[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <687016ef6a2c_114eb42941d@willemb.c.googlers.com.notmuch>
Date: Thu, 10 Jul 2025 15:39:27 -0400
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
To: Eric Dumazet <edumazet@...gle.com>,
"David S . Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
Neal Cardwell <ncardwell@...gle.com>
Cc: Simon Horman <horms@...nel.org>,
Kuniyuki Iwashima <kuniyu@...gle.com>,
Willem de Bruijn <willemb@...gle.com>,
netdev@...r.kernel.org,
eric.dumazet@...il.com,
Eric Dumazet <edumazet@...gle.com>
Subject: Re: [PATCH net-next] selftests/net: packetdrill: add --mss option to
three tests
Eric Dumazet wrote:
> Three tests are cooking GSO packets but do not provide
> gso_size information to the kernel, triggering this message:
>
> TCP: tun0: Driver has suspect GRO implementation, TCP performance may be compromised.
>
> Add --mss option to avoid this warning.
>
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
Reviewed-by: Willem de Bruijn <willemb@...gle.com>
This is because of this in packetdrill linux_tun_write, I suppose:
if (packet->tcp && packet->mss) {
if (packet->ipv4)
gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
else
gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
gso.gso_size = packet->mss;
}
Which seems to imply that setting mss means that all generated packets
use GRO. Which tracks with the commit that introduced it: commit
58a7865c47e3 ("net-test: packetdrill: add GRO support to packetdrill").
Which includes
+++ b/gtests/net/packetdrill/config.h
@@ -73,6 +73,7 @@ struct config {
u32 speed; /* speed reported by tun driver;
* may require special tun driver
*/
+ int mss; /* gso_size for GRO packets to tun device */
Powered by blists - more mailing lists