netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jon Kohler <jon@nutanix.com>
To: netdev@vger.kernel.org,
	Willem de Bruijn <willemdebruijn.kernel@gmail.com>,
	Jason Wang <jasowang@redhat.com>,
	Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	linux-kernel@vger.kernel.org (open list)
Cc: Jon Kohler <jon@nutanix.com>
Subject: [PATCH net-next v2 9/9] tun: flush deferred skb free list before bulk NAPI cache get
Date: Tue, 25 Nov 2025 13:00:36 -0700	[thread overview]
Message-ID: <20251125200041.1565663-10-jon@nutanix.com> (raw)
In-Reply-To: <20251125200041.1565663-1-jon@nutanix.com>

Call skb_defer_free_flush() immediately before invoking
napi_skb_cache_get_bulk() in the XDP batch path. This ensures any
deferred skb frees are processed so that the NAPI skb cache is refilled
just in time for use. Keeping the cache warm helps reduce unnecessary
IPIs during heavy transmit workloads.

Signed-off-by: Jon Kohler <jon@nutanix.com>
---
 drivers/net/tun.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index b48a66b39e0a..7d7f1ddcb707 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -2589,6 +2589,12 @@ static int tun_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 		rcu_read_lock();
 		bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 
+		/* Attempt to flush deferred free list immediately
+		 * prior to bulk get, which will help repopulate the local
+		 * cache and help reduce the amount of IPIs a TX hot core
+		 * will receive when the defer list grows high.
+		 */
+		skb_defer_free_flush();
 		num_skbs = napi_skb_cache_get_bulk(skbs, n);
 
 		for (i = 0; i < num_skbs; i++) {
-- 
2.43.0


  parent reply	other threads:[~2025-11-25 19:18 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-25 20:00 [PATCH net-next v2 0/9] tun: optimize SKB allocation with NAPI cache Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 1/9] tun: cleanup out label in tun_xdp_one Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 2/9] tun: correct drop statistics " Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 3/9] tun: correct drop statistics in tun_put_user Jon Kohler
2025-11-29  3:07   ` Willem de Bruijn
2025-12-02 16:40     ` Jon Kohler
2025-12-02 21:34       ` Willem de Bruijn
2025-12-02 21:36         ` Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 4/9] tun: correct drop statistics in tun_get_user Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 5/9] tun: use bulk NAPI cache allocation in tun_xdp_one Jon Kohler
2025-11-28  3:02   ` Jason Wang
2025-12-02 16:49     ` Jon Kohler
2025-12-02 17:32       ` Jesper Dangaard Brouer
2025-12-02 17:45         ` Jon Kohler
2025-12-03  4:10           ` Jason Wang
2025-12-03  4:34             ` Jon Kohler
2025-12-03  6:40               ` Jason Wang
2025-12-03  8:47         ` Sebastian Andrzej Siewior
2025-12-03 15:35           ` Jon Kohler
2025-12-05  7:58             ` Sebastian Andrzej Siewior
2025-12-05 13:21               ` Jesper Dangaard Brouer
2025-12-05 16:56                 ` Jon Kohler
2025-12-08 11:04                 ` Sebastian Andrzej Siewior
2025-11-25 20:00 ` [PATCH net-next v2 6/9] tun: use napi_build_skb in __tun_build_skb Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 7/9] tun: use napi_consume_skb() in tun_put_user Jon Kohler
2025-11-25 20:00 ` [PATCH net-next v2 8/9] net: core: export skb_defer_free_flush Jon Kohler
2025-11-25 20:00 ` Jon Kohler [this message]
2025-11-29  3:08 ` [PATCH net-next v2 0/9] tun: optimize SKB allocation with NAPI cache Willem de Bruijn
2025-12-02 16:38   ` Jon Kohler

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251125200041.1565663-10-jon@nutanix.com \
    --to=jon@nutanix.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=jasowang@redhat.com \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=willemdebruijn.kernel@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).