtio Spec v1.2:
+https://docs.oasis-open.org/virtio/virtio/v1.2/virtio-v1.2.html
+
+Check for later versions of the spec as well.
diff --git a/MAINTAINERS b/MAINTAINERS
index d247b1137178..a3d7c8945762 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -22057,6 +22057,7 @@ S: Maintained
F: Documentation/ABI/testing/sysfs-bus-vdpa
F: Documentation/ABI/testing/sysfs-class-vduse
F: Documentation/devicetree/bindings/virtio/
+F: Documentation/driver-api/virtio/
F: drivers/block/virtio_blk.c
F: drivers/crypto/virtio/
F: drivers/net/virtio_net.c
@@ -22077,6 +22078,10 @@ IFCVF VIRTIO DATA PATH ACCELERATOR
R: Zhu Lingshan <lingshan.zhu@intel.com>
F: drivers/vdpa/ifcvf/
+SNET DPU VIRTIO DATA PATH ACCELERATOR
+R: Alvaro Karsz <alvaro.karsz@solid-run.com>
+F: drivers/vdpa/solidrun/
+
VIRTIO BALLOON
M: "Michael S. Tsirkin" <mst@redhat.com>
M: David Hildenbrand <david@redhat.com>
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index dc6e9b989910..2723eede6f21 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -15,6 +15,7 @@
#include <linux/blk-mq.h>
#include <linux/blk-mq-virtio.h>
#include <linux/numa.h>
+#include <linux/vmalloc.h>
#include <uapi/linux/virtio_ring.h>
#define PART_BITS 4
@@ -80,22 +81,51 @@ struct virtio_blk {
int num_vqs;
int io_queues[HCTX_MAX_TYPES];
struct virtio_blk_vq *vqs;
+
+ /* For zoned device */
+ unsigned int zone_sectors;
};
struct virtblk_req {
+ /* Out header */
struct virtio_blk_outhdr out_hdr;
- u8 status;
+
+ /* In header */
+ union {
+ u8 status;
+
+ /*
+ * The zone append command has an extended in header.
+ * The status field in zone_append_in_hdr must have
+ * the same offset in virtblk_req as the non-zoned
+ * status field above.
+ */
+ struct {
+ u8 status;
+ u8 reserved[7];
+ __le64 append_sector;
+ } zone_append_in_hdr;
+ };
+
+ size_t in_hdr_len;
+
struct sg_table sg_table;
struct scatterlist sg[];
};
-static inline blk_status_t virtblk_result(struct virtblk_req *vbr)
+static inline blk_status_t virtblk_result(u8 status)
{
- switch (vbr->status) {
+ switch (status) {
case VIRTIO_BLK_S_OK:
return BLK_STS_OK;
case VIRTIO_BLK_S_UNSUPP:
return BLK_STS_NOTSUPP;
+ case VIRTIO_BLK_S_ZONE_OPEN_RESOURCE:
+ return BLK_STS_ZONE_OPEN_RESOURCE;
+ case VIRTIO_BLK_S_ZONE_ACTIVE_RESOURCE:
+ return BLK_STS_ZONE_ACTIVE_RESOURCE;
+ case VIRTIO_BLK_S_IOERR:
+ case VIRTIO_BLK_S_ZONE_UNALIGNED_WP:
default:
return BLK_STS_IOERR;
}
@@ -111,11 +141,11 @@ static inline struct virtio_blk_vq *get_virtio_blk_vq(struct blk_mq_hw_ctx *hctx
static int virtblk_add_req(struct virtqueue *vq, struct virtblk_req *vbr)
{
- struct scatterlist hdr, status, *sgs[3];
+ struct scatterlist out_hdr, in_hdr, *sgs[3];
unsigned int num_out = 0, num_in = 0;
- sg_init_one(&hdr, &vbr->out_hdr, sizeof(vbr->out_hdr));
- sgs[num_out++] = &hdr;
+ sg_init_one(&out_hdr, &vbr->out_hdr, sizeof(vbr->out_hdr));
+ sgs[num_out++] = &out_hdr;
if (vbr->sg_table.nents) {
if (vbr->out_hdr.type & cpu_to_virtio32(vq->vdev, VIRTIO_BLK_T_OUT))
@@ -124,8 +154,8 @@ static int virtblk_add_req(struct virtqueue *vq, struct virtblk_req *vbr)
sgs[num_out + num_in++] = vbr->sg_table.sgl;
}
- sg_init_one(&status, &vbr->status, sizeof(vbr->status));
- sgs[num_out + num_in++] = &status;
+ sg_init_one(&in_hdr, &vbr->status, vbr->in_hdr_len);
+ sgs[num_out + num_in++] = &in_hdr;
return virtqueue_add_sgs(vq, sgs, num_out, num_in, vbr, GFP_ATOMIC);
}
@@ -212,21 +242,22 @@ static blk_status_t virtblk_setup_cmd(struct virtio_device *vdev,
struct request *req,
struct virtblk_req *vbr)
{
+ size_t in_hdr_len = sizeof(vbr->status);
bool unmap = false;
u32 type;
+ u64 sector = 0;
- vbr->out_hdr.sector = 0;
+ /* Set fields for all request types */
+ vbr->out_hdr.ioprio = cpu_to_virtio32(vdev, req_get_ioprio(req));
switch (req_op(req)) {
case REQ_OP_READ:
type = VIRTIO_BLK_T_IN;
- vbr->out_hdr.sector = cpu_to_virtio64(vdev,
- blk_rq_pos(req));
+ sector = blk_rq_pos(req);
break;
case REQ_OP_WRITE:
type = VIRTIO_BLK_T_OUT;
- vbr->out_hdr.sector = cpu_to_virtio64(vdev,
- blk_rq_pos(req));
+ sector = blk_rq_pos(req);
break;
case REQ_OP_FLUSH:
type = VIRTIO_BLK_T_FLUSH;
@@ -241,16 +272,42 @@ static blk_status_t virtblk_setup_cmd(struct virtio_device *vdev,
case REQ_OP_SECURE_ERASE:
type = VIRTIO_BLK_T_SECURE_ERASE;
break;
- case REQ_OP_DRV_IN:
- type = VIRTIO_BLK_T_GET_ID;
+ case REQ_OP_ZONE_OPEN:
+ type = VIRTIO_BLK_T_ZONE_OPEN;
+ sector = blk_rq_pos(req);
+ break;
+ case REQ_OP_ZONE_CLOSE:
+ type = VIRTIO_BLK_T_ZONE_CLOSE;
+ sector = blk_rq_pos(req);
+ break;
+ case REQ_OP_ZONE_FINISH:
+ type = VIRTIO_BLK_T_ZONE_FINISH;
+ sector = blk_rq_pos(req);
break;
+ case REQ_OP_ZONE_APPEND:
+ type = VIRTIO_BLK_T_ZONE_APPEND;
+ sector = blk_rq_pos(req);
+ in_hdr_len = sizeof(vbr->zone_append_in_hdr);
+ break;
+ case REQ_OP_ZONE_RESET:
+ type = VIRTIO_BLK_T_ZONE_RESET;
+ sector = blk_rq_pos(req);
+ break;
+ case REQ_OP_ZONE_RESET_ALL:
+ type = VIRTIO_BLK_T_ZONE_RESET_ALL;
+ break;
+ case REQ_OP_DRV_IN:
+ /* Out header already filled in, nothing to do */
+ return 0;
default:
WARN_ON_ONCE(1);
return BLK_STS_IOERR;
}
+ /* Set fields for non-REQ_OP_DRV_IN request types */
+ vbr->in_hdr_len = in_hdr_len;
vbr->out_hdr.type = cpu_to_virtio32(vdev, type);
- vbr->out_hdr.ioprio = cpu_to_virtio32(vdev, req_get_ioprio(req));
+ vbr->out_hdr.sector = cpu_to_virtio64(vdev, sector);
if (type == VIRTIO_BLK_T_DISCARD || type == VIRTIO_BLK_T_WRITE_ZEROES ||
type == VIRTIO_BLK_T_SECURE_ERASE) {
@@ -264,39 +321,74 @@ static blk_status_t virtblk_setup_cmd(struct virtio_device *vdev,
static inline void virtblk_request_done(struct request *req)
{
struct virtblk_req *vbr = blk_mq_rq_to_pdu(req);
+ blk_status_t status = virtblk_result(vbr->status);
virtblk_unmap_data(req, vbr);
virtblk_cleanup_cmd(req);
- blk_mq_end_request(req, virtblk_result(vbr));
+
+ if (req_op(req) == REQ_OP_ZONE_APPEND)
+ req->__sector = le64_to_cpu(vbr->zone_append_in_hdr.append_sector);
+
+ blk_mq_end_request(req, status);
+}
+
+static void virtblk_complete_batch(struct io_comp_batch *iob)
+{
+ struct request *req;
+
+ rq_list_for_each(&iob->req_list, req) {
+ virtblk_unmap_data(req, blk_mq_rq_to_pdu(req));
+ virtblk_cleanup_cmd(req);
+ }
+ blk_mq_end_request_batch(iob);
+}
+
+static int virtblk_handle_req(struct virtio_blk_vq *vq,
+ struct io_comp_batch *iob)
+{
+ struct virtblk_req *vbr;
+ int req_done = 0;
+ unsigned int len;
+
+ while ((vbr = virtqueue_get_buf(vq->vq, &len)) != NULL) {
+ struct request *req = blk_mq_rq_from_pdu(vbr);
+
+ if (likely(!blk_should_fake_timeout(req->q)) &&
+ !blk_mq_complete_request_remote(req) &&
+ !blk_mq_add_to_batch(req, iob, vbr->status,
+ virtblk_complete_batch))
+ virtblk_request_done(req);
+ req_done++;
+ }
+
+ return req_done;
}
static void virtblk_done(struct virtqueue *vq)
{
struct virtio_blk *vblk = vq->vdev->priv;
- bool req_done = false;
- int qid = vq->index;
- struct virtblk_req *vbr;
+ struct virtio_blk_vq *vblk_vq = &vblk->vqs[vq->index];
+ int req_done = 0;
unsigned long flags;
- unsigned int len;
+ DEFINE_IO_COMP_BATCH(iob);
- spin_lock_irqsave(&vblk->vqs[qid].lock, flags);
+ spin_lock_irqsave(&vblk_vq->lock, flags);
do {
virtqueue_disable_cb(vq);
- while ((vbr = virtqueue_get_buf(vblk->vqs[qid].vq, &len)) != NULL) {
- struct request *req = blk_mq_rq_from_pdu(vbr);
+ req_done += virtblk_handle_req(vblk_vq, &iob);
- if (likely(!blk_should_fake_timeout(req->q)))
- blk_mq_complete_request(req);
- req_done = true;
- }
if (unlikely(virtqueue_is_broken(vq)))
break;
} while (!virtqueue_enable_cb(vq));
- /* In case queue is stopped waiting for more buffers. */
- if (req_done)
+ if (req_done) {
+ if (!rq_list_empty(iob.req_list))
+ iob.complete(&iob);
+
+ /* In case queue is stopped waiting for more buffers. */
blk_mq_start_stopped_hw_queues(vblk->disk->queue, true);
- spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
+ }
+ spin_unlock_irqrestore(&vblk_vq->lock, flags);
}
static void virtio_commit_rqs(struct blk_mq_hw_ctx *hctx)
@@ -455,6 +547,275 @@ static void virtio_queue_rqs(struct request **rqlist)
*rqlist = requeue_list;
}
+#ifdef CONFIG_BLK_DEV_ZONED
+static void *virtblk_alloc_report_buffer(struct virtio_blk *vblk,
+ unsigned int nr_zones,
+ unsigned int zone_sectors,
+ size_t *buflen)
+{
+ struct request_queue *q = vblk->disk->queue;
+ size_t bufsize;
+ void *buf;
+
+ nr_zones = min_t(unsigned int, nr_zones,
+ get_capacity(vblk->disk) >> ilog2(zone_sectors));
+
+ bufsize = sizeof(struct virtio_blk_zone_report) +
+ nr_zones * sizeof(struct virtio_blk_zone_descriptor);
+ bufsize = min_t(size_t, bufsize,
+ queue_max_hw_sectors(q) << SECTOR_SHIFT);
+ bufsize = min_t(size_t, bufsize, queue_max_segments(q) << PAGE_SHIFT);
+
+ while (bufsize >= sizeof(struct virtio_blk_zone_report)) {
+ buf = __vmalloc(bufsize, GFP_KERNEL | __GFP_NORETRY);
+ if (buf) {
+ *buflen = bufsize;
+ return buf;
+ }
+ bufsize >>= 1;
+ }
+
+ return NULL;
+}
+
+static int virtblk_submit_zone_report(struct virtio_blk *vblk,
+ char *report_buf, size_t report_len,
+ sector_t sector)
+{
+ struct request_queue *q = vblk->disk->queue;
+ struct request *req;
+ struct virtblk_req *vbr;
+ int err;
+
+ req = blk_mq_alloc_request(q, REQ_OP_DRV_IN, 0);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ vbr = blk_mq_rq_to_pdu(req);
+ vbr->in_hdr_len = sizeof(vbr->status);
+ vbr->out_hdr.type = cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_ZONE_REPORT);
+ vbr->out_hdr.sector = cpu_to_virtio64(vblk->vdev, sector);
+
+ err = blk_rq_map_kern(q, req, report_buf, report_len, GFP_KERNEL);
+ if (err)
+ goto out;
+
+ blk_execute_rq(req, false);
+ err = blk_status_to_errno(virtblk_result(vbr->status));
+out:
+ blk_mq_free_request(req);
+ return err;
+}
+
+static int virtblk_parse_zone(struct virtio_blk *vblk,
+ struct virtio_blk_zone_descriptor *entry,
+ unsigned int idx, unsigned int zone_sectors,
+ report_zones_cb cb, void *data)
+{
+ struct blk_zone zone = { };
+
+ if (entry->z_type != VIRTIO_BLK_ZT_SWR &&
+ entry->z_type != VIRTIO_BLK_ZT_SWP &&
+ entry->z_type != VIRTIO_BLK_ZT_CONV) {
+ dev_err(&vblk->vdev->dev, "invalid zone type %#x\n",
+ entry->z_type);
+ return -EINVAL;
+ }
+
+ zone.type = entry->z_type;
+ zone.cond = entry->z_state;
+ zone.len = zone_sectors;
+ zone.capacity = le64_to_cpu(entry->z_cap);
+ zone.start = le64_to_cpu(entry->z_start);
+ if (zone.cond == BLK_ZONE_COND_FULL)
+ zone.wp = zone.start + zone.len;
+ else
+ zone.wp = le64_to_cpu(entry->z_wp);
+
+ return cb(&zone, idx, data);
+}
+
+static int virtblk_report_zones(struct gendisk *disk, sector_t sector,
+ unsigned int nr_zones, report_zones_cb cb,
+ void *data)
+{
+ struct virtio_blk *vblk = disk->private_data;
+ struct virtio_blk_zone_report *report;
+ unsigned int zone_sectors = vblk->zone_sectors;
+ unsigned int nz, i;
+ int ret, zone_idx = 0;
+ size_t buflen;
+
+ if (WARN_ON_ONCE(!vblk->zone_sectors))
+ return -EOPNOTSUPP;
+
+ report = virtblk_alloc_report_buffer(vblk, nr_zones,
+ zone_sectors, &buflen);
+ if (!report)
+ return -ENOMEM;
+
+ while (zone_idx < nr_zones && sector < get_capacity(vblk->disk)) {
+ memset(report, 0, buflen);
+
+ ret = virtblk_submit_zone_report(vblk, (char *)report,
+ buflen, sector);
+ if (ret) {
+ if (ret > 0)
+ ret = -EIO;
+ goto out_free;
+ }
+ nz = min((unsigned int)le64_to_cpu(report->nr_zones), nr_zones);
+ if (!nz)
+ break;
+
+ for (i = 0; i < nz && zone_idx < nr_zones; i++) {
+ ret = virtblk_parse_zone(vblk, &report->zones[i],
+ zone_idx, zone_sectors, cb, data);
+ if (ret)
+ goto out_free;
+ sector = le64_to_cpu(report->zones[i].z_start) + zone_sectors;
+ zone_idx++;
+ }
+ }
+
+ if (zone_idx > 0)
+ ret = zone_idx;
+ else
+ ret = -EINVAL;
+out_free:
+ kvfree(report);
+ return ret;
+}
+
+static void virtblk_revalidate_zones(struct virtio_blk *vblk)
+{
+ u8 model;
+
+ if (!vblk->zone_sectors)
+ return;
+
+ virtio_cread(vblk->vdev, struct virtio_blk_config,
+ zoned.model, &model);
+ if (!blk_revalidate_disk_zones(vblk->disk, NULL))
+ set_capacity_and_notify(vblk->disk, 0);
+}
+
+static int virtblk_probe_zoned_device(struct virtio_device *vdev,
+ struct virtio_blk *vblk,
+ struct request_queue *q)
+{
+ u32 v;
+ u8 model;
+ int ret;
+
+ virtio_cread(vdev, struct virtio_blk_config,
+ zoned.model, &model);
+
+ switch (model) {
+ case VIRTIO_BLK_Z_NONE:
+ return 0;
+ case VIRTIO_BLK_Z_HM:
+ break;
+ case VIRTIO_BLK_Z_HA:
+ /*
+ * Present the host-aware device as a regular drive.
+ * TODO It is possible to add an option to make it appear
+ * in the system as a zoned drive.
+ */
+ return 0;
+ default:
+ dev_err(&vdev->dev, "unsupported zone model %d\n", model);
+ return -EINVAL;
+ }
+
+ dev_dbg(&vdev->dev, "probing host-managed zoned device\n");
+
+ disk_set_zoned(vblk->disk, BLK_ZONED_HM);
+ blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
+
+ virtio_cread(vdev, struct virtio_blk_config,
+ zoned.max_open_zones, &v);
+ disk_set_max_open_zones(vblk->disk, le32_to_cpu(v));
+
+ dev_dbg(&vdev->dev, "max open zones = %u\n", le32_to_cpu(v));
+
+ virtio_cread(vdev, struct virtio_blk_config,
+ zoned.max_active_zones, &v);
+ disk_set_max_active_zones(vblk->disk, le32_to_cpu(v));
+ dev_dbg(&vdev->dev, "max active zones = %u\n", le32_to_cpu(v));
+
+ virtio_cread(vdev, struct virtio_blk_config,
+ zoned.write_granularity, &v);
+ if (!v) {
+ dev_warn(&vdev->dev, "zero write granularity reported\n");
+ return -ENODEV;
+ }
+ blk_queue_physical_block_size(q, le32_to_cpu(v));
+ blk_queue_io_min(q, le32_to_cpu(v));
+
+ dev_dbg(&vdev->dev, "write granularity = %u\n", le32_to_cpu(v));
+
+ /*
+ * virtio ZBD specification doesn't require zones to be a power of
+ * two sectors in size, but the code in this driver expects that.
+ */
+ virtio_cread(vdev, struct virtio_blk_config, zoned.zone_sectors, &v);
+ vblk->zone_sectors = le32_to_cpu(v);
+ if (vblk->zone_sectors == 0 || !is_power_of_2(vblk->zone_sectors)) {
+ dev_err(&vdev->dev,
+ "zoned device with non power of two zone size %u\n",
+ vblk->zone_sectors);
+ return -ENODEV;
+ }
+ dev_dbg(&vdev->dev, "zone sectors = %u\n", vblk->zone_sectors);
+
+ if (virtio_has_feature(vdev, VIRTIO_BLK_F_DISCARD)) {
+ dev_warn(&vblk->vdev->dev,
+ "ignoring negotiated F_DISCARD for zoned device\n");
+ blk_queue_max_discard_sectors(q, 0);
+ }
+
+ ret = blk_revalidate_disk_zones(vblk->disk, NULL);
+ if (!ret) {
+ virtio_cread(vdev, struct virtio_blk_config,
+ zoned.max_append_sectors, &v);
+ if (!v) {
+ dev_warn(&vdev->dev, "zero max_append_sectors reported\n");
+ return -ENODEV;
+ }
+ blk_queue_max_zone_append_sectors(q, le32_to_cpu(v));
+ dev_dbg(&vdev->dev, "max append sectors = %u\n", le32_to_cpu(v));
+ }
+
+ return ret;
+}
+
+static inline bool virtblk_has_zoned_feature(struct virtio_device *vdev)
+{
+ return virtio_has_feature(vdev, VIRTIO_BLK_F_ZONED);
+}
+#else
+
+/*
+ * Zoned block device support is not configured in this kernel.
+ * We only need to define a few symbols to avoid compilation errors.
+ */
+#define virtblk_report_zones NULL
+static inline void virtblk_revalidate_zones(struct virtio_blk *vblk)
+{
+}
+static inline int virtblk_probe_zoned_device(struct virtio_device *vdev,
+ struct virtio_blk *vblk, struct request_queue *q)
+{
+ return -EOPNOTSUPP;
+}
+
+static inline bool virtblk_has_zoned_feature(struct virtio_device *vdev)
+{
+ return false;
+}
+#endif /* CONFIG_BLK_DEV_ZONED */
+
/* return id (s/n) string for *disk to *id_str
*/
static int virtblk_get_id(struct gendisk *disk, char *id_str)
@@ -462,18 +823,24 @@ static int virtblk_get_id(struct gendisk *disk, char *id_str)
struct virtio_blk *vblk = disk->private_data;
struct request_queue *q = vblk->disk->queue;
struct request *req;
+ struct virtblk_req *vbr;
int err;
req = blk_mq_alloc_request(q, REQ_OP_DRV_IN, 0);
if (IS_ERR(req))
return PTR_ERR(req);
+ vbr = blk_mq_rq_to_pdu(req);
+ vbr->in_hdr_len = sizeof(vbr->status);
+ vbr->out_hdr.type = cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_GET_ID);
+ vbr->out_hdr.sector = 0;
+
err = blk_rq_map_kern(q, req, id_str, VIRTIO_BLK_ID_BYTES, GFP_KERNEL);
if (err)
goto out;
blk_execute_rq(req, false);
- err = blk_status_to_errno(virtblk_result(blk_mq_rq_to_pdu(req)));
+ err = blk_status_to_errno(virtblk_result(vbr->status));
out:
blk_mq_free_request(req);
return err;
@@ -524,6 +891,7 @@ static const struct block_device_operations virtblk_fops = {
.owner = THIS_MODULE,
.getgeo = virtblk_getgeo,
.free_disk = virtblk_free_disk,
+ .report_zones = virtblk_report_zones,
};
static int index_to_minor(int index)
@@ -594,6 +962,7 @@ static void virtblk_config_changed_work(struct work_struct *work)
struct virtio_blk *vblk =
container_of(work, struct virtio_blk, config_work);
+ virtblk_revalidate_zones(vblk);
virtblk_update_capacity(vblk, true);
}
@@ -835,36 +1204,15 @@ static void virtblk_map_queues(struct blk_mq_tag_set *set)
}
}
-static void virtblk_complete_batch(struct io_comp_batch *iob)
-{
- struct request *req;
-
- rq_list_for_each(&iob->req_list, req) {
- virtblk_unmap_data(req, blk_mq_rq_to_pdu(req));
- virtblk_cleanup_cmd(req);
- }
- blk_mq_end_request_batch(iob);
-}
-
static int virtblk_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob)
{
struct virtio_blk *vblk = hctx->queue->queuedata;
struct virtio_blk_vq *vq = get_virtio_blk_vq(hctx);
- struct virtblk_req *vbr;
unsigned long flags;
- unsigned int len;
int found = 0;
spin_lock_irqsave(&vq->lock, flags);
-
- while ((vbr = virtqueue_get_buf(vq->vq, &len)) != NULL) {
- struct request *req = blk_mq_rq_from_pdu(vbr);
-
- found++;
- if (!blk_mq_add_to_batch(req, iob, vbr->status,
- virtblk_complete_batch))
- blk_mq_complete_request(req);
- }
+ found = virtblk_handle_req(vq, iob);
if (found)
blk_mq_start_stopped_hw_queues(vblk->disk->queue, true);
@@ -1150,6 +1498,15 @@ static int virtblk_probe(struct virtio_device *vdev)
virtblk_update_capacity(vblk, false);
virtio_device_ready(vdev);
+ if (virtblk_has_zoned_feature(vdev)) {
+ err = virtblk_probe_zoned_device(vdev, vblk, q);
+ if (err)
+ goto out_cleanup_disk;
+ }
+
+ dev_info(&vdev->dev, "blk config size: %zu\n",
+ sizeof(struct virtio_blk_config));
+
err = device_add_disk(&vdev->dev, vblk->disk, virtblk_attr_groups);
if (err)
goto out_cleanup_disk;
@@ -1251,6 +1608,9 @@ static unsigned int features[] = {
VIRTIO_BLK_F_FLUSH, VIRTIO_BLK_F_TOPOLOGY, VIRTIO_BLK_F_CONFIG_WCE,
VIRTIO_BLK_F_MQ, VIRTIO_BLK_F_DISCARD, VIRTIO_BLK_F_WRITE_ZEROES,
VIRTIO_BLK_F_SECURE_ERASE,
+#ifdef CONFIG_BLK_DEV_ZONED
+ VIRTIO_BLK_F_ZONED,
+#endif /* CONFIG_BLK_DEV_ZONED */
};
static struct virtio_driver virtio_blk = {
diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
index 20da455d2ef6..a92eb172f0e7 100644
--- a/drivers/nvdimm/virtio_pmem.c
+++ b/drivers/nvdimm/virtio_pmem.c
@@ -32,7 +32,6 @@ static int init_vq(struct virtio_pmem *vpmem)
static int virtio_pmem_probe(struct virtio_device *vdev)
{
struct nd_region_desc ndr_desc = {};
- int nid = dev_to_node(&vdev->dev);
struct nd_region *nd_region;
struct virtio_pmem *vpmem;
struct resource res;
@@ -79,7 +78,15 @@ static int virtio_pmem_probe(struct virtio_device *vdev)
dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
ndr_desc.res = &res;
- ndr_desc.numa_node = nid;
+
+ ndr_desc.numa_node = memory_add_physaddr_to_nid(res.start);
+ ndr_desc.target_node = phys_to_target_node(res.start);
+ if (ndr_desc.target_node == NUMA_NO_NODE) {
+ ndr_desc.target_node = ndr_desc.numa_node;
+ dev_dbg(&vdev->dev, "changing target node from %d to %d",
+ NUMA_NO_NODE, ndr_desc.target_node);
+ }
+
ndr_desc.flush = async_pmem_flush;
ndr_desc.provider_data = vdev;
set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 494fa46f5767..44cab813bf95 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -5366,6 +5366,14 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AMD, 0x7901, quirk_no_flr);
DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x1502, quirk_no_flr);
DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x1503, quirk_no_flr);
+/* FLR may cause the SolidRun SNET DPU (rev 0x1) to hang */
+static void quirk_no_flr_snet(struct pci_dev *dev)
+{
+ if (dev->revision == 0x1)
+ quirk_no_flr(dev);
+}
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SOLIDRUN, 0x1000, quirk_no_flr_snet);
+
static void quirk_no_ext_tags(struct pci_dev *pdev)
{
struct pci_host_bridge *bridge = pci_find_host_bridge(pdev->bus);
diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
index b221c3c99320..c5558c45ab3a 100644
--- a/drivers/scsi/virtio_scsi.c
+++ b/drivers/scsi/virtio_scsi.c
@@ -330,7 +330,7 @@ static void virtscsi_handle_param_change(struct virtio_scsi *vscsi,
scsi_device_put(sdev);
}
-static void virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
+static int virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
{
struct scsi_device *sdev;
struct Scsi_Host *shost = virtio_scsi_host(vscsi->vdev);
@@ -338,6 +338,11 @@ static void virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
int result, inquiry_len, inq_result_len = 256;
char *inq_result = kmalloc(inq_result_len, GFP_KERNEL);
+ if (!inq_result) {
+ kfree(inq_result);
+ return -ENOMEM;
+ }
+
shost_for_each_device(sdev, shost) {
inquiry_len = sdev->inquiry_len ? sdev->inquiry_len : 36;
@@ -366,6 +371,7 @@ static void virtscsi_rescan_hotunplug(struct virtio_scsi *vscsi)
}
kfree(inq_result);
+ return 0;
}
static void virtscsi_handle_event(struct work_struct *work)
@@ -377,9 +383,13 @@ static void virtscsi_handle_event(struct work_struct *work)
if (event->event &
cpu_to_virtio32(vscsi->vdev, VIRTIO_SCSI_T_EVENTS_MISSED)) {
+ int ret;
+
event->event &= ~cpu_to_virtio32(vscsi->vdev,
VIRTIO_SCSI_T_EVENTS_MISSED);
- virtscsi_rescan_hotunplug(vscsi);
+ ret = virtscsi_rescan_hotunplug(vscsi);
+ if (ret)
+ return;
scsi_scan_host(virtio_scsi_host(vscsi->vdev));
}
diff --git a/drivers/vdpa/Kconfig b/drivers/vdpa/Kconfig
index 50f45d037611..cd6ad92f3f05 100644
--- a/drivers/vdpa/Kconfig
+++ b/drivers/vdpa/Kconfig
@@ -71,6 +71,18 @@ config MLX5_VDPA_NET
be executed by the hardware. It also supports a variety of stateless
offloads depending on the actual device used and firmware version.
+config MLX5_VDPA_STEERING_DEBUG
+ bool "expose steering counters on debugfs"
+ select MLX5_VDPA
+ help
+ Expose RX steering counters in debugfs to aid in debugging. For each VLAN
+ or non VLAN interface, two hardware counters are added to the RX flow
+ table: one for unicast and one for multicast.
+ The counters counts the number of packets and bytes and exposes them in
+ debugfs. Once can read the counters using, e.g.:
+ cat /sys/kernel/debug/mlx5/mlx5_core.sf.1/vdpa-0/rx/untagged/ucast/packets
+ cat /sys/kernel/debug/mlx5/mlx5_core.sf.1/vdpa-0/rx/untagged/mcast/bytes
+
config VP_VDPA
tristate "Virtio PCI bridge vDPA driver"
select VIRTIO_PCI_LIB
@@ -86,4 +98,22 @@ config ALIBABA_ENI_VDPA
VDPA driver for Alibaba ENI (Elastic Network Interface) which is built upon
virtio 0.9.5 specification.
+ config SNET_VDPA
+ tristate "SolidRun's vDPA driver for SolidNET"
+ depends on PCI_MSI && PCI_IOV && (HWMON || HWMON=n)
+
+ # This driver MAY create a HWMON device.
+ # Depending on (HWMON || HWMON=n) ensures that:
+ # If HWMON=n the driver can be compiled either as a module or built-in.
+ # If HWMON=y the driver can be compiled either as a module or built-in.
+ # If HWMON=m the driver is forced to be compiled as a module.
+ # By doing so, IS_ENABLED can be used instead of IS_REACHABLE
+
+ help
+ vDPA driver for SolidNET DPU.
+ With this driver, the VirtIO dataplane can be
+ offloaded to a SolidNET DPU.
+ This driver includes a HW monitor device that
+ reads health values from the DPU.
+
endif # VDPA
diff --git a/drivers/vdpa/Makefile b/drivers/vdpa/Makefile
index 15665563a7f4..59396ff2a318 100644
--- a/drivers/vdpa/Makefile
+++ b/drivers/vdpa/Makefile
@@ -6,3 +6,4 @@ obj-$(CONFIG_IFCVF) += ifcvf/
obj-$(CONFIG_MLX5_VDPA) += mlx5/
obj-$(CONFIG_VP_VDPA) += virtio_pci/
obj-$(CONFIG_ALIBABA_ENI_VDPA) += alibaba/
+obj-$(CONFIG_SNET_VDPA) += solidrun/
diff --git a/drivers/vdpa/ifcvf/ifcvf_base.c b/drivers/vdpa/ifcvf/ifcvf_base.c
index 3e4486bfa0b7..5563b3a773c7 100644
--- a/drivers/vdpa/ifcvf/ifcvf_base.c
+++ b/drivers/vdpa/ifcvf/ifcvf_base.c
@@ -10,11 +10,6 @@
#include "ifcvf_base.h"
-struct ifcvf_adapter *vf_to_adapter(struct ifcvf_hw *hw)
-{
- return container_of(hw, struct ifcvf_adapter, vf);
-}
-
u16 ifcvf_set_vq_vector(struct ifcvf_hw *hw, u16 qid, int vector)
{
struct virtio_pci_common_cfg __iomem *cfg = hw->common_cfg;
@@ -37,8 +32,6 @@ u16 ifcvf_set_config_vector(struct ifcvf_hw *hw, int vector)
static void __iomem *get_cap_addr(struct ifcvf_hw *hw,
struct virtio_pci_cap *cap)
{
- struct ifcvf_adapter *ifcvf;
- struct pci_dev *pdev;
u32 length, offset;
u8 bar;
@@ -46,17 +39,14 @@ static void __iomem *get_cap_addr(struct ifcvf_hw *hw,
offset = le32_to_cpu(cap->offset);
bar = cap->bar;
- ifcvf= vf_to_adapter(hw);
- pdev = ifcvf->pdev;
-
if (bar >= IFCVF_PCI_MAX_RESOURCE) {
- IFCVF_DBG(pdev,
+ IFCVF_DBG(hw->pdev,
"Invalid bar number %u to get capabilities\n", bar);
return NULL;
}
- if (offset + length > pci_resource_len(pdev, bar)) {
- IFCVF_DBG(pdev,
+ if (offset + length > pci_resource_len(hw->pdev, bar)) {
+ IFCVF_DBG(hw->pdev,
"offset(%u) + len(%u) overflows bar%u's capability\n",
offset, length, bar);
return NULL;
@@ -92,6 +82,7 @@ int ifcvf_init_hw(struct ifcvf_hw *hw, struct pci_dev *pdev)
IFCVF_ERR(pdev, "Failed to read PCI capability list\n");
return -EIO;
}
+ hw->pdev = pdev;
while (pos) {
ret = ifcvf_read_config_range(pdev, (u32 *)&cap,
@@ -215,15 +206,13 @@ u64 ifcvf_get_hw_features(struct ifcvf_hw *hw)
u64 ifcvf_get_features(struct ifcvf_hw *hw)
{
- return hw->hw_features;
+ return hw->dev_features;
}
int ifcvf_verify_min_features(struct ifcvf_hw *hw, u64 features)
{
- struct ifcvf_adapter *ifcvf = vf_to_adapter(hw);
-
if (!(features & BIT_ULL(VIRTIO_F_ACCESS_PLATFORM)) && features) {
- IFCVF_ERR(ifcvf->pdev, "VIRTIO_F_ACCESS_PLATFORM is not negotiated\n");
+ IFCVF_ERR(hw->pdev, "VIRTIO_F_ACCESS_PLATFORM is not negotiated\n");
return -EINVAL;
}
@@ -232,13 +221,11 @@ int ifcvf_verify_min_features(struct ifcvf_hw *hw, u64 features)
u32 ifcvf_get_config_size(struct ifcvf_hw *hw)
{
- struct ifcvf_adapter *adapter;
u32 net_config_size = sizeof(struct virtio_net_config);
u32 blk_config_size = sizeof(struct virtio_blk_config);
u32 cap_size = hw->cap_dev_config_size;
u32 config_size;
- adapter = vf_to_adapter(hw);
/* If the onboard device config space size is greater than
* the size of struct virtio_net/blk_config, only the spec
* implementing contents size is returned, this is very
@@ -253,7 +240,7 @@ u32 ifcvf_get_config_size(struct ifcvf_hw *hw)
break;
default:
config_size = 0;
- IFCVF_ERR(adapter->pdev, "VIRTIO ID %u not supported\n", hw->dev_type);
+ IFCVF_ERR(hw->pdev, "VIRTIO ID %u not supported\n", hw->dev_type);
}
return config_size;
@@ -301,14 +288,11 @@ static void ifcvf_set_features(struct ifcvf_hw *hw, u64 features)
static int ifcvf_config_features(struct ifcvf_hw *hw)
{
- struct ifcvf_adapter *ifcvf;
-
- ifcvf = vf_to_adapter(hw);
ifcvf_set_features(hw, hw->req_features);
ifcvf_add_status(hw, VIRTIO_CONFIG_S_FEATURES_OK);
if (!(ifcvf_get_status(hw) & VIRTIO_CONFIG_S_FEATURES_OK)) {
- IFCVF_ERR(ifcvf->pdev, "Failed to set FEATURES_OK status\n");
+ IFCVF_ERR(hw->pdev, "Failed to set FEATURES_OK status\n");
return -EIO;
}
diff --git a/drivers/vdpa/ifcvf/ifcvf_base.h b/drivers/vdpa/ifcvf/ifcvf_base.h
index f5563f665cc6..c20d1c40214e 100644
--- a/drivers/vdpa/ifcvf/ifcvf_base.h
+++ b/drivers/vdpa/ifcvf/ifcvf_base.h
@@ -19,6 +19,7 @@
#include <uapi/linux/virtio_blk.h>
#include <uapi/linux/virtio_config.h>
#include <uapi/linux/virtio_pci.h>
+#include <uapi/linux/vdpa.h>
#define N3000_DEVICE_ID 0x1041
#define N3000_SUBSYS_DEVICE_ID 0x001A
@@ -38,9 +39,6 @@
#define IFCVF_DBG(pdev, fmt, ...) dev_dbg(&pdev->dev, fmt, ##__VA_ARGS__)
#define IFCVF_INFO(pdev, fmt, ...) dev_info(&pdev->dev, fmt, ##__VA_ARGS__)
-#define ifcvf_private_to_vf(adapter) \
- (&((struct ifcvf_adapter *)adapter)->vf)
-
/* all vqs and config interrupt has its own vector */
#define MSIX_VECTOR_PER_VQ_AND_CONFIG 1
/* all vqs share a vector, and config interrupt has a separate vector */
@@ -78,6 +76,8 @@ struct ifcvf_hw {
u32 dev_type;
u64 req_features;
u64 hw_features;
+ /* provisioned device features */
+ u64 dev_features;
struct virtio_pci_common_cfg __iomem *common_cfg;
void __iomem *dev_cfg;
struct vring_info vring[IFCVF_MAX_QUEUES];
@@ -89,12 +89,13 @@ struct ifcvf_hw {
u16 nr_vring;
/* VIRTIO_PCI_CAP_DEVICE_CFG size */
u32 cap_dev_config_size;
+ struct pci_dev *pdev;
};
struct ifcvf_adapter {
struct vdpa_device vdpa;
struct pci_dev *pdev;
- struct ifcvf_hw vf;
+ struct ifcvf_hw *vf;
};
struct ifcvf_vring_lm_cfg {
@@ -109,6 +110,7 @@ struct ifcvf_lm_cfg {
struct ifcvf_vdpa_mgmt_dev {
struct vdpa_mgmt_dev mdev;
+ struct ifcvf_hw vf;
struct ifcvf_adapter *adapter;
struct pci_dev *pdev;
};
diff --git a/drivers/vdpa/ifcvf/ifcvf_main.c b/drivers/vdpa/ifcvf/ifcvf_main.c
index 44b29289aa19..7f78c47e40d6 100644
--- a/drivers/vdpa/ifcvf/ifcvf_main.c
+++ b/drivers/vdpa/ifcvf/ifcvf_main.c
@@ -69,10 +69,9 @@ static void ifcvf_free_irq_vectors(void *data)
pci_free_irq_vectors(data);
}
-static void ifcvf_free_per_vq_irq(struct ifcvf_adapter *adapter)
+static void ifcvf_free_per_vq_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
int i;
for (i = 0; i < vf->nr_vring; i++) {
@@ -83,10 +82,9 @@ static void ifcvf_free_per_vq_irq(struct ifcvf_adapter *adapter)
}
}
-static void ifcvf_free_vqs_reused_irq(struct ifcvf_adapter *adapter)
+static void ifcvf_free_vqs_reused_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
if (vf->vqs_reused_irq != -EINVAL) {
devm_free_irq(&pdev->dev, vf->vqs_reused_irq, vf);
@@ -95,20 +93,17 @@ static void ifcvf_free_vqs_reused_irq(struct ifcvf_adapter *adapter)
}
-static void ifcvf_free_vq_irq(struct ifcvf_adapter *adapter)
+static void ifcvf_free_vq_irq(struct ifcvf_hw *vf)
{
- struct ifcvf_hw *vf = &adapter->vf;
-
if (vf->msix_vector_status == MSIX_VECTOR_PER_VQ_AND_CONFIG)
- ifcvf_free_per_vq_irq(adapter);
+ ifcvf_free_per_vq_irq(vf);
else
- ifcvf_free_vqs_reused_irq(adapter);
+ ifcvf_free_vqs_reused_irq(vf);
}
-static void ifcvf_free_config_irq(struct ifcvf_adapter *adapter)
+static void ifcvf_free_config_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
if (vf->config_irq == -EINVAL)
return;
@@ -123,12 +118,12 @@ static void ifcvf_free_config_irq(struct ifcvf_adapter *adapter)
}
}
-static void ifcvf_free_irq(struct ifcvf_adapter *adapter)
+static void ifcvf_free_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
+ struct pci_dev *pdev = vf->pdev;
- ifcvf_free_vq_irq(adapter);
- ifcvf_free_config_irq(adapter);
+ ifcvf_free_vq_irq(vf);
+ ifcvf_free_config_irq(vf);
ifcvf_free_irq_vectors(pdev);
}
@@ -137,10 +132,9 @@ static void ifcvf_free_irq(struct ifcvf_adapter *adapter)
* It returns the number of allocated vectors, negative
* return value when fails.
*/
-static int ifcvf_alloc_vectors(struct ifcvf_adapter *adapter)
+static int ifcvf_alloc_vectors(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
int max_intr, ret;
/* all queues and config interrupt */
@@ -160,10 +154,9 @@ static int ifcvf_alloc_vectors(struct ifcvf_adapter *adapter)
return ret;
}
-static int ifcvf_request_per_vq_irq(struct ifcvf_adapter *adapter)
+static int ifcvf_request_per_vq_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
int i, vector, ret, irq;
vf->vqs_reused_irq = -EINVAL;
@@ -190,15 +183,14 @@ static int ifcvf_request_per_vq_irq(struct ifcvf_adapter *adapter)
return 0;
err:
- ifcvf_free_irq(adapter);
+ ifcvf_free_irq(vf);
return -EFAULT;
}
-static int ifcvf_request_vqs_reused_irq(struct ifcvf_adapter *adapter)
+static int ifcvf_request_vqs_reused_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
int i, vector, ret, irq;
vector = 0;
@@ -224,15 +216,14 @@ static int ifcvf_request_vqs_reused_irq(struct ifcvf_adapter *adapter)
return 0;
err:
- ifcvf_free_irq(adapter);
+ ifcvf_free_irq(vf);
return -EFAULT;
}
-static int ifcvf_request_dev_irq(struct ifcvf_adapter *adapter)
+static int ifcvf_request_dev_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
int i, vector, ret, irq;
vector = 0;
@@ -265,29 +256,27 @@ static int ifcvf_request_dev_irq(struct ifcvf_adapter *adapter)
return 0;
err:
- ifcvf_free_irq(adapter);
+ ifcvf_free_irq(vf);
return -EFAULT;
}
-static int ifcvf_request_vq_irq(struct ifcvf_adapter *adapter)
+static int ifcvf_request_vq_irq(struct ifcvf_hw *vf)
{
- struct ifcvf_hw *vf = &adapter->vf;
int ret;
if (vf->msix_vector_status == MSIX_VECTOR_PER_VQ_AND_CONFIG)
- ret = ifcvf_request_per_vq_irq(adapter);
+ ret = ifcvf_request_per_vq_irq(vf);
else
- ret = ifcvf_request_vqs_reused_irq(adapter);
+ ret = ifcvf_request_vqs_reused_irq(vf);
return ret;
}
-static int ifcvf_request_config_irq(struct ifcvf_adapter *adapter)
+static int ifcvf_request_config_irq(struct ifcvf_hw *vf)
{
- struct pci_dev *pdev = adapter->pdev;
- struct ifcvf_hw *vf = &adapter->vf;
+ struct pci_dev *pdev = vf->pdev;
int config_vector, ret;
if (vf->msix_vector_status == MSIX_VECTOR_PER_VQ_AND_CONFIG)
@@ -320,17 +309,16 @@ static int ifcvf_request_config_irq(struct ifcvf_adapter *adapter)
return 0;
err:
- ifcvf_free_irq(adapter);
+ ifcvf_free_irq(vf);
return -EFAULT;
}
-static int ifcvf_request_irq(struct ifcvf_adapter *adapter)
+static int ifcvf_request_irq(struct ifcvf_hw *vf)
{
- struct ifcvf_hw *vf = &adapter->vf;
int nvectors, ret, max_intr;
- nvectors = ifcvf_alloc_vectors(adapter);
+ nvectors = ifcvf_alloc_vectors(vf);
if (nvectors <= 0)
return -EFAULT;
@@ -341,16 +329,16 @@ static int ifcvf_request_irq(struct ifcvf_adapter *adapter)
if (nvectors == 1) {
vf->msix_vector_status = MSIX_VECTOR_DEV_SHARED;
- ret = ifcvf_request_dev_irq(adapter);
+ ret = ifcvf_request_dev_irq(vf);
return ret;
}
- ret = ifcvf_request_vq_irq(adapter);
+ ret = ifcvf_request_vq_irq(vf);
if (ret)
return ret;
- ret = ifcvf_request_config_irq(adapter);
+ ret = ifcvf_request_config_irq(vf);
if (ret)
return ret;
@@ -358,9 +346,9 @@ static int ifcvf_request_irq(struct ifcvf_adapter *adapter)
return 0;
}
-static int ifcvf_start_datapath(void *private)
+static int ifcvf_start_datapath(struct ifcvf_adapter *adapter)
{
- struct ifcvf_hw *vf = ifcvf_private_to_vf(private);
+ struct ifcvf_hw *vf = adapter->vf;
u8 status;
int ret;
@@ -374,9 +362,9 @@ static int ifcvf_start_datapath(void *private)
return ret;
}
-static int ifcvf_stop_datapath(void *private)
+static int ifcvf_stop_datapath(struct ifcvf_adapter *adapter)
{
- struct ifcvf_hw *vf = ifcvf_private_to_vf(private);
+ struct ifcvf_hw *vf = adapter->vf;
int i;
for (i = 0; i < vf->nr_vring; i++)
@@ -389,7 +377,7 @@ static int ifcvf_stop_datapath(void *private)
static void ifcvf_reset_vring(struct ifcvf_adapter *adapter)
{
- struct ifcvf_hw *vf = ifcvf_private_to_vf(adapter);
+ struct ifcvf_hw *vf = adapter->vf;
int i;
for (i = 0; i < vf->nr_vring; i++) {
@@ -414,7 +402,7 @@ static struct ifcvf_hw *vdpa_to_vf(struct vdpa_device *vdpa_dev)
{
struct ifcvf_adapter *adapter = vdpa_to_adapter(vdpa_dev);
- return &adapter->vf;
+ return adapter->vf;
}
static u64 ifcvf_vdpa_get_device_features(struct vdpa_device *vdpa_dev)
@@ -479,7 +467,7 @@ static void ifcvf_vdpa_set_status(struct vdpa_device *vdpa_dev, u8 status)
if ((status & VIRTIO_CONFIG_S_DRIVER_OK) &&
!(status_old & VIRTIO_CONFIG_S_DRIVER_OK)) {
- ret = ifcvf_request_irq(adapter);
+ ret = ifcvf_request_irq(vf);
if (ret) {
status = ifcvf_get_status(vf);
status |= VIRTIO_CONFIG_S_FAILED;
@@ -511,7 +499,7 @@ static int ifcvf_vdpa_reset(struct vdpa_device *vdpa_dev)
if (status_old & VIRTIO_CONFIG_S_DRIVER_OK) {
ifcvf_stop_datapath(adapter);
- ifcvf_free_irq(adapter);
+ ifcvf_free_irq(vf);
}
ifcvf_reset_vring(adapter);
@@ -755,17 +743,37 @@ static int ifcvf_vdpa_dev_add(struct vdpa_mgmt_dev *mdev, const char *name,
struct vdpa_device *vdpa_dev;
struct pci_dev *pdev;
struct ifcvf_hw *vf;
+ u64 device_features;
int ret;
ifcvf_mgmt_dev = container_of(mdev, struct ifcvf_vdpa_mgmt_dev, mdev);
- if (!ifcvf_mgmt_dev->adapter)
- return -EOPNOTSUPP;
+ vf = &ifcvf_mgmt_dev->vf;
+ pdev = vf->pdev;
+ adapter = vdpa_alloc_device(struct ifcvf_adapter, vdpa,
+ &pdev->dev, &ifc_vdpa_ops, 1, 1, NULL, false);
+ if (IS_ERR(adapter)) {
+ IFCVF_ERR(pdev, "Failed to allocate vDPA structure");
+ return PTR_ERR(adapter);
+ }
- adapter = ifcvf_mgmt_dev->adapter;
- vf = &adapter->vf;
- pdev = adapter->pdev;
+ ifcvf_mgmt_dev->adapter = adapter;
+ adapter->pdev = pdev;
+ adapter->vdpa.dma_dev = &pdev->dev;
+ adapter->vdpa.mdev = mdev;
+ adapter->vf = vf;
vdpa_dev = &adapter->vdpa;
+ device_features = vf->hw_features;
+ if (config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) {
+ if (config->device_features & ~device_features) {
+ IFCVF_ERR(pdev, "The provisioned features 0x%llx are not supported by this device with features 0x%llx\n",
+ config->device_features, device_features);
+ return -EINVAL;
+ }
+ device_features &= config->device_features;
+ }
+ vf->dev_features = device_features;
+
if (name)
ret = dev_set_name(&vdpa_dev->dev, "%s", name);
else
@@ -781,7 +789,6 @@ static int ifcvf_vdpa_dev_add(struct vdpa_mgmt_dev *mdev, const char *name,
return 0;
}
-
static void ifcvf_vdpa_dev_del(struct vdpa_mgmt_dev *mdev, struct vdpa_device *dev)
{
struct ifcvf_vdpa_mgmt_dev *ifcvf_mgmt_dev;
@@ -800,7 +807,6 @@ static int ifcvf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
{
struct ifcvf_vdpa_mgmt_dev *ifcvf_mgmt_dev;
struct device *dev = &pdev->dev;
- struct ifcvf_adapter *adapter;
struct ifcvf_hw *vf;
u32 dev_type;
int ret, i;
@@ -831,20 +837,16 @@ static int ifcvf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
}
pci_set_master(pdev);
-
- adapter = vdpa_alloc_device(struct ifcvf_adapter, vdpa,
- dev, &ifc_vdpa_ops, 1, 1, NULL, false);
- if (IS_ERR(adapter)) {
- IFCVF_ERR(pdev, "Failed to allocate vDPA structure");
- return PTR_ERR(adapter);
+ ifcvf_mgmt_dev = kzalloc(sizeof(struct ifcvf_vdpa_mgmt_dev), GFP_KERNEL);
+ if (!ifcvf_mgmt_dev) {
+ IFCVF_ERR(pdev, "Failed to alloc memory for the vDPA management device\n");
+ return -ENOMEM;
}
- vf = &adapter->vf;
+ vf = &ifcvf_mgmt_dev->vf;
vf->dev_type = get_dev_type(pdev);
vf->base = pcim_iomap_table(pdev);
-
- adapter->pdev = pdev;
- adapter->vdpa.dma_dev = &pdev->dev;
+ vf->pdev = pdev;
ret = ifcvf_init_hw(vf, pdev);
if (ret) {
@@ -858,16 +860,6 @@ static int ifcvf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
vf->hw_features = ifcvf_get_hw_features(vf);
vf->config_size = ifcvf_get_config_size(vf);
- ifcvf_mgmt_dev = kzalloc(sizeof(struct ifcvf_vdpa_mgmt_dev), GFP_KERNEL);
- if (!ifcvf_mgmt_dev) {
- IFCVF_ERR(pdev, "Failed to alloc memory for the vDPA management device\n");
- return -ENOMEM;
- }
-
- ifcvf_mgmt_dev->mdev.ops = &ifcvf_vdpa_mgmt_dev_ops;
- ifcvf_mgmt_dev->mdev.device = dev;
- ifcvf_mgmt_dev->adapter = adapter;
-
dev_type = get_dev_type(pdev);
switch (dev_type) {
case VIRTIO_ID_NET:
@@ -882,11 +874,11 @@ static int ifcvf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
goto err;
}
+ ifcvf_mgmt_dev->mdev.ops = &ifcvf_vdpa_mgmt_dev_ops;
+ ifcvf_mgmt_dev->mdev.device = dev;
ifcvf_mgmt_dev->mdev.max_supported_vqs = vf->nr_vring;
ifcvf_mgmt_dev->mdev.supported_features = vf->hw_features;
-
- adapter->vdpa.mdev = &ifcvf_mgmt_dev->mdev;
-
+ ifcvf_mgmt_dev->mdev.config_attr_mask = (1 << VDPA_ATTR_DEV_FEATURES);
ret = vdpa_mgmtdev_register(&ifcvf_mgmt_dev->mdev);
if (ret) {
diff --git a/drivers/vdpa/mlx5/Makefile b/drivers/vdpa/mlx5/Makefile
index f717978c83bf..e791394c33e3 100644
--- a/drivers/vdpa/mlx5/Makefile
+++ b/drivers/vdpa/mlx5/Makefile
@@ -1,4 +1,4 @@
subdir-ccflags-y += -I$(srctree)/drivers/vdpa/mlx5/core
obj-$(CONFIG_MLX5_VDPA_NET) += mlx5_vdpa.o
-mlx5_vdpa-$(CONFIG_MLX5_VDPA_NET) += net/mlx5_vnet.o core/resources.o core/mr.o
+mlx5_vdpa-$(CONFIG_MLX5_VDPA_NET) += net/mlx5_vnet.o core/resources.o core/mr.o net/debug.o
diff --git a/drivers/vdpa/mlx5/core/mr.c b/drivers/vdpa/mlx5/core/mr.c
index 0a1e0b0dc37e..03e543229791 100644
--- a/drivers/vdpa/mlx5/core/mr.c
+++ b/drivers/vdpa/mlx5/core/mr.c
@@ -503,7 +503,6 @@ void mlx5_vdpa_destroy_mr(struct mlx5_vdpa_dev *mvdev)
else
destroy_dma_mr(mvdev, mr);
- memset(mr, 0, sizeof(*mr));
mr->initialized = false;
out:
mutex_unlock(&mr->mkey_mtx);
diff --git a/drivers/vdpa/mlx5/core/resources.c b/drivers/vdpa/mlx5/core/resources.c
index 9800f9bec225..d5a59c9035fb 100644
--- a/drivers/vdpa/mlx5/core/resources.c
+++ b/drivers/vdpa/mlx5/core/resources.c
@@ -213,7 +213,7 @@ int mlx5_vdpa_create_mkey(struct mlx5_vdpa_dev *mvdev, u32 *mkey, u32 *in,
return err;
mkey_index = MLX5_GET(create_mkey_out, lout, mkey_index);
- *mkey |= mlx5_idx_to_mkey(mkey_index);
+ *mkey = mlx5_idx_to_mkey(mkey_index);
return 0;
}
@@ -233,6 +233,7 @@ static int init_ctrl_vq(struct mlx5_vdpa_dev *mvdev)
if (!mvdev->cvq.iotlb)
return -ENOMEM;
+ spin_lock_init(&mvdev->cvq.iommu_lock);
vringh_set_iotlb(&mvdev->cvq.vring, mvdev->cvq.iotlb, &mvdev->cvq.iommu_lock);
return 0;
diff --git a/drivers/vdpa/mlx5/net/debug.c b/drivers/vdpa/mlx5/net/debug.c
new file mode 100644
index 000000000000..60d6ac68cdc4
--- /dev/null
+++ b/drivers/vdpa/mlx5/net/debug.c
@@ -0,0 +1,152 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
+
+#include <linux/debugfs.h>
+#include <linux/mlx5/fs.h>
+#include "mlx5_vnet.h"
+
+static int tirn_show(struct seq_file *file, void *priv)
+{
+ struct mlx5_vdpa_net *ndev = file->private;
+
+ seq_printf(file, "0x%x\n", ndev->res.tirn);
+ return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(tirn);
+
+void mlx5_vdpa_remove_tirn(struct mlx5_vdpa_net *ndev)
+{
+ if (ndev->debugfs)
+ debugfs_remove(ndev->res.tirn_dent);
+}
+
+void mlx5_vdpa_add_tirn(struct mlx5_vdpa_net *ndev)
+{
+ ndev->res.tirn_dent = debugfs_create_file("tirn", 0444, ndev->rx_dent,
+ ndev, &tirn_fops);
+}
+
+static int rx_flow_table_show(struct seq_file *file, void *priv)
+{
+ struct mlx5_vdpa_net *ndev = file->private;
+
+ seq_printf(file, "0x%x\n", mlx5_flow_table_id(ndev->rxft));
+ return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(rx_flow_table);
+
+void mlx5_vdpa_remove_rx_flow_table(struct mlx5_vdpa_net *ndev)
+{
+ if (ndev->debugfs)
+ debugfs_remove(ndev->rx_table_dent);
+}
+
+void mlx5_vdpa_add_rx_flow_table(struct mlx5_vdpa_net *ndev)
+{
+ ndev->rx_table_dent = debugfs_create_file("table_id", 0444, ndev->rx_dent,
+ ndev, &rx_flow_table_fops);
+}
+
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+static int packets_show(struct seq_file *file, void *priv)
+{
+ struct mlx5_vdpa_counter *counter = file->private;
+ u64 packets;
+ u64 bytes;
+ int err;
+
+ err = mlx5_fc_query(counter->mdev, counter->counter, &packets, &bytes);
+ if (err)
+ return err;
+
+ seq_printf(file, "0x%llx\n", packets);
+ return 0;
+}
+
+static int bytes_show(struct seq_file *file, void *priv)
+{
+ struct mlx5_vdpa_counter *counter = file->private;
+ u64 packets;
+ u64 bytes;
+ int err;
+
+ err = mlx5_fc_query(counter->mdev, counter->counter, &packets, &bytes);
+ if (err)
+ return err;
+
+ seq_printf(file, "0x%llx\n", bytes);
+ return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(packets);
+DEFINE_SHOW_ATTRIBUTE(bytes);
+
+static void add_counter_node(struct mlx5_vdpa_counter *counter,
+ struct dentry *parent)
+{
+ debugfs_create_file("packets", 0444, parent, counter,
+ &packets_fops);
+ debugfs_create_file("bytes", 0444, parent, counter,
+ &bytes_fops);
+}
+
+void mlx5_vdpa_add_rx_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node)
+{
+ static const char *ut = "untagged";
+ char vidstr[9];
+ u16 vid;
+
+ node->ucast_counter.mdev = ndev->mvdev.mdev;
+ node->mcast_counter.mdev = ndev->mvdev.mdev;
+ if (node->tagged) {
+ vid = key2vid(node->macvlan);
+ snprintf(vidstr, sizeof(vidstr), "0x%x", vid);
+ } else {
+ strcpy(vidstr, ut);
+ }
+
+ node->dent = debugfs_create_dir(vidstr, ndev->rx_dent);
+ if (IS_ERR(node->dent)) {
+ node->dent = NULL;
+ return;
+ }
+
+ node->ucast_counter.dent = debugfs_create_dir("ucast", node->dent);
+ if (IS_ERR(node->ucast_counter.dent))
+ return;
+
+ add_counter_node(&node->ucast_counter, node->ucast_counter.dent);
+
+ node->mcast_counter.dent = debugfs_create_dir("mcast", node->dent);
+ if (IS_ERR(node->mcast_counter.dent))
+ return;
+
+ add_counter_node(&node->mcast_counter, node->mcast_counter.dent);
+}
+
+void mlx5_vdpa_remove_rx_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node)
+{
+ if (node->dent && ndev->debugfs)
+ debugfs_remove_recursive(node->dent);
+}
+#endif
+
+void mlx5_vdpa_add_debugfs(struct mlx5_vdpa_net *ndev)
+{
+ struct mlx5_core_dev *mdev;
+
+ mdev = ndev->mvdev.mdev;
+ ndev->debugfs = debugfs_create_dir(dev_name(&ndev->mvdev.vdev.dev),
+ mlx5_debugfs_get_dev_root(mdev));
+ if (!IS_ERR(ndev->debugfs))
+ ndev->rx_dent = debugfs_create_dir("rx", ndev->debugfs);
+}
+
+void mlx5_vdpa_remove_debugfs(struct dentry *dbg)
+{
+ debugfs_remove_recursive(dbg);
+}
diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index 3a6dbbc6440d..3a0e721aef05 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -18,15 +18,12 @@
#include <linux/mlx5/mlx5_ifc_vdpa.h>
#include <linux/mlx5/mpfs.h>
#include "mlx5_vdpa.h"
+#include "mlx5_vnet.h"
MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
MODULE_DESCRIPTION("Mellanox VDPA driver");
MODULE_LICENSE("Dual BSD/GPL");
-#define to_mlx5_vdpa_ndev(__mvdev) \
- container_of(__mvdev, struct mlx5_vdpa_net, mvdev)
-#define to_mvdev(__vdev) container_of((__vdev), struct mlx5_vdpa_dev, vdev)
-
#define VALID_FEATURES_MASK \
(BIT_ULL(VIRTIO_NET_F_CSUM) | BIT_ULL(VIRTIO_NET_F_GUEST_CSUM) | \
BIT_ULL(VIRTIO_NET_F_CTRL_GUEST_OFFLOADS) | BIT_ULL(VIRTIO_NET_F_MTU) | BIT_ULL(VIRTIO_NET_F_MAC) | \
@@ -50,14 +47,6 @@ MODULE_LICENSE("Dual BSD/GPL");
#define MLX5V_UNTAGGED 0x1000
-struct mlx5_vdpa_net_resources {
- u32 tisn;
- u32 tdn;
- u32 tirn;
- u32 rqtn;
- bool valid;
-};
-
struct mlx5_vdpa_cq_buf {
struct mlx5_frag_buf_ctrl fbc;
struct mlx5_frag_buf frag_buf;
@@ -146,38 +135,6 @@ static bool is_index_valid(struct mlx5_vdpa_dev *mvdev, u16 idx)
return idx <= mvdev->max_idx;
}
-#define MLX5V_MACVLAN_SIZE 256
-
-struct mlx5_vdpa_net {
- struct mlx5_vdpa_dev mvdev;
- struct mlx5_vdpa_net_resources res;
- struct virtio_net_config config;
- struct mlx5_vdpa_virtqueue *vqs;
- struct vdpa_callback *event_cbs;
-
- /* Serialize vq resources creation and destruction. This is required
- * since memory map might change and we need to destroy and create
- * resources while driver in operational.
- */
- struct rw_semaphore reslock;
- struct mlx5_flow_table *rxft;
- bool setup;
- u32 cur_num_vqs;
- u32 rqt_size;
- bool nb_registered;
- struct notifier_block nb;
- struct vdpa_callback config_cb;
- struct mlx5_vdpa_wq_ent cvq_ent;
- struct hlist_head macvlan_hash[MLX5V_MACVLAN_SIZE];
-};
-
-struct macvlan_node {
- struct hlist_node hlist;
- struct mlx5_flow_handle *ucast_rule;
- struct mlx5_flow_handle *mcast_rule;
- u64 macvlan;
-};
-
static void free_resources(struct mlx5_vdpa_net *ndev);
static void init_mvqs(struct mlx5_vdpa_net *ndev);
static int setup_driver(struct mlx5_vdpa_dev *mvdev);
@@ -1431,36 +1388,85 @@ static int create_tir(struct mlx5_vdpa_net *ndev)
err = mlx5_vdpa_create_tir(&ndev->mvdev, in, &ndev->res.tirn);
kfree(in);
+ if (err)
+ return err;
+
+ mlx5_vdpa_add_tirn(ndev);
return err;
}
static void destroy_tir(struct mlx5_vdpa_net *ndev)
{
+ mlx5_vdpa_remove_tirn(ndev);
mlx5_vdpa_destroy_tir(&ndev->mvdev, ndev->res.tirn);
}
#define MAX_STEERING_ENT 0x8000
#define MAX_STEERING_GROUPS 2
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+ #define NUM_DESTS 2
+#else
+ #define NUM_DESTS 1
+#endif
+
+static int add_steering_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node,
+ struct mlx5_flow_act *flow_act,
+ struct mlx5_flow_destination *dests)
+{
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+ int err;
+
+ node->ucast_counter.counter = mlx5_fc_create(ndev->mvdev.mdev, false);
+ if (IS_ERR(node->ucast_counter.counter))
+ return PTR_ERR(node->ucast_counter.counter);
+
+ node->mcast_counter.counter = mlx5_fc_create(ndev->mvdev.mdev, false);
+ if (IS_ERR(node->mcast_counter.counter)) {
+ err = PTR_ERR(node->mcast_counter.counter);
+ goto err_mcast_counter;
+ }
+
+ dests[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
+ flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_COUNT;
+ return 0;
+
+err_mcast_counter:
+ mlx5_fc_destroy(ndev->mvdev.mdev, node->ucast_counter.counter);
+ return err;
+#else
+ return 0;
+#endif
+}
+
+static void remove_steering_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node)
+{
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+ mlx5_fc_destroy(ndev->mvdev.mdev, node->mcast_counter.counter);
+ mlx5_fc_destroy(ndev->mvdev.mdev, node->ucast_counter.counter);
+#endif
+}
+
static int mlx5_vdpa_add_mac_vlan_rules(struct mlx5_vdpa_net *ndev, u8 *mac,
- u16 vid, bool tagged,
- struct mlx5_flow_handle **ucast,
- struct mlx5_flow_handle **mcast)
+ struct macvlan_node *node)
{
- struct mlx5_flow_destination dest = {};
+ struct mlx5_flow_destination dests[NUM_DESTS] = {};
struct mlx5_flow_act flow_act = {};
- struct mlx5_flow_handle *rule;
struct mlx5_flow_spec *spec;
void *headers_c;
void *headers_v;
u8 *dmac_c;
u8 *dmac_v;
int err;
+ u16 vid;
spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
if (!spec)
return -ENOMEM;
+ vid = key2vid(node->macvlan);
spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers);
headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers);
@@ -1472,44 +1478,58 @@ static int mlx5_vdpa_add_mac_vlan_rules(struct mlx5_vdpa_net *ndev, u8 *mac,
MLX5_SET(fte_match_set_lyr_2_4, headers_c, cvlan_tag, 1);
MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, first_vid);
}
- if (tagged) {
+ if (node->tagged) {
MLX5_SET(fte_match_set_lyr_2_4, headers_v, cvlan_tag, 1);
MLX5_SET(fte_match_set_lyr_2_4, headers_v, first_vid, vid);
}
flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
- dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR;
- dest.tir_num = ndev->res.tirn;
- rule = mlx5_add_flow_rules(ndev->rxft, spec, &flow_act, &dest, 1);
- if (IS_ERR(rule))
- return PTR_ERR(rule);
+ dests[0].type = MLX5_FLOW_DESTINATION_TYPE_TIR;
+ dests[0].tir_num = ndev->res.tirn;
+ err = add_steering_counters(ndev, node, &flow_act, dests);
+ if (err)
+ goto out_free;
+
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+ dests[1].counter_id = mlx5_fc_id(node->ucast_counter.counter);
+#endif
+ node->ucast_rule = mlx5_add_flow_rules(ndev->rxft, spec, &flow_act, dests, NUM_DESTS);
+ if (IS_ERR(node->ucast_rule)) {
+ err = PTR_ERR(node->ucast_rule);
+ goto err_ucast;
+ }
- *ucast = rule;
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+ dests[1].counter_id = mlx5_fc_id(node->mcast_counter.counter);
+#endif
memset(dmac_c, 0, ETH_ALEN);
memset(dmac_v, 0, ETH_ALEN);
dmac_c[0] = 1;
dmac_v[0] = 1;
- rule = mlx5_add_flow_rules(ndev->rxft, spec, &flow_act, &dest, 1);
- kvfree(spec);
- if (IS_ERR(rule)) {
- err = PTR_ERR(rule);
+ node->mcast_rule = mlx5_add_flow_rules(ndev->rxft, spec, &flow_act, dests, NUM_DESTS);
+ if (IS_ERR(node->mcast_rule)) {
+ err = PTR_ERR(node->mcast_rule);
goto err_mcast;
}
-
- *mcast = rule;
+ kvfree(spec);
+ mlx5_vdpa_add_rx_counters(ndev, node);
return 0;
err_mcast:
- mlx5_del_flow_rules(*ucast);
+ mlx5_del_flow_rules(node->ucast_rule);
+err_ucast:
+ remove_steering_counters(ndev, node);
+out_free:
+ kvfree(spec);
return err;
}
static void mlx5_vdpa_del_mac_vlan_rules(struct mlx5_vdpa_net *ndev,
- struct mlx5_flow_handle *ucast,
- struct mlx5_flow_handle *mcast)
+ struct macvlan_node *node)
{
- mlx5_del_flow_rules(ucast);
- mlx5_del_flow_rules(mcast);
+ mlx5_vdpa_remove_rx_counters(ndev, node);
+ mlx5_del_flow_rules(node->ucast_rule);
+ mlx5_del_flow_rules(node->mcast_rule);
}
static u64 search_val(u8 *mac, u16 vlan, bool tagged)
@@ -1543,14 +1563,14 @@ static struct macvlan_node *mac_vlan_lookup(struct mlx5_vdpa_net *ndev, u64 valu
return NULL;
}
-static int mac_vlan_add(struct mlx5_vdpa_net *ndev, u8 *mac, u16 vlan, bool tagged) // vlan -> vid
+static int mac_vlan_add(struct mlx5_vdpa_net *ndev, u8 *mac, u16 vid, bool tagged)
{
struct macvlan_node *ptr;
u64 val;
u32 idx;
int err;
- val = search_val(mac, vlan, tagged);
+ val = search_val(mac, vid, tagged);
if (mac_vlan_lookup(ndev, val))
return -EEXIST;
@@ -1558,12 +1578,13 @@ static int mac_vlan_add(struct mlx5_vdpa_net *ndev, u8 *mac, u16 vlan, bool tagg
if (!ptr)
return -ENOMEM;
- err = mlx5_vdpa_add_mac_vlan_rules(ndev, ndev->config.mac, vlan, tagged,
- &ptr->ucast_rule, &ptr->mcast_rule);
+ ptr->tagged = tagged;
+ ptr->macvlan = val;
+ ptr->ndev = ndev;
+ err = mlx5_vdpa_add_mac_vlan_rules(ndev, ndev->config.mac, ptr);
if (err)
goto err_add;
- ptr->macvlan = val;
idx = hash_64(val, 8);
hlist_add_head(&ptr->hlist, &ndev->macvlan_hash[idx]);
return 0;
@@ -1582,7 +1603,8 @@ static void mac_vlan_del(struct mlx5_vdpa_net *ndev, u8 *mac, u16 vlan, bool tag
return;
hlist_del(&ptr->hlist);
- mlx5_vdpa_del_mac_vlan_rules(ndev, ptr->ucast_rule, ptr->mcast_rule);
+ mlx5_vdpa_del_mac_vlan_rules(ndev, ptr);
+ remove_steering_counters(ndev, ptr);
kfree(ptr);
}
@@ -1595,7 +1617,8 @@ static void clear_mac_vlan_table(struct mlx5_vdpa_net *ndev)
for (i = 0; i < MLX5V_MACVLAN_SIZE; i++) {
hlist_for_each_entry_safe(pos, n, &ndev->macvlan_hash[i], hlist) {
hlist_del(&pos->hlist);
- mlx5_vdpa_del_mac_vlan_rules(ndev, pos->ucast_rule, pos->mcast_rule);
+ mlx5_vdpa_del_mac_vlan_rules(ndev, pos);
+ remove_steering_counters(ndev, pos);
kfree(pos);
}
}
@@ -1621,6 +1644,7 @@ static int setup_steering(struct mlx5_vdpa_net *ndev)
mlx5_vdpa_warn(&ndev->mvdev, "failed to create flow table\n");
return PTR_ERR(ndev->rxft);
}
+ mlx5_vdpa_add_rx_flow_table(ndev);
err = mac_vlan_add(ndev, ndev->config.mac, 0, false);
if (err)
@@ -1629,6 +1653,7 @@ static int setup_steering(struct mlx5_vdpa_net *ndev)
return 0;
err_add:
+ mlx5_vdpa_remove_rx_flow_table(ndev);
mlx5_destroy_flow_table(ndev->rxft);
return err;
}
@@ -1636,6 +1661,7 @@ err_add:
static void teardown_steering(struct mlx5_vdpa_net *ndev)
{
clear_mac_vlan_table(ndev);
+ mlx5_vdpa_remove_rx_flow_table(ndev);
mlx5_destroy_flow_table(ndev->rxft);
}
@@ -2183,6 +2209,7 @@ static u64 get_supported_features(struct mlx5_core_dev *mdev)
mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_STATUS);
mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_MTU);
mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_CTRL_VLAN);
+ mlx_vdpa_features |= BIT_ULL(VIRTIO_NET_F_MAC);
return mlx_vdpa_features;
}
@@ -2655,6 +2682,16 @@ static int mlx5_vdpa_set_map(struct vdpa_device *vdev, unsigned int asid,
return err;
}
+static struct device *mlx5_get_vq_dma_dev(struct vdpa_device *vdev, u16 idx)
+{
+ struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev);
+
+ if (is_ctrl_vq_idx(mvdev, idx))
+ return &vdev->dev;
+
+ return mvdev->vdev.dma_dev;
+}
+
static void mlx5_vdpa_free(struct vdpa_device *vdev)
{
struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev);
@@ -2870,6 +2907,7 @@ static const struct vdpa_config_ops mlx5_vdpa_ops = {
.get_generation = mlx5_vdpa_get_generation,
.set_map = mlx5_vdpa_set_map,
.set_group_asid = mlx5_set_group_asid,
+ .get_vq_dma_dev = mlx5_get_vq_dma_dev,
.free = mlx5_vdpa_free,
.suspend = mlx5_vdpa_suspend,
};
@@ -3009,6 +3047,8 @@ static int event_handler(struct notifier_block *nb, unsigned long event, void *p
struct mlx5_vdpa_wq_ent *wqent;
if (event == MLX5_EVENT_TYPE_PORT_CHANGE) {
+ if (!(ndev->mvdev.actual_features & BIT_ULL(VIRTIO_NET_F_STATUS)))
+ return NOTIFY_DONE;
switch (eqe->sub_type) {
case MLX5_PORT_CHANGE_SUBTYPE_DOWN:
case MLX5_PORT_CHANGE_SUBTYPE_ACTIVE:
@@ -3060,6 +3100,7 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
struct mlx5_vdpa_dev *mvdev;
struct mlx5_vdpa_net *ndev;
struct mlx5_core_dev *mdev;
+ u64 device_features;
u32 max_vqs;
u16 mtu;
int err;
@@ -3068,6 +3109,24 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
return -ENOSPC;
mdev = mgtdev->madev->mdev;
+ device_features = mgtdev->mgtdev.supported_features;
+ if (add_config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) {
+ if (add_config->device_features & ~device_features) {
+ dev_warn(mdev->device,
+ "The provisioned features 0x%llx are not supported by this device with features 0x%llx\n",
+ add_config->device_features, device_features);
+ return -EINVAL;
+ }
+ device_features &= add_config->device_features;
+ }
+ if (!(device_features & BIT_ULL(VIRTIO_F_VERSION_1) &&
+ device_features & BIT_ULL(VIRTIO_F_ACCESS_PLATFORM))) {
+ dev_warn(mdev->device,
+ "Must provision minimum features 0x%llx for this device",
+ BIT_ULL(VIRTIO_F_VERSION_1) | BIT_ULL(VIRTIO_F_ACCESS_PLATFORM));
+ return -EOPNOTSUPP;
+ }
+
if (!(MLX5_CAP_DEV_VDPA_EMULATION(mdev, virtio_queue_type) &
MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT)) {
dev_warn(mdev->device, "missing support for split virtqueues\n");
@@ -3096,7 +3155,6 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
if (IS_ERR(ndev))
return PTR_ERR(ndev);
- ndev->mvdev.mlx_features = mgtdev->mgtdev.supported_features;
ndev->mvdev.max_vqs = max_vqs;
mvdev = &ndev->mvdev;
mvdev->mdev = mdev;
@@ -3118,20 +3176,26 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
goto err_alloc;
}
- err = query_mtu(mdev, &mtu);
- if (err)
- goto err_alloc;
+ if (device_features & BIT_ULL(VIRTIO_NET_F_MTU)) {
+ err = query_mtu(mdev, &mtu);
+ if (err)
+ goto err_alloc;
- ndev->config.mtu = cpu_to_mlx5vdpa16(mvdev, mtu);
+ ndev->config.mtu = cpu_to_mlx5vdpa16(mvdev, mtu);
+ }
- if (get_link_state(mvdev))
- ndev->config.status |= cpu_to_mlx5vdpa16(mvdev, VIRTIO_NET_S_LINK_UP);
- else
- ndev->config.status &= cpu_to_mlx5vdpa16(mvdev, ~VIRTIO_NET_S_LINK_UP);
+ if (device_features & BIT_ULL(VIRTIO_NET_F_STATUS)) {
+ if (get_link_state(mvdev))
+ ndev->config.status |= cpu_to_mlx5vdpa16(mvdev, VIRTIO_NET_S_LINK_UP);
+ else
+ ndev->config.status &= cpu_to_mlx5vdpa16(mvdev, ~VIRTIO_NET_S_LINK_UP);
+ }
if (add_config->mask & (1 << VDPA_ATTR_DEV_NET_CFG_MACADDR)) {
memcpy(ndev->config.mac, add_config->net.mac, ETH_ALEN);
- } else {
+ /* No bother setting mac address in config if not going to provision _F_MAC */
+ } else if ((add_config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) == 0 ||
+ device_features & BIT_ULL(VIRTIO_NET_F_MAC)) {
err = mlx5_query_nic_vport_mac_address(mdev, 0, 0, config->mac);
if (err)
goto err_alloc;
@@ -3142,11 +3206,26 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
err = mlx5_mpfs_add_mac(pfmdev, config->mac);
if (err)
goto err_alloc;
-
- ndev->mvdev.mlx_features |= BIT_ULL(VIRTIO_NET_F_MAC);
+ } else if ((add_config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) == 0) {
+ /*
+ * We used to clear _F_MAC feature bit if seeing
+ * zero mac address when device features are not
+ * specifically provisioned. Keep the behaviour
+ * so old scripts do not break.
+ */
+ device_features &= ~BIT_ULL(VIRTIO_NET_F_MAC);
+ } else if (device_features & BIT_ULL(VIRTIO_NET_F_MAC)) {
+ /* Don't provision zero mac address for _F_MAC */
+ mlx5_vdpa_warn(&ndev->mvdev,
+ "No mac address provisioned?\n");
+ err = -EINVAL;
+ goto err_alloc;
}
- config->max_virtqueue_pairs = cpu_to_mlx5vdpa16(mvdev, max_vqs / 2);
+ if (device_features & BIT_ULL(VIRTIO_NET_F_MQ))
+ config->max_virtqueue_pairs = cpu_to_mlx5vdpa16(mvdev, max_vqs / 2);
+
+ ndev->mvdev.mlx_features = device_features;
mvdev->vdev.dma_dev = &mdev->pdev->dev;
err = mlx5_vdpa_alloc_resources(&ndev->mvdev);
if (err)
@@ -3178,6 +3257,7 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name,
if (err)
goto err_reg;
+ mlx5_vdpa_add_debugfs(ndev);
mgtdev->ndev = ndev;
return 0;
@@ -3204,6 +3284,8 @@ static void mlx5_vdpa_dev_del(struct vdpa_mgmt_dev *v_mdev, struct vdpa_device *
struct mlx5_vdpa_net *ndev = to_mlx5_vdpa_ndev(mvdev);
struct workqueue_struct *wq;
+ mlx5_vdpa_remove_debugfs(ndev->debugfs);
+ ndev->debugfs = NULL;
if (ndev->nb_registered) {
ndev->nb_registered = false;
mlx5_notifier_unregister(mvdev->mdev, &ndev->nb);
@@ -3243,7 +3325,8 @@ static int mlx5v_probe(struct auxiliary_device *adev,
mgtdev->mgtdev.id_table = id_table;
mgtdev->mgtdev.config_attr_mask = BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MACADDR) |
BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MAX_VQP) |
- BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MTU);
+ BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MTU) |
+ BIT_ULL(VDPA_ATTR_DEV_FEATURES);
mgtdev->mgtdev.max_supported_vqs =
MLX5_CAP_DEV_VDPA_EMULATION(mdev, max_num_virtio_queues) + 1;
mgtdev->mgtdev.supported_features = get_supported_features(mdev);
diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.h b/drivers/vdpa/mlx5/net/mlx5_vnet.h
new file mode 100644
index 000000000000..c90a89e1de4d
--- /dev/null
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.h
@@ -0,0 +1,94 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
+
+#ifndef __MLX5_VNET_H__
+#define __MLX5_VNET_H__
+
+#include "mlx5_vdpa.h"
+
+#define to_mlx5_vdpa_ndev(__mvdev) \
+ container_of(__mvdev, struct mlx5_vdpa_net, mvdev)
+#define to_mvdev(__vdev) container_of((__vdev), struct mlx5_vdpa_dev, vdev)
+
+struct mlx5_vdpa_net_resources {
+ u32 tisn;
+ u32 tdn;
+ u32 tirn;
+ u32 rqtn;
+ bool valid;
+ struct dentry *tirn_dent;
+};
+
+#define MLX5V_MACVLAN_SIZE 256
+
+static inline u16 key2vid(u64 key)
+{
+ return (u16)(key >> 48) & 0xfff;
+}
+
+struct mlx5_vdpa_net {
+ struct mlx5_vdpa_dev mvdev;
+ struct mlx5_vdpa_net_resources res;
+ struct virtio_net_config config;
+ struct mlx5_vdpa_virtqueue *vqs;
+ struct vdpa_callback *event_cbs;
+
+ /* Serialize vq resources creation and destruction. This is required
+ * since memory map might change and we need to destroy and create
+ * resources while driver in operational.
+ */
+ struct rw_semaphore reslock;
+ struct mlx5_flow_table *rxft;
+ struct dentry *rx_dent;
+ struct dentry *rx_table_dent;
+ bool setup;
+ u32 cur_num_vqs;
+ u32 rqt_size;
+ bool nb_registered;
+ struct notifier_block nb;
+ struct vdpa_callback config_cb;
+ struct mlx5_vdpa_wq_ent cvq_ent;
+ struct hlist_head macvlan_hash[MLX5V_MACVLAN_SIZE];
+ struct dentry *debugfs;
+};
+
+struct mlx5_vdpa_counter {
+ struct mlx5_fc *counter;
+ struct dentry *dent;
+ struct mlx5_core_dev *mdev;
+};
+
+struct macvlan_node {
+ struct hlist_node hlist;
+ struct mlx5_flow_handle *ucast_rule;
+ struct mlx5_flow_handle *mcast_rule;
+ u64 macvlan;
+ struct mlx5_vdpa_net *ndev;
+ bool tagged;
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+ struct dentry *dent;
+ struct mlx5_vdpa_counter ucast_counter;
+ struct mlx5_vdpa_counter mcast_counter;
+#endif
+};
+
+void mlx5_vdpa_add_debugfs(struct mlx5_vdpa_net *ndev);
+void mlx5_vdpa_remove_debugfs(struct dentry *dbg);
+void mlx5_vdpa_add_rx_flow_table(struct mlx5_vdpa_net *ndev);
+void mlx5_vdpa_remove_rx_flow_table(struct mlx5_vdpa_net *ndev);
+void mlx5_vdpa_add_tirn(struct mlx5_vdpa_net *ndev);
+void mlx5_vdpa_remove_tirn(struct mlx5_vdpa_net *ndev);
+#if defined(CONFIG_MLX5_VDPA_STEERING_DEBUG)
+void mlx5_vdpa_add_rx_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node);
+void mlx5_vdpa_remove_rx_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node);
+#else
+static inline void mlx5_vdpa_add_rx_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node) {}
+static inline void mlx5_vdpa_remove_rx_counters(struct mlx5_vdpa_net *ndev,
+ struct macvlan_node *node) {}
+#endif
+
+
+#endif /* __MLX5_VNET_H__ */
diff --git a/drivers/vdpa/solidrun/Makefile b/drivers/vdpa/solidrun/Makefile
new file mode 100644
index 000000000000..c0aa3415bf7b
--- /dev/null
+++ b/drivers/vdpa/solidrun/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_SNET_VDPA) += snet_vdpa.o
+snet_vdpa-$(CONFIG_SNET_VDPA) += snet_main.o
+ifdef CONFIG_HWMON
+snet_vdpa-$(CONFIG_SNET_VDPA) += snet_hwmon.o
+endif
diff --git a/drivers/vdpa/solidrun/snet_hwmon.c b/drivers/vdpa/solidrun/snet_hwmon.c
new file mode 100644
index 000000000000..e695e36ff753
--- /dev/null
+++ b/drivers/vdpa/solidrun/snet_hwmon.c
@@ -0,0 +1,188 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * SolidRun DPU driver for control plane
+ *
+ * Copyright (C) 2022 SolidRun
+ *
+ * Author: Alvaro Karsz <alvaro.karsz@solid-run.com>
+ *
+ */
+#include <linux/hwmon.h>
+
+#include "snet_vdpa.h"
+
+/* Monitor offsets */
+#define SNET_MON_TMP0_IN_OFF 0x00
+#define SNET_MON_TMP0_MAX_OFF 0x08
+#define SNET_MON_TMP0_CRIT_OFF 0x10
+#define SNET_MON_TMP1_IN_OFF 0x18
+#define SNET_MON_TMP1_CRIT_OFF 0x20
+#define SNET_MON_CURR_IN_OFF 0x28
+#define SNET_MON_CURR_MAX_OFF 0x30
+#define SNET_MON_CURR_CRIT_OFF 0x38
+#define SNET_MON_PWR_IN_OFF 0x40
+#define SNET_MON_VOLT_IN_OFF 0x48
+#define SNET_MON_VOLT_CRIT_OFF 0x50
+#define SNET_MON_VOLT_LCRIT_OFF 0x58
+
+static void snet_hwmon_read_reg(struct psnet *psnet, u32 reg, long *out)
+{
+ *out = psnet_read64(psnet, psnet->cfg.hwmon_off + reg);
+}
+
+static umode_t snet_howmon_is_visible(const void *data,
+ enum hwmon_sensor_types type,
+ u32 attr, int channel)
+{
+ return 0444;
+}
+
+static int snet_howmon_read(struct device *dev, enum hwmon_sensor_types type,
+ u32 attr, int channel, long *val)
+{
+ struct psnet *psnet = dev_get_drvdata(dev);
+ int ret = 0;
+
+ switch (type) {
+ case hwmon_in:
+ switch (attr) {
+ case hwmon_in_lcrit:
+ snet_hwmon_read_reg(psnet, SNET_MON_VOLT_LCRIT_OFF, val);
+ break;
+ case hwmon_in_crit:
+ snet_hwmon_read_reg(psnet, SNET_MON_VOLT_CRIT_OFF, val);
+ break;
+ case hwmon_in_input:
+ snet_hwmon_read_reg(psnet, SNET_MON_VOLT_IN_OFF, val);
+ break;
+ default:
+ ret = -EOPNOTSUPP;
+ break;
+ }
+ break;
+
+ case hwmon_power:
+ switch (attr) {
+ case hwmon_power_input:
+ snet_hwmon_read_reg(psnet, SNET_MON_PWR_IN_OFF, val);
+ break;
+
+ default:
+ ret = -EOPNOTSUPP;
+ break;
+ }
+ break;
+
+ case hwmon_curr:
+ switch (attr) {
+ case hwmon_curr_input:
+ snet_hwmon_read_reg(psnet, SNET_MON_CURR_IN_OFF, val);
+ break;
+ case hwmon_curr_max:
+ snet_hwmon_read_reg(psnet, SNET_MON_CURR_MAX_OFF, val);
+ break;
+ case hwmon_curr_crit:
+ snet_hwmon_read_reg(psnet, SNET_MON_CURR_CRIT_OFF, val);
+ break;
+ default:
+ ret = -EOPNOTSUPP;
+ break;
+ }
+ break;
+
+ case hwmon_temp:
+ switch (attr) {
+ case hwmon_temp_input:
+ if (channel == 0)
+ snet_hwmon_read_reg(psnet, SNET_MON_TMP0_IN_OFF, val);
+ else
+ snet_hwmon_read_reg(psnet, SNET_MON_TMP1_IN_OFF, val);
+ break;
+ case hwmon_temp_max:
+ if (channel == 0)
+ snet_hwmon_read_reg(psnet, SNET_MON_TMP0_MAX_OFF, val);
+ else
+ ret = -EOPNOTSUPP;
+ break;
+ case hwmon_temp_crit:
+ if (channel == 0)
+ snet_hwmon_read_reg(psnet, SNET_MON_TMP0_CRIT_OFF, val);
+ else
+ snet_hwmon_read_reg(psnet, SNET_MON_TMP1_CRIT_OFF, val);
+ break;
+
+ default:
+ ret = -EOPNOTSUPP;
+ break;
+ }
+ break;
+
+ default:
+ ret = -EOPNOTSUPP;
+ break;
+ }
+ return ret;
+}
+
+static int snet_hwmon_read_string(struct device *dev,
+ enum hwmon_sensor_types type, u32 attr,
+ int channel, const char **str)
+{
+ int ret = 0;
+
+ switch (type) {
+ case hwmon_in:
+ *str = "main_vin";
+ break;
+ case hwmon_power:
+ *str = "soc_pin";
+ break;
+ case hwmon_curr:
+ *str = "soc_iin";
+ break;
+ case hwmon_temp:
+ if (channel == 0)
+ *str = "power_stage_temp";
+ else
+ *str = "ic_junction_temp";
+ break;
+ default:
+ ret = -EOPNOTSUPP;
+ break;
+ }
+ return ret;
+}
+
+static const struct hwmon_ops snet_hwmon_ops = {
+ .is_visible = snet_howmon_is_visible,
+ .read = snet_howmon_read,
+ .read_string = snet_hwmon_read_string
+};
+
+static const struct hwmon_channel_info *snet_hwmon_info[] = {
+ HWMON_CHANNEL_INFO(temp, HWMON_T_INPUT | HWMON_T_MAX | HWMON_T_CRIT | HWMON_T_LABEL,
+ HWMON_T_INPUT | HWMON_T_CRIT | HWMON_T_LABEL),
+ HWMON_CHANNEL_INFO(power, HWMON_P_INPUT | HWMON_P_LABEL),
+ HWMON_CHANNEL_INFO(curr, HWMON_C_INPUT | HWMON_C_MAX | HWMON_C_CRIT | HWMON_C_LABEL),
+ HWMON_CHANNEL_INFO(in, HWMON_I_INPUT | HWMON_I_CRIT | HWMON_I_LCRIT | HWMON_I_LABEL),
+ NULL
+};
+
+static const struct hwmon_chip_info snet_hwmono_info = {
+ .ops = &snet_hwmon_ops,
+ .info = snet_hwmon_info,
+};
+
+/* Create an HW monitor device */
+void psnet_create_hwmon(struct pci_dev *pdev)
+{
+ struct device *hwmon;
+ struct psnet *psnet = pci_get_drvdata(pdev);
+
+ snprintf(psnet->hwmon_name, SNET_NAME_SIZE, "snet_%s", pci_name(pdev));
+ hwmon = devm_hwmon_device_register_with_info(&pdev->dev, psnet->hwmon_name, psnet,
+ &snet_hwmono_info, NULL);
+ /* The monitor is not mandatory, Just alert user in case of an error */
+ if (IS_ERR(hwmon))
+ SNET_WARN(pdev, "Failed to create SNET hwmon, error %ld\n", PTR_ERR(hwmon));
+}
diff --git a/drivers/vdpa/solidrun/snet_main.c b/drivers/vdpa/solidrun/snet_main.c
new file mode 100644
index 000000000000..68de727398ed
--- /dev/null
+++ b/drivers/vdpa/solidrun/snet_main.c
@@ -0,0 +1,1111 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * SolidRun DPU driver for control plane
+ *
+ * Copyright (C) 2022 SolidRun
+ *
+ * Author: Alvaro Karsz <alvaro.karsz@solid-run.com>
+ *
+ */
+#include <linux/iopoll.h>
+
+#include "snet_vdpa.h"
+
+/* SNET DPU device ID */
+#define SNET_DEVICE_ID 0x1000
+/* SNET signature */
+#define SNET_SIGNATURE 0xD0D06363
+/* Max. config version that we can work with */
+#define SNET_CFG_VERSION 0x1
+/* Queue align */
+#define SNET_QUEUE_ALIGNMENT PAGE_SIZE
+/* Kick value to notify that new data is available */
+#define SNET_KICK_VAL 0x1
+#define SNET_CONFIG_OFF 0x0
+/* ACK timeout for a message */
+#define SNET_ACK_TIMEOUT 2000000
+/* How long we are willing to wait for a SNET device */
+#define SNET_DETECT_TIMEOUT 5000000
+/* How long should we wait for the DPU to read our config */
+#define SNET_READ_CFG_TIMEOUT 3000000
+/* Size of configs written to the DPU */
+#define SNET_GENERAL_CFG_LEN 36
+#define SNET_GENERAL_CFG_VQ_LEN 40
+
+enum snet_msg {
+ SNET_MSG_DESTROY = 1,
+};
+
+static struct snet *vdpa_to_snet(struct vdpa_device *vdpa)
+{
+ return container_of(vdpa, struct snet, vdpa);
+}
+
+static int snet_wait_for_msg_ack(struct snet *snet)
+{
+ struct pci_dev *pdev = snet->pdev;
+ int ret;
+ u32 val;
+
+ /* The DPU will clear the messages offset once messages
+ * are processed.
+ */
+ ret = readx_poll_timeout(ioread32, snet->bar + snet->psnet->cfg.msg_off,
+ val, !val, 10, SNET_ACK_TIMEOUT);
+ if (ret)
+ SNET_WARN(pdev, "Timeout waiting for message ACK\n");
+
+ return ret;
+}
+
+/* Sends a message to the DPU.
+ * If blocking is set, the function will return once the
+ * message was processed by the DPU (or timeout).
+ */
+static int snet_send_msg(struct snet *snet, u32 msg, bool blocking)
+{
+ int ret = 0;
+
+ /* Make sure the DPU acked last message before issuing a new one */
+ ret = snet_wait_for_msg_ack(snet);
+ if (ret)
+ return ret;
+
+ /* Write the message */
+ snet_write32(snet, snet->psnet->cfg.msg_off, msg);
+
+ if (blocking)
+ ret = snet_wait_for_msg_ack(snet);
+ else /* If non-blocking, flush the write by issuing a read */
+ snet_read32(snet, snet->psnet->cfg.msg_off);
+
+ return ret;
+}
+
+static irqreturn_t snet_cfg_irq_hndlr(int irq, void *data)
+{
+ struct snet *snet = data;
+ /* Call callback if any */
+ if (snet->cb.callback)
+ return snet->cb.callback(snet->cb.private);
+
+ return IRQ_HANDLED;
+}
+
+static irqreturn_t snet_vq_irq_hndlr(int irq, void *data)
+{
+ struct snet_vq *vq = data;
+ /* Call callback if any */
+ if (vq->cb.callback)
+ return vq->cb.callback(vq->cb.private);
+
+ return IRQ_HANDLED;
+}
+
+static void snet_free_irqs(struct snet *snet)
+{
+ struct psnet *psnet = snet->psnet;
+ struct pci_dev *pdev;
+ u32 i;
+
+ /* Which Device allcoated the IRQs? */
+ if (PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_IRQ_PF))
+ pdev = snet->pdev->physfn;
+ else
+ pdev = snet->pdev;
+
+ /* Free config's IRQ */
+ if (snet->cfg_irq != -1) {
+ devm_free_irq(&pdev->dev, snet->cfg_irq, snet);
+ snet->cfg_irq = -1;
+ }
+ /* Free VQ IRQs */
+ for (i = 0; i < snet->cfg->vq_num; i++) {
+ if (snet->vqs[i] && snet->vqs[i]->irq != -1) {
+ devm_free_irq(&pdev->dev, snet->vqs[i]->irq, snet->vqs[i]);
+ snet->vqs[i]->irq = -1;
+ }
+ }
+
+ /* IRQ vectors are freed when the pci remove callback is called */
+}
+
+static int snet_set_vq_address(struct vdpa_device *vdev, u16 idx, u64 desc_area,
+ u64 driver_area, u64 device_area)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ /* save received parameters in vqueue sturct */
+ snet->vqs[idx]->desc_area = desc_area;
+ snet->vqs[idx]->driver_area = driver_area;
+ snet->vqs[idx]->device_area = device_area;
+
+ return 0;
+}
+
+static void snet_set_vq_num(struct vdpa_device *vdev, u16 idx, u32 num)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ /* save num in vqueue */
+ snet->vqs[idx]->num = num;
+}
+
+static void snet_kick_vq(struct vdpa_device *vdev, u16 idx)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ /* not ready - ignore */
+ if (!snet->vqs[idx]->ready)
+ return;
+
+ iowrite32(SNET_KICK_VAL, snet->vqs[idx]->kick_ptr);
+}
+
+static void snet_set_vq_cb(struct vdpa_device *vdev, u16 idx, struct vdpa_callback *cb)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ snet->vqs[idx]->cb.callback = cb->callback;
+ snet->vqs[idx]->cb.private = cb->private;
+}
+
+static void snet_set_vq_ready(struct vdpa_device *vdev, u16 idx, bool ready)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ snet->vqs[idx]->ready = ready;
+}
+
+static bool snet_get_vq_ready(struct vdpa_device *vdev, u16 idx)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet->vqs[idx]->ready;
+}
+
+static int snet_set_vq_state(struct vdpa_device *vdev, u16 idx, const struct vdpa_vq_state *state)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ /* Setting the VQ state is not supported.
+ * If the asked state is the same as the initial one
+ * we can ignore it.
+ */
+ if (SNET_HAS_FEATURE(snet, VIRTIO_F_RING_PACKED)) {
+ const struct vdpa_vq_state_packed *p = &state->packed;
+
+ if (p->last_avail_counter == 1 && p->last_used_counter == 1 &&
+ p->last_avail_idx == 0 && p->last_used_idx == 0)
+ return 0;
+ } else {
+ const struct vdpa_vq_state_split *s = &state->split;
+
+ if (s->avail_index == 0)
+ return 0;
+ }
+
+ return -EOPNOTSUPP;
+}
+
+static int snet_get_vq_state(struct vdpa_device *vdev, u16 idx, struct vdpa_vq_state *state)
+{
+ /* Not supported */
+ return -EOPNOTSUPP;
+}
+
+static int snet_get_vq_irq(struct vdpa_device *vdev, u16 idx)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet->vqs[idx]->irq;
+}
+
+static u32 snet_get_vq_align(struct vdpa_device *vdev)
+{
+ return (u32)SNET_QUEUE_ALIGNMENT;
+}
+
+static int snet_reset_dev(struct snet *snet)
+{
+ struct pci_dev *pdev = snet->pdev;
+ int ret = 0;
+ u32 i;
+
+ /* If status is 0, nothing to do */
+ if (!snet->status)
+ return 0;
+
+ /* If DPU started, send a destroy message */
+ if (snet->status & VIRTIO_CONFIG_S_DRIVER_OK)
+ ret = snet_send_msg(snet, SNET_MSG_DESTROY, true);
+
+ /* Clear VQs */
+ for (i = 0; i < snet->cfg->vq_num; i++) {
+ if (!snet->vqs[i])
+ continue;
+ snet->vqs[i]->cb.callback = NULL;
+ snet->vqs[i]->cb.private = NULL;
+ snet->vqs[i]->desc_area = 0;
+ snet->vqs[i]->device_area = 0;
+ snet->vqs[i]->driver_area = 0;
+ snet->vqs[i]->ready = false;
+ }
+
+ /* Clear config callback */
+ snet->cb.callback = NULL;
+ snet->cb.private = NULL;
+ /* Free IRQs */
+ snet_free_irqs(snet);
+ /* Reset status */
+ snet->status = 0;
+ snet->dpu_ready = false;
+
+ if (ret)
+ SNET_WARN(pdev, "Incomplete reset to SNET[%u] device\n", snet->sid);
+ else
+ SNET_DBG(pdev, "Reset SNET[%u] device\n", snet->sid);
+
+ return 0;
+}
+
+static int snet_reset(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet_reset_dev(snet);
+}
+
+static size_t snet_get_config_size(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return (size_t)snet->cfg->cfg_size;
+}
+
+static u64 snet_get_features(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet->cfg->features;
+}
+
+static int snet_set_drv_features(struct vdpa_device *vdev, u64 features)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ snet->negotiated_features = snet->cfg->features & features;
+ return 0;
+}
+
+static u64 snet_get_drv_features(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet->negotiated_features;
+}
+
+static u16 snet_get_vq_num_max(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return (u16)snet->cfg->vq_size;
+}
+
+static void snet_set_config_cb(struct vdpa_device *vdev, struct vdpa_callback *cb)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ snet->cb.callback = cb->callback;
+ snet->cb.private = cb->private;
+}
+
+static u32 snet_get_device_id(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet->cfg->virtio_id;
+}
+
+static u32 snet_get_vendor_id(struct vdpa_device *vdev)
+{
+ return (u32)PCI_VENDOR_ID_SOLIDRUN;
+}
+
+static u8 snet_get_status(struct vdpa_device *vdev)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+
+ return snet->status;
+}
+
+static int snet_write_conf(struct snet *snet)
+{
+ u32 off, i, tmp;
+ int ret;
+
+ /* No need to write the config twice */
+ if (snet->dpu_ready)
+ return true;
+
+ /* Snet data :
+ *
+ * General data: SNET_GENERAL_CFG_LEN bytes long
+ * 0 0x4 0x8 0xC 0x10 0x14 0x1C 0x24
+ * | MAGIC NUMBER | CFG VER | SNET SID | NUMBER OF QUEUES | IRQ IDX | FEATURES | RSVD |
+ *
+ * For every VQ: SNET_GENERAL_CFG_VQ_LEN bytes long
+ * 0 0x4 0x8
+ * | VQ SID AND QUEUE SIZE | IRQ Index |
+ * | DESC AREA |
+ * | DEVICE AREA |
+ * | DRIVER AREA |
+ * | RESERVED |
+ *
+ * Magic number should be written last, this is the DPU indication that the data is ready
+ */
+
+ /* Init offset */
+ off = snet->psnet->cfg.host_cfg_off;
+
+ /* Ignore magic number for now */
+ off += 4;
+ snet_write32(snet, off, snet->psnet->negotiated_cfg_ver);
+ off += 4;
+ snet_write32(snet, off, snet->sid);
+ off += 4;
+ snet_write32(snet, off, snet->cfg->vq_num);
+ off += 4;
+ snet_write32(snet, off, snet->cfg_irq_idx);
+ off += 4;
+ snet_write64(snet, off, snet->negotiated_features);
+ off += 8;
+ /* Ignore reserved */
+ off += 8;
+ /* Write VQs */
+ for (i = 0 ; i < snet->cfg->vq_num ; i++) {
+ tmp = (i << 16) | (snet->vqs[i]->num & 0xFFFF);
+ snet_write32(snet, off, tmp);
+ off += 4;
+ snet_write32(snet, off, snet->vqs[i]->irq_idx);
+ off += 4;
+ snet_write64(snet, off, snet->vqs[i]->desc_area);
+ off += 8;
+ snet_write64(snet, off, snet->vqs[i]->device_area);
+ off += 8;
+ snet_write64(snet, off, snet->vqs[i]->driver_area);
+ off += 8;
+ /* Ignore reserved */
+ off += 8;
+ }
+
+ /* Clear snet messages address for this device */
+ snet_write32(snet, snet->psnet->cfg.msg_off, 0);
+ /* Write magic number - data is ready */
+ snet_write32(snet, snet->psnet->cfg.host_cfg_off, SNET_SIGNATURE);
+
+ /* The DPU will ACK the config by clearing the signature */
+ ret = readx_poll_timeout(ioread32, snet->bar + snet->psnet->cfg.host_cfg_off,
+ tmp, !tmp, 10, SNET_READ_CFG_TIMEOUT);
+ if (ret) {
+ SNET_ERR(snet->pdev, "Timeout waiting for the DPU to read the config\n");
+ return false;
+ }
+
+ /* set DPU flag */
+ snet->dpu_ready = true;
+
+ return true;
+}
+
+static int snet_request_irqs(struct pci_dev *pdev, struct snet *snet)
+{
+ int ret, i, irq;
+
+ /* Request config IRQ */
+ irq = pci_irq_vector(pdev, snet->cfg_irq_idx);
+ ret = devm_request_irq(&pdev->dev, irq, snet_cfg_irq_hndlr, 0,
+ snet->cfg_irq_name, snet);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to request IRQ\n");
+ return ret;
+ }
+ snet->cfg_irq = irq;
+
+ /* Request IRQ for every VQ */
+ for (i = 0; i < snet->cfg->vq_num; i++) {
+ irq = pci_irq_vector(pdev, snet->vqs[i]->irq_idx);
+ ret = devm_request_irq(&pdev->dev, irq, snet_vq_irq_hndlr, 0,
+ snet->vqs[i]->irq_name, snet->vqs[i]);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to request IRQ\n");
+ return ret;
+ }
+ snet->vqs[i]->irq = irq;
+ }
+ return 0;
+}
+
+static void snet_set_status(struct vdpa_device *vdev, u8 status)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ struct psnet *psnet = snet->psnet;
+ struct pci_dev *pdev = snet->pdev;
+ int ret;
+ bool pf_irqs;
+
+ if (status == snet->status)
+ return;
+
+ if ((status & VIRTIO_CONFIG_S_DRIVER_OK) &&
+ !(snet->status & VIRTIO_CONFIG_S_DRIVER_OK)) {
+ /* Request IRQs */
+ pf_irqs = PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_IRQ_PF);
+ ret = snet_request_irqs(pf_irqs ? pdev->physfn : pdev, snet);
+ if (ret)
+ goto set_err;
+
+ /* Write config to the DPU */
+ if (snet_write_conf(snet)) {
+ SNET_INFO(pdev, "Create SNET[%u] device\n", snet->sid);
+ } else {
+ snet_free_irqs(snet);
+ goto set_err;
+ }
+ }
+
+ /* Save the new status */
+ snet->status = status;
+ return;
+
+set_err:
+ snet->status |= VIRTIO_CONFIG_S_FAILED;
+}
+
+static void snet_get_config(struct vdpa_device *vdev, unsigned int offset,
+ void *buf, unsigned int len)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ void __iomem *cfg_ptr = snet->cfg->virtio_cfg + offset;
+ u8 *buf_ptr = buf;
+ u32 i;
+
+ /* check for offset error */
+ if (offset + len > snet->cfg->cfg_size)
+ return;
+
+ /* Write into buffer */
+ for (i = 0; i < len; i++)
+ *buf_ptr++ = ioread8(cfg_ptr + i);
+}
+
+static void snet_set_config(struct vdpa_device *vdev, unsigned int offset,
+ const void *buf, unsigned int len)
+{
+ struct snet *snet = vdpa_to_snet(vdev);
+ void __iomem *cfg_ptr = snet->cfg->virtio_cfg + offset;
+ const u8 *buf_ptr = buf;
+ u32 i;
+
+ /* check for offset error */
+ if (offset + len > snet->cfg->cfg_size)
+ return;
+
+ /* Write into PCI BAR */
+ for (i = 0; i < len; i++)
+ iowrite8(*buf_ptr++, cfg_ptr + i);
+}
+
+static const struct vdpa_config_ops snet_config_ops = {
+ .set_vq_address = snet_set_vq_address,
+ .set_vq_num = snet_set_vq_num,
+ .kick_vq = snet_kick_vq,
+ .set_vq_cb = snet_set_vq_cb,
+ .set_vq_ready = snet_set_vq_ready,
+ .get_vq_ready = snet_get_vq_ready,
+ .set_vq_state = snet_set_vq_state,
+ .get_vq_state = snet_get_vq_state,
+ .get_vq_irq = snet_get_vq_irq,
+ .get_vq_align = snet_get_vq_align,
+ .reset = snet_reset,
+ .get_config_size = snet_get_config_size,
+ .get_device_features = snet_get_features,
+ .set_driver_features = snet_set_drv_features,
+ .get_driver_features = snet_get_drv_features,
+ .get_vq_num_min = snet_get_vq_num_max,
+ .get_vq_num_max = snet_get_vq_num_max,
+ .set_config_cb = snet_set_config_cb,
+ .get_device_id = snet_get_device_id,
+ .get_vendor_id = snet_get_vendor_id,
+ .get_status = snet_get_status,
+ .set_status = snet_set_status,
+ .get_config = snet_get_config,
+ .set_config = snet_set_config,
+};
+
+static int psnet_open_pf_bar(struct pci_dev *pdev, struct psnet *psnet)
+{
+ char name[50];
+ int ret, i, mask = 0;
+ /* We don't know which BAR will be used to communicate..
+ * We will map every bar with len > 0.
+ *
+ * Later, we will discover the BAR and unmap all other BARs.
+ */
+ for (i = 0; i < PCI_STD_NUM_BARS; i++) {
+ if (pci_resource_len(pdev, i))
+ mask |= (1 << i);
+ }
+
+ /* No BAR can be used.. */
+ if (!mask) {
+ SNET_ERR(pdev, "Failed to find a PCI BAR\n");
+ return -ENODEV;
+ }
+
+ snprintf(name, sizeof(name), "psnet[%s]-bars", pci_name(pdev));
+ ret = pcim_iomap_regions(pdev, mask, name);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to request and map PCI BARs\n");
+ return ret;
+ }
+
+ for (i = 0; i < PCI_STD_NUM_BARS; i++) {
+ if (mask & (1 << i))
+ psnet->bars[i] = pcim_iomap_table(pdev)[i];
+ }
+
+ return 0;
+}
+
+static int snet_open_vf_bar(struct pci_dev *pdev, struct snet *snet)
+{
+ char name[50];
+ int ret;
+
+ snprintf(name, sizeof(name), "snet[%s]-bar", pci_name(pdev));
+ /* Request and map BAR */
+ ret = pcim_iomap_regions(pdev, BIT(snet->psnet->cfg.vf_bar), name);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to request and map PCI BAR for a VF\n");
+ return ret;
+ }
+
+ snet->bar = pcim_iomap_table(pdev)[snet->psnet->cfg.vf_bar];
+
+ return 0;
+}
+
+static void snet_free_cfg(struct snet_cfg *cfg)
+{
+ u32 i;
+
+ if (!cfg->devs)
+ return;
+
+ /* Free devices */
+ for (i = 0; i < cfg->devices_num; i++) {
+ if (!cfg->devs[i])
+ break;
+
+ kfree(cfg->devs[i]);
+ }
+ /* Free pointers to devices */
+ kfree(cfg->devs);
+}
+
+/* Detect which BAR is used for communication with the device. */
+static int psnet_detect_bar(struct psnet *psnet, u32 off)
+{
+ unsigned long exit_time;
+ int i;
+
+ exit_time = jiffies + usecs_to_jiffies(SNET_DETECT_TIMEOUT);
+
+ /* SNET DPU will write SNET's signature when the config is ready. */
+ while (time_before(jiffies, exit_time)) {
+ for (i = 0; i < PCI_STD_NUM_BARS; i++) {
+ /* Is this BAR mapped? */
+ if (!psnet->bars[i])
+ continue;
+
+ if (ioread32(psnet->bars[i] + off) == SNET_SIGNATURE)
+ return i;
+ }
+ usleep_range(1000, 10000);
+ }
+
+ return -ENODEV;
+}
+
+static void psnet_unmap_unused_bars(struct pci_dev *pdev, struct psnet *psnet)
+{
+ int i, mask = 0;
+
+ for (i = 0; i < PCI_STD_NUM_BARS; i++) {
+ if (psnet->bars[i] && i != psnet->barno)
+ mask |= (1 << i);
+ }
+
+ if (mask)
+ pcim_iounmap_regions(pdev, mask);
+}
+
+/* Read SNET config from PCI BAR */
+static int psnet_read_cfg(struct pci_dev *pdev, struct psnet *psnet)
+{
+ struct snet_cfg *cfg = &psnet->cfg;
+ u32 i, off;
+ int barno;
+
+ /* Move to where the config starts */
+ off = SNET_CONFIG_OFF;
+
+ /* Find BAR used for communication */
+ barno = psnet_detect_bar(psnet, off);
+ if (barno < 0) {
+ SNET_ERR(pdev, "SNET config is not ready.\n");
+ return barno;
+ }
+
+ /* Save used BAR number and unmap all other BARs */
+ psnet->barno = barno;
+ SNET_DBG(pdev, "Using BAR number %d\n", barno);
+
+ psnet_unmap_unused_bars(pdev, psnet);
+
+ /* load config from BAR */
+ cfg->key = psnet_read32(psnet, off);
+ off += 4;
+ cfg->cfg_size = psnet_read32(psnet, off);
+ off += 4;
+ cfg->cfg_ver = psnet_read32(psnet, off);
+ off += 4;
+ /* The negotiated config version is the lower one between this driver's config
+ * and the DPU's.
+ */
+ psnet->negotiated_cfg_ver = min_t(u32, cfg->cfg_ver, SNET_CFG_VERSION);
+ SNET_DBG(pdev, "SNET config version %u\n", psnet->negotiated_cfg_ver);
+
+ cfg->vf_num = psnet_read32(psnet, off);
+ off += 4;
+ cfg->vf_bar = psnet_read32(psnet, off);
+ off += 4;
+ cfg->host_cfg_off = psnet_read32(psnet, off);
+ off += 4;
+ cfg->max_size_host_cfg = psnet_read32(psnet, off);
+ off += 4;
+ cfg->virtio_cfg_off = psnet_read32(psnet, off);
+ off += 4;
+ cfg->kick_off = psnet_read32(psnet, off);
+ off += 4;
+ cfg->hwmon_off = psnet_read32(psnet, off);
+ off += 4;
+ cfg->msg_off = psnet_read32(psnet, off);
+ off += 4;
+ cfg->flags = psnet_read32(psnet, off);
+ off += 4;
+ /* Ignore Reserved */
+ off += sizeof(cfg->rsvd);
+
+ cfg->devices_num = psnet_read32(psnet, off);
+ off += 4;
+ /* Allocate memory to hold pointer to the devices */
+ cfg->devs = kcalloc(cfg->devices_num, sizeof(void *), GFP_KERNEL);
+ if (!cfg->devs)
+ return -ENOMEM;
+
+ /* Load device configuration from BAR */
+ for (i = 0; i < cfg->devices_num; i++) {
+ cfg->devs[i] = kzalloc(sizeof(*cfg->devs[i]), GFP_KERNEL);
+ if (!cfg->devs[i]) {
+ snet_free_cfg(cfg);
+ return -ENOMEM;
+ }
+ /* Read device config */
+ cfg->devs[i]->virtio_id = psnet_read32(psnet, off);
+ off += 4;
+ cfg->devs[i]->vq_num = psnet_read32(psnet, off);
+ off += 4;
+ cfg->devs[i]->vq_size = psnet_read32(psnet, off);
+ off += 4;
+ cfg->devs[i]->vfid = psnet_read32(psnet, off);
+ off += 4;
+ cfg->devs[i]->features = psnet_read64(psnet, off);
+ off += 8;
+ /* Ignore Reserved */
+ off += sizeof(cfg->devs[i]->rsvd);
+
+ cfg->devs[i]->cfg_size = psnet_read32(psnet, off);
+ off += 4;
+
+ /* Is the config witten to the DPU going to be too big? */
+ if (SNET_GENERAL_CFG_LEN + SNET_GENERAL_CFG_VQ_LEN * cfg->devs[i]->vq_num >
+ cfg->max_size_host_cfg) {
+ SNET_ERR(pdev, "Failed to read SNET config, the config is too big..\n");
+ snet_free_cfg(cfg);
+ return -EINVAL;
+ }
+ }
+ return 0;
+}
+
+static int psnet_alloc_irq_vector(struct pci_dev *pdev, struct psnet *psnet)
+{
+ int ret = 0;
+ u32 i, irq_num = 0;
+
+ /* Let's count how many IRQs we need, 1 for every VQ + 1 for config change */
+ for (i = 0; i < psnet->cfg.devices_num; i++)
+ irq_num += psnet->cfg.devs[i]->vq_num + 1;
+
+ ret = pci_alloc_irq_vectors(pdev, irq_num, irq_num, PCI_IRQ_MSIX);
+ if (ret != irq_num) {
+ SNET_ERR(pdev, "Failed to allocate IRQ vectors\n");
+ return ret;
+ }
+ SNET_DBG(pdev, "Allocated %u IRQ vectors from physical function\n", irq_num);
+
+ return 0;
+}
+
+static int snet_alloc_irq_vector(struct pci_dev *pdev, struct snet_dev_cfg *snet_cfg)
+{
+ int ret = 0;
+ u32 irq_num;
+
+ /* We want 1 IRQ for every VQ + 1 for config change events */
+ irq_num = snet_cfg->vq_num + 1;
+
+ ret = pci_alloc_irq_vectors(pdev, irq_num, irq_num, PCI_IRQ_MSIX);
+ if (ret <= 0) {
+ SNET_ERR(pdev, "Failed to allocate IRQ vectors\n");
+ return ret;
+ }
+
+ return 0;
+}
+
+static void snet_free_vqs(struct snet *snet)
+{
+ u32 i;
+
+ if (!snet->vqs)
+ return;
+
+ for (i = 0 ; i < snet->cfg->vq_num ; i++) {
+ if (!snet->vqs[i])
+ break;
+
+ kfree(snet->vqs[i]);
+ }
+ kfree(snet->vqs);
+}
+
+static int snet_build_vqs(struct snet *snet)
+{
+ u32 i;
+ /* Allocate the VQ pointers array */
+ snet->vqs = kcalloc(snet->cfg->vq_num, sizeof(void *), GFP_KERNEL);
+ if (!snet->vqs)
+ return -ENOMEM;
+
+ /* Allocate the VQs */
+ for (i = 0; i < snet->cfg->vq_num; i++) {
+ snet->vqs[i] = kzalloc(sizeof(*snet->vqs[i]), GFP_KERNEL);
+ if (!snet->vqs[i]) {
+ snet_free_vqs(snet);
+ return -ENOMEM;
+ }
+ /* Reset IRQ num */
+ snet->vqs[i]->irq = -1;
+ /* VQ serial ID */
+ snet->vqs[i]->sid = i;
+ /* Kick address - every VQ gets 4B */
+ snet->vqs[i]->kick_ptr = snet->bar + snet->psnet->cfg.kick_off +
+ snet->vqs[i]->sid * 4;
+ /* Clear kick address for this VQ */
+ iowrite32(0, snet->vqs[i]->kick_ptr);
+ }
+ return 0;
+}
+
+static int psnet_get_next_irq_num(struct psnet *psnet)
+{
+ int irq;
+
+ spin_lock(&psnet->lock);
+ irq = psnet->next_irq++;
+ spin_unlock(&psnet->lock);
+
+ return irq;
+}
+
+static void snet_reserve_irq_idx(struct pci_dev *pdev, struct snet *snet)
+{
+ struct psnet *psnet = snet->psnet;
+ int i;
+
+ /* one IRQ for every VQ, and one for config changes */
+ snet->cfg_irq_idx = psnet_get_next_irq_num(psnet);
+ snprintf(snet->cfg_irq_name, SNET_NAME_SIZE, "snet[%s]-cfg[%d]",
+ pci_name(pdev), snet->cfg_irq_idx);
+
+ for (i = 0; i < snet->cfg->vq_num; i++) {
+ /* Get next free IRQ ID */
+ snet->vqs[i]->irq_idx = psnet_get_next_irq_num(psnet);
+ /* Write IRQ name */
+ snprintf(snet->vqs[i]->irq_name, SNET_NAME_SIZE, "snet[%s]-vq[%d]",
+ pci_name(pdev), snet->vqs[i]->irq_idx);
+ }
+}
+
+/* Find a device config based on virtual function id */
+static struct snet_dev_cfg *snet_find_dev_cfg(struct snet_cfg *cfg, u32 vfid)
+{
+ u32 i;
+
+ for (i = 0; i < cfg->devices_num; i++) {
+ if (cfg->devs[i]->vfid == vfid)
+ return cfg->devs[i];
+ }
+ /* Oppss.. no config found.. */
+ return NULL;
+}
+
+/* Probe function for a physical PCI function */
+static int snet_vdpa_probe_pf(struct pci_dev *pdev)
+{
+ struct psnet *psnet;
+ int ret = 0;
+ bool pf_irqs = false;
+
+ ret = pcim_enable_device(pdev);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to enable PCI device\n");
+ return ret;
+ }
+
+ /* Allocate a PCI physical function device */
+ psnet = kzalloc(sizeof(*psnet), GFP_KERNEL);
+ if (!psnet)
+ return -ENOMEM;
+
+ /* Init PSNET spinlock */
+ spin_lock_init(&psnet->lock);
+
+ pci_set_master(pdev);
+ pci_set_drvdata(pdev, psnet);
+
+ /* Open SNET MAIN BAR */
+ ret = psnet_open_pf_bar(pdev, psnet);
+ if (ret)
+ goto free_psnet;
+
+ /* Try to read SNET's config from PCI BAR */
+ ret = psnet_read_cfg(pdev, psnet);
+ if (ret)
+ goto free_psnet;
+
+ /* If SNET_CFG_FLAG_IRQ_PF flag is set, we should use
+ * PF MSI-X vectors
+ */
+ pf_irqs = PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_IRQ_PF);
+
+ if (pf_irqs) {
+ ret = psnet_alloc_irq_vector(pdev, psnet);
+ if (ret)
+ goto free_cfg;
+ }
+
+ SNET_DBG(pdev, "Enable %u virtual functions\n", psnet->cfg.vf_num);
+ ret = pci_enable_sriov(pdev, psnet->cfg.vf_num);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to enable SR-IOV\n");
+ goto free_irq;
+ }
+
+ /* Create HW monitor device */
+ if (PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_HWMON)) {
+#if IS_ENABLED(CONFIG_HWMON)
+ psnet_create_hwmon(pdev);
+#else
+ SNET_WARN(pdev, "Can't start HWMON, CONFIG_HWMON is not enabled\n");
+#endif
+ }
+
+ return 0;
+
+free_irq:
+ if (pf_irqs)
+ pci_free_irq_vectors(pdev);
+free_cfg:
+ snet_free_cfg(&psnet->cfg);
+free_psnet:
+ kfree(psnet);
+ return ret;
+}
+
+/* Probe function for a virtual PCI function */
+static int snet_vdpa_probe_vf(struct pci_dev *pdev)
+{
+ struct pci_dev *pdev_pf = pdev->physfn;
+ struct psnet *psnet = pci_get_drvdata(pdev_pf);
+ struct snet_dev_cfg *dev_cfg;
+ struct snet *snet;
+ u32 vfid;
+ int ret;
+ bool pf_irqs = false;
+
+ /* Get virtual function id.
+ * (the DPU counts the VFs from 1)
+ */
+ ret = pci_iov_vf_id(pdev);
+ if (ret < 0) {
+ SNET_ERR(pdev, "Failed to find a VF id\n");
+ return ret;
+ }
+ vfid = ret + 1;
+
+ /* Find the snet_dev_cfg based on vfid */
+ dev_cfg = snet_find_dev_cfg(&psnet->cfg, vfid);
+ if (!dev_cfg) {
+ SNET_WARN(pdev, "Failed to find a VF config..\n");
+ return -ENODEV;
+ }
+
+ /* Which PCI device should allocate the IRQs?
+ * If the SNET_CFG_FLAG_IRQ_PF flag set, the PF device allocates the IRQs
+ */
+ pf_irqs = PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_IRQ_PF);
+
+ ret = pcim_enable_device(pdev);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to enable PCI VF device\n");
+ return ret;
+ }
+
+ /* Request for MSI-X IRQs */
+ if (!pf_irqs) {
+ ret = snet_alloc_irq_vector(pdev, dev_cfg);
+ if (ret)
+ return ret;
+ }
+
+ /* Allocate vdpa device */
+ snet = vdpa_alloc_device(struct snet, vdpa, &pdev->dev, &snet_config_ops, 1, 1, NULL,
+ false);
+ if (!snet) {
+ SNET_ERR(pdev, "Failed to allocate a vdpa device\n");
+ ret = -ENOMEM;
+ goto free_irqs;
+ }
+
+ /* Save pci device pointer */
+ snet->pdev = pdev;
+ snet->psnet = psnet;
+ snet->cfg = dev_cfg;
+ snet->dpu_ready = false;
+ snet->sid = vfid;
+ /* Reset IRQ value */
+ snet->cfg_irq = -1;
+
+ ret = snet_open_vf_bar(pdev, snet);
+ if (ret)
+ goto put_device;
+
+ /* Create a VirtIO config pointer */
+ snet->cfg->virtio_cfg = snet->bar + snet->psnet->cfg.virtio_cfg_off;
+
+ pci_set_master(pdev);
+ pci_set_drvdata(pdev, snet);
+
+ ret = snet_build_vqs(snet);
+ if (ret)
+ goto put_device;
+
+ /* Reserve IRQ indexes,
+ * The IRQs may be requested and freed multiple times,
+ * but the indexes won't change.
+ */
+ snet_reserve_irq_idx(pf_irqs ? pdev_pf : pdev, snet);
+
+ /*set DMA device*/
+ snet->vdpa.dma_dev = &pdev->dev;
+
+ /* Register VDPA device */
+ ret = vdpa_register_device(&snet->vdpa, snet->cfg->vq_num);
+ if (ret) {
+ SNET_ERR(pdev, "Failed to register vdpa device\n");
+ goto free_vqs;
+ }
+
+ return 0;
+
+free_vqs:
+ snet_free_vqs(snet);
+put_device:
+ put_device(&snet->vdpa.dev);
+free_irqs:
+ if (!pf_irqs)
+ pci_free_irq_vectors(pdev);
+ return ret;
+}
+
+static int snet_vdpa_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+ if (pdev->is_virtfn)
+ return snet_vdpa_probe_vf(pdev);
+ else
+ return snet_vdpa_probe_pf(pdev);
+}
+
+static void snet_vdpa_remove_pf(struct pci_dev *pdev)
+{
+ struct psnet *psnet = pci_get_drvdata(pdev);
+
+ pci_disable_sriov(pdev);
+ /* If IRQs are allocated from the PF, we should free the IRQs */
+ if (PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_IRQ_PF))
+ pci_free_irq_vectors(pdev);
+
+ snet_free_cfg(&psnet->cfg);
+ kfree(psnet);
+}
+
+static void snet_vdpa_remove_vf(struct pci_dev *pdev)
+{
+ struct snet *snet = pci_get_drvdata(pdev);
+ struct psnet *psnet = snet->psnet;
+
+ vdpa_unregister_device(&snet->vdpa);
+ snet_free_vqs(snet);
+ /* If IRQs are allocated from the VF, we should free the IRQs */
+ if (!PSNET_FLAG_ON(psnet, SNET_CFG_FLAG_IRQ_PF))
+ pci_free_irq_vectors(pdev);
+}
+
+static void snet_vdpa_remove(struct pci_dev *pdev)
+{
+ if (pdev->is_virtfn)
+ snet_vdpa_remove_vf(pdev);
+ else
+ snet_vdpa_remove_pf(pdev);
+}
+
+static struct pci_device_id snet_driver_pci_ids[] = {
+ { PCI_DEVICE_SUB(PCI_VENDOR_ID_SOLIDRUN, SNET_DEVICE_ID,
+ PCI_VENDOR_ID_SOLIDRUN, SNET_DEVICE_ID) },
+ { 0 },
+};
+
+MODULE_DEVICE_TABLE(pci, snet_driver_pci_ids);
+
+static struct pci_driver snet_vdpa_driver = {
+ .name = "snet-vdpa-driver",
+ .id_table = snet_driver_pci_ids,
+ .probe = snet_vdpa_probe,
+ .remove = snet_vdpa_remove,
+};
+
+module_pci_driver(snet_vdpa_driver);
+
+MODULE_AUTHOR("Alvaro Karsz <alvaro.karsz@solid-run.com>");
+MODULE_DESCRIPTION("SolidRun vDPA driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/vdpa/solidrun/snet_vdpa.h b/drivers/vdpa/solidrun/snet_vdpa.h
new file mode 100644
index 000000000000..b7f34169053f
--- /dev/null
+++ b/drivers/vdpa/solidrun/snet_vdpa.h
@@ -0,0 +1,194 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * SolidRun DPU driver for control plane
+ *
+ * Copyright (C) 2022 SolidRun
+ *
+ * Author: Alvaro Karsz <alvaro.karsz@solid-run.com>
+ *
+ */
+#ifndef _SNET_VDPA_H_
+#define _SNET_VDPA_H_
+
+#include <linux/vdpa.h>
+#include <linux/pci.h>
+
+#define SNET_NAME_SIZE 256
+
+#define SNET_ERR(pdev, fmt, ...) dev_err(&(pdev)->dev, "%s"fmt, "snet_vdpa: ", ##__VA_ARGS__)
+#define SNET_WARN(pdev, fmt, ...) dev_warn(&(pdev)->dev, "%s"fmt, "snet_vdpa: ", ##__VA_ARGS__)
+#define SNET_INFO(pdev, fmt, ...) dev_info(&(pdev)->dev, "%s"fmt, "snet_vdpa: ", ##__VA_ARGS__)
+#define SNET_DBG(pdev, fmt, ...) dev_dbg(&(pdev)->dev, "%s"fmt, "snet_vdpa: ", ##__VA_ARGS__)
+#define SNET_HAS_FEATURE(s, f) ((s)->negotiated_features & BIT_ULL(f))
+/* VQ struct */
+struct snet_vq {
+ /* VQ callback */
+ struct vdpa_callback cb;
+ /* desc base address */
+ u64 desc_area;
+ /* device base address */
+ u64 device_area;
+ /* driver base address */
+ u64 driver_area;
+ /* Queue size */
+ u32 num;
+ /* Serial ID for VQ */
+ u32 sid;
+ /* is ready flag */
+ bool ready;
+ /* IRQ number */
+ u32 irq;
+ /* IRQ index, DPU uses this to parse data from MSI-X table */
+ u32 irq_idx;
+ /* IRQ name */
+ char irq_name[SNET_NAME_SIZE];
+ /* pointer to mapped PCI BAR register used by this VQ to kick */
+ void __iomem *kick_ptr;
+};
+
+struct snet {
+ /* vdpa device */
+ struct vdpa_device vdpa;
+ /* Config callback */
+ struct vdpa_callback cb;
+ /* array of virqueues */
+ struct snet_vq **vqs;
+ /* Used features */
+ u64 negotiated_features;
+ /* Device serial ID */
+ u32 sid;
+ /* device status */
+ u8 status;
+ /* boolean indicating if snet config was passed to the device */
+ bool dpu_ready;
+ /* IRQ number */
+ u32 cfg_irq;
+ /* IRQ index, DPU uses this to parse data from MSI-X table */
+ u32 cfg_irq_idx;
+ /* IRQ name */
+ char cfg_irq_name[SNET_NAME_SIZE];
+ /* BAR to access the VF */
+ void __iomem *bar;
+ /* PCI device */
+ struct pci_dev *pdev;
+ /* Pointer to snet pdev parent device */
+ struct psnet *psnet;
+ /* Pointer to snet config device */
+ struct snet_dev_cfg *cfg;
+};
+
+struct snet_dev_cfg {
+ /* Device ID following VirtIO spec. */
+ u32 virtio_id;
+ /* Number of VQs for this device */
+ u32 vq_num;
+ /* Size of every VQ */
+ u32 vq_size;
+ /* Virtual Function id */
+ u32 vfid;
+ /* Device features, following VirtIO spec */
+ u64 features;
+ /* Reserved for future usage */
+ u32 rsvd[6];
+ /* VirtIO device specific config size */
+ u32 cfg_size;
+ /* VirtIO device specific config address */
+ void __iomem *virtio_cfg;
+} __packed;
+
+struct snet_cfg {
+ /* Magic key */
+ u32 key;
+ /* Size of total config in bytes */
+ u32 cfg_size;
+ /* Config version */
+ u32 cfg_ver;
+ /* Number of Virtual Functions to create */
+ u32 vf_num;
+ /* BAR to use for the VFs */
+ u32 vf_bar;
+ /* Where should we write the SNET's config */
+ u32 host_cfg_off;
+ /* Max. allowed size for a SNET's config */
+ u32 max_size_host_cfg;
+ /* VirtIO config offset in BAR */
+ u32 virtio_cfg_off;
+ /* Offset in PCI BAR for VQ kicks */
+ u32 kick_off;
+ /* Offset in PCI BAR for HW monitoring */
+ u32 hwmon_off;
+ /* Offset in PCI BAR for SNET messages */
+ u32 msg_off;
+ /* Config general flags - enum snet_cfg_flags */
+ u32 flags;
+ /* Reserved for future usage */
+ u32 rsvd[6];
+ /* Number of snet devices */
+ u32 devices_num;
+ /* The actual devices */
+ struct snet_dev_cfg **devs;
+} __packed;
+
+/* SolidNET PCIe device, one device per PCIe physical function */
+struct psnet {
+ /* PCI BARs */
+ void __iomem *bars[PCI_STD_NUM_BARS];
+ /* Negotiated config version */
+ u32 negotiated_cfg_ver;
+ /* Next IRQ index to use in case when the IRQs are allocated from this device */
+ u32 next_irq;
+ /* BAR number used to communicate with the device */
+ u8 barno;
+ /* spinlock to protect data that can be changed by SNET devices */
+ spinlock_t lock;
+ /* Pointer to the device's config read from BAR */
+ struct snet_cfg cfg;
+ /* Name of monitor device */
+ char hwmon_name[SNET_NAME_SIZE];
+};
+
+enum snet_cfg_flags {
+ /* Create a HWMON device */
+ SNET_CFG_FLAG_HWMON = BIT(0),
+ /* USE IRQs from the physical function */
+ SNET_CFG_FLAG_IRQ_PF = BIT(1),
+};
+
+#define PSNET_FLAG_ON(p, f) ((p)->cfg.flags & (f))
+
+static inline u32 psnet_read32(struct psnet *psnet, u32 off)
+{
+ return ioread32(psnet->bars[psnet->barno] + off);
+}
+
+static inline u32 snet_read32(struct snet *snet, u32 off)
+{
+ return ioread32(snet->bar + off);
+}
+
+static inline void snet_write32(struct snet *snet, u32 off, u32 val)
+{
+ iowrite32(val, snet->bar + off);
+}
+
+static inline u64 psnet_read64(struct psnet *psnet, u32 off)
+{
+ u64 val;
+ /* 64bits are written in 2 halves, low part first */
+ val = (u64)psnet_read32(psnet, off);
+ val |= ((u64)psnet_read32(psnet, off + 4) << 32);
+ return val;
+}
+
+static inline void snet_write64(struct snet *snet, u32 off, u64 val)
+{
+ /* The DPU expects a 64bit integer in 2 halves, the low part first */
+ snet_write32(snet, off, (u32)val);
+ snet_write32(snet, off + 4, (u32)(val >> 32));
+}
+
+#if IS_ENABLED(CONFIG_HWMON)
+void psnet_create_hwmon(struct pci_dev *pdev);
+#endif
+
+#endif //_SNET_VDPA_H_
diff --git a/drivers/vdpa/vdpa.c b/drivers/vdpa/vdpa.c
index 8ef7aa1365cc..965e32529eb8 100644
--- a/drivers/vdpa/vdpa.c
+++ b/drivers/vdpa/vdpa.c
@@ -39,6 +39,11 @@ static int vdpa_dev_probe(struct device *d)
u32 max_num, min_num = 1;
int ret = 0;
+ d->dma_mask = &d->coherent_dma_mask;
+ ret = dma_set_mask_and_coherent(d, DMA_BIT_MASK(64));
+ if (ret)
+ return ret;
+
max_num = ops->get_vq_num_max(vdev);
if (ops->get_vq_num_min)
min_num = ops->get_vq_num_min(vdev);
@@ -460,12 +465,28 @@ static int vdpa_nl_mgmtdev_handle_fill(struct sk_buff *msg, const struct vdpa_mg
return 0;
}
+static u64 vdpa_mgmtdev_get_classes(const struct vdpa_mgmt_dev *mdev,
+ unsigned int *nclasses)
+{
+ u64 supported_classes = 0;
+ unsigned int n = 0;
+
+ for (int i = 0; mdev->id_table[i].device; i++) {
+ if (mdev->id_table[i].device > 63)
+ continue;
+ supported_classes |= BIT_ULL(mdev->id_table[i].device);
+ n++;
+ }
+ if (nclasses)
+ *nclasses = n;
+
+ return supported_classes;
+}
+
static int vdpa_mgmtdev_fill(const struct vdpa_mgmt_dev *mdev, struct sk_buff *msg,
u32 portid, u32 seq, int flags)
{
- u64 supported_classes = 0;
void *hdr;
- int i = 0;
int err;
hdr = genlmsg_put(msg, portid, seq, &vdpa_nl_family, flags, VDPA_CMD_MGMTDEV_NEW);
@@ -475,14 +496,9 @@ static int vdpa_mgmtdev_fill(const struct vdpa_mgmt_dev *mdev, struct sk_buff *m
if (err)
goto msg_err;
- while (mdev->id_table[i].device) {
- if (mdev->id_table[i].device <= 63)
- supported_classes |= BIT_ULL(mdev->id_table[i].device);
- i++;
- }
-
if (nla_put_u64_64bit(msg, VDPA_ATTR_MGMTDEV_SUPPORTED_CLASSES,
- supported_classes, VDPA_ATTR_UNSPEC)) {
+ vdpa_mgmtdev_get_classes(mdev, NULL),
+ VDPA_ATTR_UNSPEC)) {
err = -EMSGSIZE;
goto msg_err;
}
@@ -566,13 +582,25 @@ out:
BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MTU) | \
BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MAX_VQP))
+/*
+ * Bitmask for all per-device features: feature bits VIRTIO_TRANSPORT_F_START
+ * through VIRTIO_TRANSPORT_F_END are unset, i.e. 0xfffffc000fffffff for
+ * all 64bit features. If the features are extended beyond 64 bits, or new
+ * "holes" are reserved for other type of features than per-device, this
+ * macro would have to be updated.
+ */
+#define VIRTIO_DEVICE_F_MASK (~0ULL << (VIRTIO_TRANSPORT_F_END + 1) | \
+ ((1ULL << VIRTIO_TRANSPORT_F_START) - 1))
+
static int vdpa_nl_cmd_dev_add_set_doit(struct sk_buff *skb, struct genl_info *info)
{
struct vdpa_dev_set_config config = {};
struct nlattr **nl_attrs = info->attrs;
struct vdpa_mgmt_dev *mdev;
+ unsigned int ncls = 0;
const u8 *macaddr;
const char *name;
+ u64 classes;
int err = 0;
if (!info->attrs[VDPA_ATTR_DEV_NAME])
@@ -601,8 +629,26 @@ static int vdpa_nl_cmd_dev_add_set_doit(struct sk_buff *skb, struct genl_info *i
config.mask |= BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MAX_VQP);
}
if (nl_attrs[VDPA_ATTR_DEV_FEATURES]) {
+ u64 missing = 0x0ULL;
+
config.device_features =
nla_get_u64(nl_attrs[VDPA_ATTR_DEV_FEATURES]);
+ if (nl_attrs[VDPA_ATTR_DEV_NET_CFG_MACADDR] &&
+ !(config.device_features & BIT_ULL(VIRTIO_NET_F_MAC)))
+ missing |= BIT_ULL(VIRTIO_NET_F_MAC);
+ if (nl_attrs[VDPA_ATTR_DEV_NET_CFG_MTU] &&
+ !(config.device_features & BIT_ULL(VIRTIO_NET_F_MTU)))
+ missing |= BIT_ULL(VIRTIO_NET_F_MTU);
+ if (nl_attrs[VDPA_ATTR_DEV_NET_CFG_MAX_VQP] &&
+ config.net.max_vq_pairs > 1 &&
+ !(config.device_features & BIT_ULL(VIRTIO_NET_F_MQ)))
+ missing |= BIT_ULL(VIRTIO_NET_F_MQ);
+ if (missing) {
+ NL_SET_ERR_MSG_FMT_MOD(info->extack,
+ "Missing features 0x%llx for provided attributes",
+ missing);
+ return -EINVAL;
+ }
config.mask |= BIT_ULL(VDPA_ATTR_DEV_FEATURES);
}
@@ -622,13 +668,33 @@ static int vdpa_nl_cmd_dev_add_set_doit(struct sk_buff *skb, struct genl_info *i
err = PTR_ERR(mdev);
goto err;
}
+
if ((config.mask & mdev->config_attr_mask) != config.mask) {
- NL_SET_ERR_MSG_MOD(info->extack,
- "All provided attributes are not supported");
+ NL_SET_ERR_MSG_FMT_MOD(info->extack,
+ "Some provided attributes are not supported: 0x%llx",
+ config.mask & ~mdev->config_attr_mask);
err = -EOPNOTSUPP;
goto err;
}
+ classes = vdpa_mgmtdev_get_classes(mdev, &ncls);
+ if (config.mask & VDPA_DEV_NET_ATTRS_MASK &&
+ !(classes & BIT_ULL(VIRTIO_ID_NET))) {
+ NL_SET_ERR_MSG_MOD(info->extack,
+ "Network class attributes provided on unsupported management device");
+ err = -EINVAL;
+ goto err;
+ }
+ if (!(config.mask & VDPA_DEV_NET_ATTRS_MASK) &&
+ config.mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES) &&
+ classes & BIT_ULL(VIRTIO_ID_NET) && ncls > 1 &&
+ config.device_features & VIRTIO_DEVICE_F_MASK) {
+ NL_SET_ERR_MSG_MOD(info->extack,
+ "Management device supports multi-class while device features specified are ambiguous");
+ err = -EINVAL;
+ goto err;
+ }
+
err = mdev->ops->dev_add(mdev, name, &config);
err:
up_write(&vdpa_dev_lock);
@@ -841,18 +907,25 @@ static int vdpa_dev_net_mac_config_fill(struct sk_buff *msg, u64 features,
sizeof(config->mac), config->mac);
}
+static int vdpa_dev_net_status_config_fill(struct sk_buff *msg, u64 features,
+ const struct virtio_net_config *config)
+{
+ u16 val_u16;
+
+ if ((features & BIT_ULL(VIRTIO_NET_F_STATUS)) == 0)
+ return 0;
+
+ val_u16 = __virtio16_to_cpu(true, config->status);
+ return nla_put_u16(msg, VDPA_ATTR_DEV_NET_STATUS, val_u16);
+}
+
static int vdpa_dev_net_config_fill(struct vdpa_device *vdev, struct sk_buff *msg)
{
struct virtio_net_config config = {};
u64 features_device;
- u16 val_u16;
vdev->config->get_config(vdev, 0, &config, sizeof(config));
- val_u16 = __virtio16_to_cpu(true, config.status);
- if (nla_put_u16(msg, VDPA_ATTR_DEV_NET_STATUS, val_u16))
- return -EMSGSIZE;
-
features_device = vdev->config->get_device_features(vdev);
if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_FEATURES, features_device,
@@ -865,6 +938,9 @@ static int vdpa_dev_net_config_fill(struct vdpa_device *vdev, struct sk_buff *ms
if (vdpa_dev_net_mac_config_fill(msg, features_device, &config))
return -EMSGSIZE;
+ if (vdpa_dev_net_status_config_fill(msg, features_device, &config))
+ return -EMSGSIZE;
+
return vdpa_dev_net_mq_config_fill(msg, features_device, &config);
}
@@ -1011,7 +1087,7 @@ static int vdpa_dev_vendor_stats_fill(struct vdpa_device *vdev,
switch (device_id) {
case VIRTIO_ID_NET:
if (index > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX) {
- NL_SET_ERR_MSG_MOD(info->extack, "queue index excceeds max value");
+ NL_SET_ERR_MSG_MOD(info->extack, "queue index exceeds max value");
err = -ERANGE;
break;
}
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c
index cb88891b44a8..6a0a65814626 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c
@@ -17,7 +17,6 @@
#include <linux/vringh.h>
#include <linux/vdpa.h>
#include <linux/vhost_iotlb.h>
-#include <linux/iova.h>
#include <uapi/linux/vdpa.h>
#include "vdpa_sim.h"
@@ -45,13 +44,6 @@ static struct vdpasim *vdpa_to_sim(struct vdpa_device *vdpa)
return container_of(vdpa, struct vdpasim, vdpa);
}
-static struct vdpasim *dev_to_sim(struct device *dev)
-{
- struct vdpa_device *vdpa = dev_to_vdpa(dev);
-
- return vdpa_to_sim(vdpa);
-}
-
static void vdpasim_vq_notify(struct vringh *vring)
{
struct vdpasim_virtqueue *vq =
@@ -66,14 +58,16 @@ static void vdpasim_vq_notify(struct vringh *vring)
static void vdpasim_queue_ready(struct vdpasim *vdpasim, unsigned int idx)
{
struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
+ uint16_t last_avail_idx = vq->vring.last_avail_idx;
- vringh_init_iotlb(&vq->vring, vdpasim->features, vq->num, false,
+ vringh_init_iotlb(&vq->vring, vdpasim->features, vq->num, true,
(struct vring_desc *)(uintptr_t)vq->desc_addr,
(struct vring_avail *)
(uintptr_t)vq->driver_addr,
(struct vring_used *)
(uintptr_t)vq->device_addr);
+ vq->vring.last_avail_idx = last_avail_idx;
vq->vring.notify = vdpasim_vq_notify;
}
@@ -104,8 +98,12 @@ static void vdpasim_do_reset(struct vdpasim *vdpasim)
&vdpasim->iommu_lock);
}
- for (i = 0; i < vdpasim->dev_attr.nas; i++)
+ for (i = 0; i < vdpasim->dev_attr.nas; i++) {
vhost_iotlb_reset(&vdpasim->iommu[i]);
+ vhost_iotlb_add_range(&vdpasim->iommu[i], 0, ULONG_MAX,
+ 0, VHOST_MAP_RW);
+ vdpasim->iommu_pt[i] = true;
+ }
vdpasim->running = true;
spin_unlock(&vdpasim->iommu_lock);
@@ -115,133 +113,6 @@ static void vdpasim_do_reset(struct vdpasim *vdpasim)
++vdpasim->generation;
}
-static int dir_to_perm(enum dma_data_direction dir)
-{
- int perm = -EFAULT;
-
- switch (dir) {
- case DMA_FROM_DEVICE:
- perm = VHOST_MAP_WO;
- break;
- case DMA_TO_DEVICE:
- perm = VHOST_MAP_RO;
- break;
- case DMA_BIDIRECTIONAL:
- perm = VHOST_MAP_RW;
- break;
- default:
- break;
- }
-
- return perm;
-}
-
-static dma_addr_t vdpasim_map_range(struct vdpasim *vdpasim, phys_addr_t paddr,
- size_t size, unsigned int perm)
-{
- struct iova *iova;
- dma_addr_t dma_addr;
- int ret;
-
- /* We set the limit_pfn to the maximum (ULONG_MAX - 1) */
- iova = alloc_iova(&vdpasim->iova, size >> iova_shift(&vdpasim->iova),
- ULONG_MAX - 1, true);
- if (!iova)
- return DMA_MAPPING_ERROR;
-
- dma_addr = iova_dma_addr(&vdpasim->iova, iova);
-
- spin_lock(&vdpasim->iommu_lock);
- ret = vhost_iotlb_add_range(&vdpasim->iommu[0], (u64)dma_addr,
- (u64)dma_addr + size - 1, (u64)paddr, perm);
- spin_unlock(&vdpasim->iommu_lock);
-
- if (ret) {
- __free_iova(&vdpasim->iova, iova);
- return DMA_MAPPING_ERROR;
- }
-
- return dma_addr;
-}
-
-static void vdpasim_unmap_range(struct vdpasim *vdpasim, dma_addr_t dma_addr,
- size_t size)
-{
- spin_lock(&vdpasim->iommu_lock);
- vhost_iotlb_del_range(&vdpasim->iommu[0], (u64)dma_addr,
- (u64)dma_addr + size - 1);
- spin_unlock(&vdpasim->iommu_lock);
-
- free_iova(&vdpasim->iova, iova_pfn(&vdpasim->iova, dma_addr));
-}
-
-static dma_addr_t vdpasim_map_page(struct device *dev, struct page *page,
- unsigned long offset, size_t size,
- enum dma_data_direction dir,
- unsigned long attrs)
-{
- struct vdpasim *vdpasim = dev_to_sim(dev);
- phys_addr_t paddr = page_to_phys(page) + offset;
- int perm = dir_to_perm(dir);
-
- if (perm < 0)
- return DMA_MAPPING_ERROR;
-
- return vdpasim_map_range(vdpasim, paddr, size, perm);
-}
-
-static void vdpasim_unmap_page(struct device *dev, dma_addr_t dma_addr,
- size_t size, enum dma_data_direction dir,
- unsigned long attrs)
-{
- struct vdpasim *vdpasim = dev_to_sim(dev);
-
- vdpasim_unmap_range(vdpasim, dma_addr, size);
-}
-
-static void *vdpasim_alloc_coherent(struct device *dev, size_t size,
- dma_addr_t *dma_addr, gfp_t flag,
- unsigned long attrs)
-{
- struct vdpasim *vdpasim = dev_to_sim(dev);
- phys_addr_t paddr;
- void *addr;
-
- addr = kmalloc(size, flag);
- if (!addr) {
- *dma_addr = DMA_MAPPING_ERROR;
- return NULL;
- }
-
- paddr = virt_to_phys(addr);
-
- *dma_addr = vdpasim_map_range(vdpasim, paddr, size, VHOST_MAP_RW);
- if (*dma_addr == DMA_MAPPING_ERROR) {
- kfree(addr);
- return NULL;
- }
-
- return addr;
-}
-
-static void vdpasim_free_coherent(struct device *dev, size_t size,
- void *vaddr, dma_addr_t dma_addr,
- unsigned long attrs)
-{
- struct vdpasim *vdpasim = dev_to_sim(dev);
-
- vdpasim_unmap_range(vdpasim, dma_addr, size);
-
- kfree(vaddr);
-}
-
-static const struct dma_map_ops vdpasim_dma_ops = {
- .map_page = vdpasim_map_page,
- .unmap_page = vdpasim_unmap_page,
- .alloc = vdpasim_alloc_coherent,
- .free = vdpasim_free_coherent,
-};
-
static const struct vdpa_config_ops vdpasim_config_ops;
static const struct vdpa_config_ops vdpasim_batch_config_ops;
@@ -249,10 +120,14 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
const struct vdpa_dev_set_config *config)
{
const struct vdpa_config_ops *ops;
+ struct vdpa_device *vdpa;
struct vdpasim *vdpasim;
struct device *dev;
int i, ret = -ENOMEM;
+ if (!dev_attr->alloc_size)
+ return ERR_PTR(-EINVAL);
+
if (config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) {
if (config->device_features &
~dev_attr->supported_features)
@@ -266,14 +141,16 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
else
ops = &vdpasim_config_ops;
- vdpasim = vdpa_alloc_device(struct vdpasim, vdpa, NULL, ops,
- dev_attr->ngroups, dev_attr->nas,
- dev_attr->name, false);
- if (IS_ERR(vdpasim)) {
- ret = PTR_ERR(vdpasim);
+ vdpa = __vdpa_alloc_device(NULL, ops,
+ dev_attr->ngroups, dev_attr->nas,
+ dev_attr->alloc_size,
+ dev_attr->name, false);
+ if (IS_ERR(vdpa)) {
+ ret = PTR_ERR(vdpa);
goto err_alloc;
}
+ vdpasim = vdpa_to_sim(vdpa);
vdpasim->dev_attr = *dev_attr;
INIT_WORK(&vdpasim->work, dev_attr->work_fn);
spin_lock_init(&vdpasim->lock);
@@ -283,7 +160,6 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
dev->dma_mask = &dev->coherent_dma_mask;
if (dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)))
goto err_iommu;
- set_dma_ops(dev, &vdpasim_dma_ops);
vdpasim->vdpa.mdev = dev_attr->mgmt_dev;
vdpasim->config = kzalloc(dev_attr->config_size, GFP_KERNEL);
@@ -300,6 +176,11 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
if (!vdpasim->iommu)
goto err_iommu;
+ vdpasim->iommu_pt = kmalloc_array(vdpasim->dev_attr.nas,
+ sizeof(*vdpasim->iommu_pt), GFP_KERNEL);
+ if (!vdpasim->iommu_pt)
+ goto err_iommu;
+
for (i = 0; i < vdpasim->dev_attr.nas; i++)
vhost_iotlb_init(&vdpasim->iommu[i], max_iotlb_entries, 0);
@@ -311,13 +192,6 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
vringh_set_iotlb(&vdpasim->vqs[i].vring, &vdpasim->iommu[0],
&vdpasim->iommu_lock);
- ret = iova_cache_get();
- if (ret)
- goto err_iommu;
-
- /* For simplicity we use an IOVA allocator with byte granularity */
- init_iova_domain(&vdpasim->iova, 1, 0);
-
vdpasim->vdpa.dma_dev = dev;
return vdpasim;
@@ -356,6 +230,12 @@ static void vdpasim_kick_vq(struct vdpa_device *vdpa, u16 idx)
struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
+ if (!vdpasim->running &&
+ (vdpasim->status & VIRTIO_CONFIG_S_DRIVER_OK)) {
+ vdpasim->pending_kick = true;
+ return;
+ }
+
if (vq->ready)
schedule_work(&vdpasim->work);
}
@@ -418,6 +298,18 @@ static int vdpasim_get_vq_state(struct vdpa_device *vdpa, u16 idx,
return 0;
}
+static int vdpasim_get_vq_stats(struct vdpa_device *vdpa, u16 idx,
+ struct sk_buff *msg,
+ struct netlink_ext_ack *extack)
+{
+ struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
+
+ if (vdpasim->dev_attr.get_stats)
+ return vdpasim->dev_attr.get_stats(vdpasim, idx,
+ msg, extack);
+ return -EOPNOTSUPP;
+}
+
static u32 vdpasim_get_vq_align(struct vdpa_device *vdpa)
{
return VDPASIM_QUEUE_ALIGN;
@@ -526,6 +418,27 @@ static int vdpasim_suspend(struct vdpa_device *vdpa)
return 0;
}
+static int vdpasim_resume(struct vdpa_device *vdpa)
+{
+ struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
+ int i;
+
+ spin_lock(&vdpasim->lock);
+ vdpasim->running = true;
+
+ if (vdpasim->pending_kick) {
+ /* Process pending descriptors */
+ for (i = 0; i < vdpasim->dev_attr.nvqs; ++i)
+ vdpasim_kick_vq(vdpa, i);
+
+ vdpasim->pending_kick = false;
+ }
+
+ spin_unlock(&vdpasim->lock);
+
+ return 0;
+}
+
static size_t vdpasim_get_config_size(struct vdpa_device *vdpa)
{
struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
@@ -621,6 +534,7 @@ static int vdpasim_set_map(struct vdpa_device *vdpa, unsigned int asid,
iommu = &vdpasim->iommu[asid];
vhost_iotlb_reset(iommu);
+ vdpasim->iommu_pt[asid] = false;
for (map = vhost_iotlb_itree_first(iotlb, start, last); map;
map = vhost_iotlb_itree_next(map, start, last)) {
@@ -649,6 +563,10 @@ static int vdpasim_dma_map(struct vdpa_device *vdpa, unsigned int asid,
return -EINVAL;
spin_lock(&vdpasim->iommu_lock);
+ if (vdpasim->iommu_pt[asid]) {
+ vhost_iotlb_reset(&vdpasim->iommu[asid]);
+ vdpasim->iommu_pt[asid] = false;
+ }
ret = vhost_iotlb_add_range_ctx(&vdpasim->iommu[asid], iova,
iova + size - 1, pa, perm, opaque);
spin_unlock(&vdpasim->iommu_lock);
@@ -664,6 +582,11 @@ static int vdpasim_dma_unmap(struct vdpa_device *vdpa, unsigned int asid,
if (asid >= vdpasim->dev_attr.nas)
return -EINVAL;
+ if (vdpasim->iommu_pt[asid]) {
+ vhost_iotlb_reset(&vdpasim->iommu[asid]);
+ vdpasim->iommu_pt[asid] = false;
+ }
+
spin_lock(&vdpasim->iommu_lock);
vhost_iotlb_del_range(&vdpasim->iommu[asid], iova, iova + size - 1);
spin_unlock(&vdpasim->iommu_lock);
@@ -683,15 +606,11 @@ static void vdpasim_free(struct vdpa_device *vdpa)
vringh_kiov_cleanup(&vdpasim->vqs[i].in_iov);
}
- if (vdpa_get_dma_dev(vdpa)) {
- put_iova_domain(&vdpasim->iova);
- iova_cache_put();
- }
-
kvfree(vdpasim->buffer);
for (i = 0; i < vdpasim->dev_attr.nas; i++)
vhost_iotlb_reset(&vdpasim->iommu[i]);
kfree(vdpasim->iommu);
+ kfree(vdpasim->iommu_pt);
kfree(vdpasim->vqs);
kfree(vdpasim->config);
}
@@ -704,6 +623,7 @@ static const struct vdpa_config_ops vdpasim_config_ops = {
.set_vq_ready = vdpasim_set_vq_ready,
.get_vq_ready = vdpasim_get_vq_ready,
.set_vq_state = vdpasim_set_vq_state,
+ .get_vendor_vq_stats = vdpasim_get_vq_stats,
.get_vq_state = vdpasim_get_vq_state,
.get_vq_align = vdpasim_get_vq_align,
.get_vq_group = vdpasim_get_vq_group,
@@ -718,6 +638,7 @@ static const struct vdpa_config_ops vdpasim_config_ops = {
.set_status = vdpasim_set_status,
.reset = vdpasim_reset,
.suspend = vdpasim_suspend,
+ .resume = vdpasim_resume,
.get_config_size = vdpasim_get_config_size,
.get_config = vdpasim_get_config,
.set_config = vdpasim_set_config,
@@ -737,6 +658,7 @@ static const struct vdpa_config_ops vdpasim_batch_config_ops = {
.set_vq_ready = vdpasim_set_vq_ready,
.get_vq_ready = vdpasim_get_vq_ready,
.set_vq_state = vdpasim_set_vq_state,
+ .get_vendor_vq_stats = vdpasim_get_vq_stats,
.get_vq_state = vdpasim_get_vq_state,
.get_vq_align = vdpasim_get_vq_align,
.get_vq_group = vdpasim_get_vq_group,
@@ -751,6 +673,7 @@ static const struct vdpa_config_ops vdpasim_batch_config_ops = {
.set_status = vdpasim_set_status,
.reset = vdpasim_reset,
.suspend = vdpasim_suspend,
+ .resume = vdpasim_resume,
.get_config_size = vdpasim_get_config_size,
.get_config = vdpasim_get_config,
.set_config = vdpasim_set_config,
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.h b/drivers/vdpa/vdpa_sim/vdpa_sim.h
index 0e78737dcc16..144858636c10 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.h
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.h
@@ -37,6 +37,7 @@ struct vdpasim_dev_attr {
struct vdpa_mgmt_dev *mgmt_dev;
const char *name;
u64 supported_features;
+ size_t alloc_size;
size_t config_size;
size_t buffer_size;
int nvqs;
@@ -47,6 +48,9 @@ struct vdpasim_dev_attr {
work_func_t work_fn;
void (*get_config)(struct vdpasim *vdpasim, void *config);
void (*set_config)(struct vdpasim *vdpasim, const void *config);
+ int (*get_stats)(struct vdpasim *vdpasim, u16 idx,
+ struct sk_buff *msg,
+ struct netlink_ext_ack *extack);
};
/* State of each vdpasim device */
@@ -60,13 +64,14 @@ struct vdpasim {
/* virtio config according to device type */
void *config;
struct vhost_iotlb *iommu;
- struct iova_domain iova;
+ bool *iommu_pt;
void *buffer;
u32 status;
u32 generation;
u64 features;
u32 groups;
bool running;
+ bool pending_kick;
/* spinlock to synchronize iommu table */
spinlock_t iommu_lock;
};
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim_blk.c b/drivers/vdpa/vdpa_sim/vdpa_sim_blk.c
index f745926237a8..5117959bed8a 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim_blk.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim_blk.c
@@ -378,6 +378,7 @@ static int vdpasim_blk_dev_add(struct vdpa_mgmt_dev *mdev, const char *name,
dev_attr.nvqs = VDPASIM_BLK_VQ_NUM;
dev_attr.ngroups = VDPASIM_BLK_GROUP_NUM;
dev_attr.nas = VDPASIM_BLK_AS_NUM;
+ dev_attr.alloc_size = sizeof(struct vdpasim);
dev_attr.config_size = sizeof(struct virtio_blk_config);
dev_attr.get_config = vdpasim_blk_get_config;
dev_attr.work_fn = vdpasim_blk_work;
diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim_net.c b/drivers/vdpa/vdpa_sim/vdpa_sim_net.c
index 584b975a98a7..862f405362de 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim_net.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim_net.c
@@ -15,6 +15,7 @@
#include <linux/etherdevice.h>
#include <linux/vringh.h>
#include <linux/vdpa.h>
+#include <net/netlink.h>
#include <uapi/linux/virtio_net.h>
#include <uapi/linux/vdpa.h>
@@ -27,6 +28,7 @@
#define VDPASIM_NET_FEATURES (VDPASIM_FEATURES | \
(1ULL << VIRTIO_NET_F_MAC) | \
+ (1ULL << VIRTIO_NET_F_STATUS) | \
(1ULL << VIRTIO_NET_F_MTU) | \
(1ULL << VIRTIO_NET_F_CTRL_VQ) | \
(1ULL << VIRTIO_NET_F_CTRL_MAC_ADDR))
@@ -36,6 +38,34 @@
#define VDPASIM_NET_AS_NUM 2
#define VDPASIM_NET_GROUP_NUM 2
+struct vdpasim_dataq_stats {
+ struct u64_stats_sync syncp;
+ u64 pkts;
+ u64 bytes;
+ u64 drops;
+ u64 errors;
+ u64 overruns;
+};
+
+struct vdpasim_cq_stats {
+ struct u64_stats_sync syncp;
+ u64 requests;
+ u64 successes;
+ u64 errors;
+};
+
+struct vdpasim_net{
+ struct vdpasim vdpasim;
+ struct vdpasim_dataq_stats tx_stats;
+ struct vdpasim_dataq_stats rx_stats;
+ struct vdpasim_cq_stats cq_stats;
+};
+
+static struct vdpasim_net *sim_to_net(struct vdpasim *vdpasim)
+{
+ return container_of(vdpasim, struct vdpasim_net, vdpasim);
+}
+
static void vdpasim_net_complete(struct vdpasim_virtqueue *vq, size_t len)
{
/* Make sure data is wrote before advancing index */
@@ -96,9 +126,11 @@ static virtio_net_ctrl_ack vdpasim_handle_ctrl_mac(struct vdpasim *vdpasim,
static void vdpasim_handle_cvq(struct vdpasim *vdpasim)
{
struct vdpasim_virtqueue *cvq = &vdpasim->vqs[2];
+ struct vdpasim_net *net = sim_to_net(vdpasim);
virtio_net_ctrl_ack status = VIRTIO_NET_ERR;
struct virtio_net_ctrl_hdr ctrl;
size_t read, write;
+ u64 requests = 0, errors = 0, successes = 0;
int err;
if (!(vdpasim->features & (1ULL << VIRTIO_NET_F_CTRL_VQ)))
@@ -114,10 +146,13 @@ static void vdpasim_handle_cvq(struct vdpasim *vdpasim)
if (err <= 0)
break;
+ ++requests;
read = vringh_iov_pull_iotlb(&cvq->vring, &cvq->in_iov, &ctrl,
sizeof(ctrl));
- if (read != sizeof(ctrl))
+ if (read != sizeof(ctrl)) {
+ ++errors;
break;
+ }
switch (ctrl.class) {
case VIRTIO_NET_CTRL_MAC:
@@ -127,6 +162,11 @@ static void vdpasim_handle_cvq(struct vdpasim *vdpasim)
break;
}
+ if (status == VIRTIO_NET_OK)
+ ++successes;
+ else
+ ++errors;
+
/* Make sure data is wrote before advancing index */
smp_wmb();
@@ -144,6 +184,12 @@ static void vdpasim_handle_cvq(struct vdpasim *vdpasim)
cvq->cb(cvq->private);
local_bh_enable();
}
+
+ u64_stats_update_begin(&net->cq_stats.syncp);
+ net->cq_stats.requests += requests;
+ net->cq_stats.errors += errors;
+ net->cq_stats.successes += successes;
+ u64_stats_update_end(&net->cq_stats.syncp);
}
static void vdpasim_net_work(struct work_struct *work)
@@ -151,8 +197,10 @@ static void vdpasim_net_work(struct work_struct *work)
struct vdpasim *vdpasim = container_of(work, struct vdpasim, work);
struct vdpasim_virtqueue *txq = &vdpasim->vqs[1];
struct vdpasim_virtqueue *rxq = &vdpasim->vqs[0];
+ struct vdpasim_net *net = sim_to_net(vdpasim);
ssize_t read, write;
- int pkts = 0;
+ u64 tx_pkts = 0, rx_pkts = 0, tx_bytes = 0, rx_bytes = 0;
+ u64 rx_drops = 0, rx_overruns = 0, rx_errors = 0, tx_errors = 0;
int err;
spin_lock(&vdpasim->lock);
@@ -171,14 +219,21 @@ static void vdpasim_net_work(struct work_struct *work)
while (true) {
err = vringh_getdesc_iotlb(&txq->vring, &txq->out_iov, NULL,
&txq->head, GFP_ATOMIC);
- if (err <= 0)
+ if (err <= 0) {
+ if (err)
+ ++tx_errors;
break;
+ }
+ ++tx_pkts;
read = vringh_iov_pull_iotlb(&txq->vring, &txq->out_iov,
vdpasim->buffer,
PAGE_SIZE);
+ tx_bytes += read;
+
if (!receive_filter(vdpasim, read)) {
+ ++rx_drops;
vdpasim_net_complete(txq, 0);
continue;
}
@@ -186,19 +241,25 @@ static void vdpasim_net_work(struct work_struct *work)
err = vringh_getdesc_iotlb(&rxq->vring, NULL, &rxq->in_iov,
&rxq->head, GFP_ATOMIC);
if (err <= 0) {
+ ++rx_overruns;
vdpasim_net_complete(txq, 0);
break;
}
write = vringh_iov_push_iotlb(&rxq->vring, &rxq->in_iov,
vdpasim->buffer, read);
- if (write <= 0)
+ if (write <= 0) {
+ ++rx_errors;
break;
+ }
+
+ ++rx_pkts;
+ rx_bytes += write;
vdpasim_net_complete(txq, 0);
vdpasim_net_complete(rxq, write);
- if (++pkts > 4) {
+ if (tx_pkts > 4) {
schedule_work(&vdpasim->work);
goto out;
}
@@ -206,6 +267,145 @@ static void vdpasim_net_work(struct work_struct *work)
out:
spin_unlock(&vdpasim->lock);
+
+ u64_stats_update_begin(&net->tx_stats.syncp);
+ net->tx_stats.pkts += tx_pkts;
+ net->tx_stats.bytes += tx_bytes;
+ net->tx_stats.errors += tx_errors;
+ u64_stats_update_end(&net->tx_stats.syncp);
+
+ u64_stats_update_begin(&net->rx_stats.syncp);
+ net->rx_stats.pkts += rx_pkts;
+ net->rx_stats.bytes += rx_bytes;
+ net->rx_stats.drops += rx_drops;
+ net->rx_stats.errors += rx_errors;
+ net->rx_stats.overruns += rx_overruns;
+ u64_stats_update_end(&net->rx_stats.syncp);
+}
+
+static int vdpasim_net_get_stats(struct vdpasim *vdpasim, u16 idx,
+ struct sk_buff *msg,
+ struct netlink_ext_ack *extack)
+{
+ struct vdpasim_net *net = sim_to_net(vdpasim);
+ u64 rx_pkts, rx_bytes, rx_errors, rx_overruns, rx_drops;
+ u64 tx_pkts, tx_bytes, tx_errors, tx_drops;
+ u64 cq_requests, cq_successes, cq_errors;
+ unsigned int start;
+ int err = -EMSGSIZE;
+
+ switch(idx) {
+ case 0:
+ do {
+ start = u64_stats_fetch_begin(&net->rx_stats.syncp);
+ rx_pkts = net->rx_stats.pkts;
+ rx_bytes = net->rx_stats.bytes;
+ rx_errors = net->rx_stats.errors;
+ rx_overruns = net->rx_stats.overruns;
+ rx_drops = net->rx_stats.drops;
+ } while (u64_stats_fetch_retry(&net->rx_stats.syncp, start));
+
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "rx packets"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ rx_pkts, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "rx bytes"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ rx_bytes, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "rx errors"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ rx_errors, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "rx overruns"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ rx_overruns, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "rx drops"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ rx_drops, VDPA_ATTR_PAD))
+ break;
+ err = 0;
+ break;
+ case 1:
+ do {
+ start = u64_stats_fetch_begin(&net->tx_stats.syncp);
+ tx_pkts = net->tx_stats.pkts;
+ tx_bytes = net->tx_stats.bytes;
+ tx_errors = net->tx_stats.errors;
+ tx_drops = net->tx_stats.drops;
+ } while (u64_stats_fetch_retry(&net->tx_stats.syncp, start));
+
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "tx packets"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ tx_pkts, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "tx bytes"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ tx_bytes, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "tx errors"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ tx_errors, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "tx drops"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ tx_drops, VDPA_ATTR_PAD))
+ break;
+ err = 0;
+ break;
+ case 2:
+ do {
+ start = u64_stats_fetch_begin(&net->cq_stats.syncp);
+ cq_requests = net->cq_stats.requests;
+ cq_successes = net->cq_stats.successes;
+ cq_errors = net->cq_stats.errors;
+ } while (u64_stats_fetch_retry(&net->cq_stats.syncp, start));
+
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "cvq requests"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ cq_requests, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "cvq successes"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ cq_successes, VDPA_ATTR_PAD))
+ break;
+ if (nla_put_string(msg, VDPA_ATTR_DEV_VENDOR_ATTR_NAME,
+ "cvq errors"))
+ break;
+ if (nla_put_u64_64bit(msg, VDPA_ATTR_DEV_VENDOR_ATTR_VALUE,
+ cq_errors, VDPA_ATTR_PAD))
+ break;
+ err = 0;
+ break;
+ default:
+ err = -EINVAL;
+ break;
+ }
+
+ return err;
}
static void vdpasim_net_get_config(struct vdpasim *vdpasim, void *config)
@@ -242,6 +442,7 @@ static int vdpasim_net_dev_add(struct vdpa_mgmt_dev *mdev, const char *name,
const struct vdpa_dev_set_config *config)
{
struct vdpasim_dev_attr dev_attr = {};
+ struct vdpasim_net *net;
struct vdpasim *simdev;
int ret;
@@ -252,9 +453,11 @@ static int vdpasim_net_dev_add(struct vdpa_mgmt_dev *mdev, const char *name,
dev_attr.nvqs = VDPASIM_NET_VQ_NUM;
dev_attr.ngroups = VDPASIM_NET_GROUP_NUM;
dev_attr.nas = VDPASIM_NET_AS_NUM;
+ dev_attr.alloc_size = sizeof(struct vdpasim_net);
dev_attr.config_size = sizeof(struct virtio_net_config);
dev_attr.get_config = vdpasim_net_get_config;
dev_attr.work_fn = vdpasim_net_work;
+ dev_attr.get_stats = vdpasim_net_get_stats;
dev_attr.buffer_size = PAGE_SIZE;
simdev = vdpasim_create(&dev_attr, config);
@@ -267,6 +470,12 @@ static int vdpasim_net_dev_add(struct vdpa_mgmt_dev *mdev, const char *name,
if (ret)
goto reg_err;
+ net = sim_to_net(simdev);
+
+ u64_stats_init(&net->tx_stats.syncp);
+ u64_stats_init(&net->rx_stats.syncp);
+ u64_stats_init(&net->cq_stats.syncp);
+
return 0;
reg_err:
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 4c538b30fd76..07181cd8d52e 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -73,7 +73,8 @@ enum {
VHOST_NET_FEATURES = VHOST_FEATURES |
(1ULL << VHOST_NET_F_VIRTIO_NET_HDR) |
(1ULL << VIRTIO_NET_F_MRG_RXBUF) |
- (1ULL << VIRTIO_F_ACCESS_PLATFORM)
+ (1ULL << VIRTIO_F_ACCESS_PLATFORM) |
+ (1ULL << VIRTIO_F_RING_RESET)
};
enum {
@@ -1645,7 +1646,7 @@ static int vhost_net_set_features(struct vhost_net *n, u64 features)
goto out_unlock;
if ((features & (1ULL << VIRTIO_F_ACCESS_PLATFORM))) {
- if (vhost_init_device_iotlb(&n->dev, true))
+ if (vhost_init_device_iotlb(&n->dev))
goto out_unlock;
}
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index d5ecb8876fc9..b244e7c0f514 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -2105,7 +2105,7 @@ static ssize_t vhost_scsi_tpg_attrib_fabric_prot_type_show(
struct vhost_scsi_tpg *tpg = container_of(se_tpg,
struct vhost_scsi_tpg, se_tpg);
- return sprintf(page, "%d\n", tpg->tv_fabric_prot_type);
+ return sysfs_emit(page, "%d\n", tpg->tv_fabric_prot_type);
}
CONFIGFS_ATTR(vhost_scsi_tpg_attrib_, fabric_prot_type);
@@ -2215,7 +2215,7 @@ static ssize_t vhost_scsi_tpg_nexus_show(struct config_item *item, char *page)
mutex_unlock(&tpg->tv_tpg_mutex);
return -ENODEV;
}
- ret = snprintf(page, PAGE_SIZE, "%s\n",
+ ret = sysfs_emit(page, "%s\n",
tv_nexus->tvn_se_sess->se_node_acl->initiatorname);
mutex_unlock(&tpg->tv_tpg_mutex);
@@ -2440,7 +2440,7 @@ static void vhost_scsi_drop_tport(struct se_wwn *wwn)
static ssize_t
vhost_scsi_wwn_version_show(struct config_item *item, char *page)
{
- return sprintf(page, "TCM_VHOST fabric module %s on %s/%s"
+ return sysfs_emit(page, "TCM_VHOST fabric module %s on %s/%s"
"on "UTS_RELEASE"\n", VHOST_SCSI_VERSION, utsname()->sysname,
utsname()->machine);
}
diff --git a/drivers/vhost/test.c b/drivers/vhost/test.c
index bc8e7fb1e635..42c955a5b211 100644
--- a/drivers/vhost/test.c
+++ b/drivers/vhost/test.c
@@ -333,13 +333,10 @@ static long vhost_test_ioctl(struct file *f, unsigned int ioctl,
return -EFAULT;
return 0;
case VHOST_SET_FEATURES:
- printk(KERN_ERR "1\n");
if (copy_from_user(&features, featurep, sizeof features))
return -EFAULT;
- printk(KERN_ERR "2\n");
if (features & ~VHOST_FEATURES)
return -EOPNOTSUPP;
- printk(KERN_ERR "3\n");
return vhost_test_set_features(n, features);
case VHOST_RESET_OWNER:
return vhost_test_reset_owner(n);
diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c
index 315f9ba47ff2..dc12dbd5b43b 100644
--- a/drivers/vhost/vdpa.c
+++ b/drivers/vhost/vdpa.c
@@ -359,6 +359,14 @@ static bool vhost_vdpa_can_suspend(const struct vhost_vdpa *v)
return ops->suspend;
}
+static bool vhost_vdpa_can_resume(const struct vhost_vdpa *v)
+{
+ struct vdpa_device *vdpa = v->vdpa;
+ const struct vdpa_config_ops *ops = vdpa->config;
+
+ return ops->resume;
+}
+
static long vhost_vdpa_get_features(struct vhost_vdpa *v, u64 __user *featurep)
{
struct vdpa_device *vdpa = v->vdpa;
@@ -498,6 +506,21 @@ static long vhost_vdpa_suspend(struct vhost_vdpa *v)
return ops->suspend(vdpa);
}
+/* After a successful return of this ioctl the device resumes processing
+ * virtqueue descriptors. The device becomes fully operational the same way it
+ * was before it was suspended.
+ */
+static long vhost_vdpa_resume(struct vhost_vdpa *v)
+{
+ struct vdpa_device *vdpa = v->vdpa;
+ const struct vdpa_config_ops *ops = vdpa->config;
+
+ if (!ops->resume)
+ return -EOPNOTSUPP;
+
+ return ops->resume(vdpa);
+}
+
static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd,
void __user *argp)
{
@@ -606,11 +629,15 @@ static long vhost_vdpa_unlocked_ioctl(struct file *filep,
if (copy_from_user(&features, featurep, sizeof(features)))
return -EFAULT;
if (features & ~(VHOST_VDPA_BACKEND_FEATURES |
- BIT_ULL(VHOST_BACKEND_F_SUSPEND)))
+ BIT_ULL(VHOST_BACKEND_F_SUSPEND) |
+ BIT_ULL(VHOST_BACKEND_F_RESUME)))
return -EOPNOTSUPP;
if ((features & BIT_ULL(VHOST_BACKEND_F_SUSPEND)) &&
!vhost_vdpa_can_suspend(v))
return -EOPNOTSUPP;
+ if ((features & BIT_ULL(VHOST_BACKEND_F_RESUME)) &&
+ !vhost_vdpa_can_resume(v))
+ return -EOPNOTSUPP;
vhost_set_backend_features(&v->vdev, features);
return 0;
}
@@ -662,6 +689,8 @@ static long vhost_vdpa_unlocked_ioctl(struct file *filep,
features = VHOST_VDPA_BACKEND_FEATURES;
if (vhost_vdpa_can_suspend(v))
features |= BIT_ULL(VHOST_BACKEND_F_SUSPEND);
+ if (vhost_vdpa_can_resume(v))
+ features |= BIT_ULL(VHOST_BACKEND_F_RESUME);
if (copy_to_user(featurep, &features, sizeof(features)))
r = -EFAULT;
break;
@@ -677,6 +706,9 @@ static long vhost_vdpa_unlocked_ioctl(struct file *filep,
case VHOST_VDPA_SUSPEND:
r = vhost_vdpa_suspend(v);
break;
+ case VHOST_VDPA_RESUME:
+ r = vhost_vdpa_resume(v);
+ break;
default:
r = vhost_dev_ioctl(&v->vdev, cmd, argp);
if (r == -ENOIOCTLCMD)
@@ -1119,8 +1151,11 @@ static int vhost_vdpa_alloc_domain(struct vhost_vdpa *v)
if (!bus)
return -EFAULT;
- if (!device_iommu_capable(dma_dev, IOMMU_CAP_CACHE_COHERENCY))
+ if (!device_iommu_capable(dma_dev, IOMMU_CAP_CACHE_COHERENCY)) {
+ dev_warn_once(&v->dev,
+ "Failed to allocate domain, device is not IOMMU cache coherent capable\n");
return -ENOTSUPP;
+ }
v->domain = iommu_domain_alloc(bus);
if (!v->domain)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 43c9770b86e5..f11bdbe4c2c5 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -1730,7 +1730,7 @@ long vhost_vring_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *arg
}
EXPORT_SYMBOL_GPL(vhost_vring_ioctl);
-int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled)
+int vhost_init_device_iotlb(struct vhost_dev *d)
{
struct vhost_iotlb *niotlb, *oiotlb;
int i;
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index 790b296271f1..1647b750169c 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -222,7 +222,7 @@ ssize_t vhost_chr_read_iter(struct vhost_dev *dev, struct iov_iter *to,
int noblock);
ssize_t vhost_chr_write_iter(struct vhost_dev *dev,
struct iov_iter *from);
-int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled);
+int vhost_init_device_iotlb(struct vhost_dev *d);
void vhost_iotlb_map_free(struct vhost_iotlb *iotlb,
struct vhost_iotlb_map *map);
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index 1f3b89c885cc..c8e6087769a1 100644
--- a/drivers/vhost/vsock.c
+++ b/drivers/vhost/vsock.c
@@ -793,7 +793,7 @@ static int vhost_vsock_set_features(struct vhost_vsock *vsock, u64 features)
}
if ((features & (1ULL << VIRTIO_F_ACCESS_PLATFORM))) {
- if (vhost_init_device_iotlb(&vsock->dev, true))
+ if (vhost_init_device_iotlb(&vsock->dev))
goto err;
}
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 723c4e29e1d3..41144b5246a8 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -202,6 +202,9 @@ struct vring_virtqueue {
/* DMA, allocation, and size information */
bool we_own_ring;
+ /* Device used for doing DMA */
+ struct device *dma_dev;
+
#ifdef DEBUG
/* They're supposed to lock for us. */
unsigned int in_use;
@@ -219,7 +222,8 @@ static struct virtqueue *__vring_new_virtqueue(unsigned int index,
bool context,
bool (*notify)(struct virtqueue *),
void (*callback)(struct virtqueue *),
- const char *name);
+ const char *name,
+ struct device *dma_dev);
static struct vring_desc_extra *vring_alloc_desc_extra(unsigned int num);
static void vring_free(struct virtqueue *_vq);
@@ -297,10 +301,11 @@ size_t virtio_max_dma_size(struct virtio_device *vdev)
EXPORT_SYMBOL_GPL(virtio_max_dma_size);
static void *vring_alloc_queue(struct virtio_device *vdev, size_t size,
- dma_addr_t *dma_handle, gfp_t flag)
+ dma_addr_t *dma_handle, gfp_t flag,
+ struct device *dma_dev)
{
if (vring_use_dma_api(vdev)) {
- return dma_alloc_coherent(vdev->dev.parent, size,
+ return dma_alloc_coherent(dma_dev, size,
dma_handle, flag);
} else {
void *queue = alloc_pages_exact(PAGE_ALIGN(size), flag);
@@ -330,10 +335,11 @@ static void *vring_alloc_queue(struct virtio_device *vdev, size_t size,
}
static void vring_free_queue(struct virtio_device *vdev, size_t size,
- void *queue, dma_addr_t dma_handle)
+ void *queue, dma_addr_t dma_handle,
+ struct device *dma_dev)
{
if (vring_use_dma_api(vdev))
- dma_free_coherent(vdev->dev.parent, size, queue, dma_handle);
+ dma_free_coherent(dma_dev, size, queue, dma_handle);
else
free_pages_exact(queue, PAGE_ALIGN(size));
}
@@ -341,11 +347,11 @@ static void vring_free_queue(struct virtio_device *vdev, size_t size,
/*
* The DMA ops on various arches are rather gnarly right now, and
* making all of the arch DMA ops work on the vring device itself
- * is a mess. For now, we use the parent device for DMA ops.
+ * is a mess.
*/
static inline struct device *vring_dma_dev(const struct vring_virtqueue *vq)
{
- return vq->vq.vdev->dev.parent;
+ return vq->dma_dev;
}
/* Map one sg entry. */
@@ -1032,11 +1038,12 @@ err_state:
}
static void vring_free_split(struct vring_virtqueue_split *vring_split,
- struct virtio_device *vdev)
+ struct virtio_device *vdev, struct device *dma_dev)
{
vring_free_queue(vdev, vring_split->queue_size_in_bytes,
vring_split->vring.desc,
- vring_split->queue_dma_addr);
+ vring_split->queue_dma_addr,
+ dma_dev);
kfree(vring_split->desc_state);
kfree(vring_split->desc_extra);
@@ -1046,7 +1053,8 @@ static int vring_alloc_queue_split(struct vring_virtqueue_split *vring_split,
struct virtio_device *vdev,
u32 num,
unsigned int vring_align,
- bool may_reduce_num)
+ bool may_reduce_num,
+ struct device *dma_dev)
{
void *queue = NULL;
dma_addr_t dma_addr;
@@ -1061,7 +1069,8 @@ static int vring_alloc_queue_split(struct vring_virtqueue_split *vring_split,
for (; num && vring_size(num, vring_align) > PAGE_SIZE; num /= 2) {
queue = vring_alloc_queue(vdev, vring_size(num, vring_align),
&dma_addr,
- GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO);
+ GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO,
+ dma_dev);
if (queue)
break;
if (!may_reduce_num)
@@ -1074,7 +1083,8 @@ static int vring_alloc_queue_split(struct vring_virtqueue_split *vring_split,
if (!queue) {
/* Try to get a single page. You are my only hope! */
queue = vring_alloc_queue(vdev, vring_size(num, vring_align),
- &dma_addr, GFP_KERNEL | __GFP_ZERO);
+ &dma_addr, GFP_KERNEL | __GFP_ZERO,
+ dma_dev);
}
if (!queue)
return -ENOMEM;
@@ -1100,21 +1110,22 @@ static struct virtqueue *vring_create_virtqueue_split(
bool context,
bool (*notify)(struct virtqueue *),
void (*callback)(struct virtqueue *),
- const char *name)
+ const char *name,
+ struct device *dma_dev)
{
struct vring_virtqueue_split vring_split = {};
struct virtqueue *vq;
int err;
err = vring_alloc_queue_split(&vring_split, vdev, num, vring_align,
- may_reduce_num);
+ may_reduce_num, dma_dev);
if (err)
return NULL;
vq = __vring_new_virtqueue(index, &vring_split, vdev, weak_barriers,
- context, notify, callback, name);
+ context, notify, callback, name, dma_dev);
if (!vq) {
- vring_free_split(&vring_split, vdev);
+ vring_free_split(&vring_split, vdev, dma_dev);
return NULL;
}
@@ -1132,7 +1143,8 @@ static int virtqueue_resize_split(struct virtqueue *_vq, u32 num)
err = vring_alloc_queue_split(&vring_split, vdev, num,
vq->split.vring_align,
- vq->split.may_reduce_num);
+ vq->split.may_reduce_num,
+ vring_dma_dev(vq));
if (err)
goto err;
@@ -1150,7 +1162,7 @@ static int virtqueue_resize_split(struct virtqueue *_vq, u32 num)
return 0;
err_state_extra:
- vring_free_split(&vring_split, vdev);
+ vring_free_split(&vring_split, vdev, vring_dma_dev(vq));
err:
virtqueue_reinit_split(vq);
return -ENOMEM;
@@ -1841,22 +1853,26 @@ static struct vring_desc_extra *vring_alloc_desc_extra(unsigned int num)
}
static void vring_free_packed(struct vring_virtqueue_packed *vring_packed,
- struct virtio_device *vdev)
+ struct virtio_device *vdev,
+ struct device *dma_dev)
{
if (vring_packed->vring.desc)
vring_free_queue(vdev, vring_packed->ring_size_in_bytes,
vring_packed->vring.desc,
- vring_packed->ring_dma_addr);
+ vring_packed->ring_dma_addr,
+ dma_dev);
if (vring_packed->vring.driver)
vring_free_queue(vdev, vring_packed->event_size_in_bytes,
vring_packed->vring.driver,
- vring_packed->driver_event_dma_addr);
+ vring_packed->driver_event_dma_addr,
+ dma_dev);
if (vring_packed->vring.device)
vring_free_queue(vdev, vring_packed->event_size_in_bytes,
vring_packed->vring.device,
- vring_packed->device_event_dma_addr);
+ vring_packed->device_event_dma_addr,
+ dma_dev);
kfree(vring_packed->desc_state);
kfree(vring_packed->desc_extra);
@@ -1864,7 +1880,7 @@ static void vring_free_packed(struct vring_virtqueue_packed *vring_packed,
static int vring_alloc_queue_packed(struct vring_virtqueue_packed *vring_packed,
struct virtio_device *vdev,
- u32 num)
+ u32 num, struct device *dma_dev)
{
struct vring_packed_desc *ring;
struct vring_packed_desc_event *driver, *device;
@@ -1875,7 +1891,8 @@ static int vring_alloc_queue_packed(struct vring_virtqueue_packed *vring_packed,
ring = vring_alloc_queue(vdev, ring_size_in_bytes,
&ring_dma_addr,
- GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO);
+ GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO,
+ dma_dev);
if (!ring)
goto err;
@@ -1887,7 +1904,8 @@ static int vring_alloc_queue_packed(struct vring_virtqueue_packed *vring_packed,
driver = vring_alloc_queue(vdev, event_size_in_bytes,
&driver_event_dma_addr,
- GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO);
+ GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO,
+ dma_dev);
if (!driver)
goto err;
@@ -1897,7 +1915,8 @@ static int vring_alloc_queue_packed(struct vring_virtqueue_packed *vring_packed,
device = vring_alloc_queue(vdev, event_size_in_bytes,
&device_event_dma_addr,
- GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO);
+ GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO,
+ dma_dev);
if (!device)
goto err;
@@ -1909,7 +1928,7 @@ static int vring_alloc_queue_packed(struct vring_virtqueue_packed *vring_packed,
return 0;
err:
- vring_free_packed(vring_packed, vdev);
+ vring_free_packed(vring_packed, vdev, dma_dev);
return -ENOMEM;
}
@@ -1987,13 +2006,14 @@ static struct virtqueue *vring_create_virtqueue_packed(
bool context,
bool (*notify)(struct virtqueue *),
void (*callback)(struct virtqueue *),
- const char *name)
+ const char *name,
+ struct device *dma_dev)
{
struct vring_virtqueue_packed vring_packed = {};
struct vring_virtqueue *vq;
int err;
- if (vring_alloc_queue_packed(&vring_packed, vdev, num))
+ if (vring_alloc_queue_packed(&vring_packed, vdev, num, dma_dev))
goto err_ring;
vq = kmalloc(sizeof(*vq), GFP_KERNEL);
@@ -2014,6 +2034,7 @@ static struct virtqueue *vring_create_virtqueue_packed(
vq->broken = false;
#endif
vq->packed_ring = true;
+ vq->dma_dev = dma_dev;
vq->use_dma_api = vring_use_dma_api(vdev);
vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
@@ -2040,7 +2061,7 @@ static struct virtqueue *vring_create_virtqueue_packed(
err_state_extra:
kfree(vq);
err_vq:
- vring_free_packed(&vring_packed, vdev);
+ vring_free_packed(&vring_packed, vdev, dma_dev);
err_ring:
return NULL;
}
@@ -2052,7 +2073,7 @@ static int virtqueue_resize_packed(struct virtqueue *_vq, u32 num)
struct virtio_device *vdev = _vq->vdev;
int err;
- if (vring_alloc_queue_packed(&vring_packed, vdev, num))
+ if (vring_alloc_queue_packed(&vring_packed, vdev, num, vring_dma_dev(vq)))
goto err_ring;
err = vring_alloc_state_extra_packed(&vring_packed);
@@ -2069,7 +2090,7 @@ static int virtqueue_resize_packed(struct virtqueue *_vq, u32 num)
return 0;
err_state_extra:
- vring_free_packed(&vring_packed, vdev);
+ vring_free_packed(&vring_packed, vdev, vring_dma_dev(vq));
err_ring:
virtqueue_reinit_packed(vq);
return -ENOMEM;
@@ -2481,7 +2502,8 @@ static struct virtqueue *__vring_new_virtqueue(unsigned int index,
bool context,
bool (*notify)(struct virtqueue *),
void (*callback)(struct virtqueue *),
- const char *name)
+ const char *name,
+ struct device *dma_dev)
{
struct vring_virtqueue *vq;
int err;
@@ -2507,6 +2529,7 @@ static struct virtqueue *__vring_new_virtqueue(unsigned int index,
#else
vq->broken = false;
#endif
+ vq->dma_dev = dma_dev;
vq->use_dma_api = vring_use_dma_api(vdev);
vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC) &&
@@ -2549,14 +2572,39 @@ struct virtqueue *vring_create_virtqueue(
if (virtio_has_feature(vdev, VIRTIO_F_RING_PACKED))
return vring_create_virtqueue_packed(index, num, vring_align,
vdev, weak_barriers, may_reduce_num,
- context, notify, callback, name);
+ context, notify, callback, name, vdev->dev.parent);
return vring_create_virtqueue_split(index, num, vring_align,
vdev, weak_barriers, may_reduce_num,
- context, notify, callback, name);
+ context, notify, callback, name, vdev->dev.parent);
}
EXPORT_SYMBOL_GPL(vring_create_virtqueue);
+struct virtqueue *vring_create_virtqueue_dma(
+ unsigned int index,
+ unsigned int num,
+ unsigned int vring_align,
+ struct virtio_device *vdev,
+ bool weak_barriers,
+ bool may_reduce_num,
+ bool context,
+ bool (*notify)(struct virtqueue *),
+ void (*callback)(struct virtqueue *),
+ const char *name,
+ struct device *dma_dev)
+{
+
+ if (virtio_has_feature(vdev, VIRTIO_F_RING_PACKED))
+ return vring_create_virtqueue_packed(index, num, vring_align,
+ vdev, weak_barriers, may_reduce_num,
+ context, notify, callback, name, dma_dev);
+
+ return vring_create_virtqueue_split(index, num, vring_align,
+ vdev, weak_barriers, may_reduce_num,
+ context, notify, callback, name, dma_dev);
+}
+EXPORT_SYMBOL_GPL(vring_create_virtqueue_dma);
+
/**
* virtqueue_resize - resize the vring of vq
* @_vq: the struct virtqueue we're talking about.
@@ -2645,7 +2693,8 @@ struct virtqueue *vring_new_virtqueue(unsigned int index,
vring_init(&vring_split.vring, num, pages, vring_align);
return __vring_new_virtqueue(index, &vring_split, vdev, weak_barriers,
- context, notify, callback, name);
+ context, notify, callback, name,
+ vdev->dev.parent);
}
EXPORT_SYMBOL_GPL(vring_new_virtqueue);
@@ -2658,17 +2707,20 @@ static void vring_free(struct virtqueue *_vq)
vring_free_queue(vq->vq.vdev,
vq->packed.ring_size_in_bytes,
vq->packed.vring.desc,
- vq->packed.ring_dma_addr);
+ vq->packed.ring_dma_addr,
+ vring_dma_dev(vq));
vring_free_queue(vq->vq.vdev,
vq->packed.event_size_in_bytes,
vq->packed.vring.driver,
- vq->packed.driver_event_dma_addr);
+ vq->packed.driver_event_dma_addr,
+ vring_dma_dev(vq));
vring_free_queue(vq->vq.vdev,
vq->packed.event_size_in_bytes,
vq->packed.vring.device,
- vq->packed.device_event_dma_addr);
+ vq->packed.device_event_dma_addr,
+ vring_dma_dev(vq));
kfree(vq->packed.desc_state);
kfree(vq->packed.desc_extra);
@@ -2676,7 +2728,8 @@ static void vring_free(struct virtqueue *_vq)
vring_free_queue(vq->vq.vdev,
vq->split.queue_size_in_bytes,
vq->split.vring.desc,
- vq->split.queue_dma_addr);
+ vq->split.queue_dma_addr,
+ vring_dma_dev(vq));
}
}
if (!vq->packed_ring) {
diff --git a/drivers/virtio/virtio_vdpa.c b/drivers/virtio/virtio_vdpa.c
index 9670cc79371d..d7f5af62ddaa 100644
--- a/drivers/virtio/virtio_vdpa.c
+++ b/drivers/virtio/virtio_vdpa.c
@@ -135,6 +135,7 @@ virtio_vdpa_setup_vq(struct virtio_device *vdev, unsigned int index,
{
struct virtio_vdpa_device *vd_dev = to_virtio_vdpa_device(vdev);
struct vdpa_device *vdpa = vd_get_vdpa(vdev);
+ struct device *dma_dev;
const struct vdpa_config_ops *ops = vdpa->config;
struct virtio_vdpa_vq_info *info;
struct vdpa_callback cb;
@@ -175,9 +176,15 @@ virtio_vdpa_setup_vq(struct virtio_device *vdev, unsigned int index,
/* Create the vring */
align = ops->get_vq_align(vdpa);
- vq = vring_create_virtqueue(index, max_num, align, vdev,
- true, may_reduce_num, ctx,
- virtio_vdpa_notify, callback, name);
+
+ if (ops->get_vq_dma_dev)
+ dma_dev = ops->get_vq_dma_dev(vdpa, index);
+ else
+ dma_dev = vdpa_get_dma_dev(vdpa);
+ vq = vring_create_virtqueue_dma(index, max_num, align, vdev,
+ true, may_reduce_num, ctx,
+ virtio_vdpa_notify, callback,
+ name, dma_dev);
if (!vq) {
err = -ENOMEM;
goto error_new_virtqueue;
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index bc8f484cdcf3..45c3d62e616d 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -3094,6 +3094,8 @@
#define PCI_VENDOR_ID_3COM_2 0xa727
+#define PCI_VENDOR_ID_SOLIDRUN 0xd063
+
#define PCI_VENDOR_ID_DIGIUM 0xd161
#define PCI_DEVICE_ID_DIGIUM_HFC4S 0xb410
diff --git a/include/linux/vdpa.h b/include/linux/vdpa.h
index 6d0f5e4e82c2..43f59ef10cc9 100644
--- a/include/linux/vdpa.h
+++ b/include/linux/vdpa.h
@@ -219,7 +219,10 @@ struct vdpa_map_file {
* @reset: Reset device
* @vdev: vdpa device
* Returns integer: success (0) or error (< 0)
- * @suspend: Suspend or resume the device (optional)
+ * @suspend: Suspend the device (optional)
+ * @vdev: vdpa device
+ * Returns integer: success (0) or error (< 0)
+ * @resume: Resume the device (optional)
* @vdev: vdpa device
* Returns integer: success (0) or error (< 0)
* @get_config_size: Get the size of the configuration space includes
@@ -282,6 +285,11 @@ struct vdpa_map_file {
* @iova: iova to be unmapped
* @size: size of the area
* Returns integer: success (0) or error (< 0)
+ * @get_vq_dma_dev: Get the dma device for a specific
+ * virtqueue (optional)
+ * @vdev: vdpa device
+ * @idx: virtqueue index
+ * Returns pointer to structure device or error (NULL)
* @free: Free resources that belongs to vDPA (optional)
* @vdev: vdpa device
*/
@@ -324,6 +332,7 @@ struct vdpa_config_ops {
void (*set_status)(struct vdpa_device *vdev, u8 status);
int (*reset)(struct vdpa_device *vdev);
int (*suspend)(struct vdpa_device *vdev);
+ int (*resume)(struct vdpa_device *vdev);
size_t (*get_config_size)(struct vdpa_device *vdev);
void (*get_config)(struct vdpa_device *vdev, unsigned int offset,
void *buf, unsigned int len);
@@ -341,6 +350,7 @@ struct vdpa_config_ops {
u64 iova, u64 size);
int (*set_group_asid)(struct vdpa_device *vdev, unsigned int group,
unsigned int asid);
+ struct device *(*get_vq_dma_dev)(struct vdpa_device *vdev, u16 idx);
/* Free device resources */
void (*free)(struct vdpa_device *vdev);
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index 4b517649cfe8..2b3438de2c4d 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -16,8 +16,10 @@ struct virtio_shm_region {
u64 len;
};
+typedef void vq_callback_t(struct virtqueue *);
+
/**
- * virtio_config_ops - operations for configuring a virtio device
+ * struct virtio_config_ops - operations for configuring a virtio device
* Note: Do not assume that a transport implements all of the operations
* getting/setting a value as a simple read/write! Generally speaking,
* any of @get/@set, @get_status/@set_status, or @get_features/
@@ -69,7 +71,8 @@ struct virtio_shm_region {
* vdev: the virtio_device
* This sends the driver feature bits to the device: it can change
* the dev->feature bits if it wants.
- * Note: despite the name this can be called any number of times.
+ * Note that despite the name this can be called any number of
+ * times.
* Returns 0 on success or error status
* @bus_name: return the bus name associated with the device (optional)
* vdev: the virtio_device
@@ -91,7 +94,6 @@ struct virtio_shm_region {
* If disable_vq_and_reset is set, then enable_vq_after_reset must also be
* set.
*/
-typedef void vq_callback_t(struct virtqueue *);
struct virtio_config_ops {
void (*get)(struct virtio_device *vdev, unsigned offset,
void *buf, unsigned len);
diff --git a/include/linux/virtio_ring.h b/include/linux/virtio_ring.h
index 8b8af1a38991..8b95b69ef694 100644
--- a/include/linux/virtio_ring.h
+++ b/include/linux/virtio_ring.h
@@ -77,6 +77,22 @@ struct virtqueue *vring_create_virtqueue(unsigned int index,
const char *name);
/*
+ * Creates a virtqueue and allocates the descriptor ring with per
+ * virtqueue DMA device.
+ */
+struct virtqueue *vring_create_virtqueue_dma(unsigned int index,
+ unsigned int num,
+ unsigned int vring_align,
+ struct virtio_device *vdev,
+ bool weak_barriers,
+ bool may_reduce_num,
+ bool ctx,
+ bool (*notify)(struct virtqueue *vq),
+ void (*callback)(struct virtqueue *vq),
+ const char *name,
+ struct device *dma_dev);
+
+/*
* Creates a virtqueue with a standard layout but a caller-allocated
* ring.
*/
diff --git a/include/linux/vringh.h b/include/linux/vringh.h
index 212892cf9822..1991a02c6431 100644
--- a/include/linux/vringh.h
+++ b/include/linux/vringh.h
@@ -92,7 +92,7 @@ struct vringh_iov {
};
/**
- * struct vringh_iov - kvec mangler.
+ * struct vringh_kiov - kvec mangler.
*
* Mangles kvec in place, and restores it.
* Remaining data is iov + i, of used - i elements.
diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h
index f9f115a7c75b..92e1b700b51c 100644
--- a/include/uapi/linux/vhost.h
+++ b/include/uapi/linux/vhost.h
@@ -180,4 +180,12 @@
*/
#define VHOST_VDPA_SUSPEND _IO(VHOST_VIRTIO, 0x7D)
+/* Resume a device so it can resume processing virtqueue requests
+ *
+ * After the return of this ioctl the device will have restored all the
+ * necessary states and it is fully operational to continue processing the
+ * virtqueue descriptors.
+ */
+#define VHOST_VDPA_RESUME _IO(VHOST_VIRTIO, 0x7E)
+
#endif
diff --git a/include/uapi/linux/vhost_types.h b/include/uapi/linux/vhost_types.h
index 53601ce2c20a..c5690a8992d8 100644
--- a/include/uapi/linux/vhost_types.h
+++ b/include/uapi/linux/vhost_types.h
@@ -163,5 +163,7 @@ struct vhost_vdpa_iova_range {
#define VHOST_BACKEND_F_IOTLB_ASID 0x3
/* Device can be suspended */
#define VHOST_BACKEND_F_SUSPEND 0x4
+/* Device can be resumed */
+#define VHOST_BACKEND_F_RESUME 0x5
#endif
diff --git a/include/uapi/linux/virtio_blk.h b/include/uapi/linux/virtio_blk.h
index 58e70b24b504..5af2a0300bb9 100644
--- a/include/uapi/linux/virtio_blk.h
+++ b/include/uapi/linux/virtio_blk.h
@@ -41,6 +41,7 @@
#define VIRTIO_BLK_F_DISCARD 13 /* DISCARD is supported */
#define VIRTIO_BLK_F_WRITE_ZEROES 14 /* WRITE ZEROES is supported */
#define VIRTIO_BLK_F_SECURE_ERASE 16 /* Secure Erase is supported */
+#define VIRTIO_BLK_F_ZONED 17 /* Zoned block device */
/* Legacy feature bits */
#ifndef VIRTIO_BLK_NO_LEGACY
@@ -137,6 +138,16 @@ struct virtio_blk_config {
/* Secure erase commands must be aligned to this number of sectors. */
__virtio32 secure_erase_sector_alignment;
+ /* Zoned block device characteristics (if VIRTIO_BLK_F_ZONED) */
+ struct virtio_blk_zoned_characteristics {
+ __le32 zone_sectors;
+ __le32 max_open_zones;
+ __le32 max_active_zones;
+ __le32 max_append_sectors;
+ __le32 write_granularity;
+ __u8 model;
+ __u8 unused2[3];
+ } zoned;
} __attribute__((packed));
/*
@@ -174,6 +185,27 @@ struct virtio_blk_config {
/* Secure erase command */
#define VIRTIO_BLK_T_SECURE_ERASE 14
+/* Zone append command */
+#define VIRTIO_BLK_T_ZONE_APPEND 15
+
+/* Report zones command */
+#define VIRTIO_BLK_T_ZONE_REPORT 16
+
+/* Open zone command */
+#define VIRTIO_BLK_T_ZONE_OPEN 18
+
+/* Close zone command */
+#define VIRTIO_BLK_T_ZONE_CLOSE 20
+
+/* Finish zone command */
+#define VIRTIO_BLK_T_ZONE_FINISH 22
+
+/* Reset zone command */
+#define VIRTIO_BLK_T_ZONE_RESET 24
+
+/* Reset All zones command */
+#define VIRTIO_BLK_T_ZONE_RESET_ALL 26
+
#ifndef VIRTIO_BLK_NO_LEGACY
/* Barrier before this op. */
#define VIRTIO_BLK_T_BARRIER 0x80000000
@@ -193,6 +225,72 @@ struct virtio_blk_outhdr {
__virtio64 sector;
};
+/*
+ * Supported zoned device models.
+ */
+
+/* Regular block device */
+#define VIRTIO_BLK_Z_NONE 0
+/* Host-managed zoned device */
+#define VIRTIO_BLK_Z_HM 1
+/* Host-aware zoned device */
+#define VIRTIO_BLK_Z_HA 2
+
+/*
+ * Zone descriptor. A part of VIRTIO_BLK_T_ZONE_REPORT command reply.
+ */
+struct virtio_blk_zone_descriptor {
+ /* Zone capacity */
+ __le64 z_cap;
+ /* The starting sector of the zone */
+ __le64 z_start;
+ /* Zone write pointer position in sectors */
+ __le64 z_wp;
+ /* Zone type */
+ __u8 z_type;
+ /* Zone state */
+ __u8 z_state;
+ __u8 reserved[38];
+};
+
+struct virtio_blk_zone_report {
+ __le64 nr_zones;
+ __u8 reserved[56];
+ struct virtio_blk_zone_descriptor zones[];
+};
+
+/*
+ * Supported zone types.
+ */
+
+/* Conventional zone */
+#define VIRTIO_BLK_ZT_CONV 1
+/* Sequential Write Required zone */
+#define VIRTIO_BLK_ZT_SWR 2
+/* Sequential Write Preferred zone */
+#define VIRTIO_BLK_ZT_SWP 3
+
+/*
+ * Zone states that are available for zones of all types.
+ */
+
+/* Not a write pointer (conventional zones only) */
+#define VIRTIO_BLK_ZS_NOT_WP 0
+/* Empty */
+#define VIRTIO_BLK_ZS_EMPTY 1
+/* Implicitly Open */
+#define VIRTIO_BLK_ZS_IOPEN 2
+/* Explicitly Open */
+#define VIRTIO_BLK_ZS_EOPEN 3
+/* Closed */
+#define VIRTIO_BLK_ZS_CLOSED 4
+/* Read-Only */
+#define VIRTIO_BLK_ZS_RDONLY 13
+/* Full */
+#define VIRTIO_BLK_ZS_FULL 14
+/* Offline */
+#define VIRTIO_BLK_ZS_OFFLINE 15
+
/* Unmap this range (only valid for write zeroes command) */
#define VIRTIO_BLK_WRITE_ZEROES_FLAG_UNMAP 0x00000001
@@ -219,4 +317,11 @@ struct virtio_scsi_inhdr {
#define VIRTIO_BLK_S_OK 0
#define VIRTIO_BLK_S_IOERR 1
#define VIRTIO_BLK_S_UNSUPP 2
+
+/* Error codes that are specific to zoned block devices */
+#define VIRTIO_BLK_S_ZONE_INVALID_CMD 3
+#define VIRTIO_BLK_S_ZONE_UNALIGNED_WP 4
+#define VIRTIO_BLK_S_ZONE_OPEN_RESOURCE 5
+#define VIRTIO_BLK_S_ZONE_ACTIVE_RESOURCE 6
+
#endif /* _LINUX_VIRTIO_BLK_H */
diff --git a/tools/virtio/Makefile b/tools/virtio/Makefile
index 1b25cc7c64bb..7b7139d97d74 100644
--- a/tools/virtio/Makefile
+++ b/tools/virtio/Makefile
@@ -4,7 +4,7 @@ test: virtio_test vringh_test
virtio_test: virtio_ring.o virtio_test.o
vringh_test: vringh_test.o vringh.o virtio_ring.o
-CFLAGS += -g -O2 -Werror -Wno-maybe-uninitialized -Wall -I. -I../include/ -I ../../usr/include/ -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -include ../../include/linux/kconfig.h
+CFLAGS += -g -O2 -Werror -Wno-maybe-uninitialized -Wall -I. -I../include/ -I ../../usr/include/ -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE -include ../../include/linux/kconfig.h -mfunction-return=thunk -fcf-protection=none -mindirect-branch-register
CFLAGS += -pthread
LDFLAGS += -pthread
vpath %.c ../../drivers/virtio ../../drivers/vhost