netdev - [PATCH net-next v3 8/9] xsk: support generic batch xmit in copy mode

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: <20251021131209.41491-9-kerneljasonxing@gmail.com>
Date: Tue, 21 Oct 2025 21:12:08 +0800
From: Jason Xing <kerneljasonxing@...il.com>
To: davem@...emloft.net,
	edumazet@...gle.com,
	kuba@...nel.org,
	pabeni@...hat.com,
	bjorn@...nel.org,
	magnus.karlsson@...el.com,
	maciej.fijalkowski@...el.com,
	jonathan.lemon@...il.com,
	sdf@...ichev.me,
	ast@...nel.org,
	daniel@...earbox.net,
	hawk@...nel.org,
	john.fastabend@...il.com,
	joe@...a.to,
	willemdebruijn.kernel@...il.com
Cc: bpf@...r.kernel.org,
	netdev@...r.kernel.org,
	Jason Xing <kernelxing@...cent.com>
Subject: [PATCH net-next v3 8/9] xsk: support generic batch xmit in copy mode

From: Jason Xing <kernelxing@...cent.com>

- Move xs->mutex into xsk_generic_xmit to prevent race condition when
  application manipulates generic_xmit_batch simultaneously.
- Enable batch xmit eventually.

Make the whole feature work eventually.

Signed-off-by: Jason Xing <kernelxing@...cent.com>
---
 net/xdp/xsk.c | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 1fa099653b7d..3741071c68fd 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -891,8 +891,6 @@ static int __xsk_generic_xmit_batch(struct xdp_sock *xs)
 	struct sk_buff *skb;
 	int err = 0;
 
-	mutex_lock(&xs->mutex);
-
 	/* Since we dropped the RCU read lock, the socket state might have changed. */
 	if (unlikely(!xsk_is_bound(xs))) {
 		err = -ENXIO;
@@ -982,21 +980,17 @@ static int __xsk_generic_xmit_batch(struct xdp_sock *xs)
 	if (sent_frame)
 		__xsk_tx_release(xs);
 
-	mutex_unlock(&xs->mutex);
 	return err;
 }
 
-static int __xsk_generic_xmit(struct sock *sk)
+static int __xsk_generic_xmit(struct xdp_sock *xs)
 {
-	struct xdp_sock *xs = xdp_sk(sk);
 	bool sent_frame = false;
 	struct xdp_desc desc;
 	struct sk_buff *skb;
 	u32 max_batch;
 	int err = 0;
 
-	mutex_lock(&xs->mutex);
-
 	/* Since we dropped the RCU read lock, the socket state might have changed. */
 	if (unlikely(!xsk_is_bound(xs))) {
 		err = -ENXIO;
@@ -1071,17 +1065,22 @@ static int __xsk_generic_xmit(struct sock *sk)
 	if (sent_frame)
 		__xsk_tx_release(xs);
 
-	mutex_unlock(&xs->mutex);
 	return err;
 }
 
 static int xsk_generic_xmit(struct sock *sk)
 {
+	struct xdp_sock *xs = xdp_sk(sk);
 	int ret;
 
 	/* Drop the RCU lock since the SKB path might sleep. */
 	rcu_read_unlock();
-	ret = __xsk_generic_xmit(sk);
+	mutex_lock(&xs->mutex);
+	if (xs->batch.generic_xmit_batch)
+		ret = __xsk_generic_xmit_batch(xs);
+	else
+		ret = __xsk_generic_xmit(xs);
+	mutex_unlock(&xs->mutex);
 	/* Reaquire RCU lock before going into common code. */
 	rcu_read_lock();
 
-- 
2.41.3