kernel/linux.git/net/core/net-procfs.c, branch v6.18.22

net: add proper RCU protection to /proc/net/ptype

2026-02-11T12:41:58+00:00

[ Upstream commit f613e8b4afea0cd17c7168e8b00e25bc8d33175d ] Yin Fengwei reported an RCU stall in ptype_seq_show() and provided a patch. Real issue is that ptype_seq_next() and ptype_seq_show() violate RCU rules. ptype_seq_show() runs under rcu_read_lock(), and reads pt->dev to get device name without any barrier. At the same time, concurrent writers can remove a packet_type structure (which is correctly freed after an RCU grace period) and clear pt->dev without an RCU grace period. Define ptype_iter_state to carry a dev pointer along seq_net_private: struct ptype_iter_state { struct seq_net_private p; struct net_device *dev; // added in this patch }; We need to record the device pointer in ptype_get_idx() and ptype_seq_next() so that ptype_seq_show() is safe against concurrent pt->dev changes. We also need to add full RCU protection in ptype_seq_next(). (Missing READ_ONCE() when reading list.next values) Many thanks to Dong Chenchen for providing a repro. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Fixes: 1d10f8a1f40b ("net-procfs: show net devices bound packet types") Fixes: c353e8983e0d ("net: introduce per netns packet chains") Reported-by: Yin Fengwei Reported-by: Dong Chenchen Closes: https://lore.kernel.org/netdev/CANn89iKRRKPnWjJmb-_3a=sq+9h6DvTQM4DBZHT5ZRGPMzQaiA@mail.gmail.com/T/#m7b80b9fc9b9267f90e0b7aad557595f686f9c50d Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Tested-by: Yin Fengwei Link: https://patch.msgid.link/20260202205217.2881198-1-edumazet@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

net: use NUMA drop counters for softnet_data.dropped

2025-09-14T18:35:17+00:00

Hosts under DOS attack can suffer from false sharing in enqueue_to_backlog() : atomic_inc(&sd->dropped). This is because sd->dropped can be touched from many cpus, possibly residing on different NUMA nodes. Generalize the sk_drop_counters infrastucture added in commit c51613fa276f ("net: add sk->sk_drop_counters") and use it to replace softnet_data.dropped with NUMA friendly softnet_data.drop_counters. This adds 64 bytes per cpu, maybe more in the future if we increase the number of counters (currently 2) per 'struct numa_drop_counters'. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250909121942.1202585-1-edumazet@google.com Signed-off-by: Jakub Kicinski

net: add data-race annotations in softnet_seq_show()

2025-04-08T19:30:55+00:00

softnet_seq_show() reads several fields that might be updated concurrently. Add READ_ONCE() and WRITE_ONCE() annotations. Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250407163602.170356-4-edumazet@google.com Signed-off-by: Jakub Kicinski

net: rps: annotate data-races around (struct sd_flow_limit)->count

2025-04-08T19:30:55+00:00

softnet_seq_show() can read fl->count while another cpu updates this field from skb_flow_limit(). Make this field an 'unsigned int', as its only consumer only deals with 32 bit. Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250407163602.170356-3-edumazet@google.com Signed-off-by: Jakub Kicinski

net: introduce per netns packet chains

2025-03-24T20:58:22+00:00

Currently network taps unbound to any interface are linked in the global ptype_all list, affecting the performance in all the network namespaces. Add per netns ptypes chains, so that in the mentioned case only the netns owning the packet socket(s) is affected. While at that drop the global ptype_all list: no in kernel user registers a tap on "any" type without specifying either the target device or the target namespace (and IMHO doing that would not make any sense). Note that this adds a conditional in the fast path (to check for per netns ptype_specific list) and increases the dataset size by a cacheline (owing the per netns lists). Reviewed-by: Sabrina Dubroca Signed-off-by: Paolo Abeni Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/ae405f98875ee87f8150c460ad162de7e466f8a7.1742494826.git.pabeni@redhat.com Signed-off-by: Jakub Kicinski

net: make softnet_data.dropped an atomic_t

2024-04-01T10:28:32+00:00

If under extreme cpu backlog pressure enqueue_to_backlog() has to drop a packet, it could do this without dirtying a cache line and potentially slowing down the target cpu. Move sd->dropped into a separate cache line, and make it atomic. In non pressure mode, this field is not touched, no need to consume valuable space in a hot cache line. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

net: move ptype_all into net_hotdata

2024-03-08T05:12:41+00:00

ptype_all is used in rx/tx fast paths. Move it to net_hotdata for better cache locality. Signed-off-by: Eric Dumazet Acked-by: Soheil Hassas Yeganeh Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240306160031.874438-5-edumazet@google.com Signed-off-by: Jakub Kicinski

net-procfs: use xarray iterator to implement /proc/net/dev

2024-02-09T03:06:09+00:00

In commit 759ab1edb56c ("net: store netdevs in an xarray") Jakub added net->dev_by_index to map ifindex to netdevices. We can get rid of the old hash table (net->dev_index_head), one patch at a time, if performance is acceptable. This patch removes unpleasant code to something more readable. As a bonus, /proc/net/dev gets netdevices sorted by their ifindex. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240207165318.3814525-1-edumazet@google.com Signed-off-by: Jakub Kicinski

net-sysfs: display two backlog queue len separately

2023-03-22T11:03:52+00:00

Sometimes we need to know which one of backlog queue can be exactly long enough to cause some latency when debugging this part is needed. Thus, we can then separate the display of both. Signed-off-by: Jason Xing Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230321015746.96994-1-kerneljasonxing@gmail.com Signed-off-by: Paolo Abeni

net: extract a few internals from netdevice.h

2022-04-08T03:32:09+00:00

There's a number of functions and static variables used under net/core/ but not from the outside. We currently dump most of them into netdevice.h. That bad for many reasons: - netdevice.h is very cluttered, hard to figure out what the APIs are; - netdevice.h is very long; - we have to touch netdevice.h more which causes expensive incremental builds. Create a header under net/core/ and move some declarations. The new header is also a bit of a catch-all but that's fine, if we create more specific headers people will likely over-think where their declaration fit best. And end up putting them in netdevice.h, again. More work should be done on splitting netdevice.h into more targeted headers, but that'd be more time consuming so small steps. Signed-off-by: Jakub Kicinski