summaryrefslogtreecommitdiffstats
path: root/net/packet
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2024-05-15 18:33:58 +0200
committerJakub Kicinski <kuba@kernel.org>2024-05-17 04:38:05 +0200
commit581073f626e387d3e7eed55c48c8495584ead7ba (patch)
tree4ef79fb695b5a7965710ba045979db1ae9cc4158 /net/packet
parentvirtio_net: Fix missed rtnl_unlock (diff)
downloadlinux-581073f626e387d3e7eed55c48c8495584ead7ba.tar.xz
linux-581073f626e387d3e7eed55c48c8495584ead7ba.zip
af_packet: do not call packet_read_pending() from tpacket_destruct_skb()
trafgen performance considerably sank on hosts with many cores after the blamed commit. packet_read_pending() is very expensive, and calling it in af_packet fast path defeats Daniel intent in commit b013840810c2 ("packet: use percpu mmap tx frame pending refcount") tpacket_destruct_skb() makes room for one packet, we can immediately wakeup a producer, no need to completely drain the tx ring. Fixes: 89ed5b519004 ("af_packet: Block execution of tasks waiting for transmit to complete in AF_PACKET") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://lore.kernel.org/r/20240515163358.4105915-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'net/packet')
-rw-r--r--net/packet/af_packet.c3
1 files changed, 1 insertions, 2 deletions
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 8c6d3fbb4ed8..ea3ebc160e25 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2522,8 +2522,7 @@ static void tpacket_destruct_skb(struct sk_buff *skb)
ts = __packet_set_timestamp(po, ph, skb);
__packet_set_status(po, ph, TP_STATUS_AVAILABLE | ts);
- if (!packet_read_pending(&po->tx_ring))
- complete(&po->skb_completion);
+ complete(&po->skb_completion);
}
sock_wfree(skb);