linux-kernel - [PATCH net-next v9 05/14] vsock: add netns and netns

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20251111-vsock-vmtest-v9-5-852787a37bed@meta.com>
Date: Tue, 11 Nov 2025 22:54:47 -0800
From: Bobby Eshleman <bobbyeshleman@...il.com>
To: Stefano Garzarella <sgarzare@...hat.com>, Shuah Khan <shuah@...nel.org>, 
 "David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
 Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, 
 Simon Horman <horms@...nel.org>, Stefan Hajnoczi <stefanha@...hat.com>, 
 "Michael S. Tsirkin" <mst@...hat.com>, Jason Wang <jasowang@...hat.com>, 
 Xuan Zhuo <xuanzhuo@...ux.alibaba.com>, 
 Eugenio Pérez <eperezma@...hat.com>, 
 "K. Y. Srinivasan" <kys@...rosoft.com>, 
 Haiyang Zhang <haiyangz@...rosoft.com>, Wei Liu <wei.liu@...nel.org>, 
 Dexuan Cui <decui@...rosoft.com>, Bryan Tan <bryan-bt.tan@...adcom.com>, 
 Vishnu Dasa <vishnu.dasa@...adcom.com>, 
 Broadcom internal kernel review list <bcm-kernel-feedback-list@...adcom.com>, 
 Bobby Eshleman <bobbyeshleman@...il.com>
Cc: virtualization@...ts.linux.dev, netdev@...r.kernel.org, 
 linux-kselftest@...r.kernel.org, linux-kernel@...r.kernel.org, 
 kvm@...r.kernel.org, linux-hyperv@...r.kernel.org, 
 Sargun Dhillon <sargun@...gun.me>, berrange@...hat.com, 
 Bobby Eshleman <bobbyeshleman@...a.com>
Subject: [PATCH net-next v9 05/14] vsock: add netns and netns_tracker to
 vsock skb cb

From: Bobby Eshleman <bobbyeshleman@...a.com>

Add a net pointer, netns_tracker, and net_mode to the vsock skb and
helpers for getting/setting them. These fields are only used by
vsock_loopback in order to avoid net-related race conditions (more info
in the loopback patch).

This extends virtio_vsock_skb_cb to 32 bytes (with
CONFIG_NET_DEV_REFCNT_TRACKER=y):

struct virtio_vsock_skb_cb {
	struct net *               net;                  /*     0     8 */
	netns_tracker              ns_tracker;           /*     8     8 */
	enum vsock_net_mode        net_mode;             /*    16     4 */
	u32                        offset;               /*    20     4 */
	bool                       reply;                /*    24     1 */
	bool                       tap_delivered;        /*    25     1 */

	/* size: 32, cachelines: 1, members: 6 */
	/* padding: 6 */
	/* last cacheline: 32 bytes */
};

Signed-off-by: Bobby Eshleman <bobbyeshleman@...a.com>
---
Changes in v9:
- update commit message to specify usage by loopback only
- add comment in virtio_vsock_skb_cb mentioning usage by vsock_loopback
- add ns_tracker to skb->cb
- removed Stefano's Reviewed-by trailer due to ns_tracker addition (not
  sure if this is the right process thing to do)

Changes in v7:
- rename `orig_net_mode` to `net_mode`
- update commit message with a more complete explanation of changes

Changes in v5:
- some diff context change due to rebase to current net-next
---
 include/linux/virtio_vsock.h | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
index 18deb3c8dab3..a3ef752cdb95 100644
--- a/include/linux/virtio_vsock.h
+++ b/include/linux/virtio_vsock.h
@@ -10,6 +10,10 @@
 #define VIRTIO_VSOCK_SKB_HEADROOM (sizeof(struct virtio_vsock_hdr))

 struct virtio_vsock_skb_cb {
+	/* net, net_mode, and ns_tracker are only used by vsock_loopback. */
+	struct net *net;
+	netns_tracker ns_tracker;
+	enum vsock_net_mode net_mode;
 	u32 offset;
 	bool reply;
 	bool tap_delivered;
@@ -130,6 +134,35 @@ static inline size_t virtio_vsock_skb_len(struct sk_buff *skb)
 	return (size_t)(skb_end_pointer(skb) - skb->head);
 }

+static inline struct net *virtio_vsock_skb_net(struct sk_buff *skb)
+{
+	return VIRTIO_VSOCK_SKB_CB(skb)->net;
+}
+
+static inline void virtio_vsock_skb_set_net(struct sk_buff *skb, struct net *net)
+{
+	get_net_track(net, &VIRTIO_VSOCK_SKB_CB(skb)->ns_tracker, GFP_KERNEL);
+	VIRTIO_VSOCK_SKB_CB(skb)->net = net;
+}
+
+static inline void virtio_vsock_skb_clear_net(struct sk_buff *skb)
+{
+	put_net_track(VIRTIO_VSOCK_SKB_CB(skb)->net,
+		      &VIRTIO_VSOCK_SKB_CB(skb)->ns_tracker);
+	VIRTIO_VSOCK_SKB_CB(skb)->net = NULL;
+}
+
+static inline enum vsock_net_mode virtio_vsock_skb_net_mode(struct sk_buff *skb)
+{
+	return VIRTIO_VSOCK_SKB_CB(skb)->net_mode;
+}
+
+static inline void virtio_vsock_skb_set_net_mode(struct sk_buff *skb,
+						      enum vsock_net_mode net_mode)
+{
+	VIRTIO_VSOCK_SKB_CB(skb)->net_mode = net_mode;
+}
+
 /* Dimension the RX SKB so that the entire thing fits exactly into
  * a single 4KiB page. This avoids wasting memory due to alloc_skb()
  * rounding up to the next page order and also means that we

-- 
2.47.3