* [PATCH net 1/4] Revert "tun/tap & vhost-net: avoid ptr_ring tail-drop when a qdisc is present"
2026-07-28 9:22 [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Simon Schippers
@ 2026-07-28 9:22 ` Simon Schippers
2026-07-28 9:22 ` [PATCH net 2/4] Revert "ptr_ring: move free-space check into separate helper" Simon Schippers
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Simon Schippers @ 2026-07-28 9:22 UTC (permalink / raw)
To: netdev
Cc: Willem de Bruijn, Jason Wang, Andrew Lunn, David S . Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Michael S . Tsirkin,
Eugenio Pérez, Simon Horman, Tim Gebauer, Brett Sheffield,
kvm, virtualization, linux-kernel, Simon Schippers
This reverts commit 1d6e569b7d0c0b2736636749e4be0a27f3cefcb3.
The commit stops the netdev queue when the ptr_ring is full instead of
dropping the packet. My own tests showed no relevant regression, but on
Brett Sheffield's librecast testbed an IPv6 multicast testcase got
slower. With 8 iperf3 TCP threads sending, the throughput dropped from
13.5 Gbit/s to 9.13 Gbit/s.
Reported-by: Brett Sheffield <brett@librecast.net>
Closes: https://lore.kernel.org/netdev/akVnoOYQOrt8k-Gu@karahi.librecast.net/
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
---
drivers/net/tun.c | 25 ++-----------------------
1 file changed, 2 insertions(+), 23 deletions(-)
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index ffbe6f13fb1f..ec5573f545af 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1018,7 +1018,6 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
struct netdev_queue *queue;
struct tun_file *tfile;
int len = skb->len;
- int ret;
rcu_read_lock();
tfile = rcu_dereference(tun->tfiles[txq]);
@@ -1073,33 +1072,13 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
nf_reset_ct(skb);
- queue = netdev_get_tx_queue(dev, txq);
-
- spin_lock(&tfile->tx_ring.producer_lock);
- ret = __ptr_ring_produce(&tfile->tx_ring, skb);
- if (!qdisc_txq_has_no_queue(queue) &&
- __ptr_ring_check_produce(&tfile->tx_ring) == -ENOSPC) {
- netif_tx_stop_queue(queue);
- /* Paired with smp_mb() in __tun_wake_queue() */
- smp_mb__after_atomic();
- if (!__ptr_ring_check_produce(&tfile->tx_ring))
- netif_tx_wake_queue(queue);
- }
- spin_unlock(&tfile->tx_ring.producer_lock);
-
- if (ret) {
- /* This should be a rare case if a qdisc is present, but
- * can happen due to lltx.
- * Since skb_tx_timestamp(), skb_orphan(),
- * run_ebpf_filter() and pskb_trim() could have tinkered
- * with the SKB, returning NETDEV_TX_BUSY is unsafe and
- * we must drop instead.
- */
+ if (ptr_ring_produce(&tfile->tx_ring, skb)) {
drop_reason = SKB_DROP_REASON_FULL_RING;
goto drop;
}
/* dev->lltx requires to do our own update of trans_start */
+ queue = netdev_get_tx_queue(dev, txq);
txq_trans_cond_update(queue);
/* Notify and wake up reader process */
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 2/4] Revert "ptr_ring: move free-space check into separate helper"
2026-07-28 9:22 [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Simon Schippers
2026-07-28 9:22 ` [PATCH net 1/4] Revert "tun/tap & vhost-net: avoid ptr_ring tail-drop when a qdisc is present" Simon Schippers
@ 2026-07-28 9:22 ` Simon Schippers
2026-07-28 9:22 ` [PATCH net 3/4] Revert "vhost-net: wake queue of tun/tap after ptr_ring consume" Simon Schippers
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Simon Schippers @ 2026-07-28 9:22 UTC (permalink / raw)
To: netdev
Cc: Willem de Bruijn, Jason Wang, Andrew Lunn, David S . Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Michael S . Tsirkin,
Eugenio Pérez, Simon Horman, Tim Gebauer, Brett Sheffield,
kvm, virtualization, linux-kernel, Simon Schippers
This reverts commit fba362c17d9d9211fc51f272156bb84fc23bdf98.
__ptr_ring_check_produce() has no users left after reverting
commit 1d6e569b7d0c ("tun/tap & vhost-net: avoid ptr_ring tail-drop
when a qdisc is present").
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
---
include/linux/ptr_ring.h | 20 ++------------------
1 file changed, 2 insertions(+), 18 deletions(-)
diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h
index c95e891903f0..d2c3629bbe45 100644
--- a/include/linux/ptr_ring.h
+++ b/include/linux/ptr_ring.h
@@ -96,20 +96,6 @@ static inline bool ptr_ring_full_bh(struct ptr_ring *r)
return ret;
}
-/* Note: callers invoking this in a loop must use a compiler barrier,
- * for example cpu_relax(). Callers must hold producer_lock.
- */
-static inline int __ptr_ring_check_produce(struct ptr_ring *r)
-{
- if (unlikely(!r->size))
- return -EINVAL;
-
- if (data_race(r->queue[r->producer]))
- return -ENOSPC;
-
- return 0;
-}
-
/* Note: callers invoking this in a loop must use a compiler barrier,
* for example cpu_relax(). Callers must hold producer_lock.
* Callers are responsible for making sure pointer that is being queued
@@ -117,10 +103,8 @@ static inline int __ptr_ring_check_produce(struct ptr_ring *r)
*/
static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr)
{
- int p = __ptr_ring_check_produce(r);
-
- if (p)
- return p;
+ if (unlikely(!r->size) || data_race(r->queue[r->producer]))
+ return -ENOSPC;
/* Make sure the pointer we are storing points to a valid data. */
/* Pairs with the dependency ordering in __ptr_ring_consume. */
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 3/4] Revert "vhost-net: wake queue of tun/tap after ptr_ring consume"
2026-07-28 9:22 [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Simon Schippers
2026-07-28 9:22 ` [PATCH net 1/4] Revert "tun/tap & vhost-net: avoid ptr_ring tail-drop when a qdisc is present" Simon Schippers
2026-07-28 9:22 ` [PATCH net 2/4] Revert "ptr_ring: move free-space check into separate helper" Simon Schippers
@ 2026-07-28 9:22 ` Simon Schippers
2026-07-28 9:22 ` [PATCH net 4/4] Revert "tun/tap: add ptr_ring consume helper with netdev queue wakeup" Simon Schippers
2026-07-28 19:22 ` [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Michael S. Tsirkin
4 siblings, 0 replies; 6+ messages in thread
From: Simon Schippers @ 2026-07-28 9:22 UTC (permalink / raw)
To: netdev
Cc: Willem de Bruijn, Jason Wang, Andrew Lunn, David S . Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Michael S . Tsirkin,
Eugenio Pérez, Simon Horman, Tim Gebauer, Brett Sheffield,
kvm, virtualization, linux-kernel, Simon Schippers
This reverts commit baf808fe4fcd35767ab732b4ab2ea80dabfd97a6.
There is no netdev queue left to wake after reverting
commit 1d6e569b7d0c ("tun/tap & vhost-net: avoid ptr_ring tail-drop
when a qdisc is present").
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
---
drivers/net/tun.c | 23 -----------------------
drivers/vhost/net.c | 21 ++++++---------------
include/linux/if_tun.h | 3 ---
3 files changed, 6 insertions(+), 41 deletions(-)
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index ec5573f545af..39abc3078097 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -3787,29 +3787,6 @@ struct ptr_ring *tun_get_tx_ring(struct file *file)
}
EXPORT_SYMBOL_GPL(tun_get_tx_ring);
-/* Callers must hold ring.consumer_lock */
-void tun_wake_queue(struct file *file, int consumed)
-{
- struct tun_file *tfile;
- struct tun_struct *tun;
-
- if (file->f_op != &tun_fops)
- return;
-
- tfile = file->private_data;
- if (!tfile)
- return;
-
- rcu_read_lock();
-
- tun = rcu_dereference(tfile->tun);
- if (tun)
- __tun_wake_queue(tun, tfile, consumed);
-
- rcu_read_unlock();
-}
-EXPORT_SYMBOL_GPL(tun_wake_queue);
-
module_init(tun_init);
module_exit(tun_cleanup);
MODULE_DESCRIPTION(DRV_DESCRIPTION);
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 3e72b9c6af0c..6949b704166d 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -176,21 +176,13 @@ static void *vhost_net_buf_consume(struct vhost_net_buf *rxq)
return ret;
}
-static int vhost_net_buf_produce(struct sock *sk,
- struct vhost_net_virtqueue *nvq)
+static int vhost_net_buf_produce(struct vhost_net_virtqueue *nvq)
{
- struct file *file = sk->sk_socket->file;
struct vhost_net_buf *rxq = &nvq->rxq;
rxq->head = 0;
- spin_lock(&nvq->rx_ring->consumer_lock);
- rxq->tail = __ptr_ring_consume_batched(nvq->rx_ring, rxq->queue,
- VHOST_NET_BATCH);
-
- if (rxq->tail)
- tun_wake_queue(file, rxq->tail);
-
- spin_unlock(&nvq->rx_ring->consumer_lock);
+ rxq->tail = ptr_ring_consume_batched(nvq->rx_ring, rxq->queue,
+ VHOST_NET_BATCH);
return rxq->tail;
}
@@ -217,15 +209,14 @@ static int vhost_net_buf_peek_len(void *ptr)
return __skb_array_len_with_tag(ptr);
}
-static int vhost_net_buf_peek(struct sock *sk,
- struct vhost_net_virtqueue *nvq)
+static int vhost_net_buf_peek(struct vhost_net_virtqueue *nvq)
{
struct vhost_net_buf *rxq = &nvq->rxq;
if (!vhost_net_buf_is_empty(rxq))
goto out;
- if (!vhost_net_buf_produce(sk, nvq))
+ if (!vhost_net_buf_produce(nvq))
return 0;
out:
@@ -1013,7 +1004,7 @@ static int peek_head_len(struct vhost_net_virtqueue *rvq, struct sock *sk)
unsigned long flags;
if (rvq->rx_ring)
- return vhost_net_buf_peek(sk, rvq);
+ return vhost_net_buf_peek(rvq);
spin_lock_irqsave(&sk->sk_receive_queue.lock, flags);
head = skb_peek(&sk->sk_receive_queue);
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 5f3e206c7a73..80166eb62f41 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -22,7 +22,6 @@ struct tun_msg_ctl {
#if defined(CONFIG_TUN) || defined(CONFIG_TUN_MODULE)
struct socket *tun_get_socket(struct file *);
struct ptr_ring *tun_get_tx_ring(struct file *file);
-void tun_wake_queue(struct file *file, int consumed);
static inline bool tun_is_xdp_frame(void *ptr)
{
@@ -56,8 +55,6 @@ static inline struct ptr_ring *tun_get_tx_ring(struct file *f)
return ERR_PTR(-EINVAL);
}
-static inline void tun_wake_queue(struct file *f, int consumed) {}
-
static inline bool tun_is_xdp_frame(void *ptr)
{
return false;
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 4/4] Revert "tun/tap: add ptr_ring consume helper with netdev queue wakeup"
2026-07-28 9:22 [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Simon Schippers
` (2 preceding siblings ...)
2026-07-28 9:22 ` [PATCH net 3/4] Revert "vhost-net: wake queue of tun/tap after ptr_ring consume" Simon Schippers
@ 2026-07-28 9:22 ` Simon Schippers
2026-07-28 19:22 ` [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Michael S. Tsirkin
4 siblings, 0 replies; 6+ messages in thread
From: Simon Schippers @ 2026-07-28 9:22 UTC (permalink / raw)
To: netdev
Cc: Willem de Bruijn, Jason Wang, Andrew Lunn, David S . Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Michael S . Tsirkin,
Eugenio Pérez, Simon Horman, Tim Gebauer, Brett Sheffield,
kvm, virtualization, linux-kernel, Simon Schippers
This reverts commit d4c22d70d7253dd727c71484c58d504f6c630343.
There is no netdev queue left to wake after reverting
commit 1d6e569b7d0c ("tun/tap & vhost-net: avoid ptr_ring tail-drop
when a qdisc is present").
Signed-off-by: Simon Schippers <simon.schippers@tu-dortmund.de>
---
drivers/net/tun.c | 61 ++++-------------------------------------------
1 file changed, 4 insertions(+), 57 deletions(-)
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 39abc3078097..fed9dfdfcc3b 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -145,8 +145,6 @@ struct tun_file {
struct list_head next;
struct tun_struct *detached;
struct ptr_ring tx_ring;
- /* Protected by tx_ring.consumer_lock */
- int cons_cnt;
struct xdp_rxq_info xdp_rxq;
};
@@ -590,13 +588,8 @@ static void __tun_detach(struct tun_file *tfile, bool clean)
rcu_assign_pointer(tun->tfiles[index],
tun->tfiles[tun->numqueues - 1]);
ntfile = rtnl_dereference(tun->tfiles[index]);
- spin_lock(&ntfile->tx_ring.consumer_lock);
ntfile->queue_index = index;
ntfile->xdp_rxq.queue_index = index;
- ntfile->cons_cnt = 0;
- if (__ptr_ring_empty(&ntfile->tx_ring))
- netif_wake_subqueue(tun->dev, index);
- spin_unlock(&ntfile->tx_ring.consumer_lock);
rcu_assign_pointer(tun->tfiles[tun->numqueues - 1],
NULL);
@@ -737,9 +730,6 @@ static int tun_attach(struct tun_struct *tun, struct file *file,
goto out;
}
- spin_lock(&tfile->tx_ring.consumer_lock);
- tfile->cons_cnt = 0;
- spin_unlock(&tfile->tx_ring.consumer_lock);
tfile->queue_index = tun->numqueues;
tfile->socket.sk->sk_shutdown &= ~RCV_SHUTDOWN;
@@ -2126,46 +2116,13 @@ static ssize_t tun_put_user(struct tun_struct *tun,
return total;
}
-/* Callers must hold ring.consumer_lock */
-static void __tun_wake_queue(struct tun_struct *tun,
- struct tun_file *tfile, int consumed)
-{
- struct netdev_queue *txq = netdev_get_tx_queue(tun->dev,
- tfile->queue_index);
-
- /* Paired with smp_mb__after_atomic() in tun_net_xmit() */
- smp_mb();
- if (netif_tx_queue_stopped(txq)) {
- tfile->cons_cnt += consumed;
- if (tfile->cons_cnt >= tfile->tx_ring.size / 2 ||
- __ptr_ring_empty(&tfile->tx_ring)) {
- netif_tx_wake_queue(txq);
- tfile->cons_cnt = 0;
- }
- }
-}
-
-static void *tun_ring_consume(struct tun_struct *tun, struct tun_file *tfile)
-{
- void *ptr;
-
- spin_lock(&tfile->tx_ring.consumer_lock);
- ptr = __ptr_ring_consume(&tfile->tx_ring);
- if (ptr)
- __tun_wake_queue(tun, tfile, 1);
-
- spin_unlock(&tfile->tx_ring.consumer_lock);
- return ptr;
-}
-
-static void *tun_ring_recv(struct tun_struct *tun, struct tun_file *tfile,
- int noblock, int *err)
+static void *tun_ring_recv(struct tun_file *tfile, int noblock, int *err)
{
DECLARE_WAITQUEUE(wait, current);
void *ptr = NULL;
int error = 0;
- ptr = tun_ring_consume(tun, tfile);
+ ptr = ptr_ring_consume(&tfile->tx_ring);
if (ptr)
goto out;
if (noblock) {
@@ -2177,7 +2134,7 @@ static void *tun_ring_recv(struct tun_struct *tun, struct tun_file *tfile,
while (1) {
set_current_state(TASK_INTERRUPTIBLE);
- ptr = tun_ring_consume(tun, tfile);
+ ptr = ptr_ring_consume(&tfile->tx_ring);
if (ptr)
break;
if (signal_pending(current)) {
@@ -2214,7 +2171,7 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
if (!ptr) {
/* Read frames from ring */
- ptr = tun_ring_recv(tun, tfile, noblock, &err);
+ ptr = tun_ring_recv(tfile, noblock, &err);
if (!ptr)
return err;
}
@@ -3669,16 +3626,6 @@ static int tun_queue_resize(struct tun_struct *tun)
dev->tx_queue_len, GFP_KERNEL,
tun_ptr_free);
- if (!ret) {
- for (i = 0; i < tun->numqueues; i++) {
- tfile = rtnl_dereference(tun->tfiles[i]);
- spin_lock(&tfile->tx_ring.consumer_lock);
- netif_wake_subqueue(tun->dev, tfile->queue_index);
- tfile->cons_cnt = 0;
- spin_unlock(&tfile->tx_ring.consumer_lock);
- }
- }
-
kfree(rings);
return ret;
}
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops"
2026-07-28 9:22 [PATCH net 0/4] Revert "tun/tap & vhost-net: apply qdisc backpressure on full ptr_ring to reduce TX drops" Simon Schippers
` (3 preceding siblings ...)
2026-07-28 9:22 ` [PATCH net 4/4] Revert "tun/tap: add ptr_ring consume helper with netdev queue wakeup" Simon Schippers
@ 2026-07-28 19:22 ` Michael S. Tsirkin
4 siblings, 0 replies; 6+ messages in thread
From: Michael S. Tsirkin @ 2026-07-28 19:22 UTC (permalink / raw)
To: Simon Schippers
Cc: netdev, Willem de Bruijn, Jason Wang, Andrew Lunn,
David S . Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Eugenio Pérez, Simon Horman, Tim Gebauer, Brett Sheffield,
kvm, virtualization, linux-kernel
On Tue, Jul 28, 2026 at 11:22:36AM +0200, Simon Schippers wrote:
> Commit 1d6e569b7d0c ("tun/tap & vhost-net: avoid ptr_ring tail-drop when
> a qdisc is present") did not show a relevant performance regression in my
> testing, but on Brett Sheffield's librecast testbed it causes a
> significant throughput drop in an IPv6 multicast testcase. The regression
> can be pinpointed to multiple iperf3 TCP threads sending: for 8 threads
> the throughput dropped from 13.5 Gbit/s to 9.13 Gbit/s.
>
> Therefore this series reverts the qdisc backpressure work.
>
> Making the backpressure opt-in via a new IFF_BACKPRESSURE flag was
> proposed in [1], but a new IFF_* flag needs more review scrutiny than is
> available at the moment, so a revert was requested instead. The opt-in
> will be resubmitted for net-next later.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Thanks!
> [1] Link: https://lore.kernel.org/netdev/20260709095511.168235-1-simon.schippers@tu-dortmund.de/
>
> Reported-by: Brett Sheffield <brett@librecast.net>
> Closes: https://lore.kernel.org/netdev/akVnoOYQOrt8k-Gu@karahi.librecast.net/
>
> Simon Schippers (4):
> Revert "tun/tap & vhost-net: avoid ptr_ring tail-drop when a qdisc is
> present"
> Revert "ptr_ring: move free-space check into separate helper"
> Revert "vhost-net: wake queue of tun/tap after ptr_ring consume"
> Revert "tun/tap: add ptr_ring consume helper with netdev queue wakeup"
>
> drivers/net/tun.c | 109 +++------------------------------------
> drivers/vhost/net.c | 21 +++-----
> include/linux/if_tun.h | 3 --
> include/linux/ptr_ring.h | 20 +------
> 4 files changed, 14 insertions(+), 139 deletions(-)
>
>
> base-commit: b14361aca6350ff7907b0e9903c7b94dc7d5d4a0
> --
> 2.43.0
^ permalink raw reply [flat|nested] 6+ messages in thread