Netdev List
 help / color / mirror / Atom feed
* [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling
@ 2026-06-28 18:20 Arseniy Krasnov
  2026-06-28 18:35 ` Michael S. Tsirkin
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: Arseniy Krasnov @ 2026-06-28 18:20 UTC (permalink / raw)
  To: Stefan Hajnoczi, Stefano Garzarella, David S. Miller,
	Eric Dumazet, Jakub Kicinski, Paolo Abeni, Michael S. Tsirkin,
	Jason Wang, Bobby Eshleman, Xuan Zhuo, Eugenio Pérez,
	Simon Horman
  Cc: kvm, virtualization, netdev, linux-kernel, oxffffaa, rulkc,
	Arseniy Krasnov

Logically it was based on TCP implementation, so to make further support
easier, rewrite it in the TCP way (like in 'tcp_sendmsg_locked()'). By
this way, patch also adds handling case when 'msg_ubuf' is already set.

Signed-off-by: Arseniy Krasnov <avkrasnov@rulkc.org>
---
 Changelog v1->v2:
 * Rebase on last 'net-next'. Don't need 'skb_zcopy_set()' now - it was
   already added.
 Changelog v2->v3:
 * Update commit message.
 * Remove one empty line.
 Changelog v3->v4:
 * Update commit message.

 net/vmw_vsock/virtio_transport_common.c | 47 ++++++++++++-------------
 1 file changed, 22 insertions(+), 25 deletions(-)

diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index 09475007165b..41c2a0b82a8e 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -328,38 +328,35 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
 	if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
 		return pkt_len;
 
-	if (info->msg) {
-		/* If zerocopy is not enabled by 'setsockopt()', we behave as
-		 * there is no MSG_ZEROCOPY flag set.
+	if (info->msg && (info->msg->msg_flags & MSG_ZEROCOPY)) {
+		/* If 'info->msg' is not NULL, this is only VIRTIO_VSOCK_OP_RW.
+		 * 'MSG_ZEROCOPY' flag handling here is based on the same flag
+		 * handling from 'tcp_sendmsg_locked()'.
 		 */
-		if (!sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY))
-			info->msg->msg_flags &= ~MSG_ZEROCOPY;
+		if (info->msg->msg_ubuf) {
+			uarg = info->msg->msg_ubuf;
+			can_zcopy = virtio_transport_can_zcopy(t_ops, info, pkt_len);
+		} else if (sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY)) {
+			uarg = msg_zerocopy_realloc(sk_vsock(vsk), pkt_len,
+						    NULL, false);
+			if (!uarg) {
+				virtio_transport_put_credit(vvs, pkt_len);
+				return -ENOMEM;
+			}
 
-		if (info->msg->msg_flags & MSG_ZEROCOPY)
 			can_zcopy = virtio_transport_can_zcopy(t_ops, info, pkt_len);
+			if (!can_zcopy)
+				uarg_to_msgzc(uarg)->zerocopy = 0;
 
+			have_uref = true;
+		}
+
+		/* 'can_zcopy' means that this transmission will be
+		 * in zerocopy way (e.g. using 'frags' array).
+		 */
 		if (can_zcopy)
 			max_skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE,
 					    (MAX_SKB_FRAGS * PAGE_SIZE));
-
-		if (info->msg->msg_flags & MSG_ZEROCOPY &&
-		    info->op == VIRTIO_VSOCK_OP_RW) {
-			uarg = info->msg->msg_ubuf;
-
-			if (!uarg) {
-				uarg = msg_zerocopy_realloc(sk_vsock(vsk),
-							    pkt_len, NULL, false);
-				if (!uarg) {
-					virtio_transport_put_credit(vvs, pkt_len);
-					return -ENOMEM;
-				}
-
-				if (!can_zcopy)
-					uarg_to_msgzc(uarg)->zerocopy = 0;
-
-				have_uref = true;
-			}
-		}
 	}
 
 	rest_len = pkt_len;
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling
  2026-06-28 18:20 [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling Arseniy Krasnov
@ 2026-06-28 18:35 ` Michael S. Tsirkin
  2026-06-30  9:41 ` Stefano Garzarella
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: Michael S. Tsirkin @ 2026-06-28 18:35 UTC (permalink / raw)
  To: Arseniy Krasnov
  Cc: Stefan Hajnoczi, Stefano Garzarella, David S. Miller,
	Eric Dumazet, Jakub Kicinski, Paolo Abeni, Jason Wang,
	Bobby Eshleman, Xuan Zhuo, Eugenio Pérez, Simon Horman, kvm,
	virtualization, netdev, linux-kernel, oxffffaa, rulkc

On Sun, Jun 28, 2026 at 09:20:52PM +0300, Arseniy Krasnov wrote:
> Logically it was based on TCP implementation, so to make further support
> easier, rewrite it in the TCP way (like in 'tcp_sendmsg_locked()'). By
> this way, patch also adds handling case when 'msg_ubuf' is already set.
> 
> Signed-off-by: Arseniy Krasnov <avkrasnov@rulkc.org>

Acked-by: Michael S. Tsirkin <mst@redhat.com>

> ---
>  Changelog v1->v2:
>  * Rebase on last 'net-next'. Don't need 'skb_zcopy_set()' now - it was
>    already added.
>  Changelog v2->v3:
>  * Update commit message.
>  * Remove one empty line.
>  Changelog v3->v4:
>  * Update commit message.
> 
>  net/vmw_vsock/virtio_transport_common.c | 47 ++++++++++++-------------
>  1 file changed, 22 insertions(+), 25 deletions(-)
> 
> diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
> index 09475007165b..41c2a0b82a8e 100644
> --- a/net/vmw_vsock/virtio_transport_common.c
> +++ b/net/vmw_vsock/virtio_transport_common.c
> @@ -328,38 +328,35 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
>  	if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
>  		return pkt_len;
>  
> -	if (info->msg) {
> -		/* If zerocopy is not enabled by 'setsockopt()', we behave as
> -		 * there is no MSG_ZEROCOPY flag set.
> +	if (info->msg && (info->msg->msg_flags & MSG_ZEROCOPY)) {
> +		/* If 'info->msg' is not NULL, this is only VIRTIO_VSOCK_OP_RW.
> +		 * 'MSG_ZEROCOPY' flag handling here is based on the same flag
> +		 * handling from 'tcp_sendmsg_locked()'.
>  		 */
> -		if (!sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY))
> -			info->msg->msg_flags &= ~MSG_ZEROCOPY;
> +		if (info->msg->msg_ubuf) {
> +			uarg = info->msg->msg_ubuf;
> +			can_zcopy = virtio_transport_can_zcopy(t_ops, info, pkt_len);
> +		} else if (sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY)) {
> +			uarg = msg_zerocopy_realloc(sk_vsock(vsk), pkt_len,
> +						    NULL, false);
> +			if (!uarg) {
> +				virtio_transport_put_credit(vvs, pkt_len);
> +				return -ENOMEM;
> +			}
>  
> -		if (info->msg->msg_flags & MSG_ZEROCOPY)
>  			can_zcopy = virtio_transport_can_zcopy(t_ops, info, pkt_len);
> +			if (!can_zcopy)
> +				uarg_to_msgzc(uarg)->zerocopy = 0;
>  
> +			have_uref = true;
> +		}
> +
> +		/* 'can_zcopy' means that this transmission will be
> +		 * in zerocopy way (e.g. using 'frags' array).
> +		 */
>  		if (can_zcopy)
>  			max_skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE,
>  					    (MAX_SKB_FRAGS * PAGE_SIZE));
> -
> -		if (info->msg->msg_flags & MSG_ZEROCOPY &&
> -		    info->op == VIRTIO_VSOCK_OP_RW) {
> -			uarg = info->msg->msg_ubuf;
> -
> -			if (!uarg) {
> -				uarg = msg_zerocopy_realloc(sk_vsock(vsk),
> -							    pkt_len, NULL, false);
> -				if (!uarg) {
> -					virtio_transport_put_credit(vvs, pkt_len);
> -					return -ENOMEM;
> -				}
> -
> -				if (!can_zcopy)
> -					uarg_to_msgzc(uarg)->zerocopy = 0;
> -
> -				have_uref = true;
> -			}
> -		}
>  	}
>  
>  	rest_len = pkt_len;
> -- 
> 2.25.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling
  2026-06-28 18:20 [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling Arseniy Krasnov
  2026-06-28 18:35 ` Michael S. Tsirkin
@ 2026-06-30  9:41 ` Stefano Garzarella
  2026-06-30 15:52 ` Paolo Abeni
  2026-06-30 20:32 ` patchwork-bot+netdevbpf
  3 siblings, 0 replies; 5+ messages in thread
From: Stefano Garzarella @ 2026-06-30  9:41 UTC (permalink / raw)
  To: Arseniy Krasnov
  Cc: Stefan Hajnoczi, David S. Miller, Eric Dumazet, Jakub Kicinski,
	Paolo Abeni, Michael S. Tsirkin, Jason Wang, Bobby Eshleman,
	Xuan Zhuo, Eugenio Pérez, Simon Horman, kvm, virtualization,
	netdev, linux-kernel, oxffffaa, rulkc

On Sun, Jun 28, 2026 at 09:20:52PM +0300, Arseniy Krasnov wrote:
>Logically it was based on TCP implementation, so to make further support
>easier, rewrite it in the TCP way (like in 'tcp_sendmsg_locked()'). By
>this way, patch also adds handling case when 'msg_ubuf' is already set.

Thanks for this!

IIUC the result will be similar of commit eb315a7d1396 ("tcp: support 
externally provided ubufs") for tcp. Maybe I would have added it to the 
commit description, anyway the patch LGTM:

Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>

>
>Signed-off-by: Arseniy Krasnov <avkrasnov@rulkc.org>
>---
> Changelog v1->v2:
> * Rebase on last 'net-next'. Don't need 'skb_zcopy_set()' now - it was
>   already added.
> Changelog v2->v3:
> * Update commit message.
> * Remove one empty line.
> Changelog v3->v4:
> * Update commit message.
>
> net/vmw_vsock/virtio_transport_common.c | 47 ++++++++++++-------------
> 1 file changed, 22 insertions(+), 25 deletions(-)
>
>diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
>index 09475007165b..41c2a0b82a8e 100644
>--- a/net/vmw_vsock/virtio_transport_common.c
>+++ b/net/vmw_vsock/virtio_transport_common.c
>@@ -328,38 +328,35 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
> 	if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
> 		return pkt_len;
>
>-	if (info->msg) {
>-		/* If zerocopy is not enabled by 'setsockopt()', we behave as
>-		 * there is no MSG_ZEROCOPY flag set.
>+	if (info->msg && (info->msg->msg_flags & MSG_ZEROCOPY)) {
>+		/* If 'info->msg' is not NULL, this is only VIRTIO_VSOCK_OP_RW.
>+		 * 'MSG_ZEROCOPY' flag handling here is based on the same flag
>+		 * handling from 'tcp_sendmsg_locked()'.
> 		 */
>-		if (!sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY))
>-			info->msg->msg_flags &= ~MSG_ZEROCOPY;
>+		if (info->msg->msg_ubuf) {
>+			uarg = info->msg->msg_ubuf;
>+			can_zcopy = virtio_transport_can_zcopy(t_ops, info, pkt_len);
>+		} else if (sock_flag(sk_vsock(vsk), SOCK_ZEROCOPY)) {
>+			uarg = msg_zerocopy_realloc(sk_vsock(vsk), pkt_len,
>+						    NULL, false);
>+			if (!uarg) {
>+				virtio_transport_put_credit(vvs, pkt_len);
>+				return -ENOMEM;
>+			}
>
>-		if (info->msg->msg_flags & MSG_ZEROCOPY)
> 			can_zcopy = virtio_transport_can_zcopy(t_ops, info, pkt_len);
>+			if (!can_zcopy)
>+				uarg_to_msgzc(uarg)->zerocopy = 0;
>
>+			have_uref = true;
>+		}
>+
>+		/* 'can_zcopy' means that this transmission will be
>+		 * in zerocopy way (e.g. using 'frags' array).
>+		 */
> 		if (can_zcopy)
> 			max_skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE,
> 					    (MAX_SKB_FRAGS * PAGE_SIZE));
>-
>-		if (info->msg->msg_flags & MSG_ZEROCOPY &&
>-		    info->op == VIRTIO_VSOCK_OP_RW) {
>-			uarg = info->msg->msg_ubuf;
>-
>-			if (!uarg) {
>-				uarg = msg_zerocopy_realloc(sk_vsock(vsk),
>-							    pkt_len, NULL, false);
>-				if (!uarg) {
>-					virtio_transport_put_credit(vvs, pkt_len);
>-					return -ENOMEM;
>-				}
>-
>-				if (!can_zcopy)
>-					uarg_to_msgzc(uarg)->zerocopy = 0;
>-
>-				have_uref = true;
>-			}
>-		}
> 	}
>
> 	rest_len = pkt_len;
>-- 
>2.25.1
>


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling
  2026-06-28 18:20 [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling Arseniy Krasnov
  2026-06-28 18:35 ` Michael S. Tsirkin
  2026-06-30  9:41 ` Stefano Garzarella
@ 2026-06-30 15:52 ` Paolo Abeni
  2026-06-30 20:32 ` patchwork-bot+netdevbpf
  3 siblings, 0 replies; 5+ messages in thread
From: Paolo Abeni @ 2026-06-30 15:52 UTC (permalink / raw)
  To: Arseniy Krasnov, Stefan Hajnoczi, Stefano Garzarella,
	David S. Miller, Eric Dumazet, Jakub Kicinski, Michael S. Tsirkin,
	Jason Wang, Bobby Eshleman, Xuan Zhuo, Eugenio Pérez,
	Simon Horman
  Cc: kvm, virtualization, netdev, linux-kernel, oxffffaa, rulkc

On 6/28/26 8:20 PM, Arseniy Krasnov wrote:
> Logically it was based on TCP implementation, so to make further support
> easier, rewrite it in the TCP way (like in 'tcp_sendmsg_locked()'). By
> this way, patch also adds handling case when 'msg_ubuf' is already set.
> 
> Signed-off-by: Arseniy Krasnov <avkrasnov@rulkc.org>

The PW bot is on holiday, no automated notifications for a while.

Applied, thanks!

/P


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling
  2026-06-28 18:20 [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling Arseniy Krasnov
                   ` (2 preceding siblings ...)
  2026-06-30 15:52 ` Paolo Abeni
@ 2026-06-30 20:32 ` patchwork-bot+netdevbpf
  3 siblings, 0 replies; 5+ messages in thread
From: patchwork-bot+netdevbpf @ 2026-06-30 20:32 UTC (permalink / raw)
  To: Arseniy Krasnov
  Cc: stefanha, sgarzare, davem, edumazet, kuba, pabeni, mst, jasowang,
	bobbyeshleman, xuanzhuo, eperezma, horms, kvm, virtualization,
	netdev, linux-kernel, oxffffaa, rulkc

Hello:

This patch was applied to netdev/net-next.git (main)
by Paolo Abeni <pabeni@redhat.com>:

On Sun, 28 Jun 2026 21:20:52 +0300 you wrote:
> Logically it was based on TCP implementation, so to make further support
> easier, rewrite it in the TCP way (like in 'tcp_sendmsg_locked()'). By
> this way, patch also adds handling case when 'msg_ubuf' is already set.
> 
> Signed-off-by: Arseniy Krasnov <avkrasnov@rulkc.org>
> ---
>  Changelog v1->v2:
>  * Rebase on last 'net-next'. Don't need 'skb_zcopy_set()' now - it was
>    already added.
>  Changelog v2->v3:
>  * Update commit message.
>  * Remove one empty line.
>  Changelog v3->v4:
>  * Update commit message.
> 
> [...]

Here is the summary with links:
  - [net-next,v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling
    https://git.kernel.org/netdev/net-next/c/f456c1922c49

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2026-06-30 20:32 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-06-28 18:20 [PATCH net-next v4] vsock/virtio: rewrite MSG_ZEROCOPY flag handling Arseniy Krasnov
2026-06-28 18:35 ` Michael S. Tsirkin
2026-06-30  9:41 ` Stefano Garzarella
2026-06-30 15:52 ` Paolo Abeni
2026-06-30 20:32 ` patchwork-bot+netdevbpf

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox