[PATCH net-next] net: remove cpu stall in txq_trans_update()

[PATCH net-next] net: remove cpu stall in txq_trans_update()

[Eric Dumazet]

txq_trans_update() currently uses txq->xmit_lock_owner
to conditionally update txq->trans_start.

For regular devices, txq->xmit_lock_owner is updated
from HARD_TX_LOCK() and HARD_TX_UNLOCK(), and this apparently
causes cpu stalls.

Using dev->lltx, which sits in a read-mostly cache-line,
and already used in HARD_TX_LOCK() and HARD_TX_UNLOCK()
helps cpu prediction.

On an AMD EPYC 7B12 dual socket server, tcp_rr with 128 threads
and 30,000 flows gets a 5 % increase in throughput.

As explained in commit 95ecba62e2fd ("net: fix races in
netdev_tx_sent_queue()/dev_watchdog()") I am planning
to no longer update txq->trans_start in the fast path
in a followup patch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 2 +-
 include/linux/netdevice.h                | 7 ++++---
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index c9fd34787c9986946c06e3d8d9de693c4438ab6a..e78de79a5d78c2e673809841e5c6d2dc35c754a1 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -427,7 +427,7 @@ static void am65_cpsw_nuss_ndo_host_tx_timeout(struct net_device *ndev,
 
 	if (netif_tx_queue_stopped(netif_txq)) {
 		/* try recover if stopped by us */
-		txq_trans_update(netif_txq);
+		txq_trans_update(ndev, netif_txq);
 		netif_tx_wake_queue(netif_txq);
 	}
 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index cf3b6445817bb9d3a142da10549ade1c49659313..0b703f0aa2043e537b7f74a4532f89f1f2890b08 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -4688,9 +4688,10 @@ static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
 /*
  * txq->trans_start can be read locklessly from dev_watchdog()
  */
-static inline void txq_trans_update(struct netdev_queue *txq)
+static inline void txq_trans_update(const struct net_device *dev,
+				    struct netdev_queue *txq)
 {
-	if (txq->xmit_lock_owner != -1)
+	if (!dev->lltx)
 		WRITE_ONCE(txq->trans_start, jiffies);
 }
 
@@ -5209,7 +5210,7 @@ static inline netdev_tx_t netdev_start_xmit(struct sk_buff *skb, struct net_devi
 
 	rc = __netdev_start_xmit(ops, skb, dev, more);
 	if (rc == NETDEV_TX_OK)
-		txq_trans_update(txq);
+		txq_trans_update(dev, txq);
 
 	return rc;
 }
-- 
2.49.0.504.g3bcea36a83-goog

Re: [PATCH net-next] net: remove cpu stall in txq_trans_update()

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Tue,  8 Apr 2025 20:27:42 +0000 you wrote:
> txq_trans_update() currently uses txq->xmit_lock_owner
> to conditionally update txq->trans_start.
> 
> For regular devices, txq->xmit_lock_owner is updated
> from HARD_TX_LOCK() and HARD_TX_UNLOCK(), and this apparently
> causes cpu stalls.
> 
> [...]

Here is the summary with links:
  - [net-next] net: remove cpu stall in txq_trans_update()
    https://git.kernel.org/netdev/net-next/c/229671ac60e2

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