| Message ID | 20240608221057.16070-1-fw@strlen.de (mailing list archive) |
|---|---|
| Headers | show |
| Series | net: flow dissector: allow explicit passing of netns | expand |
On Sun, Jun 9, 2024 at 12:20 AM Florian Westphal <fw@strlen.de> wrote: > > Change since last version: > fix kdoc comment warning reported by kbuild robot, no other changes, > thus retaining RvB tags from Eric and Willem. > v1: https://lore.kernel.org/netdev/20240607083205.3000-1-fw@strlen.de/ Thanks Florian Reviewed-by: Eric Dumazet <edumazet@google.com>
Hello: This series was applied to netdev/net-next.git (main) by Jakub Kicinski <kuba@kernel.org>: On Sun, 9 Jun 2024 00:10:38 +0200 you wrote: > Change since last version: > fix kdoc comment warning reported by kbuild robot, no other changes, > thus retaining RvB tags from Eric and Willem. > v1: https://lore.kernel.org/netdev/20240607083205.3000-1-fw@strlen.de/ > > Years ago flow dissector gained ability to delegate flow dissection > to a bpf program, scoped per netns. > > [...] Here is the summary with links: - [net-next,v2,1/2] net: add and use skb_get_hash_net https://git.kernel.org/netdev/net-next/c/b975d3ee5962 - [net-next,v2,2/2] net: add and use __skb_get_hash_symmetric_net https://git.kernel.org/netdev/net-next/c/d1dab4f71d37 You are awesome, thank you!
Hi, This series got applied to net-next. But I can trigger this splat via nftables/tests/shell in net.git (6.10-rc). As well as in -stable 6.1.x: Jun 26 02:19:26 curiosity kernel: [ 1211.840595] ------------[ cut here ]------------ Jun 26 02:19:26 curiosity kernel: [ 1211.840605] WARNING: CPU: 0 PID: 70274 at net/core/flow_dissector.c:1016 __skb_flow_dissect+0x107e/0x2860 [...] Jun 26 02:19:26 curiosity kernel: [ 1211.841240] CPU: 0 PID: 70274 Comm: socat Not tainted 6.1.93+ #18 I think that turning this into DEBUG_NET_WARN_ON_ONCE as Willem suggested provides a workaround for net.git until Florian's fixes in net-next hit -stable. Would you accept such a patch? Thanks. On Sun, Jun 09, 2024 at 12:10:38AM +0200, Florian Westphal wrote: > Change since last version: > fix kdoc comment warning reported by kbuild robot, no other changes, > thus retaining RvB tags from Eric and Willem. > v1: https://lore.kernel.org/netdev/20240607083205.3000-1-fw@strlen.de/ > > Years ago flow dissector gained ability to delegate flow dissection > to a bpf program, scoped per netns. > > The netns is derived from skb->dev, and if that is not available, from > skb->sk. If neither is set, we hit a (benign) WARN_ON_ONCE(). > > This WARN_ON_ONCE can be triggered from netfilter. > Known skb origins are nf_send_reset and ipv4 stack generated IGMP > messages. > > Lets allow callers to pass the current netns explicitly and make > nf_tables use those instead. > > This targets net-next instead of net because the WARN is benign and this > is not a regression. > > Florian Westphal (2): > net: add and use skb_get_hash_net > net: add and use __skb_get_hash_symmetric_net > > include/linux/skbuff.h | 20 +++++++++++++++++--- > net/core/flow_dissector.c | 21 ++++++++++++++------- > net/netfilter/nf_tables_trace.c | 2 +- > net/netfilter/nft_hash.c | 3 ++- > 4 files changed, 34 insertions(+), 12 deletions(-) > > -- > 2.44.2 >
On Thu, 2024-06-27 at 01:49 +0200, Pablo Neira Ayuso wrote: > This series got applied to net-next. > > But I can trigger this splat via nftables/tests/shell in net.git (6.10-rc). > > As well as in -stable 6.1.x: > > Jun 26 02:19:26 curiosity kernel: [ 1211.840595] ------------[ cut here ]------------ > Jun 26 02:19:26 curiosity kernel: [ 1211.840605] WARNING: CPU: 0 PID: 70274 at net/core/flow_dissector.c:1016 __skb_flow_dissect+0x107e/0x2860 > [...] > Jun 26 02:19:26 curiosity kernel: [ 1211.841240] CPU: 0 PID: 70274 Comm: socat Not tainted 6.1.93+ #18 > > I think that turning this into DEBUG_NET_WARN_ON_ONCE as Willem > suggested provides a workaround for net.git until Florian's fixes in > net-next hit -stable. > > Would you accept such a patch? FWISW I think it makes sense. Thanks, Paolo