linux/Documentation/networking/ip-sysctl.rst

3821 lines
123 KiB

net: Retire DCCP socket. DCCP was orphaned in 2021 by commit 054c4610bd05 ("MAINTAINERS: dccp: move Gerrit Renker to CREDITS"), which noted that the last maintainer had been inactive for five years. In recent years, it has become a playground for syzbot, and most changes to DCCP have been odd bug fixes triggered by syzbot. Apart from that, the only changes have been driven by treewide or networking API updates or adjustments related to TCP. Thus, in 2023, we announced we would remove DCCP in 2025 via commit b144fcaf46d4 ("dccp: Print deprecation notice."). Since then, only one individual has contacted the netdev mailing list. [0] There is ongoing research for Multipath DCCP. The repository is hosted on GitHub [1], and development is not taking place through the upstream community. While the repository is published under the GPLv2 license, the scheduling part remains proprietary, with a LICENSE file [2] stating: "This is not Open Source software." The researcher mentioned a plan to address the licensing issue, upstream the patches, and step up as a maintainer, but there has been no further communication since then. Maintaining DCCP for a decade without any real users has become a burden. Therefore, it's time to remove it. Removing DCCP will also provide significant benefits to TCP. It allows us to freely reorganize the layout of struct inet_connection_sock, which is currently shared with DCCP, and optimize it to reduce the number of cachelines accessed in the TCP fast path. Note that we keep DCCP netfilter modules as requested. [3] Link: https://lore.kernel.org/netdev/20230710182253.81446-1-kuniyu@amazon.com/T/#u #[0] Link: https://github.com/telekom/mp-dccp #[1] Link: https://github.com/telekom/mp-dccp/blob/mpdccp_v03_k5.10/net/dccp/non_gpl_scheduler/LICENSE #[2] Link: https://lore.kernel.org/netdev/Z_VQ0KlCRkqYWXa-@calendula/ #[3] Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Acked-by: Paul Moore <paul@paul-moore.com> (LSM and SELinux) Acked-by: Casey Schaufler <casey@schaufler-ca.com> Link: https://patch.msgid.link/20250410023921.11307-3-kuniyu@amazon.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-04-09 19:36:45 -07:00
ipv4: introduce ip_dst_mtu_maybe_forward and protect forwarding path against pmtu spoofing While forwarding we should not use the protocol path mtu to calculate the mtu for a forwarded packet but instead use the interface mtu. We mark forwarded skbs in ip_forward with IPSKB_FORWARDED, which was introduced for multicast forwarding. But as it does not conflict with our usage in unicast code path it is perfect for reuse. I moved the functions ip_sk_accept_pmtu, ip_sk_use_pmtu and ip_skb_dst_mtu along with the new ip_dst_mtu_maybe_forward to net/ip.h to fix circular dependencies because of IPSKB_FORWARDED. Because someone might have written a software which does probe destinations manually and expects the kernel to honour those path mtus I introduced a new per-namespace "ip_forward_use_pmtu" knob so someone can disable this new behaviour. We also still use mtus which are locked on a route for forwarding. The reason for this change is, that path mtus information can be injected into the kernel via e.g. icmp_err protocol handler without verification of local sockets. As such, this could cause the IPv4 forwarding path to wrongfully emit fragmentation needed notifications or start to fragment packets along a path. Tunnel and ipsec output paths clear IPCB again, thus IPSKB_FORWARDED won't be set and further fragmentation logic will use the path mtu to determine the fragmentation size. They also recheck packet size with help of path mtu discovery and report appropriate errors. Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: David Miller <davem@davemloft.net> Cc: John Heffner <johnwheffner@gmail.com> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-09 10:01:15 +01:00
ipv4: introduce ip_dst_mtu_maybe_forward and protect forwarding path against pmtu spoofing While forwarding we should not use the protocol path mtu to calculate the mtu for a forwarded packet but instead use the interface mtu. We mark forwarded skbs in ip_forward with IPSKB_FORWARDED, which was introduced for multicast forwarding. But as it does not conflict with our usage in unicast code path it is perfect for reuse. I moved the functions ip_sk_accept_pmtu, ip_sk_use_pmtu and ip_skb_dst_mtu along with the new ip_dst_mtu_maybe_forward to net/ip.h to fix circular dependencies because of IPSKB_FORWARDED. Because someone might have written a software which does probe destinations manually and expects the kernel to honour those path mtus I introduced a new per-namespace "ip_forward_use_pmtu" knob so someone can disable this new behaviour. We also still use mtus which are locked on a route for forwarding. The reason for this change is, that path mtus information can be injected into the kernel via e.g. icmp_err protocol handler without verification of local sockets. As such, this could cause the IPv4 forwarding path to wrongfully emit fragmentation needed notifications or start to fragment packets along a path. Tunnel and ipsec output paths clear IPCB again, thus IPSKB_FORWARDED won't be set and further fragmentation logic will use the path mtu to determine the fragmentation size. They also recheck packet size with help of path mtu discovery and report appropriate errors. Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: David Miller <davem@davemloft.net> Cc: John Heffner <johnwheffner@gmail.com> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-09 10:01:15 +01:00
ipv4: introduce ip_dst_mtu_maybe_forward and protect forwarding path against pmtu spoofing While forwarding we should not use the protocol path mtu to calculate the mtu for a forwarded packet but instead use the interface mtu. We mark forwarded skbs in ip_forward with IPSKB_FORWARDED, which was introduced for multicast forwarding. But as it does not conflict with our usage in unicast code path it is perfect for reuse. I moved the functions ip_sk_accept_pmtu, ip_sk_use_pmtu and ip_skb_dst_mtu along with the new ip_dst_mtu_maybe_forward to net/ip.h to fix circular dependencies because of IPSKB_FORWARDED. Because someone might have written a software which does probe destinations manually and expects the kernel to honour those path mtus I introduced a new per-namespace "ip_forward_use_pmtu" knob so someone can disable this new behaviour. We also still use mtus which are locked on a route for forwarding. The reason for this change is, that path mtus information can be injected into the kernel via e.g. icmp_err protocol handler without verification of local sockets. As such, this could cause the IPv4 forwarding path to wrongfully emit fragmentation needed notifications or start to fragment packets along a path. Tunnel and ipsec output paths clear IPCB again, thus IPSKB_FORWARDED won't be set and further fragmentation logic will use the path mtu to determine the fragmentation size. They also recheck packet size with help of path mtu discovery and report appropriate errors. Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: David Miller <davem@davemloft.net> Cc: John Heffner <johnwheffner@gmail.com> Cc: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-09 10:01:15 +01:00
net: ipv4: Consider failed nexthops in multipath routes Multipath route lookups should consider knowledge about next hops and not select a hop that is known to be failed. Example: [h2] [h3] 15.0.0.5 | | 3| 3| [SP1] [SP2]--+ 1 2 1 2 | | /-------------+ | | \ / | | X | | / \ | | / \---------------\ | 1 2 1 2 12.0.0.2 [TOR1] 3-----------------3 [TOR2] 12.0.0.3 4 4 \ / \ / \ / -------| |-----/ 1 2 [TOR3] 3| | [h1] 12.0.0.1 host h1 with IP 12.0.0.1 has 2 paths to host h3 at 15.0.0.5: root@h1:~# ip ro ls ... 12.0.0.0/24 dev swp1 proto kernel scope link src 12.0.0.1 15.0.0.0/16 nexthop via 12.0.0.2 dev swp1 weight 1 nexthop via 12.0.0.3 dev swp1 weight 1 ... If the link between tor3 and tor1 is down and the link between tor1 and tor2 then tor1 is effectively cut-off from h1. Yet the route lookups in h1 are alternating between the 2 routes: ping 15.0.0.5 gets one and ssh 15.0.0.5 gets the other. Connections that attempt to use the 12.0.0.2 nexthop fail since that neighbor is not reachable: root@h1:~# ip neigh show ... 12.0.0.3 dev swp1 lladdr 00:02:00:00:00:1b REACHABLE 12.0.0.2 dev swp1 FAILED ... The failed path can be avoided by considering known neighbor information when selecting next hops. If the neighbor lookup fails we have no knowledge about the nexthop, so give it a shot. If there is an entry then only select the nexthop if the state is sane. This is similar to what fib_detect_death does. To maintain backward compatibility use of the neighbor information is based on a new sysctl, fib_multipath_use_neigh. Signed-off-by: David Ahern <dsa@cumulusnetworks.com> Reviewed-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-07 07:21:00 -07:00
net: ipv4: Consider failed nexthops in multipath routes Multipath route lookups should consider knowledge about next hops and not select a hop that is known to be failed. Example: [h2] [h3] 15.0.0.5 | | 3| 3| [SP1] [SP2]--+ 1 2 1 2 | | /-------------+ | | \ / | | X | | / \ | | / \---------------\ | 1 2 1 2 12.0.0.2 [TOR1] 3-----------------3 [TOR2] 12.0.0.3 4 4 \ / \ / \ / -------| |-----/ 1 2 [TOR3] 3| | [h1] 12.0.0.1 host h1 with IP 12.0.0.1 has 2 paths to host h3 at 15.0.0.5: root@h1:~# ip ro ls ... 12.0.0.0/24 dev swp1 proto kernel scope link src 12.0.0.1 15.0.0.0/16 nexthop via 12.0.0.2 dev swp1 weight 1 nexthop via 12.0.0.3 dev swp1 weight 1 ... If the link between tor3 and tor1 is down and the link between tor1 and tor2 then tor1 is effectively cut-off from h1. Yet the route lookups in h1 are alternating between the 2 routes: ping 15.0.0.5 gets one and ssh 15.0.0.5 gets the other. Connections that attempt to use the 12.0.0.2 nexthop fail since that neighbor is not reachable: root@h1:~# ip neigh show ... 12.0.0.3 dev swp1 lladdr 00:02:00:00:00:1b REACHABLE 12.0.0.2 dev swp1 FAILED ... The failed path can be avoided by considering known neighbor information when selecting next hops. If the neighbor lookup fails we have no knowledge about the nexthop, so give it a shot. If there is an entry then only select the nexthop if the state is sane. This is similar to what fib_detect_death does. To maintain backward compatibility use of the neighbor information is based on a new sysctl, fib_multipath_use_neigh. Signed-off-by: David Ahern <dsa@cumulusnetworks.com> Reviewed-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-07 07:21:00 -07:00
net: ipv4: Consider failed nexthops in multipath routes Multipath route lookups should consider knowledge about next hops and not select a hop that is known to be failed. Example: [h2] [h3] 15.0.0.5 | | 3| 3| [SP1] [SP2]--+ 1 2 1 2 | | /-------------+ | | \ / | | X | | / \ | | / \---------------\ | 1 2 1 2 12.0.0.2 [TOR1] 3-----------------3 [TOR2] 12.0.0.3 4 4 \ / \ / \ / -------| |-----/ 1 2 [TOR3] 3| | [h1] 12.0.0.1 host h1 with IP 12.0.0.1 has 2 paths to host h3 at 15.0.0.5: root@h1:~# ip ro ls ... 12.0.0.0/24 dev swp1 proto kernel scope link src 12.0.0.1 15.0.0.0/16 nexthop via 12.0.0.2 dev swp1 weight 1 nexthop via 12.0.0.3 dev swp1 weight 1 ... If the link between tor3 and tor1 is down and the link between tor1 and tor2 then tor1 is effectively cut-off from h1. Yet the route lookups in h1 are alternating between the 2 routes: ping 15.0.0.5 gets one and ssh 15.0.0.5 gets the other. Connections that attempt to use the 12.0.0.2 nexthop fail since that neighbor is not reachable: root@h1:~# ip neigh show ... 12.0.0.3 dev swp1 lladdr 00:02:00:00:00:1b REACHABLE 12.0.0.2 dev swp1 FAILED ... The failed path can be avoided by considering known neighbor information when selecting next hops. If the neighbor lookup fails we have no knowledge about the nexthop, so give it a shot. If there is an entry then only select the nexthop if the state is sane. This is similar to what fib_detect_death does. To maintain backward compatibility use of the neighbor information is based on a new sysctl, fib_multipath_use_neigh. Signed-off-by: David Ahern <dsa@cumulusnetworks.com> Reviewed-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-07 07:21:00 -07:00
net: ipv4: Add a sysctl to set multipath hash seed When calculating hashes for the purpose of multipath forwarding, both IPv4 and IPv6 code currently fall back on flow_hash_from_keys(). That uses a randomly-generated seed. That's a fine choice by default, but unfortunately some deployments may need a tighter control over the seed used. In this patch, make the seed configurable by adding a new sysctl key, net.ipv4.fib_multipath_hash_seed to control the seed. This seed is used specifically for multipath forwarding and not for the other concerns that flow_hash_from_keys() is used for, such as queue selection. Expose the knob as sysctl because other such settings, such as headers to hash, are also handled that way. Like those, the multipath hash seed is a per-netns variable. Despite being placed in the net.ipv4 namespace, the multipath seed sysctl is used for both IPv4 and IPv6, similarly to e.g. a number of TCP variables. The seed used by flow_hash_from_keys() is a 128-bit quantity. However it seems that usually the seed is a much more modest value. 32 bits seem typical (Cisco, Cumulus), some systems go even lower. For that reason, and to decouple the user interface from implementation details, go with a 32-bit quantity, which is then quadruplicated to form the siphash key. Signed-off-by: Petr Machata <petrm@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org> Reviewed-by: David Ahern <dsahern@kernel.org> Link: https://lore.kernel.org/r/20240607151357.421181-3-petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-06-07 17:13:54 +02:00
ipv4: Allow amount of dirty memory from fib resizing to be controllable fib_trie implementation calls synchronize_rcu when a certain amount of pages are dirty from freed entries. The number of pages was determined experimentally in 2009 (commit c3059477fce2d). At the current setting, synchronize_rcu is called often -- 51 times in a second in one test with an average of an 8 msec delay adding a fib entry. The total impact is a lot of slow down modifying the fib. This is seen in the output of 'time' - the difference between real time and sys+user. For example, using 720,022 single path routes and 'ip -batch'[1]: $ time ./ip -batch ipv4/routes-1-hops real 0m14.214s user 0m2.513s sys 0m6.783s So roughly 35% of the actual time to install the routes is from the ip command getting scheduled out, most notably due to synchronize_rcu (this is observed using 'perf sched timehist'). This patch makes the amount of dirty memory configurable between 64k where the synchronize_rcu is called often (small, low end systems that are memory sensitive) to 64M where synchronize_rcu is called rarely during a large FIB change (for high end systems with lots of memory). The default is 512kB which corresponds to the current setting of 128 pages with a 4kB page size. As an example, at 16MB the worst interval shows 4 calls to synchronize_rcu in a second blocking for up to 30 msec in a single instance, and a total of almost 100 msec across the 4 calls in the second. The trade off is allowing FIB entries to consume more memory in a given time window but but with much better fib insertion rates (~30% increase in prefixes/sec). With this patch and net.ipv4.fib_sync_mem set to 16MB, the same batch file runs in: $ time ./ip -batch ipv4/routes-1-hops real 0m9.692s user 0m2.491s sys 0m6.769s So the dead time is reduced to about 1/2 second or <5% of the real time. [1] 'ip' modified to not request ACK messages which improves route insertion times by about 20% Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-03-20 09:18:59 -07:00
ipv4: Allow amount of dirty memory from fib resizing to be controllable fib_trie implementation calls synchronize_rcu when a certain amount of pages are dirty from freed entries. The number of pages was determined experimentally in 2009 (commit c3059477fce2d). At the current setting, synchronize_rcu is called often -- 51 times in a second in one test with an average of an 8 msec delay adding a fib entry. The total impact is a lot of slow down modifying the fib. This is seen in the output of 'time' - the difference between real time and sys+user. For example, using 720,022 single path routes and 'ip -batch'[1]: $ time ./ip -batch ipv4/routes-1-hops real 0m14.214s user 0m2.513s sys 0m6.783s So roughly 35% of the actual time to install the routes is from the ip command getting scheduled out, most notably due to synchronize_rcu (this is observed using 'perf sched timehist'). This patch makes the amount of dirty memory configurable between 64k where the synchronize_rcu is called often (small, low end systems that are memory sensitive) to 64M where synchronize_rcu is called rarely during a large FIB change (for high end systems with lots of memory). The default is 512kB which corresponds to the current setting of 128 pages with a 4kB page size. As an example, at 16MB the worst interval shows 4 calls to synchronize_rcu in a second blocking for up to 30 msec in a single instance, and a total of almost 100 msec across the 4 calls in the second. The trade off is allowing FIB entries to consume more memory in a given time window but but with much better fib insertion rates (~30% increase in prefixes/sec). With this patch and net.ipv4.fib_sync_mem set to 16MB, the same batch file runs in: $ time ./ip -batch ipv4/routes-1-hops real 0m9.692s user 0m2.491s sys 0m6.769s So the dead time is reduced to about 1/2 second or <5% of the real time. [1] 'ip' modified to not request ACK messages which improves route insertion times by about 20% Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-03-20 09:18:59 -07:00
neighbor: Improve garbage collection The existing garbage collection algorithm has a number of problems: 1. The gc algorithm will not evict PERMANENT entries as those entries are managed by userspace, yet the existing algorithm walks the entire hash table which means it always considers PERMANENT entries when looking for entries to evict. In some use cases (e.g., EVPN) there can be tens of thousands of PERMANENT entries leading to wasted CPU cycles when gc kicks in. As an example, with 32k permanent entries, neigh_alloc has been observed taking more than 4 msec per invocation. 2. Currently, when the number of neighbor entries hits gc_thresh2 and the last flush for the table was more than 5 seconds ago gc kicks in walks the entire hash table evicting *all* entries not in PERMANENT or REACHABLE state and not marked as externally learned. There is no discriminator on when the neigh entry was created or if it just moved from REACHABLE to another NUD_VALID state (e.g., NUD_STALE). It is possible for entries to be created or for established neighbor entries to be moved to STALE (e.g., an external node sends an ARP request) right before the 5 second window lapses: -----|---------x|----------|----- t-5 t t+5 If that happens those entries are evicted during gc causing unnecessary thrashing on neighbor entries and userspace caches trying to track them. Further, this contradicts the description of gc_thresh2 which says "Entries older than 5 seconds will be cleared". One workaround is to make gc_thresh2 == gc_thresh3 but that negates the whole point of having separate thresholds. 3. Clearing *all* neigh non-PERMANENT/REACHABLE/externally learned entries when gc_thresh2 is exceeded is over kill and contributes to trashing especially during startup. This patch addresses these problems as follows: 1. Use of a separate list_head to track entries that can be garbage collected along with a separate counter. PERMANENT entries are not added to this list. The gc_thresh parameters are only compared to the new counter, not the total entries in the table. The forced_gc function is updated to only walk this new gc_list looking for entries to evict. 2. Entries are added to the list head at the tail and removed from the front. 3. Entries are only evicted if they were last updated more than 5 seconds ago, adhering to the original intent of gc_thresh2. 4. Forced gc is stopped once the number of gc_entries drops below gc_thresh2. 5. Since gc checks do not apply to PERMANENT entries, gc levels are skipped when allocating a new neighbor for a PERMANENT entry. By extension this means there are no explicit limits on the number of PERMANENT entries that can be created, but this is no different than FIB entries or FDB entries. Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-12-07 12:24:57 -08:00
neigh: new unresolved queue limits Le mercredi 09 novembre 2011 à 16:21 -0500, David Miller a écrit : > From: David Miller <davem@davemloft.net> > Date: Wed, 09 Nov 2011 16:16:44 -0500 (EST) > > > From: Eric Dumazet <eric.dumazet@gmail.com> > > Date: Wed, 09 Nov 2011 12:14:09 +0100 > > > >> unres_qlen is the number of frames we are able to queue per unresolved > >> neighbour. Its default value (3) was never changed and is responsible > >> for strange drops, especially if IP fragments are used, or multiple > >> sessions start in parallel. Even a single tcp flow can hit this limit. > > ... > > > > Ok, I've applied this, let's see what happens :-) > > Early answer, build fails. > > Please test build this patch with DECNET enabled and resubmit. The > decnet neigh layer still refers to the removed ->queue_len member. > > Thanks. Ouch, this was fixed on one machine yesterday, but not the other one I used this morning, sorry. [PATCH V5 net-next] neigh: new unresolved queue limits unres_qlen is the number of frames we are able to queue per unresolved neighbour. Its default value (3) was never changed and is responsible for strange drops, especially if IP fragments are used, or multiple sessions start in parallel. Even a single tcp flow can hit this limit. $ arp -d 192.168.20.108 ; ping -c 2 -s 8000 192.168.20.108 PING 192.168.20.108 (192.168.20.108) 8000(8028) bytes of data. 8008 bytes from 192.168.20.108: icmp_seq=2 ttl=64 time=0.322 ms Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-09 12:07:14 +00:00
neigh: new unresolved queue limits Le mercredi 09 novembre 2011 à 16:21 -0500, David Miller a écrit : > From: David Miller <davem@davemloft.net> > Date: Wed, 09 Nov 2011 16:16:44 -0500 (EST) > > > From: Eric Dumazet <eric.dumazet@gmail.com> > > Date: Wed, 09 Nov 2011 12:14:09 +0100 > > > >> unres_qlen is the number of frames we are able to queue per unresolved > >> neighbour. Its default value (3) was never changed and is responsible > >> for strange drops, especially if IP fragments are used, or multiple > >> sessions start in parallel. Even a single tcp flow can hit this limit. > > ... > > > > Ok, I've applied this, let's see what happens :-) > > Early answer, build fails. > > Please test build this patch with DECNET enabled and resubmit. The > decnet neigh layer still refers to the removed ->queue_len member. > > Thanks. Ouch, this was fixed on one machine yesterday, but not the other one I used this morning, sorry. [PATCH V5 net-next] neigh: new unresolved queue limits unres_qlen is the number of frames we are able to queue per unresolved neighbour. Its default value (3) was never changed and is responsible for strange drops, especially if IP fragments are used, or multiple sessions start in parallel. Even a single tcp flow can hit this limit. $ arp -d 192.168.20.108 ; ping -c 2 -s 8000 192.168.20.108 PING 192.168.20.108 (192.168.20.108) 8000(8028) bytes of data. 8008 bytes from 192.168.20.108: icmp_seq=2 ttl=64 time=0.322 ms Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-09 12:07:14 +00:00
neigh: new unresolved queue limits Le mercredi 09 novembre 2011 à 16:21 -0500, David Miller a écrit : > From: David Miller <davem@davemloft.net> > Date: Wed, 09 Nov 2011 16:16:44 -0500 (EST) > > > From: Eric Dumazet <eric.dumazet@gmail.com> > > Date: Wed, 09 Nov 2011 12:14:09 +0100 > > > >> unres_qlen is the number of frames we are able to queue per unresolved > >> neighbour. Its default value (3) was never changed and is responsible > >> for strange drops, especially if IP fragments are used, or multiple > >> sessions start in parallel. Even a single tcp flow can hit this limit. > > ... > > > > Ok, I've applied this, let's see what happens :-) > > Early answer, build fails. > > Please test build this patch with DECNET enabled and resubmit. The > decnet neigh layer still refers to the removed ->queue_len member. > > Thanks. Ouch, this was fixed on one machine yesterday, but not the other one I used this morning, sorry. [PATCH V5 net-next] neigh: new unresolved queue limits unres_qlen is the number of frames we are able to queue per unresolved neighbour. Its default value (3) was never changed and is responsible for strange drops, especially if IP fragments are used, or multiple sessions start in parallel. Even a single tcp flow can hit this limit. $ arp -d 192.168.20.108 ; ping -c 2 -s 8000 192.168.20.108 PING 192.168.20.108 (192.168.20.108) 8000(8028) bytes of data. 8008 bytes from 192.168.20.108: icmp_seq=2 ttl=64 time=0.322 ms Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-09 12:07:14 +00:00
neigh: new unresolved queue limits Le mercredi 09 novembre 2011 à 16:21 -0500, David Miller a écrit : > From: David Miller <davem@davemloft.net> > Date: Wed, 09 Nov 2011 16:16:44 -0500 (EST) > > > From: Eric Dumazet <eric.dumazet@gmail.com> > > Date: Wed, 09 Nov 2011 12:14:09 +0100 > > > >> unres_qlen is the number of frames we are able to queue per unresolved > >> neighbour. Its default value (3) was never changed and is responsible > >> for strange drops, especially if IP fragments are used, or multiple > >> sessions start in parallel. Even a single tcp flow can hit this limit. > > ... > > > > Ok, I've applied this, let's see what happens :-) > > Early answer, build fails. > > Please test build this patch with DECNET enabled and resubmit. The > decnet neigh layer still refers to the removed ->queue_len member. > > Thanks. Ouch, this was fixed on one machine yesterday, but not the other one I used this morning, sorry. [PATCH V5 net-next] neigh: new unresolved queue limits unres_qlen is the number of frames we are able to queue per unresolved neighbour. Its default value (3) was never changed and is responsible for strange drops, especially if IP fragments are used, or multiple sessions start in parallel. Even a single tcp flow can hit this limit. $ arp -d 192.168.20.108 ; ping -c 2 -s 8000 192.168.20.108 PING 192.168.20.108 (192.168.20.108) 8000(8028) bytes of data. 8008 bytes from 192.168.20.108: icmp_seq=2 ttl=64 time=0.322 ms Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-09 12:07:14 +00:00
net: ipv4: Emit notification when fib hardware flags are changed After installing a route to the kernel, user space receives an acknowledgment, which means the route was installed in the kernel, but not necessarily in hardware. The asynchronous nature of route installation in hardware can lead to a routing daemon advertising a route before it was actually installed in hardware. This can result in packet loss or mis-routed packets until the route is installed in hardware. It is also possible for a route already installed in hardware to change its action and therefore its flags. For example, a host route that is trapping packets can be "promoted" to perform decapsulation following the installation of an IPinIP/VXLAN tunnel. Emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/RTM_F_TRAP flags are changed. The aim is to provide an indication to user-space (e.g., routing daemons) about the state of the route in hardware. Introduce a sysctl that controls this behavior. Keep the default value at 0 (i.e., do not emit notifications) for several reasons: - Multiple RTM_NEWROUTE notification per-route might confuse existing routing daemons. - Convergence reasons in routing daemons. - The extra notifications will negatively impact the insertion rate. - Not all users are interested in these notifications. Signed-off-by: Amit Cohen <amcohen@nvidia.com> Acked-by: Roopa Prabhu <roopa@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-02-01 21:47:52 +02:00
net: ipv4: Emit notification when fib hardware flags are changed After installing a route to the kernel, user space receives an acknowledgment, which means the route was installed in the kernel, but not necessarily in hardware. The asynchronous nature of route installation in hardware can lead to a routing daemon advertising a route before it was actually installed in hardware. This can result in packet loss or mis-routed packets until the route is installed in hardware. It is also possible for a route already installed in hardware to change its action and therefore its flags. For example, a host route that is trapping packets can be "promoted" to perform decapsulation following the installation of an IPinIP/VXLAN tunnel. Emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/RTM_F_TRAP flags are changed. The aim is to provide an indication to user-space (e.g., routing daemons) about the state of the route in hardware. Introduce a sysctl that controls this behavior. Keep the default value at 0 (i.e., do not emit notifications) for several reasons: - Multiple RTM_NEWROUTE notification per-route might confuse existing routing daemons. - Convergence reasons in routing daemons. - The extra notifications will negatively impact the insertion rate. - Not all users are interested in these notifications. Signed-off-by: Amit Cohen <amcohen@nvidia.com> Acked-by: Roopa Prabhu <roopa@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-02-01 21:47:52 +02:00
net: ipv4: Emit notification when fib hardware flags are changed After installing a route to the kernel, user space receives an acknowledgment, which means the route was installed in the kernel, but not necessarily in hardware. The asynchronous nature of route installation in hardware can lead to a routing daemon advertising a route before it was actually installed in hardware. This can result in packet loss or mis-routed packets until the route is installed in hardware. It is also possible for a route already installed in hardware to change its action and therefore its flags. For example, a host route that is trapping packets can be "promoted" to perform decapsulation following the installation of an IPinIP/VXLAN tunnel. Emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/RTM_F_TRAP flags are changed. The aim is to provide an indication to user-space (e.g., routing daemons) about the state of the route in hardware. Introduce a sysctl that controls this behavior. Keep the default value at 0 (i.e., do not emit notifications) for several reasons: - Multiple RTM_NEWROUTE notification per-route might confuse existing routing daemons. - Convergence reasons in routing daemons. - The extra notifications will negatively impact the insertion rate. - Not all users are interested in these notifications. Signed-off-by: Amit Cohen <amcohen@nvidia.com> Acked-by: Roopa Prabhu <roopa@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-02-01 21:47:52 +02:00
inet: frags: use rhashtables for reassembly units Some applications still rely on IP fragmentation, and to be fair linux reassembly unit is not working under any serious load. It uses static hash tables of 1024 buckets, and up to 128 items per bucket (!!!) A work queue is supposed to garbage collect items when host is under memory pressure, and doing a hash rebuild, changing seed used in hash computations. This work queue blocks softirqs for up to 25 ms when doing a hash rebuild, occurring every 5 seconds if host is under fire. Then there is the problem of sharing this hash table for all netns. It is time to switch to rhashtables, and allocate one of them per netns to speedup netns dismantle, since this is a critical metric these days. Lookup is now using RCU. A followup patch will even remove the refcount hold/release left from prior implementation and save a couple of atomic operations. Before this patch, 16 cpus (16 RX queue NIC) could not handle more than 1 Mpps frags DDOS. After the patch, I reach 9 Mpps without any tuning, and can use up to 2GB of storage for the fragments (exact number depends on frags being evicted after timeout) $ grep FRAG /proc/net/sockstat FRAG: inuse 1966916 memory 2140004608 A followup patch will change the limits for 64bit arches. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Kirill Tkhai <ktkhai@virtuozzo.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: Florian Westphal <fw@strlen.de> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Alexander Aring <alex.aring@gmail.com> Cc: Stefan Schmidt <stefan@osg.samsung.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-31 12:58:49 -07:00
inet: frags: use rhashtables for reassembly units Some applications still rely on IP fragmentation, and to be fair linux reassembly unit is not working under any serious load. It uses static hash tables of 1024 buckets, and up to 128 items per bucket (!!!) A work queue is supposed to garbage collect items when host is under memory pressure, and doing a hash rebuild, changing seed used in hash computations. This work queue blocks softirqs for up to 25 ms when doing a hash rebuild, occurring every 5 seconds if host is under fire. Then there is the problem of sharing this hash table for all netns. It is time to switch to rhashtables, and allocate one of them per netns to speedup netns dismantle, since this is a critical metric these days. Lookup is now using RCU. A followup patch will even remove the refcount hold/release left from prior implementation and save a couple of atomic operations. Before this patch, 16 cpus (16 RX queue NIC) could not handle more than 1 Mpps frags DDOS. After the patch, I reach 9 Mpps without any tuning, and can use up to 2GB of storage for the fragments (exact number depends on frags being evicted after timeout) $ grep FRAG /proc/net/sockstat FRAG: inuse 1966916 memory 2140004608 A followup patch will change the limits for 64bit arches. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Kirill Tkhai <ktkhai@virtuozzo.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: Florian Westphal <fw@strlen.de> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Alexander Aring <alex.aring@gmail.com> Cc: Stefan Schmidt <stefan@osg.samsung.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-31 12:58:49 -07:00
tcp: change tcp_adv_win_scale and tcp_rmem[2] tcp_adv_win_scale default value is 2, meaning we expect a good citizen skb to have skb->len / skb->truesize ratio of 75% (3/4) In 2.6 kernels we (mis)accounted for typical MSS=1460 frame : 1536 + 64 + 256 = 1856 'estimated truesize', and 1856 * 3/4 = 1392. So these skbs were considered as not bloated. With recent truesize fixes, a typical MSS=1460 frame truesize is now the more precise : 2048 + 256 = 2304. But 2304 * 3/4 = 1728. So these skb are not good citizen anymore, because 1460 < 1728 (GRO can escape this problem because it build skbs with a too low truesize.) This also means tcp advertises a too optimistic window for a given allocated rcvspace : When receiving frames, sk_rmem_alloc can hit sk_rcvbuf limit and we call tcp_prune_queue()/tcp_collapse() too often, especially when application is slow to drain its receive queue or in case of losses (netperf is fast, scp is slow). This is a major latency source. We should adjust the len/truesize ratio to 50% instead of 75% This patch : 1) changes tcp_adv_win_scale default to 1 instead of 2 2) increase tcp_rmem[2] limit from 4MB to 6MB to take into account better truesize tracking and to allow autotuning tcp receive window to reach same value than before. Note that same amount of kernel memory is consumed compared to 2.6 kernels. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Tom Herbert <therbert@google.com> Cc: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-02 02:28:41 +00:00
tcp: auto corking With the introduction of TCP Small Queues, TSO auto sizing, and TCP pacing, we can implement Automatic Corking in the kernel, to help applications doing small write()/sendmsg() to TCP sockets. Idea is to change tcp_push() to check if the current skb payload is under skb optimal size (a multiple of MSS bytes) If under 'size_goal', and at least one packet is still in Qdisc or NIC TX queues, set the TCP Small Queue Throttled bit, so that the push will be delayed up to TX completion time. This delay might allow the application to coalesce more bytes in the skb in following write()/sendmsg()/sendfile() system calls. The exact duration of the delay is depending on the dynamics of the system, and might be zero if no packet for this flow is actually held in Qdisc or NIC TX ring. Using FQ/pacing is a way to increase the probability of autocorking being triggered. Add a new sysctl (/proc/sys/net/ipv4/tcp_autocorking) to control this feature and default it to 1 (enabled) Add a new SNMP counter : nstat -a | grep TcpExtTCPAutoCorking This counter is incremented every time we detected skb was under used and its flush was deferred. Tested: Interesting effects when using line buffered commands under ssh. Excellent performance results in term of cpu usage and total throughput. lpq83:~# echo 1 >/proc/sys/net/ipv4/tcp_autocorking lpq83:~# perf stat ./super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128 9410.39 Performance counter stats for './super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128': 35209.439626 task-clock # 2.901 CPUs utilized 2,294 context-switches # 0.065 K/sec 101 CPU-migrations # 0.003 K/sec 4,079 page-faults # 0.116 K/sec 97,923,241,298 cycles # 2.781 GHz [83.31%] 51,832,908,236 stalled-cycles-frontend # 52.93% frontend cycles idle [83.30%] 25,697,986,603 stalled-cycles-backend # 26.24% backend cycles idle [66.70%] 102,225,978,536 instructions # 1.04 insns per cycle # 0.51 stalled cycles per insn [83.38%] 18,657,696,819 branches # 529.906 M/sec [83.29%] 91,679,646 branch-misses # 0.49% of all branches [83.40%] 12.136204899 seconds time elapsed lpq83:~# echo 0 >/proc/sys/net/ipv4/tcp_autocorking lpq83:~# perf stat ./super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128 6624.89 Performance counter stats for './super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128': 40045.864494 task-clock # 3.301 CPUs utilized 171 context-switches # 0.004 K/sec 53 CPU-migrations # 0.001 K/sec 4,080 page-faults # 0.102 K/sec 111,340,458,645 cycles # 2.780 GHz [83.34%] 61,778,039,277 stalled-cycles-frontend # 55.49% frontend cycles idle [83.31%] 29,295,522,759 stalled-cycles-backend # 26.31% backend cycles idle [66.67%] 108,654,349,355 instructions # 0.98 insns per cycle # 0.57 stalled cycles per insn [83.34%] 19,552,170,748 branches # 488.244 M/sec [83.34%] 157,875,417 branch-misses # 0.81% of all branches [83.34%] 12.130267788 seconds time elapsed Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-05 22:36:05 -08:00
tcp: auto corking With the introduction of TCP Small Queues, TSO auto sizing, and TCP pacing, we can implement Automatic Corking in the kernel, to help applications doing small write()/sendmsg() to TCP sockets. Idea is to change tcp_push() to check if the current skb payload is under skb optimal size (a multiple of MSS bytes) If under 'size_goal', and at least one packet is still in Qdisc or NIC TX queues, set the TCP Small Queue Throttled bit, so that the push will be delayed up to TX completion time. This delay might allow the application to coalesce more bytes in the skb in following write()/sendmsg()/sendfile() system calls. The exact duration of the delay is depending on the dynamics of the system, and might be zero if no packet for this flow is actually held in Qdisc or NIC TX ring. Using FQ/pacing is a way to increase the probability of autocorking being triggered. Add a new sysctl (/proc/sys/net/ipv4/tcp_autocorking) to control this feature and default it to 1 (enabled) Add a new SNMP counter : nstat -a | grep TcpExtTCPAutoCorking This counter is incremented every time we detected skb was under used and its flush was deferred. Tested: Interesting effects when using line buffered commands under ssh. Excellent performance results in term of cpu usage and total throughput. lpq83:~# echo 1 >/proc/sys/net/ipv4/tcp_autocorking lpq83:~# perf stat ./super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128 9410.39 Performance counter stats for './super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128': 35209.439626 task-clock # 2.901 CPUs utilized 2,294 context-switches # 0.065 K/sec 101 CPU-migrations # 0.003 K/sec 4,079 page-faults # 0.116 K/sec 97,923,241,298 cycles # 2.781 GHz [83.31%] 51,832,908,236 stalled-cycles-frontend # 52.93% frontend cycles idle [83.30%] 25,697,986,603 stalled-cycles-backend # 26.24% backend cycles idle [66.70%] 102,225,978,536 instructions # 1.04 insns per cycle # 0.51 stalled cycles per insn [83.38%] 18,657,696,819 branches # 529.906 M/sec [83.29%] 91,679,646 branch-misses # 0.49% of all branches [83.40%] 12.136204899 seconds time elapsed lpq83:~# echo 0 >/proc/sys/net/ipv4/tcp_autocorking lpq83:~# perf stat ./super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128 6624.89 Performance counter stats for './super_netperf 4 -t TCP_STREAM -H lpq84 -- -m 128': 40045.864494 task-clock # 3.301 CPUs utilized 171 context-switches # 0.004 K/sec 53 CPU-migrations # 0.001 K/sec 4,080 page-faults # 0.102 K/sec 111,340,458,645 cycles # 2.780 GHz [83.34%] 61,778,039,277 stalled-cycles-frontend # 55.49% frontend cycles idle [83.31%] 29,295,522,759 stalled-cycles-backend # 26.31% backend cycles idle [66.67%] 108,654,349,355 instructions # 0.98 insns per cycle # 0.57 stalled cycles per insn [83.34%] 19,552,170,748 branches # 488.244 M/sec [83.34%] 157,875,417 branch-misses # 0.81% of all branches [83.34%] 12.130267788 seconds time elapsed Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-05 22:36:05 -08:00
tcp: early retransmit This patch implements RFC 5827 early retransmit (ER) for TCP. It reduces DUPACK threshold (dupthresh) if outstanding packets are less than 4 to recover losses by fast recovery instead of timeout. While the algorithm is simple, small but frequent network reordering makes this feature dangerous: the connection repeatedly enter false recovery and degrade performance. Therefore we implement a mitigation suggested in the appendix of the RFC that delays entering fast recovery by a small interval, i.e., RTT/4. Currently ER is conservative and is disabled for the rest of the connection after the first reordering event. A large scale web server experiment on the performance impact of ER is summarized in section 6 of the paper "Proportional Rate Reduction for TCP”, IMC 2011. http://conferences.sigcomm.org/imc/2011/docs/p155.pdf Note that Linux has a similar feature called THIN_DUPACK. The differences are THIN_DUPACK do not mitigate reorderings and is only used after slow start. Currently ER is disabled if THIN_DUPACK is enabled. I would be happy to merge THIN_DUPACK feature with ER if people think it's a good idea. ER is enabled by sysctl_tcp_early_retrans: 0: Disables ER 1: Reduce dupthresh to packets_out - 1 when outstanding packets < 4. 2: (Default) reduce dupthresh like mode 1. In addition, delay entering fast recovery by RTT/4. Note: mode 2 is implemented in the third part of this patch series. Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-02 13:30:03 +00:00
tcp: early retransmit This patch implements RFC 5827 early retransmit (ER) for TCP. It reduces DUPACK threshold (dupthresh) if outstanding packets are less than 4 to recover losses by fast recovery instead of timeout. While the algorithm is simple, small but frequent network reordering makes this feature dangerous: the connection repeatedly enter false recovery and degrade performance. Therefore we implement a mitigation suggested in the appendix of the RFC that delays entering fast recovery by a small interval, i.e., RTT/4. Currently ER is conservative and is disabled for the rest of the connection after the first reordering event. A large scale web server experiment on the performance impact of ER is summarized in section 6 of the paper "Proportional Rate Reduction for TCP”, IMC 2011. http://conferences.sigcomm.org/imc/2011/docs/p155.pdf Note that Linux has a similar feature called THIN_DUPACK. The differences are THIN_DUPACK do not mitigate reorderings and is only used after slow start. Currently ER is disabled if THIN_DUPACK is enabled. I would be happy to merge THIN_DUPACK feature with ER if people think it's a good idea. ER is enabled by sysctl_tcp_early_retrans: 0: Disables ER 1: Reduce dupthresh to packets_out - 1 when outstanding packets < 4. 2: (Default) reduce dupthresh like mode 1. In addition, delay entering fast recovery by RTT/4. Note: mode 2 is implemented in the third part of this patch series. Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-02 13:30:03 +00:00
tcp: Tail loss probe (TLP) This patch series implement the Tail loss probe (TLP) algorithm described in http://tools.ietf.org/html/draft-dukkipati-tcpm-tcp-loss-probe-01. The first patch implements the basic algorithm. TLP's goal is to reduce tail latency of short transactions. It achieves this by converting retransmission timeouts (RTOs) occuring due to tail losses (losses at end of transactions) into fast recovery. TLP transmits one packet in two round-trips when a connection is in Open state and isn't receiving any ACKs. The transmitted packet, aka loss probe, can be either new or a retransmission. When there is tail loss, the ACK from a loss probe triggers FACK/early-retransmit based fast recovery, thus avoiding a costly RTO. In the absence of loss, there is no change in the connection state. PTO stands for probe timeout. It is a timer event indicating that an ACK is overdue and triggers a loss probe packet. The PTO value is set to max(2*SRTT, 10ms) and is adjusted to account for delayed ACK timer when there is only one oustanding packet. TLP Algorithm On transmission of new data in Open state: -> packets_out > 1: schedule PTO in max(2*SRTT, 10ms). -> packets_out == 1: schedule PTO in max(2*RTT, 1.5*RTT + 200ms) -> PTO = min(PTO, RTO) Conditions for scheduling PTO: -> Connection is in Open state. -> Connection is either cwnd limited or no new data to send. -> Number of probes per tail loss episode is limited to one. -> Connection is SACK enabled. When PTO fires: new_segment_exists: -> transmit new segment. -> packets_out++. cwnd remains same. no_new_packet: -> retransmit the last segment. Its ACK triggers FACK or early retransmit based recovery. ACK path: -> rearm RTO at start of ACK processing. -> reschedule PTO if need be. In addition, the patch includes a small variation to the Early Retransmit (ER) algorithm, such that ER and TLP together can in principle recover any N-degree of tail loss through fast recovery. TLP is controlled by the same sysctl as ER, tcp_early_retrans sysctl. tcp_early_retrans==0; disables TLP and ER. ==1; enables RFC5827 ER. ==2; delayed ER. ==3; TLP and delayed ER. [DEFAULT] ==4; TLP only. The TLP patch series have been extensively tested on Google Web servers. It is most effective for short Web trasactions, where it reduced RTOs by 15% and improved HTTP response time (average by 6%, 99th percentile by 10%). The transmitted probes account for <0.5% of the overall transmissions. Signed-off-by: Nandita Dukkipati <nanditad@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-11 10:00:43 +00:00
tcp: early retransmit This patch implements RFC 5827 early retransmit (ER) for TCP. It reduces DUPACK threshold (dupthresh) if outstanding packets are less than 4 to recover losses by fast recovery instead of timeout. While the algorithm is simple, small but frequent network reordering makes this feature dangerous: the connection repeatedly enter false recovery and degrade performance. Therefore we implement a mitigation suggested in the appendix of the RFC that delays entering fast recovery by a small interval, i.e., RTT/4. Currently ER is conservative and is disabled for the rest of the connection after the first reordering event. A large scale web server experiment on the performance impact of ER is summarized in section 6 of the paper "Proportional Rate Reduction for TCP”, IMC 2011. http://conferences.sigcomm.org/imc/2011/docs/p155.pdf Note that Linux has a similar feature called THIN_DUPACK. The differences are THIN_DUPACK do not mitigate reorderings and is only used after slow start. Currently ER is disabled if THIN_DUPACK is enabled. I would be happy to merge THIN_DUPACK feature with ER if people think it's a good idea. ER is enabled by sysctl_tcp_early_retrans: 0: Disables ER 1: Reduce dupthresh to packets_out - 1 when outstanding packets < 4. 2: (Default) reduce dupthresh like mode 1. In addition, delay entering fast recovery by RTT/4. Note: mode 2 is implemented in the third part of this patch series. Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-02 13:30:03 +00:00
tcp: accecn: AccECN negotiation Accurate ECN negotiation parts based on the specification: https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt Accurate ECN is negotiated using ECE, CWR and AE flags in the TCP header. TCP falls back into using RFC3168 ECN if one of the ends supports only RFC3168-style ECN. The AccECN negotiation includes reflecting IP ECN field value seen in SYN and SYNACK back using the same bits as negotiation to allow responding to SYN CE marks and to detect ECN field mangling. CE marks should not occur currently because SYN=1 segments are sent with Non-ECT in IP ECN field (but proposal exists to remove this restriction). Reflecting SYN IP ECN field in SYNACK is relatively simple. Reflecting SYNACK IP ECN field in the final/third ACK of the handshake is more challenging. Linux TCP code is not well prepared for using the final/third ACK a signalling channel which makes things somewhat complicated here. tcp_ecn sysctl can be used to select the highest ECN variant (Accurate ECN, ECN, No ECN) that is attemped to be negotiated and requested for incoming connection and outgoing connection: TCP_ECN_IN_NOECN_OUT_NOECN, TCP_ECN_IN_ECN_OUT_ECN, TCP_ECN_IN_ECN_OUT_NOECN, TCP_ECN_IN_ACCECN_OUT_ACCECN, TCP_ECN_IN_ACCECN_OUT_ECN, and TCP_ECN_IN_ACCECN_OUT_NOECN. After this patch, the size of tcp_request_sock remains unchanged and no new holes are added. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_request_sock { [...] u32 rcv_nxt; /* 352 4 */ u8 syn_tos; /* 356 1 */ /* size: 360, cachelines: 6, members: 16 */ } [AFTER THIS PATCH] struct tcp_request_sock { [...] u32 rcv_nxt; /* 352 4 */ u8 syn_tos; /* 356 1 */ bool accecn_ok; /* 357 1 */ u8 syn_ect_snt:2; /* 358: 0 1 */ u8 syn_ect_rcv:2; /* 358: 2 1 */ u8 accecn_fail_mode:4; /* 358: 4 1 */ /* size: 360, cachelines: 6, members: 20 */ } After this patch, the size of tcp_sock remains unchanged and no new holes are added. Also, 4 bits of the existing 2-byte hole are exploited. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_sock { [...] u8 dup_ack_counter:2; /* 2761: 0 1 */ u8 tlp_retrans:1; /* 2761: 2 1 */ u8 unused:5; /* 2761: 3 1 */ u8 thin_lto:1; /* 2762: 0 1 */ u8 fastopen_connect:1; /* 2762: 1 1 */ u8 fastopen_no_cookie:1; /* 2762: 2 1 */ u8 fastopen_client_fail:2; /* 2762: 3 1 */ u8 frto:1; /* 2762: 5 1 */ /* XXX 2 bits hole, try to pack */ [...] u8 keepalive_probes; /* 2765 1 */ /* XXX 2 bytes hole, try to pack */ [...] /* size: 3200, cachelines: 50, members: 164 */ } [AFTER THIS PATCH] struct tcp_sock { [...] u8 dup_ack_counter:2; /* 2761: 0 1 */ u8 tlp_retrans:1; /* 2761: 2 1 */ u8 syn_ect_snt:2; /* 2761: 3 1 */ u8 syn_ect_rcv:2; /* 2761: 5 1 */ u8 thin_lto:1; /* 2761: 7 1 */ u8 fastopen_connect:1; /* 2762: 0 1 */ u8 fastopen_no_cookie:1; /* 2762: 1 1 */ u8 fastopen_client_fail:2; /* 2762: 2 1 */ u8 frto:1; /* 2762: 4 1 */ /* XXX 3 bits hole, try to pack */ [...] u8 keepalive_probes; /* 2765 1 */ u8 accecn_fail_mode:4; /* 2766: 0 1 */ /* XXX 4 bits hole, try to pack */ /* XXX 1 byte hole, try to pack */ [...] /* size: 3200, cachelines: 50, members: 166 */ } Signed-off-by: Ilpo Järvinen <ij@kernel.org> Co-developed-by: Olivier Tilmans <olivier.tilmans@nokia.com> Signed-off-by: Olivier Tilmans <olivier.tilmans@nokia.com> Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250916082434.100722-3-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-09-16 10:24:26 +02:00
tcp: accecn: AccECN negotiation Accurate ECN negotiation parts based on the specification: https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt Accurate ECN is negotiated using ECE, CWR and AE flags in the TCP header. TCP falls back into using RFC3168 ECN if one of the ends supports only RFC3168-style ECN. The AccECN negotiation includes reflecting IP ECN field value seen in SYN and SYNACK back using the same bits as negotiation to allow responding to SYN CE marks and to detect ECN field mangling. CE marks should not occur currently because SYN=1 segments are sent with Non-ECT in IP ECN field (but proposal exists to remove this restriction). Reflecting SYN IP ECN field in SYNACK is relatively simple. Reflecting SYNACK IP ECN field in the final/third ACK of the handshake is more challenging. Linux TCP code is not well prepared for using the final/third ACK a signalling channel which makes things somewhat complicated here. tcp_ecn sysctl can be used to select the highest ECN variant (Accurate ECN, ECN, No ECN) that is attemped to be negotiated and requested for incoming connection and outgoing connection: TCP_ECN_IN_NOECN_OUT_NOECN, TCP_ECN_IN_ECN_OUT_ECN, TCP_ECN_IN_ECN_OUT_NOECN, TCP_ECN_IN_ACCECN_OUT_ACCECN, TCP_ECN_IN_ACCECN_OUT_ECN, and TCP_ECN_IN_ACCECN_OUT_NOECN. After this patch, the size of tcp_request_sock remains unchanged and no new holes are added. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_request_sock { [...] u32 rcv_nxt; /* 352 4 */ u8 syn_tos; /* 356 1 */ /* size: 360, cachelines: 6, members: 16 */ } [AFTER THIS PATCH] struct tcp_request_sock { [...] u32 rcv_nxt; /* 352 4 */ u8 syn_tos; /* 356 1 */ bool accecn_ok; /* 357 1 */ u8 syn_ect_snt:2; /* 358: 0 1 */ u8 syn_ect_rcv:2; /* 358: 2 1 */ u8 accecn_fail_mode:4; /* 358: 4 1 */ /* size: 360, cachelines: 6, members: 20 */ } After this patch, the size of tcp_sock remains unchanged and no new holes are added. Also, 4 bits of the existing 2-byte hole are exploited. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_sock { [...] u8 dup_ack_counter:2; /* 2761: 0 1 */ u8 tlp_retrans:1; /* 2761: 2 1 */ u8 unused:5; /* 2761: 3 1 */ u8 thin_lto:1; /* 2762: 0 1 */ u8 fastopen_connect:1; /* 2762: 1 1 */ u8 fastopen_no_cookie:1; /* 2762: 2 1 */ u8 fastopen_client_fail:2; /* 2762: 3 1 */ u8 frto:1; /* 2762: 5 1 */ /* XXX 2 bits hole, try to pack */ [...] u8 keepalive_probes; /* 2765 1 */ /* XXX 2 bytes hole, try to pack */ [...] /* size: 3200, cachelines: 50, members: 164 */ } [AFTER THIS PATCH] struct tcp_sock { [...] u8 dup_ack_counter:2; /* 2761: 0 1 */ u8 tlp_retrans:1; /* 2761: 2 1 */ u8 syn_ect_snt:2; /* 2761: 3 1 */ u8 syn_ect_rcv:2; /* 2761: 5 1 */ u8 thin_lto:1; /* 2761: 7 1 */ u8 fastopen_connect:1; /* 2762: 0 1 */ u8 fastopen_no_cookie:1; /* 2762: 1 1 */ u8 fastopen_client_fail:2; /* 2762: 2 1 */ u8 frto:1; /* 2762: 4 1 */ /* XXX 3 bits hole, try to pack */ [...] u8 keepalive_probes; /* 2765 1 */ u8 accecn_fail_mode:4; /* 2766: 0 1 */ /* XXX 4 bits hole, try to pack */ /* XXX 1 byte hole, try to pack */ [...] /* size: 3200, cachelines: 50, members: 166 */ } Signed-off-by: Ilpo Järvinen <ij@kernel.org> Co-developed-by: Olivier Tilmans <olivier.tilmans@nokia.com> Signed-off-by: Olivier Tilmans <olivier.tilmans@nokia.com> Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250916082434.100722-3-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-09-16 10:24:26 +02:00
tcp: accecn: AccECN negotiation Accurate ECN negotiation parts based on the specification: https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt Accurate ECN is negotiated using ECE, CWR and AE flags in the TCP header. TCP falls back into using RFC3168 ECN if one of the ends supports only RFC3168-style ECN. The AccECN negotiation includes reflecting IP ECN field value seen in SYN and SYNACK back using the same bits as negotiation to allow responding to SYN CE marks and to detect ECN field mangling. CE marks should not occur currently because SYN=1 segments are sent with Non-ECT in IP ECN field (but proposal exists to remove this restriction). Reflecting SYN IP ECN field in SYNACK is relatively simple. Reflecting SYNACK IP ECN field in the final/third ACK of the handshake is more challenging. Linux TCP code is not well prepared for using the final/third ACK a signalling channel which makes things somewhat complicated here. tcp_ecn sysctl can be used to select the highest ECN variant (Accurate ECN, ECN, No ECN) that is attemped to be negotiated and requested for incoming connection and outgoing connection: TCP_ECN_IN_NOECN_OUT_NOECN, TCP_ECN_IN_ECN_OUT_ECN, TCP_ECN_IN_ECN_OUT_NOECN, TCP_ECN_IN_ACCECN_OUT_ACCECN, TCP_ECN_IN_ACCECN_OUT_ECN, and TCP_ECN_IN_ACCECN_OUT_NOECN. After this patch, the size of tcp_request_sock remains unchanged and no new holes are added. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_request_sock { [...] u32 rcv_nxt; /* 352 4 */ u8 syn_tos; /* 356 1 */ /* size: 360, cachelines: 6, members: 16 */ } [AFTER THIS PATCH] struct tcp_request_sock { [...] u32 rcv_nxt; /* 352 4 */ u8 syn_tos; /* 356 1 */ bool accecn_ok; /* 357 1 */ u8 syn_ect_snt:2; /* 358: 0 1 */ u8 syn_ect_rcv:2; /* 358: 2 1 */ u8 accecn_fail_mode:4; /* 358: 4 1 */ /* size: 360, cachelines: 6, members: 20 */ } After this patch, the size of tcp_sock remains unchanged and no new holes are added. Also, 4 bits of the existing 2-byte hole are exploited. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_sock { [...] u8 dup_ack_counter:2; /* 2761: 0 1 */ u8 tlp_retrans:1; /* 2761: 2 1 */ u8 unused:5; /* 2761: 3 1 */ u8 thin_lto:1; /* 2762: 0 1 */ u8 fastopen_connect:1; /* 2762: 1 1 */ u8 fastopen_no_cookie:1; /* 2762: 2 1 */ u8 fastopen_client_fail:2; /* 2762: 3 1 */ u8 frto:1; /* 2762: 5 1 */ /* XXX 2 bits hole, try to pack */ [...] u8 keepalive_probes; /* 2765 1 */ /* XXX 2 bytes hole, try to pack */ [...] /* size: 3200, cachelines: 50, members: 164 */ } [AFTER THIS PATCH] struct tcp_sock { [...] u8 dup_ack_counter:2; /* 2761: 0 1 */ u8 tlp_retrans:1; /* 2761: 2 1 */ u8 syn_ect_snt:2; /* 2761: 3 1 */ u8 syn_ect_rcv:2; /* 2761: 5 1 */ u8 thin_lto:1; /* 2761: 7 1 */ u8 fastopen_connect:1; /* 2762: 0 1 */ u8 fastopen_no_cookie:1; /* 2762: 1 1 */ u8 fastopen_client_fail:2; /* 2762: 2 1 */ u8 frto:1; /* 2762: 4 1 */ /* XXX 3 bits hole, try to pack */ [...] u8 keepalive_probes; /* 2765 1 */ u8 accecn_fail_mode:4; /* 2766: 0 1 */ /* XXX 4 bits hole, try to pack */ /* XXX 1 byte hole, try to pack */ [...] /* size: 3200, cachelines: 50, members: 166 */ } Signed-off-by: Ilpo Järvinen <ij@kernel.org> Co-developed-by: Olivier Tilmans <olivier.tilmans@nokia.com> Signed-off-by: Olivier Tilmans <olivier.tilmans@nokia.com> Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250916082434.100722-3-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-09-16 10:24:26 +02:00
tcp: accecn: AccECN option The Accurate ECN allows echoing back the sum of bytes for each IP ECN field value in the received packets using AccECN option. This change implements AccECN option tx & rx side processing without option send control related features that are added by a later change. Based on specification: https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt (Some features of the spec will be added in the later changes rather than in this one). A full-length AccECN option is always attempted but if it does not fit, the minimum length is selected based on the counters that have changed since the last update. The AccECN option (with 24-bit fields) often ends in odd sizes so the option write code tries to take advantage of some nop used to pad the other TCP options. The delivered_ecn_bytes pairs with received_ecn_bytes similar to how delivered_ce pairs with received_ce. In contrast to ACE field, however, the option is not always available to update delivered_ecn_bytes. For ACK w/o AccECN option, the delivered bytes calculated based on the cumulative ACK+SACK information are assigned to one of the counters using an estimation heuristic to select the most likely ECN byte counter. Any estimation error is corrected when the next AccECN option arrives. It may occur that the heuristic gets too confused when there are enough different byte counter deltas between ACKs with the AccECN option in which case the heuristic just gives up on updating the counters for a while. tcp_ecn_option sysctl can be used to select option sending mode for AccECN: TCP_ECN_OPTION_DISABLED, TCP_ECN_OPTION_MINIMUM, and TCP_ECN_OPTION_FULL. This patch increases the size of tcp_info struct, as there is no existing holes for new u32 variables. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_info { [...] __u32 tcpi_total_rto_time; /* 244 4 */ /* size: 248, cachelines: 4, members: 61 */ } [AFTER THIS PATCH] struct tcp_info { [...] __u32 tcpi_total_rto_time; /* 244 4 */ __u32 tcpi_received_ce; /* 248 4 */ __u32 tcpi_delivered_e1_bytes; /* 252 4 */ __u32 tcpi_delivered_e0_bytes; /* 256 4 */ __u32 tcpi_delivered_ce_bytes; /* 260 4 */ __u32 tcpi_received_e1_bytes; /* 264 4 */ __u32 tcpi_received_e0_bytes; /* 268 4 */ __u32 tcpi_received_ce_bytes; /* 272 4 */ /* size: 280, cachelines: 5, members: 68 */ } This patch uses the existing 1-byte holes in the tcp_sock_write_txrx group for new u8 members, but adds a 4-byte hole in tcp_sock_write_rx group after the new u32 delivered_ecn_bytes[3] member. Therefore, the group size of tcp_sock_write_rx is increased from 96 to 112. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_sock { [...] u8 received_ce_pending:4; /* 2522: 0 1 */ u8 unused2:4; /* 2522: 4 1 */ /* XXX 1 byte hole, try to pack */ [...] u32 rcv_rtt_last_tsecr; /* 2668 4 */ [...] __cacheline_group_end__tcp_sock_write_rx[0]; /* 2728 0 */ [...] /* size: 3200, cachelines: 50, members: 167 */ } [AFTER THIS PATCH] struct tcp_sock { [...] u8 received_ce_pending:4;/* 2522: 0 1 */ u8 unused2:4; /* 2522: 4 1 */ u8 accecn_minlen:2; /* 2523: 0 1 */ u8 est_ecnfield:2; /* 2523: 2 1 */ u8 unused3:4; /* 2523: 4 1 */ [...] u32 rcv_rtt_last_tsecr; /* 2668 4 */ u32 delivered_ecn_bytes[3];/* 2672 12 */ /* XXX 4 bytes hole, try to pack */ [...] __cacheline_group_end__tcp_sock_write_rx[0]; /* 2744 0 */ [...] /* size: 3200, cachelines: 50, members: 171 */ } Signed-off-by: Ilpo Järvinen <ij@kernel.org> Signed-off-by: Neal Cardwell <ncardwell@google.com> Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250916082434.100722-7-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-09-16 10:24:30 +02:00
tcp: accecn: AccECN option The Accurate ECN allows echoing back the sum of bytes for each IP ECN field value in the received packets using AccECN option. This change implements AccECN option tx & rx side processing without option send control related features that are added by a later change. Based on specification: https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt (Some features of the spec will be added in the later changes rather than in this one). A full-length AccECN option is always attempted but if it does not fit, the minimum length is selected based on the counters that have changed since the last update. The AccECN option (with 24-bit fields) often ends in odd sizes so the option write code tries to take advantage of some nop used to pad the other TCP options. The delivered_ecn_bytes pairs with received_ecn_bytes similar to how delivered_ce pairs with received_ce. In contrast to ACE field, however, the option is not always available to update delivered_ecn_bytes. For ACK w/o AccECN option, the delivered bytes calculated based on the cumulative ACK+SACK information are assigned to one of the counters using an estimation heuristic to select the most likely ECN byte counter. Any estimation error is corrected when the next AccECN option arrives. It may occur that the heuristic gets too confused when there are enough different byte counter deltas between ACKs with the AccECN option in which case the heuristic just gives up on updating the counters for a while. tcp_ecn_option sysctl can be used to select option sending mode for AccECN: TCP_ECN_OPTION_DISABLED, TCP_ECN_OPTION_MINIMUM, and TCP_ECN_OPTION_FULL. This patch increases the size of tcp_info struct, as there is no existing holes for new u32 variables. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_info { [...] __u32 tcpi_total_rto_time; /* 244 4 */ /* size: 248, cachelines: 4, members: 61 */ } [AFTER THIS PATCH] struct tcp_info { [...] __u32 tcpi_total_rto_time; /* 244 4 */ __u32 tcpi_received_ce; /* 248 4 */ __u32 tcpi_delivered_e1_bytes; /* 252 4 */ __u32 tcpi_delivered_e0_bytes; /* 256 4 */ __u32 tcpi_delivered_ce_bytes; /* 260 4 */ __u32 tcpi_received_e1_bytes; /* 264 4 */ __u32 tcpi_received_e0_bytes; /* 268 4 */ __u32 tcpi_received_ce_bytes; /* 272 4 */ /* size: 280, cachelines: 5, members: 68 */ } This patch uses the existing 1-byte holes in the tcp_sock_write_txrx group for new u8 members, but adds a 4-byte hole in tcp_sock_write_rx group after the new u32 delivered_ecn_bytes[3] member. Therefore, the group size of tcp_sock_write_rx is increased from 96 to 112. Below are the pahole outcomes before and after this patch: [BEFORE THIS PATCH] struct tcp_sock { [...] u8 received_ce_pending:4; /* 2522: 0 1 */ u8 unused2:4; /* 2522: 4 1 */ /* XXX 1 byte hole, try to pack */ [...] u32 rcv_rtt_last_tsecr; /* 2668 4 */ [...] __cacheline_group_end__tcp_sock_write_rx[0]; /* 2728 0 */ [...] /* size: 3200, cachelines: 50, members: 167 */ } [AFTER THIS PATCH] struct tcp_sock { [...] u8 received_ce_pending:4;/* 2522: 0 1 */ u8 unused2:4; /* 2522: 4 1 */ u8 accecn_minlen:2; /* 2523: 0 1 */ u8 est_ecnfield:2; /* 2523: 2 1 */ u8 unused3:4; /* 2523: 4 1 */ [...] u32 rcv_rtt_last_tsecr; /* 2668 4 */ u32 delivered_ecn_bytes[3];/* 2672 12 */ /* XXX 4 bytes hole, try to pack */ [...] __cacheline_group_end__tcp_sock_write_rx[0]; /* 2744 0 */ [...] /* size: 3200, cachelines: 50, members: 171 */ } Signed-off-by: Ilpo Järvinen <ij@kernel.org> Signed-off-by: Neal Cardwell <ncardwell@google.com> Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250916082434.100722-7-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-09-16 10:24:30 +02:00
tcp: accecn: AccECN option send control Instead of sending the option in every ACK, limit sending to those ACKs where the option is necessary: - Handshake - "Change-triggered ACK" + the ACK following it. The 2nd ACK is necessary to unambiguously indicate which of the ECN byte counters in increasing. The first ACK has two counters increasing due to the ecnfield edge. - ACKs with CE to allow CEP delta validations to take advantage of the option. - Force option to be sent every at least once per 2^22 bytes. The check is done using the bit edges of the byte counters (avoids need for extra variables). - AccECN option beacon to send a few times per RTT even if nothing in the ECN state requires that. The default is 3 times per RTT, and its period can be set via sysctl_tcp_ecn_option_beacon. Below are the pahole outcomes before and after this patch, in which the group size of tcp_sock_write_tx is increased from 89 to 97 due to the new u64 accecn_opt_tstamp member: [BEFORE THIS PATCH] struct tcp_sock { [...] u64 tcp_wstamp_ns; /* 2488 8 */ struct list_head tsorted_sent_queue; /* 2496 16 */ [...] __cacheline_group_end__tcp_sock_write_tx[0]; /* 2521 0 */ __cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2521 0 */ u8 nonagle:4; /* 2521: 0 1 */ u8 rate_app_limited:1; /* 2521: 4 1 */ /* XXX 3 bits hole, try to pack */ /* Force alignment to the next boundary: */ u8 :0; u8 received_ce_pending:4;/* 2522: 0 1 */ u8 unused2:4; /* 2522: 4 1 */ u8 accecn_minlen:2; /* 2523: 0 1 */ u8 est_ecnfield:2; /* 2523: 2 1 */ u8 unused3:4; /* 2523: 4 1 */ [...] __cacheline_group_end__tcp_sock_write_txrx[0]; /* 2628 0 */ [...] /* size: 3200, cachelines: 50, members: 171 */ } [AFTER THIS PATCH] struct tcp_sock { [...] u64 tcp_wstamp_ns; /* 2488 8 */ u64 accecn_opt_tstamp; /* 2596 8 */ struct list_head tsorted_sent_queue; /* 2504 16 */ [...] __cacheline_group_end__tcp_sock_write_tx[0]; /* 2529 0 */ __cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2529 0 */ u8 nonagle:4; /* 2529: 0 1 */ u8 rate_app_limited:1; /* 2529: 4 1 */ /* XXX 3 bits hole, try to pack */ /* Force alignment to the next boundary: */ u8 :0; u8 received_ce_pending:4;/* 2530: 0 1 */ u8 unused2:4; /* 2530: 4 1 */ u8 accecn_minlen:2; /* 2531: 0 1 */ u8 est_ecnfield:2; /* 2531: 2 1 */ u8 accecn_opt_demand:2; /* 2531: 4 1 */ u8 prev_ecnfield:2; /* 2531: 6 1 */ [...] __cacheline_group_end__tcp_sock_write_txrx[0]; /* 2636 0 */ [...] /* size: 3200, cachelines: 50, members: 173 */ } Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> Co-developed-by: Ilpo Järvinen <ij@kernel.org> Signed-off-by: Ilpo Järvinen <ij@kernel.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250916082434.100722-8-chia-yu.chang@nokia-bell-labs.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2025-09-16 10:24:31 +02:00
tcp: add rfc3168, section 6.1.1.1. fallback This work as a follow-up of commit f7b3bec6f516 ("net: allow setting ecn via routing table") and adds RFC3168 section 6.1.1.1. fallback for outgoing ECN connections. In other words, this work adds a retry with a non-ECN setup SYN packet, as suggested from the RFC on the first timeout: [...] A host that receives no reply to an ECN-setup SYN within the normal SYN retransmission timeout interval MAY resend the SYN and any subsequent SYN retransmissions with CWR and ECE cleared. [...] Schematic client-side view when assuming the server is in tcp_ecn=2 mode, that is, Linux default since 2009 via commit 255cac91c3c9 ("tcp: extend ECN sysctl to allow server-side only ECN"): 1) Normal ECN-capable path: SYN ECE CWR -----> <----- SYN ACK ECE ACK -----> 2) Path with broken middlebox, when client has fallback: SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN -----> <----- SYN ACK ACK -----> In case we would not have the fallback implemented, the middlebox drop point would basically end up as: SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) In any case, it's rather a smaller percentage of sites where there would occur such additional setup latency: it was found in end of 2014 that ~56% of IPv4 and 65% of IPv6 servers of Alexa 1 million list would negotiate ECN (aka tcp_ecn=2 default), 0.42% of these webservers will fail to connect when trying to negotiate with ECN (tcp_ecn=1) due to timeouts, which the fallback would mitigate with a slight latency trade-off. Recent related paper on this topic: Brian Trammell, Mirja Kühlewind, Damiano Boppart, Iain Learmonth, Gorry Fairhurst, and Richard Scheffenegger: "Enabling Internet-Wide Deployment of Explicit Congestion Notification." Proc. PAM 2015, New York. http://ecn.ethz.ch/ecn-pam15.pdf Thus, when net.ipv4.tcp_ecn=1 is being set, the patch will perform RFC3168, section 6.1.1.1. fallback on timeout. For users explicitly not wanting this which can be in DC use case, we add a net.ipv4.tcp_ecn_fallback knob that allows for disabling the fallback. tp->ecn_flags are not being cleared in tcp_ecn_clear_syn() on output, but rather we let tcp_ecn_rcv_synack() take that over on input path in case a SYN ACK ECE was delayed. Thus a spurious SYN retransmission will not prevent ECN being negotiated eventually in that case. Reference: https://www.ietf.org/proceedings/92/slides/slides-92-iccrg-1.pdf Reference: https://www.ietf.org/proceedings/89/slides/slides-89-tsvarea-1.pdf Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mirja Kühlewind <mirja.kuehlewind@tik.ee.ethz.ch> Signed-off-by: Brian Trammell <trammell@tik.ee.ethz.ch> Cc: Eric Dumazet <edumazet@google.com> Cc: Dave That <dave.taht@gmail.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-19 21:04:22 +02:00
tcp: add rfc3168, section 6.1.1.1. fallback This work as a follow-up of commit f7b3bec6f516 ("net: allow setting ecn via routing table") and adds RFC3168 section 6.1.1.1. fallback for outgoing ECN connections. In other words, this work adds a retry with a non-ECN setup SYN packet, as suggested from the RFC on the first timeout: [...] A host that receives no reply to an ECN-setup SYN within the normal SYN retransmission timeout interval MAY resend the SYN and any subsequent SYN retransmissions with CWR and ECE cleared. [...] Schematic client-side view when assuming the server is in tcp_ecn=2 mode, that is, Linux default since 2009 via commit 255cac91c3c9 ("tcp: extend ECN sysctl to allow server-side only ECN"): 1) Normal ECN-capable path: SYN ECE CWR -----> <----- SYN ACK ECE ACK -----> 2) Path with broken middlebox, when client has fallback: SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN -----> <----- SYN ACK ACK -----> In case we would not have the fallback implemented, the middlebox drop point would basically end up as: SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) In any case, it's rather a smaller percentage of sites where there would occur such additional setup latency: it was found in end of 2014 that ~56% of IPv4 and 65% of IPv6 servers of Alexa 1 million list would negotiate ECN (aka tcp_ecn=2 default), 0.42% of these webservers will fail to connect when trying to negotiate with ECN (tcp_ecn=1) due to timeouts, which the fallback would mitigate with a slight latency trade-off. Recent related paper on this topic: Brian Trammell, Mirja Kühlewind, Damiano Boppart, Iain Learmonth, Gorry Fairhurst, and Richard Scheffenegger: "Enabling Internet-Wide Deployment of Explicit Congestion Notification." Proc. PAM 2015, New York. http://ecn.ethz.ch/ecn-pam15.pdf Thus, when net.ipv4.tcp_ecn=1 is being set, the patch will perform RFC3168, section 6.1.1.1. fallback on timeout. For users explicitly not wanting this which can be in DC use case, we add a net.ipv4.tcp_ecn_fallback knob that allows for disabling the fallback. tp->ecn_flags are not being cleared in tcp_ecn_clear_syn() on output, but rather we let tcp_ecn_rcv_synack() take that over on input path in case a SYN ACK ECE was delayed. Thus a spurious SYN retransmission will not prevent ECN being negotiated eventually in that case. Reference: https://www.ietf.org/proceedings/92/slides/slides-92-iccrg-1.pdf Reference: https://www.ietf.org/proceedings/89/slides/slides-89-tsvarea-1.pdf Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mirja Kühlewind <mirja.kuehlewind@tik.ee.ethz.ch> Signed-off-by: Brian Trammell <trammell@tik.ee.ethz.ch> Cc: Eric Dumazet <edumazet@google.com> Cc: Dave That <dave.taht@gmail.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-19 21:04:22 +02:00
tcp: implement RFC5682 F-RTO This patch implements F-RTO (foward RTO recovery): When the first retransmission after timeout is acknowledged, F-RTO sends new data instead of old data. If the next ACK acknowledges some never-retransmitted data, then the timeout was spurious and the congestion state is reverted. Otherwise if the next ACK selectively acknowledges the new data, then the timeout was genuine and the loss recovery continues. This idea applies to recurring timeouts as well. While F-RTO sends different data during timeout recovery, it does not (and should not) change the congestion control. The implementaion follows the three steps of SACK enhanced algorithm (section 3) in RFC5682. Step 1 is in tcp_enter_loss(). Step 2 and 3 are in tcp_process_loss(). The basic version is not supported because SACK enhanced version also works for non-SACK connections. The new implementation is functionally in parity with the old F-RTO implementation except the one case where it increases undo events: In addition to the RFC algorithm, a spurious timeout may be detected without sending data in step 2, as long as the SACK confirms not all the original data are dropped. When this happens, the sender will undo the cwnd and perhaps enter fast recovery instead. This additional check increases the F-RTO undo events by 5x compared to the prior implementation on Google Web servers, since the sender often does not have new data to send for HTTP. Note F-RTO may detect spurious timeout before Eifel with timestamps does so. Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-20 13:33:00 +00:00
tcp: implement RFC5682 F-RTO This patch implements F-RTO (foward RTO recovery): When the first retransmission after timeout is acknowledged, F-RTO sends new data instead of old data. If the next ACK acknowledges some never-retransmitted data, then the timeout was spurious and the congestion state is reverted. Otherwise if the next ACK selectively acknowledges the new data, then the timeout was genuine and the loss recovery continues. This idea applies to recurring timeouts as well. While F-RTO sends different data during timeout recovery, it does not (and should not) change the congestion control. The implementaion follows the three steps of SACK enhanced algorithm (section 3) in RFC5682. Step 1 is in tcp_enter_loss(). Step 2 and 3 are in tcp_process_loss(). The basic version is not supported because SACK enhanced version also works for non-SACK connections. The new implementation is functionally in parity with the old F-RTO implementation except the one case where it increases undo events: In addition to the RFC algorithm, a spurious timeout may be detected without sending data in step 2, as long as the SACK confirms not all the original data are dropped. When this happens, the sender will undo the cwnd and perhaps enter fast recovery instead. This additional check increases the F-RTO undo events by 5x compared to the prior implementation on Google Web servers, since the sender often does not have new data to send for HTTP. Note F-RTO may detect spurious timeout before Eifel with timestamps does so. Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-20 13:33:00 +00:00
tcp: implement RFC5682 F-RTO This patch implements F-RTO (foward RTO recovery): When the first retransmission after timeout is acknowledged, F-RTO sends new data instead of old data. If the next ACK acknowledges some never-retransmitted data, then the timeout was spurious and the congestion state is reverted. Otherwise if the next ACK selectively acknowledges the new data, then the timeout was genuine and the loss recovery continues. This idea applies to recurring timeouts as well. While F-RTO sends different data during timeout recovery, it does not (and should not) change the congestion control. The implementaion follows the three steps of SACK enhanced algorithm (section 3) in RFC5682. Step 1 is in tcp_enter_loss(). Step 2 and 3 are in tcp_process_loss(). The basic version is not supported because SACK enhanced version also works for non-SACK connections. The new implementation is functionally in parity with the old F-RTO implementation except the one case where it increases undo events: In addition to the RFC algorithm, a spurious timeout may be detected without sending data in step 2, as long as the SACK confirms not all the original data are dropped. When this happens, the sender will undo the cwnd and perhaps enter fast recovery instead. This additional check increases the F-RTO undo events by 5x compared to the prior implementation on Google Web servers, since the sender often does not have new data to send for HTTP. Note F-RTO may detect spurious timeout before Eifel with timestamps does so. Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-20 13:33:00 +00:00
tcp: helpers to mitigate ACK loops by rate-limiting out-of-window dupacks Helpers for mitigating ACK loops by rate-limiting dupacks sent in response to incoming out-of-window packets. This patch includes: - rate-limiting logic - sysctl to control how often we allow dupacks to out-of-window packets - SNMP counter for cases where we rate-limited our dupack sending The rate-limiting logic in this patch decides to not send dupacks in response to out-of-window segments if (a) they are SYNs or pure ACKs and (b) the remote endpoint is sending them faster than the configured rate limit. We rate-limit our responses rather than blocking them entirely or resetting the connection, because legitimate connections can rely on dupacks in response to some out-of-window segments. For example, zero window probes are typically sent with a sequence number that is below the current window, and ZWPs thus expect to thus elicit a dupack in response. We allow dupacks in response to TCP segments with data, because these may be spurious retransmissions for which the remote endpoint wants to receive DSACKs. This is safe because segments with data can't realistically be part of ACK loops, which by their nature consist of each side sending pure/data-less ACKs to each other. The dupack interval is controlled by a new sysctl knob, tcp_invalid_ratelimit, given in milliseconds, in case an administrator needs to dial this upward in the face of a high-rate DoS attack. The name and units are chosen to be analogous to the existing analogous knob for ICMP, icmp_ratelimit. The default value for tcp_invalid_ratelimit is 500ms, which allows at most one such dupack per 500ms. This is chosen to be 2x faster than the 1-second minimum RTO interval allowed by RFC 6298 (section 2, rule 2.4). We allow the extra 2x factor because network delay variations can cause packets sent at 1 second intervals to be compressed and arrive much closer. Reported-by: Avery Fay <avery@mixpanel.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-06 16:04:38 -05:00
tcp: track min RTT using windowed min-filter Kathleen Nichols' algorithm for tracking the minimum RTT of a data stream over some measurement window. It uses constant space and constant time per update. Yet it almost always delivers the same minimum as an implementation that has to keep all the data in the window. The measurement window is tunable via sysctl.net.ipv4.tcp_min_rtt_wlen with a default value of 5 minutes. The algorithm keeps track of the best, 2nd best & 3rd best min values, maintaining an invariant that the measurement time of the n'th best >= n-1'th best. It also makes sure that the three values are widely separated in the time window since that bounds the worse case error when that data is monotonically increasing over the window. Upon getting a new min, we can forget everything earlier because it has no value - the new min is less than everything else in the window by definition and it's the most recent. So we restart fresh on every new min and overwrites the 2nd & 3rd choices. The same property holds for the 2nd & 3rd best. Therefore we have to maintain two invariants to maximize the information in the samples, one on values (1st.v <= 2nd.v <= 3rd.v) and the other on times (now-win <=1st.t <= 2nd.t <= 3rd.t <= now). These invariants determine the structure of the code The RTT input to the windowed filter is the minimum RTT measured from ACK or SACK, or as the last resort from TCP timestamps. The accessor tcp_min_rtt() returns the minimum RTT seen in the window. ~0U indicates it is not available. The minimum is 1usec even if the true RTT is below that. Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-16 21:57:42 -07:00
tcp: track min RTT using windowed min-filter Kathleen Nichols' algorithm for tracking the minimum RTT of a data stream over some measurement window. It uses constant space and constant time per update. Yet it almost always delivers the same minimum as an implementation that has to keep all the data in the window. The measurement window is tunable via sysctl.net.ipv4.tcp_min_rtt_wlen with a default value of 5 minutes. The algorithm keeps track of the best, 2nd best & 3rd best min values, maintaining an invariant that the measurement time of the n'th best >= n-1'th best. It also makes sure that the three values are widely separated in the time window since that bounds the worse case error when that data is monotonically increasing over the window. Upon getting a new min, we can forget everything earlier because it has no value - the new min is less than everything else in the window by definition and it's the most recent. So we restart fresh on every new min and overwrites the 2nd & 3rd choices. The same property holds for the 2nd & 3rd best. Therefore we have to maintain two invariants to maximize the information in the samples, one on values (1st.v <= 2nd.v <= 3rd.v) and the other on times (now-win <=1st.t <= 2nd.t <= 3rd.t <= now). These invariants determine the structure of the code The RTT input to the windowed filter is the minimum RTT measured from ACK or SACK, or as the last resort from TCP timestamps. The accessor tcp_min_rtt() returns the minimum RTT seen in the window. ~0U indicates it is not available. The minimum is 1usec even if the true RTT is below that. Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-16 21:57:42 -07:00
tcp: add net.ipv4.tcp_rcvbuf_low_rtt This is a follow up of commit aa251c84636c ("tcp: fix too slow tcp_rcvbuf_grow() action") which brought again the issue that I tried to fix in commit 65c5287892e9 ("tcp: fix sk_rcvbuf overshoot") We also recently increased tcp_rmem[2] to 32 MB in commit 572be9bf9d0d ("tcp: increase tcp_rmem[2] to 32 MB") Idea of this patch is to not let tcp_rcvbuf_grow() grow sk->sk_rcvbuf too fast for small RTT flows. If sk->sk_rcvbuf is too big, this can force NIC driver to not recycle pages from their page pool, and also can cause cache evictions for DDIO enabled cpus/NIC, as receivers are usually slower than senders. Add net.ipv4.tcp_rcvbuf_low_rtt sysctl, set by default to 1000 usec (1 ms) If RTT if smaller than the sysctl value, use the RTT/tcp_rcvbuf_low_rtt ratio to control sk_rcvbuf inflation. Tested: Pair of hosts with a 200Gbit IDPF NIC. Using netperf/netserver Client initiates 8 TCP bulk flows, asking netserver to use CPU #10 only. super_netperf 8 -H server -T,10 -l 30 On server, use perf -e tcp:tcp_rcvbuf_grow while test is running. Before: sysctl -w net.ipv4.tcp_rcvbuf_low_rtt=1 perf record -a -e tcp:tcp_rcvbuf_grow sleep 30 ; perf script|tail -20|cut -c30-230 1153.051201: tcp:tcp_rcvbuf_grow: time=398 rtt_us=382 copied=6905856 inq=180224 space=6115328 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25600000 famil 1153.138752: tcp:tcp_rcvbuf_grow: time=446 rtt_us=413 copied=5529600 inq=180224 space=4505600 ooo=0 scaling_ratio=240 rcvbuf=23068672 rcv_ssthresh=21571860 window_clamp=21626880 rcv_wnd=21286912 famil 1153.361484: tcp:tcp_rcvbuf_grow: time=415 rtt_us=380 copied=7061504 inq=204800 space=6725632 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25600000 famil 1153.457642: tcp:tcp_rcvbuf_grow: time=483 rtt_us=421 copied=5885952 inq=720896 space=4407296 ooo=0 scaling_ratio=240 rcvbuf=23763511 rcv_ssthresh=22223271 window_clamp=22278291 rcv_wnd=21430272 famil 1153.466002: tcp:tcp_rcvbuf_grow: time=308 rtt_us=281 copied=3244032 inq=180224 space=2883584 ooo=0 scaling_ratio=240 rcvbuf=44854314 rcv_ssthresh=41992059 window_clamp=42050919 rcv_wnd=41713664 famil 1153.747792: tcp:tcp_rcvbuf_grow: time=394 rtt_us=332 copied=4460544 inq=585728 space=3063808 ooo=0 scaling_ratio=240 rcvbuf=44854314 rcv_ssthresh=41992059 window_clamp=42050919 rcv_wnd=41373696 famil 1154.260747: tcp:tcp_rcvbuf_grow: time=652 rtt_us=226 copied=10977280 inq=737280 space=9486336 ooo=0 scaling_ratio=240 rcvbuf=31165538 rcv_ssthresh=29197743 window_clamp=29217691 rcv_wnd=28368896 fami 1154.375019: tcp:tcp_rcvbuf_grow: time=461 rtt_us=443 copied=7573504 inq=507904 space=6856704 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25288704 famil 1154.463072: tcp:tcp_rcvbuf_grow: time=494 rtt_us=408 copied=7983104 inq=200704 space=7065600 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25579520 famil 1154.474658: tcp:tcp_rcvbuf_grow: time=507 rtt_us=459 copied=5586944 inq=540672 space=4718592 ooo=0 scaling_ratio=240 rcvbuf=17852266 rcv_ssthresh=16692999 window_clamp=16736499 rcv_wnd=16056320 famil 1154.584657: tcp:tcp_rcvbuf_grow: time=494 rtt_us=427 copied=8126464 inq=204800 space=7782400 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25878235 window_clamp=25937095 rcv_wnd=25600000 famil 1154.702117: tcp:tcp_rcvbuf_grow: time=480 rtt_us=406 copied=5734400 inq=180224 space=5349376 ooo=0 scaling_ratio=240 rcvbuf=23068672 rcv_ssthresh=21571860 window_clamp=21626880 rcv_wnd=21286912 famil 1155.941595: tcp:tcp_rcvbuf_grow: time=717 rtt_us=670 copied=11042816 inq=3784704 space=7159808 ooo=0 scaling_ratio=240 rcvbuf=19581357 rcv_ssthresh=18333222 window_clamp=18357522 rcv_wnd=14614528 fam 1156.384735: tcp:tcp_rcvbuf_grow: time=529 rtt_us=473 copied=9011200 inq=180224 space=7258112 ooo=0 scaling_ratio=240 rcvbuf=19581357 rcv_ssthresh=18333222 window_clamp=18357522 rcv_wnd=18018304 famil 1157.821676: tcp:tcp_rcvbuf_grow: time=529 rtt_us=272 copied=8224768 inq=602112 space=6545408 ooo=0 scaling_ratio=240 rcvbuf=67000000 rcv_ssthresh=62793576 window_clamp=62812500 rcv_wnd=62115840 famil 1158.906379: tcp:tcp_rcvbuf_grow: time=710 rtt_us=445 copied=11845632 inq=540672 space=10240000 ooo=0 scaling_ratio=240 rcvbuf=31165538 rcv_ssthresh=29205935 window_clamp=29217691 rcv_wnd=28536832 fam 1164.600160: tcp:tcp_rcvbuf_grow: time=841 rtt_us=430 copied=12976128 inq=1290240 space=11304960 ooo=0 scaling_ratio=240 rcvbuf=31165538 rcv_ssthresh=29212591 window_clamp=29217691 rcv_wnd=27856896 fa 1165.163572: tcp:tcp_rcvbuf_grow: time=845 rtt_us=800 copied=12632064 inq=540672 space=7921664 ooo=0 scaling_ratio=240 rcvbuf=27666235 rcv_ssthresh=25912795 window_clamp=25937095 rcv_wnd=25260032 fami 1165.653464: tcp:tcp_rcvbuf_grow: time=388 rtt_us=309 copied=4493312 inq=180224 space=3874816 ooo=0 scaling_ratio=240 rcvbuf=44854314 rcv_ssthresh=41995899 window_clamp=42050919 rcv_wnd=41713664 famil 1166.651211: tcp:tcp_rcvbuf_grow: time=556 rtt_us=553 copied=6328320 inq=540672 space=5554176 ooo=0 scaling_ratio=240 rcvbuf=23068672 rcv_ssthresh=21571860 window_clamp=21626880 rcv_wnd=20946944 famil After: sysctl -w net.ipv4.tcp_rcvbuf_low_rtt=1000 perf record -a -e tcp:tcp_rcvbuf_grow sleep 30 ; perf script|tail -20|cut -c30-230 1457.053149: tcp:tcp_rcvbuf_grow: time=128 rtt_us=24 copied=1441792 inq=40960 space=1269760 ooo=0 scaling_ratio=240 rcvbuf=2960741 rcv_ssthresh=2605474 window_clamp=2775694 rcv_wnd=2568192 family=AF_I 1458.000778: tcp:tcp_rcvbuf_grow: time=128 rtt_us=31 copied=1441792 inq=24576 space=1400832 ooo=0 scaling_ratio=240 rcvbuf=3060163 rcv_ssthresh=2810042 window_clamp=2868902 rcv_wnd=2674688 family=AF_I 1458.088059: tcp:tcp_rcvbuf_grow: time=190 rtt_us=110 copied=3227648 inq=385024 space=2781184 ooo=0 scaling_ratio=240 rcvbuf=6728240 rcv_ssthresh=6252705 window_clamp=6307725 rcv_wnd=5799936 family=AF 1458.148549: tcp:tcp_rcvbuf_grow: time=232 rtt_us=129 copied=3956736 inq=237568 space=2842624 ooo=0 scaling_ratio=240 rcvbuf=6731333 rcv_ssthresh=6252705 window_clamp=6310624 rcv_wnd=5918720 family=AF 1458.466861: tcp:tcp_rcvbuf_grow: time=193 rtt_us=83 copied=2949120 inq=180224 space=2457600 ooo=0 scaling_ratio=240 rcvbuf=5751438 rcv_ssthresh=5357689 window_clamp=5391973 rcv_wnd=5054464 family=AF_ 1458.775476: tcp:tcp_rcvbuf_grow: time=257 rtt_us=127 copied=4304896 inq=352256 space=3346432 ooo=0 scaling_ratio=240 rcvbuf=8067131 rcv_ssthresh=7523275 window_clamp=7562935 rcv_wnd=7061504 family=AF 1458.776631: tcp:tcp_rcvbuf_grow: time=200 rtt_us=96 copied=3260416 inq=143360 space=2768896 ooo=0 scaling_ratio=240 rcvbuf=6397256 rcv_ssthresh=5938567 window_clamp=5997427 rcv_wnd=5828608 family=AF_ 1459.707973: tcp:tcp_rcvbuf_grow: time=215 rtt_us=96 copied=2506752 inq=163840 space=1388544 ooo=0 scaling_ratio=240 rcvbuf=3068867 rcv_ssthresh=2768282 window_clamp=2877062 rcv_wnd=2555904 family=AF_ 1460.246494: tcp:tcp_rcvbuf_grow: time=231 rtt_us=80 copied=3756032 inq=204800 space=3117056 ooo=0 scaling_ratio=240 rcvbuf=7288091 rcv_ssthresh=6773725 window_clamp=6832585 rcv_wnd=6471680 family=AF_ 1460.714596: tcp:tcp_rcvbuf_grow: time=270 rtt_us=110 copied=4714496 inq=311296 space=3719168 ooo=0 scaling_ratio=240 rcvbuf=8957739 rcv_ssthresh=8339020 window_clamp=8397880 rcv_wnd=7933952 family=AF 1462.029977: tcp:tcp_rcvbuf_grow: time=101 rtt_us=19 copied=1105920 inq=40960 space=1036288 ooo=0 scaling_ratio=240 rcvbuf=2338970 rcv_ssthresh=2091684 window_clamp=2192784 rcv_wnd=1986560 family=AF_I 1462.802385: tcp:tcp_rcvbuf_grow: time=89 rtt_us=45 copied=1069056 inq=0 space=1064960 ooo=0 scaling_ratio=240 rcvbuf=2338970 rcv_ssthresh=2091684 window_clamp=2192784 rcv_wnd=2035712 family=AF_INET6 1462.918648: tcp:tcp_rcvbuf_grow: time=105 rtt_us=33 copied=1441792 inq=180224 space=1069056 ooo=0 scaling_ratio=240 rcvbuf=2383282 rcv_ssthresh=2091684 window_clamp=2234326 rcv_wnd=1896448 family=AF_ 1463.222533: tcp:tcp_rcvbuf_grow: time=273 rtt_us=144 copied=4603904 inq=385024 space=3469312 ooo=0 scaling_ratio=240 rcvbuf=8422564 rcv_ssthresh=7891053 window_clamp=7896153 rcv_wnd=7409664 family=AF 1466.519312: tcp:tcp_rcvbuf_grow: time=130 rtt_us=23 copied=1343488 inq=0 space=1261568 ooo=0 scaling_ratio=240 rcvbuf=2780158 rcv_ssthresh=2493778 window_clamp=2606398 rcv_wnd=2494464 family=AF_INET6 1466.681003: tcp:tcp_rcvbuf_grow: time=128 rtt_us=21 copied=1441792 inq=12288 space=1343488 ooo=0 scaling_ratio=240 rcvbuf=2932027 rcv_ssthresh=2578555 window_clamp=2748775 rcv_wnd=2568192 family=AF_I 1470.689959: tcp:tcp_rcvbuf_grow: time=255 rtt_us=122 copied=3932160 inq=204800 space=3551232 ooo=0 scaling_ratio=240 rcvbuf=8182038 rcv_ssthresh=7647384 window_clamp=7670660 rcv_wnd=7442432 family=AF 1471.754154: tcp:tcp_rcvbuf_grow: time=188 rtt_us=95 copied=2138112 inq=577536 space=1429504 ooo=0 scaling_ratio=240 rcvbuf=3113650 rcv_ssthresh=2806426 window_clamp=2919046 rcv_wnd=2248704 family=AF_ 1476.813542: tcp:tcp_rcvbuf_grow: time=269 rtt_us=99 copied=3088384 inq=180224 space=2564096 ooo=0 scaling_ratio=240 rcvbuf=6219470 rcv_ssthresh=5771893 window_clamp=5830753 rcv_wnd=5509120 family=AF_ 1477.738309: tcp:tcp_rcvbuf_grow: time=166 rtt_us=54 copied=1777664 inq=180224 space=1417216 ooo=0 scaling_ratio=240 rcvbuf=3117118 rcv_ssthresh=2874958 window_clamp=2922298 rcv_wnd=2613248 family=AF_ We can see sk_rcvbuf values are much smaller, and that rtt_us (estimation of rtt from a receiver point of view) is kept small, instead of being bloated. No difference in throughput. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com> Tested-by: Paolo Abeni <pabeni@redhat.com> Link: https://patch.msgid.link/20251119084813.3684576-3-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-11-19 08:48:13 +00:00
tcp: remove obsolete and unused RFC3517/RFC6675 loss recovery code RACK-TLP loss detection has been enabled as the default loss detection algorithm for Linux TCP since 2018, in: commit b38a51fec1c1 ("tcp: disable RFC6675 loss detection") In case users ran into unexpected bugs or performance regressions, that commit allowed Linux system administrators to revert to using RFC3517/RFC6675 loss recovery by setting net.ipv4.tcp_recovery to 0. In the seven years since 2018, our team has not heard reports of anyone reverting Linux TCP to use RFC3517/RFC6675 loss recovery, and we can't find any record in web searches of such a revert. RACK-TLP was published as a standards-track RFC, RFC8985, in February 2021. Several other major TCP implementations have default-enabled RACK-TLP at this point as well. RACK-TLP offers several significant performance advantages over RFC3517/RFC6675 loss recovery, including much better performance in the common cases of tail drops, lost retransmissions, and reordering. It is now time to remove the obsolete and unused RFC3517/RFC6675 loss recovery code. This will allow a substantial simplification of the Linux TCP code base, and removes 12 bytes of state in every tcp_sock for 64-bit machines (8 bytes on 32-bit machines). To arrange the commits in reasonable sizes, this patch series is split into 3 commits. The following 2 commits remove bookkeeping state and code that is no longer needed after this removal of RFC3517/RFC6675 loss recovery. Suggested-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Yuchung Cheng <ycheng@google.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20250615001435.2390793-2-ncardwell.sw@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-06-14 20:14:33 -04:00
tcp: add net.ipv4.tcp_comp_sack_rtt_percent TCP SACK compression has been added in 2018 in commit 5d9f4262b7ea ("tcp: add SACK compression"). It is working great for WAN flows (with large RTT). Wifi in particular gets a significant boost _when_ ACK are suppressed. Add a new sysctl so that we can tune the very conservative 5 % value that has been used so far in this formula, so that small RTT flows can benefit from this feature. delay = min ( 5 % of RTT, 1 ms) This patch adds new tcp_comp_sack_rtt_percent sysctl to ease experiments and tuning. Given that we cap the delay to 1ms (tcp_comp_sack_delay_ns sysctl), set the default value to 33 %. Quoting Neal Cardwell ( https://lore.kernel.org/netdev/CADVnQymZ1tFnEA1Q=vtECs0=Db7zHQ8=+WCQtnhHFVbEOzjVnQ@mail.gmail.com/ ) The rationale for 33% is basically to try to facilitate pipelining, where there are always at least 3 ACKs and 3 GSO/TSO skbs per SRTT, so that the path can maintain a budget for 3 full-sized GSO/TSO skbs "in flight" at all times: + 1 skb in the qdisc waiting to be sent by the NIC next + 1 skb being sent by the NIC (being serialized by the NIC out onto the wire) + 1 skb being received and aggregated by the receiver machine's aggregation mechanism (some combination of LRO, GRO, and sack compression) Note that this is basically the same magic number (3) and the same rationales as: (a) tcp_tso_should_defer() ensuring that we defer sending data for no longer than cwnd/tcp_tso_win_divisor (where tcp_tso_win_divisor = 3), and (b) bbr_quantization_budget() ensuring that cwnd is at least 3 GSO/TSO skbs to maintain pipelining and full throughput at low RTTs Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Link: https://patch.msgid.link/20251106115236.3450026-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-11-06 11:52:36 +00:00
tcp: add net.ipv4.tcp_comp_sack_rtt_percent TCP SACK compression has been added in 2018 in commit 5d9f4262b7ea ("tcp: add SACK compression"). It is working great for WAN flows (with large RTT). Wifi in particular gets a significant boost _when_ ACK are suppressed. Add a new sysctl so that we can tune the very conservative 5 % value that has been used so far in this formula, so that small RTT flows can benefit from this feature. delay = min ( 5 % of RTT, 1 ms) This patch adds new tcp_comp_sack_rtt_percent sysctl to ease experiments and tuning. Given that we cap the delay to 1ms (tcp_comp_sack_delay_ns sysctl), set the default value to 33 %. Quoting Neal Cardwell ( https://lore.kernel.org/netdev/CADVnQymZ1tFnEA1Q=vtECs0=Db7zHQ8=+WCQtnhHFVbEOzjVnQ@mail.gmail.com/ ) The rationale for 33% is basically to try to facilitate pipelining, where there are always at least 3 ACKs and 3 GSO/TSO skbs per SRTT, so that the path can maintain a budget for 3 full-sized GSO/TSO skbs "in flight" at all times: + 1 skb in the qdisc waiting to be sent by the NIC next + 1 skb being sent by the NIC (being serialized by the NIC out onto the wire) + 1 skb being received and aggregated by the receiver machine's aggregation mechanism (some combination of LRO, GRO, and sack compression) Note that this is basically the same magic number (3) and the same rationales as: (a) tcp_tso_should_defer() ensuring that we defer sending data for no longer than cwnd/tcp_tso_win_divisor (where tcp_tso_win_divisor = 3), and (b) bbr_quantization_budget() ensuring that cwnd is at least 3 GSO/TSO skbs to maintain pipelining and full throughput at low RTTs Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Link: https://patch.msgid.link/20251106115236.3450026-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-11-06 11:52:36 +00:00
tcp: defer regular ACK while processing socket backlog This idea came after a particular workload requested the quickack attribute set on routes, and a performance drop was noticed for large bulk transfers. For high throughput flows, it is best to use one cpu running the user thread issuing socket system calls, and a separate cpu to process incoming packets from BH context. (With TSO/GRO, bottleneck is usually the 'user' cpu) Problem is the user thread can spend a lot of time while holding the socket lock, forcing BH handler to queue most of incoming packets in the socket backlog. Whenever the user thread releases the socket lock, it must first process all accumulated packets in the backlog, potentially adding latency spikes. Due to flood mitigation, having too many packets in the backlog increases chance of unexpected drops. Backlog processing unfortunately shifts a fair amount of cpu cycles from the BH cpu to the 'user' cpu, thus reducing max throughput. This patch takes advantage of the backlog processing, and the fact that ACK are mostly cumulative. The idea is to detect we are in the backlog processing and defer all eligible ACK into a single one, sent from tcp_release_cb(). This saves cpu cycles on both sides, and network resources. Performance of a single TCP flow on a 200Gbit NIC: - Throughput is increased by 20% (100Gbit -> 120Gbit). - Number of generated ACK per second shrinks from 240,000 to 40,000. - Number of backlog drops per second shrinks from 230 to 0. Benchmark context: - Regular netperf TCP_STREAM (no zerocopy) - Intel(R) Xeon(R) Platinum 8481C (Saphire Rapids) - MAX_SKB_FRAGS = 17 (~60KB per GRO packet) This feature is guarded by a new sysctl, and enabled by default: /proc/sys/net/ipv4/tcp_backlog_ack_defer Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Acked-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2023-09-11 17:05:31 +00:00
tcp: defer regular ACK while processing socket backlog This idea came after a particular workload requested the quickack attribute set on routes, and a performance drop was noticed for large bulk transfers. For high throughput flows, it is best to use one cpu running the user thread issuing socket system calls, and a separate cpu to process incoming packets from BH context. (With TSO/GRO, bottleneck is usually the 'user' cpu) Problem is the user thread can spend a lot of time while holding the socket lock, forcing BH handler to queue most of incoming packets in the socket backlog. Whenever the user thread releases the socket lock, it must first process all accumulated packets in the backlog, potentially adding latency spikes. Due to flood mitigation, having too many packets in the backlog increases chance of unexpected drops. Backlog processing unfortunately shifts a fair amount of cpu cycles from the BH cpu to the 'user' cpu, thus reducing max throughput. This patch takes advantage of the backlog processing, and the fact that ACK are mostly cumulative. The idea is to detect we are in the backlog processing and defer all eligible ACK into a single one, sent from tcp_release_cb(). This saves cpu cycles on both sides, and network resources. Performance of a single TCP flow on a 200Gbit NIC: - Throughput is increased by 20% (100Gbit -> 120Gbit). - Number of generated ACK per second shrinks from 240,000 to 40,000. - Number of backlog drops per second shrinks from 230 to 0. Benchmark context: - Regular netperf TCP_STREAM (no zerocopy) - Intel(R) Xeon(R) Platinum 8481C (Saphire Rapids) - MAX_SKB_FRAGS = 17 (~60KB per GRO packet) This feature is guarded by a new sysctl, and enabled by default: /proc/sys/net/ipv4/tcp_backlog_ack_defer Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Acked-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2023-09-11 17:05:31 +00:00
tcp: defer regular ACK while processing socket backlog This idea came after a particular workload requested the quickack attribute set on routes, and a performance drop was noticed for large bulk transfers. For high throughput flows, it is best to use one cpu running the user thread issuing socket system calls, and a separate cpu to process incoming packets from BH context. (With TSO/GRO, bottleneck is usually the 'user' cpu) Problem is the user thread can spend a lot of time while holding the socket lock, forcing BH handler to queue most of incoming packets in the socket backlog. Whenever the user thread releases the socket lock, it must first process all accumulated packets in the backlog, potentially adding latency spikes. Due to flood mitigation, having too many packets in the backlog increases chance of unexpected drops. Backlog processing unfortunately shifts a fair amount of cpu cycles from the BH cpu to the 'user' cpu, thus reducing max throughput. This patch takes advantage of the backlog processing, and the fact that ACK are mostly cumulative. The idea is to detect we are in the backlog processing and defer all eligible ACK into a single one, sent from tcp_release_cb(). This saves cpu cycles on both sides, and network resources. Performance of a single TCP flow on a 200Gbit NIC: - Throughput is increased by 20% (100Gbit -> 120Gbit). - Number of generated ACK per second shrinks from 240,000 to 40,000. - Number of backlog drops per second shrinks from 230 to 0. Benchmark context: - Regular netperf TCP_STREAM (no zerocopy) - Intel(R) Xeon(R) Platinum 8481C (Saphire Rapids) - MAX_SKB_FRAGS = 17 (~60KB per GRO packet) This feature is guarded by a new sysctl, and enabled by default: /proc/sys/net/ipv4/tcp_backlog_ack_defer Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Acked-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2023-09-11 17:05:31 +00:00
net/tcp_fastopen: Disable active side TFO in certain scenarios Middlebox firewall issues can potentially cause server's data being blackholed after a successful 3WHS using TFO. Following are the related reports from Apple: https://www.nanog.org/sites/default/files/Paasch_Network_Support.pdf Slide 31 identifies an issue where the client ACK to the server's data sent during a TFO'd handshake is dropped. C ---> syn-data ---> S C <--- syn/ack ----- S C (accept & write) C <---- data ------- S C ----- ACK -> X S [retry and timeout] https://www.ietf.org/proceedings/94/slides/slides-94-tcpm-13.pdf Slide 5 shows a similar situation that the server's data gets dropped after 3WHS. C ---- syn-data ---> S C <--- syn/ack ----- S C ---- ack --------> S S (accept & write) C? X <- data ------ S [retry and timeout] This is the worst failure b/c the client can not detect such behavior to mitigate the situation (such as disabling TFO). Failing to proceed, the application (e.g., SSL library) may simply timeout and retry with TFO again, and the process repeats indefinitely. The proposed solution is to disable active TFO globally under the following circumstances: 1. client side TFO socket detects out of order FIN 2. client side TFO socket receives out of order RST We disable active side TFO globally for 1hr at first. Then if it happens again, we disable it for 2h, then 4h, 8h, ... And we reset the timeout to 1hr if a client side TFO sockets not opened on loopback has successfully received data segs from server. And we examine this condition during close(). The rational behind it is that when such firewall issue happens, application running on the client should eventually close the socket as it is not able to get the data it is expecting. Or application running on the server should close the socket as it is not able to receive any response from client. In both cases, out of order FIN or RST will get received on the client given that the firewall will not block them as no data are in those frames. And we want to disable active TFO globally as it helps if the middle box is very close to the client and most of the connections are likely to fail. Also, add a debug sysctl: tcp_fastopen_blackhole_detect_timeout_sec: the initial timeout to use when firewall blackhole issue happens. This can be set and read. When setting it to 0, it means to disable the active disable logic. Signed-off-by: Wei Wang <weiwan@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-20 14:45:46 -07:00
net/tcp_fastopen: Disable active side TFO in certain scenarios Middlebox firewall issues can potentially cause server's data being blackholed after a successful 3WHS using TFO. Following are the related reports from Apple: https://www.nanog.org/sites/default/files/Paasch_Network_Support.pdf Slide 31 identifies an issue where the client ACK to the server's data sent during a TFO'd handshake is dropped. C ---> syn-data ---> S C <--- syn/ack ----- S C (accept & write) C <---- data ------- S C ----- ACK -> X S [retry and timeout] https://www.ietf.org/proceedings/94/slides/slides-94-tcpm-13.pdf Slide 5 shows a similar situation that the server's data gets dropped after 3WHS. C ---- syn-data ---> S C <--- syn/ack ----- S C ---- ack --------> S S (accept & write) C? X <- data ------ S [retry and timeout] This is the worst failure b/c the client can not detect such behavior to mitigate the situation (such as disabling TFO). Failing to proceed, the application (e.g., SSL library) may simply timeout and retry with TFO again, and the process repeats indefinitely. The proposed solution is to disable active TFO globally under the following circumstances: 1. client side TFO socket detects out of order FIN 2. client side TFO socket receives out of order RST We disable active side TFO globally for 1hr at first. Then if it happens again, we disable it for 2h, then 4h, 8h, ... And we reset the timeout to 1hr if a client side TFO sockets not opened on loopback has successfully received data segs from server. And we examine this condition during close(). The rational behind it is that when such firewall issue happens, application running on the client should eventually close the socket as it is not able to get the data it is expecting. Or application running on the server should close the socket as it is not able to receive any response from client. In both cases, out of order FIN or RST will get received on the client given that the firewall will not block them as no data are in those frames. And we want to disable active TFO globally as it helps if the middle box is very close to the client and most of the connections are likely to fail. Also, add a debug sysctl: tcp_fastopen_blackhole_detect_timeout_sec: the initial timeout to use when firewall blackhole issue happens. This can be set and read. When setting it to 0, it means to disable the active disable logic. Signed-off-by: Wei Wang <weiwan@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-20 14:45:46 -07:00
tcp: TSO packets automatic sizing After hearing many people over past years complaining against TSO being bursty or even buggy, we are proud to present automatic sizing of TSO packets. One part of the problem is that tcp_tso_should_defer() uses an heuristic relying on upcoming ACKS instead of a timer, but more generally, having big TSO packets makes little sense for low rates, as it tends to create micro bursts on the network, and general consensus is to reduce the buffering amount. This patch introduces a per socket sk_pacing_rate, that approximates the current sending rate, and allows us to size the TSO packets so that we try to send one packet every ms. This field could be set by other transports. Patch has no impact for high speed flows, where having large TSO packets makes sense to reach line rate. For other flows, this helps better packet scheduling and ACK clocking. This patch increases performance of TCP flows in lossy environments. A new sysctl (tcp_min_tso_segs) is added, to specify the minimal size of a TSO packet (default being 2). A follow-up patch will provide a new packet scheduler (FQ), using sk_pacing_rate as an input to perform optional per flow pacing. This explains why we chose to set sk_pacing_rate to twice the current rate, allowing 'slow start' ramp up. sk_pacing_rate = 2 * cwnd * mss / srtt v2: Neal Cardwell reported a suspect deferring of last two segments on initial write of 10 MSS, I had to change tcp_tso_should_defer() to take into account tp->xmit_size_goal_segs Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Van Jacobson <vanj@google.com> Cc: Tom Herbert <therbert@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-27 05:46:32 -07:00
tcp: TSO packets automatic sizing After hearing many people over past years complaining against TSO being bursty or even buggy, we are proud to present automatic sizing of TSO packets. One part of the problem is that tcp_tso_should_defer() uses an heuristic relying on upcoming ACKS instead of a timer, but more generally, having big TSO packets makes little sense for low rates, as it tends to create micro bursts on the network, and general consensus is to reduce the buffering amount. This patch introduces a per socket sk_pacing_rate, that approximates the current sending rate, and allows us to size the TSO packets so that we try to send one packet every ms. This field could be set by other transports. Patch has no impact for high speed flows, where having large TSO packets makes sense to reach line rate. For other flows, this helps better packet scheduling and ACK clocking. This patch increases performance of TCP flows in lossy environments. A new sysctl (tcp_min_tso_segs) is added, to specify the minimal size of a TSO packet (default being 2). A follow-up patch will provide a new packet scheduler (FQ), using sk_pacing_rate as an input to perform optional per flow pacing. This explains why we chose to set sk_pacing_rate to twice the current rate, allowing 'slow start' ramp up. sk_pacing_rate = 2 * cwnd * mss / srtt v2: Neal Cardwell reported a suspect deferring of last two segments on initial write of 10 MSS, I had to change tcp_tso_should_defer() to take into account tp->xmit_size_goal_segs Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Van Jacobson <vanj@google.com> Cc: Tom Herbert <therbert@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-27 05:46:32 -07:00
tcp: TSO packets automatic sizing After hearing many people over past years complaining against TSO being bursty or even buggy, we are proud to present automatic sizing of TSO packets. One part of the problem is that tcp_tso_should_defer() uses an heuristic relying on upcoming ACKS instead of a timer, but more generally, having big TSO packets makes little sense for low rates, as it tends to create micro bursts on the network, and general consensus is to reduce the buffering amount. This patch introduces a per socket sk_pacing_rate, that approximates the current sending rate, and allows us to size the TSO packets so that we try to send one packet every ms. This field could be set by other transports. Patch has no impact for high speed flows, where having large TSO packets makes sense to reach line rate. For other flows, this helps better packet scheduling and ACK clocking. This patch increases performance of TCP flows in lossy environments. A new sysctl (tcp_min_tso_segs) is added, to specify the minimal size of a TSO packet (default being 2). A follow-up patch will provide a new packet scheduler (FQ), using sk_pacing_rate as an input to perform optional per flow pacing. This explains why we chose to set sk_pacing_rate to twice the current rate, allowing 'slow start' ramp up. sk_pacing_rate = 2 * cwnd * mss / srtt v2: Neal Cardwell reported a suspect deferring of last two segments on initial write of 10 MSS, I had to change tcp_tso_should_defer() to take into account tp->xmit_size_goal_segs Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Van Jacobson <vanj@google.com> Cc: Tom Herbert <therbert@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-27 05:46:32 -07:00
tcp: adjust TSO packet sizes based on min_rtt Back when tcp_tso_autosize() and TCP pacing were introduced, our focus was really to reduce burst sizes for long distance flows. The simple heuristic of using sk_pacing_rate/1024 has worked well, but can lead to too small packets for hosts in the same rack/cluster, when thousands of flows compete for the bottleneck. Neal Cardwell had the idea of making the TSO burst size a function of both sk_pacing_rate and tcp_min_rtt() Indeed, for local flows, sending bigger bursts is better to reduce cpu costs, as occasional losses can be repaired quite fast. This patch is based on Neal Cardwell implementation done more than two years ago. bbr is adjusting max_pacing_rate based on measured bandwidth, while cubic would over estimate max_pacing_rate. /proc/sys/net/ipv4/tcp_tso_rtt_log can be used to tune or disable this new feature, in logarithmic steps. Tested: 100Gbit NIC, two hosts in the same rack, 4K MTU. 600 flows rate-limited to 20000000 bytes per second. Before patch: (TSO sizes would be limited to 20000000/1024/4096 -> 4 segments per TSO) ~# echo 0 >/proc/sys/net/ipv4/tcp_tso_rtt_log ~# nstat -n;perf stat ./super_netperf 600 -H otrv6 -l 20 -- -K dctcp -q 20000000;nstat|egrep "TcpInSegs|TcpOutSegs|TcpRetransSegs|Delivered" 96005 Performance counter stats for './super_netperf 600 -H otrv6 -l 20 -- -K dctcp -q 20000000': 65,945.29 msec task-clock # 2.845 CPUs utilized 1,314,632 context-switches # 19935.279 M/sec 5,292 cpu-migrations # 80.249 M/sec 940,641 page-faults # 14264.023 M/sec 201,117,030,926 cycles # 3049769.216 GHz (83.45%) 17,699,435,405 stalled-cycles-frontend # 8.80% frontend cycles idle (83.48%) 136,584,015,071 stalled-cycles-backend # 67.91% backend cycles idle (83.44%) 53,809,530,436 instructions # 0.27 insn per cycle # 2.54 stalled cycles per insn (83.36%) 9,062,315,523 branches # 137422329.563 M/sec (83.22%) 153,008,621 branch-misses # 1.69% of all branches (83.32%) 23.182970846 seconds time elapsed TcpInSegs 15648792 0.0 TcpOutSegs 58659110 0.0 # Average of 3.7 4K segments per TSO packet TcpExtTCPDelivered 58654791 0.0 TcpExtTCPDeliveredCE 19 0.0 After patch: ~# echo 9 >/proc/sys/net/ipv4/tcp_tso_rtt_log ~# nstat -n;perf stat ./super_netperf 600 -H otrv6 -l 20 -- -K dctcp -q 20000000;nstat|egrep "TcpInSegs|TcpOutSegs|TcpRetransSegs|Delivered" 96046 Performance counter stats for './super_netperf 600 -H otrv6 -l 20 -- -K dctcp -q 20000000': 48,982.58 msec task-clock # 2.104 CPUs utilized 186,014 context-switches # 3797.599 M/sec 3,109 cpu-migrations # 63.472 M/sec 941,180 page-faults # 19214.814 M/sec 153,459,763,868 cycles # 3132982.807 GHz (83.56%) 12,069,861,356 stalled-cycles-frontend # 7.87% frontend cycles idle (83.32%) 120,485,917,953 stalled-cycles-backend # 78.51% backend cycles idle (83.24%) 36,803,672,106 instructions # 0.24 insn per cycle # 3.27 stalled cycles per insn (83.18%) 5,947,266,275 branches # 121417383.427 M/sec (83.64%) 87,984,616 branch-misses # 1.48% of all branches (83.43%) 23.281200256 seconds time elapsed TcpInSegs 1434706 0.0 TcpOutSegs 58883378 0.0 # Average of 41 4K segments per TSO packet TcpExtTCPDelivered 58878971 0.0 TcpExtTCPDeliveredCE 9664 0.0 Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Link: https://lore.kernel.org/r/20220309015757.2532973-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-08 17:57:57 -08:00
net-tcp: extend tcp_tw_reuse sysctl to enable loopback only optimization This changes the /proc/sys/net/ipv4/tcp_tw_reuse from a boolean to an integer. It now takes the values 0, 1 and 2, where 0 and 1 behave as before, while 2 enables timewait socket reuse only for sockets that we can prove are loopback connections: ie. bound to 'lo' interface or where one of source or destination IPs is 127.0.0.0/8, ::ffff:127.0.0.0/104 or ::1. This enables quicker reuse of ephemeral ports for loopback connections - where tcp_tw_reuse is 100% safe from a protocol perspective (this assumes no artificially induced packet loss on 'lo'). This also makes estblishing many loopback connections *much* faster (allocating ports out of the first half of the ephemeral port range is significantly faster, then allocating from the second half) Without this change in a 32K ephemeral port space my sample program (it just establishes and closes [::1]:ephemeral -> [::1]:server_port connections in a tight loop) fails after 32765 connections in 24 seconds. With it enabled 50000 connections only take 4.7 seconds. This is particularly problematic for IPv6 where we only have one local address and cannot play tricks with varying source IP from 127.0.0.0/8 pool. Signed-off-by: Maciej Żenczykowski <maze@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Wei Wang <weiwan@google.com> Change-Id: I0377961749979d0301b7b62871a32a4b34b654e1 Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-06-03 10:41:17 -07:00
net-tcp: extend tcp_tw_reuse sysctl to enable loopback only optimization This changes the /proc/sys/net/ipv4/tcp_tw_reuse from a boolean to an integer. It now takes the values 0, 1 and 2, where 0 and 1 behave as before, while 2 enables timewait socket reuse only for sockets that we can prove are loopback connections: ie. bound to 'lo' interface or where one of source or destination IPs is 127.0.0.0/8, ::ffff:127.0.0.0/104 or ::1. This enables quicker reuse of ephemeral ports for loopback connections - where tcp_tw_reuse is 100% safe from a protocol perspective (this assumes no artificially induced packet loss on 'lo'). This also makes estblishing many loopback connections *much* faster (allocating ports out of the first half of the ephemeral port range is significantly faster, then allocating from the second half) Without this change in a 32K ephemeral port space my sample program (it just establishes and closes [::1]:ephemeral -> [::1]:server_port connections in a tight loop) fails after 32765 connections in 24 seconds. With it enabled 50000 connections only take 4.7 seconds. This is particularly problematic for IPv6 where we only have one local address and cannot play tricks with varying source IP from 127.0.0.0/8 pool. Signed-off-by: Maciej Żenczykowski <maze@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Wei Wang <weiwan@google.com> Change-Id: I0377961749979d0301b7b62871a32a4b34b654e1 Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-06-03 10:41:17 -07:00
tcp: Add sysctl to configure TIME-WAIT reuse delay Today we have a hardcoded delay of 1 sec before a TIME-WAIT socket can be reused by reopening a connection. This is a safe choice based on an assumption that the other TCP timestamp clock frequency, which is unknown to us, may be as low as 1 Hz (RFC 7323, section 5.4). However, this means that in the presence of short lived connections with an RTT of couple of milliseconds, the time during which a 4-tuple is blocked from reuse can be orders of magnitude longer that the connection lifetime. Combined with a reduced pool of ephemeral ports, when using IP_LOCAL_PORT_RANGE to share an egress IP address between hosts [1], the long TIME-WAIT reuse delay can lead to port exhaustion, where all available 4-tuples are tied up in TIME-WAIT state. Turn the reuse delay into a per-netns setting so that sysadmins can make more aggressive assumptions about remote TCP timestamp clock frequency and shorten the delay in order to allow connections to reincarnate faster. Note that applications can completely bypass the TIME-WAIT delay protection already today by locking the local port with bind() before connecting. Such immediate connection reuse may result in PAWS failing to detect old duplicate segments, leaving us with just the sequence number check as a safety net. This new configurable offers a trade off where the sysadmin can balance between the risk of PAWS detection failing to act versus exhausting ports by having sockets tied up in TIME-WAIT state for too long. [1] https://lpc.events/event/16/contributions/1349/ Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Link: https://patch.msgid.link/20241209-jakub-krn-909-poc-msec-tw-tstamp-v2-2-66aca0eed03e@cloudflare.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-12-09 20:38:04 +01:00
tcp: enforce receive buffer memory limits by allowing the tcp window to shrink Under certain circumstances, the tcp receive buffer memory limit set by autotuning (sk_rcvbuf) is increased due to incoming data packets as a result of the window not closing when it should be. This can result in the receive buffer growing all the way up to tcp_rmem[2], even for tcp sessions with a low BDP. To reproduce: Connect a TCP session with the receiver doing nothing and the sender sending small packets (an infinite loop of socket send() with 4 bytes of payload with a sleep of 1 ms in between each send()). This will cause the tcp receive buffer to grow all the way up to tcp_rmem[2]. As a result, a host can have individual tcp sessions with receive buffers of size tcp_rmem[2], and the host itself can reach tcp_mem limits, causing the host to go into tcp memory pressure mode. The fundamental issue is the relationship between the granularity of the window scaling factor and the number of byte ACKed back to the sender. This problem has previously been identified in RFC 7323, appendix F [1]. The Linux kernel currently adheres to never shrinking the window. In addition to the overallocation of memory mentioned above, the current behavior is functionally incorrect, because once tcp_rmem[2] is reached when no remediations remain (i.e. tcp collapse fails to free up any more memory and there are no packets to prune from the out-of-order queue), the receiver will drop in-window packets resulting in retransmissions and an eventual timeout of the tcp session. A receive buffer full condition should instead result in a zero window and an indefinite wait. In practice, this problem is largely hidden for most flows. It is not applicable to mice flows. Elephant flows can send data fast enough to "overrun" the sk_rcvbuf limit (in a single ACK), triggering a zero window. But this problem does show up for other types of flows. Examples are websockets and other type of flows that send small amounts of data spaced apart slightly in time. In these cases, we directly encounter the problem described in [1]. RFC 7323, section 2.4 [2], says there are instances when a retracted window can be offered, and that TCP implementations MUST ensure that they handle a shrinking window, as specified in RFC 1122, section 4.2.2.16 [3]. All prior RFCs on the topic of tcp window management have made clear that sender must accept a shrunk window from the receiver, including RFC 793 [4] and RFC 1323 [5]. This patch implements the functionality to shrink the tcp window when necessary to keep the right edge within the memory limit by autotuning (sk_rcvbuf). This new functionality is enabled with the new sysctl: net.ipv4.tcp_shrink_window Additional information can be found at: https://blog.cloudflare.com/unbounded-memory-usage-by-tcp-for-receive-buffers-and-how-we-fixed-it/ [1] https://www.rfc-editor.org/rfc/rfc7323#appendix-F [2] https://www.rfc-editor.org/rfc/rfc7323#section-2.4 [3] https://www.rfc-editor.org/rfc/rfc1122#page-91 [4] https://www.rfc-editor.org/rfc/rfc793 [5] https://www.rfc-editor.org/rfc/rfc1323 Signed-off-by: Mike Freemon <mfreemon@cloudflare.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-06-11 22:05:24 -05:00
tcp: enforce receive buffer memory limits by allowing the tcp window to shrink Under certain circumstances, the tcp receive buffer memory limit set by autotuning (sk_rcvbuf) is increased due to incoming data packets as a result of the window not closing when it should be. This can result in the receive buffer growing all the way up to tcp_rmem[2], even for tcp sessions with a low BDP. To reproduce: Connect a TCP session with the receiver doing nothing and the sender sending small packets (an infinite loop of socket send() with 4 bytes of payload with a sleep of 1 ms in between each send()). This will cause the tcp receive buffer to grow all the way up to tcp_rmem[2]. As a result, a host can have individual tcp sessions with receive buffers of size tcp_rmem[2], and the host itself can reach tcp_mem limits, causing the host to go into tcp memory pressure mode. The fundamental issue is the relationship between the granularity of the window scaling factor and the number of byte ACKed back to the sender. This problem has previously been identified in RFC 7323, appendix F [1]. The Linux kernel currently adheres to never shrinking the window. In addition to the overallocation of memory mentioned above, the current behavior is functionally incorrect, because once tcp_rmem[2] is reached when no remediations remain (i.e. tcp collapse fails to free up any more memory and there are no packets to prune from the out-of-order queue), the receiver will drop in-window packets resulting in retransmissions and an eventual timeout of the tcp session. A receive buffer full condition should instead result in a zero window and an indefinite wait. In practice, this problem is largely hidden for most flows. It is not applicable to mice flows. Elephant flows can send data fast enough to "overrun" the sk_rcvbuf limit (in a single ACK), triggering a zero window. But this problem does show up for other types of flows. Examples are websockets and other type of flows that send small amounts of data spaced apart slightly in time. In these cases, we directly encounter the problem described in [1]. RFC 7323, section 2.4 [2], says there are instances when a retracted window can be offered, and that TCP implementations MUST ensure that they handle a shrinking window, as specified in RFC 1122, section 4.2.2.16 [3]. All prior RFCs on the topic of tcp window management have made clear that sender must accept a shrunk window from the receiver, including RFC 793 [4] and RFC 1323 [5]. This patch implements the functionality to shrink the tcp window when necessary to keep the right edge within the memory limit by autotuning (sk_rcvbuf). This new functionality is enabled with the new sysctl: net.ipv4.tcp_shrink_window Additional information can be found at: https://blog.cloudflare.com/unbounded-memory-usage-by-tcp-for-receive-buffers-and-how-we-fixed-it/ [1] https://www.rfc-editor.org/rfc/rfc7323#appendix-F [2] https://www.rfc-editor.org/rfc/rfc7323#section-2.4 [3] https://www.rfc-editor.org/rfc/rfc1122#page-91 [4] https://www.rfc-editor.org/rfc/rfc793 [5] https://www.rfc-editor.org/rfc/rfc1323 Signed-off-by: Mike Freemon <mfreemon@cloudflare.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-06-11 22:05:24 -05:00
tcp: enforce receive buffer memory limits by allowing the tcp window to shrink Under certain circumstances, the tcp receive buffer memory limit set by autotuning (sk_rcvbuf) is increased due to incoming data packets as a result of the window not closing when it should be. This can result in the receive buffer growing all the way up to tcp_rmem[2], even for tcp sessions with a low BDP. To reproduce: Connect a TCP session with the receiver doing nothing and the sender sending small packets (an infinite loop of socket send() with 4 bytes of payload with a sleep of 1 ms in between each send()). This will cause the tcp receive buffer to grow all the way up to tcp_rmem[2]. As a result, a host can have individual tcp sessions with receive buffers of size tcp_rmem[2], and the host itself can reach tcp_mem limits, causing the host to go into tcp memory pressure mode. The fundamental issue is the relationship between the granularity of the window scaling factor and the number of byte ACKed back to the sender. This problem has previously been identified in RFC 7323, appendix F [1]. The Linux kernel currently adheres to never shrinking the window. In addition to the overallocation of memory mentioned above, the current behavior is functionally incorrect, because once tcp_rmem[2] is reached when no remediations remain (i.e. tcp collapse fails to free up any more memory and there are no packets to prune from the out-of-order queue), the receiver will drop in-window packets resulting in retransmissions and an eventual timeout of the tcp session. A receive buffer full condition should instead result in a zero window and an indefinite wait. In practice, this problem is largely hidden for most flows. It is not applicable to mice flows. Elephant flows can send data fast enough to "overrun" the sk_rcvbuf limit (in a single ACK), triggering a zero window. But this problem does show up for other types of flows. Examples are websockets and other type of flows that send small amounts of data spaced apart slightly in time. In these cases, we directly encounter the problem described in [1]. RFC 7323, section 2.4 [2], says there are instances when a retracted window can be offered, and that TCP implementations MUST ensure that they handle a shrinking window, as specified in RFC 1122, section 4.2.2.16 [3]. All prior RFCs on the topic of tcp window management have made clear that sender must accept a shrunk window from the receiver, including RFC 793 [4] and RFC 1323 [5]. This patch implements the functionality to shrink the tcp window when necessary to keep the right edge within the memory limit by autotuning (sk_rcvbuf). This new functionality is enabled with the new sysctl: net.ipv4.tcp_shrink_window Additional information can be found at: https://blog.cloudflare.com/unbounded-memory-usage-by-tcp-for-receive-buffers-and-how-we-fixed-it/ [1] https://www.rfc-editor.org/rfc/rfc7323#appendix-F [2] https://www.rfc-editor.org/rfc/rfc7323#section-2.4 [3] https://www.rfc-editor.org/rfc/rfc1122#page-91 [4] https://www.rfc-editor.org/rfc/rfc793 [5] https://www.rfc-editor.org/rfc/rfc1323 Signed-off-by: Mike Freemon <mfreemon@cloudflare.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2023-06-11 22:05:24 -05:00
tcp: TCP_NOTSENT_LOWAT socket option Idea of this patch is to add optional limitation of number of unsent bytes in TCP sockets, to reduce usage of kernel memory. TCP receiver might announce a big window, and TCP sender autotuning might allow a large amount of bytes in write queue, but this has little performance impact if a large part of this buffering is wasted : Write queue needs to be large only to deal with large BDP, not necessarily to cope with scheduling delays (incoming ACKS make room for the application to queue more bytes) For most workloads, using a value of 128 KB or less is OK to give applications enough time to react to POLLOUT events in time (or being awaken in a blocking sendmsg()) This patch adds two ways to set the limit : 1) Per socket option TCP_NOTSENT_LOWAT 2) A sysctl (/proc/sys/net/ipv4/tcp_notsent_lowat) for sockets not using TCP_NOTSENT_LOWAT socket option (or setting a zero value) Default value being UINT_MAX (0xFFFFFFFF), meaning this has no effect. This changes poll()/select()/epoll() to report POLLOUT only if number of unsent bytes is below tp->nosent_lowat Note this might increase number of sendmsg()/sendfile() calls when using non blocking sockets, and increase number of context switches for blocking sockets. Note this is not related to SO_SNDLOWAT (as SO_SNDLOWAT is defined as : Specify the minimum number of bytes in the buffer until the socket layer will pass the data to the protocol) Tested: netperf sessions, and watching /proc/net/protocols "memory" column for TCP With 200 concurrent netperf -t TCP_STREAM sessions, amount of kernel memory used by TCP buffers shrinks by ~55 % (20567 pages instead of 45458) lpq83:~# echo -1 >/proc/sys/net/ipv4/tcp_notsent_lowat lpq83:~# (super_netperf 200 -t TCP_STREAM -H remote -l 90 &); sleep 60 ; grep TCP /proc/net/protocols TCPv6 1880 2 45458 no 208 yes ipv6 y y y y y y y y y y y y y n y y y y y TCP 1696 508 45458 no 208 yes kernel y y y y y y y y y y y y y n y y y y y lpq83:~# echo 131072 >/proc/sys/net/ipv4/tcp_notsent_lowat lpq83:~# (super_netperf 200 -t TCP_STREAM -H remote -l 90 &); sleep 60 ; grep TCP /proc/net/protocols TCPv6 1880 2 20567 no 208 yes ipv6 y y y y y y y y y y y y y n y y y y y TCP 1696 508 20567 no 208 yes kernel y y y y y y y y y y y y y n y y y y y Using 128KB has no bad effect on the throughput or cpu usage of a single flow, although there is an increase of context switches. A bonus is that we hold socket lock for a shorter amount of time and should improve latencies of ACK processing. lpq83:~# echo -1 >/proc/sys/net/ipv4/tcp_notsent_lowat lpq83:~# perf stat -e context-switches ./netperf -H 7.7.7.84 -t omni -l 20 -c -i10,3 OMNI Send TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.7.84 () port 0 AF_INET : +/-2.500% @ 99% conf. Local Remote Local Elapsed Throughput Throughput Local Local Remote Remote Local Remote Service Send Socket Recv Socket Send Time Units CPU CPU CPU CPU Service Service Demand Size Size Size (sec) Util Util Util Util Demand Demand Units Final Final % Method % Method 1651584 6291456 16384 20.00 17447.90 10^6bits/s 3.13 S -1.00 U 0.353 -1.000 usec/KB Performance counter stats for './netperf -H 7.7.7.84 -t omni -l 20 -c -i10,3': 412,514 context-switches 200.034645535 seconds time elapsed lpq83:~# echo 131072 >/proc/sys/net/ipv4/tcp_notsent_lowat lpq83:~# perf stat -e context-switches ./netperf -H 7.7.7.84 -t omni -l 20 -c -i10,3 OMNI Send TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.7.84 () port 0 AF_INET : +/-2.500% @ 99% conf. Local Remote Local Elapsed Throughput Throughput Local Local Remote Remote Local Remote Service Send Socket Recv Socket Send Time Units CPU CPU CPU CPU Service Service Demand Size Size Size (sec) Util Util Util Util Demand Demand Units Final Final % Method % Method 1593240 6291456 16384 20.00 17321.16 10^6bits/s 3.35 S -1.00 U 0.381 -1.000 usec/KB Performance counter stats for './netperf -H 7.7.7.84 -t omni -l 20 -c -i10,3': 2,675,818 context-switches 200.029651391 seconds time elapsed Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neal Cardwell <ncardwell@google.com> Cc: Yuchung Cheng <ycheng@google.com> Acked-By: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-22 20:27:07 -07:00
tcp: TCP Small Queues This introduce TSQ (TCP Small Queues) TSQ goal is to reduce number of TCP packets in xmit queues (qdisc & device queues), to reduce RTT and cwnd bias, part of the bufferbloat problem. sk->sk_wmem_alloc not allowed to grow above a given limit, allowing no more than ~128KB [1] per tcp socket in qdisc/dev layers at a given time. TSO packets are sized/capped to half the limit, so that we have two TSO packets in flight, allowing better bandwidth use. As a side effect, setting the limit to 40000 automatically reduces the standard gso max limit (65536) to 40000/2 : It can help to reduce latencies of high prio packets, having smaller TSO packets. This means we divert sock_wfree() to a tcp_wfree() handler, to queue/send following frames when skb_orphan() [2] is called for the already queued skbs. Results on my dev machines (tg3/ixgbe nics) are really impressive, using standard pfifo_fast, and with or without TSO/GSO. Without reduction of nominal bandwidth, we have reduction of buffering per bulk sender : < 1ms on Gbit (instead of 50ms with TSO) < 8ms on 100Mbit (instead of 132 ms) I no longer have 4 MBytes backlogged in qdisc by a single netperf session, and both side socket autotuning no longer use 4 Mbytes. As skb destructor cannot restart xmit itself ( as qdisc lock might be taken at this point ), we delegate the work to a tasklet. We use one tasklest per cpu for performance reasons. If tasklet finds a socket owned by the user, it sets TSQ_OWNED flag. This flag is tested in a new protocol method called from release_sock(), to eventually send new segments. [1] New /proc/sys/net/ipv4/tcp_limit_output_bytes tunable [2] skb_orphan() is usually called at TX completion time, but some drivers call it in their start_xmit() handler. These drivers should at least use BQL, or else a single TCP session can still fill the whole NIC TX ring, since TSQ will have no effect. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Dave Taht <dave.taht@bufferbloat.net> Cc: Tom Herbert <therbert@google.com> Cc: Matt Mathis <mattmathis@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Nandita Dukkipati <nanditad@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-11 05:50:31 +00:00
tcp: TCP Small Queues This introduce TSQ (TCP Small Queues) TSQ goal is to reduce number of TCP packets in xmit queues (qdisc & device queues), to reduce RTT and cwnd bias, part of the bufferbloat problem. sk->sk_wmem_alloc not allowed to grow above a given limit, allowing no more than ~128KB [1] per tcp socket in qdisc/dev layers at a given time. TSO packets are sized/capped to half the limit, so that we have two TSO packets in flight, allowing better bandwidth use. As a side effect, setting the limit to 40000 automatically reduces the standard gso max limit (65536) to 40000/2 : It can help to reduce latencies of high prio packets, having smaller TSO packets. This means we divert sock_wfree() to a tcp_wfree() handler, to queue/send following frames when skb_orphan() [2] is called for the already queued skbs. Results on my dev machines (tg3/ixgbe nics) are really impressive, using standard pfifo_fast, and with or without TSO/GSO. Without reduction of nominal bandwidth, we have reduction of buffering per bulk sender : < 1ms on Gbit (instead of 50ms with TSO) < 8ms on 100Mbit (instead of 132 ms) I no longer have 4 MBytes backlogged in qdisc by a single netperf session, and both side socket autotuning no longer use 4 Mbytes. As skb destructor cannot restart xmit itself ( as qdisc lock might be taken at this point ), we delegate the work to a tasklet. We use one tasklest per cpu for performance reasons. If tasklet finds a socket owned by the user, it sets TSQ_OWNED flag. This flag is tested in a new protocol method called from release_sock(), to eventually send new segments. [1] New /proc/sys/net/ipv4/tcp_limit_output_bytes tunable [2] skb_orphan() is usually called at TX completion time, but some drivers call it in their start_xmit() handler. These drivers should at least use BQL, or else a single TCP session can still fill the whole NIC TX ring, since TSQ will have no effect. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Dave Taht <dave.taht@bufferbloat.net> Cc: Tom Herbert <therbert@google.com> Cc: Matt Mathis <mattmathis@google.com> Cc: Yuchung Cheng <ycheng@google.com> Cc: Nandita Dukkipati <nanditad@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-11 05:50:31 +00:00
tcp: Introduce optional per-netns ehash. The more sockets we have in the hash table, the longer we spend looking up the socket. While running a number of small workloads on the same host, they penalise each other and cause performance degradation. The root cause might be a single workload that consumes much more resources than the others. It often happens on a cloud service where different workloads share the same computing resource. On EC2 c5.24xlarge instance (196 GiB memory and 524288 (1Mi / 2) ehash entries), after running iperf3 in different netns, creating 24Mi sockets without data transfer in the root netns causes about 10% performance regression for the iperf3's connection. thash_entries sockets length Gbps 524288 1 1 50.7 24Mi 48 45.1 It is basically related to the length of the list of each hash bucket. For testing purposes to see how performance drops along the length, I set 131072 (1Mi / 8) to thash_entries, and here's the result. thash_entries sockets length Gbps 131072 1 1 50.7 1Mi 8 49.9 2Mi 16 48.9 4Mi 32 47.3 8Mi 64 44.6 16Mi 128 40.6 24Mi 192 36.3 32Mi 256 32.5 40Mi 320 27.0 48Mi 384 25.0 To resolve the socket lookup degradation, we introduce an optional per-netns hash table for TCP, but it's just ehash, and we still share the global bhash, bhash2 and lhash2. With a smaller ehash, we can look up non-listener sockets faster and isolate such noisy neighbours. In addition, we can reduce lock contention. We can control the ehash size by a new sysctl knob. However, depending on workloads, it will require very sensitive tuning, so we disable the feature by default (net.ipv4.tcp_child_ehash_entries == 0). Moreover, we can fall back to using the global ehash in case we fail to allocate enough memory for a new ehash. The maximum size is 16Mi, which is large enough that even if we have 48Mi sockets, the average list length is 3, and regression would be less than 1%. We can check the current ehash size by another read-only sysctl knob, net.ipv4.tcp_ehash_entries. A negative value means the netns shares the global ehash (per-netns ehash is disabled or failed to allocate memory). # dmesg | cut -d ' ' -f 5- | grep "established hash" TCP established hash table entries: 524288 (order: 10, 4194304 bytes, vmalloc hugepage) # sysctl net.ipv4.tcp_ehash_entries net.ipv4.tcp_ehash_entries = 524288 # can be changed by thash_entries # sysctl net.ipv4.tcp_child_ehash_entries net.ipv4.tcp_child_ehash_entries = 0 # disabled by default # ip netns add test1 # ip netns exec test1 sysctl net.ipv4.tcp_ehash_entries net.ipv4.tcp_ehash_entries = -524288 # share the global ehash # sysctl -w net.ipv4.tcp_child_ehash_entries=100 net.ipv4.tcp_child_ehash_entries = 100 # ip netns add test2 # ip netns exec test2 sysctl net.ipv4.tcp_ehash_entries net.ipv4.tcp_ehash_entries = 128 # own a per-netns ehash with 2^n buckets When more than two processes in the same netns create per-netns ehash concurrently with different sizes, we need to guarantee the size in one of the following ways: 1) Share the global ehash and create per-netns ehash First, unshare() with tcp_child_ehash_entries==0. It creates dedicated netns sysctl knobs where we can safely change tcp_child_ehash_entries and clone()/unshare() to create a per-netns ehash. 2) Control write on sysctl by BPF We can use BPF_PROG_TYPE_CGROUP_SYSCTL to allow/deny read/write on sysctl knobs. Note that the global ehash allocated at the boot time is spread over available NUMA nodes, but inet_pernet_hashinfo_alloc() will allocate pages for each per-netns ehash depending on the current process's NUMA policy. By default, the allocation is done in the local node only, so the per-netns hash table could fully reside on a random node. Thus, depending on the NUMA policy the netns is created with and the CPU the current thread is running on, we could see some performance differences for highly optimised networking applications. Note also that the default values of two sysctl knobs depend on the ehash size and should be tuned carefully: tcp_max_tw_buckets : tcp_child_ehash_entries / 2 tcp_max_syn_backlog : max(128, tcp_child_ehash_entries / 128) As a bonus, we can dismantle netns faster. Currently, while destroying netns, we call inet_twsk_purge(), which walks through the global ehash. It can be potentially big because it can have many sockets other than TIME_WAIT in all netns. Splitting ehash changes that situation, where it's only necessary for inet_twsk_purge() to clean up TIME_WAIT sockets in each netns. With regard to this, we do not free the per-netns ehash in inet_twsk_kill() to avoid UAF while iterating the per-netns ehash in inet_twsk_purge(). Instead, we do it in tcp_sk_exit_batch() after calling tcp_twsk_purge() to keep it protocol-family-independent. In the future, we could optimise ehash lookup/iteration further by removing netns comparison for the per-netns ehash. Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-09-07 18:10:22 -07:00
udp: Introduce optional per-netns hash table. The maximum hash table size is 64K due to the nature of the protocol. [0] It's smaller than TCP, and fewer sockets can cause a performance drop. On an EC2 c5.24xlarge instance (192 GiB memory), after running iperf3 in different netns, creating 32Mi sockets without data transfer in the root netns causes regression for the iperf3's connection. uhash_entries sockets length Gbps 64K 1 1 5.69 1Mi 16 5.27 2Mi 32 4.90 4Mi 64 4.09 8Mi 128 2.96 16Mi 256 2.06 32Mi 512 1.12 The per-netns hash table breaks the lengthy lists into shorter ones. It is useful on a multi-tenant system with thousands of netns. With smaller hash tables, we can look up sockets faster, isolate noisy neighbours, and reduce lock contention. The max size of the per-netns table is 64K as well. This is because the possible hash range by udp_hashfn() always fits in 64K within the same netns and we cannot make full use of the whole buckets larger than 64K. /* 0 < num < 64K -> X < hash < X + 64K */ (num + net_hash_mix(net)) & mask; Also, the min size is 128. We use a bitmap to search for an available port in udp_lib_get_port(). To keep the bitmap on the stack and not fire the CONFIG_FRAME_WARN error at build time, we round up the table size to 128. The sysctl usage is the same with TCP: $ dmesg | cut -d ' ' -f 6- | grep "UDP hash" UDP hash table entries: 65536 (order: 9, 2097152 bytes, vmalloc) # sysctl net.ipv4.udp_hash_entries net.ipv4.udp_hash_entries = 65536 # can be changed by uhash_entries # sysctl net.ipv4.udp_child_hash_entries net.ipv4.udp_child_hash_entries = 0 # disabled by default # ip netns add test1 # ip netns exec test1 sysctl net.ipv4.udp_hash_entries net.ipv4.udp_hash_entries = -65536 # share the global table # sysctl -w net.ipv4.udp_child_hash_entries=100 net.ipv4.udp_child_hash_entries = 100 # ip netns add test2 # ip netns exec test2 sysctl net.ipv4.udp_hash_entries net.ipv4.udp_hash_entries = 128 # own a per-netns table with 2^n buckets We could optimise the hash table lookup/iteration further by removing the netns comparison for the per-netns one in the future. Also, we could optimise the sparse udp_hslot layout by putting it in udp_table. [0]: https://lore.kernel.org/netdev/4ACC2815.7010101@gmail.com/ Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-11-14 13:57:57 -08:00
udp: Introduce optional per-netns hash table. The maximum hash table size is 64K due to the nature of the protocol. [0] It's smaller than TCP, and fewer sockets can cause a performance drop. On an EC2 c5.24xlarge instance (192 GiB memory), after running iperf3 in different netns, creating 32Mi sockets without data transfer in the root netns causes regression for the iperf3's connection. uhash_entries sockets length Gbps 64K 1 1 5.69 1Mi 16 5.27 2Mi 32 4.90 4Mi 64 4.09 8Mi 128 2.96 16Mi 256 2.06 32Mi 512 1.12 The per-netns hash table breaks the lengthy lists into shorter ones. It is useful on a multi-tenant system with thousands of netns. With smaller hash tables, we can look up sockets faster, isolate noisy neighbours, and reduce lock contention. The max size of the per-netns table is 64K as well. This is because the possible hash range by udp_hashfn() always fits in 64K within the same netns and we cannot make full use of the whole buckets larger than 64K. /* 0 < num < 64K -> X < hash < X + 64K */ (num + net_hash_mix(net)) & mask; Also, the min size is 128. We use a bitmap to search for an available port in udp_lib_get_port(). To keep the bitmap on the stack and not fire the CONFIG_FRAME_WARN error at build time, we round up the table size to 128. The sysctl usage is the same with TCP: $ dmesg | cut -d ' ' -f 6- | grep "UDP hash" UDP hash table entries: 65536 (order: 9, 2097152 bytes, vmalloc) # sysctl net.ipv4.udp_hash_entries net.ipv4.udp_hash_entries = 65536 # can be changed by uhash_entries # sysctl net.ipv4.udp_child_hash_entries net.ipv4.udp_child_hash_entries = 0 # disabled by default # ip netns add test1 # ip netns exec test1 sysctl net.ipv4.udp_hash_entries net.ipv4.udp_hash_entries = -65536 # share the global table # sysctl -w net.ipv4.udp_child_hash_entries=100 net.ipv4.udp_child_hash_entries = 100 # ip netns add test2 # ip netns exec test2 sysctl net.ipv4.udp_hash_entries net.ipv4.udp_hash_entries = 128 # own a per-netns table with 2^n buckets We could optimise the hash table lookup/iteration further by removing the netns comparison for the per-netns one in the future. Also, we could optimise the sparse udp_hslot layout by putting it in udp_table. [0]: https://lore.kernel.org/netdev/4ACC2815.7010101@gmail.com/ Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-11-14 13:57:57 -08:00
inet: add ip_local_port_step_width sysctl to improve port usage distribution With the current port selection algorithm, ports after a reserved port range or long time used port are used more often than others [1]. This causes an uneven port usage distribution. This combines with cloud environments blocking connections between the application server and the database server if there was a previous connection with the same source port, leading to connectivity problems between applications on cloud environments. The real issue here is that these firewalls cannot cope with standards-compliant port reuse. This is a workaround for such situations and an improvement on the distribution of ports selected. The proposed solution is to implement a variant of RFC 6056 Algorithm 5. The step size is selected randomly on every connect() call ensuring it is a coprime with respect to the size of the range of ports we want to scan. This way, we can ensure that all ports within the range are scanned before returning an error. To enable this algorithm, the user must configure the new sysctl option "net.ipv4.ip_local_port_step_width". In addition, on graphs generated we can observe that the distribution of source ports is more even with the proposed approach. [2] [1] https://0xffsoftware.com/port_graph_current_alg.html [2] https://0xffsoftware.com/port_graph_random_step_alg.html Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de> Link: https://patch.msgid.link/20260309023946.5473-2-fmancera@suse.de Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2026-03-09 03:39:45 +01:00
tcp: bind(0) remove the SO_REUSEADDR restriction when ephemeral ports are exhausted. Commit aacd9289af8b82f5fb01bcdd53d0e3406d1333c7 ("tcp: bind() use stronger condition for bind_conflict") introduced a restriction to forbid to bind SO_REUSEADDR enabled sockets to the same (addr, port) tuple in order to assign ports dispersedly so that we can connect to the same remote host. The change results in accelerating port depletion so that we fail to bind sockets to the same local port even if we want to connect to the different remote hosts. You can reproduce this issue by following instructions below. 1. # sysctl -w net.ipv4.ip_local_port_range="32768 32768" 2. set SO_REUSEADDR to two sockets. 3. bind two sockets to (localhost, 0) and the latter fails. Therefore, when ephemeral ports are exhausted, bind(0) should fallback to the legacy behaviour to enable the SO_REUSEADDR option and make it possible to connect to different remote (addr, port) tuples. This patch allows us to bind SO_REUSEADDR enabled sockets to the same (addr, port) only when net.ipv4.ip_autobind_reuse is set 1 and all ephemeral ports are exhausted. This also allows connect() and listen() to share ports in the following way and may break some applications. So the ip_autobind_reuse is 0 by default and disables the feature. 1. setsockopt(sk1, SO_REUSEADDR) 2. setsockopt(sk2, SO_REUSEADDR) 3. bind(sk1, saddr, 0) 4. bind(sk2, saddr, 0) 5. connect(sk1, daddr) 6. listen(sk2) If it is set 1, we can fully utilize the 4-tuples, but we should use IP_BIND_ADDRESS_NO_PORT for bind()+connect() as possible. The notable thing is that if all sockets bound to the same port have both SO_REUSEADDR and SO_REUSEPORT enabled, we can bind sockets to an ephemeral port and also do listen(). Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-03-10 17:05:25 +09:00
tcp: bind(0) remove the SO_REUSEADDR restriction when ephemeral ports are exhausted. Commit aacd9289af8b82f5fb01bcdd53d0e3406d1333c7 ("tcp: bind() use stronger condition for bind_conflict") introduced a restriction to forbid to bind SO_REUSEADDR enabled sockets to the same (addr, port) tuple in order to assign ports dispersedly so that we can connect to the same remote host. The change results in accelerating port depletion so that we fail to bind sockets to the same local port even if we want to connect to the different remote hosts. You can reproduce this issue by following instructions below. 1. # sysctl -w net.ipv4.ip_local_port_range="32768 32768" 2. set SO_REUSEADDR to two sockets. 3. bind two sockets to (localhost, 0) and the latter fails. Therefore, when ephemeral ports are exhausted, bind(0) should fallback to the legacy behaviour to enable the SO_REUSEADDR option and make it possible to connect to different remote (addr, port) tuples. This patch allows us to bind SO_REUSEADDR enabled sockets to the same (addr, port) only when net.ipv4.ip_autobind_reuse is set 1 and all ephemeral ports are exhausted. This also allows connect() and listen() to share ports in the following way and may break some applications. So the ip_autobind_reuse is 0 by default and disables the feature. 1. setsockopt(sk1, SO_REUSEADDR) 2. setsockopt(sk2, SO_REUSEADDR) 3. bind(sk1, saddr, 0) 4. bind(sk2, saddr, 0) 5. connect(sk1, daddr) 6. listen(sk2) If it is set 1, we can fully utilize the 4-tuples, but we should use IP_BIND_ADDRESS_NO_PORT for bind()+connect() as possible. The notable thing is that if all sockets bound to the same port have both SO_REUSEADDR and SO_REUSEPORT enabled, we can bind sockets to an ephemeral port and also do listen(). Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-03-10 17:05:25 +09:00
2025-10-27 10:22:30 +02:00
net: arp: introduce arp_evict_nocarrier sysctl parameter This change introduces a new sysctl parameter, arp_evict_nocarrier. When set (default) the ARP cache will be cleared on a NOCARRIER event. This new option has been defaulted to '1' which maintains existing behavior. Clearing the ARP cache on NOCARRIER is relatively new, introduced by: commit 859bd2ef1fc1110a8031b967ee656c53a6260a76 Author: David Ahern <dsahern@gmail.com> Date: Thu Oct 11 20:33:49 2018 -0700 net: Evict neighbor entries on carrier down The reason for this changes is to prevent the ARP cache from being cleared when a wireless device roams. Specifically for wireless roams the ARP cache should not be cleared because the underlying network has not changed. Clearing the ARP cache in this case can introduce significant delays sending out packets after a roam. A user reported such a situation here: https://lore.kernel.org/linux-wireless/CACsRnHWa47zpx3D1oDq9JYnZWniS8yBwW1h0WAVZ6vrbwL_S0w@mail.gmail.com/ After some investigation it was found that the kernel was holding onto packets until ARP finished which resulted in this 1 second delay. It was also found that the first ARP who-has was never responded to, which is actually what caues the delay. This change is more or less working around this behavior, but again, there is no reason to clear the cache on a roam anyways. As for the unanswered who-has, we know the packet made it OTA since it was seen while monitoring. Why it never received a response is unknown. In any case, since this is a problem on the AP side of things all that can be done is to work around it until it is solved. Some background on testing/reproducing the packet delay: Hardware: - 2 access points configured for Fast BSS Transition (Though I don't see why regular reassociation wouldn't have the same behavior) - Wireless station running IWD as supplicant - A device on network able to respond to pings (I used one of the APs) Procedure: - Connect to first AP - Ping once to establish an ARP entry - Start a tcpdump - Roam to second AP - Wait for operstate UP event, and note the timestamp - Start pinging Results: Below is the tcpdump after UP. It was recorded the interface went UP at 10:42:01.432875. 10:42:01.461871 ARP, Request who-has 192.168.254.1 tell 192.168.254.71, length 28 10:42:02.497976 ARP, Request who-has 192.168.254.1 tell 192.168.254.71, length 28 10:42:02.507162 ARP, Reply 192.168.254.1 is-at ac:86:74:55:b0:20, length 46 10:42:02.507185 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 1, length 64 10:42:02.507205 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 2, length 64 10:42:02.507212 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 3, length 64 10:42:02.507219 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 4, length 64 10:42:02.507225 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 5, length 64 10:42:02.507232 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 6, length 64 10:42:02.515373 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 1, length 64 10:42:02.521399 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 2, length 64 10:42:02.521612 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 3, length 64 10:42:02.521941 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 4, length 64 10:42:02.522419 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 5, length 64 10:42:02.523085 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 6, length 64 You can see the first ARP who-has went out very quickly after UP, but was never responded to. Nearly a second later the kernel retries and gets a response. Only then do the ping packets go out. If an ARP entry is manually added prior to UP (after the cache is cleared) it is seen that the first ping is never responded to, so its not only an issue with ARP but with data packets in general. As mentioned prior, the wireless interface was also monitored to verify the ping/ARP packet made it OTA which was observed to be true. Signed-off-by: James Prestwood <prestwoj@gmail.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-11-01 10:36:28 -07:00
net: arp: introduce arp_evict_nocarrier sysctl parameter This change introduces a new sysctl parameter, arp_evict_nocarrier. When set (default) the ARP cache will be cleared on a NOCARRIER event. This new option has been defaulted to '1' which maintains existing behavior. Clearing the ARP cache on NOCARRIER is relatively new, introduced by: commit 859bd2ef1fc1110a8031b967ee656c53a6260a76 Author: David Ahern <dsahern@gmail.com> Date: Thu Oct 11 20:33:49 2018 -0700 net: Evict neighbor entries on carrier down The reason for this changes is to prevent the ARP cache from being cleared when a wireless device roams. Specifically for wireless roams the ARP cache should not be cleared because the underlying network has not changed. Clearing the ARP cache in this case can introduce significant delays sending out packets after a roam. A user reported such a situation here: https://lore.kernel.org/linux-wireless/CACsRnHWa47zpx3D1oDq9JYnZWniS8yBwW1h0WAVZ6vrbwL_S0w@mail.gmail.com/ After some investigation it was found that the kernel was holding onto packets until ARP finished which resulted in this 1 second delay. It was also found that the first ARP who-has was never responded to, which is actually what caues the delay. This change is more or less working around this behavior, but again, there is no reason to clear the cache on a roam anyways. As for the unanswered who-has, we know the packet made it OTA since it was seen while monitoring. Why it never received a response is unknown. In any case, since this is a problem on the AP side of things all that can be done is to work around it until it is solved. Some background on testing/reproducing the packet delay: Hardware: - 2 access points configured for Fast BSS Transition (Though I don't see why regular reassociation wouldn't have the same behavior) - Wireless station running IWD as supplicant - A device on network able to respond to pings (I used one of the APs) Procedure: - Connect to first AP - Ping once to establish an ARP entry - Start a tcpdump - Roam to second AP - Wait for operstate UP event, and note the timestamp - Start pinging Results: Below is the tcpdump after UP. It was recorded the interface went UP at 10:42:01.432875. 10:42:01.461871 ARP, Request who-has 192.168.254.1 tell 192.168.254.71, length 28 10:42:02.497976 ARP, Request who-has 192.168.254.1 tell 192.168.254.71, length 28 10:42:02.507162 ARP, Reply 192.168.254.1 is-at ac:86:74:55:b0:20, length 46 10:42:02.507185 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 1, length 64 10:42:02.507205 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 2, length 64 10:42:02.507212 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 3, length 64 10:42:02.507219 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 4, length 64 10:42:02.507225 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 5, length 64 10:42:02.507232 IP 192.168.254.71 > 192.168.254.1: ICMP echo request, id 52792, seq 6, length 64 10:42:02.515373 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 1, length 64 10:42:02.521399 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 2, length 64 10:42:02.521612 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 3, length 64 10:42:02.521941 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 4, length 64 10:42:02.522419 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 5, length 64 10:42:02.523085 IP 192.168.254.1 > 192.168.254.71: ICMP echo reply, id 52792, seq 6, length 64 You can see the first ARP who-has went out very quickly after UP, but was never responded to. Nearly a second later the kernel retries and gets a response. Only then do the ping packets go out. If an ARP entry is manually added prior to UP (after the cache is cleared) it is seen that the first ping is never responded to, so its not only an issue with ARP but with data packets in general. As mentioned prior, the wireless interface was also monitored to verify the ping/ARP packet made it OTA which was observed to be true. Signed-off-by: James Prestwood <prestwoj@gmail.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-11-01 10:36:28 -07:00
net: evaluate net.ipvX.conf.all.ignore_routes_with_linkdown Introduced in 0eeb075fad73, the "ignore_routes_with_linkdown" sysctl ignores a route whose interface is down. It is provided as a per-interface sysctl. However, while a "all" variant is exposed, it was a noop since it was never evaluated. We use the usual "or" logic for this kind of sysctls. Tested with: ip link add type veth # veth0 + veth1 ip link add type veth # veth1 + veth2 ip link set up dev veth0 ip link set up dev veth1 # link-status paired with veth0 ip link set up dev veth2 ip link set up dev veth3 # link-status paired with veth2 # First available path ip -4 addr add 203.0.113.${uts#H}/24 dev veth0 ip -6 addr add 2001:db8:1::${uts#H}/64 dev veth0 # Second available path ip -4 addr add 192.0.2.${uts#H}/24 dev veth2 ip -6 addr add 2001:db8:2::${uts#H}/64 dev veth2 # More specific route through first path ip -4 route add 198.51.100.0/25 via 203.0.113.254 # via veth0 ip -6 route add 2001:db8:3::/56 via 2001:db8:1::ff # via veth0 # Less specific route through second path ip -4 route add 198.51.100.0/24 via 192.0.2.254 # via veth2 ip -6 route add 2001:db8:3::/48 via 2001:db8:2::ff # via veth2 # H1: enable on "all" # H2: enable on "veth0" for v in ipv4 ipv6; do case $uts in H1) sysctl -qw net.${v}.conf.all.ignore_routes_with_linkdown=1 ;; H2) sysctl -qw net.${v}.conf.veth0.ignore_routes_with_linkdown=1 ;; esac done set -xe # When veth0 is up, best route is through veth0 ip -o route get 198.51.100.1 | grep -Fw veth0 ip -o route get 2001:db8:3::1 | grep -Fw veth0 # When veth0 is down, best route should be through veth2 on H1/H2, # but on veth0 on H2 ip link set down dev veth1 # down veth0 ip route show [ $uts != H3 ] || ip -o route get 198.51.100.1 | grep -Fw veth0 [ $uts != H3 ] || ip -o route get 2001:db8:3::1 | grep -Fw veth0 [ $uts = H3 ] || ip -o route get 198.51.100.1 | grep -Fw veth2 [ $uts = H3 ] || ip -o route get 2001:db8:3::1 | grep -Fw veth2 Without this patch, the two last lines would fail on H1 (the one using the "all" sysctl). With the patch, everything succeeds as expected. Also document the sysctl in `ip-sysctl.rst`. Fixes: 0eeb075fad73 ("net: ipv4 sysctl option to ignore routes when nexthop link is down") Signed-off-by: Vincent Bernat <vincent@bernat.ch> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-11-07 20:35:13 +01:00
ipv6: Implement automatic flow label generation on transmit Automatically generate flow labels for IPv6 packets on transmit. The flow label is computed based on skb_get_hash. The flow label will only automatically be set when it is zero otherwise (i.e. flow label manager hasn't set one). This supports the transmit side functionality of RFC 6438. Added an IPv6 sysctl auto_flowlabels to enable/disable this behavior system wide, and added IPV6_AUTOFLOWLABEL socket option to enable this functionality per socket. By default, auto flowlabels are disabled to avoid possible conflicts with flow label manager, however if this feature proves useful we may want to enable it by default. It should also be noted that FreeBSD has already implemented automatic flow labels (including the sysctl and socket option). In FreeBSD, automatic flow labels default to enabled. Performance impact: Running super_netperf with 200 flows for TCP_RR and UDP_RR for IPv6. Note that in UDP case, __skb_get_hash will be called for every packet with explains slight regression. In the TCP case the hash is saved in the socket so there is no regression. Automatic flow labels disabled: TCP_RR: 86.53% CPU utilization 127/195/322 90/95/99% latencies 1.40498e+06 tps UDP_RR: 90.70% CPU utilization 118/168/243 90/95/99% latencies 1.50309e+06 tps Automatic flow labels enabled: TCP_RR: 85.90% CPU utilization 128/199/337 90/95/99% latencies 1.40051e+06 UDP_RR 92.61% CPU utilization 115/164/236 90/95/99% latencies 1.4687e+06 Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-01 21:33:10 -07:00
ipv6: Implement automatic flow label generation on transmit Automatically generate flow labels for IPv6 packets on transmit. The flow label is computed based on skb_get_hash. The flow label will only automatically be set when it is zero otherwise (i.e. flow label manager hasn't set one). This supports the transmit side functionality of RFC 6438. Added an IPv6 sysctl auto_flowlabels to enable/disable this behavior system wide, and added IPV6_AUTOFLOWLABEL socket option to enable this functionality per socket. By default, auto flowlabels are disabled to avoid possible conflicts with flow label manager, however if this feature proves useful we may want to enable it by default. It should also be noted that FreeBSD has already implemented automatic flow labels (including the sysctl and socket option). In FreeBSD, automatic flow labels default to enabled. Performance impact: Running super_netperf with 200 flows for TCP_RR and UDP_RR for IPv6. Note that in UDP case, __skb_get_hash will be called for every packet with explains slight regression. In the TCP case the hash is saved in the socket so there is no regression. Automatic flow labels disabled: TCP_RR: 86.53% CPU utilization 127/195/322 90/95/99% latencies 1.40498e+06 tps UDP_RR: 90.70% CPU utilization 118/168/243 90/95/99% latencies 1.50309e+06 tps Automatic flow labels enabled: TCP_RR: 85.90% CPU utilization 128/199/337 90/95/99% latencies 1.40051e+06 UDP_RR 92.61% CPU utilization 115/164/236 90/95/99% latencies 1.4687e+06 Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-01 21:33:10 -07:00
ipv6: Flow label state ranges This patch divides the IPv6 flow label space into two ranges: 0-7ffff is reserved for flow label manager, 80000-fffff will be used for creating auto flow labels (per RFC6438). This only affects how labels are set on transmit, it does not affect receive. This range split can be disbaled by systcl. Background: IPv6 flow labels have been an unmitigated disappointment thus far in the lifetime of IPv6. Support in HW devices to use them for ECMP is lacking, and OSes don't turn them on by default. If we had these we could get much better hashing in IPv6 networks without resorting to DPI, possibly eliminating some of the motivations to to define new encaps in UDP just for getting ECMP. Unfortunately, the initial specfications of IPv6 did not clarify how they are to be used. There has always been a vague concept that these can be used for ECMP, flow hashing, etc. and we do now have a good standard how to this in RFC6438. The problem is that flow labels can be either stateful or stateless (as in RFC6438), and we are presented with the possibility that a stateless label may collide with a stateful one. Attempts to split the flow label space were rejected in IETF. When we added support in Linux for RFC6438, we could not turn on flow labels by default due to this conflict. This patch splits the flow label space and should give us a path to enabling auto flow labels by default for all IPv6 packets. This is an API change so we need to consider compatibility with existing deployment. The stateful range is chosen to be the lower values in hopes that most uses would have chosen small numbers. Once we resolve the stateless/stateful issue, we can proceed to look at enabling RFC6438 flow labels by default (starting with scaled testing). Signed-off-by: Tom Herbert <tom@herbertland.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-29 15:33:21 -07:00
ipv6: Flow label state ranges This patch divides the IPv6 flow label space into two ranges: 0-7ffff is reserved for flow label manager, 80000-fffff will be used for creating auto flow labels (per RFC6438). This only affects how labels are set on transmit, it does not affect receive. This range split can be disbaled by systcl. Background: IPv6 flow labels have been an unmitigated disappointment thus far in the lifetime of IPv6. Support in HW devices to use them for ECMP is lacking, and OSes don't turn them on by default. If we had these we could get much better hashing in IPv6 networks without resorting to DPI, possibly eliminating some of the motivations to to define new encaps in UDP just for getting ECMP. Unfortunately, the initial specfications of IPv6 did not clarify how they are to be used. There has always been a vague concept that these can be used for ECMP, flow hashing, etc. and we do now have a good standard how to this in RFC6438. The problem is that flow labels can be either stateful or stateless (as in RFC6438), and we are presented with the possibility that a stateless label may collide with a stateful one. Attempts to split the flow label space were rejected in IETF. When we added support in Linux for RFC6438, we could not turn on flow labels by default due to this conflict. This patch splits the flow label space and should give us a path to enabling auto flow labels by default for all IPv6 packets. This is an API change so we need to consider compatibility with existing deployment. The stateful range is chosen to be the lower values in hopes that most uses would have chosen small numbers. Once we resolve the stateless/stateful issue, we can proceed to look at enabling RFC6438 flow labels by default (starting with scaled testing). Signed-off-by: Tom Herbert <tom@herbertland.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-29 15:33:21 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert <tom@quantonium.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-10-30 14:16:00 -07:00
net: ipv6: Emit notification when fib hardware flags are changed After installing a route to the kernel, user space receives an acknowledgment, which means the route was installed in the kernel, but not necessarily in hardware. The asynchronous nature of route installation in hardware can lead to a routing daemon advertising a route before it was actually installed in hardware. This can result in packet loss or mis-routed packets until the route is installed in hardware. It is also possible for a route already installed in hardware to change its action and therefore its flags. For example, a host route that is trapping packets can be "promoted" to perform decapsulation following the installation of an IPinIP/VXLAN tunnel. Emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/RTM_F_TRAP flags are changed. The aim is to provide an indication to user-space (e.g., routing daemons) about the state of the route in hardware. Introduce a sysctl that controls this behavior. Keep the default value at 0 (i.e., do not emit notifications) for several reasons: - Multiple RTM_NEWROUTE notification per-route might confuse existing routing daemons. - Convergence reasons in routing daemons. - The extra notifications will negatively impact the insertion rate. - Not all users are interested in these notifications. Move fib6_info_hw_flags_set() to C file because it is no longer a short function. Signed-off-by: Amit Cohen <amcohen@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-02-01 21:47:55 +02:00
net: ipv6: Emit notification when fib hardware flags are changed After installing a route to the kernel, user space receives an acknowledgment, which means the route was installed in the kernel, but not necessarily in hardware. The asynchronous nature of route installation in hardware can lead to a routing daemon advertising a route before it was actually installed in hardware. This can result in packet loss or mis-routed packets until the route is installed in hardware. It is also possible for a route already installed in hardware to change its action and therefore its flags. For example, a host route that is trapping packets can be "promoted" to perform decapsulation following the installation of an IPinIP/VXLAN tunnel. Emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/RTM_F_TRAP flags are changed. The aim is to provide an indication to user-space (e.g., routing daemons) about the state of the route in hardware. Introduce a sysctl that controls this behavior. Keep the default value at 0 (i.e., do not emit notifications) for several reasons: - Multiple RTM_NEWROUTE notification per-route might confuse existing routing daemons. - Convergence reasons in routing daemons. - The extra notifications will negatively impact the insertion rate. - Not all users are interested in these notifications. Move fib6_info_hw_flags_set() to C file because it is no longer a short function. Signed-off-by: Amit Cohen <amcohen@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-02-01 21:47:55 +02:00
net: ipv6: Emit notification when fib hardware flags are changed After installing a route to the kernel, user space receives an acknowledgment, which means the route was installed in the kernel, but not necessarily in hardware. The asynchronous nature of route installation in hardware can lead to a routing daemon advertising a route before it was actually installed in hardware. This can result in packet loss or mis-routed packets until the route is installed in hardware. It is also possible for a route already installed in hardware to change its action and therefore its flags. For example, a host route that is trapping packets can be "promoted" to perform decapsulation following the installation of an IPinIP/VXLAN tunnel. Emit RTM_NEWROUTE notifications whenever RTM_F_OFFLOAD/RTM_F_TRAP flags are changed. The aim is to provide an indication to user-space (e.g., routing daemons) about the state of the route in hardware. Introduce a sysctl that controls this behavior. Keep the default value at 0 (i.e., do not emit notifications) for several reasons: - Multiple RTM_NEWROUTE notification per-route might confuse existing routing daemons. - Convergence reasons in routing daemons. - The extra notifications will negatively impact the insertion rate. - Not all users are interested in these notifications. Move fib6_info_hw_flags_set() to C file because it is no longer a short function. Signed-off-by: Amit Cohen <amcohen@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-02-01 21:47:55 +02:00
ipv6: ioam: Documentation for new IOAM sysctls Add documentation for new IOAM sysctls: - ioam6_id and ioam6_id_wide: two per-namespace sysctls - ioam6_enabled, ioam6_id and ioam6_id_wide: three per-interface sysctls Example of IOAM configuration based on the following simple topology: _____ _____ _____ | | eth0 eth0 | | eth1 eth0 | | | A |.----------.| B |.----------.| C | |_____| |_____| |_____| 1) Node and interface IDs can be configured for IOAM: # IOAM ID of A = 1, IOAM ID of A.eth0 = 11 (A) sysctl -w net.ipv6.ioam6_id=1 (A) sysctl -w net.ipv6.conf.eth0.ioam6_id=11 # IOAM ID of B = 2, IOAM ID of B.eth0 = 21, IOAM ID of B.eth1 = 22 (B) sysctl -w net.ipv6.ioam6_id=2 (B) sysctl -w net.ipv6.conf.eth0.ioam6_id=21 (B) sysctl -w net.ipv6.conf.eth1.ioam6_id=22 # IOAM ID of C = 3, IOAM ID of C.eth0 = 31 (C) sysctl -w net.ipv6.ioam6_id=3 (C) sysctl -w net.ipv6.conf.eth0.ioam6_id=31 Note that "_wide" IDs equivalents can be configured the same way. 2) Each node can be configured to form an IOAM domain. For instance, we allow IOAM from A to C only (not the reverse path), i.e. enable IOAM on ingress for B.eth0 and C.eth0: (B) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 (C) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 3) An IOAM domain (e.g. ID=123) is defined and made known to each node: (A) ip ioam namespace add 123 (B) ip ioam namespace add 123 (C) ip ioam namespace add 123 4) Finally, an IOAM Pre-allocated Trace can be inserted in traffic sent by A when C (e.g. db02::2) is the destination: (A) ip -6 route add db02::2/128 encap ioam6 trace type 0x800000 ns 123 size 12 dev eth0 Signed-off-by: Justin Iurman <justin.iurman@uliege.be> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-07-20 21:43:00 +02:00
ipv6: ioam: Documentation for new IOAM sysctls Add documentation for new IOAM sysctls: - ioam6_id and ioam6_id_wide: two per-namespace sysctls - ioam6_enabled, ioam6_id and ioam6_id_wide: three per-interface sysctls Example of IOAM configuration based on the following simple topology: _____ _____ _____ | | eth0 eth0 | | eth1 eth0 | | | A |.----------.| B |.----------.| C | |_____| |_____| |_____| 1) Node and interface IDs can be configured for IOAM: # IOAM ID of A = 1, IOAM ID of A.eth0 = 11 (A) sysctl -w net.ipv6.ioam6_id=1 (A) sysctl -w net.ipv6.conf.eth0.ioam6_id=11 # IOAM ID of B = 2, IOAM ID of B.eth0 = 21, IOAM ID of B.eth1 = 22 (B) sysctl -w net.ipv6.ioam6_id=2 (B) sysctl -w net.ipv6.conf.eth0.ioam6_id=21 (B) sysctl -w net.ipv6.conf.eth1.ioam6_id=22 # IOAM ID of C = 3, IOAM ID of C.eth0 = 31 (C) sysctl -w net.ipv6.ioam6_id=3 (C) sysctl -w net.ipv6.conf.eth0.ioam6_id=31 Note that "_wide" IDs equivalents can be configured the same way. 2) Each node can be configured to form an IOAM domain. For instance, we allow IOAM from A to C only (not the reverse path), i.e. enable IOAM on ingress for B.eth0 and C.eth0: (B) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 (C) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 3) An IOAM domain (e.g. ID=123) is defined and made known to each node: (A) ip ioam namespace add 123 (B) ip ioam namespace add 123 (C) ip ioam namespace add 123 4) Finally, an IOAM Pre-allocated Trace can be inserted in traffic sent by A when C (e.g. db02::2) is the destination: (A) ip -6 route add db02::2/128 encap ioam6 trace type 0x800000 ns 123 size 12 dev eth0 Signed-off-by: Justin Iurman <justin.iurman@uliege.be> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-07-20 21:43:00 +02:00
ipv6: ioam: Documentation for new IOAM sysctls Add documentation for new IOAM sysctls: - ioam6_id and ioam6_id_wide: two per-namespace sysctls - ioam6_enabled, ioam6_id and ioam6_id_wide: three per-interface sysctls Example of IOAM configuration based on the following simple topology: _____ _____ _____ | | eth0 eth0 | | eth1 eth0 | | | A |.----------.| B |.----------.| C | |_____| |_____| |_____| 1) Node and interface IDs can be configured for IOAM: # IOAM ID of A = 1, IOAM ID of A.eth0 = 11 (A) sysctl -w net.ipv6.ioam6_id=1 (A) sysctl -w net.ipv6.conf.eth0.ioam6_id=11 # IOAM ID of B = 2, IOAM ID of B.eth0 = 21, IOAM ID of B.eth1 = 22 (B) sysctl -w net.ipv6.ioam6_id=2 (B) sysctl -w net.ipv6.conf.eth0.ioam6_id=21 (B) sysctl -w net.ipv6.conf.eth1.ioam6_id=22 # IOAM ID of C = 3, IOAM ID of C.eth0 = 31 (C) sysctl -w net.ipv6.ioam6_id=3 (C) sysctl -w net.ipv6.conf.eth0.ioam6_id=31 Note that "_wide" IDs equivalents can be configured the same way. 2) Each node can be configured to form an IOAM domain. For instance, we allow IOAM from A to C only (not the reverse path), i.e. enable IOAM on ingress for B.eth0 and C.eth0: (B) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 (C) sysctl -w net.ipv6.conf.eth0.ioam6_enabled=1 3) An IOAM domain (e.g. ID=123) is defined and made known to each node: (A) ip ioam namespace add 123 (B) ip ioam namespace add 123 (C) ip ioam namespace add 123 4) Finally, an IOAM Pre-allocated Trace can be inserted in traffic sent by A when C (e.g. db02::2) is the destination: (A) ip -6 route add db02::2/128 encap ioam6 trace type 0x800000 ns 123 size 12 dev eth0 Signed-off-by: Justin Iurman <justin.iurman@uliege.be> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-07-20 21:43:00 +02:00
ipv6: Send ICMPv6 RSes only when RAs are accepted This patch improves the logic determining when to send ICMPv6 Router Solicitations, so that they are 1) always sent when the kernel is accepting Router Advertisements, and 2) never sent when the kernel is not accepting RAs. In other words, the operational setting of the "accept_ra" sysctl is used. The change also makes the special "Hybrid Router" forwarding mode ("forwarding" sysctl set to 2) operate exactly the same as the standard Router mode (forwarding=1). The only difference between the two was that RSes was being sent in the Hybrid Router mode only. The sysctl documentation describing the special Hybrid Router mode has therefore been removed. Rationale for the change: Currently, the value of forwarding sysctl is the only thing determining whether or not to send RSes. If it has the value 0 or 2, they are sent, otherwise they are not. This leads to inconsistent behaviour in the following cases: * accept_ra=0, forwarding=0 * accept_ra=0, forwarding=2 * accept_ra=1, forwarding=2 * accept_ra=2, forwarding=1 In the first three cases, the kernel will send RSes, even though it will not accept any RAs received in reply. In the last case, it will not send any RSes, even though it will accept and process any RAs received. (Most routers will send unsolicited RAs periodically, so suppressing RSes in the last case will merely delay auto-configuration, not prevent it.) Also, it is my opinion that having the forwarding sysctl control RS sending behaviour (completely independent of whether RAs are being accepted or not) is simply not what most users would intuitively expect to be the case. Signed-off-by: Tore Anderson <tore@fud.no> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-28 23:47:33 +00:00
net: allow user to set metric on default route learned via Router Advertisement For IPv4, default route is learned via DHCPv4 and user is allowed to change metric using config etc/network/interfaces. But for IPv6, default route can be learned via RA, for which, currently a fixed metric value 1024 is used. Ideally, user should be able to configure metric on default route for IPv6 similar to IPv4. This patch adds sysctl for the same. Logs: For IPv4: Config in etc/network/interfaces: auto eth0 iface eth0 inet dhcp metric 4261413864 IPv4 Kernel Route Table: $ ip route list default via 172.21.47.1 dev eth0 metric 4261413864 FRR Table, if a static route is configured: [In real scenario, it is useful to prefer BGP learned default route over DHCPv4 default route.] Codes: K - kernel route, C - connected, S - static, R - RIP, O - OSPF, I - IS-IS, B - BGP, P - PIM, E - EIGRP, N - NHRP, T - Table, v - VNC, V - VNC-Direct, A - Babel, D - SHARP, > - selected route, * - FIB route S>* 0.0.0.0/0 [20/0] is directly connected, eth0, 00:00:03 K 0.0.0.0/0 [254/1000] via 172.21.47.1, eth0, 6d08h51m i.e. User can prefer Default Router learned via Routing Protocol in IPv4. Similar behavior is not possible for IPv6, without this fix. After fix [for IPv6]: sudo sysctl -w net.ipv6.conf.eth0.net.ipv6.conf.eth0.ra_defrtr_metric=1996489705 IP monitor: [When IPv6 RA is received] default via fe80::xx16:xxxx:feb3:ce8e dev eth0 proto ra metric 1996489705 pref high Kernel IPv6 routing table $ ip -6 route list default via fe80::be16:65ff:feb3:ce8e dev eth0 proto ra metric 1996489705 expires 21sec hoplimit 64 pref high FRR Table, if a static route is configured: [In real scenario, it is useful to prefer BGP learned default route over IPv6 RA default route.] Codes: K - kernel route, C - connected, S - static, R - RIPng, O - OSPFv3, I - IS-IS, B - BGP, N - NHRP, T - Table, v - VNC, V - VNC-Direct, A - Babel, D - SHARP, > - selected route, * - FIB route S>* ::/0 [20/0] is directly connected, eth0, 00:00:06 K ::/0 [119/1001] via fe80::xx16:xxxx:feb3:ce8e, eth0, 6d07h43m If the metric is changed later, the effect will be seen only when next IPv6 RA is received, because the default route must be fully controlled by RA msg. Below metric is changed from 1996489705 to 1996489704. $ sudo sysctl -w net.ipv6.conf.eth0.ra_defrtr_metric=1996489704 net.ipv6.conf.eth0.ra_defrtr_metric = 1996489704 IP monitor: [On next IPv6 RA msg, Kernel deletes prev route and installs new route with updated metric] Deleted default via fe80::xx16:xxxx:feb3:ce8e dev eth0 proto ra metric 1996489705 expires 3sec hoplimit 64 pref high default via fe80::xx16:xxxx:feb3:ce8e dev eth0 proto ra metric 1996489704 pref high Signed-off-by: Praveen Chaudhary <pchaudhary@linkedin.com> Signed-off-by: Zhenggen Xu <zxu@linkedin.com> Reviewed-by: David Ahern <dsahern@kernel.org> Link: https://lore.kernel.org/r/20210125214430.24079-1-pchaudhary@linkedin.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-01-25 13:44:30 -08:00
net: change accept_ra_min_rtr_lft to affect all RA lifetimes accept_ra_min_rtr_lft only considered the lifetime of the default route and discarded entire RAs accordingly. This change renames accept_ra_min_rtr_lft to accept_ra_min_lft, and applies the value to individual RA sections; in particular, router lifetime, PIO preferred lifetime, and RIO lifetime. If any of those lifetimes are lower than the configured value, the specific RA section is ignored. In order for the sysctl to be useful to Android, it should really apply to all lifetimes in the RA, since that is what determines the minimum frequency at which RAs must be processed by the kernel. Android uses hardware offloads to drop RAs for a fraction of the minimum of all lifetimes present in the RA (some networks have very frequent RAs (5s) with high lifetimes (2h)). Despite this, we have encountered networks that set the router lifetime to 30s which results in very frequent CPU wakeups. Instead of disabling IPv6 (and dropping IPv6 ethertype in the WiFi firmware) entirely on such networks, it seems better to ignore the misconfigured routers while still processing RAs from other IPv6 routers on the same network (i.e. to support IoT applications). The previous implementation dropped the entire RA based on router lifetime. This turned out to be hard to expand to the other lifetimes present in the RA in a consistent manner; dropping the entire RA based on RIO/PIO lifetimes would essentially require parsing the whole thing twice. Fixes: 1671bcfd76fd ("net: add sysctl accept_ra_min_rtr_lft") Cc: Lorenzo Colitti <lorenzo@google.com> Signed-off-by: Patrick Rohr <prohr@google.com> Reviewed-by: Maciej Żenczykowski <maze@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Link: https://lore.kernel.org/r/20230726230701.919212-1-prohr@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-07-26 16:07:01 -07:00
net: change accept_ra_min_rtr_lft to affect all RA lifetimes accept_ra_min_rtr_lft only considered the lifetime of the default route and discarded entire RAs accordingly. This change renames accept_ra_min_rtr_lft to accept_ra_min_lft, and applies the value to individual RA sections; in particular, router lifetime, PIO preferred lifetime, and RIO lifetime. If any of those lifetimes are lower than the configured value, the specific RA section is ignored. In order for the sysctl to be useful to Android, it should really apply to all lifetimes in the RA, since that is what determines the minimum frequency at which RAs must be processed by the kernel. Android uses hardware offloads to drop RAs for a fraction of the minimum of all lifetimes present in the RA (some networks have very frequent RAs (5s) with high lifetimes (2h)). Despite this, we have encountered networks that set the router lifetime to 30s which results in very frequent CPU wakeups. Instead of disabling IPv6 (and dropping IPv6 ethertype in the WiFi firmware) entirely on such networks, it seems better to ignore the misconfigured routers while still processing RAs from other IPv6 routers on the same network (i.e. to support IoT applications). The previous implementation dropped the entire RA based on router lifetime. This turned out to be hard to expand to the other lifetimes present in the RA in a consistent manner; dropping the entire RA based on RIO/PIO lifetimes would essentially require parsing the whole thing twice. Fixes: 1671bcfd76fd ("net: add sysctl accept_ra_min_rtr_lft") Cc: Lorenzo Colitti <lorenzo@google.com> Signed-off-by: Patrick Rohr <prohr@google.com> Reviewed-by: Maciej Żenczykowski <maze@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Link: https://lore.kernel.org/r/20230726230701.919212-1-prohr@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-07-26 16:07:01 -07:00
ipv6: Send ICMPv6 RSes only when RAs are accepted This patch improves the logic determining when to send ICMPv6 Router Solicitations, so that they are 1) always sent when the kernel is accepting Router Advertisements, and 2) never sent when the kernel is not accepting RAs. In other words, the operational setting of the "accept_ra" sysctl is used. The change also makes the special "Hybrid Router" forwarding mode ("forwarding" sysctl set to 2) operate exactly the same as the standard Router mode (forwarding=1). The only difference between the two was that RSes was being sent in the Hybrid Router mode only. The sysctl documentation describing the special Hybrid Router mode has therefore been removed. Rationale for the change: Currently, the value of forwarding sysctl is the only thing determining whether or not to send RSes. If it has the value 0 or 2, they are sent, otherwise they are not. This leads to inconsistent behaviour in the following cases: * accept_ra=0, forwarding=0 * accept_ra=0, forwarding=2 * accept_ra=1, forwarding=2 * accept_ra=2, forwarding=1 In the first three cases, the kernel will send RSes, even though it will not accept any RAs received in reply. In the last case, it will not send any RSes, even though it will accept and process any RAs received. (Most routers will send unsolicited RAs periodically, so suppressing RSes in the last case will merely delay auto-configuration, not prevent it.) Also, it is my opinion that having the forwarding sysctl control RS sending behaviour (completely independent of whether RAs are being accepted or not) is simply not what most users would intuitively expect to be the case. Signed-off-by: Tore Anderson <tore@fud.no> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-28 23:47:33 +00:00
ipv6: Send ICMPv6 RSes only when RAs are accepted This patch improves the logic determining when to send ICMPv6 Router Solicitations, so that they are 1) always sent when the kernel is accepting Router Advertisements, and 2) never sent when the kernel is not accepting RAs. In other words, the operational setting of the "accept_ra" sysctl is used. The change also makes the special "Hybrid Router" forwarding mode ("forwarding" sysctl set to 2) operate exactly the same as the standard Router mode (forwarding=1). The only difference between the two was that RSes was being sent in the Hybrid Router mode only. The sysctl documentation describing the special Hybrid Router mode has therefore been removed. Rationale for the change: Currently, the value of forwarding sysctl is the only thing determining whether or not to send RSes. If it has the value 0 or 2, they are sent, otherwise they are not. This leads to inconsistent behaviour in the following cases: * accept_ra=0, forwarding=0 * accept_ra=0, forwarding=2 * accept_ra=1, forwarding=2 * accept_ra=2, forwarding=1 In the first three cases, the kernel will send RSes, even though it will not accept any RAs received in reply. In the last case, it will not send any RSes, even though it will accept and process any RAs received. (Most routers will send unsolicited RAs periodically, so suppressing RSes in the last case will merely delay auto-configuration, not prevent it.) Also, it is my opinion that having the forwarding sysctl control RS sending behaviour (completely independent of whether RAs are being accepted or not) is simply not what most users would intuitively expect to be the case. Signed-off-by: Tore Anderson <tore@fud.no> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-28 23:47:33 +00:00
net: ipv6/addrconf: clamp preferred_lft to the minimum required If the preferred lifetime was less than the minimum required lifetime, ipv6_create_tempaddr would error out without creating any new address. On my machine and network, this error happened immediately with the preferred lifetime set to 5 seconds or less, after a few minutes with the preferred lifetime set to 6 seconds, and not at all with the preferred lifetime set to 7 seconds. During my investigation, I found a Stack Exchange post from another person who seems to have had the same problem: They stopped getting new addresses if they lowered the preferred lifetime below 3 seconds, and they didn't really know why. The preferred lifetime is a preference, not a hard requirement. The kernel does not strictly forbid new connections on a deprecated address, nor does it guarantee that the address will be disposed of the instant its total valid lifetime expires. So rather than disable IPv6 privacy extensions altogether if the minimum required lifetime swells above the preferred lifetime, it is more in keeping with the user's intent to increase the temporary address's lifetime to the minimum necessary for the current network conditions. With these fixes, setting the preferred lifetime to 5 or 6 seconds "just works" because the extra fraction of a second is practically unnoticeable. It's even possible to reduce the time before deprecation to 1 or 2 seconds by setting /proc/sys/net/ipv6/conf/*/regen_min_advance and /proc/sys/net/ipv6/conf/*/dad_transmits to 0. I realize that that is a pretty niche use case, but I know at least one person who would gladly sacrifice performance and convenience to be sure that they are getting the maximum possible level of privacy. Link: https://serverfault.com/a/1031168/310447 Signed-off-by: Alex Henrie <alexhenrie24@gmail.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2024-02-13 23:26:32 -07:00
net: ipv6: Make address flushing on ifdown optional Currently, all ipv6 addresses are flushed when the interface is configured down, including global, static addresses: $ ip -6 addr show dev eth1 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 2100:1::2/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe79:34bd/64 scope link valid_lft forever preferred_lft forever $ ip link set dev eth1 down $ ip -6 addr show dev eth1 << nothing; all addresses have been flushed>> Add a new sysctl to make this behavior optional. The new setting defaults to flush all addresses to maintain backwards compatibility. When the set global addresses with no expire times are not flushed on an admin down. The sysctl is per-interface or system-wide for all interfaces $ sysctl -w net.ipv6.conf.eth1.keep_addr_on_down=1 or $ sysctl -w net.ipv6.conf.all.keep_addr_on_down=1 Will keep addresses on eth1 on an admin down. $ ip -6 addr show dev eth1 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 2100:1::2/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe79:34bd/64 scope link valid_lft forever preferred_lft forever $ ip link set dev eth1 down $ ip -6 addr show dev eth1 3: eth1: <BROADCAST,MULTICAST> mtu 1500 state DOWN qlen 1000 inet6 2100:1::2/120 scope global tentative valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe79:34bd/64 scope link tentative valid_lft forever preferred_lft forever Signed-off-by: David Ahern <dsa@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-24 09:25:37 -08:00
net: ipv6: Make address flushing on ifdown optional Currently, all ipv6 addresses are flushed when the interface is configured down, including global, static addresses: $ ip -6 addr show dev eth1 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 2100:1::2/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe79:34bd/64 scope link valid_lft forever preferred_lft forever $ ip link set dev eth1 down $ ip -6 addr show dev eth1 << nothing; all addresses have been flushed>> Add a new sysctl to make this behavior optional. The new setting defaults to flush all addresses to maintain backwards compatibility. When the set global addresses with no expire times are not flushed on an admin down. The sysctl is per-interface or system-wide for all interfaces $ sysctl -w net.ipv6.conf.eth1.keep_addr_on_down=1 or $ sysctl -w net.ipv6.conf.all.keep_addr_on_down=1 Will keep addresses on eth1 on an admin down. $ ip -6 addr show dev eth1 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000 inet6 2100:1::2/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe79:34bd/64 scope link valid_lft forever preferred_lft forever $ ip link set dev eth1 down $ ip -6 addr show dev eth1 3: eth1: <BROADCAST,MULTICAST> mtu 1500 state DOWN qlen 1000 inet6 2100:1::2/120 scope global tentative valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe79:34bd/64 scope link tentative valid_lft forever preferred_lft forever Signed-off-by: David Ahern <dsa@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-24 09:25:37 -08:00
net: ipv6: sysctl to specify IPv6 ND traffic class Add a per-device sysctl to specify the default traffic class to use for kernel originated IPv6 Neighbour Discovery packets. Currently this includes: - Router Solicitation (ICMPv6 type 133) ndisc_send_rs() -> ndisc_send_skb() -> ip6_nd_hdr() - Neighbour Solicitation (ICMPv6 type 135) ndisc_send_ns() -> ndisc_send_skb() -> ip6_nd_hdr() - Neighbour Advertisement (ICMPv6 type 136) ndisc_send_na() -> ndisc_send_skb() -> ip6_nd_hdr() - Redirect (ICMPv6 type 137) ndisc_send_redirect() -> ndisc_send_skb() -> ip6_nd_hdr() and if the kernel ever gets around to generating RA's, it would presumably also include: - Router Advertisement (ICMPv6 type 134) (radvd daemon could pick up on the kernel setting and use it) Interface drivers may examine the Traffic Class value and translate the DiffServ Code Point into a link-layer appropriate traffic prioritization scheme. An example of mapping IETF DSCP values to IEEE 802.11 User Priority values can be found here: https://tools.ietf.org/html/draft-ietf-tsvwg-ieee-802-11 The expected primary use case is to properly prioritize ND over wifi. Testing: jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 0 jzem22:~# echo -1 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass -bash: echo: write error: Invalid argument jzem22:~# echo 256 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass -bash: echo: write error: Invalid argument jzem22:~# echo 0 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# echo 255 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 255 jzem22:~# echo 34 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 34 jzem22:~# echo $[0xDC] > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# tcpdump -v -i eth0 icmp6 and src host jzem22.pgc and dst host fe80::1 tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 262144 bytes IP6 (class 0xdc, hlim 255, next-header ICMPv6 (58) payload length: 24) jzem22.pgc > fe80::1: [icmp6 sum ok] ICMP6, neighbor advertisement, length 24, tgt is jzem22.pgc, Flags [solicited] (based on original change written by Erik Kline, with minor changes) v2: fix 'suspicious rcu_dereference_check() usage' by explicitly grabbing the rcu_read_lock. Cc: Lorenzo Colitti <lorenzo@google.com> Signed-off-by: Erik Kline <ek@google.com> Signed-off-by: Maciej Żenczykowski <maze@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-07 21:52:09 -08:00
net: ipv6: sysctl to specify IPv6 ND traffic class Add a per-device sysctl to specify the default traffic class to use for kernel originated IPv6 Neighbour Discovery packets. Currently this includes: - Router Solicitation (ICMPv6 type 133) ndisc_send_rs() -> ndisc_send_skb() -> ip6_nd_hdr() - Neighbour Solicitation (ICMPv6 type 135) ndisc_send_ns() -> ndisc_send_skb() -> ip6_nd_hdr() - Neighbour Advertisement (ICMPv6 type 136) ndisc_send_na() -> ndisc_send_skb() -> ip6_nd_hdr() - Redirect (ICMPv6 type 137) ndisc_send_redirect() -> ndisc_send_skb() -> ip6_nd_hdr() and if the kernel ever gets around to generating RA's, it would presumably also include: - Router Advertisement (ICMPv6 type 134) (radvd daemon could pick up on the kernel setting and use it) Interface drivers may examine the Traffic Class value and translate the DiffServ Code Point into a link-layer appropriate traffic prioritization scheme. An example of mapping IETF DSCP values to IEEE 802.11 User Priority values can be found here: https://tools.ietf.org/html/draft-ietf-tsvwg-ieee-802-11 The expected primary use case is to properly prioritize ND over wifi. Testing: jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 0 jzem22:~# echo -1 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass -bash: echo: write error: Invalid argument jzem22:~# echo 256 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass -bash: echo: write error: Invalid argument jzem22:~# echo 0 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# echo 255 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 255 jzem22:~# echo 34 > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# cat /proc/sys/net/ipv6/conf/eth0/ndisc_tclass 34 jzem22:~# echo $[0xDC] > /proc/sys/net/ipv6/conf/eth0/ndisc_tclass jzem22:~# tcpdump -v -i eth0 icmp6 and src host jzem22.pgc and dst host fe80::1 tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 262144 bytes IP6 (class 0xdc, hlim 255, next-header ICMPv6 (58) payload length: 24) jzem22.pgc > fe80::1: [icmp6 sum ok] ICMP6, neighbor advertisement, length 24, tgt is jzem22.pgc, Flags [solicited] (based on original change written by Erik Kline, with minor changes) v2: fix 'suspicious rcu_dereference_check() usage' by explicitly grabbing the rcu_read_lock. Cc: Lorenzo Colitti <lorenzo@google.com> Signed-off-by: Erik Kline <ek@google.com> Signed-off-by: Maciej Żenczykowski <maze@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-07 21:52:09 -08:00
net: ipv6: Add a sysctl to make optimistic addresses useful candidates Add a sysctl that causes an interface's optimistic addresses to be considered equivalent to other non-deprecated addresses for source address selection purposes. Preferred addresses will still take precedence over optimistic addresses, subject to other ranking in the source address selection algorithm. This is useful where different interfaces are connected to different networks from different ISPs (e.g., a cell network and a home wifi network). The current behaviour complies with RFC 3484/6724, and it makes sense if the host has only one interface, or has multiple interfaces on the same network (same or cooperating administrative domain(s), but not in the multiple distinct networks case. For example, if a mobile device has an IPv6 address on an LTE network and then connects to IPv6-enabled wifi, while the wifi IPv6 address is undergoing DAD, IPv6 connections will try use the wifi default route with the LTE IPv6 address, and will get stuck until they time out. Also, because optimistic nodes can receive frames, issue an RTM_NEWADDR as soon as DAD starts (with the IFA_F_OPTIMSTIC flag appropriately set). A second RTM_NEWADDR is sent if DAD completes (the address flags have changed), otherwise an RTM_DELADDR is sent. Also: add an entry in ip-sysctl.txt for optimistic_dad. Signed-off-by: Erik Kline <ek@google.com> Acked-by: Lorenzo Colitti <lorenzo@google.com> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-28 18:11:14 +09:00
net: ipv6: Add a sysctl to make optimistic addresses useful candidates Add a sysctl that causes an interface's optimistic addresses to be considered equivalent to other non-deprecated addresses for source address selection purposes. Preferred addresses will still take precedence over optimistic addresses, subject to other ranking in the source address selection algorithm. This is useful where different interfaces are connected to different networks from different ISPs (e.g., a cell network and a home wifi network). The current behaviour complies with RFC 3484/6724, and it makes sense if the host has only one interface, or has multiple interfaces on the same network (same or cooperating administrative domain(s), but not in the multiple distinct networks case. For example, if a mobile device has an IPv6 address on an LTE network and then connects to IPv6-enabled wifi, while the wifi IPv6 address is undergoing DAD, IPv6 connections will try use the wifi default route with the LTE IPv6 address, and will get stuck until they time out. Also, because optimistic nodes can receive frames, issue an RTM_NEWADDR as soon as DAD starts (with the IFA_F_OPTIMSTIC flag appropriately set). A second RTM_NEWADDR is sent if DAD completes (the address flags have changed), otherwise an RTM_DELADDR is sent. Also: add an entry in ip-sysctl.txt for optimistic_dad. Signed-off-by: Erik Kline <ek@google.com> Acked-by: Lorenzo Colitti <lorenzo@google.com> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-28 18:11:14 +09:00
net: ipv6: Add a sysctl to make optimistic addresses useful candidates Add a sysctl that causes an interface's optimistic addresses to be considered equivalent to other non-deprecated addresses for source address selection purposes. Preferred addresses will still take precedence over optimistic addresses, subject to other ranking in the source address selection algorithm. This is useful where different interfaces are connected to different networks from different ISPs (e.g., a cell network and a home wifi network). The current behaviour complies with RFC 3484/6724, and it makes sense if the host has only one interface, or has multiple interfaces on the same network (same or cooperating administrative domain(s), but not in the multiple distinct networks case. For example, if a mobile device has an IPv6 address on an LTE network and then connects to IPv6-enabled wifi, while the wifi IPv6 address is undergoing DAD, IPv6 connections will try use the wifi default route with the LTE IPv6 address, and will get stuck until they time out. Also, because optimistic nodes can receive frames, issue an RTM_NEWADDR as soon as DAD starts (with the IFA_F_OPTIMSTIC flag appropriately set). A second RTM_NEWADDR is sent if DAD completes (the address flags have changed), otherwise an RTM_DELADDR is sent. Also: add an entry in ip-sysctl.txt for optimistic_dad. Signed-off-by: Erik Kline <ek@google.com> Acked-by: Lorenzo Colitti <lorenzo@google.com> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-28 18:11:14 +09:00
net: ipv6: Add a sysctl to make optimistic addresses useful candidates Add a sysctl that causes an interface's optimistic addresses to be considered equivalent to other non-deprecated addresses for source address selection purposes. Preferred addresses will still take precedence over optimistic addresses, subject to other ranking in the source address selection algorithm. This is useful where different interfaces are connected to different networks from different ISPs (e.g., a cell network and a home wifi network). The current behaviour complies with RFC 3484/6724, and it makes sense if the host has only one interface, or has multiple interfaces on the same network (same or cooperating administrative domain(s), but not in the multiple distinct networks case. For example, if a mobile device has an IPv6 address on an LTE network and then connects to IPv6-enabled wifi, while the wifi IPv6 address is undergoing DAD, IPv6 connections will try use the wifi default route with the LTE IPv6 address, and will get stuck until they time out. Also, because optimistic nodes can receive frames, issue an RTM_NEWADDR as soon as DAD starts (with the IFA_F_OPTIMSTIC flag appropriately set). A second RTM_NEWADDR is sent if DAD completes (the address flags have changed), otherwise an RTM_DELADDR is sent. Also: add an entry in ip-sysctl.txt for optimistic_dad. Signed-off-by: Erik Kline <ek@google.com> Acked-by: Lorenzo Colitti <lorenzo@google.com> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-28 18:11:14 +09:00
ipv6: icmp: Add RFC 5837 support Add the ability to append the incoming IP interface information to ICMPv6 error messages in accordance with RFC 5837 and RFC 4884. This is required for more meaningful traceroute results in unnumbered networks. The feature is disabled by default and controlled via a new sysctl ("net.ipv6.icmp.errors_extension_mask") which accepts a bitmask of ICMP extensions to append to ICMP error messages. Currently, only a single value is supported, but the interface and the implementation should be able to support more extensions, if needed. Clone the skb and copy the relevant data portions before modifying the skb as the caller of icmp6_send() still owns the skb after the function returns. This should be fine since by default ICMP error messages are rate limited to 1000 per second and no more than 1 per second per specific host. Trim or pad the packet to 128 bytes before appending the ICMP extension structure in order to be compatible with legacy applications that assume that the ICMP extension structure always starts at this offset (the minimum length specified by RFC 4884). Since commit 20e1954fe238 ("ipv6: RFC 4884 partial support for SIT/GRE tunnels") it is possible for icmp6_send() to be called with an skb that already contains ICMP extensions. This can happen when we receive an ICMPv4 message with extensions from a tunnel and translate it to an ICMPv6 message towards an IPv6 host in the overlay network. I could not find an RFC that supports this behavior, but it makes sense to not overwrite the original extensions that were appended to the packet. Therefore, avoid appending extensions if the length field in the provided ICMPv6 header is already filled. Export netdev_copy_name() using EXPORT_IPV6_MOD_GPL() to make it available to IPv6 when it is built as a module. Reviewed-by: Petr Machata <petrm@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20251027082232.232571-3-idosch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-10-27 10:22:31 +02:00
sctp: add pf_expose per netns and sock and asoc As said in rfc7829, section 3, point 12: The SCTP stack SHOULD expose the PF state of its destination addresses to the ULP as well as provide the means to notify the ULP of state transitions of its destination addresses from active to PF, and vice versa. However, it is recommended that an SCTP stack implementing SCTP-PF also allows for the ULP to be kept ignorant of the PF state of its destinations and the associated state transitions, thus allowing for retention of the simpler state transition model of [RFC4960] in the ULP. Not only does it allow to expose the PF state to ULP, but also allow to ignore sctp-pf to ULP. So this patch is to add pf_expose per netns, sock and asoc. And in sctp_assoc_control_transport(), ulp_notify will be set to false if asoc->expose is not 'enabled' in next patch. It also allows a user to change pf_expose per netns by sysctl, and pf_expose per sock and asoc will be initialized with it. Note that pf_expose also works for SCTP_GET_PEER_ADDR_INFO sockopt, to not allow a user to query the state of a sctp-pf peer address when pf_expose is 'disabled', as said in section 7.3. v1->v2: - Fix a build warning noticed by Nathan Chancellor. v2->v3: - set pf_expose to UNUSED by default to keep compatible with old applications. v3->v4: - add a new entry for pf_expose on ip-sysctl.txt, as Marcelo suggested. - change this patch to 1/5, and move sctp_assoc_control_transport change into 2/5, as Marcelo suggested. - use SCTP_PF_EXPOSE_UNSET instead of SCTP_PF_EXPOSE_UNUSED, and set SCTP_PF_EXPOSE_UNSET to 0 in enum, as Marcelo suggested. Signed-off-by: Xin Long <lucien.xin@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:20:32 +08:00
sctp: add pf_expose per netns and sock and asoc As said in rfc7829, section 3, point 12: The SCTP stack SHOULD expose the PF state of its destination addresses to the ULP as well as provide the means to notify the ULP of state transitions of its destination addresses from active to PF, and vice versa. However, it is recommended that an SCTP stack implementing SCTP-PF also allows for the ULP to be kept ignorant of the PF state of its destinations and the associated state transitions, thus allowing for retention of the simpler state transition model of [RFC4960] in the ULP. Not only does it allow to expose the PF state to ULP, but also allow to ignore sctp-pf to ULP. So this patch is to add pf_expose per netns, sock and asoc. And in sctp_assoc_control_transport(), ulp_notify will be set to false if asoc->expose is not 'enabled' in next patch. It also allows a user to change pf_expose per netns by sysctl, and pf_expose per sock and asoc will be initialized with it. Note that pf_expose also works for SCTP_GET_PEER_ADDR_INFO sockopt, to not allow a user to query the state of a sctp-pf peer address when pf_expose is 'disabled', as said in section 7.3. v1->v2: - Fix a build warning noticed by Nathan Chancellor. v2->v3: - set pf_expose to UNUSED by default to keep compatible with old applications. v3->v4: - add a new entry for pf_expose on ip-sysctl.txt, as Marcelo suggested. - change this patch to 1/5, and move sctp_assoc_control_transport change into 2/5, as Marcelo suggested. - use SCTP_PF_EXPOSE_UNSET instead of SCTP_PF_EXPOSE_UNUSED, and set SCTP_PF_EXPOSE_UNSET to 0 in enum, as Marcelo suggested. Signed-off-by: Xin Long <lucien.xin@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:20:32 +08:00
sctp: add pf_expose per netns and sock and asoc As said in rfc7829, section 3, point 12: The SCTP stack SHOULD expose the PF state of its destination addresses to the ULP as well as provide the means to notify the ULP of state transitions of its destination addresses from active to PF, and vice versa. However, it is recommended that an SCTP stack implementing SCTP-PF also allows for the ULP to be kept ignorant of the PF state of its destinations and the associated state transitions, thus allowing for retention of the simpler state transition model of [RFC4960] in the ULP. Not only does it allow to expose the PF state to ULP, but also allow to ignore sctp-pf to ULP. So this patch is to add pf_expose per netns, sock and asoc. And in sctp_assoc_control_transport(), ulp_notify will be set to false if asoc->expose is not 'enabled' in next patch. It also allows a user to change pf_expose per netns by sysctl, and pf_expose per sock and asoc will be initialized with it. Note that pf_expose also works for SCTP_GET_PEER_ADDR_INFO sockopt, to not allow a user to query the state of a sctp-pf peer address when pf_expose is 'disabled', as said in section 7.3. v1->v2: - Fix a build warning noticed by Nathan Chancellor. v2->v3: - set pf_expose to UNUSED by default to keep compatible with old applications. v3->v4: - add a new entry for pf_expose on ip-sysctl.txt, as Marcelo suggested. - change this patch to 1/5, and move sctp_assoc_control_transport change into 2/5, as Marcelo suggested. - use SCTP_PF_EXPOSE_UNSET instead of SCTP_PF_EXPOSE_UNUSED, and set SCTP_PF_EXPOSE_UNSET to 0 in enum, as Marcelo suggested. Signed-off-by: Xin Long <lucien.xin@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:20:32 +08:00
sctp: add pf_expose per netns and sock and asoc As said in rfc7829, section 3, point 12: The SCTP stack SHOULD expose the PF state of its destination addresses to the ULP as well as provide the means to notify the ULP of state transitions of its destination addresses from active to PF, and vice versa. However, it is recommended that an SCTP stack implementing SCTP-PF also allows for the ULP to be kept ignorant of the PF state of its destinations and the associated state transitions, thus allowing for retention of the simpler state transition model of [RFC4960] in the ULP. Not only does it allow to expose the PF state to ULP, but also allow to ignore sctp-pf to ULP. So this patch is to add pf_expose per netns, sock and asoc. And in sctp_assoc_control_transport(), ulp_notify will be set to false if asoc->expose is not 'enabled' in next patch. It also allows a user to change pf_expose per netns by sysctl, and pf_expose per sock and asoc will be initialized with it. Note that pf_expose also works for SCTP_GET_PEER_ADDR_INFO sockopt, to not allow a user to query the state of a sctp-pf peer address when pf_expose is 'disabled', as said in section 7.3. v1->v2: - Fix a build warning noticed by Nathan Chancellor. v2->v3: - set pf_expose to UNUSED by default to keep compatible with old applications. v3->v4: - add a new entry for pf_expose on ip-sysctl.txt, as Marcelo suggested. - change this patch to 1/5, and move sctp_assoc_control_transport change into 2/5, as Marcelo suggested. - use SCTP_PF_EXPOSE_UNSET instead of SCTP_PF_EXPOSE_UNUSED, and set SCTP_PF_EXPOSE_UNSET to 0 in enum, as Marcelo suggested. Signed-off-by: Xin Long <lucien.xin@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:20:32 +08:00
sctp: Convert cookie authentication to use HMAC-SHA256 Convert SCTP cookies to use HMAC-SHA256, instead of the previous choice of the legacy algorithms HMAC-MD5 and HMAC-SHA1. Simplify and optimize the code by using the HMAC-SHA256 library instead of crypto_shash, and by preparing the HMAC key when it is generated instead of per-operation. This doesn't break compatibility, since the cookie format is an implementation detail, not part of the SCTP protocol itself. Note that the cookie size doesn't change either. The HMAC field was already 32 bytes, even though previously at most 20 bytes were actually compared. 32 bytes exactly fits an untruncated HMAC-SHA256 value. So, although we could safely truncate the MAC to something slightly shorter, for now just keep the cookie size the same. I also considered SipHash, but that would generate only 8-byte MACs. An 8-byte MAC *might* suffice here. However, there's quite a lot of information in the SCTP cookies: more than in TCP SYN cookies. So absent an analysis that occasional forgeries of all that information is okay in SCTP, I errored on the side of caution. Remove HMAC-MD5 and HMAC-SHA1 as options, since the new HMAC-SHA256 option is just better. It's faster as well as more secure. For example, benchmarking on x86_64, cookie authentication is now nearly 3x as fast as the previous default choice and implementation of HMAC-MD5. Also just make the kernel always support cookie authentication if SCTP is supported at all, rather than making it optional in the build. (It was sort of optional before, but it didn't really work properly. E.g., a kernel with CONFIG_SCTP_COOKIE_HMAC_MD5=n still supported HMAC-MD5 cookie authentication if CONFIG_CRYPTO_HMAC and CONFIG_CRYPTO_MD5 happened to be enabled in the kconfig for other reasons.) Acked-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Eric Biggers <ebiggers@kernel.org> Link: https://patch.msgid.link/20250818205426.30222-5-ebiggers@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-08-18 13:54:25 -07:00
sctp: Convert cookie authentication to use HMAC-SHA256 Convert SCTP cookies to use HMAC-SHA256, instead of the previous choice of the legacy algorithms HMAC-MD5 and HMAC-SHA1. Simplify and optimize the code by using the HMAC-SHA256 library instead of crypto_shash, and by preparing the HMAC key when it is generated instead of per-operation. This doesn't break compatibility, since the cookie format is an implementation detail, not part of the SCTP protocol itself. Note that the cookie size doesn't change either. The HMAC field was already 32 bytes, even though previously at most 20 bytes were actually compared. 32 bytes exactly fits an untruncated HMAC-SHA256 value. So, although we could safely truncate the MAC to something slightly shorter, for now just keep the cookie size the same. I also considered SipHash, but that would generate only 8-byte MACs. An 8-byte MAC *might* suffice here. However, there's quite a lot of information in the SCTP cookies: more than in TCP SYN cookies. So absent an analysis that occasional forgeries of all that information is okay in SCTP, I errored on the side of caution. Remove HMAC-MD5 and HMAC-SHA1 as options, since the new HMAC-SHA256 option is just better. It's faster as well as more secure. For example, benchmarking on x86_64, cookie authentication is now nearly 3x as fast as the previous default choice and implementation of HMAC-MD5. Also just make the kernel always support cookie authentication if SCTP is supported at all, rather than making it optional in the build. (It was sort of optional before, but it didn't really work properly. E.g., a kernel with CONFIG_SCTP_COOKIE_HMAC_MD5=n still supported HMAC-MD5 cookie authentication if CONFIG_CRYPTO_HMAC and CONFIG_CRYPTO_MD5 happened to be enabled in the kconfig for other reasons.) Acked-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: Eric Biggers <ebiggers@kernel.org> Link: https://patch.msgid.link/20250818205426.30222-5-ebiggers@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-08-18 13:54:25 -07:00