[v7,2/6] virtio-pmem: Add virtio pmem driver
diff mbox series

Message ID 20190426050039.17460-3-pagupta@redhat.com
State Superseded
Headers show
Series
  • virtio pmem driver
Related show

Commit Message

Pankaj Gupta April 26, 2019, 5 a.m. UTC
This patch adds virtio-pmem driver for KVM guest.

Guest reads the persistent memory range information from
Qemu over VIRTIO and registers it on nvdimm_bus. It also
creates a nd_region object with the persistent memory
range information so that existing 'nvdimm/pmem' driver
can reserve this into system memory map. This way
'virtio-pmem' driver uses existing functionality of pmem
driver to register persistent memory compatible for DAX
capable filesystems.

This also provides function to perform guest flush over
VIRTIO from 'pmem' driver when userspace performs flush
on DAX memory range.

Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
---
 drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
 drivers/virtio/Kconfig           |  10 +++
 drivers/virtio/Makefile          |   1 +
 drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
 include/linux/virtio_pmem.h      |  60 ++++++++++++++++
 include/uapi/linux/virtio_ids.h  |   1 +
 include/uapi/linux/virtio_pmem.h |  10 +++
 7 files changed, 314 insertions(+)
 create mode 100644 drivers/nvdimm/virtio_pmem.c
 create mode 100644 drivers/virtio/pmem.c
 create mode 100644 include/linux/virtio_pmem.h
 create mode 100644 include/uapi/linux/virtio_pmem.h

Comments

Yuval Shaia April 30, 2019, 5:53 a.m. UTC | #1
On Fri, Apr 26, 2019 at 10:30:35AM +0530, Pankaj Gupta wrote:
> This patch adds virtio-pmem driver for KVM guest.
> 
> Guest reads the persistent memory range information from
> Qemu over VIRTIO and registers it on nvdimm_bus. It also
> creates a nd_region object with the persistent memory
> range information so that existing 'nvdimm/pmem' driver
> can reserve this into system memory map. This way
> 'virtio-pmem' driver uses existing functionality of pmem
> driver to register persistent memory compatible for DAX
> capable filesystems.
> 
> This also provides function to perform guest flush over
> VIRTIO from 'pmem' driver when userspace performs flush
> on DAX memory range.
> 
> Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> ---
>  drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
>  drivers/virtio/Kconfig           |  10 +++
>  drivers/virtio/Makefile          |   1 +
>  drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
>  include/linux/virtio_pmem.h      |  60 ++++++++++++++++
>  include/uapi/linux/virtio_ids.h  |   1 +
>  include/uapi/linux/virtio_pmem.h |  10 +++
>  7 files changed, 314 insertions(+)
>  create mode 100644 drivers/nvdimm/virtio_pmem.c
>  create mode 100644 drivers/virtio/pmem.c
>  create mode 100644 include/linux/virtio_pmem.h
>  create mode 100644 include/uapi/linux/virtio_pmem.h
> 
> diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> new file mode 100644
> index 000000000000..66b582f751a3
> --- /dev/null
> +++ b/drivers/nvdimm/virtio_pmem.c
> @@ -0,0 +1,114 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * virtio_pmem.c: Virtio pmem Driver
> + *
> + * Discovers persistent memory range information
> + * from host and provides a virtio based flushing
> + * interface.
> + */
> +#include <linux/virtio_pmem.h>
> +#include "nd.h"
> +
> + /* The interrupt handler */
> +void host_ack(struct virtqueue *vq)
> +{
> +	unsigned int len;
> +	unsigned long flags;
> +	struct virtio_pmem_request *req, *req_buf;
> +	struct virtio_pmem *vpmem = vq->vdev->priv;
> +
> +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +	while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> +		req->done = true;
> +		wake_up(&req->host_acked);
> +
> +		if (!list_empty(&vpmem->req_list)) {
> +			req_buf = list_first_entry(&vpmem->req_list,
> +					struct virtio_pmem_request, list);
> +			list_del(&vpmem->req_list);
> +			req_buf->wq_buf_avail = true;
> +			wake_up(&req_buf->wq_buf);
> +		}
> +	}
> +	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +}
> +EXPORT_SYMBOL_GPL(host_ack);
> +
> + /* The request submission function */
> +int virtio_pmem_flush(struct nd_region *nd_region)
> +{
> +	int err;
> +	unsigned long flags;
> +	struct scatterlist *sgs[2], sg, ret;
> +	struct virtio_device *vdev = nd_region->provider_data;
> +	struct virtio_pmem *vpmem = vdev->priv;
> +	struct virtio_pmem_request *req;
> +
> +	might_sleep();
> +	req = kmalloc(sizeof(*req), GFP_KERNEL);
> +	if (!req)
> +		return -ENOMEM;
> +
> +	req->done = req->wq_buf_avail = false;
> +	strcpy(req->name, "FLUSH");
> +	init_waitqueue_head(&req->host_acked);
> +	init_waitqueue_head(&req->wq_buf);
> +	sg_init_one(&sg, req->name, strlen(req->name));
> +	sgs[0] = &sg;
> +	sg_init_one(&ret, &req->ret, sizeof(req->ret));
> +	sgs[1] = &ret;
> +
> +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> +	if (err) {
> +		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> +
> +		list_add_tail(&vpmem->req_list, &req->list);
> +		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +
> +		/* When host has read buffer, this completes via host_ack */
> +		wait_event(req->wq_buf, req->wq_buf_avail);
> +		spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +	}
> +	err = virtqueue_kick(vpmem->req_vq);
> +	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +
> +	if (!err) {
> +		err = -EIO;
> +		goto ret;
> +	}
> +	/* When host has read buffer, this completes via host_ack */
> +	wait_event(req->host_acked, req->done);
> +	err = req->ret;
> +ret:
> +	kfree(req);
> +	return err;
> +};
> +
> + /* The asynchronous flush callback function */
> +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
> +{
> +	int rc = 0;
> +
> +	/* Create child bio for asynchronous flush and chain with
> +	 * parent bio. Otherwise directly call nd_region flush.
> +	 */
> +	if (bio && bio->bi_iter.bi_sector != -1) {
> +		struct bio *child = bio_alloc(GFP_ATOMIC, 0);
> +
> +		if (!child)
> +			return -ENOMEM;
> +		bio_copy_dev(child, bio);
> +		child->bi_opf = REQ_PREFLUSH;
> +		child->bi_iter.bi_sector = -1;
> +		bio_chain(child, bio);
> +		submit_bio(child);
> +	} else {
> +		if (virtio_pmem_flush(nd_region))
> +			rc = -EIO;
> +	}
> +
> +	return rc;
> +};
> +EXPORT_SYMBOL_GPL(async_pmem_flush);
> +MODULE_LICENSE("GPL");
> diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> index 35897649c24f..9f634a2ed638 100644
> --- a/drivers/virtio/Kconfig
> +++ b/drivers/virtio/Kconfig
> @@ -42,6 +42,16 @@ config VIRTIO_PCI_LEGACY
>  
>  	  If unsure, say Y.
>  
> +config VIRTIO_PMEM
> +	tristate "Support for virtio pmem driver"
> +	depends on VIRTIO
> +	depends on LIBNVDIMM
> +	help
> +	This driver provides support for virtio based flushing interface
> +	for persistent memory range.
> +
> +	If unsure, say M.
> +
>  config VIRTIO_BALLOON
>  	tristate "Virtio balloon driver"
>  	depends on VIRTIO
> diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> index 3a2b5c5dcf46..143ce91eabe9 100644
> --- a/drivers/virtio/Makefile
> +++ b/drivers/virtio/Makefile
> @@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
>  virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
>  obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
>  obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> +obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
> diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
> new file mode 100644
> index 000000000000..309788628e41
> --- /dev/null
> +++ b/drivers/virtio/pmem.c
> @@ -0,0 +1,118 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * virtio_pmem.c: Virtio pmem Driver
> + *
> + * Discovers persistent memory range information
> + * from host and registers the virtual pmem device
> + * with libnvdimm core.
> + */
> +#include <linux/virtio_pmem.h>
> +#include <../../drivers/nvdimm/nd.h>
> +
> +static struct virtio_device_id id_table[] = {
> +	{ VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
> +	{ 0 },
> +};
> +
> + /* Initialize virt queue */
> +static int init_vq(struct virtio_pmem *vpmem)
> +{
> +	/* single vq */
> +	vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
> +				host_ack, "flush_queue");
> +	if (IS_ERR(vpmem->req_vq))
> +		return PTR_ERR(vpmem->req_vq);
> +
> +	spin_lock_init(&vpmem->pmem_lock);
> +	INIT_LIST_HEAD(&vpmem->req_list);
> +
> +	return 0;
> +};
> +
> +static int virtio_pmem_probe(struct virtio_device *vdev)
> +{
> +	int err = 0;
> +	struct resource res;
> +	struct virtio_pmem *vpmem;
> +	struct nd_region_desc ndr_desc = {};
> +	int nid = dev_to_node(&vdev->dev);
> +	struct nd_region *nd_region;
> +
> +	if (!vdev->config->get) {
> +		dev_err(&vdev->dev, "%s failure: config access disabled\n",
> +			__func__);
> +		return -EINVAL;
> +	}
> +
> +	vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
> +	if (!vpmem) {
> +		err = -ENOMEM;
> +		goto out_err;
> +	}
> +
> +	vpmem->vdev = vdev;
> +	vdev->priv = vpmem;
> +	err = init_vq(vpmem);
> +	if (err)
> +		goto out_err;
> +
> +	virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> +			start, &vpmem->start);
> +	virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> +			size, &vpmem->size);
> +
> +	res.start = vpmem->start;
> +	res.end   = vpmem->start + vpmem->size-1;
> +	vpmem->nd_desc.provider_name = "virtio-pmem";
> +	vpmem->nd_desc.module = THIS_MODULE;
> +
> +	vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
> +						&vpmem->nd_desc);
> +	if (!vpmem->nvdimm_bus)
> +		goto out_vq;
> +
> +	dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
> +
> +	ndr_desc.res = &res;
> +	ndr_desc.numa_node = nid;
> +	ndr_desc.flush = async_pmem_flush;
> +	set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
> +	set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
> +	nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus, &ndr_desc);
> +
> +	if (!nd_region)
> +		goto out_nd;
> +	nd_region->provider_data =  dev_to_virtio

Delete extra space here ----------^^
I think this will let you join the two lines.

> +					(nd_region->dev.parent->parent);
> +	return 0;
> +out_nd:
> +	err = -ENXIO;
> +	nvdimm_bus_unregister(vpmem->nvdimm_bus);
> +out_vq:
> +	vdev->config->del_vqs(vdev);
> +out_err:
> +	dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
> +	return err;
> +}
> +
> +static void virtio_pmem_remove(struct virtio_device *vdev)
> +{
> +	struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
> +
> +	nvdimm_bus_unregister(nvdimm_bus);
> +	vdev->config->del_vqs(vdev);
> +	vdev->config->reset(vdev);
> +}
> +
> +static struct virtio_driver virtio_pmem_driver = {
> +	.driver.name		= KBUILD_MODNAME,
> +	.driver.owner		= THIS_MODULE,
> +	.id_table		= id_table,
> +	.probe			= virtio_pmem_probe,
> +	.remove			= virtio_pmem_remove,
> +};
> +
> +module_virtio_driver(virtio_pmem_driver);
> +MODULE_DEVICE_TABLE(virtio, id_table);
> +MODULE_DESCRIPTION("Virtio pmem driver");
> +MODULE_LICENSE("GPL");
> diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
> new file mode 100644
> index 000000000000..ab1da877575d
> --- /dev/null
> +++ b/include/linux/virtio_pmem.h
> @@ -0,0 +1,60 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * virtio_pmem.h: virtio pmem Driver
> + *
> + * Discovers persistent memory range information
> + * from host and provides a virtio based flushing
> + * interface.
> + **/
> +
> +#ifndef _LINUX_VIRTIO_PMEM_H
> +#define _LINUX_VIRTIO_PMEM_H
> +
> +#include <linux/virtio_ids.h>
> +#include <linux/module.h>
> +#include <linux/virtio_config.h>
> +#include <uapi/linux/virtio_pmem.h>
> +#include <linux/libnvdimm.h>
> +#include <linux/spinlock.h>
> +
> +struct virtio_pmem_request {
> +	/* Host return status corresponding to flush request */
> +	int ret;
> +
> +	/* command name*/
> +	char name[16];
> +
> +	/* Wait queue to process deferred work after ack from host */
> +	wait_queue_head_t host_acked;
> +	bool done;
> +
> +	/* Wait queue to process deferred work after virt queue buffer avail */
> +	wait_queue_head_t wq_buf;
> +	bool wq_buf_avail;
> +	struct list_head list;
> +};
> +
> +struct virtio_pmem {
> +	struct virtio_device *vdev;
> +
> +	/* Virtio pmem request queue */
> +	struct virtqueue *req_vq;
> +
> +	/* nvdimm bus registers virtio pmem device */
> +	struct nvdimm_bus *nvdimm_bus;
> +	struct nvdimm_bus_descriptor nd_desc;
> +
> +	/* List to store deferred work if virtqueue is full */
> +	struct list_head req_list;
> +
> +	/* Synchronize virtqueue data */
> +	spinlock_t pmem_lock;
> +
> +	/* Memory region information */
> +	uint64_t start;
> +	uint64_t size;
> +};
> +
> +void host_ack(struct virtqueue *vq);
> +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio);
> +#endif
> diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h
> index 6d5c3b2d4f4d..32b2f94d1f58 100644
> --- a/include/uapi/linux/virtio_ids.h
> +++ b/include/uapi/linux/virtio_ids.h
> @@ -43,5 +43,6 @@
>  #define VIRTIO_ID_INPUT        18 /* virtio input */
>  #define VIRTIO_ID_VSOCK        19 /* virtio vsock transport */
>  #define VIRTIO_ID_CRYPTO       20 /* virtio crypto */
> +#define VIRTIO_ID_PMEM         27 /* virtio pmem */
>  
>  #endif /* _LINUX_VIRTIO_IDS_H */
> diff --git a/include/uapi/linux/virtio_pmem.h b/include/uapi/linux/virtio_pmem.h
> new file mode 100644
> index 000000000000..fa3f7d52717a
> --- /dev/null
> +++ b/include/uapi/linux/virtio_pmem.h
> @@ -0,0 +1,10 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +
> +#ifndef _UAPI_LINUX_VIRTIO_PMEM_H
> +#define _UAPI_LINUX_VIRTIO_PMEM_H
> +
> +struct virtio_pmem_config {
> +	__le64 start;
> +	__le64 size;
> +};
> +#endif

Suggesting to fix the above minor formatting error.

With this:

Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>

> -- 
> 2.20.1
> 
>
Pankaj Gupta April 30, 2019, 6:06 a.m. UTC | #2
> 
> On Fri, Apr 26, 2019 at 10:30:35AM +0530, Pankaj Gupta wrote:
> > This patch adds virtio-pmem driver for KVM guest.
> > 
> > Guest reads the persistent memory range information from
> > Qemu over VIRTIO and registers it on nvdimm_bus. It also
> > creates a nd_region object with the persistent memory
> > range information so that existing 'nvdimm/pmem' driver
> > can reserve this into system memory map. This way
> > 'virtio-pmem' driver uses existing functionality of pmem
> > driver to register persistent memory compatible for DAX
> > capable filesystems.
> > 
> > This also provides function to perform guest flush over
> > VIRTIO from 'pmem' driver when userspace performs flush
> > on DAX memory range.
> > 
> > Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> > ---
> >  drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
> >  drivers/virtio/Kconfig           |  10 +++
> >  drivers/virtio/Makefile          |   1 +
> >  drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
> >  include/linux/virtio_pmem.h      |  60 ++++++++++++++++
> >  include/uapi/linux/virtio_ids.h  |   1 +
> >  include/uapi/linux/virtio_pmem.h |  10 +++
> >  7 files changed, 314 insertions(+)
> >  create mode 100644 drivers/nvdimm/virtio_pmem.c
> >  create mode 100644 drivers/virtio/pmem.c
> >  create mode 100644 include/linux/virtio_pmem.h
> >  create mode 100644 include/uapi/linux/virtio_pmem.h
> > 
> > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > new file mode 100644
> > index 000000000000..66b582f751a3
> > --- /dev/null
> > +++ b/drivers/nvdimm/virtio_pmem.c
> > @@ -0,0 +1,114 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * virtio_pmem.c: Virtio pmem Driver
> > + *
> > + * Discovers persistent memory range information
> > + * from host and provides a virtio based flushing
> > + * interface.
> > + */
> > +#include <linux/virtio_pmem.h>
> > +#include "nd.h"
> > +
> > + /* The interrupt handler */
> > +void host_ack(struct virtqueue *vq)
> > +{
> > +	unsigned int len;
> > +	unsigned long flags;
> > +	struct virtio_pmem_request *req, *req_buf;
> > +	struct virtio_pmem *vpmem = vq->vdev->priv;
> > +
> > +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +	while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> > +		req->done = true;
> > +		wake_up(&req->host_acked);
> > +
> > +		if (!list_empty(&vpmem->req_list)) {
> > +			req_buf = list_first_entry(&vpmem->req_list,
> > +					struct virtio_pmem_request, list);
> > +			list_del(&vpmem->req_list);
> > +			req_buf->wq_buf_avail = true;
> > +			wake_up(&req_buf->wq_buf);
> > +		}
> > +	}
> > +	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +}
> > +EXPORT_SYMBOL_GPL(host_ack);
> > +
> > + /* The request submission function */
> > +int virtio_pmem_flush(struct nd_region *nd_region)
> > +{
> > +	int err;
> > +	unsigned long flags;
> > +	struct scatterlist *sgs[2], sg, ret;
> > +	struct virtio_device *vdev = nd_region->provider_data;
> > +	struct virtio_pmem *vpmem = vdev->priv;
> > +	struct virtio_pmem_request *req;
> > +
> > +	might_sleep();
> > +	req = kmalloc(sizeof(*req), GFP_KERNEL);
> > +	if (!req)
> > +		return -ENOMEM;
> > +
> > +	req->done = req->wq_buf_avail = false;
> > +	strcpy(req->name, "FLUSH");
> > +	init_waitqueue_head(&req->host_acked);
> > +	init_waitqueue_head(&req->wq_buf);
> > +	sg_init_one(&sg, req->name, strlen(req->name));
> > +	sgs[0] = &sg;
> > +	sg_init_one(&ret, &req->ret, sizeof(req->ret));
> > +	sgs[1] = &ret;
> > +
> > +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> > +	if (err) {
> > +		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> > +
> > +		list_add_tail(&vpmem->req_list, &req->list);
> > +		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +
> > +		/* When host has read buffer, this completes via host_ack */
> > +		wait_event(req->wq_buf, req->wq_buf_avail);
> > +		spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +	}
> > +	err = virtqueue_kick(vpmem->req_vq);
> > +	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +
> > +	if (!err) {
> > +		err = -EIO;
> > +		goto ret;
> > +	}
> > +	/* When host has read buffer, this completes via host_ack */
> > +	wait_event(req->host_acked, req->done);
> > +	err = req->ret;
> > +ret:
> > +	kfree(req);
> > +	return err;
> > +};
> > +
> > + /* The asynchronous flush callback function */
> > +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
> > +{
> > +	int rc = 0;
> > +
> > +	/* Create child bio for asynchronous flush and chain with
> > +	 * parent bio. Otherwise directly call nd_region flush.
> > +	 */
> > +	if (bio && bio->bi_iter.bi_sector != -1) {
> > +		struct bio *child = bio_alloc(GFP_ATOMIC, 0);
> > +
> > +		if (!child)
> > +			return -ENOMEM;
> > +		bio_copy_dev(child, bio);
> > +		child->bi_opf = REQ_PREFLUSH;
> > +		child->bi_iter.bi_sector = -1;
> > +		bio_chain(child, bio);
> > +		submit_bio(child);
> > +	} else {
> > +		if (virtio_pmem_flush(nd_region))
> > +			rc = -EIO;
> > +	}
> > +
> > +	return rc;
> > +};
> > +EXPORT_SYMBOL_GPL(async_pmem_flush);
> > +MODULE_LICENSE("GPL");
> > diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> > index 35897649c24f..9f634a2ed638 100644
> > --- a/drivers/virtio/Kconfig
> > +++ b/drivers/virtio/Kconfig
> > @@ -42,6 +42,16 @@ config VIRTIO_PCI_LEGACY
> >  
> >  	  If unsure, say Y.
> >  
> > +config VIRTIO_PMEM
> > +	tristate "Support for virtio pmem driver"
> > +	depends on VIRTIO
> > +	depends on LIBNVDIMM
> > +	help
> > +	This driver provides support for virtio based flushing interface
> > +	for persistent memory range.
> > +
> > +	If unsure, say M.
> > +
> >  config VIRTIO_BALLOON
> >  	tristate "Virtio balloon driver"
> >  	depends on VIRTIO
> > diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> > index 3a2b5c5dcf46..143ce91eabe9 100644
> > --- a/drivers/virtio/Makefile
> > +++ b/drivers/virtio/Makefile
> > @@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
> >  virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
> >  obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
> >  obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> > +obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
> > diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
> > new file mode 100644
> > index 000000000000..309788628e41
> > --- /dev/null
> > +++ b/drivers/virtio/pmem.c
> > @@ -0,0 +1,118 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * virtio_pmem.c: Virtio pmem Driver
> > + *
> > + * Discovers persistent memory range information
> > + * from host and registers the virtual pmem device
> > + * with libnvdimm core.
> > + */
> > +#include <linux/virtio_pmem.h>
> > +#include <../../drivers/nvdimm/nd.h>
> > +
> > +static struct virtio_device_id id_table[] = {
> > +	{ VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
> > +	{ 0 },
> > +};
> > +
> > + /* Initialize virt queue */
> > +static int init_vq(struct virtio_pmem *vpmem)
> > +{
> > +	/* single vq */
> > +	vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
> > +				host_ack, "flush_queue");
> > +	if (IS_ERR(vpmem->req_vq))
> > +		return PTR_ERR(vpmem->req_vq);
> > +
> > +	spin_lock_init(&vpmem->pmem_lock);
> > +	INIT_LIST_HEAD(&vpmem->req_list);
> > +
> > +	return 0;
> > +};
> > +
> > +static int virtio_pmem_probe(struct virtio_device *vdev)
> > +{
> > +	int err = 0;
> > +	struct resource res;
> > +	struct virtio_pmem *vpmem;
> > +	struct nd_region_desc ndr_desc = {};
> > +	int nid = dev_to_node(&vdev->dev);
> > +	struct nd_region *nd_region;
> > +
> > +	if (!vdev->config->get) {
> > +		dev_err(&vdev->dev, "%s failure: config access disabled\n",
> > +			__func__);
> > +		return -EINVAL;
> > +	}
> > +
> > +	vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
> > +	if (!vpmem) {
> > +		err = -ENOMEM;
> > +		goto out_err;
> > +	}
> > +
> > +	vpmem->vdev = vdev;
> > +	vdev->priv = vpmem;
> > +	err = init_vq(vpmem);
> > +	if (err)
> > +		goto out_err;
> > +
> > +	virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > +			start, &vpmem->start);
> > +	virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > +			size, &vpmem->size);
> > +
> > +	res.start = vpmem->start;
> > +	res.end   = vpmem->start + vpmem->size-1;
> > +	vpmem->nd_desc.provider_name = "virtio-pmem";
> > +	vpmem->nd_desc.module = THIS_MODULE;
> > +
> > +	vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
> > +						&vpmem->nd_desc);
> > +	if (!vpmem->nvdimm_bus)
> > +		goto out_vq;
> > +
> > +	dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
> > +
> > +	ndr_desc.res = &res;
> > +	ndr_desc.numa_node = nid;
> > +	ndr_desc.flush = async_pmem_flush;
> > +	set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
> > +	set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
> > +	nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus, &ndr_desc);
> > +
> > +	if (!nd_region)
> > +		goto out_nd;
> > +	nd_region->provider_data =  dev_to_virtio
> 
> Delete extra space here ----------^^

ah. Sure. 

> I think this will let you join the two lines.
> 
> > +					(nd_region->dev.parent->parent);
> > +	return 0;
> > +out_nd:
> > +	err = -ENXIO;
> > +	nvdimm_bus_unregister(vpmem->nvdimm_bus);
> > +out_vq:
> > +	vdev->config->del_vqs(vdev);
> > +out_err:
> > +	dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
> > +	return err;
> > +}
> > +
> > +static void virtio_pmem_remove(struct virtio_device *vdev)
> > +{
> > +	struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
> > +
> > +	nvdimm_bus_unregister(nvdimm_bus);
> > +	vdev->config->del_vqs(vdev);
> > +	vdev->config->reset(vdev);
> > +}
> > +
> > +static struct virtio_driver virtio_pmem_driver = {
> > +	.driver.name		= KBUILD_MODNAME,
> > +	.driver.owner		= THIS_MODULE,
> > +	.id_table		= id_table,
> > +	.probe			= virtio_pmem_probe,
> > +	.remove			= virtio_pmem_remove,
> > +};
> > +
> > +module_virtio_driver(virtio_pmem_driver);
> > +MODULE_DEVICE_TABLE(virtio, id_table);
> > +MODULE_DESCRIPTION("Virtio pmem driver");
> > +MODULE_LICENSE("GPL");
> > diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
> > new file mode 100644
> > index 000000000000..ab1da877575d
> > --- /dev/null
> > +++ b/include/linux/virtio_pmem.h
> > @@ -0,0 +1,60 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/*
> > + * virtio_pmem.h: virtio pmem Driver
> > + *
> > + * Discovers persistent memory range information
> > + * from host and provides a virtio based flushing
> > + * interface.
> > + **/
> > +
> > +#ifndef _LINUX_VIRTIO_PMEM_H
> > +#define _LINUX_VIRTIO_PMEM_H
> > +
> > +#include <linux/virtio_ids.h>
> > +#include <linux/module.h>
> > +#include <linux/virtio_config.h>
> > +#include <uapi/linux/virtio_pmem.h>
> > +#include <linux/libnvdimm.h>
> > +#include <linux/spinlock.h>
> > +
> > +struct virtio_pmem_request {
> > +	/* Host return status corresponding to flush request */
> > +	int ret;
> > +
> > +	/* command name*/
> > +	char name[16];
> > +
> > +	/* Wait queue to process deferred work after ack from host */
> > +	wait_queue_head_t host_acked;
> > +	bool done;
> > +
> > +	/* Wait queue to process deferred work after virt queue buffer avail */
> > +	wait_queue_head_t wq_buf;
> > +	bool wq_buf_avail;
> > +	struct list_head list;
> > +};
> > +
> > +struct virtio_pmem {
> > +	struct virtio_device *vdev;
> > +
> > +	/* Virtio pmem request queue */
> > +	struct virtqueue *req_vq;
> > +
> > +	/* nvdimm bus registers virtio pmem device */
> > +	struct nvdimm_bus *nvdimm_bus;
> > +	struct nvdimm_bus_descriptor nd_desc;
> > +
> > +	/* List to store deferred work if virtqueue is full */
> > +	struct list_head req_list;
> > +
> > +	/* Synchronize virtqueue data */
> > +	spinlock_t pmem_lock;
> > +
> > +	/* Memory region information */
> > +	uint64_t start;
> > +	uint64_t size;
> > +};
> > +
> > +void host_ack(struct virtqueue *vq);
> > +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio);
> > +#endif
> > diff --git a/include/uapi/linux/virtio_ids.h
> > b/include/uapi/linux/virtio_ids.h
> > index 6d5c3b2d4f4d..32b2f94d1f58 100644
> > --- a/include/uapi/linux/virtio_ids.h
> > +++ b/include/uapi/linux/virtio_ids.h
> > @@ -43,5 +43,6 @@
> >  #define VIRTIO_ID_INPUT        18 /* virtio input */
> >  #define VIRTIO_ID_VSOCK        19 /* virtio vsock transport */
> >  #define VIRTIO_ID_CRYPTO       20 /* virtio crypto */
> > +#define VIRTIO_ID_PMEM         27 /* virtio pmem */
> >  
> >  #endif /* _LINUX_VIRTIO_IDS_H */
> > diff --git a/include/uapi/linux/virtio_pmem.h
> > b/include/uapi/linux/virtio_pmem.h
> > new file mode 100644
> > index 000000000000..fa3f7d52717a
> > --- /dev/null
> > +++ b/include/uapi/linux/virtio_pmem.h
> > @@ -0,0 +1,10 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +
> > +#ifndef _UAPI_LINUX_VIRTIO_PMEM_H
> > +#define _UAPI_LINUX_VIRTIO_PMEM_H
> > +
> > +struct virtio_pmem_config {
> > +	__le64 start;
> > +	__le64 size;
> > +};
> > +#endif
> 
> Suggesting to fix the above minor formatting error.
> 
> With this:
> 
> Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>

Thank You!

Best regards,
Pankaj

> 
> > --
> > 2.20.1
> > 
> > 
> 
>
Dan Williams May 7, 2019, 3:35 p.m. UTC | #3
Hi Pankaj,

Some minor file placement comments below.

On Thu, Apr 25, 2019 at 10:02 PM Pankaj Gupta <pagupta@redhat.com> wrote:
>
> This patch adds virtio-pmem driver for KVM guest.
>
> Guest reads the persistent memory range information from
> Qemu over VIRTIO and registers it on nvdimm_bus. It also
> creates a nd_region object with the persistent memory
> range information so that existing 'nvdimm/pmem' driver
> can reserve this into system memory map. This way
> 'virtio-pmem' driver uses existing functionality of pmem
> driver to register persistent memory compatible for DAX
> capable filesystems.
>
> This also provides function to perform guest flush over
> VIRTIO from 'pmem' driver when userspace performs flush
> on DAX memory range.
>
> Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> ---
>  drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
>  drivers/virtio/Kconfig           |  10 +++
>  drivers/virtio/Makefile          |   1 +
>  drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
>  include/linux/virtio_pmem.h      |  60 ++++++++++++++++
>  include/uapi/linux/virtio_ids.h  |   1 +
>  include/uapi/linux/virtio_pmem.h |  10 +++
>  7 files changed, 314 insertions(+)
>  create mode 100644 drivers/nvdimm/virtio_pmem.c
>  create mode 100644 drivers/virtio/pmem.c
>  create mode 100644 include/linux/virtio_pmem.h
>  create mode 100644 include/uapi/linux/virtio_pmem.h
>
> diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> new file mode 100644
> index 000000000000..66b582f751a3
> --- /dev/null
> +++ b/drivers/nvdimm/virtio_pmem.c
> @@ -0,0 +1,114 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * virtio_pmem.c: Virtio pmem Driver
> + *
> + * Discovers persistent memory range information
> + * from host and provides a virtio based flushing
> + * interface.
> + */
> +#include <linux/virtio_pmem.h>
> +#include "nd.h"
> +
> + /* The interrupt handler */
> +void host_ack(struct virtqueue *vq)
> +{
> +       unsigned int len;
> +       unsigned long flags;
> +       struct virtio_pmem_request *req, *req_buf;
> +       struct virtio_pmem *vpmem = vq->vdev->priv;
> +
> +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +       while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> +               req->done = true;
> +               wake_up(&req->host_acked);
> +
> +               if (!list_empty(&vpmem->req_list)) {
> +                       req_buf = list_first_entry(&vpmem->req_list,
> +                                       struct virtio_pmem_request, list);
> +                       list_del(&vpmem->req_list);
> +                       req_buf->wq_buf_avail = true;
> +                       wake_up(&req_buf->wq_buf);
> +               }
> +       }
> +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +}
> +EXPORT_SYMBOL_GPL(host_ack);
> +
> + /* The request submission function */
> +int virtio_pmem_flush(struct nd_region *nd_region)
> +{
> +       int err;
> +       unsigned long flags;
> +       struct scatterlist *sgs[2], sg, ret;
> +       struct virtio_device *vdev = nd_region->provider_data;
> +       struct virtio_pmem *vpmem = vdev->priv;
> +       struct virtio_pmem_request *req;
> +
> +       might_sleep();
> +       req = kmalloc(sizeof(*req), GFP_KERNEL);
> +       if (!req)
> +               return -ENOMEM;
> +
> +       req->done = req->wq_buf_avail = false;
> +       strcpy(req->name, "FLUSH");
> +       init_waitqueue_head(&req->host_acked);
> +       init_waitqueue_head(&req->wq_buf);
> +       sg_init_one(&sg, req->name, strlen(req->name));
> +       sgs[0] = &sg;
> +       sg_init_one(&ret, &req->ret, sizeof(req->ret));
> +       sgs[1] = &ret;
> +
> +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +       err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> +       if (err) {
> +               dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> +
> +               list_add_tail(&vpmem->req_list, &req->list);
> +               spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +
> +               /* When host has read buffer, this completes via host_ack */
> +               wait_event(req->wq_buf, req->wq_buf_avail);
> +               spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +       }
> +       err = virtqueue_kick(vpmem->req_vq);
> +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +
> +       if (!err) {
> +               err = -EIO;
> +               goto ret;
> +       }
> +       /* When host has read buffer, this completes via host_ack */
> +       wait_event(req->host_acked, req->done);
> +       err = req->ret;
> +ret:
> +       kfree(req);
> +       return err;
> +};
> +
> + /* The asynchronous flush callback function */
> +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
> +{
> +       int rc = 0;
> +
> +       /* Create child bio for asynchronous flush and chain with
> +        * parent bio. Otherwise directly call nd_region flush.
> +        */
> +       if (bio && bio->bi_iter.bi_sector != -1) {
> +               struct bio *child = bio_alloc(GFP_ATOMIC, 0);
> +
> +               if (!child)
> +                       return -ENOMEM;
> +               bio_copy_dev(child, bio);
> +               child->bi_opf = REQ_PREFLUSH;
> +               child->bi_iter.bi_sector = -1;
> +               bio_chain(child, bio);
> +               submit_bio(child);
> +       } else {
> +               if (virtio_pmem_flush(nd_region))
> +                       rc = -EIO;
> +       }
> +
> +       return rc;
> +};
> +EXPORT_SYMBOL_GPL(async_pmem_flush);
> +MODULE_LICENSE("GPL");
> diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> index 35897649c24f..9f634a2ed638 100644
> --- a/drivers/virtio/Kconfig
> +++ b/drivers/virtio/Kconfig
> @@ -42,6 +42,16 @@ config VIRTIO_PCI_LEGACY
>
>           If unsure, say Y.
>
> +config VIRTIO_PMEM
> +       tristate "Support for virtio pmem driver"
> +       depends on VIRTIO
> +       depends on LIBNVDIMM
> +       help
> +       This driver provides support for virtio based flushing interface
> +       for persistent memory range.
> +
> +       If unsure, say M.
> +
>  config VIRTIO_BALLOON
>         tristate "Virtio balloon driver"
>         depends on VIRTIO
> diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> index 3a2b5c5dcf46..143ce91eabe9 100644
> --- a/drivers/virtio/Makefile
> +++ b/drivers/virtio/Makefile
> @@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
>  virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
>  obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
>  obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> +obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
> diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
> new file mode 100644
> index 000000000000..309788628e41
> --- /dev/null
> +++ b/drivers/virtio/pmem.c

It's not clear to me why this driver is located in drivers/virtio/

> @@ -0,0 +1,118 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * virtio_pmem.c: Virtio pmem Driver
> + *
> + * Discovers persistent memory range information
> + * from host and registers the virtual pmem device
> + * with libnvdimm core.
> + */
> +#include <linux/virtio_pmem.h>
> +#include <../../drivers/nvdimm/nd.h>

...especially because it seems to require nvdimm internals.

However I don't see why that header is included.

In any event lets move this to drivers/nvdimm/virtio.c to live
alongside the other generic bus provider drivers/nvdimm/e820.c.

> +
> +static struct virtio_device_id id_table[] = {
> +       { VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
> +       { 0 },
> +};
> +
> + /* Initialize virt queue */
> +static int init_vq(struct virtio_pmem *vpmem)
> +{
> +       /* single vq */
> +       vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
> +                               host_ack, "flush_queue");
> +       if (IS_ERR(vpmem->req_vq))
> +               return PTR_ERR(vpmem->req_vq);
> +
> +       spin_lock_init(&vpmem->pmem_lock);
> +       INIT_LIST_HEAD(&vpmem->req_list);
> +
> +       return 0;
> +};
> +
> +static int virtio_pmem_probe(struct virtio_device *vdev)
> +{
> +       int err = 0;
> +       struct resource res;
> +       struct virtio_pmem *vpmem;
> +       struct nd_region_desc ndr_desc = {};
> +       int nid = dev_to_node(&vdev->dev);
> +       struct nd_region *nd_region;
> +
> +       if (!vdev->config->get) {
> +               dev_err(&vdev->dev, "%s failure: config access disabled\n",
> +                       __func__);
> +               return -EINVAL;
> +       }
> +
> +       vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
> +       if (!vpmem) {
> +               err = -ENOMEM;
> +               goto out_err;
> +       }
> +
> +       vpmem->vdev = vdev;
> +       vdev->priv = vpmem;
> +       err = init_vq(vpmem);
> +       if (err)
> +               goto out_err;
> +
> +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> +                       start, &vpmem->start);
> +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> +                       size, &vpmem->size);
> +
> +       res.start = vpmem->start;
> +       res.end   = vpmem->start + vpmem->size-1;
> +       vpmem->nd_desc.provider_name = "virtio-pmem";
> +       vpmem->nd_desc.module = THIS_MODULE;
> +
> +       vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
> +                                               &vpmem->nd_desc);
> +       if (!vpmem->nvdimm_bus)
> +               goto out_vq;
> +
> +       dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
> +
> +       ndr_desc.res = &res;
> +       ndr_desc.numa_node = nid;
> +       ndr_desc.flush = async_pmem_flush;
> +       set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
> +       set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
> +       nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus, &ndr_desc);
> +
> +       if (!nd_region)
> +               goto out_nd;
> +       nd_region->provider_data =  dev_to_virtio
> +                                       (nd_region->dev.parent->parent);
> +       return 0;
> +out_nd:
> +       err = -ENXIO;
> +       nvdimm_bus_unregister(vpmem->nvdimm_bus);
> +out_vq:
> +       vdev->config->del_vqs(vdev);
> +out_err:
> +       dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
> +       return err;
> +}
> +
> +static void virtio_pmem_remove(struct virtio_device *vdev)
> +{
> +       struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
> +
> +       nvdimm_bus_unregister(nvdimm_bus);
> +       vdev->config->del_vqs(vdev);
> +       vdev->config->reset(vdev);
> +}
> +
> +static struct virtio_driver virtio_pmem_driver = {
> +       .driver.name            = KBUILD_MODNAME,
> +       .driver.owner           = THIS_MODULE,
> +       .id_table               = id_table,
> +       .probe                  = virtio_pmem_probe,
> +       .remove                 = virtio_pmem_remove,
> +};
> +
> +module_virtio_driver(virtio_pmem_driver);
> +MODULE_DEVICE_TABLE(virtio, id_table);
> +MODULE_DESCRIPTION("Virtio pmem driver");
> +MODULE_LICENSE("GPL");
> diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
> new file mode 100644
> index 000000000000..ab1da877575d
> --- /dev/null
> +++ b/include/linux/virtio_pmem.h

Why is this a global header?

Seems it can move to drivers/nvdimm/virtio.h.

Also, I'd like to get a virtio ack from Michael (mst@redhat.com)
before taking this through the nvdimm tree.

> @@ -0,0 +1,60 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * virtio_pmem.h: virtio pmem Driver
> + *
> + * Discovers persistent memory range information
> + * from host and provides a virtio based flushing
> + * interface.
> + **/
> +
> +#ifndef _LINUX_VIRTIO_PMEM_H
> +#define _LINUX_VIRTIO_PMEM_H
> +
> +#include <linux/virtio_ids.h>
> +#include <linux/module.h>
> +#include <linux/virtio_config.h>
> +#include <uapi/linux/virtio_pmem.h>
> +#include <linux/libnvdimm.h>
> +#include <linux/spinlock.h>
> +
> +struct virtio_pmem_request {
> +       /* Host return status corresponding to flush request */
> +       int ret;
> +
> +       /* command name*/
> +       char name[16];
> +
> +       /* Wait queue to process deferred work after ack from host */
> +       wait_queue_head_t host_acked;
> +       bool done;
> +
> +       /* Wait queue to process deferred work after virt queue buffer avail */
> +       wait_queue_head_t wq_buf;
> +       bool wq_buf_avail;
> +       struct list_head list;
> +};
> +
> +struct virtio_pmem {
> +       struct virtio_device *vdev;
> +
> +       /* Virtio pmem request queue */
> +       struct virtqueue *req_vq;
> +
> +       /* nvdimm bus registers virtio pmem device */
> +       struct nvdimm_bus *nvdimm_bus;
> +       struct nvdimm_bus_descriptor nd_desc;
> +
> +       /* List to store deferred work if virtqueue is full */
> +       struct list_head req_list;
> +
> +       /* Synchronize virtqueue data */
> +       spinlock_t pmem_lock;
> +
> +       /* Memory region information */
> +       uint64_t start;
> +       uint64_t size;
> +};
> +
> +void host_ack(struct virtqueue *vq);
> +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio);
> +#endif
> diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h
> index 6d5c3b2d4f4d..32b2f94d1f58 100644
> --- a/include/uapi/linux/virtio_ids.h
> +++ b/include/uapi/linux/virtio_ids.h
> @@ -43,5 +43,6 @@
>  #define VIRTIO_ID_INPUT        18 /* virtio input */
>  #define VIRTIO_ID_VSOCK        19 /* virtio vsock transport */
>  #define VIRTIO_ID_CRYPTO       20 /* virtio crypto */
> +#define VIRTIO_ID_PMEM         27 /* virtio pmem */
>
>  #endif /* _LINUX_VIRTIO_IDS_H */
> diff --git a/include/uapi/linux/virtio_pmem.h b/include/uapi/linux/virtio_pmem.h
> new file mode 100644
> index 000000000000..fa3f7d52717a
> --- /dev/null
> +++ b/include/uapi/linux/virtio_pmem.h
> @@ -0,0 +1,10 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +
> +#ifndef _UAPI_LINUX_VIRTIO_PMEM_H
> +#define _UAPI_LINUX_VIRTIO_PMEM_H
> +
> +struct virtio_pmem_config {
> +       __le64 start;
> +       __le64 size;
> +};
> +#endif
> --
> 2.20.1
>
Jakub Staroń May 7, 2019, 8:25 p.m. UTC | #4
On 4/25/19 10:00 PM, Pankaj Gupta wrote:

> +void host_ack(struct virtqueue *vq)
> +{
> +	unsigned int len;
> +	unsigned long flags;
> +	struct virtio_pmem_request *req, *req_buf;
> +	struct virtio_pmem *vpmem = vq->vdev->priv;
> +
> +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +	while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> +		req->done = true;
> +		wake_up(&req->host_acked);
> +
> +		if (!list_empty(&vpmem->req_list)) {
> +			req_buf = list_first_entry(&vpmem->req_list,
> +					struct virtio_pmem_request, list);
> +			list_del(&vpmem->req_list);

Shouldn't it be rather `list_del(vpmem->req_list.next)`? We are trying to unlink
first element of the list and `vpmem->req_list` is just the list head.

> +int virtio_pmem_flush(struct nd_region *nd_region)
> +{
> +	int err;
> +	unsigned long flags;
> +	struct scatterlist *sgs[2], sg, ret;
> +	struct virtio_device *vdev = nd_region->provider_data;
> +	struct virtio_pmem *vpmem = vdev->priv;
> +	struct virtio_pmem_request *req;
> +
> +	might_sleep();
> +	req = kmalloc(sizeof(*req), GFP_KERNEL);
> +	if (!req)
> +		return -ENOMEM;
> +
> +	req->done = req->wq_buf_avail = false;
> +	strcpy(req->name, "FLUSH");
> +	init_waitqueue_head(&req->host_acked);
> +	init_waitqueue_head(&req->wq_buf);
> +	sg_init_one(&sg, req->name, strlen(req->name));
> +	sgs[0] = &sg;
> +	sg_init_one(&ret, &req->ret, sizeof(req->ret));
> +	sgs[1] = &ret;
> +
> +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> +	if (err) {
> +		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> +
> +		list_add_tail(&vpmem->req_list, &req->list);
> +		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +
> +		/* When host has read buffer, this completes via host_ack */
> +		wait_event(req->wq_buf, req->wq_buf_avail);
> +		spin_lock_irqsave(&vpmem->pmem_lock, flags);
> +	}

Aren't the arguments in `list_add_tail` swapped? The element we are adding should
be first, the list should be second. Also, shouldn't we resubmit the request after
waking up from `wait_event(req->wq_buf, req->wq_buf_avail)`?

I propose rewriting it like that:

diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
index 66b582f751a3..ff0556b04e86 100644
--- a/drivers/nvdimm/virtio_pmem.c
+++ b/drivers/nvdimm/virtio_pmem.c
@@ -25,7 +25,7 @@ void host_ack(struct virtqueue *vq)
 		if (!list_empty(&vpmem->req_list)) {
 			req_buf = list_first_entry(&vpmem->req_list,
 					struct virtio_pmem_request, list);
-			list_del(&vpmem->req_list);
+			list_del(vpmem->req_list.next);
 			req_buf->wq_buf_avail = true;
 			wake_up(&req_buf->wq_buf);
 		}
@@ -59,17 +59,33 @@ int virtio_pmem_flush(struct nd_region *nd_region)
 	sgs[1] = &ret;
 
 	spin_lock_irqsave(&vpmem->pmem_lock, flags);
-	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
-	if (err) {
-		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
+	/*
+	 * If virtqueue_add_sgs returns -ENOSPC then req_vq virtual queue does not
+	 * have free descriptor slots. We add the request to req_list and wait
+	 * for host_ack to wake us up when free slots are available.
+	 */
+	while ((err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC)) == -ENOSPC) {
+		dev_err(&vdev->dev, "failed to send command to virtio pmem device, no free slots in the virtqueue, postponing request\n");
+		req->wq_buf_avail = false;
 
-		list_add_tail(&vpmem->req_list, &req->list);
+		list_add_tail(&req->list, &vpmem->req_list);
 		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
 
 		/* When host has read buffer, this completes via host_ack */
 		wait_event(req->wq_buf, req->wq_buf_avail);
 		spin_lock_irqsave(&vpmem->pmem_lock, flags);
 	}
+
+	/*
+	 * virtqueue_add_sgs failed with error different than -ENOSPC, we can't
+	 * do anything about that.
+	 */
+	if (err) {
+		dev_info(&vdev->dev, "failed to send command to virtio pmem device, error code %d\n", err);
+		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
+		err = -EIO;
+		goto ret;
+	}
 	err = virtqueue_kick(vpmem->req_vq);
 	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);


Let me know if it looks reasonable to you.

Thank you,
Jakub Staron
Pankaj Gupta May 8, 2019, 11:12 a.m. UTC | #5
> 
> On 4/25/19 10:00 PM, Pankaj Gupta wrote:
> 
> > +void host_ack(struct virtqueue *vq)
> > +{
> > +	unsigned int len;
> > +	unsigned long flags;
> > +	struct virtio_pmem_request *req, *req_buf;
> > +	struct virtio_pmem *vpmem = vq->vdev->priv;
> > +
> > +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +	while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> > +		req->done = true;
> > +		wake_up(&req->host_acked);
> > +
> > +		if (!list_empty(&vpmem->req_list)) {
> > +			req_buf = list_first_entry(&vpmem->req_list,
> > +					struct virtio_pmem_request, list);
> > +			list_del(&vpmem->req_list);
> 
> Shouldn't it be rather `list_del(vpmem->req_list.next)`? We are trying to
> unlink
> first element of the list and `vpmem->req_list` is just the list head.

This looks correct. We are not deleting head but first entry in 'req_list'
which is device corresponding list of pending requests.

Please see below:

/**
 * Retrieve the first list entry for the given list pointer.
 *
 * Example:
 * struct foo *first;
 * first = list_first_entry(&bar->list_of_foos, struct foo, list_of_foos);
 *
 * @param ptr The list head
 * @param type Data type of the list element to retrieve
 * @param member Member name of the struct list_head field in the list element.
 * @return A pointer to the first list element.
 */
#define list_first_entry(ptr, type, member) \
    list_entry((ptr)->next, type, member)

> 
> > +int virtio_pmem_flush(struct nd_region *nd_region)
> > +{
> > +	int err;
> > +	unsigned long flags;
> > +	struct scatterlist *sgs[2], sg, ret;
> > +	struct virtio_device *vdev = nd_region->provider_data;
> > +	struct virtio_pmem *vpmem = vdev->priv;
> > +	struct virtio_pmem_request *req;
> > +
> > +	might_sleep();
> > +	req = kmalloc(sizeof(*req), GFP_KERNEL);
> > +	if (!req)
> > +		return -ENOMEM;
> > +
> > +	req->done = req->wq_buf_avail = false;
> > +	strcpy(req->name, "FLUSH");
> > +	init_waitqueue_head(&req->host_acked);
> > +	init_waitqueue_head(&req->wq_buf);
> > +	sg_init_one(&sg, req->name, strlen(req->name));
> > +	sgs[0] = &sg;
> > +	sg_init_one(&ret, &req->ret, sizeof(req->ret));
> > +	sgs[1] = &ret;
> > +
> > +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> > +	if (err) {
> > +		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> > +
> > +		list_add_tail(&vpmem->req_list, &req->list);
> > +		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +
> > +		/* When host has read buffer, this completes via host_ack */
> > +		wait_event(req->wq_buf, req->wq_buf_avail);
> > +		spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +	}
> 
> Aren't the arguments in `list_add_tail` swapped? The element we are adding

No, this is intentional. 'vpmem->req_list' maintains a list of pending requests
for entire pmem device.  'req->list'is per request list and maintains pending
request on virtio queue add failure. I think we don't need this list.

> should
> be first, the list should be second. Also, shouldn't we resubmit the request
> after
> waking up from `wait_event(req->wq_buf, req->wq_buf_avail)`?

Yes. we should. Good point.

> 
> I propose rewriting it like that:
> 
> diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> index 66b582f751a3..ff0556b04e86 100644
> --- a/drivers/nvdimm/virtio_pmem.c
> +++ b/drivers/nvdimm/virtio_pmem.c
> @@ -25,7 +25,7 @@ void host_ack(struct virtqueue *vq)
>  		if (!list_empty(&vpmem->req_list)) {
>  			req_buf = list_first_entry(&vpmem->req_list,
>  					struct virtio_pmem_request, list);
> -			list_del(&vpmem->req_list);
> +			list_del(vpmem->req_list.next);

Don't think its correct.

>  			req_buf->wq_buf_avail = true;
>  			wake_up(&req_buf->wq_buf);
>  		}
> @@ -59,17 +59,33 @@ int virtio_pmem_flush(struct nd_region *nd_region)
>  	sgs[1] = &ret;
>  
>  	spin_lock_irqsave(&vpmem->pmem_lock, flags);
> -	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> -	if (err) {
> -		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> +	/*
> +	 * If virtqueue_add_sgs returns -ENOSPC then req_vq virtual queue does not
> +	 * have free descriptor slots. We add the request to req_list and wait
> +	 * for host_ack to wake us up when free slots are available.
> +	 */
> +	while ((err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC))
> == -ENOSPC) {
> +		dev_err(&vdev->dev, "failed to send command to virtio pmem device, no free
> slots in the virtqueue, postponing request\n");
> +		req->wq_buf_avail = false;
>  
> -		list_add_tail(&vpmem->req_list, &req->list);
> +		list_add_tail(&req->list, &vpmem->req_list);
>  		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
>  
>  		/* When host has read buffer, this completes via host_ack */
>  		wait_event(req->wq_buf, req->wq_buf_avail);
>  		spin_lock_irqsave(&vpmem->pmem_lock, flags);
>  	}
> +
> +	/*
> +	 * virtqueue_add_sgs failed with error different than -ENOSPC, we can't
> +	 * do anything about that.
> +	 */
> +	if (err) {
> +		dev_info(&vdev->dev, "failed to send command to virtio pmem device, error
> code %d\n", err);
> +		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> +		err = -EIO;
> +		goto ret;
> +	}
>  	err = virtqueue_kick(vpmem->req_vq);
>  	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> 
> 
> Let me know if it looks reasonable to you.

Don't think this is fulfilling entire logic correctly. But thanks, I spotted a bug in my code :)
Will fix it. 

> 
> Thank you,
> Jakub Staron
> 
>
Pankaj Gupta May 8, 2019, 11:19 a.m. UTC | #6
Hi Dan,

Thank you for the review. Please see my reply inline.

> 
> Hi Pankaj,
> 
> Some minor file placement comments below.

Sure.

> 
> On Thu, Apr 25, 2019 at 10:02 PM Pankaj Gupta <pagupta@redhat.com> wrote:
> >
> > This patch adds virtio-pmem driver for KVM guest.
> >
> > Guest reads the persistent memory range information from
> > Qemu over VIRTIO and registers it on nvdimm_bus. It also
> > creates a nd_region object with the persistent memory
> > range information so that existing 'nvdimm/pmem' driver
> > can reserve this into system memory map. This way
> > 'virtio-pmem' driver uses existing functionality of pmem
> > driver to register persistent memory compatible for DAX
> > capable filesystems.
> >
> > This also provides function to perform guest flush over
> > VIRTIO from 'pmem' driver when userspace performs flush
> > on DAX memory range.
> >
> > Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> > ---
> >  drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
> >  drivers/virtio/Kconfig           |  10 +++
> >  drivers/virtio/Makefile          |   1 +
> >  drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
> >  include/linux/virtio_pmem.h      |  60 ++++++++++++++++
> >  include/uapi/linux/virtio_ids.h  |   1 +
> >  include/uapi/linux/virtio_pmem.h |  10 +++
> >  7 files changed, 314 insertions(+)
> >  create mode 100644 drivers/nvdimm/virtio_pmem.c
> >  create mode 100644 drivers/virtio/pmem.c
> >  create mode 100644 include/linux/virtio_pmem.h
> >  create mode 100644 include/uapi/linux/virtio_pmem.h
> >
> > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > new file mode 100644
> > index 000000000000..66b582f751a3
> > --- /dev/null
> > +++ b/drivers/nvdimm/virtio_pmem.c
> > @@ -0,0 +1,114 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * virtio_pmem.c: Virtio pmem Driver
> > + *
> > + * Discovers persistent memory range information
> > + * from host and provides a virtio based flushing
> > + * interface.
> > + */
> > +#include <linux/virtio_pmem.h>
> > +#include "nd.h"
> > +
> > + /* The interrupt handler */
> > +void host_ack(struct virtqueue *vq)
> > +{
> > +       unsigned int len;
> > +       unsigned long flags;
> > +       struct virtio_pmem_request *req, *req_buf;
> > +       struct virtio_pmem *vpmem = vq->vdev->priv;
> > +
> > +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +       while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> > +               req->done = true;
> > +               wake_up(&req->host_acked);
> > +
> > +               if (!list_empty(&vpmem->req_list)) {
> > +                       req_buf = list_first_entry(&vpmem->req_list,
> > +                                       struct virtio_pmem_request, list);
> > +                       list_del(&vpmem->req_list);
> > +                       req_buf->wq_buf_avail = true;
> > +                       wake_up(&req_buf->wq_buf);
> > +               }
> > +       }
> > +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +}
> > +EXPORT_SYMBOL_GPL(host_ack);
> > +
> > + /* The request submission function */
> > +int virtio_pmem_flush(struct nd_region *nd_region)
> > +{
> > +       int err;
> > +       unsigned long flags;
> > +       struct scatterlist *sgs[2], sg, ret;
> > +       struct virtio_device *vdev = nd_region->provider_data;
> > +       struct virtio_pmem *vpmem = vdev->priv;
> > +       struct virtio_pmem_request *req;
> > +
> > +       might_sleep();
> > +       req = kmalloc(sizeof(*req), GFP_KERNEL);
> > +       if (!req)
> > +               return -ENOMEM;
> > +
> > +       req->done = req->wq_buf_avail = false;
> > +       strcpy(req->name, "FLUSH");
> > +       init_waitqueue_head(&req->host_acked);
> > +       init_waitqueue_head(&req->wq_buf);
> > +       sg_init_one(&sg, req->name, strlen(req->name));
> > +       sgs[0] = &sg;
> > +       sg_init_one(&ret, &req->ret, sizeof(req->ret));
> > +       sgs[1] = &ret;
> > +
> > +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +       err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> > +       if (err) {
> > +               dev_err(&vdev->dev, "failed to send command to virtio pmem
> > device\n");
> > +
> > +               list_add_tail(&vpmem->req_list, &req->list);
> > +               spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +
> > +               /* When host has read buffer, this completes via host_ack
> > */
> > +               wait_event(req->wq_buf, req->wq_buf_avail);
> > +               spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > +       }
> > +       err = virtqueue_kick(vpmem->req_vq);
> > +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > +
> > +       if (!err) {
> > +               err = -EIO;
> > +               goto ret;
> > +       }
> > +       /* When host has read buffer, this completes via host_ack */
> > +       wait_event(req->host_acked, req->done);
> > +       err = req->ret;
> > +ret:
> > +       kfree(req);
> > +       return err;
> > +};
> > +
> > + /* The asynchronous flush callback function */
> > +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
> > +{
> > +       int rc = 0;
> > +
> > +       /* Create child bio for asynchronous flush and chain with
> > +        * parent bio. Otherwise directly call nd_region flush.
> > +        */
> > +       if (bio && bio->bi_iter.bi_sector != -1) {
> > +               struct bio *child = bio_alloc(GFP_ATOMIC, 0);
> > +
> > +               if (!child)
> > +                       return -ENOMEM;
> > +               bio_copy_dev(child, bio);
> > +               child->bi_opf = REQ_PREFLUSH;
> > +               child->bi_iter.bi_sector = -1;
> > +               bio_chain(child, bio);
> > +               submit_bio(child);
> > +       } else {
> > +               if (virtio_pmem_flush(nd_region))
> > +                       rc = -EIO;
> > +       }
> > +
> > +       return rc;
> > +};
> > +EXPORT_SYMBOL_GPL(async_pmem_flush);
> > +MODULE_LICENSE("GPL");
> > diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> > index 35897649c24f..9f634a2ed638 100644
> > --- a/drivers/virtio/Kconfig
> > +++ b/drivers/virtio/Kconfig
> > @@ -42,6 +42,16 @@ config VIRTIO_PCI_LEGACY
> >
> >           If unsure, say Y.
> >
> > +config VIRTIO_PMEM
> > +       tristate "Support for virtio pmem driver"
> > +       depends on VIRTIO
> > +       depends on LIBNVDIMM
> > +       help
> > +       This driver provides support for virtio based flushing interface
> > +       for persistent memory range.
> > +
> > +       If unsure, say M.
> > +
> >  config VIRTIO_BALLOON
> >         tristate "Virtio balloon driver"
> >         depends on VIRTIO
> > diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> > index 3a2b5c5dcf46..143ce91eabe9 100644
> > --- a/drivers/virtio/Makefile
> > +++ b/drivers/virtio/Makefile
> > @@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
> >  virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
> >  obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
> >  obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> > +obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
> > diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
> > new file mode 100644
> > index 000000000000..309788628e41
> > --- /dev/null
> > +++ b/drivers/virtio/pmem.c
> 
> It's not clear to me why this driver is located in drivers/virtio/

Like other VIRTIO drivers, I placed it initially in drivers/virtio directory.

> 
> > @@ -0,0 +1,118 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * virtio_pmem.c: Virtio pmem Driver
> > + *
> > + * Discovers persistent memory range information
> > + * from host and registers the virtual pmem device
> > + * with libnvdimm core.
> > + */
> > +#include <linux/virtio_pmem.h>
> > +#include <../../drivers/nvdimm/nd.h>
> 
> ...especially because it seems to require nvdimm internals.
> 
> However I don't see why that header is included.

Removed.

> 
> In any event lets move this to drivers/nvdimm/virtio.c to live
> alongside the other generic bus provider drivers/nvdimm/e820.c.

o.k. Makes sense.

> 
> > +
> > +static struct virtio_device_id id_table[] = {
> > +       { VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
> > +       { 0 },
> > +};
> > +
> > + /* Initialize virt queue */
> > +static int init_vq(struct virtio_pmem *vpmem)
> > +{
> > +       /* single vq */
> > +       vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
> > +                               host_ack, "flush_queue");
> > +       if (IS_ERR(vpmem->req_vq))
> > +               return PTR_ERR(vpmem->req_vq);
> > +
> > +       spin_lock_init(&vpmem->pmem_lock);
> > +       INIT_LIST_HEAD(&vpmem->req_list);
> > +
> > +       return 0;
> > +};
> > +
> > +static int virtio_pmem_probe(struct virtio_device *vdev)
> > +{
> > +       int err = 0;
> > +       struct resource res;
> > +       struct virtio_pmem *vpmem;
> > +       struct nd_region_desc ndr_desc = {};
> > +       int nid = dev_to_node(&vdev->dev);
> > +       struct nd_region *nd_region;
> > +
> > +       if (!vdev->config->get) {
> > +               dev_err(&vdev->dev, "%s failure: config access disabled\n",
> > +                       __func__);
> > +               return -EINVAL;
> > +       }
> > +
> > +       vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
> > +       if (!vpmem) {
> > +               err = -ENOMEM;
> > +               goto out_err;
> > +       }
> > +
> > +       vpmem->vdev = vdev;
> > +       vdev->priv = vpmem;
> > +       err = init_vq(vpmem);
> > +       if (err)
> > +               goto out_err;
> > +
> > +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > +                       start, &vpmem->start);
> > +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > +                       size, &vpmem->size);
> > +
> > +       res.start = vpmem->start;
> > +       res.end   = vpmem->start + vpmem->size-1;
> > +       vpmem->nd_desc.provider_name = "virtio-pmem";
> > +       vpmem->nd_desc.module = THIS_MODULE;
> > +
> > +       vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
> > +                                               &vpmem->nd_desc);
> > +       if (!vpmem->nvdimm_bus)
> > +               goto out_vq;
> > +
> > +       dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
> > +
> > +       ndr_desc.res = &res;
> > +       ndr_desc.numa_node = nid;
> > +       ndr_desc.flush = async_pmem_flush;
> > +       set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
> > +       set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
> > +       nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus,
> > &ndr_desc);
> > +
> > +       if (!nd_region)
> > +               goto out_nd;
> > +       nd_region->provider_data =  dev_to_virtio
> > +                                       (nd_region->dev.parent->parent);
> > +       return 0;
> > +out_nd:
> > +       err = -ENXIO;
> > +       nvdimm_bus_unregister(vpmem->nvdimm_bus);
> > +out_vq:
> > +       vdev->config->del_vqs(vdev);
> > +out_err:
> > +       dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
> > +       return err;
> > +}
> > +
> > +static void virtio_pmem_remove(struct virtio_device *vdev)
> > +{
> > +       struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
> > +
> > +       nvdimm_bus_unregister(nvdimm_bus);
> > +       vdev->config->del_vqs(vdev);
> > +       vdev->config->reset(vdev);
> > +}
> > +
> > +static struct virtio_driver virtio_pmem_driver = {
> > +       .driver.name            = KBUILD_MODNAME,
> > +       .driver.owner           = THIS_MODULE,
> > +       .id_table               = id_table,
> > +       .probe                  = virtio_pmem_probe,
> > +       .remove                 = virtio_pmem_remove,
> > +};
> > +
> > +module_virtio_driver(virtio_pmem_driver);
> > +MODULE_DEVICE_TABLE(virtio, id_table);
> > +MODULE_DESCRIPTION("Virtio pmem driver");
> > +MODULE_LICENSE("GPL");
> > diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
> > new file mode 100644
> > index 000000000000..ab1da877575d
> > --- /dev/null
> > +++ b/include/linux/virtio_pmem.h
> 
> Why is this a global header?

This is where other virtio driver headers are also placed.
I think this is to access uapi config file in :

./include/uapi/linux/virtio_pmem.h

Is it okay if we keep 'virtio_pmem.h' in global header?
  
> 
> Seems it can move to drivers/nvdimm/virtio.h.
> 
> Also, I'd like to get a virtio ack from Michael (mst@redhat.com)
> before taking this through the nvdimm tree.

Sure, Will post v8 with the suggestions.

Thanks,
Pankaj

> 
> > @@ -0,0 +1,60 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/*
> > + * virtio_pmem.h: virtio pmem Driver
> > + *
> > + * Discovers persistent memory range information
> > + * from host and provides a virtio based flushing
> > + * interface.
> > + **/
> > +
> > +#ifndef _LINUX_VIRTIO_PMEM_H
> > +#define _LINUX_VIRTIO_PMEM_H
> > +
> > +#include <linux/virtio_ids.h>
> > +#include <linux/module.h>
> > +#include <linux/virtio_config.h>
> > +#include <uapi/linux/virtio_pmem.h>
> > +#include <linux/libnvdimm.h>
> > +#include <linux/spinlock.h>
> > +
> > +struct virtio_pmem_request {
> > +       /* Host return status corresponding to flush request */
> > +       int ret;
> > +
> > +       /* command name*/
> > +       char name[16];
> > +
> > +       /* Wait queue to process deferred work after ack from host */
> > +       wait_queue_head_t host_acked;
> > +       bool done;
> > +
> > +       /* Wait queue to process deferred work after virt queue buffer
> > avail */
> > +       wait_queue_head_t wq_buf;
> > +       bool wq_buf_avail;
> > +       struct list_head list;
> > +};
> > +
> > +struct virtio_pmem {
> > +       struct virtio_device *vdev;
> > +
> > +       /* Virtio pmem request queue */
> > +       struct virtqueue *req_vq;
> > +
> > +       /* nvdimm bus registers virtio pmem device */
> > +       struct nvdimm_bus *nvdimm_bus;
> > +       struct nvdimm_bus_descriptor nd_desc;
> > +
> > +       /* List to store deferred work if virtqueue is full */
> > +       struct list_head req_list;
> > +
> > +       /* Synchronize virtqueue data */
> > +       spinlock_t pmem_lock;
> > +
> > +       /* Memory region information */
> > +       uint64_t start;
> > +       uint64_t size;
> > +};
> > +
> > +void host_ack(struct virtqueue *vq);
> > +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio);
> > +#endif
> > diff --git a/include/uapi/linux/virtio_ids.h
> > b/include/uapi/linux/virtio_ids.h
> > index 6d5c3b2d4f4d..32b2f94d1f58 100644
> > --- a/include/uapi/linux/virtio_ids.h
> > +++ b/include/uapi/linux/virtio_ids.h
> > @@ -43,5 +43,6 @@
> >  #define VIRTIO_ID_INPUT        18 /* virtio input */
> >  #define VIRTIO_ID_VSOCK        19 /* virtio vsock transport */
> >  #define VIRTIO_ID_CRYPTO       20 /* virtio crypto */
> > +#define VIRTIO_ID_PMEM         27 /* virtio pmem */
> >
> >  #endif /* _LINUX_VIRTIO_IDS_H */
> > diff --git a/include/uapi/linux/virtio_pmem.h
> > b/include/uapi/linux/virtio_pmem.h
> > new file mode 100644
> > index 000000000000..fa3f7d52717a
> > --- /dev/null
> > +++ b/include/uapi/linux/virtio_pmem.h
> > @@ -0,0 +1,10 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +
> > +#ifndef _UAPI_LINUX_VIRTIO_PMEM_H
> > +#define _UAPI_LINUX_VIRTIO_PMEM_H
> > +
> > +struct virtio_pmem_config {
> > +       __le64 start;
> > +       __le64 size;
> > +};
> > +#endif
> > --
> > 2.20.1
> >
> 
>
Pankaj Gupta May 8, 2019, 3:23 p.m. UTC | #7
> > 
> > > +int virtio_pmem_flush(struct nd_region *nd_region)
> > > +{
> > > +        int err;
> > > +        unsigned long flags;
> > > +        struct scatterlist *sgs[2], sg, ret;
> > > +        struct virtio_device *vdev = nd_region->provider_data;
> > > +        struct virtio_pmem *vpmem = vdev->priv;
> > > +        struct virtio_pmem_request *req;
> > > +
> > > +        might_sleep();
> > > +        req = kmalloc(sizeof(*req), GFP_KERNEL);
> > > +        if (!req)
> > > +                return -ENOMEM;
> > > +
> > > +        req->done = req->wq_buf_avail = false;
> > > +        strcpy(req->name, "FLUSH");
> > > +        init_waitqueue_head(&req->host_acked);
> > > +        init_waitqueue_head(&req->wq_buf);
> > > +        sg_init_one(&sg, req->name, strlen(req->name));
> > > +        sgs[0] = &sg;
> > > +        sg_init_one(&ret, &req->ret, sizeof(req->ret));
> > > +        sgs[1] = &ret;
> > > +
> > > +        spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > +        err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> > > +        if (err) {
> > > +                dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
> > > +
> > > +                list_add_tail(&vpmem->req_list, &req->list);
> > > +                spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > +
> > > +                /* When host has read buffer, this completes via host_ack */
> > > +                wait_event(req->wq_buf, req->wq_buf_avail);
> > > +                spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > +        }
> > 
> > Aren't the arguments in `list_add_tail` swapped? The element we are adding
> 

Yes, arguments for 'list_add_tail' should be swapped.

list_add_tail(&req->list, &vpmem->req_list);


Thank you,
Pankaj
Jakub Staroń May 8, 2019, 7:05 p.m. UTC | #8
On 5/8/19 4:12 AM, Pankaj Gupta wrote:
> 
>>
>> On 4/25/19 10:00 PM, Pankaj Gupta wrote:
>>
>>> +void host_ack(struct virtqueue *vq)
>>> +{
>>> +	unsigned int len;
>>> +	unsigned long flags;
>>> +	struct virtio_pmem_request *req, *req_buf;
>>> +	struct virtio_pmem *vpmem = vq->vdev->priv;
>>> +
>>> +	spin_lock_irqsave(&vpmem->pmem_lock, flags);
>>> +	while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
>>> +		req->done = true;
>>> +		wake_up(&req->host_acked);
>>> +
>>> +		if (!list_empty(&vpmem->req_list)) {
>>> +			req_buf = list_first_entry(&vpmem->req_list,
>>> +					struct virtio_pmem_request, list);
>>> +			list_del(&vpmem->req_list);
>>
>> Shouldn't it be rather `list_del(vpmem->req_list.next)`? We are trying to
>> unlink
>> first element of the list and `vpmem->req_list` is just the list head.
> 
> This looks correct. We are not deleting head but first entry in 'req_list'
> which is device corresponding list of pending requests.
> 
> Please see below:
> 
> /**
>  * Retrieve the first list entry for the given list pointer.
>  *
>  * Example:
>  * struct foo *first;
>  * first = list_first_entry(&bar->list_of_foos, struct foo, list_of_foos);
>  *
>  * @param ptr The list head
>  * @param type Data type of the list element to retrieve
>  * @param member Member name of the struct list_head field in the list element.
>  * @return A pointer to the first list element.
>  */
> #define list_first_entry(ptr, type, member) \
>     list_entry((ptr)->next, type, member)

Please look at this StackOverflow question:
https://stackoverflow.com/questions/19675419/deleting-first-element-of-a-list-h-list

Author asks about deleting first element of the queue. In our case
(and also in the question's author case), `vpmem->req_list` is not element
of any request struct and not an element of the list. It's just a list head storing 
`next` and `prev` pointers which are then pointing to respectively first and
last element of the list. We want to unlink the first element of the list,
so we need to pass pointer to the first element of the list to
the `list_del` function - that is, the `vpmem->req_list.next`.

Thank you,
Jakub Staron
Dan Williams May 10, 2019, 11:33 p.m. UTC | #9
On Wed, May 8, 2019 at 4:19 AM Pankaj Gupta <pagupta@redhat.com> wrote:
>
>
> Hi Dan,
>
> Thank you for the review. Please see my reply inline.
>
> >
> > Hi Pankaj,
> >
> > Some minor file placement comments below.
>
> Sure.
>
> >
> > On Thu, Apr 25, 2019 at 10:02 PM Pankaj Gupta <pagupta@redhat.com> wrote:
> > >
> > > This patch adds virtio-pmem driver for KVM guest.
> > >
> > > Guest reads the persistent memory range information from
> > > Qemu over VIRTIO and registers it on nvdimm_bus. It also
> > > creates a nd_region object with the persistent memory
> > > range information so that existing 'nvdimm/pmem' driver
> > > can reserve this into system memory map. This way
> > > 'virtio-pmem' driver uses existing functionality of pmem
> > > driver to register persistent memory compatible for DAX
> > > capable filesystems.
> > >
> > > This also provides function to perform guest flush over
> > > VIRTIO from 'pmem' driver when userspace performs flush
> > > on DAX memory range.
> > >
> > > Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> > > ---
> > >  drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
> > >  drivers/virtio/Kconfig           |  10 +++
> > >  drivers/virtio/Makefile          |   1 +
> > >  drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
> > >  include/linux/virtio_pmem.h      |  60 ++++++++++++++++
> > >  include/uapi/linux/virtio_ids.h  |   1 +
> > >  include/uapi/linux/virtio_pmem.h |  10 +++
> > >  7 files changed, 314 insertions(+)
> > >  create mode 100644 drivers/nvdimm/virtio_pmem.c
> > >  create mode 100644 drivers/virtio/pmem.c
> > >  create mode 100644 include/linux/virtio_pmem.h
> > >  create mode 100644 include/uapi/linux/virtio_pmem.h
> > >
> > > diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
> > > new file mode 100644
> > > index 000000000000..66b582f751a3
> > > --- /dev/null
> > > +++ b/drivers/nvdimm/virtio_pmem.c
> > > @@ -0,0 +1,114 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * virtio_pmem.c: Virtio pmem Driver
> > > + *
> > > + * Discovers persistent memory range information
> > > + * from host and provides a virtio based flushing
> > > + * interface.
> > > + */
> > > +#include <linux/virtio_pmem.h>
> > > +#include "nd.h"
> > > +
> > > + /* The interrupt handler */
> > > +void host_ack(struct virtqueue *vq)
> > > +{
> > > +       unsigned int len;
> > > +       unsigned long flags;
> > > +       struct virtio_pmem_request *req, *req_buf;
> > > +       struct virtio_pmem *vpmem = vq->vdev->priv;
> > > +
> > > +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > +       while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> > > +               req->done = true;
> > > +               wake_up(&req->host_acked);
> > > +
> > > +               if (!list_empty(&vpmem->req_list)) {
> > > +                       req_buf = list_first_entry(&vpmem->req_list,
> > > +                                       struct virtio_pmem_request, list);
> > > +                       list_del(&vpmem->req_list);
> > > +                       req_buf->wq_buf_avail = true;
> > > +                       wake_up(&req_buf->wq_buf);
> > > +               }
> > > +       }
> > > +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > +}
> > > +EXPORT_SYMBOL_GPL(host_ack);
> > > +
> > > + /* The request submission function */
> > > +int virtio_pmem_flush(struct nd_region *nd_region)
> > > +{
> > > +       int err;
> > > +       unsigned long flags;
> > > +       struct scatterlist *sgs[2], sg, ret;
> > > +       struct virtio_device *vdev = nd_region->provider_data;
> > > +       struct virtio_pmem *vpmem = vdev->priv;
> > > +       struct virtio_pmem_request *req;
> > > +
> > > +       might_sleep();
> > > +       req = kmalloc(sizeof(*req), GFP_KERNEL);
> > > +       if (!req)
> > > +               return -ENOMEM;
> > > +
> > > +       req->done = req->wq_buf_avail = false;
> > > +       strcpy(req->name, "FLUSH");
> > > +       init_waitqueue_head(&req->host_acked);
> > > +       init_waitqueue_head(&req->wq_buf);
> > > +       sg_init_one(&sg, req->name, strlen(req->name));
> > > +       sgs[0] = &sg;
> > > +       sg_init_one(&ret, &req->ret, sizeof(req->ret));
> > > +       sgs[1] = &ret;
> > > +
> > > +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > +       err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
> > > +       if (err) {
> > > +               dev_err(&vdev->dev, "failed to send command to virtio pmem
> > > device\n");
> > > +
> > > +               list_add_tail(&vpmem->req_list, &req->list);
> > > +               spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > +
> > > +               /* When host has read buffer, this completes via host_ack
> > > */
> > > +               wait_event(req->wq_buf, req->wq_buf_avail);
> > > +               spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > +       }
> > > +       err = virtqueue_kick(vpmem->req_vq);
> > > +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > +
> > > +       if (!err) {
> > > +               err = -EIO;
> > > +               goto ret;
> > > +       }
> > > +       /* When host has read buffer, this completes via host_ack */
> > > +       wait_event(req->host_acked, req->done);
> > > +       err = req->ret;
> > > +ret:
> > > +       kfree(req);
> > > +       return err;
> > > +};
> > > +
> > > + /* The asynchronous flush callback function */
> > > +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
> > > +{
> > > +       int rc = 0;
> > > +
> > > +       /* Create child bio for asynchronous flush and chain with
> > > +        * parent bio. Otherwise directly call nd_region flush.
> > > +        */
> > > +       if (bio && bio->bi_iter.bi_sector != -1) {
> > > +               struct bio *child = bio_alloc(GFP_ATOMIC, 0);
> > > +
> > > +               if (!child)
> > > +                       return -ENOMEM;
> > > +               bio_copy_dev(child, bio);
> > > +               child->bi_opf = REQ_PREFLUSH;
> > > +               child->bi_iter.bi_sector = -1;
> > > +               bio_chain(child, bio);
> > > +               submit_bio(child);
> > > +       } else {
> > > +               if (virtio_pmem_flush(nd_region))
> > > +                       rc = -EIO;
> > > +       }
> > > +
> > > +       return rc;
> > > +};
> > > +EXPORT_SYMBOL_GPL(async_pmem_flush);
> > > +MODULE_LICENSE("GPL");
> > > diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> > > index 35897649c24f..9f634a2ed638 100644
> > > --- a/drivers/virtio/Kconfig
> > > +++ b/drivers/virtio/Kconfig
> > > @@ -42,6 +42,16 @@ config VIRTIO_PCI_LEGACY
> > >
> > >           If unsure, say Y.
> > >
> > > +config VIRTIO_PMEM
> > > +       tristate "Support for virtio pmem driver"
> > > +       depends on VIRTIO
> > > +       depends on LIBNVDIMM
> > > +       help
> > > +       This driver provides support for virtio based flushing interface
> > > +       for persistent memory range.
> > > +
> > > +       If unsure, say M.
> > > +
> > >  config VIRTIO_BALLOON
> > >         tristate "Virtio balloon driver"
> > >         depends on VIRTIO
> > > diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> > > index 3a2b5c5dcf46..143ce91eabe9 100644
> > > --- a/drivers/virtio/Makefile
> > > +++ b/drivers/virtio/Makefile
> > > @@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
> > >  virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
> > >  obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
> > >  obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> > > +obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
> > > diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
> > > new file mode 100644
> > > index 000000000000..309788628e41
> > > --- /dev/null
> > > +++ b/drivers/virtio/pmem.c
> >
> > It's not clear to me why this driver is located in drivers/virtio/
>
> Like other VIRTIO drivers, I placed it initially in drivers/virtio directory.
>
> >
> > > @@ -0,0 +1,118 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * virtio_pmem.c: Virtio pmem Driver
> > > + *
> > > + * Discovers persistent memory range information
> > > + * from host and registers the virtual pmem device
> > > + * with libnvdimm core.
> > > + */
> > > +#include <linux/virtio_pmem.h>
> > > +#include <../../drivers/nvdimm/nd.h>
> >
> > ...especially because it seems to require nvdimm internals.
> >
> > However I don't see why that header is included.
>
> Removed.
>
> >
> > In any event lets move this to drivers/nvdimm/virtio.c to live
> > alongside the other generic bus provider drivers/nvdimm/e820.c.
>
> o.k. Makes sense.
>
> >
> > > +
> > > +static struct virtio_device_id id_table[] = {
> > > +       { VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
> > > +       { 0 },
> > > +};
> > > +
> > > + /* Initialize virt queue */
> > > +static int init_vq(struct virtio_pmem *vpmem)
> > > +{
> > > +       /* single vq */
> > > +       vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
> > > +                               host_ack, "flush_queue");
> > > +       if (IS_ERR(vpmem->req_vq))
> > > +               return PTR_ERR(vpmem->req_vq);
> > > +
> > > +       spin_lock_init(&vpmem->pmem_lock);
> > > +       INIT_LIST_HEAD(&vpmem->req_list);
> > > +
> > > +       return 0;
> > > +};
> > > +
> > > +static int virtio_pmem_probe(struct virtio_device *vdev)
> > > +{
> > > +       int err = 0;
> > > +       struct resource res;
> > > +       struct virtio_pmem *vpmem;
> > > +       struct nd_region_desc ndr_desc = {};
> > > +       int nid = dev_to_node(&vdev->dev);
> > > +       struct nd_region *nd_region;
> > > +
> > > +       if (!vdev->config->get) {
> > > +               dev_err(&vdev->dev, "%s failure: config access disabled\n",
> > > +                       __func__);
> > > +               return -EINVAL;
> > > +       }
> > > +
> > > +       vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
> > > +       if (!vpmem) {
> > > +               err = -ENOMEM;
> > > +               goto out_err;
> > > +       }
> > > +
> > > +       vpmem->vdev = vdev;
> > > +       vdev->priv = vpmem;
> > > +       err = init_vq(vpmem);
> > > +       if (err)
> > > +               goto out_err;
> > > +
> > > +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > > +                       start, &vpmem->start);
> > > +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > > +                       size, &vpmem->size);
> > > +
> > > +       res.start = vpmem->start;
> > > +       res.end   = vpmem->start + vpmem->size-1;
> > > +       vpmem->nd_desc.provider_name = "virtio-pmem";
> > > +       vpmem->nd_desc.module = THIS_MODULE;
> > > +
> > > +       vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
> > > +                                               &vpmem->nd_desc);
> > > +       if (!vpmem->nvdimm_bus)
> > > +               goto out_vq;
> > > +
> > > +       dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
> > > +
> > > +       ndr_desc.res = &res;
> > > +       ndr_desc.numa_node = nid;
> > > +       ndr_desc.flush = async_pmem_flush;
> > > +       set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
> > > +       set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
> > > +       nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus,
> > > &ndr_desc);
> > > +
> > > +       if (!nd_region)
> > > +               goto out_nd;
> > > +       nd_region->provider_data =  dev_to_virtio
> > > +                                       (nd_region->dev.parent->parent);
> > > +       return 0;
> > > +out_nd:
> > > +       err = -ENXIO;
> > > +       nvdimm_bus_unregister(vpmem->nvdimm_bus);
> > > +out_vq:
> > > +       vdev->config->del_vqs(vdev);
> > > +out_err:
> > > +       dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
> > > +       return err;
> > > +}
> > > +
> > > +static void virtio_pmem_remove(struct virtio_device *vdev)
> > > +{
> > > +       struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
> > > +
> > > +       nvdimm_bus_unregister(nvdimm_bus);
> > > +       vdev->config->del_vqs(vdev);
> > > +       vdev->config->reset(vdev);
> > > +}
> > > +
> > > +static struct virtio_driver virtio_pmem_driver = {
> > > +       .driver.name            = KBUILD_MODNAME,
> > > +       .driver.owner           = THIS_MODULE,
> > > +       .id_table               = id_table,
> > > +       .probe                  = virtio_pmem_probe,
> > > +       .remove                 = virtio_pmem_remove,
> > > +};
> > > +
> > > +module_virtio_driver(virtio_pmem_driver);
> > > +MODULE_DEVICE_TABLE(virtio, id_table);
> > > +MODULE_DESCRIPTION("Virtio pmem driver");
> > > +MODULE_LICENSE("GPL");
> > > diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
> > > new file mode 100644
> > > index 000000000000..ab1da877575d
> > > --- /dev/null
> > > +++ b/include/linux/virtio_pmem.h
> >
> > Why is this a global header?
>
> This is where other virtio driver headers are also placed.
> I think this is to access uapi config file in :
>
> ./include/uapi/linux/virtio_pmem.h
>
> Is it okay if we keep 'virtio_pmem.h' in global header?

No, I don't think so. While virtio_console.h and virtio_net.h make
sense as global headers because they are consumed from multiple
drivers, there is no need for virtio_caif.h, for example, to be a
global header. I see no practical reason that the private details of
virtio_pmem.h need to be made available outside of the virtio_pmem.c
consumer.
Pankaj Gupta May 11, 2019, 1:26 a.m. UTC | #10
> >
> > Hi Dan,
> >
> > Thank you for the review. Please see my reply inline.
> >
> > >
> > > Hi Pankaj,
> > >
> > > Some minor file placement comments below.
> >
> > Sure.
> >
> > >
> > > On Thu, Apr 25, 2019 at 10:02 PM Pankaj Gupta <pagupta@redhat.com> wrote:
> > > >
> > > > This patch adds virtio-pmem driver for KVM guest.
> > > >
> > > > Guest reads the persistent memory range information from
> > > > Qemu over VIRTIO and registers it on nvdimm_bus. It also
> > > > creates a nd_region object with the persistent memory
> > > > range information so that existing 'nvdimm/pmem' driver
> > > > can reserve this into system memory map. This way
> > > > 'virtio-pmem' driver uses existing functionality of pmem
> > > > driver to register persistent memory compatible for DAX
> > > > capable filesystems.
> > > >
> > > > This also provides function to perform guest flush over
> > > > VIRTIO from 'pmem' driver when userspace performs flush
> > > > on DAX memory range.
> > > >
> > > > Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> > > > ---
> > > >  drivers/nvdimm/virtio_pmem.c     | 114 +++++++++++++++++++++++++++++
> > > >  drivers/virtio/Kconfig           |  10 +++
> > > >  drivers/virtio/Makefile          |   1 +
> > > >  drivers/virtio/pmem.c            | 118 +++++++++++++++++++++++++++++++
> > > >  include/linux/virtio_pmem.h      |  60 ++++++++++++++++
> > > >  include/uapi/linux/virtio_ids.h  |   1 +
> > > >  include/uapi/linux/virtio_pmem.h |  10 +++
> > > >  7 files changed, 314 insertions(+)
> > > >  create mode 100644 drivers/nvdimm/virtio_pmem.c
> > > >  create mode 100644 drivers/virtio/pmem.c
> > > >  create mode 100644 include/linux/virtio_pmem.h
> > > >  create mode 100644 include/uapi/linux/virtio_pmem.h
> > > >
> > > > diff --git a/drivers/nvdimm/virtio_pmem.c
> > > > b/drivers/nvdimm/virtio_pmem.c
> > > > new file mode 100644
> > > > index 000000000000..66b582f751a3
> > > > --- /dev/null
> > > > +++ b/drivers/nvdimm/virtio_pmem.c
> > > > @@ -0,0 +1,114 @@
> > > > +// SPDX-License-Identifier: GPL-2.0
> > > > +/*
> > > > + * virtio_pmem.c: Virtio pmem Driver
> > > > + *
> > > > + * Discovers persistent memory range information
> > > > + * from host and provides a virtio based flushing
> > > > + * interface.
> > > > + */
> > > > +#include <linux/virtio_pmem.h>
> > > > +#include "nd.h"
> > > > +
> > > > + /* The interrupt handler */
> > > > +void host_ack(struct virtqueue *vq)
> > > > +{
> > > > +       unsigned int len;
> > > > +       unsigned long flags;
> > > > +       struct virtio_pmem_request *req, *req_buf;
> > > > +       struct virtio_pmem *vpmem = vq->vdev->priv;
> > > > +
> > > > +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > > +       while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
> > > > +               req->done = true;
> > > > +               wake_up(&req->host_acked);
> > > > +
> > > > +               if (!list_empty(&vpmem->req_list)) {
> > > > +                       req_buf = list_first_entry(&vpmem->req_list,
> > > > +                                       struct virtio_pmem_request,
> > > > list);
> > > > +                       list_del(&vpmem->req_list);
> > > > +                       req_buf->wq_buf_avail = true;
> > > > +                       wake_up(&req_buf->wq_buf);
> > > > +               }
> > > > +       }
> > > > +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > > +}
> > > > +EXPORT_SYMBOL_GPL(host_ack);
> > > > +
> > > > + /* The request submission function */
> > > > +int virtio_pmem_flush(struct nd_region *nd_region)
> > > > +{
> > > > +       int err;
> > > > +       unsigned long flags;
> > > > +       struct scatterlist *sgs[2], sg, ret;
> > > > +       struct virtio_device *vdev = nd_region->provider_data;
> > > > +       struct virtio_pmem *vpmem = vdev->priv;
> > > > +       struct virtio_pmem_request *req;
> > > > +
> > > > +       might_sleep();
> > > > +       req = kmalloc(sizeof(*req), GFP_KERNEL);
> > > > +       if (!req)
> > > > +               return -ENOMEM;
> > > > +
> > > > +       req->done = req->wq_buf_avail = false;
> > > > +       strcpy(req->name, "FLUSH");
> > > > +       init_waitqueue_head(&req->host_acked);
> > > > +       init_waitqueue_head(&req->wq_buf);
> > > > +       sg_init_one(&sg, req->name, strlen(req->name));
> > > > +       sgs[0] = &sg;
> > > > +       sg_init_one(&ret, &req->ret, sizeof(req->ret));
> > > > +       sgs[1] = &ret;
> > > > +
> > > > +       spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > > +       err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req,
> > > > GFP_ATOMIC);
> > > > +       if (err) {
> > > > +               dev_err(&vdev->dev, "failed to send command to virtio
> > > > pmem
> > > > device\n");
> > > > +
> > > > +               list_add_tail(&vpmem->req_list, &req->list);
> > > > +               spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > > +
> > > > +               /* When host has read buffer, this completes via
> > > > host_ack
> > > > */
> > > > +               wait_event(req->wq_buf, req->wq_buf_avail);
> > > > +               spin_lock_irqsave(&vpmem->pmem_lock, flags);
> > > > +       }
> > > > +       err = virtqueue_kick(vpmem->req_vq);
> > > > +       spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
> > > > +
> > > > +       if (!err) {
> > > > +               err = -EIO;
> > > > +               goto ret;
> > > > +       }
> > > > +       /* When host has read buffer, this completes via host_ack */
> > > > +       wait_event(req->host_acked, req->done);
> > > > +       err = req->ret;
> > > > +ret:
> > > > +       kfree(req);
> > > > +       return err;
> > > > +};
> > > > +
> > > > + /* The asynchronous flush callback function */
> > > > +int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
> > > > +{
> > > > +       int rc = 0;
> > > > +
> > > > +       /* Create child bio for asynchronous flush and chain with
> > > > +        * parent bio. Otherwise directly call nd_region flush.
> > > > +        */
> > > > +       if (bio && bio->bi_iter.bi_sector != -1) {
> > > > +               struct bio *child = bio_alloc(GFP_ATOMIC, 0);
> > > > +
> > > > +               if (!child)
> > > > +                       return -ENOMEM;
> > > > +               bio_copy_dev(child, bio);
> > > > +               child->bi_opf = REQ_PREFLUSH;
> > > > +               child->bi_iter.bi_sector = -1;
> > > > +               bio_chain(child, bio);
> > > > +               submit_bio(child);
> > > > +       } else {
> > > > +               if (virtio_pmem_flush(nd_region))
> > > > +                       rc = -EIO;
> > > > +       }
> > > > +
> > > > +       return rc;
> > > > +};
> > > > +EXPORT_SYMBOL_GPL(async_pmem_flush);
> > > > +MODULE_LICENSE("GPL");
> > > > diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> > > > index 35897649c24f..9f634a2ed638 100644
> > > > --- a/drivers/virtio/Kconfig
> > > > +++ b/drivers/virtio/Kconfig
> > > > @@ -42,6 +42,16 @@ config VIRTIO_PCI_LEGACY
> > > >
> > > >           If unsure, say Y.
> > > >
> > > > +config VIRTIO_PMEM
> > > > +       tristate "Support for virtio pmem driver"
> > > > +       depends on VIRTIO
> > > > +       depends on LIBNVDIMM
> > > > +       help
> > > > +       This driver provides support for virtio based flushing
> > > > interface
> > > > +       for persistent memory range.
> > > > +
> > > > +       If unsure, say M.
> > > > +
> > > >  config VIRTIO_BALLOON
> > > >         tristate "Virtio balloon driver"
> > > >         depends on VIRTIO
> > > > diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> > > > index 3a2b5c5dcf46..143ce91eabe9 100644
> > > > --- a/drivers/virtio/Makefile
> > > > +++ b/drivers/virtio/Makefile
> > > > @@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
> > > >  virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
> > > >  obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
> > > >  obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> > > > +obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
> > > > diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
> > > > new file mode 100644
> > > > index 000000000000..309788628e41
> > > > --- /dev/null
> > > > +++ b/drivers/virtio/pmem.c
> > >
> > > It's not clear to me why this driver is located in drivers/virtio/
> >
> > Like other VIRTIO drivers, I placed it initially in drivers/virtio
> > directory.
> >
> > >
> > > > @@ -0,0 +1,118 @@
> > > > +// SPDX-License-Identifier: GPL-2.0
> > > > +/*
> > > > + * virtio_pmem.c: Virtio pmem Driver
> > > > + *
> > > > + * Discovers persistent memory range information
> > > > + * from host and registers the virtual pmem device
> > > > + * with libnvdimm core.
> > > > + */
> > > > +#include <linux/virtio_pmem.h>
> > > > +#include <../../drivers/nvdimm/nd.h>
> > >
> > > ...especially because it seems to require nvdimm internals.
> > >
> > > However I don't see why that header is included.
> >
> > Removed.
> >
> > >
> > > In any event lets move this to drivers/nvdimm/virtio.c to live
> > > alongside the other generic bus provider drivers/nvdimm/e820.c.
> >
> > o.k. Makes sense.
> >
> > >
> > > > +
> > > > +static struct virtio_device_id id_table[] = {
> > > > +       { VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
> > > > +       { 0 },
> > > > +};
> > > > +
> > > > + /* Initialize virt queue */
> > > > +static int init_vq(struct virtio_pmem *vpmem)
> > > > +{
> > > > +       /* single vq */
> > > > +       vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
> > > > +                               host_ack, "flush_queue");
> > > > +       if (IS_ERR(vpmem->req_vq))
> > > > +               return PTR_ERR(vpmem->req_vq);
> > > > +
> > > > +       spin_lock_init(&vpmem->pmem_lock);
> > > > +       INIT_LIST_HEAD(&vpmem->req_list);
> > > > +
> > > > +       return 0;
> > > > +};
> > > > +
> > > > +static int virtio_pmem_probe(struct virtio_device *vdev)
> > > > +{
> > > > +       int err = 0;
> > > > +       struct resource res;
> > > > +       struct virtio_pmem *vpmem;
> > > > +       struct nd_region_desc ndr_desc = {};
> > > > +       int nid = dev_to_node(&vdev->dev);
> > > > +       struct nd_region *nd_region;
> > > > +
> > > > +       if (!vdev->config->get) {
> > > > +               dev_err(&vdev->dev, "%s failure: config access
> > > > disabled\n",
> > > > +                       __func__);
> > > > +               return -EINVAL;
> > > > +       }
> > > > +
> > > > +       vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
> > > > +       if (!vpmem) {
> > > > +               err = -ENOMEM;
> > > > +               goto out_err;
> > > > +       }
> > > > +
> > > > +       vpmem->vdev = vdev;
> > > > +       vdev->priv = vpmem;
> > > > +       err = init_vq(vpmem);
> > > > +       if (err)
> > > > +               goto out_err;
> > > > +
> > > > +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > > > +                       start, &vpmem->start);
> > > > +       virtio_cread(vpmem->vdev, struct virtio_pmem_config,
> > > > +                       size, &vpmem->size);
> > > > +
> > > > +       res.start = vpmem->start;
> > > > +       res.end   = vpmem->start + vpmem->size-1;
> > > > +       vpmem->nd_desc.provider_name = "virtio-pmem";
> > > > +       vpmem->nd_desc.module = THIS_MODULE;
> > > > +
> > > > +       vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
> > > > +                                               &vpmem->nd_desc);
> > > > +       if (!vpmem->nvdimm_bus)
> > > > +               goto out_vq;
> > > > +
> > > > +       dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
> > > > +
> > > > +       ndr_desc.res = &res;
> > > > +       ndr_desc.numa_node = nid;
> > > > +       ndr_desc.flush = async_pmem_flush;
> > > > +       set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
> > > > +       set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
> > > > +       nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus,
> > > > &ndr_desc);
> > > > +
> > > > +       if (!nd_region)
> > > > +               goto out_nd;
> > > > +       nd_region->provider_data =  dev_to_virtio
> > > > +
> > > > (nd_region->dev.parent->parent);
> > > > +       return 0;
> > > > +out_nd:
> > > > +       err = -ENXIO;
> > > > +       nvdimm_bus_unregister(vpmem->nvdimm_bus);
> > > > +out_vq:
> > > > +       vdev->config->del_vqs(vdev);
> > > > +out_err:
> > > > +       dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
> > > > +       return err;
> > > > +}
> > > > +
> > > > +static void virtio_pmem_remove(struct virtio_device *vdev)
> > > > +{
> > > > +       struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
> > > > +
> > > > +       nvdimm_bus_unregister(nvdimm_bus);
> > > > +       vdev->config->del_vqs(vdev);
> > > > +       vdev->config->reset(vdev);
> > > > +}
> > > > +
> > > > +static struct virtio_driver virtio_pmem_driver = {
> > > > +       .driver.name            = KBUILD_MODNAME,
> > > > +       .driver.owner           = THIS_MODULE,
> > > > +       .id_table               = id_table,
> > > > +       .probe                  = virtio_pmem_probe,
> > > > +       .remove                 = virtio_pmem_remove,
> > > > +};
> > > > +
> > > > +module_virtio_driver(virtio_pmem_driver);
> > > > +MODULE_DEVICE_TABLE(virtio, id_table);
> > > > +MODULE_DESCRIPTION("Virtio pmem driver");
> > > > +MODULE_LICENSE("GPL");
> > > > diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
> > > > new file mode 100644
> > > > index 000000000000..ab1da877575d
> > > > --- /dev/null
> > > > +++ b/include/linux/virtio_pmem.h
> > >
> > > Why is this a global header?
> >
> > This is where other virtio driver headers are also placed.
> > I think this is to access uapi config file in :
> >
> > ./include/uapi/linux/virtio_pmem.h
> >
> > Is it okay if we keep 'virtio_pmem.h' in global header?
> 
> No, I don't think so. While virtio_console.h and virtio_net.h make
> sense as global headers because they are consumed from multiple
> drivers, there is no need for virtio_caif.h, for example, to be a
> global header. I see no practical reason that the private details of
> virtio_pmem.h need to be made available outside of the virtio_pmem.c
> consumer.

o.k. Will move it.

Best regards,
Pankaj
> 
>

Patch
diff mbox series

diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
new file mode 100644
index 000000000000..66b582f751a3
--- /dev/null
+++ b/drivers/nvdimm/virtio_pmem.c
@@ -0,0 +1,114 @@ 
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * virtio_pmem.c: Virtio pmem Driver
+ *
+ * Discovers persistent memory range information
+ * from host and provides a virtio based flushing
+ * interface.
+ */
+#include <linux/virtio_pmem.h>
+#include "nd.h"
+
+ /* The interrupt handler */
+void host_ack(struct virtqueue *vq)
+{
+	unsigned int len;
+	unsigned long flags;
+	struct virtio_pmem_request *req, *req_buf;
+	struct virtio_pmem *vpmem = vq->vdev->priv;
+
+	spin_lock_irqsave(&vpmem->pmem_lock, flags);
+	while ((req = virtqueue_get_buf(vq, &len)) != NULL) {
+		req->done = true;
+		wake_up(&req->host_acked);
+
+		if (!list_empty(&vpmem->req_list)) {
+			req_buf = list_first_entry(&vpmem->req_list,
+					struct virtio_pmem_request, list);
+			list_del(&vpmem->req_list);
+			req_buf->wq_buf_avail = true;
+			wake_up(&req_buf->wq_buf);
+		}
+	}
+	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
+}
+EXPORT_SYMBOL_GPL(host_ack);
+
+ /* The request submission function */
+int virtio_pmem_flush(struct nd_region *nd_region)
+{
+	int err;
+	unsigned long flags;
+	struct scatterlist *sgs[2], sg, ret;
+	struct virtio_device *vdev = nd_region->provider_data;
+	struct virtio_pmem *vpmem = vdev->priv;
+	struct virtio_pmem_request *req;
+
+	might_sleep();
+	req = kmalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	req->done = req->wq_buf_avail = false;
+	strcpy(req->name, "FLUSH");
+	init_waitqueue_head(&req->host_acked);
+	init_waitqueue_head(&req->wq_buf);
+	sg_init_one(&sg, req->name, strlen(req->name));
+	sgs[0] = &sg;
+	sg_init_one(&ret, &req->ret, sizeof(req->ret));
+	sgs[1] = &ret;
+
+	spin_lock_irqsave(&vpmem->pmem_lock, flags);
+	err = virtqueue_add_sgs(vpmem->req_vq, sgs, 1, 1, req, GFP_ATOMIC);
+	if (err) {
+		dev_err(&vdev->dev, "failed to send command to virtio pmem device\n");
+
+		list_add_tail(&vpmem->req_list, &req->list);
+		spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
+
+		/* When host has read buffer, this completes via host_ack */
+		wait_event(req->wq_buf, req->wq_buf_avail);
+		spin_lock_irqsave(&vpmem->pmem_lock, flags);
+	}
+	err = virtqueue_kick(vpmem->req_vq);
+	spin_unlock_irqrestore(&vpmem->pmem_lock, flags);
+
+	if (!err) {
+		err = -EIO;
+		goto ret;
+	}
+	/* When host has read buffer, this completes via host_ack */
+	wait_event(req->host_acked, req->done);
+	err = req->ret;
+ret:
+	kfree(req);
+	return err;
+};
+
+ /* The asynchronous flush callback function */
+int async_pmem_flush(struct nd_region *nd_region, struct bio *bio)
+{
+	int rc = 0;
+
+	/* Create child bio for asynchronous flush and chain with
+	 * parent bio. Otherwise directly call nd_region flush.
+	 */
+	if (bio && bio->bi_iter.bi_sector != -1) {
+		struct bio *child = bio_alloc(GFP_ATOMIC, 0);
+
+		if (!child)
+			return -ENOMEM;
+		bio_copy_dev(child, bio);
+		child->bi_opf = REQ_PREFLUSH;
+		child->bi_iter.bi_sector = -1;
+		bio_chain(child, bio);
+		submit_bio(child);
+	} else {
+		if (virtio_pmem_flush(nd_region))
+			rc = -EIO;
+	}
+
+	return rc;
+};
+EXPORT_SYMBOL_GPL(async_pmem_flush);
+MODULE_LICENSE("GPL");
diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
index 35897649c24f..9f634a2ed638 100644
--- a/drivers/virtio/Kconfig
+++ b/drivers/virtio/Kconfig
@@ -42,6 +42,16 @@  config VIRTIO_PCI_LEGACY
 
 	  If unsure, say Y.
 
+config VIRTIO_PMEM
+	tristate "Support for virtio pmem driver"
+	depends on VIRTIO
+	depends on LIBNVDIMM
+	help
+	This driver provides support for virtio based flushing interface
+	for persistent memory range.
+
+	If unsure, say M.
+
 config VIRTIO_BALLOON
 	tristate "Virtio balloon driver"
 	depends on VIRTIO
diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
index 3a2b5c5dcf46..143ce91eabe9 100644
--- a/drivers/virtio/Makefile
+++ b/drivers/virtio/Makefile
@@ -6,3 +6,4 @@  virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
 virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
 obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
 obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
+obj-$(CONFIG_VIRTIO_PMEM) += pmem.o ../nvdimm/virtio_pmem.o
diff --git a/drivers/virtio/pmem.c b/drivers/virtio/pmem.c
new file mode 100644
index 000000000000..309788628e41
--- /dev/null
+++ b/drivers/virtio/pmem.c
@@ -0,0 +1,118 @@ 
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * virtio_pmem.c: Virtio pmem Driver
+ *
+ * Discovers persistent memory range information
+ * from host and registers the virtual pmem device
+ * with libnvdimm core.
+ */
+#include <linux/virtio_pmem.h>
+#include <../../drivers/nvdimm/nd.h>
+
+static struct virtio_device_id id_table[] = {
+	{ VIRTIO_ID_PMEM, VIRTIO_DEV_ANY_ID },
+	{ 0 },
+};
+
+ /* Initialize virt queue */
+static int init_vq(struct virtio_pmem *vpmem)
+{
+	/* single vq */
+	vpmem->req_vq = virtio_find_single_vq(vpmem->vdev,
+				host_ack, "flush_queue");
+	if (IS_ERR(vpmem->req_vq))
+		return PTR_ERR(vpmem->req_vq);
+
+	spin_lock_init(&vpmem->pmem_lock);
+	INIT_LIST_HEAD(&vpmem->req_list);
+
+	return 0;
+};
+
+static int virtio_pmem_probe(struct virtio_device *vdev)
+{
+	int err = 0;
+	struct resource res;
+	struct virtio_pmem *vpmem;
+	struct nd_region_desc ndr_desc = {};
+	int nid = dev_to_node(&vdev->dev);
+	struct nd_region *nd_region;
+
+	if (!vdev->config->get) {
+		dev_err(&vdev->dev, "%s failure: config access disabled\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	vpmem = devm_kzalloc(&vdev->dev, sizeof(*vpmem), GFP_KERNEL);
+	if (!vpmem) {
+		err = -ENOMEM;
+		goto out_err;
+	}
+
+	vpmem->vdev = vdev;
+	vdev->priv = vpmem;
+	err = init_vq(vpmem);
+	if (err)
+		goto out_err;
+
+	virtio_cread(vpmem->vdev, struct virtio_pmem_config,
+			start, &vpmem->start);
+	virtio_cread(vpmem->vdev, struct virtio_pmem_config,
+			size, &vpmem->size);
+
+	res.start = vpmem->start;
+	res.end   = vpmem->start + vpmem->size-1;
+	vpmem->nd_desc.provider_name = "virtio-pmem";
+	vpmem->nd_desc.module = THIS_MODULE;
+
+	vpmem->nvdimm_bus = nvdimm_bus_register(&vdev->dev,
+						&vpmem->nd_desc);
+	if (!vpmem->nvdimm_bus)
+		goto out_vq;
+
+	dev_set_drvdata(&vdev->dev, vpmem->nvdimm_bus);
+
+	ndr_desc.res = &res;
+	ndr_desc.numa_node = nid;
+	ndr_desc.flush = async_pmem_flush;
+	set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
+	set_bit(ND_REGION_ASYNC, &ndr_desc.flags);
+	nd_region = nvdimm_pmem_region_create(vpmem->nvdimm_bus, &ndr_desc);
+
+	if (!nd_region)
+		goto out_nd;
+	nd_region->provider_data =  dev_to_virtio
+					(nd_region->dev.parent->parent);
+	return 0;
+out_nd:
+	err = -ENXIO;
+	nvdimm_bus_unregister(vpmem->nvdimm_bus);
+out_vq:
+	vdev->config->del_vqs(vdev);
+out_err:
+	dev_err(&vdev->dev, "failed to register virtio pmem memory\n");
+	return err;
+}
+
+static void virtio_pmem_remove(struct virtio_device *vdev)
+{
+	struct nvdimm_bus *nvdimm_bus = dev_get_drvdata(&vdev->dev);
+
+	nvdimm_bus_unregister(nvdimm_bus);
+	vdev->config->del_vqs(vdev);
+	vdev->config->reset(vdev);
+}
+
+static struct virtio_driver virtio_pmem_driver = {
+	.driver.name		= KBUILD_MODNAME,
+	.driver.owner		= THIS_MODULE,
+	.id_table		= id_table,
+	.probe			= virtio_pmem_probe,
+	.remove			= virtio_pmem_remove,
+};
+
+module_virtio_driver(virtio_pmem_driver);
+MODULE_DEVICE_TABLE(virtio, id_table);
+MODULE_DESCRIPTION("Virtio pmem driver");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/virtio_pmem.h b/include/linux/virtio_pmem.h
new file mode 100644
index 000000000000..ab1da877575d
--- /dev/null
+++ b/include/linux/virtio_pmem.h
@@ -0,0 +1,60 @@ 
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * virtio_pmem.h: virtio pmem Driver
+ *
+ * Discovers persistent memory range information
+ * from host and provides a virtio based flushing
+ * interface.
+ **/
+
+#ifndef _LINUX_VIRTIO_PMEM_H
+#define _LINUX_VIRTIO_PMEM_H
+
+#include <linux/virtio_ids.h>
+#include <linux/module.h>
+#include <linux/virtio_config.h>
+#include <uapi/linux/virtio_pmem.h>
+#include <linux/libnvdimm.h>
+#include <linux/spinlock.h>
+
+struct virtio_pmem_request {
+	/* Host return status corresponding to flush request */
+	int ret;
+
+	/* command name*/
+	char name[16];
+
+	/* Wait queue to process deferred work after ack from host */
+	wait_queue_head_t host_acked;
+	bool done;
+
+	/* Wait queue to process deferred work after virt queue buffer avail */
+	wait_queue_head_t wq_buf;
+	bool wq_buf_avail;
+	struct list_head list;
+};
+
+struct virtio_pmem {
+	struct virtio_device *vdev;
+
+	/* Virtio pmem request queue */
+	struct virtqueue *req_vq;
+
+	/* nvdimm bus registers virtio pmem device */
+	struct nvdimm_bus *nvdimm_bus;
+	struct nvdimm_bus_descriptor nd_desc;
+
+	/* List to store deferred work if virtqueue is full */
+	struct list_head req_list;
+
+	/* Synchronize virtqueue data */
+	spinlock_t pmem_lock;
+
+	/* Memory region information */
+	uint64_t start;
+	uint64_t size;
+};
+
+void host_ack(struct virtqueue *vq);
+int async_pmem_flush(struct nd_region *nd_region, struct bio *bio);
+#endif
diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h
index 6d5c3b2d4f4d..32b2f94d1f58 100644
--- a/include/uapi/linux/virtio_ids.h
+++ b/include/uapi/linux/virtio_ids.h
@@ -43,5 +43,6 @@ 
 #define VIRTIO_ID_INPUT        18 /* virtio input */
 #define VIRTIO_ID_VSOCK        19 /* virtio vsock transport */
 #define VIRTIO_ID_CRYPTO       20 /* virtio crypto */
+#define VIRTIO_ID_PMEM         27 /* virtio pmem */
 
 #endif /* _LINUX_VIRTIO_IDS_H */
diff --git a/include/uapi/linux/virtio_pmem.h b/include/uapi/linux/virtio_pmem.h
new file mode 100644
index 000000000000..fa3f7d52717a
--- /dev/null
+++ b/include/uapi/linux/virtio_pmem.h
@@ -0,0 +1,10 @@ 
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _UAPI_LINUX_VIRTIO_PMEM_H
+#define _UAPI_LINUX_VIRTIO_PMEM_H
+
+struct virtio_pmem_config {
+	__le64 start;
+	__le64 size;
+};
+#endif