On 09/11/2013 09:54 AM, gregkh@xxxxxxxxxxxxxxxxxxx wrote:
> This is a note to let you know that I've just added the patch titled
>
> net: revert 8728c544a9c ("net: dev_pick_tx() fix")
>
> to the 3.10-stable tree which can be found at:
> http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
>
> The filename of the patch is:
> net-revert-8728c544a9c-net-dev_pick_tx-fix.patch
> and it can be found in the queue-3.10 subdirectory.
>
> If you, or anyone else, feels it should not be added to the stable tree,
> please let <stable@xxxxxxxxxxxxxxx> know about it.
>
>
> From e8b87e3d4dd56f2edc4fa09bc63681af9d4be0c5 Mon Sep 17 00:00:00 2001
> From: Eric Dumazet <edumazet@xxxxxxxxxx>
> Date: Wed, 28 Aug 2013 18:10:43 -0700
> Subject: net: revert 8728c544a9c ("net: dev_pick_tx() fix")
>
> From: Eric Dumazet <edumazet@xxxxxxxxxx>
>
> [ Upstream commit 702821f4ea6f68db18aa1de7d8ed62c6ba586a64 ]
>
> commit 8728c544a9cbdc ("net: dev_pick_tx() fix") and commit
> b6fe83e9525a ("bonding: refine IFF_XMIT_DST_RELEASE capability")
> are quite incompatible : Queue selection is disabled because skb
> dst was dropped before entering bonding device.
>
> This causes major performance regression, mainly because TCP packets
> for a given flow can be sent to multiple queues.
>
> This is particularly visible when using the new FQ packet scheduler
> with MQ + FQ setup on the slaves.
>
> We can safely revert the first commit now that 416186fbf8c5b
> ("net: Split core bits of netdev_pick_tx into __netdev_pick_tx")
> properly caps the queue_index.
>
> Reported-by: Xi Wang <xii@xxxxxxxxxx>
> Diagnosed-by: Xi Wang <xii@xxxxxxxxxx>
> Signed-off-by: Eric Dumazet <edumazet@xxxxxxxxxx>
> Cc: Tom Herbert <therbert@xxxxxxxxxx>
> Cc: Alexander Duyck <alexander.h.duyck@xxxxxxxxx>
> Cc: Denys Fedorysychenko <nuclearcat@xxxxxxxxxxxxxx>
> Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> ---
> net/core/flow_dissector.c | 11 +++--------
> 1 file changed, 3 insertions(+), 8 deletions(-)
>
> --- a/net/core/flow_dissector.c
> +++ b/net/core/flow_dissector.c
> @@ -345,14 +345,9 @@ u16 __netdev_pick_tx(struct net_device *
> if (new_index < 0)
> new_index = skb_tx_hash(dev, skb);
>
> - if (queue_index != new_index && sk) {
> - struct dst_entry *dst =
> - rcu_dereference_check(sk->sk_dst_cache, 1);
> -
> - if (dst && skb_dst(skb) == dst)
> - sk_tx_queue_set(sk, queue_index);
> -
> - }
> + if (queue_index != new_index && sk &&
> + rcu_access_pointer(sk->sk_dst_cache))
> + sk_tx_queue_set(sk, queue_index);
>
> queue_index = new_index;
> }
>
>
> Patches currently in stable-queue which might be from edumazet@xxxxxxxxxx are
>
> queue-3.10/ip_gre-fix-ipgre_header-to-return-correct-offset-mime-version-1.0.patch
> queue-3.10/tcp-cubic-fix-overflow-error-in-bictcp_update.patch
> queue-3.10/tcp-cubic-fix-bug-in-bictcp_acked.patch
> queue-3.10/net-revert-8728c544a9c-net-dev_pick_tx-fix.patch
> queue-3.10/net-ipv6-tcp-fix-potential-use-after-free-in-tcp_v6_do_rcv.patch
> queue-3.10/net-check-net.core.somaxconn-sysctl-values.patch
> queue-3.10/htb-fix-sign-extension-bug.patch
> queue-3.10/fib_trie-remove-potential-out-of-bound-access.patch
This is flawed. As per a patch Eric Dumazet sent out to netdev it should
be sk_tx_queue_set(sk, new_index). The queue_index is the old value.
Thanks,
Alex
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html