diff mbox series

[RFC,v1,05/12] vsock/virtio: non-linear skb support

Message ID b3060caf-df19-f1df-6d27-4e58f894c417@sberdevices.ru
State New
Headers
Series vsock: MSG_ZEROCOPY flag support |

Commit Message

Arseniy Krasnov Feb. 6, 2023, 6:58 a.m. UTC 
  Use pages of non-linear skb as buffers in virtio tx queue.

Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
---
 net/vmw_vsock/virtio_transport.c | 31 +++++++++++++++++++++++++------
 1 file changed, 25 insertions(+), 6 deletions(-)

-- 
2.25.1

Comments

Stefano Garzarella Feb. 16, 2023, 2:18 p.m. UTC | #1 
On Mon, Feb 06, 2023 at 06:58:24AM +0000, Arseniy Krasnov wrote:
>Use pages of non-linear skb as buffers in virtio tx queue.
>
>Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
>---
> net/vmw_vsock/virtio_transport.c | 31 +++++++++++++++++++++++++------
> 1 file changed, 25 insertions(+), 6 deletions(-)
>
>diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
>index 28b5a8e8e094..b8a7d6dc9f46 100644
>--- a/net/vmw_vsock/virtio_transport.c
>+++ b/net/vmw_vsock/virtio_transport.c
>@@ -100,7 +100,8 @@ virtio_transport_send_pkt_work(struct work_struct *work)
> 	vq = vsock->vqs[VSOCK_VQ_TX];
>
> 	for (;;) {
>-		struct scatterlist hdr, buf, *sgs[2];
>+		struct scatterlist *sgs[MAX_SKB_FRAGS + 1];
>+		struct scatterlist bufs[MAX_SKB_FRAGS + 1];

+ 1 is for the header, right?
I'd add a comment just to be clear ;-)

> 		int ret, in_sg = 0, out_sg = 0;
> 		struct sk_buff *skb;
> 		bool reply;
>@@ -111,12 +112,30 @@ virtio_transport_send_pkt_work(struct work_struct *work)
>
> 		virtio_transport_deliver_tap_pkt(skb);
> 		reply = virtio_vsock_skb_reply(skb);
>+		sg_init_one(&bufs[0], virtio_vsock_hdr(skb), sizeof(*virtio_vsock_hdr(skb)));
>+		sgs[out_sg++] = &bufs[0];
>+
>+		if (skb_is_nonlinear(skb)) {
>+			int i;
>+
>+			for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
>+				struct page *data_page = skb_shinfo(skb)->frags[i].bv_page;
>+
>+				/* We will use 'page_to_virt()' for userspace page here,
>+				 * because virtio layer will call 'virt_to_phys()' later
>+				 * to fill buffer descriptor. We don't touch memory at
>+				 * "virtual" address of this page.
>+				 */

IIUC data_page is a user page, so since we are exposing it to the host,
I think we should pin it.

Is data_page always a user page, or can it be a kernel page when skb is 
nonlinear?

Thanks,
Stefano

>+				sg_init_one(&bufs[i + 1],
>+					    page_to_virt(data_page), PAGE_SIZE);
>+				sgs[out_sg++] = &bufs[i + 1];
>+			}
>+		} else {
>+			if (skb->len > 0) {
>+				sg_init_one(&bufs[1], skb->data, skb->len);
>+				sgs[out_sg++] = &bufs[1];
>+			}
>
>-		sg_init_one(&hdr, virtio_vsock_hdr(skb), sizeof(*virtio_vsock_hdr(skb)));
>-		sgs[out_sg++] = &hdr;
>-		if (skb->len > 0) {
>-			sg_init_one(&buf, skb->data, skb->len);
>-			sgs[out_sg++] = &buf;
> 		}
>
> 		ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, skb, GFP_KERNEL);
>-- 
>2.25.1
Arseniy Krasnov Feb. 20, 2023, 9:02 a.m. UTC | #2 
On 16.02.2023 17:18, Stefano Garzarella wrote:
> On Mon, Feb 06, 2023 at 06:58:24AM +0000, Arseniy Krasnov wrote:
>> Use pages of non-linear skb as buffers in virtio tx queue.
>>
>> Signed-off-by: Arseniy Krasnov <AVKrasnov@sberdevices.ru>
>> ---
>> net/vmw_vsock/virtio_transport.c | 31 +++++++++++++++++++++++++------
>> 1 file changed, 25 insertions(+), 6 deletions(-)
>>
>> diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
>> index 28b5a8e8e094..b8a7d6dc9f46 100644
>> --- a/net/vmw_vsock/virtio_transport.c
>> +++ b/net/vmw_vsock/virtio_transport.c
>> @@ -100,7 +100,8 @@ virtio_transport_send_pkt_work(struct work_struct *work)
>>     vq = vsock->vqs[VSOCK_VQ_TX];
>>
>>     for (;;) {
>> -        struct scatterlist hdr, buf, *sgs[2];
>> +        struct scatterlist *sgs[MAX_SKB_FRAGS + 1];
>> +        struct scatterlist bufs[MAX_SKB_FRAGS + 1];
> 
> + 1 is for the header, right?
> I'd add a comment just to be clear ;-)
> 
>>         int ret, in_sg = 0, out_sg = 0;
>>         struct sk_buff *skb;
>>         bool reply;
>> @@ -111,12 +112,30 @@ virtio_transport_send_pkt_work(struct work_struct *work)
>>
>>         virtio_transport_deliver_tap_pkt(skb);
>>         reply = virtio_vsock_skb_reply(skb);
>> +        sg_init_one(&bufs[0], virtio_vsock_hdr(skb), sizeof(*virtio_vsock_hdr(skb)));
>> +        sgs[out_sg++] = &bufs[0];
>> +
>> +        if (skb_is_nonlinear(skb)) {
>> +            int i;
>> +
>> +            for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
>> +                struct page *data_page = skb_shinfo(skb)->frags[i].bv_page;
>> +
>> +                /* We will use 'page_to_virt()' for userspace page here,
>> +                 * because virtio layer will call 'virt_to_phys()' later
>> +                 * to fill buffer descriptor. We don't touch memory at
>> +                 * "virtual" address of this page.
>> +                 */
> 
> IIUC data_page is a user page, so since we are exposing it to the host,
> I think we should pin it.
> 
> Is data_page always a user page, or can it be a kernel page when skb is nonlinear?

By default it is user page, but...may be it is possible to send page of mapped
file with MAP_SHARED flags(i think it this case it will be page from page cache)

> 
> Thanks,
> Stefano
> 
>> +                sg_init_one(&bufs[i + 1],
>> +                        page_to_virt(data_page), PAGE_SIZE);
>> +                sgs[out_sg++] = &bufs[i + 1];
>> +            }
>> +        } else {
>> +            if (skb->len > 0) {
>> +                sg_init_one(&bufs[1], skb->data, skb->len);
>> +                sgs[out_sg++] = &bufs[1];
>> +            }
>>
>> -        sg_init_one(&hdr, virtio_vsock_hdr(skb), sizeof(*virtio_vsock_hdr(skb)));
>> -        sgs[out_sg++] = &hdr;
>> -        if (skb->len > 0) {
>> -            sg_init_one(&buf, skb->data, skb->len);
>> -            sgs[out_sg++] = &buf;
>>         }
>>
>>         ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, skb, GFP_KERNEL);
>> -- 
>> 2.25.1
>
diff mbox series

Patch

diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index 28b5a8e8e094..b8a7d6dc9f46 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -100,7 +100,8 @@  virtio_transport_send_pkt_work(struct work_struct *work)
 	vq = vsock->vqs[VSOCK_VQ_TX];
 
 	for (;;) {
-		struct scatterlist hdr, buf, *sgs[2];
+		struct scatterlist *sgs[MAX_SKB_FRAGS + 1];
+		struct scatterlist bufs[MAX_SKB_FRAGS + 1];
 		int ret, in_sg = 0, out_sg = 0;
 		struct sk_buff *skb;
 		bool reply;
@@ -111,12 +112,30 @@  virtio_transport_send_pkt_work(struct work_struct *work)
 
 		virtio_transport_deliver_tap_pkt(skb);
 		reply = virtio_vsock_skb_reply(skb);
+		sg_init_one(&bufs[0], virtio_vsock_hdr(skb), sizeof(*virtio_vsock_hdr(skb)));
+		sgs[out_sg++] = &bufs[0];
+
+		if (skb_is_nonlinear(skb)) {
+			int i;
+
+			for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+				struct page *data_page = skb_shinfo(skb)->frags[i].bv_page;
+
+				/* We will use 'page_to_virt()' for userspace page here,
+				 * because virtio layer will call 'virt_to_phys()' later
+				 * to fill buffer descriptor. We don't touch memory at
+				 * "virtual" address of this page.
+				 */
+				sg_init_one(&bufs[i + 1],
+					    page_to_virt(data_page), PAGE_SIZE);
+				sgs[out_sg++] = &bufs[i + 1];
+			}
+		} else {
+			if (skb->len > 0) {
+				sg_init_one(&bufs[1], skb->data, skb->len);
+				sgs[out_sg++] = &bufs[1];
+			}
 
-		sg_init_one(&hdr, virtio_vsock_hdr(skb), sizeof(*virtio_vsock_hdr(skb)));
-		sgs[out_sg++] = &hdr;
-		if (skb->len > 0) {
-			sg_init_one(&buf, skb->data, skb->len);
-			sgs[out_sg++] = &buf;
 		}
 
 		ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, skb, GFP_KERNEL);