[PATCH v3 15/16] virtio_pci_modern: reduce number of mappings

["Michael S. Tsirkin" <mst@redhat.com>]

We don't know the # of VQs that drivers are going to use so it's hard to
predict how much memory we'll need to map. However, the relevant
capability does give us an upper limit.
If that's below a page, we can reduce the number of required
mappings by mapping it all once ahead of the time.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 drivers/virtio/virtio_pci_common.h |  3 ++
 drivers/virtio/virtio_pci_modern.c | 57 ++++++++++++++++++++++++++++++++------
 2 files changed, 52 insertions(+), 8 deletions(-)

diff --git a/drivers/virtio/virtio_pci_common.h b/drivers/virtio/virtio_pci_common.h
index 610c43f..d391805 100644
--- a/drivers/virtio/virtio_pci_common.h
+++ b/drivers/virtio/virtio_pci_common.h
@@ -62,8 +62,11 @@ struct virtio_pci_device {
 	struct virtio_pci_common_cfg __iomem *common;
 	/* Device-specific data (non-legacy mode)  */
 	void __iomem *device;
+	/* Base of vq notifications (non-legacy mode). */
+	void __iomem *notify_base;
 
 	/* So we can sanity-check accesses. */
+	size_t notify_len;
 	size_t device_len;
 
 	/* Capability for when we need to map notifications per-vq. */
diff --git a/drivers/virtio/virtio_pci_modern.c b/drivers/virtio/virtio_pci_modern.c
index 5e0d309..9c000e9a 100644
--- a/drivers/virtio/virtio_pci_modern.c
+++ b/drivers/virtio/virtio_pci_modern.c
@@ -333,10 +333,26 @@ static struct virtqueue *setup_vq(struct virtio_pci_device *vp_dev,
 	iowrite64_twopart(virt_to_phys(virtqueue_get_used(vq)),
 			  &cfg->queue_used_lo, &cfg->queue_used_hi);
 
-	vq->priv = (void __force *)map_capability(vp_dev->pci_dev,
-				  vp_dev->notify_map_cap, 2, 2,
-				  off * vp_dev->notify_offset_multiplier, 2,
-				  NULL);
+	if (vp_dev->notify_base) {
+		/* offset should not wrap */
+		if ((u64)off * vp_dev->notify_offset_multiplier + 2
+		    > vp_dev->notify_len) {
+			dev_warn(&vp_dev->pci_dev->dev,
+				 "bad notification offset %u (x %u) "
+				 "for queue %u > %zd",
+				 off, vp_dev->notify_offset_multiplier,
+				 index, vp_dev->notify_len);
+			err = -EINVAL;
+			goto err_map_notify;
+		}
+		vq->priv = (void __force *)vp_dev->notify_base +
+			off * vp_dev->notify_offset_multiplier;
+	} else {
+		vq->priv = (void __force *)map_capability(vp_dev->pci_dev,
+					  vp_dev->notify_map_cap, 2, 2,
+					  off * vp_dev->notify_offset_multiplier, 2,
+					  NULL);
+	}
 
 	if (!vq->priv) {
 		err = -ENOMEM;
@@ -355,7 +371,8 @@ static struct virtqueue *setup_vq(struct virtio_pci_device *vp_dev,
 	return vq;
 
 err_assign_vector:
-	pci_iounmap(vp_dev->pci_dev, (void __iomem __force *)vq->priv);
+	if (!vp_dev->notify_base)
+		pci_iounmap(vp_dev->pci_dev, (void __iomem __force *)vq->priv);
 err_map_notify:
 	vring_del_virtqueue(vq);
 err_new_queue:
@@ -400,7 +417,8 @@ static void del_vq(struct virtio_pci_vq_info *info)
 		ioread16(&vp_dev->common->queue_msix_vector);
 	}
 
-	pci_iounmap(vp_dev->pci_dev, (void __force __iomem *)vq->priv);
+	if (!vp_dev->notify_base)
+		pci_iounmap(vp_dev->pci_dev, (void __force __iomem *)vq->priv);
 
 	vring_del_virtqueue(vq);
 
@@ -536,6 +554,7 @@ int virtio_pci_modern_probe(struct virtio_pci_device *vp_dev)
 	struct pci_dev *pci_dev = vp_dev->pci_dev;
 	int err, common, isr, notify, device;
 	u32 notify_length;
+	u32 notify_offset;
 
 	check_offsets();
 
@@ -602,13 +621,30 @@ int virtio_pci_modern_probe(struct virtio_pci_device *vp_dev)
 			      notify + offsetof(struct virtio_pci_notify_cap,
 						notify_off_multiplier),
 			      &vp_dev->notify_offset_multiplier);
-	/* Read notify length from config space. */
+	/* Read notify length and offset from config space. */
 	pci_read_config_dword(pci_dev,
 			      notify + offsetof(struct virtio_pci_notify_cap,
 						cap.length),
 			      &notify_length);
 
-	vp_dev->notify_map_cap = notify;
+	pci_read_config_dword(pci_dev,
+			      notify + offsetof(struct virtio_pci_notify_cap,
+						cap.length),
+			      &notify_offset);
+
+	/* We don't know how many VQs we'll map, ahead of the time.
+	 * If notify length is small, map it all now.
+	 * Otherwise, map each VQ individually later.
+	 */
+	if ((u64)notify_length + (notify_offset % PAGE_SIZE) <= PAGE_SIZE) {
+		vp_dev->notify_base = map_capability(pci_dev, notify, 2, 2,
+						     0, notify_length,
+						     &vp_dev->notify_len);
+		if (!vp_dev->notify_base)
+			goto err_map_notify;
+	} else {
+		vp_dev->notify_map_cap = notify;
+	}
 
 	/* Again, we don't know how much we should map, but PAGE_SIZE
 	 * is more than enough for all existing devices.
@@ -630,6 +666,9 @@ int virtio_pci_modern_probe(struct virtio_pci_device *vp_dev)
 	return 0;
 
 err_map_device:
+	if (vp_dev->notify_base)
+		pci_iounmap(pci_dev, vp_dev->notify_base);
+err_map_notify:
 	pci_iounmap(pci_dev, vp_dev->isr);
 err_map_isr:
 	pci_iounmap(pci_dev, vp_dev->common);
@@ -643,6 +682,8 @@ void virtio_pci_modern_remove(struct virtio_pci_device *vp_dev)
 
 	if (vp_dev->device)
 		pci_iounmap(pci_dev, vp_dev->device);
+	if (vp_dev->notify_base)
+		pci_iounmap(pci_dev, vp_dev->notify_base);
 	pci_iounmap(pci_dev, vp_dev->isr);
 	pci_iounmap(pci_dev, vp_dev->common);
 }
-- 
MST