diff options
Diffstat (limited to 'net/ipv4')
128 files changed, 14655 insertions, 7382 deletions
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index f96bd489b362..e983bb0c5012 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -10,11 +10,11 @@ config IP_MULTICAST intend to participate in the MBONE, a high bandwidth network on top of the Internet which carries audio and video broadcasts. More information about the MBONE is on the WWW at - <http://www.savetz.com/mbone/>. For most people, it's safe to say N. + <https://www.savetz.com/mbone/>. For most people, it's safe to say N. config IP_ADVANCED_ROUTER bool "IP: advanced router" - ---help--- + help If you intend to run your Linux box mostly as a router, i.e. as a computer that forwards and redistributes network packets, say Y; you will then be presented with several options that allow more precise @@ -49,14 +49,14 @@ config IP_ADVANCED_ROUTER Note that some distributions enable it in startup scripts. For details about rp_filter strict and loose mode read - <file:Documentation/networking/ip-sysctl.txt>. + <file:Documentation/networking/ip-sysctl.rst>. If unsure, say N here. config IP_FIB_TRIE_STATS bool "FIB TRIE statistics" depends on IP_ADVANCED_ROUTER - ---help--- + help Keep track of statistics on structure of FIB TRIE table. Useful for testing and measuring TRIE performance. @@ -64,7 +64,7 @@ config IP_MULTIPLE_TABLES bool "IP: policy routing" depends on IP_ADVANCED_ROUTER select FIB_RULES - ---help--- + help Normally, a router decides what to do with a received packet based solely on the packet's final destination address. If you say Y here, the Linux router will also be able to take the packet's source @@ -73,7 +73,7 @@ config IP_MULTIPLE_TABLES If you need more information, see the Linux Advanced Routing and Traffic Control documentation at - <http://lartc.org/howto/lartc.rpdb.html> + <https://lartc.org/howto/lartc.rpdb.html> If unsure, say N. @@ -117,7 +117,7 @@ config IP_PNP config IP_PNP_DHCP bool "IP: DHCP support" depends on IP_PNP - ---help--- + help If you want your Linux box to mount its whole root file system (the one containing the directory /) from some other computer over the net via NFS and you want the IP address of your computer to be @@ -129,12 +129,12 @@ config IP_PNP_DHCP If unsure, say Y. Note that if you want to use DHCP, a DHCP server must be operating on your network. Read - <file:Documentation/filesystems/nfs/nfsroot.txt> for details. + <file:Documentation/admin-guide/nfs/nfsroot.rst> for details. config IP_PNP_BOOTP bool "IP: BOOTP support" depends on IP_PNP - ---help--- + help If you want your Linux box to mount its whole root file system (the one containing the directory /) from some other computer over the net via NFS and you want the IP address of your computer to be @@ -144,7 +144,7 @@ config IP_PNP_BOOTP does BOOTP itself, providing all necessary information on the kernel command line, you can say N here. If unsure, say Y. Note that if you want to use BOOTP, a BOOTP server must be operating on your network. - Read <file:Documentation/filesystems/nfs/nfsroot.txt> for details. + Read <file:Documentation/admin-guide/nfs/nfsroot.rst> for details. config IP_PNP_RARP bool "IP: RARP support" @@ -157,13 +157,13 @@ config IP_PNP_RARP older protocol which is being obsoleted by BOOTP and DHCP), say Y here. Note that if you want to use RARP, a RARP server must be operating on your network. Read - <file:Documentation/filesystems/nfs/nfsroot.txt> for details. + <file:Documentation/admin-guide/nfs/nfsroot.rst> for details. config NET_IPIP tristate "IP: tunneling" select INET_TUNNEL select NET_IP_TUNNEL - ---help--- + help Tunneling means encapsulating data of one protocol type within another protocol and sending it over a channel that understands the encapsulating protocol. This particular tunneling driver implements @@ -267,7 +267,7 @@ config IP_PIMSM_V2 config SYN_COOKIES bool "IP: TCP syncookie support" - ---help--- + help Normal TCP/IP networking is open to an attack known as "SYN flooding". This denial-of-service attack prevents legitimate remote users from being able to connect to your computer during an ongoing @@ -280,7 +280,7 @@ config SYN_COOKIES continue to connect, even when your machine is under attack. There is no need for the legitimate users to change their TCP/IP software; SYN cookies work transparently to them. For technical information - about SYN cookies, check out <http://cr.yp.to/syncookies.html>. + about SYN cookies, check out <https://cr.yp.to/syncookies.html>. If you are SYN flooded, the source address reported by the kernel is likely to have been forged by the attacker; it is only reported as @@ -303,10 +303,11 @@ config SYN_COOKIES config NET_IPVTI tristate "Virtual (secure) IP: tunneling" + depends on IPV6 || IPV6=n select INET_TUNNEL select NET_IP_TUNNEL select XFRM - ---help--- + help Tunneling means encapsulating data of one protocol type within another protocol and sending it over a channel that understands the encapsulating protocol. This can be used with xfrm mode tunnel to give @@ -320,9 +321,8 @@ config NET_UDP_TUNNEL config NET_FOU tristate "IP: Foo (IP protocols) over UDP" - select XFRM select NET_UDP_TUNNEL - ---help--- + help Foo over UDP allows any IP protocol to be directly encapsulated over UDP include tunnels (IPIP, GRE, SIT). By encapsulating in UDP network mechanisms and optimizations for UDP (such as ECMP @@ -332,36 +332,38 @@ config NET_FOU_IP_TUNNELS bool "IP: FOU encapsulation of IP tunnels" depends on NET_IPIP || NET_IPGRE || IPV6_SIT select NET_FOU - ---help--- + help Allow configuration of FOU or GUE encapsulation for IP tunnels. When this option is enabled IP tunnels can be configured to use FOU or GUE encapsulation. config INET_AH tristate "IP: AH transformation" - select XFRM_ALGO - select CRYPTO - select CRYPTO_HMAC - select CRYPTO_MD5 - select CRYPTO_SHA1 - ---help--- - Support for IPsec AH. + select XFRM_AH + help + Support for IPsec AH (Authentication Header). + + AH can be used with various authentication algorithms. Besides + enabling AH support itself, this option enables the generic + implementations of the algorithms that RFC 8221 lists as MUST be + implemented. If you need any other algorithms, you'll need to enable + them in the crypto API. You should also enable accelerated + implementations of any needed algorithms when available. If unsure, say Y. config INET_ESP tristate "IP: ESP transformation" - select XFRM_ALGO - select CRYPTO - select CRYPTO_AUTHENC - select CRYPTO_HMAC - select CRYPTO_MD5 - select CRYPTO_CBC - select CRYPTO_SHA1 - select CRYPTO_DES - select CRYPTO_ECHAINIV - ---help--- - Support for IPsec ESP. + select XFRM_ESP + help + Support for IPsec ESP (Encapsulating Security Payload). + + ESP can be used with various encryption and authentication algorithms. + Besides enabling ESP support itself, this option enables the generic + implementations of the algorithms that RFC 8221 lists as MUST be + implemented. If you need any other algorithms, you'll need to enable + them in the crypto API. You should also enable accelerated + implementations of any needed algorithms when available. If unsure, say Y. @@ -370,7 +372,7 @@ config INET_ESP_OFFLOAD depends on INET_ESP select XFRM_OFFLOAD default n - ---help--- + help Support for ESP transformation offload. This makes sense only if this system really does IPsec and want to do it with high throughput. A typical desktop system does not @@ -383,6 +385,7 @@ config INET_ESPINTCP depends on XFRM && INET_ESP select STREAM_PARSER select NET_SOCK_MSG + select XFRM_ESPINTCP help Support for RFC 8229 encapsulation of ESP and IKE over TCP/IPv4 sockets. @@ -393,7 +396,7 @@ config INET_IPCOMP tristate "IP: IPComp transformation" select INET_XFRM_TUNNEL select XFRM_IPCOMP - ---help--- + help Support for IP Payload Compression Protocol (IPComp) (RFC3173), typically needed for IPsec. @@ -411,7 +414,7 @@ config INET_TUNNEL config INET_DIAG tristate "INET: socket monitoring interface" default y - ---help--- + help Support for INET (TCP, DCCP, etc) socket monitoring interface used by native Linux tools such as ss. ss is included in iproute2, currently downloadable at: @@ -428,7 +431,7 @@ config INET_UDP_DIAG tristate "UDP: socket monitoring interface" depends on INET_DIAG && (IPV6 || IPV6=n) default n - ---help--- + help Support for UDP socket monitoring interface used by the ss tool. If unsure, say Y. @@ -436,7 +439,7 @@ config INET_RAW_DIAG tristate "RAW: socket monitoring interface" depends on INET_DIAG && (IPV6 || IPV6=n) default n - ---help--- + help Support for RAW socket monitoring interface used by the ss tool. If unsure, say Y. @@ -444,7 +447,7 @@ config INET_DIAG_DESTROY bool "INET: allow privileged process to administratively close sockets" depends on INET_DIAG default n - ---help--- + help Provides a SOCK_DESTROY operation that allows privileged processes (e.g., a connection manager or a network administration tool such as ss) to close sockets opened by other processes. Closing a socket in @@ -455,7 +458,7 @@ config INET_DIAG_DESTROY menuconfig TCP_CONG_ADVANCED bool "TCP: advanced congestion control" - ---help--- + help Support for selection of various TCP congestion control modules. @@ -469,7 +472,7 @@ if TCP_CONG_ADVANCED config TCP_CONG_BIC tristate "Binary Increase Congestion (BIC) control" default m - ---help--- + help BIC-TCP is a sender-side only change that ensures a linear RTT fairness under large windows while offering both scalability and bounded TCP-friendliness. The protocol combines two schemes @@ -483,7 +486,7 @@ config TCP_CONG_BIC config TCP_CONG_CUBIC tristate "CUBIC TCP" default y - ---help--- + help This is version 2.0 of BIC-TCP which uses a cubic growth function among other techniques. See http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/cubic-paper.pdf @@ -491,7 +494,7 @@ config TCP_CONG_CUBIC config TCP_CONG_WESTWOOD tristate "TCP Westwood+" default m - ---help--- + help TCP Westwood+ is a sender-side only modification of the TCP Reno protocol stack that optimizes the performance of TCP congestion control. It is based on end-to-end bandwidth estimation to set @@ -505,7 +508,7 @@ config TCP_CONG_WESTWOOD config TCP_CONG_HTCP tristate "H-TCP" default m - ---help--- + help H-TCP is a send-side only modifications of the TCP Reno protocol stack that optimizes the performance of TCP congestion control for high speed network links. It uses a @@ -516,17 +519,17 @@ config TCP_CONG_HTCP config TCP_CONG_HSTCP tristate "High Speed TCP" default n - ---help--- + help Sally Floyd's High Speed TCP (RFC 3649) congestion control. A modification to TCP's congestion control mechanism for use with large congestion windows. A table indicates how much to increase the congestion window by when an ACK is received. - For more detail see http://www.icir.org/floyd/hstcp.html + For more detail see https://www.icir.org/floyd/hstcp.html config TCP_CONG_HYBLA tristate "TCP-Hybla congestion control algorithm" default n - ---help--- + help TCP-Hybla is a sender-side only change that eliminates penalization of long-RTT, large-bandwidth connections, like when satellite legs are involved, especially when sharing a common bottleneck with normal @@ -535,7 +538,7 @@ config TCP_CONG_HYBLA config TCP_CONG_VEGAS tristate "TCP Vegas" default n - ---help--- + help TCP Vegas is a sender-side only change to TCP that anticipates the onset of congestion by estimating the bandwidth. TCP Vegas adjusts the sending rate by modifying the congestion @@ -545,7 +548,7 @@ config TCP_CONG_VEGAS config TCP_CONG_NV tristate "TCP NV" default n - ---help--- + help TCP NV is a follow up to TCP Vegas. It has been modified to deal with 10G networks, measurement noise introduced by LRO, GRO and interrupt coalescence. In addition, it will decrease its cwnd multiplicatively @@ -561,7 +564,7 @@ config TCP_CONG_NV config TCP_CONG_SCALABLE tristate "Scalable TCP" default n - ---help--- + help Scalable TCP is a sender-side only change to TCP which uses a MIMD congestion control algorithm which has some nice scaling properties, though is known to have fairness issues. @@ -570,7 +573,7 @@ config TCP_CONG_SCALABLE config TCP_CONG_LP tristate "TCP Low Priority" default n - ---help--- + help TCP Low Priority (TCP-LP), a distributed algorithm whose goal is to utilize only the excess network bandwidth as compared to the ``fair share`` of bandwidth as targeted by TCP. @@ -579,7 +582,7 @@ config TCP_CONG_LP config TCP_CONG_VENO tristate "TCP Veno" default n - ---help--- + help TCP Veno is a sender-side only enhancement of TCP to obtain better throughput over wireless networks. TCP Veno makes use of state distinguishing to circumvent the difficult judgment of the packet loss @@ -591,7 +594,7 @@ config TCP_CONG_YEAH tristate "YeAH TCP" select TCP_CONG_VEGAS default n - ---help--- + help YeAH-TCP is a sender-side high-speed enabled TCP congestion control algorithm, which uses a mixed loss/delay approach to compute the congestion window. It's design goals target high efficiency, @@ -604,7 +607,7 @@ config TCP_CONG_YEAH config TCP_CONG_ILLINOIS tristate "TCP Illinois" default n - ---help--- + help TCP-Illinois is a sender-side modification of TCP Reno for high speed long delay links. It uses round-trip-time to adjust the alpha and beta parameters to achieve a higher average @@ -616,7 +619,7 @@ config TCP_CONG_ILLINOIS config TCP_CONG_DCTCP tristate "DataCenter TCP (DCTCP)" default n - ---help--- + help DCTCP leverages Explicit Congestion Notification (ECN) in the network to provide multi-bit feedback to the end hosts. It is designed to provide: @@ -637,7 +640,7 @@ config TCP_CONG_DCTCP config TCP_CONG_CDG tristate "CAIA Delay-Gradient (CDG)" default n - ---help--- + help CAIA Delay-Gradient (CDG) is a TCP congestion control that modifies the TCP sender in order to: @@ -653,17 +656,17 @@ config TCP_CONG_CDG config TCP_CONG_BBR tristate "BBR TCP" default n - ---help--- + help BBR (Bottleneck Bandwidth and RTT) TCP congestion control aims to maximize network utilization and minimize queues. It builds an explicit - model of the the bottleneck delivery rate and path round-trip - propagation delay. It tolerates packet loss and delay unrelated to - congestion. It can operate over LAN, WAN, cellular, wifi, or cable - modem links. It can coexist with flows that use loss-based congestion - control, and can operate with shallow buffers, deep buffers, - bufferbloat, policers, or AQM schemes that do not provide a delay - signal. It requires the fq ("Fair Queue") pacing packet scheduler. + model of the bottleneck delivery rate and path round-trip propagation + delay. It tolerates packet loss and delay unrelated to congestion. It + can operate over LAN, WAN, cellular, wifi, or cable modem links. It can + coexist with flows that use loss-based congestion control, and can + operate with shallow buffers, deep buffers, bufferbloat, policers, or + AQM schemes that do not provide a delay signal. It requires the fq + ("Fair Queue") pacing packet scheduler. choice prompt "Default TCP congestion control" @@ -732,7 +735,7 @@ config TCP_MD5SIG bool "TCP: MD5 Signature Option support (RFC2385)" select CRYPTO select CRYPTO_MD5 - ---help--- + help RFC2385 specifies a method of giving MD5 protection to TCP sessions. Its main (only?) use is to protect BGP sessions between core routers on the Internet. diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 9d97bace13c8..bbdd9c44f14e 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -14,7 +14,7 @@ obj-y := route.o inetpeer.o protocol.o \ udp_offload.o arp.o icmp.o devinet.o af_inet.o igmp.o \ fib_frontend.o fib_semantics.o fib_trie.o fib_notifier.o \ inet_fragment.o ping.o ip_tunnel_core.o gre_offload.o \ - metrics.o netlink.o nexthop.o + metrics.o netlink.o nexthop.o udp_tunnel_stub.o obj-$(CONFIG_BPFILTER) += bpfilter/ @@ -29,6 +29,7 @@ gre-y := gre_demux.o obj-$(CONFIG_NET_FOU) += fou.o obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o obj-$(CONFIG_NET_IPGRE) += ip_gre.o +udp_tunnel-y := udp_tunnel_core.o udp_tunnel_nic.o obj-$(CONFIG_NET_UDP_TUNNEL) += udp_tunnel.o obj-$(CONFIG_NET_IPVTI) += ip_vti.o obj-$(CONFIG_SYN_COOKIES) += syncookies.o @@ -61,6 +62,7 @@ obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o obj-$(CONFIG_NET_SOCK_MSG) += tcp_bpf.o +obj-$(CONFIG_BPF_SYSCALL) += udp_bpf.o obj-$(CONFIG_NETLABEL) += cipso_ipv4.o obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 2fe295432c24..4728087c42a5 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -99,6 +99,7 @@ #include <net/route.h> #include <net/ip_fib.h> #include <net/inet_connection_sock.h> +#include <net/gro.h> #include <net/tcp.h> #include <net/udp.h> #include <net/udplite.h> @@ -116,6 +117,7 @@ #include <linux/mroute.h> #endif #include <net/l3mdev.h> +#include <net/compat.h> #include <trace/events/sock.h> @@ -132,13 +134,9 @@ void inet_sock_destruct(struct sock *sk) struct inet_sock *inet = inet_sk(sk); __skb_queue_purge(&sk->sk_receive_queue); - if (sk->sk_rx_skb_cache) { - __kfree_skb(sk->sk_rx_skb_cache); - sk->sk_rx_skb_cache = NULL; - } __skb_queue_purge(&sk->sk_error_queue); - sk_mem_reclaim(sk); + sk_mem_reclaim_final(sk); if (sk->sk_type == SOCK_STREAM && sk->sk_state != TCP_CLOSE) { pr_err("Attempt to release TCP socket in state %d %p\n", @@ -150,14 +148,14 @@ void inet_sock_destruct(struct sock *sk) return; } - WARN_ON(atomic_read(&sk->sk_rmem_alloc)); - WARN_ON(refcount_read(&sk->sk_wmem_alloc)); - WARN_ON(sk->sk_wmem_queued); - WARN_ON(sk->sk_forward_alloc); + WARN_ON_ONCE(atomic_read(&sk->sk_rmem_alloc)); + WARN_ON_ONCE(refcount_read(&sk->sk_wmem_alloc)); + WARN_ON_ONCE(sk->sk_wmem_queued); + WARN_ON_ONCE(sk_forward_alloc_get(sk)); kfree(rcu_dereference_protected(inet->inet_opt, 1)); dst_release(rcu_dereference_protected(sk->sk_dst_cache, 1)); - dst_release(sk->sk_rx_dst); + dst_release(rcu_dereference_protected(sk->sk_rx_dst, 1)); sk_refcnt_debug_dec(sk); } EXPORT_SYMBOL(inet_sock_destruct); @@ -219,7 +217,7 @@ int inet_listen(struct socket *sock, int backlog) * because the socket was in TCP_LISTEN state previously but * was shutdown() rather than close(). */ - tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen; + tcp_fastopen = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen); if ((tcp_fastopen & TFO_SERVER_WO_SOCKOPT1) && (tcp_fastopen & TFO_SERVER_ENABLE) && !inet_csk(sk)->icsk_accept_queue.fastopenq.max_qlen) { @@ -227,7 +225,7 @@ int inet_listen(struct socket *sock, int backlog) tcp_fastopen_init_key_once(sock_net(sk)); } - err = inet_csk_listen_start(sk, backlog); + err = inet_csk_listen_start(sk); if (err) goto out; tcp_call_bpf(sk, BPF_SOCK_OPS_TCP_LISTEN_CB, 0, NULL); @@ -317,7 +315,7 @@ lookup_protocol: WARN_ON(!answer_prot->slab); - err = -ENOBUFS; + err = -ENOMEM; sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot, kern); if (!sk) goto out; @@ -337,7 +335,7 @@ lookup_protocol: inet->hdrincl = 1; } - if (net->ipv4.sysctl_ip_no_pmtu_disc) + if (READ_ONCE(net->ipv4.sysctl_ip_no_pmtu_disc)) inet->pmtudisc = IP_PMTUDISC_DONT; else inet->pmtudisc = IP_PMTUDISC_WANT; @@ -410,6 +408,9 @@ int inet_release(struct socket *sock) if (sk) { long timeout; + if (!sk->sk_kern_sock) + BPF_CGROUP_RUN_PROG_INET_SOCK_RELEASE(sk); + /* Applications forget to leave groups before exiting */ ip_mc_drop_socket(sk); @@ -434,6 +435,7 @@ EXPORT_SYMBOL(inet_release); int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) { struct sock *sk = sock->sk; + u32 flags = BIND_WITH_LOCK; int err; /* If the socket has its own bind function then use it. (RAW) */ @@ -446,16 +448,17 @@ int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) /* BPF prog is run before any checks are done so that if the prog * changes context in a wrong way it will be caught. */ - err = BPF_CGROUP_RUN_PROG_INET4_BIND(sk, uaddr); + err = BPF_CGROUP_RUN_PROG_INET_BIND_LOCK(sk, uaddr, + CGROUP_INET4_BIND, &flags); if (err) return err; - return __inet_bind(sk, uaddr, addr_len, false, true); + return __inet_bind(sk, uaddr, addr_len, flags); } EXPORT_SYMBOL(inet_bind); int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, - bool force_bind_address_no_port, bool with_lock) + u32 flags) { struct sockaddr_in *addr = (struct sockaddr_in *)uaddr; struct inet_sock *inet = inet_sk(sk); @@ -486,16 +489,14 @@ int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, * is temporarily down) */ err = -EADDRNOTAVAIL; - if (!inet_can_nonlocal_bind(net, inet) && - addr->sin_addr.s_addr != htonl(INADDR_ANY) && - chk_addr_ret != RTN_LOCAL && - chk_addr_ret != RTN_MULTICAST && - chk_addr_ret != RTN_BROADCAST) + if (!inet_addr_valid_or_nonlocal(net, inet, addr->sin_addr.s_addr, + chk_addr_ret)) goto out; snum = ntohs(addr->sin_port); err = -EACCES; - if (snum && inet_port_requires_bind_service(net, snum) && + if (!(flags & BIND_NO_CAP_NET_BIND_SERVICE) && + snum && inet_port_requires_bind_service(net, snum) && !ns_capable(net->user_ns, CAP_NET_BIND_SERVICE)) goto out; @@ -506,7 +507,7 @@ int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, * would be illegal to use them (multicast/broadcast) in * which case the sending device address is used. */ - if (with_lock) + if (flags & BIND_WITH_LOCK) lock_sock(sk); /* Check these errors (active socket, double bind). */ @@ -520,16 +521,20 @@ int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, /* Make sure we are allowed to bind here. */ if (snum || !(inet->bind_address_no_port || - force_bind_address_no_port)) { + (flags & BIND_FORCE_ADDRESS_NO_PORT))) { if (sk->sk_prot->get_port(sk, snum)) { inet->inet_saddr = inet->inet_rcv_saddr = 0; err = -EADDRINUSE; goto out_release_sock; } - err = BPF_CGROUP_RUN_PROG_INET4_POST_BIND(sk); - if (err) { - inet->inet_saddr = inet->inet_rcv_saddr = 0; - goto out_release_sock; + if (!(flags & BIND_FROM_BPF)) { + err = BPF_CGROUP_RUN_PROG_INET4_POST_BIND(sk); + if (err) { + inet->inet_saddr = inet->inet_rcv_saddr = 0; + if (sk->sk_prot->put_port) + sk->sk_prot->put_port(sk); + goto out_release_sock; + } } } @@ -543,7 +548,7 @@ int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len, sk_dst_reset(sk); err = 0; out_release_sock: - if (with_lock) + if (flags & BIND_WITH_LOCK) release_sock(sk); out: return err; @@ -553,22 +558,27 @@ int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags) { struct sock *sk = sock->sk; + const struct proto *prot; int err; if (addr_len < sizeof(uaddr->sa_family)) return -EINVAL; + + /* IPV6_ADDRFORM can change sk->sk_prot under us. */ + prot = READ_ONCE(sk->sk_prot); + if (uaddr->sa_family == AF_UNSPEC) - return sk->sk_prot->disconnect(sk, flags); + return prot->disconnect(sk, flags); if (BPF_CGROUP_PRE_CONNECT_ENABLED(sk)) { - err = sk->sk_prot->pre_connect(sk, uaddr, addr_len); + err = prot->pre_connect(sk, uaddr, addr_len); if (err) return err; } - if (!inet_sk(sk)->inet_num && inet_autobind(sk)) + if (data_race(!inet_sk(sk)->inet_num) && inet_autobind(sk)) return -EAGAIN; - return sk->sk_prot->connect(sk, uaddr, addr_len); + return prot->connect(sk, uaddr, addr_len); } EXPORT_SYMBOL(inet_dgram_connect); @@ -729,10 +739,11 @@ EXPORT_SYMBOL(inet_stream_connect); int inet_accept(struct socket *sock, struct socket *newsock, int flags, bool kern) { - struct sock *sk1 = sock->sk; + struct sock *sk1 = sock->sk, *sk2; int err = -EINVAL; - struct sock *sk2 = sk1->sk_prot->accept(sk1, flags, &err, kern); + /* IPV6_ADDRFORM can change sk->sk_prot under us. */ + sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, flags, &err, kern); if (!sk2) goto do_err; @@ -743,6 +754,8 @@ int inet_accept(struct socket *sock, struct socket *newsock, int flags, (TCPF_ESTABLISHED | TCPF_SYN_RECV | TCPF_CLOSE_WAIT | TCPF_CLOSE))); + if (test_bit(SOCK_SUPPORT_ZC, &sock->flags)) + set_bit(SOCK_SUPPORT_ZC, &newsock->flags); sock_graft(sk2, newsock); newsock->state = SS_CONNECTED; @@ -753,32 +766,39 @@ do_err: } EXPORT_SYMBOL(inet_accept); - /* * This does both peername and sockname. */ int inet_getname(struct socket *sock, struct sockaddr *uaddr, - int peer) + int peer) { struct sock *sk = sock->sk; struct inet_sock *inet = inet_sk(sk); DECLARE_SOCKADDR(struct sockaddr_in *, sin, uaddr); sin->sin_family = AF_INET; + lock_sock(sk); if (peer) { if (!inet->inet_dport || (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_SYN_SENT)) && - peer == 1)) + peer == 1)) { + release_sock(sk); return -ENOTCONN; + } sin->sin_port = inet->inet_dport; sin->sin_addr.s_addr = inet->inet_daddr; + BPF_CGROUP_RUN_SA_PROG(sk, (struct sockaddr *)sin, + CGROUP_INET4_GETPEERNAME); } else { __be32 addr = inet->inet_rcv_saddr; if (!addr) addr = inet->inet_saddr; sin->sin_port = inet->inet_sport; sin->sin_addr.s_addr = addr; + BPF_CGROUP_RUN_SA_PROG(sk, (struct sockaddr *)sin, + CGROUP_INET4_GETSOCKNAME); } + release_sock(sk); memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); return sizeof(*sin); } @@ -789,7 +809,7 @@ int inet_send_prepare(struct sock *sk) sock_rps_record_flow(sk); /* We may need to bind the socket. */ - if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind && + if (data_race(!inet_sk(sk)->inet_num) && !sk->sk_prot->no_autobind && inet_autobind(sk)) return -EAGAIN; @@ -813,18 +833,21 @@ ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags) { struct sock *sk = sock->sk; + const struct proto *prot; if (unlikely(inet_send_prepare(sk))) return -EAGAIN; - if (sk->sk_prot->sendpage) - return sk->sk_prot->sendpage(sk, page, offset, size, flags); + /* IPV6_ADDRFORM can change sk->sk_prot under us. */ + prot = READ_ONCE(sk->sk_prot); + if (prot->sendpage) + return prot->sendpage(sk, page, offset, size, flags); return sock_no_sendpage(sock, page, offset, size, flags); } EXPORT_SYMBOL(inet_sendpage); INDIRECT_CALLABLE_DECLARE(int udp_recvmsg(struct sock *, struct msghdr *, - size_t, int, int, int *)); + size_t, int, int *)); int inet_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, int flags) { @@ -836,8 +859,7 @@ int inet_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, sock_rps_record_flow(sk); err = INDIRECT_CALL_2(sk->sk_prot->recvmsg, tcp_recvmsg, udp_recvmsg, - sk, msg, size, flags & MSG_DONTWAIT, - flags & ~MSG_DONTWAIT, &addr_len); + sk, msg, size, flags, &addr_len); if (err >= 0) msg->msg_namelen = addr_len; return err; @@ -872,7 +894,7 @@ int inet_shutdown(struct socket *sock, int how) err = -ENOTCONN; /* Hack to wake up other listeners, who can poll for EPOLLHUP, even on eg. unconnected UDP sockets -- RR */ - /* fall through */ + fallthrough; default: sk->sk_shutdown |= how; if (sk->sk_prot->shutdown) @@ -886,7 +908,7 @@ int inet_shutdown(struct socket *sock, int how) case TCP_LISTEN: if (!(how & RCV_SHUTDOWN)) break; - /* fall through */ + fallthrough; case TCP_SYN_SENT: err = sk->sk_prot->disconnect(sk, O_NONBLOCK); sock->state = err ? SS_DISCONNECTING : SS_UNCONNECTED; @@ -939,10 +961,10 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCGIFNETMASK: case SIOCGIFDSTADDR: case SIOCGIFPFLAGS: - if (copy_from_user(&ifr, p, sizeof(struct ifreq))) + if (get_user_ifreq(&ifr, NULL, p)) return -EFAULT; err = devinet_ioctl(net, cmd, &ifr); - if (!err && copy_to_user(p, &ifr, sizeof(struct ifreq))) + if (!err && put_user_ifreq(&ifr, p)) err = -EFAULT; break; @@ -952,7 +974,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCSIFDSTADDR: case SIOCSIFPFLAGS: case SIOCSIFFLAGS: - if (copy_from_user(&ifr, p, sizeof(struct ifreq))) + if (get_user_ifreq(&ifr, NULL, p)) return -EFAULT; err = devinet_ioctl(net, cmd, &ifr); break; @@ -968,17 +990,42 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) EXPORT_SYMBOL(inet_ioctl); #ifdef CONFIG_COMPAT +static int inet_compat_routing_ioctl(struct sock *sk, unsigned int cmd, + struct compat_rtentry __user *ur) +{ + compat_uptr_t rtdev; + struct rtentry rt; + + if (copy_from_user(&rt.rt_dst, &ur->rt_dst, + 3 * sizeof(struct sockaddr)) || + get_user(rt.rt_flags, &ur->rt_flags) || + get_user(rt.rt_metric, &ur->rt_metric) || + get_user(rt.rt_mtu, &ur->rt_mtu) || + get_user(rt.rt_window, &ur->rt_window) || + get_user(rt.rt_irtt, &ur->rt_irtt) || + get_user(rtdev, &ur->rt_dev)) + return -EFAULT; + + rt.rt_dev = compat_ptr(rtdev); + return ip_rt_ioctl(sock_net(sk), cmd, &rt); +} + static int inet_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { + void __user *argp = compat_ptr(arg); struct sock *sk = sock->sk; - int err = -ENOIOCTLCMD; - if (sk->sk_prot->compat_ioctl) - err = sk->sk_prot->compat_ioctl(sk, cmd, arg); - - return err; + switch (cmd) { + case SIOCADDRT: + case SIOCDELRT: + return inet_compat_routing_ioctl(sk, cmd, argp); + default: + if (!sk->sk_prot->compat_ioctl) + return -ENOIOCTLCMD; + return sk->sk_prot->compat_ioctl(sk, cmd, arg); + } } -#endif +#endif /* CONFIG_COMPAT */ const struct proto_ops inet_stream_ops = { .family = PF_INET, @@ -1004,12 +1051,11 @@ const struct proto_ops inet_stream_ops = { .sendpage = inet_sendpage, .splice_read = tcp_splice_read, .read_sock = tcp_read_sock, + .read_skb = tcp_read_skb, .sendmsg_locked = tcp_sendmsg_locked, .sendpage_locked = tcp_sendpage_locked, .peek_len = tcp_peek_len, #ifdef CONFIG_COMPAT - .compat_setsockopt = compat_sock_common_setsockopt, - .compat_getsockopt = compat_sock_common_getsockopt, .compat_ioctl = inet_compat_ioctl, #endif .set_rcvlowat = tcp_set_rcvlowat, @@ -1033,13 +1079,12 @@ const struct proto_ops inet_dgram_ops = { .setsockopt = sock_common_setsockopt, .getsockopt = sock_common_getsockopt, .sendmsg = inet_sendmsg, + .read_skb = udp_read_skb, .recvmsg = inet_recvmsg, .mmap = sock_no_mmap, .sendpage = inet_sendpage, .set_peek_off = sk_set_peek_off, #ifdef CONFIG_COMPAT - .compat_setsockopt = compat_sock_common_setsockopt, - .compat_getsockopt = compat_sock_common_getsockopt, .compat_ioctl = inet_compat_ioctl, #endif }; @@ -1070,8 +1115,6 @@ static const struct proto_ops inet_sockraw_ops = { .mmap = sock_no_mmap, .sendpage = inet_sendpage, #ifdef CONFIG_COMPAT - .compat_setsockopt = compat_sock_common_setsockopt, - .compat_getsockopt = compat_sock_common_getsockopt, .compat_ioctl = inet_compat_ioctl, #endif }; @@ -1187,6 +1230,7 @@ EXPORT_SYMBOL(inet_unregister_protosw); static int inet_sk_reselect_saddr(struct sock *sk) { + struct inet_bind_hashbucket *prev_addr_hashbucket; struct inet_sock *inet = inet_sk(sk); __be32 old_saddr = inet->inet_saddr; __be32 daddr = inet->inet_daddr; @@ -1194,6 +1238,7 @@ static int inet_sk_reselect_saddr(struct sock *sk) struct rtable *rt; __be32 new_saddr; struct ip_options_rcu *inet_opt; + int err; inet_opt = rcu_dereference_protected(inet->inet_opt, lockdep_sock_is_held(sk)); @@ -1202,26 +1247,40 @@ static int inet_sk_reselect_saddr(struct sock *sk) /* Query new route. */ fl4 = &inet->cork.fl.u.ip4; - rt = ip_route_connect(fl4, daddr, 0, RT_CONN_FLAGS(sk), - sk->sk_bound_dev_if, sk->sk_protocol, - inet->inet_sport, inet->inet_dport, sk); + rt = ip_route_connect(fl4, daddr, 0, sk->sk_bound_dev_if, + sk->sk_protocol, inet->inet_sport, + inet->inet_dport, sk); if (IS_ERR(rt)) return PTR_ERR(rt); - sk_setup_caps(sk, &rt->dst); - new_saddr = fl4->saddr; - if (new_saddr == old_saddr) + if (new_saddr == old_saddr) { + sk_setup_caps(sk, &rt->dst); return 0; + } + + prev_addr_hashbucket = + inet_bhashfn_portaddr(tcp_or_dccp_get_hashinfo(sk), sk, + sock_net(sk), inet->inet_num); - if (sock_net(sk)->ipv4.sysctl_ip_dynaddr > 1) { + inet->inet_saddr = inet->inet_rcv_saddr = new_saddr; + + err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk); + if (err) { + inet->inet_saddr = old_saddr; + inet->inet_rcv_saddr = old_saddr; + ip_rt_put(rt); + return err; + } + + sk_setup_caps(sk, &rt->dst); + + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_ip_dynaddr) > 1) { pr_info("%s(): shifting inet->saddr from %pI4 to %pI4\n", __func__, &old_saddr, &new_saddr); } - inet->inet_saddr = inet->inet_rcv_saddr = new_saddr; - /* * XXX The only one ugly spot where we need to * XXX really change the sockets identity after @@ -1270,7 +1329,7 @@ int inet_sk_rebuild_header(struct sock *sk) * Other protocols have to map its equivalent state to TCP_SYN_SENT. * DCCP maps its DCCP_REQUESTING state to TCP_SYN_SENT. -acme */ - if (!sock_net(sk)->ipv4.sysctl_ip_dynaddr || + if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_ip_dynaddr) || sk->sk_state != TCP_SYN_SENT || (sk->sk_userlocks & SOCK_BINDADDR_LOCK) || (err = inet_sk_reselect_saddr(sk)) != 0) @@ -1344,8 +1403,11 @@ struct sk_buff *inet_gso_segment(struct sk_buff *skb, } ops = rcu_dereference(inet_offloads[proto]); - if (likely(ops && ops->callbacks.gso_segment)) + if (likely(ops && ops->callbacks.gso_segment)) { segs = ops->callbacks.gso_segment(skb, features); + if (!segs) + skb->network_header = skb_mac_header(skb) + nhoff - skb->head; + } if (IS_ERR_OR_NULL(segs)) goto out; @@ -1389,7 +1451,6 @@ struct sk_buff *inet_gso_segment(struct sk_buff *skb, out: return segs; } -EXPORT_SYMBOL(inet_gso_segment); static struct sk_buff *ipip_gso_segment(struct sk_buff *skb, netdev_features_t features) @@ -1400,10 +1461,6 @@ static struct sk_buff *ipip_gso_segment(struct sk_buff *skb, return inet_gso_segment(skb, features); } -INDIRECT_CALLABLE_DECLARE(struct sk_buff *tcp4_gro_receive(struct list_head *, - struct sk_buff *)); -INDIRECT_CALLABLE_DECLARE(struct sk_buff *udp4_gro_receive(struct list_head *, - struct sk_buff *)); struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) { const struct net_offload *ops; @@ -1418,28 +1475,24 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) off = skb_gro_offset(skb); hlen = off + sizeof(*iph); - iph = skb_gro_header_fast(skb, off); - if (skb_gro_header_hard(skb, hlen)) { - iph = skb_gro_header_slow(skb, hlen, off); - if (unlikely(!iph)) - goto out; - } + iph = skb_gro_header(skb, hlen, off); + if (unlikely(!iph)) + goto out; proto = iph->protocol; - rcu_read_lock(); ops = rcu_dereference(inet_offloads[proto]); if (!ops || !ops->callbacks.gro_receive) - goto out_unlock; + goto out; if (*(u8 *)iph != 0x45) - goto out_unlock; + goto out; if (ip_is_fragment(iph)) - goto out_unlock; + goto out; if (unlikely(ip_fast_csum((u8 *)iph, 5))) - goto out_unlock; + goto out; id = ntohl(*(__be32 *)&iph->id); flush = (u16)((ntohl(*(__be32 *)iph) ^ skb_gro_len(skb)) | (id & ~IP_DF)); @@ -1516,15 +1569,11 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) pp = indirect_call_gro_receive(tcp4_gro_receive, udp4_gro_receive, ops->callbacks.gro_receive, head, skb); -out_unlock: - rcu_read_unlock(); - out: skb_gro_flush_final(skb, pp, flush); return pp; } -EXPORT_SYMBOL(inet_gro_receive); static struct sk_buff *ipip_gro_receive(struct list_head *head, struct sk_buff *skb) @@ -1576,8 +1625,6 @@ int inet_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len) return -EINVAL; } -INDIRECT_CALLABLE_DECLARE(int tcp4_gro_complete(struct sk_buff *, int)); -INDIRECT_CALLABLE_DECLARE(int udp4_gro_complete(struct sk_buff *, int)); int inet_gro_complete(struct sk_buff *skb, int nhoff) { __be16 newlen = htons(skb->len - nhoff); @@ -1594,10 +1641,9 @@ int inet_gro_complete(struct sk_buff *skb, int nhoff) csum_replace2(&iph->check, iph->tot_len, newlen); iph->tot_len = newlen; - rcu_read_lock(); ops = rcu_dereference(inet_offloads[proto]); if (WARN_ON(!ops || !ops->callbacks.gro_complete)) - goto out_unlock; + goto out; /* Only need to add sizeof(*iph) to get to the next hdr below * because any hdr with option will have been flushed in @@ -1607,12 +1653,9 @@ int inet_gro_complete(struct sk_buff *skb, int nhoff) tcp4_gro_complete, udp4_gro_complete, skb, nhoff + sizeof(*iph)); -out_unlock: - rcu_read_unlock(); - +out: return err; } -EXPORT_SYMBOL(inet_gro_complete); static int ipip_gro_complete(struct sk_buff *skb, int nhoff) { @@ -1641,12 +1684,6 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, } EXPORT_SYMBOL_GPL(inet_ctl_sock_create); -u64 snmp_get_cpu_field(void __percpu *mib, int cpu, int offt) -{ - return *(((unsigned long *)per_cpu_ptr(mib, cpu)) + offt); -} -EXPORT_SYMBOL_GPL(snmp_get_cpu_field); - unsigned long snmp_fold_field(void __percpu *mib, int offt) { unsigned long res = 0; @@ -1695,40 +1732,26 @@ EXPORT_SYMBOL_GPL(snmp_fold_field64); #ifdef CONFIG_IP_MULTICAST static const struct net_protocol igmp_protocol = { .handler = igmp_rcv, - .netns_ok = 1, }; #endif -/* thinking of making this const? Don't. - * early_demux can change based on sysctl. - */ -static struct net_protocol tcp_protocol = { - .early_demux = tcp_v4_early_demux, - .early_demux_handler = tcp_v4_early_demux, +static const struct net_protocol tcp_protocol = { .handler = tcp_v4_rcv, .err_handler = tcp_v4_err, .no_policy = 1, - .netns_ok = 1, .icmp_strict_tag_validation = 1, }; -/* thinking of making this const? Don't. - * early_demux can change based on sysctl. - */ -static struct net_protocol udp_protocol = { - .early_demux = udp_v4_early_demux, - .early_demux_handler = udp_v4_early_demux, +static const struct net_protocol udp_protocol = { .handler = udp_rcv, .err_handler = udp_err, .no_policy = 1, - .netns_ok = 1, }; static const struct net_protocol icmp_protocol = { .handler = icmp_rcv, .err_handler = icmp_err, .no_policy = 1, - .netns_ok = 1, }; static __net_init int ipv4_mib_init_net(struct net *net) @@ -1793,6 +1816,10 @@ static __net_exit void ipv4_mib_exit_net(struct net *net) free_percpu(net->mib.net_statistics); free_percpu(net->mib.ip_statistics); free_percpu(net->mib.tcp_statistics); +#ifdef CONFIG_MPTCP + /* allocated on demand, see mptcp_init_sock() */ + free_percpu(net->mib.mptcp_statistics); +#endif } static __net_initdata struct pernet_operations ipv4_mib_ops = { @@ -1831,6 +1858,7 @@ static __net_init int inet_init_net(struct net *net) net->ipv4.sysctl_ip_early_demux = 1; net->ipv4.sysctl_udp_early_demux = 1; net->ipv4.sysctl_tcp_early_demux = 1; + net->ipv4.sysctl_nexthop_compat_mode = 1; #ifdef CONFIG_SYSCTL net->ipv4.sysctl_ip_prot_sock = PROT_SOCK; #endif @@ -1842,6 +1870,8 @@ static __net_init int inet_init_net(struct net *net) net->ipv4.sysctl_igmp_llm_reports = 1; net->ipv4.sysctl_igmp_qrv = 2; + net->ipv4.sysctl_fib_notify_on_flag_change = 0; + return 0; } @@ -1910,10 +1940,12 @@ static int __init inet_init(void) { struct inet_protosw *q; struct list_head *r; - int rc = -EINVAL; + int rc; sock_skb_cb_check_size(sizeof(struct inet_skb_parm)); + raw_hashinfo_init(&raw_v4_hashinfo); + rc = proto_register(&tcp_prot, 1); if (rc) goto out; @@ -1974,6 +2006,10 @@ static int __init inet_init(void) ip_init(); + /* Initialise per-cpu ipv4 mibs */ + if (init_ipv4_mibs()) + panic("%s: Cannot init ipv4 mibs\n", __func__); + /* Setup TCP slab cache for open requests. */ tcp_init(); @@ -2004,12 +2040,6 @@ static int __init inet_init(void) if (init_inet_pernet_ops()) pr_crit("%s: Cannot init ipv4 inet pernet ops\n", __func__); - /* - * Initialise per-cpu ipv4 mibs - */ - - if (init_ipv4_mibs()) - pr_crit("%s: Cannot init ipv4 mibs\n", __func__); ipv4_proc_init(); diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 974179b3b314..ee4e578c7f20 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -107,7 +107,7 @@ static int ip_clear_mutable_options(const struct iphdr *iph, __be32 *daddr) if (optlen < 6) return -EINVAL; memcpy(daddr, optptr+optlen-4, 4); - /* Fall through */ + fallthrough; default: memset(optptr, 0, optlen); } @@ -141,7 +141,7 @@ static void ah_output_done(struct crypto_async_request *base, int err) } kfree(AH_SKB_CB(skb)->tmp); - xfrm_output_resume(skb, err); + xfrm_output_resume(skb->sk, skb, err); } static int ah_output(struct xfrm_state *x, struct sk_buff *skb) @@ -450,6 +450,7 @@ static int ah4_err(struct sk_buff *skb, u32 info) case ICMP_DEST_UNREACH: if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED) return 0; + break; case ICMP_REDIRECT: break; default: @@ -470,30 +471,38 @@ static int ah4_err(struct sk_buff *skb, u32 info) return 0; } -static int ah_init_state(struct xfrm_state *x) +static int ah_init_state(struct xfrm_state *x, struct netlink_ext_ack *extack) { struct ah_data *ahp = NULL; struct xfrm_algo_desc *aalg_desc; struct crypto_ahash *ahash; - if (!x->aalg) + if (!x->aalg) { + NL_SET_ERR_MSG(extack, "AH requires a state with an AUTH algorithm"); goto error; + } - if (x->encap) + if (x->encap) { + NL_SET_ERR_MSG(extack, "AH is not compatible with encapsulation"); goto error; + } ahp = kzalloc(sizeof(*ahp), GFP_KERNEL); if (!ahp) return -ENOMEM; ahash = crypto_alloc_ahash(x->aalg->alg_name, 0, 0); - if (IS_ERR(ahash)) + if (IS_ERR(ahash)) { + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); goto error; + } ahp->ahash = ahash; if (crypto_ahash_setkey(ahash, x->aalg->alg_key, - (x->aalg->alg_key_len + 7) / 8)) + (x->aalg->alg_key_len + 7) / 8)) { + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); goto error; + } /* * Lookup the algorithm description maintained by xfrm_algo, @@ -506,10 +515,7 @@ static int ah_init_state(struct xfrm_state *x) if (aalg_desc->uinfo.auth.icv_fullbits/8 != crypto_ahash_digestsize(ahash)) { - pr_info("%s: %s digestsize %u != %hu\n", - __func__, x->aalg->alg_name, - crypto_ahash_digestsize(ahash), - aalg_desc->uinfo.auth.icv_fullbits / 8); + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); goto error; } @@ -554,7 +560,6 @@ static int ah4_rcv_cb(struct sk_buff *skb, int err) static const struct xfrm_type ah_type = { - .description = "AH4", .owner = THIS_MODULE, .proto = IPPROTO_AH, .flags = XFRM_TYPE_REPLAY_PROT, diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 05eb42f347e8..4f7237661afb 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -125,6 +125,7 @@ static int arp_constructor(struct neighbour *neigh); static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb); static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb); static void parp_redo(struct sk_buff *skb); +static int arp_is_multicast(const void *pkey); static const struct neigh_ops arp_generic_ops = { .family = AF_INET, @@ -156,6 +157,7 @@ struct neigh_table arp_tbl = { .key_eq = arp_key_eq, .constructor = arp_constructor, .proxy_redo = parp_redo, + .is_multicast = arp_is_multicast, .id = "arp_cache", .parms = { .tbl = &arp_tbl, @@ -166,6 +168,7 @@ struct neigh_table arp_tbl = { [NEIGH_VAR_RETRANS_TIME] = 1 * HZ, [NEIGH_VAR_BASE_REACHABLE_TIME] = 30 * HZ, [NEIGH_VAR_DELAY_PROBE_TIME] = 5 * HZ, + [NEIGH_VAR_INTERVAL_PROBE_TIME_MS] = 5 * HZ, [NEIGH_VAR_GC_STALETIME] = 60 * HZ, [NEIGH_VAR_QUEUE_LEN_BYTES] = SK_WMEM_MAX, [NEIGH_VAR_PROXY_QLEN] = 64, @@ -291,7 +294,7 @@ static int arp_constructor(struct neighbour *neigh) static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb) { dst_link_failure(skb); - kfree_skb(skb); + kfree_skb_reason(skb, SKB_DROP_REASON_NEIGH_FAILED); } /* Create and send an arp packet. */ @@ -426,6 +429,26 @@ static int arp_ignore(struct in_device *in_dev, __be32 sip, __be32 tip) return !inet_confirm_addr(net, in_dev, sip, tip, scope); } +static int arp_accept(struct in_device *in_dev, __be32 sip) +{ + struct net *net = dev_net(in_dev->dev); + int scope = RT_SCOPE_LINK; + + switch (IN_DEV_ARP_ACCEPT(in_dev)) { + case 0: /* Don't create new entries from garp */ + return 0; + case 1: /* Create new entries from garp */ + return 1; + case 2: /* Create a neighbor in the arp table only if sip + * is in the same subnet as an address configured + * on the interface that received the garp message + */ + return !!inet_confirm_addr(net, in_dev, sip, 0, scope); + default: + return 0; + } +} + static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev) { struct rtable *rt; @@ -865,12 +888,12 @@ static int arp_process(struct net *net, struct sock *sk, struct sk_buff *skb) n = __neigh_lookup(&arp_tbl, &sip, dev, 0); addr_type = -1; - if (n || IN_DEV_ARP_ACCEPT(in_dev)) { + if (n || arp_accept(in_dev, sip)) { is_garp = arp_is_garp(net, dev, &addr_type, arp->ar_op, sip, tip, sha, tha); } - if (IN_DEV_ARP_ACCEPT(in_dev)) { + if (arp_accept(in_dev, sip)) { /* Unsolicited ARP is not accepted by default. It is possible, that this option should be enabled for some devices (strip is candidate) @@ -928,6 +951,10 @@ static void parp_redo(struct sk_buff *skb) arp_process(dev_net(skb->dev), NULL, skb); } +static int arp_is_multicast(const void *pkey) +{ + return ipv4_is_multicast(*((__be32 *)pkey)); +} /* * Receive an arp request from the device layer. @@ -1102,7 +1129,7 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev) r->arp_flags = arp_state_to_flags(neigh); read_unlock_bh(&neigh->lock); r->arp_ha.sa_family = dev->type; - strlcpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); + strscpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); err = 0; } neigh_release(neigh); @@ -1110,13 +1137,18 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev) return err; } -static int arp_invalidate(struct net_device *dev, __be32 ip) +int arp_invalidate(struct net_device *dev, __be32 ip, bool force) { struct neighbour *neigh = neigh_lookup(&arp_tbl, &ip, dev); int err = -ENXIO; struct neigh_table *tbl = &arp_tbl; if (neigh) { + if ((neigh->nud_state & NUD_VALID) && !force) { + neigh_release(neigh); + return 0; + } + if (neigh->nud_state & ~NUD_NOARP) err = neigh_update(neigh, NULL, NUD_FAILED, NEIGH_UPDATE_F_OVERRIDE| @@ -1163,7 +1195,7 @@ static int arp_req_delete(struct net *net, struct arpreq *r, if (!dev) return -EINVAL; } - return arp_invalidate(dev, ip); + return arp_invalidate(dev, ip, true); } /* @@ -1181,7 +1213,7 @@ int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) case SIOCSARP: if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM; - /* fall through */ + fallthrough; case SIOCGARP: err = copy_from_user(&r, arg, sizeof(struct arpreq)); if (err) @@ -1241,6 +1273,8 @@ static int arp_netdev_event(struct notifier_block *this, unsigned long event, { struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct netdev_notifier_change_info *change_info; + struct in_device *in_dev; + bool evict_nocarrier; switch (event) { case NETDEV_CHANGEADDR: @@ -1251,7 +1285,14 @@ static int arp_netdev_event(struct notifier_block *this, unsigned long event, change_info = ptr; if (change_info->flags_changed & IFF_NOARP) neigh_changeaddr(&arp_tbl, dev); - if (!netif_carrier_ok(dev)) + + in_dev = __in_dev_get_rtnl(dev); + if (!in_dev) + evict_nocarrier = true; + else + evict_nocarrier = IN_DEV_ARP_EVICT_NOCARRIER(in_dev); + + if (evict_nocarrier && !netif_carrier_ok(dev)) neigh_carrier_down(&arp_tbl, dev); break; default: @@ -1284,24 +1325,9 @@ static struct packet_type arp_packet_type __read_mostly = { .func = arp_rcv, }; -static int arp_proc_init(void); - -void __init arp_init(void) -{ - neigh_table_init(NEIGH_ARP_TABLE, &arp_tbl); - - dev_add_pack(&arp_packet_type); - arp_proc_init(); -#ifdef CONFIG_SYSCTL - neigh_sysctl_register(NULL, &arp_tbl.parms, NULL); -#endif - register_netdevice_notifier(&arp_netdev_notifier); -} - #ifdef CONFIG_PROC_FS #if IS_ENABLED(CONFIG_AX25) -/* ------------------------------------------------------------------------ */ /* * ax25 -> ASCII conversion */ @@ -1407,16 +1433,13 @@ static void *arp_seq_start(struct seq_file *seq, loff_t *pos) return neigh_seq_start(seq, pos, &arp_tbl, NEIGH_SEQ_SKIP_NOARP); } -/* ------------------------------------------------------------------------ */ - static const struct seq_operations arp_seq_ops = { .start = arp_seq_start, .next = neigh_seq_next, .stop = neigh_seq_stop, .show = arp_seq_show, }; - -/* ------------------------------------------------------------------------ */ +#endif /* CONFIG_PROC_FS */ static int __net_init arp_net_init(struct net *net) { @@ -1436,16 +1459,14 @@ static struct pernet_operations arp_net_ops = { .exit = arp_net_exit, }; -static int __init arp_proc_init(void) +void __init arp_init(void) { - return register_pernet_subsys(&arp_net_ops); -} - -#else /* CONFIG_PROC_FS */ + neigh_table_init(NEIGH_ARP_TABLE, &arp_tbl); -static int __init arp_proc_init(void) -{ - return 0; + dev_add_pack(&arp_packet_type); + register_pernet_subsys(&arp_net_ops); +#ifdef CONFIG_SYSCTL + neigh_sysctl_register(NULL, &arp_tbl.parms, NULL); +#endif + register_netdevice_notifier(&arp_netdev_notifier); } - -#endif /* CONFIG_PROC_FS */ diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index 574972bc7299..6da16ae6a962 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -1,24 +1,18 @@ // SPDX-License-Identifier: GPL-2.0 /* Copyright (c) 2019 Facebook */ +#include <linux/init.h> #include <linux/types.h> #include <linux/bpf_verifier.h> #include <linux/bpf.h> #include <linux/btf.h> +#include <linux/btf_ids.h> #include <linux/filter.h> #include <net/tcp.h> +#include <net/bpf_sk_storage.h> -static u32 optional_ops[] = { - offsetof(struct tcp_congestion_ops, init), - offsetof(struct tcp_congestion_ops, release), - offsetof(struct tcp_congestion_ops, set_state), - offsetof(struct tcp_congestion_ops, cwnd_event), - offsetof(struct tcp_congestion_ops, in_ack_event), - offsetof(struct tcp_congestion_ops, pkts_acked), - offsetof(struct tcp_congestion_ops, min_tso_segs), - offsetof(struct tcp_congestion_ops, sndbuf_expand), - offsetof(struct tcp_congestion_ops, cong_control), -}; +/* "extern" is to avoid sparse warning. It is only used in bpf_struct_ops.c. */ +extern struct bpf_struct_ops bpf_tcp_congestion_ops; static u32 unsupported_ops[] = { offsetof(struct tcp_congestion_ops, get_info), @@ -45,18 +39,6 @@ static int bpf_tcp_ca_init(struct btf *btf) return 0; } -static bool is_optional(u32 member_offset) -{ - unsigned int i; - - for (i = 0; i < ARRAY_SIZE(optional_ops); i++) { - if (member_offset == optional_ops[i]) - return true; - } - - return false; -} - static bool is_unsupported(u32 member_offset) { unsigned int i; @@ -76,14 +58,7 @@ static bool bpf_tcp_ca_is_valid_access(int off, int size, const struct bpf_prog *prog, struct bpf_insn_access_aux *info) { - if (off < 0 || off >= sizeof(__u64) * MAX_BPF_FUNC_ARGS) - return false; - if (type != BPF_READ) - return false; - if (off % size != 0) - return false; - - if (!btf_ctx_access(off, size, type, prog, info)) + if (!bpf_tracing_btf_ctx_access(off, size, type, prog, info)) return false; if (info->reg_type == PTR_TO_BTF_ID && info->btf_id == sock_id) @@ -94,14 +69,17 @@ static bool bpf_tcp_ca_is_valid_access(int off, int size, } static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, + const struct btf *btf, const struct btf_type *t, int off, int size, enum bpf_access_type atype, - u32 *next_btf_id) + u32 *next_btf_id, + enum bpf_type_flag *flag) { size_t end; if (atype == BPF_READ) - return btf_struct_access(log, t, off, size, atype, next_btf_id); + return btf_struct_access(log, btf, t, off, size, atype, next_btf_id, + flag); if (t != tcp_sock_type) { bpf_log(log, "only read is supported\n"); @@ -109,6 +87,12 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, } switch (off) { + case offsetof(struct sock, sk_pacing_rate): + end = offsetofend(struct sock, sk_pacing_rate); + break; + case offsetof(struct sock, sk_pacing_status): + end = offsetofend(struct sock, sk_pacing_status); + break; case bpf_ctx_range(struct inet_connection_sock, icsk_ca_priv): end = offsetofend(struct inet_connection_sock, icsk_ca_priv); break; @@ -140,7 +124,7 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, return -EACCES; } - return NOT_INIT; + return 0; } BPF_CALL_2(bpf_tcp_send_ack, struct tcp_sock *, tp, u32, rcv_nxt) @@ -156,10 +140,23 @@ static const struct bpf_func_proto bpf_tcp_send_ack_proto = { /* In case we want to report error later */ .ret_type = RET_INTEGER, .arg1_type = ARG_PTR_TO_BTF_ID, + .arg1_btf_id = &tcp_sock_id, .arg2_type = ARG_ANYTHING, - .btf_id = &tcp_sock_id, }; +static u32 prog_ops_moff(const struct bpf_prog *prog) +{ + const struct btf_member *m; + const struct btf_type *t; + u32 midx; + + midx = prog->expected_attach_type; + t = bpf_tcp_congestion_ops.type; + m = &btf_type_member(t)[midx]; + + return __btf_member_bit_offset(t, m) / 8; +} + static const struct bpf_func_proto * bpf_tcp_ca_get_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) @@ -167,11 +164,52 @@ bpf_tcp_ca_get_func_proto(enum bpf_func_id func_id, switch (func_id) { case BPF_FUNC_tcp_send_ack: return &bpf_tcp_send_ack_proto; + case BPF_FUNC_sk_storage_get: + return &bpf_sk_storage_get_proto; + case BPF_FUNC_sk_storage_delete: + return &bpf_sk_storage_delete_proto; + case BPF_FUNC_setsockopt: + /* Does not allow release() to call setsockopt. + * release() is called when the current bpf-tcp-cc + * is retiring. It is not allowed to call + * setsockopt() to make further changes which + * may potentially allocate new resources. + */ + if (prog_ops_moff(prog) != + offsetof(struct tcp_congestion_ops, release)) + return &bpf_sk_setsockopt_proto; + return NULL; + case BPF_FUNC_getsockopt: + /* Since get/setsockopt is usually expected to + * be available together, disable getsockopt for + * release also to avoid usage surprise. + * The bpf-tcp-cc already has a more powerful way + * to read tcp_sock from the PTR_TO_BTF_ID. + */ + if (prog_ops_moff(prog) != + offsetof(struct tcp_congestion_ops, release)) + return &bpf_sk_getsockopt_proto; + return NULL; + case BPF_FUNC_ktime_get_coarse_ns: + return &bpf_ktime_get_coarse_ns_proto; default: return bpf_base_func_proto(func_id); } } +BTF_SET8_START(bpf_tcp_ca_check_kfunc_ids) +BTF_ID_FLAGS(func, tcp_reno_ssthresh) +BTF_ID_FLAGS(func, tcp_reno_cong_avoid) +BTF_ID_FLAGS(func, tcp_reno_undo_cwnd) +BTF_ID_FLAGS(func, tcp_slow_start) +BTF_ID_FLAGS(func, tcp_cong_avoid_ai) +BTF_SET8_END(bpf_tcp_ca_check_kfunc_ids) + +static const struct btf_kfunc_id_set bpf_tcp_ca_kfunc_set = { + .owner = THIS_MODULE, + .set = &bpf_tcp_ca_check_kfunc_ids, +}; + static const struct bpf_verifier_ops bpf_tcp_ca_verifier_ops = { .get_func_proto = bpf_tcp_ca_get_func_proto, .is_valid_access = bpf_tcp_ca_is_valid_access, @@ -184,14 +222,12 @@ static int bpf_tcp_ca_init_member(const struct btf_type *t, { const struct tcp_congestion_ops *utcp_ca; struct tcp_congestion_ops *tcp_ca; - size_t tcp_ca_name_len; - int prog_fd; u32 moff; utcp_ca = (const struct tcp_congestion_ops *)udata; tcp_ca = (struct tcp_congestion_ops *)kdata; - moff = btf_member_bit_offset(t, member) / 8; + moff = __btf_member_bit_offset(t, member) / 8; switch (moff) { case offsetof(struct tcp_congestion_ops, flags): if (utcp_ca->flags & ~TCP_CONG_MASK) @@ -199,31 +235,21 @@ static int bpf_tcp_ca_init_member(const struct btf_type *t, tcp_ca->flags = utcp_ca->flags; return 1; case offsetof(struct tcp_congestion_ops, name): - tcp_ca_name_len = strnlen(utcp_ca->name, sizeof(utcp_ca->name)); - if (!tcp_ca_name_len || - tcp_ca_name_len == sizeof(utcp_ca->name)) + if (bpf_obj_name_cpy(tcp_ca->name, utcp_ca->name, + sizeof(tcp_ca->name)) <= 0) return -EINVAL; if (tcp_ca_find(utcp_ca->name)) return -EEXIST; - memcpy(tcp_ca->name, utcp_ca->name, sizeof(tcp_ca->name)); return 1; } - if (!btf_type_resolve_func_ptr(btf_vmlinux, member->type, NULL)) - return 0; - - /* Ensure bpf_prog is provided for compulsory func ptr */ - prog_fd = (int)(*(unsigned long *)(udata + moff)); - if (!prog_fd && !is_optional(moff) && !is_unsupported(moff)) - return -EINVAL; - return 0; } static int bpf_tcp_ca_check_member(const struct btf_type *t, const struct btf_member *member) { - if (is_unsupported(btf_member_bit_offset(t, member) / 8)) + if (is_unsupported(__btf_member_bit_offset(t, member) / 8)) return -ENOTSUPP; return 0; } @@ -238,9 +264,6 @@ static void bpf_tcp_ca_unreg(void *kdata) tcp_unregister_congestion_control(kdata); } -/* Avoid sparse warning. It is only used in bpf_struct_ops.c. */ -extern struct bpf_struct_ops bpf_tcp_congestion_ops; - struct bpf_struct_ops bpf_tcp_congestion_ops = { .verifier_ops = &bpf_tcp_ca_verifier_ops, .reg = bpf_tcp_ca_reg, @@ -250,3 +273,9 @@ struct bpf_struct_ops bpf_tcp_congestion_ops = { .init = bpf_tcp_ca_init, .name = "tcp_congestion_ops", }; + +static int __init bpf_tcp_ca_kfunc_init(void) +{ + return register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &bpf_tcp_ca_kfunc_set); +} +late_initcall(bpf_tcp_ca_kfunc_init); diff --git a/net/ipv4/bpfilter/sockopt.c b/net/ipv4/bpfilter/sockopt.c index 0480918bfc7c..1b34cb9a7708 100644 --- a/net/ipv4/bpfilter/sockopt.c +++ b/net/ipv4/bpfilter/sockopt.c @@ -12,18 +12,16 @@ struct bpfilter_umh_ops bpfilter_ops; EXPORT_SYMBOL_GPL(bpfilter_ops); -static void bpfilter_umh_cleanup(struct umh_info *info) +void bpfilter_umh_cleanup(struct umd_info *info) { - mutex_lock(&bpfilter_ops.lock); - bpfilter_ops.stop = true; fput(info->pipe_to_umh); fput(info->pipe_from_umh); - info->pid = 0; - mutex_unlock(&bpfilter_ops.lock); + put_pid(info->tgid); + info->tgid = NULL; } +EXPORT_SYMBOL_GPL(bpfilter_umh_cleanup); -static int bpfilter_mbox_request(struct sock *sk, int optname, - char __user *optval, +static int bpfilter_mbox_request(struct sock *sk, int optname, sockptr_t optval, unsigned int optlen, bool is_set) { int err; @@ -38,7 +36,11 @@ static int bpfilter_mbox_request(struct sock *sk, int optname, goto out; } } - if (bpfilter_ops.stop) { + if (bpfilter_ops.info.tgid && + thread_group_exited(bpfilter_ops.info.tgid)) + bpfilter_umh_cleanup(&bpfilter_ops.info); + + if (!bpfilter_ops.info.tgid) { err = bpfilter_ops.start(); if (err) goto out; @@ -49,7 +51,7 @@ out: return err; } -int bpfilter_ip_set_sockopt(struct sock *sk, int optname, char __user *optval, +int bpfilter_ip_set_sockopt(struct sock *sk, int optname, sockptr_t optval, unsigned int optlen) { return bpfilter_mbox_request(sk, optname, optval, optlen, true); @@ -63,15 +65,15 @@ int bpfilter_ip_get_sockopt(struct sock *sk, int optname, char __user *optval, if (get_user(len, optlen)) return -EFAULT; - return bpfilter_mbox_request(sk, optname, optval, len, false); + return bpfilter_mbox_request(sk, optname, USER_SOCKPTR(optval), len, + false); } static int __init bpfilter_sockopt_init(void) { mutex_init(&bpfilter_ops.lock); - bpfilter_ops.stop = true; - bpfilter_ops.info.cmdline = "bpfilter_umh"; - bpfilter_ops.info.cleanup = &bpfilter_umh_cleanup; + bpfilter_ops.info.tgid = NULL; + bpfilter_ops.info.driver_name = "bpfilter_umh"; return 0; } diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 0bd10a1f477f..6cd3b6c559f0 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -10,9 +10,9 @@ * * The CIPSO draft specification can be found in the kernel's Documentation * directory as well as the following URL: - * http://tools.ietf.org/id/draft-ietf-cipso-ipsecurity-01.txt + * https://tools.ietf.org/id/draft-ietf-cipso-ipsecurity-01.txt * The FIPS-188 specification can be found at the following URL: - * http://www.itl.nist.gov/fipspubs/fip188.htm + * https://www.itl.nist.gov/fipspubs/fip188.htm * * Author: Paul Moore <paul.moore@hp.com> */ @@ -73,7 +73,7 @@ struct cipso_v4_map_cache_entry { static struct cipso_v4_map_cache_bkt *cipso_v4_cache; /* Restricted bitmap (tag #1) flags */ -int cipso_v4_rbm_optfmt = 0; +int cipso_v4_rbm_optfmt; int cipso_v4_rbm_strictvalid = 1; /* @@ -187,8 +187,7 @@ static int __init cipso_v4_cache_init(void) * cipso_v4_cache_invalidate - Invalidates the current CIPSO cache * * Description: - * Invalidates and frees any entries in the CIPSO cache. Returns zero on - * success and negative values on failure. + * Invalidates and frees any entries in the CIPSO cache. * */ void cipso_v4_cache_invalidate(void) @@ -240,7 +239,7 @@ static int cipso_v4_cache_check(const unsigned char *key, struct cipso_v4_map_cache_entry *prev_entry = NULL; u32 hash; - if (!cipso_v4_cache_enabled) + if (!READ_ONCE(cipso_v4_cache_enabled)) return -ENOENT; hash = cipso_v4_map_cache_hash(key, key_len); @@ -283,7 +282,7 @@ static int cipso_v4_cache_check(const unsigned char *key, /** * cipso_v4_cache_add - Add an entry to the CIPSO cache - * @skb: the packet + * @cipso_ptr: pointer to CIPSO IP option * @secattr: the packet's security attributes * * Description: @@ -297,13 +296,14 @@ static int cipso_v4_cache_check(const unsigned char *key, int cipso_v4_cache_add(const unsigned char *cipso_ptr, const struct netlbl_lsm_secattr *secattr) { + int bkt_size = READ_ONCE(cipso_v4_cache_bucketsize); int ret_val = -EPERM; u32 bkt; struct cipso_v4_map_cache_entry *entry = NULL; struct cipso_v4_map_cache_entry *old_entry = NULL; u32 cipso_ptr_len; - if (!cipso_v4_cache_enabled || cipso_v4_cache_bucketsize <= 0) + if (!READ_ONCE(cipso_v4_cache_enabled) || bkt_size <= 0) return 0; cipso_ptr_len = cipso_ptr[1]; @@ -323,7 +323,7 @@ int cipso_v4_cache_add(const unsigned char *cipso_ptr, bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETS - 1); spin_lock_bh(&cipso_v4_cache[bkt].lock); - if (cipso_v4_cache[bkt].size < cipso_v4_cache_bucketsize) { + if (cipso_v4_cache[bkt].size < bkt_size) { list_add(&entry->list, &cipso_v4_cache[bkt].list); cipso_v4_cache[bkt].size += 1; } else { @@ -472,6 +472,7 @@ void cipso_v4_doi_free(struct cipso_v4_doi *doi_def) kfree(doi_def->map.std->lvl.local); kfree(doi_def->map.std->cat.cipso); kfree(doi_def->map.std->cat.local); + kfree(doi_def->map.std); break; } kfree(doi_def); @@ -498,7 +499,7 @@ static void cipso_v4_doi_free_rcu(struct rcu_head *entry) /** * cipso_v4_doi_remove - Remove an existing DOI from the CIPSO protocol engine * @doi: the DOI value - * @audit_secid: the LSM secid to use in the audit message + * @audit_info: NetLabel audit information * * Description: * Removes a DOI definition from the CIPSO engine. The NetLabel routines will @@ -519,16 +520,10 @@ int cipso_v4_doi_remove(u32 doi, struct netlbl_audit *audit_info) ret_val = -ENOENT; goto doi_remove_return; } - if (!refcount_dec_and_test(&doi_def->refcount)) { - spin_unlock(&cipso_v4_doi_list_lock); - ret_val = -EBUSY; - goto doi_remove_return; - } list_del_rcu(&doi_def->list); spin_unlock(&cipso_v4_doi_list_lock); - cipso_v4_cache_invalidate(); - call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu); + cipso_v4_doi_putdef(doi_def); ret_val = 0; doi_remove_return: @@ -585,9 +580,6 @@ void cipso_v4_doi_putdef(struct cipso_v4_doi *doi_def) if (!refcount_dec_and_test(&doi_def->refcount)) return; - spin_lock(&cipso_v4_doi_list_lock); - list_del_rcu(&doi_def->list); - spin_unlock(&cipso_v4_doi_list_lock); cipso_v4_cache_invalidate(); call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu); @@ -1162,7 +1154,7 @@ static void cipso_v4_gentag_hdr(const struct cipso_v4_doi *doi_def, { buf[0] = IPOPT_CIPSO; buf[1] = CIPSO_V4_HDR_LEN + len; - *(__be32 *)&buf[2] = htonl(doi_def->doi); + put_unaligned_be32(doi_def->doi, &buf[2]); } /** @@ -1208,7 +1200,8 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def, /* This will send packets using the "optimized" format when * possible as specified in section 3.4.2.6 of the * CIPSO draft. */ - if (cipso_v4_rbm_optfmt && ret_val > 0 && ret_val <= 10) + if (READ_ONCE(cipso_v4_rbm_optfmt) && ret_val > 0 && + ret_val <= 10) tag_len = 14; else tag_len = 4 + ret_val; @@ -1258,7 +1251,8 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def, return ret_val; } - secattr->flags |= NETLBL_SECATTR_MLS_CAT; + if (secattr->attr.mls.cat) + secattr->flags |= NETLBL_SECATTR_MLS_CAT; } return 0; @@ -1439,7 +1433,8 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def, return ret_val; } - secattr->flags |= NETLBL_SECATTR_MLS_CAT; + if (secattr->attr.mls.cat) + secattr->flags |= NETLBL_SECATTR_MLS_CAT; } return 0; @@ -1533,6 +1528,7 @@ unsigned char *cipso_v4_optptr(const struct sk_buff *skb) /** * cipso_v4_validate - Validate a CIPSO option + * @skb: the packet * @option: the start of the option, on error it is set to point to the error * * Description: @@ -1609,7 +1605,7 @@ int cipso_v4_validate(const struct sk_buff *skb, unsigned char **option) * all the CIPSO validations here but it doesn't * really specify _exactly_ what we need to validate * ... so, just make it a sysctl tunable. */ - if (cipso_v4_rbm_strictvalid) { + if (READ_ONCE(cipso_v4_rbm_strictvalid)) { if (cipso_v4_map_lvl_valid(doi_def, tag[3]) < 0) { err_offset = opt_iter + 3; @@ -2064,7 +2060,7 @@ void cipso_v4_sock_delattr(struct sock *sk) /** * cipso_v4_req_delattr - Delete the CIPSO option from a request socket - * @reg: the request socket + * @req: the request socket * * Description: * Removes the CIPSO option from a request socket, if present. @@ -2156,6 +2152,7 @@ int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr) /** * cipso_v4_skbuff_setattr - Set the CIPSO option on a packet * @skb: the packet + * @doi_def: the DOI structure * @secattr: the security attributes * * Description: diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index 4a8550c49202..4d1af0cd7d99 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -9,7 +9,6 @@ #include <linux/types.h> #include <linux/module.h> -#include <linux/ip.h> #include <linux/in.h> #include <net/ip.h> #include <net/sock.h> @@ -43,12 +42,13 @@ int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len oif = inet->mc_index; if (!saddr) saddr = inet->mc_addr; + } else if (!oif) { + oif = inet->uc_index; } fl4 = &inet->cork.fl.u.ip4; - rt = ip_route_connect(fl4, usin->sin_addr.s_addr, saddr, - RT_CONN_FLAGS(sk), oif, - sk->sk_protocol, - inet->inet_sport, usin->sin_port, sk); + rt = ip_route_connect(fl4, usin->sin_addr.s_addr, saddr, oif, + sk->sk_protocol, inet->inet_sport, + usin->sin_port, sk); if (IS_ERR(rt)) { err = PTR_ERR(rt); if (err == -ENETUNREACH) @@ -70,10 +70,10 @@ int __ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len } inet->inet_daddr = fl4->daddr; inet->inet_dport = usin->sin_port; - reuseport_has_conns(sk, true); + reuseport_has_conns_set(sk); sk->sk_state = TCP_ESTABLISHED; sk_set_txhash(sk); - inet->inet_id = prandom_u32(); + inet->inet_id = get_random_u16(); sk_dst_set(sk, &rt->dst); err = 0; diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index e4632bd2026d..e8b9a9202fec 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -75,6 +75,7 @@ static struct ipv4_devconf ipv4_devconf = { [IPV4_DEVCONF_SHARED_MEDIA - 1] = 1, [IPV4_DEVCONF_IGMPV2_UNSOLICITED_REPORT_INTERVAL - 1] = 10000 /*ms*/, [IPV4_DEVCONF_IGMPV3_UNSOLICITED_REPORT_INTERVAL - 1] = 1000 /*ms*/, + [IPV4_DEVCONF_ARP_EVICT_NOCARRIER - 1] = 1, }, }; @@ -87,6 +88,7 @@ static struct ipv4_devconf ipv4_devconf_dflt = { [IPV4_DEVCONF_ACCEPT_SOURCE_ROUTE - 1] = 1, [IPV4_DEVCONF_IGMPV2_UNSOLICITED_REPORT_INTERVAL - 1] = 10000 /*ms*/, [IPV4_DEVCONF_IGMPV3_UNSOLICITED_REPORT_INTERVAL - 1] = 1000 /*ms*/, + [IPV4_DEVCONF_ARP_EVICT_NOCARRIER - 1] = 1, }, }; @@ -102,6 +104,7 @@ static const struct nla_policy ifa_ipv4_policy[IFA_MAX+1] = { [IFA_FLAGS] = { .type = NLA_U32 }, [IFA_RT_PRIORITY] = { .type = NLA_U32 }, [IFA_TARGET_NETNSID] = { .type = NLA_S32 }, + [IFA_PROTO] = { .type = NLA_U8 }, }; struct inet_fill_args { @@ -215,7 +218,7 @@ static void devinet_sysctl_unregister(struct in_device *idev) static struct in_ifaddr *inet_alloc_ifa(void) { - return kzalloc(sizeof(struct in_ifaddr), GFP_KERNEL); + return kzalloc(sizeof(struct in_ifaddr), GFP_KERNEL_ACCOUNT); } static void inet_rcu_free_ifa(struct rcu_head *head) @@ -241,7 +244,7 @@ void in_dev_finish_destroy(struct in_device *idev) #ifdef NET_REFCNT_DEBUG pr_debug("%s: %p=%s\n", __func__, idev, dev ? dev->name : "NIL"); #endif - dev_put(dev); + netdev_put(dev, &idev->dev_tracker); if (!idev->dead) pr_err("Freeing alive in_device %p\n", idev); else @@ -269,13 +272,14 @@ static struct in_device *inetdev_init(struct net_device *dev) if (IPV4_DEVCONF(in_dev->cnf, FORWARDING)) dev_disable_lro(dev); /* Reference in_dev->dev */ - dev_hold(dev); + netdev_hold(dev, &in_dev->dev_tracker, GFP_KERNEL); /* Account for reference dev->ip_ptr (below) */ refcount_set(&in_dev->refcnt, 1); err = devinet_sysctl_register(in_dev); if (err) { in_dev->dead = 1; + neigh_parms_release(&arp_tbl, in_dev->arp_parms); in_dev_put(in_dev); in_dev = NULL; goto out; @@ -532,10 +536,8 @@ static int __inet_insert_ifa(struct in_ifaddr *ifa, struct nlmsghdr *nlh, return ret; } - if (!(ifa->ifa_flags & IFA_F_SECONDARY)) { - prandom_seed((__force u32) ifa->ifa_local); + if (!(ifa->ifa_flags & IFA_F_SECONDARY)) ifap = last_primary; - } rcu_assign_pointer(ifa->ifa_next, *ifap); rcu_assign_pointer(*ifap, ifa); @@ -614,12 +616,15 @@ struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, return NULL; } -static int ip_mc_config(struct sock *sk, bool join, const struct in_ifaddr *ifa) +static int ip_mc_autojoin_config(struct net *net, bool join, + const struct in_ifaddr *ifa) { +#if defined(CONFIG_IP_MULTICAST) struct ip_mreqn mreq = { .imr_multiaddr.s_addr = ifa->ifa_address, .imr_ifindex = ifa->ifa_dev->dev->ifindex, }; + struct sock *sk = net->ipv4.mc_autojoin_sk; int ret; ASSERT_RTNL(); @@ -632,6 +637,9 @@ static int ip_mc_config(struct sock *sk, bool join, const struct in_ifaddr *ifa) release_sock(sk); return ret; +#else + return -EOPNOTSUPP; +#endif } static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, @@ -643,8 +651,7 @@ static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, struct in_device *in_dev; struct ifaddrmsg *ifm; struct in_ifaddr *ifa; - - int err = -EINVAL; + int err; ASSERT_RTNL(); @@ -675,7 +682,7 @@ static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, continue; if (ipv4_is_multicast(ifa->ifa_address)) - ip_mc_config(net->ipv4.mc_autojoin_sk, false, ifa); + ip_mc_autojoin_config(net, false, ifa); __inet_del_ifa(in_dev, ifap, 1, nlh, NETLINK_CB(skb).portid); return 0; } @@ -874,13 +881,16 @@ static struct in_ifaddr *rtm_to_ifaddr(struct net *net, struct nlmsghdr *nlh, ifa->ifa_broadcast = nla_get_in_addr(tb[IFA_BROADCAST]); if (tb[IFA_LABEL]) - nla_strlcpy(ifa->ifa_label, tb[IFA_LABEL], IFNAMSIZ); + nla_strscpy(ifa->ifa_label, tb[IFA_LABEL], IFNAMSIZ); else memcpy(ifa->ifa_label, dev->name, IFNAMSIZ); if (tb[IFA_RT_PRIORITY]) ifa->ifa_rt_priority = nla_get_u32(tb[IFA_RT_PRIORITY]); + if (tb[IFA_PROTO]) + ifa->ifa_proto = nla_get_u8(tb[IFA_PROTO]); + if (tb[IFA_CACHEINFO]) { struct ifa_cacheinfo *ci; @@ -940,8 +950,7 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, */ set_ifa_lifetime(ifa, valid_lft, prefered_lft); if (ifa->ifa_flags & IFA_F_MCAUTOJOIN) { - int ret = ip_mc_config(net->ipv4.mc_autojoin_sk, - true, ifa); + int ret = ip_mc_autojoin_config(net, true, ifa); if (ret < 0) { inet_free_ifa(ifa); @@ -1238,7 +1247,7 @@ out: return ret; } -static int inet_gifconf(struct net_device *dev, char __user *buf, int len, int size) +int inet_gifconf(struct net_device *dev, char __user *buf, int len, int size) { struct in_device *in_dev = __in_dev_get_rtnl(dev); const struct in_ifaddr *ifa; @@ -1566,11 +1575,11 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, } } ip_mc_up(in_dev); - /* fall through */ + fallthrough; case NETDEV_CHANGEADDR: if (!IN_DEV_ARP_NOTIFY(in_dev)) break; - /* fall through */ + fallthrough; case NETDEV_NOTIFY_PEERS: /* Send gratuitous ARP to notify of link change */ inetdev_send_gratuitous_arp(dev, in_dev); @@ -1588,7 +1597,7 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, if (inetdev_valid_mtu(dev->mtu)) break; /* disable IP when MTU is not enough */ - /* fall through */ + fallthrough; case NETDEV_UNREGISTER: inetdev_destroy(in_dev); break; @@ -1618,6 +1627,7 @@ static size_t inet_nlmsg_size(void) + nla_total_size(4) /* IFA_BROADCAST */ + nla_total_size(IFNAMSIZ) /* IFA_LABEL */ + nla_total_size(4) /* IFA_FLAGS */ + + nla_total_size(1) /* IFA_PROTO */ + nla_total_size(4) /* IFA_RT_PRIORITY */ + nla_total_size(sizeof(struct ifa_cacheinfo)); /* IFA_CACHEINFO */ } @@ -1692,6 +1702,8 @@ static int inet_fill_ifaddr(struct sk_buff *skb, struct in_ifaddr *ifa, nla_put_in_addr(skb, IFA_BROADCAST, ifa->ifa_broadcast)) || (ifa->ifa_label[0] && nla_put_string(skb, IFA_LABEL, ifa->ifa_label)) || + (ifa->ifa_proto && + nla_put_u8(skb, IFA_PROTO, ifa->ifa_proto)) || nla_put_u32(skb, IFA_FLAGS, ifa->ifa_flags) || (ifa->ifa_rt_priority && nla_put_u32(skb, IFA_RT_PRIORITY, ifa->ifa_rt_priority)) || @@ -1945,16 +1957,17 @@ static const struct nla_policy inet_af_policy[IFLA_INET_MAX+1] = { }; static int inet_validate_link_af(const struct net_device *dev, - const struct nlattr *nla) + const struct nlattr *nla, + struct netlink_ext_ack *extack) { struct nlattr *a, *tb[IFLA_INET_MAX+1]; int err, rem; - if (dev && !__in_dev_get_rcu(dev)) + if (dev && !__in_dev_get_rtnl(dev)) return -EAFNOSUPPORT; err = nla_parse_nested_deprecated(tb, IFLA_INET_MAX, nla, - inet_af_policy, NULL); + inet_af_policy, extack); if (err < 0) return err; @@ -1973,9 +1986,10 @@ static int inet_validate_link_af(const struct net_device *dev, return 0; } -static int inet_set_link_af(struct net_device *dev, const struct nlattr *nla) +static int inet_set_link_af(struct net_device *dev, const struct nlattr *nla, + struct netlink_ext_ack *extack) { - struct in_device *in_dev = __in_dev_get_rcu(dev); + struct in_device *in_dev = __in_dev_get_rtnl(dev); struct nlattr *a, *tb[IFLA_INET_MAX+1]; int rem; @@ -1983,7 +1997,7 @@ static int inet_set_link_af(struct net_device *dev, const struct nlattr *nla) return -EAFNOSUPPORT; if (nla_parse_nested_deprecated(tb, IFLA_INET_MAX, nla, NULL, NULL) < 0) - BUG(); + return -EINVAL; if (tb[IFLA_INET_CONF]) { nla_for_each_nested(a, tb[IFLA_INET_CONF], rem) @@ -2361,8 +2375,7 @@ static int devinet_conf_ifindex(struct net *net, struct ipv4_devconf *cnf) } static int devinet_conf_proc(struct ctl_table *ctl, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { int old_value = *(int *)ctl->data; int ret = proc_dointvec(ctl, write, buffer, lenp, ppos); @@ -2414,17 +2427,20 @@ static int devinet_conf_proc(struct ctl_table *ctl, int write, } static int devinet_sysctl_forward(struct ctl_table *ctl, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { int *valp = ctl->data; int val = *valp; loff_t pos = *ppos; - int ret = proc_dointvec(ctl, write, buffer, lenp, ppos); + struct net *net = ctl->extra2; + int ret; - if (write && *valp != val) { - struct net *net = ctl->extra2; + if (write && !ns_capable(net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + ret = proc_dointvec(ctl, write, buffer, lenp, ppos); + + if (write && *valp != val) { if (valp != &IPV4_DEVCONF_DFLT(net, FORWARDING)) { if (!rtnl_trylock()) { /* Restore the original values before restarting */ @@ -2458,8 +2474,7 @@ static int devinet_sysctl_forward(struct ctl_table *ctl, int write, } static int ipv4_doint_and_flush(struct ctl_table *ctl, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { int *valp = ctl->data; int val = *valp; @@ -2524,6 +2539,8 @@ static struct devinet_sysctl_table { DEVINET_SYSCTL_RW_ENTRY(ARP_IGNORE, "arp_ignore"), DEVINET_SYSCTL_RW_ENTRY(ARP_ACCEPT, "arp_accept"), DEVINET_SYSCTL_RW_ENTRY(ARP_NOTIFY, "arp_notify"), + DEVINET_SYSCTL_RW_ENTRY(ARP_EVICT_NOCARRIER, + "arp_evict_nocarrier"), DEVINET_SYSCTL_RW_ENTRY(PROXY_ARP_PVLAN, "proxy_arp_pvlan"), DEVINET_SYSCTL_RW_ENTRY(FORCE_IGMP_VERSION, "force_igmp_version"), @@ -2554,7 +2571,7 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name, struct devinet_sysctl_table *t; char path[sizeof("net/ipv4/conf/") + IFNAMSIZ]; - t = kmemdup(&devinet_sysctl, sizeof(*t), GFP_KERNEL); + t = kmemdup(&devinet_sysctl, sizeof(*t), GFP_KERNEL_ACCOUNT); if (!t) goto out; @@ -2579,7 +2596,7 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name, free: kfree(t); out: - return -ENOBUFS; + return -ENOMEM; } static void __devinet_sysctl_unregister(struct net *net, @@ -2664,11 +2681,28 @@ static __net_init int devinet_init_net(struct net *net) tbl[0].extra2 = net; #endif - if ((!IS_ENABLED(CONFIG_SYSCTL) || - sysctl_devconf_inherit_init_net != 2) && - !net_eq(net, &init_net)) { - memcpy(all, init_net.ipv4.devconf_all, sizeof(ipv4_devconf)); - memcpy(dflt, init_net.ipv4.devconf_dflt, sizeof(ipv4_devconf_dflt)); + if (!net_eq(net, &init_net)) { + switch (net_inherit_devconf()) { + case 3: + /* copy from the current netns */ + memcpy(all, current->nsproxy->net_ns->ipv4.devconf_all, + sizeof(ipv4_devconf)); + memcpy(dflt, + current->nsproxy->net_ns->ipv4.devconf_dflt, + sizeof(ipv4_devconf_dflt)); + break; + case 0: + case 1: + /* copy from init_net */ + memcpy(all, init_net.ipv4.devconf_all, + sizeof(ipv4_devconf)); + memcpy(dflt, init_net.ipv4.devconf_dflt, + sizeof(ipv4_devconf_dflt)); + break; + case 2: + /* use compiled values */ + break; + } } #ifdef CONFIG_SYSCTL @@ -2746,8 +2780,6 @@ void __init devinet_init(void) INIT_HLIST_HEAD(&inet_addr_lst[i]); register_pernet_subsys(&devinet_ops); - - register_gifconf(PF_INET, inet_gifconf); register_netdevice_notifier(&ip_netdev_notifier); queue_delayed_work(system_power_efficient_wq, &check_lifetime_work, 0); diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 103c7d599a3c..52c8047efedb 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -97,7 +97,6 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, static void esp_ssg_unref(struct xfrm_state *x, void *tmp) { - struct esp_output_extra *extra = esp_tmp_extra(tmp); struct crypto_aead *aead = x->data; int extralen = 0; u8 *iv; @@ -105,9 +104,8 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp) struct scatterlist *sg; if (x->props.flags & XFRM_STATE_ESN) - extralen += sizeof(*extra); + extralen += sizeof(struct esp_output_extra); - extra = esp_tmp_extra(tmp); iv = esp_tmp_iv(aead, tmp, extralen); req = esp_tmp_req(aead, iv); @@ -136,6 +134,7 @@ static void esp_free_tcp_sk(struct rcu_head *head) static struct sock *esp_find_tcp_sk(struct xfrm_state *x) { struct xfrm_encap_tmpl *encap = x->encap; + struct net *net = xs_net(x); struct esp_tcp_sk *esk; __be16 sport, dport; struct sock *nsk; @@ -162,7 +161,7 @@ static struct sock *esp_find_tcp_sk(struct xfrm_state *x) } spin_unlock_bh(&x->lock); - sk = inet_lookup_established(xs_net(x), &tcp_hashinfo, x->id.daddr.a4, + sk = inet_lookup_established(net, net->ipv4.tcp_death_row.hashinfo, x->id.daddr.a4, dport, x->props.saddr.a4, sport, 0); if (!sk) return ERR_PTR(-ENOENT); @@ -279,7 +278,7 @@ static void esp_output_done(struct crypto_async_request *base, int err) x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) esp_output_tail_tcp(x, skb); else - xfrm_output_resume(skb, err); + xfrm_output_resume(skb->sk, skb, err); } } @@ -309,7 +308,7 @@ static struct ip_esp_hdr *esp_output_set_extra(struct sk_buff *skb, struct esp_output_extra *extra) { /* For ESN we move the header forward by 4 bytes to - * accomodate the high bits. We will move it back after + * accommodate the high bits. We will move it back after * encryption. */ if ((x->props.flags & XFRM_STATE_ESN)) { @@ -341,22 +340,6 @@ static void esp_output_done_esn(struct crypto_async_request *base, int err) esp_output_done(base, err); } -static void esp_output_fill_trailer(u8 *tail, int tfclen, int plen, __u8 proto) -{ - /* Fill padding... */ - if (tfclen) { - memset(tail, 0, tfclen); - tail += tfclen; - } - do { - int i; - for (i = 0; i < plen - 2; i++) - tail[i] = i + 1; - } while (0); - tail[plen - 2] = plen - 2; - tail[plen - 1] = proto; -} - static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb, int encap_type, struct esp_info *esp, @@ -459,7 +442,6 @@ static int esp_output_encap(struct xfrm_state *x, struct sk_buff *skb, int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *esp) { u8 *tail; - u8 *vaddr; int nfrags; int esph_offset; struct page *page; @@ -474,6 +456,10 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * return err; } + if (ALIGN(tailen, L1_CACHE_BYTES) > PAGE_SIZE || + ALIGN(skb->data_len, L1_CACHE_BYTES) > PAGE_SIZE) + goto cow; + if (!skb_cloned(skb)) { if (tailen <= skb_tailroom(skb)) { nfrags = 1; @@ -501,14 +487,10 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * page = pfrag->page; get_page(page); - vaddr = kmap_atomic(page); - - tail = vaddr + pfrag->offset; + tail = page_address(page) + pfrag->offset; esp_output_fill_trailer(tail, esp->tfclen, esp->plen, esp->proto); - kunmap_atomic(vaddr); - nfrags = skb_shinfo(skb)->nr_frags; __skb_fill_page_desc(skb, nfrags, page, pfrag->offset, @@ -521,9 +503,7 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * nfrags++; - skb->len += tailen; - skb->data_len += tailen; - skb->truesize += tailen; + skb_len_add(skb, tailen); if (sk && sk_fullsock(sk)) refcount_add(tailen, &sk->sk_wmem_alloc); @@ -724,7 +704,6 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) static inline int esp_remove_trailer(struct sk_buff *skb) { struct xfrm_state *x = xfrm_input_state(skb); - struct xfrm_offload *xo = xfrm_offload(skb); struct crypto_aead *aead = x->data; int alen, hlen, elen; int padlen, trimlen; @@ -736,11 +715,6 @@ static inline int esp_remove_trailer(struct sk_buff *skb) hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); elen = skb->len - hlen; - if (xo && (xo->flags & XFRM_ESP_NO_TRAILER)) { - ret = xo->proto; - goto out; - } - if (skb_copy_bits(skb, skb->len - alen - 2, nexthdr, 2)) BUG(); @@ -775,7 +749,7 @@ int esp_input_done2(struct sk_buff *skb, int err) int hlen = sizeof(struct ip_esp_hdr) + crypto_aead_ivsize(aead); int ihl; - if (!xo || (xo && !(xo->flags & CRYPTO_DONE))) + if (!xo || !(xo->flags & CRYPTO_DONE)) kfree(ESP_SKB_CB(skb)->tmp); if (unlikely(err)) @@ -875,7 +849,7 @@ static void esp_input_set_header(struct sk_buff *skb, __be32 *seqhi) struct ip_esp_hdr *esph; /* For ESN we move the header forward by 4 bytes to - * accomodate the high bits. We will move it back after + * accommodate the high bits. We will move it back after * decryption. */ if ((x->props.flags & XFRM_STATE_ESN)) { @@ -1003,6 +977,7 @@ static int esp4_err(struct sk_buff *skb, u32 info) case ICMP_DEST_UNREACH: if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED) return 0; + break; case ICMP_REDIRECT: break; default: @@ -1033,16 +1008,17 @@ static void esp_destroy(struct xfrm_state *x) crypto_free_aead(aead); } -static int esp_init_aead(struct xfrm_state *x) +static int esp_init_aead(struct xfrm_state *x, struct netlink_ext_ack *extack) { char aead_name[CRYPTO_MAX_ALG_NAME]; struct crypto_aead *aead; int err; - err = -ENAMETOOLONG; if (snprintf(aead_name, CRYPTO_MAX_ALG_NAME, "%s(%s)", - x->geniv, x->aead->alg_name) >= CRYPTO_MAX_ALG_NAME) - goto error; + x->geniv, x->aead->alg_name) >= CRYPTO_MAX_ALG_NAME) { + NL_SET_ERR_MSG(extack, "Algorithm name is too long"); + return -ENAMETOOLONG; + } aead = crypto_alloc_aead(aead_name, 0, 0); err = PTR_ERR(aead); @@ -1060,11 +1036,15 @@ static int esp_init_aead(struct xfrm_state *x) if (err) goto error; + return 0; + error: + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); return err; } -static int esp_init_authenc(struct xfrm_state *x) +static int esp_init_authenc(struct xfrm_state *x, + struct netlink_ext_ack *extack) { struct crypto_aead *aead; struct crypto_authenc_key_param *param; @@ -1075,10 +1055,6 @@ static int esp_init_authenc(struct xfrm_state *x) unsigned int keylen; int err; - err = -EINVAL; - if (!x->ealg) - goto error; - err = -ENAMETOOLONG; if ((x->props.flags & XFRM_STATE_ESN)) { @@ -1087,22 +1063,28 @@ static int esp_init_authenc(struct xfrm_state *x) x->geniv ?: "", x->geniv ? "(" : "", x->aalg ? x->aalg->alg_name : "digest_null", x->ealg->alg_name, - x->geniv ? ")" : "") >= CRYPTO_MAX_ALG_NAME) + x->geniv ? ")" : "") >= CRYPTO_MAX_ALG_NAME) { + NL_SET_ERR_MSG(extack, "Algorithm name is too long"); goto error; + } } else { if (snprintf(authenc_name, CRYPTO_MAX_ALG_NAME, "%s%sauthenc(%s,%s)%s", x->geniv ?: "", x->geniv ? "(" : "", x->aalg ? x->aalg->alg_name : "digest_null", x->ealg->alg_name, - x->geniv ? ")" : "") >= CRYPTO_MAX_ALG_NAME) + x->geniv ? ")" : "") >= CRYPTO_MAX_ALG_NAME) { + NL_SET_ERR_MSG(extack, "Algorithm name is too long"); goto error; + } } aead = crypto_alloc_aead(authenc_name, 0, 0); err = PTR_ERR(aead); - if (IS_ERR(aead)) + if (IS_ERR(aead)) { + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); goto error; + } x->data = aead; @@ -1132,17 +1114,16 @@ static int esp_init_authenc(struct xfrm_state *x) err = -EINVAL; if (aalg_desc->uinfo.auth.icv_fullbits / 8 != crypto_aead_authsize(aead)) { - pr_info("ESP: %s digestsize %u != %hu\n", - x->aalg->alg_name, - crypto_aead_authsize(aead), - aalg_desc->uinfo.auth.icv_fullbits / 8); + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); goto free_key; } err = crypto_aead_setauthsize( aead, x->aalg->alg_trunc_len / 8); - if (err) + if (err) { + NL_SET_ERR_MSG(extack, "Kernel was unable to initialize cryptographic operations"); goto free_key; + } } param->enckeylen = cpu_to_be32((x->ealg->alg_key_len + 7) / 8); @@ -1157,7 +1138,7 @@ error: return err; } -static int esp_init_state(struct xfrm_state *x) +static int esp_init_state(struct xfrm_state *x, struct netlink_ext_ack *extack) { struct crypto_aead *aead; u32 align; @@ -1165,10 +1146,14 @@ static int esp_init_state(struct xfrm_state *x) x->data = NULL; - if (x->aead) - err = esp_init_aead(x); - else - err = esp_init_authenc(x); + if (x->aead) { + err = esp_init_aead(x, extack); + } else if (x->ealg) { + err = esp_init_authenc(x, extack); + } else { + NL_SET_ERR_MSG(extack, "ESP: AEAD or CRYPT must be provided"); + err = -EINVAL; + } if (err) goto error; @@ -1186,6 +1171,7 @@ static int esp_init_state(struct xfrm_state *x) switch (encap->encap_type) { default: + NL_SET_ERR_MSG(extack, "Unsupported encapsulation type for ESP"); err = -EINVAL; goto error; case UDP_ENCAP_ESPINUDP: @@ -1219,7 +1205,6 @@ static int esp4_rcv_cb(struct sk_buff *skb, int err) static const struct xfrm_type esp_type = { - .description = "ESP4", .owner = THIS_MODULE, .proto = IPPROTO_ESP, .flags = XFRM_TYPE_REPLAY_PROT, diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c index e2e219c7854a..170152772d33 100644 --- a/net/ipv4/esp4_offload.c +++ b/net/ipv4/esp4_offload.c @@ -16,6 +16,7 @@ #include <crypto/authenc.h> #include <linux/err.h> #include <linux/module.h> +#include <net/gro.h> #include <net/ip.h> #include <net/xfrm.h> #include <net/esp.h> @@ -33,12 +34,11 @@ static struct sk_buff *esp4_gro_receive(struct list_head *head, struct xfrm_state *x; __be32 seq; __be32 spi; - int err; if (!pskb_pull(skb, offset)) return NULL; - if ((err = xfrm_parse_spi(skb, IPPROTO_ESP, &spi, &seq)) != 0) + if (xfrm_parse_spi(skb, IPPROTO_ESP, &spi, &seq) != 0) goto out; xo = xfrm_offload(skb); @@ -63,10 +63,8 @@ static struct sk_buff *esp4_gro_receive(struct list_head *head, sp->olen++; xo = xfrm_offload(skb); - if (!xo) { - xfrm_state_put(x); + if (!xo) goto out_reset; - } } xo->flags |= XFRM_GRO; @@ -112,8 +110,10 @@ static struct sk_buff *xfrm4_tunnel_gso_segment(struct xfrm_state *x, struct sk_buff *skb, netdev_features_t features) { - __skb_push(skb, skb->mac_len); - return skb_mac_gso_segment(skb, features); + __be16 type = x->inner_mode.family == AF_INET6 ? htons(ETH_P_IPV6) + : htons(ETH_P_IP); + + return skb_eth_gso_segment(skb, features, type); } static struct sk_buff *xfrm4_transport_gso_segment(struct xfrm_state *x, @@ -132,6 +132,47 @@ static struct sk_buff *xfrm4_transport_gso_segment(struct xfrm_state *x, return segs; } +static struct sk_buff *xfrm4_beet_gso_segment(struct xfrm_state *x, + struct sk_buff *skb, + netdev_features_t features) +{ + struct xfrm_offload *xo = xfrm_offload(skb); + struct sk_buff *segs = ERR_PTR(-EINVAL); + const struct net_offload *ops; + u8 proto = xo->proto; + + skb->transport_header += x->props.header_len; + + if (x->sel.family != AF_INET6) { + if (proto == IPPROTO_BEETPH) { + struct ip_beet_phdr *ph = + (struct ip_beet_phdr *)skb->data; + + skb->transport_header += ph->hdrlen * 8; + proto = ph->nexthdr; + } else { + skb->transport_header -= IPV4_BEET_PHMAXLEN; + } + } else { + __be16 frag; + + skb->transport_header += + ipv6_skip_exthdr(skb, 0, &proto, &frag); + if (proto == IPPROTO_TCP) + skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV4; + } + + if (proto == IPPROTO_IPV6) + skb_shinfo(skb)->gso_type |= SKB_GSO_IPXIP4; + + __skb_pull(skb, skb_transport_offset(skb)); + ops = rcu_dereference(inet_offloads[proto]); + if (likely(ops && ops->callbacks.gso_segment)) + segs = ops->callbacks.gso_segment(skb, features); + + return segs; +} + static struct sk_buff *xfrm4_outer_mode_gso_segment(struct xfrm_state *x, struct sk_buff *skb, netdev_features_t features) @@ -141,6 +182,8 @@ static struct sk_buff *xfrm4_outer_mode_gso_segment(struct xfrm_state *x, return xfrm4_tunnel_gso_segment(x, skb, features); case XFRM_MODE_TRANSPORT: return xfrm4_transport_gso_segment(x, skb, features); + case XFRM_MODE_BEET: + return xfrm4_beet_gso_segment(x, skb, features); } return ERR_PTR(-EOPNOTSUPP); @@ -179,10 +222,12 @@ static struct sk_buff *esp4_gso_segment(struct sk_buff *skb, if ((!(skb->dev->gso_partial_features & NETIF_F_HW_ESP) && !(features & NETIF_F_HW_ESP)) || x->xso.dev != skb->dev) - esp_features = features & ~(NETIF_F_SG | NETIF_F_CSUM_MASK); + esp_features = features & ~(NETIF_F_SG | NETIF_F_CSUM_MASK | + NETIF_F_SCTP_CRC); else if (!(features & NETIF_F_HW_ESP_TX_CSUM) && !(skb->dev->gso_partial_features & NETIF_F_HW_ESP_TX_CSUM)) - esp_features = features & ~NETIF_F_CSUM_MASK; + esp_features = features & ~(NETIF_F_CSUM_MASK | + NETIF_F_SCTP_CRC); xo->flags |= XFRM_GSO_SEGMENT; @@ -247,7 +292,7 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb, netdev_features_ esp.esph = ip_esp_hdr(skb); - if (!hw_offload || (hw_offload && !skb_is_gso(skb))) { + if (!hw_offload || !skb_is_gso(skb)) { esp.nfrags = esp_output_head(x, skb, &esp); if (esp.nfrags < 0) return esp.nfrags; @@ -274,8 +319,17 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb, netdev_features_ ip_hdr(skb)->tot_len = htons(skb->len); ip_send_check(ip_hdr(skb)); - if (hw_offload) + if (hw_offload) { + if (!skb_ext_add(skb, SKB_EXT_SEC_PATH)) + return -ENOMEM; + + xo = xfrm_offload(skb); + if (!xo) + return -EINVAL; + + xo->flags |= XFRM_XMIT; return 0; + } err = esp_output_tail(x, skb, &esp); if (err) @@ -294,7 +348,6 @@ static const struct net_offload esp4_offload = { }; static const struct xfrm_type_offload esp_type_offload = { - .description = "ESP4 OFFLOAD", .owner = THIS_MODULE, .proto = IPPROTO_ESP, .input_tail = esp_input_tail, @@ -323,3 +376,4 @@ module_exit(esp4_offload_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Steffen Klassert <steffen.klassert@secunet.com>"); MODULE_ALIAS_XFRM_OFFLOAD_TYPE(AF_INET, XFRM_PROTO_ESP); +MODULE_DESCRIPTION("IPV4 GSO/GRO offload support"); diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 577db1d50a24..f361d3d56be2 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -32,6 +32,7 @@ #include <linux/list.h> #include <linux/slab.h> +#include <net/inet_dscp.h> #include <net/ip.h> #include <net/protocol.h> #include <net/route.h> @@ -290,9 +291,9 @@ __be32 fib_compute_spec_dst(struct sk_buff *skb) bool vmark = in_dev && IN_DEV_SRC_VMARK(in_dev); struct flowi4 fl4 = { .flowi4_iif = LOOPBACK_IFINDEX, - .flowi4_oif = l3mdev_master_ifindex_rcu(dev), + .flowi4_l3mdev = l3mdev_master_ifindex_rcu(dev), .daddr = ip_hdr(skb)->saddr, - .flowi4_tos = RT_TOS(ip_hdr(skb)->tos), + .flowi4_tos = ip_hdr(skb)->tos & IPTOS_RT_MASK, .flowi4_scope = scope, .flowi4_mark = vmark ? skb->mark : 0, }; @@ -309,17 +310,18 @@ bool fib_info_nh_uses_dev(struct fib_info *fi, const struct net_device *dev) { bool dev_match = false; #ifdef CONFIG_IP_ROUTE_MULTIPATH - int ret; + if (unlikely(fi->nh)) { + dev_match = nexthop_uses_dev(fi->nh, dev); + } else { + int ret; - for (ret = 0; ret < fib_info_num_path(fi); ret++) { - const struct fib_nh_common *nhc = fib_info_nhc(fi, ret); + for (ret = 0; ret < fib_info_num_path(fi); ret++) { + const struct fib_nh_common *nhc = fib_info_nhc(fi, ret); - if (nhc->nhc_dev == dev) { - dev_match = true; - break; - } else if (l3mdev_master_ifindex_rcu(nhc->nhc_dev) == dev->ifindex) { - dev_match = true; - break; + if (nhc_l3mdev_matches_dev(nhc, dev)) { + dev_match = true; + break; + } } } #else @@ -351,9 +353,8 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, bool dev_match; fl4.flowi4_oif = 0; - fl4.flowi4_iif = l3mdev_master_ifindex_rcu(dev); - if (!fl4.flowi4_iif) - fl4.flowi4_iif = oif ? : LOOPBACK_IFINDEX; + fl4.flowi4_l3mdev = l3mdev_master_ifindex_rcu(dev); + fl4.flowi4_iif = oif ? : LOOPBACK_IFINDEX; fl4.daddr = src; fl4.saddr = dst; fl4.flowi4_tos = tos; @@ -361,6 +362,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, fl4.flowi4_tun_key.tun_id = 0; fl4.flowi4_flags = 0; fl4.flowi4_uid = sock_net_uid(net, NULL); + fl4.flowi4_multipath_hash = 0; no_addr = idev->ifa_list == NULL; @@ -369,6 +371,8 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, fl4.flowi4_proto = 0; fl4.fl4_sport = 0; fl4.fl4_dport = 0; + } else { + swap(fl4.fl4_sport, fl4.fl4_dport); } if (fib_lookup(net, &fl4, &res, 0)) @@ -432,6 +436,9 @@ int fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, if (net->ipv4.fib_has_custom_local_routes || fib4_has_custom_rules(net)) goto full_check; + /* Within the same container, it is regarded as a martian source, + * and the same host but different containers are not. + */ if (inet_lookup_ifaddr_rcu(net, src)) return -EINVAL; @@ -694,7 +701,7 @@ int fib_gw_from_via(struct fib_config *cfg, struct nlattr *nla, cfg->fc_gw4 = *((__be32 *)via->rtvia_addr); break; case AF_INET6: -#ifdef CONFIG_IPV6 +#if IS_ENABLED(CONFIG_IPV6) if (alen != sizeof(struct in6_addr)) { NL_SET_ERR_MSG(extack, "Invalid IPv6 address in RTA_VIA"); return -EINVAL; @@ -731,8 +738,16 @@ static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, memset(cfg, 0, sizeof(*cfg)); rtm = nlmsg_data(nlh); + + if (!inet_validate_dscp(rtm->rtm_tos)) { + NL_SET_ERR_MSG(extack, + "Invalid dsfield (tos): ECN bits must be 0"); + err = -EINVAL; + goto errout; + } + cfg->fc_dscp = inet_dsfield_to_dscp(rtm->rtm_tos); + cfg->fc_dst_len = rtm->rtm_dst_len; - cfg->fc_tos = rtm->rtm_tos; cfg->fc_table = rtm->rtm_table; cfg->fc_protocol = rtm->rtm_protocol; cfg->fc_scope = rtm->rtm_scope; @@ -823,7 +838,7 @@ static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, if (has_gw && has_via) { NL_SET_ERR_MSG(extack, "Nexthop configuration can not contain both GATEWAY and VIA"); - goto errout; + return -EINVAL; } return 0; @@ -918,7 +933,6 @@ int ip_valid_fib_dump_req(struct net *net, const struct nlmsghdr *nlh, else filter->dump_exceptions = false; - filter->dump_all_families = (rtm->rtm_family == AF_UNSPEC); filter->flags = rtm->rtm_flags; filter->protocol = rtm->rtm_protocol; filter->rt_type = rtm->rtm_type; @@ -990,14 +1004,16 @@ static int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) if (filter.table_id) { tb = fib_get_table(net, filter.table_id); if (!tb) { - if (filter.dump_all_families) + if (rtnl_msg_family(cb->nlh) != PF_INET) return skb->len; NL_SET_ERR_MSG(cb->extack, "ipv4: FIB table does not exist"); return -ENOENT; } + rcu_read_lock(); err = fib_table_dump(tb, skb, cb, &filter); + rcu_read_unlock(); return skb->len ? : err; } @@ -1107,9 +1123,11 @@ void fib_add_ifaddr(struct in_ifaddr *ifa) return; /* Add broadcast address, if it is explicitly assigned. */ - if (ifa->ifa_broadcast && ifa->ifa_broadcast != htonl(0xFFFFFFFF)) + if (ifa->ifa_broadcast && ifa->ifa_broadcast != htonl(0xFFFFFFFF)) { fib_magic(RTM_NEWROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim, 0); + arp_invalidate(dev, ifa->ifa_broadcast, false); + } if (!ipv4_is_zeronet(prefix) && !(ifa->ifa_flags & IFA_F_SECONDARY) && (prefix != addr || ifa->ifa_prefixlen < 32)) { @@ -1119,12 +1137,11 @@ void fib_add_ifaddr(struct in_ifaddr *ifa) prefix, ifa->ifa_prefixlen, prim, ifa->ifa_rt_priority); - /* Add network specific broadcasts, when it takes a sense */ + /* Add the network broadcast address, when it makes sense */ if (ifa->ifa_prefixlen < 31) { - fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix, 32, - prim, 0); fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix | ~mask, 32, prim, 0); + arp_invalidate(dev, prefix | ~mask, false); } } } @@ -1367,13 +1384,13 @@ static void nl_fib_input(struct sk_buff *skb) return; nlh = nlmsg_hdr(skb); - frn = (struct fib_result_nl *) nlmsg_data(nlh); + frn = nlmsg_data(nlh); nl_fib_lookup(net, frn); portid = NETLINK_CB(skb).portid; /* netlink portid */ NETLINK_CB(skb).portid = 0; /* from kernel */ NETLINK_CB(skb).dst_group = 0; /* unicast */ - netlink_unicast(net->ipv4.fibnl, skb, portid, MSG_DONTWAIT); + nlmsg_unicast(net->ipv4.fibnl, skb, portid); } static int __net_init nl_fib_lookup_init(struct net *net) @@ -1408,7 +1425,7 @@ static void fib_disable_ip(struct net_device *dev, unsigned long event, static int fib_inetaddr_event(struct notifier_block *this, unsigned long event, void *ptr) { - struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; + struct in_ifaddr *ifa = ptr; struct net_device *dev = ifa->ifa_dev->dev; struct net *net = dev_net(dev); @@ -1513,6 +1530,12 @@ static int __net_init ip_fib_net_init(struct net *net) if (err) return err; +#ifdef CONFIG_IP_ROUTE_MULTIPATH + /* Default to 3-tuple */ + net->ipv4.sysctl_fib_multipath_hash_fields = + FIB_MULTIPATH_HASH_FIELD_DEFAULT_MASK; +#endif + /* Avoid false sharing : Use at least a full cache line */ size = max_t(size_t, size, L1_CACHE_BYTES); @@ -1538,7 +1561,7 @@ static void ip_fib_net_exit(struct net *net) { int i; - rtnl_lock(); + ASSERT_RTNL(); #ifdef CONFIG_IP_MULTIPLE_TABLES RCU_INIT_POINTER(net->ipv4.fib_main, NULL); RCU_INIT_POINTER(net->ipv4.fib_default, NULL); @@ -1563,7 +1586,7 @@ static void ip_fib_net_exit(struct net *net) #ifdef CONFIG_IP_MULTIPLE_TABLES fib4_rules_exit(net); #endif - rtnl_unlock(); + kfree(net->ipv4.fib_table_hash); fib4_notifier_exit(net); } @@ -1573,7 +1596,7 @@ static int __net_init fib_net_init(struct net *net) int error; #ifdef CONFIG_IP_ROUTE_CLASSID - net->ipv4.fib_num_tclassid_users = 0; + atomic_set(&net->ipv4.fib_num_tclassid_users, 0); #endif error = ip_fib_net_init(net); if (error < 0) @@ -1590,7 +1613,9 @@ out: out_proc: nl_fib_lookup_exit(net); out_nlfl: + rtnl_lock(); ip_fib_net_exit(net); + rtnl_unlock(); goto out; } @@ -1598,12 +1623,23 @@ static void __net_exit fib_net_exit(struct net *net) { fib_proc_exit(net); nl_fib_lookup_exit(net); - ip_fib_net_exit(net); +} + +static void __net_exit fib_net_exit_batch(struct list_head *net_list) +{ + struct net *net; + + rtnl_lock(); + list_for_each_entry(net, net_list, exit_list) + ip_fib_net_exit(net); + + rtnl_unlock(); } static struct pernet_operations fib_net_ops = { .init = fib_net_init, .exit = fib_net_exit, + .exit_batch = fib_net_exit_batch, }; void __init ip_fib_init(void) diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h index c092e9a55790..f9b9e26c32c1 100644 --- a/net/ipv4/fib_lookup.h +++ b/net/ipv4/fib_lookup.h @@ -4,27 +4,28 @@ #include <linux/types.h> #include <linux/list.h> +#include <net/inet_dscp.h> #include <net/ip_fib.h> #include <net/nexthop.h> struct fib_alias { struct hlist_node fa_list; struct fib_info *fa_info; - u8 fa_tos; + dscp_t fa_dscp; u8 fa_type; u8 fa_state; u8 fa_slen; u32 tb_id; s16 fa_default; - u8 offload:1, - trap:1, - unused:6; + u8 offload; + u8 trap; + u8 offload_failed; struct rcu_head rcu; }; #define FA_S_ACCESSED 0x01 -/* Dont write on fa_state unless needed, to keep it shared on all cpus */ +/* Don't write on fa_state unless needed, to keep it shared on all cpus */ static inline void fib_alias_accessed(struct fib_alias *fa) { if (!(fa->fa_state & FA_S_ACCESSED)) @@ -35,13 +36,14 @@ static inline void fib_alias_accessed(struct fib_alias *fa) void fib_release_info(struct fib_info *); struct fib_info *fib_create_info(struct fib_config *cfg, struct netlink_ext_ack *extack); -int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, +int fib_nh_match(struct net *net, struct fib_config *cfg, struct fib_info *fi, struct netlink_ext_ack *extack); bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi); int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, - struct fib_rt_info *fri, unsigned int flags); + const struct fib_rt_info *fri, unsigned int flags); void rtmsg_fib(int event, __be32 key, struct fib_alias *fa, int dst_len, u32 tb_id, const struct nl_info *info, unsigned int nlm_flags); +size_t fib_nlmsg_size(struct fib_info *fi); static inline void fib_result_assign(struct fib_result *res, struct fib_info *fi) diff --git a/net/ipv4/fib_notifier.c b/net/ipv4/fib_notifier.c index 0c28bd469a68..0e23ade74493 100644 --- a/net/ipv4/fib_notifier.c +++ b/net/ipv4/fib_notifier.c @@ -6,7 +6,6 @@ #include <linux/export.h> #include <net/net_namespace.h> #include <net/fib_notifier.h> -#include <net/netns/ipv4.h> #include <net/ip_fib.h> int call_fib4_notifier(struct notifier_block *nb, diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index f99e3bac5cab..513f475c6a53 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -23,18 +23,20 @@ #include <linux/list.h> #include <linux/rcupdate.h> #include <linux/export.h> +#include <net/inet_dscp.h> #include <net/ip.h> #include <net/route.h> #include <net/tcp.h> #include <net/ip_fib.h> #include <net/nexthop.h> #include <net/fib_rules.h> +#include <linux/indirect_call_wrapper.h> struct fib4_rule { struct fib_rule common; u8 dst_len; u8 src_len; - u8 tos; + dscp_t dscp; __be32 src; __be32 srcmask; __be32 dst; @@ -48,7 +50,7 @@ static bool fib4_rule_matchall(const struct fib_rule *rule) { struct fib4_rule *r = container_of(rule, struct fib4_rule, common); - if (r->dst_len || r->src_len || r->tos) + if (r->dst_len || r->src_len || r->dscp) return false; return fib_rule_matchall(rule); } @@ -103,8 +105,9 @@ int __fib_lookup(struct net *net, struct flowi4 *flp, } EXPORT_SYMBOL_GPL(__fib_lookup); -static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, - int flags, struct fib_lookup_arg *arg) +INDIRECT_CALLABLE_SCOPE int fib4_rule_action(struct fib_rule *rule, + struct flowi *flp, int flags, + struct fib_lookup_arg *arg) { int err = -EAGAIN; struct fib_table *tbl; @@ -138,9 +141,11 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, return err; } -static bool fib4_rule_suppress(struct fib_rule *rule, struct fib_lookup_arg *arg) +INDIRECT_CALLABLE_SCOPE bool fib4_rule_suppress(struct fib_rule *rule, + int flags, + struct fib_lookup_arg *arg) { - struct fib_result *result = (struct fib_result *) arg->result; + struct fib_result *result = arg->result; struct net_device *dev = NULL; if (result->fi) { @@ -169,7 +174,8 @@ suppress_route: return true; } -static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) +INDIRECT_CALLABLE_SCOPE int fib4_rule_match(struct fib_rule *rule, + struct flowi *fl, int flags) { struct fib4_rule *r = (struct fib4_rule *) rule; struct flowi4 *fl4 = &fl->u.ip4; @@ -180,7 +186,7 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) ((daddr ^ r->dst) & r->dstmask)) return 0; - if (r->tos && (r->tos != fl4->flowi4_tos)) + if (r->dscp && r->dscp != inet_dsfield_to_dscp(fl4->flowi4_tos)) return 0; if (rule->ip_proto && (rule->ip_proto != fl4->flowi4_proto)) @@ -211,11 +217,6 @@ static struct fib_table *fib_empty_table(struct net *net) return NULL; } -static const struct nla_policy fib4_rule_policy[FRA_MAX+1] = { - FRA_GENERIC_POLICY, - [FRA_FLOW] = { .type = NLA_U32 }, -}; - static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct fib_rule_hdr *frh, struct nlattr **tb, @@ -225,10 +226,17 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, int err = -EINVAL; struct fib4_rule *rule4 = (struct fib4_rule *) rule; + if (!inet_validate_dscp(frh->tos)) { + NL_SET_ERR_MSG(extack, + "Invalid dsfield (tos): ECN bits must be 0"); + goto errout; + } + /* IPv4 currently doesn't handle high order DSCP bits correctly */ if (frh->tos & ~IPTOS_TOS_MASK) { NL_SET_ERR_MSG(extack, "Invalid tos"); goto errout; } + rule4->dscp = inet_dsfield_to_dscp(frh->tos); /* split local/main if they are not already split */ err = fib_unmerge(net); @@ -259,7 +267,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_FLOW]) { rule4->tclassid = nla_get_u32(tb[FRA_FLOW]); if (rule4->tclassid) - net->ipv4.fib_num_tclassid_users++; + atomic_inc(&net->ipv4.fib_num_tclassid_users); } #endif @@ -270,7 +278,6 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, rule4->srcmask = inet_make_mask(rule4->src_len); rule4->dst_len = frh->dst_len; rule4->dstmask = inet_make_mask(rule4->dst_len); - rule4->tos = frh->tos; net->ipv4.fib_has_custom_rules = true; @@ -291,7 +298,7 @@ static int fib4_rule_delete(struct fib_rule *rule) #ifdef CONFIG_IP_ROUTE_CLASSID if (((struct fib4_rule *)rule)->tclassid) - net->ipv4.fib_num_tclassid_users--; + atomic_dec(&net->ipv4.fib_num_tclassid_users); #endif net->ipv4.fib_has_custom_rules = true; @@ -313,7 +320,7 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->dst_len && (rule4->dst_len != frh->dst_len)) return 0; - if (frh->tos && (rule4->tos != frh->tos)) + if (frh->tos && inet_dscp_to_dsfield(rule4->dscp) != frh->tos) return 0; #ifdef CONFIG_IP_ROUTE_CLASSID @@ -337,7 +344,7 @@ static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->dst_len = rule4->dst_len; frh->src_len = rule4->src_len; - frh->tos = rule4->tos; + frh->tos = inet_dscp_to_dsfield(rule4->dscp); if ((rule4->dst_len && nla_put_in_addr(skb, FRA_DST, rule4->dst)) || @@ -381,7 +388,6 @@ static const struct fib_rules_ops __net_initconst fib4_rules_ops_template = { .nlmsg_payload = fib4_rule_nlmsg_payload, .flush_cache = fib4_rule_flush_cache, .nlgroup = RTNLGRP_IPV4_RULE, - .policy = fib4_rule_policy, .owner = THIS_MODULE, }; diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index a803cdd9400a..f721c308248b 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -29,8 +29,10 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/netlink.h> +#include <linux/hash.h> #include <net/arp.h> +#include <net/inet_dscp.h> #include <net/ip.h> #include <net/protocol.h> #include <net/route.h> @@ -51,6 +53,7 @@ static DEFINE_SPINLOCK(fib_info_lock); static struct hlist_head *fib_info_hash; static struct hlist_head *fib_info_laddrhash; static unsigned int fib_info_hash_size; +static unsigned int fib_info_hash_bits; static unsigned int fib_info_cnt; #define DEVINDEX_HASHBITS 8 @@ -208,9 +211,7 @@ static void rt_fibinfo_free_cpus(struct rtable __rcu * __percpu *rtp) void fib_nh_common_release(struct fib_nh_common *nhc) { - if (nhc->nhc_dev) - dev_put(nhc->nhc_dev); - + netdev_put(nhc->nhc_dev, &nhc->nhc_dev_tracker); lwtstate_put(nhc->nhc_lwtstate); rt_fibinfo_free_cpus(nhc->nhc_pcpu_rth_output); rt_fibinfo_free(&nhc->nhc_rth_input); @@ -222,7 +223,7 @@ void fib_nh_release(struct net *net, struct fib_nh *fib_nh) { #ifdef CONFIG_IP_ROUTE_CLASSID if (fib_nh->nh_tclassid) - net->ipv4.fib_num_tclassid_users--; + atomic_dec(&net->ipv4.fib_num_tclassid_users); #endif fib_nh_common_release(&fib_nh->nh_common); } @@ -251,7 +252,6 @@ void free_fib_info(struct fib_info *fi) pr_warn("Freeing alive fib_info %p\n", fi); return; } - fib_info_cnt--; call_rcu(&fi->rcu, free_fib_info_rcu); } @@ -260,8 +260,12 @@ EXPORT_SYMBOL_GPL(free_fib_info); void fib_release_info(struct fib_info *fi) { spin_lock_bh(&fib_info_lock); - if (fi && --fi->fib_treeref == 0) { + if (fi && refcount_dec_and_test(&fi->fib_treeref)) { hlist_del(&fi->fib_hash); + + /* Paired with READ_ONCE() in fib_create_info(). */ + WRITE_ONCE(fib_info_cnt, fib_info_cnt - 1); + if (fi->fib_prefsrc) hlist_del(&fi->fib_lhash); if (fi->nh) { @@ -318,11 +322,15 @@ static inline int nh_comp(struct fib_info *fi, struct fib_info *ofi) static inline unsigned int fib_devindex_hashfn(unsigned int val) { - unsigned int mask = DEVINDEX_HASHSIZE - 1; + return hash_32(val, DEVINDEX_HASHBITS); +} + +static struct hlist_head * +fib_info_devhash_bucket(const struct net_device *dev) +{ + u32 val = net_hash_mix(dev_net(dev)) ^ dev->ifindex; - return (val ^ - (val >> DEVINDEX_HASHBITS) ^ - (val >> (DEVINDEX_HASHBITS * 2))) & mask; + return &fib_info_devhash[fib_devindex_hashfn(val)]; } static unsigned int fib_info_hashfn_1(int init_val, u8 protocol, u8 scope, @@ -432,12 +440,11 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev) { struct hlist_head *head; struct fib_nh *nh; - unsigned int hash; spin_lock(&fib_info_lock); - hash = fib_devindex_hashfn(dev->ifindex); - head = &fib_info_devhash[hash]; + head = fib_info_devhash_bucket(dev); + hlist_for_each_entry(nh, head, nh_hash) { if (nh->fib_nh_dev == dev && nh->fib_nh_gw4 == gw && @@ -452,7 +459,7 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev) return -1; } -static inline size_t fib_nlmsg_size(struct fib_info *fi) +size_t fib_nlmsg_size(struct fib_info *fi) { size_t payload = NLMSG_ALIGN(sizeof(struct rtmsg)) + nla_total_size(4) /* RTA_TABLE */ @@ -517,10 +524,11 @@ void rtmsg_fib(int event, __be32 key, struct fib_alias *fa, fri.tb_id = tb_id; fri.dst = key; fri.dst_len = dst_len; - fri.tos = fa->fa_tos; + fri.dscp = fa->fa_dscp; fri.type = fa->fa_type; - fri.offload = fa->offload; - fri.trap = fa->trap; + fri.offload = READ_ONCE(fa->offload); + fri.trap = READ_ONCE(fa->trap); + fri.offload_failed = READ_ONCE(fa->offload_failed); err = fib_dump_info(skb, info->portid, seq, event, &fri, nlm_flags); if (err < 0) { /* -EMSGSIZE implies BUG in fib_nlmsg_size() */ @@ -570,8 +578,9 @@ static int fib_detect_death(struct fib_info *fi, int order, return 1; } -int fib_nh_common_init(struct fib_nh_common *nhc, struct nlattr *encap, - u16 encap_type, void *cfg, gfp_t gfp_flags, +int fib_nh_common_init(struct net *net, struct fib_nh_common *nhc, + struct nlattr *encap, u16 encap_type, + void *cfg, gfp_t gfp_flags, struct netlink_ext_ack *extack) { int err; @@ -589,8 +598,9 @@ int fib_nh_common_init(struct fib_nh_common *nhc, struct nlattr *encap, err = -EINVAL; goto lwt_failure; } - err = lwtunnel_build_state(encap_type, encap, nhc->nhc_family, - cfg, &lwtstate, extack); + err = lwtunnel_build_state(net, encap_type, encap, + nhc->nhc_family, cfg, &lwtstate, + extack); if (err) goto lwt_failure; @@ -614,7 +624,7 @@ int fib_nh_init(struct net *net, struct fib_nh *nh, nh->fib_nh_family = AF_INET; - err = fib_nh_common_init(&nh->nh_common, cfg->fc_encap, + err = fib_nh_common_init(net, &nh->nh_common, cfg->fc_encap, cfg->fc_encap_type, cfg, GFP_KERNEL, extack); if (err) return err; @@ -631,7 +641,7 @@ int fib_nh_init(struct net *net, struct fib_nh *nh, #ifdef CONFIG_IP_ROUTE_CLASSID nh->nh_tclassid = cfg->fc_flow; if (nh->nh_tclassid) - net->ipv4.fib_num_tclassid_users++; + atomic_inc(&net->ipv4.fib_num_tclassid_users); #endif #ifdef CONFIG_IP_ROUTE_MULTIPATH nh->fib_nh_weight = nh_weight; @@ -661,6 +671,19 @@ static int fib_count_nexthops(struct rtnexthop *rtnh, int remaining, return nhs; } +static int fib_gw_from_attr(__be32 *gw, struct nlattr *nla, + struct netlink_ext_ack *extack) +{ + if (nla_len(nla) < sizeof(*gw)) { + NL_SET_ERR_MSG(extack, "Invalid IPv4 address in RTA_GATEWAY"); + return -EINVAL; + } + + *gw = nla_get_in_addr(nla); + + return 0; +} + /* only called when fib_nh is integrated into fib_info */ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh, int remaining, struct fib_config *cfg, @@ -703,7 +726,11 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh, return -EINVAL; } if (nla) { - fib_cfg.fc_gw4 = nla_get_in_addr(nla); + ret = fib_gw_from_attr(&fib_cfg.fc_gw4, nla, + extack); + if (ret) + goto errout; + if (fib_cfg.fc_gw4) fib_cfg.fc_gw_family = AF_INET; } else if (nlav) { @@ -713,10 +740,18 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh, } nla = nla_find(attrs, attrlen, RTA_FLOW); - if (nla) + if (nla) { + if (nla_len(nla) < sizeof(u32)) { + NL_SET_ERR_MSG(extack, "Invalid RTA_FLOW"); + return -EINVAL; + } fib_cfg.fc_flow = nla_get_u32(nla); + } fib_cfg.fc_encap = nla_find(attrs, attrlen, RTA_ENCAP); + /* RTA_ENCAP_TYPE length checked in + * lwtunnel_valid_encap_type_attr + */ nla = nla_find(attrs, attrlen, RTA_ENCAP_TYPE); if (nla) fib_cfg.fc_encap_type = nla_get_u16(nla); @@ -814,7 +849,7 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh, #endif /* CONFIG_IP_ROUTE_MULTIPATH */ -static int fib_encap_match(u16 encap_type, +static int fib_encap_match(struct net *net, u16 encap_type, struct nlattr *encap, const struct fib_nh *nh, const struct fib_config *cfg, @@ -826,7 +861,7 @@ static int fib_encap_match(u16 encap_type, if (encap_type == LWTUNNEL_ENCAP_NONE) return 0; - ret = lwtunnel_build_state(encap_type, encap, AF_INET, + ret = lwtunnel_build_state(net, encap_type, encap, AF_INET, cfg, &lwtstate, extack); if (!ret) { result = lwtunnel_cmp_encap(lwtstate, nh->fib_nh_lws); @@ -836,7 +871,7 @@ static int fib_encap_match(u16 encap_type, return result; } -int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, +int fib_nh_match(struct net *net, struct fib_config *cfg, struct fib_info *fi, struct netlink_ext_ack *extack) { #ifdef CONFIG_IP_ROUTE_MULTIPATH @@ -853,12 +888,17 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, return 1; } + /* cannot match on nexthop object attributes */ + if (fi->nh) + return 1; + if (cfg->fc_oif || cfg->fc_gw_family) { - struct fib_nh *nh = fib_info_nh(fi, 0); + struct fib_nh *nh; + nh = fib_info_nh(fi, 0); if (cfg->fc_encap) { - if (fib_encap_match(cfg->fc_encap_type, cfg->fc_encap, - nh, cfg, extack)) + if (fib_encap_match(net, cfg->fc_encap_type, + cfg->fc_encap, nh, cfg, extack)) return 1; } #ifdef CONFIG_IP_ROUTE_CLASSID @@ -901,6 +941,7 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, attrlen = rtnh_attrlen(rtnh); if (attrlen > 0) { struct nlattr *nla, *nlav, *attrs = rtnh_attrs(rtnh); + int err; nla = nla_find(attrs, attrlen, RTA_GATEWAY); nlav = nla_find(attrs, attrlen, RTA_VIA); @@ -911,12 +952,17 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, } if (nla) { + __be32 gw; + + err = fib_gw_from_attr(&gw, nla, extack); + if (err) + return err; + if (nh->fib_nh_gw_family != AF_INET || - nla_get_in_addr(nla) != nh->fib_nh_gw4) + gw != nh->fib_nh_gw4) return 1; } else if (nlav) { struct fib_config cfg2; - int err; err = fib_gw_from_via(&cfg2, nlav, extack); if (err) @@ -939,8 +985,14 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi, #ifdef CONFIG_IP_ROUTE_CLASSID nla = nla_find(attrs, attrlen, RTA_FLOW); - if (nla && nla_get_u32(nla) != nh->nh_tclassid) - return 1; + if (nla) { + if (nla_len(nla) < sizeof(u32)) { + NL_SET_ERR_MSG(extack, "Invalid RTA_FLOW"); + return -EINVAL; + } + if (nla_get_u32(nla) != nh->nh_tclassid) + return 1; + } #endif } @@ -971,7 +1023,7 @@ bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi) char tmp[TCP_CA_NAME_MAX]; bool ecn_ca = false; - nla_strlcpy(tmp, nla, sizeof(tmp)); + nla_strscpy(tmp, nla, sizeof(tmp)); val = tcp_ca_get_key_by_name(fi->fib_net, tmp, &ecn_ca); } else { if (nla_len(nla) != sizeof(u32)) @@ -1005,7 +1057,8 @@ static int fib_check_nh_v6_gw(struct net *net, struct fib_nh *nh, err = ipv6_stub->fib6_nh_init(net, &fib6_nh, &cfg, GFP_KERNEL, extack); if (!err) { nh->fib_nh_dev = fib6_nh.fib_nh_dev; - dev_hold(nh->fib_nh_dev); + netdev_hold(nh->fib_nh_dev, &nh->fib_nh_dev_tracker, + GFP_KERNEL); nh->fib_nh_oif = nh->fib_nh_dev->ifindex; nh->fib_nh_scope = RT_SCOPE_LINK; @@ -1089,7 +1142,7 @@ static int fib_check_nh_v4_gw(struct net *net, struct fib_nh *nh, u32 table, if (!netif_carrier_ok(dev)) nh->fib_nh_flags |= RTNH_F_LINKDOWN; nh->fib_nh_dev = dev; - dev_hold(dev); + netdev_hold(dev, &nh->fib_nh_dev_tracker, GFP_ATOMIC); nh->fib_nh_scope = RT_SCOPE_LINK; return 0; } @@ -1107,7 +1160,7 @@ static int fib_check_nh_v4_gw(struct net *net, struct fib_nh *nh, u32 table, if (fl4.flowi4_scope < RT_SCOPE_LINK) fl4.flowi4_scope = RT_SCOPE_LINK; - if (table) + if (table && table != RT_TABLE_MAIN) tbl = fib_get_table(net, table); if (tbl) @@ -1143,7 +1196,7 @@ static int fib_check_nh_v4_gw(struct net *net, struct fib_nh *nh, u32 table, "No egress device for nexthop gateway"); goto out; } - dev_hold(dev); + netdev_hold(dev, &nh->fib_nh_dev_tracker, GFP_ATOMIC); if (!netif_carrier_ok(dev)) nh->fib_nh_flags |= RTNH_F_LINKDOWN; err = (dev->flags & IFF_UP) ? 0 : -ENETDOWN; @@ -1177,7 +1230,7 @@ static int fib_check_nh_nongw(struct net *net, struct fib_nh *nh, } nh->fib_nh_dev = in_dev->dev; - dev_hold(nh->fib_nh_dev); + netdev_hold(nh->fib_nh_dev, &nh->fib_nh_dev_tracker, GFP_ATOMIC); nh->fib_nh_scope = RT_SCOPE_HOST; if (!netif_carrier_ok(nh->fib_nh_dev)) nh->fib_nh_flags |= RTNH_F_LINKDOWN; @@ -1202,34 +1255,13 @@ int fib_check_nh(struct net *net, struct fib_nh *nh, u32 table, u8 scope, return err; } -static inline unsigned int fib_laddr_hashfn(__be32 val) +static struct hlist_head * +fib_info_laddrhash_bucket(const struct net *net, __be32 val) { - unsigned int mask = (fib_info_hash_size - 1); + u32 slot = hash_32(net_hash_mix(net) ^ (__force u32)val, + fib_info_hash_bits); - return ((__force u32)val ^ - ((__force u32)val >> 7) ^ - ((__force u32)val >> 14)) & mask; -} - -static struct hlist_head *fib_info_hash_alloc(int bytes) -{ - if (bytes <= PAGE_SIZE) - return kzalloc(bytes, GFP_KERNEL); - else - return (struct hlist_head *) - __get_free_pages(GFP_KERNEL | __GFP_ZERO, - get_order(bytes)); -} - -static void fib_info_hash_free(struct hlist_head *hash, int bytes) -{ - if (!hash) - return; - - if (bytes <= PAGE_SIZE) - kfree(hash); - else - free_pages((unsigned long) hash, get_order(bytes)); + return &fib_info_laddrhash[slot]; } static void fib_info_hash_move(struct hlist_head *new_info_hash, @@ -1238,12 +1270,13 @@ static void fib_info_hash_move(struct hlist_head *new_info_hash, { struct hlist_head *old_info_hash, *old_laddrhash; unsigned int old_size = fib_info_hash_size; - unsigned int i, bytes; + unsigned int i; spin_lock_bh(&fib_info_lock); old_info_hash = fib_info_hash; old_laddrhash = fib_info_laddrhash; fib_info_hash_size = new_size; + fib_info_hash_bits = ilog2(new_size); for (i = 0; i < old_size; i++) { struct hlist_head *head = &fib_info_hash[i]; @@ -1261,27 +1294,25 @@ static void fib_info_hash_move(struct hlist_head *new_info_hash, } fib_info_hash = new_info_hash; + fib_info_laddrhash = new_laddrhash; for (i = 0; i < old_size; i++) { - struct hlist_head *lhead = &fib_info_laddrhash[i]; + struct hlist_head *lhead = &old_laddrhash[i]; struct hlist_node *n; struct fib_info *fi; hlist_for_each_entry_safe(fi, n, lhead, fib_lhash) { struct hlist_head *ldest; - unsigned int new_hash; - new_hash = fib_laddr_hashfn(fi->fib_prefsrc); - ldest = &new_laddrhash[new_hash]; + ldest = fib_info_laddrhash_bucket(fi->fib_net, + fi->fib_prefsrc); hlist_add_head(&fi->fib_lhash, ldest); } } - fib_info_laddrhash = new_laddrhash; spin_unlock_bh(&fib_info_lock); - bytes = old_size * sizeof(struct hlist_head *); - fib_info_hash_free(old_info_hash, bytes); - fib_info_hash_free(old_laddrhash, bytes); + kvfree(old_info_hash); + kvfree(old_laddrhash); } __be32 fib_info_update_nhc_saddr(struct net *net, struct fib_nh_common *nhc, @@ -1370,7 +1401,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg, if (!cfg->fc_mx) { fi = fib_find_info_nh(net, cfg); if (fi) { - fi->fib_treeref++; + refcount_inc(&fi->fib_treeref); return fi; } } @@ -1392,23 +1423,25 @@ struct fib_info *fib_create_info(struct fib_config *cfg, #endif err = -ENOBUFS; - if (fib_info_cnt >= fib_info_hash_size) { + + /* Paired with WRITE_ONCE() in fib_release_info() */ + if (READ_ONCE(fib_info_cnt) >= fib_info_hash_size) { unsigned int new_size = fib_info_hash_size << 1; struct hlist_head *new_info_hash; struct hlist_head *new_laddrhash; - unsigned int bytes; + size_t bytes; if (!new_size) new_size = 16; - bytes = new_size * sizeof(struct hlist_head *); - new_info_hash = fib_info_hash_alloc(bytes); - new_laddrhash = fib_info_hash_alloc(bytes); + bytes = (size_t)new_size * sizeof(struct hlist_head *); + new_info_hash = kvzalloc(bytes, GFP_KERNEL); + new_laddrhash = kvzalloc(bytes, GFP_KERNEL); if (!new_info_hash || !new_laddrhash) { - fib_info_hash_free(new_info_hash, bytes); - fib_info_hash_free(new_laddrhash, bytes); - } else + kvfree(new_info_hash); + kvfree(new_laddrhash); + } else { fib_info_hash_move(new_info_hash, new_laddrhash, new_size); - + } if (!fib_info_hash_size) goto failure; } @@ -1424,7 +1457,6 @@ struct fib_info *fib_create_info(struct fib_config *cfg, return ERR_PTR(err); } - fib_info_cnt++; fi->fib_net = net; fi->fib_protocol = cfg->fc_protocol; fi->fib_scope = cfg->fc_scope; @@ -1507,6 +1539,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg, err = -ENODEV; if (!nh->fib_nh_dev) goto failure; + netdev_tracker_alloc(nh->fib_nh_dev, &nh->fib_nh_dev_tracker, + GFP_KERNEL); } else { int linkdown = 0; @@ -1544,19 +1578,20 @@ link_it: if (ofi) { fi->fib_dead = 1; free_fib_info(fi); - ofi->fib_treeref++; + refcount_inc(&ofi->fib_treeref); return ofi; } - fi->fib_treeref++; + refcount_set(&fi->fib_treeref, 1); refcount_set(&fi->fib_clntref, 1); spin_lock_bh(&fib_info_lock); + fib_info_cnt++; hlist_add_head(&fi->fib_hash, &fib_info_hash[fib_info_hashfn(fi)]); if (fi->fib_prefsrc) { struct hlist_head *head; - head = &fib_info_laddrhash[fib_laddr_hashfn(fi->fib_prefsrc)]; + head = fib_info_laddrhash_bucket(net, fi->fib_prefsrc); hlist_add_head(&fi->fib_lhash, head); } if (fi->nh) { @@ -1564,12 +1599,10 @@ link_it: } else { change_nexthops(fi) { struct hlist_head *head; - unsigned int hash; if (!nexthop_nh->fib_nh_dev) continue; - hash = fib_devindex_hashfn(nexthop_nh->fib_nh_dev->ifindex); - head = &fib_info_devhash[hash]; + head = fib_info_devhash_bucket(nexthop_nh->fib_nh_dev); hlist_add_head(&nexthop_nh->nh_hash, head); } endfor_nexthops(fi) } @@ -1639,9 +1672,8 @@ int fib_nexthop_info(struct sk_buff *skb, const struct fib_nh_common *nhc, break; } - *flags |= (nhc->nhc_flags & RTNH_F_ONLINK); - if (nhc->nhc_flags & RTNH_F_OFFLOAD) - *flags |= RTNH_F_OFFLOAD; + *flags |= (nhc->nhc_flags & + (RTNH_F_ONLINK | RTNH_F_OFFLOAD | RTNH_F_TRAP)); if (!skip_oif && nhc->nhc_dev && nla_put_u32(skb, RTA_OIF, nhc->nhc_dev->ifindex)) @@ -1661,7 +1693,7 @@ EXPORT_SYMBOL_GPL(fib_nexthop_info); #if IS_ENABLED(CONFIG_IP_ROUTE_MULTIPATH) || IS_ENABLED(CONFIG_IPV6) int fib_add_nexthop(struct sk_buff *skb, const struct fib_nh_common *nhc, - int nh_weight, u8 rt_family) + int nh_weight, u8 rt_family, u32 nh_tclassid) { const struct net_device *dev = nhc->nhc_dev; struct rtnexthop *rtnh; @@ -1679,6 +1711,9 @@ int fib_add_nexthop(struct sk_buff *skb, const struct fib_nh_common *nhc, rtnh->rtnh_flags = flags; + if (nh_tclassid && nla_put_u32(skb, RTA_FLOW, nh_tclassid)) + goto nla_put_failure; + /* length of rtnetlink header + attributes */ rtnh->rtnh_len = nlmsg_get_pos(skb) - (void *)rtnh; @@ -1706,14 +1741,13 @@ static int fib_add_multipath(struct sk_buff *skb, struct fib_info *fi) } for_nexthops(fi) { - if (fib_add_nexthop(skb, &nh->nh_common, nh->fib_nh_weight, - AF_INET) < 0) - goto nla_put_failure; + u32 nh_tclassid = 0; #ifdef CONFIG_IP_ROUTE_CLASSID - if (nh->nh_tclassid && - nla_put_u32(skb, RTA_FLOW, nh->nh_tclassid)) - goto nla_put_failure; + nh_tclassid = nh->nh_tclassid; #endif + if (fib_add_nexthop(skb, &nh->nh_common, nh->fib_nh_weight, + AF_INET, nh_tclassid) < 0) + goto nla_put_failure; } endfor_nexthops(fi); mp_end: @@ -1732,7 +1766,7 @@ static int fib_add_multipath(struct sk_buff *skb, struct fib_info *fi) #endif int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event, - struct fib_rt_info *fri, unsigned int flags) + const struct fib_rt_info *fri, unsigned int flags) { unsigned int nhs = fib_info_num_path(fri->fi); struct fib_info *fi = fri->fi; @@ -1748,7 +1782,7 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event, rtm->rtm_family = AF_INET; rtm->rtm_dst_len = fri->dst_len; rtm->rtm_src_len = 0; - rtm->rtm_tos = fri->tos; + rtm->rtm_tos = inet_dscp_to_dsfield(fri->dscp); if (tb_id < 256) rtm->rtm_table = tb_id; else @@ -1778,6 +1812,8 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event, goto nla_put_failure; if (nexthop_is_blackhole(fi->nh)) rtm->rtm_type = RTN_BLACKHOLE; + if (!READ_ONCE(fi->fib_net->ipv4.sysctl_nexthop_compat_mode)) + goto offload; } if (nhs == 1) { @@ -1803,10 +1839,13 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event, goto nla_put_failure; } +offload: if (fri->offload) rtm->rtm_flags |= RTM_F_OFFLOAD; if (fri->trap) rtm->rtm_flags |= RTM_F_TRAP; + if (fri->offload_failed) + rtm->rtm_flags |= RTM_F_OFFLOAD_FAILED; nlmsg_end(skb, nlh); return 0; @@ -1824,16 +1863,16 @@ nla_put_failure: */ int fib_sync_down_addr(struct net_device *dev, __be32 local) { - int ret = 0; - unsigned int hash = fib_laddr_hashfn(local); - struct hlist_head *head = &fib_info_laddrhash[hash]; int tb_id = l3mdev_fib_table(dev) ? : RT_TABLE_MAIN; struct net *net = dev_net(dev); + struct hlist_head *head; struct fib_info *fi; + int ret = 0; if (!fib_info_laddrhash || local == 0) return 0; + head = fib_info_laddrhash_bucket(net, local); hlist_for_each_entry(fi, head, fib_lhash) { if (!net_eq(fi->fib_net, net) || fi->fib_tb_id != tb_id) @@ -1867,6 +1906,7 @@ static int call_fib_nh_notifiers(struct fib_nh *nh, (nh->fib_nh_flags & RTNH_F_DEAD)) return call_fib4_notifiers(dev_net(nh->fib_nh_dev), event_type, &info.info); + break; default: break; } @@ -1914,8 +1954,7 @@ void fib_nhc_update_mtu(struct fib_nh_common *nhc, u32 new, u32 orig) void fib_sync_mtu(struct net_device *dev, u32 orig_mtu) { - unsigned int hash = fib_devindex_hashfn(dev->ifindex); - struct hlist_head *head = &fib_info_devhash[hash]; + struct hlist_head *head = fib_info_devhash_bucket(dev); struct fib_nh *nh; hlist_for_each_entry(nh, head, nh_hash) { @@ -1934,12 +1973,11 @@ void fib_sync_mtu(struct net_device *dev, u32 orig_mtu) */ int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force) { - int ret = 0; - int scope = RT_SCOPE_NOWHERE; + struct hlist_head *head = fib_info_devhash_bucket(dev); struct fib_info *prev_fi = NULL; - unsigned int hash = fib_devindex_hashfn(dev->ifindex); - struct hlist_head *head = &fib_info_devhash[hash]; + int scope = RT_SCOPE_NOWHERE; struct fib_nh *nh; + int ret = 0; if (force) scope = -1; @@ -1962,7 +2000,7 @@ int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force) case NETDEV_DOWN: case NETDEV_UNREGISTER: nexthop_nh->fib_nh_flags |= RTNH_F_DEAD; - /* fall through */ + fallthrough; case NETDEV_CHANGE: nexthop_nh->fib_nh_flags |= RTNH_F_LINKDOWN; break; @@ -1984,7 +2022,7 @@ int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force) case NETDEV_DOWN: case NETDEV_UNREGISTER: fi->fib_flags |= RTNH_F_DEAD; - /* fall through */ + fallthrough; case NETDEV_CHANGE: fi->fib_flags |= RTNH_F_LINKDOWN; break; @@ -2008,35 +2046,36 @@ static void fib_select_default(const struct flowi4 *flp, struct fib_result *res) int order = -1, last_idx = -1; struct fib_alias *fa, *fa1 = NULL; u32 last_prio = res->fi->fib_priority; - u8 last_tos = 0; + dscp_t last_dscp = 0; hlist_for_each_entry_rcu(fa, fa_head, fa_list) { struct fib_info *next_fi = fa->fa_info; - struct fib_nh *nh; + struct fib_nh_common *nhc; if (fa->fa_slen != slen) continue; - if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos) + if (fa->fa_dscp && + fa->fa_dscp != inet_dsfield_to_dscp(flp->flowi4_tos)) continue; if (fa->tb_id != tb->tb_id) continue; if (next_fi->fib_priority > last_prio && - fa->fa_tos == last_tos) { - if (last_tos) + fa->fa_dscp == last_dscp) { + if (last_dscp) continue; break; } if (next_fi->fib_flags & RTNH_F_DEAD) continue; - last_tos = fa->fa_tos; + last_dscp = fa->fa_dscp; last_prio = next_fi->fib_priority; if (next_fi->fib_scope != res->scope || fa->fa_type != RTN_UNICAST) continue; - nh = fib_info_nh(next_fi, 0); - if (!nh->fib_nh_gw4 || nh->fib_nh_scope != RT_SCOPE_LINK) + nhc = fib_info_nhc(next_fi, 0); + if (!nhc->nhc_gw_family || nhc->nhc_scope != RT_SCOPE_LINK) continue; fib_alias_accessed(fa); @@ -2084,7 +2123,6 @@ out: int fib_sync_up(struct net_device *dev, unsigned char nh_flags) { struct fib_info *prev_fi; - unsigned int hash; struct hlist_head *head; struct fib_nh *nh; int ret; @@ -2100,8 +2138,7 @@ int fib_sync_up(struct net_device *dev, unsigned char nh_flags) } prev_fi = NULL; - hash = fib_devindex_hashfn(dev->ifindex); - head = &fib_info_devhash[hash]; + head = fib_info_devhash_bucket(dev); ret = 0; hlist_for_each_entry(nh, head, nh_hash) { @@ -2180,7 +2217,7 @@ void fib_select_multipath(struct fib_result *res, int hash) } change_nexthops(fi) { - if (net->ipv4.sysctl_fib_multipath_use_neigh) { + if (READ_ONCE(net->ipv4.sysctl_fib_multipath_use_neigh)) { if (!fib_good_nh(nexthop_nh)) continue; if (!first) { @@ -2203,7 +2240,7 @@ void fib_select_multipath(struct fib_result *res, int hash) void fib_select_path(struct net *net, struct fib_result *res, struct flowi4 *fl4, const struct sk_buff *skb) { - if (fl4->flowi4_oif && !(fl4->flowi4_flags & FLOWI_FLAG_SKIP_NH_OIF)) + if (fl4->flowi4_oif) goto check_saddr; #ifdef CONFIG_IP_ROUTE_MULTIPATH diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index ff0c24371e33..452ff177e4da 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -13,7 +13,7 @@ * * An experimental study of compression methods for dynamic tries * Stefan Nilsson and Matti Tikkanen. Algorithmica, 33(1):19-33, 2002. - * http://www.csc.kth.se/~snilsson/software/dyntrie2/ + * https://www.csc.kth.se/~snilsson/software/dyntrie2/ * * IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson * IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999 @@ -35,9 +35,6 @@ * Paul E. McKenney <paulmck@us.ibm.com> * Patrick McHardy <kaber@trash.net> */ - -#define VERSION "0.409" - #include <linux/cache.h> #include <linux/uaccess.h> #include <linux/bitops.h> @@ -64,6 +61,7 @@ #include <linux/vmalloc.h> #include <linux/notifier.h> #include <net/net_namespace.h> +#include <net/inet_dscp.h> #include <net/ip.h> #include <net/protocol.h> #include <net/route.h> @@ -84,7 +82,7 @@ static int call_fib_entry_notifier(struct notifier_block *nb, .dst = dst, .dst_len = dst_len, .fi = fa->fa_info, - .tos = fa->fa_tos, + .dscp = fa->fa_dscp, .type = fa->fa_type, .tb_id = fa->tb_id, }; @@ -101,7 +99,7 @@ static int call_fib_entry_notifiers(struct net *net, .dst = dst, .dst_len = dst_len, .fi = fa->fa_info, - .tos = fa->fa_tos, + .dscp = fa->fa_dscp, .type = fa->fa_type, .tb_id = fa->tb_id, }; @@ -304,8 +302,6 @@ static inline void alias_free_mem_rcu(struct fib_alias *fa) call_rcu(&fa->rcu, __alias_free_mem); } -#define TNODE_KMALLOC_MAX \ - ilog2((PAGE_SIZE - TNODE_SIZE(0)) / sizeof(struct key_vector *)) #define TNODE_VMALLOC_MAX \ ilog2((SIZE_MAX - TNODE_SIZE(0)) / sizeof(struct key_vector *)) @@ -502,7 +498,7 @@ static void tnode_free(struct key_vector *tn) tn = container_of(head, struct tnode, rcu)->kv; } - if (tnode_free_size >= sysctl_fib_sync_mem) { + if (tnode_free_size >= READ_ONCE(sysctl_fib_sync_mem)) { tnode_free_size = 0; synchronize_rcu(); } @@ -978,13 +974,13 @@ static struct key_vector *fib_find_node(struct trie *t, return n; } -/* Return the first fib alias matching TOS with +/* Return the first fib alias matching DSCP with * priority less than or equal to PRIO. * If 'find_first' is set, return the first matching - * fib alias, regardless of TOS and priority. + * fib alias, regardless of DSCP and priority. */ static struct fib_alias *fib_find_alias(struct hlist_head *fah, u8 slen, - u8 tos, u32 prio, u32 tb_id, + dscp_t dscp, u32 prio, u32 tb_id, bool find_first) { struct fib_alias *fa; @@ -993,6 +989,10 @@ static struct fib_alias *fib_find_alias(struct hlist_head *fah, u8 slen, return NULL; hlist_for_each_entry(fa, fah, fa_list) { + /* Avoid Sparse warning when using dscp_t in inequalities */ + u8 __fa_dscp = inet_dscp_to_dsfield(fa->fa_dscp); + u8 __dscp = inet_dscp_to_dsfield(dscp); + if (fa->fa_slen < slen) continue; if (fa->fa_slen != slen) @@ -1003,9 +1003,9 @@ static struct fib_alias *fib_find_alias(struct hlist_head *fah, u8 slen, break; if (find_first) return fa; - if (fa->fa_tos > tos) + if (__fa_dscp > __dscp) continue; - if (fa->fa_info->fib_priority >= prio || fa->fa_tos < tos) + if (fa->fa_info->fib_priority >= prio || __fa_dscp < __dscp) return fa; } @@ -1032,7 +1032,7 @@ fib_find_matching_alias(struct net *net, const struct fib_rt_info *fri) hlist_for_each_entry_rcu(fa, &l->leaf, fa_list) { if (fa->fa_slen == slen && fa->tb_id == fri->tb_id && - fa->fa_tos == fri->tos && fa->fa_info == fri->fi && + fa->fa_dscp == fri->dscp && fa->fa_info == fri->fi && fa->fa_type == fri->type) return fa; } @@ -1042,7 +1042,10 @@ fib_find_matching_alias(struct net *net, const struct fib_rt_info *fri) void fib_alias_hw_flags_set(struct net *net, const struct fib_rt_info *fri) { + u8 fib_notify_on_flag_change; struct fib_alias *fa_match; + struct sk_buff *skb; + int err; rcu_read_lock(); @@ -1050,9 +1053,48 @@ void fib_alias_hw_flags_set(struct net *net, const struct fib_rt_info *fri) if (!fa_match) goto out; - fa_match->offload = fri->offload; - fa_match->trap = fri->trap; + /* These are paired with the WRITE_ONCE() happening in this function. + * The reason is that we are only protected by RCU at this point. + */ + if (READ_ONCE(fa_match->offload) == fri->offload && + READ_ONCE(fa_match->trap) == fri->trap && + READ_ONCE(fa_match->offload_failed) == fri->offload_failed) + goto out; + + WRITE_ONCE(fa_match->offload, fri->offload); + WRITE_ONCE(fa_match->trap, fri->trap); + + fib_notify_on_flag_change = READ_ONCE(net->ipv4.sysctl_fib_notify_on_flag_change); + + /* 2 means send notifications only if offload_failed was changed. */ + if (fib_notify_on_flag_change == 2 && + READ_ONCE(fa_match->offload_failed) == fri->offload_failed) + goto out; + WRITE_ONCE(fa_match->offload_failed, fri->offload_failed); + + if (!fib_notify_on_flag_change) + goto out; + + skb = nlmsg_new(fib_nlmsg_size(fa_match->fa_info), GFP_ATOMIC); + if (!skb) { + err = -ENOBUFS; + goto errout; + } + + err = fib_dump_info(skb, 0, 0, RTM_NEWROUTE, fri, 0); + if (err < 0) { + /* -EMSGSIZE implies BUG in fib_nlmsg_size() */ + WARN_ON(err == -EMSGSIZE); + kfree_skb(skb); + goto errout; + } + + rtnl_notify(skb, net, 0, RTNLGRP_IPV4_ROUTE, NULL, GFP_ATOMIC); + goto out; + +errout: + rtnl_set_sk_err(net, RTNLGRP_IPV4_ROUTE, err); out: rcu_read_unlock(); } @@ -1180,7 +1222,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, struct fib_info *fi; u8 plen = cfg->fc_dst_len; u8 slen = KEYLENGTH - plen; - u8 tos = cfg->fc_tos; + dscp_t dscp; u32 key; int err; @@ -1197,12 +1239,13 @@ int fib_table_insert(struct net *net, struct fib_table *tb, goto err; } + dscp = cfg->fc_dscp; l = fib_find_node(t, &tp, key); - fa = l ? fib_find_alias(&l->leaf, slen, tos, fi->fib_priority, + fa = l ? fib_find_alias(&l->leaf, slen, dscp, fi->fib_priority, tb->tb_id, false) : NULL; /* Now fa, if non-NULL, points to the first fib alias - * with the same keys [prefix,tos,priority], if such key already + * with the same keys [prefix,dscp,priority], if such key already * exists or to the node before which we will insert new one. * * If fa is NULL, we will need to allocate a new one and @@ -1210,7 +1253,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, * of the new alias. */ - if (fa && fa->fa_tos == tos && + if (fa && fa->fa_dscp == dscp && fa->fa_info->fib_priority == fi->fib_priority) { struct fib_alias *fa_first, *fa_match; @@ -1230,7 +1273,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, hlist_for_each_entry_from(fa, fa_list) { if ((fa->fa_slen != slen) || (fa->tb_id != tb->tb_id) || - (fa->fa_tos != tos)) + (fa->fa_dscp != dscp)) break; if (fa->fa_info->fib_priority != fi->fib_priority) break; @@ -1258,7 +1301,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, goto out; fi_drop = fa->fa_info; - new_fa->fa_tos = fa->fa_tos; + new_fa->fa_dscp = fa->fa_dscp; new_fa->fa_info = fi; new_fa->fa_type = cfg->fc_type; state = fa->fa_state; @@ -1268,6 +1311,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, new_fa->fa_default = -1; new_fa->offload = 0; new_fa->trap = 0; + new_fa->offload_failed = 0; hlist_replace_rcu(&fa->fa_list, &new_fa->fa_list); @@ -1320,7 +1364,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, goto out; new_fa->fa_info = fi; - new_fa->fa_tos = tos; + new_fa->fa_dscp = dscp; new_fa->fa_type = cfg->fc_type; new_fa->fa_state = 0; new_fa->fa_slen = slen; @@ -1328,6 +1372,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, new_fa->fa_default = -1; new_fa->offload = 0; new_fa->trap = 0; + new_fa->offload_failed = 0; /* Insert new entry to the list. */ err = fib_insert_alias(t, tp, l, new_fa, fa, key); @@ -1376,6 +1421,23 @@ static inline t_key prefix_mismatch(t_key key, struct key_vector *n) return (key ^ prefix) & (prefix | -prefix); } +bool fib_lookup_good_nhc(const struct fib_nh_common *nhc, int fib_flags, + const struct flowi4 *flp) +{ + if (nhc->nhc_flags & RTNH_F_DEAD) + return false; + + if (ip_ignore_linkdown(nhc->nhc_dev) && + nhc->nhc_flags & RTNH_F_LINKDOWN && + !(fib_flags & FIB_LOOKUP_IGNORE_LINKSTATE)) + return false; + + if (flp->flowi4_oif && flp->flowi4_oif != nhc->nhc_oif) + return false; + + return true; +} + /* should be called with rcu_read_lock */ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp, struct fib_result *res, int fib_flags) @@ -1508,13 +1570,15 @@ found: /* Step 3: Process the leaf, if that fails fall back to backtracing */ hlist_for_each_entry_rcu(fa, &n->leaf, fa_list) { struct fib_info *fi = fa->fa_info; + struct fib_nh_common *nhc; int nhsel, err; if ((BITS_PER_LONG > KEYLENGTH) || (fa->fa_slen < KEYLENGTH)) { if (index >= (1ul << fa->fa_slen)) continue; } - if (fa->fa_tos && fa->fa_tos != flp->flowi4_tos) + if (fa->fa_dscp && + inet_dscp_to_dsfield(fa->fa_dscp) != flp->flowi4_tos) continue; if (fi->fib_dead) continue; @@ -1533,26 +1597,25 @@ out_reject: if (fi->fib_flags & RTNH_F_DEAD) continue; - if (unlikely(fi->nh && nexthop_is_blackhole(fi->nh))) { - err = fib_props[RTN_BLACKHOLE].error; - goto out_reject; + if (unlikely(fi->nh)) { + if (nexthop_is_blackhole(fi->nh)) { + err = fib_props[RTN_BLACKHOLE].error; + goto out_reject; + } + + nhc = nexthop_get_nhc_lookup(fi->nh, fib_flags, flp, + &nhsel); + if (nhc) + goto set_result; + goto miss; } for (nhsel = 0; nhsel < fib_info_num_path(fi); nhsel++) { - struct fib_nh_common *nhc = fib_info_nhc(fi, nhsel); + nhc = fib_info_nhc(fi, nhsel); - if (nhc->nhc_flags & RTNH_F_DEAD) - continue; - if (ip_ignore_linkdown(nhc->nhc_dev) && - nhc->nhc_flags & RTNH_F_LINKDOWN && - !(fib_flags & FIB_LOOKUP_IGNORE_LINKSTATE)) + if (!fib_lookup_good_nhc(nhc, fib_flags, flp)) continue; - if (!(flp->flowi4_flags & FLOWI_FLAG_SKIP_NH_OIF)) { - if (flp->flowi4_oif && - flp->flowi4_oif != nhc->nhc_oif) - continue; - } - +set_result: if (!(fib_flags & FIB_LOOKUP_NOREF)) refcount_inc(&fi->fib_clntref); @@ -1573,6 +1636,7 @@ out_reject: return err; } } +miss: #ifdef CONFIG_IP_FIB_TRIE_STATS this_cpu_inc(stats->semantic_match_miss); #endif @@ -1650,7 +1714,7 @@ int fib_table_delete(struct net *net, struct fib_table *tb, struct key_vector *l, *tp; u8 plen = cfg->fc_dst_len; u8 slen = KEYLENGTH - plen; - u8 tos = cfg->fc_tos; + dscp_t dscp; u32 key; key = ntohl(cfg->fc_dst); @@ -1662,11 +1726,13 @@ int fib_table_delete(struct net *net, struct fib_table *tb, if (!l) return -ESRCH; - fa = fib_find_alias(&l->leaf, slen, tos, 0, tb->tb_id, false); + dscp = cfg->fc_dscp; + fa = fib_find_alias(&l->leaf, slen, dscp, 0, tb->tb_id, false); if (!fa) return -ESRCH; - pr_debug("Deleting %08x/%d tos=%d t=%p\n", key, plen, tos, t); + pr_debug("Deleting %08x/%d dsfield=0x%02x t=%p\n", key, plen, + inet_dscp_to_dsfield(dscp), t); fa_to_delete = NULL; hlist_for_each_entry_from(fa, fa_list) { @@ -1674,7 +1740,7 @@ int fib_table_delete(struct net *net, struct fib_table *tb, if ((fa->fa_slen != slen) || (fa->tb_id != tb->tb_id) || - (fa->fa_tos != tos)) + (fa->fa_dscp != dscp)) break; if ((!cfg->fc_type || fa->fa_type == cfg->fc_type) && @@ -1684,7 +1750,7 @@ int fib_table_delete(struct net *net, struct fib_table *tb, fi->fib_prefsrc == cfg->fc_prefsrc) && (!cfg->fc_protocol || fi->fib_protocol == cfg->fc_protocol) && - fib_nh_match(cfg, fi, extack) == 0 && + fib_nh_match(net, cfg, fi, extack) == 0 && fib_metrics_match(cfg, fi)) { fa_to_delete = fa; break; @@ -1848,7 +1914,7 @@ struct fib_table *fib_trie_unmerge(struct fib_table *oldtb) while ((l = leaf_walk_rcu(&tp, key)) != NULL) { struct key_vector *local_l = NULL, *local_tp; - hlist_for_each_entry_rcu(fa, &l->leaf, fa_list) { + hlist_for_each_entry(fa, &l->leaf, fa_list) { struct fib_alias *new_fa; if (local_tb->tb_id != fa->tb_id) @@ -2084,15 +2150,6 @@ static void __fib_info_notify_update(struct net *net, struct fib_table *tb, rtmsg_fib(RTM_NEWROUTE, htonl(n->key), fa, KEYLENGTH - fa->fa_slen, tb->tb_id, info, NLM_F_REPLACE); - - /* call_fib_entry_notifiers will be removed when - * in-kernel notifier is implemented and supported - * for nexthop objects - */ - call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_REPLACE, - n->key, - KEYLENGTH - fa->fa_slen, fa, - NULL); } } } @@ -2105,7 +2162,8 @@ void fib_info_notify_update(struct net *net, struct nl_info *info) struct hlist_head *head = &net->ipv4.fib_table_hash[h]; struct fib_table *tb; - hlist_for_each_entry_rcu(tb, head, tb_hlist) + hlist_for_each_entry_rcu(tb, head, tb_hlist, + lockdep_rtnl_is_held()) __fib_info_notify_update(net, tb, info); } } @@ -2250,10 +2308,11 @@ static int fn_trie_dump_leaf(struct key_vector *l, struct fib_table *tb, fri.tb_id = tb->tb_id; fri.dst = xkey; fri.dst_len = KEYLENGTH - fa->fa_slen; - fri.tos = fa->fa_tos; + fri.dscp = fa->fa_dscp; fri.type = fa->fa_type; - fri.offload = fa->offload; - fri.trap = fa->trap; + fri.offload = READ_ONCE(fa->offload); + fri.trap = READ_ONCE(fa->trap); + fri.offload_failed = READ_ONCE(fa->offload_failed); err = fib_dump_info(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, @@ -2334,11 +2393,11 @@ void __init fib_trie_init(void) { fn_alias_kmem = kmem_cache_create("ip_fib_alias", sizeof(struct fib_alias), - 0, SLAB_PANIC, NULL); + 0, SLAB_PANIC | SLAB_ACCOUNT, NULL); trie_leaf_kmem = kmem_cache_create("ip_fib_trie", LEAF_SIZE, - 0, SLAB_PANIC, NULL); + 0, SLAB_PANIC | SLAB_ACCOUNT, NULL); } struct fib_table *fib_trie_table(u32 id, struct fib_table *alias) @@ -2569,7 +2628,7 @@ static void fib_table_print(struct seq_file *seq, struct fib_table *tb) static int fib_triestat_seq_show(struct seq_file *seq, void *v) { - struct net *net = (struct net *)seq->private; + struct net *net = seq->private; unsigned int h; seq_printf(seq, @@ -2577,6 +2636,7 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v) " %zd bytes, size of tnode: %zd bytes.\n", LEAF_SIZE, TNODE_SIZE(0)); + rcu_read_lock(); for (h = 0; h < FIB_TABLE_HASHSZ; h++) { struct hlist_head *head = &net->ipv4.fib_table_hash[h]; struct fib_table *tb; @@ -2596,7 +2656,9 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v) trie_show_usage(seq, t->stats); #endif } + cond_resched_rcu(); } + rcu_read_unlock(); return 0; } @@ -2758,8 +2820,9 @@ static int fib_trie_seq_show(struct seq_file *seq, void *v) fa->fa_info->fib_scope), rtn_type(buf2, sizeof(buf2), fa->fa_type)); - if (fa->fa_tos) - seq_printf(seq, " tos=%d", fa->fa_tos); + if (fa->fa_dscp) + seq_printf(seq, " tos=%d", + inet_dscp_to_dsfield(fa->fa_dscp)); seq_putc(seq, '\n'); } } diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c index dcc79ff54b41..0c3c6d0cee29 100644 --- a/net/ipv4/fou.c +++ b/net/ipv4/fou.c @@ -9,13 +9,13 @@ #include <linux/types.h> #include <linux/kernel.h> #include <net/genetlink.h> +#include <net/gro.h> #include <net/gue.h> #include <net/fou.h> #include <net/ip.h> #include <net/protocol.h> #include <net/udp.h> #include <net/udp_tunnel.h> -#include <net/xfrm.h> #include <uapi/linux/fou.h> #include <uapi/linux/genetlink.h> @@ -230,14 +230,14 @@ static struct sk_buff *fou_gro_receive(struct sock *sk, struct list_head *head, struct sk_buff *skb) { + const struct net_offload __rcu **offloads; u8 proto = fou_from_sock(sk)->protocol; - const struct net_offload **offloads; const struct net_offload *ops; struct sk_buff *pp = NULL; /* We can clear the encap_mark for FOU as we are essentially doing * one of two possible things. We are either adding an L4 tunnel - * header to the outer L3 tunnel header, or we are are simply + * header to the outer L3 tunnel header, or we are simply * treating the GRE tunnel header as though it is a UDP protocol * specific header such as VXLAN or GENEVE. */ @@ -246,41 +246,35 @@ static struct sk_buff *fou_gro_receive(struct sock *sk, /* Flag this frame as already having an outer encap header */ NAPI_GRO_CB(skb)->is_fou = 1; - rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]); if (!ops || !ops->callbacks.gro_receive) - goto out_unlock; + goto out; pp = call_gro_receive(ops->callbacks.gro_receive, head, skb); -out_unlock: - rcu_read_unlock(); - +out: return pp; } static int fou_gro_complete(struct sock *sk, struct sk_buff *skb, int nhoff) { - const struct net_offload *ops; + const struct net_offload __rcu **offloads; u8 proto = fou_from_sock(sk)->protocol; + const struct net_offload *ops; int err = -ENOSYS; - const struct net_offload **offloads; - rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]); if (WARN_ON(!ops || !ops->callbacks.gro_complete)) - goto out_unlock; + goto out; err = ops->callbacks.gro_complete(skb, nhoff); skb_set_inner_mac_header(skb, nhoff); -out_unlock: - rcu_read_unlock(); - +out: return err; } @@ -311,7 +305,7 @@ static struct sk_buff *gue_gro_receive(struct sock *sk, struct list_head *head, struct sk_buff *skb) { - const struct net_offload **offloads; + const struct net_offload __rcu **offloads; const struct net_offload *ops; struct sk_buff *pp = NULL; struct sk_buff *p; @@ -329,12 +323,9 @@ static struct sk_buff *gue_gro_receive(struct sock *sk, off = skb_gro_offset(skb); len = off + sizeof(*guehdr); - guehdr = skb_gro_header_fast(skb, off); - if (skb_gro_header_hard(skb, len)) { - guehdr = skb_gro_header_slow(skb, len, off); - if (unlikely(!guehdr)) - goto out; - } + guehdr = skb_gro_header(skb, len, off); + if (unlikely(!guehdr)) + goto out; switch (guehdr->version) { case 0: @@ -429,7 +420,7 @@ next_proto: /* We can clear the encap_mark for GUE as we are essentially doing * one of two possible things. We are either adding an L4 tunnel - * header to the outer L3 tunnel header, or we are are simply + * header to the outer L3 tunnel header, or we are simply * treating the GRE tunnel header as though it is a UDP protocol * specific header such as VXLAN or GENEVE. */ @@ -438,17 +429,14 @@ next_proto: /* Flag this frame as already having an outer encap header */ NAPI_GRO_CB(skb)->is_fou = 1; - rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]); if (WARN_ON_ONCE(!ops || !ops->callbacks.gro_receive)) - goto out_unlock; + goto out; pp = call_gro_receive(ops->callbacks.gro_receive, head, skb); flush = 0; -out_unlock: - rcu_read_unlock(); out: skb_gro_flush_final_remcsum(skb, pp, flush, &grc); @@ -457,8 +445,8 @@ out: static int gue_gro_complete(struct sock *sk, struct sk_buff *skb, int nhoff) { - const struct net_offload **offloads; struct guehdr *guehdr = (struct guehdr *)(skb->data + nhoff); + const struct net_offload __rcu **offloads; const struct net_offload *ops; unsigned int guehlen = 0; u8 proto; @@ -485,18 +473,16 @@ static int gue_gro_complete(struct sock *sk, struct sk_buff *skb, int nhoff) return err; } - rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]); if (WARN_ON(!ops || !ops->callbacks.gro_complete)) - goto out_unlock; + goto out; err = ops->callbacks.gro_complete(skb, nhoff + guehlen); skb_set_inner_mac_header(skb, nhoff + guehlen); -out_unlock: - rcu_read_unlock(); +out: return err; } @@ -911,7 +897,7 @@ static int fou_nl_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } -static const struct genl_ops fou_nl_ops[] = { +static const struct genl_small_ops fou_nl_ops[] = { { .cmd = FOU_CMD_ADD, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, @@ -940,8 +926,9 @@ static struct genl_family fou_nl_family __ro_after_init = { .policy = fou_nl_policy, .netnsok = true, .module = THIS_MODULE, - .ops = fou_nl_ops, - .n_ops = ARRAY_SIZE(fou_nl_ops), + .small_ops = fou_nl_ops, + .n_small_ops = ARRAY_SIZE(fou_nl_ops), + .resv_start_op = FOU_CMD_GET + 1, }; size_t fou_encap_hlen(struct ip_tunnel_encap *e) @@ -1304,3 +1291,4 @@ module_init(fou_init); module_exit(fou_fini); MODULE_AUTHOR("Tom Herbert <therbert@google.com>"); MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Foo over UDP"); diff --git a/net/ipv4/gre_demux.c b/net/ipv4/gre_demux.c index 66fdbfe5447c..cbb2b4bb0dfa 100644 --- a/net/ipv4/gre_demux.c +++ b/net/ipv4/gre_demux.c @@ -128,7 +128,7 @@ int gre_parse_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, * to 0 and sets the configured key in the * inner erspan header field */ - if (greh->protocol == htons(ETH_P_ERSPAN) || + if ((greh->protocol == htons(ETH_P_ERSPAN) && hdr_len != 4) || greh->protocol == htons(ETH_P_ERSPAN2)) { struct erspan_base_hdr *ershdr; @@ -195,7 +195,6 @@ static int gre_err(struct sk_buff *skb, u32 info) static const struct net_protocol net_gre_protocol = { .handler = gre_rcv, .err_handler = gre_err, - .netns_ok = 1, }; static int __init gre_init(void) diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c index 2e6d1b7a7bc9..2b9cb5398335 100644 --- a/net/ipv4/gre_offload.c +++ b/net/ipv4/gre_offload.c @@ -10,17 +10,18 @@ #include <linux/init.h> #include <net/protocol.h> #include <net/gre.h> +#include <net/gro.h> static struct sk_buff *gre_gso_segment(struct sk_buff *skb, netdev_features_t features) { int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb); + bool need_csum, offload_csum, gso_partial, need_ipsec; struct sk_buff *segs = ERR_PTR(-EINVAL); u16 mac_offset = skb->mac_header; __be16 protocol = skb->protocol; u16 mac_len = skb->mac_len; int gre_offset, outer_hlen; - bool need_csum, gso_partial; if (!skb->encapsulation) goto out; @@ -44,6 +45,13 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, skb->encap_hdr_csum = need_csum; features &= skb->dev->hw_enc_features; + if (need_csum) + features &= ~NETIF_F_SCTP_CRC; + + need_ipsec = skb_dst(skb) && dst_xfrm(skb_dst(skb)); + /* Try to offload checksum if possible */ + offload_csum = !!(need_csum && !need_ipsec && + (skb->dev->features & NETIF_F_HW_CSUM)); /* segment inner packet. */ segs = skb_mac_gso_segment(skb, features); @@ -98,7 +106,13 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, } *(pcsum + 1) = 0; - *pcsum = gso_make_checksum(skb, 0); + if (skb->encapsulation || !offload_csum) { + *pcsum = gso_make_checksum(skb, 0); + } else { + skb->ip_summed = CHECKSUM_PARTIAL; + skb->csum_start = skb_transport_header(skb) - skb->head; + skb->csum_offset = sizeof(*greh); + } } while ((skb = skb->next)); out: return segs; @@ -123,12 +137,9 @@ static struct sk_buff *gre_gro_receive(struct list_head *head, off = skb_gro_offset(skb); hlen = off + sizeof(*greh); - greh = skb_gro_header_fast(skb, off); - if (skb_gro_header_hard(skb, hlen)) { - greh = skb_gro_header_slow(skb, hlen, off); - if (unlikely(!greh)) - goto out; - } + greh = skb_gro_header(skb, hlen, off); + if (unlikely(!greh)) + goto out; /* Only support version 0 and K (key), C (csum) flags. Note that * although the support for the S (seq#) flag can be added easily @@ -149,10 +160,9 @@ static struct sk_buff *gre_gro_receive(struct list_head *head, type = greh->protocol; - rcu_read_lock(); ptype = gro_find_receive_by_type(type); if (!ptype) - goto out_unlock; + goto out; grehlen = GRE_HEADER_SECTION; @@ -166,13 +176,13 @@ static struct sk_buff *gre_gro_receive(struct list_head *head, if (skb_gro_header_hard(skb, hlen)) { greh = skb_gro_header_slow(skb, hlen, off); if (unlikely(!greh)) - goto out_unlock; + goto out; } /* Don't bother verifying checksum if we're going to flush anyway. */ if ((greh->flags & GRE_CSUM) && !NAPI_GRO_CB(skb)->flush) { if (skb_gro_checksum_simple_validate(skb)) - goto out_unlock; + goto out; skb_gro_checksum_try_convert(skb, IPPROTO_GRE, null_compute_pseudo); @@ -216,8 +226,6 @@ static struct sk_buff *gre_gro_receive(struct list_head *head, pp = call_gro_receive(ptype->callbacks.gro_receive, head, skb); flush = 0; -out_unlock: - rcu_read_unlock(); out: skb_gro_flush_final(skb, pp, flush); @@ -242,13 +250,10 @@ static int gre_gro_complete(struct sk_buff *skb, int nhoff) if (greh->flags & GRE_CSUM) grehlen += GRE_HEADER_SECTION; - rcu_read_lock(); ptype = gro_find_complete_by_type(type); if (ptype) err = ptype->callbacks.gro_complete(skb, nhoff + grehlen); - rcu_read_unlock(); - skb_set_inner_mac_header(skb, nhoff + grehlen); return err; diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index f369e7ce685b..d5d745c3e345 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -186,30 +186,20 @@ EXPORT_SYMBOL(icmp_err_convert); */ struct icmp_control { - bool (*handler)(struct sk_buff *skb); + enum skb_drop_reason (*handler)(struct sk_buff *skb); short error; /* This ICMP is classed as an error message */ }; static const struct icmp_control icmp_pointers[NR_ICMP_TYPES+1]; -/* - * The ICMP socket(s). This is the most convenient way to flow control - * our ICMP output as well as maintain a clean interface throughout - * all layers. All Socketless IP sends will soon be gone. - * - * On SMP we have one ICMP socket per-cpu. - */ -static struct sock *icmp_sk(struct net *net) -{ - return this_cpu_read(*net->ipv4.icmp_sk); -} +static DEFINE_PER_CPU(struct sock *, ipv4_icmp_sk); /* Called with BH disabled */ static inline struct sock *icmp_xmit_lock(struct net *net) { struct sock *sk; - sk = icmp_sk(net); + sk = this_cpu_read(ipv4_icmp_sk); if (unlikely(!spin_trylock(&sk->sk_lock.slock))) { /* This can happen if the output path signals a @@ -217,11 +207,13 @@ static inline struct sock *icmp_xmit_lock(struct net *net) */ return NULL; } + sock_net_set(sk, net); return sk; } static inline void icmp_xmit_unlock(struct sock *sk) { + sock_net_set(sk, &init_net); spin_unlock(&sk->sk_lock.slock); } @@ -239,7 +231,7 @@ static struct { /** * icmp_global_allow - Are we allowed to send one more ICMP message ? * - * Uses a token bucket to limit our ICMP messages to sysctl_icmp_msgs_per_sec. + * Uses a token bucket to limit our ICMP messages to ~sysctl_icmp_msgs_per_sec. * Returns false if we reached the limit and can not send another packet. * Note: called with BH disabled */ @@ -261,13 +253,17 @@ bool icmp_global_allow(void) spin_lock(&icmp_global.lock); delta = min_t(u32, now - icmp_global.stamp, HZ); if (delta >= HZ / 50) { - incr = sysctl_icmp_msgs_per_sec * delta / HZ ; + incr = READ_ONCE(sysctl_icmp_msgs_per_sec) * delta / HZ; if (incr) WRITE_ONCE(icmp_global.stamp, now); } - credit = min_t(u32, icmp_global.credit + incr, sysctl_icmp_msgs_burst); + credit = min_t(u32, icmp_global.credit + incr, + READ_ONCE(sysctl_icmp_msgs_burst)); if (credit) { - credit--; + /* We want to use a credit of one in average, but need to randomize + * it for security reasons. + */ + credit = max_t(int, credit - prandom_u32_max(3), 0); rc = true; } WRITE_ONCE(icmp_global.credit, credit); @@ -286,7 +282,7 @@ static bool icmpv4_mask_allow(struct net *net, int type, int code) return true; /* Limit if icmp type is enabled in ratemask. */ - if (!((1 << type) & net->ipv4.sysctl_icmp_ratemask)) + if (!((1 << type) & READ_ONCE(net->ipv4.sysctl_icmp_ratemask))) return true; return false; @@ -324,7 +320,8 @@ static bool icmpv4_xrlim_allow(struct net *net, struct rtable *rt, vif = l3mdev_master_ifindex(dst->dev); peer = inet_getpeer_v4(net->ipv4.peers, fl4->daddr, vif, 1); - rc = inet_peer_xrlim_allow(peer, net->ipv4.sysctl_icmp_ratelimit); + rc = inet_peer_xrlim_allow(peer, + READ_ONCE(net->ipv4.sysctl_icmp_ratelimit)); if (peer) inet_putpeer(peer); out: @@ -347,12 +344,12 @@ void icmp_out_count(struct net *net, unsigned char type) static int icmp_glue_bits(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb) { - struct icmp_bxm *icmp_param = (struct icmp_bxm *)from; + struct icmp_bxm *icmp_param = from; __wsum csum; csum = skb_copy_and_csum_bits(icmp_param->skb, icmp_param->offset + offset, - to, len, 0); + to, len); skb->csum = csum_block_add(skb->csum, csum, odd); if (icmp_pointers[icmp_param->data.icmph.type].error) @@ -360,14 +357,13 @@ static int icmp_glue_bits(void *from, char *to, int offset, int len, int odd, return 0; } -static void icmp_push_reply(struct icmp_bxm *icmp_param, +static void icmp_push_reply(struct sock *sk, + struct icmp_bxm *icmp_param, struct flowi4 *fl4, struct ipcm_cookie *ipc, struct rtable **rt) { - struct sock *sk; struct sk_buff *skb; - sk = icmp_sk(dev_net((*rt)->dst.dev)); if (ip_append_data(sk, fl4, icmp_glue_bits, icmp_param, icmp_param->data_len+icmp_param->head_len, icmp_param->head_len, @@ -376,15 +372,15 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param, ip_flush_pending_frames(sk); } else if ((skb = skb_peek(&sk->sk_write_queue)) != NULL) { struct icmphdr *icmph = icmp_hdr(skb); - __wsum csum = 0; + __wsum csum; struct sk_buff *skb1; + csum = csum_partial_copy_nocheck((void *)&icmp_param->data, + (char *)icmph, + icmp_param->head_len); skb_queue_walk(&sk->sk_write_queue, skb1) { csum = csum_add(csum, skb1->csum); } - csum = csum_partial_copy_nocheck((void *)&icmp_param->data, - (char *)icmph, - icmp_param->head_len, csum); icmph->checksum = csum_fold(csum); skb->ip_summed = CHECKSUM_NONE; ip_push_pending_frames(sk, fl4); @@ -427,7 +423,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) ipcm_init(&ipc); inet->tos = ip_hdr(skb)->tos; - sk->sk_mark = mark; + ipc.sockc.mark = mark; daddr = ipc.addr = ip_hdr(skb)->saddr; saddr = fib_compute_spec_dst(skb); @@ -444,12 +440,12 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos); fl4.flowi4_proto = IPPROTO_ICMP; fl4.flowi4_oif = l3mdev_master_ifindex(skb->dev); - security_skb_classify_flow(skb, flowi4_to_flowi(&fl4)); + security_skb_classify_flow(skb, flowi4_to_flowi_common(&fl4)); rt = ip_route_output_key(net, &fl4); if (IS_ERR(rt)) goto out_unlock; if (icmpv4_xrlim_allow(net, rt, &fl4, type, code)) - icmp_push_reply(icmp_param, &fl4, &ipc, &rt); + icmp_push_reply(sk, icmp_param, &fl4, &ipc, &rt); ip_rt_put(rt); out_unlock: icmp_xmit_unlock(sk); @@ -457,6 +453,23 @@ out_bh_enable: local_bh_enable(); } +/* + * The device used for looking up which routing table to use for sending an ICMP + * error is preferably the source whenever it is set, which should ensure the + * icmp error can be sent to the source host, else lookup using the routing + * table of the destination device, else use the main routing table (index 0). + */ +static struct net_device *icmp_get_route_lookup_dev(struct sk_buff *skb) +{ + struct net_device *route_lookup_dev = NULL; + + if (skb->dev) + route_lookup_dev = skb->dev; + else if (skb_dst(skb)) + route_lookup_dev = skb_dst(skb)->dev; + return route_lookup_dev; +} + static struct rtable *icmp_route_lookup(struct net *net, struct flowi4 *fl4, struct sk_buff *skb_in, @@ -465,6 +478,7 @@ static struct rtable *icmp_route_lookup(struct net *net, int type, int code, struct icmp_bxm *param) { + struct net_device *route_lookup_dev; struct rtable *rt, *rt2; struct flowi4 fl4_dec; int err; @@ -479,9 +493,10 @@ static struct rtable *icmp_route_lookup(struct net *net, fl4->flowi4_proto = IPPROTO_ICMP; fl4->fl4_icmp_type = type; fl4->fl4_icmp_code = code; - fl4->flowi4_oif = l3mdev_master_ifindex(skb_dst(skb_in)->dev); + route_lookup_dev = icmp_get_route_lookup_dev(skb_in); + fl4->flowi4_oif = l3mdev_master_ifindex(route_lookup_dev); - security_skb_classify_flow(skb_in, flowi4_to_flowi(fl4)); + security_skb_classify_flow(skb_in, flowi4_to_flowi_common(fl4)); rt = ip_route_output_key_hash(net, fl4, skb_in); if (IS_ERR(rt)) return rt; @@ -503,7 +518,7 @@ static struct rtable *icmp_route_lookup(struct net *net, if (err) goto relookup_failed; - if (inet_addr_type_dev_table(net, skb_dst(skb_in)->dev, + if (inet_addr_type_dev_table(net, route_lookup_dev, fl4_dec.saddr) == RTN_LOCAL) { rt2 = __ip_route_output_key(net, &fl4_dec); if (IS_ERR(rt2)) @@ -679,7 +694,7 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, rcu_read_lock(); if (rt_is_input_route(rt) && - net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr) + READ_ONCE(net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr)) dev = dev_get_by_index_rcu(net, inet_iif(skb_in)); if (dev) @@ -690,9 +705,9 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, rcu_read_unlock(); } - tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) | + tos = icmp_pointers[type].error ? (RT_TOS(iph->tos) | IPTOS_PREC_INTERNETCONTROL) : - iph->tos; + iph->tos; mark = IP4_REPLY_MARK(net, skb_in->mark); if (__ip_options_echo(net, &icmp_param.replyopts.opt.opt, skb_in, opt)) @@ -710,10 +725,10 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, icmp_param.skb = skb_in; icmp_param.offset = skb_network_offset(skb_in); inet_sk(sk)->tos = tos; - sk->sk_mark = mark; ipcm_init(&ipc); ipc.addr = iph->saddr; ipc.opt = &icmp_param.replyopts.opt; + ipc.sockc.mark = mark; rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos, mark, type, code, &icmp_param); @@ -737,7 +752,14 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, icmp_param.data_len = room; icmp_param.head_len = sizeof(struct icmphdr); - icmp_push_reply(&icmp_param, &fl4, &ipc, &rt); + /* if we don't have a source address at this point, fall back to the + * dummy address instead of sending out a packet with a source address + * of 0.0.0.0 + */ + if (!fl4.saddr) + fl4.saddr = htonl(INADDR_DUMMY); + + icmp_push_reply(sk, &icmp_param, &fl4, &ipc, &rt); ende: ip_rt_put(rt); out_unlock: @@ -753,13 +775,14 @@ EXPORT_SYMBOL(__icmp_send); void icmp_ndo_send(struct sk_buff *skb_in, int type, int code, __be32 info) { struct sk_buff *cloned_skb = NULL; + struct ip_options opts = { 0 }; enum ip_conntrack_info ctinfo; struct nf_conn *ct; __be32 orig_ip; ct = nf_ct_get(skb_in, &ctinfo); if (!ct || !(ct->status & IPS_SRC_NAT)) { - icmp_send(skb_in, type, code, info); + __icmp_send(skb_in, type, code, info, &opts); return; } @@ -774,7 +797,7 @@ void icmp_ndo_send(struct sk_buff *skb_in, int type, int code, __be32 info) orig_ip = ip_hdr(skb_in)->saddr; ip_hdr(skb_in)->saddr = ct->tuplehash[0].tuple.src.u3.ip; - icmp_send(skb_in, type, code, info); + __icmp_send(skb_in, type, code, info, &opts); ip_hdr(skb_in)->saddr = orig_ip; out: consume_skb(cloned_skb); @@ -784,7 +807,7 @@ EXPORT_SYMBOL(icmp_ndo_send); static void icmp_socket_deliver(struct sk_buff *skb, u32 info) { - const struct iphdr *iph = (const struct iphdr *) skb->data; + const struct iphdr *iph = (const struct iphdr *)skb->data; const struct net_protocol *ipprot; int protocol = iph->protocol; @@ -818,8 +841,9 @@ static bool icmp_tag_validation(int proto) * ICMP_PARAMETERPROB. */ -static bool icmp_unreach(struct sk_buff *skb) +static enum skb_drop_reason icmp_unreach(struct sk_buff *skb) { + enum skb_drop_reason reason = SKB_NOT_DROPPED_YET; const struct iphdr *iph; struct icmphdr *icmph; struct net *net; @@ -839,8 +863,10 @@ static bool icmp_unreach(struct sk_buff *skb) icmph = icmp_hdr(skb); iph = (const struct iphdr *)skb->data; - if (iph->ihl < 5) /* Mangled header, drop. */ + if (iph->ihl < 5) { /* Mangled header, drop. */ + reason = SKB_DROP_REASON_IP_INHDR; goto out_err; + } switch (icmph->type) { case ICMP_DEST_UNREACH: @@ -853,9 +879,9 @@ static bool icmp_unreach(struct sk_buff *skb) case ICMP_FRAG_NEEDED: /* for documentation of the ip_no_pmtu_disc * values please see - * Documentation/networking/ip-sysctl.txt + * Documentation/networking/ip-sysctl.rst */ - switch (net->ipv4.sysctl_ip_no_pmtu_disc) { + switch (READ_ONCE(net->ipv4.sysctl_ip_no_pmtu_disc)) { default: net_dbg_ratelimited("%pI4: fragmentation needed and DF set\n", &iph->daddr); @@ -865,7 +891,7 @@ static bool icmp_unreach(struct sk_buff *skb) case 3: if (!icmp_tag_validation(iph->protocol)) goto out; - /* fall through */ + fallthrough; case 0: info = ntohs(icmph->un.frag.mtu); } @@ -908,7 +934,7 @@ static bool icmp_unreach(struct sk_buff *skb) * get the other vendor to fix their kit. */ - if (!net->ipv4.sysctl_icmp_ignore_bogus_error_responses && + if (!READ_ONCE(net->ipv4.sysctl_icmp_ignore_bogus_error_responses) && inet_addr_type_dev_table(net, skb->dev, iph->daddr) == RTN_BROADCAST) { net_warn_ratelimited("%pI4 sent an invalid ICMP type %u, code %u error to a broadcast: %pI4 on %s\n", &ip_hdr(skb)->saddr, @@ -920,10 +946,10 @@ static bool icmp_unreach(struct sk_buff *skb) icmp_socket_deliver(skb, info); out: - return true; + return reason; out_err: __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); - return false; + return reason ?: SKB_DROP_REASON_NOT_SPECIFIED; } @@ -931,24 +957,24 @@ out_err: * Handle ICMP_REDIRECT. */ -static bool icmp_redirect(struct sk_buff *skb) +static enum skb_drop_reason icmp_redirect(struct sk_buff *skb) { if (skb->len < sizeof(struct iphdr)) { __ICMP_INC_STATS(dev_net(skb->dev), ICMP_MIB_INERRORS); - return false; + return SKB_DROP_REASON_PKT_TOO_SMALL; } if (!pskb_may_pull(skb, sizeof(struct iphdr))) { /* there aught to be a stat */ - return false; + return SKB_DROP_REASON_NOMEM; } icmp_socket_deliver(skb, ntohl(icmp_hdr(skb)->un.gateway)); - return true; + return SKB_NOT_DROPPED_YET; } /* - * Handle ICMP_ECHO ("ping") requests. + * Handle ICMP_ECHO ("ping") and ICMP_EXT_ECHO ("PROBE") requests. * * RFC 1122: 3.2.2.6 MUST have an echo server that answers ICMP echo * requests. @@ -956,28 +982,143 @@ static bool icmp_redirect(struct sk_buff *skb) * included in the reply. * RFC 1812: 4.3.3.6 SHOULD have a config option for silently ignoring * echo requests, MUST have default=NOT. + * RFC 8335: 8 MUST have a config option to enable/disable ICMP + * Extended Echo Functionality, MUST be disabled by default * See also WRT handling of options once they are done and working. */ -static bool icmp_echo(struct sk_buff *skb) +static enum skb_drop_reason icmp_echo(struct sk_buff *skb) { + struct icmp_bxm icmp_param; struct net *net; net = dev_net(skb_dst(skb)->dev); - if (!net->ipv4.sysctl_icmp_echo_ignore_all) { - struct icmp_bxm icmp_param; + /* should there be an ICMP stat for ignored echos? */ + if (READ_ONCE(net->ipv4.sysctl_icmp_echo_ignore_all)) + return SKB_NOT_DROPPED_YET; + + icmp_param.data.icmph = *icmp_hdr(skb); + icmp_param.skb = skb; + icmp_param.offset = 0; + icmp_param.data_len = skb->len; + icmp_param.head_len = sizeof(struct icmphdr); - icmp_param.data.icmph = *icmp_hdr(skb); + if (icmp_param.data.icmph.type == ICMP_ECHO) icmp_param.data.icmph.type = ICMP_ECHOREPLY; - icmp_param.skb = skb; - icmp_param.offset = 0; - icmp_param.data_len = skb->len; - icmp_param.head_len = sizeof(struct icmphdr); - icmp_reply(&icmp_param, skb); + else if (!icmp_build_probe(skb, &icmp_param.data.icmph)) + return SKB_NOT_DROPPED_YET; + + icmp_reply(&icmp_param, skb); + return SKB_NOT_DROPPED_YET; +} + +/* Helper for icmp_echo and icmpv6_echo_reply. + * Searches for net_device that matches PROBE interface identifier + * and builds PROBE reply message in icmphdr. + * + * Returns false if PROBE responses are disabled via sysctl + */ + +bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr) +{ + struct icmp_ext_hdr *ext_hdr, _ext_hdr; + struct icmp_ext_echo_iio *iio, _iio; + struct net *net = dev_net(skb->dev); + struct net_device *dev; + char buff[IFNAMSIZ]; + u16 ident_len; + u8 status; + + if (!READ_ONCE(net->ipv4.sysctl_icmp_echo_enable_probe)) + return false; + + /* We currently only support probing interfaces on the proxy node + * Check to ensure L-bit is set + */ + if (!(ntohs(icmphdr->un.echo.sequence) & 1)) + return false; + /* Clear status bits in reply message */ + icmphdr->un.echo.sequence &= htons(0xFF00); + if (icmphdr->type == ICMP_EXT_ECHO) + icmphdr->type = ICMP_EXT_ECHOREPLY; + else + icmphdr->type = ICMPV6_EXT_ECHO_REPLY; + ext_hdr = skb_header_pointer(skb, 0, sizeof(_ext_hdr), &_ext_hdr); + /* Size of iio is class_type dependent. + * Only check header here and assign length based on ctype in the switch statement + */ + iio = skb_header_pointer(skb, sizeof(_ext_hdr), sizeof(iio->extobj_hdr), &_iio); + if (!ext_hdr || !iio) + goto send_mal_query; + if (ntohs(iio->extobj_hdr.length) <= sizeof(iio->extobj_hdr) || + ntohs(iio->extobj_hdr.length) > sizeof(_iio)) + goto send_mal_query; + ident_len = ntohs(iio->extobj_hdr.length) - sizeof(iio->extobj_hdr); + iio = skb_header_pointer(skb, sizeof(_ext_hdr), + sizeof(iio->extobj_hdr) + ident_len, &_iio); + if (!iio) + goto send_mal_query; + + status = 0; + dev = NULL; + switch (iio->extobj_hdr.class_type) { + case ICMP_EXT_ECHO_CTYPE_NAME: + if (ident_len >= IFNAMSIZ) + goto send_mal_query; + memset(buff, 0, sizeof(buff)); + memcpy(buff, &iio->ident.name, ident_len); + dev = dev_get_by_name(net, buff); + break; + case ICMP_EXT_ECHO_CTYPE_INDEX: + if (ident_len != sizeof(iio->ident.ifindex)) + goto send_mal_query; + dev = dev_get_by_index(net, ntohl(iio->ident.ifindex)); + break; + case ICMP_EXT_ECHO_CTYPE_ADDR: + if (ident_len < sizeof(iio->ident.addr.ctype3_hdr) || + ident_len != sizeof(iio->ident.addr.ctype3_hdr) + + iio->ident.addr.ctype3_hdr.addrlen) + goto send_mal_query; + switch (ntohs(iio->ident.addr.ctype3_hdr.afi)) { + case ICMP_AFI_IP: + if (iio->ident.addr.ctype3_hdr.addrlen != sizeof(struct in_addr)) + goto send_mal_query; + dev = ip_dev_find(net, iio->ident.addr.ip_addr.ipv4_addr); + break; +#if IS_ENABLED(CONFIG_IPV6) + case ICMP_AFI_IP6: + if (iio->ident.addr.ctype3_hdr.addrlen != sizeof(struct in6_addr)) + goto send_mal_query; + dev = ipv6_stub->ipv6_dev_find(net, &iio->ident.addr.ip_addr.ipv6_addr, dev); + dev_hold(dev); + break; +#endif + default: + goto send_mal_query; + } + break; + default: + goto send_mal_query; } - /* should there be an ICMP stat for ignored echos? */ + if (!dev) { + icmphdr->code = ICMP_EXT_CODE_NO_IF; + return true; + } + /* Fill bits in reply message */ + if (dev->flags & IFF_UP) + status |= ICMP_EXT_ECHOREPLY_ACTIVE; + if (__in_dev_get_rcu(dev) && __in_dev_get_rcu(dev)->ifa_list) + status |= ICMP_EXT_ECHOREPLY_IPV4; + if (!list_empty(&rcu_dereference(dev->ip6_ptr)->addr_list)) + status |= ICMP_EXT_ECHOREPLY_IPV6; + dev_put(dev); + icmphdr->un.echo.sequence |= htons(status); + return true; +send_mal_query: + icmphdr->code = ICMP_EXT_CODE_MAL_QUERY; return true; } +EXPORT_SYMBOL_GPL(icmp_build_probe); /* * Handle ICMP Timestamp requests. @@ -986,7 +1127,7 @@ static bool icmp_echo(struct sk_buff *skb) * MUST be accurate to a few minutes. * MUST be updated at least at 15Hz. */ -static bool icmp_timestamp(struct sk_buff *skb) +static enum skb_drop_reason icmp_timestamp(struct sk_buff *skb) { struct icmp_bxm icmp_param; /* @@ -1011,17 +1152,17 @@ static bool icmp_timestamp(struct sk_buff *skb) icmp_param.data_len = 0; icmp_param.head_len = sizeof(struct icmphdr) + 12; icmp_reply(&icmp_param, skb); - return true; + return SKB_NOT_DROPPED_YET; out_err: __ICMP_INC_STATS(dev_net(skb_dst(skb)->dev), ICMP_MIB_INERRORS); - return false; + return SKB_DROP_REASON_PKT_TOO_SMALL; } -static bool icmp_discard(struct sk_buff *skb) +static enum skb_drop_reason icmp_discard(struct sk_buff *skb) { /* pretend it was a success */ - return true; + return SKB_NOT_DROPPED_YET; } /* @@ -1029,18 +1170,20 @@ static bool icmp_discard(struct sk_buff *skb) */ int icmp_rcv(struct sk_buff *skb) { - struct icmphdr *icmph; + enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED; struct rtable *rt = skb_rtable(skb); struct net *net = dev_net(rt->dst.dev); - bool success; + struct icmphdr *icmph; if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { struct sec_path *sp = skb_sec_path(skb); int nh; if (!(sp && sp->xvec[sp->len - 1]->props.flags & - XFRM_STATE_ICMP)) + XFRM_STATE_ICMP)) { + reason = SKB_DROP_REASON_XFRM_POLICY; goto drop; + } if (!pskb_may_pull(skb, sizeof(*icmph) + sizeof(struct iphdr))) goto drop; @@ -1048,8 +1191,11 @@ int icmp_rcv(struct sk_buff *skb) nh = skb_network_offset(skb); skb_set_network_header(skb, sizeof(*icmph)); - if (!xfrm4_policy_check_reverse(NULL, XFRM_POLICY_IN, skb)) + if (!xfrm4_policy_check_reverse(NULL, XFRM_POLICY_IN, + skb)) { + reason = SKB_DROP_REASON_XFRM_POLICY; goto drop; + } skb_set_network_header(skb, nh); } @@ -1065,15 +1211,31 @@ int icmp_rcv(struct sk_buff *skb) icmph = icmp_hdr(skb); ICMPMSGIN_INC_STATS(net, icmph->type); + + /* Check for ICMP Extended Echo (PROBE) messages */ + if (icmph->type == ICMP_EXT_ECHO) { + /* We can't use icmp_pointers[].handler() because it is an array of + * size NR_ICMP_TYPES + 1 (19 elements) and PROBE has code 42. + */ + reason = icmp_echo(skb); + goto reason_check; + } + + if (icmph->type == ICMP_EXT_ECHOREPLY) { + reason = ping_rcv(skb); + goto reason_check; + } + /* * 18 is the highest 'known' ICMP type. Anything else is a mystery * * RFC 1122: 3.2.2 Unknown ICMP messages types MUST be silently * discarded. */ - if (icmph->type > NR_ICMP_TYPES) + if (icmph->type > NR_ICMP_TYPES) { + reason = SKB_DROP_REASON_UNHANDLED_PROTO; goto error; - + } /* * Parse the ICMP message @@ -1088,34 +1250,96 @@ int icmp_rcv(struct sk_buff *skb) */ if ((icmph->type == ICMP_ECHO || icmph->type == ICMP_TIMESTAMP) && - net->ipv4.sysctl_icmp_echo_ignore_broadcasts) { + READ_ONCE(net->ipv4.sysctl_icmp_echo_ignore_broadcasts)) { + reason = SKB_DROP_REASON_INVALID_PROTO; goto error; } if (icmph->type != ICMP_ECHO && icmph->type != ICMP_TIMESTAMP && icmph->type != ICMP_ADDRESS && icmph->type != ICMP_ADDRESSREPLY) { + reason = SKB_DROP_REASON_INVALID_PROTO; goto error; } } - success = icmp_pointers[icmph->type].handler(skb); - - if (success) { + reason = icmp_pointers[icmph->type].handler(skb); +reason_check: + if (!reason) { consume_skb(skb); return NET_RX_SUCCESS; } drop: - kfree_skb(skb); + kfree_skb_reason(skb, reason); return NET_RX_DROP; csum_error: + reason = SKB_DROP_REASON_ICMP_CSUM; __ICMP_INC_STATS(net, ICMP_MIB_CSUMERRORS); error: __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); goto drop; } +static bool ip_icmp_error_rfc4884_validate(const struct sk_buff *skb, int off) +{ + struct icmp_extobj_hdr *objh, _objh; + struct icmp_ext_hdr *exth, _exth; + u16 olen; + + exth = skb_header_pointer(skb, off, sizeof(_exth), &_exth); + if (!exth) + return false; + if (exth->version != 2) + return true; + + if (exth->checksum && + csum_fold(skb_checksum(skb, off, skb->len - off, 0))) + return false; + + off += sizeof(_exth); + while (off < skb->len) { + objh = skb_header_pointer(skb, off, sizeof(_objh), &_objh); + if (!objh) + return false; + + olen = ntohs(objh->length); + if (olen < sizeof(_objh)) + return false; + + off += olen; + if (off > skb->len) + return false; + } + + return true; +} + +void ip_icmp_error_rfc4884(const struct sk_buff *skb, + struct sock_ee_data_rfc4884 *out, + int thlen, int off) +{ + int hlen; + + /* original datagram headers: end of icmph to payload (skb->data) */ + hlen = -skb_transport_offset(skb) - thlen; + + /* per rfc 4884: minimal datagram length of 128 bytes */ + if (off < 128 || off < hlen) + return; + + /* kernel has stripped headers: return payload offset in bytes */ + off -= hlen; + if (off + sizeof(struct icmp_ext_hdr) > skb->len) + return; + + out->len = off; + + if (!ip_icmp_error_rfc4884_validate(skb, off)) + out->flags |= SO_EE_RFC4884_FLAG_INVALID; +} +EXPORT_SYMBOL_GPL(ip_icmp_error_rfc4884); + int icmp_err(struct sk_buff *skb, u32 info) { struct iphdr *iph = (struct iphdr *)skb->data; @@ -1216,48 +1440,11 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = { }, }; -static void __net_exit icmp_sk_exit(struct net *net) -{ - int i; - - for_each_possible_cpu(i) - inet_ctl_sock_destroy(*per_cpu_ptr(net->ipv4.icmp_sk, i)); - free_percpu(net->ipv4.icmp_sk); - net->ipv4.icmp_sk = NULL; -} - static int __net_init icmp_sk_init(struct net *net) { - int i, err; - - net->ipv4.icmp_sk = alloc_percpu(struct sock *); - if (!net->ipv4.icmp_sk) - return -ENOMEM; - - for_each_possible_cpu(i) { - struct sock *sk; - - err = inet_ctl_sock_create(&sk, PF_INET, - SOCK_RAW, IPPROTO_ICMP, net); - if (err < 0) - goto fail; - - *per_cpu_ptr(net->ipv4.icmp_sk, i) = sk; - - /* Enough space for 2 64K ICMP packets, including - * sk_buff/skb_shared_info struct overhead. - */ - sk->sk_sndbuf = 2 * SKB_TRUESIZE(64 * 1024); - - /* - * Speedup sock_wfree() - */ - sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); - inet_sk(sk)->pmtudisc = IP_PMTUDISC_DONT; - } - /* Control parameters for ECHO replies. */ net->ipv4.sysctl_icmp_echo_ignore_all = 0; + net->ipv4.sysctl_icmp_echo_enable_probe = 0; net->ipv4.sysctl_icmp_echo_ignore_broadcasts = 1; /* Control parameter - ignore bogus broadcast responses? */ @@ -1280,18 +1467,36 @@ static int __net_init icmp_sk_init(struct net *net) net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr = 0; return 0; - -fail: - icmp_sk_exit(net); - return err; } static struct pernet_operations __net_initdata icmp_sk_ops = { .init = icmp_sk_init, - .exit = icmp_sk_exit, }; int __init icmp_init(void) { + int err, i; + + for_each_possible_cpu(i) { + struct sock *sk; + + err = inet_ctl_sock_create(&sk, PF_INET, + SOCK_RAW, IPPROTO_ICMP, &init_net); + if (err < 0) + return err; + + per_cpu(ipv4_icmp_sk, i) = sk; + + /* Enough space for 2 64K ICMP packets, including + * sk_buff/skb_shared_info struct overhead. + */ + sk->sk_sndbuf = 2 * SKB_TRUESIZE(64 * 1024); + + /* + * Speedup sock_wfree() + */ + sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); + inet_sk(sk)->pmtudisc = IP_PMTUDISC_DONT; + } return register_pernet_subsys(&icmp_sk_ops); } diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 3b9c7a2725a9..81be3e0f0e70 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -107,8 +107,6 @@ #ifdef CONFIG_IP_MULTICAST /* Parameter names and values are taken from igmp-v2-06 draft */ -#define IGMP_V2_UNSOLICITED_REPORT_INTERVAL (10*HZ) -#define IGMP_V3_UNSOLICITED_REPORT_INTERVAL (1*HZ) #define IGMP_QUERY_INTERVAL (125*HZ) #define IGMP_QUERY_RESPONSE_INTERVAL (10*HZ) @@ -215,7 +213,7 @@ static void igmp_stop_timer(struct ip_mc_list *im) /* It must be called with locked im->lock */ static void igmp_start_timer(struct ip_mc_list *im, int max_delay) { - int tv = prandom_u32() % max_delay; + int tv = prandom_u32_max(max_delay); im->tm_running = 1; if (!mod_timer(&im->timer, jiffies+tv+2)) @@ -224,7 +222,7 @@ static void igmp_start_timer(struct ip_mc_list *im, int max_delay) static void igmp_gq_start_timer(struct in_device *in_dev) { - int tv = prandom_u32() % in_dev->mr_maxdelay; + int tv = prandom_u32_max(in_dev->mr_maxdelay); unsigned long exp = jiffies + tv + 2; if (in_dev->mr_gq_running && @@ -238,7 +236,7 @@ static void igmp_gq_start_timer(struct in_device *in_dev) static void igmp_ifc_start_timer(struct in_device *in_dev, int delay) { - int tv = prandom_u32() % delay; + int tv = prandom_u32_max(delay); if (!mod_timer(&in_dev->mr_ifc_timer, jiffies+tv+2)) in_dev_hold(in_dev); @@ -469,7 +467,8 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, if (pmc->multiaddr == IGMP_ALL_HOSTS) return skb; - if (ipv4_is_local_multicast(pmc->multiaddr) && !net->ipv4.sysctl_igmp_llm_reports) + if (ipv4_is_local_multicast(pmc->multiaddr) && + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) return skb; mtu = READ_ONCE(dev->mtu); @@ -595,7 +594,7 @@ static int igmpv3_send_report(struct in_device *in_dev, struct ip_mc_list *pmc) if (pmc->multiaddr == IGMP_ALL_HOSTS) continue; if (ipv4_is_local_multicast(pmc->multiaddr) && - !net->ipv4.sysctl_igmp_llm_reports) + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) continue; spin_lock_bh(&pmc->lock); if (pmc->sfcount[MCAST_EXCLUDE]) @@ -738,7 +737,8 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, if (type == IGMPV3_HOST_MEMBERSHIP_REPORT) return igmpv3_send_report(in_dev, pmc); - if (ipv4_is_local_multicast(group) && !net->ipv4.sysctl_igmp_llm_reports) + if (ipv4_is_local_multicast(group) && + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) return 0; if (type == IGMP_HOST_LEAVE_MESSAGE) @@ -805,10 +805,17 @@ static void igmp_gq_timer_expire(struct timer_list *t) static void igmp_ifc_timer_expire(struct timer_list *t) { struct in_device *in_dev = from_timer(in_dev, t, mr_ifc_timer); + u32 mr_ifc_count; igmpv3_send_cr(in_dev); - if (in_dev->mr_ifc_count) { - in_dev->mr_ifc_count--; +restart: + mr_ifc_count = READ_ONCE(in_dev->mr_ifc_count); + + if (mr_ifc_count) { + if (cmpxchg(&in_dev->mr_ifc_count, + mr_ifc_count, + mr_ifc_count - 1) != mr_ifc_count) + goto restart; igmp_ifc_start_timer(in_dev, unsolicited_report_interval(in_dev)); } @@ -820,7 +827,7 @@ static void igmp_ifc_event(struct in_device *in_dev) struct net *net = dev_net(in_dev->dev); if (IGMP_V1_SEEN(in_dev) || IGMP_V2_SEEN(in_dev)) return; - in_dev->mr_ifc_count = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; + WRITE_ONCE(in_dev->mr_ifc_count, in_dev->mr_qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv)); igmp_ifc_start_timer(in_dev, 1); } @@ -915,7 +922,8 @@ static bool igmp_heard_report(struct in_device *in_dev, __be32 group) if (group == IGMP_ALL_HOSTS) return false; - if (ipv4_is_local_multicast(group) && !net->ipv4.sysctl_igmp_llm_reports) + if (ipv4_is_local_multicast(group) && + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) return false; rcu_read_lock(); @@ -959,7 +967,7 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, in_dev->mr_qri; } /* cancel the interface change timer */ - in_dev->mr_ifc_count = 0; + WRITE_ONCE(in_dev->mr_ifc_count, 0); if (del_timer(&in_dev->mr_ifc_timer)) __in_dev_put(in_dev); /* clear deleted report items */ @@ -1001,7 +1009,7 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, * received value was zero, use the default or statically * configured value. */ - in_dev->mr_qrv = ih3->qrv ?: net->ipv4.sysctl_igmp_qrv; + in_dev->mr_qrv = ih3->qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); in_dev->mr_qi = IGMPV3_QQIC(ih3->qqic)*HZ ?: IGMP_QUERY_INTERVAL; /* RFC3376, 8.3. Query Response Interval: @@ -1040,7 +1048,7 @@ static bool igmp_heard_query(struct in_device *in_dev, struct sk_buff *skb, if (im->multiaddr == IGMP_ALL_HOSTS) continue; if (ipv4_is_local_multicast(im->multiaddr) && - !net->ipv4.sysctl_igmp_llm_reports) + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) continue; spin_lock_bh(&im->lock); if (im->tm_running) @@ -1181,7 +1189,7 @@ static void igmpv3_add_delrec(struct in_device *in_dev, struct ip_mc_list *im, pmc->interface = im->interface; in_dev_hold(in_dev); pmc->multiaddr = im->multiaddr; - pmc->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; + pmc->crcount = in_dev->mr_qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); pmc->sfmode = im->sfmode; if (pmc->sfmode == MCAST_INCLUDE) { struct ip_sf_list *psf; @@ -1232,9 +1240,11 @@ static void igmpv3_del_delrec(struct in_device *in_dev, struct ip_mc_list *im) swap(im->tomb, pmc->tomb); swap(im->sources, pmc->sources); for (psf = im->sources; psf; psf = psf->sf_next) - psf->sf_crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; + psf->sf_crcount = in_dev->mr_qrv ?: + READ_ONCE(net->ipv4.sysctl_igmp_qrv); } else { - im->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; + im->crcount = in_dev->mr_qrv ?: + READ_ONCE(net->ipv4.sysctl_igmp_qrv); } in_dev_put(pmc->interface); kfree_pmc(pmc); @@ -1291,7 +1301,8 @@ static void __igmp_group_dropped(struct ip_mc_list *im, gfp_t gfp) #ifdef CONFIG_IP_MULTICAST if (im->multiaddr == IGMP_ALL_HOSTS) return; - if (ipv4_is_local_multicast(im->multiaddr) && !net->ipv4.sysctl_igmp_llm_reports) + if (ipv4_is_local_multicast(im->multiaddr) && + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) return; reporter = im->reporter; @@ -1333,13 +1344,14 @@ static void igmp_group_added(struct ip_mc_list *im) #ifdef CONFIG_IP_MULTICAST if (im->multiaddr == IGMP_ALL_HOSTS) return; - if (ipv4_is_local_multicast(im->multiaddr) && !net->ipv4.sysctl_igmp_llm_reports) + if (ipv4_is_local_multicast(im->multiaddr) && + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) return; if (in_dev->dead) return; - im->unsolicit_count = net->ipv4.sysctl_igmp_qrv; + im->unsolicit_count = READ_ONCE(net->ipv4.sysctl_igmp_qrv); if (IGMP_V1_SEEN(in_dev) || IGMP_V2_SEEN(in_dev)) { spin_lock_bh(&im->lock); igmp_start_timer(im, IGMP_INITIAL_REPORT_DELAY); @@ -1353,7 +1365,7 @@ static void igmp_group_added(struct ip_mc_list *im) * IN() to IN(A). */ if (im->sfmode == MCAST_EXCLUDE) - im->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; + im->crcount = in_dev->mr_qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); igmp_ifc_event(in_dev); #endif @@ -1637,7 +1649,7 @@ static void ip_mc_rejoin_groups(struct in_device *in_dev) if (im->multiaddr == IGMP_ALL_HOSTS) continue; if (ipv4_is_local_multicast(im->multiaddr) && - !net->ipv4.sysctl_igmp_llm_reports) + !READ_ONCE(net->ipv4.sysctl_igmp_llm_reports)) continue; /* a failover is happening and switches @@ -1726,7 +1738,7 @@ void ip_mc_down(struct in_device *in_dev) igmp_group_dropped(pmc); #ifdef CONFIG_IP_MULTICAST - in_dev->mr_ifc_count = 0; + WRITE_ONCE(in_dev->mr_ifc_count, 0); if (del_timer(&in_dev->mr_ifc_timer)) __in_dev_put(in_dev); in_dev->mr_gq_running = 0; @@ -1744,7 +1756,7 @@ static void ip_mc_reset(struct in_device *in_dev) in_dev->mr_qi = IGMP_QUERY_INTERVAL; in_dev->mr_qri = IGMP_QUERY_RESPONSE_INTERVAL; - in_dev->mr_qrv = net->ipv4.sysctl_igmp_qrv; + in_dev->mr_qrv = READ_ONCE(net->ipv4.sysctl_igmp_qrv); } #else static void ip_mc_reset(struct in_device *in_dev) @@ -1803,6 +1815,7 @@ void ip_mc_destroy_dev(struct in_device *in_dev) while ((i = rtnl_dereference(in_dev->mc_list)) != NULL) { in_dev->mc_list = i->next_rcu; in_dev->mc_count--; + ip_mc_clear_src(i); ip_ma_put(i); } } @@ -1877,7 +1890,7 @@ static int ip_mc_del1_src(struct ip_mc_list *pmc, int sfmode, #ifdef CONFIG_IP_MULTICAST if (psf->sf_oldin && !IGMP_V1_SEEN(in_dev) && !IGMP_V2_SEEN(in_dev)) { - psf->sf_crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; + psf->sf_crcount = in_dev->mr_qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); psf->sf_next = pmc->tomb; pmc->tomb = psf; rv = 1; @@ -1941,8 +1954,8 @@ static int ip_mc_del_src(struct in_device *in_dev, __be32 *pmca, int sfmode, /* filter mode change */ pmc->sfmode = MCAST_INCLUDE; #ifdef CONFIG_IP_MULTICAST - pmc->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; - in_dev->mr_ifc_count = pmc->crcount; + pmc->crcount = in_dev->mr_qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); + WRITE_ONCE(in_dev->mr_ifc_count, pmc->crcount); for (psf = pmc->sources; psf; psf = psf->sf_next) psf->sf_crcount = 0; igmp_ifc_event(pmc->interface); @@ -2120,8 +2133,8 @@ static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode, #ifdef CONFIG_IP_MULTICAST /* else no filters; keep old mode for reports */ - pmc->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv; - in_dev->mr_ifc_count = pmc->crcount; + pmc->crcount = in_dev->mr_qrv ?: READ_ONCE(net->ipv4.sysctl_igmp_qrv); + WRITE_ONCE(in_dev->mr_ifc_count, pmc->crcount); for (psf = pmc->sources; psf; psf = psf->sf_next) psf->sf_crcount = 0; igmp_ifc_event(in_dev); @@ -2186,7 +2199,7 @@ static int __ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr, count++; } err = -ENOBUFS; - if (count >= net->ipv4.sysctl_igmp_max_memberships) + if (count >= READ_ONCE(net->ipv4.sysctl_igmp_max_memberships)) goto done; iml = sock_kmalloc(sk, sizeof(*iml), GFP_KERNEL); if (!iml) @@ -2234,7 +2247,7 @@ static int ip_mc_leave_src(struct sock *sk, struct ip_mc_socklist *iml, iml->sfmode, psf->sl_count, psf->sl_addr, 0); RCU_INIT_POINTER(iml->sflist, NULL); /* decrease mem now to avoid the memleak warning */ - atomic_sub(IP_SFLSIZE(psf->sl_max), &sk->sk_omem_alloc); + atomic_sub(struct_size(psf, sl_addr, psf->sl_max), &sk->sk_omem_alloc); kfree_rcu(psf, rcu); return err; } @@ -2373,7 +2386,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct } /* else, add a new source to the filter */ - if (psl && psl->sl_count >= net->ipv4.sysctl_igmp_max_msf) { + if (psl && psl->sl_count >= READ_ONCE(net->ipv4.sysctl_igmp_max_msf)) { err = -ENOBUFS; goto done; } @@ -2383,7 +2396,8 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct if (psl) count += psl->sl_max; - newpsl = sock_kmalloc(sk, IP_SFLSIZE(count), GFP_KERNEL); + newpsl = sock_kmalloc(sk, struct_size(newpsl, sl_addr, count), + GFP_KERNEL); if (!newpsl) { err = -ENOBUFS; goto done; @@ -2394,10 +2408,12 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct for (i = 0; i < psl->sl_count; i++) newpsl->sl_addr[i] = psl->sl_addr[i]; /* decrease mem now to avoid the memleak warning */ - atomic_sub(IP_SFLSIZE(psl->sl_max), &sk->sk_omem_alloc); - kfree_rcu(psl, rcu); + atomic_sub(struct_size(psl, sl_addr, psl->sl_max), + &sk->sk_omem_alloc); } rcu_assign_pointer(pmc->sflist, newpsl); + if (psl) + kfree_rcu(psl, rcu); psl = newpsl; } rv = 1; /* > 0 for insert logic below if sl_count is 0 */ @@ -2469,19 +2485,22 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) goto done; } if (msf->imsf_numsrc) { - newpsl = sock_kmalloc(sk, IP_SFLSIZE(msf->imsf_numsrc), - GFP_KERNEL); + newpsl = sock_kmalloc(sk, struct_size(newpsl, sl_addr, + msf->imsf_numsrc), + GFP_KERNEL); if (!newpsl) { err = -ENOBUFS; goto done; } newpsl->sl_max = newpsl->sl_count = msf->imsf_numsrc; - memcpy(newpsl->sl_addr, msf->imsf_slist, - msf->imsf_numsrc * sizeof(msf->imsf_slist[0])); + memcpy(newpsl->sl_addr, msf->imsf_slist_flex, + flex_array_size(msf, imsf_slist_flex, msf->imsf_numsrc)); err = ip_mc_add_src(in_dev, &msf->imsf_multiaddr, msf->imsf_fmode, newpsl->sl_count, newpsl->sl_addr, 0); if (err) { - sock_kfree_s(sk, newpsl, IP_SFLSIZE(newpsl->sl_max)); + sock_kfree_s(sk, newpsl, + struct_size(newpsl, sl_addr, + newpsl->sl_max)); goto done; } } else { @@ -2494,12 +2513,15 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) (void) ip_mc_del_src(in_dev, &msf->imsf_multiaddr, pmc->sfmode, psl->sl_count, psl->sl_addr, 0); /* decrease mem now to avoid the memleak warning */ - atomic_sub(IP_SFLSIZE(psl->sl_max), &sk->sk_omem_alloc); - kfree_rcu(psl, rcu); - } else + atomic_sub(struct_size(psl, sl_addr, psl->sl_max), + &sk->sk_omem_alloc); + } else { (void) ip_mc_del_src(in_dev, &msf->imsf_multiaddr, pmc->sfmode, 0, NULL, 0); + } rcu_assign_pointer(pmc->sflist, newpsl); + if (psl) + kfree_rcu(psl, rcu); pmc->sfmode = msf->imsf_fmode; err = 0; done: @@ -2507,11 +2529,10 @@ done: err = ip_mc_leave_group(sk, &imr); return err; } - int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf, - struct ip_msfilter __user *optval, int __user *optlen) + sockptr_t optval, sockptr_t optlen) { - int err, len, count, copycount; + int err, len, count, copycount, msf_size; struct ip_mreqn imr; __be32 addr = msf->imsf_multiaddr; struct ip_mc_socklist *pmc; @@ -2546,20 +2567,22 @@ int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf, msf->imsf_fmode = pmc->sfmode; psl = rtnl_dereference(pmc->sflist); if (!psl) { - len = 0; count = 0; } else { count = psl->sl_count; } copycount = count < msf->imsf_numsrc ? count : msf->imsf_numsrc; - len = copycount * sizeof(psl->sl_addr[0]); + len = flex_array_size(psl, sl_addr, copycount); msf->imsf_numsrc = count; - if (put_user(IP_MSFILTER_SIZE(copycount), optlen) || - copy_to_user(optval, msf, IP_MSFILTER_SIZE(0))) { + msf_size = IP_MSFILTER_SIZE(copycount); + if (copy_to_sockptr(optlen, &msf_size, sizeof(int)) || + copy_to_sockptr(optval, msf, IP_MSFILTER_SIZE(0))) { return -EFAULT; } if (len && - copy_to_user(&optval->imsf_slist[0], psl->sl_addr, len)) + copy_to_sockptr_offset(optval, + offsetof(struct ip_msfilter, imsf_slist_flex), + psl->sl_addr, len)) return -EFAULT; return 0; done: @@ -2567,9 +2590,9 @@ done: } int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf, - struct group_filter __user *optval, int __user *optlen) + sockptr_t optval, size_t ss_offset) { - int err, i, count, copycount; + int i, count, copycount; struct sockaddr_in *psin; __be32 addr; struct ip_mc_socklist *pmc; @@ -2585,24 +2608,18 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf, if (!ipv4_is_multicast(addr)) return -EINVAL; - err = -EADDRNOTAVAIL; - for_each_pmc_rtnl(inet, pmc) { if (pmc->multi.imr_multiaddr.s_addr == addr && pmc->multi.imr_ifindex == gsf->gf_interface) break; } if (!pmc) /* must have a prior join */ - goto done; + return -EADDRNOTAVAIL; gsf->gf_fmode = pmc->sfmode; psl = rtnl_dereference(pmc->sflist); count = psl ? psl->sl_count : 0; copycount = count < gsf->gf_numsrc ? count : gsf->gf_numsrc; gsf->gf_numsrc = count; - if (put_user(GROUP_FILTER_SIZE(copycount), optlen) || - copy_to_user(optval, gsf, GROUP_FILTER_SIZE(0))) { - return -EFAULT; - } for (i = 0; i < copycount; i++) { struct sockaddr_storage ss; @@ -2610,12 +2627,12 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf, memset(&ss, 0, sizeof(ss)); psin->sin_family = AF_INET; psin->sin_addr.s_addr = psl->sl_addr[i]; - if (copy_to_user(&optval->gf_slist[i], &ss, sizeof(ss))) + if (copy_to_sockptr_offset(optval, ss_offset, + &ss, sizeof(ss))) return -EFAULT; + ss_offset += sizeof(ss); } return 0; -done: - return err; } /* @@ -2722,6 +2739,7 @@ int ip_check_mc_rcu(struct in_device *in_dev, __be32 mc_addr, __be32 src_addr, u rv = 1; } else if (im) { if (src_addr) { + spin_lock_bh(&im->lock); for (psf = im->sources; psf; psf = psf->sf_next) { if (psf->sf_inaddr == src_addr) break; @@ -2732,6 +2750,7 @@ int ip_check_mc_rcu(struct in_device *in_dev, __be32 mc_addr, __be32 src_addr, u im->sfcount[MCAST_EXCLUDE]; else rv = im->sfcount[MCAST_EXCLUDE] != 0; + spin_unlock_bh(&im->lock); } else rv = 1; /* unspecified source; tentatively allow */ } @@ -2831,7 +2850,7 @@ static int igmp_mc_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "Idx\tDevice : Count Querier\tGroup Users Timer\tReporter\n"); else { - struct ip_mc_list *im = (struct ip_mc_list *)v; + struct ip_mc_list *im = v; struct igmp_mc_iter_state *state = igmp_mc_seq_private(seq); char *querier; long delta; @@ -2975,7 +2994,7 @@ static void igmp_mcf_seq_stop(struct seq_file *seq, void *v) static int igmp_mcf_seq_show(struct seq_file *seq, void *v) { - struct ip_sf_list *psf = (struct ip_sf_list *)v; + struct ip_sf_list *psf = v; struct igmp_mcf_iter_state *state = igmp_mcf_seq_private(seq); if (v == SEQ_START_TOKEN) { diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index d545fb99a8a1..4e84ed21d16f 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -24,17 +24,19 @@ #include <net/addrconf.h> #if IS_ENABLED(CONFIG_IPV6) -/* match_wildcard == true: IPV6_ADDR_ANY equals to any IPv6 addresses if IPv6 - * only, and any IPv4 addresses if not IPv6 only - * match_wildcard == false: addresses must be exactly the same, i.e. - * IPV6_ADDR_ANY only equals to IPV6_ADDR_ANY, - * and 0.0.0.0 equals to 0.0.0.0 only +/* match_sk*_wildcard == true: IPV6_ADDR_ANY equals to any IPv6 addresses + * if IPv6 only, and any IPv4 addresses + * if not IPv6 only + * match_sk*_wildcard == false: addresses must be exactly the same, i.e. + * IPV6_ADDR_ANY only equals to IPV6_ADDR_ANY, + * and 0.0.0.0 equals to 0.0.0.0 only */ static bool ipv6_rcv_saddr_equal(const struct in6_addr *sk1_rcv_saddr6, const struct in6_addr *sk2_rcv_saddr6, __be32 sk1_rcv_saddr, __be32 sk2_rcv_saddr, bool sk1_ipv6only, bool sk2_ipv6only, - bool match_wildcard) + bool match_sk1_wildcard, + bool match_sk2_wildcard) { int addr_type = ipv6_addr_type(sk1_rcv_saddr6); int addr_type2 = sk2_rcv_saddr6 ? ipv6_addr_type(sk2_rcv_saddr6) : IPV6_ADDR_MAPPED; @@ -44,8 +46,8 @@ static bool ipv6_rcv_saddr_equal(const struct in6_addr *sk1_rcv_saddr6, if (!sk2_ipv6only) { if (sk1_rcv_saddr == sk2_rcv_saddr) return true; - if (!sk1_rcv_saddr || !sk2_rcv_saddr) - return match_wildcard; + return (match_sk1_wildcard && !sk1_rcv_saddr) || + (match_sk2_wildcard && !sk2_rcv_saddr); } return false; } @@ -53,11 +55,11 @@ static bool ipv6_rcv_saddr_equal(const struct in6_addr *sk1_rcv_saddr6, if (addr_type == IPV6_ADDR_ANY && addr_type2 == IPV6_ADDR_ANY) return true; - if (addr_type2 == IPV6_ADDR_ANY && match_wildcard && + if (addr_type2 == IPV6_ADDR_ANY && match_sk2_wildcard && !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED)) return true; - if (addr_type == IPV6_ADDR_ANY && match_wildcard && + if (addr_type == IPV6_ADDR_ANY && match_sk1_wildcard && !(sk1_ipv6only && addr_type2 == IPV6_ADDR_MAPPED)) return true; @@ -69,18 +71,19 @@ static bool ipv6_rcv_saddr_equal(const struct in6_addr *sk1_rcv_saddr6, } #endif -/* match_wildcard == true: 0.0.0.0 equals to any IPv4 addresses - * match_wildcard == false: addresses must be exactly the same, i.e. - * 0.0.0.0 only equals to 0.0.0.0 +/* match_sk*_wildcard == true: 0.0.0.0 equals to any IPv4 addresses + * match_sk*_wildcard == false: addresses must be exactly the same, i.e. + * 0.0.0.0 only equals to 0.0.0.0 */ static bool ipv4_rcv_saddr_equal(__be32 sk1_rcv_saddr, __be32 sk2_rcv_saddr, - bool sk2_ipv6only, bool match_wildcard) + bool sk2_ipv6only, bool match_sk1_wildcard, + bool match_sk2_wildcard) { if (!sk2_ipv6only) { if (sk1_rcv_saddr == sk2_rcv_saddr) return true; - if (!sk1_rcv_saddr || !sk2_rcv_saddr) - return match_wildcard; + return (match_sk1_wildcard && !sk1_rcv_saddr) || + (match_sk2_wildcard && !sk2_rcv_saddr); } return false; } @@ -96,10 +99,12 @@ bool inet_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2, sk2->sk_rcv_saddr, ipv6_only_sock(sk), ipv6_only_sock(sk2), + match_wildcard, match_wildcard); #endif return ipv4_rcv_saddr_equal(sk->sk_rcv_saddr, sk2->sk_rcv_saddr, - ipv6_only_sock(sk2), match_wildcard); + ipv6_only_sock(sk2), match_wildcard, + match_wildcard); } EXPORT_SYMBOL(inet_rcv_saddr_equal); @@ -125,63 +130,172 @@ void inet_get_local_port_range(struct net *net, int *low, int *high) } EXPORT_SYMBOL(inet_get_local_port_range); +static bool inet_use_bhash2_on_bind(const struct sock *sk) +{ +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) { + int addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr); + + return addr_type != IPV6_ADDR_ANY && + addr_type != IPV6_ADDR_MAPPED; + } +#endif + return sk->sk_rcv_saddr != htonl(INADDR_ANY); +} + +static bool inet_bind_conflict(const struct sock *sk, struct sock *sk2, + kuid_t sk_uid, bool relax, + bool reuseport_cb_ok, bool reuseport_ok) +{ + int bound_dev_if2; + + if (sk == sk2) + return false; + + bound_dev_if2 = READ_ONCE(sk2->sk_bound_dev_if); + + if (!sk->sk_bound_dev_if || !bound_dev_if2 || + sk->sk_bound_dev_if == bound_dev_if2) { + if (sk->sk_reuse && sk2->sk_reuse && + sk2->sk_state != TCP_LISTEN) { + if (!relax || (!reuseport_ok && sk->sk_reuseport && + sk2->sk_reuseport && reuseport_cb_ok && + (sk2->sk_state == TCP_TIME_WAIT || + uid_eq(sk_uid, sock_i_uid(sk2))))) + return true; + } else if (!reuseport_ok || !sk->sk_reuseport || + !sk2->sk_reuseport || !reuseport_cb_ok || + (sk2->sk_state != TCP_TIME_WAIT && + !uid_eq(sk_uid, sock_i_uid(sk2)))) { + return true; + } + } + return false; +} + +static bool inet_bhash2_conflict(const struct sock *sk, + const struct inet_bind2_bucket *tb2, + kuid_t sk_uid, + bool relax, bool reuseport_cb_ok, + bool reuseport_ok) +{ + struct sock *sk2; + + sk_for_each_bound_bhash2(sk2, &tb2->owners) { + if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) + continue; + + if (inet_bind_conflict(sk, sk2, sk_uid, relax, + reuseport_cb_ok, reuseport_ok)) + return true; + } + return false; +} + +/* This should be called only when the tb and tb2 hashbuckets' locks are held */ static int inet_csk_bind_conflict(const struct sock *sk, const struct inet_bind_bucket *tb, + const struct inet_bind2_bucket *tb2, /* may be null */ bool relax, bool reuseport_ok) { - struct sock *sk2; - bool reuse = sk->sk_reuse; - bool reuseport = !!sk->sk_reuseport && reuseport_ok; + bool reuseport_cb_ok; + struct sock_reuseport *reuseport_cb; kuid_t uid = sock_i_uid((struct sock *)sk); + rcu_read_lock(); + reuseport_cb = rcu_dereference(sk->sk_reuseport_cb); + /* paired with WRITE_ONCE() in __reuseport_(add|detach)_closed_sock */ + reuseport_cb_ok = !reuseport_cb || READ_ONCE(reuseport_cb->num_closed_socks); + rcu_read_unlock(); + /* * Unlike other sk lookup places we do not check * for sk_net here, since _all_ the socks listed - * in tb->owners list belong to the same net - the - * one this bucket belongs to. + * in tb->owners and tb2->owners list belong + * to the same net - the one this bucket belongs to. */ - sk_for_each_bound(sk2, &tb->owners) { - if (sk != sk2 && - (!sk->sk_bound_dev_if || - !sk2->sk_bound_dev_if || - sk->sk_bound_dev_if == sk2->sk_bound_dev_if)) { - if ((!reuse || !sk2->sk_reuse || - sk2->sk_state == TCP_LISTEN) && - (!reuseport || !sk2->sk_reuseport || - rcu_access_pointer(sk->sk_reuseport_cb) || - (sk2->sk_state != TCP_TIME_WAIT && - !uid_eq(uid, sock_i_uid(sk2))))) { - if (inet_rcv_saddr_equal(sk, sk2, true)) - break; - } - if (!relax && reuse && sk2->sk_reuse && - sk2->sk_state != TCP_LISTEN) { - if (inet_rcv_saddr_equal(sk, sk2, true)) - break; - } - } + if (!inet_use_bhash2_on_bind(sk)) { + struct sock *sk2; + + sk_for_each_bound(sk2, &tb->owners) + if (inet_bind_conflict(sk, sk2, uid, relax, + reuseport_cb_ok, reuseport_ok) && + inet_rcv_saddr_equal(sk, sk2, true)) + return true; + + return false; } - return sk2 != NULL; + + /* Conflicts with an existing IPV6_ADDR_ANY (if ipv6) or INADDR_ANY (if + * ipv4) should have been checked already. We need to do these two + * checks separately because their spinlocks have to be acquired/released + * independently of each other, to prevent possible deadlocks + */ + return tb2 && inet_bhash2_conflict(sk, tb2, uid, relax, reuseport_cb_ok, + reuseport_ok); +} + +/* Determine if there is a bind conflict with an existing IPV6_ADDR_ANY (if ipv6) or + * INADDR_ANY (if ipv4) socket. + * + * Caller must hold bhash hashbucket lock with local bh disabled, to protect + * against concurrent binds on the port for addr any + */ +static bool inet_bhash2_addr_any_conflict(const struct sock *sk, int port, int l3mdev, + bool relax, bool reuseport_ok) +{ + kuid_t uid = sock_i_uid((struct sock *)sk); + const struct net *net = sock_net(sk); + struct sock_reuseport *reuseport_cb; + struct inet_bind_hashbucket *head2; + struct inet_bind2_bucket *tb2; + bool reuseport_cb_ok; + + rcu_read_lock(); + reuseport_cb = rcu_dereference(sk->sk_reuseport_cb); + /* paired with WRITE_ONCE() in __reuseport_(add|detach)_closed_sock */ + reuseport_cb_ok = !reuseport_cb || READ_ONCE(reuseport_cb->num_closed_socks); + rcu_read_unlock(); + + head2 = inet_bhash2_addr_any_hashbucket(sk, net, port); + + spin_lock(&head2->lock); + + inet_bind_bucket_for_each(tb2, &head2->chain) + if (inet_bind2_bucket_match_addr_any(tb2, net, port, l3mdev, sk)) + break; + + if (tb2 && inet_bhash2_conflict(sk, tb2, uid, relax, reuseport_cb_ok, + reuseport_ok)) { + spin_unlock(&head2->lock); + return true; + } + + spin_unlock(&head2->lock); + return false; } /* * Find an open port number for the socket. Returns with the - * inet_bind_hashbucket lock held. + * inet_bind_hashbucket locks held if successful. */ static struct inet_bind_hashbucket * -inet_csk_find_open_port(struct sock *sk, struct inet_bind_bucket **tb_ret, int *port_ret) +inet_csk_find_open_port(const struct sock *sk, struct inet_bind_bucket **tb_ret, + struct inet_bind2_bucket **tb2_ret, + struct inet_bind_hashbucket **head2_ret, int *port_ret) { - struct inet_hashinfo *hinfo = sk->sk_prot->h.hashinfo; - int port = 0; - struct inet_bind_hashbucket *head; + struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk); + int i, low, high, attempt_half, port, l3mdev; + struct inet_bind_hashbucket *head, *head2; struct net *net = sock_net(sk); - int i, low, high, attempt_half; + struct inet_bind2_bucket *tb2; struct inet_bind_bucket *tb; u32 remaining, offset; - int l3mdev; + bool relax = false; l3mdev = inet_sk_bound_l3mdev(sk); +ports_exhausted: attempt_half = (sk->sk_reuse == SK_CAN_REUSE) ? 1 : 0; other_half_scan: inet_get_local_port_range(net, &low, &high); @@ -200,7 +314,7 @@ other_half_scan: if (likely(remaining > 1)) remaining &= ~1U; - offset = prandom_u32() % remaining; + offset = prandom_u32_max(remaining); /* __inet_hash_connect() favors ports having @low parity * We do the opposite to not pollute connect() users. */ @@ -216,11 +330,20 @@ other_parity_scan: head = &hinfo->bhash[inet_bhashfn(net, port, hinfo->bhash_size)]; spin_lock_bh(&head->lock); + if (inet_use_bhash2_on_bind(sk)) { + if (inet_bhash2_addr_any_conflict(sk, port, l3mdev, relax, false)) + goto next_port; + } + + head2 = inet_bhashfn_portaddr(hinfo, sk, net, port); + spin_lock(&head2->lock); + tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk); inet_bind_bucket_for_each(tb, &head->chain) - if (net_eq(ib_net(tb), net) && tb->l3mdev == l3mdev && - tb->port == port) { - if (!inet_csk_bind_conflict(sk, tb, false, false)) + if (inet_bind_bucket_match(tb, net, port, l3mdev)) { + if (!inet_csk_bind_conflict(sk, tb, tb2, + relax, false)) goto success; + spin_unlock(&head2->lock); goto next_port; } tb = NULL; @@ -239,10 +362,18 @@ next_port: attempt_half = 2; goto other_half_scan; } + + if (READ_ONCE(net->ipv4.sysctl_ip_autobind_reuse) && !relax) { + /* We still have a chance to connect to different destinations */ + relax = true; + goto ports_exhausted; + } return NULL; success: *port_ret = port; *tb_ret = tb; + *tb2_ret = tb2; + *head2_ret = head2; return head; } @@ -273,61 +404,18 @@ static inline int sk_reuseport_match(struct inet_bind_bucket *tb, tb->fast_rcv_saddr, sk->sk_rcv_saddr, tb->fast_ipv6_only, - ipv6_only_sock(sk), true); + ipv6_only_sock(sk), true, false); #endif return ipv4_rcv_saddr_equal(tb->fast_rcv_saddr, sk->sk_rcv_saddr, - ipv6_only_sock(sk), true); + ipv6_only_sock(sk), true, false); } -/* Obtain a reference to a local port for the given sock, - * if snum is zero it means select any available local port. - * We try to allocate an odd port (and leave even ports for connect()) - */ -int inet_csk_get_port(struct sock *sk, unsigned short snum) +void inet_csk_update_fastreuse(struct inet_bind_bucket *tb, + struct sock *sk) { - bool reuse = sk->sk_reuse && sk->sk_state != TCP_LISTEN; - struct inet_hashinfo *hinfo = sk->sk_prot->h.hashinfo; - int ret = 1, port = snum; - struct inet_bind_hashbucket *head; - struct net *net = sock_net(sk); - struct inet_bind_bucket *tb = NULL; kuid_t uid = sock_i_uid(sk); - int l3mdev; - - l3mdev = inet_sk_bound_l3mdev(sk); - - if (!port) { - head = inet_csk_find_open_port(sk, &tb, &port); - if (!head) - return ret; - if (!tb) - goto tb_not_found; - goto success; - } - head = &hinfo->bhash[inet_bhashfn(net, port, - hinfo->bhash_size)]; - spin_lock_bh(&head->lock); - inet_bind_bucket_for_each(tb, &head->chain) - if (net_eq(ib_net(tb), net) && tb->l3mdev == l3mdev && - tb->port == port) - goto tb_found; -tb_not_found: - tb = inet_bind_bucket_create(hinfo->bind_bucket_cachep, - net, head, port, l3mdev); - if (!tb) - goto fail_unlock; -tb_found: - if (!hlist_empty(&tb->owners)) { - if (sk->sk_reuse == SK_FORCE_REUSE) - goto success; + bool reuse = sk->sk_reuse && sk->sk_state != TCP_LISTEN; - if ((tb->fastreuse > 0 && reuse) || - sk_reuseport_match(tb, sk)) - goto success; - if (inet_csk_bind_conflict(sk, tb, true, true)) - goto fail_unlock; - } -success: if (hlist_empty(&tb->owners)) { tb->fastreuse = reuse; if (sk->sk_reuseport) { @@ -371,12 +459,105 @@ success: tb->fastreuseport = 0; } } +} + +/* Obtain a reference to a local port for the given sock, + * if snum is zero it means select any available local port. + * We try to allocate an odd port (and leave even ports for connect()) + */ +int inet_csk_get_port(struct sock *sk, unsigned short snum) +{ + struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk); + bool reuse = sk->sk_reuse && sk->sk_state != TCP_LISTEN; + bool found_port = false, check_bind_conflict = true; + bool bhash_created = false, bhash2_created = false; + struct inet_bind_hashbucket *head, *head2; + struct inet_bind2_bucket *tb2 = NULL; + struct inet_bind_bucket *tb = NULL; + bool head2_lock_acquired = false; + int ret = 1, port = snum, l3mdev; + struct net *net = sock_net(sk); + + l3mdev = inet_sk_bound_l3mdev(sk); + + if (!port) { + head = inet_csk_find_open_port(sk, &tb, &tb2, &head2, &port); + if (!head) + return ret; + + head2_lock_acquired = true; + + if (tb && tb2) + goto success; + found_port = true; + } else { + head = &hinfo->bhash[inet_bhashfn(net, port, + hinfo->bhash_size)]; + spin_lock_bh(&head->lock); + inet_bind_bucket_for_each(tb, &head->chain) + if (inet_bind_bucket_match(tb, net, port, l3mdev)) + break; + } + + if (!tb) { + tb = inet_bind_bucket_create(hinfo->bind_bucket_cachep, net, + head, port, l3mdev); + if (!tb) + goto fail_unlock; + bhash_created = true; + } + + if (!found_port) { + if (!hlist_empty(&tb->owners)) { + if (sk->sk_reuse == SK_FORCE_REUSE || + (tb->fastreuse > 0 && reuse) || + sk_reuseport_match(tb, sk)) + check_bind_conflict = false; + } + + if (check_bind_conflict && inet_use_bhash2_on_bind(sk)) { + if (inet_bhash2_addr_any_conflict(sk, port, l3mdev, true, true)) + goto fail_unlock; + } + + head2 = inet_bhashfn_portaddr(hinfo, sk, net, port); + spin_lock(&head2->lock); + head2_lock_acquired = true; + tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk); + } + + if (!tb2) { + tb2 = inet_bind2_bucket_create(hinfo->bind2_bucket_cachep, + net, head2, port, l3mdev, sk); + if (!tb2) + goto fail_unlock; + bhash2_created = true; + } + + if (!found_port && check_bind_conflict) { + if (inet_csk_bind_conflict(sk, tb, tb2, true, true)) + goto fail_unlock; + } + +success: + inet_csk_update_fastreuse(tb, sk); + if (!inet_csk(sk)->icsk_bind_hash) - inet_bind_hash(sk, tb, port); + inet_bind_hash(sk, tb, tb2, port); WARN_ON(inet_csk(sk)->icsk_bind_hash != tb); + WARN_ON(inet_csk(sk)->icsk_bind2_hash != tb2); ret = 0; fail_unlock: + if (ret) { + if (bhash_created) + inet_bind_bucket_destroy(hinfo->bind_bucket_cachep, tb); + if (bhash2_created) + inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, + tb2); + } + if (head2_lock_acquired) + spin_unlock(&head2->lock); spin_unlock_bh(&head->lock); return ret; } @@ -500,7 +681,8 @@ out: atomic_read(&newsk->sk_rmem_alloc)); mem_cgroup_sk_alloc(newsk); if (newsk->sk_memcg && amt) - mem_cgroup_charge_skmem(newsk->sk_memcg, amt); + mem_cgroup_charge_skmem(newsk->sk_memcg, amt, + GFP_KERNEL | __GFP_NOFAIL); release_sock(newsk); } @@ -538,7 +720,7 @@ void inet_csk_clear_xmit_timers(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); - icsk->icsk_pending = icsk->icsk_ack.pending = icsk->icsk_ack.blocked = 0; + icsk->icsk_pending = icsk->icsk_ack.pending = 0; sk_stop_timer(sk, &icsk->icsk_retransmit_timer); sk_stop_timer(sk, &icsk->icsk_delack_timer); @@ -576,7 +758,7 @@ struct dst_entry *inet_csk_route_req(const struct sock *sk, (opt && opt->opt.srr) ? opt->opt.faddr : ireq->ir_rmt_addr, ireq->ir_loc_addr, ireq->ir_rmt_port, htons(ireq->ir_num), sk->sk_uid); - security_req_classify_flow(req, flowi4_to_flowi(fl4)); + security_req_classify_flow(req, flowi4_to_flowi_common(fl4)); rt = ip_route_output_flow(net, fl4, sk); if (IS_ERR(rt)) goto no_route; @@ -614,7 +796,7 @@ struct dst_entry *inet_csk_route_child_sock(const struct sock *sk, (opt && opt->opt.srr) ? opt->opt.faddr : ireq->ir_rmt_addr, ireq->ir_loc_addr, ireq->ir_rmt_port, htons(ireq->ir_num), sk->sk_uid); - security_req_classify_flow(req, flowi4_to_flowi(fl4)); + security_req_classify_flow(req, flowi4_to_flowi_common(fl4)); rt = ip_route_output_flow(net, fl4, sk); if (IS_ERR(rt)) goto no_route; @@ -631,20 +813,19 @@ no_route: EXPORT_SYMBOL_GPL(inet_csk_route_child_sock); /* Decide when to expire the request and when to resend SYN-ACK */ -static inline void syn_ack_recalc(struct request_sock *req, const int thresh, - const int max_retries, - const u8 rskq_defer_accept, - int *expire, int *resend) +static void syn_ack_recalc(struct request_sock *req, + const int max_syn_ack_retries, + const u8 rskq_defer_accept, + int *expire, int *resend) { if (!rskq_defer_accept) { - *expire = req->num_timeout >= thresh; + *expire = req->num_timeout >= max_syn_ack_retries; *resend = 1; return; } - *expire = req->num_timeout >= thresh && - (!inet_rsk(req)->acked || req->num_timeout >= max_retries); - /* - * Do not resend while waiting for data after ACK, + *expire = req->num_timeout >= max_syn_ack_retries && + (!inet_rsk(req)->acked || req->num_timeout >= rskq_defer_accept); + /* Do not resend while waiting for data after ACK, * start to resend on end of deferring period to give * last chance for data or ACK to create established socket. */ @@ -662,17 +843,78 @@ int inet_rtx_syn_ack(const struct sock *parent, struct request_sock *req) } EXPORT_SYMBOL(inet_rtx_syn_ack); +static struct request_sock *inet_reqsk_clone(struct request_sock *req, + struct sock *sk) +{ + struct sock *req_sk, *nreq_sk; + struct request_sock *nreq; + + nreq = kmem_cache_alloc(req->rsk_ops->slab, GFP_ATOMIC | __GFP_NOWARN); + if (!nreq) { + __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE); + + /* paired with refcount_inc_not_zero() in reuseport_migrate_sock() */ + sock_put(sk); + return NULL; + } + + req_sk = req_to_sk(req); + nreq_sk = req_to_sk(nreq); + + memcpy(nreq_sk, req_sk, + offsetof(struct sock, sk_dontcopy_begin)); + memcpy(&nreq_sk->sk_dontcopy_end, &req_sk->sk_dontcopy_end, + req->rsk_ops->obj_size - offsetof(struct sock, sk_dontcopy_end)); + + sk_node_init(&nreq_sk->sk_node); + nreq_sk->sk_tx_queue_mapping = req_sk->sk_tx_queue_mapping; +#ifdef CONFIG_SOCK_RX_QUEUE_MAPPING + nreq_sk->sk_rx_queue_mapping = req_sk->sk_rx_queue_mapping; +#endif + nreq_sk->sk_incoming_cpu = req_sk->sk_incoming_cpu; + + nreq->rsk_listener = sk; + + /* We need not acquire fastopenq->lock + * because the child socket is locked in inet_csk_listen_stop(). + */ + if (sk->sk_protocol == IPPROTO_TCP && tcp_rsk(nreq)->tfo_listener) + rcu_assign_pointer(tcp_sk(nreq->sk)->fastopen_rsk, nreq); + + return nreq; +} + +static void reqsk_queue_migrated(struct request_sock_queue *queue, + const struct request_sock *req) +{ + if (req->num_timeout == 0) + atomic_inc(&queue->young); + atomic_inc(&queue->qlen); +} + +static void reqsk_migrate_reset(struct request_sock *req) +{ + req->saved_syn = NULL; +#if IS_ENABLED(CONFIG_IPV6) + inet_rsk(req)->ipv6_opt = NULL; + inet_rsk(req)->pktopts = NULL; +#else + inet_rsk(req)->ireq_opt = NULL; +#endif +} + /* return true if req was found in the ehash table */ static bool reqsk_queue_unlink(struct request_sock *req) { - struct inet_hashinfo *hashinfo = req_to_sk(req)->sk_prot->h.hashinfo; + struct sock *sk = req_to_sk(req); bool found = false; - if (sk_hashed(req_to_sk(req))) { + if (sk_hashed(sk)) { + struct inet_hashinfo *hashinfo = tcp_or_dccp_get_hashinfo(sk); spinlock_t *lock = inet_ehash_lockp(hashinfo, req->rsk_hash); spin_lock(lock); - found = __sk_nulls_del_node_init_rcu(req_to_sk(req)); + found = __sk_nulls_del_node_init_rcu(sk); spin_unlock(lock); } if (timer_pending(&req->rsk_timer) && del_timer_sync(&req->rsk_timer)) @@ -680,12 +922,15 @@ static bool reqsk_queue_unlink(struct request_sock *req) return found; } -void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) +bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) { - if (reqsk_queue_unlink(req)) { + bool unlinked = reqsk_queue_unlink(req); + + if (unlinked) { reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); reqsk_put(req); } + return unlinked; } EXPORT_SYMBOL(inet_csk_reqsk_queue_drop); @@ -699,19 +944,41 @@ EXPORT_SYMBOL(inet_csk_reqsk_queue_drop_and_put); static void reqsk_timer_handler(struct timer_list *t) { struct request_sock *req = from_timer(req, t, rsk_timer); + struct request_sock *nreq = NULL, *oreq = req; struct sock *sk_listener = req->rsk_listener; - struct net *net = sock_net(sk_listener); - struct inet_connection_sock *icsk = inet_csk(sk_listener); - struct request_sock_queue *queue = &icsk->icsk_accept_queue; - int qlen, expire = 0, resend = 0; - int max_retries, thresh; - u8 defer_accept; + struct inet_connection_sock *icsk; + struct request_sock_queue *queue; + struct net *net; + int max_syn_ack_retries, qlen, expire = 0, resend = 0; + + if (inet_sk_state_load(sk_listener) != TCP_LISTEN) { + struct sock *nsk; + + nsk = reuseport_migrate_sock(sk_listener, req_to_sk(req), NULL); + if (!nsk) + goto drop; + + nreq = inet_reqsk_clone(req, nsk); + if (!nreq) + goto drop; + + /* The new timer for the cloned req can decrease the 2 + * by calling inet_csk_reqsk_queue_drop_and_put(), so + * hold another count to prevent use-after-free and + * call reqsk_put() just before return. + */ + refcount_set(&nreq->rsk_refcnt, 2 + 1); + timer_setup(&nreq->rsk_timer, reqsk_timer_handler, TIMER_PINNED); + reqsk_queue_migrated(&inet_csk(nsk)->icsk_accept_queue, req); - if (inet_sk_state_load(sk_listener) != TCP_LISTEN) - goto drop; + req = nreq; + sk_listener = nsk; + } - max_retries = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_synack_retries; - thresh = max_retries; + icsk = inet_csk(sk_listener); + net = sock_net(sk_listener); + max_syn_ack_retries = icsk->icsk_syn_retries ? : + READ_ONCE(net->ipv4.sysctl_tcp_synack_retries); /* Normally all the openreqs are young and become mature * (i.e. converted to established socket) for first timeout. * If synack was not acknowledged for 1 second, it means @@ -729,37 +996,61 @@ static void reqsk_timer_handler(struct timer_list *t) * embrions; and abort old ones without pity, if old * ones are about to clog our table. */ + queue = &icsk->icsk_accept_queue; qlen = reqsk_queue_len(queue); if ((qlen << 1) > max(8U, READ_ONCE(sk_listener->sk_max_ack_backlog))) { int young = reqsk_queue_len_young(queue) << 1; - while (thresh > 2) { + while (max_syn_ack_retries > 2) { if (qlen < young) break; - thresh--; + max_syn_ack_retries--; young <<= 1; } } - defer_accept = READ_ONCE(queue->rskq_defer_accept); - if (defer_accept) - max_retries = defer_accept; - syn_ack_recalc(req, thresh, max_retries, defer_accept, + syn_ack_recalc(req, max_syn_ack_retries, READ_ONCE(queue->rskq_defer_accept), &expire, &resend); req->rsk_ops->syn_ack_timeout(req); if (!expire && (!resend || !inet_rtx_syn_ack(sk_listener, req) || inet_rsk(req)->acked)) { - unsigned long timeo; - if (req->num_timeout++ == 0) atomic_dec(&queue->young); - timeo = min(TCP_TIMEOUT_INIT << req->num_timeout, TCP_RTO_MAX); - mod_timer(&req->rsk_timer, jiffies + timeo); + mod_timer(&req->rsk_timer, jiffies + reqsk_timeout(req, TCP_RTO_MAX)); + + if (!nreq) + return; + + if (!inet_ehash_insert(req_to_sk(nreq), req_to_sk(oreq), NULL)) { + /* delete timer */ + inet_csk_reqsk_queue_drop(sk_listener, nreq); + goto no_ownership; + } + + __NET_INC_STATS(net, LINUX_MIB_TCPMIGRATEREQSUCCESS); + reqsk_migrate_reset(oreq); + reqsk_queue_removed(&inet_csk(oreq->rsk_listener)->icsk_accept_queue, oreq); + reqsk_put(oreq); + + reqsk_put(nreq); return; } + + /* Even if we can clone the req, we may need not retransmit any more + * SYN+ACKs (nreq->num_timeout > max_syn_ack_retries, etc), or another + * CPU may win the "own_req" race so that inet_ehash_insert() fails. + */ + if (nreq) { + __NET_INC_STATS(net, LINUX_MIB_TCPMIGRATEREQFAILURE); +no_ownership: + reqsk_migrate_reset(nreq); + reqsk_queue_removed(queue, nreq); + __reqsk_free(nreq); + } + drop: - inet_csk_reqsk_queue_drop_and_put(sk_listener, req); + inet_csk_reqsk_queue_drop_and_put(oreq->rsk_listener, oreq); } static void reqsk_queue_hash_req(struct request_sock *req, @@ -768,7 +1059,7 @@ static void reqsk_queue_hash_req(struct request_sock *req, timer_setup(&req->rsk_timer, reqsk_timer_handler, TIMER_PINNED); mod_timer(&req->rsk_timer, jiffies + timeout); - inet_ehash_insert(req_to_sk(req), NULL); + inet_ehash_insert(req_to_sk(req), NULL, NULL); /* before letting lookups find us, make sure all req fields * are committed to memory and refcnt initialized. */ @@ -815,6 +1106,7 @@ struct sock *inet_csk_clone_lock(const struct sock *sk, inet_sk_set_state(newsk, TCP_SYN_RECV); newicsk->icsk_bind_hash = NULL; + newicsk->icsk_bind2_hash = NULL; inet_sk(newsk)->inet_dport = inet_rsk(req)->ir_rmt_port; inet_sk(newsk)->inet_num = inet_rsk(req)->ir_num; @@ -832,6 +1124,7 @@ struct sock *inet_csk_clone_lock(const struct sock *sk, newicsk->icsk_retransmits = 0; newicsk->icsk_backoff = 0; newicsk->icsk_probes_out = 0; + newicsk->icsk_probes_tstamp = 0; /* Deinitialize accept_queue to trap illegal accesses. */ memset(&newicsk->icsk_accept_queue, 0, sizeof(newicsk->icsk_accept_queue)); @@ -869,7 +1162,7 @@ void inet_csk_destroy_sock(struct sock *sk) sk_refcnt_debug_release(sk); - percpu_counter_dec(sk->sk_prot->orphan_count); + this_cpu_dec(*sk->sk_prot->orphan_count); sock_put(sk); } @@ -884,15 +1177,12 @@ void inet_csk_prepare_forced_close(struct sock *sk) /* sk_clone_lock locked the socket and set refcnt to 2 */ bh_unlock_sock(sk); sock_put(sk); - - /* The below has to be done to allow calling inet_csk_destroy_sock */ - sock_set_flag(sk, SOCK_DEAD); - percpu_counter_inc(sk->sk_prot->orphan_count); + inet_csk_prepare_for_destroy_sock(sk); inet_sk(sk)->inet_num = 0; } EXPORT_SYMBOL(inet_csk_prepare_forced_close); -int inet_csk_listen_start(struct sock *sk, int backlog) +int inet_csk_listen_start(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); struct inet_sock *inet = inet_sk(sk); @@ -903,6 +1193,9 @@ int inet_csk_listen_start(struct sock *sk, int backlog) sk->sk_ack_backlog = 0; inet_csk_delack_init(sk); + if (sk->sk_txrehash == SOCK_TXREHASH_DEFAULT) + sk->sk_txrehash = READ_ONCE(sock_net(sk)->core.sysctl_txrehash); + /* There is race window here: we announce ourselves listening, * but this transition is still not validated by get_port(). * It is OK, because this socket enters to hash table only @@ -931,7 +1224,7 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req, sock_orphan(child); - percpu_counter_inc(sk->sk_prot->orphan_count); + this_cpu_inc(*sk->sk_prot->orphan_count); if (sk->sk_protocol == IPPROTO_TCP && tcp_rsk(req)->tfo_listener) { BUG_ON(rcu_access_pointer(tcp_sk(child)->fastopen_rsk) != req); @@ -977,12 +1270,42 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child, struct request_sock *req, bool own_req) { if (own_req) { - inet_csk_reqsk_queue_drop(sk, req); - reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); - if (inet_csk_reqsk_queue_add(sk, req, child)) + inet_csk_reqsk_queue_drop(req->rsk_listener, req); + reqsk_queue_removed(&inet_csk(req->rsk_listener)->icsk_accept_queue, req); + + if (sk != req->rsk_listener) { + /* another listening sk has been selected, + * migrate the req to it. + */ + struct request_sock *nreq; + + /* hold a refcnt for the nreq->rsk_listener + * which is assigned in inet_reqsk_clone() + */ + sock_hold(sk); + nreq = inet_reqsk_clone(req, sk); + if (!nreq) { + inet_child_forget(sk, req, child); + goto child_put; + } + + refcount_set(&nreq->rsk_refcnt, 1); + if (inet_csk_reqsk_queue_add(sk, nreq, child)) { + __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQSUCCESS); + reqsk_migrate_reset(req); + reqsk_put(req); + return child; + } + + __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE); + reqsk_migrate_reset(nreq); + __reqsk_free(nreq); + } else if (inet_csk_reqsk_queue_add(sk, req, child)) { return child; + } } /* Too bad, another child took ownership of the request, undo. */ +child_put: bh_unlock_sock(child); sock_put(child); return NULL; @@ -1008,14 +1331,40 @@ void inet_csk_listen_stop(struct sock *sk) * of the variants now. --ANK */ while ((req = reqsk_queue_remove(queue, sk)) != NULL) { - struct sock *child = req->sk; + struct sock *child = req->sk, *nsk; + struct request_sock *nreq; local_bh_disable(); bh_lock_sock(child); WARN_ON(sock_owned_by_user(child)); sock_hold(child); + nsk = reuseport_migrate_sock(sk, child, NULL); + if (nsk) { + nreq = inet_reqsk_clone(req, nsk); + if (nreq) { + refcount_set(&nreq->rsk_refcnt, 1); + + if (inet_csk_reqsk_queue_add(nsk, nreq, child)) { + __NET_INC_STATS(sock_net(nsk), + LINUX_MIB_TCPMIGRATEREQSUCCESS); + reqsk_migrate_reset(req); + } else { + __NET_INC_STATS(sock_net(nsk), + LINUX_MIB_TCPMIGRATEREQFAILURE); + reqsk_migrate_reset(nreq); + __reqsk_free(nreq); + } + + /* inet_csk_reqsk_queue_add() has already + * called inet_child_forget() on failure case. + */ + goto skip_child_forget; + } + } + inet_child_forget(sk, req, child); +skip_child_forget: reqsk_put(req); bh_unlock_sock(child); local_bh_enable(); @@ -1050,34 +1399,6 @@ void inet_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr) } EXPORT_SYMBOL_GPL(inet_csk_addr2sockaddr); -#ifdef CONFIG_COMPAT -int inet_csk_compat_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - const struct inet_connection_sock *icsk = inet_csk(sk); - - if (icsk->icsk_af_ops->compat_getsockopt) - return icsk->icsk_af_ops->compat_getsockopt(sk, level, optname, - optval, optlen); - return icsk->icsk_af_ops->getsockopt(sk, level, optname, - optval, optlen); -} -EXPORT_SYMBOL_GPL(inet_csk_compat_getsockopt); - -int inet_csk_compat_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) -{ - const struct inet_connection_sock *icsk = inet_csk(sk); - - if (icsk->icsk_af_ops->compat_setsockopt) - return icsk->icsk_af_ops->compat_setsockopt(sk, level, optname, - optval, optlen); - return icsk->icsk_af_ops->setsockopt(sk, level, optname, - optval, optlen); -} -EXPORT_SYMBOL_GPL(inet_csk_compat_setsockopt); -#endif - static struct dst_entry *inet_csk_rebuild_route(struct sock *sk, struct flowi *fl) { const struct inet_sock *inet = inet_sk(sk); diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 8c8377568a78..b812eb36f0e3 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -23,6 +23,7 @@ #include <net/inet_hashtables.h> #include <net/inet_timewait_sock.h> #include <net/inet6_hashtables.h> +#include <net/bpf_sk_storage.h> #include <net/netlink.h> #include <linux/inet.h> @@ -42,12 +43,20 @@ struct inet_diag_entry { u16 userlocks; u32 ifindex; u32 mark; +#ifdef CONFIG_SOCK_CGROUP_DATA + u64 cgroup_id; +#endif }; static DEFINE_MUTEX(inet_diag_table_mutex); static const struct inet_diag_handler *inet_diag_lock_handler(int proto) { + if (proto < 0 || proto >= IPPROTO_MAX) { + mutex_lock(&inet_diag_table_mutex); + return ERR_PTR(-ENOENT); + } + if (!inet_diag_table[proto]) sock_load_diag_module(AF_INET, proto); @@ -116,6 +125,7 @@ int inet_diag_msg_attrs_fill(struct sock *sk, struct sk_buff *skb, bool net_admin) { const struct inet_sock *inet = inet_sk(sk); + struct inet_diag_sockopt inet_sockopt; if (nla_put_u8(skb, INET_DIAG_SHUTDOWN, sk->sk_shutdown)) goto errout; @@ -161,35 +171,86 @@ int inet_diag_msg_attrs_fill(struct sock *sk, struct sk_buff *skb, goto errout; } +#ifdef CONFIG_SOCK_CGROUP_DATA + if (nla_put_u64_64bit(skb, INET_DIAG_CGROUP_ID, + cgroup_id(sock_cgroup_ptr(&sk->sk_cgrp_data)), + INET_DIAG_PAD)) + goto errout; +#endif + r->idiag_uid = from_kuid_munged(user_ns, sock_i_uid(sk)); r->idiag_inode = sock_i_ino(sk); + memset(&inet_sockopt, 0, sizeof(inet_sockopt)); + inet_sockopt.recverr = inet->recverr; + inet_sockopt.is_icsk = inet->is_icsk; + inet_sockopt.freebind = inet->freebind; + inet_sockopt.hdrincl = inet->hdrincl; + inet_sockopt.mc_loop = inet->mc_loop; + inet_sockopt.transparent = inet->transparent; + inet_sockopt.mc_all = inet->mc_all; + inet_sockopt.nodefrag = inet->nodefrag; + inet_sockopt.bind_address_no_port = inet->bind_address_no_port; + inet_sockopt.recverr_rfc4884 = inet->recverr_rfc4884; + inet_sockopt.defer_connect = inet->defer_connect; + if (nla_put(skb, INET_DIAG_SOCKOPT, sizeof(inet_sockopt), + &inet_sockopt)) + goto errout; + return 0; errout: return 1; } EXPORT_SYMBOL_GPL(inet_diag_msg_attrs_fill); +static int inet_diag_parse_attrs(const struct nlmsghdr *nlh, int hdrlen, + struct nlattr **req_nlas) +{ + struct nlattr *nla; + int remaining; + + nlmsg_for_each_attr(nla, nlh, hdrlen, remaining) { + int type = nla_type(nla); + + if (type == INET_DIAG_REQ_PROTOCOL && nla_len(nla) != sizeof(u32)) + return -EINVAL; + + if (type < __INET_DIAG_REQ_MAX) + req_nlas[type] = nla; + } + return 0; +} + +static int inet_diag_get_protocol(const struct inet_diag_req_v2 *req, + const struct inet_diag_dump_data *data) +{ + if (data->req_nlas[INET_DIAG_REQ_PROTOCOL]) + return nla_get_u32(data->req_nlas[INET_DIAG_REQ_PROTOCOL]); + return req->sdiag_protocol; +} + +#define MAX_DUMP_ALLOC_SIZE (KMALLOC_MAX_SIZE - SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) + int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, - struct sk_buff *skb, const struct inet_diag_req_v2 *req, - struct user_namespace *user_ns, - u32 portid, u32 seq, u16 nlmsg_flags, - const struct nlmsghdr *unlh, - bool net_admin) + struct sk_buff *skb, struct netlink_callback *cb, + const struct inet_diag_req_v2 *req, + u16 nlmsg_flags, bool net_admin) { const struct tcp_congestion_ops *ca_ops; const struct inet_diag_handler *handler; + struct inet_diag_dump_data *cb_data; int ext = req->idiag_ext; struct inet_diag_msg *r; struct nlmsghdr *nlh; struct nlattr *attr; void *info = NULL; - handler = inet_diag_table[req->sdiag_protocol]; + cb_data = cb->data; + handler = inet_diag_table[inet_diag_get_protocol(req, cb_data)]; BUG_ON(!handler); - nlh = nlmsg_put(skb, portid, seq, unlh->nlmsg_type, sizeof(*r), - nlmsg_flags); + nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, + cb->nlh->nlmsg_type, sizeof(*r), nlmsg_flags); if (!nlh) return -EMSGSIZE; @@ -200,15 +261,18 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, r->idiag_state = sk->sk_state; r->idiag_timer = 0; r->idiag_retrans = 0; + r->idiag_expires = 0; - if (inet_diag_msg_attrs_fill(sk, skb, r, ext, user_ns, net_admin)) + if (inet_diag_msg_attrs_fill(sk, skb, r, ext, + sk_user_ns(NETLINK_CB(cb->skb).sk), + net_admin)) goto errout; if (ext & (1 << (INET_DIAG_MEMINFO - 1))) { struct inet_diag_meminfo minfo = { .idiag_rmem = sk_rmem_alloc_get(sk), .idiag_wmem = READ_ONCE(sk->sk_wmem_queued), - .idiag_fmem = sk->sk_forward_alloc, + .idiag_fmem = sk_forward_alloc_get(sk), .idiag_tmem = sk_wmem_alloc_get(sk), }; @@ -251,9 +315,6 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, r->idiag_retrans = icsk->icsk_probes_out; r->idiag_expires = jiffies_delta_to_msecs(sk->sk_timer.expires - jiffies); - } else { - r->idiag_timer = 0; - r->idiag_expires = 0; } if ((ext & (1 << (INET_DIAG_INFO - 1))) && handler->idiag_info_size) { @@ -298,6 +359,48 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, goto errout; } + /* Keep it at the end for potential retry with a larger skb, + * or else do best-effort fitting, which is only done for the + * first_nlmsg. + */ + if (cb_data->bpf_stg_diag) { + bool first_nlmsg = ((unsigned char *)nlh == skb->data); + unsigned int prev_min_dump_alloc; + unsigned int total_nla_size = 0; + unsigned int msg_len; + int err; + + msg_len = skb_tail_pointer(skb) - (unsigned char *)nlh; + err = bpf_sk_storage_diag_put(cb_data->bpf_stg_diag, sk, skb, + INET_DIAG_SK_BPF_STORAGES, + &total_nla_size); + + if (!err) + goto out; + + total_nla_size += msg_len; + prev_min_dump_alloc = cb->min_dump_alloc; + if (total_nla_size > prev_min_dump_alloc) + cb->min_dump_alloc = min_t(u32, total_nla_size, + MAX_DUMP_ALLOC_SIZE); + + if (!first_nlmsg) + goto errout; + + if (cb->min_dump_alloc > prev_min_dump_alloc) + /* Retry with pskb_expand_head() with + * __GFP_DIRECT_RECLAIM + */ + goto errout; + + WARN_ON_ONCE(total_nla_size <= prev_min_dump_alloc); + + /* Send what we have for this sk + * and move on to the next sk in the following + * dump() + */ + } + out: nlmsg_end(skb, nlh); return 0; @@ -308,30 +411,19 @@ errout: } EXPORT_SYMBOL_GPL(inet_sk_diag_fill); -static int inet_csk_diag_fill(struct sock *sk, - struct sk_buff *skb, - const struct inet_diag_req_v2 *req, - struct user_namespace *user_ns, - u32 portid, u32 seq, u16 nlmsg_flags, - const struct nlmsghdr *unlh, - bool net_admin) -{ - return inet_sk_diag_fill(sk, inet_csk(sk), skb, req, user_ns, - portid, seq, nlmsg_flags, unlh, net_admin); -} - static int inet_twsk_diag_fill(struct sock *sk, struct sk_buff *skb, - u32 portid, u32 seq, u16 nlmsg_flags, - const struct nlmsghdr *unlh) + struct netlink_callback *cb, + u16 nlmsg_flags, bool net_admin) { struct inet_timewait_sock *tw = inet_twsk(sk); struct inet_diag_msg *r; struct nlmsghdr *nlh; long tmo; - nlh = nlmsg_put(skb, portid, seq, unlh->nlmsg_type, sizeof(*r), - nlmsg_flags); + nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, cb->nlh->nlmsg_type, + sizeof(*r), nlmsg_flags); if (!nlh) return -EMSGSIZE; @@ -350,21 +442,27 @@ static int inet_twsk_diag_fill(struct sock *sk, r->idiag_uid = 0; r->idiag_inode = 0; + if (net_admin && nla_put_u32(skb, INET_DIAG_MARK, + tw->tw_mark)) { + nlmsg_cancel(skb, nlh); + return -EMSGSIZE; + } + nlmsg_end(skb, nlh); return 0; } static int inet_req_diag_fill(struct sock *sk, struct sk_buff *skb, - u32 portid, u32 seq, u16 nlmsg_flags, - const struct nlmsghdr *unlh, bool net_admin) + struct netlink_callback *cb, + u16 nlmsg_flags, bool net_admin) { struct request_sock *reqsk = inet_reqsk(sk); struct inet_diag_msg *r; struct nlmsghdr *nlh; long tmo; - nlh = nlmsg_put(skb, portid, seq, unlh->nlmsg_type, sizeof(*r), - nlmsg_flags); + nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, + cb->nlh->nlmsg_type, sizeof(*r), nlmsg_flags); if (!nlh) return -EMSGSIZE; @@ -385,29 +483,28 @@ static int inet_req_diag_fill(struct sock *sk, struct sk_buff *skb, r->idiag_inode = 0; if (net_admin && nla_put_u32(skb, INET_DIAG_MARK, - inet_rsk(reqsk)->ir_mark)) + inet_rsk(reqsk)->ir_mark)) { + nlmsg_cancel(skb, nlh); return -EMSGSIZE; + } nlmsg_end(skb, nlh); return 0; } static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, + struct netlink_callback *cb, const struct inet_diag_req_v2 *r, - struct user_namespace *user_ns, - u32 portid, u32 seq, u16 nlmsg_flags, - const struct nlmsghdr *unlh, bool net_admin) + u16 nlmsg_flags, bool net_admin) { if (sk->sk_state == TCP_TIME_WAIT) - return inet_twsk_diag_fill(sk, skb, portid, seq, - nlmsg_flags, unlh); + return inet_twsk_diag_fill(sk, skb, cb, nlmsg_flags, net_admin); if (sk->sk_state == TCP_NEW_SYN_RECV) - return inet_req_diag_fill(sk, skb, portid, seq, - nlmsg_flags, unlh, net_admin); + return inet_req_diag_fill(sk, skb, cb, nlmsg_flags, net_admin); - return inet_csk_diag_fill(sk, skb, r, user_ns, portid, seq, - nlmsg_flags, unlh, net_admin); + return inet_sk_diag_fill(sk, inet_csk(sk), skb, cb, r, nlmsg_flags, + net_admin); } struct sock *inet_diag_find_one_icsk(struct net *net, @@ -455,10 +552,10 @@ struct sock *inet_diag_find_one_icsk(struct net *net, EXPORT_SYMBOL_GPL(inet_diag_find_one_icsk); int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, - struct sk_buff *in_skb, - const struct nlmsghdr *nlh, + struct netlink_callback *cb, const struct inet_diag_req_v2 *req) { + struct sk_buff *in_skb = cb->skb; bool net_admin = netlink_net_capable(in_skb, CAP_NET_ADMIN); struct net *net = sock_net(in_skb->sk); struct sk_buff *rep; @@ -475,19 +572,13 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo, goto out; } - err = sk_diag_fill(sk, rep, req, - sk_user_ns(NETLINK_CB(in_skb).sk), - NETLINK_CB(in_skb).portid, - nlh->nlmsg_seq, 0, nlh, net_admin); + err = sk_diag_fill(sk, rep, cb, req, 0, net_admin); if (err < 0) { WARN_ON(err == -EMSGSIZE); nlmsg_free(rep); goto out; } - err = netlink_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid, - MSG_DONTWAIT); - if (err > 0) - err = 0; + err = nlmsg_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid); out: if (sk) @@ -499,20 +590,35 @@ EXPORT_SYMBOL_GPL(inet_diag_dump_one_icsk); static int inet_diag_cmd_exact(int cmd, struct sk_buff *in_skb, const struct nlmsghdr *nlh, + int hdrlen, const struct inet_diag_req_v2 *req) { const struct inet_diag_handler *handler; - int err; + struct inet_diag_dump_data dump_data; + int err, protocol; - handler = inet_diag_lock_handler(req->sdiag_protocol); - if (IS_ERR(handler)) + memset(&dump_data, 0, sizeof(dump_data)); + err = inet_diag_parse_attrs(nlh, hdrlen, dump_data.req_nlas); + if (err) + return err; + + protocol = inet_diag_get_protocol(req, &dump_data); + + handler = inet_diag_lock_handler(protocol); + if (IS_ERR(handler)) { err = PTR_ERR(handler); - else if (cmd == SOCK_DIAG_BY_FAMILY) - err = handler->dump_one(in_skb, nlh, req); - else if (cmd == SOCK_DESTROY && handler->destroy) + } else if (cmd == SOCK_DIAG_BY_FAMILY) { + struct netlink_callback cb = { + .nlh = nlh, + .skb = in_skb, + .data = &dump_data, + }; + err = handler->dump_one(&cb, req); + } else if (cmd == SOCK_DESTROY && handler->destroy) { err = handler->destroy(in_skb, req); - else + } else { err = -EOPNOTSUPP; + } inet_diag_unlock_handler(handler); return err; @@ -638,6 +744,16 @@ static int inet_diag_bc_run(const struct nlattr *_bc, yes = 0; break; } +#ifdef CONFIG_SOCK_CGROUP_DATA + case INET_DIAG_BC_CGROUP_COND: { + u64 cgroup_id; + + cgroup_id = get_unaligned((const u64 *)(op + 1)); + if (cgroup_id != entry->cgroup_id) + yes = 0; + break; + } +#endif } if (yes) { @@ -686,8 +802,14 @@ int inet_diag_bc_sk(const struct nlattr *bc, struct sock *sk) entry.mark = sk->sk_mark; else if (sk->sk_state == TCP_NEW_SYN_RECV) entry.mark = inet_rsk(inet_reqsk(sk))->ir_mark; + else if (sk->sk_state == TCP_TIME_WAIT) + entry.mark = inet_twsk(sk)->tw_mark; else entry.mark = 0; +#ifdef CONFIG_SOCK_CGROUP_DATA + entry.cgroup_id = sk_fullsock(sk) ? + cgroup_id(sock_cgroup_ptr(&sk->sk_cgrp_data)) : 0; +#endif return inet_diag_bc_run(bc, &entry); } @@ -777,6 +899,15 @@ static bool valid_markcond(const struct inet_diag_bc_op *op, int len, return len >= *min_len; } +#ifdef CONFIG_SOCK_CGROUP_DATA +static bool valid_cgroupcond(const struct inet_diag_bc_op *op, int len, + int *min_len) +{ + *min_len += sizeof(u64); + return len >= *min_len; +} +#endif + static int inet_diag_bc_audit(const struct nlattr *attr, const struct sk_buff *skb) { @@ -819,6 +950,12 @@ static int inet_diag_bc_audit(const struct nlattr *attr, if (!valid_markcond(bc, len, &min_len)) return -EINVAL; break; +#ifdef CONFIG_SOCK_CGROUP_DATA + case INET_DIAG_BC_CGROUP_COND: + if (!valid_cgroupcond(bc, len, &min_len)) + return -EINVAL; + break; +#endif case INET_DIAG_BC_AUTO: case INET_DIAG_BC_JMP: case INET_DIAG_BC_NOP: @@ -843,23 +980,6 @@ static int inet_diag_bc_audit(const struct nlattr *attr, return len == 0 ? 0 : -EINVAL; } -static int inet_csk_diag_dump(struct sock *sk, - struct sk_buff *skb, - struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, - const struct nlattr *bc, - bool net_admin) -{ - if (!inet_diag_bc_sk(bc, sk)) - return 0; - - return inet_csk_diag_fill(sk, skb, r, - sk_user_ns(NETLINK_CB(cb->skb).sk), - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, NLM_F_MULTI, cb->nlh, - net_admin); -} - static void twsk_build_assert(void) { BUILD_BUG_ON(offsetof(struct inet_timewait_sock, tw_family) != @@ -888,14 +1008,17 @@ static void twsk_build_assert(void) void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, struct nlattr *bc) + const struct inet_diag_req_v2 *r) { bool net_admin = netlink_net_capable(cb->skb, CAP_NET_ADMIN); + struct inet_diag_dump_data *cb_data = cb->data; struct net *net = sock_net(skb->sk); u32 idiag_states = r->idiag_states; int i, num, s_i, s_num; + struct nlattr *bc; struct sock *sk; + bc = cb_data->inet_diag_nla_bc; if (idiag_states & TCPF_SYN_RECV) idiag_states |= TCPF_NEW_SYN_RECV; s_i = cb->args[1]; @@ -905,12 +1028,13 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb, if (!(idiag_states & TCPF_LISTEN) || r->id.idiag_dport) goto skip_listen_ht; - for (i = s_i; i < INET_LHTABLE_SIZE; i++) { + for (i = s_i; i <= hashinfo->lhash2_mask; i++) { struct inet_listen_hashbucket *ilb; struct hlist_nulls_node *node; num = 0; - ilb = &hashinfo->listening_hash[i]; + ilb = &hashinfo->lhash2[i]; + spin_lock(&ilb->lock); sk_nulls_for_each(sk, node, &ilb->nulls_head) { struct inet_sock *inet = inet_sk(sk); @@ -931,8 +1055,12 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb, r->id.idiag_sport) goto next_listen; - if (inet_csk_diag_dump(sk, skb, cb, r, - bc, net_admin) < 0) { + if (!inet_diag_bc_sk(bc, sk)) + goto next_listen; + + if (inet_sk_diag_fill(sk, inet_csk(sk), skb, + cb, r, NLM_F_MULTI, + net_admin) < 0) { spin_unlock(&ilb->lock); goto done; } @@ -1010,11 +1138,8 @@ next_normal: res = 0; for (idx = 0; idx < accum; idx++) { if (res >= 0) { - res = sk_diag_fill(sk_arr[idx], skb, r, - sk_user_ns(NETLINK_CB(cb->skb).sk), - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, NLM_F_MULTI, - cb->nlh, net_admin); + res = sk_diag_fill(sk_arr[idx], skb, cb, r, + NLM_F_MULTI, net_admin); if (res < 0) num = num_arr[idx]; } @@ -1038,31 +1163,101 @@ out: EXPORT_SYMBOL_GPL(inet_diag_dump_icsk); static int __inet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, - struct nlattr *bc) + const struct inet_diag_req_v2 *r) { + struct inet_diag_dump_data *cb_data = cb->data; const struct inet_diag_handler *handler; - int err = 0; + u32 prev_min_dump_alloc; + int protocol, err = 0; - handler = inet_diag_lock_handler(r->sdiag_protocol); + protocol = inet_diag_get_protocol(r, cb_data); + +again: + prev_min_dump_alloc = cb->min_dump_alloc; + handler = inet_diag_lock_handler(protocol); if (!IS_ERR(handler)) - handler->dump(skb, cb, r, bc); + handler->dump(skb, cb, r); else err = PTR_ERR(handler); inet_diag_unlock_handler(handler); + /* The skb is not large enough to fit one sk info and + * inet_sk_diag_fill() has requested for a larger skb. + */ + if (!skb->len && cb->min_dump_alloc > prev_min_dump_alloc) { + err = pskb_expand_head(skb, 0, cb->min_dump_alloc, GFP_KERNEL); + if (!err) + goto again; + } + return err ? : skb->len; } static int inet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb) { - int hdrlen = sizeof(struct inet_diag_req_v2); - struct nlattr *bc = NULL; + return __inet_diag_dump(skb, cb, nlmsg_data(cb->nlh)); +} + +static int __inet_diag_dump_start(struct netlink_callback *cb, int hdrlen) +{ + const struct nlmsghdr *nlh = cb->nlh; + struct inet_diag_dump_data *cb_data; + struct sk_buff *skb = cb->skb; + struct nlattr *nla; + int err; - if (nlmsg_attrlen(cb->nlh, hdrlen)) - bc = nlmsg_find_attr(cb->nlh, hdrlen, INET_DIAG_REQ_BYTECODE); + cb_data = kzalloc(sizeof(*cb_data), GFP_KERNEL); + if (!cb_data) + return -ENOMEM; + + err = inet_diag_parse_attrs(nlh, hdrlen, cb_data->req_nlas); + if (err) { + kfree(cb_data); + return err; + } + nla = cb_data->inet_diag_nla_bc; + if (nla) { + err = inet_diag_bc_audit(nla, skb); + if (err) { + kfree(cb_data); + return err; + } + } + + nla = cb_data->inet_diag_nla_bpf_stgs; + if (nla) { + struct bpf_sk_storage_diag *bpf_stg_diag; + + bpf_stg_diag = bpf_sk_storage_diag_alloc(nla); + if (IS_ERR(bpf_stg_diag)) { + kfree(cb_data); + return PTR_ERR(bpf_stg_diag); + } + cb_data->bpf_stg_diag = bpf_stg_diag; + } + + cb->data = cb_data; + return 0; +} - return __inet_diag_dump(skb, cb, nlmsg_data(cb->nlh), bc); +static int inet_diag_dump_start(struct netlink_callback *cb) +{ + return __inet_diag_dump_start(cb, sizeof(struct inet_diag_req_v2)); +} + +static int inet_diag_dump_start_compat(struct netlink_callback *cb) +{ + return __inet_diag_dump_start(cb, sizeof(struct inet_diag_req)); +} + +static int inet_diag_dump_done(struct netlink_callback *cb) +{ + struct inet_diag_dump_data *cb_data = cb->data; + + bpf_sk_storage_diag_free(cb_data->bpf_stg_diag); + kfree(cb->data); + + return 0; } static int inet_diag_type2proto(int type) @@ -1081,9 +1276,7 @@ static int inet_diag_dump_compat(struct sk_buff *skb, struct netlink_callback *cb) { struct inet_diag_req *rc = nlmsg_data(cb->nlh); - int hdrlen = sizeof(struct inet_diag_req); struct inet_diag_req_v2 req; - struct nlattr *bc = NULL; req.sdiag_family = AF_UNSPEC; /* compatibility */ req.sdiag_protocol = inet_diag_type2proto(cb->nlh->nlmsg_type); @@ -1091,10 +1284,7 @@ static int inet_diag_dump_compat(struct sk_buff *skb, req.idiag_states = rc->idiag_states; req.id = rc->id; - if (nlmsg_attrlen(cb->nlh, hdrlen)) - bc = nlmsg_find_attr(cb->nlh, hdrlen, INET_DIAG_REQ_BYTECODE); - - return __inet_diag_dump(skb, cb, &req, bc); + return __inet_diag_dump(skb, cb, &req); } static int inet_diag_get_exact_compat(struct sk_buff *in_skb, @@ -1109,7 +1299,8 @@ static int inet_diag_get_exact_compat(struct sk_buff *in_skb, req.idiag_states = rc->idiag_states; req.id = rc->id; - return inet_diag_cmd_exact(SOCK_DIAG_BY_FAMILY, in_skb, nlh, &req); + return inet_diag_cmd_exact(SOCK_DIAG_BY_FAMILY, in_skb, nlh, + sizeof(struct inet_diag_req), &req); } static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh) @@ -1122,22 +1313,12 @@ static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh) return -EINVAL; if (nlh->nlmsg_flags & NLM_F_DUMP) { - if (nlmsg_attrlen(nlh, hdrlen)) { - struct nlattr *attr; - int err; - - attr = nlmsg_find_attr(nlh, hdrlen, - INET_DIAG_REQ_BYTECODE); - err = inet_diag_bc_audit(attr, skb); - if (err) - return err; - } - { - struct netlink_dump_control c = { - .dump = inet_diag_dump_compat, - }; - return netlink_dump_start(net->diag_nlsk, skb, nlh, &c); - } + struct netlink_dump_control c = { + .start = inet_diag_dump_start_compat, + .done = inet_diag_dump_done, + .dump = inet_diag_dump_compat, + }; + return netlink_dump_start(net->diag_nlsk, skb, nlh, &c); } return inet_diag_get_exact_compat(skb, nlh); @@ -1153,25 +1334,16 @@ static int inet_diag_handler_cmd(struct sk_buff *skb, struct nlmsghdr *h) if (h->nlmsg_type == SOCK_DIAG_BY_FAMILY && h->nlmsg_flags & NLM_F_DUMP) { - if (nlmsg_attrlen(h, hdrlen)) { - struct nlattr *attr; - int err; - - attr = nlmsg_find_attr(h, hdrlen, - INET_DIAG_REQ_BYTECODE); - err = inet_diag_bc_audit(attr, skb); - if (err) - return err; - } - { - struct netlink_dump_control c = { - .dump = inet_diag_dump, - }; - return netlink_dump_start(net->diag_nlsk, skb, h, &c); - } + struct netlink_dump_control c = { + .start = inet_diag_dump_start, + .done = inet_diag_dump_done, + .dump = inet_diag_dump, + }; + return netlink_dump_start(net->diag_nlsk, skb, h, &c); } - return inet_diag_cmd_exact(h->nlmsg_type, skb, h, nlmsg_data(h)); + return inet_diag_cmd_exact(h->nlmsg_type, skb, h, hdrlen, + nlmsg_data(h)); } static diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 10d31733297d..c9f9ac5013a7 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -145,12 +145,16 @@ static void inet_frags_free_cb(void *ptr, void *arg) inet_frag_destroy(fq); } -static void fqdir_work_fn(struct work_struct *work) +static LLIST_HEAD(fqdir_free_list); + +static void fqdir_free_fn(struct work_struct *work) { - struct fqdir *fqdir = container_of(work, struct fqdir, destroy_work); - struct inet_frags *f = fqdir->f; + struct llist_node *kill_list; + struct fqdir *fqdir, *tmp; + struct inet_frags *f; - rhashtable_free_and_destroy(&fqdir->rhashtable, inet_frags_free_cb, NULL); + /* Atomically snapshot the list of fqdirs to free */ + kill_list = llist_del_all(&fqdir_free_list); /* We need to make sure all ongoing call_rcu(..., inet_frag_destroy_rcu) * have completed, since they need to dereference fqdir. @@ -158,10 +162,25 @@ static void fqdir_work_fn(struct work_struct *work) */ rcu_barrier(); - if (refcount_dec_and_test(&f->refcnt)) - complete(&f->completion); + llist_for_each_entry_safe(fqdir, tmp, kill_list, free_list) { + f = fqdir->f; + if (refcount_dec_and_test(&f->refcnt)) + complete(&f->completion); - kfree(fqdir); + kfree(fqdir); + } +} + +static DECLARE_WORK(fqdir_free_work, fqdir_free_fn); + +static void fqdir_work_fn(struct work_struct *work) +{ + struct fqdir *fqdir = container_of(work, struct fqdir, destroy_work); + + rhashtable_free_and_destroy(&fqdir->rhashtable, inet_frags_free_cb, NULL); + + if (llist_add(&fqdir->free_list, &fqdir_free_list)) + queue_work(system_wq, &fqdir_free_work); } int fqdir_init(struct fqdir **fqdirp, struct inet_frags *f, struct net *net) @@ -184,10 +203,22 @@ int fqdir_init(struct fqdir **fqdirp, struct inet_frags *f, struct net *net) } EXPORT_SYMBOL(fqdir_init); +static struct workqueue_struct *inet_frag_wq; + +static int __init inet_frag_wq_init(void) +{ + inet_frag_wq = create_workqueue("inet_frag_wq"); + if (!inet_frag_wq) + panic("Could not create inet frag workq"); + return 0; +} + +pure_initcall(inet_frag_wq_init); + void fqdir_exit(struct fqdir *fqdir) { INIT_WORK(&fqdir->destroy_work, fqdir_work_fn); - queue_work(system_wq, &fqdir->destroy_work); + queue_work(inet_frag_wq, &fqdir->destroy_work); } EXPORT_SYMBOL(fqdir_exit); @@ -204,9 +235,9 @@ void inet_frag_kill(struct inet_frag_queue *fq) /* The RCU read lock provides a memory barrier * guaranteeing that if fqdir->dead is false then * the hash table destruction will not start until - * after we unlock. Paired with inet_frags_exit_net(). + * after we unlock. Paired with fqdir_pre_exit(). */ - if (!fqdir->dead) { + if (!READ_ONCE(fqdir->dead)) { rhashtable_remove_fast(&fqdir->rhashtable, &fq->node, fqdir->f->rhash_params); refcount_dec(&fq->refcnt); @@ -321,9 +352,11 @@ static struct inet_frag_queue *inet_frag_create(struct fqdir *fqdir, /* TODO : call from rcu_read_lock() and no longer use refcount_inc_not_zero() */ struct inet_frag_queue *inet_frag_find(struct fqdir *fqdir, void *key) { + /* This pairs with WRITE_ONCE() in fqdir_pre_exit(). */ + long high_thresh = READ_ONCE(fqdir->high_thresh); struct inet_frag_queue *fq = NULL, *prev; - if (!fqdir->high_thresh || frag_mem_limit(fqdir) > fqdir->high_thresh) + if (!high_thresh || frag_mem_limit(fqdir) > high_thresh) return NULL; rcu_read_lock(); @@ -477,7 +510,7 @@ EXPORT_SYMBOL(inet_frag_reasm_prepare); void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head, void *reasm_data, bool try_coalesce) { - struct sk_buff **nextp = (struct sk_buff **)reasm_data; + struct sk_buff **nextp = reasm_data; struct rb_node *rbn; struct sk_buff *fp; int sum_truesize; @@ -539,6 +572,7 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head, skb_mark_not_on_list(head); head->prev = NULL; head->tstamp = q->stamp; + head->mono_delivery_time = q->mono_delivery_time; } EXPORT_SYMBOL(inet_frag_reasm_finish); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 2bbaaf0c7176..d3dc28156622 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -20,6 +20,9 @@ #include <net/addrconf.h> #include <net/inet_connection_sock.h> #include <net/inet_hashtables.h> +#if IS_ENABLED(CONFIG_IPV6) +#include <net/inet6_hashtables.h> +#endif #include <net/secure_seq.h> #include <net/ip.h> #include <net/tcp.h> @@ -89,12 +92,79 @@ void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket } } +bool inet_bind_bucket_match(const struct inet_bind_bucket *tb, const struct net *net, + unsigned short port, int l3mdev) +{ + return net_eq(ib_net(tb), net) && tb->port == port && + tb->l3mdev == l3mdev; +} + +static void inet_bind2_bucket_init(struct inet_bind2_bucket *tb, + struct net *net, + struct inet_bind_hashbucket *head, + unsigned short port, int l3mdev, + const struct sock *sk) +{ + write_pnet(&tb->ib_net, net); + tb->l3mdev = l3mdev; + tb->port = port; +#if IS_ENABLED(CONFIG_IPV6) + tb->family = sk->sk_family; + if (sk->sk_family == AF_INET6) + tb->v6_rcv_saddr = sk->sk_v6_rcv_saddr; + else +#endif + tb->rcv_saddr = sk->sk_rcv_saddr; + INIT_HLIST_HEAD(&tb->owners); + hlist_add_head(&tb->node, &head->chain); +} + +struct inet_bind2_bucket *inet_bind2_bucket_create(struct kmem_cache *cachep, + struct net *net, + struct inet_bind_hashbucket *head, + unsigned short port, + int l3mdev, + const struct sock *sk) +{ + struct inet_bind2_bucket *tb = kmem_cache_alloc(cachep, GFP_ATOMIC); + + if (tb) + inet_bind2_bucket_init(tb, net, head, port, l3mdev, sk); + + return tb; +} + +/* Caller must hold hashbucket lock for this tb with local BH disabled */ +void inet_bind2_bucket_destroy(struct kmem_cache *cachep, struct inet_bind2_bucket *tb) +{ + if (hlist_empty(&tb->owners)) { + __hlist_del(&tb->node); + kmem_cache_free(cachep, tb); + } +} + +static bool inet_bind2_bucket_addr_match(const struct inet_bind2_bucket *tb2, + const struct sock *sk) +{ +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family != tb2->family) + return false; + + if (sk->sk_family == AF_INET6) + return ipv6_addr_equal(&tb2->v6_rcv_saddr, + &sk->sk_v6_rcv_saddr); +#endif + return tb2->rcv_saddr == sk->sk_rcv_saddr; +} + void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb, - const unsigned short snum) + struct inet_bind2_bucket *tb2, unsigned short port) { - inet_sk(sk)->inet_num = snum; + inet_sk(sk)->inet_num = port; sk_add_bind_node(sk, &tb->owners); inet_csk(sk)->icsk_bind_hash = tb; + sk_add_bind2_node(sk, &tb2->owners); + inet_csk(sk)->icsk_bind2_hash = tb2; } /* @@ -102,11 +172,15 @@ void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb, */ static void __inet_put_port(struct sock *sk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; - const int bhash = inet_bhashfn(sock_net(sk), inet_sk(sk)->inet_num, - hashinfo->bhash_size); - struct inet_bind_hashbucket *head = &hashinfo->bhash[bhash]; + struct inet_hashinfo *hashinfo = tcp_or_dccp_get_hashinfo(sk); + struct inet_bind_hashbucket *head, *head2; + struct net *net = sock_net(sk); struct inet_bind_bucket *tb; + int bhash; + + bhash = inet_bhashfn(net, inet_sk(sk)->inet_num, hashinfo->bhash_size); + head = &hashinfo->bhash[bhash]; + head2 = inet_bhashfn_portaddr(hashinfo, sk, net, inet_sk(sk)->inet_num); spin_lock(&head->lock); tb = inet_csk(sk)->icsk_bind_hash; @@ -114,6 +188,17 @@ static void __inet_put_port(struct sock *sk) inet_csk(sk)->icsk_bind_hash = NULL; inet_sk(sk)->inet_num = 0; inet_bind_bucket_destroy(hashinfo->bind_bucket_cachep, tb); + + spin_lock(&head2->lock); + if (inet_csk(sk)->icsk_bind2_hash) { + struct inet_bind2_bucket *tb2 = inet_csk(sk)->icsk_bind2_hash; + + __sk_del_bind2_node(sk); + inet_csk(sk)->icsk_bind2_hash = NULL; + inet_bind2_bucket_destroy(hashinfo->bind2_bucket_cachep, tb2); + } + spin_unlock(&head2->lock); + spin_unlock(&head->lock); } @@ -127,17 +212,26 @@ EXPORT_SYMBOL(inet_put_port); int __inet_inherit_port(const struct sock *sk, struct sock *child) { - struct inet_hashinfo *table = sk->sk_prot->h.hashinfo; + struct inet_hashinfo *table = tcp_or_dccp_get_hashinfo(sk); unsigned short port = inet_sk(child)->inet_num; - const int bhash = inet_bhashfn(sock_net(sk), port, - table->bhash_size); - struct inet_bind_hashbucket *head = &table->bhash[bhash]; + struct inet_bind_hashbucket *head, *head2; + bool created_inet_bind_bucket = false; + struct net *net = sock_net(sk); + bool update_fastreuse = false; + struct inet_bind2_bucket *tb2; struct inet_bind_bucket *tb; - int l3mdev; + int bhash, l3mdev; + + bhash = inet_bhashfn(net, port, table->bhash_size); + head = &table->bhash[bhash]; + head2 = inet_bhashfn_portaddr(table, child, net, port); spin_lock(&head->lock); + spin_lock(&head2->lock); tb = inet_csk(sk)->icsk_bind_hash; - if (unlikely(!tb)) { + tb2 = inet_csk(sk)->icsk_bind2_hash; + if (unlikely(!tb || !tb2)) { + spin_unlock(&head2->lock); spin_unlock(&head->lock); return -ENOENT; } @@ -150,24 +244,49 @@ int __inet_inherit_port(const struct sock *sk, struct sock *child) * as that of the child socket. We have to look up or * create a new bind bucket for the child here. */ inet_bind_bucket_for_each(tb, &head->chain) { - if (net_eq(ib_net(tb), sock_net(sk)) && - tb->l3mdev == l3mdev && tb->port == port) + if (inet_bind_bucket_match(tb, net, port, l3mdev)) break; } if (!tb) { tb = inet_bind_bucket_create(table->bind_bucket_cachep, - sock_net(sk), head, port, - l3mdev); + net, head, port, l3mdev); if (!tb) { + spin_unlock(&head2->lock); spin_unlock(&head->lock); return -ENOMEM; } + created_inet_bind_bucket = true; + } + update_fastreuse = true; + + goto bhash2_find; + } else if (!inet_bind2_bucket_addr_match(tb2, child)) { + l3mdev = inet_sk_bound_l3mdev(sk); + +bhash2_find: + tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, child); + if (!tb2) { + tb2 = inet_bind2_bucket_create(table->bind2_bucket_cachep, + net, head2, port, + l3mdev, child); + if (!tb2) + goto error; } } - inet_bind_hash(child, tb, port); + if (update_fastreuse) + inet_csk_update_fastreuse(tb, child); + inet_bind_hash(child, tb, tb2, port); + spin_unlock(&head2->lock); spin_unlock(&head->lock); return 0; + +error: + if (created_inet_bind_bucket) + inet_bind_bucket_destroy(table->bind_bucket_cachep, tb); + spin_unlock(&head2->lock); + spin_unlock(&head->lock); + return -ENOMEM; } EXPORT_SYMBOL_GPL(__inet_inherit_port); @@ -189,45 +308,9 @@ inet_lhash2_bucket_sk(struct inet_hashinfo *h, struct sock *sk) return inet_lhash2_bucket(h, hash); } -static void inet_hash2(struct inet_hashinfo *h, struct sock *sk) -{ - struct inet_listen_hashbucket *ilb2; - - if (!h->lhash2) - return; - - ilb2 = inet_lhash2_bucket_sk(h, sk); - - spin_lock(&ilb2->lock); - if (sk->sk_reuseport && sk->sk_family == AF_INET6) - hlist_add_tail_rcu(&inet_csk(sk)->icsk_listen_portaddr_node, - &ilb2->head); - else - hlist_add_head_rcu(&inet_csk(sk)->icsk_listen_portaddr_node, - &ilb2->head); - ilb2->count++; - spin_unlock(&ilb2->lock); -} - -static void inet_unhash2(struct inet_hashinfo *h, struct sock *sk) -{ - struct inet_listen_hashbucket *ilb2; - - if (!h->lhash2 || - WARN_ON_ONCE(hlist_unhashed(&inet_csk(sk)->icsk_listen_portaddr_node))) - return; - - ilb2 = inet_lhash2_bucket_sk(h, sk); - - spin_lock(&ilb2->lock); - hlist_del_init_rcu(&inet_csk(sk)->icsk_listen_portaddr_node); - ilb2->count--; - spin_unlock(&ilb2->lock); -} - static inline int compute_score(struct sock *sk, struct net *net, const unsigned short hnum, const __be32 daddr, - const int dif, const int sdif, bool exact_dif) + const int dif, const int sdif) { int score = -1; @@ -238,14 +321,31 @@ static inline int compute_score(struct sock *sk, struct net *net, if (!inet_sk_bound_dev_eq(net, sk->sk_bound_dev_if, dif, sdif)) return -1; + score = sk->sk_bound_dev_if ? 2 : 1; - score = sk->sk_family == PF_INET ? 2 : 1; + if (sk->sk_family == PF_INET) + score++; if (READ_ONCE(sk->sk_incoming_cpu) == raw_smp_processor_id()) score++; } return score; } +static inline struct sock *lookup_reuseport(struct net *net, struct sock *sk, + struct sk_buff *skb, int doff, + __be32 saddr, __be16 sport, + __be32 daddr, unsigned short hnum) +{ + struct sock *reuse_sk = NULL; + u32 phash; + + if (sk->sk_reuseport) { + phash = inet_ehashfn(net, daddr, hnum, saddr, sport); + reuse_sk = reuseport_select_sock(sk, phash, skb, doff); + } + return reuse_sk; +} + /* * Here are some nice properties to exploit here. The BSD API * does not allow a listening sock to specify the remote port nor the @@ -261,25 +361,18 @@ static struct sock *inet_lhash2_lookup(struct net *net, const __be32 daddr, const unsigned short hnum, const int dif, const int sdif) { - bool exact_dif = inet_exact_dif_match(net, skb); - struct inet_connection_sock *icsk; struct sock *sk, *result = NULL; + struct hlist_nulls_node *node; int score, hiscore = 0; - u32 phash = 0; - inet_lhash2_for_each_icsk_rcu(icsk, &ilb2->head) { - sk = (struct sock *)icsk; - score = compute_score(sk, net, hnum, daddr, - dif, sdif, exact_dif); + sk_nulls_for_each_rcu(sk, node, &ilb2->nulls_head) { + score = compute_score(sk, net, hnum, daddr, dif, sdif); if (score > hiscore) { - if (sk->sk_reuseport) { - phash = inet_ehashfn(net, daddr, hnum, - saddr, sport); - result = reuseport_select_sock(sk, phash, - skb, doff); - if (result) - return result; - } + result = lookup_reuseport(net, sk, skb, doff, + saddr, sport, daddr, hnum); + if (result) + return result; + result = sk; hiscore = score; } @@ -288,6 +381,29 @@ static struct sock *inet_lhash2_lookup(struct net *net, return result; } +static inline struct sock *inet_lookup_run_bpf(struct net *net, + struct inet_hashinfo *hashinfo, + struct sk_buff *skb, int doff, + __be32 saddr, __be16 sport, + __be32 daddr, u16 hnum, const int dif) +{ + struct sock *sk, *reuse_sk; + bool no_reuseport; + + if (hashinfo != net->ipv4.tcp_death_row.hashinfo) + return NULL; /* only TCP is supported */ + + no_reuseport = bpf_sk_lookup_run_v4(net, IPPROTO_TCP, saddr, sport, + daddr, hnum, dif, &sk); + if (no_reuseport || IS_ERR_OR_NULL(sk)) + return sk; + + reuse_sk = lookup_reuseport(net, sk, skb, doff, saddr, sport, daddr, hnum); + if (reuse_sk) + sk = reuse_sk; + return sk; +} + struct sock *__inet_lookup_listener(struct net *net, struct inet_hashinfo *hashinfo, struct sk_buff *skb, int doff, @@ -299,6 +415,14 @@ struct sock *__inet_lookup_listener(struct net *net, struct sock *result = NULL; unsigned int hash2; + /* Lookup redirect from BPF */ + if (static_branch_unlikely(&bpf_sk_lookup_enabled)) { + result = inet_lookup_run_bpf(net, hashinfo, skb, doff, + saddr, sport, daddr, hnum, dif); + if (result) + goto done; + } + hash2 = ipv4_portaddr_hash(net, daddr, hnum); ilb2 = inet_lhash2_bucket(hashinfo, hash2); @@ -364,13 +488,11 @@ begin: sk_nulls_for_each_rcu(sk, node, &head->chain) { if (sk->sk_hash != hash) continue; - if (likely(INET_MATCH(sk, net, acookie, - saddr, daddr, ports, dif, sdif))) { + if (likely(inet_match(net, sk, acookie, ports, dif, sdif))) { if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt))) goto out; - if (unlikely(!INET_MATCH(sk, net, acookie, - saddr, daddr, ports, - dif, sdif))) { + if (unlikely(!inet_match(net, sk, acookie, + ports, dif, sdif))) { sock_gen_put(sk); goto begin; } @@ -419,8 +541,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, if (sk2->sk_hash != hash) continue; - if (likely(INET_MATCH(sk2, net, acookie, - saddr, daddr, ports, dif, sdif))) { + if (likely(inet_match(net, sk2, acookie, ports, dif, sdif))) { if (sk2->sk_state == TCP_TIME_WAIT) { tw = inet_twsk(sk2); if (twsk_unique(sk, sk2, twp)) @@ -458,7 +579,7 @@ not_unique: return -EADDRNOTAVAIL; } -static u32 inet_sk_port_offset(const struct sock *sk) +static u64 inet_sk_port_offset(const struct sock *sk) { const struct inet_sock *inet = inet_sk(sk); @@ -467,14 +588,54 @@ static u32 inet_sk_port_offset(const struct sock *sk) inet->inet_dport); } -/* insert a socket into ehash, and eventually remove another one - * (The another one can be a SYN_RECV or TIMEWAIT +/* Searches for an exsiting socket in the ehash bucket list. + * Returns true if found, false otherwise. */ -bool inet_ehash_insert(struct sock *sk, struct sock *osk) +static bool inet_ehash_lookup_by_sk(struct sock *sk, + struct hlist_nulls_head *list) { - struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; - struct hlist_nulls_head *list; + const __portpair ports = INET_COMBINED_PORTS(sk->sk_dport, sk->sk_num); + const int sdif = sk->sk_bound_dev_if; + const int dif = sk->sk_bound_dev_if; + const struct hlist_nulls_node *node; + struct net *net = sock_net(sk); + struct sock *esk; + + INET_ADDR_COOKIE(acookie, sk->sk_daddr, sk->sk_rcv_saddr); + + sk_nulls_for_each_rcu(esk, node, list) { + if (esk->sk_hash != sk->sk_hash) + continue; + if (sk->sk_family == AF_INET) { + if (unlikely(inet_match(net, esk, acookie, + ports, dif, sdif))) { + return true; + } + } +#if IS_ENABLED(CONFIG_IPV6) + else if (sk->sk_family == AF_INET6) { + if (unlikely(inet6_match(net, esk, + &sk->sk_v6_daddr, + &sk->sk_v6_rcv_saddr, + ports, dif, sdif))) { + return true; + } + } +#endif + } + return false; +} + +/* Insert a socket into ehash, and eventually remove another one + * (The another one can be a SYN_RECV or TIMEWAIT) + * If an existing socket already exists, socket sk is not inserted, + * and sets found_dup_sk parameter to true. + */ +bool inet_ehash_insert(struct sock *sk, struct sock *osk, bool *found_dup_sk) +{ + struct inet_hashinfo *hashinfo = tcp_or_dccp_get_hashinfo(sk); struct inet_ehash_bucket *head; + struct hlist_nulls_head *list; spinlock_t *lock; bool ret = true; @@ -489,21 +650,28 @@ bool inet_ehash_insert(struct sock *sk, struct sock *osk) if (osk) { WARN_ON_ONCE(sk->sk_hash != osk->sk_hash); ret = sk_nulls_del_node_init_rcu(osk); + } else if (found_dup_sk) { + *found_dup_sk = inet_ehash_lookup_by_sk(sk, list); + if (*found_dup_sk) + ret = false; } + if (ret) __sk_nulls_add_node_rcu(sk, list); + spin_unlock(lock); + return ret; } -bool inet_ehash_nolisten(struct sock *sk, struct sock *osk) +bool inet_ehash_nolisten(struct sock *sk, struct sock *osk, bool *found_dup_sk) { - bool ok = inet_ehash_insert(sk, osk); + bool ok = inet_ehash_insert(sk, osk, found_dup_sk); if (ok) { sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); } else { - percpu_counter_inc(sk->sk_prot->orphan_count); + this_cpu_inc(*sk->sk_prot->orphan_count); inet_sk_set_state(sk, TCP_CLOSE); sock_set_flag(sk, SOCK_DEAD); inet_csk_destroy_sock(sk); @@ -537,34 +705,34 @@ static int inet_reuseport_add_sock(struct sock *sk, int __inet_hash(struct sock *sk, struct sock *osk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; - struct inet_listen_hashbucket *ilb; + struct inet_hashinfo *hashinfo = tcp_or_dccp_get_hashinfo(sk); + struct inet_listen_hashbucket *ilb2; int err = 0; if (sk->sk_state != TCP_LISTEN) { - inet_ehash_nolisten(sk, osk); + local_bh_disable(); + inet_ehash_nolisten(sk, osk, NULL); + local_bh_enable(); return 0; } WARN_ON(!sk_unhashed(sk)); - ilb = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)]; + ilb2 = inet_lhash2_bucket_sk(hashinfo, sk); - spin_lock(&ilb->lock); + spin_lock(&ilb2->lock); if (sk->sk_reuseport) { - err = inet_reuseport_add_sock(sk, ilb); + err = inet_reuseport_add_sock(sk, ilb2); if (err) goto unlock; } if (IS_ENABLED(CONFIG_IPV6) && sk->sk_reuseport && sk->sk_family == AF_INET6) - __sk_nulls_add_node_tail_rcu(sk, &ilb->nulls_head); + __sk_nulls_add_node_tail_rcu(sk, &ilb2->nulls_head); else - __sk_nulls_add_node_rcu(sk, &ilb->nulls_head); - inet_hash2(hashinfo, sk); - ilb->count++; + __sk_nulls_add_node_rcu(sk, &ilb2->nulls_head); sock_set_flag(sk, SOCK_RCU_FREE); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); unlock: - spin_unlock(&ilb->lock); + spin_unlock(&ilb2->lock); return err; } @@ -574,11 +742,8 @@ int inet_hash(struct sock *sk) { int err = 0; - if (sk->sk_state != TCP_CLOSE) { - local_bh_disable(); + if (sk->sk_state != TCP_CLOSE) err = __inet_hash(sk, NULL); - local_bh_enable(); - } return err; } @@ -586,51 +751,187 @@ EXPORT_SYMBOL_GPL(inet_hash); void inet_unhash(struct sock *sk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; - struct inet_listen_hashbucket *ilb = NULL; - spinlock_t *lock; + struct inet_hashinfo *hashinfo = tcp_or_dccp_get_hashinfo(sk); if (sk_unhashed(sk)) return; if (sk->sk_state == TCP_LISTEN) { - ilb = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)]; - lock = &ilb->lock; + struct inet_listen_hashbucket *ilb2; + + ilb2 = inet_lhash2_bucket_sk(hashinfo, sk); + /* Don't disable bottom halves while acquiring the lock to + * avoid circular locking dependency on PREEMPT_RT. + */ + spin_lock(&ilb2->lock); + if (sk_unhashed(sk)) { + spin_unlock(&ilb2->lock); + return; + } + + if (rcu_access_pointer(sk->sk_reuseport_cb)) + reuseport_stop_listen_sock(sk); + + __sk_nulls_del_node_init_rcu(sk); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); + spin_unlock(&ilb2->lock); } else { - lock = inet_ehash_lockp(hashinfo, sk->sk_hash); - } - spin_lock_bh(lock); - if (sk_unhashed(sk)) - goto unlock; + spinlock_t *lock = inet_ehash_lockp(hashinfo, sk->sk_hash); - if (rcu_access_pointer(sk->sk_reuseport_cb)) - reuseport_detach_sock(sk); - if (ilb) { - inet_unhash2(hashinfo, sk); - ilb->count--; + spin_lock_bh(lock); + if (sk_unhashed(sk)) { + spin_unlock_bh(lock); + return; + } + __sk_nulls_del_node_init_rcu(sk); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); + spin_unlock_bh(lock); } - __sk_nulls_del_node_init_rcu(sk); - sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); -unlock: - spin_unlock_bh(lock); } EXPORT_SYMBOL_GPL(inet_unhash); +static bool inet_bind2_bucket_match(const struct inet_bind2_bucket *tb, + const struct net *net, unsigned short port, + int l3mdev, const struct sock *sk) +{ +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family != tb->family) + return false; + + if (sk->sk_family == AF_INET6) + return net_eq(ib2_net(tb), net) && tb->port == port && + tb->l3mdev == l3mdev && + ipv6_addr_equal(&tb->v6_rcv_saddr, &sk->sk_v6_rcv_saddr); + else +#endif + return net_eq(ib2_net(tb), net) && tb->port == port && + tb->l3mdev == l3mdev && tb->rcv_saddr == sk->sk_rcv_saddr; +} + +bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const struct net *net, + unsigned short port, int l3mdev, const struct sock *sk) +{ +#if IS_ENABLED(CONFIG_IPV6) + struct in6_addr addr_any = {}; + + if (sk->sk_family != tb->family) + return false; + + if (sk->sk_family == AF_INET6) + return net_eq(ib2_net(tb), net) && tb->port == port && + tb->l3mdev == l3mdev && + ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any); + else +#endif + return net_eq(ib2_net(tb), net) && tb->port == port && + tb->l3mdev == l3mdev && tb->rcv_saddr == 0; +} + +/* The socket's bhash2 hashbucket spinlock must be held when this is called */ +struct inet_bind2_bucket * +inet_bind2_bucket_find(const struct inet_bind_hashbucket *head, const struct net *net, + unsigned short port, int l3mdev, const struct sock *sk) +{ + struct inet_bind2_bucket *bhash2 = NULL; + + inet_bind_bucket_for_each(bhash2, &head->chain) + if (inet_bind2_bucket_match(bhash2, net, port, l3mdev, sk)) + break; + + return bhash2; +} + +struct inet_bind_hashbucket * +inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, int port) +{ + struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk); + u32 hash; +#if IS_ENABLED(CONFIG_IPV6) + struct in6_addr addr_any = {}; + + if (sk->sk_family == AF_INET6) + hash = ipv6_portaddr_hash(net, &addr_any, port); + else +#endif + hash = ipv4_portaddr_hash(net, 0, port); + + return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)]; +} + +int inet_bhash2_update_saddr(struct inet_bind_hashbucket *prev_saddr, struct sock *sk) +{ + struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk); + struct inet_bind2_bucket *tb2, *new_tb2; + int l3mdev = inet_sk_bound_l3mdev(sk); + struct inet_bind_hashbucket *head2; + int port = inet_sk(sk)->inet_num; + struct net *net = sock_net(sk); + + /* Allocate a bind2 bucket ahead of time to avoid permanently putting + * the bhash2 table in an inconsistent state if a new tb2 bucket + * allocation fails. + */ + new_tb2 = kmem_cache_alloc(hinfo->bind2_bucket_cachep, GFP_ATOMIC); + if (!new_tb2) + return -ENOMEM; + + head2 = inet_bhashfn_portaddr(hinfo, sk, net, port); + + if (prev_saddr) { + spin_lock_bh(&prev_saddr->lock); + __sk_del_bind2_node(sk); + inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, + inet_csk(sk)->icsk_bind2_hash); + spin_unlock_bh(&prev_saddr->lock); + } + + spin_lock_bh(&head2->lock); + tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk); + if (!tb2) { + tb2 = new_tb2; + inet_bind2_bucket_init(tb2, net, head2, port, l3mdev, sk); + } + sk_add_bind2_node(sk, &tb2->owners); + inet_csk(sk)->icsk_bind2_hash = tb2; + spin_unlock_bh(&head2->lock); + + if (tb2 != new_tb2) + kmem_cache_free(hinfo->bind2_bucket_cachep, new_tb2); + + return 0; +} +EXPORT_SYMBOL_GPL(inet_bhash2_update_saddr); + +/* RFC 6056 3.3.4. Algorithm 4: Double-Hash Port Selection Algorithm + * Note that we use 32bit integers (vs RFC 'short integers') + * because 2^16 is not a multiple of num_ephemeral and this + * property might be used by clever attacker. + * RFC claims using TABLE_LENGTH=10 buckets gives an improvement, though + * attacks were since demonstrated, thus we use 65536 instead to really + * give more isolation and privacy, at the expense of 256kB of kernel + * memory. + */ +#define INET_TABLE_PERTURB_SHIFT 16 +#define INET_TABLE_PERTURB_SIZE (1 << INET_TABLE_PERTURB_SHIFT) +static u32 *table_perturb; + int __inet_hash_connect(struct inet_timewait_death_row *death_row, - struct sock *sk, u32 port_offset, + struct sock *sk, u64 port_offset, int (*check_established)(struct inet_timewait_death_row *, struct sock *, __u16, struct inet_timewait_sock **)) { struct inet_hashinfo *hinfo = death_row->hashinfo; + struct inet_bind_hashbucket *head, *head2; struct inet_timewait_sock *tw = NULL; - struct inet_bind_hashbucket *head; int port = inet_sk(sk)->inet_num; struct net *net = sock_net(sk); + struct inet_bind2_bucket *tb2; struct inet_bind_bucket *tb; + bool tb_created = false; u32 remaining, offset; int ret, i, low, high; - static u32 hint; int l3mdev; + u32 index; if (port) { head = &hinfo->bhash[inet_bhashfn(net, port, @@ -638,7 +939,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, tb = inet_csk(sk)->icsk_bind_hash; spin_lock_bh(&head->lock); if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) { - inet_ehash_nolisten(sk, NULL); + inet_ehash_nolisten(sk, NULL, NULL); spin_unlock_bh(&head->lock); return 0; } @@ -657,7 +958,13 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, if (likely(remaining > 1)) remaining &= ~1U; - offset = (hint + port_offset) % remaining; + get_random_sleepable_once(table_perturb, + INET_TABLE_PERTURB_SIZE * sizeof(*table_perturb)); + index = port_offset & (INET_TABLE_PERTURB_SIZE - 1); + + offset = READ_ONCE(table_perturb[index]) + (port_offset >> 32); + offset %= remaining; + /* In first pass we try ports of @low parity. * inet_csk_get_port() does the opposite choice. */ @@ -677,8 +984,7 @@ other_parity_scan: * the established check is already unique enough. */ inet_bind_bucket_for_each(tb, &head->chain) { - if (net_eq(ib_net(tb), net) && tb->l3mdev == l3mdev && - tb->port == port) { + if (inet_bind_bucket_match(tb, net, port, l3mdev)) { if (tb->fastreuse >= 0 || tb->fastreuseport >= 0) goto next_port; @@ -696,6 +1002,7 @@ other_parity_scan: spin_unlock_bh(&head->lock); return -ENOMEM; } + tb_created = true; tb->fastreuse = -1; tb->fastreuseport = -1; goto ok; @@ -711,13 +1018,36 @@ next_port: return -EADDRNOTAVAIL; ok: - hint += i + 2; + /* Find the corresponding tb2 bucket since we need to + * add the socket to the bhash2 table as well + */ + head2 = inet_bhashfn_portaddr(hinfo, sk, net, port); + spin_lock(&head2->lock); + + tb2 = inet_bind2_bucket_find(head2, net, port, l3mdev, sk); + if (!tb2) { + tb2 = inet_bind2_bucket_create(hinfo->bind2_bucket_cachep, net, + head2, port, l3mdev, sk); + if (!tb2) + goto error; + } + + /* Here we want to add a little bit of randomness to the next source + * port that will be chosen. We use a max() with a random here so that + * on low contention the randomness is maximal and on high contention + * it may be inexistent. + */ + i = max_t(int, i, prandom_u32_max(8) * 2); + WRITE_ONCE(table_perturb[index], READ_ONCE(table_perturb[index]) + i + 2); /* Head lock still held and bh's disabled */ - inet_bind_hash(sk, tb, port); + inet_bind_hash(sk, tb, tb2, port); + + spin_unlock(&head2->lock); + if (sk_unhashed(sk)) { inet_sk(sk)->inet_sport = htons(port); - inet_ehash_nolisten(sk, (struct sock *)tw); + inet_ehash_nolisten(sk, (struct sock *)tw, NULL); } if (tw) inet_twsk_bind_unhash(tw, hinfo); @@ -726,6 +1056,13 @@ ok: inet_twsk_deschedule_put(tw); local_bh_enable(); return 0; + +error: + spin_unlock(&head2->lock); + if (tb_created) + inet_bind_bucket_destroy(hinfo->bind_bucket_cachep, tb); + spin_unlock_bh(&head->lock); + return -ENOMEM; } /* @@ -734,7 +1071,7 @@ ok: int inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk) { - u32 port_offset = 0; + u64 port_offset = 0; if (!inet_sk(sk)->inet_num) port_offset = inet_sk_port_offset(sk); @@ -743,29 +1080,14 @@ int inet_hash_connect(struct inet_timewait_death_row *death_row, } EXPORT_SYMBOL_GPL(inet_hash_connect); -void inet_hashinfo_init(struct inet_hashinfo *h) -{ - int i; - - for (i = 0; i < INET_LHTABLE_SIZE; i++) { - spin_lock_init(&h->listening_hash[i].lock); - INIT_HLIST_NULLS_HEAD(&h->listening_hash[i].nulls_head, - i + LISTENING_NULLS_BASE); - h->listening_hash[i].count = 0; - } - - h->lhash2 = NULL; -} -EXPORT_SYMBOL_GPL(inet_hashinfo_init); - static void init_hashinfo_lhash2(struct inet_hashinfo *h) { int i; for (i = 0; i <= h->lhash2_mask; i++) { spin_lock_init(&h->lhash2[i].lock); - INIT_HLIST_HEAD(&h->lhash2[i].head); - h->lhash2[i].count = 0; + INIT_HLIST_NULLS_HEAD(&h->lhash2[i].nulls_head, + i + LISTENING_NULLS_BASE); } } @@ -784,6 +1106,14 @@ void __init inet_hashinfo2_init(struct inet_hashinfo *h, const char *name, low_limit, high_limit); init_hashinfo_lhash2(h); + + /* this one is used for source ports of outgoing connections */ + table_perturb = alloc_large_system_hash("Table-perturb", + sizeof(*table_perturb), + INET_TABLE_PERTURB_SIZE, + 0, 0, NULL, NULL, + INET_TABLE_PERTURB_SIZE, + INET_TABLE_PERTURB_SIZE); } int inet_hashinfo2_init_mod(struct inet_hashinfo *h) @@ -825,3 +1155,50 @@ int inet_ehash_locks_alloc(struct inet_hashinfo *hashinfo) return 0; } EXPORT_SYMBOL_GPL(inet_ehash_locks_alloc); + +struct inet_hashinfo *inet_pernet_hashinfo_alloc(struct inet_hashinfo *hashinfo, + unsigned int ehash_entries) +{ + struct inet_hashinfo *new_hashinfo; + int i; + + new_hashinfo = kmemdup(hashinfo, sizeof(*hashinfo), GFP_KERNEL); + if (!new_hashinfo) + goto err; + + new_hashinfo->ehash = vmalloc_huge(ehash_entries * sizeof(struct inet_ehash_bucket), + GFP_KERNEL_ACCOUNT); + if (!new_hashinfo->ehash) + goto free_hashinfo; + + new_hashinfo->ehash_mask = ehash_entries - 1; + + if (inet_ehash_locks_alloc(new_hashinfo)) + goto free_ehash; + + for (i = 0; i < ehash_entries; i++) + INIT_HLIST_NULLS_HEAD(&new_hashinfo->ehash[i].chain, i); + + new_hashinfo->pernet = true; + + return new_hashinfo; + +free_ehash: + vfree(new_hashinfo->ehash); +free_hashinfo: + kfree(new_hashinfo); +err: + return NULL; +} +EXPORT_SYMBOL_GPL(inet_pernet_hashinfo_alloc); + +void inet_pernet_hashinfo_free(struct inet_hashinfo *hashinfo) +{ + if (!hashinfo->pernet) + return; + + inet_ehash_locks_free(hashinfo); + vfree(hashinfo->ehash); + kfree(hashinfo); +} +EXPORT_SYMBOL_GPL(inet_pernet_hashinfo_free); diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index c411c87ae865..66fc940f9521 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -59,7 +59,7 @@ static void inet_twsk_kill(struct inet_timewait_sock *tw) inet_twsk_bind_unhash(tw, hashinfo); spin_unlock(&bhead->lock); - atomic_dec(&tw->tw_dr->tw_count); + refcount_dec(&tw->tw_dr->tw_refcount); inet_twsk_put(tw); } @@ -145,10 +145,6 @@ static void tw_timer_handler(struct timer_list *t) { struct inet_timewait_sock *tw = from_timer(tw, t, tw_timer); - if (tw->tw_kill) - __NET_INC_STATS(twsk_net(tw), LINUX_MIB_TIMEWAITKILLED); - else - __NET_INC_STATS(twsk_net(tw), LINUX_MIB_TIMEWAITED); inet_twsk_kill(tw); } @@ -158,7 +154,8 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, { struct inet_timewait_sock *tw; - if (atomic_read(&dr->tw_count) >= dr->sysctl_max_tw_buckets) + if (refcount_read(&dr->tw_refcount) - 1 >= + READ_ONCE(dr->sysctl_max_tw_buckets)) return NULL; tw = kmem_cache_alloc(sk->sk_prot_creator->twsk_prot->twsk_slab, @@ -244,10 +241,13 @@ void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm) * of PAWS. */ - tw->tw_kill = timeo <= 4*HZ; if (!rearm) { + bool kill = timeo <= 4*HZ; + + __NET_INC_STATS(twsk_net(tw), kill ? LINUX_MIB_TIMEWAITKILLED : + LINUX_MIB_TIMEWAITED); BUG_ON(mod_timer(&tw->tw_timer, jiffies + timeo)); - atomic_inc(&tw->tw_dr->tw_count); + refcount_inc(&tw->tw_dr->tw_refcount); } else { mod_timer_pending(&tw->tw_timer, jiffies + timeo); } @@ -268,18 +268,31 @@ restart_rcu: rcu_read_lock(); restart: sk_nulls_for_each_rcu(sk, node, &head->chain) { - if (sk->sk_state != TCP_TIME_WAIT) + if (sk->sk_state != TCP_TIME_WAIT) { + /* A kernel listener socket might not hold refcnt for net, + * so reqsk_timer_handler() could be fired after net is + * freed. Userspace listener and reqsk never exist here. + */ + if (unlikely(sk->sk_state == TCP_NEW_SYN_RECV && + hashinfo->pernet)) { + struct request_sock *req = inet_reqsk(sk); + + inet_csk_reqsk_queue_drop_and_put(req->rsk_listener, req); + } + continue; + } + tw = inet_twsk(sk); if ((tw->tw_family != family) || - refcount_read(&twsk_net(tw)->count)) + refcount_read(&twsk_net(tw)->ns.count)) continue; if (unlikely(!refcount_inc_not_zero(&tw->tw_refcnt))) continue; if (unlikely((tw->tw_family != family) || - refcount_read(&twsk_net(tw)->count))) { + refcount_read(&twsk_net(tw)->ns.count))) { inet_twsk_put(tw); goto restart; } diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index ff327a62c9ce..e9fed83e9b3c 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -65,7 +65,7 @@ EXPORT_SYMBOL_GPL(inet_peer_base_init); #define PEER_MAX_GC 32 /* Exported for sysctl_net_ipv4. */ -int inet_peer_threshold __read_mostly = 65536 + 128; /* start to throw entries more +int inet_peer_threshold __read_mostly; /* start to throw entries more * aggressively at this stage */ int inet_peer_minttl __read_mostly = 120 * HZ; /* TTL under high load: 120 sec */ int inet_peer_maxttl __read_mostly = 10 * 60 * HZ; /* usual time to live: 10 min */ @@ -73,20 +73,13 @@ int inet_peer_maxttl __read_mostly = 10 * 60 * HZ; /* usual time to live: 10 min /* Called from ip_output.c:ip_init */ void __init inet_initpeers(void) { - struct sysinfo si; + u64 nr_entries; - /* Use the straight interface to information about memory. */ - si_meminfo(&si); - /* The values below were suggested by Alexey Kuznetsov - * <kuznet@ms2.inr.ac.ru>. I don't have any opinion about the values - * myself. --SAW - */ - if (si.totalram <= (32768*1024)/PAGE_SIZE) - inet_peer_threshold >>= 1; /* max pool size about 1MB on IA32 */ - if (si.totalram <= (16384*1024)/PAGE_SIZE) - inet_peer_threshold >>= 1; /* about 512KB */ - if (si.totalram <= (8192*1024)/PAGE_SIZE) - inet_peer_threshold >>= 2; /* about 128KB */ + /* 1% of physical memory */ + nr_entries = div64_ul((u64)totalram_pages() << PAGE_SHIFT, + 100 * L1_CACHE_ALIGN(sizeof(struct inet_peer))); + + inet_peer_threshold = clamp_val(nr_entries, 4096, 65536 + 128); peer_cachep = kmem_cache_create("inet_peer_cache", sizeof(struct inet_peer), @@ -148,16 +141,20 @@ static void inet_peer_gc(struct inet_peer_base *base, struct inet_peer *gc_stack[], unsigned int gc_cnt) { + int peer_threshold, peer_maxttl, peer_minttl; struct inet_peer *p; __u32 delta, ttl; int i; - if (base->total >= inet_peer_threshold) + peer_threshold = READ_ONCE(inet_peer_threshold); + peer_maxttl = READ_ONCE(inet_peer_maxttl); + peer_minttl = READ_ONCE(inet_peer_minttl); + + if (base->total >= peer_threshold) ttl = 0; /* be aggressive */ else - ttl = inet_peer_maxttl - - (inet_peer_maxttl - inet_peer_minttl) / HZ * - base->total / inet_peer_threshold * HZ; + ttl = peer_maxttl - (peer_maxttl - peer_minttl) / HZ * + base->total / peer_threshold * HZ; for (i = 0; i < gc_cnt; i++) { p = gc_stack[i]; diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 00ec819f949b..e18931a6d153 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -79,7 +79,7 @@ static int ip_forward_finish(struct net *net, struct sock *sk, struct sk_buff *s if (unlikely(opt->optlen)) ip_forward_options(skb); - skb->tstamp = 0; + skb_clear_tstamp(skb); return dst_output(net, sk, skb); } @@ -90,6 +90,7 @@ int ip_forward(struct sk_buff *skb) struct rtable *rt; /* Route we use */ struct ip_options *opt = &(IPCB(skb)->opt); struct net *net; + SKB_DR(reason); /* that should never happen */ if (skb->pkt_type != PACKET_HOST) @@ -101,8 +102,10 @@ int ip_forward(struct sk_buff *skb) if (skb_warn_if_lro(skb)) goto drop; - if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb)) + if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb)) { + SKB_DR_SET(reason, XFRM_POLICY); goto drop; + } if (IPCB(skb)->opt.router_alert && ip_call_ra_chain(skb)) return NET_RX_SUCCESS; @@ -118,8 +121,10 @@ int ip_forward(struct sk_buff *skb) if (ip_hdr(skb)->ttl <= 1) goto too_many_hops; - if (!xfrm4_route_forward(skb)) + if (!xfrm4_route_forward(skb)) { + SKB_DR_SET(reason, XFRM_POLICY); goto drop; + } rt = skb_rtable(skb); @@ -132,6 +137,7 @@ int ip_forward(struct sk_buff *skb) IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); + SKB_DR_SET(reason, PKT_TOO_BIG); goto drop; } @@ -151,7 +157,7 @@ int ip_forward(struct sk_buff *skb) !skb_sec_path(skb)) ip_rt_send_redirect(skb); - if (net->ipv4.sysctl_ip_fwd_update_priority) + if (READ_ONCE(net->ipv4.sysctl_ip_fwd_update_priority)) skb->priority = rt_tos2priority(iph->tos); return NF_HOOK(NFPROTO_IPV4, NF_INET_FORWARD, @@ -169,7 +175,8 @@ too_many_hops: /* Tell the sender its packet died... */ __IP_INC_STATS(net, IPSTATS_MIB_INHDRERRORS); icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0); + SKB_DR_SET(reason, IP_INHDR); drop: - kfree_skb(skb); + kfree_skb_reason(skb, reason); return NET_RX_DROP; } diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index cfeb8890f94e..fb153569889e 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -144,7 +144,8 @@ static void ip_expire(struct timer_list *t) rcu_read_lock(); - if (qp->q.fqdir->dead) + /* Paired with WRITE_ONCE() in fqdir_pre_exit(). */ + if (READ_ONCE(qp->q.fqdir->dead)) goto out_rcu_unlock; spin_lock(&qp->q.lock); @@ -348,6 +349,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) qp->iif = dev->ifindex; qp->q.stamp = skb->tstamp; + qp->q.mono_delivery_time = skb->mono_delivery_time; qp->q.meat += skb->len; qp->ecn |= ecn; add_frag_mem_limit(qp->q.fqdir, skb->truesize); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 8274f98c511c..f866d6282b2b 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -107,6 +107,8 @@ module_param(log_ecn_error, bool, 0644); MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN"); static struct rtnl_link_ops ipgre_link_ops __read_mostly; +static const struct header_ops ipgre_header_ops; + static int ipgre_tunnel_init(struct net_device *dev); static void erspan_build_header(struct sk_buff *skb, u32 id, u32 index, @@ -248,6 +250,15 @@ static void gre_err(struct sk_buff *skb, u32 info) ipgre_err(skb, info, &tpi); } +static bool is_erspan_type1(int gre_hdr_len) +{ + /* Both ERSPAN type I (version 0) and type II (version 1) use + * protocol 0x88BE, but the type I has only 4-byte GRE header, + * while type II has 8-byte. + */ + return gre_hdr_len == 4; +} + static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, int gre_hdr_len) { @@ -262,17 +273,26 @@ static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, int len; itn = net_generic(net, erspan_net_id); - iph = ip_hdr(skb); - ershdr = (struct erspan_base_hdr *)(skb->data + gre_hdr_len); - ver = ershdr->ver; - - tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, - tpi->flags | TUNNEL_KEY, - iph->saddr, iph->daddr, tpi->key); + if (is_erspan_type1(gre_hdr_len)) { + ver = 0; + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, + tpi->flags | TUNNEL_NO_KEY, + iph->saddr, iph->daddr, 0); + } else { + ershdr = (struct erspan_base_hdr *)(skb->data + gre_hdr_len); + ver = ershdr->ver; + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, + tpi->flags | TUNNEL_KEY, + iph->saddr, iph->daddr, tpi->key); + } if (tunnel) { - len = gre_hdr_len + erspan_hdr_len(ver); + if (is_erspan_type1(gre_hdr_len)) + len = gre_hdr_len; + else + len = gre_hdr_len + erspan_hdr_len(ver); + if (unlikely(!pskb_may_pull(skb, len))) return PACKET_REJECT; @@ -346,7 +366,10 @@ static int __ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi, raw_proto, false) < 0) goto drop; - if (tunnel->dev->type != ARPHRD_NONE) + /* Special case for ipgre_header_parse(), which expects the + * mac_header to point to the outer IP header. + */ + if (tunnel->dev->header_ops == &ipgre_header_ops) skb_pop_mac_header(skb); else skb_reset_mac_header(skb); @@ -436,14 +459,12 @@ static void __gre_xmit(struct sk_buff *skb, struct net_device *dev, __be16 proto) { struct ip_tunnel *tunnel = netdev_priv(dev); - - if (tunnel->parms.o_flags & TUNNEL_SEQ) - tunnel->o_seqno++; + __be16 flags = tunnel->parms.o_flags; /* Push GRE header. */ gre_build_header(skb, tunnel->tun_hlen, - tunnel->parms.o_flags, proto, tunnel->parms.o_key, - htonl(tunnel->o_seqno)); + flags, proto, tunnel->parms.o_key, + (flags & TUNNEL_SEQ) ? htonl(atomic_fetch_inc(&tunnel->o_seqno)) : 0); ip_tunnel_xmit(skb, dev, tnl_params, tnl_params->protocol); } @@ -481,7 +502,7 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev, (TUNNEL_CSUM | TUNNEL_KEY | TUNNEL_SEQ); gre_build_header(skb, tunnel_hlen, flags, proto, tunnel_id_to_key32(tun_info->key.tun_id), - (flags & TUNNEL_SEQ) ? htonl(tunnel->o_seqno++) : 0); + (flags & TUNNEL_SEQ) ? htonl(atomic_fetch_inc(&tunnel->o_seqno)) : 0); ip_md_tunnel_xmit(skb, dev, IPPROTO_GRE, tunnel_hlen); @@ -503,7 +524,6 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) int tunnel_hlen; int version; int nhoff; - int thoff; tun_info = skb_tunnel_info(skb); if (unlikely(!tun_info || !(tun_info->mode & IP_TUNNEL_INFO_TX) || @@ -537,10 +557,16 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) (ntohs(ip_hdr(skb)->tot_len) > skb->len - nhoff)) truncate = true; - thoff = skb_transport_header(skb) - skb_mac_header(skb); - if (skb->protocol == htons(ETH_P_IPV6) && - (ntohs(ipv6_hdr(skb)->payload_len) > skb->len - thoff)) - truncate = true; + if (skb->protocol == htons(ETH_P_IPV6)) { + int thoff; + + if (skb_transport_header_was_set(skb)) + thoff = skb_transport_header(skb) - skb_mac_header(skb); + else + thoff = nhoff + sizeof(struct ipv6hdr); + if (ntohs(ipv6_hdr(skb)->payload_len) > skb->len - thoff) + truncate = true; + } if (version == 1) { erspan_build_header(skb, ntohl(tunnel_id_to_key32(key->tun_id)), @@ -558,7 +584,7 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) } gre_build_header(skb, 8, TUNNEL_SEQ, - proto, 0, htonl(tunnel->o_seqno++)); + proto, 0, htonl(atomic_fetch_inc(&tunnel->o_seqno))); ip_md_tunnel_xmit(skb, dev, IPPROTO_GRE, tunnel_hlen); @@ -581,8 +607,9 @@ static int gre_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb) key = &info->key; ip_tunnel_init_flow(&fl4, IPPROTO_GRE, key->u.ipv4.dst, key->u.ipv4.src, - tunnel_id_to_key32(key->tun_id), key->tos, 0, - skb->mark, skb_get_hash(skb)); + tunnel_id_to_key32(key->tun_id), + key->tos & ~INET_ECN_MASK, dev_net(dev), 0, + skb->mark, skb_get_hash(skb), key->flow_flags); rt = ip_route_output_key(dev_net(dev), &fl4); if (IS_ERR(rt)) return PTR_ERR(rt); @@ -607,9 +634,7 @@ static netdev_tx_t ipgre_xmit(struct sk_buff *skb, } if (dev->header_ops) { - /* Need space for new headers */ - if (skb_cow_head(skb, dev->needed_headroom - - (tunnel->hlen + sizeof(struct iphdr)))) + if (skb_cow_head(skb, 0)) goto free_skb; tnl_params = (const struct iphdr *)skb->data; @@ -619,6 +644,10 @@ static netdev_tx_t ipgre_xmit(struct sk_buff *skb, */ skb_pull(skb, tunnel->hlen + sizeof(struct iphdr)); skb_reset_mac_header(skb); + + if (skb->ip_summed == CHECKSUM_PARTIAL && + skb_checksum_start(skb) < skb->data) + goto free_skb; } else { if (skb_cow_head(skb, dev->needed_headroom)) goto free_skb; @@ -665,7 +694,10 @@ static netdev_tx_t erspan_xmit(struct sk_buff *skb, } /* Push ERSPAN header */ - if (tunnel->erspan_ver == 1) { + if (tunnel->erspan_ver == 0) { + proto = htons(ETH_P_ERSPAN); + tunnel->parms.o_flags &= ~TUNNEL_SEQ; + } else if (tunnel->erspan_ver == 1) { erspan_build_header(skb, ntohl(tunnel->parms.o_key), tunnel->index, truncate, true); @@ -720,6 +752,7 @@ free_skb: static void ipgre_link_update(struct net_device *dev, bool set_mtu) { struct ip_tunnel *tunnel = netdev_priv(dev); + __be16 flags; int len; len = tunnel->tun_hlen; @@ -727,65 +760,57 @@ static void ipgre_link_update(struct net_device *dev, bool set_mtu) len = tunnel->tun_hlen - len; tunnel->hlen = tunnel->hlen + len; - dev->needed_headroom = dev->needed_headroom + len; + if (dev->header_ops) + dev->hard_header_len += len; + else + dev->needed_headroom += len; + if (set_mtu) dev->mtu = max_t(int, dev->mtu - len, 68); - if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) { - if (!(tunnel->parms.o_flags & TUNNEL_CSUM) || - tunnel->encap.type == TUNNEL_ENCAP_NONE) { - dev->features |= NETIF_F_GSO_SOFTWARE; - dev->hw_features |= NETIF_F_GSO_SOFTWARE; - } else { - dev->features &= ~NETIF_F_GSO_SOFTWARE; - dev->hw_features &= ~NETIF_F_GSO_SOFTWARE; - } - dev->features |= NETIF_F_LLTX; - } else { + flags = tunnel->parms.o_flags; + + if (flags & TUNNEL_SEQ || + (flags & TUNNEL_CSUM && tunnel->encap.type != TUNNEL_ENCAP_NONE)) { + dev->features &= ~NETIF_F_GSO_SOFTWARE; dev->hw_features &= ~NETIF_F_GSO_SOFTWARE; - dev->features &= ~(NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE); + } else { + dev->features |= NETIF_F_GSO_SOFTWARE; + dev->hw_features |= NETIF_F_GSO_SOFTWARE; } } -static int ipgre_tunnel_ioctl(struct net_device *dev, - struct ifreq *ifr, int cmd) +static int ipgre_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, + int cmd) { - struct ip_tunnel_parm p; int err; - if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) - return -EFAULT; - if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { - if (p.iph.version != 4 || p.iph.protocol != IPPROTO_GRE || - p.iph.ihl != 5 || (p.iph.frag_off & htons(~IP_DF)) || - ((p.i_flags | p.o_flags) & (GRE_VERSION | GRE_ROUTING))) + if (p->iph.version != 4 || p->iph.protocol != IPPROTO_GRE || + p->iph.ihl != 5 || (p->iph.frag_off & htons(~IP_DF)) || + ((p->i_flags | p->o_flags) & (GRE_VERSION | GRE_ROUTING))) return -EINVAL; } - p.i_flags = gre_flags_to_tnl_flags(p.i_flags); - p.o_flags = gre_flags_to_tnl_flags(p.o_flags); + p->i_flags = gre_flags_to_tnl_flags(p->i_flags); + p->o_flags = gre_flags_to_tnl_flags(p->o_flags); - err = ip_tunnel_ioctl(dev, &p, cmd); + err = ip_tunnel_ctl(dev, p, cmd); if (err) return err; if (cmd == SIOCCHGTUNNEL) { struct ip_tunnel *t = netdev_priv(dev); - t->parms.i_flags = p.i_flags; - t->parms.o_flags = p.o_flags; + t->parms.i_flags = p->i_flags; + t->parms.o_flags = p->o_flags; if (strcmp(dev->rtnl_link_ops->kind, "erspan")) ipgre_link_update(dev, true); } - p.i_flags = gre_tnl_flags_to_gre_flags(p.i_flags); - p.o_flags = gre_tnl_flags_to_gre_flags(p.o_flags); - - if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p))) - return -EFAULT; - + p->i_flags = gre_tnl_flags_to_gre_flags(p->i_flags); + p->o_flags = gre_tnl_flags_to_gre_flags(p->o_flags); return 0; } @@ -903,10 +928,11 @@ static const struct net_device_ops ipgre_netdev_ops = { .ndo_stop = ipgre_close, #endif .ndo_start_xmit = ipgre_xmit, - .ndo_do_ioctl = ipgre_tunnel_ioctl, + .ndo_siocdevprivate = ip_tunnel_siocdevprivate, .ndo_change_mtu = ip_tunnel_change_mtu, - .ndo_get_stats64 = ip_tunnel_get_stats64, + .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip_tunnel_get_iflink, + .ndo_tunnel_ctl = ipgre_tunnel_ctl, }; #define GRE_FEATURES (NETIF_F_SG | \ @@ -924,32 +950,30 @@ static void ipgre_tunnel_setup(struct net_device *dev) static void __gre_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel; + __be16 flags; tunnel = netdev_priv(dev); tunnel->tun_hlen = gre_calc_hlen(tunnel->parms.o_flags); tunnel->parms.iph.protocol = IPPROTO_GRE; tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen; + dev->needed_headroom = tunnel->hlen + sizeof(tunnel->parms.iph); - dev->features |= GRE_FEATURES; + dev->features |= GRE_FEATURES | NETIF_F_LLTX; dev->hw_features |= GRE_FEATURES; - if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) { - /* TCP offload with GRE SEQ is not supported, nor - * can we support 2 levels of outer headers requiring - * an update. - */ - if (!(tunnel->parms.o_flags & TUNNEL_CSUM) || - (tunnel->encap.type == TUNNEL_ENCAP_NONE)) { - dev->features |= NETIF_F_GSO_SOFTWARE; - dev->hw_features |= NETIF_F_GSO_SOFTWARE; - } + flags = tunnel->parms.o_flags; - /* Can use a lockless transmit, unless we generate - * output sequences - */ - dev->features |= NETIF_F_LLTX; - } + /* TCP offload with GRE SEQ is not supported, nor can we support 2 + * levels of outer headers requiring an update. + */ + if (flags & TUNNEL_SEQ) + return; + if (flags & TUNNEL_CSUM && tunnel->encap.type != TUNNEL_ENCAP_NONE) + return; + + dev->features |= NETIF_F_GSO_SOFTWARE; + dev->hw_features |= NETIF_F_GSO_SOFTWARE; } static int ipgre_tunnel_init(struct net_device *dev) @@ -959,7 +983,7 @@ static int ipgre_tunnel_init(struct net_device *dev) __gre_tunnel_init(dev); - memcpy(dev->dev_addr, &iph->saddr, 4); + __dev_addr_set(dev, &iph->saddr, 4); memcpy(dev->broadcast, &iph->daddr, 4); dev->flags = IFF_NOARP; @@ -973,10 +997,14 @@ static int ipgre_tunnel_init(struct net_device *dev) return -EINVAL; dev->flags = IFF_BROADCAST; dev->header_ops = &ipgre_header_ops; + dev->hard_header_len = tunnel->hlen + sizeof(*iph); + dev->needed_headroom = 0; } #endif } else if (!tunnel->collect_md) { dev->header_ops = &ipgre_header_ops; + dev->hard_header_len = tunnel->hlen + sizeof(*iph); + dev->needed_headroom = 0; } return ip_tunnel_init(dev); @@ -1066,7 +1094,11 @@ static int erspan_validate(struct nlattr *tb[], struct nlattr *data[], if (ret) return ret; - /* ERSPAN should only have GRE sequence and key flag */ + if (data[IFLA_GRE_ERSPAN_VER] && + nla_get_u8(data[IFLA_GRE_ERSPAN_VER]) == 0) + return 0; + + /* ERSPAN type II/III should only have GRE sequence and key flag */ if (data[IFLA_GRE_OFLAGS]) flags |= nla_get_be16(data[IFLA_GRE_OFLAGS]); if (data[IFLA_GRE_IFLAGS]) @@ -1153,10 +1185,28 @@ static int ipgre_netlink_parms(struct net_device *dev, if (data[IFLA_GRE_FWMARK]) *fwmark = nla_get_u32(data[IFLA_GRE_FWMARK]); + return 0; +} + +static int erspan_netlink_parms(struct net_device *dev, + struct nlattr *data[], + struct nlattr *tb[], + struct ip_tunnel_parm *parms, + __u32 *fwmark) +{ + struct ip_tunnel *t = netdev_priv(dev); + int err; + + err = ipgre_netlink_parms(dev, data, tb, parms, fwmark); + if (err) + return err; + if (!data) + return 0; + if (data[IFLA_GRE_ERSPAN_VER]) { t->erspan_ver = nla_get_u8(data[IFLA_GRE_ERSPAN_VER]); - if (t->erspan_ver != 1 && t->erspan_ver != 2) + if (t->erspan_ver > 2) return -EINVAL; } @@ -1232,7 +1282,7 @@ static const struct net_device_ops gre_tap_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = ip_tunnel_change_mtu, - .ndo_get_stats64 = ip_tunnel_get_stats64, + .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip_tunnel_get_iflink, .ndo_fill_metadata_dst = gre_fill_metadata_dst, }; @@ -1241,7 +1291,11 @@ static int erspan_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - tunnel->tun_hlen = 8; + if (tunnel->erspan_ver == 0) + tunnel->tun_hlen = 4; /* 4-byte GRE hdr. */ + else + tunnel->tun_hlen = 8; /* 8-byte GRE hdr. */ + tunnel->parms.iph.protocol = IPPROTO_GRE; tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen + erspan_hdr_len(tunnel->erspan_ver); @@ -1261,7 +1315,7 @@ static const struct net_device_ops erspan_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = ip_tunnel_change_mtu, - .ndo_get_stats64 = ip_tunnel_get_stats64, + .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip_tunnel_get_iflink, .ndo_fill_metadata_dst = gre_fill_metadata_dst, }; @@ -1276,45 +1330,70 @@ static void ipgre_tap_setup(struct net_device *dev) ip_tunnel_setup(dev, gre_tap_net_id); } -static int ipgre_newlink(struct net *src_net, struct net_device *dev, - struct nlattr *tb[], struct nlattr *data[], - struct netlink_ext_ack *extack) +static int +ipgre_newlink_encap_setup(struct net_device *dev, struct nlattr *data[]) { - struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; - __u32 fwmark = 0; - int err; if (ipgre_netlink_encap_parms(data, &ipencap)) { struct ip_tunnel *t = netdev_priv(dev); - err = ip_tunnel_encap_setup(t, &ipencap); + int err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) return err; } + return 0; +} + +static int ipgre_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[], + struct netlink_ext_ack *extack) +{ + struct ip_tunnel_parm p; + __u32 fwmark = 0; + int err; + + err = ipgre_newlink_encap_setup(dev, data); + if (err) + return err; + err = ipgre_netlink_parms(dev, data, tb, &p, &fwmark); if (err < 0) return err; return ip_tunnel_newlink(dev, tb, &p, fwmark); } +static int erspan_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[], + struct netlink_ext_ack *extack) +{ + struct ip_tunnel_parm p; + __u32 fwmark = 0; + int err; + + err = ipgre_newlink_encap_setup(dev, data); + if (err) + return err; + + err = erspan_netlink_parms(dev, data, tb, &p, &fwmark); + if (err) + return err; + return ip_tunnel_newlink(dev, tb, &p, fwmark); +} + static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); - struct ip_tunnel_encap ipencap; __u32 fwmark = t->fwmark; struct ip_tunnel_parm p; int err; - if (ipgre_netlink_encap_parms(data, &ipencap)) { - err = ip_tunnel_encap_setup(t, &ipencap); - - if (err < 0) - return err; - } + err = ipgre_newlink_encap_setup(dev, data); + if (err) + return err; err = ipgre_netlink_parms(dev, data, tb, &p, &fwmark); if (err < 0) @@ -1327,8 +1406,34 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], t->parms.i_flags = p.i_flags; t->parms.o_flags = p.o_flags; - if (strcmp(dev->rtnl_link_ops->kind, "erspan")) - ipgre_link_update(dev, !tb[IFLA_MTU]); + ipgre_link_update(dev, !tb[IFLA_MTU]); + + return 0; +} + +static int erspan_changelink(struct net_device *dev, struct nlattr *tb[], + struct nlattr *data[], + struct netlink_ext_ack *extack) +{ + struct ip_tunnel *t = netdev_priv(dev); + __u32 fwmark = t->fwmark; + struct ip_tunnel_parm p; + int err; + + err = ipgre_newlink_encap_setup(dev, data); + if (err) + return err; + + err = erspan_netlink_parms(dev, data, tb, &p, &fwmark); + if (err < 0) + return err; + + err = ip_tunnel_changelink(dev, tb, &p, fwmark); + if (err < 0) + return err; + + t->parms.i_flags = p.i_flags; + t->parms.o_flags = p.o_flags; return 0; } @@ -1387,8 +1492,8 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) struct ip_tunnel_parm *p = &t->parms; __be16 o_flags = p->o_flags; - if (t->erspan_ver == 1 || t->erspan_ver == 2) { - if (!t->collect_md) + if (t->erspan_ver <= 2) { + if (t->erspan_ver != 0 && !t->collect_md) o_flags |= TUNNEL_KEY; if (nla_put_u8(skb, IFLA_GRE_ERSPAN_VER, t->erspan_ver)) @@ -1397,7 +1502,7 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) if (t->erspan_ver == 1) { if (nla_put_u32(skb, IFLA_GRE_ERSPAN_INDEX, t->index)) goto nla_put_failure; - } else { + } else if (t->erspan_ver == 2) { if (nla_put_u8(skb, IFLA_GRE_ERSPAN_DIR, t->dir)) goto nla_put_failure; if (nla_put_u16(skb, IFLA_GRE_ERSPAN_HWID, t->hwid)) @@ -1519,8 +1624,8 @@ static struct rtnl_link_ops erspan_link_ops __read_mostly = { .priv_size = sizeof(struct ip_tunnel), .setup = erspan_setup, .validate = erspan_validate, - .newlink = ipgre_newlink, - .changelink = ipgre_changelink, + .newlink = erspan_newlink, + .changelink = erspan_changelink, .dellink = ip_tunnel_dellink, .get_size = ipgre_get_size, .fill_info = ipgre_fill_info, diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index aa438c6758a7..1b512390b3cf 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -196,7 +196,8 @@ resubmit: if (ipprot) { if (!ipprot->no_policy) { if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { - kfree_skb(skb); + kfree_skb_reason(skb, + SKB_DROP_REASON_XFRM_POLICY); return; } nf_reset_ct(skb); @@ -215,7 +216,7 @@ resubmit: icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PROT_UNREACH, 0); } - kfree_skb(skb); + kfree_skb_reason(skb, SKB_DROP_REASON_IP_NOPROTO); } else { __IP_INC_STATS(net, IPSTATS_MIB_INDELIVERS); consume_skb(skb); @@ -225,6 +226,7 @@ resubmit: static int ip_local_deliver_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { + skb_clear_delivery_time(skb); __skb_pull(skb, skb_network_header_len(skb)); rcu_read_lock(); @@ -253,6 +255,7 @@ int ip_local_deliver(struct sk_buff *skb) net, NULL, skb, skb->dev, NULL, ip_local_deliver_finish); } +EXPORT_SYMBOL(ip_local_deliver); static inline bool ip_rcv_options(struct sk_buff *skb, struct net_device *dev) { @@ -309,16 +312,17 @@ static bool ip_can_use_hint(const struct sk_buff *skb, const struct iphdr *iph, ip_hdr(hint)->tos == iph->tos; } -INDIRECT_CALLABLE_DECLARE(int udp_v4_early_demux(struct sk_buff *)); -INDIRECT_CALLABLE_DECLARE(int tcp_v4_early_demux(struct sk_buff *)); +int tcp_v4_early_demux(struct sk_buff *skb); +int udp_v4_early_demux(struct sk_buff *skb); static int ip_rcv_finish_core(struct net *net, struct sock *sk, struct sk_buff *skb, struct net_device *dev, const struct sk_buff *hint) { const struct iphdr *iph = ip_hdr(skb); - int (*edemux)(struct sk_buff *skb); + int err, drop_reason; struct rtable *rt; - int err; + + drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; if (ip_can_use_hint(skb, iph, hint)) { err = ip_route_use_hint(skb, iph->daddr, iph->saddr, iph->tos, @@ -327,21 +331,29 @@ static int ip_rcv_finish_core(struct net *net, struct sock *sk, goto drop_error; } - if (net->ipv4.sysctl_ip_early_demux && + if (READ_ONCE(net->ipv4.sysctl_ip_early_demux) && !skb_dst(skb) && !skb->sk && !ip_is_fragment(iph)) { - const struct net_protocol *ipprot; - int protocol = iph->protocol; - - ipprot = rcu_dereference(inet_protos[protocol]); - if (ipprot && (edemux = READ_ONCE(ipprot->early_demux))) { - err = INDIRECT_CALL_2(edemux, tcp_v4_early_demux, - udp_v4_early_demux, skb); - if (unlikely(err)) - goto drop_error; - /* must reload iph, skb->head might have changed */ - iph = ip_hdr(skb); + switch (iph->protocol) { + case IPPROTO_TCP: + if (READ_ONCE(net->ipv4.sysctl_tcp_early_demux)) { + tcp_v4_early_demux(skb); + + /* must reload iph, skb->head might have changed */ + iph = ip_hdr(skb); + } + break; + case IPPROTO_UDP: + if (READ_ONCE(net->ipv4.sysctl_udp_early_demux)) { + err = udp_v4_early_demux(skb); + if (unlikely(err)) + goto drop_error; + + /* must reload iph, skb->head might have changed */ + iph = ip_hdr(skb); + } + break; } } @@ -395,19 +407,23 @@ static int ip_rcv_finish_core(struct net *net, struct sock *sk, * so-called "hole-196" attack) so do it for both. */ if (in_dev && - IN_DEV_ORCONF(in_dev, DROP_UNICAST_IN_L2_MULTICAST)) + IN_DEV_ORCONF(in_dev, DROP_UNICAST_IN_L2_MULTICAST)) { + drop_reason = SKB_DROP_REASON_UNICAST_IN_L2_MULTICAST; goto drop; + } } return NET_RX_SUCCESS; drop: - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); return NET_RX_DROP; drop_error: - if (err == -EXDEV) + if (err == -EXDEV) { + drop_reason = SKB_DROP_REASON_IP_RPFILTER; __NET_INC_STATS(net, LINUX_MIB_IPRPFILTER); + } goto drop; } @@ -435,13 +451,17 @@ static int ip_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb) static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) { const struct iphdr *iph; + int drop_reason; u32 len; /* When the interface is in promisc. mode, drop all the crap * that it receives, do not try to analyse it. */ - if (skb->pkt_type == PACKET_OTHERHOST) + if (skb->pkt_type == PACKET_OTHERHOST) { + dev_core_stats_rx_otherhost_dropped_inc(skb->dev); + drop_reason = SKB_DROP_REASON_OTHERHOST; goto drop; + } __IP_UPD_PO_STATS(net, IPSTATS_MIB_IN, skb->len); @@ -451,6 +471,7 @@ static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) goto out; } + drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; if (!pskb_may_pull(skb, sizeof(struct iphdr))) goto inhdr_error; @@ -487,6 +508,7 @@ static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) len = ntohs(iph->tot_len); if (skb->len < len) { + drop_reason = SKB_DROP_REASON_PKT_TOO_SMALL; __IP_INC_STATS(net, IPSTATS_MIB_INTRUNCATEDPKTS); goto drop; } else if (len < (iph->ihl*4)) @@ -509,16 +531,20 @@ static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) IPCB(skb)->iif = skb->skb_iif; /* Must drop socket now because of tproxy. */ - skb_orphan(skb); + if (!skb_sk_is_prefetched(skb)) + skb_orphan(skb); return skb; csum_error: + drop_reason = SKB_DROP_REASON_IP_CSUM; __IP_INC_STATS(net, IPSTATS_MIB_CSUMERRORS); inhdr_error: + if (drop_reason == SKB_DROP_REASON_NOT_SPECIFIED) + drop_reason = SKB_DROP_REASON_IP_INHDR; __IP_INC_STATS(net, IPSTATS_MIB_INHDRERRORS); drop: - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); out: return NULL; } diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index ddaa01ec2bce..a9e22a098872 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -42,39 +42,26 @@ */ void ip_options_build(struct sk_buff *skb, struct ip_options *opt, - __be32 daddr, struct rtable *rt, int is_frag) + __be32 daddr, struct rtable *rt) { unsigned char *iph = skb_network_header(skb); memcpy(&(IPCB(skb)->opt), opt, sizeof(struct ip_options)); - memcpy(iph+sizeof(struct iphdr), opt->__data, opt->optlen); + memcpy(iph + sizeof(struct iphdr), opt->__data, opt->optlen); opt = &(IPCB(skb)->opt); if (opt->srr) - memcpy(iph+opt->srr+iph[opt->srr+1]-4, &daddr, 4); + memcpy(iph + opt->srr + iph[opt->srr + 1] - 4, &daddr, 4); - if (!is_frag) { - if (opt->rr_needaddr) - ip_rt_get_source(iph+opt->rr+iph[opt->rr+2]-5, skb, rt); - if (opt->ts_needaddr) - ip_rt_get_source(iph+opt->ts+iph[opt->ts+2]-9, skb, rt); - if (opt->ts_needtime) { - __be32 midtime; + if (opt->rr_needaddr) + ip_rt_get_source(iph + opt->rr + iph[opt->rr + 2] - 5, skb, rt); + if (opt->ts_needaddr) + ip_rt_get_source(iph + opt->ts + iph[opt->ts + 2] - 9, skb, rt); + if (opt->ts_needtime) { + __be32 midtime; - midtime = inet_current_timestamp(); - memcpy(iph+opt->ts+iph[opt->ts+2]-5, &midtime, 4); - } - return; - } - if (opt->rr) { - memset(iph+opt->rr, IPOPT_NOP, iph[opt->rr+1]); - opt->rr = 0; - opt->rr_needaddr = 0; - } - if (opt->ts) { - memset(iph+opt->ts, IPOPT_NOP, iph[opt->ts+1]); - opt->ts = 0; - opt->ts_needaddr = opt->ts_needtime = 0; + midtime = inet_current_timestamp(); + memcpy(iph + opt->ts + iph[opt->ts + 2] - 5, &midtime, 4); } } @@ -495,39 +482,47 @@ EXPORT_SYMBOL(ip_options_compile); void ip_options_undo(struct ip_options *opt) { if (opt->srr) { - unsigned char *optptr = opt->__data+opt->srr-sizeof(struct iphdr); - memmove(optptr+7, optptr+3, optptr[1]-7); - memcpy(optptr+3, &opt->faddr, 4); + unsigned char *optptr = opt->__data + opt->srr - sizeof(struct iphdr); + + memmove(optptr + 7, optptr + 3, optptr[1] - 7); + memcpy(optptr + 3, &opt->faddr, 4); } if (opt->rr_needaddr) { - unsigned char *optptr = opt->__data+opt->rr-sizeof(struct iphdr); + unsigned char *optptr = opt->__data + opt->rr - sizeof(struct iphdr); + optptr[2] -= 4; - memset(&optptr[optptr[2]-1], 0, 4); + memset(&optptr[optptr[2] - 1], 0, 4); } if (opt->ts) { - unsigned char *optptr = opt->__data+opt->ts-sizeof(struct iphdr); + unsigned char *optptr = opt->__data + opt->ts - sizeof(struct iphdr); + if (opt->ts_needtime) { optptr[2] -= 4; - memset(&optptr[optptr[2]-1], 0, 4); - if ((optptr[3]&0xF) == IPOPT_TS_PRESPEC) + memset(&optptr[optptr[2] - 1], 0, 4); + if ((optptr[3] & 0xF) == IPOPT_TS_PRESPEC) optptr[2] -= 4; } if (opt->ts_needaddr) { optptr[2] -= 4; - memset(&optptr[optptr[2]-1], 0, 4); + memset(&optptr[optptr[2] - 1], 0, 4); } } } -static struct ip_options_rcu *ip_options_get_alloc(const int optlen) +int ip_options_get(struct net *net, struct ip_options_rcu **optp, + sockptr_t data, int optlen) { - return kzalloc(sizeof(struct ip_options_rcu) + ((optlen + 3) & ~3), + struct ip_options_rcu *opt; + + opt = kzalloc(sizeof(struct ip_options_rcu) + ((optlen + 3) & ~3), GFP_KERNEL); -} + if (!opt) + return -ENOMEM; + if (optlen && copy_from_sockptr(opt->opt.__data, data, optlen)) { + kfree(opt); + return -EFAULT; + } -static int ip_options_get_finish(struct net *net, struct ip_options_rcu **optp, - struct ip_options_rcu *opt, int optlen) -{ while (optlen & 3) opt->opt.__data[optlen++] = IPOPT_END; opt->opt.optlen = optlen; @@ -540,32 +535,6 @@ static int ip_options_get_finish(struct net *net, struct ip_options_rcu **optp, return 0; } -int ip_options_get_from_user(struct net *net, struct ip_options_rcu **optp, - unsigned char __user *data, int optlen) -{ - struct ip_options_rcu *opt = ip_options_get_alloc(optlen); - - if (!opt) - return -ENOMEM; - if (optlen && copy_from_user(opt->opt.__data, data, optlen)) { - kfree(opt); - return -EFAULT; - } - return ip_options_get_finish(net, optp, opt, optlen); -} - -int ip_options_get(struct net *net, struct ip_options_rcu **optp, - unsigned char *data, int optlen) -{ - struct ip_options_rcu *opt = ip_options_get_alloc(optlen); - - if (!opt) - return -ENOMEM; - if (optlen) - memcpy(opt->opt.__data, data, optlen); - return ip_options_get_finish(net, optp, opt, optlen); -} - void ip_forward_options(struct sk_buff *skb) { struct ip_options *opt = &(IPCB(skb)->opt); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index d84819893db9..922c87ef1ab5 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -34,7 +34,7 @@ * Andi Kleen : Replace ip_reply with ip_send_reply. * Andi Kleen : Split fast and slow ip_build_xmit path * for decreased register pressure on x86 - * and more readibility. + * and more readability. * Marc Boucher : When call_out_firewall returns FW_QUEUE, * silently drop skb instead of failing with -EPERM. * Detlev Wengorz : Copy protocol for fragments. @@ -74,6 +74,7 @@ #include <net/icmp.h> #include <net/checksum.h> #include <net/inetpeer.h> +#include <net/inet_ecn.h> #include <net/lwtunnel.h> #include <linux/bpf-cgroup.h> #include <linux/igmp.h> @@ -142,7 +143,8 @@ static inline int ip_select_ttl(struct inet_sock *inet, struct dst_entry *dst) * */ int ip_build_and_send_pkt(struct sk_buff *skb, const struct sock *sk, - __be32 saddr, __be32 daddr, struct ip_options_rcu *opt) + __be32 saddr, __be32 daddr, struct ip_options_rcu *opt, + u8 tos) { struct inet_sock *inet = inet_sk(sk); struct rtable *rt = skb_rtable(skb); @@ -155,22 +157,29 @@ int ip_build_and_send_pkt(struct sk_buff *skb, const struct sock *sk, iph = ip_hdr(skb); iph->version = 4; iph->ihl = 5; - iph->tos = inet->tos; + iph->tos = tos; iph->ttl = ip_select_ttl(inet, &rt->dst); iph->daddr = (opt && opt->opt.srr ? opt->opt.faddr : daddr); iph->saddr = saddr; iph->protocol = sk->sk_protocol; - if (ip_dont_fragment(sk, &rt->dst)) { + /* Do not bother generating IPID for small packets (eg SYNACK) */ + if (skb->len <= IPV4_MIN_MTU || ip_dont_fragment(sk, &rt->dst)) { iph->frag_off = htons(IP_DF); iph->id = 0; } else { iph->frag_off = 0; - __ip_select_ident(net, iph, 1); + /* TCP packets here are SYNACK with fat IPv4/TCP options. + * Avoid using the hashed IP ident generator. + */ + if (sk->sk_protocol == IPPROTO_TCP) + iph->id = (__force __be16)get_random_u16(); + else + __ip_select_ident(net, iph, 1); } if (opt && opt->opt.optlen) { iph->ihl += opt->opt.optlen>>2; - ip_options_build(skb, &opt->opt, daddr, rt, 0); + ip_options_build(skb, &opt->opt, daddr, rt); } skb->priority = sk->sk_priority; @@ -196,19 +205,10 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s } else if (rt->rt_type == RTN_BROADCAST) IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTBCAST, skb->len); - /* Be paranoid, rather than too clever. */ if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { - struct sk_buff *skb2; - - skb2 = skb_realloc_headroom(skb, LL_RESERVED_SPACE(dev)); - if (!skb2) { - kfree_skb(skb); + skb = skb_expand_head(skb, hh_len); + if (!skb) return -ENOMEM; - } - if (skb->sk) - skb_set_owner_w(skb2, skb->sk); - consume_skb(skb); - skb = skb2; } if (lwtunnel_xmit_redirect(dst->lwtstate)) { @@ -233,7 +233,7 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s net_dbg_ratelimited("%s: No header cache and no neighbour!\n", __func__); - kfree_skb(skb); + kfree_skb_reason(skb, SKB_DROP_REASON_NEIGH_CREATEFAIL); return -EINVAL; } @@ -260,10 +260,10 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk, * interface with a smaller MTU. * - Arriving GRO skb (or GSO skb in a virtualized environment) that is * bridged to a NETIF_F_TSO tunnel stacked over an interface with an - * insufficent MTU. + * insufficient MTU. */ features = netif_skb_features(skb); - BUILD_BUG_ON(sizeof(*IPCB(skb)) > SKB_SGO_CB_OFFSET); + BUILD_BUG_ON(sizeof(*IPCB(skb)) > SKB_GSO_CB_OFFSET); segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) { kfree_skb(skb); @@ -300,7 +300,7 @@ static int __ip_finish_output(struct net *net, struct sock *sk, struct sk_buff * if (skb_is_gso(skb)) return ip_finish_output_gso(net, sk, skb, mtu); - if (skb->len > mtu || (IPCB(skb)->flags & IPSKB_FRAG_PMTU)) + if (skb->len > mtu || IPCB(skb)->frag_max_size) return ip_fragment(net, sk, skb, mtu, ip_finish_output2); return ip_finish_output2(net, sk, skb); @@ -317,7 +317,7 @@ static int ip_finish_output(struct net *net, struct sock *sk, struct sk_buff *sk case NET_XMIT_CN: return __ip_finish_output(net, sk, skb) ? : ret; default: - kfree_skb(skb); + kfree_skb_reason(skb, SKB_DROP_REASON_BPF_CGROUP_EGRESS); return ret; } } @@ -333,11 +333,11 @@ static int ip_mc_finish_output(struct net *net, struct sock *sk, switch (ret) { case NET_XMIT_CN: do_cn = true; - /* fall through */ + fallthrough; case NET_XMIT_SUCCESS: break; default: - kfree_skb(skb); + kfree_skb_reason(skb, SKB_DROP_REASON_BPF_CGROUP_EGRESS); return ret; } @@ -432,6 +432,7 @@ int ip_output(struct net *net, struct sock *sk, struct sk_buff *skb) ip_finish_output, !(IPCB(skb)->flags & IPSKB_REROUTED)); } +EXPORT_SYMBOL(ip_output); /* * copy saddr and daddr, possibly using 64bit load/stores @@ -443,8 +444,9 @@ static void ip_copy_addrs(struct iphdr *iph, const struct flowi4 *fl4) { BUILD_BUG_ON(offsetof(typeof(*fl4), daddr) != offsetof(typeof(*fl4), saddr) + sizeof(fl4->saddr)); - memcpy(&iph->saddr, &fl4->saddr, - sizeof(fl4->saddr) + sizeof(fl4->daddr)); + + iph->saddr = fl4->saddr; + iph->daddr = fl4->daddr; } /* Note: skb->sk can be different from sk, in case of tunnels */ @@ -517,7 +519,7 @@ packet_routed: if (inet_opt && inet_opt->opt.optlen) { iph->ihl += inet_opt->opt.optlen >> 2; - ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt, 0); + ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt); } ip_select_ident_segs(net, skb, sk, @@ -534,11 +536,17 @@ packet_routed: no_route: rcu_read_unlock(); IP_INC_STATS(net, IPSTATS_MIB_OUTNOROUTES); - kfree_skb(skb); + kfree_skb_reason(skb, SKB_DROP_REASON_IP_OUTNOROUTES); return -EHOSTUNREACH; } EXPORT_SYMBOL(__ip_queue_xmit); +int ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl) +{ + return __ip_queue_xmit(sk, skb, fl, inet_sk(sk)->tos); +} +EXPORT_SYMBOL(ip_queue_xmit); + static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) { to->pkt_type = from->pkt_type; @@ -605,18 +613,6 @@ void ip_fraglist_init(struct sk_buff *skb, struct iphdr *iph, } EXPORT_SYMBOL(ip_fraglist_init); -static void ip_fraglist_ipcb_prepare(struct sk_buff *skb, - struct ip_fraglist_iter *iter) -{ - struct sk_buff *to = iter->frag; - - /* Copy the flags to each fragment. */ - IPCB(to)->flags = IPCB(skb)->flags; - - if (iter->offset == 0) - ip_options_fragment(to); -} - void ip_fraglist_prepare(struct sk_buff *skb, struct ip_fraglist_iter *iter) { unsigned int hlen = iter->hlen; @@ -662,7 +658,7 @@ void ip_frag_init(struct sk_buff *skb, unsigned int hlen, EXPORT_SYMBOL(ip_frag_init); static void ip_frag_ipcb(struct sk_buff *from, struct sk_buff *to, - bool first_frag, struct ip_frag_state *state) + bool first_frag) { /* Copy the flags to each fragment. */ IPCB(to)->flags = IPCB(from)->flags; @@ -683,7 +679,6 @@ struct sk_buff *ip_frag_next(struct sk_buff *skb, struct ip_frag_state *state) struct sk_buff *skb2; struct iphdr *iph; - len = state->left; /* IF: it doesn't fit, use 'mtu' - the data space left */ if (len > state->mtu) len = state->mtu; @@ -766,6 +761,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, { struct iphdr *iph; struct sk_buff *skb2; + bool mono_delivery_time = skb->mono_delivery_time; struct rtable *rt = skb_rtable(skb); unsigned int mtu, hlen, ll_rs; struct ip_fraglist_iter iter; @@ -841,11 +837,23 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, /* Prepare header of the next frame, * before previous one went down. */ if (iter.frag) { - ip_fraglist_ipcb_prepare(skb, &iter); + bool first_frag = (iter.offset == 0); + + IPCB(iter.frag)->flags = IPCB(skb)->flags; ip_fraglist_prepare(skb, &iter); + if (first_frag && IPCB(skb)->opt.optlen) { + /* ipcb->opt is not populated for frags + * coming from __ip_make_skb(), + * ip_options_fragment() needs optlen + */ + IPCB(iter.frag)->opt.optlen = + IPCB(skb)->opt.optlen; + ip_options_fragment(iter.frag); + ip_send_check(iter.iph); + } } - skb->tstamp = tstamp; + skb_set_delivery_time(skb, tstamp, mono_delivery_time); err = output(net, sk, skb); if (!err) @@ -896,12 +904,12 @@ slow_path: err = PTR_ERR(skb2); goto fail; } - ip_frag_ipcb(skb, skb2, first_frag, &state); + ip_frag_ipcb(skb, skb2, first_frag); /* * Put this fragment into the sending queue. */ - skb2->tstamp = tstamp; + skb_set_delivery_time(skb2, tstamp, mono_delivery_time); err = output(net, sk, skb2); if (err) goto fail; @@ -961,7 +969,6 @@ static int __ip_append_data(struct sock *sk, struct inet_sock *inet = inet_sk(sk); struct ubuf_info *uarg = NULL; struct sk_buff *skb; - struct ip_options *opt = cork->opt; int hh_len; int exthdrlen; @@ -969,6 +976,7 @@ static int __ip_append_data(struct sock *sk, int copy; int err; int offset = 0; + bool zc = false; unsigned int maxfraglen, fragheaderlen, maxnonfragsize; int csummode = CHECKSUM_NONE; struct rtable *rt = (struct rtable *)cork->dst; @@ -984,13 +992,13 @@ static int __ip_append_data(struct sock *sk, if (cork->tx_flags & SKBTX_ANY_SW_TSTAMP && sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID) - tskey = sk->sk_tskey++; + tskey = atomic_inc_return(&sk->sk_tskey) - 1; hh_len = LL_RESERVED_SPACE(rt->dst.dev); fragheaderlen = sizeof(struct iphdr) + (opt ? opt->optlen : 0); maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen; - maxnonfragsize = ip_sk_ignore_df(sk) ? 0xFFFF : mtu; + maxnonfragsize = ip_sk_ignore_df(sk) ? IP_MAX_MTU : mtu; if (cork->length + length > maxnonfragsize - fragheaderlen) { ip_local_error(sk, EMSGSIZE, fl4->daddr, inet->inet_dport, @@ -1009,17 +1017,35 @@ static int __ip_append_data(struct sock *sk, (!exthdrlen || (rt->dst.dev->features & NETIF_F_HW_ESP_TX_CSUM))) csummode = CHECKSUM_PARTIAL; - if (flags & MSG_ZEROCOPY && length && sock_flag(sk, SOCK_ZEROCOPY)) { - uarg = sock_zerocopy_realloc(sk, length, skb_zcopy(skb)); - if (!uarg) - return -ENOBUFS; - extra_uref = !skb_zcopy(skb); /* only ref on new uarg */ - if (rt->dst.dev->features & NETIF_F_SG && - csummode == CHECKSUM_PARTIAL) { - paged = true; - } else { - uarg->zerocopy = 0; - skb_zcopy_set(skb, uarg, &extra_uref); + if ((flags & MSG_ZEROCOPY) && length) { + struct msghdr *msg = from; + + if (getfrag == ip_generic_getfrag && msg->msg_ubuf) { + if (skb_zcopy(skb) && msg->msg_ubuf != skb_zcopy(skb)) + return -EINVAL; + + /* Leave uarg NULL if can't zerocopy, callers should + * be able to handle it. + */ + if ((rt->dst.dev->features & NETIF_F_SG) && + csummode == CHECKSUM_PARTIAL) { + paged = true; + zc = true; + uarg = msg->msg_ubuf; + } + } else if (sock_flag(sk, SOCK_ZEROCOPY)) { + uarg = msg_zerocopy_realloc(sk, length, skb_zcopy(skb)); + if (!uarg) + return -ENOBUFS; + extra_uref = !skb_zcopy(skb); /* only ref on new uarg */ + if (rt->dst.dev->features & NETIF_F_SG && + csummode == CHECKSUM_PARTIAL) { + paged = true; + zc = true; + } else { + uarg_to_msgzc(uarg)->zerocopy = 0; + skb_zcopy_set(skb, uarg, &extra_uref); + } } } @@ -1045,7 +1071,7 @@ static int __ip_append_data(struct sock *sk, unsigned int datalen; unsigned int fraglen; unsigned int fraggap; - unsigned int alloclen; + unsigned int alloclen, alloc_extra; unsigned int pagedlen; struct sk_buff *skb_prev; alloc_new_skb: @@ -1065,17 +1091,8 @@ alloc_new_skb: fraglen = datalen + fragheaderlen; pagedlen = 0; - if ((flags & MSG_MORE) && - !(rt->dst.dev->features&NETIF_F_SG)) - alloclen = mtu; - else if (!paged) - alloclen = fraglen; - else { - alloclen = min_t(int, fraglen, MAX_HEADER); - pagedlen = fraglen - alloclen; - } - - alloclen += exthdrlen; + alloc_extra = hh_len + 15; + alloc_extra += exthdrlen; /* The last fragment gets additional space at tail. * Note, with MSG_MORE we overallocate on fragments, @@ -1083,17 +1100,30 @@ alloc_new_skb: * the last. */ if (datalen == length + fraggap) - alloclen += rt->dst.trailer_len; + alloc_extra += rt->dst.trailer_len; + + if ((flags & MSG_MORE) && + !(rt->dst.dev->features&NETIF_F_SG)) + alloclen = mtu; + else if (!paged && + (fraglen + alloc_extra < SKB_MAX_ALLOC || + !(rt->dst.dev->features & NETIF_F_SG))) + alloclen = fraglen; + else { + alloclen = fragheaderlen + transhdrlen; + pagedlen = datalen - transhdrlen; + } + + alloclen += alloc_extra; if (transhdrlen) { - skb = sock_alloc_send_skb(sk, - alloclen + hh_len + 15, + skb = sock_alloc_send_skb(sk, alloclen, (flags & MSG_DONTWAIT), &err); } else { skb = NULL; if (refcount_read(&sk->sk_wmem_alloc) + wmem_alloc_delta <= 2 * sk->sk_sndbuf) - skb = alloc_skb(alloclen + hh_len + 15, + skb = alloc_skb(alloclen, sk->sk_allocation); if (unlikely(!skb)) err = -ENOBUFS; @@ -1120,7 +1150,7 @@ alloc_new_skb: if (fraggap) { skb->csum = skb_copy_and_csum_bits( skb_prev, maxfraglen, - data + transhdrlen, fraggap, 0); + data + transhdrlen, fraggap); skb_prev->csum = csum_sub(skb_prev->csum, skb->csum); data += fraggap; @@ -1176,13 +1206,14 @@ alloc_new_skb: err = -EFAULT; goto error; } - } else if (!uarg || !uarg->zerocopy) { + } else if (!zc) { int i = skb_shinfo(skb)->nr_frags; err = -ENOMEM; if (!sk_page_frag_refill(sk, pfrag)) goto error; + skb_zcopy_downgrade_managed(skb); if (!skb_can_coalesce(skb, i, pfrag->page, pfrag->offset)) { err = -EMSGSIZE; @@ -1202,9 +1233,7 @@ alloc_new_skb: pfrag->offset += copy; skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy); - skb->len += copy; - skb->data_len += copy; - skb->truesize += copy; + skb_len_add(skb, copy); wmem_alloc_delta += copy; } else { err = skb_zerocopy_iter_dgram(skb, from, copy); @@ -1222,8 +1251,7 @@ alloc_new_skb: error_efault: err = -EFAULT; error: - if (uarg) - sock_zerocopy_put_abort(uarg, extra_uref); + net_zcopy_put_abort(uarg, extra_uref); cork->length -= length; IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTDISCARDS); refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc); @@ -1345,7 +1373,7 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, if (cork->flags & IPCORK_OPT) opt = cork->opt; - if (!(rt->dst.dev->features&NETIF_F_SG)) + if (!(rt->dst.dev->features & NETIF_F_SG)) return -EOPNOTSUPP; hh_len = LL_RESERVED_SPACE(rt->dst.dev); @@ -1405,7 +1433,7 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, skb->csum = skb_copy_and_csum_bits(skb_prev, maxfraglen, skb_transport_header(skb), - fraggap, 0); + fraggap); skb_prev->csum = csum_sub(skb_prev->csum, skb->csum); pskb_trim_unique(skb_prev, maxfraglen); @@ -1432,9 +1460,7 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, skb->csum = csum_block_add(skb->csum, csum, skb->len); } - skb->len += len; - skb->data_len += len; - skb->truesize += len; + skb_len_add(skb, len); refcount_add(len, &sk->sk_wmem_alloc); offset += len; size -= len; @@ -1530,8 +1556,8 @@ struct sk_buff *__ip_make_skb(struct sock *sk, ip_select_ident(net, skb, sk); if (opt) { - iph->ihl += opt->optlen>>2; - ip_options_build(skb, opt, cork->addr, rt, 0); + iph->ihl += opt->optlen >> 2; + ip_options_build(skb, opt, cork->addr, rt); } skb->priority = (cork->tos != -1) ? cork->priority: sk->sk_priority; @@ -1642,7 +1668,7 @@ static int ip_reply_glue_bits(void *dptr, char *to, int offset, { __wsum csum; - csum = csum_partial_copy_nocheck(dptr+offset, to, len, 0); + csum = csum_partial_copy_nocheck(dptr+offset, to, len); skb->csum = csum_block_add(skb->csum, csum, odd); return 0; } @@ -1692,17 +1718,17 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, daddr, saddr, tcp_hdr(skb)->source, tcp_hdr(skb)->dest, arg->uid); - security_skb_classify_flow(skb, flowi4_to_flowi(&fl4)); - rt = ip_route_output_key(net, &fl4); + security_skb_classify_flow(skb, flowi4_to_flowi_common(&fl4)); + rt = ip_route_output_flow(net, &fl4, sk); if (IS_ERR(rt)) return; - inet_sk(sk)->tos = arg->tos; + inet_sk(sk)->tos = arg->tos & ~INET_ECN_MASK; sk->sk_protocol = ip_hdr(skb)->protocol; sk->sk_bound_dev_if = arg->bound_dev_if; - sk->sk_sndbuf = sysctl_wmem_default; - sk->sk_mark = fl4.flowi4_mark; + sk->sk_sndbuf = READ_ONCE(sysctl_wmem_default); + ipc.sockc.mark = fl4.flowi4_mark; err = ip_append_data(sk, &fl4, ip_reply_glue_bits, arg->iov->iov_base, len, 0, &ipc, &rt, MSG_DONTWAIT); if (unlikely(err)) { @@ -1717,6 +1743,7 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, arg->csumoffset) = csum_fold(csum_add(nskb->csum, arg->csum)); nskb->ip_summed = CHECKSUM_NONE; + nskb->mono_delivery_time = !!transmit_time; ip_push_pending_frames(sk, &fl4); } out: diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index aa3fd61818c4..6e19cad154f5 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -280,7 +280,8 @@ int ip_cmsg_send(struct sock *sk, struct msghdr *msg, struct ipcm_cookie *ipc, err = cmsg->cmsg_len - sizeof(struct cmsghdr); /* Our caller is responsible for freeing ipc->opt */ - err = ip_options_get(net, &ipc->opt, CMSG_DATA(cmsg), + err = ip_options_get(net, &ipc->opt, + KERNEL_SOCKPTR(CMSG_DATA(cmsg)), err < 40 ? err : 40); if (err) return err; @@ -389,6 +390,18 @@ int ip_ra_control(struct sock *sk, unsigned char on, return 0; } +static void ipv4_icmp_error_rfc4884(const struct sk_buff *skb, + struct sock_ee_data_rfc4884 *out) +{ + switch (icmp_hdr(skb)->type) { + case ICMP_DEST_UNREACH: + case ICMP_TIME_EXCEEDED: + case ICMP_PARAMETERPROB: + ip_icmp_error_rfc4884(skb, out, sizeof(struct icmphdr), + icmp_hdr(skb)->un.reserved[1] * 4); + } +} + void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, __be16 port, u32 info, u8 *payload) { @@ -411,6 +424,9 @@ void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, serr->port = port; if (skb_pull(skb, payload - skb->data)) { + if (inet_sk(sk)->recverr_rfc4884) + ipv4_icmp_error_rfc4884(skb, &serr->ee.ee_rfc4884); + skb_reset_transport_header(skb); if (sock_queue_err_skb(sk, skb) == 0) return; @@ -560,6 +576,61 @@ out: return err; } +void __ip_sock_set_tos(struct sock *sk, int val) +{ + if (sk->sk_type == SOCK_STREAM) { + val &= ~INET_ECN_MASK; + val |= inet_sk(sk)->tos & INET_ECN_MASK; + } + if (inet_sk(sk)->tos != val) { + inet_sk(sk)->tos = val; + sk->sk_priority = rt_tos2priority(val); + sk_dst_reset(sk); + } +} + +void ip_sock_set_tos(struct sock *sk, int val) +{ + lock_sock(sk); + __ip_sock_set_tos(sk, val); + release_sock(sk); +} +EXPORT_SYMBOL(ip_sock_set_tos); + +void ip_sock_set_freebind(struct sock *sk) +{ + lock_sock(sk); + inet_sk(sk)->freebind = true; + release_sock(sk); +} +EXPORT_SYMBOL(ip_sock_set_freebind); + +void ip_sock_set_recverr(struct sock *sk) +{ + lock_sock(sk); + inet_sk(sk)->recverr = true; + release_sock(sk); +} +EXPORT_SYMBOL(ip_sock_set_recverr); + +int ip_sock_set_mtu_discover(struct sock *sk, int val) +{ + if (val < IP_PMTUDISC_DONT || val > IP_PMTUDISC_OMIT) + return -EINVAL; + lock_sock(sk); + inet_sk(sk)->pmtudisc = val; + release_sock(sk); + return 0; +} +EXPORT_SYMBOL(ip_sock_set_mtu_discover); + +void ip_sock_set_pktinfo(struct sock *sk) +{ + lock_sock(sk); + inet_sk(sk)->cmsg_flags |= IP_CMSG_PKTINFO; + release_sock(sk); +} +EXPORT_SYMBOL(ip_sock_set_pktinfo); /* * Socket option code for IP. This is the end of the line after any @@ -587,8 +658,238 @@ static bool setsockopt_needs_rtnl(int optname) return false; } -static int do_ip_setsockopt(struct sock *sk, int level, - int optname, char __user *optval, unsigned int optlen) +static int set_mcast_msfilter(struct sock *sk, int ifindex, + int numsrc, int fmode, + struct sockaddr_storage *group, + struct sockaddr_storage *list) +{ + struct ip_msfilter *msf; + struct sockaddr_in *psin; + int err, i; + + msf = kmalloc(IP_MSFILTER_SIZE(numsrc), GFP_KERNEL); + if (!msf) + return -ENOBUFS; + + psin = (struct sockaddr_in *)group; + if (psin->sin_family != AF_INET) + goto Eaddrnotavail; + msf->imsf_multiaddr = psin->sin_addr.s_addr; + msf->imsf_interface = 0; + msf->imsf_fmode = fmode; + msf->imsf_numsrc = numsrc; + for (i = 0; i < numsrc; ++i) { + psin = (struct sockaddr_in *)&list[i]; + + if (psin->sin_family != AF_INET) + goto Eaddrnotavail; + msf->imsf_slist_flex[i] = psin->sin_addr.s_addr; + } + err = ip_mc_msfilter(sk, msf, ifindex); + kfree(msf); + return err; + +Eaddrnotavail: + kfree(msf); + return -EADDRNOTAVAIL; +} + +static int copy_group_source_from_sockptr(struct group_source_req *greqs, + sockptr_t optval, int optlen) +{ + if (in_compat_syscall()) { + struct compat_group_source_req gr32; + + if (optlen != sizeof(gr32)) + return -EINVAL; + if (copy_from_sockptr(&gr32, optval, sizeof(gr32))) + return -EFAULT; + greqs->gsr_interface = gr32.gsr_interface; + greqs->gsr_group = gr32.gsr_group; + greqs->gsr_source = gr32.gsr_source; + } else { + if (optlen != sizeof(*greqs)) + return -EINVAL; + if (copy_from_sockptr(greqs, optval, sizeof(*greqs))) + return -EFAULT; + } + + return 0; +} + +static int do_mcast_group_source(struct sock *sk, int optname, + sockptr_t optval, int optlen) +{ + struct group_source_req greqs; + struct ip_mreq_source mreqs; + struct sockaddr_in *psin; + int omode, add, err; + + err = copy_group_source_from_sockptr(&greqs, optval, optlen); + if (err) + return err; + + if (greqs.gsr_group.ss_family != AF_INET || + greqs.gsr_source.ss_family != AF_INET) + return -EADDRNOTAVAIL; + + psin = (struct sockaddr_in *)&greqs.gsr_group; + mreqs.imr_multiaddr = psin->sin_addr.s_addr; + psin = (struct sockaddr_in *)&greqs.gsr_source; + mreqs.imr_sourceaddr = psin->sin_addr.s_addr; + mreqs.imr_interface = 0; /* use index for mc_source */ + + if (optname == MCAST_BLOCK_SOURCE) { + omode = MCAST_EXCLUDE; + add = 1; + } else if (optname == MCAST_UNBLOCK_SOURCE) { + omode = MCAST_EXCLUDE; + add = 0; + } else if (optname == MCAST_JOIN_SOURCE_GROUP) { + struct ip_mreqn mreq; + + psin = (struct sockaddr_in *)&greqs.gsr_group; + mreq.imr_multiaddr = psin->sin_addr; + mreq.imr_address.s_addr = 0; + mreq.imr_ifindex = greqs.gsr_interface; + err = ip_mc_join_group_ssm(sk, &mreq, MCAST_INCLUDE); + if (err && err != -EADDRINUSE) + return err; + greqs.gsr_interface = mreq.imr_ifindex; + omode = MCAST_INCLUDE; + add = 1; + } else /* MCAST_LEAVE_SOURCE_GROUP */ { + omode = MCAST_INCLUDE; + add = 0; + } + return ip_mc_source(add, omode, sk, &mreqs, greqs.gsr_interface); +} + +static int ip_set_mcast_msfilter(struct sock *sk, sockptr_t optval, int optlen) +{ + struct group_filter *gsf = NULL; + int err; + + if (optlen < GROUP_FILTER_SIZE(0)) + return -EINVAL; + if (optlen > READ_ONCE(sysctl_optmem_max)) + return -ENOBUFS; + + gsf = memdup_sockptr(optval, optlen); + if (IS_ERR(gsf)) + return PTR_ERR(gsf); + + /* numsrc >= (4G-140)/128 overflow in 32 bits */ + err = -ENOBUFS; + if (gsf->gf_numsrc >= 0x1ffffff || + gsf->gf_numsrc > READ_ONCE(sock_net(sk)->ipv4.sysctl_igmp_max_msf)) + goto out_free_gsf; + + err = -EINVAL; + if (GROUP_FILTER_SIZE(gsf->gf_numsrc) > optlen) + goto out_free_gsf; + + err = set_mcast_msfilter(sk, gsf->gf_interface, gsf->gf_numsrc, + gsf->gf_fmode, &gsf->gf_group, + gsf->gf_slist_flex); +out_free_gsf: + kfree(gsf); + return err; +} + +static int compat_ip_set_mcast_msfilter(struct sock *sk, sockptr_t optval, + int optlen) +{ + const int size0 = offsetof(struct compat_group_filter, gf_slist_flex); + struct compat_group_filter *gf32; + unsigned int n; + void *p; + int err; + + if (optlen < size0) + return -EINVAL; + if (optlen > READ_ONCE(sysctl_optmem_max) - 4) + return -ENOBUFS; + + p = kmalloc(optlen + 4, GFP_KERNEL); + if (!p) + return -ENOMEM; + gf32 = p + 4; /* we want ->gf_group and ->gf_slist_flex aligned */ + + err = -EFAULT; + if (copy_from_sockptr(gf32, optval, optlen)) + goto out_free_gsf; + + /* numsrc >= (4G-140)/128 overflow in 32 bits */ + n = gf32->gf_numsrc; + err = -ENOBUFS; + if (n >= 0x1ffffff) + goto out_free_gsf; + + err = -EINVAL; + if (offsetof(struct compat_group_filter, gf_slist_flex[n]) > optlen) + goto out_free_gsf; + + /* numsrc >= (4G-140)/128 overflow in 32 bits */ + err = -ENOBUFS; + if (n > READ_ONCE(sock_net(sk)->ipv4.sysctl_igmp_max_msf)) + goto out_free_gsf; + err = set_mcast_msfilter(sk, gf32->gf_interface, n, gf32->gf_fmode, + &gf32->gf_group, gf32->gf_slist_flex); +out_free_gsf: + kfree(p); + return err; +} + +static int ip_mcast_join_leave(struct sock *sk, int optname, + sockptr_t optval, int optlen) +{ + struct ip_mreqn mreq = { }; + struct sockaddr_in *psin; + struct group_req greq; + + if (optlen < sizeof(struct group_req)) + return -EINVAL; + if (copy_from_sockptr(&greq, optval, sizeof(greq))) + return -EFAULT; + + psin = (struct sockaddr_in *)&greq.gr_group; + if (psin->sin_family != AF_INET) + return -EINVAL; + mreq.imr_multiaddr = psin->sin_addr; + mreq.imr_ifindex = greq.gr_interface; + if (optname == MCAST_JOIN_GROUP) + return ip_mc_join_group(sk, &mreq); + return ip_mc_leave_group(sk, &mreq); +} + +static int compat_ip_mcast_join_leave(struct sock *sk, int optname, + sockptr_t optval, int optlen) +{ + struct compat_group_req greq; + struct ip_mreqn mreq = { }; + struct sockaddr_in *psin; + + if (optlen < sizeof(struct compat_group_req)) + return -EINVAL; + if (copy_from_sockptr(&greq, optval, sizeof(greq))) + return -EFAULT; + + psin = (struct sockaddr_in *)&greq.gr_group; + if (psin->sin_family != AF_INET) + return -EINVAL; + mreq.imr_multiaddr = psin->sin_addr; + mreq.imr_ifindex = greq.gr_interface; + + if (optname == MCAST_JOIN_GROUP) + return ip_mc_join_group(sk, &mreq); + return ip_mc_leave_group(sk, &mreq); +} + +DEFINE_STATIC_KEY_FALSE(ip4_min_ttl); + +int do_ip_setsockopt(struct sock *sk, int level, int optname, + sockptr_t optval, unsigned int optlen) { struct inet_sock *inet = inet_sk(sk); struct net *net = sock_net(sk); @@ -620,13 +921,14 @@ static int do_ip_setsockopt(struct sock *sk, int level, case IP_RECVORIGDSTADDR: case IP_CHECKSUM: case IP_RECVFRAGSIZE: + case IP_RECVERR_RFC4884: if (optlen >= sizeof(int)) { - if (get_user(val, (int __user *) optval)) + if (copy_from_sockptr(&val, optval, sizeof(val))) return -EFAULT; } else if (optlen >= sizeof(char)) { unsigned char ucval; - if (get_user(ucval, (unsigned char __user *) optval)) + if (copy_from_sockptr(&ucval, optval, sizeof(ucval))) return -EFAULT; val = (int) ucval; } @@ -642,7 +944,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, err = 0; if (needs_rtnl) rtnl_lock(); - lock_sock(sk); + sockopt_lock_sock(sk); switch (optname) { case IP_OPTIONS: @@ -651,8 +953,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, if (optlen > 40) goto e_inval; - err = ip_options_get_from_user(sock_net(sk), &opt, - optval, optlen); + err = ip_options_get(sock_net(sk), &opt, optval, optlen); if (err) break; old = rcu_dereference_protected(inet->inet_opt, @@ -743,15 +1044,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, inet->cmsg_flags &= ~IP_CMSG_RECVFRAGSIZE; break; case IP_TOS: /* This sets both TOS and Precedence */ - if (sk->sk_type == SOCK_STREAM) { - val &= ~INET_ECN_MASK; - val |= inet->tos & INET_ECN_MASK; - } - if (inet->tos != val) { - inet->tos = val; - sk->sk_priority = rt_tos2priority(val); - sk_dst_reset(sk); - } + __ip_sock_set_tos(sk, val); break; case IP_TTL: if (optlen < 1) @@ -787,6 +1080,11 @@ static int do_ip_setsockopt(struct sock *sk, int level, if (!val) skb_queue_purge(&sk->sk_error_queue); break; + case IP_RECVERR_RFC4884: + if (val < 0 || val > 1) + goto e_inval; + inet->recverr_rfc4884 = !!val; + break; case IP_MULTICAST_TTL: if (sk->sk_type == SOCK_STREAM) goto e_inval; @@ -828,8 +1126,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, dev_put(dev); err = -EINVAL; - if (sk->sk_bound_dev_if && - (!midx || midx != sk->sk_bound_dev_if)) + if (sk->sk_bound_dev_if && midx != sk->sk_bound_dev_if) break; inet->uc_index = ifindex; @@ -853,17 +1150,17 @@ static int do_ip_setsockopt(struct sock *sk, int level, err = -EFAULT; if (optlen >= sizeof(struct ip_mreqn)) { - if (copy_from_user(&mreq, optval, sizeof(mreq))) + if (copy_from_sockptr(&mreq, optval, sizeof(mreq))) break; } else { memset(&mreq, 0, sizeof(mreq)); if (optlen >= sizeof(struct ip_mreq)) { - if (copy_from_user(&mreq, optval, - sizeof(struct ip_mreq))) + if (copy_from_sockptr(&mreq, optval, + sizeof(struct ip_mreq))) break; } else if (optlen >= sizeof(struct in_addr)) { - if (copy_from_user(&mreq.imr_address, optval, - sizeof(struct in_addr))) + if (copy_from_sockptr(&mreq.imr_address, optval, + sizeof(struct in_addr))) break; } } @@ -893,7 +1190,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, err = -EINVAL; if (sk->sk_bound_dev_if && mreq.imr_ifindex != sk->sk_bound_dev_if && - (!midx || midx != sk->sk_bound_dev_if)) + midx != sk->sk_bound_dev_if) break; inet->mc_index = mreq.imr_ifindex; @@ -915,11 +1212,12 @@ static int do_ip_setsockopt(struct sock *sk, int level, goto e_inval; err = -EFAULT; if (optlen >= sizeof(struct ip_mreqn)) { - if (copy_from_user(&mreq, optval, sizeof(mreq))) + if (copy_from_sockptr(&mreq, optval, sizeof(mreq))) break; } else { memset(&mreq, 0, sizeof(mreq)); - if (copy_from_user(&mreq, optval, sizeof(struct ip_mreq))) + if (copy_from_sockptr(&mreq, optval, + sizeof(struct ip_mreq))) break; } @@ -935,18 +1233,18 @@ static int do_ip_setsockopt(struct sock *sk, int level, if (optlen < IP_MSFILTER_SIZE(0)) goto e_inval; - if (optlen > sysctl_optmem_max) { + if (optlen > READ_ONCE(sysctl_optmem_max)) { err = -ENOBUFS; break; } - msf = memdup_user(optval, optlen); + msf = memdup_sockptr(optval, optlen); if (IS_ERR(msf)) { err = PTR_ERR(msf); break; } /* numsrc >= (1G-4) overflow in 32 bits */ if (msf->imsf_numsrc >= 0x3ffffffcU || - msf->imsf_numsrc > net->ipv4.sysctl_igmp_max_msf) { + msf->imsf_numsrc > READ_ONCE(net->ipv4.sysctl_igmp_max_msf)) { kfree(msf); err = -ENOBUFS; break; @@ -970,7 +1268,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, if (optlen != sizeof(struct ip_mreq_source)) goto e_inval; - if (copy_from_user(&mreqs, optval, sizeof(mreqs))) { + if (copy_from_sockptr(&mreqs, optval, sizeof(mreqs))) { err = -EFAULT; break; } @@ -1000,145 +1298,24 @@ static int do_ip_setsockopt(struct sock *sk, int level, } case MCAST_JOIN_GROUP: case MCAST_LEAVE_GROUP: - { - struct group_req greq; - struct sockaddr_in *psin; - struct ip_mreqn mreq; - - if (optlen < sizeof(struct group_req)) - goto e_inval; - err = -EFAULT; - if (copy_from_user(&greq, optval, sizeof(greq))) - break; - psin = (struct sockaddr_in *)&greq.gr_group; - if (psin->sin_family != AF_INET) - goto e_inval; - memset(&mreq, 0, sizeof(mreq)); - mreq.imr_multiaddr = psin->sin_addr; - mreq.imr_ifindex = greq.gr_interface; - - if (optname == MCAST_JOIN_GROUP) - err = ip_mc_join_group(sk, &mreq); + if (in_compat_syscall()) + err = compat_ip_mcast_join_leave(sk, optname, optval, + optlen); else - err = ip_mc_leave_group(sk, &mreq); + err = ip_mcast_join_leave(sk, optname, optval, optlen); break; - } case MCAST_JOIN_SOURCE_GROUP: case MCAST_LEAVE_SOURCE_GROUP: case MCAST_BLOCK_SOURCE: case MCAST_UNBLOCK_SOURCE: - { - struct group_source_req greqs; - struct ip_mreq_source mreqs; - struct sockaddr_in *psin; - int omode, add; - - if (optlen != sizeof(struct group_source_req)) - goto e_inval; - if (copy_from_user(&greqs, optval, sizeof(greqs))) { - err = -EFAULT; - break; - } - if (greqs.gsr_group.ss_family != AF_INET || - greqs.gsr_source.ss_family != AF_INET) { - err = -EADDRNOTAVAIL; - break; - } - psin = (struct sockaddr_in *)&greqs.gsr_group; - mreqs.imr_multiaddr = psin->sin_addr.s_addr; - psin = (struct sockaddr_in *)&greqs.gsr_source; - mreqs.imr_sourceaddr = psin->sin_addr.s_addr; - mreqs.imr_interface = 0; /* use index for mc_source */ - - if (optname == MCAST_BLOCK_SOURCE) { - omode = MCAST_EXCLUDE; - add = 1; - } else if (optname == MCAST_UNBLOCK_SOURCE) { - omode = MCAST_EXCLUDE; - add = 0; - } else if (optname == MCAST_JOIN_SOURCE_GROUP) { - struct ip_mreqn mreq; - - psin = (struct sockaddr_in *)&greqs.gsr_group; - mreq.imr_multiaddr = psin->sin_addr; - mreq.imr_address.s_addr = 0; - mreq.imr_ifindex = greqs.gsr_interface; - err = ip_mc_join_group_ssm(sk, &mreq, MCAST_INCLUDE); - if (err && err != -EADDRINUSE) - break; - greqs.gsr_interface = mreq.imr_ifindex; - omode = MCAST_INCLUDE; - add = 1; - } else /* MCAST_LEAVE_SOURCE_GROUP */ { - omode = MCAST_INCLUDE; - add = 0; - } - err = ip_mc_source(add, omode, sk, &mreqs, - greqs.gsr_interface); + err = do_mcast_group_source(sk, optname, optval, optlen); break; - } case MCAST_MSFILTER: - { - struct sockaddr_in *psin; - struct ip_msfilter *msf = NULL; - struct group_filter *gsf = NULL; - int msize, i, ifindex; - - if (optlen < GROUP_FILTER_SIZE(0)) - goto e_inval; - if (optlen > sysctl_optmem_max) { - err = -ENOBUFS; - break; - } - gsf = memdup_user(optval, optlen); - if (IS_ERR(gsf)) { - err = PTR_ERR(gsf); - break; - } - - /* numsrc >= (4G-140)/128 overflow in 32 bits */ - if (gsf->gf_numsrc >= 0x1ffffff || - gsf->gf_numsrc > net->ipv4.sysctl_igmp_max_msf) { - err = -ENOBUFS; - goto mc_msf_out; - } - if (GROUP_FILTER_SIZE(gsf->gf_numsrc) > optlen) { - err = -EINVAL; - goto mc_msf_out; - } - msize = IP_MSFILTER_SIZE(gsf->gf_numsrc); - msf = kmalloc(msize, GFP_KERNEL); - if (!msf) { - err = -ENOBUFS; - goto mc_msf_out; - } - ifindex = gsf->gf_interface; - psin = (struct sockaddr_in *)&gsf->gf_group; - if (psin->sin_family != AF_INET) { - err = -EADDRNOTAVAIL; - goto mc_msf_out; - } - msf->imsf_multiaddr = psin->sin_addr.s_addr; - msf->imsf_interface = 0; - msf->imsf_fmode = gsf->gf_fmode; - msf->imsf_numsrc = gsf->gf_numsrc; - err = -EADDRNOTAVAIL; - for (i = 0; i < gsf->gf_numsrc; ++i) { - psin = (struct sockaddr_in *)&gsf->gf_slist[i]; - - if (psin->sin_family != AF_INET) - goto mc_msf_out; - msf->imsf_slist[i] = psin->sin_addr.s_addr; - } - kfree(gsf); - gsf = NULL; - - err = ip_mc_msfilter(sk, msf, ifindex); -mc_msf_out: - kfree(msf); - kfree(gsf); + if (in_compat_syscall()) + err = compat_ip_set_mcast_msfilter(sk, optval, optlen); + else + err = ip_set_mcast_msfilter(sk, optval, optlen); break; - } case IP_MULTICAST_ALL: if (optlen < 1) goto e_inval; @@ -1156,14 +1333,14 @@ mc_msf_out: case IP_IPSEC_POLICY: case IP_XFRM_POLICY: err = -EPERM; - if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) + if (!sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) break; err = xfrm_user_policy(sk, optname, optval, optlen); break; case IP_TRANSPARENT: - if (!!val && !ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) && - !ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) { + if (!!val && !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) && + !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) { err = -EPERM; break; } @@ -1177,20 +1354,27 @@ mc_msf_out: goto e_inval; if (val < 0 || val > 255) goto e_inval; - inet->min_ttl = val; + + if (val) + static_branch_enable(&ip4_min_ttl); + + /* tcp_v4_err() and tcp_v4_rcv() might read min_ttl + * while we are changint it. + */ + WRITE_ONCE(inet->min_ttl, val); break; default: err = -ENOPROTOOPT; break; } - release_sock(sk); + sockopt_release_sock(sk); if (needs_rtnl) rtnl_unlock(); return err; e_inval: - release_sock(sk); + sockopt_release_sock(sk); if (needs_rtnl) rtnl_unlock(); return -EINVAL; @@ -1237,8 +1421,8 @@ void ipv4_pktinfo_prepare(const struct sock *sk, struct sk_buff *skb) skb_dst_drop(skb); } -int ip_setsockopt(struct sock *sk, int level, - int optname, char __user *optval, unsigned int optlen) +int ip_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval, + unsigned int optlen) { int err; @@ -1263,34 +1447,6 @@ int ip_setsockopt(struct sock *sk, int level, } EXPORT_SYMBOL(ip_setsockopt); -#ifdef CONFIG_COMPAT -int compat_ip_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) -{ - int err; - - if (level != SOL_IP) - return -ENOPROTOOPT; - - if (optname >= MCAST_JOIN_GROUP && optname <= MCAST_MSFILTER) - return compat_mc_setsockopt(sk, level, optname, optval, optlen, - ip_setsockopt); - - err = do_ip_setsockopt(sk, level, optname, optval, optlen); -#ifdef CONFIG_NETFILTER - /* we need to exclude all possible ENOPROTOOPTs except default case */ - if (err == -ENOPROTOOPT && optname != IP_HDRINCL && - optname != IP_IPSEC_POLICY && - optname != IP_XFRM_POLICY && - !ip_mroute_opt(optname)) - err = compat_nf_setsockopt(sk, PF_INET, optname, optval, - optlen); -#endif - return err; -} -EXPORT_SYMBOL(compat_ip_setsockopt); -#endif - /* * Get the options. Note for future reference. The GET of IP options gets * the _received_ ones. The set sets the _sent_ ones. @@ -1306,8 +1462,70 @@ static bool getsockopt_needs_rtnl(int optname) return false; } -static int do_ip_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen, unsigned int flags) +static int ip_get_mcast_msfilter(struct sock *sk, sockptr_t optval, + sockptr_t optlen, int len) +{ + const int size0 = offsetof(struct group_filter, gf_slist_flex); + struct group_filter gsf; + int num, gsf_size; + int err; + + if (len < size0) + return -EINVAL; + if (copy_from_sockptr(&gsf, optval, size0)) + return -EFAULT; + + num = gsf.gf_numsrc; + err = ip_mc_gsfget(sk, &gsf, optval, + offsetof(struct group_filter, gf_slist_flex)); + if (err) + return err; + if (gsf.gf_numsrc < num) + num = gsf.gf_numsrc; + gsf_size = GROUP_FILTER_SIZE(num); + if (copy_to_sockptr(optlen, &gsf_size, sizeof(int)) || + copy_to_sockptr(optval, &gsf, size0)) + return -EFAULT; + return 0; +} + +static int compat_ip_get_mcast_msfilter(struct sock *sk, sockptr_t optval, + sockptr_t optlen, int len) +{ + const int size0 = offsetof(struct compat_group_filter, gf_slist_flex); + struct compat_group_filter gf32; + struct group_filter gf; + int num; + int err; + + if (len < size0) + return -EINVAL; + if (copy_from_sockptr(&gf32, optval, size0)) + return -EFAULT; + + gf.gf_interface = gf32.gf_interface; + gf.gf_fmode = gf32.gf_fmode; + num = gf.gf_numsrc = gf32.gf_numsrc; + gf.gf_group = gf32.gf_group; + + err = ip_mc_gsfget(sk, &gf, optval, + offsetof(struct compat_group_filter, gf_slist_flex)); + if (err) + return err; + if (gf.gf_numsrc < num) + num = gf.gf_numsrc; + len = GROUP_FILTER_SIZE(num) - (sizeof(gf) - sizeof(gf32)); + if (copy_to_sockptr(optlen, &len, sizeof(int)) || + copy_to_sockptr_offset(optval, offsetof(struct compat_group_filter, gf_fmode), + &gf.gf_fmode, sizeof(gf.gf_fmode)) || + copy_to_sockptr_offset(optval, offsetof(struct compat_group_filter, gf_numsrc), + &gf.gf_numsrc, sizeof(gf.gf_numsrc))) + return -EFAULT; + return 0; +} + +int do_ip_getsockopt(struct sock *sk, int level, int optname, + sockptr_t optval, sockptr_t optlen) { struct inet_sock *inet = inet_sk(sk); bool needs_rtnl = getsockopt_needs_rtnl(optname); @@ -1320,14 +1538,14 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, if (ip_mroute_opt(optname)) return ip_mroute_getsockopt(sk, optname, optval, optlen); - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; if (len < 0) return -EINVAL; if (needs_rtnl) rtnl_lock(); - lock_sock(sk); + sockopt_lock_sock(sk); switch (optname) { case IP_OPTIONS: @@ -1343,17 +1561,19 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, memcpy(optbuf, &inet_opt->opt, sizeof(struct ip_options) + inet_opt->opt.optlen); - release_sock(sk); + sockopt_release_sock(sk); - if (opt->optlen == 0) - return put_user(0, optlen); + if (opt->optlen == 0) { + len = 0; + return copy_to_sockptr(optlen, &len, sizeof(int)); + } ip_options_undo(opt); len = min_t(unsigned int, len, opt->optlen); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, opt->__data, len)) + if (copy_to_sockptr(optval, opt->__data, len)) return -EFAULT; return 0; } @@ -1391,7 +1611,7 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, { struct net *net = sock_net(sk); val = (inet->uc_ttl == -1 ? - net->ipv4.sysctl_ip_default_ttl : + READ_ONCE(net->ipv4.sysctl_ip_default_ttl) : inet->uc_ttl); break; } @@ -1417,7 +1637,7 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, dst_release(dst); } if (!val) { - release_sock(sk); + sockopt_release_sock(sk); return -ENOTCONN; } break; @@ -1425,6 +1645,9 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, case IP_RECVERR: val = inet->recverr; break; + case IP_RECVERR_RFC4884: + val = inet->recverr_rfc4884; + break; case IP_MULTICAST_TTL: val = inet->mc_ttl; break; @@ -1439,11 +1662,11 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, struct in_addr addr; len = min_t(unsigned int, len, sizeof(struct in_addr)); addr.s_addr = inet->mc_addr; - release_sock(sk); + sockopt_release_sock(sk); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &addr, len)) + if (copy_to_sockptr(optval, &addr, len)) return -EFAULT; return 0; } @@ -1455,31 +1678,20 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, err = -EINVAL; goto out; } - if (copy_from_user(&msf, optval, IP_MSFILTER_SIZE(0))) { + if (copy_from_sockptr(&msf, optval, IP_MSFILTER_SIZE(0))) { err = -EFAULT; goto out; } - err = ip_mc_msfget(sk, &msf, - (struct ip_msfilter __user *)optval, optlen); + err = ip_mc_msfget(sk, &msf, optval, optlen); goto out; } case MCAST_MSFILTER: - { - struct group_filter gsf; - - if (len < GROUP_FILTER_SIZE(0)) { - err = -EINVAL; - goto out; - } - if (copy_from_user(&gsf, optval, GROUP_FILTER_SIZE(0))) { - err = -EFAULT; - goto out; - } - err = ip_mc_gsfget(sk, &gsf, - (struct group_filter __user *)optval, - optlen); + if (in_compat_syscall()) + err = compat_ip_get_mcast_msfilter(sk, optval, optlen, + len); + else + err = ip_get_mcast_msfilter(sk, optval, optlen, len); goto out; - } case IP_MULTICAST_ALL: val = inet->mc_all; break; @@ -1487,14 +1699,20 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, { struct msghdr msg; - release_sock(sk); + sockopt_release_sock(sk); if (sk->sk_type != SOCK_STREAM) return -ENOPROTOOPT; - msg.msg_control = (__force void *) optval; + if (optval.is_kernel) { + msg.msg_control_is_user = false; + msg.msg_control = optval.kernel; + } else { + msg.msg_control_is_user = true; + msg.msg_control_user = optval.user; + } msg.msg_controllen = len; - msg.msg_flags = flags; + msg.msg_flags = in_compat_syscall() ? MSG_CMSG_COMPAT : 0; if (inet->cmsg_flags & IP_CMSG_PKTINFO) { struct in_pktinfo info; @@ -1513,7 +1731,7 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, put_cmsg(&msg, SOL_IP, IP_TOS, sizeof(tos), &tos); } len -= msg.msg_controllen; - return put_user(len, optlen); + return copy_to_sockptr(optlen, &len, sizeof(int)); } case IP_FREEBIND: val = inet->freebind; @@ -1525,29 +1743,29 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, val = inet->min_ttl; break; default: - release_sock(sk); + sockopt_release_sock(sk); return -ENOPROTOOPT; } - release_sock(sk); + sockopt_release_sock(sk); if (len < sizeof(int) && len > 0 && val >= 0 && val <= 255) { unsigned char ucval = (unsigned char)val; len = 1; - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &ucval, 1)) + if (copy_to_sockptr(optval, &ucval, 1)) return -EFAULT; } else { len = min_t(unsigned int, sizeof(int), len); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &val, len)) + if (copy_to_sockptr(optval, &val, len)) return -EFAULT; } return 0; out: - release_sock(sk); + sockopt_release_sock(sk); if (needs_rtnl) rtnl_unlock(); return err; @@ -1558,7 +1776,9 @@ int ip_getsockopt(struct sock *sk, int level, { int err; - err = do_ip_getsockopt(sk, level, optname, optval, optlen, 0); + err = do_ip_getsockopt(sk, level, optname, + USER_SOCKPTR(optval), USER_SOCKPTR(optlen)); + #if IS_ENABLED(CONFIG_BPFILTER_UMH) if (optname >= BPFILTER_IPT_SO_GET_INFO && optname < BPFILTER_IPT_GET_MAX) @@ -1582,41 +1802,3 @@ int ip_getsockopt(struct sock *sk, int level, return err; } EXPORT_SYMBOL(ip_getsockopt); - -#ifdef CONFIG_COMPAT -int compat_ip_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - int err; - - if (optname == MCAST_MSFILTER) - return compat_mc_getsockopt(sk, level, optname, optval, optlen, - ip_getsockopt); - - err = do_ip_getsockopt(sk, level, optname, optval, optlen, - MSG_CMSG_COMPAT); - -#if IS_ENABLED(CONFIG_BPFILTER_UMH) - if (optname >= BPFILTER_IPT_SO_GET_INFO && - optname < BPFILTER_IPT_GET_MAX) - err = bpfilter_ip_get_sockopt(sk, optname, optval, optlen); -#endif -#ifdef CONFIG_NETFILTER - /* we need to exclude all possible ENOPROTOOPTs except default case */ - if (err == -ENOPROTOOPT && optname != IP_PKTOPTIONS && - !ip_mroute_opt(optname)) { - int len; - - if (get_user(len, optlen)) - return -EFAULT; - - err = compat_nf_getsockopt(sk, PF_INET, optname, optval, &len); - if (err >= 0) - err = put_user(len, optlen); - return err; - } -#endif - return err; -} -EXPORT_SYMBOL(compat_ip_getsockopt); -#endif diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 74e1d964a615..019f3b0839c5 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -85,9 +85,10 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, __be32 remote, __be32 local, __be32 key) { - unsigned int hash; struct ip_tunnel *t, *cand = NULL; struct hlist_head *head; + struct net_device *ndev; + unsigned int hash; hash = ip_tunnel_hash(key, remote); head = &itn->tunnels[hash]; @@ -142,11 +143,8 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, cand = t; } - if (flags & TUNNEL_NO_KEY) - goto skip_key_lookup; - hlist_for_each_entry_rcu(t, head, hash_node) { - if (t->parms.i_key != key || + if ((!(flags & TUNNEL_NO_KEY) && t->parms.i_key != key) || t->parms.iph.saddr != 0 || t->parms.iph.daddr != 0 || !(t->dev->flags & IFF_UP)) @@ -158,7 +156,6 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, cand = t; } -skip_key_lookup: if (cand) return cand; @@ -166,8 +163,9 @@ skip_key_lookup: if (t && t->dev->flags & IFF_UP) return t; - if (itn->fb_tunnel_dev && itn->fb_tunnel_dev->flags & IFF_UP) - return netdev_priv(itn->fb_tunnel_dev); + ndev = READ_ONCE(itn->fb_tunnel_dev); + if (ndev && ndev->flags & IFF_UP) + return netdev_priv(ndev); return NULL; } @@ -244,7 +242,7 @@ static struct net_device *__ip_tunnel_create(struct net *net, if (parms->name[0]) { if (!dev_valid_name(parms->name)) goto failed; - strlcpy(name, parms->name, IFNAMSIZ); + strscpy(name, parms->name, IFNAMSIZ); } else { if (strlen(ops->kind) > (IFNAMSIZ - 3)) goto failed; @@ -296,8 +294,8 @@ static int ip_tunnel_bind_dev(struct net_device *dev) ip_tunnel_init_flow(&fl4, iph->protocol, iph->daddr, iph->saddr, tunnel->parms.o_key, - RT_TOS(iph->tos), tunnel->parms.link, - tunnel->fwmark, 0); + RT_TOS(iph->tos), dev_net(dev), + tunnel->parms.link, tunnel->fwmark, 0, 0); rt = ip_route_output_key(tunnel->net, &fl4); if (!IS_ERR(rt)) { @@ -319,7 +317,7 @@ static int ip_tunnel_bind_dev(struct net_device *dev) } dev->needed_headroom = t_hlen + hlen; - mtu -= (dev->hard_header_len + t_hlen); + mtu -= t_hlen + (dev->type == ARPHRD_ETHER ? dev->hard_header_len : 0); if (mtu < IPV4_MIN_MTU) mtu = IPV4_MIN_MTU; @@ -349,7 +347,10 @@ static struct ip_tunnel *ip_tunnel_create(struct net *net, nt = netdev_priv(dev); t_hlen = nt->hlen + sizeof(struct iphdr); dev->min_mtu = ETH_MIN_MTU; - dev->max_mtu = IP_MAX_MTU - dev->hard_header_len - t_hlen; + dev->max_mtu = IP_MAX_MTU - t_hlen; + if (dev->type == ARPHRD_ETHER) + dev->max_mtu -= dev->hard_header_len; + ip_tunnel_add(itn, nt); return nt; @@ -362,7 +363,6 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst, bool log_ecn_error) { - struct pcpu_sw_netstats *tstats; const struct iphdr *iph = ip_hdr(skb); int err; @@ -390,7 +390,7 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, tunnel->i_seqno = ntohl(tpi->seq) + 1; } - skb_reset_network_header(skb); + skb_set_network_header(skb, (tunnel->dev->type == ARPHRD_ETHER) ? ETH_HLEN : 0); err = IP_ECN_decapsulate(iph, skb); if (unlikely(err)) { @@ -404,12 +404,7 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, } } - tstats = this_cpu_ptr(tunnel->dev->tstats); - u64_stats_update_begin(&tstats->syncp); - tstats->rx_packets++; - tstats->rx_bytes += skb->len; - u64_stats_update_end(&tstats->syncp); - + dev_sw_netstats_rx_add(tunnel->dev, skb->len); skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(tunnel->dev))); if (tunnel->dev->type == ARPHRD_ETHER) { @@ -496,13 +491,15 @@ static int tnl_update_pmtu(struct net_device *dev, struct sk_buff *skb, int mtu; tunnel_hlen = md ? tunnel_hlen : tunnel->hlen; - pkt_size = skb->len - tunnel_hlen - dev->hard_header_len; + pkt_size = skb->len - tunnel_hlen; + pkt_size -= dev->type == ARPHRD_ETHER ? dev->hard_header_len : 0; - if (df) - mtu = dst_mtu(&rt->dst) - dev->hard_header_len - - sizeof(struct iphdr) - tunnel_hlen; - else + if (df) { + mtu = dst_mtu(&rt->dst) - (sizeof(struct iphdr) + tunnel_hlen); + mtu -= dev->type == ARPHRD_ETHER ? dev->hard_header_len : 0; + } else { mtu = skb_valid_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu; + } if (skb_valid_dst(skb)) skb_dst_update_pmtu_no_confirm(skb, mtu); @@ -511,8 +508,7 @@ static int tnl_update_pmtu(struct net_device *dev, struct sk_buff *skb, if (!skb_is_gso(skb) && (inner_iph->frag_off & htons(IP_DF)) && mtu < pkt_size) { - memset(IPCB(skb), 0, sizeof(*IPCB(skb))); - icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); + icmp_ndo_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); return -E2BIG; } } @@ -536,7 +532,7 @@ static int tnl_update_pmtu(struct net_device *dev, struct sk_buff *skb, if (!skb_is_gso(skb) && mtu >= IPV6_MIN_MTU && mtu < pkt_size) { - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); + icmpv6_ndo_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); return -E2BIG; } } @@ -574,7 +570,8 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, } ip_tunnel_init_flow(&fl4, proto, key->u.ipv4.dst, key->u.ipv4.src, tunnel_id_to_key32(key->tun_id), RT_TOS(tos), - 0, skb->mark, skb_get_hash(skb)); + dev_net(dev), 0, skb->mark, skb_get_hash(skb), + key->flow_flags); if (tunnel->encap.type != TUNNEL_ENCAP_NONE) goto tx_error; @@ -616,9 +613,6 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, ttl = ip4_dst_hoplimit(&rt->dst); } - if (!df && skb->protocol == htons(ETH_P_IP)) - df = inner_iph->frag_off & htons(IP_DF); - headroom += LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len; if (headroom > dev->needed_headroom) dev->needed_headroom = headroom; @@ -648,6 +642,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, const struct iphdr *inner_iph; unsigned int max_headroom; /* The extra header space needed */ struct rtable *rt = NULL; /* Route to the other host */ + __be16 payload_protocol; bool use_cache = false; struct flowi4 fl4; bool md = false; @@ -658,6 +653,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, inner_iph = (const struct iphdr *)skb_inner_network_header(skb); connected = (tunnel->parms.iph.daddr != 0); + payload_protocol = skb_protocol(skb, true); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); @@ -677,13 +673,12 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, dst = tun_info->key.u.ipv4.dst; md = true; connected = true; - } - else if (skb->protocol == htons(ETH_P_IP)) { + } else if (payload_protocol == htons(ETH_P_IP)) { rt = skb_rtable(skb); dst = rt_nexthop(rt, inner_iph->daddr); } #if IS_ENABLED(CONFIG_IPV6) - else if (skb->protocol == htons(ETH_P_IPV6)) { + else if (payload_protocol == htons(ETH_P_IPV6)) { const struct in6_addr *addr6; struct neighbour *neigh; bool do_tx_error_icmp; @@ -723,18 +718,19 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, tos = tnl_params->tos; if (tos & 0x1) { tos &= ~0x1; - if (skb->protocol == htons(ETH_P_IP)) { + if (payload_protocol == htons(ETH_P_IP)) { tos = inner_iph->tos; connected = false; - } else if (skb->protocol == htons(ETH_P_IPV6)) { + } else if (payload_protocol == htons(ETH_P_IPV6)) { tos = ipv6_get_dsfield((const struct ipv6hdr *)inner_iph); connected = false; } } ip_tunnel_init_flow(&fl4, protocol, dst, tnl_params->saddr, - tunnel->parms.o_key, RT_TOS(tos), tunnel->parms.link, - tunnel->fwmark, skb_get_hash(skb)); + tunnel->parms.o_key, RT_TOS(tos), + dev_net(dev), tunnel->parms.link, + tunnel->fwmark, skb_get_hash(skb), 0); if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0) goto tx_error; @@ -770,8 +766,11 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, goto tx_error; } - if (tnl_update_pmtu(dev, skb, rt, tnl_params->frag_off, inner_iph, - 0, 0, false)) { + df = tnl_params->frag_off; + if (payload_protocol == htons(ETH_P_IP) && !tunnel->ignore_df) + df |= (inner_iph->frag_off & htons(IP_DF)); + + if (tnl_update_pmtu(dev, skb, rt, df, inner_iph, 0, 0, false)) { ip_rt_put(rt); goto tx_error; } @@ -789,20 +788,16 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, tos = ip_tunnel_ecn_encap(tos, inner_iph, skb); ttl = tnl_params->ttl; if (ttl == 0) { - if (skb->protocol == htons(ETH_P_IP)) + if (payload_protocol == htons(ETH_P_IP)) ttl = inner_iph->ttl; #if IS_ENABLED(CONFIG_IPV6) - else if (skb->protocol == htons(ETH_P_IPV6)) + else if (payload_protocol == htons(ETH_P_IPV6)) ttl = ((const struct ipv6hdr *)inner_iph)->hop_limit; #endif else ttl = ip4_dst_hoplimit(&rt->dst); } - df = tnl_params->frag_off; - if (skb->protocol == htons(ETH_P_IP) && !tunnel->ignore_df) - df |= (inner_iph->frag_off&htons(IP_DF)); - max_headroom = LL_RESERVED_SPACE(rt->dst.dev) + sizeof(struct iphdr) + rt->dst.header_len + ip_encap_hlen(&tunnel->encap); if (max_headroom > dev->needed_headroom) @@ -842,7 +837,7 @@ static void ip_tunnel_update(struct ip_tunnel_net *itn, t->parms.i_key = p->i_key; t->parms.o_key = p->o_key; if (dev->type != ARPHRD_ETHER) { - memcpy(dev->dev_addr, &p->iph.saddr, 4); + __dev_addr_set(dev, &p->iph.saddr, 4); memcpy(dev->broadcast, &p->iph.daddr, 4); } ip_tunnel_add(itn, t); @@ -864,7 +859,7 @@ static void ip_tunnel_update(struct ip_tunnel_net *itn, netdev_state_change(dev); } -int ip_tunnel_ioctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) +int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) { int err = 0; struct ip_tunnel *t = netdev_priv(dev); @@ -964,13 +959,31 @@ int ip_tunnel_ioctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) done: return err; } -EXPORT_SYMBOL_GPL(ip_tunnel_ioctl); +EXPORT_SYMBOL_GPL(ip_tunnel_ctl); + +int ip_tunnel_siocdevprivate(struct net_device *dev, struct ifreq *ifr, + void __user *data, int cmd) +{ + struct ip_tunnel_parm p; + int err; + + if (copy_from_user(&p, data, sizeof(p))) + return -EFAULT; + err = dev->netdev_ops->ndo_tunnel_ctl(dev, &p, cmd); + if (!err && copy_to_user(data, &p, sizeof(p))) + return -EFAULT; + return err; +} +EXPORT_SYMBOL_GPL(ip_tunnel_siocdevprivate); int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict) { struct ip_tunnel *tunnel = netdev_priv(dev); int t_hlen = tunnel->hlen + sizeof(struct iphdr); - int max_mtu = IP_MAX_MTU - dev->hard_header_len - t_hlen; + int max_mtu = IP_MAX_MTU - t_hlen; + + if (dev->type == ARPHRD_ETHER) + max_mtu -= dev->hard_header_len; if (new_mtu < ETH_MIN_MTU) return -EINVAL; @@ -1054,7 +1067,7 @@ int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id, memset(&parms, 0, sizeof(parms)); if (devname) - strlcpy(parms.name, devname, IFNAMSIZ); + strscpy(parms.name, devname, IFNAMSIZ); rtnl_lock(); itn->fb_tunnel_dev = __ip_tunnel_create(net, ops, &parms); @@ -1147,10 +1160,12 @@ int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[], mtu = ip_tunnel_bind_dev(dev); if (tb[IFLA_MTU]) { - unsigned int max = IP_MAX_MTU - dev->hard_header_len - nt->hlen; + unsigned int max = IP_MAX_MTU - (nt->hlen + sizeof(struct iphdr)); + + if (dev->type == ARPHRD_ETHER) + max -= dev->hard_header_len; - mtu = clamp(dev->mtu, (unsigned int)ETH_MIN_MTU, - (unsigned int)(max - sizeof(struct iphdr))); + mtu = clamp(dev->mtu, (unsigned int)ETH_MIN_MTU, max); } err = dev_set_mtu(dev, mtu); @@ -1249,9 +1264,9 @@ void ip_tunnel_uninit(struct net_device *dev) struct ip_tunnel_net *itn; itn = net_generic(net, tunnel->ip_tnl_net_id); - /* fb_tunnel_dev will be unregisted in net-exit call. */ - if (itn->fb_tunnel_dev != dev) - ip_tunnel_del(itn, netdev_priv(dev)); + ip_tunnel_del(itn, netdev_priv(dev)); + if (itn->fb_tunnel_dev == dev) + WRITE_ONCE(itn->fb_tunnel_dev, NULL); dst_cache_reset(&tunnel->dst_cache); } diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 47f8b947eef1..92c02c886fe7 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -25,6 +25,7 @@ #include <net/protocol.h> #include <net/ip_tunnels.h> #include <net/ip6_tunnel.h> +#include <net/ip6_checksum.h> #include <net/arp.h> #include <net/checksum.h> #include <net/dsfield.h> @@ -184,35 +185,249 @@ int iptunnel_handle_offloads(struct sk_buff *skb, } EXPORT_SYMBOL_GPL(iptunnel_handle_offloads); -/* Often modified stats are per cpu, other are shared (netdev->stats) */ -void ip_tunnel_get_stats64(struct net_device *dev, - struct rtnl_link_stats64 *tot) +/** + * iptunnel_pmtud_build_icmp() - Build ICMP error message for PMTUD + * @skb: Original packet with L2 header + * @mtu: MTU value for ICMP error + * + * Return: length on success, negative error code if message couldn't be built. + */ +static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) +{ + const struct iphdr *iph = ip_hdr(skb); + struct icmphdr *icmph; + struct iphdr *niph; + struct ethhdr eh; + int len, err; + + if (!pskb_may_pull(skb, ETH_HLEN + sizeof(struct iphdr))) + return -EINVAL; + + skb_copy_bits(skb, skb_mac_offset(skb), &eh, ETH_HLEN); + pskb_pull(skb, ETH_HLEN); + skb_reset_network_header(skb); + + err = pskb_trim(skb, 576 - sizeof(*niph) - sizeof(*icmph)); + if (err) + return err; + + len = skb->len + sizeof(*icmph); + err = skb_cow(skb, sizeof(*niph) + sizeof(*icmph) + ETH_HLEN); + if (err) + return err; + + icmph = skb_push(skb, sizeof(*icmph)); + *icmph = (struct icmphdr) { + .type = ICMP_DEST_UNREACH, + .code = ICMP_FRAG_NEEDED, + .checksum = 0, + .un.frag.__unused = 0, + .un.frag.mtu = htons(mtu), + }; + icmph->checksum = ip_compute_csum(icmph, len); + skb_reset_transport_header(skb); + + niph = skb_push(skb, sizeof(*niph)); + *niph = (struct iphdr) { + .ihl = sizeof(*niph) / 4u, + .version = 4, + .tos = 0, + .tot_len = htons(len + sizeof(*niph)), + .id = 0, + .frag_off = htons(IP_DF), + .ttl = iph->ttl, + .protocol = IPPROTO_ICMP, + .saddr = iph->daddr, + .daddr = iph->saddr, + }; + ip_send_check(niph); + skb_reset_network_header(skb); + + skb->ip_summed = CHECKSUM_NONE; + + eth_header(skb, skb->dev, ntohs(eh.h_proto), eh.h_source, eh.h_dest, 0); + skb_reset_mac_header(skb); + + return skb->len; +} + +/** + * iptunnel_pmtud_check_icmp() - Trigger ICMP reply if needed and allowed + * @skb: Buffer being sent by encapsulation, L2 headers expected + * @mtu: Network MTU for path + * + * Return: 0 for no ICMP reply, length if built, negative value on error. + */ +static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) +{ + const struct icmphdr *icmph = icmp_hdr(skb); + const struct iphdr *iph = ip_hdr(skb); + + if (mtu < 576 || iph->frag_off != htons(IP_DF)) + return 0; + + if (ipv4_is_lbcast(iph->daddr) || ipv4_is_multicast(iph->daddr) || + ipv4_is_zeronet(iph->saddr) || ipv4_is_loopback(iph->saddr) || + ipv4_is_lbcast(iph->saddr) || ipv4_is_multicast(iph->saddr)) + return 0; + + if (iph->protocol == IPPROTO_ICMP && icmp_is_err(icmph->type)) + return 0; + + return iptunnel_pmtud_build_icmp(skb, mtu); +} + +#if IS_ENABLED(CONFIG_IPV6) +/** + * iptunnel_pmtud_build_icmpv6() - Build ICMPv6 error message for PMTUD + * @skb: Original packet with L2 header + * @mtu: MTU value for ICMPv6 error + * + * Return: length on success, negative error code if message couldn't be built. + */ +static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) +{ + const struct ipv6hdr *ip6h = ipv6_hdr(skb); + struct icmp6hdr *icmp6h; + struct ipv6hdr *nip6h; + struct ethhdr eh; + int len, err; + __wsum csum; + + if (!pskb_may_pull(skb, ETH_HLEN + sizeof(struct ipv6hdr))) + return -EINVAL; + + skb_copy_bits(skb, skb_mac_offset(skb), &eh, ETH_HLEN); + pskb_pull(skb, ETH_HLEN); + skb_reset_network_header(skb); + + err = pskb_trim(skb, IPV6_MIN_MTU - sizeof(*nip6h) - sizeof(*icmp6h)); + if (err) + return err; + + len = skb->len + sizeof(*icmp6h); + err = skb_cow(skb, sizeof(*nip6h) + sizeof(*icmp6h) + ETH_HLEN); + if (err) + return err; + + icmp6h = skb_push(skb, sizeof(*icmp6h)); + *icmp6h = (struct icmp6hdr) { + .icmp6_type = ICMPV6_PKT_TOOBIG, + .icmp6_code = 0, + .icmp6_cksum = 0, + .icmp6_mtu = htonl(mtu), + }; + skb_reset_transport_header(skb); + + nip6h = skb_push(skb, sizeof(*nip6h)); + *nip6h = (struct ipv6hdr) { + .priority = 0, + .version = 6, + .flow_lbl = { 0 }, + .payload_len = htons(len), + .nexthdr = IPPROTO_ICMPV6, + .hop_limit = ip6h->hop_limit, + .saddr = ip6h->daddr, + .daddr = ip6h->saddr, + }; + skb_reset_network_header(skb); + + csum = csum_partial(icmp6h, len, 0); + icmp6h->icmp6_cksum = csum_ipv6_magic(&nip6h->saddr, &nip6h->daddr, len, + IPPROTO_ICMPV6, csum); + + skb->ip_summed = CHECKSUM_NONE; + + eth_header(skb, skb->dev, ntohs(eh.h_proto), eh.h_source, eh.h_dest, 0); + skb_reset_mac_header(skb); + + return skb->len; +} + +/** + * iptunnel_pmtud_check_icmpv6() - Trigger ICMPv6 reply if needed and allowed + * @skb: Buffer being sent by encapsulation, L2 headers expected + * @mtu: Network MTU for path + * + * Return: 0 for no ICMPv6 reply, length if built, negative value on error. + */ +static int iptunnel_pmtud_check_icmpv6(struct sk_buff *skb, int mtu) { - int i; + const struct ipv6hdr *ip6h = ipv6_hdr(skb); + int stype = ipv6_addr_type(&ip6h->saddr); + u8 proto = ip6h->nexthdr; + __be16 frag_off; + int offset; - netdev_stats_to_stats64(tot, &dev->stats); + if (mtu < IPV6_MIN_MTU) + return 0; - for_each_possible_cpu(i) { - const struct pcpu_sw_netstats *tstats = - per_cpu_ptr(dev->tstats, i); - u64 rx_packets, rx_bytes, tx_packets, tx_bytes; - unsigned int start; + if (stype == IPV6_ADDR_ANY || stype == IPV6_ADDR_MULTICAST || + stype == IPV6_ADDR_LOOPBACK) + return 0; - do { - start = u64_stats_fetch_begin_irq(&tstats->syncp); - rx_packets = tstats->rx_packets; - tx_packets = tstats->tx_packets; - rx_bytes = tstats->rx_bytes; - tx_bytes = tstats->tx_bytes; - } while (u64_stats_fetch_retry_irq(&tstats->syncp, start)); + offset = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &proto, + &frag_off); + if (offset < 0 || (frag_off & htons(~0x7))) + return 0; - tot->rx_packets += rx_packets; - tot->tx_packets += tx_packets; - tot->rx_bytes += rx_bytes; - tot->tx_bytes += tx_bytes; + if (proto == IPPROTO_ICMPV6) { + struct icmp6hdr *icmp6h; + + if (!pskb_may_pull(skb, skb_network_header(skb) + + offset + 1 - skb->data)) + return 0; + + icmp6h = (struct icmp6hdr *)(skb_network_header(skb) + offset); + if (icmpv6_is_err(icmp6h->icmp6_type) || + icmp6h->icmp6_type == NDISC_REDIRECT) + return 0; } + + return iptunnel_pmtud_build_icmpv6(skb, mtu); } -EXPORT_SYMBOL_GPL(ip_tunnel_get_stats64); +#endif /* IS_ENABLED(CONFIG_IPV6) */ + +/** + * skb_tunnel_check_pmtu() - Check, update PMTU and trigger ICMP reply as needed + * @skb: Buffer being sent by encapsulation, L2 headers expected + * @encap_dst: Destination for tunnel encapsulation (outer IP) + * @headroom: Encapsulation header size, bytes + * @reply: Build matching ICMP or ICMPv6 message as a result + * + * L2 tunnel implementations that can carry IP and can be directly bridged + * (currently UDP tunnels) can't always rely on IP forwarding paths to handle + * PMTU discovery. In the bridged case, ICMP or ICMPv6 messages need to be built + * based on payload and sent back by the encapsulation itself. + * + * For routable interfaces, we just need to update the PMTU for the destination. + * + * Return: 0 if ICMP error not needed, length if built, negative value on error + */ +int skb_tunnel_check_pmtu(struct sk_buff *skb, struct dst_entry *encap_dst, + int headroom, bool reply) +{ + u32 mtu = dst_mtu(encap_dst) - headroom; + + if ((skb_is_gso(skb) && skb_gso_validate_network_len(skb, mtu)) || + (!skb_is_gso(skb) && (skb->len - skb_network_offset(skb)) <= mtu)) + return 0; + + skb_dst_update_pmtu_no_confirm(skb, mtu); + + if (!reply || skb->pkt_type == PACKET_HOST) + return 0; + + if (skb->protocol == htons(ETH_P_IP)) + return iptunnel_pmtud_check_icmp(skb, mtu); + +#if IS_ENABLED(CONFIG_IPV6) + if (skb->protocol == htons(ETH_P_IPV6)) + return iptunnel_pmtud_check_icmpv6(skb, mtu); +#endif + return 0; +} +EXPORT_SYMBOL(skb_tunnel_check_pmtu); static const struct nla_policy ip_tun_policy[LWTUNNEL_IP_MAX + 1] = { [LWTUNNEL_IP_UNSPEC] = { .strict_start_type = LWTUNNEL_IP_OPTS }, @@ -309,6 +524,7 @@ static int ip_tun_parse_opts_vxlan(struct nlattr *attr, attr = tb[LWTUNNEL_IP_OPT_VXLAN_GBP]; md->gbp = nla_get_u32(attr); + md->gbp &= VXLAN_GBP_MASK; info->key.tun_flags |= TUNNEL_VXLAN_OPT; } @@ -367,8 +583,9 @@ static int ip_tun_parse_opts_erspan(struct nlattr *attr, static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, struct netlink_ext_ack *extack) { - int err, rem, opt_len, opts_len = 0, type = 0; + int err, rem, opt_len, opts_len = 0; struct nlattr *nla; + __be16 type = 0; if (!attr) return 0; @@ -432,7 +649,7 @@ static int ip_tun_set_opts(struct nlattr *attr, struct ip_tunnel_info *info, return ip_tun_parse_opts(attr, info, extack); } -static int ip_tun_build_state(struct nlattr *attr, +static int ip_tun_build_state(struct net *net, struct nlattr *attr, unsigned int family, const void *cfg, struct lwtunnel_state **ts, struct netlink_ext_ack *extack) @@ -719,7 +936,7 @@ static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX + 1] = { [LWTUNNEL_IP6_OPTS] = { .type = NLA_NESTED }, }; -static int ip6_tun_build_state(struct nlattr *attr, +static int ip6_tun_build_state(struct net *net, struct nlattr *attr, unsigned int family, const void *cfg, struct lwtunnel_state **ts, struct netlink_ext_ack *extack) @@ -844,3 +1061,88 @@ void ip_tunnel_unneed_metadata(void) static_branch_dec(&ip_tunnel_metadata_cnt); } EXPORT_SYMBOL_GPL(ip_tunnel_unneed_metadata); + +/* Returns either the correct skb->protocol value, or 0 if invalid. */ +__be16 ip_tunnel_parse_protocol(const struct sk_buff *skb) +{ + if (skb_network_header(skb) >= skb->head && + (skb_network_header(skb) + sizeof(struct iphdr)) <= skb_tail_pointer(skb) && + ip_hdr(skb)->version == 4) + return htons(ETH_P_IP); + if (skb_network_header(skb) >= skb->head && + (skb_network_header(skb) + sizeof(struct ipv6hdr)) <= skb_tail_pointer(skb) && + ipv6_hdr(skb)->version == 6) + return htons(ETH_P_IPV6); + return 0; +} +EXPORT_SYMBOL(ip_tunnel_parse_protocol); + +const struct header_ops ip_tunnel_header_ops = { .parse_protocol = ip_tunnel_parse_protocol }; +EXPORT_SYMBOL(ip_tunnel_header_ops); + +/* This function returns true when ENCAP attributes are present in the nl msg */ +bool ip_tunnel_netlink_encap_parms(struct nlattr *data[], + struct ip_tunnel_encap *encap) +{ + bool ret = false; + + memset(encap, 0, sizeof(*encap)); + + if (!data) + return ret; + + if (data[IFLA_IPTUN_ENCAP_TYPE]) { + ret = true; + encap->type = nla_get_u16(data[IFLA_IPTUN_ENCAP_TYPE]); + } + + if (data[IFLA_IPTUN_ENCAP_FLAGS]) { + ret = true; + encap->flags = nla_get_u16(data[IFLA_IPTUN_ENCAP_FLAGS]); + } + + if (data[IFLA_IPTUN_ENCAP_SPORT]) { + ret = true; + encap->sport = nla_get_be16(data[IFLA_IPTUN_ENCAP_SPORT]); + } + + if (data[IFLA_IPTUN_ENCAP_DPORT]) { + ret = true; + encap->dport = nla_get_be16(data[IFLA_IPTUN_ENCAP_DPORT]); + } + + return ret; +} +EXPORT_SYMBOL_GPL(ip_tunnel_netlink_encap_parms); + +void ip_tunnel_netlink_parms(struct nlattr *data[], + struct ip_tunnel_parm *parms) +{ + if (data[IFLA_IPTUN_LINK]) + parms->link = nla_get_u32(data[IFLA_IPTUN_LINK]); + + if (data[IFLA_IPTUN_LOCAL]) + parms->iph.saddr = nla_get_be32(data[IFLA_IPTUN_LOCAL]); + + if (data[IFLA_IPTUN_REMOTE]) + parms->iph.daddr = nla_get_be32(data[IFLA_IPTUN_REMOTE]); + + if (data[IFLA_IPTUN_TTL]) { + parms->iph.ttl = nla_get_u8(data[IFLA_IPTUN_TTL]); + if (parms->iph.ttl) + parms->iph.frag_off = htons(IP_DF); + } + + if (data[IFLA_IPTUN_TOS]) + parms->iph.tos = nla_get_u8(data[IFLA_IPTUN_TOS]); + + if (!data[IFLA_IPTUN_PMTUDISC] || nla_get_u8(data[IFLA_IPTUN_PMTUDISC])) + parms->iph.frag_off = htons(IP_DF); + + if (data[IFLA_IPTUN_FLAGS]) + parms->i_flags = nla_get_be16(data[IFLA_IPTUN_FLAGS]); + + if (data[IFLA_IPTUN_PROTO]) + parms->iph.protocol = nla_get_u8(data[IFLA_IPTUN_PROTO]); +} +EXPORT_SYMBOL_GPL(ip_tunnel_netlink_parms); diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 37cddd18f282..8c2bd1d9ddce 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -91,16 +91,10 @@ static int vti_rcv_proto(struct sk_buff *skb) return vti_rcv(skb, 0, false); } -static int vti_rcv_tunnel(struct sk_buff *skb) -{ - return vti_rcv(skb, ip_hdr(skb)->saddr, true); -} - static int vti_rcv_cb(struct sk_buff *skb, int err) { unsigned short family; struct net_device *dev; - struct pcpu_sw_netstats *tstats; struct xfrm_state *x; const struct xfrm_mode *inner_mode; struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4; @@ -143,13 +137,7 @@ static int vti_rcv_cb(struct sk_buff *skb, int err) skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev))); skb->dev = dev; - - tstats = this_cpu_ptr(dev->tstats); - - u64_stats_update_begin(&tstats->syncp); - tstats->rx_packets++; - tstats->rx_bytes += skb->len; - u64_stats_update_end(&tstats->syncp); + dev_sw_netstats_rx_add(dev, skb->len); return 0; } @@ -187,26 +175,51 @@ static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev, int mtu; if (!dst) { - struct rtable *rt; - - fl->u.ip4.flowi4_oif = dev->ifindex; - fl->u.ip4.flowi4_flags |= FLOWI_FLAG_ANYSRC; - rt = __ip_route_output_key(dev_net(dev), &fl->u.ip4); - if (IS_ERR(rt)) { + switch (skb->protocol) { + case htons(ETH_P_IP): { + struct rtable *rt; + + fl->u.ip4.flowi4_oif = dev->ifindex; + fl->u.ip4.flowi4_flags |= FLOWI_FLAG_ANYSRC; + rt = __ip_route_output_key(dev_net(dev), &fl->u.ip4); + if (IS_ERR(rt)) { + dev->stats.tx_carrier_errors++; + goto tx_error_icmp; + } + dst = &rt->dst; + skb_dst_set(skb, dst); + break; + } +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + fl->u.ip6.flowi6_oif = dev->ifindex; + fl->u.ip6.flowi6_flags |= FLOWI_FLAG_ANYSRC; + dst = ip6_route_output(dev_net(dev), NULL, &fl->u.ip6); + if (dst->error) { + dst_release(dst); + dst = NULL; + dev->stats.tx_carrier_errors++; + goto tx_error_icmp; + } + skb_dst_set(skb, dst); + break; +#endif + default: dev->stats.tx_carrier_errors++; goto tx_error_icmp; } - dst = &rt->dst; - skb_dst_set(skb, dst); } dst_hold(dst); - dst = xfrm_lookup(tunnel->net, dst, fl, NULL, 0); + dst = xfrm_lookup_route(tunnel->net, dst, fl, NULL, 0); if (IS_ERR(dst)) { dev->stats.tx_carrier_errors++; goto tx_error_icmp; } + if (dst->flags & DST_XFRM_QUEUE) + goto xmit; + if (!vti_state_check(dst->xfrm, parms->iph.daddr, parms->iph.saddr)) { dev->stats.tx_carrier_errors++; dst_release(dst); @@ -225,19 +238,22 @@ static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev, if (skb->len > mtu) { skb_dst_update_pmtu_no_confirm(skb, mtu); if (skb->protocol == htons(ETH_P_IP)) { - icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, - htonl(mtu)); + if (!(ip_hdr(skb)->frag_off & htons(IP_DF))) + goto xmit; + icmp_ndo_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, + htonl(mtu)); } else { if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); + icmpv6_ndo_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); } dst_release(dst); goto tx_error; } +xmit: skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev))); skb_dst_set(skb, dst); skb->dev = skb_dst(skb)->dev; @@ -335,6 +351,7 @@ static int vti4_err(struct sk_buff *skb, u32 info) case ICMP_DEST_UNREACH: if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED) return 0; + break; case ICMP_REDIRECT: break; default: @@ -356,38 +373,31 @@ static int vti4_err(struct sk_buff *skb, u32 info) } static int -vti_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +vti_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) { int err = 0; - struct ip_tunnel_parm p; - - if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) - return -EFAULT; if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { - if (p.iph.version != 4 || p.iph.protocol != IPPROTO_IPIP || - p.iph.ihl != 5) + if (p->iph.version != 4 || p->iph.protocol != IPPROTO_IPIP || + p->iph.ihl != 5) return -EINVAL; } - if (!(p.i_flags & GRE_KEY)) - p.i_key = 0; - if (!(p.o_flags & GRE_KEY)) - p.o_key = 0; + if (!(p->i_flags & GRE_KEY)) + p->i_key = 0; + if (!(p->o_flags & GRE_KEY)) + p->o_key = 0; - p.i_flags = VTI_ISVTI; + p->i_flags = VTI_ISVTI; - err = ip_tunnel_ioctl(dev, &p, cmd); + err = ip_tunnel_ctl(dev, p, cmd); if (err) return err; if (cmd != SIOCDELTUNNEL) { - p.i_flags |= GRE_KEY; - p.o_flags |= GRE_KEY; + p->i_flags |= GRE_KEY; + p->o_flags |= GRE_KEY; } - - if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p))) - return -EFAULT; return 0; } @@ -395,15 +405,17 @@ static const struct net_device_ops vti_netdev_ops = { .ndo_init = vti_tunnel_init, .ndo_uninit = ip_tunnel_uninit, .ndo_start_xmit = vti_tunnel_xmit, - .ndo_do_ioctl = vti_tunnel_ioctl, + .ndo_siocdevprivate = ip_tunnel_siocdevprivate, .ndo_change_mtu = ip_tunnel_change_mtu, - .ndo_get_stats64 = ip_tunnel_get_stats64, + .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip_tunnel_get_iflink, + .ndo_tunnel_ctl = vti_tunnel_ctl, }; static void vti_tunnel_setup(struct net_device *dev) { dev->netdev_ops = &vti_netdev_ops; + dev->header_ops = &ip_tunnel_header_ops; dev->type = ARPHRD_TUNNEL; ip_tunnel_setup(dev, vti_net_id); } @@ -413,7 +425,7 @@ static int vti_tunnel_init(struct net_device *dev) struct ip_tunnel *tunnel = netdev_priv(dev); struct iphdr *iph = &tunnel->parms.iph; - memcpy(dev->dev_addr, &iph->saddr, 4); + __dev_addr_set(dev, &iph->saddr, 4); memcpy(dev->broadcast, &iph->daddr, 4); dev->flags = IFF_NOARP; @@ -458,11 +470,31 @@ static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = { .priority = 100, }; -static struct xfrm_tunnel ipip_handler __read_mostly = { +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) +static int vti_rcv_tunnel(struct sk_buff *skb) +{ + XFRM_SPI_SKB_CB(skb)->family = AF_INET; + XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr); + + return vti_input(skb, IPPROTO_IPIP, ip_hdr(skb)->saddr, 0, false); +} + +static struct xfrm_tunnel vti_ipip_handler __read_mostly = { + .handler = vti_rcv_tunnel, + .cb_handler = vti_rcv_cb, + .err_handler = vti4_err, + .priority = 0, +}; + +#if IS_ENABLED(CONFIG_IPV6) +static struct xfrm_tunnel vti_ipip6_handler __read_mostly = { .handler = vti_rcv_tunnel, + .cb_handler = vti_rcv_cb, .err_handler = vti4_err, .priority = 0, }; +#endif +#endif static int __net_init vti_init_net(struct net *net) { @@ -632,10 +664,17 @@ static int __init vti_init(void) if (err < 0) goto xfrm_proto_comp_failed; +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) msg = "ipip tunnel"; - err = xfrm4_tunnel_register(&ipip_handler, AF_INET); + err = xfrm4_tunnel_register(&vti_ipip_handler, AF_INET); + if (err < 0) + goto xfrm_tunnel_ipip_failed; +#if IS_ENABLED(CONFIG_IPV6) + err = xfrm4_tunnel_register(&vti_ipip6_handler, AF_INET6); if (err < 0) - goto xfrm_tunnel_failed; + goto xfrm_tunnel_ipip6_failed; +#endif +#endif msg = "netlink interface"; err = rtnl_link_register(&vti_link_ops); @@ -645,8 +684,14 @@ static int __init vti_init(void) return err; rtnl_link_failed: - xfrm4_tunnel_deregister(&ipip_handler, AF_INET); -xfrm_tunnel_failed: +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) +#if IS_ENABLED(CONFIG_IPV6) + xfrm4_tunnel_deregister(&vti_ipip6_handler, AF_INET6); +xfrm_tunnel_ipip6_failed: +#endif + xfrm4_tunnel_deregister(&vti_ipip_handler, AF_INET); +xfrm_tunnel_ipip_failed: +#endif xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP); xfrm_proto_comp_failed: xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH); @@ -662,7 +707,12 @@ pernet_dev_failed: static void __exit vti_fini(void) { rtnl_link_unregister(&vti_link_ops); - xfrm4_tunnel_deregister(&ipip_handler, AF_INET); +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) +#if IS_ENABLED(CONFIG_IPV6) + xfrm4_tunnel_deregister(&vti_ipip6_handler, AF_INET6); +#endif + xfrm4_tunnel_deregister(&vti_ipip_handler, AF_INET); +#endif xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP); xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH); xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP); diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c index 59bfa3825810..5a4fb2539b08 100644 --- a/net/ipv4/ipcomp.c +++ b/net/ipv4/ipcomp.c @@ -31,6 +31,7 @@ static int ipcomp4_err(struct sk_buff *skb, u32 info) case ICMP_DEST_UNREACH: if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED) return 0; + break; case ICMP_REDIRECT: break; default: @@ -72,6 +73,7 @@ static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x) t->props.flags = x->props.flags; t->props.extra_flags = x->props.extra_flags; memcpy(&t->mark, &x->mark, sizeof(t->mark)); + t->if_id = x->if_id; if (xfrm_init_state(t)) goto error; @@ -115,7 +117,8 @@ out: return err; } -static int ipcomp4_init_state(struct xfrm_state *x) +static int ipcomp4_init_state(struct xfrm_state *x, + struct netlink_ext_ack *extack) { int err = -EINVAL; @@ -127,17 +130,20 @@ static int ipcomp4_init_state(struct xfrm_state *x) x->props.header_len += sizeof(struct iphdr); break; default: + NL_SET_ERR_MSG(extack, "Unsupported XFRM mode for IPcomp"); goto out; } - err = ipcomp_init_state(x); + err = ipcomp_init_state(x, extack); if (err) goto out; if (x->props.mode == XFRM_MODE_TUNNEL) { err = ipcomp_tunnel_attach(x); - if (err) + if (err) { + NL_SET_ERR_MSG(extack, "Kernel error: failed to initialize the associated state"); goto out; + } } err = 0; @@ -151,7 +157,6 @@ static int ipcomp4_rcv_cb(struct sk_buff *skb, int err) } static const struct xfrm_type ipcomp_type = { - .description = "IPCOMP4", .owner = THIS_MODULE, .proto = IPPROTO_COMP, .init_state = ipcomp4_init_state, diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 4438f6b12335..e90bc0aa85c7 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -61,7 +61,6 @@ #include <linux/export.h> #include <net/net_namespace.h> #include <net/arp.h> -#include <net/dsa.h> #include <net/ip.h> #include <net/ipconfig.h> #include <net/route.h> @@ -218,9 +217,9 @@ static int __init ic_open_devs(void) last = &ic_first_dev; rtnl_lock(); - /* bring loopback and DSA master network devices up first */ + /* bring loopback device up first */ for_each_netdev(&init_net, dev) { - if (!(dev->flags & IFF_LOOPBACK) && !netdev_uses_dsa(dev)) + if (!(dev->flags & IFF_LOOPBACK)) continue; if (dev_change_flags(dev, dev->flags | IFF_UP, NULL) < 0) pr_err("IP-Config: Failed to open %s\n", dev->name); @@ -263,6 +262,11 @@ static int __init ic_open_devs(void) dev->name, able, d->xid); } } + /* Devices with a complex topology like SFP ethernet interfaces needs + * the rtnl_lock at init. The carrier wait-loop must therefore run + * without holding it. + */ + rtnl_unlock(); /* no point in waiting if we could not bring up at least one device */ if (!ic_first_dev) @@ -275,9 +279,13 @@ static int __init ic_open_devs(void) msecs_to_jiffies(carrier_timeout * 1000))) { int wait, elapsed; + rtnl_lock(); for_each_netdev(&init_net, dev) - if (ic_is_init_dev(dev) && netif_carrier_ok(dev)) + if (ic_is_init_dev(dev) && netif_carrier_ok(dev)) { + rtnl_unlock(); goto have_carrier; + } + rtnl_unlock(); msleep(1); @@ -290,7 +298,6 @@ static int __init ic_open_devs(void) next_msg = jiffies + msecs_to_jiffies(20000); } have_carrier: - rtnl_unlock(); *last = NULL; @@ -305,17 +312,34 @@ have_carrier: return 0; } +/* Close all network interfaces except the one we've autoconfigured, and its + * lowers, in case it's a stacked virtual interface. + */ static void __init ic_close_devs(void) { + struct net_device *selected_dev = ic_dev ? ic_dev->dev : NULL; struct ic_device *d, *next; struct net_device *dev; rtnl_lock(); next = ic_first_dev; while ((d = next)) { + bool bring_down = (d != ic_dev); + struct net_device *lower; + struct list_head *iter; + next = d->next; dev = d->dev; - if (d != ic_dev && !netdev_uses_dsa(dev)) { + + if (selected_dev) { + netdev_for_each_lower_dev(selected_dev, lower, iter) { + if (dev == lower) { + bring_down = false; + break; + } + } + } + if (bring_down) { pr_debug("IP-Config: Downing %s\n", dev->name); dev_change_flags(dev, d->flags, NULL); } @@ -870,7 +894,7 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d /* - * Copy BOOTP-supplied string if not already set. + * Copy BOOTP-supplied string */ static int __init ic_bootp_string(char *dest, char *src, int len, int max) { @@ -919,12 +943,15 @@ static void __init ic_do_bootp_ext(u8 *ext) } break; case 12: /* Host name */ - ic_bootp_string(utsname()->nodename, ext+1, *ext, - __NEW_UTS_LEN); - ic_host_name_set = 1; + if (!ic_host_name_set) { + ic_bootp_string(utsname()->nodename, ext+1, *ext, + __NEW_UTS_LEN); + ic_host_name_set = 1; + } break; case 15: /* Domain name (DNS) */ - ic_bootp_string(ic_domain, ext+1, *ext, sizeof(ic_domain)); + if (!ic_domain[0]) + ic_bootp_string(ic_domain, ext+1, *ext, sizeof(ic_domain)); break; case 17: /* Root path */ if (!root_server_path[0]) @@ -1407,6 +1434,7 @@ __be32 __init root_nfs_parse_addr(char *name) static int __init wait_for_devices(void) { int i; + bool try_init_devs = true; for (i = 0; i < DEVICE_WAIT_MAX; i++) { struct net_device *dev; @@ -1425,6 +1453,11 @@ static int __init wait_for_devices(void) rtnl_unlock(); if (found) return 0; + if (try_init_devs && + (ROOT_DEV == Root_NFS || ROOT_DEV == Root_CIFS)) { + try_init_devs = false; + wait_for_init_devices_probe(); + } ssleep(1); } return -ENODEV; @@ -1441,7 +1474,7 @@ static int __init ip_auto_config(void) int retries = CONF_OPEN_RETRIES; #endif int err; - unsigned int i; + unsigned int i, count; /* Initialise all name servers and NTP servers to NONE (but only if the * "ip=" or "nfsaddrs=" kernel command line parameters weren't decoded, @@ -1575,7 +1608,7 @@ static int __init ip_auto_config(void) if (ic_dev_mtu) pr_cont(", mtu=%d", ic_dev_mtu); /* Name servers (if any): */ - for (i = 0; i < CONF_NAMESERVERS_MAX; i++) { + for (i = 0, count = 0; i < CONF_NAMESERVERS_MAX; i++) { if (ic_nameservers[i] != NONE) { if (i == 0) pr_info(" nameserver%u=%pI4", @@ -1583,12 +1616,14 @@ static int __init ip_auto_config(void) else pr_cont(", nameserver%u=%pI4", i, &ic_nameservers[i]); + + count++; } - if (i + 1 == CONF_NAMESERVERS_MAX) + if ((i + 1 == CONF_NAMESERVERS_MAX) && count > 0) pr_cont("\n"); } /* NTP servers (if any): */ - for (i = 0; i < CONF_NTP_SERVERS_MAX; i++) { + for (i = 0, count = 0; i < CONF_NTP_SERVERS_MAX; i++) { if (ic_ntp_servers[i] != NONE) { if (i == 0) pr_info(" ntpserver%u=%pI4", @@ -1596,8 +1631,10 @@ static int __init ip_auto_config(void) else pr_cont(", ntpserver%u=%pI4", i, &ic_ntp_servers[i]); + + count++; } - if (i + 1 == CONF_NTP_SERVERS_MAX) + if ((i + 1 == CONF_NTP_SERVERS_MAX) && count > 0) pr_cont("\n"); } #endif /* !SILENT */ @@ -1621,7 +1658,7 @@ late_initcall(ip_auto_config); /* * Decode any IP configuration options in the "ip=" or "nfsaddrs=" kernel - * command line parameter. See Documentation/filesystems/nfs/nfsroot.txt. + * command line parameter. See Documentation/admin-guide/nfs/nfsroot.rst. */ static int __init ic_proto_name(char *name) { @@ -1728,15 +1765,15 @@ static int __init ip_auto_config_setup(char *addrs) case 4: if ((dp = strchr(ip, '.'))) { *dp++ = '\0'; - strlcpy(utsname()->domainname, dp, + strscpy(utsname()->domainname, dp, sizeof(utsname()->domainname)); } - strlcpy(utsname()->nodename, ip, + strscpy(utsname()->nodename, ip, sizeof(utsname()->nodename)); ic_host_name_set = 1; break; case 5: - strlcpy(user_dev_name, ip, sizeof(user_dev_name)); + strscpy(user_dev_name, ip, sizeof(user_dev_name)); break; case 6: if (ic_proto_name(ip) == 0 && @@ -1783,7 +1820,7 @@ __setup("nfsaddrs=", nfsaddrs_config_setup); static int __init vendor_class_identifier_setup(char *addrs) { - if (strlcpy(vendor_class_identifier, addrs, + if (strscpy(vendor_class_identifier, addrs, sizeof(vendor_class_identifier)) >= sizeof(vendor_class_identifier)) pr_warn("DHCP: vendorclass too long, truncated to \"%s\"\n", diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 2f01cf6fa0de..180f9daf5bec 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -242,6 +242,8 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto) if (!tun_dst) return 0; } + skb_reset_mac_header(skb); + return ip_tunnel_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error); } @@ -327,41 +329,29 @@ static bool ipip_tunnel_ioctl_verify_protocol(u8 ipproto) } static int -ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +ipip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) { - int err = 0; - struct ip_tunnel_parm p; - - if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) - return -EFAULT; - if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { - if (p.iph.version != 4 || - !ipip_tunnel_ioctl_verify_protocol(p.iph.protocol) || - p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF))) + if (p->iph.version != 4 || + !ipip_tunnel_ioctl_verify_protocol(p->iph.protocol) || + p->iph.ihl != 5 || (p->iph.frag_off & htons(~IP_DF))) return -EINVAL; } - p.i_key = p.o_key = 0; - p.i_flags = p.o_flags = 0; - err = ip_tunnel_ioctl(dev, &p, cmd); - if (err) - return err; - - if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p))) - return -EFAULT; - - return 0; + p->i_key = p->o_key = 0; + p->i_flags = p->o_flags = 0; + return ip_tunnel_ctl(dev, p, cmd); } static const struct net_device_ops ipip_netdev_ops = { .ndo_init = ipip_tunnel_init, .ndo_uninit = ip_tunnel_uninit, .ndo_start_xmit = ipip_tunnel_xmit, - .ndo_do_ioctl = ipip_tunnel_ioctl, + .ndo_siocdevprivate = ip_tunnel_siocdevprivate, .ndo_change_mtu = ip_tunnel_change_mtu, - .ndo_get_stats64 = ip_tunnel_get_stats64, + .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip_tunnel_get_iflink, + .ndo_tunnel_ctl = ipip_tunnel_ctl, }; #define IPIP_FEATURES (NETIF_F_SG | \ @@ -373,6 +363,7 @@ static const struct net_device_ops ipip_netdev_ops = { static void ipip_tunnel_setup(struct net_device *dev) { dev->netdev_ops = &ipip_netdev_ops; + dev->header_ops = &ip_tunnel_header_ops; dev->type = ARPHRD_TUNNEL; dev->flags = IFF_NOARP; @@ -389,7 +380,7 @@ static int ipip_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4); + __dev_addr_set(dev, &tunnel->parms.iph.saddr, 4); memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4); tunnel->tun_hlen = 0; @@ -426,29 +417,7 @@ static void ipip_netlink_parms(struct nlattr *data[], if (!data) return; - if (data[IFLA_IPTUN_LINK]) - parms->link = nla_get_u32(data[IFLA_IPTUN_LINK]); - - if (data[IFLA_IPTUN_LOCAL]) - parms->iph.saddr = nla_get_in_addr(data[IFLA_IPTUN_LOCAL]); - - if (data[IFLA_IPTUN_REMOTE]) - parms->iph.daddr = nla_get_in_addr(data[IFLA_IPTUN_REMOTE]); - - if (data[IFLA_IPTUN_TTL]) { - parms->iph.ttl = nla_get_u8(data[IFLA_IPTUN_TTL]); - if (parms->iph.ttl) - parms->iph.frag_off = htons(IP_DF); - } - - if (data[IFLA_IPTUN_TOS]) - parms->iph.tos = nla_get_u8(data[IFLA_IPTUN_TOS]); - - if (data[IFLA_IPTUN_PROTO]) - parms->iph.protocol = nla_get_u8(data[IFLA_IPTUN_PROTO]); - - if (!data[IFLA_IPTUN_PMTUDISC] || nla_get_u8(data[IFLA_IPTUN_PMTUDISC])) - parms->iph.frag_off = htons(IP_DF); + ip_tunnel_netlink_parms(data, parms); if (data[IFLA_IPTUN_COLLECT_METADATA]) *collect_md = true; @@ -457,40 +426,6 @@ static void ipip_netlink_parms(struct nlattr *data[], *fwmark = nla_get_u32(data[IFLA_IPTUN_FWMARK]); } -/* This function returns true when ENCAP attributes are present in the nl msg */ -static bool ipip_netlink_encap_parms(struct nlattr *data[], - struct ip_tunnel_encap *ipencap) -{ - bool ret = false; - - memset(ipencap, 0, sizeof(*ipencap)); - - if (!data) - return ret; - - if (data[IFLA_IPTUN_ENCAP_TYPE]) { - ret = true; - ipencap->type = nla_get_u16(data[IFLA_IPTUN_ENCAP_TYPE]); - } - - if (data[IFLA_IPTUN_ENCAP_FLAGS]) { - ret = true; - ipencap->flags = nla_get_u16(data[IFLA_IPTUN_ENCAP_FLAGS]); - } - - if (data[IFLA_IPTUN_ENCAP_SPORT]) { - ret = true; - ipencap->sport = nla_get_be16(data[IFLA_IPTUN_ENCAP_SPORT]); - } - - if (data[IFLA_IPTUN_ENCAP_DPORT]) { - ret = true; - ipencap->dport = nla_get_be16(data[IFLA_IPTUN_ENCAP_DPORT]); - } - - return ret; -} - static int ipip_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) @@ -500,7 +435,7 @@ static int ipip_newlink(struct net *src_net, struct net_device *dev, struct ip_tunnel_encap ipencap; __u32 fwmark = 0; - if (ipip_netlink_encap_parms(data, &ipencap)) { + if (ip_tunnel_netlink_encap_parms(data, &ipencap)) { int err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) @@ -521,7 +456,7 @@ static int ipip_changelink(struct net_device *dev, struct nlattr *tb[], bool collect_md; __u32 fwmark = t->fwmark; - if (ipip_netlink_encap_parms(data, &ipencap)) { + if (ip_tunnel_netlink_encap_parms(data, &ipencap)) { int err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) @@ -698,7 +633,7 @@ out: rtnl_link_failed: #if IS_ENABLED(CONFIG_MPLS) - xfrm4_tunnel_deregister(&mplsip_handler, AF_INET); + xfrm4_tunnel_deregister(&mplsip_handler, AF_MPLS); xfrm_tunnel_mplsip_failed: #endif diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 6e68def66822..e04544ac4b45 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -77,7 +77,12 @@ struct ipmr_result { * Note that the changes are semaphored via rtnl_lock. */ -static DEFINE_RWLOCK(mrt_lock); +static DEFINE_SPINLOCK(mrt_lock); + +static struct net_device *vif_dev_read(const struct vif_device *vif) +{ + return rcu_dereference(vif->dev); +} /* Multicast router control variables */ @@ -100,17 +105,19 @@ static void ipmr_free_table(struct mr_table *mrt); static void ip_mr_forward(struct net *net, struct mr_table *mrt, struct net_device *dev, struct sk_buff *skb, struct mfc_cache *cache, int local); -static int ipmr_cache_report(struct mr_table *mrt, +static int ipmr_cache_report(const struct mr_table *mrt, struct sk_buff *pkt, vifi_t vifi, int assert); static void mroute_netlink_event(struct mr_table *mrt, struct mfc_cache *mfc, int cmd); -static void igmpmsg_netlink_event(struct mr_table *mrt, struct sk_buff *pkt); +static void igmpmsg_netlink_event(const struct mr_table *mrt, struct sk_buff *pkt); static void mroute_clean_tables(struct mr_table *mrt, int flags); static void ipmr_expire_process(struct timer_list *t); #ifdef CONFIG_IP_MROUTE_MULTIPLE_TABLES -#define ipmr_for_each_table(mrt, net) \ - list_for_each_entry_rcu(mrt, &net->ipv4.mr_tables, list) +#define ipmr_for_each_table(mrt, net) \ + list_for_each_entry_rcu(mrt, &net->ipv4.mr_tables, list, \ + lockdep_rtnl_is_held() || \ + list_empty(&net->ipv4.mr_tables)) static struct mr_table *ipmr_mr_table_iter(struct net *net, struct mr_table *mrt) @@ -193,10 +200,6 @@ static int ipmr_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) return 1; } -static const struct nla_policy ipmr_rule_policy[FRA_MAX + 1] = { - FRA_GENERIC_POLICY, -}; - static int ipmr_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct fib_rule_hdr *frh, struct nlattr **tb, struct netlink_ext_ack *extack) @@ -229,7 +232,6 @@ static const struct fib_rules_ops __net_initconst ipmr_rules_ops_template = { .compare = ipmr_rule_compare, .fill = ipmr_rule_fill, .nlgroup = RTNLGRP_IPV4_RULE, - .policy = ipmr_rule_policy, .owner = THIS_MODULE, }; @@ -259,7 +261,9 @@ static int __net_init ipmr_rules_init(struct net *net) return 0; err2: + rtnl_lock(); ipmr_free_table(mrt); + rtnl_unlock(); err1: fib_rules_unregister(ops); return err; @@ -269,13 +273,12 @@ static void __net_exit ipmr_rules_exit(struct net *net) { struct mr_table *mrt, *next; - rtnl_lock(); + ASSERT_RTNL(); list_for_each_entry_safe(mrt, next, &net->ipv4.mr_tables, list) { list_del(&mrt->list); ipmr_free_table(mrt); } fib_rules_unregister(net->ipv4.mr_rules_ops); - rtnl_unlock(); } static int ipmr_rules_dump(struct net *net, struct notifier_block *nb, @@ -331,10 +334,9 @@ static int __net_init ipmr_rules_init(struct net *net) static void __net_exit ipmr_rules_exit(struct net *net) { - rtnl_lock(); + ASSERT_RTNL(); ipmr_free_table(net->ipv4.mrt); net->ipv4.mrt = NULL; - rtnl_unlock(); } static int ipmr_rules_dump(struct net *net, struct notifier_block *nb, @@ -359,7 +361,7 @@ static inline int ipmr_hash_cmp(struct rhashtable_compare_arg *arg, const void *ptr) { const struct mfc_cache_cmp_arg *cmparg = arg->key; - struct mfc_cache *c = (struct mfc_cache *)ptr; + const struct mfc_cache *c = ptr; return cmparg->mfc_mcastgrp != c->mfc_mcastgrp || cmparg->mfc_origin != c->mfc_origin; @@ -419,37 +421,6 @@ static void ipmr_free_table(struct mr_table *mrt) /* Service routines creating virtual interfaces: DVMRP tunnels and PIMREG */ -static void ipmr_del_tunnel(struct net_device *dev, struct vifctl *v) -{ - struct net *net = dev_net(dev); - - dev_close(dev); - - dev = __dev_get_by_name(net, "tunl0"); - if (dev) { - const struct net_device_ops *ops = dev->netdev_ops; - struct ifreq ifr; - struct ip_tunnel_parm p; - - memset(&p, 0, sizeof(p)); - p.iph.daddr = v->vifc_rmt_addr.s_addr; - p.iph.saddr = v->vifc_lcl_addr.s_addr; - p.iph.version = 4; - p.iph.ihl = 5; - p.iph.protocol = IPPROTO_IPIP; - sprintf(p.name, "dvmrp%d", v->vifc_vifi); - ifr.ifr_ifru.ifru_data = (__force void __user *)&p; - - if (ops->ndo_do_ioctl) { - mm_segment_t oldfs = get_fs(); - - set_fs(KERNEL_DS); - ops->ndo_do_ioctl(dev, &ifr, SIOCDELTUNNEL); - set_fs(oldfs); - } - } -} - /* Initialize ipmr pimreg/tunnel in_device */ static bool ipmr_init_vif_indev(const struct net_device *dev) { @@ -469,51 +440,52 @@ static bool ipmr_init_vif_indev(const struct net_device *dev) static struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) { - struct net_device *dev; - - dev = __dev_get_by_name(net, "tunl0"); - - if (dev) { - const struct net_device_ops *ops = dev->netdev_ops; - int err; - struct ifreq ifr; - struct ip_tunnel_parm p; + struct net_device *tunnel_dev, *new_dev; + struct ip_tunnel_parm p = { }; + int err; - memset(&p, 0, sizeof(p)); - p.iph.daddr = v->vifc_rmt_addr.s_addr; - p.iph.saddr = v->vifc_lcl_addr.s_addr; - p.iph.version = 4; - p.iph.ihl = 5; - p.iph.protocol = IPPROTO_IPIP; - sprintf(p.name, "dvmrp%d", v->vifc_vifi); - ifr.ifr_ifru.ifru_data = (__force void __user *)&p; + tunnel_dev = __dev_get_by_name(net, "tunl0"); + if (!tunnel_dev) + goto out; - if (ops->ndo_do_ioctl) { - mm_segment_t oldfs = get_fs(); + p.iph.daddr = v->vifc_rmt_addr.s_addr; + p.iph.saddr = v->vifc_lcl_addr.s_addr; + p.iph.version = 4; + p.iph.ihl = 5; + p.iph.protocol = IPPROTO_IPIP; + sprintf(p.name, "dvmrp%d", v->vifc_vifi); - set_fs(KERNEL_DS); - err = ops->ndo_do_ioctl(dev, &ifr, SIOCADDTUNNEL); - set_fs(oldfs); - } else { - err = -EOPNOTSUPP; - } - dev = NULL; + if (!tunnel_dev->netdev_ops->ndo_tunnel_ctl) + goto out; + err = tunnel_dev->netdev_ops->ndo_tunnel_ctl(tunnel_dev, &p, + SIOCADDTUNNEL); + if (err) + goto out; - if (err == 0 && - (dev = __dev_get_by_name(net, p.name)) != NULL) { - dev->flags |= IFF_MULTICAST; - if (!ipmr_init_vif_indev(dev)) - goto failure; - if (dev_open(dev, NULL)) - goto failure; - dev_hold(dev); - } - } - return dev; + new_dev = __dev_get_by_name(net, p.name); + if (!new_dev) + goto out; -failure: - unregister_netdevice(dev); - return NULL; + new_dev->flags |= IFF_MULTICAST; + if (!ipmr_init_vif_indev(new_dev)) + goto out_unregister; + if (dev_open(new_dev, NULL)) + goto out_unregister; + dev_hold(new_dev); + err = dev_set_allmulti(new_dev, 1); + if (err) { + dev_close(new_dev); + tunnel_dev->netdev_ops->ndo_tunnel_ctl(tunnel_dev, &p, + SIOCDELTUNNEL); + dev_put(new_dev); + new_dev = ERR_PTR(err); + } + return new_dev; + +out_unregister: + unregister_netdevice(new_dev); +out: + return ERR_PTR(-ENOBUFS); } #if defined(CONFIG_IP_PIMSM_V1) || defined(CONFIG_IP_PIMSM_V2) @@ -534,11 +506,15 @@ static netdev_tx_t reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) return err; } - read_lock(&mrt_lock); dev->stats.tx_bytes += skb->len; dev->stats.tx_packets++; - ipmr_cache_report(mrt, skb, mrt->mroute_reg_vif_num, IGMPMSG_WHOLEPKT); - read_unlock(&mrt_lock); + rcu_read_lock(); + + /* Pairs with WRITE_ONCE() in vif_add() and vif_delete() */ + ipmr_cache_report(mrt, skb, READ_ONCE(mrt->mroute_reg_vif_num), + IGMPMSG_WHOLEPKT); + + rcu_read_unlock(); kfree_skb(skb); return NETDEV_TX_OK; } @@ -605,6 +581,7 @@ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, { struct net_device *reg_dev = NULL; struct iphdr *encap; + int vif_num; encap = (struct iphdr *)(skb_transport_header(skb) + pimlen); /* Check that: @@ -617,11 +594,10 @@ static int __pim_rcv(struct mr_table *mrt, struct sk_buff *skb, ntohs(encap->tot_len) + pimlen > skb->len) return 1; - read_lock(&mrt_lock); - if (mrt->mroute_reg_vif_num >= 0) - reg_dev = mrt->vif_table[mrt->mroute_reg_vif_num].dev; - read_unlock(&mrt_lock); - + /* Pairs with WRITE_ONCE() in vif_add()/vid_delete() */ + vif_num = READ_ONCE(mrt->mroute_reg_vif_num); + if (vif_num >= 0) + reg_dev = vif_dev_read(&mrt->vif_table[vif_num]); if (!reg_dev) return 1; @@ -647,10 +623,11 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) static int call_ipmr_vif_entry_notifiers(struct net *net, enum fib_event_type event_type, struct vif_device *vif, + struct net_device *vif_dev, vifi_t vif_index, u32 tb_id) { return mr_call_vif_notifiers(net, RTNL_FAMILY_IPMR, event_type, - vif, vif_index, tb_id, + vif, vif_dev, vif_index, tb_id, &net->ipv4.ipmr_seq); } @@ -664,7 +641,10 @@ static int call_ipmr_mfc_entry_notifiers(struct net *net, /** * vif_delete - Delete a VIF entry + * @mrt: Table to delete from + * @vifi: VIF identifier to delete * @notify: Set to 1, if the caller is a notifier_call + * @head: if unregistering the VIF, place it on this queue */ static int vif_delete(struct mr_table *mrt, int vifi, int notify, struct list_head *head) @@ -679,22 +659,19 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify, v = &mrt->vif_table[vifi]; - if (VIF_EXISTS(mrt, vifi)) - call_ipmr_vif_entry_notifiers(net, FIB_EVENT_VIF_DEL, v, vifi, - mrt->id); - - write_lock_bh(&mrt_lock); - dev = v->dev; - v->dev = NULL; - - if (!dev) { - write_unlock_bh(&mrt_lock); + dev = rtnl_dereference(v->dev); + if (!dev) return -EADDRNOTAVAIL; - } - if (vifi == mrt->mroute_reg_vif_num) - mrt->mroute_reg_vif_num = -1; + spin_lock(&mrt_lock); + call_ipmr_vif_entry_notifiers(net, FIB_EVENT_VIF_DEL, v, dev, + vifi, mrt->id); + RCU_INIT_POINTER(v->dev, NULL); + if (vifi == mrt->mroute_reg_vif_num) { + /* Pairs with READ_ONCE() in ipmr_cache_report() and reg_vif_xmit() */ + WRITE_ONCE(mrt->mroute_reg_vif_num, -1); + } if (vifi + 1 == mrt->maxvif) { int tmp; @@ -702,10 +679,10 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify, if (VIF_EXISTS(mrt, tmp)) break; } - mrt->maxvif = tmp+1; + WRITE_ONCE(mrt->maxvif, tmp + 1); } - write_unlock_bh(&mrt_lock); + spin_unlock(&mrt_lock); dev_set_allmulti(dev, -1); @@ -721,7 +698,7 @@ static int vif_delete(struct mr_table *mrt, int vifi, int notify, if (v->flags & (VIFF_TUNNEL | VIFF_REGISTER) && !notify) unregister_netdevice_queue(dev, head); - dev_put(dev); + netdev_put(dev, &v->dev_tracker); return 0; } @@ -807,7 +784,7 @@ out: spin_unlock(&mfc_unres_lock); } -/* Fill oifs list. It is called under write locked mrt_lock. */ +/* Fill oifs list. It is called under locked mrt_lock. */ static void ipmr_update_thresholds(struct mr_table *mrt, struct mr_mfc *cache, unsigned char *ttls) { @@ -865,14 +842,8 @@ static int vif_add(struct net *net, struct mr_table *mrt, break; case VIFF_TUNNEL: dev = ipmr_new_tunnel(net, vifc); - if (!dev) - return -ENOBUFS; - err = dev_set_allmulti(dev, 1); - if (err) { - ipmr_del_tunnel(dev, vifc); - dev_put(dev); - return err; - } + if (IS_ERR(dev)) + return PTR_ERR(dev); break; case VIFF_USE_IFINDEX: case 0: @@ -925,14 +896,18 @@ static int vif_add(struct net *net, struct mr_table *mrt, v->remote = vifc->vifc_rmt_addr.s_addr; /* And finish update writing critical data */ - write_lock_bh(&mrt_lock); - v->dev = dev; - if (v->flags & VIFF_REGISTER) - mrt->mroute_reg_vif_num = vifi; + spin_lock(&mrt_lock); + rcu_assign_pointer(v->dev, dev); + netdev_tracker_alloc(dev, &v->dev_tracker, GFP_ATOMIC); + if (v->flags & VIFF_REGISTER) { + /* Pairs with READ_ONCE() in ipmr_cache_report() and reg_vif_xmit() */ + WRITE_ONCE(mrt->mroute_reg_vif_num, vifi); + } if (vifi+1 > mrt->maxvif) - mrt->maxvif = vifi+1; - write_unlock_bh(&mrt_lock); - call_ipmr_vif_entry_notifiers(net, FIB_EVENT_VIF_ADD, v, vifi, mrt->id); + WRITE_ONCE(mrt->maxvif, vifi + 1); + spin_unlock(&mrt_lock); + call_ipmr_vif_entry_notifiers(net, FIB_EVENT_VIF_ADD, v, dev, + vifi, mrt->id); return 0; } @@ -1029,16 +1004,18 @@ static void ipmr_cache_resolve(struct net *net, struct mr_table *mrt, rtnl_unicast(skb, net, NETLINK_CB(skb).portid); } else { + rcu_read_lock(); ip_mr_forward(net, mrt, skb->dev, skb, c, 0); + rcu_read_unlock(); } } } /* Bounce a cache query up to mrouted and netlink. * - * Called under mrt_lock. + * Called under rcu_read_lock(). */ -static int ipmr_cache_report(struct mr_table *mrt, +static int ipmr_cache_report(const struct mr_table *mrt, struct sk_buff *pkt, vifi_t vifi, int assert) { const int ihl = ip_hdrlen(pkt); @@ -1069,10 +1046,16 @@ static int ipmr_cache_report(struct mr_table *mrt, memcpy(msg, skb_network_header(pkt), sizeof(struct iphdr)); msg->im_msgtype = assert; msg->im_mbz = 0; - if (assert == IGMPMSG_WRVIFWHOLE) + if (assert == IGMPMSG_WRVIFWHOLE) { msg->im_vif = vifi; - else - msg->im_vif = mrt->mroute_reg_vif_num; + msg->im_vif_hi = vifi >> 8; + } else { + /* Pairs with WRITE_ONCE() in vif_add() and vif_delete() */ + int vif_num = READ_ONCE(mrt->mroute_reg_vif_num); + + msg->im_vif = vif_num; + msg->im_vif_hi = vif_num >> 8; + } ip_hdr(skb)->ihl = sizeof(struct iphdr) >> 2; ip_hdr(skb)->tot_len = htons(ntohs(ip_hdr(pkt)->tot_len) + sizeof(struct iphdr)); @@ -1085,6 +1068,7 @@ static int ipmr_cache_report(struct mr_table *mrt, ip_hdr(skb)->protocol = 0; msg = (struct igmpmsg *)skb_network_header(skb); msg->im_vif = vifi; + msg->im_vif_hi = vifi >> 8; skb_dst_set(skb, dst_clone(skb_dst(pkt))); /* Add our header */ igmp = skb_put(skb, sizeof(struct igmphdr)); @@ -1095,10 +1079,8 @@ static int ipmr_cache_report(struct mr_table *mrt, skb->transport_header = skb->network_header; } - rcu_read_lock(); mroute_sk = rcu_dereference(mrt->mroute_sk); if (!mroute_sk) { - rcu_read_unlock(); kfree_skb(skb); return -EINVAL; } @@ -1107,7 +1089,7 @@ static int ipmr_cache_report(struct mr_table *mrt, /* Deliver to mrouted */ ret = sock_queue_rcv_skb(mroute_sk, skb); - rcu_read_unlock(); + if (ret < 0) { net_warn_ratelimited("mroute: pending queue full, dropping entries\n"); kfree_skb(skb); @@ -1117,6 +1099,7 @@ static int ipmr_cache_report(struct mr_table *mrt, } /* Queue a packet for resolution. It gets locked cache entry! */ +/* Called under rcu_read_lock() */ static int ipmr_cache_unresolved(struct mr_table *mrt, vifi_t vifi, struct sk_buff *skb, struct net_device *dev) { @@ -1229,12 +1212,12 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, mfc->mfcc_mcastgrp.s_addr, parent); rcu_read_unlock(); if (c) { - write_lock_bh(&mrt_lock); + spin_lock(&mrt_lock); c->_c.mfc_parent = mfc->mfcc_parent; ipmr_update_thresholds(mrt, &c->_c, mfc->mfcc_ttls); if (!mrtsock) c->_c.mfc_flags |= MFC_STATIC; - write_unlock_bh(&mrt_lock); + spin_unlock(&mrt_lock); call_ipmr_mfc_entry_notifiers(net, FIB_EVENT_ENTRY_REPLACE, c, mrt->id); mroute_netlink_event(mrt, c, RTM_NEWROUTE); @@ -1344,7 +1327,7 @@ static void mroute_clean_tables(struct mr_table *mrt, int flags) } /* called from ip_ra_control(), before an RCU grace period, - * we dont need to call synchronize_rcu() here + * we don't need to call synchronize_rcu() here */ static void mrtsock_destruct(struct sock *sk) { @@ -1372,7 +1355,7 @@ static void mrtsock_destruct(struct sock *sk) * MOSPF/PIM router set up we can clean this up. */ -int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, +int ip_mroute_setsockopt(struct sock *sk, int optname, sockptr_t optval, unsigned int optlen) { struct net *net = sock_net(sk); @@ -1444,7 +1427,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, ret = -EINVAL; break; } - if (copy_from_user(&vif, optval, sizeof(vif))) { + if (copy_from_sockptr(&vif, optval, sizeof(vif))) { ret = -EFAULT; break; } @@ -1465,14 +1448,14 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, case MRT_ADD_MFC: case MRT_DEL_MFC: parent = -1; - /* fall through */ + fallthrough; case MRT_ADD_MFC_PROXY: case MRT_DEL_MFC_PROXY: if (optlen != sizeof(mfc)) { ret = -EINVAL; break; } - if (copy_from_user(&mfc, optval, sizeof(mfc))) { + if (copy_from_sockptr(&mfc, optval, sizeof(mfc))) { ret = -EFAULT; break; } @@ -1490,7 +1473,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, ret = -EINVAL; break; } - if (get_user(val, (int __user *)optval)) { + if (copy_from_sockptr(&val, optval, sizeof(val))) { ret = -EFAULT; break; } @@ -1502,7 +1485,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, ret = -EINVAL; break; } - if (get_user(val, (int __user *)optval)) { + if (copy_from_sockptr(&val, optval, sizeof(val))) { ret = -EFAULT; break; } @@ -1517,7 +1500,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, ret = -EINVAL; break; } - if (get_user(val, (int __user *)optval)) { + if (copy_from_sockptr(&val, optval, sizeof(val))) { ret = -EFAULT; break; } @@ -1539,7 +1522,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, ret = -EINVAL; break; } - if (get_user(uval, (u32 __user *)optval)) { + if (copy_from_sockptr(&uval, optval, sizeof(uval))) { ret = -EFAULT; break; } @@ -1565,7 +1548,8 @@ out: } /* Getsock opt support for the multicast routing system. */ -int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int __user *optlen) +int ip_mroute_getsockopt(struct sock *sk, int optname, sockptr_t optval, + sockptr_t optlen) { int olr; int val; @@ -1596,14 +1580,14 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int return -ENOPROTOOPT; } - if (get_user(olr, optlen)) + if (copy_from_sockptr(&olr, optlen, sizeof(int))) return -EFAULT; olr = min_t(unsigned int, olr, sizeof(int)); if (olr < 0) return -EINVAL; - if (put_user(olr, optlen)) + if (copy_to_sockptr(optlen, &olr, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &val, olr)) + if (copy_to_sockptr(optval, &val, olr)) return -EFAULT; return 0; } @@ -1629,20 +1613,20 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) if (vr.vifi >= mrt->maxvif) return -EINVAL; vr.vifi = array_index_nospec(vr.vifi, mrt->maxvif); - read_lock(&mrt_lock); + rcu_read_lock(); vif = &mrt->vif_table[vr.vifi]; if (VIF_EXISTS(mrt, vr.vifi)) { - vr.icount = vif->pkt_in; - vr.ocount = vif->pkt_out; - vr.ibytes = vif->bytes_in; - vr.obytes = vif->bytes_out; - read_unlock(&mrt_lock); + vr.icount = READ_ONCE(vif->pkt_in); + vr.ocount = READ_ONCE(vif->pkt_out); + vr.ibytes = READ_ONCE(vif->bytes_in); + vr.obytes = READ_ONCE(vif->bytes_out); + rcu_read_unlock(); if (copy_to_user(arg, &vr, sizeof(vr))) return -EFAULT; return 0; } - read_unlock(&mrt_lock); + rcu_read_unlock(); return -EADDRNOTAVAIL; case SIOCGETSGCNT: if (copy_from_user(&sr, arg, sizeof(sr))) @@ -1704,20 +1688,20 @@ int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg) if (vr.vifi >= mrt->maxvif) return -EINVAL; vr.vifi = array_index_nospec(vr.vifi, mrt->maxvif); - read_lock(&mrt_lock); + rcu_read_lock(); vif = &mrt->vif_table[vr.vifi]; if (VIF_EXISTS(mrt, vr.vifi)) { - vr.icount = vif->pkt_in; - vr.ocount = vif->pkt_out; - vr.ibytes = vif->bytes_in; - vr.obytes = vif->bytes_out; - read_unlock(&mrt_lock); + vr.icount = READ_ONCE(vif->pkt_in); + vr.ocount = READ_ONCE(vif->pkt_out); + vr.ibytes = READ_ONCE(vif->bytes_in); + vr.obytes = READ_ONCE(vif->bytes_out); + rcu_read_unlock(); if (copy_to_user(arg, &vr, sizeof(vr))) return -EFAULT; return 0; } - read_unlock(&mrt_lock); + rcu_read_unlock(); return -EADDRNOTAVAIL; case SIOCGETSGCNT: if (copy_from_user(&sr, arg, sizeof(sr))) @@ -1757,7 +1741,7 @@ static int ipmr_device_event(struct notifier_block *this, unsigned long event, v ipmr_for_each_table(mrt, net) { v = &mrt->vif_table[0]; for (ct = 0; ct < mrt->maxvif; ct++, v++) { - if (v->dev == dev) + if (rcu_access_pointer(v->dev) == dev) vif_delete(mrt, ct, 1, NULL); } } @@ -1835,26 +1819,28 @@ static bool ipmr_forward_offloaded(struct sk_buff *skb, struct mr_table *mrt, } #endif -/* Processing handlers for ipmr_forward */ +/* Processing handlers for ipmr_forward, under rcu_read_lock() */ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, int in_vifi, struct sk_buff *skb, int vifi) { const struct iphdr *iph = ip_hdr(skb); struct vif_device *vif = &mrt->vif_table[vifi]; + struct net_device *vif_dev; struct net_device *dev; struct rtable *rt; struct flowi4 fl4; int encap = 0; - if (!vif->dev) + vif_dev = vif_dev_read(vif); + if (!vif_dev) goto out_free; if (vif->flags & VIFF_REGISTER) { - vif->pkt_out++; - vif->bytes_out += skb->len; - vif->dev->stats.tx_bytes += skb->len; - vif->dev->stats.tx_packets++; + WRITE_ONCE(vif->pkt_out, vif->pkt_out + 1); + WRITE_ONCE(vif->bytes_out, vif->bytes_out + skb->len); + vif_dev->stats.tx_bytes += skb->len; + vif_dev->stats.tx_packets++; ipmr_cache_report(mrt, skb, vifi, IGMPMSG_WHOLEPKT); goto out_free; } @@ -1899,8 +1885,8 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, goto out_free; } - vif->pkt_out++; - vif->bytes_out += skb->len; + WRITE_ONCE(vif->pkt_out, vif->pkt_out + 1); + WRITE_ONCE(vif->bytes_out, vif->bytes_out + skb->len); skb_dst_drop(skb); skb_dst_set(skb, &rt->dst); @@ -1912,8 +1898,8 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, if (vif->flags & VIFF_TUNNEL) { ip_encap(net, skb, vif->local, vif->remote); /* FIXME: extra output firewall step used to be here. --RR */ - vif->dev->stats.tx_packets++; - vif->dev->stats.tx_bytes += skb->len; + vif_dev->stats.tx_packets++; + vif_dev->stats.tx_bytes += skb->len; } IPCB(skb)->flags |= IPSKB_FORWARDED; @@ -1937,18 +1923,20 @@ out_free: kfree_skb(skb); } -static int ipmr_find_vif(struct mr_table *mrt, struct net_device *dev) +/* Called with mrt_lock or rcu_read_lock() */ +static int ipmr_find_vif(const struct mr_table *mrt, struct net_device *dev) { int ct; - - for (ct = mrt->maxvif-1; ct >= 0; ct--) { - if (mrt->vif_table[ct].dev == dev) + /* Pairs with WRITE_ONCE() in vif_delete()/vif_add() */ + for (ct = READ_ONCE(mrt->maxvif) - 1; ct >= 0; ct--) { + if (rcu_access_pointer(mrt->vif_table[ct].dev) == dev) break; } return ct; } /* "local" means that we should preserve one skb (for local delivery) */ +/* Called uner rcu_read_lock() */ static void ip_mr_forward(struct net *net, struct mr_table *mrt, struct net_device *dev, struct sk_buff *skb, struct mfc_cache *c, int local) @@ -1965,7 +1953,7 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, if (c->mfc_origin == htonl(INADDR_ANY) && true_vifi >= 0) { struct mfc_cache *cache_proxy; - /* For an (*,G) entry, we only check that the incomming + /* For an (*,G) entry, we only check that the incoming * interface is part of the static tree. */ cache_proxy = mr_mfc_find_any_parent(mrt, vif); @@ -1975,7 +1963,7 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, } /* Wrong interface: drop packet and (maybe) send PIM assert. */ - if (mrt->vif_table[vif].dev != dev) { + if (rcu_access_pointer(mrt->vif_table[vif].dev) != dev) { if (rt_is_output_route(skb_rtable(skb))) { /* It is our own packet, looped back. * Very complicated situation... @@ -2014,8 +2002,10 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, } forward: - mrt->vif_table[vif].pkt_in++; - mrt->vif_table[vif].bytes_in += skb->len; + WRITE_ONCE(mrt->vif_table[vif].pkt_in, + mrt->vif_table[vif].pkt_in + 1); + WRITE_ONCE(mrt->vif_table[vif].bytes_in, + mrt->vif_table[vif].bytes_in + skb->len); /* Forward the frame */ if (c->mfc_origin == htonl(INADDR_ANY) && @@ -2146,7 +2136,7 @@ int ip_mr_input(struct sk_buff *skb) raw_rcv(mroute_sk, skb); return 0; } - } + } } /* already under rcu_read_lock() */ @@ -2171,22 +2161,14 @@ int ip_mr_input(struct sk_buff *skb) skb = skb2; } - read_lock(&mrt_lock); vif = ipmr_find_vif(mrt, dev); - if (vif >= 0) { - int err2 = ipmr_cache_unresolved(mrt, vif, skb, dev); - read_unlock(&mrt_lock); - - return err2; - } - read_unlock(&mrt_lock); + if (vif >= 0) + return ipmr_cache_unresolved(mrt, vif, skb, dev); kfree_skb(skb); return -ENODEV; } - read_lock(&mrt_lock); ip_mr_forward(net, mrt, dev, skb, cache, local); - read_unlock(&mrt_lock); if (local) return ip_local_deliver(skb); @@ -2283,18 +2265,15 @@ int ipmr_get_route(struct net *net, struct sk_buff *skb, int vif = -1; dev = skb->dev; - read_lock(&mrt_lock); if (dev) vif = ipmr_find_vif(mrt, dev); if (vif < 0) { - read_unlock(&mrt_lock); rcu_read_unlock(); return -ENODEV; } skb2 = skb_realloc_headroom(skb, sizeof(struct iphdr)); if (!skb2) { - read_unlock(&mrt_lock); rcu_read_unlock(); return -ENOMEM; } @@ -2308,14 +2287,11 @@ int ipmr_get_route(struct net *net, struct sk_buff *skb, iph->daddr = daddr; iph->version = 0; err = ipmr_cache_unresolved(mrt, vif, skb2, dev); - read_unlock(&mrt_lock); rcu_read_unlock(); return err; } - read_lock(&mrt_lock); err = mr_fill_mroute(mrt, skb, &cache->_c, rtm); - read_unlock(&mrt_lock); rcu_read_unlock(); return err; } @@ -2427,6 +2403,7 @@ static size_t igmpmsg_netlink_msgsize(size_t payloadlen) + nla_total_size(4) /* IPMRA_CREPORT_VIF_ID */ + nla_total_size(4) /* IPMRA_CREPORT_SRC_ADDR */ + nla_total_size(4) /* IPMRA_CREPORT_DST_ADDR */ + + nla_total_size(4) /* IPMRA_CREPORT_TABLE */ /* IPMRA_CREPORT_PKT */ + nla_total_size(payloadlen) ; @@ -2434,7 +2411,7 @@ static size_t igmpmsg_netlink_msgsize(size_t payloadlen) return len; } -static void igmpmsg_netlink_event(struct mr_table *mrt, struct sk_buff *pkt) +static void igmpmsg_netlink_event(const struct mr_table *mrt, struct sk_buff *pkt) { struct net *net = read_pnet(&mrt->net); struct nlmsghdr *nlh; @@ -2458,11 +2435,12 @@ static void igmpmsg_netlink_event(struct mr_table *mrt, struct sk_buff *pkt) rtgenm = nlmsg_data(nlh); rtgenm->rtgen_family = RTNL_FAMILY_IPMR; if (nla_put_u8(skb, IPMRA_CREPORT_MSGTYPE, msg->im_msgtype) || - nla_put_u32(skb, IPMRA_CREPORT_VIF_ID, msg->im_vif) || + nla_put_u32(skb, IPMRA_CREPORT_VIF_ID, msg->im_vif | (msg->im_vif_hi << 8)) || nla_put_in_addr(skb, IPMRA_CREPORT_SRC_ADDR, msg->im_src.s_addr) || nla_put_in_addr(skb, IPMRA_CREPORT_DST_ADDR, - msg->im_dst.s_addr)) + msg->im_dst.s_addr) || + nla_put_u32(skb, IPMRA_CREPORT_TABLE, mrt->id)) goto nla_put_failure; nla = nla_reserve(skb, IPMRA_CREPORT_PKT, payloadlen); @@ -2611,7 +2589,7 @@ static int ipmr_rtm_dumproute(struct sk_buff *skb, struct netlink_callback *cb) mrt = ipmr_get_table(sock_net(skb->sk), filter.table_id); if (!mrt) { - if (filter.dump_all_families) + if (rtnl_msg_family(cb->nlh) != RTNL_FAMILY_IPMR) return skb->len; NL_SET_ERR_MSG(cb->extack, "ipv4: MR table does not exist"); @@ -2773,18 +2751,21 @@ static bool ipmr_fill_table(struct mr_table *mrt, struct sk_buff *skb) static bool ipmr_fill_vif(struct mr_table *mrt, u32 vifid, struct sk_buff *skb) { + struct net_device *vif_dev; struct nlattr *vif_nest; struct vif_device *vif; + vif = &mrt->vif_table[vifid]; + vif_dev = rtnl_dereference(vif->dev); /* if the VIF doesn't exist just continue */ - if (!VIF_EXISTS(mrt, vifid)) + if (!vif_dev) return true; - vif = &mrt->vif_table[vifid]; vif_nest = nla_nest_start_noflag(skb, IPMRA_VIF); if (!vif_nest) return false; - if (nla_put_u32(skb, IPMRA_VIFA_IFINDEX, vif->dev->ifindex) || + + if (nla_put_u32(skb, IPMRA_VIFA_IFINDEX, vif_dev->ifindex) || nla_put_u32(skb, IPMRA_VIFA_VIF_ID, vifid) || nla_put_u16(skb, IPMRA_VIFA_FLAGS, vif->flags) || nla_put_u64_64bit(skb, IPMRA_VIFA_BYTES_IN, vif->bytes_in, @@ -2916,7 +2897,7 @@ out: */ static void *ipmr_vif_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(mrt_lock) + __acquires(RCU) { struct mr_vif_iter *iter = seq->private; struct net *net = seq_file_net(seq); @@ -2928,14 +2909,14 @@ static void *ipmr_vif_seq_start(struct seq_file *seq, loff_t *pos) iter->mrt = mrt; - read_lock(&mrt_lock); + rcu_read_lock(); return mr_vif_seq_start(seq, pos); } static void ipmr_vif_seq_stop(struct seq_file *seq, void *v) - __releases(mrt_lock) + __releases(RCU) { - read_unlock(&mrt_lock); + rcu_read_unlock(); } static int ipmr_vif_seq_show(struct seq_file *seq, void *v) @@ -2948,9 +2929,11 @@ static int ipmr_vif_seq_show(struct seq_file *seq, void *v) "Interface BytesIn PktsIn BytesOut PktsOut Flags Local Remote\n"); } else { const struct vif_device *vif = v; - const char *name = vif->dev ? - vif->dev->name : "none"; + const struct net_device *vif_dev; + const char *name; + vif_dev = vif_dev_read(vif); + name = vif_dev ? vif_dev->name : "none"; seq_printf(seq, "%2td %-10s %8ld %7ld %8ld %7ld %05X %08X %08X\n", vif - mrt->vif_table, @@ -3032,7 +3015,6 @@ static const struct seq_operations ipmr_mfc_seq_ops = { #ifdef CONFIG_IP_PIMSM_V2 static const struct net_protocol pim_protocol = { .handler = pim_rcv, - .netns_ok = 1, }; #endif @@ -3047,7 +3029,7 @@ static int ipmr_dump(struct net *net, struct notifier_block *nb, struct netlink_ext_ack *extack) { return mr_dump(net, nb, RTNL_FAMILY_IPMR, ipmr_rules_dump, - ipmr_mr_table_iter, &mrt_lock, extack); + ipmr_mr_table_iter, extack); } static const struct fib_notifier_ops ipmr_notifier_ops_template = { @@ -3105,7 +3087,9 @@ static int __net_init ipmr_net_init(struct net *net) proc_cache_fail: remove_proc_entry("ip_mr_vif", net->proc_net); proc_vif_fail: + rtnl_lock(); ipmr_rules_exit(net); + rtnl_unlock(); #endif ipmr_rules_fail: ipmr_notifier_exit(net); @@ -3120,12 +3104,22 @@ static void __net_exit ipmr_net_exit(struct net *net) remove_proc_entry("ip_mr_vif", net->proc_net); #endif ipmr_notifier_exit(net); - ipmr_rules_exit(net); +} + +static void __net_exit ipmr_net_exit_batch(struct list_head *net_list) +{ + struct net *net; + + rtnl_lock(); + list_for_each_entry(net, net_list, exit_list) + ipmr_rules_exit(net); + rtnl_unlock(); } static struct pernet_operations ipmr_net_ops = { .init = ipmr_net_init, .exit = ipmr_net_exit, + .exit_batch = ipmr_net_exit_batch, }; int __init ip_mr_init(void) diff --git a/net/ipv4/ipmr_base.c b/net/ipv4/ipmr_base.c index aa8738a91210..271dc03fc6db 100644 --- a/net/ipv4/ipmr_base.c +++ b/net/ipv4/ipmr_base.c @@ -13,7 +13,7 @@ void vif_device_init(struct vif_device *v, unsigned short flags, unsigned short get_iflink_mask) { - v->dev = NULL; + RCU_INIT_POINTER(v->dev, NULL); v->bytes_in = 0; v->bytes_out = 0; v->pkt_in = 0; @@ -208,6 +208,7 @@ EXPORT_SYMBOL(mr_mfc_seq_next); int mr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, struct mr_mfc *c, struct rtmsg *rtm) { + struct net_device *vif_dev; struct rta_mfc_stats mfcs; struct nlattr *mp_attr; struct rtnexthop *nhp; @@ -220,10 +221,13 @@ int mr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, return -ENOENT; } - if (VIF_EXISTS(mrt, c->mfc_parent) && - nla_put_u32(skb, RTA_IIF, - mrt->vif_table[c->mfc_parent].dev->ifindex) < 0) + rcu_read_lock(); + vif_dev = rcu_dereference(mrt->vif_table[c->mfc_parent].dev); + if (vif_dev && nla_put_u32(skb, RTA_IIF, vif_dev->ifindex) < 0) { + rcu_read_unlock(); return -EMSGSIZE; + } + rcu_read_unlock(); if (c->mfc_flags & MFC_OFFLOAD) rtm->rtm_flags |= RTNH_F_OFFLOAD; @@ -232,23 +236,27 @@ int mr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, if (!mp_attr) return -EMSGSIZE; + rcu_read_lock(); for (ct = c->mfc_un.res.minvif; ct < c->mfc_un.res.maxvif; ct++) { - if (VIF_EXISTS(mrt, ct) && c->mfc_un.res.ttls[ct] < 255) { - struct vif_device *vif; + struct vif_device *vif = &mrt->vif_table[ct]; + + vif_dev = rcu_dereference(vif->dev); + if (vif_dev && c->mfc_un.res.ttls[ct] < 255) { nhp = nla_reserve_nohdr(skb, sizeof(*nhp)); if (!nhp) { + rcu_read_unlock(); nla_nest_cancel(skb, mp_attr); return -EMSGSIZE; } nhp->rtnh_flags = 0; nhp->rtnh_hops = c->mfc_un.res.ttls[ct]; - vif = &mrt->vif_table[ct]; - nhp->rtnh_ifindex = vif->dev->ifindex; + nhp->rtnh_ifindex = vif_dev->ifindex; nhp->rtnh_len = sizeof(*nhp); } } + rcu_read_unlock(); nla_nest_end(skb, mp_attr); @@ -275,13 +283,14 @@ static bool mr_mfc_uses_dev(const struct mr_table *mrt, int ct; for (ct = c->mfc_un.res.minvif; ct < c->mfc_un.res.maxvif; ct++) { - if (VIF_EXISTS(mrt, ct) && c->mfc_un.res.ttls[ct] < 255) { - const struct vif_device *vif; - - vif = &mrt->vif_table[ct]; - if (vif->dev == dev) - return true; - } + const struct net_device *vif_dev; + const struct vif_device *vif; + + vif = &mrt->vif_table[ct]; + vif_dev = rcu_access_pointer(vif->dev); + if (vif_dev && c->mfc_un.res.ttls[ct] < 255 && + vif_dev == dev) + return true; } return false; } @@ -390,7 +399,6 @@ int mr_dump(struct net *net, struct notifier_block *nb, unsigned short family, struct netlink_ext_ack *extack), struct mr_table *(*mr_iter)(struct net *net, struct mr_table *mrt), - rwlock_t *mrt_lock, struct netlink_ext_ack *extack) { struct mr_table *mrt; @@ -402,22 +410,25 @@ int mr_dump(struct net *net, struct notifier_block *nb, unsigned short family, for (mrt = mr_iter(net, NULL); mrt; mrt = mr_iter(net, mrt)) { struct vif_device *v = &mrt->vif_table[0]; + struct net_device *vif_dev; struct mr_mfc *mfc; int vifi; /* Notifiy on table VIF entries */ - read_lock(mrt_lock); + rcu_read_lock(); for (vifi = 0; vifi < mrt->maxvif; vifi++, v++) { - if (!v->dev) + vif_dev = rcu_dereference(v->dev); + if (!vif_dev) continue; err = mr_call_vif_notifier(nb, family, - FIB_EVENT_VIF_ADD, - v, vifi, mrt->id, extack); + FIB_EVENT_VIF_ADD, v, + vif_dev, vifi, + mrt->id, extack); if (err) break; } - read_unlock(mrt_lock); + rcu_read_unlock(); if (err) return err; diff --git a/net/ipv4/metrics.c b/net/ipv4/metrics.c index 3205d5f7c8c9..25ea6ac44db9 100644 --- a/net/ipv4/metrics.c +++ b/net/ipv4/metrics.c @@ -31,7 +31,7 @@ static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx, if (type == RTAX_CC_ALGO) { char tmp[TCP_CA_NAME_MAX]; - nla_strlcpy(tmp, nla, sizeof(tmp)); + nla_strscpy(tmp, nla, sizeof(tmp)); val = tcp_ca_get_key_by_name(net, tmp, &ecn_ca); if (val == TCP_CA_UNSPEC) { NL_SET_ERR_MSG(extack, "Unknown tcp congestion algorithm"); diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index a058213b77a7..bd135165482a 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -17,17 +17,20 @@ #include <net/netfilter/nf_queue.h> /* route_me_harder function, used by iptable_nat, iptable_mangle + ip_queue */ -int ip_route_me_harder(struct net *net, struct sk_buff *skb, unsigned int addr_type) +int ip_route_me_harder(struct net *net, struct sock *sk, struct sk_buff *skb, unsigned int addr_type) { const struct iphdr *iph = ip_hdr(skb); struct rtable *rt; struct flowi4 fl4 = {}; __be32 saddr = iph->saddr; - const struct sock *sk = skb_to_full_sk(skb); - __u8 flags = sk ? inet_sk_flowi_flags(sk) : 0; + __u8 flags; struct net_device *dev = skb_dst(skb)->dev; + struct flow_keys flkeys; unsigned int hh_len; + sk = sk_to_full_sk(sk); + flags = sk ? inet_sk_flowi_flags(sk) : 0; + if (addr_type == RTN_UNSPEC) addr_type = inet_addr_type_dev_table(net, dev, saddr); if (addr_type == RTN_LOCAL || addr_type == RTN_UNICAST) @@ -42,10 +45,10 @@ int ip_route_me_harder(struct net *net, struct sk_buff *skb, unsigned int addr_t fl4.saddr = saddr; fl4.flowi4_tos = RT_TOS(iph->tos); fl4.flowi4_oif = sk ? sk->sk_bound_dev_if : 0; - if (!fl4.flowi4_oif) - fl4.flowi4_oif = l3mdev_master_ifindex(dev); + fl4.flowi4_l3mdev = l3mdev_master_ifindex(dev); fl4.flowi4_mark = skb->mark; fl4.flowi4_flags = flags; + fib4_rules_early_flow_dissect(net, skb, &fl4, &flkeys); rt = ip_route_output_key(net, &fl4); if (IS_ERR(rt)) return PTR_ERR(rt); diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index f17b402111ce..aab384126f61 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -58,14 +58,6 @@ config NF_TABLES_ARP endif # NF_TABLES -config NF_FLOW_TABLE_IPV4 - tristate "Netfilter flow table IPv4 module" - depends on NF_FLOW_TABLE - help - This option adds the flow table IPv4 support. - - To compile it as a module, choose M here. - config NF_DUP_IPV4 tristate "Netfilter IPv4 packet duplication to alternate destination" depends on !NF_CONNTRACK || NF_CONNTRACK @@ -76,12 +68,18 @@ config NF_DUP_IPV4 config NF_LOG_ARP tristate "ARP packet logging" default m if NETFILTER_ADVANCED=n - select NF_LOG_COMMON + select NF_LOG_SYSLOG + help + This is a backwards-compat option for the user's convenience + (e.g. when running oldconfig). It selects CONFIG_NF_LOG_SYSLOG. config NF_LOG_IPV4 tristate "IPv4 packet logging" default m if NETFILTER_ADVANCED=n - select NF_LOG_COMMON + select NF_LOG_SYSLOG + help + This is a backwards-compat option for the user's convenience + (e.g. when running oldconfig). It selects CONFIG_NF_LOG_SYSLOG. config NF_REJECT_IPV4 tristate "IPv4 packet rejection" @@ -94,7 +92,7 @@ config NF_NAT_SNMP_BASIC depends on NETFILTER_ADVANCED default NF_NAT && NF_CONNTRACK_SNMP select ASN1 - ---help--- + help This module implements an Application Layer Gateway (ALG) for SNMP payloads. In conjunction with NAT, it allows a network @@ -146,7 +144,7 @@ config IP_NF_MATCH_ECN tristate '"ecn" match support' depends on NETFILTER_ADVANCED select NETFILTER_XT_MATCH_ECN - ---help--- + help This is a backwards-compat option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_MATCH_ECN. @@ -155,7 +153,7 @@ config IP_NF_MATCH_RPFILTER tristate '"rpfilter" reverse path filter match support' depends on NETFILTER_ADVANCED depends on IP_NF_MANGLE || IP_NF_RAW - ---help--- + help This option allows you to match packets whose replies would go out via the interface the packet came in. @@ -166,7 +164,7 @@ config IP_NF_MATCH_TTL tristate '"ttl" match support' depends on NETFILTER_ADVANCED select NETFILTER_XT_MATCH_HL - ---help--- + help This is a backwards-compat option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_MATCH_HL. @@ -234,7 +232,7 @@ config IP_NF_TARGET_NETMAP tristate "NETMAP target support" depends on NETFILTER_ADVANCED select NETFILTER_XT_TARGET_NETMAP - ---help--- + help This is a backwards-compat option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_TARGET_NETMAP. @@ -243,7 +241,7 @@ config IP_NF_TARGET_REDIRECT tristate "REDIRECT target support" depends on NETFILTER_ADVANCED select NETFILTER_XT_TARGET_REDIRECT - ---help--- + help This is a backwards-compat option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_TARGET_REDIRECT. @@ -279,7 +277,7 @@ config IP_NF_TARGET_ECN tristate "ECN target support" depends on IP_NF_MANGLE depends on NETFILTER_ADVANCED - ---help--- + help This option adds a `ECN' target, which can be used in the iptables mangle table. @@ -294,7 +292,7 @@ config IP_NF_TARGET_TTL tristate '"TTL" target support' depends on NETFILTER_ADVANCED && IP_NF_MANGLE select NETFILTER_XT_TARGET_HL - ---help--- + help This is a backwards-compatible option for the user's convenience (e.g. when running oldconfig). It selects CONFIG_NETFILTER_XT_TARGET_HL. diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 7c497c78105f..93bad1184251 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -9,10 +9,6 @@ obj-$(CONFIG_NF_DEFRAG_IPV4) += nf_defrag_ipv4.o obj-$(CONFIG_NF_SOCKET_IPV4) += nf_socket_ipv4.o obj-$(CONFIG_NF_TPROXY_IPV4) += nf_tproxy_ipv4.o -# logging -obj-$(CONFIG_NF_LOG_ARP) += nf_log_arp.o -obj-$(CONFIG_NF_LOG_IPV4) += nf_log_ipv4.o - # reject obj-$(CONFIG_NF_REJECT_IPV4) += nf_reject_ipv4.o @@ -28,9 +24,6 @@ obj-$(CONFIG_NFT_REJECT_IPV4) += nft_reject_ipv4.o obj-$(CONFIG_NFT_FIB_IPV4) += nft_fib_ipv4.o obj-$(CONFIG_NFT_DUP_IPV4) += nft_dup_ipv4.o -# flow table support -obj-$(CONFIG_NF_FLOW_TABLE_IPV4) += nf_flow_table_ipv4.o - # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index f1f78a742b36..ffc0cab7cf18 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -179,10 +179,11 @@ struct arpt_entry *arpt_next_entry(const struct arpt_entry *entry) return (void *)entry + entry->next_offset; } -unsigned int arpt_do_table(struct sk_buff *skb, - const struct nf_hook_state *state, - struct xt_table *table) +unsigned int arpt_do_table(void *priv, + struct sk_buff *skb, + const struct nf_hook_state *state) { + const struct xt_table *table = priv; unsigned int hook = state->hook; static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); unsigned int verdict = NF_DROP; @@ -713,7 +714,7 @@ static int copy_entries_to_user(unsigned int total_size, return ret; } -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT static void compat_standard_from_user(void *dst, const void *src) { int v = *(compat_int_t *)src; @@ -787,8 +788,7 @@ static int compat_table_info(const struct xt_table_info *info, } #endif -static int get_info(struct net *net, void __user *user, - const int *len, int compat) +static int get_info(struct net *net, void __user *user, const int *len) { char name[XT_TABLE_MAXNAMELEN]; struct xt_table *t; @@ -801,18 +801,18 @@ static int get_info(struct net *net, void __user *user, return -EFAULT; name[XT_TABLE_MAXNAMELEN-1] = '\0'; -#ifdef CONFIG_COMPAT - if (compat) +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) xt_compat_lock(NFPROTO_ARP); #endif t = xt_request_find_table_lock(net, NFPROTO_ARP, name); if (!IS_ERR(t)) { struct arpt_getinfo info; const struct xt_table_info *private = t->private; -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT struct xt_table_info tmp; - if (compat) { + if (in_compat_syscall()) { ret = compat_table_info(private, &tmp); xt_compat_flush_offsets(NFPROTO_ARP); private = &tmp; @@ -836,8 +836,8 @@ static int get_info(struct net *net, void __user *user, module_put(t->me); } else ret = PTR_ERR(t); -#ifdef CONFIG_COMPAT - if (compat) +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) xt_compat_unlock(NFPROTO_ARP); #endif return ret; @@ -948,8 +948,7 @@ static int __do_replace(struct net *net, const char *name, return ret; } -static int do_replace(struct net *net, const void __user *user, - unsigned int len) +static int do_replace(struct net *net, sockptr_t arg, unsigned int len) { int ret; struct arpt_replace tmp; @@ -957,7 +956,7 @@ static int do_replace(struct net *net, const void __user *user, void *loc_cpu_entry; struct arpt_entry *iter; - if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) + if (copy_from_sockptr(&tmp, arg, sizeof(tmp)) != 0) return -EFAULT; /* overflow check */ @@ -973,8 +972,8 @@ static int do_replace(struct net *net, const void __user *user, return -ENOMEM; loc_cpu_entry = newinfo->entries; - if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), - tmp.size) != 0) { + if (copy_from_sockptr_offset(loc_cpu_entry, arg, sizeof(tmp), + tmp.size) != 0) { ret = -EFAULT; goto free_newinfo; } @@ -997,8 +996,7 @@ static int do_replace(struct net *net, const void __user *user, return ret; } -static int do_add_counters(struct net *net, const void __user *user, - unsigned int len, int compat) +static int do_add_counters(struct net *net, sockptr_t arg, unsigned int len) { unsigned int i; struct xt_counters_info tmp; @@ -1009,7 +1007,7 @@ static int do_add_counters(struct net *net, const void __user *user, struct arpt_entry *iter; unsigned int addend; - paddc = xt_copy_counters_from_user(user, len, &tmp, compat); + paddc = xt_copy_counters(arg, len, &tmp); if (IS_ERR(paddc)) return PTR_ERR(paddc); @@ -1047,7 +1045,7 @@ static int do_add_counters(struct net *net, const void __user *user, return ret; } -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT struct compat_arpt_replace { char name[XT_TABLE_MAXNAMELEN]; u32 valid_hooks; @@ -1057,7 +1055,7 @@ struct compat_arpt_replace { u32 underflow[NF_ARP_NUMHOOKS]; u32 num_counters; compat_uptr_t counters; - struct compat_arpt_entry entries[0]; + struct compat_arpt_entry entries[]; }; static inline void compat_release_entry(struct compat_arpt_entry *e) @@ -1196,6 +1194,8 @@ static int translate_compat_table(struct net *net, if (!newinfo) goto out_unlock; + memset(newinfo->entries, 0, size); + newinfo->number = compatr->num_entries; for (i = 0; i < NF_ARP_NUMHOOKS; i++) { newinfo->hook_entry[i] = compatr->hook_entry[i]; @@ -1246,8 +1246,7 @@ out_unlock: return ret; } -static int compat_do_replace(struct net *net, void __user *user, - unsigned int len) +static int compat_do_replace(struct net *net, sockptr_t arg, unsigned int len) { int ret; struct compat_arpt_replace tmp; @@ -1255,7 +1254,7 @@ static int compat_do_replace(struct net *net, void __user *user, void *loc_cpu_entry; struct arpt_entry *iter; - if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) + if (copy_from_sockptr(&tmp, arg, sizeof(tmp)) != 0) return -EFAULT; /* overflow check */ @@ -1271,7 +1270,8 @@ static int compat_do_replace(struct net *net, void __user *user, return -ENOMEM; loc_cpu_entry = newinfo->entries; - if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { + if (copy_from_sockptr_offset(loc_cpu_entry, arg, sizeof(tmp), + tmp.size) != 0) { ret = -EFAULT; goto free_newinfo; } @@ -1294,30 +1294,6 @@ static int compat_do_replace(struct net *net, void __user *user, return ret; } -static int compat_do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, - unsigned int len) -{ - int ret; - - if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) - return -EPERM; - - switch (cmd) { - case ARPT_SO_SET_REPLACE: - ret = compat_do_replace(sock_net(sk), user, len); - break; - - case ARPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sock_net(sk), user, len, 1); - break; - - default: - ret = -EINVAL; - } - - return ret; -} - static int compat_copy_entry_to_user(struct arpt_entry *e, void __user **dstptr, compat_uint_t *size, struct xt_counters *counters, @@ -1383,7 +1359,7 @@ static int compat_copy_entries_to_user(unsigned int total_size, struct compat_arpt_get_entries { char name[XT_TABLE_MAXNAMELEN]; compat_uint_t size; - struct compat_arpt_entry entrytable[0]; + struct compat_arpt_entry entrytable[]; }; static int compat_get_entries(struct net *net, @@ -1425,32 +1401,10 @@ static int compat_get_entries(struct net *net, xt_compat_unlock(NFPROTO_ARP); return ret; } - -static int do_arpt_get_ctl(struct sock *, int, void __user *, int *); - -static int compat_do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, - int *len) -{ - int ret; - - if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) - return -EPERM; - - switch (cmd) { - case ARPT_SO_GET_INFO: - ret = get_info(sock_net(sk), user, len, 1); - break; - case ARPT_SO_GET_ENTRIES: - ret = compat_get_entries(sock_net(sk), user, len); - break; - default: - ret = do_arpt_get_ctl(sk, cmd, user, len); - } - return ret; -} #endif -static int do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) +static int do_arpt_set_ctl(struct sock *sk, int cmd, sockptr_t arg, + unsigned int len) { int ret; @@ -1459,11 +1413,16 @@ static int do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned switch (cmd) { case ARPT_SO_SET_REPLACE: - ret = do_replace(sock_net(sk), user, len); +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) + ret = compat_do_replace(sock_net(sk), arg, len); + else +#endif + ret = do_replace(sock_net(sk), arg, len); break; case ARPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sock_net(sk), user, len, 0); + ret = do_add_counters(sock_net(sk), arg, len); break; default: @@ -1482,11 +1441,16 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len switch (cmd) { case ARPT_SO_GET_INFO: - ret = get_info(sock_net(sk), user, len, 0); + ret = get_info(sock_net(sk), user, len); break; case ARPT_SO_GET_ENTRIES: - ret = get_entries(sock_net(sk), user, len); +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) + ret = compat_get_entries(sock_net(sk), user, len); + else +#endif + ret = get_entries(sock_net(sk), user, len); break; case ARPT_SO_GET_REVISION_TARGET: { @@ -1536,10 +1500,11 @@ static void __arpt_unregister_table(struct net *net, struct xt_table *table) int arpt_register_table(struct net *net, const struct xt_table *table, const struct arpt_replace *repl, - const struct nf_hook_ops *ops, - struct xt_table **res) + const struct nf_hook_ops *template_ops) { - int ret; + struct nf_hook_ops *ops; + unsigned int num_ops; + int ret, i; struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; void *loc_cpu_entry; @@ -1553,36 +1518,60 @@ int arpt_register_table(struct net *net, memcpy(loc_cpu_entry, repl->entries, repl->size); ret = translate_table(net, newinfo, loc_cpu_entry, repl); - if (ret != 0) - goto out_free; + if (ret != 0) { + xt_free_table_info(newinfo); + return ret; + } new_table = xt_register_table(net, table, &bootstrap, newinfo); if (IS_ERR(new_table)) { - ret = PTR_ERR(new_table); - goto out_free; + xt_free_table_info(newinfo); + return PTR_ERR(new_table); } - /* set res now, will see skbs right after nf_register_net_hooks */ - WRITE_ONCE(*res, new_table); + num_ops = hweight32(table->valid_hooks); + if (num_ops == 0) { + ret = -EINVAL; + goto out_free; + } - ret = nf_register_net_hooks(net, ops, hweight32(table->valid_hooks)); - if (ret != 0) { - __arpt_unregister_table(net, new_table); - *res = NULL; + ops = kmemdup(template_ops, sizeof(*ops) * num_ops, GFP_KERNEL); + if (!ops) { + ret = -ENOMEM; + goto out_free; } + for (i = 0; i < num_ops; i++) + ops[i].priv = new_table; + + new_table->ops = ops; + + ret = nf_register_net_hooks(net, ops, num_ops); + if (ret != 0) + goto out_free; + return ret; out_free: - xt_free_table_info(newinfo); + __arpt_unregister_table(net, new_table); return ret; } -void arpt_unregister_table(struct net *net, struct xt_table *table, - const struct nf_hook_ops *ops) +void arpt_unregister_table_pre_exit(struct net *net, const char *name) +{ + struct xt_table *table = xt_find_table(net, NFPROTO_ARP, name); + + if (table) + nf_unregister_net_hooks(net, table->ops, hweight32(table->valid_hooks)); +} +EXPORT_SYMBOL(arpt_unregister_table_pre_exit); + +void arpt_unregister_table(struct net *net, const char *name) { - nf_unregister_net_hooks(net, ops, hweight32(table->valid_hooks)); - __arpt_unregister_table(net, table); + struct xt_table *table = xt_find_table(net, NFPROTO_ARP, name); + + if (table) + __arpt_unregister_table(net, table); } /* The built-in targets: standard (NULL) and error. */ @@ -1591,7 +1580,7 @@ static struct xt_target arpt_builtin_tg[] __read_mostly = { .name = XT_STANDARD_TARGET, .targetsize = sizeof(int), .family = NFPROTO_ARP, -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT .compatsize = sizeof(compat_int_t), .compat_from_user = compat_standard_from_user, .compat_to_user = compat_standard_to_user, @@ -1610,15 +1599,9 @@ static struct nf_sockopt_ops arpt_sockopts = { .set_optmin = ARPT_BASE_CTL, .set_optmax = ARPT_SO_SET_MAX+1, .set = do_arpt_set_ctl, -#ifdef CONFIG_COMPAT - .compat_set = compat_do_arpt_set_ctl, -#endif .get_optmin = ARPT_BASE_CTL, .get_optmax = ARPT_SO_GET_MAX+1, .get = do_arpt_get_ctl, -#ifdef CONFIG_COMPAT - .compat_get = compat_do_arpt_get_ctl, -#endif .owner = THIS_MODULE, }; diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index c216b9ad3bb2..78cd5ee24448 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -18,82 +18,72 @@ MODULE_DESCRIPTION("arptables filter table"); #define FILTER_VALID_HOOKS ((1 << NF_ARP_IN) | (1 << NF_ARP_OUT) | \ (1 << NF_ARP_FORWARD)) -static int __net_init arptable_filter_table_init(struct net *net); - static const struct xt_table packet_filter = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_ARP, .priority = NF_IP_PRI_FILTER, - .table_init = arptable_filter_table_init, }; -/* The work comes in here from netfilter.c */ -static unsigned int -arptable_filter_hook(void *priv, struct sk_buff *skb, - const struct nf_hook_state *state) -{ - return arpt_do_table(skb, state, state->net->ipv4.arptable_filter); -} - static struct nf_hook_ops *arpfilter_ops __read_mostly; -static int __net_init arptable_filter_table_init(struct net *net) +static int arptable_filter_table_init(struct net *net) { struct arpt_replace *repl; int err; - if (net->ipv4.arptable_filter) - return 0; - repl = arpt_alloc_initial_table(&packet_filter); if (repl == NULL) return -ENOMEM; - err = arpt_register_table(net, &packet_filter, repl, arpfilter_ops, - &net->ipv4.arptable_filter); + err = arpt_register_table(net, &packet_filter, repl, arpfilter_ops); kfree(repl); return err; } +static void __net_exit arptable_filter_net_pre_exit(struct net *net) +{ + arpt_unregister_table_pre_exit(net, "filter"); +} + static void __net_exit arptable_filter_net_exit(struct net *net) { - if (!net->ipv4.arptable_filter) - return; - arpt_unregister_table(net, net->ipv4.arptable_filter, arpfilter_ops); - net->ipv4.arptable_filter = NULL; + arpt_unregister_table(net, "filter"); } static struct pernet_operations arptable_filter_net_ops = { .exit = arptable_filter_net_exit, + .pre_exit = arptable_filter_net_pre_exit, }; static int __init arptable_filter_init(void) { - int ret; + int ret = xt_register_template(&packet_filter, + arptable_filter_table_init); + + if (ret < 0) + return ret; - arpfilter_ops = xt_hook_ops_alloc(&packet_filter, arptable_filter_hook); - if (IS_ERR(arpfilter_ops)) + arpfilter_ops = xt_hook_ops_alloc(&packet_filter, arpt_do_table); + if (IS_ERR(arpfilter_ops)) { + xt_unregister_template(&packet_filter); return PTR_ERR(arpfilter_ops); + } ret = register_pernet_subsys(&arptable_filter_net_ops); if (ret < 0) { + xt_unregister_template(&packet_filter); kfree(arpfilter_ops); return ret; } - ret = arptable_filter_table_init(&init_net); - if (ret) { - unregister_pernet_subsys(&arptable_filter_net_ops); - kfree(arpfilter_ops); - } - return ret; } static void __exit arptable_filter_fini(void) { unregister_pernet_subsys(&arptable_filter_net_ops); + xt_unregister_template(&packet_filter); kfree(arpfilter_ops); } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 10b91ebdf213..2ed7c58b471a 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -222,10 +222,11 @@ struct ipt_entry *ipt_next_entry(const struct ipt_entry *entry) /* Returns one of the generic firewall policies, like NF_ACCEPT. */ unsigned int -ipt_do_table(struct sk_buff *skb, - const struct nf_hook_state *state, - struct xt_table *table) +ipt_do_table(void *priv, + struct sk_buff *skb, + const struct nf_hook_state *state) { + const struct xt_table *table = priv; unsigned int hook = state->hook; static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); const struct iphdr *ip; @@ -868,7 +869,7 @@ copy_entries_to_user(unsigned int total_size, return ret; } -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT static void compat_standard_from_user(void *dst, const void *src) { int v = *(compat_int_t *)src; @@ -944,8 +945,7 @@ static int compat_table_info(const struct xt_table_info *info, } #endif -static int get_info(struct net *net, void __user *user, - const int *len, int compat) +static int get_info(struct net *net, void __user *user, const int *len) { char name[XT_TABLE_MAXNAMELEN]; struct xt_table *t; @@ -958,18 +958,18 @@ static int get_info(struct net *net, void __user *user, return -EFAULT; name[XT_TABLE_MAXNAMELEN-1] = '\0'; -#ifdef CONFIG_COMPAT - if (compat) +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) xt_compat_lock(AF_INET); #endif t = xt_request_find_table_lock(net, AF_INET, name); if (!IS_ERR(t)) { struct ipt_getinfo info; const struct xt_table_info *private = t->private; -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT struct xt_table_info tmp; - if (compat) { + if (in_compat_syscall()) { ret = compat_table_info(private, &tmp); xt_compat_flush_offsets(AF_INET); private = &tmp; @@ -994,8 +994,8 @@ static int get_info(struct net *net, void __user *user, module_put(t->me); } else ret = PTR_ERR(t); -#ifdef CONFIG_COMPAT - if (compat) +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) xt_compat_unlock(AF_INET); #endif return ret; @@ -1103,7 +1103,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, } static int -do_replace(struct net *net, const void __user *user, unsigned int len) +do_replace(struct net *net, sockptr_t arg, unsigned int len) { int ret; struct ipt_replace tmp; @@ -1111,7 +1111,7 @@ do_replace(struct net *net, const void __user *user, unsigned int len) void *loc_cpu_entry; struct ipt_entry *iter; - if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) + if (copy_from_sockptr(&tmp, arg, sizeof(tmp)) != 0) return -EFAULT; /* overflow check */ @@ -1127,8 +1127,8 @@ do_replace(struct net *net, const void __user *user, unsigned int len) return -ENOMEM; loc_cpu_entry = newinfo->entries; - if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), - tmp.size) != 0) { + if (copy_from_sockptr_offset(loc_cpu_entry, arg, sizeof(tmp), + tmp.size) != 0) { ret = -EFAULT; goto free_newinfo; } @@ -1152,8 +1152,7 @@ do_replace(struct net *net, const void __user *user, unsigned int len) } static int -do_add_counters(struct net *net, const void __user *user, - unsigned int len, int compat) +do_add_counters(struct net *net, sockptr_t arg, unsigned int len) { unsigned int i; struct xt_counters_info tmp; @@ -1164,7 +1163,7 @@ do_add_counters(struct net *net, const void __user *user, struct ipt_entry *iter; unsigned int addend; - paddc = xt_copy_counters_from_user(user, len, &tmp, compat); + paddc = xt_copy_counters(arg, len, &tmp); if (IS_ERR(paddc)) return PTR_ERR(paddc); @@ -1201,7 +1200,7 @@ do_add_counters(struct net *net, const void __user *user, return ret; } -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT struct compat_ipt_replace { char name[XT_TABLE_MAXNAMELEN]; u32 valid_hooks; @@ -1211,7 +1210,7 @@ struct compat_ipt_replace { u32 underflow[NF_INET_NUMHOOKS]; u32 num_counters; compat_uptr_t counters; /* struct xt_counters * */ - struct compat_ipt_entry entries[0]; + struct compat_ipt_entry entries[]; }; static int @@ -1430,6 +1429,8 @@ translate_compat_table(struct net *net, if (!newinfo) goto out_unlock; + memset(newinfo->entries, 0, size); + newinfo->number = compatr->num_entries; for (i = 0; i < NF_INET_NUMHOOKS; i++) { newinfo->hook_entry[i] = compatr->hook_entry[i]; @@ -1486,7 +1487,7 @@ out_unlock: } static int -compat_do_replace(struct net *net, void __user *user, unsigned int len) +compat_do_replace(struct net *net, sockptr_t arg, unsigned int len) { int ret; struct compat_ipt_replace tmp; @@ -1494,7 +1495,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) void *loc_cpu_entry; struct ipt_entry *iter; - if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) + if (copy_from_sockptr(&tmp, arg, sizeof(tmp)) != 0) return -EFAULT; /* overflow check */ @@ -1510,8 +1511,8 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return -ENOMEM; loc_cpu_entry = newinfo->entries; - if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), - tmp.size) != 0) { + if (copy_from_sockptr_offset(loc_cpu_entry, arg, sizeof(tmp), + tmp.size) != 0) { ret = -EFAULT; goto free_newinfo; } @@ -1534,35 +1535,10 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return ret; } -static int -compat_do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, - unsigned int len) -{ - int ret; - - if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) - return -EPERM; - - switch (cmd) { - case IPT_SO_SET_REPLACE: - ret = compat_do_replace(sock_net(sk), user, len); - break; - - case IPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sock_net(sk), user, len, 1); - break; - - default: - ret = -EINVAL; - } - - return ret; -} - struct compat_ipt_get_entries { char name[XT_TABLE_MAXNAMELEN]; compat_uint_t size; - struct compat_ipt_entry entrytable[0]; + struct compat_ipt_entry entrytable[]; }; static int @@ -1634,33 +1610,10 @@ compat_get_entries(struct net *net, struct compat_ipt_get_entries __user *uptr, xt_compat_unlock(AF_INET); return ret; } - -static int do_ipt_get_ctl(struct sock *, int, void __user *, int *); - -static int -compat_do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) -{ - int ret; - - if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) - return -EPERM; - - switch (cmd) { - case IPT_SO_GET_INFO: - ret = get_info(sock_net(sk), user, len, 1); - break; - case IPT_SO_GET_ENTRIES: - ret = compat_get_entries(sock_net(sk), user, len); - break; - default: - ret = do_ipt_get_ctl(sk, cmd, user, len); - } - return ret; -} #endif static int -do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) +do_ipt_set_ctl(struct sock *sk, int cmd, sockptr_t arg, unsigned int len) { int ret; @@ -1669,11 +1622,16 @@ do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) switch (cmd) { case IPT_SO_SET_REPLACE: - ret = do_replace(sock_net(sk), user, len); +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) + ret = compat_do_replace(sock_net(sk), arg, len); + else +#endif + ret = do_replace(sock_net(sk), arg, len); break; case IPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sock_net(sk), user, len, 0); + ret = do_add_counters(sock_net(sk), arg, len); break; default: @@ -1693,11 +1651,16 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) switch (cmd) { case IPT_SO_GET_INFO: - ret = get_info(sock_net(sk), user, len, 0); + ret = get_info(sock_net(sk), user, len); break; case IPT_SO_GET_ENTRIES: - ret = get_entries(sock_net(sk), user, len); +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT + if (in_compat_syscall()) + ret = compat_get_entries(sock_net(sk), user, len); + else +#endif + ret = get_entries(sock_net(sk), user, len); break; case IPT_SO_GET_REVISION_MATCH: @@ -1754,9 +1717,11 @@ static void __ipt_unregister_table(struct net *net, struct xt_table *table) int ipt_register_table(struct net *net, const struct xt_table *table, const struct ipt_replace *repl, - const struct nf_hook_ops *ops, struct xt_table **res) + const struct nf_hook_ops *template_ops) { - int ret; + struct nf_hook_ops *ops; + unsigned int num_ops; + int ret, i; struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; void *loc_cpu_entry; @@ -1770,39 +1735,65 @@ int ipt_register_table(struct net *net, const struct xt_table *table, memcpy(loc_cpu_entry, repl->entries, repl->size); ret = translate_table(net, newinfo, loc_cpu_entry, repl); - if (ret != 0) - goto out_free; + if (ret != 0) { + xt_free_table_info(newinfo); + return ret; + } new_table = xt_register_table(net, table, &bootstrap, newinfo); if (IS_ERR(new_table)) { - ret = PTR_ERR(new_table); - goto out_free; + xt_free_table_info(newinfo); + return PTR_ERR(new_table); } - /* set res now, will see skbs right after nf_register_net_hooks */ - WRITE_ONCE(*res, new_table); - if (!ops) + /* No template? No need to do anything. This is used by 'nat' table, it registers + * with the nat core instead of the netfilter core. + */ + if (!template_ops) return 0; - ret = nf_register_net_hooks(net, ops, hweight32(table->valid_hooks)); - if (ret != 0) { - __ipt_unregister_table(net, new_table); - *res = NULL; + num_ops = hweight32(table->valid_hooks); + if (num_ops == 0) { + ret = -EINVAL; + goto out_free; + } + + ops = kmemdup(template_ops, sizeof(*ops) * num_ops, GFP_KERNEL); + if (!ops) { + ret = -ENOMEM; + goto out_free; } + for (i = 0; i < num_ops; i++) + ops[i].priv = new_table; + + new_table->ops = ops; + + ret = nf_register_net_hooks(net, ops, num_ops); + if (ret != 0) + goto out_free; + return ret; out_free: - xt_free_table_info(newinfo); + __ipt_unregister_table(net, new_table); return ret; } -void ipt_unregister_table(struct net *net, struct xt_table *table, - const struct nf_hook_ops *ops) +void ipt_unregister_table_pre_exit(struct net *net, const char *name) { - if (ops) - nf_unregister_net_hooks(net, ops, hweight32(table->valid_hooks)); - __ipt_unregister_table(net, table); + struct xt_table *table = xt_find_table(net, NFPROTO_IPV4, name); + + if (table) + nf_unregister_net_hooks(net, table->ops, hweight32(table->valid_hooks)); +} + +void ipt_unregister_table_exit(struct net *net, const char *name) +{ + struct xt_table *table = xt_find_table(net, NFPROTO_IPV4, name); + + if (table) + __ipt_unregister_table(net, table); } /* Returns 1 if the type and code is matched by the range, 0 otherwise */ @@ -1856,7 +1847,7 @@ static struct xt_target ipt_builtin_tg[] __read_mostly = { .name = XT_STANDARD_TARGET, .targetsize = sizeof(int), .family = NFPROTO_IPV4, -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT .compatsize = sizeof(compat_int_t), .compat_from_user = compat_standard_from_user, .compat_to_user = compat_standard_to_user, @@ -1875,15 +1866,9 @@ static struct nf_sockopt_ops ipt_sockopts = { .set_optmin = IPT_BASE_CTL, .set_optmax = IPT_SO_SET_MAX+1, .set = do_ipt_set_ctl, -#ifdef CONFIG_COMPAT - .compat_set = compat_do_ipt_set_ctl, -#endif .get_optmin = IPT_BASE_CTL, .get_optmax = IPT_SO_GET_MAX+1, .get = do_ipt_get_ctl, -#ifdef CONFIG_COMPAT - .compat_get = compat_do_ipt_get_ctl, -#endif .owner = THIS_MODULE, }; @@ -1957,7 +1942,8 @@ static void __exit ip_tables_fini(void) } EXPORT_SYMBOL(ipt_register_table); -EXPORT_SYMBOL(ipt_unregister_table); +EXPORT_SYMBOL(ipt_unregister_table_pre_exit); +EXPORT_SYMBOL(ipt_unregister_table_exit); EXPORT_SYMBOL(ipt_do_table); module_init(ip_tables_init); module_exit(ip_tables_fini); diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index f8755a4ae9d4..f8e176c77d1c 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -3,7 +3,7 @@ * (C) 2003-2004 by Harald Welte <laforge@netfilter.org> * based on ideas of Fabio Olive Leite <olive@unixforge.org> * - * Development of this code funded by SuSE Linux AG, http://www.suse.com/ + * Development of this code funded by SuSE Linux AG, https://www.suse.com/ */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/module.h> @@ -66,11 +66,22 @@ struct clusterip_net { /* lock protects the configs list */ spinlock_t lock; + bool clusterip_deprecated_warning; #ifdef CONFIG_PROC_FS struct proc_dir_entry *procdir; /* mutex protects the config->pde*/ struct mutex mutex; #endif + unsigned int hook_users; +}; + +static unsigned int clusterip_arp_mangle(void *priv, struct sk_buff *skb, const struct nf_hook_state *state); + +static const struct nf_hook_ops cip_arp_ops = { + .hook = clusterip_arp_mangle, + .pf = NFPROTO_ARP, + .hooknum = NF_ARP_OUT, + .priority = -1 }; static unsigned int clusterip_net_id __read_mostly; @@ -458,6 +469,7 @@ clusterip_tg(struct sk_buff *skb, const struct xt_action_param *par) static int clusterip_tg_check(const struct xt_tgchk_param *par) { struct ipt_clusterip_tgt_info *cipinfo = par->targinfo; + struct clusterip_net *cn = clusterip_pernet(par->net); const struct ipt_entry *e = par->entryinfo; struct clusterip_config *config; int ret, i; @@ -467,6 +479,9 @@ static int clusterip_tg_check(const struct xt_tgchk_param *par) return -EOPNOTSUPP; } + if (cn->hook_users == UINT_MAX) + return -EOVERFLOW; + if (cipinfo->hash_mode != CLUSTERIP_HASHMODE_SIP && cipinfo->hash_mode != CLUSTERIP_HASHMODE_SIP_SPT && cipinfo->hash_mode != CLUSTERIP_HASHMODE_SIP_SPT_DPT) { @@ -505,8 +520,11 @@ static int clusterip_tg_check(const struct xt_tgchk_param *par) if (IS_ERR(config)) return PTR_ERR(config); } - } else if (memcmp(&config->clustermac, &cipinfo->clustermac, ETH_ALEN)) + } else if (memcmp(&config->clustermac, &cipinfo->clustermac, ETH_ALEN)) { + clusterip_config_entry_put(config); + clusterip_config_put(config); return -EINVAL; + } ret = nf_ct_netns_get(par->net, par->family); if (ret < 0) { @@ -517,10 +535,23 @@ static int clusterip_tg_check(const struct xt_tgchk_param *par) return ret; } - if (!par->net->xt.clusterip_deprecated_warning) { + if (cn->hook_users == 0) { + ret = nf_register_net_hook(par->net, &cip_arp_ops); + + if (ret < 0) { + clusterip_config_entry_put(config); + clusterip_config_put(config); + nf_ct_netns_put(par->net, par->family); + return ret; + } + } + + cn->hook_users++; + + if (!cn->clusterip_deprecated_warning) { pr_info("ipt_CLUSTERIP is deprecated and it will removed soon, " "use xt_cluster instead\n"); - par->net->xt.clusterip_deprecated_warning = true; + cn->clusterip_deprecated_warning = true; } cipinfo->config = config; @@ -531,6 +562,7 @@ static int clusterip_tg_check(const struct xt_tgchk_param *par) static void clusterip_tg_destroy(const struct xt_tgdtor_param *par) { const struct ipt_clusterip_tgt_info *cipinfo = par->targinfo; + struct clusterip_net *cn = clusterip_pernet(par->net); /* if no more entries are referencing the config, remove it * from the list and destroy the proc entry */ @@ -539,9 +571,13 @@ static void clusterip_tg_destroy(const struct xt_tgdtor_param *par) clusterip_config_put(cipinfo->config); nf_ct_netns_put(par->net, par->family); + cn->hook_users--; + + if (cn->hook_users == 0) + nf_unregister_net_hook(par->net, &cip_arp_ops); } -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT struct compat_ipt_clusterip_tgt_info { u_int32_t flags; @@ -553,7 +589,7 @@ struct compat_ipt_clusterip_tgt_info u_int32_t hash_initval; compat_uptr_t config; }; -#endif /* CONFIG_COMPAT */ +#endif /* CONFIG_NETFILTER_XTABLES_COMPAT */ static struct xt_target clusterip_tg_reg __read_mostly = { .name = "CLUSTERIP", @@ -563,9 +599,9 @@ static struct xt_target clusterip_tg_reg __read_mostly = { .destroy = clusterip_tg_destroy, .targetsize = sizeof(struct ipt_clusterip_tgt_info), .usersize = offsetof(struct ipt_clusterip_tgt_info, config), -#ifdef CONFIG_COMPAT +#ifdef CONFIG_NETFILTER_XTABLES_COMPAT .compatsize = sizeof(struct compat_ipt_clusterip_tgt_info), -#endif /* CONFIG_COMPAT */ +#endif /* CONFIG_NETFILTER_XTABLES_COMPAT */ .me = THIS_MODULE }; @@ -602,9 +638,8 @@ static void arp_print(struct arp_payload *payload) #endif static unsigned int -arp_mangle(void *priv, - struct sk_buff *skb, - const struct nf_hook_state *state) +clusterip_arp_mangle(void *priv, struct sk_buff *skb, + const struct nf_hook_state *state) { struct arphdr *arp = arp_hdr(skb); struct arp_payload *payload; @@ -654,13 +689,6 @@ arp_mangle(void *priv, return NF_ACCEPT; } -static const struct nf_hook_ops cip_arp_ops = { - .hook = arp_mangle, - .pf = NFPROTO_ARP, - .hooknum = NF_ARP_OUT, - .priority = -1 -}; - /*********************************************************************** * PROC DIR HANDLING ***********************************************************************/ @@ -748,7 +776,7 @@ static int clusterip_proc_open(struct inode *inode, struct file *file) if (!ret) { struct seq_file *sf = file->private_data; - struct clusterip_config *c = PDE_DATA(inode); + struct clusterip_config *c = pde_data(inode); sf->private = c; @@ -760,7 +788,7 @@ static int clusterip_proc_open(struct inode *inode, struct file *file) static int clusterip_proc_release(struct inode *inode, struct file *file) { - struct clusterip_config *c = PDE_DATA(inode); + struct clusterip_config *c = pde_data(inode); int ret; ret = seq_release(inode, file); @@ -774,7 +802,7 @@ static int clusterip_proc_release(struct inode *inode, struct file *file) static ssize_t clusterip_proc_write(struct file *file, const char __user *input, size_t size, loff_t *ofs) { - struct clusterip_config *c = PDE_DATA(file_inode(file)); + struct clusterip_config *c = pde_data(file_inode(file)); #define PROC_WRITELEN 10 char buffer[PROC_WRITELEN+1]; unsigned long nodenum; @@ -817,20 +845,14 @@ static const struct proc_ops clusterip_proc_ops = { static int clusterip_net_init(struct net *net) { struct clusterip_net *cn = clusterip_pernet(net); - int ret; INIT_LIST_HEAD(&cn->configs); spin_lock_init(&cn->lock); - ret = nf_register_net_hook(net, &cip_arp_ops); - if (ret < 0) - return ret; - #ifdef CONFIG_PROC_FS cn->procdir = proc_mkdir("ipt_CLUSTERIP", net->proc_net); if (!cn->procdir) { - nf_unregister_net_hook(net, &cip_arp_ops); pr_err("Unable to proc dir entry\n"); return -ENOMEM; } @@ -850,7 +872,6 @@ static void clusterip_net_exit(struct net *net) cn->procdir = NULL; mutex_unlock(&cn->mutex); #endif - nf_unregister_net_hook(net, &cip_arp_ops); } static struct pernet_operations clusterip_net_ops = { diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index e16b98ee6266..4b8840734762 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -56,7 +56,8 @@ reject_tg(struct sk_buff *skb, const struct xt_action_param *par) nf_send_unreach(skb, ICMP_PKT_FILTERED, hook); break; case IPT_TCP_RESET: - nf_send_reset(xt_net(par), skb, hook); + nf_send_reset(xt_net(par), par->state->sk, skb, hook); + break; case IPT_ICMP_ECHOREPLY: /* Doesn't happen. */ break; diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c index 748dc3ce58d3..f2984c7eef40 100644 --- a/net/ipv4/netfilter/ipt_SYNPROXY.c +++ b/net/ipv4/netfilter/ipt_SYNPROXY.c @@ -118,3 +118,4 @@ module_exit(synproxy_tg4_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); +MODULE_DESCRIPTION("Intercept TCP connections and establish them using syncookies"); diff --git a/net/ipv4/netfilter/ipt_rpfilter.c b/net/ipv4/netfilter/ipt_rpfilter.c index cc23f1ce239c..ded5bef02f77 100644 --- a/net/ipv4/netfilter/ipt_rpfilter.c +++ b/net/ipv4/netfilter/ipt_rpfilter.c @@ -33,7 +33,6 @@ static bool rpfilter_lookup_reverse(struct net *net, struct flowi4 *fl4, const struct net_device *dev, u8 flags) { struct fib_result res; - int ret __maybe_unused; if (fib_lookup(net, fl4, &res, FIB_LOOKUP_IGNORE_LINKSTATE)) return false; @@ -76,9 +75,10 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) flow.daddr = iph->saddr; flow.saddr = rpfilter_get_saddr(iph->daddr); flow.flowi4_mark = info->flags & XT_RPFILTER_VALID_MARK ? skb->mark : 0; - flow.flowi4_tos = RT_TOS(iph->tos); + flow.flowi4_tos = iph->tos & IPTOS_RT_MASK; flow.flowi4_scope = RT_SCOPE_UNIVERSE; - flow.flowi4_oif = l3mdev_master_ifindex_rcu(xt_in(par)); + flow.flowi4_l3mdev = l3mdev_master_ifindex_rcu(xt_in(par)); + flow.flowi4_uid = sock_net_uid(xt_net(par), NULL); return rpfilter_lookup_reverse(xt_net(par), &flow, xt_in(par), info->flags) ^ invert; } diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 9d54b4017e50..b9062f4552ac 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -19,7 +19,6 @@ MODULE_DESCRIPTION("iptables filter table"); #define FILTER_VALID_HOOKS ((1 << NF_INET_LOCAL_IN) | \ (1 << NF_INET_FORWARD) | \ (1 << NF_INET_LOCAL_OUT)) -static int __net_init iptable_filter_table_init(struct net *net); static const struct xt_table packet_filter = { .name = "filter", @@ -27,30 +26,19 @@ static const struct xt_table packet_filter = { .me = THIS_MODULE, .af = NFPROTO_IPV4, .priority = NF_IP_PRI_FILTER, - .table_init = iptable_filter_table_init, }; -static unsigned int -iptable_filter_hook(void *priv, struct sk_buff *skb, - const struct nf_hook_state *state) -{ - return ipt_do_table(skb, state, state->net->ipv4.iptable_filter); -} - static struct nf_hook_ops *filter_ops __read_mostly; /* Default to forward because I got too much mail already. */ static bool forward __read_mostly = true; module_param(forward, bool, 0000); -static int __net_init iptable_filter_table_init(struct net *net) +static int iptable_filter_table_init(struct net *net) { struct ipt_replace *repl; int err; - if (net->ipv4.iptable_filter) - return 0; - repl = ipt_alloc_initial_table(&packet_filter); if (repl == NULL) return -ENOMEM; @@ -58,51 +46,63 @@ static int __net_init iptable_filter_table_init(struct net *net) ((struct ipt_standard *)repl->entries)[1].target.verdict = forward ? -NF_ACCEPT - 1 : -NF_DROP - 1; - err = ipt_register_table(net, &packet_filter, repl, filter_ops, - &net->ipv4.iptable_filter); + err = ipt_register_table(net, &packet_filter, repl, filter_ops); kfree(repl); return err; } static int __net_init iptable_filter_net_init(struct net *net) { - if (net == &init_net || !forward) + if (!forward) return iptable_filter_table_init(net); return 0; } +static void __net_exit iptable_filter_net_pre_exit(struct net *net) +{ + ipt_unregister_table_pre_exit(net, "filter"); +} + static void __net_exit iptable_filter_net_exit(struct net *net) { - if (!net->ipv4.iptable_filter) - return; - ipt_unregister_table(net, net->ipv4.iptable_filter, filter_ops); - net->ipv4.iptable_filter = NULL; + ipt_unregister_table_exit(net, "filter"); } static struct pernet_operations iptable_filter_net_ops = { .init = iptable_filter_net_init, + .pre_exit = iptable_filter_net_pre_exit, .exit = iptable_filter_net_exit, }; static int __init iptable_filter_init(void) { - int ret; + int ret = xt_register_template(&packet_filter, + iptable_filter_table_init); + + if (ret < 0) + return ret; - filter_ops = xt_hook_ops_alloc(&packet_filter, iptable_filter_hook); - if (IS_ERR(filter_ops)) + filter_ops = xt_hook_ops_alloc(&packet_filter, ipt_do_table); + if (IS_ERR(filter_ops)) { + xt_unregister_template(&packet_filter); return PTR_ERR(filter_ops); + } ret = register_pernet_subsys(&iptable_filter_net_ops); - if (ret < 0) + if (ret < 0) { + xt_unregister_template(&packet_filter); kfree(filter_ops); + return ret; + } - return ret; + return 0; } static void __exit iptable_filter_fini(void) { unregister_pernet_subsys(&iptable_filter_net_ops); + xt_unregister_template(&packet_filter); kfree(filter_ops); } diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index bb9266ea3785..3abb430af9e6 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -25,19 +25,16 @@ MODULE_DESCRIPTION("iptables mangle table"); (1 << NF_INET_LOCAL_OUT) | \ (1 << NF_INET_POST_ROUTING)) -static int __net_init iptable_mangle_table_init(struct net *net); - static const struct xt_table packet_mangler = { .name = "mangle", .valid_hooks = MANGLE_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV4, .priority = NF_IP_PRI_MANGLE, - .table_init = iptable_mangle_table_init, }; static unsigned int -ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state) +ipt_mangle_out(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) { unsigned int ret; const struct iphdr *iph; @@ -53,7 +50,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state) daddr = iph->daddr; tos = iph->tos; - ret = ipt_do_table(skb, state, state->net->ipv4.iptable_mangle); + ret = ipt_do_table(priv, skb, state); /* Reroute for ANY change. */ if (ret != NF_DROP && ret != NF_STOLEN) { iph = ip_hdr(skb); @@ -62,7 +59,7 @@ ipt_mangle_out(struct sk_buff *skb, const struct nf_hook_state *state) iph->daddr != daddr || skb->mark != mark || iph->tos != tos) { - err = ip_route_me_harder(state->net, skb, RTN_UNSPEC); + err = ip_route_me_harder(state->net, state->sk, skb, RTN_UNSPEC); if (err < 0) ret = NF_DROP_ERR(err); } @@ -78,68 +75,67 @@ iptable_mangle_hook(void *priv, const struct nf_hook_state *state) { if (state->hook == NF_INET_LOCAL_OUT) - return ipt_mangle_out(skb, state); - return ipt_do_table(skb, state, state->net->ipv4.iptable_mangle); + return ipt_mangle_out(priv, skb, state); + return ipt_do_table(priv, skb, state); } static struct nf_hook_ops *mangle_ops __read_mostly; -static int __net_init iptable_mangle_table_init(struct net *net) +static int iptable_mangle_table_init(struct net *net) { struct ipt_replace *repl; int ret; - if (net->ipv4.iptable_mangle) - return 0; - repl = ipt_alloc_initial_table(&packet_mangler); if (repl == NULL) return -ENOMEM; - ret = ipt_register_table(net, &packet_mangler, repl, mangle_ops, - &net->ipv4.iptable_mangle); + ret = ipt_register_table(net, &packet_mangler, repl, mangle_ops); kfree(repl); return ret; } +static void __net_exit iptable_mangle_net_pre_exit(struct net *net) +{ + ipt_unregister_table_pre_exit(net, "mangle"); +} + static void __net_exit iptable_mangle_net_exit(struct net *net) { - if (!net->ipv4.iptable_mangle) - return; - ipt_unregister_table(net, net->ipv4.iptable_mangle, mangle_ops); - net->ipv4.iptable_mangle = NULL; + ipt_unregister_table_exit(net, "mangle"); } static struct pernet_operations iptable_mangle_net_ops = { + .pre_exit = iptable_mangle_net_pre_exit, .exit = iptable_mangle_net_exit, }; static int __init iptable_mangle_init(void) { - int ret; + int ret = xt_register_template(&packet_mangler, + iptable_mangle_table_init); + if (ret < 0) + return ret; mangle_ops = xt_hook_ops_alloc(&packet_mangler, iptable_mangle_hook); if (IS_ERR(mangle_ops)) { + xt_unregister_template(&packet_mangler); ret = PTR_ERR(mangle_ops); return ret; } ret = register_pernet_subsys(&iptable_mangle_net_ops); if (ret < 0) { + xt_unregister_template(&packet_mangler); kfree(mangle_ops); return ret; } - ret = iptable_mangle_table_init(&init_net); - if (ret) { - unregister_pernet_subsys(&iptable_mangle_net_ops); - kfree(mangle_ops); - } - return ret; } static void __exit iptable_mangle_fini(void) { unregister_pernet_subsys(&iptable_mangle_net_ops); + xt_unregister_template(&packet_mangler); kfree(mangle_ops); } diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index ad33687b7444..56f6ecc43451 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -13,7 +13,11 @@ #include <net/netfilter/nf_nat.h> -static int __net_init iptable_nat_table_init(struct net *net); +struct iptable_nat_pernet { + struct nf_hook_ops *nf_nat_ops; +}; + +static unsigned int iptable_nat_net_id __read_mostly; static const struct xt_table nf_nat_ipv4_table = { .name = "nat", @@ -23,37 +27,29 @@ static const struct xt_table nf_nat_ipv4_table = { (1 << NF_INET_LOCAL_IN), .me = THIS_MODULE, .af = NFPROTO_IPV4, - .table_init = iptable_nat_table_init, }; -static unsigned int iptable_nat_do_chain(void *priv, - struct sk_buff *skb, - const struct nf_hook_state *state) -{ - return ipt_do_table(skb, state, state->net->ipv4.nat_table); -} - static const struct nf_hook_ops nf_nat_ipv4_ops[] = { { - .hook = iptable_nat_do_chain, + .hook = ipt_do_table, .pf = NFPROTO_IPV4, .hooknum = NF_INET_PRE_ROUTING, .priority = NF_IP_PRI_NAT_DST, }, { - .hook = iptable_nat_do_chain, + .hook = ipt_do_table, .pf = NFPROTO_IPV4, .hooknum = NF_INET_POST_ROUTING, .priority = NF_IP_PRI_NAT_SRC, }, { - .hook = iptable_nat_do_chain, + .hook = ipt_do_table, .pf = NFPROTO_IPV4, .hooknum = NF_INET_LOCAL_OUT, .priority = NF_IP_PRI_NAT_DST, }, { - .hook = iptable_nat_do_chain, + .hook = ipt_do_table, .pf = NFPROTO_IPV4, .hooknum = NF_INET_LOCAL_IN, .priority = NF_IP_PRI_NAT_SRC, @@ -62,86 +58,112 @@ static const struct nf_hook_ops nf_nat_ipv4_ops[] = { static int ipt_nat_register_lookups(struct net *net) { + struct iptable_nat_pernet *xt_nat_net; + struct nf_hook_ops *ops; + struct xt_table *table; int i, ret; + xt_nat_net = net_generic(net, iptable_nat_net_id); + table = xt_find_table(net, NFPROTO_IPV4, "nat"); + if (WARN_ON_ONCE(!table)) + return -ENOENT; + + ops = kmemdup(nf_nat_ipv4_ops, sizeof(nf_nat_ipv4_ops), GFP_KERNEL); + if (!ops) + return -ENOMEM; + for (i = 0; i < ARRAY_SIZE(nf_nat_ipv4_ops); i++) { - ret = nf_nat_ipv4_register_fn(net, &nf_nat_ipv4_ops[i]); + ops[i].priv = table; + ret = nf_nat_ipv4_register_fn(net, &ops[i]); if (ret) { while (i) - nf_nat_ipv4_unregister_fn(net, &nf_nat_ipv4_ops[--i]); + nf_nat_ipv4_unregister_fn(net, &ops[--i]); + kfree(ops); return ret; } } + xt_nat_net->nf_nat_ops = ops; return 0; } static void ipt_nat_unregister_lookups(struct net *net) { + struct iptable_nat_pernet *xt_nat_net = net_generic(net, iptable_nat_net_id); + struct nf_hook_ops *ops = xt_nat_net->nf_nat_ops; int i; + if (!ops) + return; + for (i = 0; i < ARRAY_SIZE(nf_nat_ipv4_ops); i++) - nf_nat_ipv4_unregister_fn(net, &nf_nat_ipv4_ops[i]); + nf_nat_ipv4_unregister_fn(net, &ops[i]); + + kfree(ops); } -static int __net_init iptable_nat_table_init(struct net *net) +static int iptable_nat_table_init(struct net *net) { struct ipt_replace *repl; int ret; - if (net->ipv4.nat_table) - return 0; - repl = ipt_alloc_initial_table(&nf_nat_ipv4_table); if (repl == NULL) return -ENOMEM; - ret = ipt_register_table(net, &nf_nat_ipv4_table, repl, - NULL, &net->ipv4.nat_table); + + ret = ipt_register_table(net, &nf_nat_ipv4_table, repl, NULL); if (ret < 0) { kfree(repl); return ret; } ret = ipt_nat_register_lookups(net); - if (ret < 0) { - ipt_unregister_table(net, net->ipv4.nat_table, NULL); - net->ipv4.nat_table = NULL; - } + if (ret < 0) + ipt_unregister_table_exit(net, "nat"); kfree(repl); return ret; } -static void __net_exit iptable_nat_net_exit(struct net *net) +static void __net_exit iptable_nat_net_pre_exit(struct net *net) { - if (!net->ipv4.nat_table) - return; ipt_nat_unregister_lookups(net); - ipt_unregister_table(net, net->ipv4.nat_table, NULL); - net->ipv4.nat_table = NULL; +} + +static void __net_exit iptable_nat_net_exit(struct net *net) +{ + ipt_unregister_table_exit(net, "nat"); } static struct pernet_operations iptable_nat_net_ops = { + .pre_exit = iptable_nat_net_pre_exit, .exit = iptable_nat_net_exit, + .id = &iptable_nat_net_id, + .size = sizeof(struct iptable_nat_pernet), }; static int __init iptable_nat_init(void) { - int ret = register_pernet_subsys(&iptable_nat_net_ops); + int ret = xt_register_template(&nf_nat_ipv4_table, + iptable_nat_table_init); + + if (ret < 0) + return ret; - if (ret) + ret = register_pernet_subsys(&iptable_nat_net_ops); + if (ret < 0) { + xt_unregister_template(&nf_nat_ipv4_table); return ret; + } - ret = iptable_nat_table_init(&init_net); - if (ret) - unregister_pernet_subsys(&iptable_nat_net_ops); return ret; } static void __exit iptable_nat_exit(void) { unregister_pernet_subsys(&iptable_nat_net_ops); + xt_unregister_template(&nf_nat_ipv4_table); } module_init(iptable_nat_init); diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 69697eb4bfc6..ca5e5b21587c 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -12,8 +12,6 @@ #define RAW_VALID_HOOKS ((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_OUT)) -static int __net_init iptable_raw_table_init(struct net *net); - static bool raw_before_defrag __read_mostly; MODULE_PARM_DESC(raw_before_defrag, "Enable raw table before defrag"); module_param(raw_before_defrag, bool, 0000); @@ -24,7 +22,6 @@ static const struct xt_table packet_raw = { .me = THIS_MODULE, .af = NFPROTO_IPV4, .priority = NF_IP_PRI_RAW, - .table_init = iptable_raw_table_init, }; static const struct xt_table packet_raw_before_defrag = { @@ -33,20 +30,11 @@ static const struct xt_table packet_raw_before_defrag = { .me = THIS_MODULE, .af = NFPROTO_IPV4, .priority = NF_IP_PRI_RAW_BEFORE_DEFRAG, - .table_init = iptable_raw_table_init, }; -/* The work comes in here from netfilter.c. */ -static unsigned int -iptable_raw_hook(void *priv, struct sk_buff *skb, - const struct nf_hook_state *state) -{ - return ipt_do_table(skb, state, state->net->ipv4.iptable_raw); -} - static struct nf_hook_ops *rawtable_ops __read_mostly; -static int __net_init iptable_raw_table_init(struct net *net) +static int iptable_raw_table_init(struct net *net) { struct ipt_replace *repl; const struct xt_table *table = &packet_raw; @@ -55,27 +43,26 @@ static int __net_init iptable_raw_table_init(struct net *net) if (raw_before_defrag) table = &packet_raw_before_defrag; - if (net->ipv4.iptable_raw) - return 0; - repl = ipt_alloc_initial_table(table); if (repl == NULL) return -ENOMEM; - ret = ipt_register_table(net, table, repl, rawtable_ops, - &net->ipv4.iptable_raw); + ret = ipt_register_table(net, table, repl, rawtable_ops); kfree(repl); return ret; } +static void __net_exit iptable_raw_net_pre_exit(struct net *net) +{ + ipt_unregister_table_pre_exit(net, "raw"); +} + static void __net_exit iptable_raw_net_exit(struct net *net) { - if (!net->ipv4.iptable_raw) - return; - ipt_unregister_table(net, net->ipv4.iptable_raw, rawtable_ops); - net->ipv4.iptable_raw = NULL; + ipt_unregister_table_exit(net, "raw"); } static struct pernet_operations iptable_raw_net_ops = { + .pre_exit = iptable_raw_net_pre_exit, .exit = iptable_raw_net_exit, }; @@ -90,22 +77,24 @@ static int __init iptable_raw_init(void) pr_info("Enabling raw table before defrag\n"); } - rawtable_ops = xt_hook_ops_alloc(table, iptable_raw_hook); - if (IS_ERR(rawtable_ops)) + ret = xt_register_template(table, + iptable_raw_table_init); + if (ret < 0) + return ret; + + rawtable_ops = xt_hook_ops_alloc(table, ipt_do_table); + if (IS_ERR(rawtable_ops)) { + xt_unregister_template(table); return PTR_ERR(rawtable_ops); + } ret = register_pernet_subsys(&iptable_raw_net_ops); if (ret < 0) { + xt_unregister_template(table); kfree(rawtable_ops); return ret; } - ret = iptable_raw_table_init(&init_net); - if (ret) { - unregister_pernet_subsys(&iptable_raw_net_ops); - kfree(rawtable_ops); - } - return ret; } @@ -113,6 +102,7 @@ static void __exit iptable_raw_fini(void) { unregister_pernet_subsys(&iptable_raw_net_ops); kfree(rawtable_ops); + xt_unregister_template(&packet_raw); } module_init(iptable_raw_init); diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c index ac633c1db97e..d885443cb267 100644 --- a/net/ipv4/netfilter/iptable_security.c +++ b/net/ipv4/netfilter/iptable_security.c @@ -25,76 +25,65 @@ MODULE_DESCRIPTION("iptables security table, for MAC rules"); (1 << NF_INET_FORWARD) | \ (1 << NF_INET_LOCAL_OUT) -static int __net_init iptable_security_table_init(struct net *net); - static const struct xt_table security_table = { .name = "security", .valid_hooks = SECURITY_VALID_HOOKS, .me = THIS_MODULE, .af = NFPROTO_IPV4, .priority = NF_IP_PRI_SECURITY, - .table_init = iptable_security_table_init, }; -static unsigned int -iptable_security_hook(void *priv, struct sk_buff *skb, - const struct nf_hook_state *state) -{ - return ipt_do_table(skb, state, state->net->ipv4.iptable_security); -} - static struct nf_hook_ops *sectbl_ops __read_mostly; -static int __net_init iptable_security_table_init(struct net *net) +static int iptable_security_table_init(struct net *net) { struct ipt_replace *repl; int ret; - if (net->ipv4.iptable_security) - return 0; - repl = ipt_alloc_initial_table(&security_table); if (repl == NULL) return -ENOMEM; - ret = ipt_register_table(net, &security_table, repl, sectbl_ops, - &net->ipv4.iptable_security); + ret = ipt_register_table(net, &security_table, repl, sectbl_ops); kfree(repl); return ret; } -static void __net_exit iptable_security_net_exit(struct net *net) +static void __net_exit iptable_security_net_pre_exit(struct net *net) { - if (!net->ipv4.iptable_security) - return; + ipt_unregister_table_pre_exit(net, "security"); +} - ipt_unregister_table(net, net->ipv4.iptable_security, sectbl_ops); - net->ipv4.iptable_security = NULL; +static void __net_exit iptable_security_net_exit(struct net *net) +{ + ipt_unregister_table_exit(net, "security"); } static struct pernet_operations iptable_security_net_ops = { + .pre_exit = iptable_security_net_pre_exit, .exit = iptable_security_net_exit, }; static int __init iptable_security_init(void) { - int ret; + int ret = xt_register_template(&security_table, + iptable_security_table_init); + + if (ret < 0) + return ret; - sectbl_ops = xt_hook_ops_alloc(&security_table, iptable_security_hook); - if (IS_ERR(sectbl_ops)) + sectbl_ops = xt_hook_ops_alloc(&security_table, ipt_do_table); + if (IS_ERR(sectbl_ops)) { + xt_unregister_template(&security_table); return PTR_ERR(sectbl_ops); + } ret = register_pernet_subsys(&iptable_security_net_ops); if (ret < 0) { + xt_unregister_template(&security_table); kfree(sectbl_ops); return ret; } - ret = iptable_security_table_init(&init_net); - if (ret) { - unregister_pernet_subsys(&iptable_security_net_ops); - kfree(sectbl_ops); - } - return ret; } @@ -102,6 +91,7 @@ static void __exit iptable_security_fini(void) { unregister_pernet_subsys(&iptable_security_net_ops); kfree(sectbl_ops); + xt_unregister_template(&security_table); } module_init(iptable_security_init); diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index 8115611aa47d..e61ea428ea18 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -106,10 +106,10 @@ static const struct nf_hook_ops ipv4_defrag_ops[] = { static void __net_exit defrag4_net_exit(struct net *net) { - if (net->nf.defrag_ipv4) { + if (net->nf.defrag_ipv4_users) { nf_unregister_net_hooks(net, ipv4_defrag_ops, ARRAY_SIZE(ipv4_defrag_ops)); - net->nf.defrag_ipv4 = false; + net->nf.defrag_ipv4_users = 0; } } @@ -131,19 +131,21 @@ int nf_defrag_ipv4_enable(struct net *net) { int err = 0; - might_sleep(); - - if (net->nf.defrag_ipv4) - return 0; - mutex_lock(&defrag4_mutex); - if (net->nf.defrag_ipv4) + if (net->nf.defrag_ipv4_users == UINT_MAX) { + err = -EOVERFLOW; goto out_unlock; + } + + if (net->nf.defrag_ipv4_users) { + net->nf.defrag_ipv4_users++; + goto out_unlock; + } err = nf_register_net_hooks(net, ipv4_defrag_ops, ARRAY_SIZE(ipv4_defrag_ops)); if (err == 0) - net->nf.defrag_ipv4 = true; + net->nf.defrag_ipv4_users = 1; out_unlock: mutex_unlock(&defrag4_mutex); @@ -151,6 +153,20 @@ int nf_defrag_ipv4_enable(struct net *net) } EXPORT_SYMBOL_GPL(nf_defrag_ipv4_enable); +void nf_defrag_ipv4_disable(struct net *net) +{ + mutex_lock(&defrag4_mutex); + if (net->nf.defrag_ipv4_users) { + net->nf.defrag_ipv4_users--; + if (net->nf.defrag_ipv4_users == 0) + nf_unregister_net_hooks(net, ipv4_defrag_ops, + ARRAY_SIZE(ipv4_defrag_ops)); + } + + mutex_unlock(&defrag4_mutex); +} +EXPORT_SYMBOL_GPL(nf_defrag_ipv4_disable); + module_init(nf_defrag_init); module_exit(nf_defrag_fini); diff --git a/net/ipv4/netfilter/nf_flow_table_ipv4.c b/net/ipv4/netfilter/nf_flow_table_ipv4.c deleted file mode 100644 index e32e41b99f0f..000000000000 --- a/net/ipv4/netfilter/nf_flow_table_ipv4.c +++ /dev/null @@ -1,36 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/module.h> -#include <linux/netfilter.h> -#include <net/netfilter/nf_flow_table.h> -#include <net/netfilter/nf_tables.h> - -static struct nf_flowtable_type flowtable_ipv4 = { - .family = NFPROTO_IPV4, - .init = nf_flow_table_init, - .setup = nf_flow_table_offload_setup, - .action = nf_flow_rule_route_ipv4, - .free = nf_flow_table_free, - .hook = nf_flow_offload_ip_hook, - .owner = THIS_MODULE, -}; - -static int __init nf_flow_ipv4_module_init(void) -{ - nft_register_flowtable_type(&flowtable_ipv4); - - return 0; -} - -static void __exit nf_flow_ipv4_module_exit(void) -{ - nft_unregister_flowtable_type(&flowtable_ipv4); -} - -module_init(nf_flow_ipv4_module_init); -module_exit(nf_flow_ipv4_module_exit); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>"); -MODULE_ALIAS_NF_FLOWTABLE(AF_INET); diff --git a/net/ipv4/netfilter/nf_log_arp.c b/net/ipv4/netfilter/nf_log_arp.c deleted file mode 100644 index 7a83f881efa9..000000000000 --- a/net/ipv4/netfilter/nf_log_arp.c +++ /dev/null @@ -1,157 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * (C) 2014 by Pablo Neira Ayuso <pablo@netfilter.org> - * - * Based on code from ebt_log from: - * - * Bart De Schuymer <bdschuym@pandora.be> - * Harald Welte <laforge@netfilter.org> - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/spinlock.h> -#include <linux/skbuff.h> -#include <linux/if_arp.h> -#include <linux/ip.h> -#include <net/route.h> - -#include <linux/netfilter.h> -#include <linux/netfilter/xt_LOG.h> -#include <net/netfilter/nf_log.h> - -static const struct nf_loginfo default_loginfo = { - .type = NF_LOG_TYPE_LOG, - .u = { - .log = { - .level = LOGLEVEL_NOTICE, - .logflags = NF_LOG_DEFAULT_MASK, - }, - }, -}; - -struct arppayload { - unsigned char mac_src[ETH_ALEN]; - unsigned char ip_src[4]; - unsigned char mac_dst[ETH_ALEN]; - unsigned char ip_dst[4]; -}; - -static void dump_arp_packet(struct nf_log_buf *m, - const struct nf_loginfo *info, - const struct sk_buff *skb, unsigned int nhoff) -{ - const struct arphdr *ah; - struct arphdr _arph; - const struct arppayload *ap; - struct arppayload _arpp; - - ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); - if (ah == NULL) { - nf_log_buf_add(m, "TRUNCATED"); - return; - } - nf_log_buf_add(m, "ARP HTYPE=%d PTYPE=0x%04x OPCODE=%d", - ntohs(ah->ar_hrd), ntohs(ah->ar_pro), ntohs(ah->ar_op)); - - /* If it's for Ethernet and the lengths are OK, then log the ARP - * payload. - */ - if (ah->ar_hrd != htons(ARPHRD_ETHER) || - ah->ar_hln != ETH_ALEN || - ah->ar_pln != sizeof(__be32)) - return; - - ap = skb_header_pointer(skb, sizeof(_arph), sizeof(_arpp), &_arpp); - if (ap == NULL) { - nf_log_buf_add(m, " INCOMPLETE [%zu bytes]", - skb->len - sizeof(_arph)); - return; - } - nf_log_buf_add(m, " MACSRC=%pM IPSRC=%pI4 MACDST=%pM IPDST=%pI4", - ap->mac_src, ap->ip_src, ap->mac_dst, ap->ip_dst); -} - -static void nf_log_arp_packet(struct net *net, u_int8_t pf, - unsigned int hooknum, const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const struct nf_loginfo *loginfo, - const char *prefix) -{ - struct nf_log_buf *m; - - /* FIXME: Disabled from containers until syslog ns is supported */ - if (!net_eq(net, &init_net) && !sysctl_nf_log_all_netns) - return; - - m = nf_log_buf_open(); - - if (!loginfo) - loginfo = &default_loginfo; - - nf_log_dump_packet_common(m, pf, hooknum, skb, in, out, loginfo, - prefix); - dump_arp_packet(m, loginfo, skb, 0); - - nf_log_buf_close(m); -} - -static struct nf_logger nf_arp_logger __read_mostly = { - .name = "nf_log_arp", - .type = NF_LOG_TYPE_LOG, - .logfn = nf_log_arp_packet, - .me = THIS_MODULE, -}; - -static int __net_init nf_log_arp_net_init(struct net *net) -{ - return nf_log_set(net, NFPROTO_ARP, &nf_arp_logger); -} - -static void __net_exit nf_log_arp_net_exit(struct net *net) -{ - nf_log_unset(net, &nf_arp_logger); -} - -static struct pernet_operations nf_log_arp_net_ops = { - .init = nf_log_arp_net_init, - .exit = nf_log_arp_net_exit, -}; - -static int __init nf_log_arp_init(void) -{ - int ret; - - ret = register_pernet_subsys(&nf_log_arp_net_ops); - if (ret < 0) - return ret; - - ret = nf_log_register(NFPROTO_ARP, &nf_arp_logger); - if (ret < 0) { - pr_err("failed to register logger\n"); - goto err1; - } - - return 0; - -err1: - unregister_pernet_subsys(&nf_log_arp_net_ops); - return ret; -} - -static void __exit nf_log_arp_exit(void) -{ - unregister_pernet_subsys(&nf_log_arp_net_ops); - nf_log_unregister(&nf_arp_logger); -} - -module_init(nf_log_arp_init); -module_exit(nf_log_arp_exit); - -MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>"); -MODULE_DESCRIPTION("Netfilter ARP packet logging"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS_NF_LOGGER(3, 0); diff --git a/net/ipv4/netfilter/nf_log_ipv4.c b/net/ipv4/netfilter/nf_log_ipv4.c deleted file mode 100644 index 4b2d49cc9f1a..000000000000 --- a/net/ipv4/netfilter/nf_log_ipv4.c +++ /dev/null @@ -1,393 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/spinlock.h> -#include <linux/skbuff.h> -#include <linux/if_arp.h> -#include <linux/ip.h> -#include <net/ipv6.h> -#include <net/icmp.h> -#include <net/udp.h> -#include <net/tcp.h> -#include <net/route.h> - -#include <linux/netfilter.h> -#include <linux/netfilter/xt_LOG.h> -#include <net/netfilter/nf_log.h> - -static const struct nf_loginfo default_loginfo = { - .type = NF_LOG_TYPE_LOG, - .u = { - .log = { - .level = LOGLEVEL_NOTICE, - .logflags = NF_LOG_DEFAULT_MASK, - }, - }, -}; - -/* One level of recursion won't kill us */ -static void dump_ipv4_packet(struct net *net, struct nf_log_buf *m, - const struct nf_loginfo *info, - const struct sk_buff *skb, unsigned int iphoff) -{ - struct iphdr _iph; - const struct iphdr *ih; - unsigned int logflags; - - if (info->type == NF_LOG_TYPE_LOG) - logflags = info->u.log.logflags; - else - logflags = NF_LOG_DEFAULT_MASK; - - ih = skb_header_pointer(skb, iphoff, sizeof(_iph), &_iph); - if (ih == NULL) { - nf_log_buf_add(m, "TRUNCATED"); - return; - } - - /* Important fields: - * TOS, len, DF/MF, fragment offset, TTL, src, dst, options. */ - /* Max length: 40 "SRC=255.255.255.255 DST=255.255.255.255 " */ - nf_log_buf_add(m, "SRC=%pI4 DST=%pI4 ", &ih->saddr, &ih->daddr); - - /* Max length: 46 "LEN=65535 TOS=0xFF PREC=0xFF TTL=255 ID=65535 " */ - nf_log_buf_add(m, "LEN=%u TOS=0x%02X PREC=0x%02X TTL=%u ID=%u ", - ntohs(ih->tot_len), ih->tos & IPTOS_TOS_MASK, - ih->tos & IPTOS_PREC_MASK, ih->ttl, ntohs(ih->id)); - - /* Max length: 6 "CE DF MF " */ - if (ntohs(ih->frag_off) & IP_CE) - nf_log_buf_add(m, "CE "); - if (ntohs(ih->frag_off) & IP_DF) - nf_log_buf_add(m, "DF "); - if (ntohs(ih->frag_off) & IP_MF) - nf_log_buf_add(m, "MF "); - - /* Max length: 11 "FRAG:65535 " */ - if (ntohs(ih->frag_off) & IP_OFFSET) - nf_log_buf_add(m, "FRAG:%u ", ntohs(ih->frag_off) & IP_OFFSET); - - if ((logflags & NF_LOG_IPOPT) && - ih->ihl * 4 > sizeof(struct iphdr)) { - const unsigned char *op; - unsigned char _opt[4 * 15 - sizeof(struct iphdr)]; - unsigned int i, optsize; - - optsize = ih->ihl * 4 - sizeof(struct iphdr); - op = skb_header_pointer(skb, iphoff+sizeof(_iph), - optsize, _opt); - if (op == NULL) { - nf_log_buf_add(m, "TRUNCATED"); - return; - } - - /* Max length: 127 "OPT (" 15*4*2chars ") " */ - nf_log_buf_add(m, "OPT ("); - for (i = 0; i < optsize; i++) - nf_log_buf_add(m, "%02X", op[i]); - nf_log_buf_add(m, ") "); - } - - switch (ih->protocol) { - case IPPROTO_TCP: - if (nf_log_dump_tcp_header(m, skb, ih->protocol, - ntohs(ih->frag_off) & IP_OFFSET, - iphoff+ih->ihl*4, logflags)) - return; - break; - case IPPROTO_UDP: - case IPPROTO_UDPLITE: - if (nf_log_dump_udp_header(m, skb, ih->protocol, - ntohs(ih->frag_off) & IP_OFFSET, - iphoff+ih->ihl*4)) - return; - break; - case IPPROTO_ICMP: { - struct icmphdr _icmph; - const struct icmphdr *ich; - static const size_t required_len[NR_ICMP_TYPES+1] - = { [ICMP_ECHOREPLY] = 4, - [ICMP_DEST_UNREACH] - = 8 + sizeof(struct iphdr), - [ICMP_SOURCE_QUENCH] - = 8 + sizeof(struct iphdr), - [ICMP_REDIRECT] - = 8 + sizeof(struct iphdr), - [ICMP_ECHO] = 4, - [ICMP_TIME_EXCEEDED] - = 8 + sizeof(struct iphdr), - [ICMP_PARAMETERPROB] - = 8 + sizeof(struct iphdr), - [ICMP_TIMESTAMP] = 20, - [ICMP_TIMESTAMPREPLY] = 20, - [ICMP_ADDRESS] = 12, - [ICMP_ADDRESSREPLY] = 12 }; - - /* Max length: 11 "PROTO=ICMP " */ - nf_log_buf_add(m, "PROTO=ICMP "); - - if (ntohs(ih->frag_off) & IP_OFFSET) - break; - - /* Max length: 25 "INCOMPLETE [65535 bytes] " */ - ich = skb_header_pointer(skb, iphoff + ih->ihl * 4, - sizeof(_icmph), &_icmph); - if (ich == NULL) { - nf_log_buf_add(m, "INCOMPLETE [%u bytes] ", - skb->len - iphoff - ih->ihl*4); - break; - } - - /* Max length: 18 "TYPE=255 CODE=255 " */ - nf_log_buf_add(m, "TYPE=%u CODE=%u ", ich->type, ich->code); - - /* Max length: 25 "INCOMPLETE [65535 bytes] " */ - if (ich->type <= NR_ICMP_TYPES && - required_len[ich->type] && - skb->len-iphoff-ih->ihl*4 < required_len[ich->type]) { - nf_log_buf_add(m, "INCOMPLETE [%u bytes] ", - skb->len - iphoff - ih->ihl*4); - break; - } - - switch (ich->type) { - case ICMP_ECHOREPLY: - case ICMP_ECHO: - /* Max length: 19 "ID=65535 SEQ=65535 " */ - nf_log_buf_add(m, "ID=%u SEQ=%u ", - ntohs(ich->un.echo.id), - ntohs(ich->un.echo.sequence)); - break; - - case ICMP_PARAMETERPROB: - /* Max length: 14 "PARAMETER=255 " */ - nf_log_buf_add(m, "PARAMETER=%u ", - ntohl(ich->un.gateway) >> 24); - break; - case ICMP_REDIRECT: - /* Max length: 24 "GATEWAY=255.255.255.255 " */ - nf_log_buf_add(m, "GATEWAY=%pI4 ", &ich->un.gateway); - /* Fall through */ - case ICMP_DEST_UNREACH: - case ICMP_SOURCE_QUENCH: - case ICMP_TIME_EXCEEDED: - /* Max length: 3+maxlen */ - if (!iphoff) { /* Only recurse once. */ - nf_log_buf_add(m, "["); - dump_ipv4_packet(net, m, info, skb, - iphoff + ih->ihl*4+sizeof(_icmph)); - nf_log_buf_add(m, "] "); - } - - /* Max length: 10 "MTU=65535 " */ - if (ich->type == ICMP_DEST_UNREACH && - ich->code == ICMP_FRAG_NEEDED) { - nf_log_buf_add(m, "MTU=%u ", - ntohs(ich->un.frag.mtu)); - } - } - break; - } - /* Max Length */ - case IPPROTO_AH: { - struct ip_auth_hdr _ahdr; - const struct ip_auth_hdr *ah; - - if (ntohs(ih->frag_off) & IP_OFFSET) - break; - - /* Max length: 9 "PROTO=AH " */ - nf_log_buf_add(m, "PROTO=AH "); - - /* Max length: 25 "INCOMPLETE [65535 bytes] " */ - ah = skb_header_pointer(skb, iphoff+ih->ihl*4, - sizeof(_ahdr), &_ahdr); - if (ah == NULL) { - nf_log_buf_add(m, "INCOMPLETE [%u bytes] ", - skb->len - iphoff - ih->ihl*4); - break; - } - - /* Length: 15 "SPI=0xF1234567 " */ - nf_log_buf_add(m, "SPI=0x%x ", ntohl(ah->spi)); - break; - } - case IPPROTO_ESP: { - struct ip_esp_hdr _esph; - const struct ip_esp_hdr *eh; - - /* Max length: 10 "PROTO=ESP " */ - nf_log_buf_add(m, "PROTO=ESP "); - - if (ntohs(ih->frag_off) & IP_OFFSET) - break; - - /* Max length: 25 "INCOMPLETE [65535 bytes] " */ - eh = skb_header_pointer(skb, iphoff+ih->ihl*4, - sizeof(_esph), &_esph); - if (eh == NULL) { - nf_log_buf_add(m, "INCOMPLETE [%u bytes] ", - skb->len - iphoff - ih->ihl*4); - break; - } - - /* Length: 15 "SPI=0xF1234567 " */ - nf_log_buf_add(m, "SPI=0x%x ", ntohl(eh->spi)); - break; - } - /* Max length: 10 "PROTO 255 " */ - default: - nf_log_buf_add(m, "PROTO=%u ", ih->protocol); - } - - /* Max length: 15 "UID=4294967295 " */ - if ((logflags & NF_LOG_UID) && !iphoff) - nf_log_dump_sk_uid_gid(net, m, skb->sk); - - /* Max length: 16 "MARK=0xFFFFFFFF " */ - if (!iphoff && skb->mark) - nf_log_buf_add(m, "MARK=0x%x ", skb->mark); - - /* Proto Max log string length */ - /* IP: 40+46+6+11+127 = 230 */ - /* TCP: 10+max(25,20+30+13+9+32+11+127) = 252 */ - /* UDP: 10+max(25,20) = 35 */ - /* UDPLITE: 14+max(25,20) = 39 */ - /* ICMP: 11+max(25, 18+25+max(19,14,24+3+n+10,3+n+10)) = 91+n */ - /* ESP: 10+max(25)+15 = 50 */ - /* AH: 9+max(25)+15 = 49 */ - /* unknown: 10 */ - - /* (ICMP allows recursion one level deep) */ - /* maxlen = IP + ICMP + IP + max(TCP,UDP,ICMP,unknown) */ - /* maxlen = 230+ 91 + 230 + 252 = 803 */ -} - -static void dump_ipv4_mac_header(struct nf_log_buf *m, - const struct nf_loginfo *info, - const struct sk_buff *skb) -{ - struct net_device *dev = skb->dev; - unsigned int logflags = 0; - - if (info->type == NF_LOG_TYPE_LOG) - logflags = info->u.log.logflags; - - if (!(logflags & NF_LOG_MACDECODE)) - goto fallback; - - switch (dev->type) { - case ARPHRD_ETHER: - nf_log_buf_add(m, "MACSRC=%pM MACDST=%pM MACPROTO=%04x ", - eth_hdr(skb)->h_source, eth_hdr(skb)->h_dest, - ntohs(eth_hdr(skb)->h_proto)); - return; - default: - break; - } - -fallback: - nf_log_buf_add(m, "MAC="); - if (dev->hard_header_len && - skb->mac_header != skb->network_header) { - const unsigned char *p = skb_mac_header(skb); - unsigned int i; - - nf_log_buf_add(m, "%02x", *p++); - for (i = 1; i < dev->hard_header_len; i++, p++) - nf_log_buf_add(m, ":%02x", *p); - } - nf_log_buf_add(m, " "); -} - -static void nf_log_ip_packet(struct net *net, u_int8_t pf, - unsigned int hooknum, const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const struct nf_loginfo *loginfo, - const char *prefix) -{ - struct nf_log_buf *m; - - /* FIXME: Disabled from containers until syslog ns is supported */ - if (!net_eq(net, &init_net) && !sysctl_nf_log_all_netns) - return; - - m = nf_log_buf_open(); - - if (!loginfo) - loginfo = &default_loginfo; - - nf_log_dump_packet_common(m, pf, hooknum, skb, in, - out, loginfo, prefix); - - if (in != NULL) - dump_ipv4_mac_header(m, loginfo, skb); - - dump_ipv4_packet(net, m, loginfo, skb, 0); - - nf_log_buf_close(m); -} - -static struct nf_logger nf_ip_logger __read_mostly = { - .name = "nf_log_ipv4", - .type = NF_LOG_TYPE_LOG, - .logfn = nf_log_ip_packet, - .me = THIS_MODULE, -}; - -static int __net_init nf_log_ipv4_net_init(struct net *net) -{ - return nf_log_set(net, NFPROTO_IPV4, &nf_ip_logger); -} - -static void __net_exit nf_log_ipv4_net_exit(struct net *net) -{ - nf_log_unset(net, &nf_ip_logger); -} - -static struct pernet_operations nf_log_ipv4_net_ops = { - .init = nf_log_ipv4_net_init, - .exit = nf_log_ipv4_net_exit, -}; - -static int __init nf_log_ipv4_init(void) -{ - int ret; - - ret = register_pernet_subsys(&nf_log_ipv4_net_ops); - if (ret < 0) - return ret; - - ret = nf_log_register(NFPROTO_IPV4, &nf_ip_logger); - if (ret < 0) { - pr_err("failed to register logger\n"); - goto err1; - } - - return 0; - -err1: - unregister_pernet_subsys(&nf_log_ipv4_net_ops); - return ret; -} - -static void __exit nf_log_ipv4_exit(void) -{ - unregister_pernet_subsys(&nf_log_ipv4_net_ops); - nf_log_unregister(&nf_ip_logger); -} - -module_init(nf_log_ipv4_init); -module_exit(nf_log_ipv4_exit); - -MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); -MODULE_DESCRIPTION("Netfilter IPv4 packet logging"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS_NF_LOGGER(AF_INET, 0); diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c index 3e2685c120c7..faee20af4856 100644 --- a/net/ipv4/netfilter/nf_nat_h323.c +++ b/net/ipv4/netfilter/nf_nat_h323.c @@ -291,20 +291,7 @@ static int nat_t120(struct sk_buff *skb, struct nf_conn *ct, exp->expectfn = nf_nat_follow_master; exp->dir = !dir; - /* Try to get same port: if not, try to change it. */ - for (; nated_port != 0; nated_port++) { - int ret; - - exp->tuple.dst.u.tcp.port = htons(nated_port); - ret = nf_ct_expect_related(exp, 0); - if (ret == 0) - break; - else if (ret != -EBUSY) { - nated_port = 0; - break; - } - } - + nated_port = nf_nat_exp_find_port(exp, nated_port); if (nated_port == 0) { /* No port available */ net_notice_ratelimited("nf_nat_h323: out of TCP ports\n"); return 0; @@ -347,20 +334,7 @@ static int nat_h245(struct sk_buff *skb, struct nf_conn *ct, if (info->sig_port[dir] == port) nated_port = ntohs(info->sig_port[!dir]); - /* Try to get same port: if not, try to change it. */ - for (; nated_port != 0; nated_port++) { - int ret; - - exp->tuple.dst.u.tcp.port = htons(nated_port); - ret = nf_ct_expect_related(exp, 0); - if (ret == 0) - break; - else if (ret != -EBUSY) { - nated_port = 0; - break; - } - } - + nated_port = nf_nat_exp_find_port(exp, nated_port); if (nated_port == 0) { /* No port available */ net_notice_ratelimited("nf_nat_q931: out of TCP ports\n"); return 0; @@ -439,20 +413,7 @@ static int nat_q931(struct sk_buff *skb, struct nf_conn *ct, if (info->sig_port[dir] == port) nated_port = ntohs(info->sig_port[!dir]); - /* Try to get same port: if not, try to change it. */ - for (; nated_port != 0; nated_port++) { - int ret; - - exp->tuple.dst.u.tcp.port = htons(nated_port); - ret = nf_ct_expect_related(exp, 0); - if (ret == 0) - break; - else if (ret != -EBUSY) { - nated_port = 0; - break; - } - } - + nated_port = nf_nat_exp_find_port(exp, nated_port); if (nated_port == 0) { /* No port available */ net_notice_ratelimited("nf_nat_ras: out of TCP ports\n"); return 0; @@ -532,20 +493,7 @@ static int nat_callforwarding(struct sk_buff *skb, struct nf_conn *ct, exp->expectfn = ip_nat_callforwarding_expect; exp->dir = !dir; - /* Try to get same port: if not, try to change it. */ - for (nated_port = ntohs(port); nated_port != 0; nated_port++) { - int ret; - - exp->tuple.dst.u.tcp.port = htons(nated_port); - ret = nf_ct_expect_related(exp, 0); - if (ret == 0) - break; - else if (ret != -EBUSY) { - nated_port = 0; - break; - } - } - + nated_port = nf_nat_exp_find_port(exp, ntohs(port)); if (nated_port == 0) { /* No port available */ net_notice_ratelimited("nf_nat_q931: out of TCP ports\n"); return 0; @@ -579,53 +527,39 @@ static struct nf_ct_helper_expectfn callforwarding_nat = { .expectfn = ip_nat_callforwarding_expect, }; +static const struct nfct_h323_nat_hooks nathooks = { + .set_h245_addr = set_h245_addr, + .set_h225_addr = set_h225_addr, + .set_sig_addr = set_sig_addr, + .set_ras_addr = set_ras_addr, + .nat_rtp_rtcp = nat_rtp_rtcp, + .nat_t120 = nat_t120, + .nat_h245 = nat_h245, + .nat_callforwarding = nat_callforwarding, + .nat_q931 = nat_q931, +}; + /****************************************************************************/ -static int __init init(void) +static int __init nf_nat_h323_init(void) { - BUG_ON(set_h245_addr_hook != NULL); - BUG_ON(set_h225_addr_hook != NULL); - BUG_ON(set_sig_addr_hook != NULL); - BUG_ON(set_ras_addr_hook != NULL); - BUG_ON(nat_rtp_rtcp_hook != NULL); - BUG_ON(nat_t120_hook != NULL); - BUG_ON(nat_h245_hook != NULL); - BUG_ON(nat_callforwarding_hook != NULL); - BUG_ON(nat_q931_hook != NULL); - - RCU_INIT_POINTER(set_h245_addr_hook, set_h245_addr); - RCU_INIT_POINTER(set_h225_addr_hook, set_h225_addr); - RCU_INIT_POINTER(set_sig_addr_hook, set_sig_addr); - RCU_INIT_POINTER(set_ras_addr_hook, set_ras_addr); - RCU_INIT_POINTER(nat_rtp_rtcp_hook, nat_rtp_rtcp); - RCU_INIT_POINTER(nat_t120_hook, nat_t120); - RCU_INIT_POINTER(nat_h245_hook, nat_h245); - RCU_INIT_POINTER(nat_callforwarding_hook, nat_callforwarding); - RCU_INIT_POINTER(nat_q931_hook, nat_q931); + RCU_INIT_POINTER(nfct_h323_nat_hook, &nathooks); nf_ct_helper_expectfn_register(&q931_nat); nf_ct_helper_expectfn_register(&callforwarding_nat); return 0; } /****************************************************************************/ -static void __exit fini(void) +static void __exit nf_nat_h323_fini(void) { - RCU_INIT_POINTER(set_h245_addr_hook, NULL); - RCU_INIT_POINTER(set_h225_addr_hook, NULL); - RCU_INIT_POINTER(set_sig_addr_hook, NULL); - RCU_INIT_POINTER(set_ras_addr_hook, NULL); - RCU_INIT_POINTER(nat_rtp_rtcp_hook, NULL); - RCU_INIT_POINTER(nat_t120_hook, NULL); - RCU_INIT_POINTER(nat_h245_hook, NULL); - RCU_INIT_POINTER(nat_callforwarding_hook, NULL); - RCU_INIT_POINTER(nat_q931_hook, NULL); + RCU_INIT_POINTER(nfct_h323_nat_hook, NULL); nf_ct_helper_expectfn_unregister(&q931_nat); nf_ct_helper_expectfn_unregister(&callforwarding_nat); synchronize_rcu(); } /****************************************************************************/ -module_init(init); -module_exit(fini); +module_init(nf_nat_h323_init); +module_exit(nf_nat_h323_fini); MODULE_AUTHOR("Jing Min Zhao <zhaojingmin@users.sourceforge.net>"); MODULE_DESCRIPTION("H.323 NAT helper"); diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c index b2aeb7bf5dac..fab357cc8559 100644 --- a/net/ipv4/netfilter/nf_nat_pptp.c +++ b/net/ipv4/netfilter/nf_nat_pptp.c @@ -3,7 +3,7 @@ * nf_nat_pptp.c * * NAT support for PPTP (Point to Point Tunneling Protocol). - * PPTP is a a protocol for creating virtual private networks. + * PPTP is a protocol for creating virtual private networks. * It is a specification defined by Microsoft and some vendors * working with Microsoft. PPTP is built on top of a modified * version of the Internet Generic Routing Encapsulation Protocol. @@ -166,9 +166,8 @@ pptp_outbound_pkt(struct sk_buff *skb, break; default: pr_debug("unknown outbound packet 0x%04x:%s\n", msg, - msg <= PPTP_MSG_MAX ? pptp_msg_name[msg] : - pptp_msg_name[0]); - /* fall through */ + pptp_msg_name(msg)); + fallthrough; case PPTP_SET_LINK_INFO: /* only need to NAT in case PAC is behind NAT box */ case PPTP_START_SESSION_REQUEST: @@ -268,10 +267,8 @@ pptp_inbound_pkt(struct sk_buff *skb, pcid_off = offsetof(union pptp_ctrl_union, setlink.peersCallID); break; default: - pr_debug("unknown inbound packet %s\n", - msg <= PPTP_MSG_MAX ? pptp_msg_name[msg] : - pptp_msg_name[0]); - /* fall through */ + pr_debug("unknown inbound packet %s\n", pptp_msg_name(msg)); + fallthrough; case PPTP_START_SESSION_REQUEST: case PPTP_START_SESSION_REPLY: case PPTP_STOP_SESSION_REQUEST: @@ -298,28 +295,24 @@ pptp_inbound_pkt(struct sk_buff *skb, return NF_ACCEPT; } +static const struct nf_nat_pptp_hook pptp_hooks = { + .outbound = pptp_outbound_pkt, + .inbound = pptp_inbound_pkt, + .exp_gre = pptp_exp_gre, + .expectfn = pptp_nat_expected, +}; + static int __init nf_nat_helper_pptp_init(void) { - BUG_ON(nf_nat_pptp_hook_outbound != NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_outbound, pptp_outbound_pkt); - - BUG_ON(nf_nat_pptp_hook_inbound != NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_inbound, pptp_inbound_pkt); - - BUG_ON(nf_nat_pptp_hook_exp_gre != NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_exp_gre, pptp_exp_gre); + WARN_ON(nf_nat_pptp_hook != NULL); + RCU_INIT_POINTER(nf_nat_pptp_hook, &pptp_hooks); - BUG_ON(nf_nat_pptp_hook_expectfn != NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_expectfn, pptp_nat_expected); return 0; } static void __exit nf_nat_helper_pptp_fini(void) { - RCU_INIT_POINTER(nf_nat_pptp_hook_expectfn, NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_exp_gre, NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_inbound, NULL); - RCU_INIT_POINTER(nf_nat_pptp_hook_outbound, NULL); + RCU_INIT_POINTER(nf_nat_pptp_hook, NULL); synchronize_rcu(); } diff --git a/net/ipv4/netfilter/nf_reject_ipv4.c b/net/ipv4/netfilter/nf_reject_ipv4.c index 2361fdac2c43..d640adcaf1b1 100644 --- a/net/ipv4/netfilter/nf_reject_ipv4.c +++ b/net/ipv4/netfilter/nf_reject_ipv4.c @@ -12,6 +12,130 @@ #include <linux/netfilter_ipv4.h> #include <linux/netfilter_bridge.h> +static int nf_reject_iphdr_validate(struct sk_buff *skb) +{ + struct iphdr *iph; + u32 len; + + if (!pskb_may_pull(skb, sizeof(struct iphdr))) + return 0; + + iph = ip_hdr(skb); + if (iph->ihl < 5 || iph->version != 4) + return 0; + + len = ntohs(iph->tot_len); + if (skb->len < len) + return 0; + else if (len < (iph->ihl*4)) + return 0; + + if (!pskb_may_pull(skb, iph->ihl*4)) + return 0; + + return 1; +} + +struct sk_buff *nf_reject_skb_v4_tcp_reset(struct net *net, + struct sk_buff *oldskb, + const struct net_device *dev, + int hook) +{ + const struct tcphdr *oth; + struct sk_buff *nskb; + struct iphdr *niph; + struct tcphdr _oth; + + if (!nf_reject_iphdr_validate(oldskb)) + return NULL; + + oth = nf_reject_ip_tcphdr_get(oldskb, &_oth, hook); + if (!oth) + return NULL; + + nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct tcphdr) + + LL_MAX_HEADER, GFP_ATOMIC); + if (!nskb) + return NULL; + + nskb->dev = (struct net_device *)dev; + + skb_reserve(nskb, LL_MAX_HEADER); + niph = nf_reject_iphdr_put(nskb, oldskb, IPPROTO_TCP, + READ_ONCE(net->ipv4.sysctl_ip_default_ttl)); + nf_reject_ip_tcphdr_put(nskb, oldskb, oth); + niph->tot_len = htons(nskb->len); + ip_send_check(niph); + + return nskb; +} +EXPORT_SYMBOL_GPL(nf_reject_skb_v4_tcp_reset); + +struct sk_buff *nf_reject_skb_v4_unreach(struct net *net, + struct sk_buff *oldskb, + const struct net_device *dev, + int hook, u8 code) +{ + struct sk_buff *nskb; + struct iphdr *niph; + struct icmphdr *icmph; + unsigned int len; + int dataoff; + __wsum csum; + u8 proto; + + if (!nf_reject_iphdr_validate(oldskb)) + return NULL; + + /* IP header checks: fragment. */ + if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET)) + return NULL; + + /* RFC says return as much as we can without exceeding 576 bytes. */ + len = min_t(unsigned int, 536, oldskb->len); + + if (!pskb_may_pull(oldskb, len)) + return NULL; + + if (pskb_trim_rcsum(oldskb, ntohs(ip_hdr(oldskb)->tot_len))) + return NULL; + + dataoff = ip_hdrlen(oldskb); + proto = ip_hdr(oldskb)->protocol; + + if (!skb_csum_unnecessary(oldskb) && + nf_reject_verify_csum(oldskb, dataoff, proto) && + nf_ip_checksum(oldskb, hook, ip_hdrlen(oldskb), proto)) + return NULL; + + nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct icmphdr) + + LL_MAX_HEADER + len, GFP_ATOMIC); + if (!nskb) + return NULL; + + nskb->dev = (struct net_device *)dev; + + skb_reserve(nskb, LL_MAX_HEADER); + niph = nf_reject_iphdr_put(nskb, oldskb, IPPROTO_ICMP, + READ_ONCE(net->ipv4.sysctl_ip_default_ttl)); + + skb_reset_transport_header(nskb); + icmph = skb_put_zero(nskb, sizeof(struct icmphdr)); + icmph->type = ICMP_DEST_UNREACH; + icmph->code = code; + + skb_put_data(nskb, skb_network_header(oldskb), len); + + csum = csum_partial((void *)icmph, len + sizeof(struct icmphdr), 0); + icmph->checksum = csum_fold(csum); + + niph->tot_len = htons(nskb->len); + ip_send_check(niph); + + return nskb; +} +EXPORT_SYMBOL_GPL(nf_reject_skb_v4_unreach); + const struct tcphdr *nf_reject_ip_tcphdr_get(struct sk_buff *oldskb, struct tcphdr *_oth, int hook) { @@ -96,8 +220,24 @@ void nf_reject_ip_tcphdr_put(struct sk_buff *nskb, const struct sk_buff *oldskb, } EXPORT_SYMBOL_GPL(nf_reject_ip_tcphdr_put); +static int nf_reject_fill_skb_dst(struct sk_buff *skb_in) +{ + struct dst_entry *dst = NULL; + struct flowi fl; + + memset(&fl, 0, sizeof(struct flowi)); + fl.u.ip4.daddr = ip_hdr(skb_in)->saddr; + nf_ip_route(dev_net(skb_in->dev), &dst, &fl, false); + if (!dst) + return -1; + + skb_dst_set(skb_in, dst); + return 0; +} + /* Send RST reply */ -void nf_send_reset(struct net *net, struct sk_buff *oldskb, int hook) +void nf_send_reset(struct net *net, struct sock *sk, struct sk_buff *oldskb, + int hook) { struct net_device *br_indev __maybe_unused; struct sk_buff *nskb; @@ -109,6 +249,10 @@ void nf_send_reset(struct net *net, struct sk_buff *oldskb, int hook) if (!oth) return; + if ((hook == NF_INET_PRE_ROUTING || hook == NF_INET_INGRESS) && + nf_reject_fill_skb_dst(oldskb) < 0) + return; + if (skb_rtable(oldskb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) return; @@ -126,8 +270,7 @@ void nf_send_reset(struct net *net, struct sk_buff *oldskb, int hook) niph = nf_reject_iphdr_put(nskb, oldskb, IPPROTO_TCP, ip4_dst_hoplimit(skb_dst(nskb))); nf_reject_ip_tcphdr_put(nskb, oldskb, oth); - - if (ip_route_me_harder(net, nskb, RTN_UNSPEC)) + if (ip_route_me_harder(net, sk, nskb, RTN_UNSPEC)) goto free_nskb; niph = ip_hdr(nskb); @@ -170,17 +313,23 @@ EXPORT_SYMBOL_GPL(nf_send_reset); void nf_send_unreach(struct sk_buff *skb_in, int code, int hook) { struct iphdr *iph = ip_hdr(skb_in); + int dataoff = ip_hdrlen(skb_in); u8 proto = iph->protocol; if (iph->frag_off & htons(IP_OFFSET)) return; - if (skb_csum_unnecessary(skb_in) || !nf_reject_verify_csum(proto)) { + if ((hook == NF_INET_PRE_ROUTING || hook == NF_INET_INGRESS) && + nf_reject_fill_skb_dst(skb_in) < 0) + return; + + if (skb_csum_unnecessary(skb_in) || + !nf_reject_verify_csum(skb_in, dataoff, proto)) { icmp_send(skb_in, ICMP_DEST_UNREACH, code, 0); return; } - if (nf_ip_checksum(skb_in, hook, ip_hdrlen(skb_in), proto) == 0) + if (nf_ip_checksum(skb_in, hook, dataoff, proto) == 0) icmp_send(skb_in, ICMP_DEST_UNREACH, code, 0); } EXPORT_SYMBOL_GPL(nf_send_unreach); diff --git a/net/ipv4/netfilter/nf_socket_ipv4.c b/net/ipv4/netfilter/nf_socket_ipv4.c index c94445b44d8c..a1350fc25838 100644 --- a/net/ipv4/netfilter/nf_socket_ipv4.c +++ b/net/ipv4/netfilter/nf_socket_ipv4.c @@ -71,8 +71,8 @@ nf_socket_get_sock_v4(struct net *net, struct sk_buff *skb, const int doff, { switch (protocol) { case IPPROTO_TCP: - return inet_lookup(net, &tcp_hashinfo, skb, doff, - saddr, sport, daddr, dport, + return inet_lookup(net, net->ipv4.tcp_death_row.hashinfo, + skb, doff, saddr, sport, daddr, dport, in->ifindex); case IPPROTO_UDP: return udp4_lib_lookup(net, saddr, sport, daddr, dport, @@ -84,11 +84,11 @@ nf_socket_get_sock_v4(struct net *net, struct sk_buff *skb, const int doff, struct sock *nf_sk_lookup_slow_v4(struct net *net, const struct sk_buff *skb, const struct net_device *indev) { - __be32 uninitialized_var(daddr), uninitialized_var(saddr); - __be16 uninitialized_var(dport), uninitialized_var(sport); + __be32 daddr, saddr; + __be16 dport, sport; const struct iphdr *iph = ip_hdr(skb); struct sk_buff *data_skb = NULL; - u8 uninitialized_var(protocol); + u8 protocol; #if IS_ENABLED(CONFIG_NF_CONNTRACK) enum ip_conntrack_info ctinfo; struct nf_conn const *ct; diff --git a/net/ipv4/netfilter/nf_tproxy_ipv4.c b/net/ipv4/netfilter/nf_tproxy_ipv4.c index b2bae0b0e42a..b22b2c745c76 100644 --- a/net/ipv4/netfilter/nf_tproxy_ipv4.c +++ b/net/ipv4/netfilter/nf_tproxy_ipv4.c @@ -79,6 +79,7 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, const struct net_device *in, const enum nf_tproxy_lookup_t lookup_type) { + struct inet_hashinfo *hinfo = net->ipv4.tcp_death_row.hashinfo; struct sock *sk; switch (protocol) { @@ -92,12 +93,10 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, switch (lookup_type) { case NF_TPROXY_LOOKUP_LISTENER: - sk = inet_lookup_listener(net, &tcp_hashinfo, skb, - ip_hdrlen(skb) + - __tcp_hdrlen(hp), - saddr, sport, - daddr, dport, - in->ifindex, 0); + sk = inet_lookup_listener(net, hinfo, skb, + ip_hdrlen(skb) + __tcp_hdrlen(hp), + saddr, sport, daddr, dport, + in->ifindex, 0); if (sk && !refcount_inc_not_zero(&sk->sk_refcnt)) sk = NULL; @@ -108,9 +107,8 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb, */ break; case NF_TPROXY_LOOKUP_ESTABLISHED: - sk = inet_lookup_established(net, &tcp_hashinfo, - saddr, sport, daddr, dport, - in->ifindex); + sk = inet_lookup_established(net, hinfo, saddr, sport, + daddr, dport, in->ifindex); break; default: BUG(); diff --git a/net/ipv4/netfilter/nft_dup_ipv4.c b/net/ipv4/netfilter/nft_dup_ipv4.c index abf89b972094..0bcd6aee6000 100644 --- a/net/ipv4/netfilter/nft_dup_ipv4.c +++ b/net/ipv4/netfilter/nft_dup_ipv4.c @@ -13,8 +13,8 @@ #include <net/netfilter/ipv4/nf_dup_ipv4.h> struct nft_dup_ipv4 { - enum nft_registers sreg_addr:8; - enum nft_registers sreg_dev:8; + u8 sreg_addr; + u8 sreg_dev; }; static void nft_dup_ipv4_eval(const struct nft_expr *expr, @@ -40,16 +40,16 @@ static int nft_dup_ipv4_init(const struct nft_ctx *ctx, if (tb[NFTA_DUP_SREG_ADDR] == NULL) return -EINVAL; - priv->sreg_addr = nft_parse_register(tb[NFTA_DUP_SREG_ADDR]); - err = nft_validate_register_load(priv->sreg_addr, sizeof(struct in_addr)); + err = nft_parse_register_load(tb[NFTA_DUP_SREG_ADDR], &priv->sreg_addr, + sizeof(struct in_addr)); if (err < 0) return err; - if (tb[NFTA_DUP_SREG_DEV] != NULL) { - priv->sreg_dev = nft_parse_register(tb[NFTA_DUP_SREG_DEV]); - return nft_validate_register_load(priv->sreg_dev, sizeof(int)); - } - return 0; + if (tb[NFTA_DUP_SREG_DEV]) + err = nft_parse_register_load(tb[NFTA_DUP_SREG_DEV], + &priv->sreg_dev, sizeof(int)); + + return err; } static int nft_dup_ipv4_dump(struct sk_buff *skb, const struct nft_expr *expr) @@ -75,6 +75,7 @@ static const struct nft_expr_ops nft_dup_ipv4_ops = { .eval = nft_dup_ipv4_eval, .init = nft_dup_ipv4_init, .dump = nft_dup_ipv4_dump, + .reduce = NFT_REDUCE_READONLY, }; static const struct nla_policy nft_dup_ipv4_policy[NFTA_DUP_MAX + 1] = { @@ -107,3 +108,4 @@ module_exit(nft_dup_ipv4_module_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Pablo Neira Ayuso <pablo@netfilter.org>"); MODULE_ALIAS_NFT_AF_EXPR(AF_INET, "dup"); +MODULE_DESCRIPTION("IPv4 nftables packet duplication support"); diff --git a/net/ipv4/netfilter/nft_fib_ipv4.c b/net/ipv4/netfilter/nft_fib_ipv4.c index ce294113dbcd..fc65d69f23e1 100644 --- a/net/ipv4/netfilter/nft_fib_ipv4.c +++ b/net/ipv4/netfilter/nft_fib_ipv4.c @@ -65,6 +65,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, struct flowi4 fl4 = { .flowi4_scope = RT_SCOPE_UNIVERSE, .flowi4_iif = LOOPBACK_IFINDEX, + .flowi4_uid = sock_net_uid(nft_net(pkt), NULL), }; const struct net_device *oif; const struct net_device *found; @@ -83,6 +84,9 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, else oif = NULL; + if (priv->flags & NFTA_FIB_F_IIF) + fl4.flowi4_l3mdev = l3mdev_master_ifindex_rcu(oif); + if (nft_hook(pkt) == NF_INET_PRE_ROUTING && nft_fib_is_loopback(pkt->skb, nft_in(pkt))) { nft_fib_store_result(dest, priv, nft_in(pkt)); @@ -112,6 +116,10 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs, fl4.daddr = iph->daddr; fl4.saddr = get_saddr(iph->saddr); } else { + if (nft_hook(pkt) == NF_INET_FORWARD && + priv->flags & NFTA_FIB_F_IIF) + fl4.flowi4_iif = nft_out(pkt)->ifindex; + fl4.daddr = iph->saddr; fl4.saddr = get_saddr(iph->daddr); } @@ -152,6 +160,7 @@ static const struct nft_expr_ops nft_fib4_type_ops = { .init = nft_fib_init, .dump = nft_fib_dump, .validate = nft_fib_validate, + .reduce = nft_fib_reduce, }; static const struct nft_expr_ops nft_fib4_ops = { @@ -161,6 +170,7 @@ static const struct nft_expr_ops nft_fib4_ops = { .init = nft_fib_init, .dump = nft_fib_dump, .validate = nft_fib_validate, + .reduce = nft_fib_reduce, }; static const struct nft_expr_ops * @@ -210,3 +220,4 @@ module_exit(nft_fib4_module_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Florian Westphal <fw@strlen.de>"); MODULE_ALIAS_NFT_AF_EXPR(2, "fib"); +MODULE_DESCRIPTION("nftables fib / ip route lookup support"); diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/ipv4/netfilter/nft_reject_ipv4.c index 7e6fd5cde50f..6cb213bb7256 100644 --- a/net/ipv4/netfilter/nft_reject_ipv4.c +++ b/net/ipv4/netfilter/nft_reject_ipv4.c @@ -27,7 +27,8 @@ static void nft_reject_ipv4_eval(const struct nft_expr *expr, nf_send_unreach(pkt->skb, priv->icmp_code, nft_hook(pkt)); break; case NFT_REJECT_TCP_RST: - nf_send_reset(nft_net(pkt), pkt->skb, nft_hook(pkt)); + nf_send_reset(nft_net(pkt), nft_sk(pkt), pkt->skb, + nft_hook(pkt)); break; default: break; @@ -44,6 +45,7 @@ static const struct nft_expr_ops nft_reject_ipv4_ops = { .init = nft_reject_init, .dump = nft_reject_dump, .validate = nft_reject_validate, + .reduce = NFT_REDUCE_READONLY, }; static struct nft_expr_type nft_reject_ipv4_type __read_mostly = { @@ -71,3 +73,4 @@ module_exit(nft_reject_ipv4_module_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); MODULE_ALIAS_NFT_AF_EXPR(AF_INET, "reject"); +MODULE_DESCRIPTION("IPv4 packet rejection for nftables"); diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c index d072c326dd64..d8ef05347fd9 100644 --- a/net/ipv4/nexthop.c +++ b/net/ipv4/nexthop.c @@ -8,6 +8,7 @@ #include <linux/nexthop.h> #include <linux/rtnetlink.h> #include <linux/slab.h> +#include <linux/vmalloc.h> #include <net/arp.h> #include <net/ipv6_stubs.h> #include <net/lwtunnel.h> @@ -16,13 +17,16 @@ #include <net/route.h> #include <net/sock.h> +#define NH_RES_DEFAULT_IDLE_TIMER (120 * HZ) +#define NH_RES_DEFAULT_UNBALANCED_TIMER 0 /* No forced rebalancing. */ + static void remove_nexthop(struct net *net, struct nexthop *nh, struct nl_info *nlinfo); #define NH_DEV_HASHBITS 8 #define NH_DEV_HASHSIZE (1U << NH_DEV_HASHBITS) -static const struct nla_policy rtm_nh_policy[NHA_MAX + 1] = { +static const struct nla_policy rtm_nh_policy_new[] = { [NHA_ID] = { .type = NLA_U32 }, [NHA_GROUP] = { .type = NLA_BINARY }, [NHA_GROUP_TYPE] = { .type = NLA_U16 }, @@ -31,10 +35,413 @@ static const struct nla_policy rtm_nh_policy[NHA_MAX + 1] = { [NHA_GATEWAY] = { .type = NLA_BINARY }, [NHA_ENCAP_TYPE] = { .type = NLA_U16 }, [NHA_ENCAP] = { .type = NLA_NESTED }, + [NHA_FDB] = { .type = NLA_FLAG }, + [NHA_RES_GROUP] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy rtm_nh_policy_get[] = { + [NHA_ID] = { .type = NLA_U32 }, +}; + +static const struct nla_policy rtm_nh_policy_dump[] = { + [NHA_OIF] = { .type = NLA_U32 }, [NHA_GROUPS] = { .type = NLA_FLAG }, [NHA_MASTER] = { .type = NLA_U32 }, + [NHA_FDB] = { .type = NLA_FLAG }, +}; + +static const struct nla_policy rtm_nh_res_policy_new[] = { + [NHA_RES_GROUP_BUCKETS] = { .type = NLA_U16 }, + [NHA_RES_GROUP_IDLE_TIMER] = { .type = NLA_U32 }, + [NHA_RES_GROUP_UNBALANCED_TIMER] = { .type = NLA_U32 }, }; +static const struct nla_policy rtm_nh_policy_dump_bucket[] = { + [NHA_ID] = { .type = NLA_U32 }, + [NHA_OIF] = { .type = NLA_U32 }, + [NHA_MASTER] = { .type = NLA_U32 }, + [NHA_RES_BUCKET] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy rtm_nh_res_bucket_policy_dump[] = { + [NHA_RES_BUCKET_NH_ID] = { .type = NLA_U32 }, +}; + +static const struct nla_policy rtm_nh_policy_get_bucket[] = { + [NHA_ID] = { .type = NLA_U32 }, + [NHA_RES_BUCKET] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy rtm_nh_res_bucket_policy_get[] = { + [NHA_RES_BUCKET_INDEX] = { .type = NLA_U16 }, +}; + +static bool nexthop_notifiers_is_empty(struct net *net) +{ + return !net->nexthop.notifier_chain.head; +} + +static void +__nh_notifier_single_info_init(struct nh_notifier_single_info *nh_info, + const struct nh_info *nhi) +{ + nh_info->dev = nhi->fib_nhc.nhc_dev; + nh_info->gw_family = nhi->fib_nhc.nhc_gw_family; + if (nh_info->gw_family == AF_INET) + nh_info->ipv4 = nhi->fib_nhc.nhc_gw.ipv4; + else if (nh_info->gw_family == AF_INET6) + nh_info->ipv6 = nhi->fib_nhc.nhc_gw.ipv6; + + nh_info->is_reject = nhi->reject_nh; + nh_info->is_fdb = nhi->fdb_nh; + nh_info->has_encap = !!nhi->fib_nhc.nhc_lwtstate; +} + +static int nh_notifier_single_info_init(struct nh_notifier_info *info, + const struct nexthop *nh) +{ + struct nh_info *nhi = rtnl_dereference(nh->nh_info); + + info->type = NH_NOTIFIER_INFO_TYPE_SINGLE; + info->nh = kzalloc(sizeof(*info->nh), GFP_KERNEL); + if (!info->nh) + return -ENOMEM; + + __nh_notifier_single_info_init(info->nh, nhi); + + return 0; +} + +static void nh_notifier_single_info_fini(struct nh_notifier_info *info) +{ + kfree(info->nh); +} + +static int nh_notifier_mpath_info_init(struct nh_notifier_info *info, + struct nh_group *nhg) +{ + u16 num_nh = nhg->num_nh; + int i; + + info->type = NH_NOTIFIER_INFO_TYPE_GRP; + info->nh_grp = kzalloc(struct_size(info->nh_grp, nh_entries, num_nh), + GFP_KERNEL); + if (!info->nh_grp) + return -ENOMEM; + + info->nh_grp->num_nh = num_nh; + info->nh_grp->is_fdb = nhg->fdb_nh; + + for (i = 0; i < num_nh; i++) { + struct nh_grp_entry *nhge = &nhg->nh_entries[i]; + struct nh_info *nhi; + + nhi = rtnl_dereference(nhge->nh->nh_info); + info->nh_grp->nh_entries[i].id = nhge->nh->id; + info->nh_grp->nh_entries[i].weight = nhge->weight; + __nh_notifier_single_info_init(&info->nh_grp->nh_entries[i].nh, + nhi); + } + + return 0; +} + +static int nh_notifier_res_table_info_init(struct nh_notifier_info *info, + struct nh_group *nhg) +{ + struct nh_res_table *res_table = rtnl_dereference(nhg->res_table); + u16 num_nh_buckets = res_table->num_nh_buckets; + unsigned long size; + u16 i; + + info->type = NH_NOTIFIER_INFO_TYPE_RES_TABLE; + size = struct_size(info->nh_res_table, nhs, num_nh_buckets); + info->nh_res_table = __vmalloc(size, GFP_KERNEL | __GFP_ZERO | + __GFP_NOWARN); + if (!info->nh_res_table) + return -ENOMEM; + + info->nh_res_table->num_nh_buckets = num_nh_buckets; + + for (i = 0; i < num_nh_buckets; i++) { + struct nh_res_bucket *bucket = &res_table->nh_buckets[i]; + struct nh_grp_entry *nhge; + struct nh_info *nhi; + + nhge = rtnl_dereference(bucket->nh_entry); + nhi = rtnl_dereference(nhge->nh->nh_info); + __nh_notifier_single_info_init(&info->nh_res_table->nhs[i], + nhi); + } + + return 0; +} + +static int nh_notifier_grp_info_init(struct nh_notifier_info *info, + const struct nexthop *nh) +{ + struct nh_group *nhg = rtnl_dereference(nh->nh_grp); + + if (nhg->hash_threshold) + return nh_notifier_mpath_info_init(info, nhg); + else if (nhg->resilient) + return nh_notifier_res_table_info_init(info, nhg); + return -EINVAL; +} + +static void nh_notifier_grp_info_fini(struct nh_notifier_info *info, + const struct nexthop *nh) +{ + struct nh_group *nhg = rtnl_dereference(nh->nh_grp); + + if (nhg->hash_threshold) + kfree(info->nh_grp); + else if (nhg->resilient) + vfree(info->nh_res_table); +} + +static int nh_notifier_info_init(struct nh_notifier_info *info, + const struct nexthop *nh) +{ + info->id = nh->id; + + if (nh->is_group) + return nh_notifier_grp_info_init(info, nh); + else + return nh_notifier_single_info_init(info, nh); +} + +static void nh_notifier_info_fini(struct nh_notifier_info *info, + const struct nexthop *nh) +{ + if (nh->is_group) + nh_notifier_grp_info_fini(info, nh); + else + nh_notifier_single_info_fini(info); +} + +static int call_nexthop_notifiers(struct net *net, + enum nexthop_event_type event_type, + struct nexthop *nh, + struct netlink_ext_ack *extack) +{ + struct nh_notifier_info info = { + .net = net, + .extack = extack, + }; + int err; + + ASSERT_RTNL(); + + if (nexthop_notifiers_is_empty(net)) + return 0; + + err = nh_notifier_info_init(&info, nh); + if (err) { + NL_SET_ERR_MSG(extack, "Failed to initialize nexthop notifier info"); + return err; + } + + err = blocking_notifier_call_chain(&net->nexthop.notifier_chain, + event_type, &info); + nh_notifier_info_fini(&info, nh); + + return notifier_to_errno(err); +} + +static int +nh_notifier_res_bucket_idle_timer_get(const struct nh_notifier_info *info, + bool force, unsigned int *p_idle_timer_ms) +{ + struct nh_res_table *res_table; + struct nh_group *nhg; + struct nexthop *nh; + int err = 0; + + /* When 'force' is false, nexthop bucket replacement is performed + * because the bucket was deemed to be idle. In this case, capable + * listeners can choose to perform an atomic replacement: The bucket is + * only replaced if it is inactive. However, if the idle timer interval + * is smaller than the interval in which a listener is querying + * buckets' activity from the device, then atomic replacement should + * not be tried. Pass the idle timer value to listeners, so that they + * could determine which type of replacement to perform. + */ + if (force) { + *p_idle_timer_ms = 0; + return 0; + } + + rcu_read_lock(); + + nh = nexthop_find_by_id(info->net, info->id); + if (!nh) { + err = -EINVAL; + goto out; + } + + nhg = rcu_dereference(nh->nh_grp); + res_table = rcu_dereference(nhg->res_table); + *p_idle_timer_ms = jiffies_to_msecs(res_table->idle_timer); + +out: + rcu_read_unlock(); + + return err; +} + +static int nh_notifier_res_bucket_info_init(struct nh_notifier_info *info, + u16 bucket_index, bool force, + struct nh_info *oldi, + struct nh_info *newi) +{ + unsigned int idle_timer_ms; + int err; + + err = nh_notifier_res_bucket_idle_timer_get(info, force, + &idle_timer_ms); + if (err) + return err; + + info->type = NH_NOTIFIER_INFO_TYPE_RES_BUCKET; + info->nh_res_bucket = kzalloc(sizeof(*info->nh_res_bucket), + GFP_KERNEL); + if (!info->nh_res_bucket) + return -ENOMEM; + + info->nh_res_bucket->bucket_index = bucket_index; + info->nh_res_bucket->idle_timer_ms = idle_timer_ms; + info->nh_res_bucket->force = force; + __nh_notifier_single_info_init(&info->nh_res_bucket->old_nh, oldi); + __nh_notifier_single_info_init(&info->nh_res_bucket->new_nh, newi); + return 0; +} + +static void nh_notifier_res_bucket_info_fini(struct nh_notifier_info *info) +{ + kfree(info->nh_res_bucket); +} + +static int __call_nexthop_res_bucket_notifiers(struct net *net, u32 nhg_id, + u16 bucket_index, bool force, + struct nh_info *oldi, + struct nh_info *newi, + struct netlink_ext_ack *extack) +{ + struct nh_notifier_info info = { + .net = net, + .extack = extack, + .id = nhg_id, + }; + int err; + + if (nexthop_notifiers_is_empty(net)) + return 0; + + err = nh_notifier_res_bucket_info_init(&info, bucket_index, force, + oldi, newi); + if (err) + return err; + + err = blocking_notifier_call_chain(&net->nexthop.notifier_chain, + NEXTHOP_EVENT_BUCKET_REPLACE, &info); + nh_notifier_res_bucket_info_fini(&info); + + return notifier_to_errno(err); +} + +/* There are three users of RES_TABLE, and NHs etc. referenced from there: + * + * 1) a collection of callbacks for NH maintenance. This operates under + * RTNL, + * 2) the delayed work that gradually balances the resilient table, + * 3) and nexthop_select_path(), operating under RCU. + * + * Both the delayed work and the RTNL block are writers, and need to + * maintain mutual exclusion. Since there are only two and well-known + * writers for each table, the RTNL code can make sure it has exclusive + * access thus: + * + * - Have the DW operate without locking; + * - synchronously cancel the DW; + * - do the writing; + * - if the write was not actually a delete, call upkeep, which schedules + * DW again if necessary. + * + * The functions that are always called from the RTNL context use + * rtnl_dereference(). The functions that can also be called from the DW do + * a raw dereference and rely on the above mutual exclusion scheme. + */ +#define nh_res_dereference(p) (rcu_dereference_raw(p)) + +static int call_nexthop_res_bucket_notifiers(struct net *net, u32 nhg_id, + u16 bucket_index, bool force, + struct nexthop *old_nh, + struct nexthop *new_nh, + struct netlink_ext_ack *extack) +{ + struct nh_info *oldi = nh_res_dereference(old_nh->nh_info); + struct nh_info *newi = nh_res_dereference(new_nh->nh_info); + + return __call_nexthop_res_bucket_notifiers(net, nhg_id, bucket_index, + force, oldi, newi, extack); +} + +static int call_nexthop_res_table_notifiers(struct net *net, struct nexthop *nh, + struct netlink_ext_ack *extack) +{ + struct nh_notifier_info info = { + .net = net, + .extack = extack, + }; + struct nh_group *nhg; + int err; + + ASSERT_RTNL(); + + if (nexthop_notifiers_is_empty(net)) + return 0; + + /* At this point, the nexthop buckets are still not populated. Only + * emit a notification with the logical nexthops, so that a listener + * could potentially veto it in case of unsupported configuration. + */ + nhg = rtnl_dereference(nh->nh_grp); + err = nh_notifier_mpath_info_init(&info, nhg); + if (err) { + NL_SET_ERR_MSG(extack, "Failed to initialize nexthop notifier info"); + return err; + } + + err = blocking_notifier_call_chain(&net->nexthop.notifier_chain, + NEXTHOP_EVENT_RES_TABLE_PRE_REPLACE, + &info); + kfree(info.nh_grp); + + return notifier_to_errno(err); +} + +static int call_nexthop_notifier(struct notifier_block *nb, struct net *net, + enum nexthop_event_type event_type, + struct nexthop *nh, + struct netlink_ext_ack *extack) +{ + struct nh_notifier_info info = { + .net = net, + .extack = extack, + }; + int err; + + err = nh_notifier_info_init(&info, nh); + if (err) + return err; + + err = nb->notifier_call(nb, event_type, &info); + nh_notifier_info_fini(&info, nh); + + return notifier_to_errno(err); +} + static unsigned int nh_dev_hashfn(unsigned int val) { unsigned int mask = NH_DEV_HASHSIZE - 1; @@ -57,15 +464,25 @@ static void nexthop_devhash_add(struct net *net, struct nh_info *nhi) hlist_add_head(&nhi->dev_hash, head); } -static void nexthop_free_mpath(struct nexthop *nh) +static void nexthop_free_group(struct nexthop *nh) { struct nh_group *nhg; int i; nhg = rcu_dereference_raw(nh->nh_grp); - for (i = 0; i < nhg->num_nh; ++i) - WARN_ON(nhg->nh_entries[i].nh); + for (i = 0; i < nhg->num_nh; ++i) { + struct nh_grp_entry *nhge = &nhg->nh_entries[i]; + + WARN_ON(!list_empty(&nhge->nh_list)); + nexthop_put(nhge->nh); + } + + WARN_ON(nhg->spare == nhg); + + if (nhg->resilient) + vfree(rcu_dereference_raw(nhg->res_table)); + kfree(nhg->spare); kfree(nhg); } @@ -90,7 +507,7 @@ void nexthop_free_rcu(struct rcu_head *head) struct nexthop *nh = container_of(head, struct nexthop, rcu); if (nh->is_group) - nexthop_free_mpath(nh); + nexthop_free_group(nh); else nexthop_free_single(nh); @@ -107,24 +524,46 @@ static struct nexthop *nexthop_alloc(void) INIT_LIST_HEAD(&nh->fi_list); INIT_LIST_HEAD(&nh->f6i_list); INIT_LIST_HEAD(&nh->grp_list); + INIT_LIST_HEAD(&nh->fdb_list); } return nh; } static struct nh_group *nexthop_grp_alloc(u16 num_nh) { - size_t sz = offsetof(struct nexthop, nh_grp) - + sizeof(struct nh_group) - + sizeof(struct nh_grp_entry) * num_nh; struct nh_group *nhg; - nhg = kzalloc(sz, GFP_KERNEL); + nhg = kzalloc(struct_size(nhg, nh_entries, num_nh), GFP_KERNEL); if (nhg) nhg->num_nh = num_nh; return nhg; } +static void nh_res_table_upkeep_dw(struct work_struct *work); + +static struct nh_res_table * +nexthop_res_table_alloc(struct net *net, u32 nhg_id, struct nh_config *cfg) +{ + const u16 num_nh_buckets = cfg->nh_grp_res_num_buckets; + struct nh_res_table *res_table; + unsigned long size; + + size = struct_size(res_table, nh_buckets, num_nh_buckets); + res_table = __vmalloc(size, GFP_KERNEL | __GFP_ZERO | __GFP_NOWARN); + if (!res_table) + return NULL; + + res_table->net = net; + res_table->nhg_id = nhg_id; + INIT_DELAYED_WORK(&res_table->upkeep_dw, &nh_res_table_upkeep_dw); + INIT_LIST_HEAD(&res_table->uw_nh_entries); + res_table->idle_timer = cfg->nh_grp_res_idle_timer; + res_table->unbalanced_timer = cfg->nh_grp_res_unbalanced_timer; + res_table->num_nh_buckets = num_nh_buckets; + return res_table; +} + static void nh_base_seq_inc(struct net *net) { while (++net->nexthop.seq == 0) @@ -173,6 +612,48 @@ static u32 nh_find_unused_id(struct net *net) return 0; } +static void nh_res_time_set_deadline(unsigned long next_time, + unsigned long *deadline) +{ + if (time_before(next_time, *deadline)) + *deadline = next_time; +} + +static clock_t nh_res_table_unbalanced_time(struct nh_res_table *res_table) +{ + if (list_empty(&res_table->uw_nh_entries)) + return 0; + return jiffies_delta_to_clock_t(jiffies - res_table->unbalanced_since); +} + +static int nla_put_nh_group_res(struct sk_buff *skb, struct nh_group *nhg) +{ + struct nh_res_table *res_table = rtnl_dereference(nhg->res_table); + struct nlattr *nest; + + nest = nla_nest_start(skb, NHA_RES_GROUP); + if (!nest) + return -EMSGSIZE; + + if (nla_put_u16(skb, NHA_RES_GROUP_BUCKETS, + res_table->num_nh_buckets) || + nla_put_u32(skb, NHA_RES_GROUP_IDLE_TIMER, + jiffies_to_clock_t(res_table->idle_timer)) || + nla_put_u32(skb, NHA_RES_GROUP_UNBALANCED_TIMER, + jiffies_to_clock_t(res_table->unbalanced_timer)) || + nla_put_u64_64bit(skb, NHA_RES_GROUP_UNBALANCED_TIME, + nh_res_table_unbalanced_time(res_table), + NHA_RES_GROUP_PAD)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + static int nla_put_nh_group(struct sk_buff *skb, struct nh_group *nhg) { struct nexthop_grp *p; @@ -181,8 +662,10 @@ static int nla_put_nh_group(struct sk_buff *skb, struct nh_group *nhg) u16 group_type = 0; int i; - if (nhg->mpath) + if (nhg->hash_threshold) group_type = NEXTHOP_GRP_TYPE_MPATH; + else if (nhg->resilient) + group_type = NEXTHOP_GRP_TYPE_RES; if (nla_put_u16(skb, NHA_GROUP_TYPE, group_type)) goto nla_put_failure; @@ -198,6 +681,9 @@ static int nla_put_nh_group(struct sk_buff *skb, struct nh_group *nhg) p += 1; } + if (nhg->resilient && nla_put_nh_group_res(skb, nhg)) + goto nla_put_failure; + return 0; nla_put_failure: @@ -230,6 +716,8 @@ static int nh_fill_node(struct sk_buff *skb, struct nexthop *nh, if (nh->is_group) { struct nh_group *nhg = rtnl_dereference(nh->nh_grp); + if (nhg->fdb_nh && nla_put_flag(skb, NHA_FDB)) + goto nla_put_failure; if (nla_put_nh_group(skb, nhg)) goto nla_put_failure; goto out; @@ -241,6 +729,9 @@ static int nh_fill_node(struct sk_buff *skb, struct nexthop *nh, if (nla_put_flag(skb, NHA_BLACKHOLE)) goto nla_put_failure; goto out; + } else if (nhi->fdb_nh) { + if (nla_put_flag(skb, NHA_FDB)) + goto nla_put_failure; } else { const struct net_device *dev; @@ -254,7 +745,7 @@ static int nh_fill_node(struct sk_buff *skb, struct nexthop *nh, case AF_INET: fib_nh = &nhi->fib_nh; if (fib_nh->fib_nh_gw_family && - nla_put_u32(skb, NHA_GATEWAY, fib_nh->fib_nh_gw4)) + nla_put_be32(skb, NHA_GATEWAY, fib_nh->fib_nh_gw4)) goto nla_put_failure; break; @@ -276,16 +767,30 @@ out: return 0; nla_put_failure: + nlmsg_cancel(skb, nlh); return -EMSGSIZE; } +static size_t nh_nlmsg_size_grp_res(struct nh_group *nhg) +{ + return nla_total_size(0) + /* NHA_RES_GROUP */ + nla_total_size(2) + /* NHA_RES_GROUP_BUCKETS */ + nla_total_size(4) + /* NHA_RES_GROUP_IDLE_TIMER */ + nla_total_size(4) + /* NHA_RES_GROUP_UNBALANCED_TIMER */ + nla_total_size_64bit(8);/* NHA_RES_GROUP_UNBALANCED_TIME */ +} + static size_t nh_nlmsg_size_grp(struct nexthop *nh) { struct nh_group *nhg = rtnl_dereference(nh->nh_grp); size_t sz = sizeof(struct nexthop_grp) * nhg->num_nh; + size_t tot = nla_total_size(sz) + + nla_total_size(2); /* NHA_GROUP_TYPE */ - return nla_total_size(sz) + - nla_total_size(2); /* NHA_GROUP_TYPE */ + if (nhg->resilient) + tot += nh_nlmsg_size_grp_res(nhg); + + return tot; } static size_t nh_nlmsg_size_single(struct nexthop *nh) @@ -360,20 +865,145 @@ errout: rtnl_set_sk_err(info->nl_net, RTNLGRP_NEXTHOP, err); } +static unsigned long nh_res_bucket_used_time(const struct nh_res_bucket *bucket) +{ + return (unsigned long)atomic_long_read(&bucket->used_time); +} + +static unsigned long +nh_res_bucket_idle_point(const struct nh_res_table *res_table, + const struct nh_res_bucket *bucket, + unsigned long now) +{ + unsigned long time = nh_res_bucket_used_time(bucket); + + /* Bucket was not used since it was migrated. The idle time is now. */ + if (time == bucket->migrated_time) + return now; + + return time + res_table->idle_timer; +} + +static unsigned long +nh_res_table_unb_point(const struct nh_res_table *res_table) +{ + return res_table->unbalanced_since + res_table->unbalanced_timer; +} + +static void nh_res_bucket_set_idle(const struct nh_res_table *res_table, + struct nh_res_bucket *bucket) +{ + unsigned long now = jiffies; + + atomic_long_set(&bucket->used_time, (long)now); + bucket->migrated_time = now; +} + +static void nh_res_bucket_set_busy(struct nh_res_bucket *bucket) +{ + atomic_long_set(&bucket->used_time, (long)jiffies); +} + +static clock_t nh_res_bucket_idle_time(const struct nh_res_bucket *bucket) +{ + unsigned long used_time = nh_res_bucket_used_time(bucket); + + return jiffies_delta_to_clock_t(jiffies - used_time); +} + +static int nh_fill_res_bucket(struct sk_buff *skb, struct nexthop *nh, + struct nh_res_bucket *bucket, u16 bucket_index, + int event, u32 portid, u32 seq, + unsigned int nlflags, + struct netlink_ext_ack *extack) +{ + struct nh_grp_entry *nhge = nh_res_dereference(bucket->nh_entry); + struct nlmsghdr *nlh; + struct nlattr *nest; + struct nhmsg *nhm; + + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*nhm), nlflags); + if (!nlh) + return -EMSGSIZE; + + nhm = nlmsg_data(nlh); + nhm->nh_family = AF_UNSPEC; + nhm->nh_flags = bucket->nh_flags; + nhm->nh_protocol = nh->protocol; + nhm->nh_scope = 0; + nhm->resvd = 0; + + if (nla_put_u32(skb, NHA_ID, nh->id)) + goto nla_put_failure; + + nest = nla_nest_start(skb, NHA_RES_BUCKET); + if (!nest) + goto nla_put_failure; + + if (nla_put_u16(skb, NHA_RES_BUCKET_INDEX, bucket_index) || + nla_put_u32(skb, NHA_RES_BUCKET_NH_ID, nhge->nh->id) || + nla_put_u64_64bit(skb, NHA_RES_BUCKET_IDLE_TIME, + nh_res_bucket_idle_time(bucket), + NHA_RES_BUCKET_PAD)) + goto nla_put_failure_nest; + + nla_nest_end(skb, nest); + nlmsg_end(skb, nlh); + return 0; + +nla_put_failure_nest: + nla_nest_cancel(skb, nest); +nla_put_failure: + nlmsg_cancel(skb, nlh); + return -EMSGSIZE; +} + +static void nexthop_bucket_notify(struct nh_res_table *res_table, + u16 bucket_index) +{ + struct nh_res_bucket *bucket = &res_table->nh_buckets[bucket_index]; + struct nh_grp_entry *nhge = nh_res_dereference(bucket->nh_entry); + struct nexthop *nh = nhge->nh_parent; + struct sk_buff *skb; + int err = -ENOBUFS; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + goto errout; + + err = nh_fill_res_bucket(skb, nh, bucket, bucket_index, + RTM_NEWNEXTHOPBUCKET, 0, 0, NLM_F_REPLACE, + NULL); + if (err < 0) { + kfree_skb(skb); + goto errout; + } + + rtnl_notify(skb, nh->net, 0, RTNLGRP_NEXTHOP, NULL, GFP_KERNEL); + return; +errout: + if (err < 0) + rtnl_set_sk_err(nh->net, RTNLGRP_NEXTHOP, err); +} + static bool valid_group_nh(struct nexthop *nh, unsigned int npaths, - struct netlink_ext_ack *extack) + bool *is_fdb, struct netlink_ext_ack *extack) { if (nh->is_group) { struct nh_group *nhg = rtnl_dereference(nh->nh_grp); - /* nested multipath (group within a group) is not - * supported - */ - if (nhg->mpath) { + /* Nesting groups within groups is not supported. */ + if (nhg->hash_threshold) { NL_SET_ERR_MSG(extack, - "Multipath group can not be a nexthop within a group"); + "Hash-threshold group can not be a nexthop within a group"); return false; } + if (nhg->resilient) { + NL_SET_ERR_MSG(extack, + "Resilient group can not be a nexthop within a group"); + return false; + } + *is_fdb = nhg->fdb_nh; } else { struct nh_info *nhi = rtnl_dereference(nh->nh_info); @@ -382,19 +1012,45 @@ static bool valid_group_nh(struct nexthop *nh, unsigned int npaths, "Blackhole nexthop can not be used in a group with more than 1 path"); return false; } + *is_fdb = nhi->fdb_nh; } return true; } -static int nh_check_attr_group(struct net *net, struct nlattr *tb[], - struct netlink_ext_ack *extack) +static int nh_check_attr_fdb_group(struct nexthop *nh, u8 *nh_family, + struct netlink_ext_ack *extack) +{ + struct nh_info *nhi; + + nhi = rtnl_dereference(nh->nh_info); + + if (!nhi->fdb_nh) { + NL_SET_ERR_MSG(extack, "FDB nexthop group can only have fdb nexthops"); + return -EINVAL; + } + + if (*nh_family == AF_UNSPEC) { + *nh_family = nhi->family; + } else if (*nh_family != nhi->family) { + NL_SET_ERR_MSG(extack, "FDB nexthop group cannot have mixed family nexthops"); + return -EINVAL; + } + + return 0; +} + +static int nh_check_attr_group(struct net *net, + struct nlattr *tb[], size_t tb_size, + u16 nh_grp_type, struct netlink_ext_ack *extack) { unsigned int len = nla_len(tb[NHA_GROUP]); + u8 nh_family = AF_UNSPEC; struct nexthop_grp *nhg; unsigned int i, j; + u8 nhg_fdb = 0; - if (len & (sizeof(struct nexthop_grp) - 1)) { + if (!len || len & (sizeof(struct nexthop_grp) - 1)) { NL_SET_ERR_MSG(extack, "Invalid length for nexthop group attribute"); return -EINVAL; @@ -421,22 +1077,40 @@ static int nh_check_attr_group(struct net *net, struct nlattr *tb[], } } + if (tb[NHA_FDB]) + nhg_fdb = 1; nhg = nla_data(tb[NHA_GROUP]); for (i = 0; i < len; ++i) { struct nexthop *nh; + bool is_fdb_nh; nh = nexthop_find_by_id(net, nhg[i].id); if (!nh) { NL_SET_ERR_MSG(extack, "Invalid nexthop id"); return -EINVAL; } - if (!valid_group_nh(nh, len, extack)) + if (!valid_group_nh(nh, len, &is_fdb_nh, extack)) + return -EINVAL; + + if (nhg_fdb && nh_check_attr_fdb_group(nh, &nh_family, extack)) + return -EINVAL; + + if (!nhg_fdb && is_fdb_nh) { + NL_SET_ERR_MSG(extack, "Non FDB nexthop group cannot have fdb nexthops"); return -EINVAL; + } } - for (i = NHA_GROUP + 1; i < __NHA_MAX; ++i) { + for (i = NHA_GROUP_TYPE + 1; i < tb_size; ++i) { if (!tb[i]) continue; - + switch (i) { + case NHA_FDB: + continue; + case NHA_RES_GROUP: + if (nh_grp_type == NEXTHOP_GRP_TYPE_RES) + continue; + break; + } NL_SET_ERR_MSG(extack, "No other attributes can be set in nexthop groups"); return -EINVAL; @@ -478,27 +1152,25 @@ static bool ipv4_good_nh(const struct fib_nh *nh) return !!(state & NUD_VALID); } -struct nexthop *nexthop_select_path(struct nexthop *nh, int hash) +static struct nexthop *nexthop_select_path_hthr(struct nh_group *nhg, int hash) { struct nexthop *rc = NULL; - struct nh_group *nhg; int i; - if (!nh->is_group) - return nh; - - nhg = rcu_dereference(nh->nh_grp); for (i = 0; i < nhg->num_nh; ++i) { struct nh_grp_entry *nhge = &nhg->nh_entries[i]; struct nh_info *nhi; - if (hash > atomic_read(&nhge->upper_bound)) + if (hash > atomic_read(&nhge->hthr.upper_bound)) continue; + nhi = rcu_dereference(nhge->nh->nh_info); + if (nhi->fdb_nh) + return nhge->nh; + /* nexthops always check if it is good and does * not rely on a sysctl for this behavior */ - nhi = rcu_dereference(nhge->nh->nh_info); switch (nhi->family) { case AF_INET: if (ipv4_good_nh(&nhi->fib_nh)) @@ -516,6 +1188,39 @@ struct nexthop *nexthop_select_path(struct nexthop *nh, int hash) return rc; } + +static struct nexthop *nexthop_select_path_res(struct nh_group *nhg, int hash) +{ + struct nh_res_table *res_table = rcu_dereference(nhg->res_table); + u16 bucket_index = hash % res_table->num_nh_buckets; + struct nh_res_bucket *bucket; + struct nh_grp_entry *nhge; + + /* nexthop_select_path() is expected to return a non-NULL value, so + * skip protocol validation and just hand out whatever there is. + */ + bucket = &res_table->nh_buckets[bucket_index]; + nh_res_bucket_set_busy(bucket); + nhge = rcu_dereference(bucket->nh_entry); + return nhge->nh; +} + +struct nexthop *nexthop_select_path(struct nexthop *nh, int hash) +{ + struct nh_group *nhg; + + if (!nh->is_group) + return nh; + + nhg = rcu_dereference(nh->nh_grp); + if (nhg->hash_threshold) + return nexthop_select_path_hthr(nhg, hash); + else if (nhg->resilient) + return nexthop_select_path_res(nhg, hash); + + /* Unreachable. */ + return NULL; +} EXPORT_SYMBOL_GPL(nexthop_select_path); int nexthop_for_each_fib6_nh(struct nexthop *nh, @@ -563,6 +1268,7 @@ int fib6_check_nexthop(struct nexthop *nh, struct fib6_config *cfg, struct netlink_ext_ack *extack) { struct nh_info *nhi; + bool is_fdb_nh; /* fib6_src is unique to a fib6_info and limits the ability to cache * routes in fib6_nh within a nexthop that is potentially shared @@ -579,10 +1285,17 @@ int fib6_check_nexthop(struct nexthop *nh, struct fib6_config *cfg, nhg = rtnl_dereference(nh->nh_grp); if (nhg->has_v4) goto no_v4_nh; + is_fdb_nh = nhg->fdb_nh; } else { nhi = rtnl_dereference(nh->nh_info); if (nhi->family == AF_INET) goto no_v4_nh; + is_fdb_nh = nhi->fdb_nh; + } + + if (is_fdb_nh) { + NL_SET_ERR_MSG(extack, "Route cannot point to a fdb nexthop"); + return -EINVAL; } return 0; @@ -611,12 +1324,9 @@ static int fib6_check_nh_list(struct nexthop *old, struct nexthop *new, return fib6_check_nexthop(new, NULL, extack); } -static int nexthop_check_scope(struct nexthop *nh, u8 scope, +static int nexthop_check_scope(struct nh_info *nhi, u8 scope, struct netlink_ext_ack *extack) { - struct nh_info *nhi; - - nhi = rtnl_dereference(nh->nh_info); if (scope == RT_SCOPE_HOST && nhi->fib_nhc.nhc_gw_family) { NL_SET_ERR_MSG(extack, "Route with host scope can not have a gateway"); @@ -638,23 +1348,38 @@ static int nexthop_check_scope(struct nexthop *nh, u8 scope, int fib_check_nexthop(struct nexthop *nh, u8 scope, struct netlink_ext_ack *extack) { + struct nh_info *nhi; int err = 0; if (nh->is_group) { struct nh_group *nhg; + nhg = rtnl_dereference(nh->nh_grp); + if (nhg->fdb_nh) { + NL_SET_ERR_MSG(extack, "Route cannot point to a fdb nexthop"); + err = -EINVAL; + goto out; + } + if (scope == RT_SCOPE_HOST) { NL_SET_ERR_MSG(extack, "Route with host scope can not have multiple nexthops"); err = -EINVAL; goto out; } - nhg = rtnl_dereference(nh->nh_grp); /* all nexthops in a group have the same scope */ - err = nexthop_check_scope(nhg->nh_entries[0].nh, scope, extack); + nhi = rtnl_dereference(nhg->nh_entries[0].nh->nh_info); + err = nexthop_check_scope(nhi, scope, extack); } else { - err = nexthop_check_scope(nh, scope, extack); + nhi = rtnl_dereference(nh->nh_info); + if (nhi->fdb_nh) { + NL_SET_ERR_MSG(extack, "Route cannot point to a fdb nexthop"); + err = -EINVAL; + goto out; + } + err = nexthop_check_scope(nhi, scope, extack); } + out: return err; } @@ -674,7 +1399,319 @@ static int fib_check_nh_list(struct nexthop *old, struct nexthop *new, return 0; } -static void nh_group_rebalance(struct nh_group *nhg) +static bool nh_res_nhge_is_balanced(const struct nh_grp_entry *nhge) +{ + return nhge->res.count_buckets == nhge->res.wants_buckets; +} + +static bool nh_res_nhge_is_ow(const struct nh_grp_entry *nhge) +{ + return nhge->res.count_buckets > nhge->res.wants_buckets; +} + +static bool nh_res_nhge_is_uw(const struct nh_grp_entry *nhge) +{ + return nhge->res.count_buckets < nhge->res.wants_buckets; +} + +static bool nh_res_table_is_balanced(const struct nh_res_table *res_table) +{ + return list_empty(&res_table->uw_nh_entries); +} + +static void nh_res_bucket_unset_nh(struct nh_res_bucket *bucket) +{ + struct nh_grp_entry *nhge; + + if (bucket->occupied) { + nhge = nh_res_dereference(bucket->nh_entry); + nhge->res.count_buckets--; + bucket->occupied = false; + } +} + +static void nh_res_bucket_set_nh(struct nh_res_bucket *bucket, + struct nh_grp_entry *nhge) +{ + nh_res_bucket_unset_nh(bucket); + + bucket->occupied = true; + rcu_assign_pointer(bucket->nh_entry, nhge); + nhge->res.count_buckets++; +} + +static bool nh_res_bucket_should_migrate(struct nh_res_table *res_table, + struct nh_res_bucket *bucket, + unsigned long *deadline, bool *force) +{ + unsigned long now = jiffies; + struct nh_grp_entry *nhge; + unsigned long idle_point; + + if (!bucket->occupied) { + /* The bucket is not occupied, its NHGE pointer is either + * NULL or obsolete. We _have to_ migrate: set force. + */ + *force = true; + return true; + } + + nhge = nh_res_dereference(bucket->nh_entry); + + /* If the bucket is populated by an underweight or balanced + * nexthop, do not migrate. + */ + if (!nh_res_nhge_is_ow(nhge)) + return false; + + /* At this point we know that the bucket is populated with an + * overweight nexthop. It needs to be migrated to a new nexthop if + * the idle timer of unbalanced timer expired. + */ + + idle_point = nh_res_bucket_idle_point(res_table, bucket, now); + if (time_after_eq(now, idle_point)) { + /* The bucket is idle. We _can_ migrate: unset force. */ + *force = false; + return true; + } + + /* Unbalanced timer of 0 means "never force". */ + if (res_table->unbalanced_timer) { + unsigned long unb_point; + + unb_point = nh_res_table_unb_point(res_table); + if (time_after(now, unb_point)) { + /* The bucket is not idle, but the unbalanced timer + * expired. We _can_ migrate, but set force anyway, + * so that drivers know to ignore activity reports + * from the HW. + */ + *force = true; + return true; + } + + nh_res_time_set_deadline(unb_point, deadline); + } + + nh_res_time_set_deadline(idle_point, deadline); + return false; +} + +static bool nh_res_bucket_migrate(struct nh_res_table *res_table, + u16 bucket_index, bool notify, + bool notify_nl, bool force) +{ + struct nh_res_bucket *bucket = &res_table->nh_buckets[bucket_index]; + struct nh_grp_entry *new_nhge; + struct netlink_ext_ack extack; + int err; + + new_nhge = list_first_entry_or_null(&res_table->uw_nh_entries, + struct nh_grp_entry, + res.uw_nh_entry); + if (WARN_ON_ONCE(!new_nhge)) + /* If this function is called, "bucket" is either not + * occupied, or it belongs to a next hop that is + * overweight. In either case, there ought to be a + * corresponding underweight next hop. + */ + return false; + + if (notify) { + struct nh_grp_entry *old_nhge; + + old_nhge = nh_res_dereference(bucket->nh_entry); + err = call_nexthop_res_bucket_notifiers(res_table->net, + res_table->nhg_id, + bucket_index, force, + old_nhge->nh, + new_nhge->nh, &extack); + if (err) { + pr_err_ratelimited("%s\n", extack._msg); + if (!force) + return false; + /* It is not possible to veto a forced replacement, so + * just clear the hardware flags from the nexthop + * bucket to indicate to user space that this bucket is + * not correctly populated in hardware. + */ + bucket->nh_flags &= ~(RTNH_F_OFFLOAD | RTNH_F_TRAP); + } + } + + nh_res_bucket_set_nh(bucket, new_nhge); + nh_res_bucket_set_idle(res_table, bucket); + + if (notify_nl) + nexthop_bucket_notify(res_table, bucket_index); + + if (nh_res_nhge_is_balanced(new_nhge)) + list_del(&new_nhge->res.uw_nh_entry); + return true; +} + +#define NH_RES_UPKEEP_DW_MINIMUM_INTERVAL (HZ / 2) + +static void nh_res_table_upkeep(struct nh_res_table *res_table, + bool notify, bool notify_nl) +{ + unsigned long now = jiffies; + unsigned long deadline; + u16 i; + + /* Deadline is the next time that upkeep should be run. It is the + * earliest time at which one of the buckets might be migrated. + * Start at the most pessimistic estimate: either unbalanced_timer + * from now, or if there is none, idle_timer from now. For each + * encountered time point, call nh_res_time_set_deadline() to + * refine the estimate. + */ + if (res_table->unbalanced_timer) + deadline = now + res_table->unbalanced_timer; + else + deadline = now + res_table->idle_timer; + + for (i = 0; i < res_table->num_nh_buckets; i++) { + struct nh_res_bucket *bucket = &res_table->nh_buckets[i]; + bool force; + + if (nh_res_bucket_should_migrate(res_table, bucket, + &deadline, &force)) { + if (!nh_res_bucket_migrate(res_table, i, notify, + notify_nl, force)) { + unsigned long idle_point; + + /* A driver can override the migration + * decision if the HW reports that the + * bucket is actually not idle. Therefore + * remark the bucket as busy again and + * update the deadline. + */ + nh_res_bucket_set_busy(bucket); + idle_point = nh_res_bucket_idle_point(res_table, + bucket, + now); + nh_res_time_set_deadline(idle_point, &deadline); + } + } + } + + /* If the group is still unbalanced, schedule the next upkeep to + * either the deadline computed above, or the minimum deadline, + * whichever comes later. + */ + if (!nh_res_table_is_balanced(res_table)) { + unsigned long now = jiffies; + unsigned long min_deadline; + + min_deadline = now + NH_RES_UPKEEP_DW_MINIMUM_INTERVAL; + if (time_before(deadline, min_deadline)) + deadline = min_deadline; + + queue_delayed_work(system_power_efficient_wq, + &res_table->upkeep_dw, deadline - now); + } +} + +static void nh_res_table_upkeep_dw(struct work_struct *work) +{ + struct delayed_work *dw = to_delayed_work(work); + struct nh_res_table *res_table; + + res_table = container_of(dw, struct nh_res_table, upkeep_dw); + nh_res_table_upkeep(res_table, true, true); +} + +static void nh_res_table_cancel_upkeep(struct nh_res_table *res_table) +{ + cancel_delayed_work_sync(&res_table->upkeep_dw); +} + +static void nh_res_group_rebalance(struct nh_group *nhg, + struct nh_res_table *res_table) +{ + int prev_upper_bound = 0; + int total = 0; + int w = 0; + int i; + + INIT_LIST_HEAD(&res_table->uw_nh_entries); + + for (i = 0; i < nhg->num_nh; ++i) + total += nhg->nh_entries[i].weight; + + for (i = 0; i < nhg->num_nh; ++i) { + struct nh_grp_entry *nhge = &nhg->nh_entries[i]; + int upper_bound; + + w += nhge->weight; + upper_bound = DIV_ROUND_CLOSEST(res_table->num_nh_buckets * w, + total); + nhge->res.wants_buckets = upper_bound - prev_upper_bound; + prev_upper_bound = upper_bound; + + if (nh_res_nhge_is_uw(nhge)) { + if (list_empty(&res_table->uw_nh_entries)) + res_table->unbalanced_since = jiffies; + list_add(&nhge->res.uw_nh_entry, + &res_table->uw_nh_entries); + } + } +} + +/* Migrate buckets in res_table so that they reference NHGE's from NHG with + * the right NH ID. Set those buckets that do not have a corresponding NHGE + * entry in NHG as not occupied. + */ +static void nh_res_table_migrate_buckets(struct nh_res_table *res_table, + struct nh_group *nhg) +{ + u16 i; + + for (i = 0; i < res_table->num_nh_buckets; i++) { + struct nh_res_bucket *bucket = &res_table->nh_buckets[i]; + u32 id = rtnl_dereference(bucket->nh_entry)->nh->id; + bool found = false; + int j; + + for (j = 0; j < nhg->num_nh; j++) { + struct nh_grp_entry *nhge = &nhg->nh_entries[j]; + + if (nhge->nh->id == id) { + nh_res_bucket_set_nh(bucket, nhge); + found = true; + break; + } + } + + if (!found) + nh_res_bucket_unset_nh(bucket); + } +} + +static void replace_nexthop_grp_res(struct nh_group *oldg, + struct nh_group *newg) +{ + /* For NH group replacement, the new NHG might only have a stub + * hash table with 0 buckets, because the number of buckets was not + * specified. For NH removal, oldg and newg both reference the same + * res_table. So in any case, in the following, we want to work + * with oldg->res_table. + */ + struct nh_res_table *old_res_table = rtnl_dereference(oldg->res_table); + unsigned long prev_unbalanced_since = old_res_table->unbalanced_since; + bool prev_has_uw = !list_empty(&old_res_table->uw_nh_entries); + + nh_res_table_cancel_upkeep(old_res_table); + nh_res_table_migrate_buckets(old_res_table, newg); + nh_res_group_rebalance(newg, old_res_table); + if (prev_has_uw && !list_empty(&old_res_table->uw_nh_entries)) + old_res_table->unbalanced_since = prev_unbalanced_since; + nh_res_table_upkeep(old_res_table, true, false); +} + +static void nh_hthr_group_rebalance(struct nh_group *nhg) { int total = 0; int w = 0; @@ -689,45 +1726,84 @@ static void nh_group_rebalance(struct nh_group *nhg) w += nhge->weight; upper_bound = DIV_ROUND_CLOSEST_ULL((u64)w << 31, total) - 1; - atomic_set(&nhge->upper_bound, upper_bound); + atomic_set(&nhge->hthr.upper_bound, upper_bound); } } -static void remove_nh_grp_entry(struct nh_grp_entry *nhge, - struct nh_group *nhg, +static void remove_nh_grp_entry(struct net *net, struct nh_grp_entry *nhge, struct nl_info *nlinfo) { + struct nh_grp_entry *nhges, *new_nhges; + struct nexthop *nhp = nhge->nh_parent; + struct netlink_ext_ack extack; struct nexthop *nh = nhge->nh; - struct nh_grp_entry *nhges; - bool found = false; - int i; + struct nh_group *nhg, *newg; + int i, j, err; WARN_ON(!nh); + nhg = rtnl_dereference(nhp->nh_grp); + newg = nhg->spare; + + /* last entry, keep it visible and remove the parent */ + if (nhg->num_nh == 1) { + remove_nexthop(net, nhp, nlinfo); + return; + } + + newg->has_v4 = false; + newg->is_multipath = nhg->is_multipath; + newg->hash_threshold = nhg->hash_threshold; + newg->resilient = nhg->resilient; + newg->fdb_nh = nhg->fdb_nh; + newg->num_nh = nhg->num_nh; + + /* copy old entries to new except the one getting removed */ nhges = nhg->nh_entries; - for (i = 0; i < nhg->num_nh; ++i) { - if (found) { - nhges[i-1].nh = nhges[i].nh; - nhges[i-1].weight = nhges[i].weight; - list_del(&nhges[i].nh_list); - list_add(&nhges[i-1].nh_list, &nhges[i-1].nh->grp_list); - } else if (nhg->nh_entries[i].nh == nh) { - found = true; + new_nhges = newg->nh_entries; + for (i = 0, j = 0; i < nhg->num_nh; ++i) { + struct nh_info *nhi; + + /* current nexthop getting removed */ + if (nhg->nh_entries[i].nh == nh) { + newg->num_nh--; + continue; } + + nhi = rtnl_dereference(nhges[i].nh->nh_info); + if (nhi->family == AF_INET) + newg->has_v4 = true; + + list_del(&nhges[i].nh_list); + new_nhges[j].nh_parent = nhges[i].nh_parent; + new_nhges[j].nh = nhges[i].nh; + new_nhges[j].weight = nhges[i].weight; + list_add(&new_nhges[j].nh_list, &new_nhges[j].nh->grp_list); + j++; } - if (WARN_ON(!found)) - return; + if (newg->hash_threshold) + nh_hthr_group_rebalance(newg); + else if (newg->resilient) + replace_nexthop_grp_res(nhg, newg); - nhg->num_nh--; - nhg->nh_entries[nhg->num_nh].nh = NULL; + rcu_assign_pointer(nhp->nh_grp, newg); - nh_group_rebalance(nhg); + list_del(&nhge->nh_list); + nexthop_put(nhge->nh); - nexthop_put(nh); + /* Removal of a NH from a resilient group is notified through + * bucket notifications. + */ + if (newg->hash_threshold) { + err = call_nexthop_notifiers(net, NEXTHOP_EVENT_REPLACE, nhp, + &extack); + if (err) + pr_err("%s\n", extack._msg); + } if (nlinfo) - nexthop_notify(RTM_NEWNEXTHOP, nhge->nh_parent, nlinfo); + nexthop_notify(RTM_NEWNEXTHOP, nhp, nlinfo); } static void remove_nexthop_from_groups(struct net *net, struct nexthop *nh, @@ -735,22 +1811,17 @@ static void remove_nexthop_from_groups(struct net *net, struct nexthop *nh, { struct nh_grp_entry *nhge, *tmp; - list_for_each_entry_safe(nhge, tmp, &nh->grp_list, nh_list) { - struct nh_group *nhg; - - list_del(&nhge->nh_list); - nhg = rtnl_dereference(nhge->nh_parent->nh_grp); - remove_nh_grp_entry(nhge, nhg, nlinfo); + list_for_each_entry_safe(nhge, tmp, &nh->grp_list, nh_list) + remove_nh_grp_entry(net, nhge, nlinfo); - /* if this group has no more entries then remove it */ - if (!nhg->num_nh) - remove_nexthop(net, nhge->nh_parent, nlinfo); - } + /* make sure all see the newly published array before releasing rtnl */ + synchronize_net(); } static void remove_nexthop_group(struct nexthop *nh, struct nl_info *nlinfo) { struct nh_group *nhg = rcu_dereference_rtnl(nh->nh_grp); + struct nh_res_table *res_table; int i, num_nh = nhg->num_nh; for (i = 0; i < num_nh; ++i) { @@ -759,10 +1830,12 @@ static void remove_nexthop_group(struct nexthop *nh, struct nl_info *nlinfo) if (WARN_ON(!nhge->nh)) continue; - list_del(&nhge->nh_list); - nexthop_put(nhge->nh); - nhge->nh = NULL; - nhg->num_nh--; + list_del_init(&nhge->nh_list); + } + + if (nhg->resilient) { + res_table = rtnl_dereference(nhg->res_table); + nh_res_table_cancel_upkeep(res_table); } } @@ -784,7 +1857,8 @@ static void __remove_nexthop_fib(struct net *net, struct nexthop *nh) list_for_each_entry_safe(f6i, tmp, &nh->f6i_list, nh_list) { /* __ip6_del_rt does a release, so do a hold here */ fib6_info_hold(f6i); - ipv6_stub->ip6_del_rt(net, f6i); + ipv6_stub->ip6_del_rt(net, f6i, + !READ_ONCE(net->ipv4.sysctl_nexthop_compat_mode)); } } @@ -809,6 +1883,8 @@ static void __remove_nexthop(struct net *net, struct nexthop *nh, static void remove_nexthop(struct net *net, struct nexthop *nh, struct nl_info *nlinfo) { + call_nexthop_notifiers(net, NEXTHOP_EVENT_DEL, nh, NULL); + /* remove from the tree */ rb_erase(&nh->rb_node, &net->nexthop.rb_root); @@ -824,23 +1900,44 @@ static void remove_nexthop(struct net *net, struct nexthop *nh, /* if any FIB entries reference this nexthop, any dst entries * need to be regenerated */ -static void nh_rt_cache_flush(struct net *net, struct nexthop *nh) +static void nh_rt_cache_flush(struct net *net, struct nexthop *nh, + struct nexthop *replaced_nh) { struct fib6_info *f6i; + struct nh_group *nhg; + int i; if (!list_empty(&nh->fi_list)) rt_cache_flush(net); list_for_each_entry(f6i, &nh->f6i_list, nh_list) ipv6_stub->fib6_update_sernum(net, f6i); + + /* if an IPv6 group was replaced, we have to release all old + * dsts to make sure all refcounts are released + */ + if (!replaced_nh->is_group) + return; + + nhg = rtnl_dereference(replaced_nh->nh_grp); + for (i = 0; i < nhg->num_nh; i++) { + struct nh_grp_entry *nhge = &nhg->nh_entries[i]; + struct nh_info *nhi = rtnl_dereference(nhge->nh->nh_info); + + if (nhi->family == AF_INET6) + ipv6_stub->fib6_nh_release_dsts(&nhi->fib6_nh); + } } static int replace_nexthop_grp(struct net *net, struct nexthop *old, - struct nexthop *new, + struct nexthop *new, const struct nh_config *cfg, struct netlink_ext_ack *extack) { + struct nh_res_table *tmp_table = NULL; + struct nh_res_table *new_res_table; + struct nh_res_table *old_res_table; struct nh_group *oldg, *newg; - int i; + int i, err; if (!new->is_group) { NL_SET_ERR_MSG(extack, "Can not replace a nexthop group with a nexthop."); @@ -850,12 +1947,67 @@ static int replace_nexthop_grp(struct net *net, struct nexthop *old, oldg = rtnl_dereference(old->nh_grp); newg = rtnl_dereference(new->nh_grp); + if (newg->hash_threshold != oldg->hash_threshold) { + NL_SET_ERR_MSG(extack, "Can not replace a nexthop group with one of a different type."); + return -EINVAL; + } + + if (newg->hash_threshold) { + err = call_nexthop_notifiers(net, NEXTHOP_EVENT_REPLACE, new, + extack); + if (err) + return err; + } else if (newg->resilient) { + new_res_table = rtnl_dereference(newg->res_table); + old_res_table = rtnl_dereference(oldg->res_table); + + /* Accept if num_nh_buckets was not given, but if it was + * given, demand that the value be correct. + */ + if (cfg->nh_grp_res_has_num_buckets && + cfg->nh_grp_res_num_buckets != + old_res_table->num_nh_buckets) { + NL_SET_ERR_MSG(extack, "Can not change number of buckets of a resilient nexthop group."); + return -EINVAL; + } + + /* Emit a pre-replace notification so that listeners could veto + * a potentially unsupported configuration. Otherwise, + * individual bucket replacement notifications would need to be + * vetoed, which is something that should only happen if the + * bucket is currently active. + */ + err = call_nexthop_res_table_notifiers(net, new, extack); + if (err) + return err; + + if (cfg->nh_grp_res_has_idle_timer) + old_res_table->idle_timer = cfg->nh_grp_res_idle_timer; + if (cfg->nh_grp_res_has_unbalanced_timer) + old_res_table->unbalanced_timer = + cfg->nh_grp_res_unbalanced_timer; + + replace_nexthop_grp_res(oldg, newg); + + tmp_table = new_res_table; + rcu_assign_pointer(newg->res_table, old_res_table); + rcu_assign_pointer(newg->spare->res_table, old_res_table); + } + /* update parents - used by nexthop code for cleanup */ for (i = 0; i < newg->num_nh; i++) newg->nh_entries[i].nh_parent = old; rcu_assign_pointer(old->nh_grp, newg); + /* Make sure concurrent readers are not using 'oldg' anymore. */ + synchronize_net(); + + if (newg->resilient) { + rcu_assign_pointer(oldg->res_table, tmp_table); + rcu_assign_pointer(oldg->spare->res_table, tmp_table); + } + for (i = 0; i < oldg->num_nh; i++) oldg->nh_entries[i].nh_parent = new; @@ -864,30 +2016,165 @@ static int replace_nexthop_grp(struct net *net, struct nexthop *old, return 0; } +static void nh_group_v4_update(struct nh_group *nhg) +{ + struct nh_grp_entry *nhges; + bool has_v4 = false; + int i; + + nhges = nhg->nh_entries; + for (i = 0; i < nhg->num_nh; i++) { + struct nh_info *nhi; + + nhi = rtnl_dereference(nhges[i].nh->nh_info); + if (nhi->family == AF_INET) + has_v4 = true; + } + nhg->has_v4 = has_v4; +} + +static int replace_nexthop_single_notify_res(struct net *net, + struct nh_res_table *res_table, + struct nexthop *old, + struct nh_info *oldi, + struct nh_info *newi, + struct netlink_ext_ack *extack) +{ + u32 nhg_id = res_table->nhg_id; + int err; + u16 i; + + for (i = 0; i < res_table->num_nh_buckets; i++) { + struct nh_res_bucket *bucket = &res_table->nh_buckets[i]; + struct nh_grp_entry *nhge; + + nhge = rtnl_dereference(bucket->nh_entry); + if (nhge->nh == old) { + err = __call_nexthop_res_bucket_notifiers(net, nhg_id, + i, true, + oldi, newi, + extack); + if (err) + goto err_notify; + } + } + + return 0; + +err_notify: + while (i-- > 0) { + struct nh_res_bucket *bucket = &res_table->nh_buckets[i]; + struct nh_grp_entry *nhge; + + nhge = rtnl_dereference(bucket->nh_entry); + if (nhge->nh == old) + __call_nexthop_res_bucket_notifiers(net, nhg_id, i, + true, newi, oldi, + extack); + } + return err; +} + +static int replace_nexthop_single_notify(struct net *net, + struct nexthop *group_nh, + struct nexthop *old, + struct nh_info *oldi, + struct nh_info *newi, + struct netlink_ext_ack *extack) +{ + struct nh_group *nhg = rtnl_dereference(group_nh->nh_grp); + struct nh_res_table *res_table; + + if (nhg->hash_threshold) { + return call_nexthop_notifiers(net, NEXTHOP_EVENT_REPLACE, + group_nh, extack); + } else if (nhg->resilient) { + res_table = rtnl_dereference(nhg->res_table); + return replace_nexthop_single_notify_res(net, res_table, + old, oldi, newi, + extack); + } + + return -EINVAL; +} + static int replace_nexthop_single(struct net *net, struct nexthop *old, struct nexthop *new, struct netlink_ext_ack *extack) { + u8 old_protocol, old_nh_flags; struct nh_info *oldi, *newi; + struct nh_grp_entry *nhge; + int err; if (new->is_group) { NL_SET_ERR_MSG(extack, "Can not replace a nexthop with a nexthop group."); return -EINVAL; } + err = call_nexthop_notifiers(net, NEXTHOP_EVENT_REPLACE, new, extack); + if (err) + return err; + + /* Hardware flags were set on 'old' as 'new' is not in the red-black + * tree. Therefore, inherit the flags from 'old' to 'new'. + */ + new->nh_flags |= old->nh_flags & (RTNH_F_OFFLOAD | RTNH_F_TRAP); + oldi = rtnl_dereference(old->nh_info); newi = rtnl_dereference(new->nh_info); newi->nh_parent = old; oldi->nh_parent = new; + old_protocol = old->protocol; + old_nh_flags = old->nh_flags; + old->protocol = new->protocol; old->nh_flags = new->nh_flags; rcu_assign_pointer(old->nh_info, newi); rcu_assign_pointer(new->nh_info, oldi); + /* Send a replace notification for all the groups using the nexthop. */ + list_for_each_entry(nhge, &old->grp_list, nh_list) { + struct nexthop *nhp = nhge->nh_parent; + + err = replace_nexthop_single_notify(net, nhp, old, oldi, newi, + extack); + if (err) + goto err_notify; + } + + /* When replacing an IPv4 nexthop with an IPv6 nexthop, potentially + * update IPv4 indication in all the groups using the nexthop. + */ + if (oldi->family == AF_INET && newi->family == AF_INET6) { + list_for_each_entry(nhge, &old->grp_list, nh_list) { + struct nexthop *nhp = nhge->nh_parent; + struct nh_group *nhg; + + nhg = rtnl_dereference(nhp->nh_grp); + nh_group_v4_update(nhg); + } + } + return 0; + +err_notify: + rcu_assign_pointer(new->nh_info, newi); + rcu_assign_pointer(old->nh_info, oldi); + old->nh_flags = old_nh_flags; + old->protocol = old_protocol; + oldi->nh_parent = old; + newi->nh_parent = new; + list_for_each_entry_continue_reverse(nhge, &old->grp_list, nh_list) { + struct nexthop *nhp = nhge->nh_parent; + + replace_nexthop_single_notify(net, nhp, old, newi, oldi, NULL); + } + call_nexthop_notifiers(net, NEXTHOP_EVENT_REPLACE, old, extack); + return err; } static void __nexthop_replace_notify(struct net *net, struct nexthop *nh, @@ -931,7 +2218,8 @@ static void nexthop_replace_notify(struct net *net, struct nexthop *nh, } static int replace_nexthop(struct net *net, struct nexthop *old, - struct nexthop *new, struct netlink_ext_ack *extack) + struct nexthop *new, const struct nh_config *cfg, + struct netlink_ext_ack *extack) { bool new_is_reject = false; struct nh_grp_entry *nhge; @@ -974,12 +2262,12 @@ static int replace_nexthop(struct net *net, struct nexthop *old, } if (old->is_group) - err = replace_nexthop_grp(net, old, new, extack); + err = replace_nexthop_grp(net, old, new, cfg, extack); else err = replace_nexthop_single(net, old, new, extack); if (!err) { - nh_rt_cache_flush(net, old); + nh_rt_cache_flush(net, old, new); __remove_nexthop(net, new, NULL); nexthop_put(new); @@ -1004,7 +2292,7 @@ static int insert_nexthop(struct net *net, struct nexthop *new_nh, while (1) { struct nexthop *nh; - next = rtnl_dereference(*pp); + next = *pp; if (!next) break; @@ -1016,7 +2304,7 @@ static int insert_nexthop(struct net *net, struct nexthop *new_nh, } else if (new_id > nh->id) { pp = &next->rb_right; } else if (replace) { - rc = replace_nexthop(net, nh, new_nh, extack); + rc = replace_nexthop(net, nh, new_nh, cfg, extack); if (!rc) { new_nh = nh; /* send notification with old nh */ replace_notify = 1; @@ -1034,14 +2322,47 @@ static int insert_nexthop(struct net *net, struct nexthop *new_nh, goto out; } + if (new_nh->is_group) { + struct nh_group *nhg = rtnl_dereference(new_nh->nh_grp); + struct nh_res_table *res_table; + + if (nhg->resilient) { + res_table = rtnl_dereference(nhg->res_table); + + /* Not passing the number of buckets is OK when + * replacing, but not when creating a new group. + */ + if (!cfg->nh_grp_res_has_num_buckets) { + NL_SET_ERR_MSG(extack, "Number of buckets not specified for nexthop group insertion"); + rc = -EINVAL; + goto out; + } + + nh_res_group_rebalance(nhg, res_table); + + /* Do not send bucket notifications, we do full + * notification below. + */ + nh_res_table_upkeep(res_table, false, false); + } + } + rb_link_node_rcu(&new_nh->rb_node, parent, pp); rb_insert_color(&new_nh->rb_node, root); - rc = 0; + + /* The initial insertion is a full notification for hash-threshold as + * well as resilient groups. + */ + rc = call_nexthop_notifiers(net, NEXTHOP_EVENT_REPLACE, new_nh, extack); + if (rc) + rb_erase(&new_nh->rb_node, &net->nexthop.rb_root); + out: if (!rc) { nh_base_seq_inc(net); nexthop_notify(RTM_NEWNEXTHOP, new_nh, &cfg->nlinfo); - if (replace_notify) + if (replace_notify && + READ_ONCE(net->ipv4.sysctl_nexthop_compat_mode)) nexthop_replace_notify(net, new_nh, &cfg->nlinfo); } @@ -1050,7 +2371,7 @@ out: /* rtnl */ /* remove all nexthops tied to a device being deleted */ -static void nexthop_flush_dev(struct net_device *dev) +static void nexthop_flush_dev(struct net_device *dev, unsigned long event) { unsigned int hash = nh_dev_hashfn(dev->ifindex); struct net *net = dev_net(dev); @@ -1062,6 +2383,10 @@ static void nexthop_flush_dev(struct net_device *dev) if (nhi->fib_nhc.nhc_dev != dev) continue; + if (nhi->reject_nh && + (event == NETDEV_DOWN || event == NETDEV_CHANGE)) + continue; + remove_nexthop(net, nhi->nh_parent, NULL); } } @@ -1085,29 +2410,45 @@ static struct nexthop *nexthop_create_group(struct net *net, { struct nlattr *grps_attr = cfg->nh_grp; struct nexthop_grp *entry = nla_data(grps_attr); + u16 num_nh = nla_len(grps_attr) / sizeof(*entry); struct nh_group *nhg; struct nexthop *nh; + int err; int i; + if (WARN_ON(!num_nh)) + return ERR_PTR(-EINVAL); + nh = nexthop_alloc(); if (!nh) return ERR_PTR(-ENOMEM); nh->is_group = 1; - nhg = nexthop_grp_alloc(nla_len(grps_attr) / sizeof(*entry)); + nhg = nexthop_grp_alloc(num_nh); if (!nhg) { kfree(nh); return ERR_PTR(-ENOMEM); } + /* spare group used for removals */ + nhg->spare = nexthop_grp_alloc(num_nh); + if (!nhg->spare) { + kfree(nhg); + kfree(nh); + return ERR_PTR(-ENOMEM); + } + nhg->spare->spare = nhg; + for (i = 0; i < nhg->num_nh; ++i) { struct nexthop *nhe; struct nh_info *nhi; nhe = nexthop_find_by_id(net, entry[i].id); - if (!nexthop_get(nhe)) + if (!nexthop_get(nhe)) { + err = -ENOENT; goto out_no_nh; + } nhi = rtnl_dereference(nhe->nh_info); if (nhi->family == AF_INET) @@ -1120,22 +2461,46 @@ static struct nexthop *nexthop_create_group(struct net *net, } if (cfg->nh_grp_type == NEXTHOP_GRP_TYPE_MPATH) { - nhg->mpath = 1; - nh_group_rebalance(nhg); + nhg->hash_threshold = 1; + nhg->is_multipath = true; + } else if (cfg->nh_grp_type == NEXTHOP_GRP_TYPE_RES) { + struct nh_res_table *res_table; + + res_table = nexthop_res_table_alloc(net, cfg->nh_id, cfg); + if (!res_table) { + err = -ENOMEM; + goto out_no_nh; + } + + rcu_assign_pointer(nhg->spare->res_table, res_table); + rcu_assign_pointer(nhg->res_table, res_table); + nhg->resilient = true; + nhg->is_multipath = true; } + WARN_ON_ONCE(nhg->hash_threshold + nhg->resilient != 1); + + if (nhg->hash_threshold) + nh_hthr_group_rebalance(nhg); + + if (cfg->nh_fdb) + nhg->fdb_nh = 1; + rcu_assign_pointer(nh->nh_grp, nhg); return nh; out_no_nh: - for (; i >= 0; --i) + for (i--; i >= 0; --i) { + list_del(&nhg->nh_entries[i].nh_list); nexthop_put(nhg->nh_entries[i].nh); + } + kfree(nhg->spare); kfree(nhg); kfree(nh); - return ERR_PTR(-ENOENT); + return ERR_PTR(err); } static int nh_create_ipv4(struct net *net, struct nexthop *nh, @@ -1148,10 +2513,11 @@ static int nh_create_ipv4(struct net *net, struct nexthop *nh, .fc_gw4 = cfg->gw.ipv4, .fc_gw_family = cfg->gw.ipv4 ? AF_INET : 0, .fc_flags = cfg->nh_flags, + .fc_nlinfo = cfg->nlinfo, .fc_encap = cfg->nh_encap, .fc_encap_type = cfg->nh_encap_type, }; - u32 tb_id = l3mdev_fib_table(cfg->dev); + u32 tb_id = (cfg->dev ? l3mdev_fib_table(cfg->dev) : RT_TABLE_MAIN); int err; err = fib_nh_init(net, fib_nh, &fib_cfg, 1, extack); @@ -1160,12 +2526,15 @@ static int nh_create_ipv4(struct net *net, struct nexthop *nh, goto out; } + if (nhi->fdb_nh) + goto out; + /* sets nh_dev if successful */ err = fib_check_nh(net, fib_nh, tb_id, 0, extack); if (!err) { nh->nh_flags = fib_nh->fib_nh_flags; fib_info_update_nhc_saddr(net, &fib_nh->nh_common, - fib_nh->fib_nh_scope); + !fib_nh->fib_nh_scope ? 0 : fib_nh->fib_nh_scope - 1); } else { fib_nh_release(net, fib_nh); } @@ -1183,8 +2552,10 @@ static int nh_create_ipv6(struct net *net, struct nexthop *nh, .fc_ifindex = cfg->nh_ifindex, .fc_gateway = cfg->gw.ipv6, .fc_flags = cfg->nh_flags, + .fc_nlinfo = cfg->nlinfo, .fc_encap = cfg->nh_encap, .fc_encap_type = cfg->nh_encap_type, + .fc_is_fdb = cfg->nh_fdb, }; int err; @@ -1194,11 +2565,15 @@ static int nh_create_ipv6(struct net *net, struct nexthop *nh, /* sets nh_dev if successful */ err = ipv6_stub->fib6_nh_init(net, fib6_nh, &fib6_cfg, GFP_KERNEL, extack); - if (err) + if (err) { + /* IPv6 is not enabled, don't call fib6_nh_release */ + if (err == -EAFNOSUPPORT) + goto out; ipv6_stub->fib6_nh_release(fib6_nh); - else + } else { nh->nh_flags = fib6_nh->fib_nh_flags; - + } +out: return err; } @@ -1226,6 +2601,9 @@ static struct nexthop *nexthop_create(struct net *net, struct nh_config *cfg, nhi->family = cfg->nh_family; nhi->fib_nhc.nhc_scope = RT_SCOPE_LINK; + if (cfg->nh_fdb) + nhi->fdb_nh = 1; + if (cfg->nh_blackhole) { nhi->reject_nh = 1; cfg->nh_ifindex = net->loopback_dev->ifindex; @@ -1247,7 +2625,8 @@ static struct nexthop *nexthop_create(struct net *net, struct nh_config *cfg, } /* add the entry to the device based hash */ - nexthop_devhash_add(net, nhi); + if (!nhi->fdb_nh) + nexthop_devhash_add(net, nhi); rcu_assign_pointer(nh->nh_info, nhi); @@ -1297,16 +2676,81 @@ static struct nexthop *nexthop_add(struct net *net, struct nh_config *cfg, return nh; } +static int rtm_nh_get_timer(struct nlattr *attr, unsigned long fallback, + unsigned long *timer_p, bool *has_p, + struct netlink_ext_ack *extack) +{ + unsigned long timer; + u32 value; + + if (!attr) { + *timer_p = fallback; + *has_p = false; + return 0; + } + + value = nla_get_u32(attr); + timer = clock_t_to_jiffies(value); + if (timer == ~0UL) { + NL_SET_ERR_MSG(extack, "Timer value too large"); + return -EINVAL; + } + + *timer_p = timer; + *has_p = true; + return 0; +} + +static int rtm_to_nh_config_grp_res(struct nlattr *res, struct nh_config *cfg, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[ARRAY_SIZE(rtm_nh_res_policy_new)] = {}; + int err; + + if (res) { + err = nla_parse_nested(tb, + ARRAY_SIZE(rtm_nh_res_policy_new) - 1, + res, rtm_nh_res_policy_new, extack); + if (err < 0) + return err; + } + + if (tb[NHA_RES_GROUP_BUCKETS]) { + cfg->nh_grp_res_num_buckets = + nla_get_u16(tb[NHA_RES_GROUP_BUCKETS]); + cfg->nh_grp_res_has_num_buckets = true; + if (!cfg->nh_grp_res_num_buckets) { + NL_SET_ERR_MSG(extack, "Number of buckets needs to be non-0"); + return -EINVAL; + } + } + + err = rtm_nh_get_timer(tb[NHA_RES_GROUP_IDLE_TIMER], + NH_RES_DEFAULT_IDLE_TIMER, + &cfg->nh_grp_res_idle_timer, + &cfg->nh_grp_res_has_idle_timer, + extack); + if (err) + return err; + + return rtm_nh_get_timer(tb[NHA_RES_GROUP_UNBALANCED_TIMER], + NH_RES_DEFAULT_UNBALANCED_TIMER, + &cfg->nh_grp_res_unbalanced_timer, + &cfg->nh_grp_res_has_unbalanced_timer, + extack); +} + static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, struct nlmsghdr *nlh, struct nh_config *cfg, struct netlink_ext_ack *extack) { struct nhmsg *nhm = nlmsg_data(nlh); - struct nlattr *tb[NHA_MAX + 1]; + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_new)]; int err; - err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy, - extack); + err = nlmsg_parse(nlh, sizeof(*nhm), tb, + ARRAY_SIZE(rtm_nh_policy_new) - 1, + rtm_nh_policy_new, extack); if (err < 0) return err; @@ -1327,17 +2771,12 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, case AF_UNSPEC: if (tb[NHA_GROUP]) break; - /* fallthrough */ + fallthrough; default: NL_SET_ERR_MSG(extack, "Invalid address family"); goto out; } - if (tb[NHA_GROUPS] || tb[NHA_MASTER]) { - NL_SET_ERR_MSG(extack, "Invalid attributes in request"); - goto out; - } - memset(cfg, 0, sizeof(*cfg)); cfg->nlflags = nlh->nlmsg_flags; cfg->nlinfo.portid = NETLINK_CB(skb).portid; @@ -1351,6 +2790,19 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, if (tb[NHA_ID]) cfg->nh_id = nla_get_u32(tb[NHA_ID]); + if (tb[NHA_FDB]) { + if (tb[NHA_OIF] || tb[NHA_BLACKHOLE] || + tb[NHA_ENCAP] || tb[NHA_ENCAP_TYPE]) { + NL_SET_ERR_MSG(extack, "Fdb attribute can not be used with encap, oif or blackhole"); + goto out; + } + if (nhm->nh_flags) { + NL_SET_ERR_MSG(extack, "Unsupported nexthop flags in ancillary header"); + goto out; + } + cfg->nh_fdb = nla_get_flag(tb[NHA_FDB]); + } + if (tb[NHA_GROUP]) { if (nhm->nh_family != AF_UNSPEC) { NL_SET_ERR_MSG(extack, "Invalid family for group"); @@ -1366,7 +2818,14 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, NL_SET_ERR_MSG(extack, "Invalid group type"); goto out; } - err = nh_check_attr_group(net, tb, extack); + err = nh_check_attr_group(net, tb, ARRAY_SIZE(tb), + cfg->nh_grp_type, extack); + if (err) + goto out; + + if (cfg->nh_grp_type == NEXTHOP_GRP_TYPE_RES) + err = rtm_to_nh_config_grp_res(tb[NHA_RES_GROUP], + cfg, extack); /* no other attributes should be set */ goto out; @@ -1374,8 +2833,8 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, if (tb[NHA_BLACKHOLE]) { if (tb[NHA_GATEWAY] || tb[NHA_OIF] || - tb[NHA_ENCAP] || tb[NHA_ENCAP_TYPE]) { - NL_SET_ERR_MSG(extack, "Blackhole attribute can not be used with gateway or oif"); + tb[NHA_ENCAP] || tb[NHA_ENCAP_TYPE] || tb[NHA_FDB]) { + NL_SET_ERR_MSG(extack, "Blackhole attribute can not be used with gateway, oif, encap or fdb"); goto out; } @@ -1384,26 +2843,28 @@ static int rtm_to_nh_config(struct net *net, struct sk_buff *skb, goto out; } - if (!tb[NHA_OIF]) { - NL_SET_ERR_MSG(extack, "Device attribute required for non-blackhole nexthops"); + if (!cfg->nh_fdb && !tb[NHA_OIF]) { + NL_SET_ERR_MSG(extack, "Device attribute required for non-blackhole and non-fdb nexthops"); goto out; } - cfg->nh_ifindex = nla_get_u32(tb[NHA_OIF]); - if (cfg->nh_ifindex) - cfg->dev = __dev_get_by_index(net, cfg->nh_ifindex); + if (!cfg->nh_fdb && tb[NHA_OIF]) { + cfg->nh_ifindex = nla_get_u32(tb[NHA_OIF]); + if (cfg->nh_ifindex) + cfg->dev = __dev_get_by_index(net, cfg->nh_ifindex); - if (!cfg->dev) { - NL_SET_ERR_MSG(extack, "Invalid device index"); - goto out; - } else if (!(cfg->dev->flags & IFF_UP)) { - NL_SET_ERR_MSG(extack, "Nexthop device is not up"); - err = -ENETDOWN; - goto out; - } else if (!netif_carrier_ok(cfg->dev)) { - NL_SET_ERR_MSG(extack, "Carrier for nexthop device is down"); - err = -ENETDOWN; - goto out; + if (!cfg->dev) { + NL_SET_ERR_MSG(extack, "Invalid device index"); + goto out; + } else if (!(cfg->dev->flags & IFF_UP)) { + NL_SET_ERR_MSG(extack, "Nexthop device is not up"); + err = -ENETDOWN; + goto out; + } else if (!netif_carrier_ok(cfg->dev)) { + NL_SET_ERR_MSG(extack, "Carrier for nexthop device is down"); + err = -ENETDOWN; + goto out; + } } err = -EINVAL; @@ -1482,49 +2943,44 @@ static int rtm_new_nexthop(struct sk_buff *skb, struct nlmsghdr *nlh, return err; } -static int nh_valid_get_del_req(struct nlmsghdr *nlh, u32 *id, - struct netlink_ext_ack *extack) +static int __nh_valid_get_del_req(const struct nlmsghdr *nlh, + struct nlattr **tb, u32 *id, + struct netlink_ext_ack *extack) { struct nhmsg *nhm = nlmsg_data(nlh); - struct nlattr *tb[NHA_MAX + 1]; - int err, i; - - err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy, - extack); - if (err < 0) - return err; - - err = -EINVAL; - for (i = 0; i < __NHA_MAX; ++i) { - if (!tb[i]) - continue; - switch (i) { - case NHA_ID: - break; - default: - NL_SET_ERR_MSG_ATTR(extack, tb[i], - "Unexpected attribute in request"); - goto out; - } - } if (nhm->nh_protocol || nhm->resvd || nhm->nh_scope || nhm->nh_flags) { NL_SET_ERR_MSG(extack, "Invalid values in header"); - goto out; + return -EINVAL; } if (!tb[NHA_ID]) { NL_SET_ERR_MSG(extack, "Nexthop id is missing"); - goto out; + return -EINVAL; } *id = nla_get_u32(tb[NHA_ID]); - if (!(*id)) + if (!(*id)) { NL_SET_ERR_MSG(extack, "Invalid nexthop id"); - else - err = 0; -out: - return err; + return -EINVAL; + } + + return 0; +} + +static int nh_valid_get_del_req(const struct nlmsghdr *nlh, u32 *id, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_get)]; + int err; + + err = nlmsg_parse(nlh, sizeof(struct nhmsg), tb, + ARRAY_SIZE(rtm_nh_policy_get) - 1, + rtm_nh_policy_get, extack); + if (err < 0) + return err; + + return __nh_valid_get_del_req(nlh, tb, id, extack); } /* rtnl */ @@ -1593,16 +3049,25 @@ errout_free: goto out; } -static bool nh_dump_filtered(struct nexthop *nh, int dev_idx, int master_idx, - bool group_filter, u8 family) +struct nh_dump_filter { + u32 nh_id; + int dev_idx; + int master_idx; + bool group_filter; + bool fdb_filter; + u32 res_bucket_nh_id; +}; + +static bool nh_dump_filtered(struct nexthop *nh, + struct nh_dump_filter *filter, u8 family) { const struct net_device *dev; const struct nh_info *nhi; - if (group_filter && !nh->is_group) + if (filter->group_filter && !nh->is_group) return true; - if (!dev_idx && !master_idx && !family) + if (!filter->dev_idx && !filter->master_idx && !family) return false; if (nh->is_group) @@ -1613,67 +3078,48 @@ static bool nh_dump_filtered(struct nexthop *nh, int dev_idx, int master_idx, return true; dev = nhi->fib_nhc.nhc_dev; - if (dev_idx && (!dev || dev->ifindex != dev_idx)) + if (filter->dev_idx && (!dev || dev->ifindex != filter->dev_idx)) return true; - if (master_idx) { + if (filter->master_idx) { struct net_device *master; if (!dev) return true; master = netdev_master_upper_dev_get((struct net_device *)dev); - if (!master || master->ifindex != master_idx) + if (!master || master->ifindex != filter->master_idx) return true; } return false; } -static int nh_valid_dump_req(const struct nlmsghdr *nlh, int *dev_idx, - int *master_idx, bool *group_filter, - struct netlink_callback *cb) +static int __nh_valid_dump_req(const struct nlmsghdr *nlh, struct nlattr **tb, + struct nh_dump_filter *filter, + struct netlink_ext_ack *extack) { - struct netlink_ext_ack *extack = cb->extack; - struct nlattr *tb[NHA_MAX + 1]; struct nhmsg *nhm; - int err, i; u32 idx; - err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy, - NULL); - if (err < 0) - return err; - - for (i = 0; i <= NHA_MAX; ++i) { - if (!tb[i]) - continue; - - switch (i) { - case NHA_OIF: - idx = nla_get_u32(tb[i]); - if (idx > INT_MAX) { - NL_SET_ERR_MSG(extack, "Invalid device index"); - return -EINVAL; - } - *dev_idx = idx; - break; - case NHA_MASTER: - idx = nla_get_u32(tb[i]); - if (idx > INT_MAX) { - NL_SET_ERR_MSG(extack, "Invalid master device index"); - return -EINVAL; - } - *master_idx = idx; - break; - case NHA_GROUPS: - *group_filter = true; - break; - default: - NL_SET_ERR_MSG(extack, "Unsupported attribute in dump request"); + if (tb[NHA_OIF]) { + idx = nla_get_u32(tb[NHA_OIF]); + if (idx > INT_MAX) { + NL_SET_ERR_MSG(extack, "Invalid device index"); + return -EINVAL; + } + filter->dev_idx = idx; + } + if (tb[NHA_MASTER]) { + idx = nla_get_u32(tb[NHA_MASTER]); + if (idx > INT_MAX) { + NL_SET_ERR_MSG(extack, "Invalid master device index"); return -EINVAL; } + filter->master_idx = idx; } + filter->group_filter = nla_get_flag(tb[NHA_GROUPS]); + filter->fdb_filter = nla_get_flag(tb[NHA_FDB]); nhm = nlmsg_data(nlh); if (nhm->nh_protocol || nhm->resvd || nhm->nh_scope || nhm->nh_flags) { @@ -1684,55 +3130,418 @@ static int nh_valid_dump_req(const struct nlmsghdr *nlh, int *dev_idx, return 0; } +static int nh_valid_dump_req(const struct nlmsghdr *nlh, + struct nh_dump_filter *filter, + struct netlink_callback *cb) +{ + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_dump)]; + int err; + + err = nlmsg_parse(nlh, sizeof(struct nhmsg), tb, + ARRAY_SIZE(rtm_nh_policy_dump) - 1, + rtm_nh_policy_dump, cb->extack); + if (err < 0) + return err; + + return __nh_valid_dump_req(nlh, tb, filter, cb->extack); +} + +struct rtm_dump_nh_ctx { + u32 idx; +}; + +static struct rtm_dump_nh_ctx * +rtm_dump_nh_ctx(struct netlink_callback *cb) +{ + struct rtm_dump_nh_ctx *ctx = (void *)cb->ctx; + + BUILD_BUG_ON(sizeof(*ctx) > sizeof(cb->ctx)); + return ctx; +} + +static int rtm_dump_walk_nexthops(struct sk_buff *skb, + struct netlink_callback *cb, + struct rb_root *root, + struct rtm_dump_nh_ctx *ctx, + int (*nh_cb)(struct sk_buff *skb, + struct netlink_callback *cb, + struct nexthop *nh, void *data), + void *data) +{ + struct rb_node *node; + int s_idx; + int err; + + s_idx = ctx->idx; + for (node = rb_first(root); node; node = rb_next(node)) { + struct nexthop *nh; + + nh = rb_entry(node, struct nexthop, rb_node); + if (nh->id < s_idx) + continue; + + ctx->idx = nh->id; + err = nh_cb(skb, cb, nh, data); + if (err) + return err; + } + + ctx->idx++; + return 0; +} + +static int rtm_dump_nexthop_cb(struct sk_buff *skb, struct netlink_callback *cb, + struct nexthop *nh, void *data) +{ + struct nhmsg *nhm = nlmsg_data(cb->nlh); + struct nh_dump_filter *filter = data; + + if (nh_dump_filtered(nh, filter, nhm->nh_family)) + return 0; + + return nh_fill_node(skb, nh, RTM_NEWNEXTHOP, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI); +} + /* rtnl */ static int rtm_dump_nexthop(struct sk_buff *skb, struct netlink_callback *cb) { - struct nhmsg *nhm = nlmsg_data(cb->nlh); - int dev_filter_idx = 0, master_idx = 0; + struct rtm_dump_nh_ctx *ctx = rtm_dump_nh_ctx(cb); struct net *net = sock_net(skb->sk); struct rb_root *root = &net->nexthop.rb_root; - bool group_filter = false; - struct rb_node *node; - int idx = 0, s_idx; + struct nh_dump_filter filter = {}; int err; - err = nh_valid_dump_req(cb->nlh, &dev_filter_idx, &master_idx, - &group_filter, cb); + err = nh_valid_dump_req(cb->nlh, &filter, cb); if (err < 0) return err; - s_idx = cb->args[0]; - for (node = rb_first(root); node; node = rb_next(node)) { - struct nexthop *nh; + err = rtm_dump_walk_nexthops(skb, cb, root, ctx, + &rtm_dump_nexthop_cb, &filter); + if (err < 0) { + if (likely(skb->len)) + goto out; + goto out_err; + } - if (idx < s_idx) - goto cont; +out: + err = skb->len; +out_err: + cb->seq = net->nexthop.seq; + nl_dump_check_consistent(cb, nlmsg_hdr(skb)); + return err; +} - nh = rb_entry(node, struct nexthop, rb_node); - if (nh_dump_filtered(nh, dev_filter_idx, master_idx, - group_filter, nhm->nh_family)) - goto cont; +static struct nexthop * +nexthop_find_group_resilient(struct net *net, u32 id, + struct netlink_ext_ack *extack) +{ + struct nh_group *nhg; + struct nexthop *nh; + + nh = nexthop_find_by_id(net, id); + if (!nh) + return ERR_PTR(-ENOENT); + + if (!nh->is_group) { + NL_SET_ERR_MSG(extack, "Not a nexthop group"); + return ERR_PTR(-EINVAL); + } + + nhg = rtnl_dereference(nh->nh_grp); + if (!nhg->resilient) { + NL_SET_ERR_MSG(extack, "Nexthop group not of type resilient"); + return ERR_PTR(-EINVAL); + } + + return nh; +} + +static int nh_valid_dump_nhid(struct nlattr *attr, u32 *nh_id_p, + struct netlink_ext_ack *extack) +{ + u32 idx; + + if (attr) { + idx = nla_get_u32(attr); + if (!idx) { + NL_SET_ERR_MSG(extack, "Invalid nexthop id"); + return -EINVAL; + } + *nh_id_p = idx; + } else { + *nh_id_p = 0; + } + + return 0; +} + +static int nh_valid_dump_bucket_req(const struct nlmsghdr *nlh, + struct nh_dump_filter *filter, + struct netlink_callback *cb) +{ + struct nlattr *res_tb[ARRAY_SIZE(rtm_nh_res_bucket_policy_dump)]; + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_dump_bucket)]; + int err; + + err = nlmsg_parse(nlh, sizeof(struct nhmsg), tb, + ARRAY_SIZE(rtm_nh_policy_dump_bucket) - 1, + rtm_nh_policy_dump_bucket, NULL); + if (err < 0) + return err; + + err = nh_valid_dump_nhid(tb[NHA_ID], &filter->nh_id, cb->extack); + if (err) + return err; + + if (tb[NHA_RES_BUCKET]) { + size_t max = ARRAY_SIZE(rtm_nh_res_bucket_policy_dump) - 1; + + err = nla_parse_nested(res_tb, max, + tb[NHA_RES_BUCKET], + rtm_nh_res_bucket_policy_dump, + cb->extack); + if (err < 0) + return err; + + err = nh_valid_dump_nhid(res_tb[NHA_RES_BUCKET_NH_ID], + &filter->res_bucket_nh_id, + cb->extack); + if (err) + return err; + } + + return __nh_valid_dump_req(nlh, tb, filter, cb->extack); +} - err = nh_fill_node(skb, nh, RTM_NEWNEXTHOP, - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, NLM_F_MULTI); +struct rtm_dump_res_bucket_ctx { + struct rtm_dump_nh_ctx nh; + u16 bucket_index; + u32 done_nh_idx; /* 1 + the index of the last fully processed NH. */ +}; + +static struct rtm_dump_res_bucket_ctx * +rtm_dump_res_bucket_ctx(struct netlink_callback *cb) +{ + struct rtm_dump_res_bucket_ctx *ctx = (void *)cb->ctx; + + BUILD_BUG_ON(sizeof(*ctx) > sizeof(cb->ctx)); + return ctx; +} + +struct rtm_dump_nexthop_bucket_data { + struct rtm_dump_res_bucket_ctx *ctx; + struct nh_dump_filter filter; +}; + +static int rtm_dump_nexthop_bucket_nh(struct sk_buff *skb, + struct netlink_callback *cb, + struct nexthop *nh, + struct rtm_dump_nexthop_bucket_data *dd) +{ + u32 portid = NETLINK_CB(cb->skb).portid; + struct nhmsg *nhm = nlmsg_data(cb->nlh); + struct nh_res_table *res_table; + struct nh_group *nhg; + u16 bucket_index; + int err; + + if (dd->ctx->nh.idx < dd->ctx->done_nh_idx) + return 0; + + nhg = rtnl_dereference(nh->nh_grp); + res_table = rtnl_dereference(nhg->res_table); + for (bucket_index = dd->ctx->bucket_index; + bucket_index < res_table->num_nh_buckets; + bucket_index++) { + struct nh_res_bucket *bucket; + struct nh_grp_entry *nhge; + + bucket = &res_table->nh_buckets[bucket_index]; + nhge = rtnl_dereference(bucket->nh_entry); + if (nh_dump_filtered(nhge->nh, &dd->filter, nhm->nh_family)) + continue; + + if (dd->filter.res_bucket_nh_id && + dd->filter.res_bucket_nh_id != nhge->nh->id) + continue; + + err = nh_fill_res_bucket(skb, nh, bucket, bucket_index, + RTM_NEWNEXTHOPBUCKET, portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + cb->extack); if (err < 0) { if (likely(skb->len)) goto out; - goto out_err; } -cont: - idx++; + } + + dd->ctx->done_nh_idx = dd->ctx->nh.idx + 1; + bucket_index = 0; + +out: + err = skb->len; +out_err: + dd->ctx->bucket_index = bucket_index; + return err; +} + +static int rtm_dump_nexthop_bucket_cb(struct sk_buff *skb, + struct netlink_callback *cb, + struct nexthop *nh, void *data) +{ + struct rtm_dump_nexthop_bucket_data *dd = data; + struct nh_group *nhg; + + if (!nh->is_group) + return 0; + + nhg = rtnl_dereference(nh->nh_grp); + if (!nhg->resilient) + return 0; + + return rtm_dump_nexthop_bucket_nh(skb, cb, nh, dd); +} + +/* rtnl */ +static int rtm_dump_nexthop_bucket(struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct rtm_dump_res_bucket_ctx *ctx = rtm_dump_res_bucket_ctx(cb); + struct rtm_dump_nexthop_bucket_data dd = { .ctx = ctx }; + struct net *net = sock_net(skb->sk); + struct nexthop *nh; + int err; + + err = nh_valid_dump_bucket_req(cb->nlh, &dd.filter, cb); + if (err) + return err; + + if (dd.filter.nh_id) { + nh = nexthop_find_group_resilient(net, dd.filter.nh_id, + cb->extack); + if (IS_ERR(nh)) + return PTR_ERR(nh); + err = rtm_dump_nexthop_bucket_nh(skb, cb, nh, &dd); + } else { + struct rb_root *root = &net->nexthop.rb_root; + + err = rtm_dump_walk_nexthops(skb, cb, root, &ctx->nh, + &rtm_dump_nexthop_bucket_cb, &dd); + } + + if (err < 0) { + if (likely(skb->len)) + goto out; + goto out_err; } out: err = skb->len; out_err: - cb->args[0] = idx; cb->seq = net->nexthop.seq; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); + return err; +} +static int nh_valid_get_bucket_req_res_bucket(struct nlattr *res, + u16 *bucket_index, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[ARRAY_SIZE(rtm_nh_res_bucket_policy_get)]; + int err; + + err = nla_parse_nested(tb, ARRAY_SIZE(rtm_nh_res_bucket_policy_get) - 1, + res, rtm_nh_res_bucket_policy_get, extack); + if (err < 0) + return err; + + if (!tb[NHA_RES_BUCKET_INDEX]) { + NL_SET_ERR_MSG(extack, "Bucket index is missing"); + return -EINVAL; + } + + *bucket_index = nla_get_u16(tb[NHA_RES_BUCKET_INDEX]); + return 0; +} + +static int nh_valid_get_bucket_req(const struct nlmsghdr *nlh, + u32 *id, u16 *bucket_index, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[ARRAY_SIZE(rtm_nh_policy_get_bucket)]; + int err; + + err = nlmsg_parse(nlh, sizeof(struct nhmsg), tb, + ARRAY_SIZE(rtm_nh_policy_get_bucket) - 1, + rtm_nh_policy_get_bucket, extack); + if (err < 0) + return err; + + err = __nh_valid_get_del_req(nlh, tb, id, extack); + if (err) + return err; + + if (!tb[NHA_RES_BUCKET]) { + NL_SET_ERR_MSG(extack, "Bucket information is missing"); + return -EINVAL; + } + + err = nh_valid_get_bucket_req_res_bucket(tb[NHA_RES_BUCKET], + bucket_index, extack); + if (err) + return err; + + return 0; +} + +/* rtnl */ +static int rtm_get_nexthop_bucket(struct sk_buff *in_skb, struct nlmsghdr *nlh, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(in_skb->sk); + struct nh_res_table *res_table; + struct sk_buff *skb = NULL; + struct nh_group *nhg; + struct nexthop *nh; + u16 bucket_index; + int err; + u32 id; + + err = nh_valid_get_bucket_req(nlh, &id, &bucket_index, extack); + if (err) + return err; + + nh = nexthop_find_group_resilient(net, id, extack); + if (IS_ERR(nh)) + return PTR_ERR(nh); + + nhg = rtnl_dereference(nh->nh_grp); + res_table = rtnl_dereference(nhg->res_table); + if (bucket_index >= res_table->num_nh_buckets) { + NL_SET_ERR_MSG(extack, "Bucket index out of bounds"); + return -ENOENT; + } + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + err = nh_fill_res_bucket(skb, nh, &res_table->nh_buckets[bucket_index], + bucket_index, RTM_NEWNEXTHOPBUCKET, + NETLINK_CB(in_skb).portid, nlh->nlmsg_seq, + 0, extack); + if (err < 0) { + WARN_ON(err == -EMSGSIZE); + goto errout_free; + } + + return rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid); + +errout_free: + kfree_skb(skb); return err; } @@ -1763,11 +3572,11 @@ static int nh_netdev_event(struct notifier_block *this, switch (event) { case NETDEV_DOWN: case NETDEV_UNREGISTER: - nexthop_flush_dev(dev); + nexthop_flush_dev(dev, event); break; case NETDEV_CHANGE: if (!(dev_get_flags(dev) & (IFF_RUNNING | IFF_LOWER_UP))) - nexthop_flush_dev(dev); + nexthop_flush_dev(dev, event); break; case NETDEV_CHANGEMTU: info_ext = ptr; @@ -1782,12 +3591,159 @@ static struct notifier_block nh_netdev_notifier = { .notifier_call = nh_netdev_event, }; -static void __net_exit nexthop_net_exit(struct net *net) +static int nexthops_dump(struct net *net, struct notifier_block *nb, + enum nexthop_event_type event_type, + struct netlink_ext_ack *extack) { + struct rb_root *root = &net->nexthop.rb_root; + struct rb_node *node; + int err = 0; + + for (node = rb_first(root); node; node = rb_next(node)) { + struct nexthop *nh; + + nh = rb_entry(node, struct nexthop, rb_node); + err = call_nexthop_notifier(nb, net, event_type, nh, extack); + if (err) + break; + } + + return err; +} + +int register_nexthop_notifier(struct net *net, struct notifier_block *nb, + struct netlink_ext_ack *extack) +{ + int err; + + rtnl_lock(); + err = nexthops_dump(net, nb, NEXTHOP_EVENT_REPLACE, extack); + if (err) + goto unlock; + err = blocking_notifier_chain_register(&net->nexthop.notifier_chain, + nb); +unlock: + rtnl_unlock(); + return err; +} +EXPORT_SYMBOL(register_nexthop_notifier); + +int unregister_nexthop_notifier(struct net *net, struct notifier_block *nb) +{ + int err; + + rtnl_lock(); + err = blocking_notifier_chain_unregister(&net->nexthop.notifier_chain, + nb); + if (err) + goto unlock; + nexthops_dump(net, nb, NEXTHOP_EVENT_DEL, NULL); +unlock: + rtnl_unlock(); + return err; +} +EXPORT_SYMBOL(unregister_nexthop_notifier); + +void nexthop_set_hw_flags(struct net *net, u32 id, bool offload, bool trap) +{ + struct nexthop *nexthop; + + rcu_read_lock(); + + nexthop = nexthop_find_by_id(net, id); + if (!nexthop) + goto out; + + nexthop->nh_flags &= ~(RTNH_F_OFFLOAD | RTNH_F_TRAP); + if (offload) + nexthop->nh_flags |= RTNH_F_OFFLOAD; + if (trap) + nexthop->nh_flags |= RTNH_F_TRAP; + +out: + rcu_read_unlock(); +} +EXPORT_SYMBOL(nexthop_set_hw_flags); + +void nexthop_bucket_set_hw_flags(struct net *net, u32 id, u16 bucket_index, + bool offload, bool trap) +{ + struct nh_res_table *res_table; + struct nh_res_bucket *bucket; + struct nexthop *nexthop; + struct nh_group *nhg; + + rcu_read_lock(); + + nexthop = nexthop_find_by_id(net, id); + if (!nexthop || !nexthop->is_group) + goto out; + + nhg = rcu_dereference(nexthop->nh_grp); + if (!nhg->resilient) + goto out; + + if (bucket_index >= nhg->res_table->num_nh_buckets) + goto out; + + res_table = rcu_dereference(nhg->res_table); + bucket = &res_table->nh_buckets[bucket_index]; + bucket->nh_flags &= ~(RTNH_F_OFFLOAD | RTNH_F_TRAP); + if (offload) + bucket->nh_flags |= RTNH_F_OFFLOAD; + if (trap) + bucket->nh_flags |= RTNH_F_TRAP; + +out: + rcu_read_unlock(); +} +EXPORT_SYMBOL(nexthop_bucket_set_hw_flags); + +void nexthop_res_grp_activity_update(struct net *net, u32 id, u16 num_buckets, + unsigned long *activity) +{ + struct nh_res_table *res_table; + struct nexthop *nexthop; + struct nh_group *nhg; + u16 i; + + rcu_read_lock(); + + nexthop = nexthop_find_by_id(net, id); + if (!nexthop || !nexthop->is_group) + goto out; + + nhg = rcu_dereference(nexthop->nh_grp); + if (!nhg->resilient) + goto out; + + /* Instead of silently ignoring some buckets, demand that the sizes + * be the same. + */ + res_table = rcu_dereference(nhg->res_table); + if (num_buckets != res_table->num_nh_buckets) + goto out; + + for (i = 0; i < num_buckets; i++) { + if (test_bit(i, activity)) + nh_res_bucket_set_busy(&res_table->nh_buckets[i]); + } + +out: + rcu_read_unlock(); +} +EXPORT_SYMBOL(nexthop_res_grp_activity_update); + +static void __net_exit nexthop_net_exit_batch(struct list_head *net_list) +{ + struct net *net; + rtnl_lock(); - flush_all_nexthops(net); + list_for_each_entry(net, net_list, exit_list) { + flush_all_nexthops(net); + kfree(net->nexthop.devhash); + } rtnl_unlock(); - kfree(net->nexthop.devhash); } static int __net_init nexthop_net_init(struct net *net) @@ -1798,13 +3754,14 @@ static int __net_init nexthop_net_init(struct net *net) net->nexthop.devhash = kzalloc(sz, GFP_KERNEL); if (!net->nexthop.devhash) return -ENOMEM; + BLOCKING_INIT_NOTIFIER_HEAD(&net->nexthop.notifier_chain); return 0; } static struct pernet_operations nexthop_net_ops = { .init = nexthop_net_init, - .exit = nexthop_net_exit, + .exit_batch = nexthop_net_exit_batch, }; static int __init nexthop_init(void) @@ -1824,6 +3781,9 @@ static int __init nexthop_init(void) rtnl_register(PF_INET6, RTM_NEWNEXTHOP, rtm_new_nexthop, NULL, 0); rtnl_register(PF_INET6, RTM_GETNEXTHOP, NULL, rtm_dump_nexthop, 0); + rtnl_register(PF_UNSPEC, RTM_GETNEXTHOPBUCKET, rtm_get_nexthop_bucket, + rtm_dump_nexthop_bucket, 0); + return 0; } subsys_initcall(nexthop_init); diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 535427292194..bde333b24837 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -33,6 +33,7 @@ #include <linux/skbuff.h> #include <linux/proc_fs.h> #include <linux/export.h> +#include <linux/bpf-cgroup.h> #include <net/sock.h> #include <net/ping.h> #include <net/udp.h> @@ -50,7 +51,7 @@ struct ping_table { struct hlist_nulls_head hash[PING_HTABLE_SIZE]; - rwlock_t lock; + spinlock_t lock; }; static struct ping_table ping_table; @@ -82,7 +83,7 @@ int ping_get_port(struct sock *sk, unsigned short ident) struct sock *sk2 = NULL; isk = inet_sk(sk); - write_lock_bh(&ping_table.lock); + spin_lock(&ping_table.lock); if (ident == 0) { u32 i; u16 result = ping_port_rover + 1; @@ -128,14 +129,15 @@ next_port: if (sk_unhashed(sk)) { pr_debug("was not hashed\n"); sock_hold(sk); - hlist_nulls_add_head(&sk->sk_nulls_node, hlist); + sock_set_flag(sk, SOCK_RCU_FREE); + hlist_nulls_add_head_rcu(&sk->sk_nulls_node, hlist); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); } - write_unlock_bh(&ping_table.lock); + spin_unlock(&ping_table.lock); return 0; fail: - write_unlock_bh(&ping_table.lock); + spin_unlock(&ping_table.lock); return 1; } EXPORT_SYMBOL_GPL(ping_get_port); @@ -153,39 +155,43 @@ void ping_unhash(struct sock *sk) struct inet_sock *isk = inet_sk(sk); pr_debug("ping_unhash(isk=%p,isk->num=%u)\n", isk, isk->inet_num); - write_lock_bh(&ping_table.lock); + spin_lock(&ping_table.lock); if (sk_hashed(sk)) { - hlist_nulls_del(&sk->sk_nulls_node); - sk_nulls_node_init(&sk->sk_nulls_node); + hlist_nulls_del_init_rcu(&sk->sk_nulls_node); sock_put(sk); isk->inet_num = 0; isk->inet_sport = 0; sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); } - write_unlock_bh(&ping_table.lock); + spin_unlock(&ping_table.lock); } EXPORT_SYMBOL_GPL(ping_unhash); +/* Called under rcu_read_lock() */ static struct sock *ping_lookup(struct net *net, struct sk_buff *skb, u16 ident) { struct hlist_nulls_head *hslot = ping_hashslot(&ping_table, net, ident); struct sock *sk = NULL; struct inet_sock *isk; struct hlist_nulls_node *hnode; - int dif = skb->dev->ifindex; + int dif, sdif; if (skb->protocol == htons(ETH_P_IP)) { + dif = inet_iif(skb); + sdif = inet_sdif(skb); pr_debug("try to find: num = %d, daddr = %pI4, dif = %d\n", (int)ident, &ip_hdr(skb)->daddr, dif); #if IS_ENABLED(CONFIG_IPV6) } else if (skb->protocol == htons(ETH_P_IPV6)) { + dif = inet6_iif(skb); + sdif = inet6_sdif(skb); pr_debug("try to find: num = %d, daddr = %pI6c, dif = %d\n", (int)ident, &ipv6_hdr(skb)->daddr, dif); #endif + } else { + return NULL; } - read_lock_bh(&ping_table.lock); - ping_portaddr_for_each_entry(sk, hnode, hslot) { isk = inet_sk(sk); @@ -220,16 +226,15 @@ static struct sock *ping_lookup(struct net *net, struct sk_buff *skb, u16 ident) continue; } - if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif) + if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif && + sk->sk_bound_dev_if != sdif) continue; - sock_hold(sk); goto exit; } sk = NULL; exit: - read_unlock_bh(&ping_table.lock); return sk; } @@ -291,12 +296,27 @@ void ping_close(struct sock *sk, long timeout) } EXPORT_SYMBOL_GPL(ping_close); +static int ping_pre_connect(struct sock *sk, struct sockaddr *uaddr, + int addr_len) +{ + /* This check is replicated from __ip4_datagram_connect() and + * intended to prevent BPF program called below from accessing bytes + * that are out of the bound specified by user in addr_len. + */ + if (addr_len < sizeof(struct sockaddr_in)) + return -EINVAL; + + return BPF_CGROUP_RUN_PROG_INET4_CONNECT_LOCK(sk, uaddr); +} + /* Checks the bind address and possibly modifies sk->sk_bound_dev_if. */ static int ping_check_bind_addr(struct sock *sk, struct inet_sock *isk, - struct sockaddr *uaddr, int addr_len) { + struct sockaddr *uaddr, int addr_len) +{ struct net *net = sock_net(sk); if (sk->sk_family == AF_INET) { struct sockaddr_in *addr = (struct sockaddr_in *) uaddr; + u32 tb_id = RT_TABLE_LOCAL; int chk_addr_ret; if (addr_len < sizeof(*addr)) @@ -310,15 +330,16 @@ static int ping_check_bind_addr(struct sock *sk, struct inet_sock *isk, pr_debug("ping_check_bind_addr(sk=%p,addr=%pI4,port=%d)\n", sk, &addr->sin_addr.s_addr, ntohs(addr->sin_port)); - chk_addr_ret = inet_addr_type(net, addr->sin_addr.s_addr); - if (addr->sin_addr.s_addr == htonl(INADDR_ANY)) - chk_addr_ret = RTN_LOCAL; + return 0; + + tb_id = l3mdev_fib_table_by_index(net, sk->sk_bound_dev_if) ? : tb_id; + chk_addr_ret = inet_addr_type_table(net, addr->sin_addr.s_addr, tb_id); - if ((!inet_can_nonlocal_bind(net, isk) && - chk_addr_ret != RTN_LOCAL) || - chk_addr_ret == RTN_MULTICAST || - chk_addr_ret == RTN_BROADCAST) + if (chk_addr_ret == RTN_MULTICAST || + chk_addr_ret == RTN_BROADCAST || + (chk_addr_ret != RTN_LOCAL && + !inet_can_nonlocal_bind(net, isk))) return -EADDRNOTAVAIL; #if IS_ENABLED(CONFIG_IPV6) @@ -351,6 +372,14 @@ static int ping_check_bind_addr(struct sock *sk, struct inet_sock *isk, return -ENODEV; } } + + if (!dev && sk->sk_bound_dev_if) { + dev = dev_get_by_index_rcu(net, sk->sk_bound_dev_if); + if (!dev) { + rcu_read_unlock(); + return -ENODEV; + } + } has_addr = pingv6_ops.ipv6_chk_addr(net, &addr->sin6_addr, dev, scoped); rcu_read_unlock(); @@ -383,20 +412,6 @@ static void ping_set_saddr(struct sock *sk, struct sockaddr *saddr) } } -static void ping_clear_saddr(struct sock *sk, int dif) -{ - sk->sk_bound_dev_if = dif; - if (sk->sk_family == AF_INET) { - struct inet_sock *isk = inet_sk(sk); - isk->inet_rcv_saddr = isk->inet_saddr = 0; -#if IS_ENABLED(CONFIG_IPV6) - } else if (sk->sk_family == AF_INET6) { - struct ipv6_pinfo *np = inet6_sk(sk); - memset(&sk->sk_v6_rcv_saddr, 0, sizeof(sk->sk_v6_rcv_saddr)); - memset(&np->saddr, 0, sizeof(np->saddr)); -#endif - } -} /* * We need our own bind because there are no privileged id's == local ports. * Moreover, we don't allow binding to multi- and broadcast addresses. @@ -420,12 +435,13 @@ int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) goto out; err = -EADDRINUSE; - ping_set_saddr(sk, uaddr); snum = ntohs(((struct sockaddr_in *)uaddr)->sin_port); if (ping_get_port(sk, snum) != 0) { - ping_clear_saddr(sk, dif); + /* Restore possibly modified sk->sk_bound_dev_if by ping_check_bind_addr(). */ + sk->sk_bound_dev_if = dif; goto out; } + ping_set_saddr(sk, uaddr); pr_debug("after bind(): num = %hu, dif = %d\n", isk->inet_num, @@ -465,7 +481,9 @@ EXPORT_SYMBOL_GPL(ping_bind); static inline int ping_supported(int family, int type, int code) { return (family == AF_INET && type == ICMP_ECHO && code == 0) || - (family == AF_INET6 && type == ICMPV6_ECHO_REQUEST && code == 0); + (family == AF_INET && type == ICMP_EXT_ECHO && code == 0) || + (family == AF_INET6 && type == ICMPV6_ECHO_REQUEST && code == 0) || + (family == AF_INET6 && type == ICMPV6_EXT_ECHO_REQUEST && code == 0); } /* @@ -583,9 +601,9 @@ void ping_err(struct sk_buff *skb, int offset, u32 info) } } sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); out: - sock_put(sk); + return; } EXPORT_SYMBOL_GPL(ping_err); @@ -597,23 +615,11 @@ EXPORT_SYMBOL_GPL(ping_err); int ping_getfrag(void *from, char *to, int offset, int fraglen, int odd, struct sk_buff *skb) { - struct pingfakehdr *pfh = (struct pingfakehdr *)from; - - if (offset == 0) { - fraglen -= sizeof(struct icmphdr); - if (fraglen < 0) - BUG(); - if (!csum_and_copy_from_iter_full(to + sizeof(struct icmphdr), - fraglen, &pfh->wcheck, - &pfh->msg->msg_iter)) - return -EFAULT; - } else if (offset < sizeof(struct icmphdr)) { - BUG(); - } else { - if (!csum_and_copy_from_iter_full(to, fraglen, &pfh->wcheck, - &pfh->msg->msg_iter)) - return -EFAULT; - } + struct pingfakehdr *pfh = from; + + if (!csum_and_copy_from_iter_full(to, fraglen, &pfh->wcheck, + &pfh->msg->msg_iter)) + return -EFAULT; #if IS_ENABLED(CONFIG_IPV6) /* For IPv6, checksum each skb as we go along, as expected by @@ -621,7 +627,7 @@ int ping_getfrag(void *from, char *to, * wcheck, it will be finalized in ping_v4_push_pending_frames. */ if (pfh->family == AF_INET6) { - skb->csum = pfh->wcheck; + skb->csum = csum_block_add(skb->csum, pfh->wcheck, odd); skb->ip_summed = CHECKSUM_NONE; pfh->wcheck = 0; } @@ -647,7 +653,8 @@ static int ping_v4_push_pending_frames(struct sock *sk, struct pingfakehdr *pfh, } int ping_common_sendmsg(int family, struct msghdr *msg, size_t len, - void *user_icmph, size_t icmph_len) { + void *user_icmph, size_t icmph_len) +{ u8 type, code; if (len > 0xFFFF) @@ -786,7 +793,10 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) inet_sk_flowi_flags(sk), faddr, saddr, 0, 0, sk->sk_uid); - security_sk_classify_flow(sk, flowi4_to_flowi(&fl4)); + fl4.fl4_icmp_type = user_icmph.type; + fl4.fl4_icmp_code = user_icmph.code; + + security_sk_classify_flow(sk, flowi4_to_flowi_common(&fl4)); rt = ip_route_output_flow(net, &fl4, sk); if (IS_ERR(rt)) { err = PTR_ERR(rt); @@ -820,7 +830,8 @@ back_from_confirm: pfh.family = AF_INET; err = ip_append_data(sk, &fl4, ping_getfrag, &pfh, len, - 0, &ipc, &rt, msg->msg_flags); + sizeof(struct icmphdr), &ipc, &rt, + msg->msg_flags); if (err) ip_flush_pending_frames(sk); else @@ -847,8 +858,8 @@ do_confirm: goto out; } -int ping_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, - int flags, int *addr_len) +int ping_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags, + int *addr_len) { struct inet_sock *isk = inet_sk(sk); int family = sk->sk_family; @@ -864,7 +875,7 @@ int ping_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, if (flags & MSG_ERRQUEUE) return inet_recv_error(sk, msg, len, addr_len); - skb = skb_recv_datagram(sk, flags, noblock, &err); + skb = skb_recv_datagram(sk, flags, &err); if (!skb) goto out; @@ -937,16 +948,24 @@ out: } EXPORT_SYMBOL_GPL(ping_recvmsg); -int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) +static enum skb_drop_reason __ping_queue_rcv_skb(struct sock *sk, + struct sk_buff *skb) { + enum skb_drop_reason reason; + pr_debug("ping_queue_rcv_skb(sk=%p,sk->num=%d,skb=%p)\n", inet_sk(sk), inet_sk(sk)->inet_num, skb); - if (sock_queue_rcv_skb(sk, skb) < 0) { - kfree_skb(skb); + if (sock_queue_rcv_skb_reason(sk, skb, &reason) < 0) { + kfree_skb_reason(skb, reason); pr_debug("ping_queue_rcv_skb -> failed\n"); - return -1; + return reason; } - return 0; + return SKB_NOT_DROPPED_YET; +} + +int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) +{ + return __ping_queue_rcv_skb(sk, skb) ? -1 : 0; } EXPORT_SYMBOL_GPL(ping_queue_rcv_skb); @@ -955,8 +974,9 @@ EXPORT_SYMBOL_GPL(ping_queue_rcv_skb); * All we need to do is get the socket. */ -bool ping_rcv(struct sk_buff *skb) +enum skb_drop_reason ping_rcv(struct sk_buff *skb) { + enum skb_drop_reason reason = SKB_DROP_REASON_NO_SOCKET; struct sock *sk; struct net *net = dev_net(skb->dev); struct icmphdr *icmph = icmp_hdr(skb); @@ -975,13 +995,15 @@ bool ping_rcv(struct sk_buff *skb) pr_debug("rcv on socket %p\n", sk); if (skb2) - ping_queue_rcv_skb(sk, skb2); - sock_put(sk); - return true; + reason = __ping_queue_rcv_skb(sk, skb2); + else + reason = SKB_DROP_REASON_NOMEM; } - pr_debug("no socket, dropping\n"); - return false; + if (reason) + pr_debug("no socket, dropping\n"); + + return reason; } EXPORT_SYMBOL_GPL(ping_rcv); @@ -990,6 +1012,7 @@ struct proto ping_prot = { .owner = THIS_MODULE, .init = ping_init_sock, .close = ping_close, + .pre_connect = ping_pre_connect, .connect = ip4_datagram_connect, .disconnect = __udp_disconnect, .setsockopt = ip_setsockopt, @@ -1002,6 +1025,7 @@ struct proto ping_prot = { .hash = ping_hash, .unhash = ping_unhash, .get_port = ping_get_port, + .put_port = ping_unhash, .obj_size = sizeof(struct inet_sock), }; EXPORT_SYMBOL(ping_prot); @@ -1060,13 +1084,13 @@ static struct sock *ping_get_idx(struct seq_file *seq, loff_t pos) } void *ping_seq_start(struct seq_file *seq, loff_t *pos, sa_family_t family) - __acquires(ping_table.lock) + __acquires(RCU) { struct ping_iter_state *state = seq->private; state->bucket = 0; state->family = family; - read_lock_bh(&ping_table.lock); + rcu_read_lock(); return *pos ? ping_get_idx(seq, *pos-1) : SEQ_START_TOKEN; } @@ -1092,9 +1116,9 @@ void *ping_seq_next(struct seq_file *seq, void *v, loff_t *pos) EXPORT_SYMBOL_GPL(ping_seq_next); void ping_seq_stop(struct seq_file *seq, void *v) - __releases(ping_table.lock) + __releases(RCU) { - read_unlock_bh(&ping_table.lock); + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(ping_seq_stop); @@ -1178,5 +1202,5 @@ void __init ping_init(void) for (i = 0; i < PING_HTABLE_SIZE; i++) INIT_HLIST_NULLS_HEAD(&ping_table.hash[i], i); - rwlock_init(&ping_table.lock); + spin_lock_init(&ping_table.lock); } diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 2580303249e2..5386f460bd20 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -32,6 +32,7 @@ #include <net/icmp.h> #include <net/protocol.h> #include <net/tcp.h> +#include <net/mptcp.h> #include <net/udp.h> #include <net/udplite.h> #include <linux/bottom_half.h> @@ -52,14 +53,14 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) struct net *net = seq->private; int orphans, sockets; - orphans = percpu_counter_sum_positive(&tcp_orphan_count); + orphans = tcp_orphan_count_sum(); sockets = proto_sockets_allocated_sum_positive(&tcp_prot); socket_seq_show(seq); seq_printf(seq, "TCP: inuse %d orphan %d tw %d alloc %d mem %ld\n", sock_prot_inuse_get(net, &tcp_prot), orphans, - atomic_read(&net->ipv4.tcp_death_row.tw_count), sockets, - proto_memory_allocated(&tcp_prot)); + refcount_read(&net->ipv4.tcp_death_row.tw_refcount) - 1, + sockets, proto_memory_allocated(&tcp_prot)); seq_printf(seq, "UDP: inuse %d mem %ld\n", sock_prot_inuse_get(net, &udp_prot), proto_memory_allocated(&udp_prot)); @@ -166,6 +167,7 @@ static const struct snmp_mib snmp4_udp_list[] = { SNMP_MIB_ITEM("SndbufErrors", UDP_MIB_SNDBUFERRORS), SNMP_MIB_ITEM("InCsumErrors", UDP_MIB_CSUMERRORS), SNMP_MIB_ITEM("IgnoredMulti", UDP_MIB_IGNOREDMULTI), + SNMP_MIB_ITEM("MemErrors", UDP_MIB_MEMERRORS), SNMP_MIB_SENTINEL }; @@ -291,6 +293,10 @@ static const struct snmp_mib snmp4_net_list[] = { SNMP_MIB_ITEM("TCPFastOpenPassiveAltKey", LINUX_MIB_TCPFASTOPENPASSIVEALTKEY), SNMP_MIB_ITEM("TcpTimeoutRehash", LINUX_MIB_TCPTIMEOUTREHASH), SNMP_MIB_ITEM("TcpDuplicateDataRehash", LINUX_MIB_TCPDUPLICATEDATAREHASH), + SNMP_MIB_ITEM("TCPDSACKRecvSegs", LINUX_MIB_TCPDSACKRECVSEGS), + SNMP_MIB_ITEM("TCPDSACKIgnoredDubious", LINUX_MIB_TCPDSACKIGNOREDDUBIOUS), + SNMP_MIB_ITEM("TCPMigrateReqSuccess", LINUX_MIB_TCPMIGRATEREQSUCCESS), + SNMP_MIB_ITEM("TCPMigrateReqFailure", LINUX_MIB_TCPMIGRATEREQFAILURE), SNMP_MIB_SENTINEL }; @@ -381,7 +387,7 @@ static int snmp_seq_show_ipstats(struct seq_file *seq, void *v) seq_printf(seq, "\nIp: %d %d", IPV4_DEVCONF_ALL(net, FORWARDING) ? 1 : 2, - net->ipv4.sysctl_ip_default_ttl); + READ_ONCE(net->ipv4.sysctl_ip_default_ttl)); BUILD_BUG_ON(offsetof(struct ipstats_mib, mibs) != 0); snmp_get_cpu_field64_batch(buff64, snmp4_ipstats_list, @@ -460,31 +466,54 @@ static int snmp_seq_show(struct seq_file *seq, void *v) */ static int netstat_seq_show(struct seq_file *seq, void *v) { - int i; + const int ip_cnt = ARRAY_SIZE(snmp4_ipextstats_list) - 1; + const int tcp_cnt = ARRAY_SIZE(snmp4_net_list) - 1; struct net *net = seq->private; + unsigned long *buff; + int i; seq_puts(seq, "TcpExt:"); - for (i = 0; snmp4_net_list[i].name; i++) + for (i = 0; i < tcp_cnt; i++) seq_printf(seq, " %s", snmp4_net_list[i].name); seq_puts(seq, "\nTcpExt:"); - for (i = 0; snmp4_net_list[i].name; i++) - seq_printf(seq, " %lu", - snmp_fold_field(net->mib.net_statistics, - snmp4_net_list[i].entry)); - + buff = kzalloc(max(tcp_cnt * sizeof(long), ip_cnt * sizeof(u64)), + GFP_KERNEL); + if (buff) { + snmp_get_cpu_field_batch(buff, snmp4_net_list, + net->mib.net_statistics); + for (i = 0; i < tcp_cnt; i++) + seq_printf(seq, " %lu", buff[i]); + } else { + for (i = 0; i < tcp_cnt; i++) + seq_printf(seq, " %lu", + snmp_fold_field(net->mib.net_statistics, + snmp4_net_list[i].entry)); + } seq_puts(seq, "\nIpExt:"); - for (i = 0; snmp4_ipextstats_list[i].name; i++) + for (i = 0; i < ip_cnt; i++) seq_printf(seq, " %s", snmp4_ipextstats_list[i].name); seq_puts(seq, "\nIpExt:"); - for (i = 0; snmp4_ipextstats_list[i].name; i++) - seq_printf(seq, " %llu", - snmp_fold_field64(net->mib.ip_statistics, - snmp4_ipextstats_list[i].entry, - offsetof(struct ipstats_mib, syncp))); - + if (buff) { + u64 *buff64 = (u64 *)buff; + + memset(buff64, 0, ip_cnt * sizeof(u64)); + snmp_get_cpu_field64_batch(buff64, snmp4_ipextstats_list, + net->mib.ip_statistics, + offsetof(struct ipstats_mib, syncp)); + for (i = 0; i < ip_cnt; i++) + seq_printf(seq, " %llu", buff64[i]); + } else { + for (i = 0; i < ip_cnt; i++) + seq_printf(seq, " %llu", + snmp_fold_field64(net->mib.ip_statistics, + snmp4_ipextstats_list[i].entry, + offsetof(struct ipstats_mib, syncp))); + } + kfree(buff); seq_putc(seq, '\n'); + mptcp_seq_show(seq); return 0; } diff --git a/net/ipv4/protocol.c b/net/ipv4/protocol.c index 9a8c0892622b..6913979948d7 100644 --- a/net/ipv4/protocol.c +++ b/net/ipv4/protocol.c @@ -31,12 +31,6 @@ EXPORT_SYMBOL(inet_offloads); int inet_add_protocol(const struct net_protocol *prot, unsigned char protocol) { - if (!prot->netns_ok) { - pr_err("Protocol %u is not namespace aware, cannot register.\n", - protocol); - return -EINVAL; - } - return !cmpxchg((const struct net_protocol **)&inet_protos[protocol], NULL, prot) ? 0 : -1; } diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 3183413ebc6c..006c1f0ed8b4 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -85,22 +85,21 @@ struct raw_frag_vec { int hlen; }; -struct raw_hashinfo raw_v4_hashinfo = { - .lock = __RW_LOCK_UNLOCKED(raw_v4_hashinfo.lock), -}; +struct raw_hashinfo raw_v4_hashinfo; EXPORT_SYMBOL_GPL(raw_v4_hashinfo); int raw_hash_sk(struct sock *sk) { struct raw_hashinfo *h = sk->sk_prot->h.raw_hash; - struct hlist_head *head; + struct hlist_nulls_head *hlist; - head = &h->ht[inet_sk(sk)->inet_num & (RAW_HTABLE_SIZE - 1)]; + hlist = &h->ht[inet_sk(sk)->inet_num & (RAW_HTABLE_SIZE - 1)]; - write_lock_bh(&h->lock); - sk_add_node(sk, head); + spin_lock(&h->lock); + __sk_nulls_add_node_rcu(sk, hlist); + sock_set_flag(sk, SOCK_RCU_FREE); + spin_unlock(&h->lock); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); - write_unlock_bh(&h->lock); return 0; } @@ -110,31 +109,26 @@ void raw_unhash_sk(struct sock *sk) { struct raw_hashinfo *h = sk->sk_prot->h.raw_hash; - write_lock_bh(&h->lock); - if (sk_del_node_init(sk)) + spin_lock(&h->lock); + if (__sk_nulls_del_node_init_rcu(sk)) sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); - write_unlock_bh(&h->lock); + spin_unlock(&h->lock); } EXPORT_SYMBOL_GPL(raw_unhash_sk); -struct sock *__raw_v4_lookup(struct net *net, struct sock *sk, - unsigned short num, __be32 raddr, __be32 laddr, - int dif, int sdif) +bool raw_v4_match(struct net *net, struct sock *sk, unsigned short num, + __be32 raddr, __be32 laddr, int dif, int sdif) { - sk_for_each_from(sk) { - struct inet_sock *inet = inet_sk(sk); - - if (net_eq(sock_net(sk), net) && inet->inet_num == num && - !(inet->inet_daddr && inet->inet_daddr != raddr) && - !(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) && - raw_sk_bound_dev_eq(net, sk->sk_bound_dev_if, dif, sdif)) - goto found; /* gotcha */ - } - sk = NULL; -found: - return sk; + struct inet_sock *inet = inet_sk(sk); + + if (net_eq(sock_net(sk), net) && inet->inet_num == num && + !(inet->inet_daddr && inet->inet_daddr != raddr) && + !(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) && + raw_sk_bound_dev_eq(net, sk->sk_bound_dev_if, dif, sdif)) + return true; + return false; } -EXPORT_SYMBOL_GPL(__raw_v4_lookup); +EXPORT_SYMBOL_GPL(raw_v4_match); /* * 0 - deliver @@ -168,23 +162,20 @@ static int icmp_filter(const struct sock *sk, const struct sk_buff *skb) */ static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash) { + struct net *net = dev_net(skb->dev); + struct hlist_nulls_head *hlist; + struct hlist_nulls_node *hnode; int sdif = inet_sdif(skb); int dif = inet_iif(skb); - struct sock *sk; - struct hlist_head *head; int delivered = 0; - struct net *net; - - read_lock(&raw_v4_hashinfo.lock); - head = &raw_v4_hashinfo.ht[hash]; - if (hlist_empty(head)) - goto out; - - net = dev_net(skb->dev); - sk = __raw_v4_lookup(net, __sk_head(head), iph->protocol, - iph->saddr, iph->daddr, dif, sdif); + struct sock *sk; - while (sk) { + hlist = &raw_v4_hashinfo.ht[hash]; + rcu_read_lock(); + sk_nulls_for_each(sk, hnode, hlist) { + if (!raw_v4_match(net, sk, iph->protocol, + iph->saddr, iph->daddr, dif, sdif)) + continue; delivered = 1; if ((iph->protocol != IPPROTO_ICMP || !icmp_filter(sk, skb)) && ip_mc_sf_allow(sk, iph->daddr, iph->saddr, @@ -195,31 +186,16 @@ static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash) if (clone) raw_rcv(sk, clone); } - sk = __raw_v4_lookup(net, sk_next(sk), iph->protocol, - iph->saddr, iph->daddr, - dif, sdif); } -out: - read_unlock(&raw_v4_hashinfo.lock); + rcu_read_unlock(); return delivered; } int raw_local_deliver(struct sk_buff *skb, int protocol) { - int hash; - struct sock *raw_sk; - - hash = protocol & (RAW_HTABLE_SIZE - 1); - raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]); - - /* If there maybe a raw socket we must check - if not we - * don't care less - */ - if (raw_sk && !raw_v4_input(skb, ip_hdr(skb), hash)) - raw_sk = NULL; - - return raw_sk != NULL; + int hash = protocol & (RAW_HTABLE_SIZE - 1); + return raw_v4_input(skb, ip_hdr(skb), hash); } static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info) @@ -260,11 +236,12 @@ static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info) err = EHOSTUNREACH; if (code > NR_ICMP_UNREACH) break; - err = icmp_err_convert[code].errno; - harderr = icmp_err_convert[code].fatal; if (code == ICMP_FRAG_NEEDED) { harderr = inet->pmtudisc != IP_PMTUDISC_DONT; err = EMSGSIZE; + } else { + err = icmp_err_convert[code].errno; + harderr = icmp_err_convert[code].fatal; } } @@ -279,37 +256,33 @@ static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info) if (inet->recverr || harderr) { sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); } } void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info) { - int hash; - struct sock *raw_sk; + struct net *net = dev_net(skb->dev); + struct hlist_nulls_head *hlist; + struct hlist_nulls_node *hnode; + int dif = skb->dev->ifindex; + int sdif = inet_sdif(skb); const struct iphdr *iph; - struct net *net; + struct sock *sk; + int hash; hash = protocol & (RAW_HTABLE_SIZE - 1); + hlist = &raw_v4_hashinfo.ht[hash]; - read_lock(&raw_v4_hashinfo.lock); - raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]); - if (raw_sk) { - int dif = skb->dev->ifindex; - int sdif = inet_sdif(skb); - + rcu_read_lock(); + sk_nulls_for_each(sk, hnode, hlist) { iph = (const struct iphdr *)skb->data; - net = dev_net(skb->dev); - - while ((raw_sk = __raw_v4_lookup(net, raw_sk, protocol, - iph->daddr, iph->saddr, - dif, sdif)) != NULL) { - raw_err(raw_sk, skb, info); - raw_sk = sk_next(raw_sk); - iph = (const struct iphdr *)skb->data; - } + if (!raw_v4_match(net, sk, iph->protocol, + iph->daddr, iph->saddr, dif, sdif)) + continue; + raw_err(sk, skb, info); } - read_unlock(&raw_v4_hashinfo.lock); + rcu_read_unlock(); } static int raw_rcv_skb(struct sock *sk, struct sk_buff *skb) @@ -478,7 +451,7 @@ static int raw_getfrag(void *from, char *to, int offset, int len, int odd, skb->csum = csum_block_add( skb->csum, csum_partial_copy_nocheck(rfv->hdr.c + offset, - to, copy, 0), + to, copy), odd); odd = 0; @@ -610,7 +583,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } else if (!ipc.oif) { ipc.oif = inet->uc_index; } else if (ipv4_is_lbcast(daddr) && inet->uc_index) { - /* oif is set, packet is to local broadcast and + /* oif is set, packet is to local broadcast * and uc_index is set. oif is most likely set * by sk_bound_dev_if. If uc_index != oif check if the * oif is an L3 master and uc_index is an L3 slave. @@ -639,7 +612,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) goto done; } - security_sk_classify_flow(sk, flowi4_to_flowi(&fl4)); + security_sk_classify_flow(sk, flowi4_to_flowi_common(&fl4)); rt = ip_route_output_flow(net, &fl4, sk); if (IS_ERR(rt)) { err = PTR_ERR(rt); @@ -716,30 +689,34 @@ static int raw_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *addr = (struct sockaddr_in *) uaddr; + struct net *net = sock_net(sk); u32 tb_id = RT_TABLE_LOCAL; int ret = -EINVAL; int chk_addr_ret; + lock_sock(sk); if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_in)) goto out; if (sk->sk_bound_dev_if) - tb_id = l3mdev_fib_table_by_index(sock_net(sk), - sk->sk_bound_dev_if) ? : tb_id; + tb_id = l3mdev_fib_table_by_index(net, + sk->sk_bound_dev_if) ? : tb_id; - chk_addr_ret = inet_addr_type_table(sock_net(sk), addr->sin_addr.s_addr, - tb_id); + chk_addr_ret = inet_addr_type_table(net, addr->sin_addr.s_addr, tb_id); ret = -EADDRNOTAVAIL; - if (addr->sin_addr.s_addr && chk_addr_ret != RTN_LOCAL && - chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST) + if (!inet_addr_valid_or_nonlocal(net, inet, addr->sin_addr.s_addr, + chk_addr_ret)) goto out; + inet->inet_rcv_saddr = inet->inet_saddr = addr->sin_addr.s_addr; if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST) inet->inet_saddr = 0; /* Use device */ sk_dst_reset(sk); ret = 0; -out: return ret; +out: + release_sock(sk); + return ret; } /* @@ -748,7 +725,7 @@ out: return ret; */ static int raw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, - int noblock, int flags, int *addr_len) + int flags, int *addr_len) { struct inet_sock *inet = inet_sk(sk); size_t copied = 0; @@ -764,7 +741,7 @@ static int raw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, goto out; } - skb = skb_recv_datagram(sk, flags, noblock, &err); + skb = skb_recv_datagram(sk, flags, &err); if (!skb) goto out; @@ -778,7 +755,7 @@ static int raw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, if (err) goto done; - sock_recv_ts_and_drops(msg, sk, skb); + sock_recv_cmsgs(msg, sk, skb); /* Copy the address. */ if (sin) { @@ -809,11 +786,11 @@ static int raw_sk_init(struct sock *sk) return 0; } -static int raw_seticmpfilter(struct sock *sk, char __user *optval, int optlen) +static int raw_seticmpfilter(struct sock *sk, sockptr_t optval, int optlen) { if (optlen > sizeof(struct icmp_filter)) optlen = sizeof(struct icmp_filter); - if (copy_from_user(&raw_sk(sk)->filter, optval, optlen)) + if (copy_from_sockptr(&raw_sk(sk)->filter, optval, optlen)) return -EFAULT; return 0; } @@ -838,7 +815,7 @@ out: return ret; } static int do_raw_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) + sockptr_t optval, unsigned int optlen) { if (optname == ICMP_FILTER) { if (inet_sk(sk)->inet_num != IPPROTO_ICMP) @@ -850,23 +827,13 @@ static int do_raw_setsockopt(struct sock *sk, int level, int optname, } static int raw_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) + sockptr_t optval, unsigned int optlen) { if (level != SOL_RAW) return ip_setsockopt(sk, level, optname, optval, optlen); return do_raw_setsockopt(sk, level, optname, optval, optlen); } -#ifdef CONFIG_COMPAT -static int compat_raw_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) -{ - if (level != SOL_RAW) - return compat_ip_setsockopt(sk, level, optname, optval, optlen); - return do_raw_setsockopt(sk, level, optname, optval, optlen); -} -#endif - static int do_raw_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -887,16 +854,6 @@ static int raw_getsockopt(struct sock *sk, int level, int optname, return do_raw_getsockopt(sk, level, optname, optval, optlen); } -#ifdef CONFIG_COMPAT -static int compat_raw_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (level != SOL_RAW) - return compat_ip_getsockopt(sk, level, optname, optval, optlen); - return do_raw_getsockopt(sk, level, optname, optval, optlen); -} -#endif - static int raw_ioctl(struct sock *sk, int cmd, unsigned long arg) { switch (cmd) { @@ -948,7 +905,7 @@ int raw_abort(struct sock *sk, int err) lock_sock(sk); sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); __udp_disconnect(sk, 0); release_sock(sk); @@ -980,52 +937,47 @@ struct proto raw_prot = { .usersize = sizeof_field(struct raw_sock, filter), .h.raw_hash = &raw_v4_hashinfo, #ifdef CONFIG_COMPAT - .compat_setsockopt = compat_raw_setsockopt, - .compat_getsockopt = compat_raw_getsockopt, .compat_ioctl = compat_raw_ioctl, #endif .diag_destroy = raw_abort, }; #ifdef CONFIG_PROC_FS -static struct sock *raw_get_first(struct seq_file *seq) +static struct sock *raw_get_first(struct seq_file *seq, int bucket) { - struct sock *sk; - struct raw_hashinfo *h = PDE_DATA(file_inode(seq->file)); + struct raw_hashinfo *h = pde_data(file_inode(seq->file)); struct raw_iter_state *state = raw_seq_private(seq); + struct hlist_nulls_head *hlist; + struct hlist_nulls_node *hnode; + struct sock *sk; - for (state->bucket = 0; state->bucket < RAW_HTABLE_SIZE; + for (state->bucket = bucket; state->bucket < RAW_HTABLE_SIZE; ++state->bucket) { - sk_for_each(sk, &h->ht[state->bucket]) + hlist = &h->ht[state->bucket]; + sk_nulls_for_each(sk, hnode, hlist) { if (sock_net(sk) == seq_file_net(seq)) - goto found; + return sk; + } } - sk = NULL; -found: - return sk; + return NULL; } static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk) { - struct raw_hashinfo *h = PDE_DATA(file_inode(seq->file)); struct raw_iter_state *state = raw_seq_private(seq); do { - sk = sk_next(sk); -try_again: - ; + sk = sk_nulls_next(sk); } while (sk && sock_net(sk) != seq_file_net(seq)); - if (!sk && ++state->bucket < RAW_HTABLE_SIZE) { - sk = sk_head(&h->ht[state->bucket]); - goto try_again; - } + if (!sk) + return raw_get_first(seq, state->bucket + 1); return sk; } static struct sock *raw_get_idx(struct seq_file *seq, loff_t pos) { - struct sock *sk = raw_get_first(seq); + struct sock *sk = raw_get_first(seq, 0); if (sk) while (pos && (sk = raw_get_next(seq, sk)) != NULL) @@ -1034,10 +986,9 @@ static struct sock *raw_get_idx(struct seq_file *seq, loff_t pos) } void *raw_seq_start(struct seq_file *seq, loff_t *pos) + __acquires(RCU) { - struct raw_hashinfo *h = PDE_DATA(file_inode(seq->file)); - - read_lock(&h->lock); + rcu_read_lock(); return *pos ? raw_get_idx(seq, *pos - 1) : SEQ_START_TOKEN; } EXPORT_SYMBOL_GPL(raw_seq_start); @@ -1047,7 +998,7 @@ void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos) struct sock *sk; if (v == SEQ_START_TOKEN) - sk = raw_get_first(seq); + sk = raw_get_first(seq, 0); else sk = raw_get_next(seq, v); ++*pos; @@ -1056,10 +1007,9 @@ void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos) EXPORT_SYMBOL_GPL(raw_seq_next); void raw_seq_stop(struct seq_file *seq, void *v) + __releases(RCU) { - struct raw_hashinfo *h = PDE_DATA(file_inode(seq->file)); - - read_unlock(&h->lock); + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(raw_seq_stop); @@ -1121,6 +1071,7 @@ static __net_initdata struct pernet_operations raw_net_ops = { int __init raw_proc_init(void) { + return register_pernet_subsys(&raw_net_ops); } diff --git a/net/ipv4/raw_diag.c b/net/ipv4/raw_diag.c index a93e7d1e1251..999321834b94 100644 --- a/net/ipv4/raw_diag.c +++ b/net/ipv4/raw_diag.c @@ -34,68 +34,69 @@ raw_get_hashinfo(const struct inet_diag_req_v2 *r) * use helper to figure it out. */ -static struct sock *raw_lookup(struct net *net, struct sock *from, - const struct inet_diag_req_v2 *req) +static bool raw_lookup(struct net *net, struct sock *sk, + const struct inet_diag_req_v2 *req) { struct inet_diag_req_raw *r = (void *)req; - struct sock *sk = NULL; if (r->sdiag_family == AF_INET) - sk = __raw_v4_lookup(net, from, r->sdiag_raw_protocol, - r->id.idiag_dst[0], - r->id.idiag_src[0], - r->id.idiag_if, 0); + return raw_v4_match(net, sk, r->sdiag_raw_protocol, + r->id.idiag_dst[0], + r->id.idiag_src[0], + r->id.idiag_if, 0); #if IS_ENABLED(CONFIG_IPV6) else - sk = __raw_v6_lookup(net, from, r->sdiag_raw_protocol, - (const struct in6_addr *)r->id.idiag_src, - (const struct in6_addr *)r->id.idiag_dst, - r->id.idiag_if, 0); + return raw_v6_match(net, sk, r->sdiag_raw_protocol, + (const struct in6_addr *)r->id.idiag_src, + (const struct in6_addr *)r->id.idiag_dst, + r->id.idiag_if, 0); #endif - return sk; + return false; } static struct sock *raw_sock_get(struct net *net, const struct inet_diag_req_v2 *r) { struct raw_hashinfo *hashinfo = raw_get_hashinfo(r); - struct sock *sk = NULL, *s; + struct hlist_nulls_head *hlist; + struct hlist_nulls_node *hnode; + struct sock *sk; int slot; if (IS_ERR(hashinfo)) return ERR_CAST(hashinfo); - read_lock(&hashinfo->lock); + rcu_read_lock(); for (slot = 0; slot < RAW_HTABLE_SIZE; slot++) { - sk_for_each(s, &hashinfo->ht[slot]) { - sk = raw_lookup(net, s, r); - if (sk) { + hlist = &hashinfo->ht[slot]; + sk_nulls_for_each(sk, hnode, hlist) { + if (raw_lookup(net, sk, r)) { /* * Grab it and keep until we fill - * diag meaage to be reported, so + * diag message to be reported, so * caller should call sock_put then. - * We can do that because we're keeping - * hashinfo->lock here. */ - sock_hold(sk); - goto out_unlock; + if (refcount_inc_not_zero(&sk->sk_refcnt)) + goto out_unlock; } } } + sk = ERR_PTR(-ENOENT); out_unlock: - read_unlock(&hashinfo->lock); + rcu_read_unlock(); - return sk ? sk : ERR_PTR(-ENOENT); + return sk; } -static int raw_diag_dump_one(struct sk_buff *in_skb, - const struct nlmsghdr *nlh, +static int raw_diag_dump_one(struct netlink_callback *cb, const struct inet_diag_req_v2 *r) { - struct net *net = sock_net(in_skb->sk); + struct sk_buff *in_skb = cb->skb; struct sk_buff *rep; struct sock *sk; + struct net *net; int err; + net = sock_net(in_skb->sk); sk = raw_sock_get(net, r); if (IS_ERR(sk)) return PTR_ERR(sk); @@ -109,10 +110,7 @@ static int raw_diag_dump_one(struct sk_buff *in_skb, return -ENOMEM; } - err = inet_sk_diag_fill(sk, NULL, rep, r, - sk_user_ns(NETLINK_CB(in_skb).sk), - NETLINK_CB(in_skb).portid, - nlh->nlmsg_seq, 0, nlh, + err = inet_sk_diag_fill(sk, NULL, rep, cb, r, 0, netlink_net_capable(in_skb, CAP_NET_ADMIN)); sock_put(sk); @@ -121,11 +119,8 @@ static int raw_diag_dump_one(struct sk_buff *in_skb, return err; } - err = netlink_unicast(net->diag_nlsk, rep, - NETLINK_CB(in_skb).portid, - MSG_DONTWAIT); - if (err > 0) - err = 0; + err = nlmsg_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid); + return err; } @@ -137,33 +132,36 @@ static int sk_diag_dump(struct sock *sk, struct sk_buff *skb, if (!inet_diag_bc_sk(bc, sk)) return 0; - return inet_sk_diag_fill(sk, NULL, skb, r, - sk_user_ns(NETLINK_CB(cb->skb).sk), - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, NLM_F_MULTI, - cb->nlh, net_admin); + return inet_sk_diag_fill(sk, NULL, skb, cb, r, NLM_F_MULTI, net_admin); } static void raw_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, struct nlattr *bc) + const struct inet_diag_req_v2 *r) { bool net_admin = netlink_net_capable(cb->skb, CAP_NET_ADMIN); struct raw_hashinfo *hashinfo = raw_get_hashinfo(r); struct net *net = sock_net(skb->sk); + struct inet_diag_dump_data *cb_data; + struct hlist_nulls_head *hlist; + struct hlist_nulls_node *hnode; int num, s_num, slot, s_slot; struct sock *sk = NULL; + struct nlattr *bc; if (IS_ERR(hashinfo)) return; + cb_data = cb->data; + bc = cb_data->inet_diag_nla_bc; s_slot = cb->args[0]; num = s_num = cb->args[1]; - read_lock(&hashinfo->lock); + rcu_read_lock(); for (slot = s_slot; slot < RAW_HTABLE_SIZE; s_num = 0, slot++) { num = 0; - sk_for_each(sk, &hashinfo->ht[slot]) { + hlist = &hashinfo->ht[slot]; + sk_nulls_for_each(sk, hnode, hlist) { struct inet_sock *inet = inet_sk(sk); if (!net_eq(sock_net(sk), net)) @@ -186,7 +184,7 @@ next: } out_unlock: - read_unlock(&hashinfo->lock); + rcu_read_unlock(); cb->args[0] = slot; cb->args[1] = num; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index ebe7060d0fc9..cd1fa9f70f1a 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -21,7 +21,7 @@ * Alan Cox : Added BSD route gw semantics * Alan Cox : Super /proc >4K * Alan Cox : MTU in route table - * Alan Cox : MSS actually. Also added the window + * Alan Cox : MSS actually. Also added the window * clamper. * Sam Lantinga : Fixed route matching in rt_del() * Alan Cox : Routing cache support. @@ -41,7 +41,7 @@ * Olaf Erb : irtt wasn't being copied right. * Bjorn Ekwall : Kerneld route support. * Alan Cox : Multicast fixed (I hope) - * Pavel Krauz : Limited broadcast fixed + * Pavel Krauz : Limited broadcast fixed * Mike McLagan : Routing by source * Alexey Kuznetsov : End of old history. Split to fib.c and * route.c and rewritten from scratch. @@ -54,21 +54,18 @@ * Robert Olsson : Added rt_cache statistics * Arnaldo C. Melo : Convert proc stuff to seq_file * Eric Dumazet : hashed spinlocks and rt_check_expire() fixes. - * Ilia Sotnikov : Ignore TOS on PMTUD and Redirect - * Ilia Sotnikov : Removed TOS from hash calculations + * Ilia Sotnikov : Ignore TOS on PMTUD and Redirect + * Ilia Sotnikov : Removed TOS from hash calculations */ #define pr_fmt(fmt) "IPv4: " fmt #include <linux/module.h> -#include <linux/uaccess.h> #include <linux/bitops.h> -#include <linux/types.h> #include <linux/kernel.h> #include <linux/mm.h> -#include <linux/string.h> +#include <linux/memblock.h> #include <linux/socket.h> -#include <linux/sockios.h> #include <linux/errno.h> #include <linux/in.h> #include <linux/inet.h> @@ -83,20 +80,18 @@ #include <linux/netfilter_ipv4.h> #include <linux/random.h> #include <linux/rcupdate.h> -#include <linux/times.h> #include <linux/slab.h> #include <linux/jhash.h> #include <net/dst.h> #include <net/dst_metadata.h> +#include <net/inet_dscp.h> #include <net/net_namespace.h> -#include <net/protocol.h> #include <net/ip.h> #include <net/route.h> #include <net/inetpeer.h> #include <net/sock.h> #include <net/ip_fib.h> #include <net/nexthop.h> -#include <net/arp.h> #include <net/tcp.h> #include <net/icmp.h> #include <net/xfrm.h> @@ -108,7 +103,6 @@ #endif #include <net/secure_seq.h> #include <net/ip_tunnels.h> -#include <net/l3mdev.h> #include "fib_lookup.h" @@ -117,15 +111,15 @@ #define RT_GC_TIMEOUT (300*HZ) +#define DEFAULT_MIN_PMTU (512 + 20 + 20) +#define DEFAULT_MTU_EXPIRES (10 * 60 * HZ) +#define DEFAULT_MIN_ADVMSS 256 static int ip_rt_max_size; static int ip_rt_redirect_number __read_mostly = 9; static int ip_rt_redirect_load __read_mostly = HZ / 50; static int ip_rt_redirect_silence __read_mostly = ((HZ / 50) << (9 + 1)); static int ip_rt_error_cost __read_mostly = HZ; static int ip_rt_error_burst __read_mostly = 5 * HZ; -static int ip_rt_mtu_expires __read_mostly = 10 * 60 * HZ; -static u32 ip_rt_min_pmtu __read_mostly = 512 + 20 + 20; -static int ip_rt_min_advmss __read_mostly = 256; static int ip_rt_gc_timeout __read_mostly = RT_GC_TIMEOUT; @@ -133,9 +127,11 @@ static int ip_rt_gc_timeout __read_mostly = RT_GC_TIMEOUT; * Interface to generic destination cache. */ -static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie); +INDIRECT_CALLABLE_SCOPE +struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie); static unsigned int ipv4_default_advmss(const struct dst_entry *dst); -static unsigned int ipv4_mtu(const struct dst_entry *dst); +INDIRECT_CALLABLE_SCOPE +unsigned int ipv4_mtu(const struct dst_entry *dst); static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst); static void ipv4_link_failure(struct sk_buff *skb); static void ip_rt_update_pmtu(struct dst_entry *dst, struct sock *sk, @@ -232,19 +228,6 @@ static const struct seq_operations rt_cache_seq_ops = { .show = rt_cache_seq_show, }; -static int rt_cache_seq_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &rt_cache_seq_ops); -} - -static const struct proc_ops rt_cache_proc_ops = { - .proc_open = rt_cache_seq_open, - .proc_read = seq_read, - .proc_lseek = seq_lseek, - .proc_release = seq_release, -}; - - static void *rt_cpu_seq_start(struct seq_file *seq, loff_t *pos) { int cpu; @@ -286,12 +269,13 @@ static int rt_cpu_seq_show(struct seq_file *seq, void *v) struct rt_cache_stat *st = v; if (v == SEQ_START_TOKEN) { - seq_printf(seq, "entries in_hit in_slow_tot in_slow_mc in_no_route in_brd in_martian_dst in_martian_src out_hit out_slow_tot out_slow_mc gc_total gc_ignored gc_goal_miss gc_dst_overflow in_hlist_search out_hlist_search\n"); + seq_puts(seq, "entries in_hit in_slow_tot in_slow_mc in_no_route in_brd in_martian_dst in_martian_src out_hit out_slow_tot out_slow_mc gc_total gc_ignored gc_goal_miss gc_dst_overflow in_hlist_search out_hlist_search\n"); return 0; } - seq_printf(seq,"%08x %08x %08x %08x %08x %08x %08x %08x " - " %08x %08x %08x %08x %08x %08x %08x %08x %08x \n", + seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x " + "%08x %08x %08x %08x %08x %08x " + "%08x %08x %08x %08x\n", dst_entries_get_slow(&ipv4_dst_ops), 0, /* st->in_hit */ st->in_slow_tot, @@ -322,19 +306,6 @@ static const struct seq_operations rt_cpu_seq_ops = { .show = rt_cpu_seq_show, }; - -static int rt_cpu_seq_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &rt_cpu_seq_ops); -} - -static const struct proc_ops rt_cpu_proc_ops = { - .proc_open = rt_cpu_seq_open, - .proc_read = seq_read, - .proc_lseek = seq_lseek, - .proc_release = seq_release, -}; - #ifdef CONFIG_IP_ROUTE_CLASSID static int rt_acct_proc_show(struct seq_file *m, void *v) { @@ -365,13 +336,13 @@ static int __net_init ip_rt_do_proc_init(struct net *net) { struct proc_dir_entry *pde; - pde = proc_create("rt_cache", 0444, net->proc_net, - &rt_cache_proc_ops); + pde = proc_create_seq("rt_cache", 0444, net->proc_net, + &rt_cache_seq_ops); if (!pde) goto err1; - pde = proc_create("rt_cache", 0444, - net->proc_net_stat, &rt_cpu_proc_ops); + pde = proc_create_seq("rt_cache", 0444, net->proc_net_stat, + &rt_cpu_seq_ops); if (!pde) goto err2; @@ -476,8 +447,10 @@ static void ipv4_confirm_neigh(const struct dst_entry *dst, const void *daddr) __ipv4_confirm_neigh(dev, *(__force u32 *)pkey); } -#define IP_IDENTS_SZ 2048u - +/* Hash tables of size 2048..262144 depending on RAM size. + * Each bucket uses 8 bytes. + */ +static u32 ip_idents_mask __read_mostly; static atomic_t *ip_idents __read_mostly; static u32 *ip_tstamps __read_mostly; @@ -485,26 +458,27 @@ static u32 *ip_tstamps __read_mostly; * if one generator is seldom used. This makes hard for an attacker * to infer how many packets were sent between two points in time. */ -u32 ip_idents_reserve(u32 hash, int segs) +static u32 ip_idents_reserve(u32 hash, int segs) { - u32 *p_tstamp = ip_tstamps + hash % IP_IDENTS_SZ; - atomic_t *p_id = ip_idents + hash % IP_IDENTS_SZ; - u32 old = READ_ONCE(*p_tstamp); - u32 now = (u32)jiffies; - u32 new, delta = 0; + u32 bucket, old, now = (u32)jiffies; + atomic_t *p_id; + u32 *p_tstamp; + u32 delta = 0; + + bucket = hash & ip_idents_mask; + p_tstamp = ip_tstamps + bucket; + p_id = ip_idents + bucket; + old = READ_ONCE(*p_tstamp); if (old != now && cmpxchg(p_tstamp, old, now) == old) delta = prandom_u32_max(now - old); - /* Do not use atomic_add_return() as it makes UBSAN unhappy */ - do { - old = (u32)atomic_read(p_id); - new = old + delta + segs; - } while (atomic_cmpxchg(p_id, old, new) != old); - - return new - segs; + /* If UBSAN reports an error there, please make sure your compiler + * supports -fno-strict-overflow before reporting it that was a bug + * in UBSAN, and it has been fixed in GCC-8. + */ + return atomic_add_return(segs + delta, p_id) - segs; } -EXPORT_SYMBOL(ip_idents_reserve); void __ip_select_ident(struct net *net, struct iphdr *iph, int segs) { @@ -524,24 +498,34 @@ void __ip_select_ident(struct net *net, struct iphdr *iph, int segs) } EXPORT_SYMBOL(__ip_select_ident); +static void ip_rt_fix_tos(struct flowi4 *fl4) +{ + __u8 tos = RT_FL_TOS(fl4); + + fl4->flowi4_tos = tos & IPTOS_RT_MASK; + if (tos & RTO_ONLINK) + fl4->flowi4_scope = RT_SCOPE_LINK; +} + static void __build_flow_key(const struct net *net, struct flowi4 *fl4, - const struct sock *sk, - const struct iphdr *iph, - int oif, u8 tos, - u8 prot, u32 mark, int flow_flags) + const struct sock *sk, const struct iphdr *iph, + int oif, __u8 tos, u8 prot, u32 mark, + int flow_flags) { + __u8 scope = RT_SCOPE_UNIVERSE; + if (sk) { const struct inet_sock *inet = inet_sk(sk); oif = sk->sk_bound_dev_if; mark = sk->sk_mark; - tos = RT_CONN_FLAGS(sk); + tos = ip_sock_rt_tos(sk); + scope = ip_sock_rt_scope(sk); prot = inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol; } - flowi4_init_output(fl4, oif, mark, tos, - RT_SCOPE_UNIVERSE, prot, - flow_flags, - iph->daddr, iph->saddr, 0, 0, + + flowi4_init_output(fl4, oif, mark, tos & IPTOS_RT_MASK, scope, + prot, flow_flags, iph->daddr, iph->saddr, 0, 0, sock_net_uid(net, sk)); } @@ -551,9 +535,9 @@ static void build_skb_flow_key(struct flowi4 *fl4, const struct sk_buff *skb, const struct net *net = dev_net(skb->dev); const struct iphdr *iph = ip_hdr(skb); int oif = skb->dev->ifindex; - u8 tos = RT_TOS(iph->tos); u8 prot = iph->protocol; u32 mark = skb->mark; + __u8 tos = iph->tos; __build_flow_key(net, fl4, sk, iph, oif, tos, prot, mark, 0); } @@ -569,7 +553,8 @@ static void build_sk_flow_key(struct flowi4 *fl4, const struct sock *sk) if (inet_opt && inet_opt->opt.srr) daddr = inet_opt->opt.faddr; flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark, - RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, + ip_sock_rt_tos(sk) & IPTOS_RT_MASK, + ip_sock_rt_scope(sk), inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol, inet_sk_flowi_flags(sk), daddr, inet->inet_saddr, 0, 0, sk->sk_uid); @@ -605,28 +590,35 @@ static void fnhe_flush_routes(struct fib_nh_exception *fnhe) } } -static struct fib_nh_exception *fnhe_oldest(struct fnhe_hash_bucket *hash) +static void fnhe_remove_oldest(struct fnhe_hash_bucket *hash) { - struct fib_nh_exception *fnhe, *oldest; + struct fib_nh_exception __rcu **fnhe_p, **oldest_p; + struct fib_nh_exception *fnhe, *oldest = NULL; - oldest = rcu_dereference(hash->chain); - for (fnhe = rcu_dereference(oldest->fnhe_next); fnhe; - fnhe = rcu_dereference(fnhe->fnhe_next)) { - if (time_before(fnhe->fnhe_stamp, oldest->fnhe_stamp)) + for (fnhe_p = &hash->chain; ; fnhe_p = &fnhe->fnhe_next) { + fnhe = rcu_dereference_protected(*fnhe_p, + lockdep_is_held(&fnhe_lock)); + if (!fnhe) + break; + if (!oldest || + time_before(fnhe->fnhe_stamp, oldest->fnhe_stamp)) { oldest = fnhe; + oldest_p = fnhe_p; + } } fnhe_flush_routes(oldest); - return oldest; + *oldest_p = oldest->fnhe_next; + kfree_rcu(oldest, rcu); } -static inline u32 fnhe_hashfun(__be32 daddr) +static u32 fnhe_hashfun(__be32 daddr) { - static u32 fnhe_hashrnd __read_mostly; - u32 hval; + static siphash_aligned_key_t fnhe_hash_key; + u64 hval; - net_get_random_once(&fnhe_hashrnd, sizeof(fnhe_hashrnd)); - hval = jhash_1word((__force u32) daddr, fnhe_hashrnd); - return hash_32(hval, FNHE_HASH_SHIFT); + net_get_random_once(&fnhe_hash_key, sizeof(fnhe_hash_key)); + hval = siphash_1u32((__force u32)daddr, &fnhe_hash_key); + return hash_64(hval, FNHE_HASH_SHIFT); } static void fill_route_from_fnhe(struct rtable *rt, struct fib_nh_exception *fnhe) @@ -695,16 +687,21 @@ static void update_or_create_fnhe(struct fib_nh_common *nhc, __be32 daddr, if (rt) fill_route_from_fnhe(rt, fnhe); } else { - if (depth > FNHE_RECLAIM_DEPTH) - fnhe = fnhe_oldest(hash); - else { - fnhe = kzalloc(sizeof(*fnhe), GFP_ATOMIC); - if (!fnhe) - goto out_unlock; - - fnhe->fnhe_next = hash->chain; - rcu_assign_pointer(hash->chain, fnhe); + /* Randomize max depth to avoid some side channels attacks. */ + int max_depth = FNHE_RECLAIM_DEPTH + + prandom_u32_max(FNHE_RECLAIM_DEPTH); + + while (depth > max_depth) { + fnhe_remove_oldest(hash); + depth--; } + + fnhe = kzalloc(sizeof(*fnhe), GFP_ATOMIC); + if (!fnhe) + goto out_unlock; + + fnhe->fnhe_next = hash->chain; + fnhe->fnhe_genid = genid; fnhe->fnhe_daddr = daddr; fnhe->fnhe_gw = gw; @@ -712,6 +709,8 @@ static void update_or_create_fnhe(struct fib_nh_common *nhc, __be32 daddr, fnhe->fnhe_mtu_locked = lock; fnhe->fnhe_expires = max(1UL, expires); + rcu_assign_pointer(hash->chain, fnhe); + /* Exception created; mark the cached routes for the nexthop * stale, so anyone caching it rechecks if this exception * applies to them. @@ -722,6 +721,7 @@ static void update_or_create_fnhe(struct fib_nh_common *nhc, __be32 daddr, for_each_possible_cpu(i) { struct rtable __rcu **prt; + prt = per_cpu_ptr(nhc->nhc_pcpu_rth_output, i); rt = rcu_dereference(*prt); if (rt) @@ -788,8 +788,10 @@ static void __ip_do_redirect(struct rtable *rt, struct sk_buff *skb, struct flow neigh_event_send(n, NULL); } else { if (fib_lookup(net, fl4, &res, 0) == 0) { - struct fib_nh_common *nhc = FIB_RES_NHC(res); + struct fib_nh_common *nhc; + fib_select_path(net, &res, fl4, skb); + nhc = FIB_RES_NHC(res); update_or_create_fnhe(nhc, fl4->daddr, new_gw, 0, false, jiffies + ip_rt_gc_timeout); @@ -825,9 +827,9 @@ static void ip_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buf const struct iphdr *iph = (const struct iphdr *) skb->data; struct net *net = dev_net(skb->dev); int oif = skb->dev->ifindex; - u8 tos = RT_TOS(iph->tos); u8 prot = iph->protocol; u32 mark = skb->mark; + __u8 tos = iph->tos; rt = (struct rtable *) dst; @@ -915,7 +917,7 @@ void ip_rt_send_redirect(struct sk_buff *skb) /* Check for load limit; set rate_last to the latest sent * redirect. */ - if (peer->rate_tokens == 0 || + if (peer->n_redirects == 0 || time_after(jiffies, (peer->rate_last + (ip_rt_redirect_load << peer->n_redirects)))) { @@ -944,6 +946,7 @@ static int ip_error(struct sk_buff *skb) struct inet_peer *peer; unsigned long now; struct net *net; + SKB_DR(reason); bool send; int code; @@ -963,10 +966,12 @@ static int ip_error(struct sk_buff *skb) if (!IN_DEV_FORWARD(in_dev)) { switch (rt->dst.error) { case EHOSTUNREACH: + SKB_DR_SET(reason, IP_INADDRERRORS); __IP_INC_STATS(net, IPSTATS_MIB_INADDRERRORS); break; case ENETUNREACH: + SKB_DR_SET(reason, IP_INNOROUTES); __IP_INC_STATS(net, IPSTATS_MIB_INNOROUTES); break; } @@ -982,6 +987,7 @@ static int ip_error(struct sk_buff *skb) break; case ENETUNREACH: code = ICMP_NET_UNREACH; + SKB_DR_SET(reason, IP_INNOROUTES); __IP_INC_STATS(net, IPSTATS_MIB_INNOROUTES); break; case EACCES: @@ -1008,38 +1014,42 @@ static int ip_error(struct sk_buff *skb) if (send) icmp_send(skb, ICMP_DEST_UNREACH, code, 0); -out: kfree_skb(skb); +out: kfree_skb_reason(skb, reason); return 0; } static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) { struct dst_entry *dst = &rt->dst; - u32 old_mtu = ipv4_mtu(dst); + struct net *net = dev_net(dst->dev); struct fib_result res; bool lock = false; + u32 old_mtu; if (ip_mtu_locked(dst)) return; + old_mtu = ipv4_mtu(dst); if (old_mtu < mtu) return; - if (mtu < ip_rt_min_pmtu) { + if (mtu < net->ipv4.ip_rt_min_pmtu) { lock = true; - mtu = min(old_mtu, ip_rt_min_pmtu); + mtu = min(old_mtu, net->ipv4.ip_rt_min_pmtu); } if (rt->rt_pmtu == mtu && !lock && - time_before(jiffies, dst->expires - ip_rt_mtu_expires / 2)) + time_before(jiffies, dst->expires - net->ipv4.ip_rt_mtu_expires / 2)) return; rcu_read_lock(); - if (fib_lookup(dev_net(dst->dev), fl4, &res, 0) == 0) { - struct fib_nh_common *nhc = FIB_RES_NHC(res); + if (fib_lookup(net, fl4, &res, 0) == 0) { + struct fib_nh_common *nhc; + fib_select_path(net, &res, fl4, NULL); + nhc = FIB_RES_NHC(res); update_or_create_fnhe(nhc, fl4->daddr, 0, mtu, lock, - jiffies + ip_rt_mtu_expires); + jiffies + net->ipv4.ip_rt_mtu_expires); } rcu_read_unlock(); } @@ -1052,19 +1062,24 @@ static void ip_rt_update_pmtu(struct dst_entry *dst, struct sock *sk, struct flowi4 fl4; ip_rt_build_flow_key(&fl4, sk, skb); + + /* Don't make lookup fail for bridged encapsulations */ + if (skb && netif_is_any_bridge_port(skb->dev)) + fl4.flowi4_oif = 0; + __ip_rt_update_pmtu(rt, &fl4, mtu); } void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu, int oif, u8 protocol) { - const struct iphdr *iph = (const struct iphdr *) skb->data; + const struct iphdr *iph = (const struct iphdr *)skb->data; struct flowi4 fl4; struct rtable *rt; u32 mark = IP4_REPLY_MARK(net, skb->mark); - __build_flow_key(net, &fl4, NULL, iph, oif, - RT_TOS(iph->tos), protocol, mark, 0); + __build_flow_key(net, &fl4, NULL, iph, oif, iph->tos, protocol, mark, + 0); rt = __ip_route_output_key(net, &fl4); if (!IS_ERR(rt)) { __ip_rt_update_pmtu(rt, &fl4, mtu); @@ -1075,7 +1090,7 @@ EXPORT_SYMBOL_GPL(ipv4_update_pmtu); static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) { - const struct iphdr *iph = (const struct iphdr *) skb->data; + const struct iphdr *iph = (const struct iphdr *)skb->data; struct flowi4 fl4; struct rtable *rt; @@ -1093,7 +1108,7 @@ static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) { - const struct iphdr *iph = (const struct iphdr *) skb->data; + const struct iphdr *iph = (const struct iphdr *)skb->data; struct flowi4 fl4; struct rtable *rt; struct dst_entry *odst = NULL; @@ -1123,7 +1138,7 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) new = true; } - __ip_rt_update_pmtu((struct rtable *) xfrm_dst_path(&rt->dst), &fl4, mtu); + __ip_rt_update_pmtu((struct rtable *)xfrm_dst_path(&rt->dst), &fl4, mtu); if (!dst_check(&rt->dst, 0)) { if (new) @@ -1148,12 +1163,11 @@ EXPORT_SYMBOL_GPL(ipv4_sk_update_pmtu); void ipv4_redirect(struct sk_buff *skb, struct net *net, int oif, u8 protocol) { - const struct iphdr *iph = (const struct iphdr *) skb->data; + const struct iphdr *iph = (const struct iphdr *)skb->data; struct flowi4 fl4; struct rtable *rt; - __build_flow_key(net, &fl4, NULL, iph, oif, - RT_TOS(iph->tos), protocol, 0, 0); + __build_flow_key(net, &fl4, NULL, iph, oif, iph->tos, protocol, 0, 0); rt = __ip_route_output_key(net, &fl4); if (!IS_ERR(rt)) { __ip_do_redirect(rt, skb, &fl4, false); @@ -1164,7 +1178,7 @@ EXPORT_SYMBOL_GPL(ipv4_redirect); void ipv4_sk_redirect(struct sk_buff *skb, struct sock *sk) { - const struct iphdr *iph = (const struct iphdr *) skb->data; + const struct iphdr *iph = (const struct iphdr *)skb->data; struct flowi4 fl4; struct rtable *rt; struct net *net = sock_net(sk); @@ -1178,7 +1192,8 @@ void ipv4_sk_redirect(struct sk_buff *skb, struct sock *sk) } EXPORT_SYMBOL_GPL(ipv4_sk_redirect); -static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie) +INDIRECT_CALLABLE_SCOPE struct dst_entry *ipv4_dst_check(struct dst_entry *dst, + u32 cookie) { struct rtable *rt = (struct rtable *) dst; @@ -1194,6 +1209,7 @@ static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie) return NULL; return dst; } +EXPORT_INDIRECT_CALLABLE(ipv4_dst_check); static void ipv4_send_dest_unreach(struct sk_buff *skb) { @@ -1245,12 +1261,12 @@ static int ip_rt_bug(struct net *net, struct sock *sk, struct sk_buff *skb) } /* - We do not cache source address of outgoing interface, - because it is used only by IP RR, TS and SRR options, - so that it out of fast path. - - BTW remember: "addr" is allowed to be not aligned - in IP options! + * We do not cache source address of outgoing interface, + * because it is used only by IP RR, TS and SRR options, + * so that it out of fast path. + * + * BTW remember: "addr" is allowed to be not aligned + * in IP options! */ void ip_rt_get_source(u8 *addr, struct sk_buff *skb, struct rtable *rt) @@ -1295,35 +1311,19 @@ static void set_class_tag(struct rtable *rt, u32 tag) static unsigned int ipv4_default_advmss(const struct dst_entry *dst) { + struct net *net = dev_net(dst->dev); unsigned int header_size = sizeof(struct tcphdr) + sizeof(struct iphdr); unsigned int advmss = max_t(unsigned int, ipv4_mtu(dst) - header_size, - ip_rt_min_advmss); + net->ipv4.ip_rt_min_advmss); return min(advmss, IPV4_MAX_PMTU - header_size); } -static unsigned int ipv4_mtu(const struct dst_entry *dst) +INDIRECT_CALLABLE_SCOPE unsigned int ipv4_mtu(const struct dst_entry *dst) { - const struct rtable *rt = (const struct rtable *) dst; - unsigned int mtu = rt->rt_pmtu; - - if (!mtu || time_after_eq(jiffies, rt->dst.expires)) - mtu = dst_metric_raw(dst, RTAX_MTU); - - if (mtu) - return mtu; - - mtu = READ_ONCE(dst->dev->mtu); - - if (unlikely(ip_mtu_locked(dst))) { - if (rt->rt_uses_gateway && mtu > 576) - mtu = 576; - } - - mtu = min_t(unsigned int, mtu, IP_MAX_MTU); - - return mtu - lwtunnel_headroom(dst->lwtstate, mtu); + return ip_dst_mtu_maybe_forward(dst, false); } +EXPORT_INDIRECT_CALLABLE(ipv4_mtu); static void ip_del_fnhe(struct fib_nh_common *nhc, __be32 daddr) { @@ -1398,7 +1398,7 @@ u32 ip_mtu_from_fib_result(struct fib_result *res, __be32 daddr) struct fib_info *fi = res->fi; u32 mtu = 0; - if (dev_net(dev)->ipv4.sysctl_ip_fwd_use_pmtu || + if (READ_ONCE(dev_net(dev)->ipv4.sysctl_ip_fwd_use_pmtu) || fi->fib_metrics->metrics[RTAX_LOCK - 1] & (1 << RTAX_MTU)) mtu = fi->fib_mtu; @@ -1499,6 +1499,7 @@ static bool rt_cache_route(struct fib_nh_common *nhc, struct rtable *rt) struct uncached_list { spinlock_t lock; struct list_head head; + struct list_head quarantine; }; static DEFINE_PER_CPU_ALIGNED(struct uncached_list, rt_uncached_list); @@ -1520,7 +1521,7 @@ void rt_del_uncached_list(struct rtable *rt) struct uncached_list *ul = rt->rt_uncached_list; spin_lock_bh(&ul->lock); - list_del(&rt->rt_uncached); + list_del_init(&rt->rt_uncached); spin_unlock_bh(&ul->lock); } } @@ -1535,19 +1536,23 @@ static void ipv4_dst_destroy(struct dst_entry *dst) void rt_flush_dev(struct net_device *dev) { - struct rtable *rt; + struct rtable *rt, *safe; int cpu; for_each_possible_cpu(cpu) { struct uncached_list *ul = &per_cpu(rt_uncached_list, cpu); + if (list_empty(&ul->head)) + continue; + spin_lock_bh(&ul->lock); - list_for_each_entry(rt, &ul->head, rt_uncached) { + list_for_each_entry_safe(rt, safe, &ul->head, rt_uncached) { if (rt->dst.dev != dev) continue; rt->dst.dev = blackhole_netdev; - dev_hold(rt->dst.dev); - dev_put(dev); + netdev_ref_replace(dev, blackhole_netdev, + &rt->dst.dev_tracker, GFP_ATOMIC); + list_move(&rt->rt_uncached, &ul->quarantine); } spin_unlock_bh(&ul->lock); } @@ -1621,13 +1626,11 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr, struct rtable *rt_dst_alloc(struct net_device *dev, unsigned int flags, u16 type, - bool nopolicy, bool noxfrm, bool will_cache) + bool noxfrm) { struct rtable *rt; rt = dst_alloc(&ipv4_dst_ops, dev, 1, DST_OBSOLETE_FORCE_CHK, - (will_cache ? 0 : DST_HOST) | - (nopolicy ? DST_NOPOLICY : 0) | (noxfrm ? DST_NOXFRM : 0)); if (rt) { @@ -1674,7 +1677,6 @@ struct rtable *rt_dst_clone(struct net_device *dev, struct rtable *rt) new_rt->rt_gw6 = rt->rt_gw6; INIT_LIST_HEAD(&new_rt->rt_uncached); - new_rt->dst.flags |= DST_HOST; new_rt->dst.input = rt->dst.input; new_rt->dst.output = rt->dst.output; new_rt->dst.error = rt->dst.error; @@ -1733,8 +1735,11 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, if (our) flags |= RTCF_LOCAL; + if (IN_DEV_ORCONF(in_dev, NOPOLICY)) + IPCB(skb)->flags |= IPSKB_NOPOLICY; + rth = rt_dst_alloc(dev_net(dev)->loopback_dev, flags, RTN_MULTICAST, - IN_DEV_CONF_GET(in_dev, NOPOLICY), false, false); + false); if (!rth) return -ENOBUFS; @@ -1750,6 +1755,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, #endif RT_CACHE_STAT_INC(in_slow_mc); + skb_dst_drop(skb); skb_dst_set(skb, &rth->dst); return 0; } @@ -1837,6 +1843,9 @@ static int __mkroute_input(struct sk_buff *skb, } } + if (IN_DEV_ORCONF(in_dev, NOPOLICY)) + IPCB(skb)->flags |= IPSKB_NOPOLICY; + fnhe = find_exception(nhc, daddr); if (do_cache) { if (fnhe) @@ -1850,8 +1859,7 @@ static int __mkroute_input(struct sk_buff *skb, } rth = rt_dst_alloc(out_dev->dev, 0, res->type, - IN_DEV_CONF_GET(in_dev, NOPOLICY), - IN_DEV_CONF_GET(out_dev, NOXFRM), do_cache); + IN_DEV_ORCONF(out_dev, NOXFRM)); if (!rth) { err = -ENOBUFS; goto cleanup; @@ -1912,15 +1920,130 @@ out: hash_keys->addrs.v4addrs.dst = key_iph->daddr; } +static u32 fib_multipath_custom_hash_outer(const struct net *net, + const struct sk_buff *skb, + bool *p_has_inner) +{ + u32 hash_fields = READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_fields); + struct flow_keys keys, hash_keys; + + if (!(hash_fields & FIB_MULTIPATH_HASH_FIELD_OUTER_MASK)) + return 0; + + memset(&hash_keys, 0, sizeof(hash_keys)); + skb_flow_dissect_flow_keys(skb, &keys, FLOW_DISSECTOR_F_STOP_AT_ENCAP); + + hash_keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_SRC_IP) + hash_keys.addrs.v4addrs.src = keys.addrs.v4addrs.src; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_DST_IP) + hash_keys.addrs.v4addrs.dst = keys.addrs.v4addrs.dst; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_IP_PROTO) + hash_keys.basic.ip_proto = keys.basic.ip_proto; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_SRC_PORT) + hash_keys.ports.src = keys.ports.src; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_DST_PORT) + hash_keys.ports.dst = keys.ports.dst; + + *p_has_inner = !!(keys.control.flags & FLOW_DIS_ENCAPSULATION); + return flow_hash_from_keys(&hash_keys); +} + +static u32 fib_multipath_custom_hash_inner(const struct net *net, + const struct sk_buff *skb, + bool has_inner) +{ + u32 hash_fields = READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_fields); + struct flow_keys keys, hash_keys; + + /* We assume the packet carries an encapsulation, but if none was + * encountered during dissection of the outer flow, then there is no + * point in calling the flow dissector again. + */ + if (!has_inner) + return 0; + + if (!(hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_MASK)) + return 0; + + memset(&hash_keys, 0, sizeof(hash_keys)); + skb_flow_dissect_flow_keys(skb, &keys, 0); + + if (!(keys.control.flags & FLOW_DIS_ENCAPSULATION)) + return 0; + + if (keys.control.addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) { + hash_keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_SRC_IP) + hash_keys.addrs.v4addrs.src = keys.addrs.v4addrs.src; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_DST_IP) + hash_keys.addrs.v4addrs.dst = keys.addrs.v4addrs.dst; + } else if (keys.control.addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { + hash_keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_SRC_IP) + hash_keys.addrs.v6addrs.src = keys.addrs.v6addrs.src; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_DST_IP) + hash_keys.addrs.v6addrs.dst = keys.addrs.v6addrs.dst; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_FLOWLABEL) + hash_keys.tags.flow_label = keys.tags.flow_label; + } + + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_IP_PROTO) + hash_keys.basic.ip_proto = keys.basic.ip_proto; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_SRC_PORT) + hash_keys.ports.src = keys.ports.src; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_DST_PORT) + hash_keys.ports.dst = keys.ports.dst; + + return flow_hash_from_keys(&hash_keys); +} + +static u32 fib_multipath_custom_hash_skb(const struct net *net, + const struct sk_buff *skb) +{ + u32 mhash, mhash_inner; + bool has_inner = true; + + mhash = fib_multipath_custom_hash_outer(net, skb, &has_inner); + mhash_inner = fib_multipath_custom_hash_inner(net, skb, has_inner); + + return jhash_2words(mhash, mhash_inner, 0); +} + +static u32 fib_multipath_custom_hash_fl4(const struct net *net, + const struct flowi4 *fl4) +{ + u32 hash_fields = READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_fields); + struct flow_keys hash_keys; + + if (!(hash_fields & FIB_MULTIPATH_HASH_FIELD_OUTER_MASK)) + return 0; + + memset(&hash_keys, 0, sizeof(hash_keys)); + hash_keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_SRC_IP) + hash_keys.addrs.v4addrs.src = fl4->saddr; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_DST_IP) + hash_keys.addrs.v4addrs.dst = fl4->daddr; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_IP_PROTO) + hash_keys.basic.ip_proto = fl4->flowi4_proto; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_SRC_PORT) + hash_keys.ports.src = fl4->fl4_sport; + if (hash_fields & FIB_MULTIPATH_HASH_FIELD_DST_PORT) + hash_keys.ports.dst = fl4->fl4_dport; + + return flow_hash_from_keys(&hash_keys); +} + /* if skb is set it will be used and fl4 can be NULL */ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4, const struct sk_buff *skb, struct flow_keys *flkeys) { u32 multipath_hash = fl4 ? fl4->flowi4_multipath_hash : 0; struct flow_keys hash_keys; - u32 mhash; + u32 mhash = 0; - switch (net->ipv4.sysctl_fib_multipath_hash_policy) { + switch (READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_policy)) { case 0: memset(&hash_keys, 0, sizeof(hash_keys)); hash_keys.control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; @@ -1930,6 +2053,7 @@ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4, hash_keys.addrs.v4addrs.src = fl4->saddr; hash_keys.addrs.v4addrs.dst = fl4->daddr; } + mhash = flow_hash_from_keys(&hash_keys); break; case 1: /* skb is currently provided only when forwarding */ @@ -1963,6 +2087,7 @@ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4, hash_keys.ports.dst = fl4->fl4_dport; hash_keys.basic.ip_proto = fl4->flowi4_proto; } + mhash = flow_hash_from_keys(&hash_keys); break; case 2: memset(&hash_keys, 0, sizeof(hash_keys)); @@ -1993,9 +2118,15 @@ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4, hash_keys.addrs.v4addrs.src = fl4->saddr; hash_keys.addrs.v4addrs.dst = fl4->daddr; } + mhash = flow_hash_from_keys(&hash_keys); + break; + case 3: + if (skb) + mhash = fib_multipath_custom_hash_skb(net, skb); + else + mhash = fib_multipath_custom_hash_fl4(net, fl4); break; } - mhash = flow_hash_from_keys(&hash_keys); if (multipath_hash) mhash = jhash_2words(mhash, multipath_hash, 0); @@ -2031,7 +2162,7 @@ int ip_route_use_hint(struct sk_buff *skb, __be32 daddr, __be32 saddr, const struct sk_buff *hint) { struct in_device *in_dev = __in_dev_get_rcu(dev); - struct rtable *rt = (struct rtable *)hint; + struct rtable *rt = skb_rtable(hint); struct net *net = dev_net(dev); int err = -EINVAL; u32 tag = 0; @@ -2062,6 +2193,19 @@ martian_source: return err; } +/* get device for dst_alloc with local routes */ +static struct net_device *ip_rt_get_dev(struct net *net, + const struct fib_result *res) +{ + struct fib_nh_common *nhc = res->fi ? res->nhc : NULL; + struct net_device *dev = NULL; + + if (nhc) + dev = l3mdev_master_dev_rcu(nhc->nhc_dev); + + return dev ? : net->loopback_dev; +} + /* * NOTE. We drop all the packets that has local source * addresses, because every properly looped back packet @@ -2096,7 +2240,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, goto out; /* Check for the most weird martians, which can be not detected - by fib_lookup. + * by fib_lookup. */ tun_info = skb_tunnel_info(skb); @@ -2137,6 +2281,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, /* * Now we are ready to route packet. */ + fl4.flowi4_l3mdev = 0; fl4.flowi4_oif = 0; fl4.flowi4_iif = dev->ifindex; fl4.flowi4_mark = skb->mark; @@ -2146,6 +2291,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, fl4.daddr = daddr; fl4.saddr = saddr; fl4.flowi4_uid = sock_net_uid(net, NULL); + fl4.flowi4_multipath_hash = 0; if (fib4_rules_early_flow_dissect(net, skb, &fl4, &_flkeys)) { flkeys = &_flkeys; @@ -2205,6 +2351,9 @@ brd_input: RT_CACHE_STAT_INC(in_brd); local_input: + if (IN_DEV_ORCONF(in_dev, NOPOLICY)) + IPCB(skb)->flags |= IPSKB_NOPOLICY; + do_cache &= res->fi && !itag; if (do_cache) { struct fib_nh_common *nhc = FIB_RES_NHC(*res); @@ -2217,9 +2366,8 @@ local_input: } } - rth = rt_dst_alloc(l3mdev_master_dev_rcu(dev) ? : net->loopback_dev, - flags | RTCF_LOCAL, res->type, - IN_DEV_CONF_GET(in_dev, NOPOLICY), false, do_cache); + rth = rt_dst_alloc(ip_rt_get_dev(net, res), + flags | RTCF_LOCAL, res->type, false); if (!rth) goto e_nobufs; @@ -2233,7 +2381,7 @@ local_input: if (res->type == RTN_UNREACHABLE) { rth->dst.input= ip_error; rth->dst.error= -err; - rth->rt_flags &= ~RTCF_LOCAL; + rth->rt_flags &= ~RTCF_LOCAL; } if (do_cache) { @@ -2284,35 +2432,20 @@ martian_source: goto out; } -int ip_route_input_noref(struct sk_buff *skb, __be32 daddr, __be32 saddr, - u8 tos, struct net_device *dev) -{ - struct fib_result res; - int err; - - tos &= IPTOS_RT_MASK; - rcu_read_lock(); - err = ip_route_input_rcu(skb, daddr, saddr, tos, dev, &res); - rcu_read_unlock(); - - return err; -} -EXPORT_SYMBOL(ip_route_input_noref); - /* called with rcu_read_lock held */ -int ip_route_input_rcu(struct sk_buff *skb, __be32 daddr, __be32 saddr, - u8 tos, struct net_device *dev, struct fib_result *res) +static int ip_route_input_rcu(struct sk_buff *skb, __be32 daddr, __be32 saddr, + u8 tos, struct net_device *dev, struct fib_result *res) { /* Multicast recognition logic is moved from route cache to here. - The problem was that too many Ethernet cards have broken/missing - hardware multicast filters :-( As result the host on multicasting - network acquires a lot of useless route cache entries, sort of - SDR messages from all the world. Now we try to get rid of them. - Really, provided software IP multicast filter is organized - reasonably (at least, hashed), it does not result in a slowdown - comparing with route cache reject entries. - Note, that multicast routers are not affected, because - route cache entry is created eventually. + * The problem was that too many Ethernet cards have broken/missing + * hardware multicast filters :-( As result the host on multicasting + * network acquires a lot of useless route cache entries, sort of + * SDR messages from all the world. Now we try to get rid of them. + * Really, provided software IP multicast filter is organized + * reasonably (at least, hashed), it does not result in a slowdown + * comparing with route cache reject entries. + * Note, that multicast routers are not affected, because + * route cache entry is created eventually. */ if (ipv4_is_multicast(daddr)) { struct in_device *in_dev = __in_dev_get_rcu(dev); @@ -2350,6 +2483,21 @@ int ip_route_input_rcu(struct sk_buff *skb, __be32 daddr, __be32 saddr, return ip_route_input_slow(skb, daddr, saddr, tos, dev, res); } +int ip_route_input_noref(struct sk_buff *skb, __be32 daddr, __be32 saddr, + u8 tos, struct net_device *dev) +{ + struct fib_result res; + int err; + + tos &= IPTOS_RT_MASK; + rcu_read_lock(); + err = ip_route_input_rcu(skb, daddr, saddr, tos, dev, &res); + rcu_read_unlock(); + + return err; +} +EXPORT_SYMBOL(ip_route_input_noref); + /* called with rcu_read_lock() */ static struct rtable *__mkroute_output(const struct fib_result *res, const struct flowi4 *fl4, int orig_oif, @@ -2442,9 +2590,7 @@ static struct rtable *__mkroute_output(const struct fib_result *res, add: rth = rt_dst_alloc(dev_out, flags, type, - IN_DEV_CONF_GET(in_dev, NOPOLICY), - IN_DEV_CONF_GET(in_dev, NOXFRM), - do_cache); + IN_DEV_ORCONF(in_dev, NOXFRM)); if (!rth) return ERR_PTR(-ENOBUFS); @@ -2482,7 +2628,6 @@ add: struct rtable *ip_route_output_key_hash(struct net *net, struct flowi4 *fl4, const struct sk_buff *skb) { - __u8 tos = RT_FL_TOS(fl4); struct fib_result res = { .type = RTN_UNSPEC, .fi = NULL, @@ -2492,9 +2637,7 @@ struct rtable *ip_route_output_key_hash(struct net *net, struct flowi4 *fl4, struct rtable *rth; fl4->flowi4_iif = LOOPBACK_IFINDEX; - fl4->flowi4_tos = tos & IPTOS_RT_MASK; - fl4->flowi4_scope = ((tos & RTO_ONLINK) ? - RT_SCOPE_LINK : RT_SCOPE_UNIVERSE); + ip_rt_fix_tos(fl4); rcu_read_lock(); rth = ip_route_output_key_hash_rcu(net, fl4, &res, skb); @@ -2525,11 +2668,11 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4, rth = ERR_PTR(-ENETUNREACH); /* I removed check for oif == dev_out->oif here. - It was wrong for two reasons: - 1. ip_dev_find(net, saddr) can return wrong iface, if saddr - is assigned to multiple interfaces. - 2. Moreover, we are allowed to send packets with saddr - of another iface. --ANK + * It was wrong for two reasons: + * 1. ip_dev_find(net, saddr) can return wrong iface, if saddr + * is assigned to multiple interfaces. + * 2. Moreover, we are allowed to send packets with saddr + * of another iface. --ANK */ if (fl4->flowi4_oif == 0 && @@ -2541,18 +2684,18 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4, goto out; /* Special hack: user can direct multicasts - and limited broadcast via necessary interface - without fiddling with IP_MULTICAST_IF or IP_PKTINFO. - This hack is not just for fun, it allows - vic,vat and friends to work. - They bind socket to loopback, set ttl to zero - and expect that it will work. - From the viewpoint of routing cache they are broken, - because we are not allowed to build multicast path - with loopback source addr (look, routing cache - cannot know, that ttl is zero, so that packet - will not leave this host and route is valid). - Luckily, this hack is good workaround. + * and limited broadcast via necessary interface + * without fiddling with IP_MULTICAST_IF or IP_PKTINFO. + * This hack is not just for fun, it allows + * vic,vat and friends to work. + * They bind socket to loopback, set ttl to zero + * and expect that it will work. + * From the viewpoint of routing cache they are broken, + * because we are not allowed to build multicast path + * with loopback source addr (look, routing cache + * cannot know, that ttl is zero, so that packet + * will not leave this host and route is valid). + * Luckily, this hack is good workaround. */ fl4->flowi4_oif = dev_out->ifindex; @@ -2612,24 +2755,23 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4, res->fi = NULL; res->table = NULL; if (fl4->flowi4_oif && - (ipv4_is_multicast(fl4->daddr) || - !netif_index_is_l3_master(net, fl4->flowi4_oif))) { + (ipv4_is_multicast(fl4->daddr) || !fl4->flowi4_l3mdev)) { /* Apparently, routing tables are wrong. Assume, - that the destination is on link. - - WHY? DW. - Because we are allowed to send to iface - even if it has NO routes and NO assigned - addresses. When oif is specified, routing - tables are looked up with only one purpose: - to catch if destination is gatewayed, rather than - direct. Moreover, if MSG_DONTROUTE is set, - we send packet, ignoring both routing tables - and ifaddr state. --ANK - - - We could make it even if oif is unknown, - likely IPv6, but we do not. + * that the destination is on link. + * + * WHY? DW. + * Because we are allowed to send to iface + * even if it has NO routes and NO assigned + * addresses. When oif is specified, routing + * tables are looked up with only one purpose: + * to catch if destination is gatewayed, rather than + * direct. Moreover, if MSG_DONTROUTE is set, + * we send packet, ignoring both routing tables + * and ifaddr state. --ANK + * + * + * We could make it even if oif is unknown, + * likely IPv6, but we do not. */ if (fl4->saddr == 0) @@ -2667,8 +2809,6 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4, fib_select_path(net, res, fl4, skb); dev_out = FIB_RES_DEV(*res); - fl4->flowi4_oif = dev_out->ifindex; - make_route: rth = __mkroute_output(res, fl4, orig_oif, dev_out, flags); @@ -2677,44 +2817,15 @@ out: return rth; } -static struct dst_entry *ipv4_blackhole_dst_check(struct dst_entry *dst, u32 cookie) -{ - return NULL; -} - -static unsigned int ipv4_blackhole_mtu(const struct dst_entry *dst) -{ - unsigned int mtu = dst_metric_raw(dst, RTAX_MTU); - - return mtu ? : dst->dev->mtu; -} - -static void ipv4_rt_blackhole_update_pmtu(struct dst_entry *dst, struct sock *sk, - struct sk_buff *skb, u32 mtu, - bool confirm_neigh) -{ -} - -static void ipv4_rt_blackhole_redirect(struct dst_entry *dst, struct sock *sk, - struct sk_buff *skb) -{ -} - -static u32 *ipv4_rt_blackhole_cow_metrics(struct dst_entry *dst, - unsigned long old) -{ - return NULL; -} - static struct dst_ops ipv4_dst_blackhole_ops = { - .family = AF_INET, - .check = ipv4_blackhole_dst_check, - .mtu = ipv4_blackhole_mtu, - .default_advmss = ipv4_default_advmss, - .update_pmtu = ipv4_rt_blackhole_update_pmtu, - .redirect = ipv4_rt_blackhole_redirect, - .cow_metrics = ipv4_rt_blackhole_cow_metrics, - .neigh_lookup = ipv4_neigh_lookup, + .family = AF_INET, + .default_advmss = ipv4_default_advmss, + .neigh_lookup = ipv4_neigh_lookup, + .check = dst_blackhole_check, + .cow_metrics = dst_blackhole_cow_metrics, + .update_pmtu = dst_blackhole_update_pmtu, + .redirect = dst_blackhole_redirect, + .mtu = dst_blackhole_mtu, }; struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_orig) @@ -2731,8 +2842,7 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or new->output = dst_discard_out; new->dev = net->loopback_dev; - if (new->dev) - dev_hold(new->dev); + netdev_hold(new->dev, &new->dev_tracker, GFP_ATOMIC); rt->rt_is_input = ort->rt_is_input; rt->rt_iif = ort->rt_iif; @@ -2765,15 +2875,65 @@ struct rtable *ip_route_output_flow(struct net *net, struct flowi4 *flp4, if (IS_ERR(rt)) return rt; - if (flp4->flowi4_proto) + if (flp4->flowi4_proto) { + flp4->flowi4_oif = rt->dst.dev->ifindex; rt = (struct rtable *)xfrm_lookup_route(net, &rt->dst, flowi4_to_flowi(flp4), sk, 0); + } return rt; } EXPORT_SYMBOL_GPL(ip_route_output_flow); +struct rtable *ip_route_output_tunnel(struct sk_buff *skb, + struct net_device *dev, + struct net *net, __be32 *saddr, + const struct ip_tunnel_info *info, + u8 protocol, bool use_cache) +{ +#ifdef CONFIG_DST_CACHE + struct dst_cache *dst_cache; +#endif + struct rtable *rt = NULL; + struct flowi4 fl4; + __u8 tos; + +#ifdef CONFIG_DST_CACHE + dst_cache = (struct dst_cache *)&info->dst_cache; + if (use_cache) { + rt = dst_cache_get_ip4(dst_cache, saddr); + if (rt) + return rt; + } +#endif + memset(&fl4, 0, sizeof(fl4)); + fl4.flowi4_mark = skb->mark; + fl4.flowi4_proto = protocol; + fl4.daddr = info->key.u.ipv4.dst; + fl4.saddr = info->key.u.ipv4.src; + tos = info->key.tos; + fl4.flowi4_tos = RT_TOS(tos); + + rt = ip_route_output_key(net, &fl4); + if (IS_ERR(rt)) { + netdev_dbg(dev, "no route to %pI4\n", &fl4.daddr); + return ERR_PTR(-ENETUNREACH); + } + if (rt->dst.dev == dev) { /* is this necessary? */ + netdev_dbg(dev, "circular route to %pI4\n", &fl4.daddr); + ip_rt_put(rt); + return ERR_PTR(-ELOOP); + } +#ifdef CONFIG_DST_CACHE + if (use_cache) + dst_cache_set_ip4(dst_cache, &rt->dst, fl4.saddr); +#endif + *saddr = fl4.saddr; + return rt; +} +EXPORT_SYMBOL_GPL(ip_route_output_tunnel); + /* called with rcu_read_lock held */ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, struct rtable *rt, u32 table_id, struct flowi4 *fl4, @@ -2817,6 +2977,9 @@ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, if (rt->dst.dev && nla_put_u32(skb, RTA_OIF, rt->dst.dev->ifindex)) goto nla_put_failure; + if (rt->dst.lwtstate && + lwtunnel_fill_encap(skb, rt->dst.lwtstate, RTA_ENCAP, RTA_ENCAP_TYPE) < 0) + goto nla_put_failure; #ifdef CONFIG_IP_ROUTE_CLASSID if (rt->dst.tclassid && nla_put_u32(skb, RTA_FLOW, rt->dst.tclassid)) @@ -3017,7 +3180,7 @@ static struct sk_buff *inet_rtm_getroute_build_skb(__be32 src, __be32 dst, udph = skb_put_zero(skb, sizeof(struct udphdr)); udph->source = sport; udph->dest = dport; - udph->len = sizeof(struct udphdr); + udph->len = htons(sizeof(struct udphdr)); udph->check = 0; break; } @@ -3167,7 +3330,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, fl4.daddr = dst; fl4.saddr = src; - fl4.flowi4_tos = rtm->rtm_tos; + fl4.flowi4_tos = rtm->rtm_tos & IPTOS_RT_MASK; fl4.flowi4_oif = tb[RTA_OIF] ? nla_get_u32(tb[RTA_OIF]) : 0; fl4.flowi4_mark = mark; fl4.flowi4_uid = uid; @@ -3191,8 +3354,9 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, fl4.flowi4_iif = iif; /* for rt_fill_info */ skb->dev = dev; skb->mark = mark; - err = ip_route_input_rcu(skb, dst, src, rtm->rtm_tos, - dev, &res); + err = ip_route_input_rcu(skb, dst, src, + rtm->rtm_tos & IPTOS_RT_MASK, dev, + &res); rt = skb_rtable(skb); if (err == 0 && rt->dst.error) @@ -3236,10 +3400,11 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, fri.tb_id = table_id; fri.dst = res.prefix; fri.dst_len = res.prefixlen; - fri.tos = fl4.flowi4_tos; + fri.dscp = inet_dsfield_to_dscp(fl4.flowi4_tos); fri.type = rt->rt_type; fri.offload = 0; fri.trap = 0; + fri.offload_failed = 0; if (res.fa_head) { struct fib_alias *fa; @@ -3248,11 +3413,11 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, if (fa->fa_slen == slen && fa->tb_id == fri.tb_id && - fa->fa_tos == fri.tos && + fa->fa_dscp == fri.dscp && fa->fa_info == res.fi && fa->fa_type == fri.type) { - fri.offload = fa->offload; - fri.trap = fa->trap; + fri.offload = READ_ONCE(fa->offload); + fri.trap = READ_ONCE(fa->trap); break; } } @@ -3291,8 +3456,7 @@ static int ip_rt_gc_elasticity __read_mostly = 8; static int ip_min_valid_pmtu __read_mostly = IPV4_MIN_MTU; static int ipv4_sysctl_rtcache_flush(struct ctl_table *__ctl, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = (struct net *)__ctl->extra1; @@ -3392,40 +3556,40 @@ static struct ctl_table ipv4_route_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { - .procname = "mtu_expires", - .data = &ip_rt_mtu_expires, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_jiffies, - }, - { - .procname = "min_pmtu", - .data = &ip_rt_min_pmtu, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &ip_min_valid_pmtu, - }, - { - .procname = "min_adv_mss", - .data = &ip_rt_min_advmss, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { } }; static const char ipv4_route_flush_procname[] = "flush"; -static struct ctl_table ipv4_route_flush_table[] = { +static struct ctl_table ipv4_route_netns_table[] = { { .procname = ipv4_route_flush_procname, .maxlen = sizeof(int), .mode = 0200, .proc_handler = ipv4_sysctl_rtcache_flush, }, + { + .procname = "min_pmtu", + .data = &init_net.ipv4.ip_rt_min_pmtu, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &ip_min_valid_pmtu, + }, + { + .procname = "mtu_expires", + .data = &init_net.ipv4.ip_rt_mtu_expires, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .procname = "min_adv_mss", + .data = &init_net.ipv4.ip_rt_min_advmss, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { }, }; @@ -3433,9 +3597,11 @@ static __net_init int sysctl_route_net_init(struct net *net) { struct ctl_table *tbl; - tbl = ipv4_route_flush_table; + tbl = ipv4_route_netns_table; if (!net_eq(net, &init_net)) { - tbl = kmemdup(tbl, sizeof(ipv4_route_flush_table), GFP_KERNEL); + int i; + + tbl = kmemdup(tbl, sizeof(ipv4_route_netns_table), GFP_KERNEL); if (!tbl) goto err_dup; @@ -3444,6 +3610,12 @@ static __net_init int sysctl_route_net_init(struct net *net) if (tbl[0].procname != ipv4_route_flush_procname) tbl[0].procname = NULL; } + + /* Update the variables to point into the current struct net + * except for the first element flush + */ + for (i = 1; i < ARRAY_SIZE(ipv4_route_netns_table) - 1; i++) + tbl[i].data += (void *)net - (void *)&init_net; } tbl[0].extra1 = net; @@ -3453,7 +3625,7 @@ static __net_init int sysctl_route_net_init(struct net *net) return 0; err_reg: - if (tbl != ipv4_route_flush_table) + if (tbl != ipv4_route_netns_table) kfree(tbl); err_dup: return -ENOMEM; @@ -3465,7 +3637,7 @@ static __net_exit void sysctl_route_net_exit(struct net *net) tbl = net->ipv4.route_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.route_hdr); - BUG_ON(tbl == ipv4_route_flush_table); + BUG_ON(tbl == ipv4_route_netns_table); kfree(tbl); } @@ -3475,11 +3647,24 @@ static __net_initdata struct pernet_operations sysctl_route_ops = { }; #endif +static __net_init int netns_ip_rt_init(struct net *net) +{ + /* Set default value for namespaceified sysctls */ + net->ipv4.ip_rt_min_pmtu = DEFAULT_MIN_PMTU; + net->ipv4.ip_rt_mtu_expires = DEFAULT_MTU_EXPIRES; + net->ipv4.ip_rt_min_advmss = DEFAULT_MIN_ADVMSS; + return 0; +} + +static struct pernet_operations __net_initdata ip_rt_ops = { + .init = netns_ip_rt_init, +}; + static __net_init int rt_genid_init(struct net *net) { atomic_set(&net->ipv4.rt_genid, 0); atomic_set(&net->fnhe_genid, 0); - atomic_set(&net->ipv4.dev_addr_genid, get_random_int()); + atomic_set(&net->ipv4.dev_addr_genid, get_random_u32()); return 0; } @@ -3518,23 +3703,31 @@ struct ip_rt_acct __percpu *ip_rt_acct __read_mostly; int __init ip_rt_init(void) { + void *idents_hash; int cpu; - ip_idents = kmalloc_array(IP_IDENTS_SZ, sizeof(*ip_idents), - GFP_KERNEL); - if (!ip_idents) - panic("IP: failed to allocate ip_idents\n"); + /* For modern hosts, this will use 2 MB of memory */ + idents_hash = alloc_large_system_hash("IP idents", + sizeof(*ip_idents) + sizeof(*ip_tstamps), + 0, + 16, /* one bucket per 64 KB */ + HASH_ZERO, + NULL, + &ip_idents_mask, + 2048, + 256*1024); + + ip_idents = idents_hash; - prandom_bytes(ip_idents, IP_IDENTS_SZ * sizeof(*ip_idents)); + get_random_bytes(ip_idents, (ip_idents_mask + 1) * sizeof(*ip_idents)); - ip_tstamps = kcalloc(IP_IDENTS_SZ, sizeof(*ip_tstamps), GFP_KERNEL); - if (!ip_tstamps) - panic("IP: failed to allocate ip_tstamps\n"); + ip_tstamps = idents_hash + (ip_idents_mask + 1) * sizeof(*ip_idents); for_each_possible_cpu(cpu) { struct uncached_list *ul = &per_cpu(rt_uncached_list, cpu); INIT_LIST_HEAD(&ul->head); + INIT_LIST_HEAD(&ul->quarantine); spin_lock_init(&ul->lock); } #ifdef CONFIG_IP_ROUTE_CLASSID @@ -3573,6 +3766,7 @@ int __init ip_rt_init(void) #ifdef CONFIG_SYSCTL register_pernet_subsys(&sysctl_route_ops); #endif + register_pernet_subsys(&ip_rt_ops); register_pernet_subsys(&rt_genid_ops); register_pernet_subsys(&ipv4_inetpeer_ops); return 0; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 9a4f6b16c9bc..942d2dfa1115 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -7,8 +7,6 @@ */ #include <linux/tcp.h> -#include <linux/slab.h> -#include <linux/random.h> #include <linux/siphash.h> #include <linux/kernel.h> #include <linux/export.h> @@ -16,7 +14,7 @@ #include <net/tcp.h> #include <net/route.h> -static siphash_key_t syncookie_secret[2] __read_mostly; +static siphash_aligned_key_t syncookie_secret[2]; #define COOKIEBITS 24 /* Upper bits store count */ #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) @@ -212,6 +210,12 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb, refcount_set(&req->rsk_refcnt, 1); tcp_sk(child)->tsoffset = tsoff; sock_rps_save_rxhash(child, skb); + + if (rsk_drop_req(req)) { + reqsk_put(req); + return child; + } + if (inet_csk_reqsk_queue_add(sk, req, child)) return child; @@ -243,12 +247,12 @@ bool cookie_timestamp_decode(const struct net *net, return true; } - if (!net->ipv4.sysctl_tcp_timestamps) + if (!READ_ONCE(net->ipv4.sysctl_tcp_timestamps)) return false; tcp_opt->sack_ok = (options & TS_OPT_SACK) ? TCP_SACK_SEEN : 0; - if (tcp_opt->sack_ok && !net->ipv4.sysctl_tcp_sack) + if (tcp_opt->sack_ok && !READ_ONCE(net->ipv4.sysctl_tcp_sack)) return false; if ((options & TS_OPT_WSCALE_MASK) == TS_OPT_WSCALE_MASK) @@ -257,7 +261,7 @@ bool cookie_timestamp_decode(const struct net *net, tcp_opt->wscale_ok = 1; tcp_opt->snd_wscale = options & TS_OPT_WSCALE_MASK; - return net->ipv4.sysctl_tcp_window_scaling != 0; + return READ_ONCE(net->ipv4.sysctl_tcp_window_scaling) != 0; } EXPORT_SYMBOL(cookie_timestamp_decode); @@ -269,13 +273,52 @@ bool cookie_ecn_ok(const struct tcp_options_received *tcp_opt, if (!ecn_ok) return false; - if (net->ipv4.sysctl_tcp_ecn) + if (READ_ONCE(net->ipv4.sysctl_tcp_ecn)) return true; return dst_feature(dst, RTAX_FEATURE_ECN); } EXPORT_SYMBOL(cookie_ecn_ok); +struct request_sock *cookie_tcp_reqsk_alloc(const struct request_sock_ops *ops, + const struct tcp_request_sock_ops *af_ops, + struct sock *sk, + struct sk_buff *skb) +{ + struct tcp_request_sock *treq; + struct request_sock *req; + +#ifdef CONFIG_MPTCP + if (sk_is_mptcp(sk)) + ops = &mptcp_subflow_request_sock_ops; +#endif + + req = inet_reqsk_alloc(ops, sk, false); + if (!req) + return NULL; + + treq = tcp_rsk(req); + + /* treq->af_specific might be used to perform TCP_MD5 lookup */ + treq->af_specific = af_ops; + + treq->syn_tos = TCP_SKB_CB(skb)->ip_dsfield; +#if IS_ENABLED(CONFIG_MPTCP) + treq->is_mptcp = sk_is_mptcp(sk); + if (treq->is_mptcp) { + int err = mptcp_subflow_init_cookie_req(req, sk, skb); + + if (err) { + reqsk_free(req); + return NULL; + } + } +#endif + + return req; +} +EXPORT_SYMBOL_GPL(cookie_tcp_reqsk_alloc); + /* On input, sk is a listener. * Output is listener if incoming packet would not create a child * NULL if memory could not be allocated. @@ -291,13 +334,14 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) __u32 cookie = ntohl(th->ack_seq) - 1; struct sock *ret = sk; struct request_sock *req; - int mss; + int full_space, mss; struct rtable *rt; __u8 rcv_wscale; struct flowi4 fl4; u32 tsoff = 0; - if (!sock_net(sk)->ipv4.sysctl_tcp_syncookies || !th->ack || th->rst) + if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies) || + !th->ack || th->rst) goto out; if (tcp_synq_no_recent_overflow(sk)) @@ -326,7 +370,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) goto out; ret = NULL; - req = inet_reqsk_alloc(&tcp_request_sock_ops, sk, false); /* for safety */ + req = cookie_tcp_reqsk_alloc(&tcp_request_sock_ops, + &tcp_request_sock_ipv4_ops, sk, skb); if (!req) goto out; @@ -350,9 +395,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) treq->snt_synack = 0; treq->tfo_listener = false; - if (IS_ENABLED(CONFIG_MPTCP)) - treq->is_mptcp = 0; - if (IS_ENABLED(CONFIG_SMC)) ireq->smc_ok = 0; @@ -381,7 +423,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) inet_sk_flowi_flags(sk), opt->srr ? opt->faddr : ireq->ir_rmt_addr, ireq->ir_loc_addr, th->source, th->dest, sk->sk_uid); - security_req_classify_flow(req, flowi4_to_flowi(&fl4)); + security_req_classify_flow(req, flowi4_to_flowi_common(&fl4)); rt = ip_route_output_key(sock_net(sk), &fl4); if (IS_ERR(rt)) { reqsk_free(req); @@ -390,8 +432,13 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) /* Try to redo what tcp_v4_send_synack did. */ req->rsk_window_clamp = tp->window_clamp ? :dst_metric(&rt->dst, RTAX_WINDOW); + /* limit the window selection if the user enforce a smaller rx buffer */ + full_space = tcp_full_space(sk); + if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && + (req->rsk_window_clamp > full_space || req->rsk_window_clamp == 0)) + req->rsk_window_clamp = full_space; - tcp_select_initial_window(sk, tcp_full_space(sk), req->mss, + tcp_select_initial_window(sk, full_space, req->mss, &req->rsk_rcv_wnd, &req->rsk_window_clamp, ireq->wscale_ok, &rcv_wscale, dst_metric(&rt->dst, RTAX_INITRWND)); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 9684af02e0a5..9b8a6db7a66b 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -6,32 +6,20 @@ * Added /proc/sys/net/ipv4 directory entry (empty =) ). [MS] */ -#include <linux/mm.h> -#include <linux/module.h> #include <linux/sysctl.h> -#include <linux/igmp.h> -#include <linux/inetdevice.h> #include <linux/seqlock.h> #include <linux/init.h> #include <linux/slab.h> -#include <linux/nsproxy.h> -#include <linux/swap.h> -#include <net/snmp.h> #include <net/icmp.h> #include <net/ip.h> -#include <net/route.h> +#include <net/ip_fib.h> #include <net/tcp.h> #include <net/udp.h> #include <net/cipso_ipv4.h> -#include <net/inet_frag.h> #include <net/ping.h> #include <net/protocol.h> #include <net/netevent.h> -static int two = 2; -static int four = 4; -static int thousand = 1000; -static int gso_max_segs = GSO_MAX_SEGS; static int tcp_retr1_max = 255; static int ip_local_port_range_min[] = { 1, 1 }; static int ip_local_port_range_max[] = { 65535, 65535 }; @@ -47,9 +35,11 @@ static int tcp_syn_retries_min = 1; static int tcp_syn_retries_max = MAX_TCP_SYNCNT; static int ip_ping_group_range_min[] = { 0, 0 }; static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX }; -static int comp_sack_nr_max = 255; static u32 u32_max_div_HZ = UINT_MAX / HZ; static int one_day_secs = 24 * 3600; +static u32 fib_multipath_hash_fields_all_mask __maybe_unused = + FIB_MULTIPATH_HASH_FIELD_ALL_MASK; +static unsigned int tcp_child_ehash_entries_max = 16 * 1024 * 1024; /* obsolete */ static int sysctl_tcp_low_latency __read_mostly; @@ -71,8 +61,7 @@ static void set_local_port_range(struct net *net, int range[2]) /* Validate changes from /proc interface. */ static int ipv4_local_port_range(struct ctl_table *table, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = container_of(table->data, struct net, ipv4.ip_local_ports.range); @@ -96,7 +85,7 @@ static int ipv4_local_port_range(struct ctl_table *table, int write, * port limit. */ if ((range[1] < range[0]) || - (range[0] < net->ipv4.sysctl_ip_prot_sock)) + (range[0] < READ_ONCE(net->ipv4.sysctl_ip_prot_sock))) ret = -EINVAL; else set_local_port_range(net, range); @@ -107,7 +96,7 @@ static int ipv4_local_port_range(struct ctl_table *table, int write, /* Validate changes from /proc interface. */ static int ipv4_privileged_ports(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = container_of(table->data, struct net, ipv4.sysctl_ip_prot_sock); @@ -122,7 +111,7 @@ static int ipv4_privileged_ports(struct ctl_table *table, int write, .extra2 = &ip_privileged_port_max, }; - pports = net->ipv4.sysctl_ip_prot_sock; + pports = READ_ONCE(net->ipv4.sysctl_ip_prot_sock); ret = proc_dointvec_minmax(&tmp, write, buffer, lenp, ppos); @@ -134,7 +123,7 @@ static int ipv4_privileged_ports(struct ctl_table *table, int write, if (range[0] < pports) ret = -EINVAL; else - net->ipv4.sysctl_ip_prot_sock = pports; + WRITE_ONCE(net->ipv4.sysctl_ip_prot_sock, pports); } return ret; @@ -168,8 +157,7 @@ static void set_ping_group_range(struct ctl_table *table, kgid_t low, kgid_t hig /* Validate changes from /proc interface. */ static int ipv4_ping_group_range(struct ctl_table *table, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct user_namespace *user_ns = current_user_ns(); int ret; @@ -204,15 +192,14 @@ static int ipv4_ping_group_range(struct ctl_table *table, int write, } static int ipv4_fwd_update_priority(struct ctl_table *table, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net; int ret; net = container_of(table->data, struct net, ipv4.sysctl_ip_fwd_update_priority); - ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + ret = proc_dou8vec_minmax(table, write, buffer, lenp, ppos); if (write && ret == 0) call_netevent_notifiers(NETEVENT_IPV4_FWD_UPDATE_PRIORITY_UPDATE, net); @@ -221,7 +208,7 @@ static int ipv4_fwd_update_priority(struct ctl_table *table, int write, } static int proc_tcp_congestion_control(struct ctl_table *ctl, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = container_of(ctl->data, struct net, ipv4.tcp_congestion_control); @@ -241,9 +228,8 @@ static int proc_tcp_congestion_control(struct ctl_table *ctl, int write, } static int proc_tcp_available_congestion_control(struct ctl_table *ctl, - int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) + int write, void *buffer, + size_t *lenp, loff_t *ppos) { struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, }; int ret; @@ -258,9 +244,8 @@ static int proc_tcp_available_congestion_control(struct ctl_table *ctl, } static int proc_allowed_congestion_control(struct ctl_table *ctl, - int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) + int write, void *buffer, + size_t *lenp, loff_t *ppos) { struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; int ret; @@ -296,8 +281,7 @@ static int sscanf_key(char *buf, __le32 *key) } static int proc_tcp_fastopen_key(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = container_of(table->data, struct net, ipv4.sysctl_tcp_fastopen); @@ -307,24 +291,16 @@ static int proc_tcp_fastopen_key(struct ctl_table *table, int write, struct ctl_table tbl = { .maxlen = ((TCP_FASTOPEN_KEY_LENGTH * 2 * TCP_FASTOPEN_KEY_MAX) + (TCP_FASTOPEN_KEY_MAX * 5)) }; - struct tcp_fastopen_context *ctx; - u32 user_key[TCP_FASTOPEN_KEY_MAX * 4]; - __le32 key[TCP_FASTOPEN_KEY_MAX * 4]; + u32 user_key[TCP_FASTOPEN_KEY_BUF_LENGTH / sizeof(u32)]; + __le32 key[TCP_FASTOPEN_KEY_BUF_LENGTH / sizeof(__le32)]; char *backup_data; - int ret, i = 0, off = 0, n_keys = 0; + int ret, i = 0, off = 0, n_keys; tbl.data = kmalloc(tbl.maxlen, GFP_KERNEL); if (!tbl.data) return -ENOMEM; - rcu_read_lock(); - ctx = rcu_dereference(net->ipv4.tcp_fastopen_ctx); - if (ctx) { - n_keys = tcp_fastopen_context_len(ctx); - memcpy(&key[0], &ctx->key[0], TCP_FASTOPEN_KEY_LENGTH * n_keys); - } - rcu_read_unlock(); - + n_keys = tcp_fastopen_get_cipher(net, NULL, (u64 *)key); if (!n_keys) { memset(&key[0], 0, TCP_FASTOPEN_KEY_LENGTH); n_keys = 1; @@ -375,64 +351,8 @@ bad_key: return ret; } -static void proc_configure_early_demux(int enabled, int protocol) -{ - struct net_protocol *ipprot; -#if IS_ENABLED(CONFIG_IPV6) - struct inet6_protocol *ip6prot; -#endif - - rcu_read_lock(); - - ipprot = rcu_dereference(inet_protos[protocol]); - if (ipprot) - ipprot->early_demux = enabled ? ipprot->early_demux_handler : - NULL; - -#if IS_ENABLED(CONFIG_IPV6) - ip6prot = rcu_dereference(inet6_protos[protocol]); - if (ip6prot) - ip6prot->early_demux = enabled ? ip6prot->early_demux_handler : - NULL; -#endif - rcu_read_unlock(); -} - -static int proc_tcp_early_demux(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - int ret = 0; - - ret = proc_dointvec(table, write, buffer, lenp, ppos); - - if (write && !ret) { - int enabled = init_net.ipv4.sysctl_tcp_early_demux; - - proc_configure_early_demux(enabled, IPPROTO_TCP); - } - - return ret; -} - -static int proc_udp_early_demux(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - int ret = 0; - - ret = proc_dointvec(table, write, buffer, lenp, ppos); - - if (write && !ret) { - int enabled = init_net.ipv4.sysctl_udp_early_demux; - - proc_configure_early_demux(enabled, IPPROTO_UDP); - } - - return ret; -} - static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table, - int write, - void __user *buffer, + int write, void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = container_of(table->data, struct net, @@ -447,8 +367,7 @@ static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table, } static int proc_tcp_available_ulp(struct ctl_table *ctl, - int write, - void __user *buffer, size_t *lenp, + int write, void *buffer, size_t *lenp, loff_t *ppos) { struct ctl_table tbl = { .maxlen = TCP_ULP_BUF_MAX, }; @@ -464,16 +383,55 @@ static int proc_tcp_available_ulp(struct ctl_table *ctl, return ret; } +static int proc_tcp_ehash_entries(struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + struct net *net = container_of(table->data, struct net, + ipv4.sysctl_tcp_child_ehash_entries); + struct inet_hashinfo *hinfo = net->ipv4.tcp_death_row.hashinfo; + int tcp_ehash_entries; + struct ctl_table tbl; + + tcp_ehash_entries = hinfo->ehash_mask + 1; + + /* A negative number indicates that the child netns + * shares the global ehash. + */ + if (!net_eq(net, &init_net) && !hinfo->pernet) + tcp_ehash_entries *= -1; + + tbl.data = &tcp_ehash_entries; + tbl.maxlen = sizeof(int); + + return proc_dointvec(&tbl, write, buffer, lenp, ppos); +} + #ifdef CONFIG_IP_ROUTE_MULTIPATH static int proc_fib_multipath_hash_policy(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, + void *buffer, size_t *lenp, loff_t *ppos) { struct net *net = container_of(table->data, struct net, ipv4.sysctl_fib_multipath_hash_policy); int ret; - ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + ret = proc_dou8vec_minmax(table, write, buffer, lenp, ppos); + if (write && ret == 0) + call_netevent_notifiers(NETEVENT_IPV4_MPATH_HASH_UPDATE, net); + + return ret; +} + +static int proc_fib_multipath_hash_fields(struct ctl_table *table, int write, + void *buffer, size_t *lenp, + loff_t *ppos) +{ + struct net *net; + int ret; + + net = container_of(table->data, struct net, + ipv4.sysctl_fib_multipath_hash_fields); + ret = proc_douintvec_minmax(table, write, buffer, lenp, ppos); if (write && ret == 0) call_netevent_notifiers(NETEVENT_IPV4_MPATH_HASH_UPDATE, net); @@ -555,18 +513,6 @@ static struct ctl_table ipv4_table[] = { }, #endif /* CONFIG_NETLABEL */ { - .procname = "tcp_available_congestion_control", - .maxlen = TCP_CA_BUF_MAX, - .mode = 0444, - .proc_handler = proc_tcp_available_congestion_control, - }, - { - .procname = "tcp_allowed_congestion_control", - .maxlen = TCP_CA_BUF_MAX, - .mode = 0644, - .proc_handler = proc_allowed_congestion_control, - }, - { .procname = "tcp_available_ulp", .maxlen = TCP_ULP_BUF_MAX, .mode = 0444, @@ -604,49 +550,61 @@ static struct ctl_table ipv4_table[] = { .extra1 = &sysctl_fib_sync_mem_min, .extra2 = &sysctl_fib_sync_mem_max, }, - { - .procname = "tcp_rx_skb_cache", - .data = &tcp_rx_skb_cache_key.key, - .mode = 0644, - .proc_handler = proc_do_static_key, - }, - { - .procname = "tcp_tx_skb_cache", - .data = &tcp_tx_skb_cache_key.key, - .mode = 0644, - .proc_handler = proc_do_static_key, - }, { } }; static struct ctl_table ipv4_net_table[] = { { - .procname = "icmp_echo_ignore_all", - .data = &init_net.ipv4.sysctl_icmp_echo_ignore_all, + .procname = "tcp_max_tw_buckets", + .data = &init_net.ipv4.tcp_death_row.sysctl_max_tw_buckets, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec }, { + .procname = "icmp_echo_ignore_all", + .data = &init_net.ipv4.sysctl_icmp_echo_ignore_all, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE + }, + { + .procname = "icmp_echo_enable_probe", + .data = &init_net.ipv4.sysctl_icmp_echo_enable_probe, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE + }, + { .procname = "icmp_echo_ignore_broadcasts", .data = &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE }, { .procname = "icmp_ignore_bogus_error_responses", .data = &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE }, { .procname = "icmp_errors_use_inbound_ifaddr", .data = &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE }, { .procname = "icmp_ratelimit", @@ -673,9 +631,9 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "raw_l3mdev_accept", .data = &init_net.ipv4.sysctl_raw_l3mdev_accept, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, @@ -683,51 +641,64 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_ecn", .data = &init_net.ipv4.sysctl_tcp_ecn, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, }, { .procname = "tcp_ecn_fallback", .data = &init_net.ipv4.sysctl_tcp_ecn_fallback, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, }, { .procname = "ip_dynaddr", .data = &init_net.ipv4.sysctl_ip_dynaddr, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "ip_early_demux", .data = &init_net.ipv4.sysctl_ip_early_demux, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "udp_early_demux", .data = &init_net.ipv4.sysctl_udp_early_demux, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_udp_early_demux + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_early_demux", .data = &init_net.ipv4.sysctl_tcp_early_demux, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + }, + { + .procname = "nexthop_compat_mode", + .data = &init_net.ipv4.sysctl_nexthop_compat_mode, + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_tcp_early_demux + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, }, { .procname = "ip_default_ttl", .data = &init_net.ipv4.sysctl_ip_default_ttl, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = &ip_ttl_min, .extra2 = &ip_ttl_max, }, @@ -748,21 +719,21 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "ip_no_pmtu_disc", .data = &init_net.ipv4.sysctl_ip_no_pmtu_disc, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "ip_forward_use_pmtu", .data = &init_net.ipv4.sysctl_ip_fwd_use_pmtu, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "ip_forward_update_priority", .data = &init_net.ipv4.sysctl_ip_fwd_update_priority, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, .proc_handler = ipv4_fwd_update_priority, .extra1 = SYSCTL_ZERO, @@ -771,31 +742,40 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "ip_nonlocal_bind", .data = &init_net.ipv4.sysctl_ip_nonlocal_bind, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, + }, + { + .procname = "ip_autobind_reuse", + .data = &init_net.ipv4.sysctl_ip_autobind_reuse, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, }, { .procname = "fwmark_reflect", .data = &init_net.ipv4.sysctl_fwmark_reflect, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_fwmark_accept", .data = &init_net.ipv4.sysctl_tcp_fwmark_accept, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, #ifdef CONFIG_NET_L3_MASTER_DEV { .procname = "tcp_l3mdev_accept", .data = &init_net.ipv4.sysctl_tcp_l3mdev_accept, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, @@ -803,9 +783,9 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_mtu_probing", .data = &init_net.ipv4.sysctl_tcp_mtu_probing, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_base_mss", @@ -850,9 +830,9 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "igmp_link_local_mcast_reports", .data = &init_net.ipv4.sysctl_igmp_llm_reports, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "igmp_max_memberships", @@ -886,6 +866,18 @@ static struct ctl_table ipv4_net_table[] = { .proc_handler = proc_tcp_congestion_control, }, { + .procname = "tcp_available_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0444, + .proc_handler = proc_tcp_available_congestion_control, + }, + { + .procname = "tcp_allowed_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0644, + .proc_handler = proc_allowed_congestion_control, + }, + { .procname = "tcp_keepalive_time", .data = &init_net.ipv4.sysctl_tcp_keepalive_time, .maxlen = sizeof(int), @@ -895,9 +887,9 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_keepalive_probes", .data = &init_net.ipv4.sysctl_tcp_keepalive_probes, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_keepalive_intvl", @@ -909,29 +901,38 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_syn_retries", .data = &init_net.ipv4.sysctl_tcp_syn_retries, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = &tcp_syn_retries_min, .extra2 = &tcp_syn_retries_max }, { .procname = "tcp_synack_retries", .data = &init_net.ipv4.sysctl_tcp_synack_retries, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, #ifdef CONFIG_SYN_COOKIES { .procname = "tcp_syncookies", .data = &init_net.ipv4.sysctl_tcp_syncookies, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, #endif { + .procname = "tcp_migrate_req", + .data = &init_net.ipv4.sysctl_tcp_migrate_req, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE + }, + { .procname = "tcp_reordering", .data = &init_net.ipv4.sysctl_tcp_reordering, .maxlen = sizeof(int), @@ -941,24 +942,24 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_retries1", .data = &init_net.ipv4.sysctl_tcp_retries1, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra2 = &tcp_retr1_max }, { .procname = "tcp_retries2", .data = &init_net.ipv4.sysctl_tcp_retries2, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_orphan_retries", .data = &init_net.ipv4.sysctl_tcp_orphan_retries, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_fin_timeout", @@ -977,18 +978,11 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_tw_reuse", .data = &init_net.ipv4.sysctl_tcp_tw_reuse, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, - .extra2 = &two, - }, - { - .procname = "tcp_max_tw_buckets", - .data = &init_net.ipv4.tcp_death_row.sysctl_max_tw_buckets, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec + .extra2 = SYSCTL_TWO, }, { .procname = "tcp_max_syn_backlog", @@ -1028,20 +1022,29 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "fib_multipath_use_neigh", .data = &init_net.ipv4.sysctl_fib_multipath_use_neigh, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, { .procname = "fib_multipath_hash_policy", .data = &init_net.ipv4.sysctl_fib_multipath_hash_policy, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_fib_multipath_hash_policy, .extra1 = SYSCTL_ZERO, - .extra2 = &two, + .extra2 = SYSCTL_THREE, + }, + { + .procname = "fib_multipath_hash_fields", + .data = &init_net.ipv4.sysctl_fib_multipath_hash_fields, + .maxlen = sizeof(u32), + .mode = 0644, + .proc_handler = proc_fib_multipath_hash_fields, + .extra1 = SYSCTL_ONE, + .extra2 = &fib_multipath_hash_fields_all_mask, }, #endif { @@ -1055,9 +1058,9 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "udp_l3mdev_accept", .data = &init_net.ipv4.sysctl_udp_l3mdev_accept, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, @@ -1065,88 +1068,88 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_sack", .data = &init_net.ipv4.sysctl_tcp_sack, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_window_scaling", .data = &init_net.ipv4.sysctl_tcp_window_scaling, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_timestamps", .data = &init_net.ipv4.sysctl_tcp_timestamps, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_early_retrans", .data = &init_net.ipv4.sysctl_tcp_early_retrans, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, - .extra2 = &four, + .extra2 = SYSCTL_FOUR, }, { .procname = "tcp_recovery", .data = &init_net.ipv4.sysctl_tcp_recovery, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_thin_linear_timeouts", .data = &init_net.ipv4.sysctl_tcp_thin_linear_timeouts, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_slow_start_after_idle", .data = &init_net.ipv4.sysctl_tcp_slow_start_after_idle, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_retrans_collapse", .data = &init_net.ipv4.sysctl_tcp_retrans_collapse, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_stdurg", .data = &init_net.ipv4.sysctl_tcp_stdurg, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_rfc1337", .data = &init_net.ipv4.sysctl_tcp_rfc1337, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_abort_on_overflow", .data = &init_net.ipv4.sysctl_tcp_abort_on_overflow, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_fack", .data = &init_net.ipv4.sysctl_tcp_fack, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_max_reordering", @@ -1158,16 +1161,16 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_dsack", .data = &init_net.ipv4.sysctl_tcp_dsack, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_app_win", .data = &init_net.ipv4.sysctl_tcp_app_win, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_adv_win_scale", @@ -1181,46 +1184,46 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_frto", .data = &init_net.ipv4.sysctl_tcp_frto, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_no_metrics_save", .data = &init_net.ipv4.sysctl_tcp_nometrics_save, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_no_ssthresh_metrics_save", .data = &init_net.ipv4.sysctl_tcp_no_ssthresh_metrics_save, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, { .procname = "tcp_moderate_rcvbuf", .data = &init_net.ipv4.sysctl_tcp_moderate_rcvbuf, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_tso_win_divisor", .data = &init_net.ipv4.sysctl_tcp_tso_win_divisor, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_workaround_signed_windows", .data = &init_net.ipv4.sysctl_tcp_workaround_signed_windows, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_limit_output_bytes", @@ -1239,11 +1242,17 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_min_tso_segs", .data = &init_net.ipv4.sysctl_tcp_min_tso_segs, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ONE, - .extra2 = &gso_max_segs, + }, + { + .procname = "tcp_tso_rtt_log", + .data = &init_net.ipv4.sysctl_tcp_tso_rtt_log, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, }, { .procname = "tcp_min_rtt_wlen", @@ -1257,9 +1266,9 @@ static struct ctl_table ipv4_net_table[] = { { .procname = "tcp_autocorking", .data = &init_net.ipv4.sysctl_tcp_autocorking, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, @@ -1277,7 +1286,7 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, - .extra2 = &thousand, + .extra2 = SYSCTL_ONE_THOUSAND, }, { .procname = "tcp_pacing_ca_ratio", @@ -1286,7 +1295,7 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, - .extra2 = &thousand, + .extra2 = SYSCTL_ONE_THOUSAND, }, { .procname = "tcp_wmem", @@ -1312,13 +1321,43 @@ static struct ctl_table ipv4_net_table[] = { .proc_handler = proc_doulongvec_minmax, }, { + .procname = "tcp_comp_sack_slack_ns", + .data = &init_net.ipv4.sysctl_tcp_comp_sack_slack_ns, + .maxlen = sizeof(unsigned long), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + }, + { .procname = "tcp_comp_sack_nr", .data = &init_net.ipv4.sysctl_tcp_comp_sack_nr, - .maxlen = sizeof(int), + .maxlen = sizeof(u8), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ZERO, - .extra2 = &comp_sack_nr_max, + }, + { + .procname = "tcp_reflect_tos", + .data = &init_net.ipv4.sysctl_tcp_reflect_tos, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, + { + .procname = "tcp_ehash_entries", + .data = &init_net.ipv4.sysctl_tcp_child_ehash_entries, + .mode = 0444, + .proc_handler = proc_tcp_ehash_entries, + }, + { + .procname = "tcp_child_ehash_entries", + .data = &init_net.ipv4.sysctl_tcp_child_ehash_entries, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_douintvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = &tcp_child_ehash_entries_max, }, { .procname = "udp_rmem_min", @@ -1336,6 +1375,15 @@ static struct ctl_table ipv4_net_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ONE }, + { + .procname = "fib_notify_on_flag_change", + .data = &init_net.ipv4.sysctl_fib_notify_on_flag_change, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, + }, { } }; @@ -1351,9 +1399,19 @@ static __net_init int ipv4_sysctl_init_net(struct net *net) if (!table) goto err_alloc; - /* Update the variables to point into the current struct net */ - for (i = 0; i < ARRAY_SIZE(ipv4_net_table) - 1; i++) - table[i].data += (void *)net - (void *)&init_net; + for (i = 0; i < ARRAY_SIZE(ipv4_net_table) - 1; i++) { + if (table[i].data) { + /* Update the variables to point into + * the current struct net + */ + table[i].data += (void *)net - (void *)&init_net; + } else { + /* Entries without data pointer are global; + * Make them read-only in non-init_net ns + */ + table[i].mode &= ~0222; + } + } } net->ipv4.ipv4_hdr = register_net_sysctl(net, "net/ipv4", table); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index eb2d80519f8e..54836a6b81d6 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -260,13 +260,13 @@ #include <linux/random.h> #include <linux/memblock.h> #include <linux/highmem.h> -#include <linux/swap.h> #include <linux/cache.h> #include <linux/err.h> #include <linux/time.h> #include <linux/slab.h> #include <linux/errqueue.h> #include <linux/static_key.h> +#include <linux/btf.h> #include <net/icmp.h> #include <net/inet_common.h> @@ -280,14 +280,22 @@ #include <asm/ioctls.h> #include <net/busy_poll.h> -struct percpu_counter tcp_orphan_count; -EXPORT_SYMBOL_GPL(tcp_orphan_count); +/* Track pending CMSGs. */ +enum { + TCP_CMSG_INQ = 1, + TCP_CMSG_TS = 2 +}; + +DEFINE_PER_CPU(unsigned int, tcp_orphan_count); +EXPORT_PER_CPU_SYMBOL_GPL(tcp_orphan_count); long sysctl_tcp_mem[3] __read_mostly; EXPORT_SYMBOL(sysctl_tcp_mem); -atomic_long_t tcp_memory_allocated; /* Current allocated memory. */ +atomic_long_t tcp_memory_allocated ____cacheline_aligned_in_smp; /* Current allocated memory. */ EXPORT_SYMBOL(tcp_memory_allocated); +DEFINE_PER_CPU(int, tcp_memory_per_cpu_fw_alloc); +EXPORT_PER_CPU_SYMBOL_GPL(tcp_memory_per_cpu_fw_alloc); #if IS_ENABLED(CONFIG_SMC) DEFINE_STATIC_KEY_FALSE(tcp_have_smc); @@ -297,7 +305,7 @@ EXPORT_SYMBOL(tcp_have_smc); /* * Current number of TCP sockets. */ -struct percpu_counter tcp_sockets_allocated; +struct percpu_counter tcp_sockets_allocated ____cacheline_aligned_in_smp; EXPORT_SYMBOL(tcp_sockets_allocated); /* @@ -318,11 +326,6 @@ struct tcp_splice_state { unsigned long tcp_memory_pressure __read_mostly; EXPORT_SYMBOL_GPL(tcp_memory_pressure); -DEFINE_STATIC_KEY_FALSE(tcp_rx_skb_cache_key); -EXPORT_SYMBOL(tcp_rx_skb_cache_key); - -DEFINE_STATIC_KEY_FALSE(tcp_tx_skb_cache_key); - void tcp_enter_memory_pressure(struct sock *sk) { unsigned long val; @@ -418,6 +421,8 @@ void tcp_init_sock(struct sock *sk) INIT_LIST_HEAD(&tp->tsorted_sent_queue); icsk->icsk_rto = TCP_TIMEOUT_INIT; + icsk->icsk_rto_min = TCP_RTO_MIN; + icsk->icsk_delack_max = TCP_DELACK_MAX; tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT); minmax_reset(&tp->rtt_min, tcp_jiffies32, ~0U); @@ -426,7 +431,7 @@ void tcp_init_sock(struct sock *sk) * algorithms that we must have the following bandaid to talk * efficiently to them. -DaveM */ - tp->snd_cwnd = TCP_INIT_CWND; + tcp_snd_cwnd_set(tp, TCP_INIT_CWND); /* There's a bubble in the pipe until at least the first ACK. */ tp->app_limited = ~0U; @@ -438,7 +443,7 @@ void tcp_init_sock(struct sock *sk) tp->snd_cwnd_clamp = ~0; tp->mss_cache = TCP_MSS_DEFAULT; - tp->reordering = sock_net(sk)->ipv4.sysctl_tcp_reordering; + tp->reordering = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reordering); tcp_assign_congestion_control(sk); tp->tsoffset = 0; @@ -449,11 +454,11 @@ void tcp_init_sock(struct sock *sk) icsk->icsk_sync_mss = tcp_sync_mss; - WRITE_ONCE(sk->sk_sndbuf, sock_net(sk)->ipv4.sysctl_tcp_wmem[1]); - WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]); + WRITE_ONCE(sk->sk_sndbuf, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[1])); + WRITE_ONCE(sk->sk_rcvbuf, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[1])); + set_bit(SOCK_SUPPORT_ZC, &sk->sk_socket->flags); sk_sockets_allocated_inc(sk); - sk->sk_route_forced_caps = NETIF_F_GSO; } EXPORT_SYMBOL(tcp_init_sock); @@ -473,12 +478,11 @@ static void tcp_tx_timestamp(struct sock *sk, u16 tsflags) } } -static inline bool tcp_stream_is_readable(const struct tcp_sock *tp, - int target, struct sock *sk) +static bool tcp_stream_is_readable(struct sock *sk, int target) { - return (READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->copied_seq) >= target) || - (sk->sk_prot->stream_memory_read ? - sk->sk_prot->stream_memory_read(sk) : false); + if (tcp_epollin_ready(sk, target)) + return true; + return sk_is_readable(sk); } /* @@ -544,17 +548,18 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) if (state != TCP_SYN_SENT && (state != TCP_SYN_RECV || rcu_access_pointer(tp->fastopen_rsk))) { int target = sock_rcvlowat(sk, 0, INT_MAX); + u16 urg_data = READ_ONCE(tp->urg_data); - if (READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq) && - !sock_flag(sk, SOCK_URGINLINE) && - tp->urg_data) + if (unlikely(urg_data) && + READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq) && + !sock_flag(sk, SOCK_URGINLINE)) target++; - if (tcp_stream_is_readable(tp, target, sk)) + if (tcp_stream_is_readable(sk, target)) mask |= EPOLLIN | EPOLLRDNORM; if (!(sk->sk_shutdown & SEND_SHUTDOWN)) { - if (sk_stream_is_writeable(sk)) { + if (__sk_stream_is_writeable(sk, 1)) { mask |= EPOLLOUT | EPOLLWRNORM; } else { /* send SIGIO later */ sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk); @@ -566,13 +571,13 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) * pairs with the input side. */ smp_mb__after_atomic(); - if (sk_stream_is_writeable(sk)) + if (__sk_stream_is_writeable(sk, 1)) mask |= EPOLLOUT | EPOLLWRNORM; } } else mask |= EPOLLOUT | EPOLLWRNORM; - if (tp->urg_data & TCP_URG_VALID) + if (urg_data & TCP_URG_VALID) mask |= EPOLLPRI; } else if (state == TCP_SYN_SENT && inet_sk(sk)->defer_connect) { /* Active TCP fastopen socket with defer_connect @@ -606,7 +611,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) unlock_sock_fast(sk, slow); break; case SIOCATMARK: - answ = tp->urg_data && + answ = READ_ONCE(tp->urg_data) && READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq); break; case SIOCOUTQ: @@ -636,7 +641,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) } EXPORT_SYMBOL(tcp_ioctl); -static inline void tcp_mark_push(struct tcp_sock *tp, struct sk_buff *skb) +void tcp_mark_push(struct tcp_sock *tp, struct sk_buff *skb) { TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_PSH; tp->pushed_seq = tp->write_seq; @@ -647,15 +652,13 @@ static inline bool forced_push(const struct tcp_sock *tp) return after(tp->write_seq, tp->pushed_seq + (tp->max_window >> 1)); } -static void skb_entail(struct sock *sk, struct sk_buff *skb) +void tcp_skb_entail(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_skb_cb *tcb = TCP_SKB_CB(skb); - skb->csum = 0; tcb->seq = tcb->end_seq = tp->write_seq; tcb->tcp_flags = TCPHDR_ACK; - tcb->sacked = 0; __skb_header_release(skb); tcp_add_write_queue_tail(sk, skb); sk_wmem_queued_add(sk, skb->truesize); @@ -686,9 +689,10 @@ static bool tcp_should_autocork(struct sock *sk, struct sk_buff *skb, int size_goal) { return skb->len < size_goal && - sock_net(sk)->ipv4.sysctl_tcp_autocorking && + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_autocorking) && !tcp_rtx_queue_empty(sk) && - refcount_read(&sk->sk_wmem_alloc) > skb->truesize; + refcount_read(&sk->sk_wmem_alloc) > skb->truesize && + tcp_skb_can_collapse_to(skb); } void tcp_push(struct sock *sk, int flags, int mss_now, @@ -850,33 +854,16 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos, } EXPORT_SYMBOL(tcp_splice_read); -struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp, - bool force_schedule) +struct sk_buff *tcp_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp, + bool force_schedule) { struct sk_buff *skb; - if (likely(!size)) { - skb = sk->sk_tx_skb_cache; - if (skb) { - skb->truesize = SKB_TRUESIZE(skb_end_offset(skb)); - sk->sk_tx_skb_cache = NULL; - pskb_trim(skb, 0); - INIT_LIST_HEAD(&skb->tcp_tsorted_anchor); - skb_shinfo(skb)->tx_flags = 0; - memset(TCP_SKB_CB(skb), 0, sizeof(struct tcp_skb_cb)); - return skb; - } - } - /* The TCP header must be at least 32-bit aligned. */ - size = ALIGN(size, 4); - - if (unlikely(tcp_under_memory_pressure(sk))) - sk_mem_reclaim_partial(sk); - - skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp); + skb = alloc_skb_fclone(size + MAX_TCP_HEADER, gfp); if (likely(skb)) { bool mem_scheduled; + skb->truesize = SKB_TRUESIZE(skb_end_offset(skb)); if (force_schedule) { mem_scheduled = true; sk_forced_mem_schedule(sk, skb->truesize); @@ -884,12 +871,8 @@ struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp, mem_scheduled = sk_wmem_schedule(sk, skb->truesize); } if (likely(mem_scheduled)) { - skb_reserve(skb, sk->sk_prot->max_header); - /* - * Make sure that we have exactly size bytes - * available to the caller, no more, no less. - */ - skb->reserved_tailroom = skb->end - skb->tail - size; + skb_reserve(skb, MAX_TCP_HEADER); + skb->ip_summed = CHECKSUM_PARTIAL; INIT_LIST_HEAD(&skb->tcp_tsorted_anchor); return skb; } @@ -911,8 +894,7 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now, return mss_now; /* Note : tcp_tso_autosize() will eventually split this later */ - new_size_goal = sk->sk_gso_max_size - 1 - MAX_TCP_HEADER; - new_size_goal = tcp_bound_to_half_wnd(tp, new_size_goal); + new_size_goal = tcp_bound_to_half_wnd(tp, sk->sk_gso_max_size); /* We try hard to avoid divides here */ size_goal = tp->gso_segs * mss_now; @@ -942,14 +924,115 @@ int tcp_send_mss(struct sock *sk, int *size_goal, int flags) * importantly be able to generate EPOLLOUT for Edge Trigger epoll() * users. */ -static void tcp_remove_empty_skb(struct sock *sk, struct sk_buff *skb) +void tcp_remove_empty_skb(struct sock *sk) { - if (skb && !skb->len) { + struct sk_buff *skb = tcp_write_queue_tail(sk); + + if (skb && TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) { tcp_unlink_write_queue(skb, sk); if (tcp_write_queue_empty(sk)) tcp_chrono_stop(sk, TCP_CHRONO_BUSY); - sk_wmem_free_skb(sk, skb); + tcp_wmem_free_skb(sk, skb); + } +} + +/* skb changing from pure zc to mixed, must charge zc */ +static int tcp_downgrade_zcopy_pure(struct sock *sk, struct sk_buff *skb) +{ + if (unlikely(skb_zcopy_pure(skb))) { + u32 extra = skb->truesize - + SKB_TRUESIZE(skb_end_offset(skb)); + + if (!sk_wmem_schedule(sk, extra)) + return -ENOMEM; + + sk_mem_charge(sk, extra); + skb_shinfo(skb)->flags &= ~SKBFL_PURE_ZEROCOPY; + } + return 0; +} + + +static int tcp_wmem_schedule(struct sock *sk, int copy) +{ + int left; + + if (likely(sk_wmem_schedule(sk, copy))) + return copy; + + /* We could be in trouble if we have nothing queued. + * Use whatever is left in sk->sk_forward_alloc and tcp_wmem[0] + * to guarantee some progress. + */ + left = sock_net(sk)->ipv4.sysctl_tcp_wmem[0] - sk->sk_wmem_queued; + if (left > 0) + sk_forced_mem_schedule(sk, min(left, copy)); + return min(copy, sk->sk_forward_alloc); +} + +static struct sk_buff *tcp_build_frag(struct sock *sk, int size_goal, int flags, + struct page *page, int offset, size_t *size) +{ + struct sk_buff *skb = tcp_write_queue_tail(sk); + struct tcp_sock *tp = tcp_sk(sk); + bool can_coalesce; + int copy, i; + + if (!skb || (copy = size_goal - skb->len) <= 0 || + !tcp_skb_can_collapse_to(skb)) { +new_segment: + if (!sk_stream_memory_free(sk)) + return NULL; + + skb = tcp_stream_alloc_skb(sk, 0, sk->sk_allocation, + tcp_rtx_and_write_queues_empty(sk)); + if (!skb) + return NULL; + +#ifdef CONFIG_TLS_DEVICE + skb->decrypted = !!(flags & MSG_SENDPAGE_DECRYPTED); +#endif + tcp_skb_entail(sk, skb); + copy = size_goal; + } + + if (copy > *size) + copy = *size; + + i = skb_shinfo(skb)->nr_frags; + can_coalesce = skb_can_coalesce(skb, i, page, offset); + if (!can_coalesce && i >= READ_ONCE(sysctl_max_skb_frags)) { + tcp_mark_push(tp, skb); + goto new_segment; + } + if (tcp_downgrade_zcopy_pure(sk, skb)) + return NULL; + + copy = tcp_wmem_schedule(sk, copy); + if (!copy) + return NULL; + + if (can_coalesce) { + skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy); + } else { + get_page(page); + skb_fill_page_desc_noacc(skb, i, page, offset, copy); } + + if (!(flags & MSG_NO_SHARED_FRAGS)) + skb_shinfo(skb)->flags |= SKBFL_SHARED_FRAG; + + skb->len += copy; + skb->data_len += copy; + skb->truesize += copy; + sk_wmem_queued_add(sk, copy); + sk_mem_charge(sk, copy); + WRITE_ONCE(tp->write_seq, tp->write_seq + copy); + TCP_SKB_CB(skb)->end_seq += copy; + tcp_skb_pcount_set(skb, 0); + + *size = copy; + return skb; } ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset, @@ -962,7 +1045,8 @@ ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset, long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT); if (IS_ENABLED(CONFIG_DEBUG_VM) && - WARN_ONCE(PageSlab(page), "page must not be a Slab one")) + WARN_ONCE(!sendpage_ok(page), + "page must not be a Slab one and have page_count > 0")) return -EINVAL; /* Wait for a connection to finish. One exception is TCP Fast Open @@ -986,59 +1070,12 @@ ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset, goto out_err; while (size > 0) { - struct sk_buff *skb = tcp_write_queue_tail(sk); - int copy, i; - bool can_coalesce; - - if (!skb || (copy = size_goal - skb->len) <= 0 || - !tcp_skb_can_collapse_to(skb)) { -new_segment: - if (!sk_stream_memory_free(sk)) - goto wait_for_sndbuf; - - skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation, - tcp_rtx_and_write_queues_empty(sk)); - if (!skb) - goto wait_for_memory; - -#ifdef CONFIG_TLS_DEVICE - skb->decrypted = !!(flags & MSG_SENDPAGE_DECRYPTED); -#endif - skb_entail(sk, skb); - copy = size_goal; - } + struct sk_buff *skb; + size_t copy = size; - if (copy > size) - copy = size; - - i = skb_shinfo(skb)->nr_frags; - can_coalesce = skb_can_coalesce(skb, i, page, offset); - if (!can_coalesce && i >= sysctl_max_skb_frags) { - tcp_mark_push(tp, skb); - goto new_segment; - } - if (!sk_wmem_schedule(sk, copy)) - goto wait_for_memory; - - if (can_coalesce) { - skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy); - } else { - get_page(page); - skb_fill_page_desc(skb, i, page, offset, copy); - } - - if (!(flags & MSG_NO_SHARED_FRAGS)) - skb_shinfo(skb)->tx_flags |= SKBTX_SHARED_FRAG; - - skb->len += copy; - skb->data_len += copy; - skb->truesize += copy; - sk_wmem_queued_add(sk, copy); - sk_mem_charge(sk, copy); - skb->ip_summed = CHECKSUM_PARTIAL; - WRITE_ONCE(tp->write_seq, tp->write_seq + copy); - TCP_SKB_CB(skb)->end_seq += copy; - tcp_skb_pcount_set(skb, 0); + skb = tcp_build_frag(sk, size_goal, flags, page, offset, ©); + if (!skb) + goto wait_for_space; if (!copied) TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_PSH; @@ -1059,9 +1096,8 @@ new_segment: tcp_push_one(sk, mss_now); continue; -wait_for_sndbuf: +wait_for_space: set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); -wait_for_memory: tcp_push(sk, flags & ~MSG_MORE, mss_now, TCP_NAGLE_PUSH, size_goal); @@ -1081,7 +1117,7 @@ out: return copied; do_error: - tcp_remove_empty_skb(sk, tcp_write_queue_tail(sk)); + tcp_remove_empty_skb(sk); if (copied) goto out; out_err: @@ -1127,16 +1163,16 @@ void tcp_free_fastopen_req(struct tcp_sock *tp) } } -static int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, - int *copied, size_t size, - struct ubuf_info *uarg) +int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, int *copied, + size_t size, struct ubuf_info *uarg) { struct tcp_sock *tp = tcp_sk(sk); struct inet_sock *inet = inet_sk(sk); struct sockaddr *uaddr = msg->msg_name; int err, flags; - if (!(sock_net(sk)->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) || + if (!(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen) & + TFO_CLIENT_ENABLE) || (uaddr && msg->msg_namelen >= sizeof(uaddr->sa_family) && uaddr->sa_family == AF_UNSPEC)) return -EOPNOTSUPP; @@ -1188,17 +1224,23 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) flags = msg->msg_flags; - if (flags & MSG_ZEROCOPY && size && sock_flag(sk, SOCK_ZEROCOPY)) { + if ((flags & MSG_ZEROCOPY) && size) { skb = tcp_write_queue_tail(sk); - uarg = sock_zerocopy_realloc(sk, size, skb_zcopy(skb)); - if (!uarg) { - err = -ENOBUFS; - goto out_err; - } - zc = sk->sk_route_caps & NETIF_F_SG; - if (!zc) - uarg->zerocopy = 0; + if (msg->msg_ubuf) { + uarg = msg->msg_ubuf; + net_zcopy_get(uarg); + zc = sk->sk_route_caps & NETIF_F_SG; + } else if (sock_flag(sk, SOCK_ZEROCOPY)) { + uarg = msg_zerocopy_realloc(sk, size, skb_zcopy(skb)); + if (!uarg) { + err = -ENOBUFS; + goto out_err; + } + zc = sk->sk_route_caps & NETIF_F_SG; + if (!zc) + uarg_to_msgzc(uarg)->zerocopy = 0; + } } if (unlikely(flags & MSG_FASTOPEN || inet_sk(sk)->defer_connect) && @@ -1272,7 +1314,7 @@ restart: new_segment: if (!sk_stream_memory_free(sk)) - goto wait_for_sndbuf; + goto wait_for_space; if (unlikely(process_backlog >= 16)) { process_backlog = 0; @@ -1280,15 +1322,14 @@ new_segment: goto restart; } first_skb = tcp_rtx_and_write_queues_empty(sk); - skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation, - first_skb); + skb = tcp_stream_alloc_skb(sk, 0, sk->sk_allocation, + first_skb); if (!skb) - goto wait_for_memory; + goto wait_for_space; process_backlog++; - skb->ip_summed = CHECKSUM_PARTIAL; - skb_entail(sk, skb); + tcp_skb_entail(sk, skb); copy = size_goal; /* All packets are restored as if they have @@ -1303,24 +1344,17 @@ new_segment: if (copy > msg_data_left(msg)) copy = msg_data_left(msg); - /* Where to copy to? */ - if (skb_availroom(skb) > 0 && !zc) { - /* We have some space in skb head. Superb! */ - copy = min_t(int, copy, skb_availroom(skb)); - err = skb_add_data_nocache(sk, skb, &msg->msg_iter, copy); - if (err) - goto do_fault; - } else if (!zc) { + if (!zc) { bool merge = true; int i = skb_shinfo(skb)->nr_frags; struct page_frag *pfrag = sk_page_frag(sk); if (!sk_page_frag_refill(sk, pfrag)) - goto wait_for_memory; + goto wait_for_space; if (!skb_can_coalesce(skb, i, pfrag->page, pfrag->offset)) { - if (i >= sysctl_max_skb_frags) { + if (i >= READ_ONCE(sysctl_max_skb_frags)) { tcp_mark_push(tp, skb); goto new_segment; } @@ -1329,8 +1363,15 @@ new_segment: copy = min_t(int, copy, pfrag->size - pfrag->offset); - if (!sk_wmem_schedule(sk, copy)) - goto wait_for_memory; + if (unlikely(skb_zcopy_pure(skb) || skb_zcopy_managed(skb))) { + if (tcp_downgrade_zcopy_pure(sk, skb)) + goto wait_for_space; + skb_zcopy_downgrade_managed(skb); + } + + copy = tcp_wmem_schedule(sk, copy); + if (!copy) + goto wait_for_space; err = skb_copy_to_page_nocache(sk, &msg->msg_iter, skb, pfrag->page, @@ -1349,6 +1390,18 @@ new_segment: } pfrag->offset += copy; } else { + /* First append to a fragless skb builds initial + * pure zerocopy skb + */ + if (!skb->len) + skb_shinfo(skb)->flags |= SKBFL_PURE_ZEROCOPY; + + if (!skb_zcopy_pure(skb)) { + copy = tcp_wmem_schedule(sk, copy); + if (!copy) + goto wait_for_space; + } + err = skb_zerocopy_iter_stream(sk, skb, msg, copy, uarg); if (err == -EMSGSIZE || err == -EEXIST) { tcp_mark_push(tp, skb); @@ -1383,9 +1436,8 @@ new_segment: tcp_push_one(sk, mss_now); continue; -wait_for_sndbuf: +wait_for_space: set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); -wait_for_memory: if (copied) tcp_push(sk, flags & ~MSG_MORE, mss_now, TCP_NAGLE_PUSH, size_goal); @@ -1403,18 +1455,16 @@ out: tcp_push(sk, flags, mss_now, tp->nonagle, size_goal); } out_nopush: - sock_zerocopy_put(uarg); + net_zcopy_put(uarg); return copied + copied_syn; do_error: - skb = tcp_write_queue_tail(sk); -do_fault: - tcp_remove_empty_skb(sk, skb); + tcp_remove_empty_skb(sk); if (copied + copied_syn) goto out; out_err: - sock_zerocopy_put_abort(uarg, true); + net_zcopy_put_abort(uarg, true); err = sk_stream_error(sk, flags, err); /* make sure we wake any epoll edge trigger waiter */ if (unlikely(tcp_rtx_and_write_queues_empty(sk) && err == -EAGAIN)) { @@ -1459,7 +1509,7 @@ static int tcp_recv_urg(struct sock *sk, struct msghdr *msg, int len, int flags) char c = tp->urg_data; if (!(flags & MSG_PEEK)) - tp->urg_data = TCP_URG_READ; + WRITE_ONCE(tp->urg_data, TCP_URG_READ); /* Read urgent data. */ msg->msg_flags |= MSG_OOB; @@ -1517,23 +1567,15 @@ static int tcp_peek_sndq(struct sock *sk, struct msghdr *msg, int len) * calculation of whether or not we must ACK for the sake of * a window update. */ -static void tcp_cleanup_rbuf(struct sock *sk, int copied) +static void __tcp_cleanup_rbuf(struct sock *sk, int copied) { struct tcp_sock *tp = tcp_sk(sk); bool time_to_ack = false; - struct sk_buff *skb = skb_peek(&sk->sk_receive_queue); - - WARN(skb && !before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq), - "cleanup rbuf bug: copied %X seq %X rcvnxt %X\n", - tp->copied_seq, TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt); - if (inet_csk_ack_scheduled(sk)) { const struct inet_connection_sock *icsk = inet_csk(sk); - /* Delayed ACKs frequently hit locked sockets during bulk - * receive. */ - if (icsk->icsk_ack.blocked || - /* Once-per-two-segments ACK was not sent by tcp_input.c */ + + if (/* Once-per-two-segments ACK was not sent by tcp_input.c */ tp->rcv_nxt - tp->rcv_wup > icsk->icsk_ack.rcv_mss || /* * If this read emptied read buffer, we send ACK, if @@ -1575,7 +1617,30 @@ static void tcp_cleanup_rbuf(struct sock *sk, int copied) tcp_send_ack(sk); } -static struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off) +void tcp_cleanup_rbuf(struct sock *sk, int copied) +{ + struct sk_buff *skb = skb_peek(&sk->sk_receive_queue); + struct tcp_sock *tp = tcp_sk(sk); + + WARN(skb && !before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq), + "cleanup rbuf bug: copied %X seq %X rcvnxt %X\n", + tp->copied_seq, TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt); + __tcp_cleanup_rbuf(sk, copied); +} + +static void tcp_eat_recv_skb(struct sock *sk, struct sk_buff *skb) +{ + __skb_unlink(skb, &sk->sk_receive_queue); + if (likely(skb->destructor == sock_rfree)) { + sock_rfree(skb); + skb->destructor = NULL; + skb->sk = NULL; + return skb_attempt_defer_free(skb); + } + __kfree_skb(skb); +} + +struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off) { struct sk_buff *skb; u32 offset; @@ -1594,10 +1659,11 @@ static struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off) * splitted a fat GRO packet, while we released socket lock * in skb_splice_bits() */ - sk_eat_skb(sk, skb); + tcp_eat_recv_skb(sk, skb); } return NULL; } +EXPORT_SYMBOL(tcp_recv_skb); /* * This routine provides an alternative to tcp_recvmsg() for routines @@ -1628,7 +1694,7 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, len = skb->len - offset; /* Stop reading if we hit a patch of urgent data */ - if (tp->urg_data) { + if (unlikely(tp->urg_data)) { u32 urg_offset = tp->urg_seq - seq; if (urg_offset < len) len = urg_offset; @@ -1640,11 +1706,13 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, if (!copied) copied = used; break; - } else if (used <= len) { - seq += used; - copied += used; - offset += used; } + if (WARN_ON_ONCE(used > len)) + used = len; + seq += used; + copied += used; + offset += used; + /* If recv_actor drops the lock (e.g. TCP splice * receive) the skb pointer might be invalid when * getting here: tcp_collapse might have deleted it @@ -1660,11 +1728,11 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, continue; } if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) { - sk_eat_skb(sk, skb); + tcp_eat_recv_skb(sk, skb); ++seq; break; } - sk_eat_skb(sk, skb); + tcp_eat_recv_skb(sk, skb); if (!desc->count) break; WRITE_ONCE(tp->copied_seq, seq); @@ -1682,6 +1750,90 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, } EXPORT_SYMBOL(tcp_read_sock); +int tcp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) +{ + struct tcp_sock *tp = tcp_sk(sk); + u32 seq = tp->copied_seq; + struct sk_buff *skb; + int copied = 0; + u32 offset; + + if (sk->sk_state == TCP_LISTEN) + return -ENOTCONN; + + while ((skb = tcp_recv_skb(sk, seq, &offset)) != NULL) { + u8 tcp_flags; + int used; + + __skb_unlink(skb, &sk->sk_receive_queue); + WARN_ON_ONCE(!skb_set_owner_sk_safe(skb, sk)); + tcp_flags = TCP_SKB_CB(skb)->tcp_flags; + used = recv_actor(sk, skb); + consume_skb(skb); + if (used < 0) { + if (!copied) + copied = used; + break; + } + seq += used; + copied += used; + + if (tcp_flags & TCPHDR_FIN) { + ++seq; + break; + } + } + WRITE_ONCE(tp->copied_seq, seq); + + tcp_rcv_space_adjust(sk); + + /* Clean up data we have read: This will do ACK frames. */ + if (copied > 0) + __tcp_cleanup_rbuf(sk, copied); + + return copied; +} +EXPORT_SYMBOL(tcp_read_skb); + +void tcp_read_done(struct sock *sk, size_t len) +{ + struct tcp_sock *tp = tcp_sk(sk); + u32 seq = tp->copied_seq; + struct sk_buff *skb; + size_t left; + u32 offset; + + if (sk->sk_state == TCP_LISTEN) + return; + + left = len; + while (left && (skb = tcp_recv_skb(sk, seq, &offset)) != NULL) { + int used; + + used = min_t(size_t, skb->len - offset, left); + seq += used; + left -= used; + + if (skb->len > offset + used) + break; + + if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) { + tcp_eat_recv_skb(sk, skb); + ++seq; + break; + } + tcp_eat_recv_skb(sk, skb); + } + WRITE_ONCE(tp->copied_seq, seq); + + tcp_rcv_space_adjust(sk); + + /* Clean up data we have read: This will do ACK frames. */ + if (left != len) + tcp_cleanup_rbuf(sk, len - left); +} +EXPORT_SYMBOL(tcp_read_done); + int tcp_peek_len(struct socket *sock) { return tcp_inq(sock->sk); @@ -1696,7 +1848,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val) if (sk->sk_userlocks & SOCK_RCVBUF_LOCK) cap = sk->sk_rcvbuf >> 1; else - cap = sock_net(sk)->ipv4.sysctl_tcp_rmem[2] >> 1; + cap = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1; val = min(val, cap); WRITE_ONCE(sk->sk_rcvlowat, val ? : 1); @@ -1715,6 +1867,20 @@ int tcp_set_rcvlowat(struct sock *sk, int val) } EXPORT_SYMBOL(tcp_set_rcvlowat); +void tcp_update_recv_tstamps(struct sk_buff *skb, + struct scm_timestamping_internal *tss) +{ + if (skb->tstamp) + tss->ts[0] = ktime_to_timespec64(skb->tstamp); + else + tss->ts[0] = (struct timespec64) {0}; + + if (skb_hwtstamps(skb)->hwtstamp) + tss->ts[2] = ktime_to_timespec64(skb_hwtstamps(skb)->hwtstamp); + else + tss->ts[2] = (struct timespec64) {0}; +} + #ifdef CONFIG_MMU static const struct vm_operations_struct tcp_vm_ops = { }; @@ -1726,7 +1892,7 @@ int tcp_mmap(struct file *file, struct socket *sock, return -EPERM; vma->vm_flags &= ~(VM_MAYWRITE | VM_MAYEXEC); - /* Instruct vm_insert_page() to not down_read(mmap_sem) */ + /* Instruct vm_insert_page() to not mmap_read_lock(mm) */ vma->vm_flags |= VM_MIXEDMAP; vma->vm_ops = &tcp_vm_ops; @@ -1734,18 +1900,309 @@ int tcp_mmap(struct file *file, struct socket *sock, } EXPORT_SYMBOL(tcp_mmap); +static skb_frag_t *skb_advance_to_frag(struct sk_buff *skb, u32 offset_skb, + u32 *offset_frag) +{ + skb_frag_t *frag; + + if (unlikely(offset_skb >= skb->len)) + return NULL; + + offset_skb -= skb_headlen(skb); + if ((int)offset_skb < 0 || skb_has_frag_list(skb)) + return NULL; + + frag = skb_shinfo(skb)->frags; + while (offset_skb) { + if (skb_frag_size(frag) > offset_skb) { + *offset_frag = offset_skb; + return frag; + } + offset_skb -= skb_frag_size(frag); + ++frag; + } + *offset_frag = 0; + return frag; +} + +static bool can_map_frag(const skb_frag_t *frag) +{ + return skb_frag_size(frag) == PAGE_SIZE && !skb_frag_off(frag); +} + +static int find_next_mappable_frag(const skb_frag_t *frag, + int remaining_in_skb) +{ + int offset = 0; + + if (likely(can_map_frag(frag))) + return 0; + + while (offset < remaining_in_skb && !can_map_frag(frag)) { + offset += skb_frag_size(frag); + ++frag; + } + return offset; +} + +static void tcp_zerocopy_set_hint_for_skb(struct sock *sk, + struct tcp_zerocopy_receive *zc, + struct sk_buff *skb, u32 offset) +{ + u32 frag_offset, partial_frag_remainder = 0; + int mappable_offset; + skb_frag_t *frag; + + /* worst case: skip to next skb. try to improve on this case below */ + zc->recv_skip_hint = skb->len - offset; + + /* Find the frag containing this offset (and how far into that frag) */ + frag = skb_advance_to_frag(skb, offset, &frag_offset); + if (!frag) + return; + + if (frag_offset) { + struct skb_shared_info *info = skb_shinfo(skb); + + /* We read part of the last frag, must recvmsg() rest of skb. */ + if (frag == &info->frags[info->nr_frags - 1]) + return; + + /* Else, we must at least read the remainder in this frag. */ + partial_frag_remainder = skb_frag_size(frag) - frag_offset; + zc->recv_skip_hint -= partial_frag_remainder; + ++frag; + } + + /* partial_frag_remainder: If part way through a frag, must read rest. + * mappable_offset: Bytes till next mappable frag, *not* counting bytes + * in partial_frag_remainder. + */ + mappable_offset = find_next_mappable_frag(frag, zc->recv_skip_hint); + zc->recv_skip_hint = mappable_offset + partial_frag_remainder; +} + +static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, + int flags, struct scm_timestamping_internal *tss, + int *cmsg_flags); +static int receive_fallback_to_copy(struct sock *sk, + struct tcp_zerocopy_receive *zc, int inq, + struct scm_timestamping_internal *tss) +{ + unsigned long copy_address = (unsigned long)zc->copybuf_address; + struct msghdr msg = {}; + struct iovec iov; + int err; + + zc->length = 0; + zc->recv_skip_hint = 0; + + if (copy_address != zc->copybuf_address) + return -EINVAL; + + err = import_single_range(READ, (void __user *)copy_address, + inq, &iov, &msg.msg_iter); + if (err) + return err; + + err = tcp_recvmsg_locked(sk, &msg, inq, MSG_DONTWAIT, + tss, &zc->msg_flags); + if (err < 0) + return err; + + zc->copybuf_len = err; + if (likely(zc->copybuf_len)) { + struct sk_buff *skb; + u32 offset; + + skb = tcp_recv_skb(sk, tcp_sk(sk)->copied_seq, &offset); + if (skb) + tcp_zerocopy_set_hint_for_skb(sk, zc, skb, offset); + } + return 0; +} + +static int tcp_copy_straggler_data(struct tcp_zerocopy_receive *zc, + struct sk_buff *skb, u32 copylen, + u32 *offset, u32 *seq) +{ + unsigned long copy_address = (unsigned long)zc->copybuf_address; + struct msghdr msg = {}; + struct iovec iov; + int err; + + if (copy_address != zc->copybuf_address) + return -EINVAL; + + err = import_single_range(READ, (void __user *)copy_address, + copylen, &iov, &msg.msg_iter); + if (err) + return err; + err = skb_copy_datagram_msg(skb, *offset, &msg, copylen); + if (err) + return err; + zc->recv_skip_hint -= copylen; + *offset += copylen; + *seq += copylen; + return (__s32)copylen; +} + +static int tcp_zc_handle_leftover(struct tcp_zerocopy_receive *zc, + struct sock *sk, + struct sk_buff *skb, + u32 *seq, + s32 copybuf_len, + struct scm_timestamping_internal *tss) +{ + u32 offset, copylen = min_t(u32, copybuf_len, zc->recv_skip_hint); + + if (!copylen) + return 0; + /* skb is null if inq < PAGE_SIZE. */ + if (skb) { + offset = *seq - TCP_SKB_CB(skb)->seq; + } else { + skb = tcp_recv_skb(sk, *seq, &offset); + if (TCP_SKB_CB(skb)->has_rxtstamp) { + tcp_update_recv_tstamps(skb, tss); + zc->msg_flags |= TCP_CMSG_TS; + } + } + + zc->copybuf_len = tcp_copy_straggler_data(zc, skb, copylen, &offset, + seq); + return zc->copybuf_len < 0 ? 0 : copylen; +} + +static int tcp_zerocopy_vm_insert_batch_error(struct vm_area_struct *vma, + struct page **pending_pages, + unsigned long pages_remaining, + unsigned long *address, + u32 *length, + u32 *seq, + struct tcp_zerocopy_receive *zc, + u32 total_bytes_to_map, + int err) +{ + /* At least one page did not map. Try zapping if we skipped earlier. */ + if (err == -EBUSY && + zc->flags & TCP_RECEIVE_ZEROCOPY_FLAG_TLB_CLEAN_HINT) { + u32 maybe_zap_len; + + maybe_zap_len = total_bytes_to_map - /* All bytes to map */ + *length + /* Mapped or pending */ + (pages_remaining * PAGE_SIZE); /* Failed map. */ + zap_page_range(vma, *address, maybe_zap_len); + err = 0; + } + + if (!err) { + unsigned long leftover_pages = pages_remaining; + int bytes_mapped; + + /* We called zap_page_range, try to reinsert. */ + err = vm_insert_pages(vma, *address, + pending_pages, + &pages_remaining); + bytes_mapped = PAGE_SIZE * (leftover_pages - pages_remaining); + *seq += bytes_mapped; + *address += bytes_mapped; + } + if (err) { + /* Either we were unable to zap, OR we zapped, retried an + * insert, and still had an issue. Either ways, pages_remaining + * is the number of pages we were unable to map, and we unroll + * some state we speculatively touched before. + */ + const int bytes_not_mapped = PAGE_SIZE * pages_remaining; + + *length -= bytes_not_mapped; + zc->recv_skip_hint += bytes_not_mapped; + } + return err; +} + +static int tcp_zerocopy_vm_insert_batch(struct vm_area_struct *vma, + struct page **pages, + unsigned int pages_to_map, + unsigned long *address, + u32 *length, + u32 *seq, + struct tcp_zerocopy_receive *zc, + u32 total_bytes_to_map) +{ + unsigned long pages_remaining = pages_to_map; + unsigned int pages_mapped; + unsigned int bytes_mapped; + int err; + + err = vm_insert_pages(vma, *address, pages, &pages_remaining); + pages_mapped = pages_to_map - (unsigned int)pages_remaining; + bytes_mapped = PAGE_SIZE * pages_mapped; + /* Even if vm_insert_pages fails, it may have partially succeeded in + * mapping (some but not all of the pages). + */ + *seq += bytes_mapped; + *address += bytes_mapped; + + if (likely(!err)) + return 0; + + /* Error: maybe zap and retry + rollback state for failed inserts. */ + return tcp_zerocopy_vm_insert_batch_error(vma, pages + pages_mapped, + pages_remaining, address, length, seq, zc, total_bytes_to_map, + err); +} + +#define TCP_VALID_ZC_MSG_FLAGS (TCP_CMSG_TS) +static void tcp_zc_finalize_rx_tstamp(struct sock *sk, + struct tcp_zerocopy_receive *zc, + struct scm_timestamping_internal *tss) +{ + unsigned long msg_control_addr; + struct msghdr cmsg_dummy; + + msg_control_addr = (unsigned long)zc->msg_control; + cmsg_dummy.msg_control = (void *)msg_control_addr; + cmsg_dummy.msg_controllen = + (__kernel_size_t)zc->msg_controllen; + cmsg_dummy.msg_flags = in_compat_syscall() + ? MSG_CMSG_COMPAT : 0; + cmsg_dummy.msg_control_is_user = true; + zc->msg_flags = 0; + if (zc->msg_control == msg_control_addr && + zc->msg_controllen == cmsg_dummy.msg_controllen) { + tcp_recv_timestamp(&cmsg_dummy, sk, tss); + zc->msg_control = (__u64) + ((uintptr_t)cmsg_dummy.msg_control); + zc->msg_controllen = + (__u64)cmsg_dummy.msg_controllen; + zc->msg_flags = (__u32)cmsg_dummy.msg_flags; + } +} + +#define TCP_ZEROCOPY_PAGE_BATCH_SIZE 32 static int tcp_zerocopy_receive(struct sock *sk, - struct tcp_zerocopy_receive *zc) + struct tcp_zerocopy_receive *zc, + struct scm_timestamping_internal *tss) { + u32 length = 0, offset, vma_len, avail_len, copylen = 0; unsigned long address = (unsigned long)zc->address; - u32 length = 0, seq, offset, zap_len; + struct page *pages[TCP_ZEROCOPY_PAGE_BATCH_SIZE]; + s32 copybuf_len = zc->copybuf_len; + struct tcp_sock *tp = tcp_sk(sk); const skb_frag_t *frags = NULL; + unsigned int pages_to_map = 0; struct vm_area_struct *vma; struct sk_buff *skb = NULL; - struct tcp_sock *tp; - int inq; + u32 seq = tp->copied_seq; + u32 total_bytes_to_map; + int inq = tcp_inq(sk); int ret; + zc->copybuf_len = 0; + zc->msg_flags = 0; + if (address & (PAGE_SIZE - 1) || address != zc->address) return -EINVAL; @@ -1754,28 +2211,44 @@ static int tcp_zerocopy_receive(struct sock *sk, sock_rps_record_flow(sk); - down_read(¤t->mm->mmap_sem); + if (inq && inq <= copybuf_len) + return receive_fallback_to_copy(sk, zc, inq, tss); - ret = -EINVAL; - vma = find_vma(current->mm, address); - if (!vma || vma->vm_start > address || vma->vm_ops != &tcp_vm_ops) - goto out; - zc->length = min_t(unsigned long, zc->length, vma->vm_end - address); - - tp = tcp_sk(sk); - seq = tp->copied_seq; - inq = tcp_inq(sk); - zc->length = min_t(u32, zc->length, inq); - zap_len = zc->length & ~(PAGE_SIZE - 1); - if (zap_len) { - zap_page_range(vma, address, zap_len); + if (inq < PAGE_SIZE) { + zc->length = 0; + zc->recv_skip_hint = inq; + if (!inq && sock_flag(sk, SOCK_DONE)) + return -EIO; + return 0; + } + + mmap_read_lock(current->mm); + + vma = vma_lookup(current->mm, address); + if (!vma || vma->vm_ops != &tcp_vm_ops) { + mmap_read_unlock(current->mm); + return -EINVAL; + } + vma_len = min_t(unsigned long, zc->length, vma->vm_end - address); + avail_len = min_t(u32, vma_len, inq); + total_bytes_to_map = avail_len & ~(PAGE_SIZE - 1); + if (total_bytes_to_map) { + if (!(zc->flags & TCP_RECEIVE_ZEROCOPY_FLAG_TLB_CLEAN_HINT)) + zap_page_range(vma, address, total_bytes_to_map); + zc->length = total_bytes_to_map; zc->recv_skip_hint = 0; } else { - zc->recv_skip_hint = zc->length; + zc->length = avail_len; + zc->recv_skip_hint = avail_len; } ret = 0; while (length + PAGE_SIZE <= zc->length) { + int mappable_offset; + struct page *page; + if (zc->recv_skip_hint < PAGE_SIZE) { + u32 offset_frag; + if (skb) { if (zc->recv_skip_hint > 0) break; @@ -1785,47 +2258,61 @@ static int tcp_zerocopy_receive(struct sock *sk, skb = tcp_recv_skb(sk, seq, &offset); } + if (TCP_SKB_CB(skb)->has_rxtstamp) { + tcp_update_recv_tstamps(skb, tss); + zc->msg_flags |= TCP_CMSG_TS; + } zc->recv_skip_hint = skb->len - offset; - offset -= skb_headlen(skb); - if ((int)offset < 0 || skb_has_frag_list(skb)) + frags = skb_advance_to_frag(skb, offset, &offset_frag); + if (!frags || offset_frag) break; - frags = skb_shinfo(skb)->frags; - while (offset) { - if (skb_frag_size(frags) > offset) - goto out; - offset -= skb_frag_size(frags); - frags++; - } } - if (skb_frag_size(frags) != PAGE_SIZE || skb_frag_off(frags)) { - int remaining = zc->recv_skip_hint; - while (remaining && (skb_frag_size(frags) != PAGE_SIZE || - skb_frag_off(frags))) { - remaining -= skb_frag_size(frags); - frags++; - } - zc->recv_skip_hint -= remaining; + mappable_offset = find_next_mappable_frag(frags, + zc->recv_skip_hint); + if (mappable_offset) { + zc->recv_skip_hint = mappable_offset; break; } - ret = vm_insert_page(vma, address + length, - skb_frag_page(frags)); - if (ret) - break; + page = skb_frag_page(frags); + prefetchw(page); + pages[pages_to_map++] = page; length += PAGE_SIZE; - seq += PAGE_SIZE; zc->recv_skip_hint -= PAGE_SIZE; frags++; + if (pages_to_map == TCP_ZEROCOPY_PAGE_BATCH_SIZE || + zc->recv_skip_hint < PAGE_SIZE) { + /* Either full batch, or we're about to go to next skb + * (and we cannot unroll failed ops across skbs). + */ + ret = tcp_zerocopy_vm_insert_batch(vma, pages, + pages_to_map, + &address, &length, + &seq, zc, + total_bytes_to_map); + if (ret) + goto out; + pages_to_map = 0; + } + } + if (pages_to_map) { + ret = tcp_zerocopy_vm_insert_batch(vma, pages, pages_to_map, + &address, &length, &seq, + zc, total_bytes_to_map); } out: - up_read(¤t->mm->mmap_sem); - if (length) { + mmap_read_unlock(current->mm); + /* Try to copy straggler data. */ + if (!ret) + copylen = tcp_zc_handle_leftover(zc, sk, skb, &seq, copybuf_len, tss); + + if (length + copylen) { WRITE_ONCE(tp->copied_seq, seq); tcp_rcv_space_adjust(sk); /* Clean up data we have read: This will do ACK frames. */ tcp_recv_skb(sk, seq, &offset); - tcp_cleanup_rbuf(sk, length); + tcp_cleanup_rbuf(sk, length + copylen); ret = 0; if (length == zc->length) zc->recv_skip_hint = 0; @@ -1838,23 +2325,9 @@ out: } #endif -static void tcp_update_recv_tstamps(struct sk_buff *skb, - struct scm_timestamping_internal *tss) -{ - if (skb->tstamp) - tss->ts[0] = ktime_to_timespec64(skb->tstamp); - else - tss->ts[0] = (struct timespec64) {0}; - - if (skb_hwtstamps(skb)->hwtstamp) - tss->ts[2] = ktime_to_timespec64(skb_hwtstamps(skb)->hwtstamp); - else - tss->ts[2] = (struct timespec64) {0}; -} - /* Similar to __sock_recv_timestamp, but does not require an skb */ -static void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk, - struct scm_timestamping_internal *tss) +void tcp_recv_timestamp(struct msghdr *msg, const struct sock *sk, + struct scm_timestamping_internal *tss) { int new_tstamp = sock_flag(sk, SOCK_TSTAMP_NEW); bool has_timestamping = false; @@ -1947,37 +2420,30 @@ static int tcp_inq_hint(struct sock *sk) * Probably, code can be easily improved even more. */ -int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, - int flags, int *addr_len) +static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, + int flags, struct scm_timestamping_internal *tss, + int *cmsg_flags) { struct tcp_sock *tp = tcp_sk(sk); int copied = 0; u32 peek_seq; u32 *seq; unsigned long used; - int err, inq; + int err; int target; /* Read at least this many bytes */ long timeo; struct sk_buff *skb, *last; u32 urg_hole = 0; - struct scm_timestamping_internal tss; - int cmsg_flags; - - if (unlikely(flags & MSG_ERRQUEUE)) - return inet_recv_error(sk, msg, len, addr_len); - - if (sk_can_busy_loop(sk) && skb_queue_empty_lockless(&sk->sk_receive_queue) && - (sk->sk_state == TCP_ESTABLISHED)) - sk_busy_loop(sk, nonblock); - - lock_sock(sk); err = -ENOTCONN; if (sk->sk_state == TCP_LISTEN) goto out; - cmsg_flags = tp->recvmsg_inq ? 1 : 0; - timeo = sock_rcvtimeo(sk, nonblock); + if (tp->recvmsg_inq) { + *cmsg_flags = TCP_CMSG_INQ; + msg->msg_get_inq = 1; + } + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); /* Urgent data needs to be handled specially. */ if (flags & MSG_OOB) @@ -2010,7 +2476,7 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, u32 offset; /* Are we at urgent data? Stop if we have read anything or have SIGURG pending. */ - if (tp->urg_data && tp->urg_seq == *seq) { + if (unlikely(tp->urg_data) && tp->urg_seq == *seq) { if (copied) break; if (signal_pending(current)) { @@ -2053,10 +2519,10 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, break; if (copied) { - if (sk->sk_err || + if (!timeo || + sk->sk_err || sk->sk_state == TCP_CLOSE || (sk->sk_shutdown & RCV_SHUTDOWN) || - !timeo || signal_pending(current)) break; } else { @@ -2090,13 +2556,11 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, } } - tcp_cleanup_rbuf(sk, copied); - if (copied >= target) { /* Do not sleep, just process backlog. */ - release_sock(sk); - lock_sock(sk); + __sk_flush_backlog(sk); } else { + tcp_cleanup_rbuf(sk, copied); sk_wait_data(sk, &timeo, last); } @@ -2116,7 +2580,7 @@ found_ok_skb: used = len; /* Do we have urgent data here? */ - if (tp->urg_data) { + if (unlikely(tp->urg_data)) { u32 urg_offset = tp->urg_seq - *seq; if (urg_offset < used) { if (!urg_offset) { @@ -2150,28 +2614,30 @@ found_ok_skb: tcp_rcv_space_adjust(sk); skip_copy: - if (tp->urg_data && after(tp->copied_seq, tp->urg_seq)) { - tp->urg_data = 0; + if (unlikely(tp->urg_data) && after(tp->copied_seq, tp->urg_seq)) { + WRITE_ONCE(tp->urg_data, 0); tcp_fast_path_check(sk); } - if (used + offset < skb->len) - continue; if (TCP_SKB_CB(skb)->has_rxtstamp) { - tcp_update_recv_tstamps(skb, &tss); - cmsg_flags |= 2; + tcp_update_recv_tstamps(skb, tss); + *cmsg_flags |= TCP_CMSG_TS; } + + if (used + offset < skb->len) + continue; + if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) goto found_fin_ok; if (!(flags & MSG_PEEK)) - sk_eat_skb(sk, skb); + tcp_eat_recv_skb(sk, skb); continue; found_fin_ok: /* Process the FIN. */ WRITE_ONCE(*seq, *seq + 1); if (!(flags & MSG_PEEK)) - sk_eat_skb(sk, skb); + tcp_eat_recv_skb(sk, skb); break; } while (len > 0); @@ -2181,22 +2647,9 @@ found_fin_ok: /* Clean up data we have read: This will do ACK frames. */ tcp_cleanup_rbuf(sk, copied); - - release_sock(sk); - - if (cmsg_flags) { - if (cmsg_flags & 2) - tcp_recv_timestamp(msg, sk, &tss); - if (cmsg_flags & 1) { - inq = tcp_inq_hint(sk); - put_cmsg(msg, SOL_TCP, TCP_CM_INQ, sizeof(inq), &inq); - } - } - return copied; out: - release_sock(sk); return err; recv_urg: @@ -2207,6 +2660,37 @@ recv_sndq: err = tcp_peek_sndq(sk, msg, len); goto out; } + +int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags, + int *addr_len) +{ + int cmsg_flags = 0, ret; + struct scm_timestamping_internal tss; + + if (unlikely(flags & MSG_ERRQUEUE)) + return inet_recv_error(sk, msg, len, addr_len); + + if (sk_can_busy_loop(sk) && + skb_queue_empty_lockless(&sk->sk_receive_queue) && + sk->sk_state == TCP_ESTABLISHED) + sk_busy_loop(sk, flags & MSG_DONTWAIT); + + lock_sock(sk); + ret = tcp_recvmsg_locked(sk, msg, len, flags, &tss, &cmsg_flags); + release_sock(sk); + + if ((cmsg_flags || msg->msg_get_inq) && ret >= 0) { + if (cmsg_flags & TCP_CMSG_TS) + tcp_recv_timestamp(msg, sk, &tss); + if (msg->msg_get_inq) { + msg->msg_inq = tcp_inq_hint(sk); + if (cmsg_flags & TCP_CMSG_INQ) + put_cmsg(msg, SOL_TCP, TCP_CM_INQ, + sizeof(msg->msg_inq), &msg->msg_inq); + } + } + return ret; +} EXPORT_SYMBOL(tcp_recvmsg); void tcp_set_state(struct sock *sk, int state) @@ -2234,6 +2718,17 @@ void tcp_set_state(struct sock *sk, int state) BUILD_BUG_ON((int)BPF_TCP_NEW_SYN_RECV != (int)TCP_NEW_SYN_RECV); BUILD_BUG_ON((int)BPF_TCP_MAX_STATES != (int)TCP_MAX_STATES); + /* bpf uapi header bpf.h defines an anonymous enum with values + * BPF_TCP_* used by bpf programs. Currently gcc built vmlinux + * is able to emit this enum in DWARF due to the above BUILD_BUG_ON. + * But clang built vmlinux does not have this enum in DWARF + * since clang removes the above code before generating IR/debuginfo. + * Let us explicitly emit the type debuginfo to ensure the + * above-mentioned anonymous enum in the vmlinux DWARF and hence BTF + * regardless of which compiler is used. + */ + BTF_TYPE_EMIT_ENUM(BPF_TCP_ESTABLISHED); + if (BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk), BPF_SOCK_OPS_STATE_CB_FLAG)) tcp_call_bpf_2arg(sk, BPF_SOCK_OPS_STATE_CB, oldstate, state); @@ -2251,7 +2746,7 @@ void tcp_set_state(struct sock *sk, int state) if (inet_csk(sk)->icsk_bind_hash && !(sk->sk_userlocks & SOCK_BINDPORT_LOCK)) inet_put_port(sk); - /* fall through */ + fallthrough; default: if (oldstate == TCP_ESTABLISHED) TCP_DEC_STATS(sock_net(sk), TCP_MIB_CURRESTAB); @@ -2323,11 +2818,37 @@ void tcp_shutdown(struct sock *sk, int how) } EXPORT_SYMBOL(tcp_shutdown); +int tcp_orphan_count_sum(void) +{ + int i, total = 0; + + for_each_possible_cpu(i) + total += per_cpu(tcp_orphan_count, i); + + return max(total, 0); +} + +static int tcp_orphan_cache; +static struct timer_list tcp_orphan_timer; +#define TCP_ORPHAN_TIMER_PERIOD msecs_to_jiffies(100) + +static void tcp_orphan_update(struct timer_list *unused) +{ + WRITE_ONCE(tcp_orphan_cache, tcp_orphan_count_sum()); + mod_timer(&tcp_orphan_timer, jiffies + TCP_ORPHAN_TIMER_PERIOD); +} + +static bool tcp_too_many_orphans(int shift) +{ + return READ_ONCE(tcp_orphan_cache) << shift > + READ_ONCE(sysctl_tcp_max_orphans); +} + bool tcp_check_oom(struct sock *sk, int shift) { bool too_many_orphans, out_of_socket_memory; - too_many_orphans = tcp_too_many_orphans(sk, shift); + too_many_orphans = tcp_too_many_orphans(shift); out_of_socket_memory = tcp_out_of_memory(sk); if (too_many_orphans) @@ -2337,13 +2858,12 @@ bool tcp_check_oom(struct sock *sk, int shift) return too_many_orphans || out_of_socket_memory; } -void tcp_close(struct sock *sk, long timeout) +void __tcp_close(struct sock *sk, long timeout) { struct sk_buff *skb; int data_was_unread = 0; int state; - lock_sock(sk); sk->sk_shutdown = SHUTDOWN_MASK; if (sk->sk_state == TCP_LISTEN) { @@ -2368,8 +2888,6 @@ void tcp_close(struct sock *sk, long timeout) __kfree_skb(skb); } - sk_mem_reclaim(sk); - /* If socket has been already reset (e.g. in tcp_reset()) - kill it. */ if (sk->sk_state == TCP_CLOSE) goto adjudge_to_death; @@ -2437,7 +2955,7 @@ adjudge_to_death: /* remove backlog if any, without releasing ownership. */ __release_sock(sk); - percpu_counter_inc(sk->sk_prot->orphan_count); + this_cpu_inc(tcp_orphan_count); /* Have we already been destroyed by a softirq or backlog? */ if (state != TCP_CLOSE && sk->sk_state == TCP_CLOSE) @@ -2477,7 +2995,6 @@ adjudge_to_death: } } if (sk->sk_state != TCP_CLOSE) { - sk_mem_reclaim(sk); if (tcp_check_oom(sk, 0)) { tcp_set_state(sk, TCP_CLOSE); tcp_send_active_reset(sk, GFP_ATOMIC); @@ -2507,6 +3024,12 @@ adjudge_to_death: out: bh_unlock_sock(sk); local_bh_enable(); +} + +void tcp_close(struct sock *sk, long timeout) +{ + lock_sock(sk); + __tcp_close(sk, timeout); release_sock(sk); sock_put(sk); } @@ -2534,7 +3057,7 @@ static void tcp_rtx_queue_purge(struct sock *sk) * list_del(&skb->tcp_tsorted_anchor) */ tcp_rtx_queue_unlink(skb, sk); - sk_wmem_free_skb(sk, skb); + tcp_wmem_free_skb(sk, skb); } } @@ -2545,16 +3068,10 @@ void tcp_write_queue_purge(struct sock *sk) tcp_chrono_stop(sk, TCP_CHRONO_BUSY); while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL) { tcp_skb_tsorted_anchor_cleanup(skb); - sk_wmem_free_skb(sk, skb); + tcp_wmem_free_skb(sk, skb); } tcp_rtx_queue_purge(sk); - skb = sk->sk_tx_skb_cache; - if (skb) { - __kfree_skb(skb); - sk->sk_tx_skb_cache = NULL; - } INIT_LIST_HEAD(&tcp_sk(sk)->tsorted_sent_queue); - sk_mem_reclaim(sk); tcp_clear_all_retrans_hints(tcp_sk(sk)); tcp_sk(sk)->packets_out = 0; inet_csk(sk)->icsk_backoff = 0; @@ -2589,12 +3106,8 @@ int tcp_disconnect(struct sock *sk, int flags) tcp_clear_xmit_timers(sk); __skb_queue_purge(&sk->sk_receive_queue); - if (sk->sk_rx_skb_cache) { - __kfree_skb(sk->sk_rx_skb_cache); - sk->sk_rx_skb_cache = NULL; - } WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); - tp->urg_data = 0; + WRITE_ONCE(tp->urg_data, 0); tcp_write_queue_purge(sk); tcp_fastopen_active_disable_ofo_check(sk); skb_rbtree_purge(&tp->out_of_order_queue); @@ -2617,13 +3130,22 @@ int tcp_disconnect(struct sock *sk, int flags) icsk->icsk_backoff = 0; icsk->icsk_probes_out = 0; + icsk->icsk_probes_tstamp = 0; icsk->icsk_rto = TCP_TIMEOUT_INIT; + icsk->icsk_rto_min = TCP_RTO_MIN; + icsk->icsk_delack_max = TCP_DELACK_MAX; tp->snd_ssthresh = TCP_INFINITE_SSTHRESH; - tp->snd_cwnd = TCP_INIT_CWND; + tcp_snd_cwnd_set(tp, TCP_INIT_CWND); tp->snd_cwnd_cnt = 0; + tp->is_cwnd_limited = 0; + tp->max_packets_out = 0; tp->window_clamp = 0; tp->delivered = 0; tp->delivered_ce = 0; + if (icsk->icsk_ca_ops->release) + icsk->icsk_ca_ops->release(sk); + memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv)); + icsk->icsk_ca_initialized = 0; tcp_set_ca_state(sk, TCP_CA_Open); tp->is_sack_reneg = 0; tcp_clear_retrans(tp); @@ -2635,8 +3157,7 @@ int tcp_disconnect(struct sock *sk, int flags) icsk->icsk_ack.rcv_mss = TCP_MIN_MSS; memset(&tp->rx_opt, 0, sizeof(tp->rx_opt)); __sk_dst_reset(sk); - dst_release(sk->sk_rx_dst); - sk->sk_rx_dst = NULL; + dst_release(xchg((__force struct dst_entry **)&sk->sk_rx_dst, NULL)); tcp_saved_syn_free(tp); tp->compressed_ack = 0; tp->segs_in = 0; @@ -2682,19 +3203,18 @@ int tcp_disconnect(struct sock *sk, int flags) sk->sk_frag.page = NULL; sk->sk_frag.offset = 0; } - - sk->sk_error_report(sk); + sk_error_report(sk); return 0; } EXPORT_SYMBOL(tcp_disconnect); static inline bool tcp_can_repair_sock(const struct sock *sk) { - return ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN) && + return sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN) && (sk->sk_state != TCP_LISTEN); } -static int tcp_repair_set_window(struct tcp_sock *tp, char __user *optbuf, int len) +static int tcp_repair_set_window(struct tcp_sock *tp, sockptr_t optbuf, int len) { struct tcp_repair_window opt; @@ -2704,7 +3224,7 @@ static int tcp_repair_set_window(struct tcp_sock *tp, char __user *optbuf, int l if (len != sizeof(opt)) return -EINVAL; - if (copy_from_user(&opt, optbuf, sizeof(opt))) + if (copy_from_sockptr(&opt, optbuf, sizeof(opt))) return -EFAULT; if (opt.max_window < opt.snd_wnd) @@ -2726,17 +3246,18 @@ static int tcp_repair_set_window(struct tcp_sock *tp, char __user *optbuf, int l return 0; } -static int tcp_repair_options_est(struct sock *sk, - struct tcp_repair_opt __user *optbuf, unsigned int len) +static int tcp_repair_options_est(struct sock *sk, sockptr_t optbuf, + unsigned int len) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_repair_opt opt; + size_t offset = 0; while (len >= sizeof(opt)) { - if (copy_from_user(&opt, optbuf, sizeof(opt))) + if (copy_from_sockptr_offset(&opt, optbuf, offset, sizeof(opt))) return -EFAULT; - optbuf++; + offset += sizeof(opt); len -= sizeof(opt); switch (opt.opt_code) { @@ -2790,11 +3311,184 @@ static void tcp_enable_tx_delay(void) } } +/* When set indicates to always queue non-full frames. Later the user clears + * this option and we transmit any pending partial frames in the queue. This is + * meant to be used alongside sendfile() to get properly filled frames when the + * user (for example) must write out headers with a write() call first and then + * use sendfile to send out the data parts. + * + * TCP_CORK can be set together with TCP_NODELAY and it is stronger than + * TCP_NODELAY. + */ +void __tcp_sock_set_cork(struct sock *sk, bool on) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (on) { + tp->nonagle |= TCP_NAGLE_CORK; + } else { + tp->nonagle &= ~TCP_NAGLE_CORK; + if (tp->nonagle & TCP_NAGLE_OFF) + tp->nonagle |= TCP_NAGLE_PUSH; + tcp_push_pending_frames(sk); + } +} + +void tcp_sock_set_cork(struct sock *sk, bool on) +{ + lock_sock(sk); + __tcp_sock_set_cork(sk, on); + release_sock(sk); +} +EXPORT_SYMBOL(tcp_sock_set_cork); + +/* TCP_NODELAY is weaker than TCP_CORK, so that this option on corked socket is + * remembered, but it is not activated until cork is cleared. + * + * However, when TCP_NODELAY is set we make an explicit push, which overrides + * even TCP_CORK for currently queued segments. + */ +void __tcp_sock_set_nodelay(struct sock *sk, bool on) +{ + if (on) { + tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF|TCP_NAGLE_PUSH; + tcp_push_pending_frames(sk); + } else { + tcp_sk(sk)->nonagle &= ~TCP_NAGLE_OFF; + } +} + +void tcp_sock_set_nodelay(struct sock *sk) +{ + lock_sock(sk); + __tcp_sock_set_nodelay(sk, true); + release_sock(sk); +} +EXPORT_SYMBOL(tcp_sock_set_nodelay); + +static void __tcp_sock_set_quickack(struct sock *sk, int val) +{ + if (!val) { + inet_csk_enter_pingpong_mode(sk); + return; + } + + inet_csk_exit_pingpong_mode(sk); + if ((1 << sk->sk_state) & (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) && + inet_csk_ack_scheduled(sk)) { + inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_PUSHED; + tcp_cleanup_rbuf(sk, 1); + if (!(val & 1)) + inet_csk_enter_pingpong_mode(sk); + } +} + +void tcp_sock_set_quickack(struct sock *sk, int val) +{ + lock_sock(sk); + __tcp_sock_set_quickack(sk, val); + release_sock(sk); +} +EXPORT_SYMBOL(tcp_sock_set_quickack); + +int tcp_sock_set_syncnt(struct sock *sk, int val) +{ + if (val < 1 || val > MAX_TCP_SYNCNT) + return -EINVAL; + + lock_sock(sk); + inet_csk(sk)->icsk_syn_retries = val; + release_sock(sk); + return 0; +} +EXPORT_SYMBOL(tcp_sock_set_syncnt); + +void tcp_sock_set_user_timeout(struct sock *sk, u32 val) +{ + lock_sock(sk); + inet_csk(sk)->icsk_user_timeout = val; + release_sock(sk); +} +EXPORT_SYMBOL(tcp_sock_set_user_timeout); + +int tcp_sock_set_keepidle_locked(struct sock *sk, int val) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (val < 1 || val > MAX_TCP_KEEPIDLE) + return -EINVAL; + + tp->keepalive_time = val * HZ; + if (sock_flag(sk, SOCK_KEEPOPEN) && + !((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) { + u32 elapsed = keepalive_time_elapsed(tp); + + if (tp->keepalive_time > elapsed) + elapsed = tp->keepalive_time - elapsed; + else + elapsed = 0; + inet_csk_reset_keepalive_timer(sk, elapsed); + } + + return 0; +} + +int tcp_sock_set_keepidle(struct sock *sk, int val) +{ + int err; + + lock_sock(sk); + err = tcp_sock_set_keepidle_locked(sk, val); + release_sock(sk); + return err; +} +EXPORT_SYMBOL(tcp_sock_set_keepidle); + +int tcp_sock_set_keepintvl(struct sock *sk, int val) +{ + if (val < 1 || val > MAX_TCP_KEEPINTVL) + return -EINVAL; + + lock_sock(sk); + tcp_sk(sk)->keepalive_intvl = val * HZ; + release_sock(sk); + return 0; +} +EXPORT_SYMBOL(tcp_sock_set_keepintvl); + +int tcp_sock_set_keepcnt(struct sock *sk, int val) +{ + if (val < 1 || val > MAX_TCP_KEEPCNT) + return -EINVAL; + + lock_sock(sk); + tcp_sk(sk)->keepalive_probes = val; + release_sock(sk); + return 0; +} +EXPORT_SYMBOL(tcp_sock_set_keepcnt); + +int tcp_set_window_clamp(struct sock *sk, int val) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (!val) { + if (sk->sk_state != TCP_CLOSE) + return -EINVAL; + tp->window_clamp = 0; + } else { + tp->window_clamp = val < SOCK_MIN_RCVBUF / 2 ? + SOCK_MIN_RCVBUF / 2 : val; + tp->rcv_ssthresh = min(tp->rcv_wnd, tp->window_clamp); + } + return 0; +} + /* * Socket option code for TCP. */ -static int do_tcp_setsockopt(struct sock *sk, int level, - int optname, char __user *optval, unsigned int optlen) +int do_tcp_setsockopt(struct sock *sk, int level, int optname, + sockptr_t optval, unsigned int optlen) { struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); @@ -2810,17 +3504,17 @@ static int do_tcp_setsockopt(struct sock *sk, int level, if (optlen < 1) return -EINVAL; - val = strncpy_from_user(name, optval, + val = strncpy_from_sockptr(name, optval, min_t(long, TCP_CA_NAME_MAX-1, optlen)); if (val < 0) return -EFAULT; name[val] = 0; - lock_sock(sk); - err = tcp_set_congestion_control(sk, name, true, true, - ns_capable(sock_net(sk)->user_ns, - CAP_NET_ADMIN)); - release_sock(sk); + sockopt_lock_sock(sk); + err = tcp_set_congestion_control(sk, name, !has_current_bpf_ctx(), + sockopt_ns_capable(sock_net(sk)->user_ns, + CAP_NET_ADMIN)); + sockopt_release_sock(sk); return err; } case TCP_ULP: { @@ -2829,16 +3523,16 @@ static int do_tcp_setsockopt(struct sock *sk, int level, if (optlen < 1) return -EINVAL; - val = strncpy_from_user(name, optval, + val = strncpy_from_sockptr(name, optval, min_t(long, TCP_ULP_NAME_MAX - 1, optlen)); if (val < 0) return -EFAULT; name[val] = 0; - lock_sock(sk); + sockopt_lock_sock(sk); err = tcp_set_ulp(sk, name); - release_sock(sk); + sockopt_release_sock(sk); return err; } case TCP_FASTOPEN_KEY: { @@ -2852,7 +3546,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, optlen != TCP_FASTOPEN_KEY_BUF_LENGTH) return -EINVAL; - if (copy_from_user(key, optval, optlen)) + if (copy_from_sockptr(key, optval, optlen)) return -EFAULT; if (optlen == TCP_FASTOPEN_KEY_BUF_LENGTH) @@ -2868,10 +3562,10 @@ static int do_tcp_setsockopt(struct sock *sk, int level, if (optlen < sizeof(int)) return -EINVAL; - if (get_user(val, (int __user *)optval)) + if (copy_from_sockptr(&val, optval, sizeof(val))) return -EFAULT; - lock_sock(sk); + sockopt_lock_sock(sk); switch (optname) { case TCP_MAXSEG: @@ -2887,20 +3581,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, break; case TCP_NODELAY: - if (val) { - /* TCP_NODELAY is weaker than TCP_CORK, so that - * this option on corked socket is remembered, but - * it is not activated until cork is cleared. - * - * However, when TCP_NODELAY is set we make - * an explicit push, which overrides even TCP_CORK - * for currently queued segments. - */ - tp->nonagle |= TCP_NAGLE_OFF|TCP_NAGLE_PUSH; - tcp_push_pending_frames(sk); - } else { - tp->nonagle &= ~TCP_NAGLE_OFF; - } + __tcp_sock_set_nodelay(sk, val); break; case TCP_THIN_LINEAR_TIMEOUTS: @@ -2944,65 +3625,40 @@ static int do_tcp_setsockopt(struct sock *sk, int level, break; case TCP_QUEUE_SEQ: - if (sk->sk_state != TCP_CLOSE) + if (sk->sk_state != TCP_CLOSE) { err = -EPERM; - else if (tp->repair_queue == TCP_SEND_QUEUE) - WRITE_ONCE(tp->write_seq, val); - else if (tp->repair_queue == TCP_RECV_QUEUE) - WRITE_ONCE(tp->rcv_nxt, val); - else + } else if (tp->repair_queue == TCP_SEND_QUEUE) { + if (!tcp_rtx_queue_empty(sk)) + err = -EPERM; + else + WRITE_ONCE(tp->write_seq, val); + } else if (tp->repair_queue == TCP_RECV_QUEUE) { + if (tp->rcv_nxt != tp->copied_seq) { + err = -EPERM; + } else { + WRITE_ONCE(tp->rcv_nxt, val); + WRITE_ONCE(tp->copied_seq, val); + } + } else { err = -EINVAL; + } break; case TCP_REPAIR_OPTIONS: if (!tp->repair) err = -EINVAL; - else if (sk->sk_state == TCP_ESTABLISHED) - err = tcp_repair_options_est(sk, - (struct tcp_repair_opt __user *)optval, - optlen); + else if (sk->sk_state == TCP_ESTABLISHED && !tp->bytes_sent) + err = tcp_repair_options_est(sk, optval, optlen); else err = -EPERM; break; case TCP_CORK: - /* When set indicates to always queue non-full frames. - * Later the user clears this option and we transmit - * any pending partial frames in the queue. This is - * meant to be used alongside sendfile() to get properly - * filled frames when the user (for example) must write - * out headers with a write() call first and then use - * sendfile to send out the data parts. - * - * TCP_CORK can be set together with TCP_NODELAY and it is - * stronger than TCP_NODELAY. - */ - if (val) { - tp->nonagle |= TCP_NAGLE_CORK; - } else { - tp->nonagle &= ~TCP_NAGLE_CORK; - if (tp->nonagle&TCP_NAGLE_OFF) - tp->nonagle |= TCP_NAGLE_PUSH; - tcp_push_pending_frames(sk); - } + __tcp_sock_set_cork(sk, val); break; case TCP_KEEPIDLE: - if (val < 1 || val > MAX_TCP_KEEPIDLE) - err = -EINVAL; - else { - tp->keepalive_time = val * HZ; - if (sock_flag(sk, SOCK_KEEPOPEN) && - !((1 << sk->sk_state) & - (TCPF_CLOSE | TCPF_LISTEN))) { - u32 elapsed = keepalive_time_elapsed(tp); - if (tp->keepalive_time > elapsed) - elapsed = tp->keepalive_time - elapsed; - else - elapsed = 0; - inet_csk_reset_keepalive_timer(sk, elapsed); - } - } + err = tcp_sock_set_keepidle_locked(sk, val); break; case TCP_KEEPINTVL: if (val < 1 || val > MAX_TCP_KEEPINTVL) @@ -3024,7 +3680,8 @@ static int do_tcp_setsockopt(struct sock *sk, int level, break; case TCP_SAVE_SYN: - if (val < 0 || val > 1) + /* 0: disable, 1: enable, 2: start from ether_header */ + if (val < 0 || val > 2) err = -EINVAL; else tp->save_syn = val; @@ -3033,8 +3690,8 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_LINGER2: if (val < 0) tp->linger2 = -1; - else if (val > net->ipv4.sysctl_tcp_fin_timeout / HZ) - tp->linger2 = 0; + else if (val > TCP_FIN_TIMEOUT_MAX / HZ) + tp->linger2 = TCP_FIN_TIMEOUT_MAX; else tp->linger2 = val * HZ; break; @@ -3047,40 +3704,17 @@ static int do_tcp_setsockopt(struct sock *sk, int level, break; case TCP_WINDOW_CLAMP: - if (!val) { - if (sk->sk_state != TCP_CLOSE) { - err = -EINVAL; - break; - } - tp->window_clamp = 0; - } else - tp->window_clamp = val < SOCK_MIN_RCVBUF / 2 ? - SOCK_MIN_RCVBUF / 2 : val; + err = tcp_set_window_clamp(sk, val); break; case TCP_QUICKACK: - if (!val) { - inet_csk_enter_pingpong_mode(sk); - } else { - inet_csk_exit_pingpong_mode(sk); - if ((1 << sk->sk_state) & - (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) && - inet_csk_ack_scheduled(sk)) { - icsk->icsk_ack.pending |= ICSK_ACK_PUSHED; - tcp_cleanup_rbuf(sk, 1); - if (!(val & 1)) - inet_csk_enter_pingpong_mode(sk); - } - } + __tcp_sock_set_quickack(sk, val); break; #ifdef CONFIG_TCP_MD5SIG case TCP_MD5SIG: case TCP_MD5SIG_EXT: - if ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) - err = tp->af_specific->md5_parse(sk, optname, optval, optlen); - else - err = -EINVAL; + err = tp->af_specific->md5_parse(sk, optname, optval, optlen); break; #endif case TCP_USER_TIMEOUT: @@ -3106,7 +3740,8 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_FASTOPEN_CONNECT: if (val > 1 || val < 0) { err = -EINVAL; - } else if (net->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) { + } else if (READ_ONCE(net->ipv4.sysctl_tcp_fastopen) & + TFO_CLIENT_ENABLE) { if (sk->sk_state == TCP_CLOSE) tp->fastopen_connect = val; else @@ -3152,34 +3787,23 @@ static int do_tcp_setsockopt(struct sock *sk, int level, break; } - release_sock(sk); + sockopt_release_sock(sk); return err; } -int tcp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, +int tcp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval, unsigned int optlen) { const struct inet_connection_sock *icsk = inet_csk(sk); if (level != SOL_TCP) - return icsk->icsk_af_ops->setsockopt(sk, level, optname, - optval, optlen); + /* Paired with WRITE_ONCE() in do_ipv6_setsockopt() and tcp_v6_connect() */ + return READ_ONCE(icsk->icsk_af_ops)->setsockopt(sk, level, optname, + optval, optlen); return do_tcp_setsockopt(sk, level, optname, optval, optlen); } EXPORT_SYMBOL(tcp_setsockopt); -#ifdef CONFIG_COMPAT -int compat_tcp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) -{ - if (level != SOL_TCP) - return inet_csk_compat_setsockopt(sk, level, optname, - optval, optlen); - return do_tcp_setsockopt(sk, level, optname, optval, optlen); -} -EXPORT_SYMBOL(compat_tcp_setsockopt); -#endif - static void tcp_get_info_chrono_stats(const struct tcp_sock *tp, struct tcp_info *info) { @@ -3225,7 +3849,7 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info) info->tcpi_max_pacing_rate = rate64; info->tcpi_reordering = tp->reordering; - info->tcpi_snd_cwnd = tp->snd_cwnd; + info->tcpi_snd_cwnd = tcp_snd_cwnd(tp); if (info->tcpi_state == TCP_LISTEN) { /* listeners aliased fields : @@ -3295,10 +3919,12 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info) tcp_get_info_chrono_stats(tp, info); info->tcpi_segs_out = tp->segs_out; - info->tcpi_segs_in = tp->segs_in; + + /* segs_in and data_segs_in can be updated from tcp_segs_in() from BH */ + info->tcpi_segs_in = READ_ONCE(tp->segs_in); + info->tcpi_data_segs_in = READ_ONCE(tp->data_segs_in); info->tcpi_min_rtt = tcp_min_rtt(tp); - info->tcpi_data_segs_in = tp->data_segs_in; info->tcpi_data_segs_out = tp->data_segs_out; info->tcpi_delivery_rate_app_limited = tp->rate_app_limited ? 1 : 0; @@ -3344,10 +3970,26 @@ static size_t tcp_opt_stats_get_size(void) nla_total_size(sizeof(u32)) + /* TCP_NLA_REORD_SEEN */ nla_total_size(sizeof(u32)) + /* TCP_NLA_SRTT */ nla_total_size(sizeof(u16)) + /* TCP_NLA_TIMEOUT_REHASH */ + nla_total_size(sizeof(u32)) + /* TCP_NLA_BYTES_NOTSENT */ + nla_total_size_64bit(sizeof(u64)) + /* TCP_NLA_EDT */ + nla_total_size(sizeof(u8)) + /* TCP_NLA_TTL */ 0; } -struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk) +/* Returns TTL or hop limit of an incoming packet from skb. */ +static u8 tcp_skb_ttl_or_hop_limit(const struct sk_buff *skb) +{ + if (skb->protocol == htons(ETH_P_IP)) + return ip_hdr(skb)->ttl; + else if (skb->protocol == htons(ETH_P_IPV6)) + return ipv6_hdr(skb)->hop_limit; + else + return 0; +} + +struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk, + const struct sk_buff *orig_skb, + const struct sk_buff *ack_skb) { const struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *stats; @@ -3378,7 +4020,7 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk) rate64 = tcp_compute_delivery_rate(tp); nla_put_u64_64bit(stats, TCP_NLA_DELIVERY_RATE, rate64, TCP_NLA_PAD); - nla_put_u32(stats, TCP_NLA_SND_CWND, tp->snd_cwnd); + nla_put_u32(stats, TCP_NLA_SND_CWND, tcp_snd_cwnd(tp)); nla_put_u32(stats, TCP_NLA_REORDERING, tp->reordering); nla_put_u32(stats, TCP_NLA_MIN_RTT, tcp_min_rtt(tp)); @@ -3399,19 +4041,26 @@ struct sk_buff *tcp_get_timestamping_opt_stats(const struct sock *sk) nla_put_u32(stats, TCP_NLA_REORD_SEEN, tp->reord_seen); nla_put_u32(stats, TCP_NLA_SRTT, tp->srtt_us >> 3); nla_put_u16(stats, TCP_NLA_TIMEOUT_REHASH, tp->timeout_rehash); + nla_put_u32(stats, TCP_NLA_BYTES_NOTSENT, + max_t(int, 0, tp->write_seq - tp->snd_nxt)); + nla_put_u64_64bit(stats, TCP_NLA_EDT, orig_skb->skb_mstamp_ns, + TCP_NLA_PAD); + if (ack_skb) + nla_put_u8(stats, TCP_NLA_TTL, + tcp_skb_ttl_or_hop_limit(ack_skb)); return stats; } -static int do_tcp_getsockopt(struct sock *sk, int level, - int optname, char __user *optval, int __user *optlen) +int do_tcp_getsockopt(struct sock *sk, int level, + int optname, sockptr_t optval, sockptr_t optlen) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); int val, len; - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; len = min_t(unsigned int, len, sizeof(int)); @@ -3443,12 +4092,13 @@ static int do_tcp_getsockopt(struct sock *sk, int level, val = keepalive_probes(tp); break; case TCP_SYNCNT: - val = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries; + val = icsk->icsk_syn_retries ? : + READ_ONCE(net->ipv4.sysctl_tcp_syn_retries); break; case TCP_LINGER2: val = tp->linger2; if (val >= 0) - val = (val ? : net->ipv4.sysctl_tcp_fin_timeout) / HZ; + val = (val ? : READ_ONCE(net->ipv4.sysctl_tcp_fin_timeout)) / HZ; break; case TCP_DEFER_ACCEPT: val = retrans_to_secs(icsk->icsk_accept_queue.rskq_defer_accept, @@ -3460,15 +4110,15 @@ static int do_tcp_getsockopt(struct sock *sk, int level, case TCP_INFO: { struct tcp_info info; - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; tcp_get_info(sk, &info); len = min_t(unsigned int, len, sizeof(info)); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &info, len)) + if (copy_to_sockptr(optval, &info, len)) return -EFAULT; return 0; } @@ -3478,7 +4128,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level, size_t sz = 0; int attr; - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; ca_ops = icsk->icsk_ca_ops; @@ -3486,9 +4136,9 @@ static int do_tcp_getsockopt(struct sock *sk, int level, sz = ca_ops->get_info(sk, ~0U, &attr, &info); len = min_t(unsigned int, len, sz); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &info, len)) + if (copy_to_sockptr(optval, &info, len)) return -EFAULT; return 0; } @@ -3497,51 +4147,44 @@ static int do_tcp_getsockopt(struct sock *sk, int level, break; case TCP_CONGESTION: - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; len = min_t(unsigned int, len, TCP_CA_NAME_MAX); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, icsk->icsk_ca_ops->name, len)) + if (copy_to_sockptr(optval, icsk->icsk_ca_ops->name, len)) return -EFAULT; return 0; case TCP_ULP: - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; len = min_t(unsigned int, len, TCP_ULP_NAME_MAX); if (!icsk->icsk_ulp_ops) { - if (put_user(0, optlen)) + len = 0; + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; return 0; } - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, icsk->icsk_ulp_ops->name, len)) + if (copy_to_sockptr(optval, icsk->icsk_ulp_ops->name, len)) return -EFAULT; return 0; case TCP_FASTOPEN_KEY: { - __u8 key[TCP_FASTOPEN_KEY_BUF_LENGTH]; - struct tcp_fastopen_context *ctx; - unsigned int key_len = 0; + u64 key[TCP_FASTOPEN_KEY_BUF_LENGTH / sizeof(u64)]; + unsigned int key_len; - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; - rcu_read_lock(); - ctx = rcu_dereference(icsk->icsk_accept_queue.fastopenq.ctx); - if (ctx) { - key_len = tcp_fastopen_context_len(ctx) * - TCP_FASTOPEN_KEY_LENGTH; - memcpy(&key[0], &ctx->key[0], key_len); - } - rcu_read_unlock(); - + key_len = tcp_fastopen_get_cipher(net, icsk, key) * + TCP_FASTOPEN_KEY_LENGTH; len = min_t(unsigned int, len, key_len); - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, key, len)) + if (copy_to_sockptr(optval, key, len)) return -EFAULT; return 0; } @@ -3567,7 +4210,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level, case TCP_REPAIR_WINDOW: { struct tcp_repair_window opt; - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; if (len != sizeof(opt)) @@ -3582,7 +4225,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level, opt.rcv_wnd = tp->rcv_wnd; opt.rcv_wup = tp->rcv_wup; - if (copy_to_user(optval, &opt, len)) + if (copy_to_sockptr(optval, &opt, len)) return -EFAULT; return 0; } @@ -3628,53 +4271,100 @@ static int do_tcp_getsockopt(struct sock *sk, int level, val = tp->save_syn; break; case TCP_SAVED_SYN: { - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; - lock_sock(sk); + sockopt_lock_sock(sk); if (tp->saved_syn) { - if (len < tp->saved_syn[0]) { - if (put_user(tp->saved_syn[0], optlen)) { - release_sock(sk); + if (len < tcp_saved_syn_len(tp->saved_syn)) { + len = tcp_saved_syn_len(tp->saved_syn); + if (copy_to_sockptr(optlen, &len, sizeof(int))) { + sockopt_release_sock(sk); return -EFAULT; } - release_sock(sk); + sockopt_release_sock(sk); return -EINVAL; } - len = tp->saved_syn[0]; - if (put_user(len, optlen)) { - release_sock(sk); + len = tcp_saved_syn_len(tp->saved_syn); + if (copy_to_sockptr(optlen, &len, sizeof(int))) { + sockopt_release_sock(sk); return -EFAULT; } - if (copy_to_user(optval, tp->saved_syn + 1, len)) { - release_sock(sk); + if (copy_to_sockptr(optval, tp->saved_syn->data, len)) { + sockopt_release_sock(sk); return -EFAULT; } tcp_saved_syn_free(tp); - release_sock(sk); + sockopt_release_sock(sk); } else { - release_sock(sk); + sockopt_release_sock(sk); len = 0; - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; } return 0; } #ifdef CONFIG_MMU case TCP_ZEROCOPY_RECEIVE: { - struct tcp_zerocopy_receive zc; + struct scm_timestamping_internal tss; + struct tcp_zerocopy_receive zc = {}; int err; - if (get_user(len, optlen)) + if (copy_from_sockptr(&len, optlen, sizeof(int))) return -EFAULT; - if (len != sizeof(zc)) + if (len < 0 || + len < offsetofend(struct tcp_zerocopy_receive, length)) return -EINVAL; - if (copy_from_user(&zc, optval, len)) + if (unlikely(len > sizeof(zc))) { + err = check_zeroed_sockptr(optval, sizeof(zc), + len - sizeof(zc)); + if (err < 1) + return err == 0 ? -EINVAL : err; + len = sizeof(zc); + if (copy_to_sockptr(optlen, &len, sizeof(int))) + return -EFAULT; + } + if (copy_from_sockptr(&zc, optval, len)) return -EFAULT; - lock_sock(sk); - err = tcp_zerocopy_receive(sk, &zc); - release_sock(sk); - if (!err && copy_to_user(optval, &zc, len)) + if (zc.reserved) + return -EINVAL; + if (zc.msg_flags & ~(TCP_VALID_ZC_MSG_FLAGS)) + return -EINVAL; + sockopt_lock_sock(sk); + err = tcp_zerocopy_receive(sk, &zc, &tss); + err = BPF_CGROUP_RUN_PROG_GETSOCKOPT_KERN(sk, level, optname, + &zc, &len, err); + sockopt_release_sock(sk); + if (len >= offsetofend(struct tcp_zerocopy_receive, msg_flags)) + goto zerocopy_rcv_cmsg; + switch (len) { + case offsetofend(struct tcp_zerocopy_receive, msg_flags): + goto zerocopy_rcv_cmsg; + case offsetofend(struct tcp_zerocopy_receive, msg_controllen): + case offsetofend(struct tcp_zerocopy_receive, msg_control): + case offsetofend(struct tcp_zerocopy_receive, flags): + case offsetofend(struct tcp_zerocopy_receive, copybuf_len): + case offsetofend(struct tcp_zerocopy_receive, copybuf_address): + case offsetofend(struct tcp_zerocopy_receive, err): + goto zerocopy_rcv_sk_err; + case offsetofend(struct tcp_zerocopy_receive, inq): + goto zerocopy_rcv_inq; + case offsetofend(struct tcp_zerocopy_receive, length): + default: + goto zerocopy_rcv_out; + } +zerocopy_rcv_cmsg: + if (zc.msg_flags & TCP_CMSG_TS) + tcp_zc_finalize_rx_tstamp(sk, &zc, &tss); + else + zc.msg_flags = 0; +zerocopy_rcv_sk_err: + if (!err) + zc.err = sock_error(sk); +zerocopy_rcv_inq: + zc.inq = tcp_inq_hint(sk); +zerocopy_rcv_out: + if (!err && copy_to_sockptr(optval, &zc, len)) err = -EFAULT; return err; } @@ -3683,37 +4373,39 @@ static int do_tcp_getsockopt(struct sock *sk, int level, return -ENOPROTOOPT; } - if (put_user(len, optlen)) + if (copy_to_sockptr(optlen, &len, sizeof(int))) return -EFAULT; - if (copy_to_user(optval, &val, len)) + if (copy_to_sockptr(optval, &val, len)) return -EFAULT; return 0; } +bool tcp_bpf_bypass_getsockopt(int level, int optname) +{ + /* TCP do_tcp_getsockopt has optimized getsockopt implementation + * to avoid extra socket lock for TCP_ZEROCOPY_RECEIVE. + */ + if (level == SOL_TCP && optname == TCP_ZEROCOPY_RECEIVE) + return true; + + return false; +} +EXPORT_SYMBOL(tcp_bpf_bypass_getsockopt); + int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { struct inet_connection_sock *icsk = inet_csk(sk); if (level != SOL_TCP) - return icsk->icsk_af_ops->getsockopt(sk, level, optname, - optval, optlen); - return do_tcp_getsockopt(sk, level, optname, optval, optlen); + /* Paired with WRITE_ONCE() in do_ipv6_setsockopt() and tcp_v6_connect() */ + return READ_ONCE(icsk->icsk_af_ops)->getsockopt(sk, level, optname, + optval, optlen); + return do_tcp_getsockopt(sk, level, optname, USER_SOCKPTR(optval), + USER_SOCKPTR(optlen)); } EXPORT_SYMBOL(tcp_getsockopt); -#ifdef CONFIG_COMPAT -int compat_tcp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (level != SOL_TCP) - return inet_csk_compat_getsockopt(sk, level, optname, - optval, optlen); - return do_tcp_getsockopt(sk, level, optname, optval, optlen); -} -EXPORT_SYMBOL(compat_tcp_getsockopt); -#endif - #ifdef CONFIG_TCP_MD5SIG static DEFINE_PER_CPU(struct tcp_md5sig_pool, tcp_md5sig_pool); static DEFINE_MUTEX(tcp_md5sig_mutex); @@ -3756,12 +4448,16 @@ static void __tcp_alloc_md5sig_pool(void) * to memory. See smp_rmb() in tcp_get_md5sig_pool() */ smp_wmb(); - tcp_md5sig_pool_populated = true; + /* Paired with READ_ONCE() from tcp_alloc_md5sig_pool() + * and tcp_get_md5sig_pool(). + */ + WRITE_ONCE(tcp_md5sig_pool_populated, true); } bool tcp_alloc_md5sig_pool(void) { - if (unlikely(!tcp_md5sig_pool_populated)) { + /* Paired with WRITE_ONCE() from __tcp_alloc_md5sig_pool() */ + if (unlikely(!READ_ONCE(tcp_md5sig_pool_populated))) { mutex_lock(&tcp_md5sig_mutex); if (!tcp_md5sig_pool_populated) { @@ -3772,7 +4468,8 @@ bool tcp_alloc_md5sig_pool(void) mutex_unlock(&tcp_md5sig_mutex); } - return tcp_md5sig_pool_populated; + /* Paired with WRITE_ONCE() from __tcp_alloc_md5sig_pool() */ + return READ_ONCE(tcp_md5sig_pool_populated); } EXPORT_SYMBOL(tcp_alloc_md5sig_pool); @@ -3788,7 +4485,8 @@ struct tcp_md5sig_pool *tcp_get_md5sig_pool(void) { local_bh_disable(); - if (tcp_md5sig_pool_populated) { + /* Paired with WRITE_ONCE() from __tcp_alloc_md5sig_pool() */ + if (READ_ONCE(tcp_md5sig_pool_populated)) { /* coupled with smp_wmb() in __tcp_alloc_md5sig_pool() */ smp_rmb(); return this_cpu_ptr(&tcp_md5sig_pool); @@ -3839,14 +4537,93 @@ EXPORT_SYMBOL(tcp_md5_hash_skb_data); int tcp_md5_hash_key(struct tcp_md5sig_pool *hp, const struct tcp_md5sig_key *key) { + u8 keylen = READ_ONCE(key->keylen); /* paired with WRITE_ONCE() in tcp_md5_do_add */ struct scatterlist sg; - sg_init_one(&sg, key->key, key->keylen); - ahash_request_set_crypt(hp->md5_req, &sg, NULL, key->keylen); - return crypto_ahash_update(hp->md5_req); + sg_init_one(&sg, key->key, keylen); + ahash_request_set_crypt(hp->md5_req, &sg, NULL, keylen); + + /* We use data_race() because tcp_md5_do_add() might change key->key under us */ + return data_race(crypto_ahash_update(hp->md5_req)); } EXPORT_SYMBOL(tcp_md5_hash_key); +/* Called with rcu_read_lock() */ +enum skb_drop_reason +tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb, + const void *saddr, const void *daddr, + int family, int dif, int sdif) +{ + /* + * This gets called for each TCP segment that arrives + * so we want to be efficient. + * We have 3 drop cases: + * o No MD5 hash and one expected. + * o MD5 hash and we're not expecting one. + * o MD5 hash and its wrong. + */ + const __u8 *hash_location = NULL; + struct tcp_md5sig_key *hash_expected; + const struct tcphdr *th = tcp_hdr(skb); + struct tcp_sock *tp = tcp_sk(sk); + int genhash, l3index; + u8 newhash[16]; + + /* sdif set, means packet ingressed via a device + * in an L3 domain and dif is set to the l3mdev + */ + l3index = sdif ? dif : 0; + + hash_expected = tcp_md5_do_lookup(sk, l3index, saddr, family); + hash_location = tcp_parse_md5sig_option(th); + + /* We've parsed the options - do we have a hash? */ + if (!hash_expected && !hash_location) + return SKB_NOT_DROPPED_YET; + + if (hash_expected && !hash_location) { + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND); + return SKB_DROP_REASON_TCP_MD5NOTFOUND; + } + + if (!hash_expected && hash_location) { + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED); + return SKB_DROP_REASON_TCP_MD5UNEXPECTED; + } + + /* Check the signature. + * To support dual stack listeners, we need to handle + * IPv4-mapped case. + */ + if (family == AF_INET) + genhash = tcp_v4_md5_hash_skb(newhash, + hash_expected, + NULL, skb); + else + genhash = tp->af_specific->calc_md5_hash(newhash, + hash_expected, + NULL, skb); + + if (genhash || memcmp(hash_location, newhash, 16) != 0) { + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE); + if (family == AF_INET) { + net_info_ratelimited("MD5 Hash failed for (%pI4, %d)->(%pI4, %d)%s L3 index %d\n", + saddr, ntohs(th->source), + daddr, ntohs(th->dest), + genhash ? " tcp_v4_calc_md5_hash failed" + : "", l3index); + } else { + net_info_ratelimited("MD5 Hash %s for [%pI6c]:%u->[%pI6c]:%u L3 index %d\n", + genhash ? "failed" : "mismatch", + saddr, ntohs(th->source), + daddr, ntohs(th->dest), l3index); + } + return SKB_DROP_REASON_TCP_MD5FAILURE; + } + return SKB_NOT_DROPPED_YET; +} +EXPORT_SYMBOL(tcp_inbound_md5_hash); + #endif void tcp_done(struct sock *sk) @@ -3878,16 +4655,24 @@ EXPORT_SYMBOL_GPL(tcp_done); int tcp_abort(struct sock *sk, int err) { - if (!sk_fullsock(sk)) { - if (sk->sk_state == TCP_NEW_SYN_RECV) { - struct request_sock *req = inet_reqsk(sk); + int state = inet_sk_state_load(sk); - local_bh_disable(); - inet_csk_reqsk_queue_drop(req->rsk_listener, req); - local_bh_enable(); - return 0; - } - return -EOPNOTSUPP; + if (state == TCP_NEW_SYN_RECV) { + struct request_sock *req = inet_reqsk(sk); + + local_bh_disable(); + inet_csk_reqsk_queue_drop(req->rsk_listener, req); + local_bh_enable(); + return 0; + } + if (state == TCP_TIME_WAIT) { + struct inet_timewait_sock *tw = inet_twsk(sk); + + refcount_inc(&tw->tw_refcnt); + local_bh_disable(); + inet_twsk_deschedule_put(tw); + local_bh_enable(); + return 0; } /* Don't race with userspace socket closes such as tcp_close. */ @@ -3906,7 +4691,7 @@ int tcp_abort(struct sock *sk, int err) sk->sk_err = err; /* This barrier is coupled with smp_rmb() in tcp_poll() */ smp_wmb(); - sk->sk_error_report(sk); + sk_error_report(sk); if (tcp_need_reset(sk->sk_state)) tcp_send_active_reset(sk, GFP_ATOMIC); tcp_done(sk); @@ -3959,15 +4744,25 @@ void __init tcp_init(void) sizeof_field(struct sk_buff, cb)); percpu_counter_init(&tcp_sockets_allocated, 0, GFP_KERNEL); - percpu_counter_init(&tcp_orphan_count, 0, GFP_KERNEL); - inet_hashinfo_init(&tcp_hashinfo); + + timer_setup(&tcp_orphan_timer, tcp_orphan_update, TIMER_DEFERRABLE); + mod_timer(&tcp_orphan_timer, jiffies + TCP_ORPHAN_TIMER_PERIOD); + inet_hashinfo2_init(&tcp_hashinfo, "tcp_listen_portaddr_hash", thash_entries, 21, /* one slot per 2 MB*/ 0, 64 * 1024); tcp_hashinfo.bind_bucket_cachep = kmem_cache_create("tcp_bind_bucket", sizeof(struct inet_bind_bucket), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + SLAB_HWCACHE_ALIGN | SLAB_PANIC | + SLAB_ACCOUNT, + NULL); + tcp_hashinfo.bind2_bucket_cachep = + kmem_cache_create("tcp_bind2_bucket", + sizeof(struct inet_bind2_bucket), 0, + SLAB_HWCACHE_ALIGN | SLAB_PANIC | + SLAB_ACCOUNT, + NULL); /* Size and allocate the main established and bind bucket * hash tables. @@ -3991,7 +4786,7 @@ void __init tcp_init(void) panic("TCP: failed to alloc ehash_locks"); tcp_hashinfo.bhash = alloc_large_system_hash("TCP bind", - sizeof(struct inet_bind_hashbucket), + 2 * sizeof(struct inet_bind_hashbucket), tcp_hashinfo.ehash_mask + 1, 17, /* one slot per 128 KB of memory */ 0, @@ -4000,11 +4795,15 @@ void __init tcp_init(void) 0, 64 * 1024); tcp_hashinfo.bhash_size = 1U << tcp_hashinfo.bhash_size; + tcp_hashinfo.bhash2 = tcp_hashinfo.bhash + tcp_hashinfo.bhash_size; for (i = 0; i < tcp_hashinfo.bhash_size; i++) { spin_lock_init(&tcp_hashinfo.bhash[i].lock); INIT_HLIST_HEAD(&tcp_hashinfo.bhash[i].chain); + spin_lock_init(&tcp_hashinfo.bhash2[i].lock); + INIT_HLIST_HEAD(&tcp_hashinfo.bhash2[i].chain); } + tcp_hashinfo.pernet = false; cnt = tcp_hashinfo.ehash_mask + 1; sysctl_tcp_max_orphans = cnt / 2; @@ -4015,11 +4814,11 @@ void __init tcp_init(void) max_wshare = min(4UL*1024*1024, limit); max_rshare = min(6UL*1024*1024, limit); - init_net.ipv4.sysctl_tcp_wmem[0] = SK_MEM_QUANTUM; + init_net.ipv4.sysctl_tcp_wmem[0] = PAGE_SIZE; init_net.ipv4.sysctl_tcp_wmem[1] = 16*1024; init_net.ipv4.sysctl_tcp_wmem[2] = max(64*1024, max_wshare); - init_net.ipv4.sysctl_tcp_rmem[0] = SK_MEM_QUANTUM; + init_net.ipv4.sysctl_tcp_rmem[0] = PAGE_SIZE; init_net.ipv4.sysctl_tcp_rmem[1] = 131072; init_net.ipv4.sysctl_tcp_rmem[2] = max(131072, max_rshare); diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index 6c4d79baff26..54eec33c6e1c 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -56,6 +56,8 @@ * otherwise TCP stack falls back to an internal pacing using one high * resolution timer per TCP socket and may use more resources. */ +#include <linux/btf.h> +#include <linux/btf_ids.h> #include <linux/module.h> #include <net/tcp.h> #include <linux/inet_diag.h> @@ -274,7 +276,7 @@ static void bbr_init_pacing_rate_from_rtt(struct sock *sk) } else { /* no RTT sample yet */ rtt_us = USEC_PER_MSEC; /* use nominal default RTT */ } - bw = (u64)tp->snd_cwnd * BW_UNIT; + bw = (u64)tcp_snd_cwnd(tp) * BW_UNIT; do_div(bw, rtt_us); sk->sk_pacing_rate = bbr_bw_to_pacing_rate(sk, bw, bbr_high_gain); } @@ -308,7 +310,7 @@ static u32 bbr_tso_segs_goal(struct sock *sk) */ bytes = min_t(unsigned long, sk->sk_pacing_rate >> READ_ONCE(sk->sk_pacing_shift), - GSO_MAX_SIZE - 1 - MAX_TCP_HEADER); + GSO_LEGACY_MAX_SIZE - 1 - MAX_TCP_HEADER); segs = max_t(u32, bytes / tp->mss_cache, bbr_min_tso_segs(sk)); return min(segs, 0x7FU); @@ -321,9 +323,9 @@ static void bbr_save_cwnd(struct sock *sk) struct bbr *bbr = inet_csk_ca(sk); if (bbr->prev_ca_state < TCP_CA_Recovery && bbr->mode != BBR_PROBE_RTT) - bbr->prior_cwnd = tp->snd_cwnd; /* this cwnd is good enough */ + bbr->prior_cwnd = tcp_snd_cwnd(tp); /* this cwnd is good enough */ else /* loss recovery or BBR_PROBE_RTT have temporarily cut cwnd */ - bbr->prior_cwnd = max(bbr->prior_cwnd, tp->snd_cwnd); + bbr->prior_cwnd = max(bbr->prior_cwnd, tcp_snd_cwnd(tp)); } static void bbr_cwnd_event(struct sock *sk, enum tcp_ca_event event) @@ -480,7 +482,7 @@ static bool bbr_set_cwnd_to_recover_or_restore( struct tcp_sock *tp = tcp_sk(sk); struct bbr *bbr = inet_csk_ca(sk); u8 prev_state = bbr->prev_ca_state, state = inet_csk(sk)->icsk_ca_state; - u32 cwnd = tp->snd_cwnd; + u32 cwnd = tcp_snd_cwnd(tp); /* An ACK for P pkts should release at most 2*P packets. We do this * in two steps. First, here we deduct the number of lost packets. @@ -518,7 +520,7 @@ static void bbr_set_cwnd(struct sock *sk, const struct rate_sample *rs, { struct tcp_sock *tp = tcp_sk(sk); struct bbr *bbr = inet_csk_ca(sk); - u32 cwnd = tp->snd_cwnd, target_cwnd = 0; + u32 cwnd = tcp_snd_cwnd(tp), target_cwnd = 0; if (!acked) goto done; /* no packet fully ACKed; just apply caps */ @@ -542,9 +544,9 @@ static void bbr_set_cwnd(struct sock *sk, const struct rate_sample *rs, cwnd = max(cwnd, bbr_cwnd_min_target); done: - tp->snd_cwnd = min(cwnd, tp->snd_cwnd_clamp); /* apply global cap */ + tcp_snd_cwnd_set(tp, min(cwnd, tp->snd_cwnd_clamp)); /* apply global cap */ if (bbr->mode == BBR_PROBE_RTT) /* drain queue, refresh min_rtt */ - tp->snd_cwnd = min(tp->snd_cwnd, bbr_cwnd_min_target); + tcp_snd_cwnd_set(tp, min(tcp_snd_cwnd(tp), bbr_cwnd_min_target)); } /* End cycle phase if it's time and/or we hit the phase's in-flight target. */ @@ -854,7 +856,7 @@ static void bbr_update_ack_aggregation(struct sock *sk, bbr->ack_epoch_acked = min_t(u32, 0xFFFFF, bbr->ack_epoch_acked + rs->acked_sacked); extra_acked = bbr->ack_epoch_acked - expected_acked; - extra_acked = min(extra_acked, tp->snd_cwnd); + extra_acked = min(extra_acked, tcp_snd_cwnd(tp)); if (extra_acked > bbr->extra_acked[bbr->extra_acked_win_idx]) bbr->extra_acked[bbr->extra_acked_win_idx] = extra_acked; } @@ -912,7 +914,7 @@ static void bbr_check_probe_rtt_done(struct sock *sk) return; bbr->min_rtt_stamp = tcp_jiffies32; /* wait a while until PROBE_RTT */ - tp->snd_cwnd = max(tp->snd_cwnd, bbr->prior_cwnd); + tcp_snd_cwnd_set(tp, max(tcp_snd_cwnd(tp), bbr->prior_cwnd)); bbr_reset_mode(sk); } @@ -945,7 +947,7 @@ static void bbr_update_min_rtt(struct sock *sk, const struct rate_sample *rs) filter_expired = after(tcp_jiffies32, bbr->min_rtt_stamp + bbr_min_rtt_win_sec * HZ); if (rs->rtt_us >= 0 && - (rs->rtt_us <= bbr->min_rtt_us || + (rs->rtt_us < bbr->min_rtt_us || (filter_expired && !rs->is_ack_delayed))) { bbr->min_rtt_us = rs->rtt_us; bbr->min_rtt_stamp = tcp_jiffies32; @@ -1041,7 +1043,7 @@ static void bbr_init(struct sock *sk) bbr->prior_cwnd = 0; tp->snd_ssthresh = TCP_INFINITE_SSTHRESH; bbr->rtt_cnt = 0; - bbr->next_rtt_delivered = 0; + bbr->next_rtt_delivered = tp->delivered; bbr->prev_ca_state = TCP_CA_Open; bbr->packet_conservation = 0; @@ -1091,7 +1093,7 @@ static u32 bbr_undo_cwnd(struct sock *sk) bbr->full_bw = 0; /* spurious slow-down; reset full pipe detection */ bbr->full_bw_cnt = 0; bbr_reset_lt_bw_sampling(sk); - return tcp_sk(sk)->snd_cwnd; + return tcp_snd_cwnd(tcp_sk(sk)); } /* Entering loss recovery, so save cwnd for when we exit or undo recovery. */ @@ -1152,9 +1154,35 @@ static struct tcp_congestion_ops tcp_bbr_cong_ops __read_mostly = { .set_state = bbr_set_state, }; +BTF_SET8_START(tcp_bbr_check_kfunc_ids) +#ifdef CONFIG_X86 +#ifdef CONFIG_DYNAMIC_FTRACE +BTF_ID_FLAGS(func, bbr_init) +BTF_ID_FLAGS(func, bbr_main) +BTF_ID_FLAGS(func, bbr_sndbuf_expand) +BTF_ID_FLAGS(func, bbr_undo_cwnd) +BTF_ID_FLAGS(func, bbr_cwnd_event) +BTF_ID_FLAGS(func, bbr_ssthresh) +BTF_ID_FLAGS(func, bbr_min_tso_segs) +BTF_ID_FLAGS(func, bbr_set_state) +#endif +#endif +BTF_SET8_END(tcp_bbr_check_kfunc_ids) + +static const struct btf_kfunc_id_set tcp_bbr_kfunc_set = { + .owner = THIS_MODULE, + .set = &tcp_bbr_check_kfunc_ids, +}; + static int __init bbr_register(void) { + int ret; + BUILD_BUG_ON(sizeof(struct bbr) > ICSK_CA_PRIV_SIZE); + + ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_bbr_kfunc_set); + if (ret < 0) + return ret; return tcp_register_congestion_control(&tcp_bbr_cong_ops); } diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c index 645cc3009e64..58358bf92e1b 100644 --- a/net/ipv4/tcp_bic.c +++ b/net/ipv4/tcp_bic.c @@ -145,12 +145,13 @@ static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) if (!tcp_is_cwnd_limited(sk)) return; - if (tcp_in_slow_start(tp)) - tcp_slow_start(tp, acked); - else { - bictcp_update(ca, tp->snd_cwnd); - tcp_cong_avoid_ai(tp, ca->cnt, 1); + if (tcp_in_slow_start(tp)) { + acked = tcp_slow_start(tp, acked); + if (!acked) + return; } + bictcp_update(ca, tcp_snd_cwnd(tp)); + tcp_cong_avoid_ai(tp, ca->cnt, acked); } /* @@ -165,16 +166,16 @@ static u32 bictcp_recalc_ssthresh(struct sock *sk) ca->epoch_start = 0; /* end of epoch */ /* Wmax and fast convergence */ - if (tp->snd_cwnd < ca->last_max_cwnd && fast_convergence) - ca->last_max_cwnd = (tp->snd_cwnd * (BICTCP_BETA_SCALE + beta)) + if (tcp_snd_cwnd(tp) < ca->last_max_cwnd && fast_convergence) + ca->last_max_cwnd = (tcp_snd_cwnd(tp) * (BICTCP_BETA_SCALE + beta)) / (2 * BICTCP_BETA_SCALE); else - ca->last_max_cwnd = tp->snd_cwnd; + ca->last_max_cwnd = tcp_snd_cwnd(tp); - if (tp->snd_cwnd <= low_window) - return max(tp->snd_cwnd >> 1U, 2U); + if (tcp_snd_cwnd(tp) <= low_window) + return max(tcp_snd_cwnd(tp) >> 1U, 2U); else - return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U); + return max((tcp_snd_cwnd(tp) * beta) / BICTCP_BETA_SCALE, 2U); } static void bictcp_state(struct sock *sk, u8 new_state) diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 8a01428f80c1..cf9c3e8f7ccb 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -10,154 +10,6 @@ #include <net/inet_common.h> #include <net/tls.h> -static bool tcp_bpf_stream_read(const struct sock *sk) -{ - struct sk_psock *psock; - bool empty = true; - - rcu_read_lock(); - psock = sk_psock(sk); - if (likely(psock)) - empty = list_empty(&psock->ingress_msg); - rcu_read_unlock(); - return !empty; -} - -static int tcp_bpf_wait_data(struct sock *sk, struct sk_psock *psock, - int flags, long timeo, int *err) -{ - DEFINE_WAIT_FUNC(wait, woken_wake_function); - int ret = 0; - - if (!timeo) - return ret; - - add_wait_queue(sk_sleep(sk), &wait); - sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); - ret = sk_wait_event(sk, &timeo, - !list_empty(&psock->ingress_msg) || - !skb_queue_empty(&sk->sk_receive_queue), &wait); - sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); - remove_wait_queue(sk_sleep(sk), &wait); - return ret; -} - -int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock, - struct msghdr *msg, int len, int flags) -{ - struct iov_iter *iter = &msg->msg_iter; - int peek = flags & MSG_PEEK; - int i, ret, copied = 0; - struct sk_msg *msg_rx; - - msg_rx = list_first_entry_or_null(&psock->ingress_msg, - struct sk_msg, list); - - while (copied != len) { - struct scatterlist *sge; - - if (unlikely(!msg_rx)) - break; - - i = msg_rx->sg.start; - do { - struct page *page; - int copy; - - sge = sk_msg_elem(msg_rx, i); - copy = sge->length; - page = sg_page(sge); - if (copied + copy > len) - copy = len - copied; - ret = copy_page_to_iter(page, sge->offset, copy, iter); - if (ret != copy) { - msg_rx->sg.start = i; - return -EFAULT; - } - - copied += copy; - if (likely(!peek)) { - sge->offset += copy; - sge->length -= copy; - sk_mem_uncharge(sk, copy); - msg_rx->sg.size -= copy; - - if (!sge->length) { - sk_msg_iter_var_next(i); - if (!msg_rx->skb) - put_page(page); - } - } else { - sk_msg_iter_var_next(i); - } - - if (copied == len) - break; - } while (i != msg_rx->sg.end); - - if (unlikely(peek)) { - msg_rx = list_next_entry(msg_rx, list); - continue; - } - - msg_rx->sg.start = i; - if (!sge->length && msg_rx->sg.start == msg_rx->sg.end) { - list_del(&msg_rx->list); - if (msg_rx->skb) - consume_skb(msg_rx->skb); - kfree(msg_rx); - } - msg_rx = list_first_entry_or_null(&psock->ingress_msg, - struct sk_msg, list); - } - - return copied; -} -EXPORT_SYMBOL_GPL(__tcp_bpf_recvmsg); - -int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, - int nonblock, int flags, int *addr_len) -{ - struct sk_psock *psock; - int copied, ret; - - psock = sk_psock_get(sk); - if (unlikely(!psock)) - return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len); - if (unlikely(flags & MSG_ERRQUEUE)) - return inet_recv_error(sk, msg, len, addr_len); - if (!skb_queue_empty(&sk->sk_receive_queue) && - sk_psock_queue_empty(psock)) - return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len); - lock_sock(sk); -msg_bytes_ready: - copied = __tcp_bpf_recvmsg(sk, psock, msg, len, flags); - if (!copied) { - int data, err = 0; - long timeo; - - timeo = sock_rcvtimeo(sk, nonblock); - data = tcp_bpf_wait_data(sk, psock, flags, timeo, &err); - if (data) { - if (!sk_psock_queue_empty(psock)) - goto msg_bytes_ready; - release_sock(sk); - sk_psock_put(sk, psock); - return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len); - } - if (err) { - ret = err; - goto out; - } - copied = -EAGAIN; - } - ret = copied; -out: - release_sock(sk); - sk_psock_put(sk, psock); - return ret; -} - static int bpf_tcp_ingress(struct sock *sk, struct sk_psock *psock, struct sk_msg *msg, u32 apply_bytes, int flags) { @@ -200,7 +52,6 @@ static int bpf_tcp_ingress(struct sock *sk, struct sk_psock *psock, if (!ret) { msg->sg.start = i; - msg->sg.size -= apply_bytes; sk_psock_queue_msg(psock, tmp); sk_psock_data_ready(sk, psock); } else { @@ -287,10 +138,9 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, struct sk_psock *psock = sk_psock_get(sk); int ret; - if (unlikely(!psock)) { - sk_msg_free(sk, msg); - return 0; - } + if (unlikely(!psock)) + return -EPIPE; + ret = ingress ? bpf_tcp_ingress(sk, psock, msg, bytes, flags) : tcp_bpf_push_locked(sk, msg, bytes, flags, false); sk_psock_put(sk, psock); @@ -298,12 +148,138 @@ int tcp_bpf_sendmsg_redir(struct sock *sk, struct sk_msg *msg, } EXPORT_SYMBOL_GPL(tcp_bpf_sendmsg_redir); +#ifdef CONFIG_BPF_SYSCALL +static int tcp_msg_wait_data(struct sock *sk, struct sk_psock *psock, + long timeo) +{ + DEFINE_WAIT_FUNC(wait, woken_wake_function); + int ret = 0; + + if (sk->sk_shutdown & RCV_SHUTDOWN) + return 1; + + if (!timeo) + return ret; + + add_wait_queue(sk_sleep(sk), &wait); + sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); + ret = sk_wait_event(sk, &timeo, + !list_empty(&psock->ingress_msg) || + !skb_queue_empty(&sk->sk_receive_queue), &wait); + sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); + remove_wait_queue(sk_sleep(sk), &wait); + return ret; +} + +static int tcp_bpf_recvmsg_parser(struct sock *sk, + struct msghdr *msg, + size_t len, + int flags, + int *addr_len) +{ + struct sk_psock *psock; + int copied; + + if (unlikely(flags & MSG_ERRQUEUE)) + return inet_recv_error(sk, msg, len, addr_len); + + psock = sk_psock_get(sk); + if (unlikely(!psock)) + return tcp_recvmsg(sk, msg, len, flags, addr_len); + + lock_sock(sk); +msg_bytes_ready: + copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + if (!copied) { + long timeo; + int data; + + if (sock_flag(sk, SOCK_DONE)) + goto out; + + if (sk->sk_err) { + copied = sock_error(sk); + goto out; + } + + if (sk->sk_shutdown & RCV_SHUTDOWN) + goto out; + + if (sk->sk_state == TCP_CLOSE) { + copied = -ENOTCONN; + goto out; + } + + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + if (!timeo) { + copied = -EAGAIN; + goto out; + } + + if (signal_pending(current)) { + copied = sock_intr_errno(timeo); + goto out; + } + + data = tcp_msg_wait_data(sk, psock, timeo); + if (data && !sk_psock_queue_empty(psock)) + goto msg_bytes_ready; + copied = -EAGAIN; + } +out: + release_sock(sk); + sk_psock_put(sk, psock); + return copied; +} + +static int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, + int flags, int *addr_len) +{ + struct sk_psock *psock; + int copied, ret; + + if (unlikely(flags & MSG_ERRQUEUE)) + return inet_recv_error(sk, msg, len, addr_len); + + psock = sk_psock_get(sk); + if (unlikely(!psock)) + return tcp_recvmsg(sk, msg, len, flags, addr_len); + if (!skb_queue_empty(&sk->sk_receive_queue) && + sk_psock_queue_empty(psock)) { + sk_psock_put(sk, psock); + return tcp_recvmsg(sk, msg, len, flags, addr_len); + } + lock_sock(sk); +msg_bytes_ready: + copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + if (!copied) { + long timeo; + int data; + + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + data = tcp_msg_wait_data(sk, psock, timeo); + if (data) { + if (!sk_psock_queue_empty(psock)) + goto msg_bytes_ready; + release_sock(sk); + sk_psock_put(sk, psock); + return tcp_recvmsg(sk, msg, len, flags, addr_len); + } + copied = -EAGAIN; + } + ret = copied; + release_sock(sk); + sk_psock_put(sk, psock); + return ret; +} + static int tcp_bpf_send_verdict(struct sock *sk, struct sk_psock *psock, struct sk_msg *msg, int *copied, int flags) { bool cork = false, enospc = sk_msg_full(msg); struct sock *sk_redir; - u32 tosend, delta = 0; + u32 tosend, origsize, sent, delta = 0; + u32 eval = __SK_NONE; int ret; more_data: @@ -347,13 +323,26 @@ more_data: case __SK_REDIRECT: sk_redir = psock->sk_redir; sk_msg_apply_bytes(psock, tosend); + if (!psock->apply_bytes) { + /* Clean up before releasing the sock lock. */ + eval = psock->eval; + psock->eval = __SK_NONE; + psock->sk_redir = NULL; + } if (psock->cork) { cork = true; psock->cork = NULL; } sk_msg_return(sk, msg, tosend); release_sock(sk); + + origsize = msg->sg.size; ret = tcp_bpf_sendmsg_redir(sk_redir, msg, tosend, flags); + sent = origsize - msg->sg.size; + + if (eval == __SK_REDIRECT) + sock_put(sk_redir); + lock_sock(sk); if (unlikely(ret < 0)) { int free = sk_msg_free_nocharge(sk, msg); @@ -386,8 +375,11 @@ more_data: } if (msg && msg->sg.data[msg->sg.start].page_link && - msg->sg.data[msg->sg.start].length) + msg->sg.data[msg->sg.start].length) { + if (eval == __SK_REDIRECT) + sk_mem_charge(sk, tosend - sent); goto more_data; + } } return ret; } @@ -528,57 +520,6 @@ out_err: return copied ? copied : err; } -static void tcp_bpf_remove(struct sock *sk, struct sk_psock *psock) -{ - struct sk_psock_link *link; - - while ((link = sk_psock_link_pop(psock))) { - sk_psock_unlink(sk, link); - sk_psock_free_link(link); - } -} - -static void tcp_bpf_unhash(struct sock *sk) -{ - void (*saved_unhash)(struct sock *sk); - struct sk_psock *psock; - - rcu_read_lock(); - psock = sk_psock(sk); - if (unlikely(!psock)) { - rcu_read_unlock(); - if (sk->sk_prot->unhash) - sk->sk_prot->unhash(sk); - return; - } - - saved_unhash = psock->saved_unhash; - tcp_bpf_remove(sk, psock); - rcu_read_unlock(); - saved_unhash(sk); -} - -static void tcp_bpf_close(struct sock *sk, long timeout) -{ - void (*saved_close)(struct sock *sk, long timeout); - struct sk_psock *psock; - - lock_sock(sk); - rcu_read_lock(); - psock = sk_psock(sk); - if (unlikely(!psock)) { - rcu_read_unlock(); - release_sock(sk); - return sk->sk_prot->close(sk, timeout); - } - - saved_close = psock->saved_close; - tcp_bpf_remove(sk, psock); - rcu_read_unlock(); - release_sock(sk); - saved_close(sk, timeout); -} - enum { TCP_BPF_IPV4, TCP_BPF_IPV6, @@ -588,6 +529,8 @@ enum { enum { TCP_BPF_BASE, TCP_BPF_TX, + TCP_BPF_RX, + TCP_BPF_TXRX, TCP_BPF_NUM_CFGS, }; @@ -599,20 +542,25 @@ static void tcp_bpf_rebuild_protos(struct proto prot[TCP_BPF_NUM_CFGS], struct proto *base) { prot[TCP_BPF_BASE] = *base; - prot[TCP_BPF_BASE].unhash = tcp_bpf_unhash; - prot[TCP_BPF_BASE].close = tcp_bpf_close; + prot[TCP_BPF_BASE].destroy = sock_map_destroy; + prot[TCP_BPF_BASE].close = sock_map_close; prot[TCP_BPF_BASE].recvmsg = tcp_bpf_recvmsg; - prot[TCP_BPF_BASE].stream_memory_read = tcp_bpf_stream_read; + prot[TCP_BPF_BASE].sock_is_readable = sk_msg_is_readable; prot[TCP_BPF_TX] = prot[TCP_BPF_BASE]; prot[TCP_BPF_TX].sendmsg = tcp_bpf_sendmsg; prot[TCP_BPF_TX].sendpage = tcp_bpf_sendpage; + + prot[TCP_BPF_RX] = prot[TCP_BPF_BASE]; + prot[TCP_BPF_RX].recvmsg = tcp_bpf_recvmsg_parser; + + prot[TCP_BPF_TXRX] = prot[TCP_BPF_TX]; + prot[TCP_BPF_TXRX].recvmsg = tcp_bpf_recvmsg_parser; } -static void tcp_bpf_check_v6_needs_rebuild(struct sock *sk, struct proto *ops) +static void tcp_bpf_check_v6_needs_rebuild(struct proto *ops) { - if (sk->sk_family == AF_INET6 && - unlikely(ops != smp_load_acquire(&tcpv6_prot_saved))) { + if (unlikely(ops != smp_load_acquire(&tcpv6_prot_saved))) { spin_lock_bh(&tcpv6_prot_lock); if (likely(ops != tcpv6_prot_saved)) { tcp_bpf_rebuild_protos(tcp_bpf_prots[TCP_BPF_IPV6], ops); @@ -627,27 +575,7 @@ static int __init tcp_bpf_v4_build_proto(void) tcp_bpf_rebuild_protos(tcp_bpf_prots[TCP_BPF_IPV4], &tcp_prot); return 0; } -core_initcall(tcp_bpf_v4_build_proto); - -static void tcp_bpf_update_sk_prot(struct sock *sk, struct sk_psock *psock) -{ - int family = sk->sk_family == AF_INET6 ? TCP_BPF_IPV6 : TCP_BPF_IPV4; - int config = psock->progs.msg_parser ? TCP_BPF_TX : TCP_BPF_BASE; - - sk_psock_update_proto(sk, psock, &tcp_bpf_prots[family][config]); -} - -static void tcp_bpf_reinit_sk_prot(struct sock *sk, struct sk_psock *psock) -{ - int family = sk->sk_family == AF_INET6 ? TCP_BPF_IPV6 : TCP_BPF_IPV4; - int config = psock->progs.msg_parser ? TCP_BPF_TX : TCP_BPF_BASE; - - /* Reinit occurs when program types change e.g. TCP_BPF_TX is removed - * or added requiring sk_prot hook updates. We keep original saved - * hooks in this case. - */ - sk->sk_prot = &tcp_bpf_prots[family][config]; -} +late_initcall(tcp_bpf_v4_build_proto); static int tcp_bpf_assert_proto_ops(struct proto *ops) { @@ -660,34 +588,56 @@ static int tcp_bpf_assert_proto_ops(struct proto *ops) ops->sendpage == tcp_sendpage ? 0 : -ENOTSUPP; } -void tcp_bpf_reinit(struct sock *sk) +int tcp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore) { - struct sk_psock *psock; - - sock_owned_by_me(sk); + int family = sk->sk_family == AF_INET6 ? TCP_BPF_IPV6 : TCP_BPF_IPV4; + int config = psock->progs.msg_parser ? TCP_BPF_TX : TCP_BPF_BASE; - rcu_read_lock(); - psock = sk_psock(sk); - tcp_bpf_reinit_sk_prot(sk, psock); - rcu_read_unlock(); -} + if (psock->progs.stream_verdict || psock->progs.skb_verdict) { + config = (config == TCP_BPF_TX) ? TCP_BPF_TXRX : TCP_BPF_RX; + } -int tcp_bpf_init(struct sock *sk) -{ - struct proto *ops = READ_ONCE(sk->sk_prot); - struct sk_psock *psock; + if (restore) { + if (inet_csk_has_ulp(sk)) { + /* TLS does not have an unhash proto in SW cases, + * but we need to ensure we stop using the sock_map + * unhash routine because the associated psock is being + * removed. So use the original unhash handler. + */ + WRITE_ONCE(sk->sk_prot->unhash, psock->saved_unhash); + tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space); + } else { + sk->sk_write_space = psock->saved_write_space; + /* Pairs with lockless read in sk_clone_lock() */ + sock_replace_proto(sk, psock->sk_proto); + } + return 0; + } - sock_owned_by_me(sk); + if (sk->sk_family == AF_INET6) { + if (tcp_bpf_assert_proto_ops(psock->sk_proto)) + return -EINVAL; - rcu_read_lock(); - psock = sk_psock(sk); - if (unlikely(!psock || psock->sk_proto || - tcp_bpf_assert_proto_ops(ops))) { - rcu_read_unlock(); - return -EINVAL; + tcp_bpf_check_v6_needs_rebuild(psock->sk_proto); } - tcp_bpf_check_v6_needs_rebuild(sk, ops); - tcp_bpf_update_sk_prot(sk, psock); - rcu_read_unlock(); + + /* Pairs with lockless read in sk_clone_lock() */ + sock_replace_proto(sk, &tcp_bpf_prots[family][config]); return 0; } +EXPORT_SYMBOL_GPL(tcp_bpf_update_proto); + +/* If a child got cloned from a listening socket that had tcp_bpf + * protocol callbacks installed, we need to restore the callbacks to + * the default ones because the child does not inherit the psock state + * that tcp_bpf callbacks expect. + */ +void tcp_bpf_clone(const struct sock *sk, struct sock *newsk) +{ + int family = sk->sk_family == AF_INET6 ? TCP_BPF_IPV6 : TCP_BPF_IPV4; + struct proto *prot = newsk->sk_prot; + + if (prot == &tcp_bpf_prots[family][TCP_BPF_BASE]) + newsk->sk_prot = sk->sk_prot_creator; +} +#endif /* CONFIG_BPF_SYSCALL */ diff --git a/net/ipv4/tcp_cdg.c b/net/ipv4/tcp_cdg.c index 709d23801823..ba4d98e510e0 100644 --- a/net/ipv4/tcp_cdg.c +++ b/net/ipv4/tcp_cdg.c @@ -161,8 +161,8 @@ static void tcp_cdg_hystart_update(struct sock *sk) LINUX_MIB_TCPHYSTARTTRAINDETECT); NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPHYSTARTTRAINCWND, - tp->snd_cwnd); - tp->snd_ssthresh = tp->snd_cwnd; + tcp_snd_cwnd(tp)); + tp->snd_ssthresh = tcp_snd_cwnd(tp); return; } } @@ -180,8 +180,8 @@ static void tcp_cdg_hystart_update(struct sock *sk) LINUX_MIB_TCPHYSTARTDELAYDETECT); NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPHYSTARTDELAYCWND, - tp->snd_cwnd); - tp->snd_ssthresh = tp->snd_cwnd; + tcp_snd_cwnd(tp)); + tp->snd_ssthresh = tcp_snd_cwnd(tp); } } } @@ -243,7 +243,7 @@ static bool tcp_cdg_backoff(struct sock *sk, u32 grad) struct cdg *ca = inet_csk_ca(sk); struct tcp_sock *tp = tcp_sk(sk); - if (prandom_u32() <= nexp_u32(grad * backoff_factor)) + if (get_random_u32() <= nexp_u32(grad * backoff_factor)) return false; if (use_ineff) { @@ -252,7 +252,7 @@ static bool tcp_cdg_backoff(struct sock *sk, u32 grad) return false; } - ca->shadow_wnd = max(ca->shadow_wnd, tp->snd_cwnd); + ca->shadow_wnd = max(ca->shadow_wnd, tcp_snd_cwnd(tp)); ca->state = CDG_BACKOFF; tcp_enter_cwr(sk); return true; @@ -285,14 +285,14 @@ static void tcp_cdg_cong_avoid(struct sock *sk, u32 ack, u32 acked) } if (!tcp_is_cwnd_limited(sk)) { - ca->shadow_wnd = min(ca->shadow_wnd, tp->snd_cwnd); + ca->shadow_wnd = min(ca->shadow_wnd, tcp_snd_cwnd(tp)); return; } - prior_snd_cwnd = tp->snd_cwnd; + prior_snd_cwnd = tcp_snd_cwnd(tp); tcp_reno_cong_avoid(sk, ack, acked); - incr = tp->snd_cwnd - prior_snd_cwnd; + incr = tcp_snd_cwnd(tp) - prior_snd_cwnd; ca->shadow_wnd = max(ca->shadow_wnd, ca->shadow_wnd + incr); } @@ -331,15 +331,15 @@ static u32 tcp_cdg_ssthresh(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); if (ca->state == CDG_BACKOFF) - return max(2U, (tp->snd_cwnd * min(1024U, backoff_beta)) >> 10); + return max(2U, (tcp_snd_cwnd(tp) * min(1024U, backoff_beta)) >> 10); if (ca->state == CDG_NONFULL && use_tolerance) - return tp->snd_cwnd; + return tcp_snd_cwnd(tp); - ca->shadow_wnd = min(ca->shadow_wnd >> 1, tp->snd_cwnd); + ca->shadow_wnd = min(ca->shadow_wnd >> 1, tcp_snd_cwnd(tp)); if (use_shadow) - return max3(2U, ca->shadow_wnd, tp->snd_cwnd >> 1); - return max(2U, tp->snd_cwnd >> 1); + return max3(2U, ca->shadow_wnd, tcp_snd_cwnd(tp) >> 1); + return max(2U, tcp_snd_cwnd(tp) >> 1); } static void tcp_cdg_cwnd_event(struct sock *sk, const enum tcp_ca_event ev) @@ -357,7 +357,7 @@ static void tcp_cdg_cwnd_event(struct sock *sk, const enum tcp_ca_event ev) ca->gradients = gradients; ca->rtt_seq = tp->snd_nxt; - ca->shadow_wnd = tp->snd_cwnd; + ca->shadow_wnd = tcp_snd_cwnd(tp); break; case CA_EVENT_COMPLETE_CWR: ca->state = CDG_UNKNOWN; @@ -375,12 +375,13 @@ static void tcp_cdg_init(struct sock *sk) struct cdg *ca = inet_csk_ca(sk); struct tcp_sock *tp = tcp_sk(sk); + ca->gradients = NULL; /* We silently fall back to window = 1 if allocation fails. */ if (window > 1) ca->gradients = kcalloc(window, sizeof(ca->gradients[0]), GFP_NOWAIT | __GFP_NOWARN); ca->rtt_seq = tp->snd_nxt; - ca->shadow_wnd = tp->snd_cwnd; + ca->shadow_wnd = tcp_snd_cwnd(tp); } static void tcp_cdg_release(struct sock *sk) @@ -388,6 +389,7 @@ static void tcp_cdg_release(struct sock *sk) struct cdg *ca = inet_csk_ca(sk); kfree(ca->gradients); + ca->gradients = NULL; } static struct tcp_congestion_ops tcp_cdg __read_mostly = { diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 3172e31987be..d3cae40749e8 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -16,6 +16,7 @@ #include <linux/gfp.h> #include <linux/jhash.h> #include <net/tcp.h> +#include <trace/events/tcp.h> static DEFINE_SPINLOCK(tcp_cong_list_lock); static LIST_HEAD(tcp_cong_list); @@ -33,6 +34,17 @@ struct tcp_congestion_ops *tcp_ca_find(const char *name) return NULL; } +void tcp_set_ca_state(struct sock *sk, const u8 ca_state) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + + trace_tcp_cong_state_set(sk, ca_state); + + if (icsk->icsk_ca_ops->set_state) + icsk->icsk_ca_ops->set_state(sk, ca_state); + icsk->icsk_ca_state = ca_state; +} + /* Must be called with rcu lock held */ static struct tcp_congestion_ops *tcp_ca_find_autoload(struct net *net, const char *name) @@ -135,7 +147,6 @@ u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca) return key; } -EXPORT_SYMBOL_GPL(tcp_ca_get_key_by_name); char *tcp_ca_get_name_by_key(u32 key, char *buffer) { @@ -151,7 +162,6 @@ char *tcp_ca_get_name_by_key(u32 key, char *buffer) return ret; } -EXPORT_SYMBOL_GPL(tcp_ca_get_name_by_key); /* Assign choice of congestion control. */ void tcp_assign_congestion_control(struct sock *sk) @@ -176,7 +186,7 @@ void tcp_assign_congestion_control(struct sock *sk) void tcp_init_congestion_control(struct sock *sk) { - const struct inet_connection_sock *icsk = inet_csk(sk); + struct inet_connection_sock *icsk = inet_csk(sk); tcp_sk(sk)->prior_ssthresh = 0; if (icsk->icsk_ca_ops->init) @@ -185,6 +195,7 @@ void tcp_init_congestion_control(struct sock *sk) INET_ECN_xmit(sk); else INET_ECN_dontxmit(sk); + icsk->icsk_ca_initialized = 1; } static void tcp_reinit_congestion_control(struct sock *sk, @@ -197,7 +208,12 @@ static void tcp_reinit_congestion_control(struct sock *sk, icsk->icsk_ca_setsockopt = 1; memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv)); - if (sk->sk_state != TCP_CLOSE) + if (ca->flags & TCP_CONG_NEEDS_ECN) + INET_ECN_xmit(sk); + else + INET_ECN_dontxmit(sk); + + if (!((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) tcp_init_congestion_control(sk); } @@ -224,6 +240,10 @@ int tcp_set_default_congestion_control(struct net *net, const char *name) ret = -ENOENT; } else if (!bpf_try_module_get(ca, ca->owner)) { ret = -EBUSY; + } else if (!net_eq(net, &init_net) && + !(ca->flags & TCP_CONG_NON_RESTRICTED)) { + /* Only init netns can set default to a restricted algorithm */ + ret = -EPERM; } else { prev = xchg(&net->ipv4.tcp_congestion_control, ca); if (prev) @@ -340,7 +360,7 @@ out: * already initialized. */ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, - bool reinit, bool cap_net_admin) + bool cap_net_admin) { struct inet_connection_sock *icsk = inet_csk(sk); const struct tcp_congestion_ops *ca; @@ -361,28 +381,14 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, goto out; } - if (!ca) { + if (!ca) err = -ENOENT; - } else if (!load) { - const struct tcp_congestion_ops *old_ca = icsk->icsk_ca_ops; - - if (bpf_try_module_get(ca, ca->owner)) { - if (reinit) { - tcp_reinit_congestion_control(sk, ca); - } else { - icsk->icsk_ca_ops = ca; - bpf_module_put(old_ca, old_ca->owner); - } - } else { - err = -EBUSY; - } - } else if (!((ca->flags & TCP_CONG_NON_RESTRICTED) || cap_net_admin)) { + else if (!((ca->flags & TCP_CONG_NON_RESTRICTED) || cap_net_admin)) err = -EPERM; - } else if (!bpf_try_module_get(ca, ca->owner)) { + else if (!bpf_try_module_get(ca, ca->owner)) err = -EBUSY; - } else { + else tcp_reinit_congestion_control(sk, ca); - } out: rcu_read_unlock(); return err; @@ -399,10 +405,10 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, */ u32 tcp_slow_start(struct tcp_sock *tp, u32 acked) { - u32 cwnd = min(tp->snd_cwnd + acked, tp->snd_ssthresh); + u32 cwnd = min(tcp_snd_cwnd(tp) + acked, tp->snd_ssthresh); - acked -= cwnd - tp->snd_cwnd; - tp->snd_cwnd = min(cwnd, tp->snd_cwnd_clamp); + acked -= cwnd - tcp_snd_cwnd(tp); + tcp_snd_cwnd_set(tp, min(cwnd, tp->snd_cwnd_clamp)); return acked; } @@ -416,7 +422,7 @@ void tcp_cong_avoid_ai(struct tcp_sock *tp, u32 w, u32 acked) /* If credits accumulated at a higher w, apply them gently now. */ if (tp->snd_cwnd_cnt >= w) { tp->snd_cwnd_cnt = 0; - tp->snd_cwnd++; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); } tp->snd_cwnd_cnt += acked; @@ -424,9 +430,9 @@ void tcp_cong_avoid_ai(struct tcp_sock *tp, u32 w, u32 acked) u32 delta = tp->snd_cwnd_cnt / w; tp->snd_cwnd_cnt -= delta * w; - tp->snd_cwnd += delta; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + delta); } - tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_cwnd_clamp); + tcp_snd_cwnd_set(tp, min(tcp_snd_cwnd(tp), tp->snd_cwnd_clamp)); } EXPORT_SYMBOL_GPL(tcp_cong_avoid_ai); @@ -451,7 +457,7 @@ void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked) return; } /* In dangerous area, increase slowly. */ - tcp_cong_avoid_ai(tp, tp->snd_cwnd, acked); + tcp_cong_avoid_ai(tp, tcp_snd_cwnd(tp), acked); } EXPORT_SYMBOL_GPL(tcp_reno_cong_avoid); @@ -460,7 +466,7 @@ u32 tcp_reno_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - return max(tp->snd_cwnd >> 1U, 2U); + return max(tcp_snd_cwnd(tp) >> 1U, 2U); } EXPORT_SYMBOL_GPL(tcp_reno_ssthresh); @@ -468,7 +474,7 @@ u32 tcp_reno_undo_cwnd(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - return max(tp->snd_cwnd, tp->prior_cwnd); + return max(tcp_snd_cwnd(tp), tp->prior_cwnd); } EXPORT_SYMBOL_GPL(tcp_reno_undo_cwnd); diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c index 8f8eefd3a3ce..768c10c1f649 100644 --- a/net/ipv4/tcp_cubic.c +++ b/net/ipv4/tcp_cubic.c @@ -25,6 +25,8 @@ */ #include <linux/mm.h> +#include <linux/btf.h> +#include <linux/btf_ids.h> #include <linux/module.h> #include <linux/math64.h> #include <net/tcp.h> @@ -104,16 +106,7 @@ struct bictcp { static inline void bictcp_reset(struct bictcp *ca) { - ca->cnt = 0; - ca->last_max_cwnd = 0; - ca->last_cwnd = 0; - ca->last_time = 0; - ca->bic_origin_point = 0; - ca->bic_K = 0; - ca->delay_min = 0; - ca->epoch_start = 0; - ca->ack_cnt = 0; - ca->tcp_cwnd = 0; + memset(ca, 0, offsetof(struct bictcp, unused)); ca->found = 0; } @@ -133,7 +126,7 @@ static inline void bictcp_hystart_reset(struct sock *sk) ca->sample_cnt = 0; } -static void bictcp_init(struct sock *sk) +static void cubictcp_init(struct sock *sk) { struct bictcp *ca = inet_csk_ca(sk); @@ -146,7 +139,7 @@ static void bictcp_init(struct sock *sk) tcp_sk(sk)->snd_ssthresh = initial_ssthresh; } -static void bictcp_cwnd_event(struct sock *sk, enum tcp_ca_event event) +static void cubictcp_cwnd_event(struct sock *sk, enum tcp_ca_event event) { if (event == CA_EVENT_TX_START) { struct bictcp *ca = inet_csk_ca(sk); @@ -328,7 +321,7 @@ tcp_friendliness: ca->cnt = max(ca->cnt, 2U); } -static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) +static void cubictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct bictcp *ca = inet_csk_ca(sk); @@ -337,17 +330,15 @@ static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) return; if (tcp_in_slow_start(tp)) { - if (hystart && after(ack, ca->end_seq)) - bictcp_hystart_reset(sk); acked = tcp_slow_start(tp, acked); if (!acked) return; } - bictcp_update(ca, tp->snd_cwnd, acked); + bictcp_update(ca, tcp_snd_cwnd(tp), acked); tcp_cong_avoid_ai(tp, ca->cnt, acked); } -static u32 bictcp_recalc_ssthresh(struct sock *sk) +static u32 cubictcp_recalc_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); struct bictcp *ca = inet_csk_ca(sk); @@ -355,16 +346,16 @@ static u32 bictcp_recalc_ssthresh(struct sock *sk) ca->epoch_start = 0; /* end of epoch */ /* Wmax and fast convergence */ - if (tp->snd_cwnd < ca->last_max_cwnd && fast_convergence) - ca->last_max_cwnd = (tp->snd_cwnd * (BICTCP_BETA_SCALE + beta)) + if (tcp_snd_cwnd(tp) < ca->last_max_cwnd && fast_convergence) + ca->last_max_cwnd = (tcp_snd_cwnd(tp) * (BICTCP_BETA_SCALE + beta)) / (2 * BICTCP_BETA_SCALE); else - ca->last_max_cwnd = tp->snd_cwnd; + ca->last_max_cwnd = tcp_snd_cwnd(tp); - return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U); + return max((tcp_snd_cwnd(tp) * beta) / BICTCP_BETA_SCALE, 2U); } -static void bictcp_state(struct sock *sk, u8 new_state) +static void cubictcp_state(struct sock *sk, u8 new_state) { if (new_state == TCP_CA_Loss) { bictcp_reset(inet_csk_ca(sk)); @@ -381,7 +372,7 @@ static void bictcp_state(struct sock *sk, u8 new_state) * We apply another 100% factor because @rate is doubled at this point. * We cap the cushion to 1ms. */ -static u32 hystart_ack_delay(struct sock *sk) +static u32 hystart_ack_delay(const struct sock *sk) { unsigned long rate; @@ -389,7 +380,7 @@ static u32 hystart_ack_delay(struct sock *sk) if (!rate) return 0; return min_t(u64, USEC_PER_MSEC, - div64_ul((u64)GSO_MAX_SIZE * 4 * USEC_PER_SEC, rate)); + div64_ul((u64)sk->sk_gso_max_size * 4 * USEC_PER_SEC, rate)); } static void hystart_update(struct sock *sk, u32 delay) @@ -398,6 +389,9 @@ static void hystart_update(struct sock *sk, u32 delay) struct bictcp *ca = inet_csk_ca(sk); u32 threshold; + if (after(tp->snd_una, ca->end_seq)) + bictcp_hystart_reset(sk); + if (hystart_detect & HYSTART_ACK_TRAIN) { u32 now = bictcp_clock_us(sk); @@ -419,23 +413,22 @@ static void hystart_update(struct sock *sk, u32 delay) ca->found = 1; pr_debug("hystart_ack_train (%u > %u) delay_min %u (+ ack_delay %u) cwnd %u\n", now - ca->round_start, threshold, - ca->delay_min, hystart_ack_delay(sk), tp->snd_cwnd); + ca->delay_min, hystart_ack_delay(sk), tcp_snd_cwnd(tp)); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPHYSTARTTRAINDETECT); NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPHYSTARTTRAINCWND, - tp->snd_cwnd); - tp->snd_ssthresh = tp->snd_cwnd; + tcp_snd_cwnd(tp)); + tp->snd_ssthresh = tcp_snd_cwnd(tp); } } } if (hystart_detect & HYSTART_DELAY) { /* obtain the minimum delay of more than sampling packets */ + if (ca->curr_rtt > delay) + ca->curr_rtt = delay; if (ca->sample_cnt < HYSTART_MIN_SAMPLES) { - if (ca->curr_rtt > delay) - ca->curr_rtt = delay; - ca->sample_cnt++; } else { if (ca->curr_rtt > ca->delay_min + @@ -445,14 +438,14 @@ static void hystart_update(struct sock *sk, u32 delay) LINUX_MIB_TCPHYSTARTDELAYDETECT); NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPHYSTARTDELAYCWND, - tp->snd_cwnd); - tp->snd_ssthresh = tp->snd_cwnd; + tcp_snd_cwnd(tp)); + tp->snd_ssthresh = tcp_snd_cwnd(tp); } } } } -static void bictcp_acked(struct sock *sk, const struct ack_sample *sample) +static void cubictcp_acked(struct sock *sk, const struct ack_sample *sample) { const struct tcp_sock *tp = tcp_sk(sk); struct bictcp *ca = inet_csk_ca(sk); @@ -476,24 +469,44 @@ static void bictcp_acked(struct sock *sk, const struct ack_sample *sample) /* hystart triggers when cwnd is larger than some threshold */ if (!ca->found && tcp_in_slow_start(tp) && hystart && - tp->snd_cwnd >= hystart_low_window) + tcp_snd_cwnd(tp) >= hystart_low_window) hystart_update(sk, delay); } static struct tcp_congestion_ops cubictcp __read_mostly = { - .init = bictcp_init, - .ssthresh = bictcp_recalc_ssthresh, - .cong_avoid = bictcp_cong_avoid, - .set_state = bictcp_state, + .init = cubictcp_init, + .ssthresh = cubictcp_recalc_ssthresh, + .cong_avoid = cubictcp_cong_avoid, + .set_state = cubictcp_state, .undo_cwnd = tcp_reno_undo_cwnd, - .cwnd_event = bictcp_cwnd_event, - .pkts_acked = bictcp_acked, + .cwnd_event = cubictcp_cwnd_event, + .pkts_acked = cubictcp_acked, .owner = THIS_MODULE, .name = "cubic", }; +BTF_SET8_START(tcp_cubic_check_kfunc_ids) +#ifdef CONFIG_X86 +#ifdef CONFIG_DYNAMIC_FTRACE +BTF_ID_FLAGS(func, cubictcp_init) +BTF_ID_FLAGS(func, cubictcp_recalc_ssthresh) +BTF_ID_FLAGS(func, cubictcp_cong_avoid) +BTF_ID_FLAGS(func, cubictcp_state) +BTF_ID_FLAGS(func, cubictcp_cwnd_event) +BTF_ID_FLAGS(func, cubictcp_acked) +#endif +#endif +BTF_SET8_END(tcp_cubic_check_kfunc_ids) + +static const struct btf_kfunc_id_set tcp_cubic_kfunc_set = { + .owner = THIS_MODULE, + .set = &tcp_cubic_check_kfunc_ids, +}; + static int __init cubictcp_register(void) { + int ret; + BUILD_BUG_ON(sizeof(struct bictcp) > ICSK_CA_PRIV_SIZE); /* Precompute a bunch of the scaling factors that are used per-packet @@ -524,6 +537,9 @@ static int __init cubictcp_register(void) /* divide by bic_scale and by constant Srtt (100ms) */ do_div(cube_factor, bic_scale * 10); + ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_cubic_kfunc_set); + if (ret < 0) + return ret; return tcp_register_congestion_control(&cubictcp); } diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c index 79f705450c16..2a6c0dd665a4 100644 --- a/net/ipv4/tcp_dctcp.c +++ b/net/ipv4/tcp_dctcp.c @@ -36,6 +36,8 @@ * Glenn Judd <glenn.judd@morganstanley.com> */ +#include <linux/btf.h> +#include <linux/btf_ids.h> #include <linux/module.h> #include <linux/mm.h> #include <net/tcp.h> @@ -104,8 +106,8 @@ static u32 dctcp_ssthresh(struct sock *sk) struct dctcp *ca = inet_csk_ca(sk); struct tcp_sock *tp = tcp_sk(sk); - ca->loss_cwnd = tp->snd_cwnd; - return max(tp->snd_cwnd - ((tp->snd_cwnd * ca->dctcp_alpha) >> 11U), 2U); + ca->loss_cwnd = tcp_snd_cwnd(tp); + return max(tcp_snd_cwnd(tp) - ((tcp_snd_cwnd(tp) * ca->dctcp_alpha) >> 11U), 2U); } static void dctcp_update_alpha(struct sock *sk, u32 flags) @@ -146,8 +148,8 @@ static void dctcp_react_to_loss(struct sock *sk) struct dctcp *ca = inet_csk_ca(sk); struct tcp_sock *tp = tcp_sk(sk); - ca->loss_cwnd = tp->snd_cwnd; - tp->snd_ssthresh = max(tp->snd_cwnd >> 1U, 2U); + ca->loss_cwnd = tcp_snd_cwnd(tp); + tp->snd_ssthresh = max(tcp_snd_cwnd(tp) >> 1U, 2U); } static void dctcp_state(struct sock *sk, u8 new_state) @@ -209,8 +211,9 @@ static size_t dctcp_get_info(struct sock *sk, u32 ext, int *attr, static u32 dctcp_cwnd_undo(struct sock *sk) { const struct dctcp *ca = inet_csk_ca(sk); + struct tcp_sock *tp = tcp_sk(sk); - return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); + return max(tcp_snd_cwnd(tp), ca->loss_cwnd); } static struct tcp_congestion_ops dctcp __read_mostly = { @@ -236,9 +239,33 @@ static struct tcp_congestion_ops dctcp_reno __read_mostly = { .name = "dctcp-reno", }; +BTF_SET8_START(tcp_dctcp_check_kfunc_ids) +#ifdef CONFIG_X86 +#ifdef CONFIG_DYNAMIC_FTRACE +BTF_ID_FLAGS(func, dctcp_init) +BTF_ID_FLAGS(func, dctcp_update_alpha) +BTF_ID_FLAGS(func, dctcp_cwnd_event) +BTF_ID_FLAGS(func, dctcp_ssthresh) +BTF_ID_FLAGS(func, dctcp_cwnd_undo) +BTF_ID_FLAGS(func, dctcp_state) +#endif +#endif +BTF_SET8_END(tcp_dctcp_check_kfunc_ids) + +static const struct btf_kfunc_id_set tcp_dctcp_kfunc_set = { + .owner = THIS_MODULE, + .set = &tcp_dctcp_check_kfunc_ids, +}; + static int __init dctcp_register(void) { + int ret; + BUILD_BUG_ON(sizeof(struct dctcp) > ICSK_CA_PRIV_SIZE); + + ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_dctcp_kfunc_set); + if (ret < 0) + return ret; return tcp_register_congestion_control(&dctcp); } diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 0d08f9e2d8d0..01b50fa79189 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -179,15 +179,23 @@ static size_t tcp_diag_get_aux_size(struct sock *sk, bool net_admin) } static void tcp_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, struct nlattr *bc) + const struct inet_diag_req_v2 *r) { - inet_diag_dump_icsk(&tcp_hashinfo, skb, cb, r, bc); + struct inet_hashinfo *hinfo; + + hinfo = sock_net(cb->skb->sk)->ipv4.tcp_death_row.hashinfo; + + inet_diag_dump_icsk(hinfo, skb, cb, r); } -static int tcp_diag_dump_one(struct sk_buff *in_skb, const struct nlmsghdr *nlh, +static int tcp_diag_dump_one(struct netlink_callback *cb, const struct inet_diag_req_v2 *req) { - return inet_diag_dump_one_icsk(&tcp_hashinfo, in_skb, nlh, req); + struct inet_hashinfo *hinfo; + + hinfo = sock_net(cb->skb->sk)->ipv4.tcp_death_row.hashinfo; + + return inet_diag_dump_one_icsk(hinfo, cb, req); } #ifdef CONFIG_INET_DIAG_DESTROY @@ -195,9 +203,13 @@ static int tcp_diag_destroy(struct sk_buff *in_skb, const struct inet_diag_req_v2 *req) { struct net *net = sock_net(in_skb->sk); - struct sock *sk = inet_diag_find_one_icsk(net, &tcp_hashinfo, req); + struct inet_hashinfo *hinfo; + struct sock *sk; int err; + hinfo = net->ipv4.tcp_death_row.hashinfo; + sk = inet_diag_find_one_icsk(net, hinfo, req); + if (IS_ERR(sk)) return PTR_ERR(sk); diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 19ad9586c720..45cc7f1ca296 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -1,13 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 -#include <linux/crypto.h> -#include <linux/err.h> -#include <linux/init.h> #include <linux/kernel.h> -#include <linux/list.h> #include <linux/tcp.h> #include <linux/rcupdate.h> -#include <linux/rculist.h> -#include <net/inetpeer.h> #include <net/tcp.h> void tcp_fastopen_init_key_once(struct net *net) @@ -38,7 +32,7 @@ static void tcp_fastopen_ctx_free(struct rcu_head *head) struct tcp_fastopen_context *ctx = container_of(head, struct tcp_fastopen_context, rcu); - kzfree(ctx); + kfree_sensitive(ctx); } void tcp_fastopen_destroy_cipher(struct sock *sk) @@ -55,12 +49,7 @@ void tcp_fastopen_ctx_destroy(struct net *net) { struct tcp_fastopen_context *ctxt; - spin_lock(&net->ipv4.tcp_fastopen_ctx_lock); - - ctxt = rcu_dereference_protected(net->ipv4.tcp_fastopen_ctx, - lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); - rcu_assign_pointer(net->ipv4.tcp_fastopen_ctx, NULL); - spin_unlock(&net->ipv4.tcp_fastopen_ctx_lock); + ctxt = xchg((__force struct tcp_fastopen_context **)&net->ipv4.tcp_fastopen_ctx, NULL); if (ctxt) call_rcu(&ctxt->rcu, tcp_fastopen_ctx_free); @@ -89,18 +78,12 @@ int tcp_fastopen_reset_cipher(struct net *net, struct sock *sk, ctx->num = 1; } - spin_lock(&net->ipv4.tcp_fastopen_ctx_lock); if (sk) { q = &inet_csk(sk)->icsk_accept_queue.fastopenq; - octx = rcu_dereference_protected(q->ctx, - lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); - rcu_assign_pointer(q->ctx, ctx); + octx = xchg((__force struct tcp_fastopen_context **)&q->ctx, ctx); } else { - octx = rcu_dereference_protected(net->ipv4.tcp_fastopen_ctx, - lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); - rcu_assign_pointer(net->ipv4.tcp_fastopen_ctx, ctx); + octx = xchg((__force struct tcp_fastopen_context **)&net->ipv4.tcp_fastopen_ctx, ctx); } - spin_unlock(&net->ipv4.tcp_fastopen_ctx_lock); if (octx) call_rcu(&octx->rcu, tcp_fastopen_ctx_free); @@ -108,6 +91,29 @@ out: return err; } +int tcp_fastopen_get_cipher(struct net *net, struct inet_connection_sock *icsk, + u64 *key) +{ + struct tcp_fastopen_context *ctx; + int n_keys = 0, i; + + rcu_read_lock(); + if (icsk) + ctx = rcu_dereference(icsk->icsk_accept_queue.fastopenq.ctx); + else + ctx = rcu_dereference(net->ipv4.tcp_fastopen_ctx); + if (ctx) { + n_keys = tcp_fastopen_context_len(ctx); + for (i = 0; i < n_keys; i++) { + put_unaligned_le64(ctx->key[i].key[0], key + (i * 2)); + put_unaligned_le64(ctx->key[i].key[1], key + (i * 2) + 1); + } + } + rcu_read_unlock(); + + return n_keys; +} + static bool __tcp_fastopen_cookie_gen_cipher(struct request_sock *req, struct sk_buff *syn, const siphash_key_t *key, @@ -266,13 +272,14 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, * The request socket is not added to the ehash * because it's been added to the accept queue directly. */ + req->timeout = tcp_timeout_init(child); inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS, - TCP_TIMEOUT_INIT, TCP_RTO_MAX); + req->timeout, TCP_RTO_MAX); refcount_set(&req->rsk_refcnt, 2); /* Now finish processing the fastopen child socket. */ - tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); + tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB, skb); tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; @@ -326,7 +333,7 @@ static bool tcp_fastopen_no_cookie(const struct sock *sk, const struct dst_entry *dst, int flag) { - return (sock_net(sk)->ipv4.sysctl_tcp_fastopen & flag) || + return (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen) & flag) || tcp_sk(sk)->fastopen_no_cookie || (dst && dst_metric(dst, RTAX_FASTOPEN_NO_COOKIE)); } @@ -341,7 +348,7 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, const struct dst_entry *dst) { bool syn_data = TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1; - int tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen; + int tcp_fastopen = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen); struct tcp_fastopen_cookie valid_foc = { .len = -1 }; struct sock *child; int ret = 0; @@ -356,8 +363,7 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, return NULL; } - if (syn_data && - tcp_fastopen_no_cookie(sk, dst, TFO_SERVER_COOKIE_NOT_REQD)) + if (tcp_fastopen_no_cookie(sk, dst, TFO_SERVER_COOKIE_NOT_REQD)) goto fastopen; if (foc->len == 0) { @@ -484,8 +490,18 @@ void tcp_fastopen_active_disable(struct sock *sk) { struct net *net = sock_net(sk); + if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen_blackhole_timeout)) + return; + + /* Paired with READ_ONCE() in tcp_fastopen_active_should_disable() */ + WRITE_ONCE(net->ipv4.tfo_active_disable_stamp, jiffies); + + /* Paired with smp_rmb() in tcp_fastopen_active_should_disable(). + * We want net->ipv4.tfo_active_disable_stamp to be updated first. + */ + smp_mb__before_atomic(); atomic_inc(&net->ipv4.tfo_active_disable_times); - net->ipv4.tfo_active_disable_stamp = jiffies; + NET_INC_STATS(net, LINUX_MIB_TCPFASTOPENBLACKHOLE); } @@ -495,18 +511,29 @@ void tcp_fastopen_active_disable(struct sock *sk) */ bool tcp_fastopen_active_should_disable(struct sock *sk) { - unsigned int tfo_bh_timeout = sock_net(sk)->ipv4.sysctl_tcp_fastopen_blackhole_timeout; - int tfo_da_times = atomic_read(&sock_net(sk)->ipv4.tfo_active_disable_times); + unsigned int tfo_bh_timeout = + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen_blackhole_timeout); unsigned long timeout; + int tfo_da_times; int multiplier; + if (!tfo_bh_timeout) + return false; + + tfo_da_times = atomic_read(&sock_net(sk)->ipv4.tfo_active_disable_times); if (!tfo_da_times) return false; - /* Limit timout to max: 2^6 * initial timeout */ + /* Paired with smp_mb__before_atomic() in tcp_fastopen_active_disable() */ + smp_rmb(); + + /* Limit timeout to max: 2^6 * initial timeout */ multiplier = 1 << min(tfo_da_times - 1, 6); - timeout = multiplier * tfo_bh_timeout * HZ; - if (time_before(jiffies, sock_net(sk)->ipv4.tfo_active_disable_stamp + timeout)) + + /* Paired with the WRITE_ONCE() in tcp_fastopen_active_disable(). */ + timeout = READ_ONCE(sock_net(sk)->ipv4.tfo_active_disable_stamp) + + multiplier * tfo_bh_timeout * HZ; + if (time_before(jiffies, timeout)) return true; /* Mark check bit so we can check for successful active TFO diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c index bfdfbb972c57..c6de5ce79ad3 100644 --- a/net/ipv4/tcp_highspeed.c +++ b/net/ipv4/tcp_highspeed.c @@ -2,7 +2,7 @@ /* * Sally Floyd's High Speed TCP (RFC 3649) congestion control * - * See http://www.icir.org/floyd/hstcp.html + * See https://www.icir.org/floyd/hstcp.html * * John Heffner <jheffner@psc.edu> */ @@ -127,22 +127,22 @@ static void hstcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) * snd_cwnd <= * hstcp_aimd_vals[ca->ai].cwnd */ - if (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd) { - while (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd && + if (tcp_snd_cwnd(tp) > hstcp_aimd_vals[ca->ai].cwnd) { + while (tcp_snd_cwnd(tp) > hstcp_aimd_vals[ca->ai].cwnd && ca->ai < HSTCP_AIMD_MAX - 1) ca->ai++; - } else if (ca->ai && tp->snd_cwnd <= hstcp_aimd_vals[ca->ai-1].cwnd) { - while (ca->ai && tp->snd_cwnd <= hstcp_aimd_vals[ca->ai-1].cwnd) + } else if (ca->ai && tcp_snd_cwnd(tp) <= hstcp_aimd_vals[ca->ai-1].cwnd) { + while (ca->ai && tcp_snd_cwnd(tp) <= hstcp_aimd_vals[ca->ai-1].cwnd) ca->ai--; } /* Do additive increase */ - if (tp->snd_cwnd < tp->snd_cwnd_clamp) { + if (tcp_snd_cwnd(tp) < tp->snd_cwnd_clamp) { /* cwnd = cwnd + a(w) / cwnd */ tp->snd_cwnd_cnt += ca->ai + 1; - if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { - tp->snd_cwnd_cnt -= tp->snd_cwnd; - tp->snd_cwnd++; + if (tp->snd_cwnd_cnt >= tcp_snd_cwnd(tp)) { + tp->snd_cwnd_cnt -= tcp_snd_cwnd(tp); + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); } } } @@ -154,7 +154,7 @@ static u32 hstcp_ssthresh(struct sock *sk) struct hstcp *ca = inet_csk_ca(sk); /* Do multiplicative decrease */ - return max(tp->snd_cwnd - ((tp->snd_cwnd * hstcp_aimd_vals[ca->ai].md) >> 8), 2U); + return max(tcp_snd_cwnd(tp) - ((tcp_snd_cwnd(tp) * hstcp_aimd_vals[ca->ai].md) >> 8), 2U); } static struct tcp_congestion_ops tcp_highspeed __read_mostly = { diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c index 88e1f011afe0..52b1f2665dfa 100644 --- a/net/ipv4/tcp_htcp.c +++ b/net/ipv4/tcp_htcp.c @@ -4,7 +4,7 @@ * R.N.Shorten, D.J.Leith: * "H-TCP: TCP for high-speed and long-distance networks" * Proc. PFLDnet, Argonne, 2004. - * http://www.hamilton.ie/net/htcp3.pdf + * https://www.hamilton.ie/net/htcp3.pdf */ #include <linux/mm.h> @@ -124,7 +124,7 @@ static void measure_achieved_throughput(struct sock *sk, ca->packetcount += sample->pkts_acked; - if (ca->packetcount >= tp->snd_cwnd - (ca->alpha >> 7 ? : 1) && + if (ca->packetcount >= tcp_snd_cwnd(tp) - (ca->alpha >> 7 ? : 1) && now - ca->lasttime >= ca->minRTT && ca->minRTT > 0) { __u32 cur_Bi = ca->packetcount * HZ / (now - ca->lasttime); @@ -225,7 +225,7 @@ static u32 htcp_recalc_ssthresh(struct sock *sk) const struct htcp *ca = inet_csk_ca(sk); htcp_param_update(sk); - return max((tp->snd_cwnd * ca->beta) >> 7, 2U); + return max((tcp_snd_cwnd(tp) * ca->beta) >> 7, 2U); } static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) @@ -242,9 +242,9 @@ static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) /* In dangerous area, increase slowly. * In theory this is tp->snd_cwnd += alpha / tp->snd_cwnd */ - if ((tp->snd_cwnd_cnt * ca->alpha)>>7 >= tp->snd_cwnd) { - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; + if ((tp->snd_cwnd_cnt * ca->alpha)>>7 >= tcp_snd_cwnd(tp)) { + if (tcp_snd_cwnd(tp) < tp->snd_cwnd_clamp) + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); tp->snd_cwnd_cnt = 0; htcp_alpha_update(ca); } else diff --git a/net/ipv4/tcp_hybla.c b/net/ipv4/tcp_hybla.c index be39327e04e6..abd7d91807e5 100644 --- a/net/ipv4/tcp_hybla.c +++ b/net/ipv4/tcp_hybla.c @@ -54,7 +54,7 @@ static void hybla_init(struct sock *sk) ca->rho2_7ls = 0; ca->snd_cwnd_cents = 0; ca->hybla_en = true; - tp->snd_cwnd = 2; + tcp_snd_cwnd_set(tp, 2); tp->snd_cwnd_clamp = 65535; /* 1st Rho measurement based on initial srtt */ @@ -62,7 +62,7 @@ static void hybla_init(struct sock *sk) /* set minimum rtt as this is the 1st ever seen */ ca->minrtt_us = tp->srtt_us; - tp->snd_cwnd = ca->rho; + tcp_snd_cwnd_set(tp, ca->rho); } static void hybla_state(struct sock *sk, u8 ca_state) @@ -137,31 +137,31 @@ static void hybla_cong_avoid(struct sock *sk, u32 ack, u32 acked) * as long as increment is estimated as (rho<<7)/window * it already is <<7 and we can easily count its fractions. */ - increment = ca->rho2_7ls / tp->snd_cwnd; + increment = ca->rho2_7ls / tcp_snd_cwnd(tp); if (increment < 128) tp->snd_cwnd_cnt++; } odd = increment % 128; - tp->snd_cwnd += increment >> 7; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + (increment >> 7)); ca->snd_cwnd_cents += odd; /* check when fractions goes >=128 and increase cwnd by 1. */ while (ca->snd_cwnd_cents >= 128) { - tp->snd_cwnd++; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); ca->snd_cwnd_cents -= 128; tp->snd_cwnd_cnt = 0; } /* check when cwnd has not been incremented for a while */ - if (increment == 0 && odd == 0 && tp->snd_cwnd_cnt >= tp->snd_cwnd) { - tp->snd_cwnd++; + if (increment == 0 && odd == 0 && tp->snd_cwnd_cnt >= tcp_snd_cwnd(tp)) { + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); tp->snd_cwnd_cnt = 0; } /* clamp down slowstart cwnd to ssthresh value. */ if (is_slowstart) - tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); + tcp_snd_cwnd_set(tp, min(tcp_snd_cwnd(tp), tp->snd_ssthresh)); - tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); + tcp_snd_cwnd_set(tp, min(tcp_snd_cwnd(tp), tp->snd_cwnd_clamp)); } static struct tcp_congestion_ops tcp_hybla __read_mostly = { diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c index 00e54873213e..c0c81a2c77fa 100644 --- a/net/ipv4/tcp_illinois.c +++ b/net/ipv4/tcp_illinois.c @@ -224,7 +224,7 @@ static void update_params(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct illinois *ca = inet_csk_ca(sk); - if (tp->snd_cwnd < win_thresh) { + if (tcp_snd_cwnd(tp) < win_thresh) { ca->alpha = ALPHA_BASE; ca->beta = BETA_BASE; } else if (ca->cnt_rtt > 0) { @@ -284,9 +284,9 @@ static void tcp_illinois_cong_avoid(struct sock *sk, u32 ack, u32 acked) * tp->snd_cwnd += alpha/tp->snd_cwnd */ delta = (tp->snd_cwnd_cnt * ca->alpha) >> ALPHA_SHIFT; - if (delta >= tp->snd_cwnd) { - tp->snd_cwnd = min(tp->snd_cwnd + delta / tp->snd_cwnd, - (u32)tp->snd_cwnd_clamp); + if (delta >= tcp_snd_cwnd(tp)) { + tcp_snd_cwnd_set(tp, min(tcp_snd_cwnd(tp) + delta / tcp_snd_cwnd(tp), + (u32)tp->snd_cwnd_clamp)); tp->snd_cwnd_cnt = 0; } } @@ -296,9 +296,11 @@ static u32 tcp_illinois_ssthresh(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct illinois *ca = inet_csk_ca(sk); + u32 decr; /* Multiplicative decrease */ - return max(tp->snd_cwnd - ((tp->snd_cwnd * ca->beta) >> BETA_SHIFT), 2U); + decr = (tcp_snd_cwnd(tp) * ca->beta) >> BETA_SHIFT; + return max(tcp_snd_cwnd(tp) - decr, 2U); } /* Extract info for Tcp socket info provided via netlink. */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6b6b57000dad..0640453fce54 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -100,6 +100,7 @@ int sysctl_tcp_max_orphans __read_mostly = NR_FILE; #define FLAG_UPDATE_TS_RECENT 0x4000 /* tcp_replace_ts_recent() */ #define FLAG_NO_CHALLENGE_ACK 0x8000 /* do not call tcp_send_challenge_ack() */ #define FLAG_ACK_MAYBE_DELAYED 0x10000 /* Likely a delayed ACK */ +#define FLAG_DSACK_TLP 0x20000 /* DSACK for tail loss probe */ #define FLAG_ACKED (FLAG_DATA_ACKED|FLAG_SYN_ACKED) #define FLAG_NOT_DUP (FLAG_DATA|FLAG_WIN_UPDATE|FLAG_ACKED) @@ -138,6 +139,69 @@ void clean_acked_data_flush(void) EXPORT_SYMBOL_GPL(clean_acked_data_flush); #endif +#ifdef CONFIG_CGROUP_BPF +static void bpf_skops_parse_hdr(struct sock *sk, struct sk_buff *skb) +{ + bool unknown_opt = tcp_sk(sk)->rx_opt.saw_unknown && + BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk), + BPF_SOCK_OPS_PARSE_UNKNOWN_HDR_OPT_CB_FLAG); + bool parse_all_opt = BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk), + BPF_SOCK_OPS_PARSE_ALL_HDR_OPT_CB_FLAG); + struct bpf_sock_ops_kern sock_ops; + + if (likely(!unknown_opt && !parse_all_opt)) + return; + + /* The skb will be handled in the + * bpf_skops_established() or + * bpf_skops_write_hdr_opt(). + */ + switch (sk->sk_state) { + case TCP_SYN_RECV: + case TCP_SYN_SENT: + case TCP_LISTEN: + return; + } + + sock_owned_by_me(sk); + + memset(&sock_ops, 0, offsetof(struct bpf_sock_ops_kern, temp)); + sock_ops.op = BPF_SOCK_OPS_PARSE_HDR_OPT_CB; + sock_ops.is_fullsock = 1; + sock_ops.sk = sk; + bpf_skops_init_skb(&sock_ops, skb, tcp_hdrlen(skb)); + + BPF_CGROUP_RUN_PROG_SOCK_OPS(&sock_ops); +} + +static void bpf_skops_established(struct sock *sk, int bpf_op, + struct sk_buff *skb) +{ + struct bpf_sock_ops_kern sock_ops; + + sock_owned_by_me(sk); + + memset(&sock_ops, 0, offsetof(struct bpf_sock_ops_kern, temp)); + sock_ops.op = bpf_op; + sock_ops.is_fullsock = 1; + sock_ops.sk = sk; + /* sk with TCP_REPAIR_ON does not have skb in tcp_finish_connect */ + if (skb) + bpf_skops_init_skb(&sock_ops, skb, tcp_hdrlen(skb)); + + BPF_CGROUP_RUN_PROG_SOCK_OPS(&sock_ops); +} +#else +static void bpf_skops_parse_hdr(struct sock *sk, struct sk_buff *skb) +{ +} + +static void bpf_skops_established(struct sock *sk, int bpf_op, + struct sk_buff *skb) +{ +} +#endif + static void tcp_gro_dev_warn(struct sock *sk, const struct sk_buff *skb, unsigned int len) { @@ -261,7 +325,8 @@ static void tcp_ecn_accept_cwr(struct sock *sk, const struct sk_buff *skb) * cwnd may be very low (even just 1 packet), so we should ACK * immediately. */ - inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW; + if (TCP_SKB_CB(skb)->seq != TCP_SKB_CB(skb)->end_seq) + inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW; } } @@ -349,7 +414,7 @@ static void tcp_sndbuf_expand(struct sock *sk) per_mss = roundup_pow_of_two(per_mss) + SKB_DATA_ALIGN(sizeof(struct sk_buff)); - nr_segs = max_t(u32, TCP_INIT_CWND, tp->snd_cwnd); + nr_segs = max_t(u32, TCP_INIT_CWND, tcp_snd_cwnd(tp)); nr_segs = max_t(u32, nr_segs, tp->reordering + 1); /* Fast Recovery (RFC 5681 3.2) : @@ -361,7 +426,7 @@ static void tcp_sndbuf_expand(struct sock *sk) if (sk->sk_sndbuf < sndmem) WRITE_ONCE(sk->sk_sndbuf, - min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2])); + min(sndmem, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[2]))); } /* 2. Tuning advertised window (window_clamp, rcv_ssthresh) @@ -390,12 +455,13 @@ static void tcp_sndbuf_expand(struct sock *sk) */ /* Slow part of check#2. */ -static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb) +static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb, + unsigned int skbtruesize) { struct tcp_sock *tp = tcp_sk(sk); /* Optimize this! */ - int truesize = tcp_win_from_space(sk, skb->truesize) >> 1; - int window = tcp_win_from_space(sk, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1; + int truesize = tcp_win_from_space(sk, skbtruesize) >> 1; + int window = tcp_win_from_space(sk, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])) >> 1; while (tp->rcv_ssthresh <= window) { if (truesize <= skb->len) @@ -407,46 +473,74 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb) return 0; } -static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb) +/* Even if skb appears to have a bad len/truesize ratio, TCP coalescing + * can play nice with us, as sk_buff and skb->head might be either + * freed or shared with up to MAX_SKB_FRAGS segments. + * Only give a boost to drivers using page frag(s) to hold the frame(s), + * and if no payload was pulled in skb->head before reaching us. + */ +static u32 truesize_adjust(bool adjust, const struct sk_buff *skb) +{ + u32 truesize = skb->truesize; + + if (adjust && !skb_headlen(skb)) { + truesize -= SKB_TRUESIZE(skb_end_offset(skb)); + /* paranoid check, some drivers might be buggy */ + if (unlikely((int)truesize < (int)skb->len)) + truesize = skb->truesize; + } + return truesize; +} + +static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb, + bool adjust) { struct tcp_sock *tp = tcp_sk(sk); int room; room = min_t(int, tp->window_clamp, tcp_space(sk)) - tp->rcv_ssthresh; + if (room <= 0) + return; + /* Check #1 */ - if (room > 0 && !tcp_under_memory_pressure(sk)) { + if (!tcp_under_memory_pressure(sk)) { + unsigned int truesize = truesize_adjust(adjust, skb); int incr; /* Check #2. Increase window, if skb with such overhead * will fit to rcvbuf in future. */ - if (tcp_win_from_space(sk, skb->truesize) <= skb->len) + if (tcp_win_from_space(sk, truesize) <= skb->len) incr = 2 * tp->advmss; else - incr = __tcp_grow_window(sk, skb); + incr = __tcp_grow_window(sk, skb, truesize); if (incr) { incr = max_t(int, incr, 2 * skb->len); tp->rcv_ssthresh += min(room, incr); inet_csk(sk)->icsk_ack.quick |= 1; } + } else { + /* Under pressure: + * Adjust rcv_ssthresh according to reserved mem + */ + tcp_adjust_rcv_ssthresh(sk); } } /* 3. Try to fixup all. It is made immediately after connection enters * established state. */ -void tcp_init_buffer_space(struct sock *sk) +static void tcp_init_buffer_space(struct sock *sk) { - int tcp_app_win = sock_net(sk)->ipv4.sysctl_tcp_app_win; + int tcp_app_win = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_app_win); struct tcp_sock *tp = tcp_sk(sk); int maxwin; if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK)) tcp_sndbuf_expand(sk); - tp->rcvq_space.space = min_t(u32, tp->rcv_wnd, TCP_INIT_CWND * tp->advmss); tcp_mstamp_refresh(tp); tp->rcvq_space.time = tp->tcp_mstamp; tp->rcvq_space.seq = tp->copied_seq; @@ -470,6 +564,8 @@ void tcp_init_buffer_space(struct sock *sk) tp->rcv_ssthresh = min(tp->rcv_ssthresh, tp->window_clamp); tp->snd_cwnd_stamp = tcp_jiffies32; + tp->rcvq_space.space = min3(tp->rcv_ssthresh, tp->rcv_wnd, + (u32)TCP_INIT_CWND * tp->advmss); } /* 4. Recalculate window clamp after socket hit its memory bounds. */ @@ -478,16 +574,17 @@ static void tcp_clamp_window(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); struct net *net = sock_net(sk); + int rmem2; icsk->icsk_ack.quick = 0; + rmem2 = READ_ONCE(net->ipv4.sysctl_tcp_rmem[2]); - if (sk->sk_rcvbuf < net->ipv4.sysctl_tcp_rmem[2] && + if (sk->sk_rcvbuf < rmem2 && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && !tcp_under_memory_pressure(sk) && sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) { WRITE_ONCE(sk->sk_rcvbuf, - min(atomic_read(&sk->sk_rmem_alloc), - net->ipv4.sysctl_tcp_rmem[2])); + min(atomic_read(&sk->sk_rmem_alloc), rmem2)); } if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss); @@ -517,7 +614,7 @@ EXPORT_SYMBOL(tcp_initialize_rcv_mss); * * The algorithm for RTT estimation w/o timestamps is based on * Dynamic Right-Sizing (DRS) by Wu Feng and Mike Fisk of LANL. - * <http://public.lanl.gov/radiant/pubs.html#DRS> + * <https://public.lanl.gov/radiant/pubs.html#DRS> * * More detail on this code can be found at * <http://staff.psc.edu/jheffner/>, @@ -628,7 +725,7 @@ void tcp_rcv_space_adjust(struct sock *sk) * <prev RTT . ><current RTT .. ><next RTT .... > */ - if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf && + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf) && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) { int rcvmem, rcvbuf; u64 rcvwin, grow; @@ -649,7 +746,7 @@ void tcp_rcv_space_adjust(struct sock *sk) do_div(rcvwin, tp->advmss); rcvbuf = min_t(u64, rcvwin * rcvmem, - sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])); if (rcvbuf > sk->sk_rcvbuf) { WRITE_ONCE(sk->sk_rcvbuf, rcvbuf); @@ -709,7 +806,6 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb) * restart window, so that we send ACKs quickly. */ tcp_incr_quickack(sk, TCP_MAX_QUICKACKS); - sk_mem_reclaim(sk); } } icsk->icsk_ack.lrcvtime = now; @@ -717,7 +813,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb) tcp_ecn_check_ce(sk, skb); if (skb->len >= 128) - tcp_grow_window(sk, skb); + tcp_grow_window(sk, skb, true); } /* Called to compute a smoothed rtt estimate. The data fed to this @@ -813,12 +909,12 @@ static void tcp_update_pacing_rate(struct sock *sk) * If snd_cwnd >= (tp->snd_ssthresh / 2), we are approaching * end of slow start and should slow down. */ - if (tp->snd_cwnd < tp->snd_ssthresh / 2) - rate *= sock_net(sk)->ipv4.sysctl_tcp_pacing_ss_ratio; + if (tcp_snd_cwnd(tp) < tp->snd_ssthresh / 2) + rate *= READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_pacing_ss_ratio); else - rate *= sock_net(sk)->ipv4.sysctl_tcp_pacing_ca_ratio; + rate *= READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_pacing_ca_ratio); - rate *= max(tp->snd_cwnd, tp->packets_out); + rate *= max(tcp_snd_cwnd(tp), tp->packets_out); if (likely(tp->srtt_us)) do_div(rate, tp->srtt_us); @@ -870,12 +966,63 @@ __u32 tcp_init_cwnd(const struct tcp_sock *tp, const struct dst_entry *dst) return min_t(__u32, cwnd, tp->snd_cwnd_clamp); } -/* Take a notice that peer is sending D-SACKs */ -static void tcp_dsack_seen(struct tcp_sock *tp) +struct tcp_sacktag_state { + /* Timestamps for earliest and latest never-retransmitted segment + * that was SACKed. RTO needs the earliest RTT to stay conservative, + * but congestion control should still get an accurate delay signal. + */ + u64 first_sackt; + u64 last_sackt; + u32 reord; + u32 sack_delivered; + int flag; + unsigned int mss_now; + struct rate_sample *rate; +}; + +/* Take a notice that peer is sending D-SACKs. Skip update of data delivery + * and spurious retransmission information if this DSACK is unlikely caused by + * sender's action: + * - DSACKed sequence range is larger than maximum receiver's window. + * - Total no. of DSACKed segments exceed the total no. of retransmitted segs. + */ +static u32 tcp_dsack_seen(struct tcp_sock *tp, u32 start_seq, + u32 end_seq, struct tcp_sacktag_state *state) { + u32 seq_len, dup_segs = 1; + + if (!before(start_seq, end_seq)) + return 0; + + seq_len = end_seq - start_seq; + /* Dubious DSACK: DSACKed range greater than maximum advertised rwnd */ + if (seq_len > tp->max_window) + return 0; + if (seq_len > tp->mss_cache) + dup_segs = DIV_ROUND_UP(seq_len, tp->mss_cache); + else if (tp->tlp_high_seq && tp->tlp_high_seq == end_seq) + state->flag |= FLAG_DSACK_TLP; + + tp->dsack_dups += dup_segs; + /* Skip the DSACK if dup segs weren't retransmitted by sender */ + if (tp->dsack_dups > tp->total_retrans) + return 0; + tp->rx_opt.sack_ok |= TCP_DSACK_SEEN; - tp->rack.dsack_seen = 1; - tp->dsack_dups++; + /* We increase the RACK ordering window in rounds where we receive + * DSACKs that may have been due to reordering causing RACK to trigger + * a spurious fast recovery. Thus RACK ignores DSACKs that happen + * without having seen reordering, or that match TLP probes (TLP + * is timer-driven, not triggered by RACK). + */ + if (tp->reord_seen && !(state->flag & FLAG_DSACK_TLP)) + tp->rack.dsack_seen = 1; + + state->flag |= FLAG_DSACKING_ACK; + /* A spurious retransmission is delivered */ + state->sack_delivered += dup_segs; + + return dup_segs; } /* It's reordering when higher sequence was delivered (i.e. sacked) before @@ -904,7 +1051,7 @@ static void tcp_check_sack_reordering(struct sock *sk, const u32 low_seq, tp->undo_marker ? tp->undo_retrans : 0); #endif tp->reordering = min_t(u32, (metric + mss - 1) / mss, - sock_net(sk)->ipv4.sysctl_tcp_max_reordering); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_max_reordering)); } /* This exciting event is worth to be remembered. 8) */ @@ -913,7 +1060,11 @@ static void tcp_check_sack_reordering(struct sock *sk, const u32 low_seq, ts ? LINUX_MIB_TCPTSREORDER : LINUX_MIB_TCPSACKREORDER); } -/* This must be called before lost_out is incremented */ + /* This must be called before lost_out or retrans_out are updated + * on a new loss, because we want to know if all skbs previously + * known to be lost have already been retransmitted, indicating + * that this newly lost skb is our next skb to retransmit. + */ static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb) { if ((!tp->retransmit_skb_hint && tp->retrans_out >= tp->lost_out) || @@ -923,42 +1074,46 @@ static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb) tp->retransmit_skb_hint = skb; } -/* Sum the number of packets on the wire we have marked as lost. - * There are two cases we care about here: - * a) Packet hasn't been marked lost (nor retransmitted), - * and this is the first loss. - * b) Packet has been marked both lost and retransmitted, - * and this means we think it was lost again. +/* Sum the number of packets on the wire we have marked as lost, and + * notify the congestion control module that the given skb was marked lost. */ -static void tcp_sum_lost(struct tcp_sock *tp, struct sk_buff *skb) +static void tcp_notify_skb_loss_event(struct tcp_sock *tp, const struct sk_buff *skb) { - __u8 sacked = TCP_SKB_CB(skb)->sacked; - - if (!(sacked & TCPCB_LOST) || - ((sacked & TCPCB_LOST) && (sacked & TCPCB_SACKED_RETRANS))) - tp->lost += tcp_skb_pcount(skb); + tp->lost += tcp_skb_pcount(skb); } -static void tcp_skb_mark_lost(struct tcp_sock *tp, struct sk_buff *skb) +void tcp_mark_skb_lost(struct sock *sk, struct sk_buff *skb) { - if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) { - tcp_verify_retransmit_hint(tp, skb); + __u8 sacked = TCP_SKB_CB(skb)->sacked; + struct tcp_sock *tp = tcp_sk(sk); + + if (sacked & TCPCB_SACKED_ACKED) + return; + tcp_verify_retransmit_hint(tp, skb); + if (sacked & TCPCB_LOST) { + if (sacked & TCPCB_SACKED_RETRANS) { + /* Account for retransmits that are lost again */ + TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; + tp->retrans_out -= tcp_skb_pcount(skb); + NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT, + tcp_skb_pcount(skb)); + tcp_notify_skb_loss_event(tp, skb); + } + } else { tp->lost_out += tcp_skb_pcount(skb); - tcp_sum_lost(tp, skb); TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; + tcp_notify_skb_loss_event(tp, skb); } } -void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, struct sk_buff *skb) +/* Updates the delivered and delivered_ce counts */ +static void tcp_count_delivered(struct tcp_sock *tp, u32 delivered, + bool ece_ack) { - tcp_verify_retransmit_hint(tp, skb); - - tcp_sum_lost(tp, skb); - if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) { - tp->lost_out += tcp_skb_pcount(skb); - TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; - } + tp->delivered += delivered; + if (ece_ack) + tp->delivered_ce += delivered; } /* This procedure tags the retransmission queue when SACKs arrive. @@ -1093,52 +1248,43 @@ static bool tcp_is_sackblock_valid(struct tcp_sock *tp, bool is_dsack, static bool tcp_check_dsack(struct sock *sk, const struct sk_buff *ack_skb, struct tcp_sack_block_wire *sp, int num_sacks, - u32 prior_snd_una) + u32 prior_snd_una, struct tcp_sacktag_state *state) { struct tcp_sock *tp = tcp_sk(sk); u32 start_seq_0 = get_unaligned_be32(&sp[0].start_seq); u32 end_seq_0 = get_unaligned_be32(&sp[0].end_seq); - bool dup_sack = false; + u32 dup_segs; if (before(start_seq_0, TCP_SKB_CB(ack_skb)->ack_seq)) { - dup_sack = true; - tcp_dsack_seen(tp); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDSACKRECV); } else if (num_sacks > 1) { u32 end_seq_1 = get_unaligned_be32(&sp[1].end_seq); u32 start_seq_1 = get_unaligned_be32(&sp[1].start_seq); - if (!after(end_seq_0, end_seq_1) && - !before(start_seq_0, start_seq_1)) { - dup_sack = true; - tcp_dsack_seen(tp); - NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPDSACKOFORECV); - } + if (after(end_seq_0, end_seq_1) || before(start_seq_0, start_seq_1)) + return false; + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDSACKOFORECV); + } else { + return false; + } + + dup_segs = tcp_dsack_seen(tp, start_seq_0, end_seq_0, state); + if (!dup_segs) { /* Skip dubious DSACK */ + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDSACKIGNOREDDUBIOUS); + return false; } + NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPDSACKRECVSEGS, dup_segs); + /* D-SACK for already forgotten data... Do dumb counting. */ - if (dup_sack && tp->undo_marker && tp->undo_retrans > 0 && + if (tp->undo_marker && tp->undo_retrans > 0 && !after(end_seq_0, prior_snd_una) && after(end_seq_0, tp->undo_marker)) - tp->undo_retrans--; + tp->undo_retrans = max_t(int, 0, tp->undo_retrans - dup_segs); - return dup_sack; + return true; } -struct tcp_sacktag_state { - u32 reord; - /* Timestamps for earliest and latest never-retransmitted segment - * that was SACKed. RTO needs the earliest RTT to stay conservative, - * but congestion control should still get an accurate delay signal. - */ - u64 first_sackt; - u64 last_sackt; - struct rate_sample *rate; - int flag; - unsigned int mss_now; -}; - /* Check if skb is fully within the SACK block. In presence of GSO skbs, * the incoming SACK may not exactly match but we can find smaller MSS * aligned portion of it that matches. Therefore we might need to fragment @@ -1208,7 +1354,7 @@ static u8 tcp_sacktag_one(struct sock *sk, if (dup_sack && (sacked & TCPCB_RETRANS)) { if (tp->undo_marker && tp->undo_retrans > 0 && after(end_seq, tp->undo_marker)) - tp->undo_retrans--; + tp->undo_retrans = max_t(int, 0, tp->undo_retrans - pcount); if ((sacked & TCPCB_SACKED_ACKED) && before(start_seq, state->reord)) state->reord = start_seq; @@ -1257,7 +1403,8 @@ static u8 tcp_sacktag_one(struct sock *sk, sacked |= TCPCB_SACKED_ACKED; state->flag |= FLAG_DATA_SACKED; tp->sacked_out += pcount; - tp->delivered += pcount; /* Out-of-order packets delivered */ + /* Out-of-order packets delivered */ + state->sack_delivered += pcount; /* Lost marker hint past SACKed? Tweak RFC3517 cnt */ if (tp->lost_skb_hint && @@ -1513,6 +1660,8 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb, (mss != tcp_skb_seglen(skb))) goto out; + if (!tcp_skb_can_collapse(prev, skb)) + goto out; len = skb->len; pcount = tcp_skb_pcount(skb); if (tcp_skb_shift(prev, skb, pcount, len)) @@ -1680,11 +1829,7 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, tcp_highest_sack_reset(sk); found_dup_sack = tcp_check_dsack(sk, ack_skb, sp_wire, - num_sacks, prior_snd_una); - if (found_dup_sack) { - state->flag |= FLAG_DSACKING_ACK; - tp->delivered++; /* A spurious retransmission is delivered */ - } + num_sacks, prior_snd_una, state); /* Eliminate too old ACKs, but take into * account more or less fresh ones, they can @@ -1885,14 +2030,14 @@ static void tcp_check_reno_reordering(struct sock *sk, const int addend) return; tp->reordering = min_t(u32, tp->packets_out + addend, - sock_net(sk)->ipv4.sysctl_tcp_max_reordering); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_max_reordering)); tp->reord_seen++; NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRENOREORDER); } /* Emulate SACKs for SACKless connection: account for a new dupack. */ -static void tcp_add_reno_sack(struct sock *sk, int num_dupack) +static void tcp_add_reno_sack(struct sock *sk, int num_dupack, bool ece_ack) { if (num_dupack) { struct tcp_sock *tp = tcp_sk(sk); @@ -1903,20 +2048,21 @@ static void tcp_add_reno_sack(struct sock *sk, int num_dupack) tcp_check_reno_reordering(sk, 0); delivered = tp->sacked_out - prior_sacked; if (delivered > 0) - tp->delivered += delivered; + tcp_count_delivered(tp, delivered, ece_ack); tcp_verify_left_out(tp); } } /* Account for ACK, ACKing some data in Reno Recovery phase. */ -static void tcp_remove_reno_sacks(struct sock *sk, int acked) +static void tcp_remove_reno_sacks(struct sock *sk, int acked, bool ece_ack) { struct tcp_sock *tp = tcp_sk(sk); if (acked > 0) { /* One ACK acked hole. The rest eat duplicate ACKs. */ - tp->delivered += max_t(int, acked - tp->sacked_out, 1); + tcp_count_delivered(tp, max_t(int, acked - tp->sacked_out, 1), + ece_ack); if (acked - 1 >= tp->sacked_out) tp->sacked_out = 0; else @@ -1949,7 +2095,8 @@ static inline void tcp_init_undo(struct tcp_sock *tp) static bool tcp_is_rack(const struct sock *sk) { - return sock_net(sk)->ipv4.sysctl_tcp_recovery & TCP_RACK_LOSS_DETECTION; + return READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_recovery) & + TCP_RACK_LOSS_DETECTION; } /* If we detect SACK reneging, forget all SACK information @@ -1993,6 +2140,7 @@ void tcp_enter_loss(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); bool new_recovery = icsk->icsk_ca_state < TCP_CA_Recovery; + u8 reordering; tcp_timeout_mark_lost(sk); @@ -2001,22 +2149,24 @@ void tcp_enter_loss(struct sock *sk) !after(tp->high_seq, tp->snd_una) || (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) { tp->prior_ssthresh = tcp_current_ssthresh(sk); - tp->prior_cwnd = tp->snd_cwnd; + tp->prior_cwnd = tcp_snd_cwnd(tp); tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); tcp_ca_event(sk, CA_EVENT_LOSS); tcp_init_undo(tp); } - tp->snd_cwnd = tcp_packets_in_flight(tp) + 1; + tcp_snd_cwnd_set(tp, tcp_packets_in_flight(tp) + 1); tp->snd_cwnd_cnt = 0; tp->snd_cwnd_stamp = tcp_jiffies32; /* Timeout in disordered state after receiving substantial DUPACKs * suggests that the degree of reordering is over-estimated. */ + reordering = READ_ONCE(net->ipv4.sysctl_tcp_reordering); if (icsk->icsk_ca_state <= TCP_CA_Disorder && - tp->sacked_out >= net->ipv4.sysctl_tcp_reordering) + tp->sacked_out >= reordering) tp->reordering = min_t(unsigned int, tp->reordering, - net->ipv4.sysctl_tcp_reordering); + reordering); + tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->snd_nxt; tcp_ecn_queue_cwr(tp); @@ -2025,7 +2175,7 @@ void tcp_enter_loss(struct sock *sk) * loss recovery is underway except recurring timeout(s) on * the same SND.UNA (sec 3.2). Disable F-RTO on path MTU probing */ - tp->frto = net->ipv4.sysctl_tcp_frto && + tp->frto = READ_ONCE(net->ipv4.sysctl_tcp_frto) && (new_recovery || icsk->icsk_retransmits) && !inet_csk(sk)->icsk_mtup.probe_size; } @@ -2042,7 +2192,8 @@ void tcp_enter_loss(struct sock *sk) */ static bool tcp_check_sack_reneging(struct sock *sk, int flag) { - if (flag & FLAG_SACK_RENEGING) { + if (flag & FLAG_SACK_RENEGING && + flag & FLAG_SND_UNA_ADVANCED) { struct tcp_sock *tp = tcp_sk(sk); unsigned long delay = max(usecs_to_jiffies(tp->srtt_us >> 4), msecs_to_jiffies(10)); @@ -2183,8 +2334,7 @@ static bool tcp_time_to_recover(struct sock *sk, int flag) } /* Detect loss in event "A" above by marking head of queue up as lost. - * For non-SACK(Reno) senders, the first "packets" number of segments - * are considered lost. For RFC3517 SACK, a segment is considered lost if it + * For RFC3517 SACK, a segment is considered lost if it * has at least tp->reordering SACKed seqments above it; "packets" refers to * the maximum SACKed segments to pass before reaching this limit. */ @@ -2192,10 +2342,9 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - int cnt, oldcnt, lost; - unsigned int mss; + int cnt; /* Use SACK to deduce losses of new sequences sent during recovery */ - const u32 loss_high = tcp_is_sack(tp) ? tp->snd_nxt : tp->high_seq; + const u32 loss_high = tp->snd_nxt; WARN_ON(packets > tp->packets_out); skb = tp->lost_skb_hint; @@ -2218,28 +2367,14 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) if (after(TCP_SKB_CB(skb)->end_seq, loss_high)) break; - oldcnt = cnt; - if (tcp_is_reno(tp) || - (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) + if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) cnt += tcp_skb_pcount(skb); - if (cnt > packets) { - if (tcp_is_sack(tp) || - (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) || - (oldcnt >= packets)) - break; - - mss = tcp_skb_mss(skb); - /* If needed, chop off the prefix to mark as lost. */ - lost = (packets - oldcnt) * mss; - if (lost < skb->len && - tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, - lost, mss, GFP_ATOMIC) < 0) - break; - cnt = packets; - } + if (cnt > packets) + break; - tcp_skb_mark_lost(tp, skb); + if (!(TCP_SKB_CB(skb)->sacked & TCPCB_LOST)) + tcp_mark_skb_lost(sk, skb); if (mark_head) break; @@ -2328,7 +2463,7 @@ static void DBGUNDO(struct sock *sk, const char *msg) pr_debug("Undo %s %pI4/%u c%u l%u ss%u/%u p%u\n", msg, &inet->inet_daddr, ntohs(inet->inet_dport), - tp->snd_cwnd, tcp_left_out(tp), + tcp_snd_cwnd(tp), tcp_left_out(tp), tp->snd_ssthresh, tp->prior_ssthresh, tp->packets_out); } @@ -2337,7 +2472,7 @@ static void DBGUNDO(struct sock *sk, const char *msg) pr_debug("Undo %s %pI6/%u c%u l%u ss%u/%u p%u\n", msg, &sk->sk_v6_daddr, ntohs(inet->inet_dport), - tp->snd_cwnd, tcp_left_out(tp), + tcp_snd_cwnd(tp), tcp_left_out(tp), tp->snd_ssthresh, tp->prior_ssthresh, tp->packets_out); } @@ -2362,7 +2497,7 @@ static void tcp_undo_cwnd_reduction(struct sock *sk, bool unmark_loss) if (tp->prior_ssthresh) { const struct inet_connection_sock *icsk = inet_csk(sk); - tp->snd_cwnd = icsk->icsk_ca_ops->undo_cwnd(sk); + tcp_snd_cwnd_set(tp, icsk->icsk_ca_ops->undo_cwnd(sk)); if (tp->prior_ssthresh > tp->snd_ssthresh) { tp->snd_ssthresh = tp->prior_ssthresh; @@ -2379,6 +2514,21 @@ static inline bool tcp_may_undo(const struct tcp_sock *tp) return tp->undo_marker && (!tp->undo_retrans || tcp_packet_delayed(tp)); } +static bool tcp_is_non_sack_preventing_reopen(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) { + /* Hold old state until something *above* high_seq + * is ACKed. For Reno it is MUST to prevent false + * fast retransmits (RFC2582). SACK TCP is safe. */ + if (!tcp_any_retrans_done(sk)) + tp->retrans_stamp = 0; + return true; + } + return false; +} + /* People celebrate: "We love our President!" */ static bool tcp_try_undo_recovery(struct sock *sk) { @@ -2401,14 +2551,8 @@ static bool tcp_try_undo_recovery(struct sock *sk) } else if (tp->rack.reo_wnd_persist) { tp->rack.reo_wnd_persist--; } - if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) { - /* Hold old state until something *above* high_seq - * is ACKed. For Reno it is MUST to prevent false - * fast retransmits (RFC2582). SACK TCP is safe. */ - if (!tcp_any_retrans_done(sk)) - tp->retrans_stamp = 0; + if (tcp_is_non_sack_preventing_reopen(sk)) return true; - } tcp_set_ca_state(sk, TCP_CA_Open); tp->is_sack_reneg = 0; return false; @@ -2444,6 +2588,8 @@ static bool tcp_try_undo_loss(struct sock *sk, bool frto_undo) NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSPURIOUSRTOS); inet_csk(sk)->icsk_retransmits = 0; + if (tcp_is_non_sack_preventing_reopen(sk)) + return true; if (frto_undo || tcp_is_sack(tp)) { tcp_set_ca_state(sk, TCP_CA_Open); tp->is_sack_reneg = 0; @@ -2459,7 +2605,7 @@ static bool tcp_try_undo_loss(struct sock *sk, bool frto_undo) * 1) If the packets in flight is larger than ssthresh, PRR spreads the * cwnd reductions across a full RTT. * 2) Otherwise PRR uses packet conservation to send as much as delivered. - * But when the retransmits are acked without further losses, PRR + * But when SND_UNA is acked without further losses, * slow starts cwnd up to ssthresh to speed up the recovery. */ static void tcp_init_cwnd_reduction(struct sock *sk) @@ -2469,14 +2615,14 @@ static void tcp_init_cwnd_reduction(struct sock *sk) tp->high_seq = tp->snd_nxt; tp->tlp_high_seq = 0; tp->snd_cwnd_cnt = 0; - tp->prior_cwnd = tp->snd_cwnd; + tp->prior_cwnd = tcp_snd_cwnd(tp); tp->prr_delivered = 0; tp->prr_out = 0; tp->snd_ssthresh = inet_csk(sk)->icsk_ca_ops->ssthresh(sk); tcp_ecn_queue_cwr(tp); } -void tcp_cwnd_reduction(struct sock *sk, int newly_acked_sacked, int flag) +void tcp_cwnd_reduction(struct sock *sk, int newly_acked_sacked, int newly_lost, int flag) { struct tcp_sock *tp = tcp_sk(sk); int sndcnt = 0; @@ -2490,17 +2636,16 @@ void tcp_cwnd_reduction(struct sock *sk, int newly_acked_sacked, int flag) u64 dividend = (u64)tp->snd_ssthresh * tp->prr_delivered + tp->prior_cwnd - 1; sndcnt = div_u64(dividend, tp->prior_cwnd) - tp->prr_out; - } else if ((flag & (FLAG_RETRANS_DATA_ACKED | FLAG_LOST_RETRANS)) == - FLAG_RETRANS_DATA_ACKED) { - sndcnt = min_t(int, delta, - max_t(int, tp->prr_delivered - tp->prr_out, - newly_acked_sacked) + 1); } else { - sndcnt = min(delta, newly_acked_sacked); + sndcnt = max_t(int, tp->prr_delivered - tp->prr_out, + newly_acked_sacked); + if (flag & FLAG_SND_UNA_ADVANCED && !newly_lost) + sndcnt++; + sndcnt = min(delta, sndcnt); } /* Force a fast retransmit upon entering fast recovery */ sndcnt = max(sndcnt, (tp->prr_out ? 0 : 1)); - tp->snd_cwnd = tcp_packets_in_flight(tp) + sndcnt; + tcp_snd_cwnd_set(tp, tcp_packets_in_flight(tp) + sndcnt); } static inline void tcp_end_cwnd_reduction(struct sock *sk) @@ -2513,7 +2658,7 @@ static inline void tcp_end_cwnd_reduction(struct sock *sk) /* Reset cwnd to ssthresh in CWR or Recovery (unless it's undone) */ if (tp->snd_ssthresh < TCP_INFINITE_SSTHRESH && (inet_csk(sk)->icsk_ca_state == TCP_CA_CWR || tp->undo_marker)) { - tp->snd_cwnd = tp->snd_ssthresh; + tcp_snd_cwnd_set(tp, tp->snd_ssthresh); tp->snd_cwnd_stamp = tcp_jiffies32; } tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR); @@ -2577,12 +2722,15 @@ static void tcp_mtup_probe_success(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); + u64 val; - /* FIXME: breaks with very large cwnd */ tp->prior_ssthresh = tcp_current_ssthresh(sk); - tp->snd_cwnd = tp->snd_cwnd * - tcp_mss_to_mtu(sk, tp->mss_cache) / - icsk->icsk_mtup.probe_size; + + val = (u64)tcp_snd_cwnd(tp) * tcp_mss_to_mtu(sk, tp->mss_cache); + do_div(val, icsk->icsk_mtup.probe_size); + DEBUG_NET_WARN_ON_ONCE((u32)val != val); + tcp_snd_cwnd_set(tp, max_t(u32, 1U, val)); + tp->snd_cwnd_cnt = 0; tp->snd_cwnd_stamp = tcp_jiffies32; tp->snd_ssthresh = tcp_current_ssthresh(sk); @@ -2602,17 +2750,26 @@ void tcp_simple_retransmit(struct sock *sk) const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - unsigned int mss = tcp_current_mss(sk); + int mss; + + /* A fastopen SYN request is stored as two separate packets within + * the retransmit queue, this is done by tcp_send_syn_data(). + * As a result simply checking the MSS of the frames in the queue + * will not work for the SYN packet. + * + * Us being here is an indication of a path MTU issue so we can + * assume that the fastopen SYN was lost and just mark all the + * frames in the retransmit queue as lost. We will use an MSS of + * -1 to mark all frames as lost, otherwise compute the current MSS. + */ + if (tp->syn_data && sk->sk_state == TCP_SYN_SENT) + mss = -1; + else + mss = tcp_current_mss(sk); skb_rbtree_walk(skb, &sk->tcp_rtx_queue) { - if (tcp_skb_seglen(skb) > mss && - !(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) { - if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) { - TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; - tp->retrans_out -= tcp_skb_pcount(skb); - } - tcp_skb_mark_lost_uncond_verify(tp, skb); - } + if (tcp_skb_seglen(skb) > mss) + tcp_mark_skb_lost(sk, skb); } tcp_clear_retrans_hints_partial(tp); @@ -2713,15 +2870,24 @@ static void tcp_process_loss(struct sock *sk, int flag, int num_dupack, * delivered. Lower inflight to clock out (re)tranmissions. */ if (after(tp->snd_nxt, tp->high_seq) && num_dupack) - tcp_add_reno_sack(sk, num_dupack); + tcp_add_reno_sack(sk, num_dupack, flag & FLAG_ECE); else if (flag & FLAG_SND_UNA_ADVANCED) tcp_reset_reno_sack(tp); } *rexmit = REXMIT_LOST; } +static bool tcp_force_fast_retransmit(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + return after(tcp_highest_sack_seq(tp), + tp->snd_una + tp->reordering * tp->mss_cache); +} + /* Undo during fast recovery after partial ACK. */ -static bool tcp_try_undo_partial(struct sock *sk, u32 prior_snd_una) +static bool tcp_try_undo_partial(struct sock *sk, u32 prior_snd_una, + bool *do_lost) { struct tcp_sock *tp = tcp_sk(sk); @@ -2746,7 +2912,9 @@ static bool tcp_try_undo_partial(struct sock *sk, u32 prior_snd_una) tcp_undo_cwnd_reduction(sk, true); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPPARTIALUNDO); tcp_try_keep_open(sk); - return true; + } else { + /* Partial ACK arrived. Force fast retransmit. */ + *do_lost = tcp_force_fast_retransmit(sk); } return false; } @@ -2763,20 +2931,13 @@ static void tcp_identify_packet_loss(struct sock *sk, int *ack_flag) } else if (tcp_is_rack(sk)) { u32 prior_retrans = tp->retrans_out; - tcp_rack_mark_lost(sk); + if (tcp_rack_mark_lost(sk)) + *ack_flag &= ~FLAG_SET_XMIT_TIMER; if (prior_retrans > tp->retrans_out) *ack_flag |= FLAG_LOST_RETRANS; } } -static bool tcp_force_fast_retransmit(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - return after(tcp_highest_sack_seq(tp), - tp->snd_una + tp->reordering * tp->mss_cache); -} - /* Process an event, which can update packets-in-flight not trivially. * Main goal of this function is to calculate new estimate for left_out, * taking into account both packets sitting in receiver's buffer and @@ -2795,6 +2956,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); int fast_rexmit = 0, flag = *ack_flag; + bool ece_ack = flag & FLAG_ECE; bool do_lost = num_dupack || ((flag & FLAG_DATA_SACKED) && tcp_force_fast_retransmit(sk)); @@ -2803,7 +2965,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, /* Now state machine starts. * A. ECE, hence prohibit cwnd undoing, the reduction is required. */ - if (flag & FLAG_ECE) + if (ece_ack) tp->prior_ssthresh = 0; /* B. In all the states check for reneging SACKs. */ @@ -2816,7 +2978,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, /* D. Check state exit conditions. State can be terminated * when high_seq is ACKed. */ if (icsk->icsk_ca_state == TCP_CA_Open) { - WARN_ON(tp->retrans_out != 0); + WARN_ON(tp->retrans_out != 0 && !tp->syn_data); tp->retrans_stamp = 0; } else if (!before(tp->snd_una, tp->high_seq)) { switch (icsk->icsk_ca_state) { @@ -2844,19 +3006,22 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, case TCP_CA_Recovery: if (!(flag & FLAG_SND_UNA_ADVANCED)) { if (tcp_is_reno(tp)) - tcp_add_reno_sack(sk, num_dupack); - } else { - if (tcp_try_undo_partial(sk, prior_snd_una)) - return; - /* Partial ACK arrived. Force fast retransmit. */ - do_lost = tcp_is_reno(tp) || - tcp_force_fast_retransmit(sk); - } - if (tcp_try_undo_dsack(sk)) { - tcp_try_keep_open(sk); + tcp_add_reno_sack(sk, num_dupack, ece_ack); + } else if (tcp_try_undo_partial(sk, prior_snd_una, &do_lost)) return; - } + + if (tcp_try_undo_dsack(sk)) + tcp_try_keep_open(sk); + tcp_identify_packet_loss(sk, ack_flag); + if (icsk->icsk_ca_state != TCP_CA_Recovery) { + if (!tcp_time_to_recover(sk, flag)) + return; + /* Undo reverts the recovery state. If loss is evident, + * starts a new recovery (e.g. reordering then loss); + */ + tcp_enter_recovery(sk, ece_ack); + } break; case TCP_CA_Loss: tcp_process_loss(sk, flag, num_dupack, rexmit); @@ -2865,12 +3030,12 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, (*ack_flag & FLAG_LOST_RETRANS))) return; /* Change state if cwnd is undone or retransmits are lost */ - /* fall through */ + fallthrough; default: if (tcp_is_reno(tp)) { if (flag & FLAG_SND_UNA_ADVANCED) tcp_reset_reno_sack(tp); - tcp_add_reno_sack(sk, num_dupack); + tcp_add_reno_sack(sk, num_dupack, ece_ack); } if (icsk->icsk_ca_state <= TCP_CA_Disorder) @@ -2888,13 +3053,13 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, tp->snd_una == tp->mtu_probe.probe_seq_start) { tcp_mtup_probe_failed(sk); /* Restores the reduction we did in tcp_mtup_probe() */ - tp->snd_cwnd++; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); tcp_simple_retransmit(sk); return; } /* Otherwise enter Recovery state */ - tcp_enter_recovery(sk, (flag & FLAG_ECE)); + tcp_enter_recovery(sk, ece_ack); fast_rexmit = 1; } @@ -2905,7 +3070,7 @@ static void tcp_fastretrans_alert(struct sock *sk, const u32 prior_snd_una, static void tcp_update_rtt_min(struct sock *sk, u32 rtt_us, const int flag) { - u32 wlen = sock_net(sk)->ipv4.sysctl_tcp_min_rtt_wlen * HZ; + u32 wlen = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_min_rtt_wlen) * HZ; struct tcp_sock *tp = tcp_sk(sk); if ((flag & FLAG_ACK_MAYBE_DELAYED) && rtt_us > tcp_min_rtt(tp)) { @@ -2944,6 +3109,8 @@ static bool tcp_ack_update_rtt(struct sock *sk, const int flag, u32 delta = tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr; if (likely(delta < INT_MAX / (USEC_PER_SEC / TCP_TS_HZ))) { + if (!delta) + delta = 1; seq_rtt_us = delta * (USEC_PER_SEC / TCP_TS_HZ); ca_rtt_us = seq_rtt_us; } @@ -3014,7 +3181,7 @@ void tcp_rearm_rto(struct sock *sk) rto = usecs_to_jiffies(max_t(int, delta_us, 1)); } tcp_reset_xmit_timer(sk, ICSK_TIME_RETRANS, rto, - TCP_RTO_MAX, tcp_rtx_queue_head(sk)); + TCP_RTO_MAX); } } @@ -3047,7 +3214,7 @@ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb) } static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, - u32 prior_snd_una) + const struct sk_buff *ack_skb, u32 prior_snd_una) { const struct skb_shared_info *shinfo; @@ -3059,7 +3226,7 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, if (!before(shinfo->tskey, prior_snd_una) && before(shinfo->tskey, tcp_sk(sk)->snd_una)) { tcp_skb_tsorted_save(skb) { - __skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK); + __skb_tstamp_tx(skb, ack_skb, NULL, sk, SCM_TSTAMP_ACK); } tcp_skb_tsorted_restore(skb); } } @@ -3068,9 +3235,9 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, * is before the ack sequence we can discard it as it's confirmed to have * arrived at the other end. */ -static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, - u32 prior_snd_una, - struct tcp_sacktag_state *sack) +static int tcp_clean_rtx_queue(struct sock *sk, const struct sk_buff *ack_skb, + u32 prior_fack, u32 prior_snd_una, + struct tcp_sacktag_state *sack, bool ece_ack) { const struct inet_connection_sock *icsk = inet_csk(sk); u64 first_ackt, last_ackt; @@ -3083,7 +3250,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, long seq_rtt_us = -1L; long ca_rtt_us = -1L; u32 pkts_acked = 0; - u32 last_in_flight = 0; bool rtt_update; int flag = 0; @@ -3095,8 +3261,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, u8 sacked = scb->sacked; u32 acked_pcount; - tcp_ack_tstamp(sk, skb, prior_snd_una); - /* Determine how many packets and what bytes were acked, tso and else */ if (after(scb->end_seq, tp->snd_una)) { if (tcp_skb_pcount(skb) == 1 || @@ -3121,7 +3285,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, if (!first_ackt) first_ackt = last_ackt; - last_in_flight = TCP_SKB_CB(skb)->tx.in_flight; if (before(start_seq, reord)) reord = start_seq; if (!after(scb->end_seq, tp->high_seq)) @@ -3131,7 +3294,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, if (sacked & TCPCB_SACKED_ACKED) { tp->sacked_out -= acked_pcount; } else if (tcp_is_sack(tp)) { - tp->delivered += acked_pcount; + tcp_count_delivered(tp, acked_pcount, ece_ack); if (!tcp_skb_spurious_retrans(tp, skb)) tcp_rack_advance(tp, sacked, scb->end_seq, tcp_skb_timestamp_us(skb)); @@ -3160,6 +3323,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, if (!fully_acked) break; + tcp_ack_tstamp(sk, skb, ack_skb, prior_snd_una); + next = skb_rb_next(skb); if (unlikely(skb == tp->retransmit_skb_hint)) tp->retransmit_skb_hint = NULL; @@ -3175,15 +3340,18 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, if (likely(between(tp->snd_up, prior_snd_una, tp->snd_una))) tp->snd_up = tp->snd_una; - if (skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) - flag |= FLAG_SACK_RENEGING; + if (skb) { + tcp_ack_tstamp(sk, skb, ack_skb, prior_snd_una); + if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) + flag |= FLAG_SACK_RENEGING; + } if (likely(first_ackt) && !(flag & FLAG_RETRANS_DATA_ACKED)) { seq_rtt_us = tcp_stamp_us_delta(tp->tcp_mstamp, first_ackt); ca_rtt_us = tcp_stamp_us_delta(tp->tcp_mstamp, last_ackt); - if (pkts_acked == 1 && last_in_flight < tp->mss_cache && - last_in_flight && !prior_sacked && fully_acked && + if (pkts_acked == 1 && fully_acked && !prior_sacked && + (tp->snd_una - prior_snd_una) < tp->mss_cache && sack->rate->prior_delivered + 1 == tp->delivered && !(flag & (FLAG_CA_ALERT | FLAG_SYN_ACKED))) { /* Conservatively mark a delayed ACK. It's typically @@ -3208,7 +3376,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, } if (tcp_is_reno(tp)) { - tcp_remove_reno_sacks(sk, pkts_acked); + tcp_remove_reno_sacks(sk, pkts_acked, ece_ack); /* If any of the cumulatively ACKed segments was * retransmitted, non-SACK case cannot confirm that @@ -3240,9 +3408,10 @@ static int tcp_clean_rtx_queue(struct sock *sk, u32 prior_fack, if (icsk->icsk_ca_ops->pkts_acked) { struct ack_sample sample = { .pkts_acked = pkts_acked, - .rtt_us = sack->rate->rtt_us, - .in_flight = last_in_flight }; + .rtt_us = sack->rate->rtt_us }; + sample.in_flight = tp->mss_cache * + (tp->delivered - sack->rate->prior_delivered); icsk->icsk_ca_ops->pkts_acked(sk, &sample); } @@ -3283,6 +3452,7 @@ static void tcp_ack_probe(struct sock *sk) return; if (!after(TCP_SKB_CB(head)->end_seq, tcp_wnd_end(tp))) { icsk->icsk_backoff = 0; + icsk->icsk_probes_tstamp = 0; inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0); /* Socket must be waked up by subsequent tcp_data_snd_check(). * This function is not for random using! @@ -3290,8 +3460,8 @@ static void tcp_ack_probe(struct sock *sk) } else { unsigned long when = tcp_probe0_when(sk, TCP_RTO_MAX); - tcp_reset_xmit_timer(sk, ICSK_TIME_PROBE0, - when, TCP_RTO_MAX, NULL); + when = tcp_clamp_probe0_to_user_timeout(sk, when); + tcp_reset_xmit_timer(sk, ICSK_TIME_PROBE0, when, TCP_RTO_MAX); } } @@ -3310,7 +3480,8 @@ static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag) * new SACK or ECE mark may first advance cwnd here and later reduce * cwnd in tcp_fastretrans_alert() based on more states. */ - if (tcp_sk(sk)->reordering > sock_net(sk)->ipv4.sysctl_tcp_reordering) + if (tcp_sk(sk)->reordering > + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reordering)) return flag & FLAG_FORWARD_PROGRESS; return flag & FLAG_DATA_ACKED; @@ -3333,7 +3504,7 @@ static void tcp_cong_control(struct sock *sk, u32 ack, u32 acked_sacked, if (tcp_in_cwnd_reduction(sk)) { /* Reduce cwnd if state mandates */ - tcp_cwnd_reduction(sk, acked_sacked, flag); + tcp_cwnd_reduction(sk, acked_sacked, rs->losses, flag); } else if (tcp_may_raise_cwnd(sk, flag)) { /* Advance cwnd if state allows */ tcp_cong_avoid(sk, ack, acked_sacked); @@ -3422,7 +3593,8 @@ static bool __tcp_oow_rate_limited(struct net *net, int mib_idx, if (*last_oow_ack_time) { s32 elapsed = (s32)(tcp_jiffies32 - *last_oow_ack_time); - if (0 <= elapsed && elapsed < net->ipv4.sysctl_tcp_invalid_ratelimit) { + if (0 <= elapsed && + elapsed < READ_ONCE(net->ipv4.sysctl_tcp_invalid_ratelimit)) { NET_INC_STATS(net, mib_idx); return true; /* rate-limited: don't send yet! */ } @@ -3452,14 +3624,11 @@ bool tcp_oow_rate_limited(struct net *net, const struct sk_buff *skb, } /* RFC 5961 7 [ACK Throttling] */ -static void tcp_send_challenge_ack(struct sock *sk, const struct sk_buff *skb) +static void tcp_send_challenge_ack(struct sock *sk) { - /* unprotected vars, we dont care of overwrites */ - static u32 challenge_timestamp; - static unsigned int challenge_count; struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); - u32 count, now; + u32 count, now, ack_limit; /* First check our per-socket dupack rate limit. */ if (__tcp_oow_rate_limited(net, @@ -3467,18 +3636,22 @@ static void tcp_send_challenge_ack(struct sock *sk, const struct sk_buff *skb) &tp->last_oow_ack_time)) return; + ack_limit = READ_ONCE(net->ipv4.sysctl_tcp_challenge_ack_limit); + if (ack_limit == INT_MAX) + goto send_ack; + /* Then check host-wide RFC 5961 rate limit. */ now = jiffies / HZ; - if (now != challenge_timestamp) { - u32 ack_limit = net->ipv4.sysctl_tcp_challenge_ack_limit; + if (now != READ_ONCE(net->ipv4.tcp_challenge_timestamp)) { u32 half = (ack_limit + 1) >> 1; - challenge_timestamp = now; - WRITE_ONCE(challenge_count, half + prandom_u32_max(ack_limit)); + WRITE_ONCE(net->ipv4.tcp_challenge_timestamp, now); + WRITE_ONCE(net->ipv4.tcp_challenge_count, half + prandom_u32_max(ack_limit)); } - count = READ_ONCE(challenge_count); + count = READ_ONCE(net->ipv4.tcp_challenge_count); if (count > 0) { - WRITE_ONCE(challenge_count, count - 1); + WRITE_ONCE(net->ipv4.tcp_challenge_count, count - 1); +send_ack: NET_INC_STATS(net, LINUX_MIB_TCPCHALLENGEACK); tcp_send_ack(sk); } @@ -3505,10 +3678,8 @@ static void tcp_replace_ts_recent(struct tcp_sock *tp, u32 seq) } } -/* This routine deals with acks during a TLP episode. - * We mark the end of a TLP episode on receiving TLP dupack or when - * ack is after tlp_high_seq. - * Ref: loss detection algorithm in draft-dukkipati-tcpm-tcp-loss-probe. +/* This routine deals with acks during a TLP episode and ends an episode by + * resetting tlp_high_seq. Ref: TLP algorithm in draft-ietf-tcpm-rack */ static void tcp_process_tlp_ack(struct sock *sk, u32 ack, int flag) { @@ -3517,7 +3688,10 @@ static void tcp_process_tlp_ack(struct sock *sk, u32 ack, int flag) if (before(ack, tp->tlp_high_seq)) return; - if (flag & FLAG_DSACKING_ACK) { + if (!tp->tlp_retrans) { + /* TLP of new data has been acknowledged */ + tp->tlp_high_seq = 0; + } else if (flag & FLAG_DSACK_TLP) { /* This DSACK means original and TLP probe arrived; no loss */ tp->tlp_high_seq = 0; } else if (after(ack, tp->tlp_high_seq)) { @@ -3575,10 +3749,9 @@ static u32 tcp_newly_delivered(struct sock *sk, u32 prior_delivered, int flag) delivered = tp->delivered - prior_delivered; NET_ADD_STATS(net, LINUX_MIB_TCPDELIVERED, delivered); - if (flag & FLAG_ECE) { - tp->delivered_ce += delivered; + if (flag & FLAG_ECE) NET_ADD_STATS(net, LINUX_MIB_TCPDELIVEREDCE, delivered); - } + return delivered; } @@ -3602,6 +3775,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) sack_state.first_sackt = 0; sack_state.rate = &rs; + sack_state.sack_delivered = 0; /* We very likely will need to access rtx queue. */ prefetch(sk->tcp_rtx_queue.rb_node); @@ -3613,8 +3787,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) /* RFC 5961 5.2 [Blind Data Injection Attack].[Mitigation] */ if (before(ack, prior_snd_una - tp->max_window)) { if (!(flag & FLAG_NO_CHALLENGE_ACK)) - tcp_send_challenge_ack(sk, skb); - return -1; + tcp_send_challenge_ack(sk); + return -SKB_DROP_REASON_TCP_TOO_OLD_ACK; } goto old_ack; } @@ -3623,7 +3797,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) * this segment (RFC793 Section 3.9). */ if (after(ack, tp->snd_nxt)) - return -1; + return -SKB_DROP_REASON_TCP_ACK_UNSENT_DATA; if (after(ack, prior_snd_una)) { flag |= FLAG_SND_UNA_ADVANCED; @@ -3677,12 +3851,25 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) ack_ev_flags |= CA_ACK_ECE; } + if (sack_state.sack_delivered) + tcp_count_delivered(tp, sack_state.sack_delivered, + flag & FLAG_ECE); + if (flag & FLAG_WIN_UPDATE) ack_ev_flags |= CA_ACK_WIN_UPDATE; tcp_in_ack_event(sk, ack_ev_flags); } + /* This is a deviation from RFC3168 since it states that: + * "When the TCP data sender is ready to set the CWR bit after reducing + * the congestion window, it SHOULD set the CWR bit only on the first + * new data packet that it transmits." + * We accept CWR on pure ACKs to be more robust + * with widely-deployed TCP implementations that do this. + */ + tcp_ecn_accept_cwr(sk, skb); + /* We passed data and got it acked, remove any soft error * log. Something worked... */ @@ -3693,18 +3880,17 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) goto no_queue; /* See if we can take anything off of the retransmit queue. */ - flag |= tcp_clean_rtx_queue(sk, prior_fack, prior_snd_una, &sack_state); + flag |= tcp_clean_rtx_queue(sk, skb, prior_fack, prior_snd_una, + &sack_state, flag & FLAG_ECE); tcp_rack_update_reo_wnd(sk, &rs); if (tp->tlp_high_seq) tcp_process_tlp_ack(sk, ack, flag); - /* If needed, reset TLP/RTO timer; RACK may later override this. */ - if (flag & FLAG_SET_XMIT_TIMER) - tcp_set_xmit_timer(sk); if (tcp_ack_is_dubious(sk, flag)) { - if (!(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP))) { + if (!(flag & (FLAG_SND_UNA_ADVANCED | + FLAG_NOT_DUP | FLAG_DSACKING_ACK))) { num_dupack = 1; /* Consider if pure acks were aggregated in tcp_add_backlog() */ if (!(flag & FLAG_DATA)) @@ -3714,6 +3900,10 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) &rexmit); } + /* If needed, reset TLP/RTO timer when RACK doesn't set. */ + if (flag & FLAG_SET_XMIT_TIMER) + tcp_set_xmit_timer(sk); + if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) sk_dst_confirm(sk); @@ -3775,7 +3965,7 @@ static void tcp_parse_fastopen_option(int len, const unsigned char *cookie, foc->exp = exp_opt; } -static void smc_parse_options(const struct tcphdr *th, +static bool smc_parse_options(const struct tcphdr *th, struct tcp_options_received *opt_rx, const unsigned char *ptr, int opsize) @@ -3784,16 +3974,19 @@ static void smc_parse_options(const struct tcphdr *th, if (static_branch_unlikely(&tcp_have_smc)) { if (th->syn && !(opsize & 1) && opsize >= TCPOLEN_EXP_SMC_BASE && - get_unaligned_be32(ptr) == TCPOPT_SMC_MAGIC) + get_unaligned_be32(ptr) == TCPOPT_SMC_MAGIC) { opt_rx->smc_ok = 1; + return true; + } } #endif + return false; } /* Try to parse the MSS option from the TCP header. Return 0 on failure, clamped * value on success. */ -static u16 tcp_parse_mss_option(const struct tcphdr *th, u16 user_mss) +u16 tcp_parse_mss_option(const struct tcphdr *th, u16 user_mss) { const unsigned char *ptr = (const unsigned char *)(th + 1); int length = (th->doff * 4) - sizeof(struct tcphdr); @@ -3832,6 +4025,7 @@ static u16 tcp_parse_mss_option(const struct tcphdr *th, u16 user_mss) } return mss; } +EXPORT_SYMBOL_GPL(tcp_parse_mss_option); /* Look for tcp options. Normally only called on SYN and SYNACK packets. * But, this can also be called on packets in the established flow when @@ -3848,6 +4042,7 @@ void tcp_parse_options(const struct net *net, ptr = (const unsigned char *)(th + 1); opt_rx->saw_tstamp = 0; + opt_rx->saw_unknown = 0; while (length > 0) { int opcode = *ptr++; @@ -3881,7 +4076,7 @@ void tcp_parse_options(const struct net *net, break; case TCPOPT_WINDOW: if (opsize == TCPOLEN_WINDOW && th->syn && - !estab && net->ipv4.sysctl_tcp_window_scaling) { + !estab && READ_ONCE(net->ipv4.sysctl_tcp_window_scaling)) { __u8 snd_wscale = *(__u8 *)ptr; opt_rx->wscale_ok = 1; if (snd_wscale > TCP_MAX_WSCALE) { @@ -3897,7 +4092,7 @@ void tcp_parse_options(const struct net *net, case TCPOPT_TIMESTAMP: if ((opsize == TCPOLEN_TIMESTAMP) && ((estab && opt_rx->tstamp_ok) || - (!estab && net->ipv4.sysctl_tcp_timestamps))) { + (!estab && READ_ONCE(net->ipv4.sysctl_tcp_timestamps)))) { opt_rx->saw_tstamp = 1; opt_rx->rcv_tsval = get_unaligned_be32(ptr); opt_rx->rcv_tsecr = get_unaligned_be32(ptr + 4); @@ -3905,7 +4100,7 @@ void tcp_parse_options(const struct net *net, break; case TCPOPT_SACK_PERM: if (opsize == TCPOLEN_SACK_PERM && th->syn && - !estab && net->ipv4.sysctl_tcp_sack) { + !estab && READ_ONCE(net->ipv4.sysctl_tcp_sack)) { opt_rx->sack_ok = TCP_SACK_SEEN; tcp_sack_reset(opt_rx); } @@ -3926,10 +4121,6 @@ void tcp_parse_options(const struct net *net, */ break; #endif - case TCPOPT_MPTCP: - mptcp_parse_option(skb, ptr, opsize, opt_rx); - break; - case TCPOPT_FASTOPEN: tcp_parse_fastopen_option( opsize - TCPOLEN_FASTOPEN_BASE, @@ -3942,15 +4133,21 @@ void tcp_parse_options(const struct net *net, */ if (opsize >= TCPOLEN_EXP_FASTOPEN_BASE && get_unaligned_be16(ptr) == - TCPOPT_FASTOPEN_MAGIC) + TCPOPT_FASTOPEN_MAGIC) { tcp_parse_fastopen_option(opsize - TCPOLEN_EXP_FASTOPEN_BASE, ptr + 2, th->syn, foc, true); - else - smc_parse_options(th, opt_rx, ptr, - opsize); + break; + } + + if (smc_parse_options(th, opt_rx, ptr, opsize)) + break; + + opt_rx->saw_unknown = 1; break; + default: + opt_rx->saw_unknown = 1; } ptr += opsize-2; length -= opsize; @@ -4111,10 +4308,16 @@ static inline bool tcp_sequence(const struct tcp_sock *tp, u32 seq, u32 end_seq) } /* When we get a reset we do this. */ -void tcp_reset(struct sock *sk) +void tcp_reset(struct sock *sk, struct sk_buff *skb) { trace_tcp_receive_reset(sk); + /* mptcp can't tell us to ignore reset pkts, + * so just ignore the return value of mptcp_incoming_options(). + */ + if (sk_is_mptcp(sk)) + mptcp_incoming_options(sk, skb); + /* We want the right error as BSD sees it (and indeed as we do). */ switch (sk->sk_state) { case TCP_SYN_SENT: @@ -4135,7 +4338,7 @@ void tcp_reset(struct sock *sk) tcp_done(sk); if (!sock_flag(sk, SOCK_DEAD)) - sk->sk_error_report(sk); + sk_error_report(sk); } /* @@ -4207,7 +4410,6 @@ void tcp_fin(struct sock *sk) skb_rbtree_purge(&tp->out_of_order_queue); if (tcp_is_sack(tp)) tcp_sack_reset(&tp->rx_opt); - sk_mem_reclaim(sk); if (!sock_flag(sk, SOCK_DEAD)) { sk->sk_state_change(sk); @@ -4238,7 +4440,7 @@ static void tcp_dsack_set(struct sock *sk, u32 seq, u32 end_seq) { struct tcp_sock *tp = tcp_sk(sk); - if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_dsack) { + if (tcp_is_sack(tp) && READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_dsack)) { int mib_idx; if (before(seq, tp->rcv_nxt)) @@ -4271,10 +4473,9 @@ static void tcp_rcv_spurious_retrans(struct sock *sk, const struct sk_buff *skb) * The receiver remembers and reflects via DSACKs. Leverage the * DSACK state and change the txhash to re-route speculatively. */ - if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq) { - sk_rethink_txhash(sk); + if (TCP_SKB_CB(skb)->seq == tcp_sk(sk)->duplicate_sack[0].start_seq && + sk_rethink_txhash(sk)) NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDUPLICATEDATAREHASH); - } } static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb) @@ -4286,7 +4487,7 @@ static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb) NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOST); tcp_enter_quickack_mode(sk, TCP_MAX_QUICKACKS); - if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_dsack) { + if (tcp_is_sack(tp) && READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_dsack)) { u32 end_seq = TCP_SKB_CB(skb)->end_seq; tcp_rcv_spurious_retrans(sk, skb); @@ -4323,10 +4524,38 @@ static void tcp_sack_maybe_coalesce(struct tcp_sock *tp) sp[i] = sp[i + 1]; continue; } - this_sack++, swalk++; + this_sack++; + swalk++; } } +static void tcp_sack_compress_send_ack(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (!tp->compressed_ack) + return; + + if (hrtimer_try_to_cancel(&tp->compressed_ack_timer) == 1) + __sock_put(sk); + + /* Since we have to send one ack finally, + * substract one from tp->compressed_ack to keep + * LINUX_MIB_TCPACKCOMPRESSED accurate. + */ + NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPACKCOMPRESSED, + tp->compressed_ack - 1); + + tp->compressed_ack = 0; + tcp_send_ack(sk); +} + +/* Reasonable amount of sack blocks included in TCP SACK option + * The max is 4, but this becomes 3 if TCP timestamps are there. + * Given that SACK packets might be lost, be conservative and use 2. + */ +#define TCP_SACK_BLOCKS_EXPECTED 2 + static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) { struct tcp_sock *tp = tcp_sk(sk); @@ -4339,6 +4568,8 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) for (this_sack = 0; this_sack < cur_sacks; this_sack++, sp++) { if (tcp_sack_extend(sp, seq, end_seq)) { + if (this_sack >= TCP_SACK_BLOCKS_EXPECTED) + tcp_sack_compress_send_ack(sk); /* Rotate this_sack to the first one. */ for (; this_sack > 0; this_sack--, sp--) swap(*sp, *(sp - 1)); @@ -4348,6 +4579,9 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) } } + if (this_sack >= TCP_SACK_BLOCKS_EXPECTED) + tcp_sack_compress_send_ack(sk); + /* Could not find an adjacent existing SACK, build a new one, * put it at the front, and shift everyone else down. We * always know there is at least one SACK present already here. @@ -4355,8 +4589,6 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq) * If the sack array is full, forget about the last one. */ if (this_sack >= TCP_NUM_SACKS) { - if (tp->compressed_ack > TCP_FASTRETRANS_THRESH) - tcp_send_ack(sk); this_sack--; tp->rx_opt.num_sacks--; sp--; @@ -4408,7 +4640,6 @@ static void tcp_sack_remove(struct tcp_sock *tp) /** * tcp_try_coalesce - try to merge skb to prior one * @sk: socket - * @dest: destination queue * @to: prior buffer * @from: buffer to add in queue * @fragstolen: pointer to boolean @@ -4466,7 +4697,7 @@ static bool tcp_ooo_try_coalesce(struct sock *sk, { bool res = tcp_try_coalesce(sk, to, from, fragstolen); - /* In case tcp_drop() is called later, update to->gso_segs */ + /* In case tcp_drop_reason() is called later, update to->gso_segs */ if (res) { u32 gso_segs = max_t(u16, 1, skb_shinfo(to)->gso_segs) + max_t(u16, 1, skb_shinfo(from)->gso_segs); @@ -4476,10 +4707,11 @@ static bool tcp_ooo_try_coalesce(struct sock *sk, return res; } -static void tcp_drop(struct sock *sk, struct sk_buff *skb) +static void tcp_drop_reason(struct sock *sk, struct sk_buff *skb, + enum skb_drop_reason reason) { sk_drops_add(sk, skb); - __kfree_skb(skb); + kfree_skb_reason(skb, reason); } /* This one checks to see if we can put data from the @@ -4509,7 +4741,7 @@ static void tcp_ofo_queue(struct sock *sk) rb_erase(&skb->rbnode, &tp->out_of_order_queue); if (unlikely(!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt))) { - tcp_drop(sk, skb); + tcp_drop_reason(sk, skb, SKB_DROP_REASON_TCP_OFO_DROP); continue; } @@ -4564,7 +4796,8 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) { NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFODROP); - tcp_drop(sk, skb); + sk->sk_data_ready(sk); + tcp_drop_reason(sk, skb, SKB_DROP_REASON_PROTO_MEM); return; } @@ -4597,7 +4830,11 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) if (tcp_ooo_try_coalesce(sk, tp->ooo_last_skb, skb, &fragstolen)) { coalesce_done: - tcp_grow_window(sk, skb); + /* For non sack flows, do not grow window to force DUPACK + * and trigger fast retransmit. + */ + if (tcp_is_sack(tp)) + tcp_grow_window(sk, skb, true); kfree_skb_partial(skb, fragstolen); skb = NULL; goto add_sack; @@ -4623,7 +4860,8 @@ coalesce_done: /* All the bits are present. Drop. */ NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFOMERGE); - tcp_drop(sk, skb); + tcp_drop_reason(sk, skb, + SKB_DROP_REASON_TCP_OFOMERGE); skb = NULL; tcp_dsack_set(sk, seq, end_seq); goto add_sack; @@ -4642,7 +4880,8 @@ coalesce_done: TCP_SKB_CB(skb1)->end_seq); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFOMERGE); - tcp_drop(sk, skb1); + tcp_drop_reason(sk, skb1, + SKB_DROP_REASON_TCP_OFOMERGE); goto merge_right; } } else if (tcp_ooo_try_coalesce(sk, skb1, @@ -4670,7 +4909,7 @@ merge_right: tcp_dsack_extend(sk, TCP_SKB_CB(skb1)->seq, TCP_SKB_CB(skb1)->end_seq); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPOFOMERGE); - tcp_drop(sk, skb1); + tcp_drop_reason(sk, skb1, SKB_DROP_REASON_TCP_OFOMERGE); } /* If there is no skb after us, we are the last_skb ! */ if (!skb1) @@ -4681,7 +4920,11 @@ add_sack: tcp_sack_new_ofo_skb(sk, seq, end_seq); end: if (skb) { - tcp_grow_window(sk, skb); + /* For non sack flows, do not grow window to force DUPACK + * and trigger fast retransmit. + */ + if (tcp_is_sack(tp)) + tcp_grow_window(sk, skb, false); skb_condense(skb); skb_set_owner_r(skb, sk); } @@ -4758,23 +5001,24 @@ err: void tcp_data_ready(struct sock *sk) { - const struct tcp_sock *tp = tcp_sk(sk); - int avail = tp->rcv_nxt - tp->copied_seq; - - if (avail < sk->sk_rcvlowat && !sock_flag(sk, SOCK_DONE)) - return; - - sk->sk_data_ready(sk); + if (tcp_epollin_ready(sk, sk->sk_rcvlowat) || sock_flag(sk, SOCK_DONE)) + sk->sk_data_ready(sk); } static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); + enum skb_drop_reason reason; bool fragstolen; int eaten; - if (sk_is_mptcp(sk)) - mptcp_incoming_options(sk, skb, &tp->rx_opt); + /* If a subflow has been reset, the packet should not continue + * to be processed, drop the packet. + */ + if (sk_is_mptcp(sk) && !mptcp_incoming_options(sk, skb)) { + __kfree_skb(skb); + return; + } if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) { __kfree_skb(skb); @@ -4783,8 +5027,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) skb_dst_drop(skb); __skb_pull(skb, tcp_hdr(skb)->doff * 4); - tcp_ecn_accept_cwr(sk, skb); - + reason = SKB_DROP_REASON_NOT_SPECIFIED; tp->rx_opt.dsack = 0; /* Queue data for delivery to the user. @@ -4793,6 +5036,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) */ if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) { if (tcp_receive_window(tp) == 0) { + reason = SKB_DROP_REASON_TCP_ZEROWINDOW; NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPZEROWINDOWDROP); goto out_of_window; } @@ -4802,7 +5046,9 @@ queue_and_out: if (skb_queue_len(&sk->sk_receive_queue) == 0) sk_forced_mem_schedule(sk, skb->truesize); else if (tcp_try_rmem_schedule(sk, skb, skb->truesize)) { + reason = SKB_DROP_REASON_PROTO_MEM; NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRCVQDROP); + sk->sk_data_ready(sk); goto drop; } @@ -4837,6 +5083,7 @@ queue_and_out: if (!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) { tcp_rcv_spurious_retrans(sk, skb); /* A retransmit, 2nd most common case. Force an immediate ack. */ + reason = SKB_DROP_REASON_TCP_OLD_DATA; NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOST); tcp_dsack_set(sk, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq); @@ -4844,13 +5091,16 @@ out_of_window: tcp_enter_quickack_mode(sk, TCP_MAX_QUICKACKS); inet_csk_schedule_ack(sk); drop: - tcp_drop(sk, skb); + tcp_drop_reason(sk, skb, reason); return; } /* Out of window. F.e. zero window probe. */ - if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt + tcp_receive_window(tp))) + if (!before(TCP_SKB_CB(skb)->seq, + tp->rcv_nxt + tcp_receive_window(tp))) { + reason = SKB_DROP_REASON_TCP_OVERWINDOW; goto out_of_window; + } if (before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) { /* Partial packet, seq < rcv_next < end_seq */ @@ -4860,6 +5110,7 @@ drop: * remembering D-SACK for its head made in previous line. */ if (!tcp_receive_window(tp)) { + reason = SKB_DROP_REASON_TCP_ZEROWINDOW; NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPZEROWINDOWDROP); goto out_of_window; } @@ -5055,7 +5306,7 @@ new_range: before(TCP_SKB_CB(skb)->end_seq, start)) { /* Do not attempt collapsing tiny skbs */ if (range_truesize != head->truesize || - end - start >= SKB_WITH_OVERHEAD(SK_MEM_QUANTUM)) { + end - start >= SKB_WITH_OVERHEAD(PAGE_SIZE)) { tcp_collapse(sk, NULL, &tp->out_of_order_queue, head, skb, start, end); } else { @@ -5101,9 +5352,9 @@ static bool tcp_prune_ofo_queue(struct sock *sk) prev = rb_prev(node); rb_erase(node, &tp->out_of_order_queue); goal -= rb_to_skb(node)->truesize; - tcp_drop(sk, rb_to_skb(node)); + tcp_drop_reason(sk, rb_to_skb(node), + SKB_DROP_REASON_TCP_OFO_QUEUE_PRUNE); if (!prev || goal <= 0) { - sk_mem_reclaim(sk); if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf && !tcp_under_memory_pressure(sk)) break; @@ -5139,7 +5390,7 @@ static int tcp_prune_queue(struct sock *sk) if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf) tcp_clamp_window(sk); else if (tcp_under_memory_pressure(sk)) - tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U * tp->advmss); + tcp_adjust_rcv_ssthresh(sk); if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf) return 0; @@ -5150,7 +5401,6 @@ static int tcp_prune_queue(struct sock *sk) skb_peek(&sk->sk_receive_queue), NULL, tp->copied_seq, tp->rcv_nxt); - sk_mem_reclaim(sk); if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf) return 0; @@ -5174,7 +5424,7 @@ static int tcp_prune_queue(struct sock *sk) return -1; } -static bool tcp_should_expand_sndbuf(const struct sock *sk) +static bool tcp_should_expand_sndbuf(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); @@ -5185,26 +5435,30 @@ static bool tcp_should_expand_sndbuf(const struct sock *sk) return false; /* If we are under global TCP memory pressure, do not expand. */ - if (tcp_under_memory_pressure(sk)) + if (tcp_under_memory_pressure(sk)) { + int unused_mem = sk_unused_reserved_mem(sk); + + /* Adjust sndbuf according to reserved mem. But make sure + * it never goes below SOCK_MIN_SNDBUF. + * See sk_stream_moderate_sndbuf() for more details. + */ + if (unused_mem > SOCK_MIN_SNDBUF) + WRITE_ONCE(sk->sk_sndbuf, unused_mem); + return false; + } /* If we are under soft global TCP memory pressure, do not expand. */ if (sk_memory_allocated(sk) >= sk_prot_mem_limits(sk, 0)) return false; /* If we filled the congestion window, do not expand. */ - if (tcp_packets_in_flight(tp) >= tp->snd_cwnd) + if (tcp_packets_in_flight(tp) >= tcp_snd_cwnd(tp)) return false; return true; } -/* When incoming ACK allowed to free some skb from write_queue, - * we remember this event in flag SOCK_QUEUE_SHRUNK and wake up socket - * on the exit from tcp input handler. - * - * PROBLEM: sndbuf expansion does not work well with largesend. - */ static void tcp_new_space(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); @@ -5214,21 +5468,28 @@ static void tcp_new_space(struct sock *sk) tp->snd_cwnd_stamp = tcp_jiffies32; } - sk->sk_write_space(sk); + INDIRECT_CALL_1(sk->sk_write_space, sk_stream_write_space, sk); } -static void tcp_check_space(struct sock *sk) +/* Caller made space either from: + * 1) Freeing skbs in rtx queues (after tp->snd_una has advanced) + * 2) Sent skbs from output queue (and thus advancing tp->snd_nxt) + * + * We might be able to generate EPOLLOUT to the application if: + * 1) Space consumed in output/rtx queues is below sk->sk_sndbuf/2 + * 2) notsent amount (tp->write_seq - tp->snd_nxt) became + * small enough that tcp_stream_memory_free() decides it + * is time to generate EPOLLOUT. + */ +void tcp_check_space(struct sock *sk) { - if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) { - sock_reset_flag(sk, SOCK_QUEUE_SHRUNK); - /* pairs with tcp_poll() */ - smp_mb(); - if (sk->sk_socket && - test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - tcp_new_space(sk); - if (!test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) - tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED); - } + /* pairs with tcp_poll() */ + smp_mb(); + if (sk->sk_socket && + test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { + tcp_new_space(sk); + if (!test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) + tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED); } } @@ -5270,20 +5531,18 @@ send_now: } if (!tcp_is_sack(tp) || - tp->compressed_ack >= sock_net(sk)->ipv4.sysctl_tcp_comp_sack_nr) + tp->compressed_ack >= READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_nr)) goto send_now; if (tp->compressed_ack_rcv_nxt != tp->rcv_nxt) { tp->compressed_ack_rcv_nxt = tp->rcv_nxt; - if (tp->compressed_ack > TCP_FASTRETRANS_THRESH) - NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPACKCOMPRESSED, - tp->compressed_ack - TCP_FASTRETRANS_THRESH); - tp->compressed_ack = 0; + tp->dup_ack_counter = 0; } - - if (++tp->compressed_ack <= TCP_FASTRETRANS_THRESH) + if (tp->dup_ack_counter < TCP_FASTRETRANS_THRESH) { + tp->dup_ack_counter++; goto send_now; - + } + tp->compressed_ack++; if (hrtimer_is_queued(&tp->compressed_ack_timer)) return; @@ -5293,11 +5552,13 @@ send_now: if (tp->srtt_us && tp->srtt_us < rtt) rtt = tp->srtt_us; - delay = min_t(unsigned long, sock_net(sk)->ipv4.sysctl_tcp_comp_sack_delay_ns, + delay = min_t(unsigned long, + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_delay_ns), rtt * (NSEC_PER_USEC >> 3)/20); sock_hold(sk); - hrtimer_start(&tp->compressed_ack_timer, ns_to_ktime(delay), - HRTIMER_MODE_REL_PINNED_SOFT); + hrtimer_start_range_ns(&tp->compressed_ack_timer, ns_to_ktime(delay), + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_slack_ns), + HRTIMER_MODE_REL_PINNED_SOFT); } static inline void tcp_ack_snd_check(struct sock *sk) @@ -5324,7 +5585,7 @@ static void tcp_check_urg(struct sock *sk, const struct tcphdr *th) struct tcp_sock *tp = tcp_sk(sk); u32 ptr = ntohs(th->urg_ptr); - if (ptr && !sock_net(sk)->ipv4.sysctl_tcp_stdurg) + if (ptr && !READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_stdurg)) ptr--; ptr += ntohl(th->seq); @@ -5377,7 +5638,7 @@ static void tcp_check_urg(struct sock *sk, const struct tcphdr *th) } } - tp->urg_data = TCP_URG_NOTYET; + WRITE_ONCE(tp->urg_data, TCP_URG_NOTYET); WRITE_ONCE(tp->urg_seq, ptr); /* Disable header prediction. */ @@ -5390,11 +5651,11 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *t struct tcp_sock *tp = tcp_sk(sk); /* Check if we get a new urgent pointer - normally not. */ - if (th->urg) + if (unlikely(th->urg)) tcp_check_urg(sk, th); /* Do we wait for any urgent data? - normally not... */ - if (tp->urg_data == TCP_URG_NOTYET) { + if (unlikely(tp->urg_data == TCP_URG_NOTYET)) { u32 ptr = tp->urg_seq - ntohl(th->seq) + (th->doff * 4) - th->syn; @@ -5403,7 +5664,7 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *t u8 tmp; if (skb_copy_bits(skb, ptr, &tmp, 1)) BUG(); - tp->urg_data = TCP_URG_VALID | tmp; + WRITE_ONCE(tp->urg_data, TCP_URG_VALID | tmp); if (!sock_flag(sk, SOCK_DEAD)) sk->sk_data_ready(sk); } @@ -5434,7 +5695,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, const struct tcphdr *th, int syn_inerr) { struct tcp_sock *tp = tcp_sk(sk); - bool rst_seq_match = false; + SKB_DR(reason); /* RFC1323: H1. Apply PAWS check first. */ if (tcp_fast_parse_options(sock_net(sk), skb, th, tp) && @@ -5446,6 +5707,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, LINUX_MIB_TCPACKSKIPPEDPAWS, &tp->last_oow_ack_time)) tcp_send_dupack(sk, skb); + SKB_DR_SET(reason, TCP_RFC7323_PAWS); goto discard; } /* Reset is accepted even if it did not pass PAWS. */ @@ -5467,8 +5729,9 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, &tp->last_oow_ack_time)) tcp_send_dupack(sk, skb); } else if (tcp_reset_check(sk, skb)) { - tcp_reset(sk); + goto reset; } + SKB_DR_SET(reason, TCP_INVALID_SEQUENCE); goto discard; } @@ -5484,9 +5747,10 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, * Send a challenge ACK */ if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt || - tcp_reset_check(sk, skb)) { - rst_seq_match = true; - } else if (tcp_is_sack(tp) && tp->rx_opt.num_sacks > 0) { + tcp_reset_check(sk, skb)) + goto reset; + + if (tcp_is_sack(tp) && tp->rx_opt.num_sacks > 0) { struct tcp_sack_block *sp = &tp->selective_acks[0]; int max_sack = sp[0].end_seq; int this_sack; @@ -5499,21 +5763,18 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, } if (TCP_SKB_CB(skb)->seq == max_sack) - rst_seq_match = true; + goto reset; } - if (rst_seq_match) - tcp_reset(sk); - else { - /* Disable TFO if RST is out-of-order - * and no data has been received - * for current active TFO socket - */ - if (tp->syn_fastopen && !tp->data_segs_in && - sk->sk_state == TCP_ESTABLISHED) - tcp_fastopen_active_disable(sk); - tcp_send_challenge_ack(sk, skb); - } + /* Disable TFO if RST is out-of-order + * and no data has been received + * for current active TFO socket + */ + if (tp->syn_fastopen && !tp->data_segs_in && + sk->sk_state == TCP_ESTABLISHED) + tcp_fastopen_active_disable(sk); + tcp_send_challenge_ack(sk); + SKB_DR_SET(reason, TCP_RESET); goto discard; } @@ -5527,14 +5788,22 @@ syn_challenge: if (syn_inerr) TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNCHALLENGE); - tcp_send_challenge_ack(sk, skb); + tcp_send_challenge_ack(sk); + SKB_DR_SET(reason, TCP_INVALID_SYN); goto discard; } + bpf_skops_parse_hdr(sk, skb); + return true; discard: - tcp_drop(sk, skb); + tcp_drop_reason(sk, skb, reason); + return false; + +reset: + tcp_reset(sk, skb); + __kfree_skb(skb); return false; } @@ -5563,6 +5832,7 @@ discard: */ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb) { + enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED; const struct tcphdr *th = (const struct tcphdr *)skb->data; struct tcp_sock *tp = tcp_sk(sk); unsigned int len = skb->len; @@ -5571,7 +5841,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb) trace_tcp_probe(sk, skb); tcp_mstamp_refresh(tp); - if (unlikely(!sk->sk_rx_dst)) + if (unlikely(!rcu_access_pointer(sk->sk_rx_dst))) inet_csk(sk)->icsk_af_ops->sk_rx_dst_set(sk, skb); /* * Header prediction. @@ -5651,6 +5921,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb) tp->rcv_rtt_last_tsecr = tp->rx_opt.rcv_tsecr; return; } else { /* Header too small */ + reason = SKB_DROP_REASON_PKT_TOO_SMALL; TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS); goto discard; } @@ -5678,6 +5949,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb) NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPHPHITS); /* Bulk data transfer: receiver */ + skb_dst_drop(skb); __skb_pull(skb, tcp_header_len); eaten = tcp_queue_rcv(sk, skb, &fragstolen); @@ -5689,6 +5961,8 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb) tcp_data_snd_check(sk); if (!inet_csk_ack_scheduled(sk)) goto no_ack; + } else { + tcp_update_wl(tp, TCP_SKB_CB(skb)->seq); } __tcp_ack_snd_check(sk, 0); @@ -5704,8 +5978,10 @@ slow_path: if (len < (th->doff << 2) || tcp_checksum_complete(skb)) goto csum_error; - if (!th->ack && !th->rst && !th->syn) + if (!th->ack && !th->rst && !th->syn) { + reason = SKB_DROP_REASON_TCP_FLAGS; goto discard; + } /* * Standard slow path. @@ -5715,9 +5991,11 @@ slow_path: return; step5: - if (tcp_ack(sk, skb, FLAG_SLOWPATH | FLAG_UPDATE_TS_RECENT) < 0) + reason = tcp_ack(sk, skb, FLAG_SLOWPATH | FLAG_UPDATE_TS_RECENT); + if ((int)reason < 0) { + reason = -reason; goto discard; - + } tcp_rcv_rtt_measure_ts(sk, skb); /* Process urgent data. */ @@ -5731,15 +6009,17 @@ step5: return; csum_error: + reason = SKB_DROP_REASON_TCP_CSUM; + trace_tcp_bad_csum(skb); TCP_INC_STATS(sock_net(sk), TCP_MIB_CSUMERRORS); TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS); discard: - tcp_drop(sk, skb); + tcp_drop_reason(sk, skb, reason); } EXPORT_SYMBOL(tcp_rcv_established); -void tcp_init_transfer(struct sock *sk, int bpf_op) +void tcp_init_transfer(struct sock *sk, int bpf_op, struct sk_buff *skb) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); @@ -5755,13 +6035,15 @@ void tcp_init_transfer(struct sock *sk, int bpf_op) * retransmission has occurred. */ if (tp->total_retrans > 1 && tp->undo_marker) - tp->snd_cwnd = 1; + tcp_snd_cwnd_set(tp, 1); else - tp->snd_cwnd = tcp_init_cwnd(tp, __sk_dst_get(sk)); + tcp_snd_cwnd_set(tp, tcp_init_cwnd(tp, __sk_dst_get(sk))); tp->snd_cwnd_stamp = tcp_jiffies32; - tcp_call_bpf(sk, bpf_op, 0, NULL); - tcp_init_congestion_control(sk); + bpf_skops_established(sk, bpf_op, skb); + /* Initialize congestion control unless BPF initialized it already: */ + if (!icsk->icsk_ca_initialized) + tcp_init_congestion_control(sk); tcp_init_buffer_space(sk); } @@ -5779,7 +6061,7 @@ void tcp_finish_connect(struct sock *sk, struct sk_buff *skb) sk_mark_napi_id(sk, skb); } - tcp_init_transfer(sk, BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB); + tcp_init_transfer(sk, BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB, skb); /* Prevent spurious tcp_cwnd_restart() on first data * packet. @@ -5838,11 +6120,9 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, tp->fastopen_client_fail = TFO_SYN_RETRANSMITTED; else tp->fastopen_client_fail = TFO_DATA_NOT_ACKED; - skb_rbtree_walk_from(data) { - if (__tcp_retransmit_skb(sk, data, 1)) - break; - } - tcp_rearm_rto(sk); + skb_rbtree_walk_from(data) + tcp_mark_skb_lost(sk, data); + tcp_xmit_retransmit_queue(sk); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPFASTOPENACTIVEFAIL); return true; @@ -5893,6 +6173,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, struct tcp_fastopen_cookie foc = { .len = -1 }; int saved_clamp = tp->rx_opt.mss_clamp; bool fastopen_fail; + SKB_DR(reason); tcp_parse_options(sock_net(sk), skb, &tp->rx_opt, 0, &foc); if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) @@ -5934,8 +6215,10 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, */ if (th->rst) { - tcp_reset(sk); - goto discard; + tcp_reset(sk, skb); +consume: + __kfree_skb(skb); + return 0; } /* rfc793: @@ -5945,9 +6228,10 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, * See note below! * --ANK(990513) */ - if (!th->syn) + if (!th->syn) { + SKB_DR_SET(reason, TCP_FLAGS); goto discard_and_undo; - + } /* rfc793: * "If the SYN bit is on ... * are acceptable then ... @@ -5990,9 +6274,6 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); tcp_initialize_rcv_mss(sk); - if (sk_is_mptcp(sk)) - mptcp_rcv_synsent(sk); - /* Remember, tcp_poll() does not lock socket! * Change state from SYN-SENT only after copied_seq * is initialized. */ @@ -6027,13 +6308,9 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tcp_enter_quickack_mode(sk, TCP_MAX_QUICKACKS); inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, TCP_DELACK_MAX, TCP_RTO_MAX); - -discard: - tcp_drop(sk, skb); - return 0; - } else { - tcp_send_ack(sk); + goto consume; } + tcp_send_ack(sk); return -1; } @@ -6045,15 +6322,16 @@ discard: * * Otherwise (no ACK) drop the segment and return." */ - + SKB_DR_SET(reason, TCP_RESET); goto discard_and_undo; } /* PAWS check. */ if (tp->rx_opt.ts_recent_stamp && tp->rx_opt.saw_tstamp && - tcp_paws_reject(&tp->rx_opt, 0)) + tcp_paws_reject(&tp->rx_opt, 0)) { + SKB_DR_SET(reason, TCP_RFC7323_PAWS); goto discard_and_undo; - + } if (th->syn) { /* We see SYN without ACK. It is attempt of * simultaneous connect with crossed SYNs. @@ -6102,7 +6380,7 @@ discard: */ return -1; #else - goto discard; + goto consume; #endif } /* "fifth, if neither of the SYN or RST bits is set then @@ -6112,7 +6390,8 @@ discard: discard_and_undo: tcp_clear_options(&tp->rx_opt); tp->rx_opt.mss_clamp = saved_clamp; - goto discard; + tcp_drop_reason(sk, skb, reason); + return 0; reset_and_undo: tcp_clear_options(&tp->rx_opt); @@ -6167,21 +6446,26 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) struct request_sock *req; int queued = 0; bool acceptable; + SKB_DR(reason); switch (sk->sk_state) { case TCP_CLOSE: + SKB_DR_SET(reason, TCP_CLOSE); goto discard; case TCP_LISTEN: if (th->ack) return 1; - if (th->rst) + if (th->rst) { + SKB_DR_SET(reason, TCP_RESET); goto discard; - + } if (th->syn) { - if (th->fin) + if (th->fin) { + SKB_DR_SET(reason, TCP_FLAGS); goto discard; + } /* It is possible that we process SYN packets from backlog, * so we need to make sure to disable BH and RCU right there. */ @@ -6196,6 +6480,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) consume_skb(skb); return 0; } + SKB_DR_SET(reason, TCP_FLAGS); goto discard; case TCP_SYN_SENT: @@ -6222,13 +6507,16 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) WARN_ON_ONCE(sk->sk_state != TCP_SYN_RECV && sk->sk_state != TCP_FIN_WAIT1); - if (!tcp_check_req(sk, skb, req, true, &req_stolen)) + if (!tcp_check_req(sk, skb, req, true, &req_stolen)) { + SKB_DR_SET(reason, TCP_FASTOPEN); goto discard; + } } - if (!th->ack && !th->rst && !th->syn) + if (!th->ack && !th->rst && !th->syn) { + SKB_DR_SET(reason, TCP_FLAGS); goto discard; - + } if (!tcp_validate_incoming(sk, skb, th, 0)) return 0; @@ -6240,7 +6528,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) if (!acceptable) { if (sk->sk_state == TCP_SYN_RECV) return 1; /* send one RST */ - tcp_send_challenge_ack(sk, skb); + tcp_send_challenge_ack(sk); + SKB_DR_SET(reason, TCP_OLD_ACK); goto discard; } switch (sk->sk_state) { @@ -6254,7 +6543,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) } else { tcp_try_undo_spurious_syn(sk); tp->retrans_stamp = 0; - tcp_init_transfer(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); + tcp_init_transfer(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB, + skb); WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); } smp_mb(); @@ -6333,7 +6623,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) inet_csk_reset_keepalive_timer(sk, tmo); } else { tcp_time_wait(sk, TCP_FIN_WAIT2, tmo); - goto discard; + goto consume; } break; } @@ -6341,7 +6631,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) case TCP_CLOSING: if (tp->snd_una == tp->write_seq) { tcp_time_wait(sk, TCP_TIME_WAIT, 0); - goto discard; + goto consume; } break; @@ -6349,7 +6639,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) if (tp->snd_una == tp->write_seq) { tcp_update_metrics(sk); tcp_done(sk); - goto discard; + goto consume; } break; } @@ -6363,11 +6653,14 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) case TCP_CLOSING: case TCP_LAST_ACK: if (!before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) { - if (sk_is_mptcp(sk)) - mptcp_incoming_options(sk, skb, &tp->rx_opt); + /* If a subflow has been reset, the packet should not + * continue to be processed, drop the packet. + */ + if (sk_is_mptcp(sk) && !mptcp_incoming_options(sk, skb)) + goto discard; break; } - /* fall through */ + fallthrough; case TCP_FIN_WAIT1: case TCP_FIN_WAIT2: /* RFC 793 says to queue data in these states, @@ -6378,11 +6671,11 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq && after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt)) { NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONDATA); - tcp_reset(sk); + tcp_reset(sk, skb); return 1; } } - /* Fall through */ + fallthrough; case TCP_ESTABLISHED: tcp_data_queue(sk, skb); queued = 1; @@ -6397,9 +6690,13 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) if (!queued) { discard: - tcp_drop(sk, skb); + tcp_drop_reason(sk, skb, reason); } return 0; + +consume: + __kfree_skb(skb); + return 0; } EXPORT_SYMBOL(tcp_rcv_state_process); @@ -6450,7 +6747,7 @@ static void tcp_ecn_create_request(struct request_sock *req, ect = !INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield); ecn_ok_dst = dst_feature(dst, DST_FEATURE_ECN_MASK); - ecn_ok = net->ipv4.sysctl_tcp_ecn || ecn_ok_dst; + ecn_ok = READ_ONCE(net->ipv4.sysctl_tcp_ecn) || ecn_ok_dst; if (((!ect || th->res1) && ecn_ok) || tcp_ca_needs_ecn(listen_sk) || (ecn_ok_dst & DST_FEATURE_ECN_CA) || @@ -6465,7 +6762,6 @@ static void tcp_openreq_init(struct request_sock *req, struct inet_request_sock *ireq = inet_rsk(req); req->rsk_rcv_wnd = 0; /* So that tcp_send_synack() knows! */ - req->cookie_ts = 0; tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq; tcp_rsk(req)->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; tcp_rsk(req)->snt_synack = 0; @@ -6482,7 +6778,8 @@ static void tcp_openreq_init(struct request_sock *req, ireq->ir_num = ntohs(tcp_hdr(skb)->dest); ireq->ir_mark = inet_request_mark(sk, skb); #if IS_ENABLED(CONFIG_SMC) - ireq->smc_ok = rx_opt->smc_ok; + ireq->smc_ok = rx_opt->smc_ok && !(tcp_sk(sk)->smc_hs_congested && + tcp_sk(sk)->smc_hs_congested(sk)); #endif } @@ -6504,6 +6801,7 @@ struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops, ireq->ireq_state = TCP_NEW_SYN_RECV; write_pnet(&ireq->ireq_net, sock_net(sk_listener)); ireq->ireq_family = sk_listener->sk_family; + req->timeout = TCP_TIMEOUT_INIT; } return req; @@ -6517,11 +6815,14 @@ static bool tcp_syn_flood_action(const struct sock *sk, const char *proto) { struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue; const char *msg = "Dropping request"; - bool want_cookie = false; struct net *net = sock_net(sk); + bool want_cookie = false; + u8 syncookies; + + syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies); #ifdef CONFIG_SYN_COOKIES - if (net->ipv4.sysctl_tcp_syncookies) { + if (syncookies) { msg = "Sending cookies"; want_cookie = true; __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPREQQFULLDOCOOKIES); @@ -6529,8 +6830,7 @@ static bool tcp_syn_flood_action(const struct sock *sk, const char *proto) #endif __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPREQQFULLDROP); - if (!queue->synflood_warned && - net->ipv4.sysctl_tcp_syncookies != 2 && + if (!queue->synflood_warned && syncookies != 2 && xchg(&queue->synflood_warned, 1) == 0) net_info_ratelimited("%s: Possible SYN flooding on port %d. %s. Check SNMP counters.\n", proto, sk->sk_num, msg); @@ -6544,13 +6844,27 @@ static void tcp_reqsk_record_syn(const struct sock *sk, { if (tcp_sk(sk)->save_syn) { u32 len = skb_network_header_len(skb) + tcp_hdrlen(skb); - u32 *copy; + struct saved_syn *saved_syn; + u32 mac_hdrlen; + void *base; + + if (tcp_sk(sk)->save_syn == 2) { /* Save full header. */ + base = skb_mac_header(skb); + mac_hdrlen = skb_mac_header_len(skb); + len += mac_hdrlen; + } else { + base = skb_network_header(skb); + mac_hdrlen = 0; + } - copy = kmalloc(len + sizeof(u32), GFP_ATOMIC); - if (copy) { - copy[0] = len; - memcpy(©[1], skb_network_header(skb), len); - req->saved_syn = copy; + saved_syn = kmalloc(struct_size(saved_syn, data, len), + GFP_ATOMIC); + if (saved_syn) { + saved_syn->mac_hdrlen = mac_hdrlen; + saved_syn->network_hdrlen = skb_network_header_len(skb); + saved_syn->tcp_hdrlen = tcp_hdrlen(skb); + memcpy(saved_syn->data, base, len); + req->saved_syn = saved_syn; } } } @@ -6565,7 +6879,7 @@ u16 tcp_get_syncookie_mss(struct request_sock_ops *rsk_ops, struct tcp_sock *tp = tcp_sk(sk); u16 mss; - if (sock_net(sk)->ipv4.sysctl_tcp_syncookies != 2 && + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_syncookies) != 2 && !inet_csk_reqsk_queue_is_full(sk)) return 0; @@ -6599,13 +6913,15 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, bool want_cookie = false; struct dst_entry *dst; struct flowi fl; + u8 syncookies; + + syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies); /* TW buckets are converted to open requests without * limitations, they conserve resources and peer is * evidently real one. */ - if ((net->ipv4.sysctl_tcp_syncookies == 2 || - inet_csk_reqsk_queue_is_full(sk)) && !isn) { + if ((syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) && !isn) { want_cookie = tcp_syn_flood_action(sk, rsk_ops->slab_name); if (!want_cookie) goto drop; @@ -6620,6 +6936,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, if (!req) goto drop; + req->syncookie = want_cookie; tcp_rsk(req)->af_specific = af_ops; tcp_rsk(req)->ts_off = 0; #if IS_ENABLED(CONFIG_MPTCP) @@ -6645,26 +6962,20 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, /* Note: tcp_v6_init_req() might override ir_iif for link locals */ inet_rsk(req)->ir_iif = inet_request_bound_dev_if(sk, skb); - af_ops->init_req(req, sk, skb); - - if (IS_ENABLED(CONFIG_MPTCP) && want_cookie) - tcp_rsk(req)->is_mptcp = 0; - - if (security_inet_conn_request(sk, skb, req)) + dst = af_ops->route_req(sk, skb, &fl, req); + if (!dst) goto drop_and_free; if (tmp_opt.tstamp_ok) tcp_rsk(req)->ts_off = af_ops->init_ts_off(net, skb); - dst = af_ops->route_req(sk, &fl, req); - if (!dst) - goto drop_and_free; - if (!want_cookie && !isn) { + int max_syn_backlog = READ_ONCE(net->ipv4.sysctl_max_syn_backlog); + /* Kill the following clause, if you dislike this way. */ - if (!net->ipv4.sysctl_tcp_syncookies && - (net->ipv4.sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) < - (net->ipv4.sysctl_max_syn_backlog >> 2)) && + if (!syncookies && + (max_syn_backlog - inet_csk_reqsk_queue_len(sk) < + (max_syn_backlog >> 2)) && !tcp_peer_is_proven(req, dst)) { /* Without syncookies last quarter of * backlog is filled with destinations, @@ -6685,13 +6996,13 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, if (want_cookie) { isn = cookie_init_sequence(af_ops, sk, skb, &req->mss); - req->cookie_ts = tmp_opt.tstamp_ok; if (!tmp_opt.tstamp_ok) inet_rsk(req)->ecn_ok = 0; } tcp_rsk(req)->snt_isn = isn; tcp_rsk(req)->txhash = net_tx_rndhash(); + tcp_rsk(req)->syn_tos = TCP_SKB_CB(skb)->ip_dsfield; tcp_openreq_init_rwin(req, sk, dst); sk_rx_queue_set(req_to_sk(req), skb); if (!want_cookie) { @@ -6700,7 +7011,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, } if (fastopen_sk) { af_ops->send_synack(fastopen_sk, dst, &fl, req, - &foc, TCP_SYNACK_FASTOPEN); + &foc, TCP_SYNACK_FASTOPEN, skb); /* Add the child socket directly into the accept queue */ if (!inet_csk_reqsk_queue_add(sk, req, fastopen_sk)) { reqsk_fastopen_remove(fastopen_sk, req, false); @@ -6713,12 +7024,14 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, sock_put(fastopen_sk); } else { tcp_rsk(req)->tfo_listener = false; - if (!want_cookie) - inet_csk_reqsk_queue_hash_add(sk, req, - tcp_timeout_init((struct sock *)req)); + if (!want_cookie) { + req->timeout = tcp_timeout_init((struct sock *)req); + inet_csk_reqsk_queue_hash_add(sk, req, req->timeout); + } af_ops->send_synack(sk, dst, &fl, req, &foc, !want_cookie ? TCP_SYNACK_NORMAL : - TCP_SYNACK_COOKIE); + TCP_SYNACK_COOKIE, + skb); if (want_cookie) { reqsk_free(req); return 0; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index df1166b76126..87d440f47a70 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -76,6 +76,7 @@ #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/inetdevice.h> +#include <linux/btf_ids.h> #include <crypto/hash.h> #include <linux/scatterlist.h> @@ -90,6 +91,8 @@ static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, struct inet_hashinfo tcp_hashinfo; EXPORT_SYMBOL(tcp_hashinfo); +static DEFINE_PER_CPU(struct sock *, ipv4_tcp_sk); + static u32 tcp_v4_init_seq(const struct sk_buff *skb) { return secure_tcp_seq(ip_hdr(skb)->daddr, @@ -105,10 +108,10 @@ static u32 tcp_v4_init_ts_off(const struct net *net, const struct sk_buff *skb) int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp) { + int reuse = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_tw_reuse); const struct inet_timewait_sock *tw = inet_twsk(sktw); const struct tcp_timewait_sock *tcptw = tcp_twsk(sktw); struct tcp_sock *tp = tcp_sk(sk); - int reuse = sock_net(sk)->ipv4.sysctl_tcp_tw_reuse; if (reuse == 2) { /* Still does not detect *everything* that goes through @@ -196,16 +199,18 @@ static int tcp_v4_pre_connect(struct sock *sk, struct sockaddr *uaddr, /* This will initiate an outgoing connection. */ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { + struct inet_bind_hashbucket *prev_addr_hashbucket = NULL; struct sockaddr_in *usin = (struct sockaddr_in *)uaddr; + struct inet_timewait_death_row *tcp_death_row; + __be32 daddr, nexthop, prev_sk_rcv_saddr; struct inet_sock *inet = inet_sk(sk); struct tcp_sock *tp = tcp_sk(sk); + struct ip_options_rcu *inet_opt; + struct net *net = sock_net(sk); __be16 orig_sport, orig_dport; - __be32 daddr, nexthop; struct flowi4 *fl4; struct rtable *rt; int err; - struct ip_options_rcu *inet_opt; - struct inet_timewait_death_row *tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row; if (addr_len < sizeof(struct sockaddr_in)) return -EINVAL; @@ -226,13 +231,12 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) orig_dport = usin->sin_port; fl4 = &inet->cork.fl.u.ip4; rt = ip_route_connect(fl4, nexthop, inet->inet_saddr, - RT_CONN_FLAGS(sk), sk->sk_bound_dev_if, - IPPROTO_TCP, - orig_sport, orig_dport, sk); + sk->sk_bound_dev_if, IPPROTO_TCP, orig_sport, + orig_dport, sk); if (IS_ERR(rt)) { err = PTR_ERR(rt); if (err == -ENETUNREACH) - IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES); + IP_INC_STATS(net, IPSTATS_MIB_OUTNOROUTES); return err; } @@ -244,10 +248,29 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (!inet_opt || !inet_opt->opt.srr) daddr = fl4->daddr; - if (!inet->inet_saddr) + tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row; + + if (!inet->inet_saddr) { + if (inet_csk(sk)->icsk_bind2_hash) { + prev_addr_hashbucket = inet_bhashfn_portaddr(tcp_death_row->hashinfo, + sk, net, inet->inet_num); + prev_sk_rcv_saddr = sk->sk_rcv_saddr; + } inet->inet_saddr = fl4->saddr; + } + sk_rcv_saddr_set(sk, inet->inet_saddr); + if (prev_addr_hashbucket) { + err = inet_bhash2_update_saddr(prev_addr_hashbucket, sk); + if (err) { + inet->inet_saddr = 0; + sk_rcv_saddr_set(sk, prev_sk_rcv_saddr); + ip_rt_put(rt); + return err; + } + } + if (tp->rx_opt.ts_recent_stamp && inet->inet_daddr != daddr) { /* Reset inherited state */ tp->rx_opt.ts_recent = 0; @@ -296,12 +319,11 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) inet->inet_daddr, inet->inet_sport, usin->sin_port)); - tp->tsoffset = secure_tcp_ts_off(sock_net(sk), - inet->inet_saddr, + tp->tsoffset = secure_tcp_ts_off(net, inet->inet_saddr, inet->inet_daddr); } - inet->inet_id = prandom_u32(); + inet->inet_id = get_random_u16(); if (tcp_fastopen_defer_connect(sk, &err)) return err; @@ -341,7 +363,7 @@ void tcp_v4_mtu_reduced(struct sock *sk) if ((1 << sk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE)) return; - mtu = tcp_sk(sk)->mtu_info; + mtu = READ_ONCE(tcp_sk(sk)->mtu_info); dst = inet_csk_update_pmtu(sk, mtu); if (!dst) return; @@ -403,6 +425,46 @@ void tcp_req_err(struct sock *sk, u32 seq, bool abort) } EXPORT_SYMBOL(tcp_req_err); +/* TCP-LD (RFC 6069) logic */ +void tcp_ld_RTO_revert(struct sock *sk, u32 seq) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + struct tcp_sock *tp = tcp_sk(sk); + struct sk_buff *skb; + s32 remaining; + u32 delta_us; + + if (sock_owned_by_user(sk)) + return; + + if (seq != tp->snd_una || !icsk->icsk_retransmits || + !icsk->icsk_backoff) + return; + + skb = tcp_rtx_queue_head(sk); + if (WARN_ON_ONCE(!skb)) + return; + + icsk->icsk_backoff--; + icsk->icsk_rto = tp->srtt_us ? __tcp_set_rto(tp) : TCP_TIMEOUT_INIT; + icsk->icsk_rto = inet_csk_rto_backoff(icsk, TCP_RTO_MAX); + + tcp_mstamp_refresh(tp); + delta_us = (u32)(tp->tcp_mstamp - tcp_skb_timestamp_us(skb)); + remaining = icsk->icsk_rto - usecs_to_jiffies(delta_us); + + if (remaining > 0) { + inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, + remaining, TCP_RTO_MAX); + } else { + /* RTO revert clocked out retransmission. + * Will retransmit now. + */ + tcp_retransmit_timer(sk); + } +} +EXPORT_SYMBOL(tcp_ld_RTO_revert); + /* * This routine is called by the ICMP module when it gets some * sort of error condition. If err < 0 then the socket should @@ -419,27 +481,23 @@ EXPORT_SYMBOL(tcp_req_err); * */ -int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) +int tcp_v4_err(struct sk_buff *skb, u32 info) { - const struct iphdr *iph = (const struct iphdr *)icmp_skb->data; - struct tcphdr *th = (struct tcphdr *)(icmp_skb->data + (iph->ihl << 2)); - struct inet_connection_sock *icsk; + const struct iphdr *iph = (const struct iphdr *)skb->data; + struct tcphdr *th = (struct tcphdr *)(skb->data + (iph->ihl << 2)); struct tcp_sock *tp; struct inet_sock *inet; - const int type = icmp_hdr(icmp_skb)->type; - const int code = icmp_hdr(icmp_skb)->code; + const int type = icmp_hdr(skb)->type; + const int code = icmp_hdr(skb)->code; struct sock *sk; - struct sk_buff *skb; struct request_sock *fastopen; u32 seq, snd_una; - s32 remaining; - u32 delta_us; int err; - struct net *net = dev_net(icmp_skb->dev); + struct net *net = dev_net(skb->dev); - sk = __inet_lookup_established(net, &tcp_hashinfo, iph->daddr, - th->dest, iph->saddr, ntohs(th->source), - inet_iif(icmp_skb), 0); + sk = __inet_lookup_established(net, net->ipv4.tcp_death_row.hashinfo, + iph->daddr, th->dest, iph->saddr, + ntohs(th->source), inet_iif(skb), 0); if (!sk) { __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); return -ENOENT; @@ -471,12 +529,14 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) if (sk->sk_state == TCP_CLOSE) goto out; - if (unlikely(iph->ttl < inet_sk(sk)->min_ttl)) { - __NET_INC_STATS(net, LINUX_MIB_TCPMINTTLDROP); - goto out; + if (static_branch_unlikely(&ip4_min_ttl)) { + /* min_ttl can be changed concurrently from do_ip_setsockopt() */ + if (unlikely(iph->ttl < READ_ONCE(inet_sk(sk)->min_ttl))) { + __NET_INC_STATS(net, LINUX_MIB_TCPMINTTLDROP); + goto out; + } } - icsk = inet_csk(sk); tp = tcp_sk(sk); /* XXX (TFO) - tp->snd_una should be ISN (tcp_create_openreq_child() */ fastopen = rcu_dereference(tp->fastopen_rsk); @@ -490,7 +550,7 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) switch (type) { case ICMP_REDIRECT: if (!sock_owned_by_user(sk)) - do_redirect(icmp_skb, sk); + do_redirect(skb, sk); goto out; case ICMP_SOURCE_QUENCH: /* Just silently ignore these. */ @@ -510,7 +570,7 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) if (sk->sk_state == TCP_LISTEN) goto out; - tp->mtu_info = info; + WRITE_ONCE(tp->mtu_info, info); if (!sock_owned_by_user(sk)) { tcp_v4_mtu_reduced(sk); } else { @@ -521,41 +581,12 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) } err = icmp_err_convert[code].errno; - /* check if icmp_skb allows revert of backoff - * (see draft-zimmermann-tcp-lcd) */ - if (code != ICMP_NET_UNREACH && code != ICMP_HOST_UNREACH) - break; - if (seq != tp->snd_una || !icsk->icsk_retransmits || - !icsk->icsk_backoff || fastopen) - break; - - if (sock_owned_by_user(sk)) - break; - - skb = tcp_rtx_queue_head(sk); - if (WARN_ON_ONCE(!skb)) - break; - - icsk->icsk_backoff--; - icsk->icsk_rto = tp->srtt_us ? __tcp_set_rto(tp) : - TCP_TIMEOUT_INIT; - icsk->icsk_rto = inet_csk_rto_backoff(icsk, TCP_RTO_MAX); - - - tcp_mstamp_refresh(tp); - delta_us = (u32)(tp->tcp_mstamp - tcp_skb_timestamp_us(skb)); - remaining = icsk->icsk_rto - - usecs_to_jiffies(delta_us); - - if (remaining > 0) { - inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, - remaining, TCP_RTO_MAX); - } else { - /* RTO revert clocked out retransmission. - * Will retransmit now */ - tcp_retransmit_timer(sk); - } - + /* check if this ICMP message allows revert of backoff. + * (see RFC 6069) + */ + if (!fastopen && + (code == ICMP_NET_UNREACH || code == ICMP_HOST_UNREACH)) + tcp_ld_RTO_revert(sk, seq); break; case ICMP_TIME_EXCEEDED: err = EHOSTUNREACH; @@ -568,15 +599,17 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) case TCP_SYN_SENT: case TCP_SYN_RECV: /* Only in fast or simultaneous open. If a fast open socket is - * is already accepted it is treated as a connected one below. + * already accepted it is treated as a connected one below. */ if (fastopen && !fastopen->sk) break; + ip_icmp_error(sk, skb, err, th->dest, info, (u8 *)th); + if (!sock_owned_by_user(sk)) { sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); tcp_done(sk); } else { @@ -604,7 +637,7 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) inet = inet_sk(sk); if (!sock_owned_by_user(sk) && inet->recverr) { sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); } else { /* Only an error on timeout */ sk->sk_err_soft = err; } @@ -646,14 +679,18 @@ EXPORT_SYMBOL(tcp_v4_send_check); * Exception: precedence violation. We do not implement it in any case. */ +#ifdef CONFIG_TCP_MD5SIG +#define OPTION_BYTES TCPOLEN_MD5SIG_ALIGNED +#else +#define OPTION_BYTES sizeof(__be32) +#endif + static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) { const struct tcphdr *th = tcp_hdr(skb); struct { struct tcphdr th; -#ifdef CONFIG_TCP_MD5SIG - __be32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; -#endif + __be32 opt[OPTION_BYTES / sizeof(__be32)]; } rep; struct ip_reply_arg arg; #ifdef CONFIG_TCP_MD5SIG @@ -723,8 +760,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) * Incoming packet is checked with md5 hash with finding key, * no RST generated if md5 hash doesn't match. */ - sk1 = __inet_lookup_listener(net, &tcp_hashinfo, NULL, 0, - ip_hdr(skb)->saddr, + sk1 = __inet_lookup_listener(net, net->ipv4.tcp_death_row.hashinfo, + NULL, 0, ip_hdr(skb)->saddr, th->source, ip_hdr(skb)->daddr, ntohs(th->source), dif, sdif); /* don't send rst if it can't find key */ @@ -761,6 +798,17 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) ip_hdr(skb)->daddr, &rep.th); } #endif + /* Can't co-exist with TCPMD5, hence check rep.opt[0] */ + if (rep.opt[0] == 0) { + __be32 mrst = mptcp_reset_option(skb); + + if (mrst) { + rep.opt[0] = mrst; + arg.iov[0].iov_len += sizeof(mrst); + rep.th.doff = arg.iov[0].iov_len / 4; + } + } + arg.csum = csum_tcpudp_nofold(ip_hdr(skb)->daddr, ip_hdr(skb)->saddr, /* XXX */ arg.iov[0].iov_len, IPPROTO_TCP, 0); @@ -783,13 +831,15 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) arg.tos = ip_hdr(skb)->tos; arg.uid = sock_net_uid(net, sk && sk_fullsock(sk) ? sk : NULL); local_bh_disable(); - ctl_sk = this_cpu_read(*net->ipv4.tcp_sk); + ctl_sk = this_cpu_read(ipv4_tcp_sk); + sock_net_set(ctl_sk, net); if (sk) { ctl_sk->sk_mark = (sk->sk_state == TCP_TIME_WAIT) ? inet_twsk(sk)->tw_mark : sk->sk_mark; ctl_sk->sk_priority = (sk->sk_state == TCP_TIME_WAIT) ? inet_twsk(sk)->tw_priority : sk->sk_priority; transmit_time = tcp_transmit_time(sk); + xfrm_sk_clone_policy(ctl_sk, sk); } ip_send_unicast_reply(ctl_sk, skb, &TCP_SKB_CB(skb)->header.h4.opt, @@ -798,6 +848,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) transmit_time); ctl_sk->sk_mark = 0; + xfrm_sk_free_policy(ctl_sk); + sock_net_set(ctl_sk, &init_net); __TCP_INC_STATS(net, TCP_MIB_OUTSEGS); __TCP_INC_STATS(net, TCP_MIB_OUTRSTS); local_bh_enable(); @@ -881,7 +933,8 @@ static void tcp_v4_send_ack(const struct sock *sk, arg.tos = tos; arg.uid = sock_net_uid(net, sk_fullsock(sk) ? sk : NULL); local_bh_disable(); - ctl_sk = this_cpu_read(*net->ipv4.tcp_sk); + ctl_sk = this_cpu_read(ipv4_tcp_sk); + sock_net_set(ctl_sk, net); ctl_sk->sk_mark = (sk->sk_state == TCP_TIME_WAIT) ? inet_twsk(sk)->tw_mark : sk->sk_mark; ctl_sk->sk_priority = (sk->sk_state == TCP_TIME_WAIT) ? @@ -894,6 +947,7 @@ static void tcp_v4_send_ack(const struct sock *sk, transmit_time); ctl_sk->sk_mark = 0; + sock_net_set(ctl_sk, &init_net); __TCP_INC_STATS(net, TCP_MIB_OUTSEGS); local_bh_enable(); } @@ -956,26 +1010,38 @@ static int tcp_v4_send_synack(const struct sock *sk, struct dst_entry *dst, struct flowi *fl, struct request_sock *req, struct tcp_fastopen_cookie *foc, - enum tcp_synack_type synack_type) + enum tcp_synack_type synack_type, + struct sk_buff *syn_skb) { const struct inet_request_sock *ireq = inet_rsk(req); struct flowi4 fl4; int err = -1; struct sk_buff *skb; + u8 tos; /* First, grab a route. */ if (!dst && (dst = inet_csk_route_req(sk, &fl4, req)) == NULL) return -1; - skb = tcp_make_synack(sk, dst, req, foc, synack_type); + skb = tcp_make_synack(sk, dst, req, foc, synack_type, syn_skb); if (skb) { __tcp_v4_send_check(skb, ireq->ir_loc_addr, ireq->ir_rmt_addr); + tos = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reflect_tos) ? + (tcp_rsk(req)->syn_tos & ~INET_ECN_MASK) | + (inet_sk(sk)->tos & INET_ECN_MASK) : + inet_sk(sk)->tos; + + if (!INET_ECN_is_capable(tos) && + tcp_bpf_ca_needs_ecn((struct sock *)req)) + tos |= INET_ECN_ECT_0; + rcu_read_lock(); err = ip_build_and_send_pkt(skb, sk, ireq->ir_loc_addr, ireq->ir_rmt_addr, - rcu_dereference(ireq->ireq_opt)); + rcu_dereference(ireq->ireq_opt), + tos); rcu_read_unlock(); err = net_xmit_eval(err); } @@ -1001,6 +1067,20 @@ static void tcp_v4_reqsk_destructor(struct request_sock *req) DEFINE_STATIC_KEY_FALSE(tcp_md5_needed); EXPORT_SYMBOL(tcp_md5_needed); +static bool better_md5_match(struct tcp_md5sig_key *old, struct tcp_md5sig_key *new) +{ + if (!old) + return true; + + /* l3index always overrides non-l3index */ + if (old->l3index && new->l3index == 0) + return false; + if (old->l3index == 0 && new->l3index) + return true; + + return old->prefixlen < new->prefixlen; +} + /* Find the Key structure for an address. */ struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk, int l3index, const union tcp_md5_addr *addr, @@ -1019,10 +1099,11 @@ struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk, int l3index, if (!md5sig) return NULL; - hlist_for_each_entry_rcu(key, &md5sig->head, node) { + hlist_for_each_entry_rcu(key, &md5sig->head, node, + lockdep_sock_is_held(sk)) { if (key->family != family) continue; - if (key->l3index && key->l3index != l3index) + if (key->flags & TCP_MD5SIG_FLAG_IFINDEX && key->l3index != l3index) continue; if (family == AF_INET) { mask = inet_make_mask(key->prefixlen); @@ -1037,8 +1118,7 @@ struct tcp_md5sig_key *__tcp_md5_do_lookup(const struct sock *sk, int l3index, match = false; } - if (match && (!best_match || - key->prefixlen > best_match->prefixlen)) + if (match && better_md5_match(best_match, key)) best_match = key; } return best_match; @@ -1048,7 +1128,7 @@ EXPORT_SYMBOL(__tcp_md5_do_lookup); static struct tcp_md5sig_key *tcp_md5_do_lookup_exact(const struct sock *sk, const union tcp_md5_addr *addr, int family, u8 prefixlen, - int l3index) + int l3index, u8 flags) { const struct tcp_sock *tp = tcp_sk(sk); struct tcp_md5sig_key *key; @@ -1064,10 +1144,13 @@ static struct tcp_md5sig_key *tcp_md5_do_lookup_exact(const struct sock *sk, if (family == AF_INET6) size = sizeof(struct in6_addr); #endif - hlist_for_each_entry_rcu(key, &md5sig->head, node) { + hlist_for_each_entry_rcu(key, &md5sig->head, node, + lockdep_sock_is_held(sk)) { if (key->family != family) continue; - if (key->l3index && key->l3index != l3index) + if ((key->flags & TCP_MD5SIG_FLAG_IFINDEX) != (flags & TCP_MD5SIG_FLAG_IFINDEX)) + continue; + if (key->l3index != l3index) continue; if (!memcmp(&key->addr, addr, size) && key->prefixlen == prefixlen) @@ -1091,7 +1174,7 @@ EXPORT_SYMBOL(tcp_v4_md5_lookup); /* This can be called on a newly created socket, from other files */ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, - int family, u8 prefixlen, int l3index, + int family, u8 prefixlen, int l3index, u8 flags, const u8 *newkey, u8 newkeylen, gfp_t gfp) { /* Add Key to the list */ @@ -1099,11 +1182,23 @@ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, struct tcp_sock *tp = tcp_sk(sk); struct tcp_md5sig_info *md5sig; - key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen, l3index); + key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen, l3index, flags); if (key) { - /* Pre-existing entry - just update that one. */ - memcpy(key->key, newkey, newkeylen); - key->keylen = newkeylen; + /* Pre-existing entry - just update that one. + * Note that the key might be used concurrently. + * data_race() is telling kcsan that we do not care of + * key mismatches, since changing MD5 key on live flows + * can lead to packet drops. + */ + data_race(memcpy(key->key, newkey, newkeylen)); + + /* Pairs with READ_ONCE() in tcp_md5_hash_key(). + * Also note that a reader could catch new key->keylen value + * but old key->key[], this is the reason we use __GFP_ZERO + * at sock_kmalloc() time below these lines. + */ + WRITE_ONCE(key->keylen, newkeylen); + return 0; } @@ -1114,12 +1209,12 @@ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, if (!md5sig) return -ENOMEM; - sk_nocaps_add(sk, NETIF_F_GSO_MASK); + sk_gso_disable(sk); INIT_HLIST_HEAD(&md5sig->head); rcu_assign_pointer(tp->md5sig_info, md5sig); } - key = sock_kmalloc(sk, sizeof(*key), gfp); + key = sock_kmalloc(sk, sizeof(*key), gfp | __GFP_ZERO); if (!key) return -ENOMEM; if (!tcp_alloc_md5sig_pool()) { @@ -1132,20 +1227,21 @@ int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, key->family = family; key->prefixlen = prefixlen; key->l3index = l3index; + key->flags = flags; memcpy(&key->addr, addr, - (family == AF_INET6) ? sizeof(struct in6_addr) : - sizeof(struct in_addr)); + (IS_ENABLED(CONFIG_IPV6) && family == AF_INET6) ? sizeof(struct in6_addr) : + sizeof(struct in_addr)); hlist_add_head_rcu(&key->node, &md5sig->head); return 0; } EXPORT_SYMBOL(tcp_md5_do_add); int tcp_md5_do_del(struct sock *sk, const union tcp_md5_addr *addr, int family, - u8 prefixlen, int l3index) + u8 prefixlen, int l3index, u8 flags) { struct tcp_md5sig_key *key; - key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen, l3index); + key = tcp_md5_do_lookup_exact(sk, addr, family, prefixlen, l3index, flags); if (!key) return -ENOENT; hlist_del_rcu(&key->node); @@ -1172,23 +1268,26 @@ static void tcp_clear_md5_list(struct sock *sk) } static int tcp_v4_parse_md5_keys(struct sock *sk, int optname, - char __user *optval, int optlen) + sockptr_t optval, int optlen) { struct tcp_md5sig cmd; struct sockaddr_in *sin = (struct sockaddr_in *)&cmd.tcpm_addr; const union tcp_md5_addr *addr; u8 prefixlen = 32; int l3index = 0; + u8 flags; if (optlen < sizeof(cmd)) return -EINVAL; - if (copy_from_user(&cmd, optval, sizeof(cmd))) + if (copy_from_sockptr(&cmd, optval, sizeof(cmd))) return -EFAULT; if (sin->sin_family != AF_INET) return -EINVAL; + flags = cmd.tcpm_flags & TCP_MD5SIG_FLAG_IFINDEX; + if (optname == TCP_MD5SIG_EXT && cmd.tcpm_flags & TCP_MD5SIG_FLAG_PREFIX) { prefixlen = cmd.tcpm_prefixlen; @@ -1196,7 +1295,7 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname, return -EINVAL; } - if (optname == TCP_MD5SIG_EXT && + if (optname == TCP_MD5SIG_EXT && cmd.tcpm_ifindex && cmd.tcpm_flags & TCP_MD5SIG_FLAG_IFINDEX) { struct net_device *dev; @@ -1217,12 +1316,12 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, int optname, addr = (union tcp_md5_addr *)&sin->sin_addr.s_addr; if (!cmd.tcpm_keylen) - return tcp_md5_do_del(sk, addr, AF_INET, prefixlen, l3index); + return tcp_md5_do_del(sk, addr, AF_INET, prefixlen, l3index, flags); if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN) return -EINVAL; - return tcp_md5_do_add(sk, addr, AF_INET, prefixlen, l3index, + return tcp_md5_do_add(sk, addr, AF_INET, prefixlen, l3index, flags, cmd.tcpm_key, cmd.tcpm_keylen, GFP_KERNEL); } @@ -1331,72 +1430,6 @@ EXPORT_SYMBOL(tcp_v4_md5_hash_skb); #endif -/* Called with rcu_read_lock() */ -static bool tcp_v4_inbound_md5_hash(const struct sock *sk, - const struct sk_buff *skb, - int dif, int sdif) -{ -#ifdef CONFIG_TCP_MD5SIG - /* - * This gets called for each TCP segment that arrives - * so we want to be efficient. - * We have 3 drop cases: - * o No MD5 hash and one expected. - * o MD5 hash and we're not expecting one. - * o MD5 hash and its wrong. - */ - const __u8 *hash_location = NULL; - struct tcp_md5sig_key *hash_expected; - const struct iphdr *iph = ip_hdr(skb); - const struct tcphdr *th = tcp_hdr(skb); - const union tcp_md5_addr *addr; - unsigned char newhash[16]; - int genhash, l3index; - - /* sdif set, means packet ingressed via a device - * in an L3 domain and dif is set to the l3mdev - */ - l3index = sdif ? dif : 0; - - addr = (union tcp_md5_addr *)&iph->saddr; - hash_expected = tcp_md5_do_lookup(sk, l3index, addr, AF_INET); - hash_location = tcp_parse_md5sig_option(th); - - /* We've parsed the options - do we have a hash? */ - if (!hash_expected && !hash_location) - return false; - - if (hash_expected && !hash_location) { - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND); - return true; - } - - if (!hash_expected && hash_location) { - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED); - return true; - } - - /* Okay, so this is hash_expected and hash_location - - * so we need to calculate the checksum. - */ - genhash = tcp_v4_md5_hash_skb(newhash, - hash_expected, - NULL, skb); - - if (genhash || memcmp(hash_location, newhash, 16) != 0) { - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE); - net_info_ratelimited("MD5 Hash failed for (%pI4, %d)->(%pI4, %d)%s L3 index %d\n", - &iph->saddr, ntohs(th->source), - &iph->daddr, ntohs(th->dest), - genhash ? " tcp_v4_calc_md5_hash failed" - : "", l3index); - return true; - } - return false; -#endif - return false; -} - static void tcp_v4_init_req(struct request_sock *req, const struct sock *sk_listener, struct sk_buff *skb) @@ -1410,9 +1443,15 @@ static void tcp_v4_init_req(struct request_sock *req, } static struct dst_entry *tcp_v4_route_req(const struct sock *sk, + struct sk_buff *skb, struct flowi *fl, - const struct request_sock *req) + struct request_sock *req) { + tcp_v4_init_req(req, sk, skb); + + if (security_inet_conn_request(sk, skb, req)) + return NULL; + return inet_csk_route_req(sk, &fl->u.ip4, req); } @@ -1432,7 +1471,6 @@ const struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = { .req_md5_lookup = tcp_v4_md5_lookup, .calc_md5_hash = tcp_v4_md5_hash_skb, #endif - .init_req = tcp_v4_init_req, #ifdef CONFIG_SYN_COOKIES .cookie_init_seq = cookie_v4_init_sequence, #endif @@ -1469,6 +1507,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, bool *own_req) { struct inet_request_sock *ireq; + bool found_dup_sk = false; struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; @@ -1504,7 +1543,13 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, inet_csk(newsk)->icsk_ext_hdr_len = 0; if (inet_opt) inet_csk(newsk)->icsk_ext_hdr_len = inet_opt->opt.optlen; - newinet->inet_id = prandom_u32(); + newinet->inet_id = get_random_u16(); + + /* Set ToS of the new socket based upon the value of incoming SYN. + * ECT bits are set later in tcp_init_transfer(). + */ + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reflect_tos)) + newinet->tos = tcp_rsk(req)->syn_tos & ~INET_ECN_MASK; if (!dst) { dst = inet_csk_route_child_sock(sk, newsk, req); @@ -1534,20 +1579,30 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, * memory, then we end up not copying the key * across. Shucks. */ - tcp_md5_do_add(newsk, addr, AF_INET, 32, l3index, + tcp_md5_do_add(newsk, addr, AF_INET, 32, l3index, key->flags, key->key, key->keylen, GFP_ATOMIC); - sk_nocaps_add(newsk, NETIF_F_GSO_MASK); + sk_gso_disable(newsk); } #endif if (__inet_inherit_port(sk, newsk) < 0) goto put_and_exit; - *own_req = inet_ehash_nolisten(newsk, req_to_sk(req_unhash)); + *own_req = inet_ehash_nolisten(newsk, req_to_sk(req_unhash), + &found_dup_sk); if (likely(*own_req)) { tcp_move_syn(newtp, req); ireq->ireq_opt = NULL; } else { newinet->inet_opt = NULL; + + if (!req_unhash && found_dup_sk) { + /* This code path should only be executed in the + * syncookie case only + */ + bh_unlock_sock(newsk); + sock_put(newsk); + newsk = NULL; + } } return newsk; @@ -1592,6 +1647,8 @@ u16 tcp_v4_get_syncookie(struct sock *sk, struct iphdr *iph, return mss; } +INDIRECT_CALLABLE_DECLARE(struct dst_entry *ipv4_dst_check(struct dst_entry *, + u32)); /* The socket must have it's spinlock held when we get * here, unless it is a TCP_LISTEN socket. * @@ -1602,24 +1659,30 @@ u16 tcp_v4_get_syncookie(struct sock *sk, struct iphdr *iph, */ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) { + enum skb_drop_reason reason; struct sock *rsk; if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */ - struct dst_entry *dst = sk->sk_rx_dst; + struct dst_entry *dst; + + dst = rcu_dereference_protected(sk->sk_rx_dst, + lockdep_sock_is_held(sk)); sock_rps_save_rxhash(sk, skb); sk_mark_napi_id(sk, skb); if (dst) { - if (inet_sk(sk)->rx_dst_ifindex != skb->skb_iif || - !dst->ops->check(dst, 0)) { + if (sk->sk_rx_dst_ifindex != skb->skb_iif || + !INDIRECT_CALL_1(dst->ops->check, ipv4_dst_check, + dst, 0)) { + RCU_INIT_POINTER(sk->sk_rx_dst, NULL); dst_release(dst); - sk->sk_rx_dst = NULL; } } tcp_rcv_established(sk, skb); return 0; } + reason = SKB_DROP_REASON_NOT_SPECIFIED; if (tcp_checksum_complete(skb)) goto csum_err; @@ -1647,7 +1710,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) reset: tcp_v4_send_reset(rsk, skb); discard: - kfree_skb(skb); + kfree_skb_reason(skb, reason); /* Be careful here. If this function gets more complicated and * gcc suffers from register pressure on the x86, sk (in %ebx) * might be destroyed here. This current version compiles correctly, @@ -1656,6 +1719,8 @@ discard: return 0; csum_err: + reason = SKB_DROP_REASON_TCP_CSUM; + trace_tcp_bad_csum(skb); TCP_INC_STATS(sock_net(sk), TCP_MIB_CSUMERRORS); TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS); goto discard; @@ -1664,6 +1729,7 @@ EXPORT_SYMBOL(tcp_v4_do_rcv); int tcp_v4_early_demux(struct sk_buff *skb) { + struct net *net = dev_net(skb->dev); const struct iphdr *iph; const struct tcphdr *th; struct sock *sk; @@ -1680,7 +1746,7 @@ int tcp_v4_early_demux(struct sk_buff *skb) if (th->doff < sizeof(struct tcphdr) / 4) return 0; - sk = __inet_lookup_established(dev_net(skb->dev), &tcp_hashinfo, + sk = __inet_lookup_established(net, net->ipv4.tcp_death_row.hashinfo, iph->saddr, th->source, iph->daddr, ntohs(th->dest), skb->skb_iif, inet_sdif(skb)); @@ -1688,21 +1754,22 @@ int tcp_v4_early_demux(struct sk_buff *skb) skb->sk = sk; skb->destructor = sock_edemux; if (sk_fullsock(sk)) { - struct dst_entry *dst = READ_ONCE(sk->sk_rx_dst); + struct dst_entry *dst = rcu_dereference(sk->sk_rx_dst); if (dst) dst = dst_check(dst, 0); if (dst && - inet_sk(sk)->rx_dst_ifindex == skb->skb_iif) + sk->sk_rx_dst_ifindex == skb->skb_iif) skb_dst_set_noref(skb, dst); } } return 0; } -bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) +bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb, + enum skb_drop_reason *reason) { - u32 limit = READ_ONCE(sk->sk_rcvbuf) + READ_ONCE(sk->sk_sndbuf); + u32 limit, tail_gso_size, tail_gso_segs; struct skb_shared_info *shinfo; const struct tcphdr *th; struct tcphdr *thtail; @@ -1710,6 +1777,7 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) unsigned int hdrlen; bool fragstolen; u32 gso_segs; + u32 gso_size; int delta; /* In case all data was pulled from skb frags (in __pskb_pull_tail()), @@ -1724,6 +1792,8 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) if (unlikely(tcp_checksum_complete(skb))) { bh_unlock_sock(sk); + trace_tcp_bad_csum(skb); + *reason = SKB_DROP_REASON_TCP_CSUM; __TCP_INC_STATS(sock_net(sk), TCP_MIB_CSUMERRORS); __TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS); return true; @@ -1735,13 +1805,6 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) */ th = (const struct tcphdr *)skb->data; hdrlen = th->doff * 4; - shinfo = skb_shinfo(skb); - - if (!shinfo->gso_size) - shinfo->gso_size = skb->len - hdrlen; - - if (!shinfo->gso_segs) - shinfo->gso_segs = 1; tail = sk->sk_backlog.tail; if (!tail) @@ -1764,13 +1827,22 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) goto no_coalesce; __skb_pull(skb, hdrlen); - if (skb_try_coalesce(tail, skb, &fragstolen, &delta)) { - thtail->window = th->window; + shinfo = skb_shinfo(skb); + gso_size = shinfo->gso_size ?: skb->len; + gso_segs = shinfo->gso_segs ?: 1; + + shinfo = skb_shinfo(tail); + tail_gso_size = shinfo->gso_size ?: (tail->len - hdrlen); + tail_gso_segs = shinfo->gso_segs ?: 1; + + if (skb_try_coalesce(tail, skb, &fragstolen, &delta)) { TCP_SKB_CB(tail)->end_seq = TCP_SKB_CB(skb)->end_seq; - if (after(TCP_SKB_CB(skb)->ack_seq, TCP_SKB_CB(tail)->ack_seq)) + if (likely(!before(TCP_SKB_CB(skb)->ack_seq, TCP_SKB_CB(tail)->ack_seq))) { TCP_SKB_CB(tail)->ack_seq = TCP_SKB_CB(skb)->ack_seq; + thtail->window = th->window; + } /* We have to update both TCP_SKB_CB(tail)->tcp_flags and * thtail->fin, so that the fast path in tcp_rcv_established() @@ -1790,11 +1862,8 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) } /* Not as strict as GRO. We only need to carry mss max value */ - skb_shinfo(tail)->gso_size = max(shinfo->gso_size, - skb_shinfo(tail)->gso_size); - - gso_segs = skb_shinfo(tail)->gso_segs + shinfo->gso_segs; - skb_shinfo(tail)->gso_segs = min_t(u32, gso_segs, 0xFFFF); + shinfo->gso_size = max(gso_size, tail_gso_size); + shinfo->gso_segs = min_t(u32, gso_segs + tail_gso_segs, 0xFFFF); sk->sk_backlog.len += delta; __NET_INC_STATS(sock_net(sk), @@ -1805,14 +1874,17 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) __skb_push(skb, hdrlen); no_coalesce: + limit = (u32)READ_ONCE(sk->sk_rcvbuf) + (u32)(READ_ONCE(sk->sk_sndbuf) >> 1); + /* Only socket owner can try to collapse/prune rx queues * to reduce memory overhead, so add a little headroom here. * Few sockets backlog are possibly concurrently non empty. */ - limit += 64*1024; + limit += 64 * 1024; if (unlikely(sk_add_backlog(sk, skb, limit))) { bh_unlock_sock(sk); + *reason = SKB_DROP_REASON_SOCKET_BACKLOG; __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPBACKLOGDROP); return true; } @@ -1863,7 +1935,7 @@ static void tcp_v4_fill_cb(struct sk_buff *skb, const struct iphdr *iph, int tcp_v4_rcv(struct sk_buff *skb) { struct net *net = dev_net(skb->dev); - struct sk_buff *skb_to_free; + enum skb_drop_reason drop_reason; int sdif = inet_sdif(skb); int dif = inet_iif(skb); const struct iphdr *iph; @@ -1872,6 +1944,7 @@ int tcp_v4_rcv(struct sk_buff *skb) struct sock *sk; int ret; + drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; if (skb->pkt_type != PACKET_HOST) goto discard_it; @@ -1883,8 +1956,10 @@ int tcp_v4_rcv(struct sk_buff *skb) th = (const struct tcphdr *)skb->data; - if (unlikely(th->doff < sizeof(struct tcphdr) / 4)) + if (unlikely(th->doff < sizeof(struct tcphdr) / 4)) { + drop_reason = SKB_DROP_REASON_PKT_TOO_SMALL; goto bad_packet; + } if (!pskb_may_pull(skb, th->doff * 4)) goto discard_it; @@ -1899,7 +1974,8 @@ int tcp_v4_rcv(struct sk_buff *skb) th = (const struct tcphdr *)skb->data; iph = ip_hdr(skb); lookup: - sk = __inet_lookup_skb(&tcp_hashinfo, skb, __tcp_hdrlen(th), th->source, + sk = __inet_lookup_skb(net->ipv4.tcp_death_row.hashinfo, + skb, __tcp_hdrlen(th), th->source, th->dest, sdif, &refcounted); if (!sk) goto no_tcp_socket; @@ -1914,7 +1990,13 @@ process: struct sock *nsk; sk = req->rsk_listener; - if (unlikely(tcp_v4_inbound_md5_hash(sk, skb, dif, sdif))) { + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) + drop_reason = SKB_DROP_REASON_XFRM_POLICY; + else + drop_reason = tcp_inbound_md5_hash(sk, skb, + &iph->saddr, &iph->daddr, + AF_INET, dif, sdif); + if (unlikely(drop_reason)) { sk_drops_add(sk, skb); reqsk_put(req); goto discard_it; @@ -1924,13 +2006,21 @@ process: goto csum_error; } if (unlikely(sk->sk_state != TCP_LISTEN)) { - inet_csk_reqsk_queue_drop_and_put(sk, req); - goto lookup; + nsk = reuseport_migrate_sock(sk, req_to_sk(req), skb); + if (!nsk) { + inet_csk_reqsk_queue_drop_and_put(sk, req); + goto lookup; + } + sk = nsk; + /* reuseport_migrate_sock() has already held one sk_refcnt + * before returning. + */ + } else { + /* We own a reference on the listener, increase it again + * as we might lose it too soon. + */ + sock_hold(sk); } - /* We own a reference on the listener, increase it again - * as we might lose it too soon. - */ - sock_hold(sk); refcounted = true; nsk = NULL; if (!tcp_filter(sk, skb)) { @@ -1938,6 +2028,8 @@ process: iph = ip_hdr(skb); tcp_v4_fill_cb(skb, iph, th); nsk = tcp_check_req(sk, skb, req, false, &req_stolen); + } else { + drop_reason = SKB_DROP_REASON_SOCKET_FILTER; } if (!nsk) { reqsk_put(req); @@ -1953,6 +2045,7 @@ process: } goto discard_and_relse; } + nf_reset_ct(skb); if (nsk == sk) { reqsk_put(req); tcp_v4_restore_cb(skb); @@ -1964,21 +2057,31 @@ process: return 0; } } - if (unlikely(iph->ttl < inet_sk(sk)->min_ttl)) { - __NET_INC_STATS(net, LINUX_MIB_TCPMINTTLDROP); - goto discard_and_relse; + + if (static_branch_unlikely(&ip4_min_ttl)) { + /* min_ttl can be changed concurrently from do_ip_setsockopt() */ + if (unlikely(iph->ttl < READ_ONCE(inet_sk(sk)->min_ttl))) { + __NET_INC_STATS(net, LINUX_MIB_TCPMINTTLDROP); + goto discard_and_relse; + } } - if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) { + drop_reason = SKB_DROP_REASON_XFRM_POLICY; goto discard_and_relse; + } - if (tcp_v4_inbound_md5_hash(sk, skb, dif, sdif)) + drop_reason = tcp_inbound_md5_hash(sk, skb, &iph->saddr, + &iph->daddr, AF_INET, dif, sdif); + if (drop_reason) goto discard_and_relse; nf_reset_ct(skb); - if (tcp_filter(sk, skb)) + if (tcp_filter(sk, skb)) { + drop_reason = SKB_DROP_REASON_SOCKET_FILTER; goto discard_and_relse; + } th = (const struct tcphdr *)skb->data; iph = ip_hdr(skb); tcp_v4_fill_cb(skb, iph, th); @@ -1996,17 +2099,12 @@ process: tcp_segs_in(tcp_sk(sk), skb); ret = 0; if (!sock_owned_by_user(sk)) { - skb_to_free = sk->sk_rx_skb_cache; - sk->sk_rx_skb_cache = NULL; ret = tcp_v4_do_rcv(sk, skb); } else { - if (tcp_add_backlog(sk, skb)) + if (tcp_add_backlog(sk, skb, &drop_reason)) goto discard_and_relse; - skb_to_free = NULL; } bh_unlock_sock(sk); - if (skb_to_free) - __kfree_skb(skb_to_free); put_and_return: if (refcounted) @@ -2015,6 +2113,7 @@ put_and_return: return ret; no_tcp_socket: + drop_reason = SKB_DROP_REASON_NO_SOCKET; if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard_it; @@ -2022,6 +2121,8 @@ no_tcp_socket: if (tcp_checksum_complete(skb)) { csum_error: + drop_reason = SKB_DROP_REASON_TCP_CSUM; + trace_tcp_bad_csum(skb); __TCP_INC_STATS(net, TCP_MIB_CSUMERRORS); bad_packet: __TCP_INC_STATS(net, TCP_MIB_INERRS); @@ -2030,8 +2131,9 @@ bad_packet: } discard_it: + SKB_DR_OR(drop_reason, NOT_SPECIFIED); /* Discard frame. */ - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); return 0; discard_and_relse: @@ -2042,6 +2144,7 @@ discard_and_relse: do_time_wait: if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { + drop_reason = SKB_DROP_REASON_XFRM_POLICY; inet_twsk_put(inet_twsk(sk)); goto discard_it; } @@ -2054,9 +2157,9 @@ do_time_wait: } switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) { case TCP_TW_SYN: { - struct sock *sk2 = inet_lookup_listener(dev_net(skb->dev), - &tcp_hashinfo, skb, - __tcp_hdrlen(th), + struct sock *sk2 = inet_lookup_listener(net, + net->ipv4.tcp_death_row.hashinfo, + skb, __tcp_hdrlen(th), iph->saddr, th->source, iph->daddr, th->dest, inet_iif(skb), @@ -2070,7 +2173,7 @@ do_time_wait: } } /* to ACK */ - /* fall through */ + fallthrough; case TCP_TW_ACK: tcp_v4_timewait_ack(sk, skb); break; @@ -2094,8 +2197,8 @@ void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) struct dst_entry *dst = skb_dst(skb); if (dst && dst_hold_safe(dst)) { - sk->sk_rx_dst = dst; - inet_sk(sk)->rx_dst_ifindex = skb->skb_iif; + rcu_assign_pointer(sk->sk_rx_dst, dst); + sk->sk_rx_dst_ifindex = skb->skb_iif; } } EXPORT_SYMBOL(inet_sk_rx_dst_set); @@ -2112,10 +2215,6 @@ const struct inet_connection_sock_af_ops ipv4_specific = { .getsockopt = ip_getsockopt, .addr2sockaddr = inet_csk_addr2sockaddr, .sockaddr_len = sizeof(struct sockaddr_in), -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_ip_setsockopt, - .compat_getsockopt = compat_ip_getsockopt, -#endif .mtu_reduced = tcp_v4_mtu_reduced, }; EXPORT_SYMBOL(ipv4_specific); @@ -2194,45 +2293,73 @@ EXPORT_SYMBOL(tcp_v4_destroy_sock); #ifdef CONFIG_PROC_FS /* Proc filesystem TCP sock list dumping. */ -/* - * Get next listener socket follow cur. If cur is NULL, get first socket - * starting from bucket given in st->bucket; when st->bucket is zero the - * very first socket in the hash table is returned. +static unsigned short seq_file_family(const struct seq_file *seq); + +static bool seq_sk_match(struct seq_file *seq, const struct sock *sk) +{ + unsigned short family = seq_file_family(seq); + + /* AF_UNSPEC is used as a match all */ + return ((family == AF_UNSPEC || family == sk->sk_family) && + net_eq(sock_net(sk), seq_file_net(seq))); +} + +/* Find a non empty bucket (starting from st->bucket) + * and return the first sk from it. + */ +static void *listening_get_first(struct seq_file *seq) +{ + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; + struct tcp_iter_state *st = seq->private; + + st->offset = 0; + for (; st->bucket <= hinfo->lhash2_mask; st->bucket++) { + struct inet_listen_hashbucket *ilb2; + struct hlist_nulls_node *node; + struct sock *sk; + + ilb2 = &hinfo->lhash2[st->bucket]; + if (hlist_nulls_empty(&ilb2->nulls_head)) + continue; + + spin_lock(&ilb2->lock); + sk_nulls_for_each(sk, node, &ilb2->nulls_head) { + if (seq_sk_match(seq, sk)) + return sk; + } + spin_unlock(&ilb2->lock); + } + + return NULL; +} + +/* Find the next sk of "cur" within the same bucket (i.e. st->bucket). + * If "cur" is the last one in the st->bucket, + * call listening_get_first() to return the first sk of the next + * non empty bucket. */ static void *listening_get_next(struct seq_file *seq, void *cur) { - struct tcp_seq_afinfo *afinfo = PDE_DATA(file_inode(seq->file)); struct tcp_iter_state *st = seq->private; - struct net *net = seq_file_net(seq); - struct inet_listen_hashbucket *ilb; + struct inet_listen_hashbucket *ilb2; struct hlist_nulls_node *node; + struct inet_hashinfo *hinfo; struct sock *sk = cur; - if (!sk) { -get_head: - ilb = &tcp_hashinfo.listening_hash[st->bucket]; - spin_lock(&ilb->lock); - sk = sk_nulls_head(&ilb->nulls_head); - st->offset = 0; - goto get_sk; - } - ilb = &tcp_hashinfo.listening_hash[st->bucket]; ++st->num; ++st->offset; sk = sk_nulls_next(sk); -get_sk: sk_nulls_for_each_from(sk, node) { - if (!net_eq(sock_net(sk), net)) - continue; - if (sk->sk_family == afinfo->family) + if (seq_sk_match(seq, sk)) return sk; } - spin_unlock(&ilb->lock); - st->offset = 0; - if (++st->bucket < INET_LHTABLE_SIZE) - goto get_head; - return NULL; + + hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; + ilb2 = &hinfo->lhash2[st->bucket]; + spin_unlock(&ilb2->lock); + ++st->bucket; + return listening_get_first(seq); } static void *listening_get_idx(struct seq_file *seq, loff_t *pos) @@ -2242,7 +2369,7 @@ static void *listening_get_idx(struct seq_file *seq, loff_t *pos) st->bucket = 0; st->offset = 0; - rc = listening_get_next(seq, NULL); + rc = listening_get_first(seq); while (rc && *pos) { rc = listening_get_next(seq, rc); @@ -2251,9 +2378,10 @@ static void *listening_get_idx(struct seq_file *seq, loff_t *pos) return rc; } -static inline bool empty_bucket(const struct tcp_iter_state *st) +static inline bool empty_bucket(struct inet_hashinfo *hinfo, + const struct tcp_iter_state *st) { - return hlist_nulls_empty(&tcp_hashinfo.ehash[st->bucket].chain); + return hlist_nulls_empty(&hinfo->ehash[st->bucket].chain); } /* @@ -2262,43 +2390,36 @@ static inline bool empty_bucket(const struct tcp_iter_state *st) */ static void *established_get_first(struct seq_file *seq) { - struct tcp_seq_afinfo *afinfo = PDE_DATA(file_inode(seq->file)); + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; struct tcp_iter_state *st = seq->private; - struct net *net = seq_file_net(seq); - void *rc = NULL; st->offset = 0; - for (; st->bucket <= tcp_hashinfo.ehash_mask; ++st->bucket) { + for (; st->bucket <= hinfo->ehash_mask; ++st->bucket) { struct sock *sk; struct hlist_nulls_node *node; - spinlock_t *lock = inet_ehash_lockp(&tcp_hashinfo, st->bucket); + spinlock_t *lock = inet_ehash_lockp(hinfo, st->bucket); /* Lockless fast path for the common case of empty buckets */ - if (empty_bucket(st)) + if (empty_bucket(hinfo, st)) continue; spin_lock_bh(lock); - sk_nulls_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) { - if (sk->sk_family != afinfo->family || - !net_eq(sock_net(sk), net)) { - continue; - } - rc = sk; - goto out; + sk_nulls_for_each(sk, node, &hinfo->ehash[st->bucket].chain) { + if (seq_sk_match(seq, sk)) + return sk; } spin_unlock_bh(lock); } -out: - return rc; + + return NULL; } static void *established_get_next(struct seq_file *seq, void *cur) { - struct tcp_seq_afinfo *afinfo = PDE_DATA(file_inode(seq->file)); - struct sock *sk = cur; - struct hlist_nulls_node *node; + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; struct tcp_iter_state *st = seq->private; - struct net *net = seq_file_net(seq); + struct hlist_nulls_node *node; + struct sock *sk = cur; ++st->num; ++st->offset; @@ -2306,12 +2427,11 @@ static void *established_get_next(struct seq_file *seq, void *cur) sk = sk_nulls_next(sk); sk_nulls_for_each_from(sk, node) { - if (sk->sk_family == afinfo->family && - net_eq(sock_net(sk), net)) + if (seq_sk_match(seq, sk)) return sk; } - spin_unlock_bh(inet_ehash_lockp(&tcp_hashinfo, st->bucket)); + spin_unlock_bh(inet_ehash_lockp(hinfo, st->bucket)); ++st->bucket; return established_get_first(seq); } @@ -2349,29 +2469,31 @@ static void *tcp_get_idx(struct seq_file *seq, loff_t pos) static void *tcp_seek_last_pos(struct seq_file *seq) { + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; struct tcp_iter_state *st = seq->private; + int bucket = st->bucket; int offset = st->offset; int orig_num = st->num; void *rc = NULL; switch (st->state) { case TCP_SEQ_STATE_LISTENING: - if (st->bucket >= INET_LHTABLE_SIZE) + if (st->bucket > hinfo->lhash2_mask) break; st->state = TCP_SEQ_STATE_LISTENING; - rc = listening_get_next(seq, NULL); - while (offset-- && rc) + rc = listening_get_first(seq); + while (offset-- && rc && bucket == st->bucket) rc = listening_get_next(seq, rc); if (rc) break; st->bucket = 0; st->state = TCP_SEQ_STATE_ESTABLISHED; - /* Fallthrough */ + fallthrough; case TCP_SEQ_STATE_ESTABLISHED: - if (st->bucket > tcp_hashinfo.ehash_mask) + if (st->bucket > hinfo->ehash_mask) break; rc = established_get_first(seq); - while (offset-- && rc) + while (offset-- && rc && bucket == st->bucket) rc = established_get_next(seq, rc); } @@ -2436,16 +2558,17 @@ EXPORT_SYMBOL(tcp_seq_next); void tcp_seq_stop(struct seq_file *seq, void *v) { + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; struct tcp_iter_state *st = seq->private; switch (st->state) { case TCP_SEQ_STATE_LISTENING: if (v != SEQ_START_TOKEN) - spin_unlock(&tcp_hashinfo.listening_hash[st->bucket].lock); + spin_unlock(&hinfo->lhash2[st->bucket].lock); break; case TCP_SEQ_STATE_ESTABLISHED: if (v) - spin_unlock_bh(inet_ehash_lockp(&tcp_hashinfo, st->bucket)); + spin_unlock_bh(inet_ehash_lockp(hinfo, st->bucket)); break; } } @@ -2533,7 +2656,7 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) jiffies_to_clock_t(icsk->icsk_rto), jiffies_to_clock_t(icsk->icsk_ack.ato), (icsk->icsk_ack.quick << 1) | inet_csk_in_pingpong_mode(sk), - tp->snd_cwnd, + tcp_snd_cwnd(tp), state == TCP_LISTEN ? fastopenq->max_qlen : (tcp_in_initial_slowstart(tp) ? -1 : tp->snd_ssthresh)); @@ -2585,6 +2708,297 @@ out: return 0; } +#ifdef CONFIG_BPF_SYSCALL +struct bpf_tcp_iter_state { + struct tcp_iter_state state; + unsigned int cur_sk; + unsigned int end_sk; + unsigned int max_sk; + struct sock **batch; + bool st_bucket_done; +}; + +struct bpf_iter__tcp { + __bpf_md_ptr(struct bpf_iter_meta *, meta); + __bpf_md_ptr(struct sock_common *, sk_common); + uid_t uid __aligned(8); +}; + +static int tcp_prog_seq_show(struct bpf_prog *prog, struct bpf_iter_meta *meta, + struct sock_common *sk_common, uid_t uid) +{ + struct bpf_iter__tcp ctx; + + meta->seq_num--; /* skip SEQ_START_TOKEN */ + ctx.meta = meta; + ctx.sk_common = sk_common; + ctx.uid = uid; + return bpf_iter_run_prog(prog, &ctx); +} + +static void bpf_iter_tcp_put_batch(struct bpf_tcp_iter_state *iter) +{ + while (iter->cur_sk < iter->end_sk) + sock_put(iter->batch[iter->cur_sk++]); +} + +static int bpf_iter_tcp_realloc_batch(struct bpf_tcp_iter_state *iter, + unsigned int new_batch_sz) +{ + struct sock **new_batch; + + new_batch = kvmalloc(sizeof(*new_batch) * new_batch_sz, + GFP_USER | __GFP_NOWARN); + if (!new_batch) + return -ENOMEM; + + bpf_iter_tcp_put_batch(iter); + kvfree(iter->batch); + iter->batch = new_batch; + iter->max_sk = new_batch_sz; + + return 0; +} + +static unsigned int bpf_iter_tcp_listening_batch(struct seq_file *seq, + struct sock *start_sk) +{ + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; + struct bpf_tcp_iter_state *iter = seq->private; + struct tcp_iter_state *st = &iter->state; + struct hlist_nulls_node *node; + unsigned int expected = 1; + struct sock *sk; + + sock_hold(start_sk); + iter->batch[iter->end_sk++] = start_sk; + + sk = sk_nulls_next(start_sk); + sk_nulls_for_each_from(sk, node) { + if (seq_sk_match(seq, sk)) { + if (iter->end_sk < iter->max_sk) { + sock_hold(sk); + iter->batch[iter->end_sk++] = sk; + } + expected++; + } + } + spin_unlock(&hinfo->lhash2[st->bucket].lock); + + return expected; +} + +static unsigned int bpf_iter_tcp_established_batch(struct seq_file *seq, + struct sock *start_sk) +{ + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; + struct bpf_tcp_iter_state *iter = seq->private; + struct tcp_iter_state *st = &iter->state; + struct hlist_nulls_node *node; + unsigned int expected = 1; + struct sock *sk; + + sock_hold(start_sk); + iter->batch[iter->end_sk++] = start_sk; + + sk = sk_nulls_next(start_sk); + sk_nulls_for_each_from(sk, node) { + if (seq_sk_match(seq, sk)) { + if (iter->end_sk < iter->max_sk) { + sock_hold(sk); + iter->batch[iter->end_sk++] = sk; + } + expected++; + } + } + spin_unlock_bh(inet_ehash_lockp(hinfo, st->bucket)); + + return expected; +} + +static struct sock *bpf_iter_tcp_batch(struct seq_file *seq) +{ + struct inet_hashinfo *hinfo = seq_file_net(seq)->ipv4.tcp_death_row.hashinfo; + struct bpf_tcp_iter_state *iter = seq->private; + struct tcp_iter_state *st = &iter->state; + unsigned int expected; + bool resized = false; + struct sock *sk; + + /* The st->bucket is done. Directly advance to the next + * bucket instead of having the tcp_seek_last_pos() to skip + * one by one in the current bucket and eventually find out + * it has to advance to the next bucket. + */ + if (iter->st_bucket_done) { + st->offset = 0; + st->bucket++; + if (st->state == TCP_SEQ_STATE_LISTENING && + st->bucket > hinfo->lhash2_mask) { + st->state = TCP_SEQ_STATE_ESTABLISHED; + st->bucket = 0; + } + } + +again: + /* Get a new batch */ + iter->cur_sk = 0; + iter->end_sk = 0; + iter->st_bucket_done = false; + + sk = tcp_seek_last_pos(seq); + if (!sk) + return NULL; /* Done */ + + if (st->state == TCP_SEQ_STATE_LISTENING) + expected = bpf_iter_tcp_listening_batch(seq, sk); + else + expected = bpf_iter_tcp_established_batch(seq, sk); + + if (iter->end_sk == expected) { + iter->st_bucket_done = true; + return sk; + } + + if (!resized && !bpf_iter_tcp_realloc_batch(iter, expected * 3 / 2)) { + resized = true; + goto again; + } + + return sk; +} + +static void *bpf_iter_tcp_seq_start(struct seq_file *seq, loff_t *pos) +{ + /* bpf iter does not support lseek, so it always + * continue from where it was stop()-ped. + */ + if (*pos) + return bpf_iter_tcp_batch(seq); + + return SEQ_START_TOKEN; +} + +static void *bpf_iter_tcp_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct bpf_tcp_iter_state *iter = seq->private; + struct tcp_iter_state *st = &iter->state; + struct sock *sk; + + /* Whenever seq_next() is called, the iter->cur_sk is + * done with seq_show(), so advance to the next sk in + * the batch. + */ + if (iter->cur_sk < iter->end_sk) { + /* Keeping st->num consistent in tcp_iter_state. + * bpf_iter_tcp does not use st->num. + * meta.seq_num is used instead. + */ + st->num++; + /* Move st->offset to the next sk in the bucket such that + * the future start() will resume at st->offset in + * st->bucket. See tcp_seek_last_pos(). + */ + st->offset++; + sock_put(iter->batch[iter->cur_sk++]); + } + + if (iter->cur_sk < iter->end_sk) + sk = iter->batch[iter->cur_sk]; + else + sk = bpf_iter_tcp_batch(seq); + + ++*pos; + /* Keeping st->last_pos consistent in tcp_iter_state. + * bpf iter does not do lseek, so st->last_pos always equals to *pos. + */ + st->last_pos = *pos; + return sk; +} + +static int bpf_iter_tcp_seq_show(struct seq_file *seq, void *v) +{ + struct bpf_iter_meta meta; + struct bpf_prog *prog; + struct sock *sk = v; + bool slow; + uid_t uid; + int ret; + + if (v == SEQ_START_TOKEN) + return 0; + + if (sk_fullsock(sk)) + slow = lock_sock_fast(sk); + + if (unlikely(sk_unhashed(sk))) { + ret = SEQ_SKIP; + goto unlock; + } + + if (sk->sk_state == TCP_TIME_WAIT) { + uid = 0; + } else if (sk->sk_state == TCP_NEW_SYN_RECV) { + const struct request_sock *req = v; + + uid = from_kuid_munged(seq_user_ns(seq), + sock_i_uid(req->rsk_listener)); + } else { + uid = from_kuid_munged(seq_user_ns(seq), sock_i_uid(sk)); + } + + meta.seq = seq; + prog = bpf_iter_get_info(&meta, false); + ret = tcp_prog_seq_show(prog, &meta, v, uid); + +unlock: + if (sk_fullsock(sk)) + unlock_sock_fast(sk, slow); + return ret; + +} + +static void bpf_iter_tcp_seq_stop(struct seq_file *seq, void *v) +{ + struct bpf_tcp_iter_state *iter = seq->private; + struct bpf_iter_meta meta; + struct bpf_prog *prog; + + if (!v) { + meta.seq = seq; + prog = bpf_iter_get_info(&meta, true); + if (prog) + (void)tcp_prog_seq_show(prog, &meta, v, 0); + } + + if (iter->cur_sk < iter->end_sk) { + bpf_iter_tcp_put_batch(iter); + iter->st_bucket_done = false; + } +} + +static const struct seq_operations bpf_iter_tcp_seq_ops = { + .show = bpf_iter_tcp_seq_show, + .start = bpf_iter_tcp_seq_start, + .next = bpf_iter_tcp_seq_next, + .stop = bpf_iter_tcp_seq_stop, +}; +#endif +static unsigned short seq_file_family(const struct seq_file *seq) +{ + const struct tcp_seq_afinfo *afinfo; + +#ifdef CONFIG_BPF_SYSCALL + /* Iterated from bpf_iter. Let the bpf prog to filter instead. */ + if (seq->op == &bpf_iter_tcp_seq_ops) + return AF_UNSPEC; +#endif + + /* Iterated from proc fs */ + afinfo = pde_data(file_inode(seq->file)); + return afinfo->family; +} + static const struct seq_operations tcp4_seq_ops = { .show = tcp4_seq_show, .start = tcp_seq_start, @@ -2625,6 +3039,20 @@ void tcp4_proc_exit(void) } #endif /* CONFIG_PROC_FS */ +/* @wake is one when sk_stream_write_space() calls us. + * This sends EPOLLOUT only if notsent_bytes is half the limit. + * This mimics the strategy used in sock_def_write_space(). + */ +bool tcp_stream_memory_free(const struct sock *sk, int wake) +{ + const struct tcp_sock *tp = tcp_sk(sk); + u32 notsent_bytes = READ_ONCE(tp->write_seq) - + READ_ONCE(tp->snd_nxt); + + return (notsent_bytes << wake) < tcp_notsent_lowat(tp); +} +EXPORT_SYMBOL(tcp_stream_memory_free); + struct proto tcp_prot = { .name = "TCP", .owner = THIS_MODULE, @@ -2639,6 +3067,7 @@ struct proto tcp_prot = { .shutdown = tcp_shutdown, .setsockopt = tcp_setsockopt, .getsockopt = tcp_getsockopt, + .bpf_bypass_getsockopt = tcp_bpf_bypass_getsockopt, .keepalive = tcp_set_keepalive, .recvmsg = tcp_recvmsg, .sendmsg = tcp_sendmsg, @@ -2648,12 +3077,19 @@ struct proto tcp_prot = { .hash = inet_hash, .unhash = inet_unhash, .get_port = inet_csk_get_port, + .put_port = inet_put_port, +#ifdef CONFIG_BPF_SYSCALL + .psock_update_sk_prot = tcp_bpf_update_proto, +#endif .enter_memory_pressure = tcp_enter_memory_pressure, .leave_memory_pressure = tcp_leave_memory_pressure, .stream_memory_free = tcp_stream_memory_free, .sockets_allocated = &tcp_sockets_allocated, .orphan_count = &tcp_orphan_count, + .memory_allocated = &tcp_memory_allocated, + .per_cpu_fw_alloc = &tcp_memory_per_cpu_fw_alloc, + .memory_pressure = &tcp_memory_pressure, .sysctl_mem = sysctl_tcp_mem, .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_tcp_wmem), @@ -2663,54 +3099,51 @@ struct proto tcp_prot = { .slab_flags = SLAB_TYPESAFE_BY_RCU, .twsk_prot = &tcp_timewait_sock_ops, .rsk_prot = &tcp_request_sock_ops, - .h.hashinfo = &tcp_hashinfo, + .h.hashinfo = NULL, .no_autobind = true, -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_tcp_setsockopt, - .compat_getsockopt = compat_tcp_getsockopt, -#endif .diag_destroy = tcp_abort, }; EXPORT_SYMBOL(tcp_prot); static void __net_exit tcp_sk_exit(struct net *net) { - int cpu; - if (net->ipv4.tcp_congestion_control) bpf_module_put(net->ipv4.tcp_congestion_control, net->ipv4.tcp_congestion_control->owner); - - for_each_possible_cpu(cpu) - inet_ctl_sock_destroy(*per_cpu_ptr(net->ipv4.tcp_sk, cpu)); - free_percpu(net->ipv4.tcp_sk); } -static int __net_init tcp_sk_init(struct net *net) +static void __net_init tcp_set_hashinfo(struct net *net) { - int res, cpu, cnt; - - net->ipv4.tcp_sk = alloc_percpu(struct sock *); - if (!net->ipv4.tcp_sk) - return -ENOMEM; - - for_each_possible_cpu(cpu) { - struct sock *sk; + struct inet_hashinfo *hinfo; + unsigned int ehash_entries; + struct net *old_net; - res = inet_ctl_sock_create(&sk, PF_INET, SOCK_RAW, - IPPROTO_TCP, net); - if (res) - goto fail; - sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); + if (net_eq(net, &init_net)) + goto fallback; - /* Please enforce IP_DF and IPID==0 for RST and - * ACK sent in SYN-RECV and TIME-WAIT state. - */ - inet_sk(sk)->pmtudisc = IP_PMTUDISC_DO; + old_net = current->nsproxy->net_ns; + ehash_entries = READ_ONCE(old_net->ipv4.sysctl_tcp_child_ehash_entries); + if (!ehash_entries) + goto fallback; - *per_cpu_ptr(net->ipv4.tcp_sk, cpu) = sk; + ehash_entries = roundup_pow_of_two(ehash_entries); + hinfo = inet_pernet_hashinfo_alloc(&tcp_hashinfo, ehash_entries); + if (!hinfo) { + pr_warn("Failed to allocate TCP ehash (entries: %u) " + "for a netns, fallback to the global one\n", + ehash_entries); +fallback: + hinfo = &tcp_hashinfo; + ehash_entries = tcp_hashinfo.ehash_mask + 1; } + net->ipv4.tcp_death_row.hashinfo = hinfo; + net->ipv4.tcp_death_row.sysctl_max_tw_buckets = ehash_entries / 2; + net->ipv4.sysctl_max_syn_backlog = max(128U, ehash_entries / 128); +} + +static int __net_init tcp_sk_init(struct net *net) +{ net->ipv4.sysctl_tcp_ecn = 2; net->ipv4.sysctl_tcp_ecn_fallback = 1; @@ -2736,11 +3169,9 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_tw_reuse = 2; net->ipv4.sysctl_tcp_no_ssthresh_metrics_save = 1; - cnt = tcp_hashinfo.ehash_mask + 1; - net->ipv4.tcp_death_row.sysctl_max_tw_buckets = cnt / 2; - net->ipv4.tcp_death_row.hashinfo = &tcp_hashinfo; + refcount_set(&net->ipv4.tcp_death_row.tw_refcount, 1); + tcp_set_hashinfo(net); - net->ipv4.sysctl_max_syn_backlog = max(128, cnt / 128); net->ipv4.sysctl_tcp_sack = 1; net->ipv4.sysctl_tcp_window_scaling = 1; net->ipv4.sysctl_tcp_timestamps = 1; @@ -2761,9 +3192,12 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_tso_win_divisor = 3; /* Default TSQ limit of 16 TSO segments */ net->ipv4.sysctl_tcp_limit_output_bytes = 16 * 65536; - /* rfc5961 challenge ack rate limiting */ - net->ipv4.sysctl_tcp_challenge_ack_limit = 1000; + + /* rfc5961 challenge ack rate limiting, per net-ns, disabled by default. */ + net->ipv4.sysctl_tcp_challenge_ack_limit = INT_MAX; + net->ipv4.sysctl_tcp_min_tso_segs = 2; + net->ipv4.sysctl_tcp_tso_rtt_log = 9; /* 2^9 = 512 usec */ net->ipv4.sysctl_tcp_min_rtt_wlen = 300; net->ipv4.sysctl_tcp_autocorking = 1; net->ipv4.sysctl_tcp_invalid_ratelimit = HZ/2; @@ -2778,10 +3212,10 @@ static int __net_init tcp_sk_init(struct net *net) sizeof(init_net.ipv4.sysctl_tcp_wmem)); } net->ipv4.sysctl_tcp_comp_sack_delay_ns = NSEC_PER_MSEC; + net->ipv4.sysctl_tcp_comp_sack_slack_ns = 100 * NSEC_PER_USEC; net->ipv4.sysctl_tcp_comp_sack_nr = 44; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; - spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); - net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 60 * 60; + net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 0; atomic_set(&net->ipv4.tfo_active_disable_times, 0); /* Reno is always built in */ @@ -2793,20 +3227,19 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.tcp_congestion_control = &tcp_reno; return 0; -fail: - tcp_sk_exit(net); - - return res; } static void __net_exit tcp_sk_exit_batch(struct list_head *net_exit_list) { struct net *net; - inet_twsk_purge(&tcp_hashinfo, AF_INET); + tcp_twsk_purge(net_exit_list, AF_INET); - list_for_each_entry(net, net_exit_list, exit_list) + list_for_each_entry(net, net_exit_list, exit_list) { + inet_pernet_hashinfo_free(net->ipv4.tcp_death_row.hashinfo); + WARN_ON_ONCE(!refcount_dec_and_test(&net->ipv4.tcp_death_row.tw_refcount)); tcp_fastopen_ctx_destroy(net); + } } static struct pernet_operations __net_initdata tcp_sk_ops = { @@ -2815,8 +3248,103 @@ static struct pernet_operations __net_initdata tcp_sk_ops = { .exit_batch = tcp_sk_exit_batch, }; +#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) +DEFINE_BPF_ITER_FUNC(tcp, struct bpf_iter_meta *meta, + struct sock_common *sk_common, uid_t uid) + +#define INIT_BATCH_SZ 16 + +static int bpf_iter_init_tcp(void *priv_data, struct bpf_iter_aux_info *aux) +{ + struct bpf_tcp_iter_state *iter = priv_data; + int err; + + err = bpf_iter_init_seq_net(priv_data, aux); + if (err) + return err; + + err = bpf_iter_tcp_realloc_batch(iter, INIT_BATCH_SZ); + if (err) { + bpf_iter_fini_seq_net(priv_data); + return err; + } + + return 0; +} + +static void bpf_iter_fini_tcp(void *priv_data) +{ + struct bpf_tcp_iter_state *iter = priv_data; + + bpf_iter_fini_seq_net(priv_data); + kvfree(iter->batch); +} + +static const struct bpf_iter_seq_info tcp_seq_info = { + .seq_ops = &bpf_iter_tcp_seq_ops, + .init_seq_private = bpf_iter_init_tcp, + .fini_seq_private = bpf_iter_fini_tcp, + .seq_priv_size = sizeof(struct bpf_tcp_iter_state), +}; + +static const struct bpf_func_proto * +bpf_iter_tcp_get_func_proto(enum bpf_func_id func_id, + const struct bpf_prog *prog) +{ + switch (func_id) { + case BPF_FUNC_setsockopt: + return &bpf_sk_setsockopt_proto; + case BPF_FUNC_getsockopt: + return &bpf_sk_getsockopt_proto; + default: + return NULL; + } +} + +static struct bpf_iter_reg tcp_reg_info = { + .target = "tcp", + .ctx_arg_info_size = 1, + .ctx_arg_info = { + { offsetof(struct bpf_iter__tcp, sk_common), + PTR_TO_BTF_ID_OR_NULL }, + }, + .get_func_proto = bpf_iter_tcp_get_func_proto, + .seq_info = &tcp_seq_info, +}; + +static void __init bpf_iter_register(void) +{ + tcp_reg_info.ctx_arg_info[0].btf_id = btf_sock_ids[BTF_SOCK_TYPE_SOCK_COMMON]; + if (bpf_iter_reg_target(&tcp_reg_info)) + pr_warn("Warning: could not register bpf iterator tcp\n"); +} + +#endif + void __init tcp_v4_init(void) { + int cpu, res; + + for_each_possible_cpu(cpu) { + struct sock *sk; + + res = inet_ctl_sock_create(&sk, PF_INET, SOCK_RAW, + IPPROTO_TCP, &init_net); + if (res) + panic("Failed to create the TCP control socket.\n"); + sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); + + /* Please enforce IP_DF and IPID==0 for RST and + * ACK sent in SYN-RECV and TIME-WAIT state. + */ + inet_sk(sk)->pmtudisc = IP_PMTUDISC_DO; + + per_cpu(ipv4_tcp_sk, cpu) = sk; + } if (register_pernet_subsys(&tcp_sk_ops)) panic("Failed to create the TCP control socket.\n"); + +#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) + bpf_iter_register(); +#endif } diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c index 8c643a4ffad1..ae36780977d2 100644 --- a/net/ipv4/tcp_lp.c +++ b/net/ipv4/tcp_lp.c @@ -63,7 +63,7 @@ enum tcp_lp_state { * @sowd: smoothed OWD << 3 * @owd_min: min OWD * @owd_max: max OWD - * @owd_max_rsv: resrved max owd + * @owd_max_rsv: reserved max owd * @remote_hz: estimated remote HZ * @remote_ref_time: remote reference time * @local_ref_time: local reference time @@ -89,6 +89,7 @@ struct lp { /** * tcp_lp_init + * @sk: socket to initialize congestion control algorithm for * * Init all required variables. * Clone the handling from Vegas module implementation. @@ -111,6 +112,7 @@ static void tcp_lp_init(struct sock *sk) /** * tcp_lp_cong_avoid + * @sk: socket to avoid congesting * * Implementation of cong_avoid. * Will only call newReno CA when away from inference. @@ -126,6 +128,7 @@ static void tcp_lp_cong_avoid(struct sock *sk, u32 ack, u32 acked) /** * tcp_lp_remote_hz_estimator + * @sk: socket which needs an estimate for the remote HZs * * Estimate remote HZ. * We keep on updating the estimated value, where original TCP-LP @@ -176,6 +179,7 @@ static u32 tcp_lp_remote_hz_estimator(struct sock *sk) /** * tcp_lp_owd_calculator + * @sk: socket to calculate one way delay for * * Calculate one way delay (in relative format). * Original implement OWD as minus of remote time difference to local time @@ -210,6 +214,8 @@ static u32 tcp_lp_owd_calculator(struct sock *sk) /** * tcp_lp_rtt_sample + * @sk: socket to add a rtt sample to + * @rtt: round trip time, which is ignored! * * Implementation or rtt_sample. * Will take the following action, @@ -254,6 +260,7 @@ static void tcp_lp_rtt_sample(struct sock *sk, u32 rtt) /** * tcp_lp_pkts_acked + * @sk: socket requiring congestion avoidance calculations * * Implementation of pkts_acked. * Deal with active drop under Early Congestion Indication. @@ -290,7 +297,7 @@ static void tcp_lp_pkts_acked(struct sock *sk, const struct ack_sample *sample) lp->flag &= ~LP_WITHIN_THR; pr_debug("TCP-LP: %05o|%5u|%5u|%15u|%15u|%15u\n", lp->flag, - tp->snd_cwnd, lp->remote_hz, lp->owd_min, lp->owd_max, + tcp_snd_cwnd(tp), lp->remote_hz, lp->owd_min, lp->owd_max, lp->sowd >> 3); if (lp->flag & LP_WITHIN_THR) @@ -298,7 +305,7 @@ static void tcp_lp_pkts_acked(struct sock *sk, const struct ack_sample *sample) /* FIXME: try to reset owd_min and owd_max here * so decrease the chance the min/max is no longer suitable - * and will usually within threshold when whithin inference */ + * and will usually within threshold when within inference */ lp->owd_min = lp->sowd >> 3; lp->owd_max = lp->sowd >> 2; lp->owd_max_rsv = lp->sowd >> 2; @@ -306,12 +313,12 @@ static void tcp_lp_pkts_acked(struct sock *sk, const struct ack_sample *sample) /* happened within inference * drop snd_cwnd into 1 */ if (lp->flag & LP_WITHIN_INF) - tp->snd_cwnd = 1U; + tcp_snd_cwnd_set(tp, 1U); /* happened after inference * cut snd_cwnd into half */ else - tp->snd_cwnd = max(tp->snd_cwnd >> 1U, 1U); + tcp_snd_cwnd_set(tp, max(tcp_snd_cwnd(tp) >> 1U, 1U)); /* record this drop time */ lp->last_drop = now; diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index 279db8822439..82f4575f9cd9 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -329,7 +329,7 @@ void tcp_update_metrics(struct sock *sk) int m; sk_dst_confirm(sk); - if (net->ipv4.sysctl_tcp_nometrics_save || !dst) + if (READ_ONCE(net->ipv4.sysctl_tcp_nometrics_save) || !dst) return; rcu_read_lock(); @@ -385,29 +385,29 @@ void tcp_update_metrics(struct sock *sk) if (tcp_in_initial_slowstart(tp)) { /* Slow start still did not finish. */ - if (!net->ipv4.sysctl_tcp_no_ssthresh_metrics_save && + if (!READ_ONCE(net->ipv4.sysctl_tcp_no_ssthresh_metrics_save) && !tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) { val = tcp_metric_get(tm, TCP_METRIC_SSTHRESH); - if (val && (tp->snd_cwnd >> 1) > val) + if (val && (tcp_snd_cwnd(tp) >> 1) > val) tcp_metric_set(tm, TCP_METRIC_SSTHRESH, - tp->snd_cwnd >> 1); + tcp_snd_cwnd(tp) >> 1); } if (!tcp_metric_locked(tm, TCP_METRIC_CWND)) { val = tcp_metric_get(tm, TCP_METRIC_CWND); - if (tp->snd_cwnd > val) + if (tcp_snd_cwnd(tp) > val) tcp_metric_set(tm, TCP_METRIC_CWND, - tp->snd_cwnd); + tcp_snd_cwnd(tp)); } } else if (!tcp_in_slow_start(tp) && icsk->icsk_ca_state == TCP_CA_Open) { /* Cong. avoidance phase, cwnd is reliable. */ - if (!net->ipv4.sysctl_tcp_no_ssthresh_metrics_save && + if (!READ_ONCE(net->ipv4.sysctl_tcp_no_ssthresh_metrics_save) && !tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) tcp_metric_set(tm, TCP_METRIC_SSTHRESH, - max(tp->snd_cwnd >> 1, tp->snd_ssthresh)); + max(tcp_snd_cwnd(tp) >> 1, tp->snd_ssthresh)); if (!tcp_metric_locked(tm, TCP_METRIC_CWND)) { val = tcp_metric_get(tm, TCP_METRIC_CWND); - tcp_metric_set(tm, TCP_METRIC_CWND, (val + tp->snd_cwnd) >> 1); + tcp_metric_set(tm, TCP_METRIC_CWND, (val + tcp_snd_cwnd(tp)) >> 1); } } else { /* Else slow start did not finish, cwnd is non-sense, @@ -418,7 +418,7 @@ void tcp_update_metrics(struct sock *sk) tcp_metric_set(tm, TCP_METRIC_CWND, (val + tp->snd_ssthresh) >> 1); } - if (!net->ipv4.sysctl_tcp_no_ssthresh_metrics_save && + if (!READ_ONCE(net->ipv4.sysctl_tcp_no_ssthresh_metrics_save) && !tcp_metric_locked(tm, TCP_METRIC_SSTHRESH)) { val = tcp_metric_get(tm, TCP_METRIC_SSTHRESH); if (val && tp->snd_ssthresh > val) @@ -428,7 +428,8 @@ void tcp_update_metrics(struct sock *sk) if (!tcp_metric_locked(tm, TCP_METRIC_REORDERING)) { val = tcp_metric_get(tm, TCP_METRIC_REORDERING); if (val < tp->reordering && - tp->reordering != net->ipv4.sysctl_tcp_reordering) + tp->reordering != + READ_ONCE(net->ipv4.sysctl_tcp_reordering)) tcp_metric_set(tm, TCP_METRIC_REORDERING, tp->reordering); } @@ -462,7 +463,7 @@ void tcp_init_metrics(struct sock *sk) if (tcp_metric_locked(tm, TCP_METRIC_CWND)) tp->snd_cwnd_clamp = tcp_metric_get(tm, TCP_METRIC_CWND); - val = net->ipv4.sysctl_tcp_no_ssthresh_metrics_save ? + val = READ_ONCE(net->ipv4.sysctl_tcp_no_ssthresh_metrics_save) ? 0 : tcp_metric_get(tm, TCP_METRIC_SSTHRESH); if (val) { tp->snd_ssthresh = val; @@ -887,7 +888,7 @@ static void tcp_metrics_flush_all(struct net *net) pp = &hb->chain; for (tm = deref_locked(*pp); tm; tm = deref_locked(*pp)) { match = net ? net_eq(tm_net(tm), net) : - !refcount_read(&tm_net(tm)->count); + !refcount_read(&tm_net(tm)->ns.count); if (match) { *pp = tm->tcpm_next; kfree_rcu(tm, rcu_head); @@ -943,7 +944,7 @@ static int tcp_metrics_nl_cmd_del(struct sk_buff *skb, struct genl_info *info) return 0; } -static const struct genl_ops tcp_metrics_nl_ops[] = { +static const struct genl_small_ops tcp_metrics_nl_ops[] = { { .cmd = TCP_METRICS_CMD_GET, .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, @@ -966,8 +967,9 @@ static struct genl_family tcp_metrics_nl_family __ro_after_init = { .policy = tcp_metrics_nl_policy, .netnsok = true, .module = THIS_MODULE, - .ops = tcp_metrics_nl_ops, - .n_ops = ARRAY_SIZE(tcp_metrics_nl_ops), + .small_ops = tcp_metrics_nl_ops, + .n_small_ops = ARRAY_SIZE(tcp_metrics_nl_ops), + .resv_start_op = TCP_METRICS_CMD_DEL + 1, }; static unsigned int tcpmhash_entries; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index ad3b56d9fa71..c375f603a16c 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -19,14 +19,7 @@ * Jorge Cwik, <jorge@laser.satlink.net> */ -#include <linux/mm.h> -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/sysctl.h> -#include <linux/workqueue.h> -#include <linux/static_key.h> #include <net/tcp.h> -#include <net/inet_common.h> #include <net/xfrm.h> #include <net/busy_poll.h> @@ -180,7 +173,7 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb, * Oh well... nobody has a sufficient solution to this * protocol bug yet. */ - if (twsk_net(tw)->ipv4.sysctl_tcp_rfc1337 == 0) { + if (!READ_ONCE(twsk_net(tw)->ipv4.sysctl_tcp_rfc1337)) { kill: inet_twsk_deschedule_put(tw); return TCP_TW_SUCCESS; @@ -254,10 +247,10 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) { const struct inet_connection_sock *icsk = inet_csk(sk); const struct tcp_sock *tp = tcp_sk(sk); + struct net *net = sock_net(sk); struct inet_timewait_sock *tw; - struct inet_timewait_death_row *tcp_death_row = &sock_net(sk)->ipv4.tcp_death_row; - tw = inet_twsk_alloc(sk, tcp_death_row, state); + tw = inet_twsk_alloc(sk, &net->ipv4.tcp_death_row, state); if (tw) { struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw); @@ -326,14 +319,14 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) /* Linkage updates. * Note that access to tw after this point is illegal. */ - inet_twsk_hashdance(tw, sk, &tcp_hashinfo); + inet_twsk_hashdance(tw, sk, net->ipv4.tcp_death_row.hashinfo); local_bh_enable(); } else { /* Sorry, if we're out of memory, just CLOSE this * socket up. We've got bigger problems than * non-graceful socket closings. */ - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPTIMEWAITOVERFLOW); + NET_INC_STATS(net, LINUX_MIB_TCPTIMEWAITOVERFLOW); } tcp_update_metrics(sk); @@ -354,6 +347,27 @@ void tcp_twsk_destructor(struct sock *sk) } EXPORT_SYMBOL_GPL(tcp_twsk_destructor); +void tcp_twsk_purge(struct list_head *net_exit_list, int family) +{ + bool purged_once = false; + struct net *net; + + list_for_each_entry(net, net_exit_list, exit_list) { + if (net->ipv4.tcp_death_row.hashinfo->pernet) { + /* Even if tw_refcount == 1, we must clean up kernel reqsk */ + inet_twsk_purge(net->ipv4.tcp_death_row.hashinfo, family); + } else if (!purged_once) { + /* The last refcount is decremented in tcp_sk_exit_batch() */ + if (refcount_read(&net->ipv4.tcp_death_row.tw_refcount) == 1) + continue; + + inet_twsk_purge(&tcp_hashinfo, family); + purged_once = true; + } + } +} +EXPORT_SYMBOL_GPL(tcp_twsk_purge); + /* Warning : This function is called without sk_listener being locked. * Be sure to read socket fields once, as their value could change under us. */ @@ -538,7 +552,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->tsoffset = treq->ts_off; #ifdef CONFIG_TCP_MD5SIG newtp->md5sig_info = NULL; /*XXX*/ - if (newtp->af_specific->md5_lookup(sk, newsk)) + if (treq->af_specific->req_md5_lookup(sk, req_to_sk(req))) newtp->tcp_header_len += TCPOLEN_MD5SIG_ALIGNED; #endif if (skb->len >= TCP_MSS_DEFAULT + newtp->tcp_header_len) @@ -548,6 +562,9 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->fastopen_req = NULL; RCU_INIT_POINTER(newtp->fastopen_rsk, NULL); + newtp->bpf_chg_cc_inprogress = 0; + tcp_bpf_clone(sk, newsk); + __TCP_INC_STATS(sock_net(sk), TCP_MIB_PASSIVEOPENS); return newsk; @@ -588,7 +605,7 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, * it can be estimated (approximately) * from another data. */ - tmp_opt.ts_recent_stamp = ktime_get_seconds() - ((TCP_TIMEOUT_INIT/HZ)<<req->num_timeout); + tmp_opt.ts_recent_stamp = ktime_get_seconds() - reqsk_timeout(req, TCP_RTO_MAX) / HZ; paws_reject = tcp_paws_reject(&tmp_opt, th->rst); } } @@ -627,8 +644,7 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, !inet_rtx_syn_ack(sk, req)) { unsigned long expires = jiffies; - expires += min(TCP_TIMEOUT_INIT << req->num_timeout, - TCP_RTO_MAX); + expires += reqsk_timeout(req, TCP_RTO_MAX); if (!fastopen) mod_timer_pending(&req->rsk_timer, expires); else @@ -772,13 +788,22 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, if (!child) goto listen_overflow; + if (own_req && rsk_drop_req(req)) { + reqsk_queue_removed(&inet_csk(req->rsk_listener)->icsk_accept_queue, req); + inet_csk_reqsk_queue_drop_and_put(req->rsk_listener, req); + return child; + } + sock_rps_save_rxhash(child, skb); tcp_synack_rtt_meas(child, req); *req_stolen = !own_req; return inet_csk_complete_hashdance(sk, child, req, own_req); listen_overflow: - if (!sock_net(sk)->ipv4.sysctl_tcp_abort_on_overflow) { + if (sk != req->rsk_listener) + __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE); + + if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_abort_on_overflow)) { inet_rsk(req)->acked = 1; return NULL; } @@ -793,11 +818,14 @@ embryonic_reset: req->rsk_ops->send_reset(sk, skb); } else if (fastopen) { /* received a valid RST pkt */ reqsk_fastopen_remove(sk, req, true); - tcp_reset(sk); + tcp_reset(sk, skb); } if (!fastopen) { - inet_csk_reqsk_queue_drop(sk, req); - __NET_INC_STATS(sock_net(sk), LINUX_MIB_EMBRYONICRSTS); + bool unlinked = inet_csk_reqsk_queue_drop(sk, req); + + if (unlinked) + __NET_INC_STATS(sock_net(sk), LINUX_MIB_EMBRYONICRSTS); + *req_stolen = !unlinked; } return NULL; } @@ -817,12 +845,13 @@ EXPORT_SYMBOL(tcp_check_req); int tcp_child_process(struct sock *parent, struct sock *child, struct sk_buff *skb) + __releases(&((child)->sk_lock.slock)) { int ret = 0; int state = child->sk_state; - /* record NAPI ID of child */ - sk_mark_napi_id(child, skb); + /* record sk_napi_id and sk_rx_queue_mapping of child. */ + sk_mark_napi_id_set(child, skb); tcp_segs_in(tcp_sk(child), skb); if (!sock_owned_by_user(child)) { diff --git a/net/ipv4/tcp_nv.c b/net/ipv4/tcp_nv.c index 95db7a11ba2a..a60662f4bdf9 100644 --- a/net/ipv4/tcp_nv.c +++ b/net/ipv4/tcp_nv.c @@ -25,7 +25,6 @@ * 1) Add mechanism to deal with reverse congestion. */ -#include <linux/mm.h> #include <linux/module.h> #include <linux/math64.h> #include <net/tcp.h> @@ -198,10 +197,10 @@ static void tcpnv_cong_avoid(struct sock *sk, u32 ack, u32 acked) } if (ca->cwnd_growth_factor < 0) { - cnt = tp->snd_cwnd << -ca->cwnd_growth_factor; + cnt = tcp_snd_cwnd(tp) << -ca->cwnd_growth_factor; tcp_cong_avoid_ai(tp, cnt, acked); } else { - cnt = max(4U, tp->snd_cwnd >> ca->cwnd_growth_factor); + cnt = max(4U, tcp_snd_cwnd(tp) >> ca->cwnd_growth_factor); tcp_cong_avoid_ai(tp, cnt, acked); } } @@ -210,7 +209,7 @@ static u32 tcpnv_recalc_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - return max((tp->snd_cwnd * nv_loss_dec_factor) >> 10, 2U); + return max((tcp_snd_cwnd(tp) * nv_loss_dec_factor) >> 10, 2U); } static void tcpnv_state(struct sock *sk, u8 new_state) @@ -258,7 +257,7 @@ static void tcpnv_acked(struct sock *sk, const struct ack_sample *sample) return; /* Stop cwnd growth if we were in catch up mode */ - if (ca->nv_catchup && tp->snd_cwnd >= nv_min_cwnd) { + if (ca->nv_catchup && tcp_snd_cwnd(tp) >= nv_min_cwnd) { ca->nv_catchup = 0; ca->nv_allow_cwnd_growth = 0; } @@ -372,7 +371,7 @@ static void tcpnv_acked(struct sock *sk, const struct ack_sample *sample) * if cwnd < max_win, grow cwnd * else leave the same */ - if (tp->snd_cwnd > max_win) { + if (tcp_snd_cwnd(tp) > max_win) { /* there is congestion, check that it is ok * to make a CA decision * 1. We should have at least nv_dec_eval_min_calls @@ -399,20 +398,20 @@ static void tcpnv_acked(struct sock *sk, const struct ack_sample *sample) ca->nv_allow_cwnd_growth = 0; tp->snd_ssthresh = (nv_ssthresh_factor * max_win) >> 3; - if (tp->snd_cwnd - max_win > 2) { + if (tcp_snd_cwnd(tp) - max_win > 2) { /* gap > 2, we do exponential cwnd decrease */ int dec; - dec = max(2U, ((tp->snd_cwnd - max_win) * + dec = max(2U, ((tcp_snd_cwnd(tp) - max_win) * nv_cong_dec_mult) >> 7); - tp->snd_cwnd -= dec; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) - dec); } else if (nv_cong_dec_mult > 0) { - tp->snd_cwnd = max_win; + tcp_snd_cwnd_set(tp, max_win); } if (ca->cwnd_growth_factor > 0) ca->cwnd_growth_factor = 0; ca->nv_no_cong_cnt = 0; - } else if (tp->snd_cwnd <= max_win - nv_pad_buffer) { + } else if (tcp_snd_cwnd(tp) <= max_win - nv_pad_buffer) { /* There is no congestion, grow cwnd if allowed*/ if (ca->nv_eval_call_cnt < nv_inc_eval_min_calls) return; @@ -445,8 +444,8 @@ static void tcpnv_acked(struct sock *sk, const struct ack_sample *sample) * (it wasn't before, if it is now is because nv * decreased it). */ - if (tp->snd_cwnd < nv_min_cwnd) - tp->snd_cwnd = nv_min_cwnd; + if (tcp_snd_cwnd(tp) < nv_min_cwnd) + tcp_snd_cwnd_set(tp, nv_min_cwnd); } } diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index e09147ac9a99..45dda7889387 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -8,6 +8,7 @@ #include <linux/indirect_call_wrapper.h> #include <linux/skbuff.h> +#include <net/gro.h> #include <net/tcp.h> #include <net/protocol.h> @@ -194,12 +195,9 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb) off = skb_gro_offset(skb); hlen = off + sizeof(*th); - th = skb_gro_header_fast(skb, off); - if (skb_gro_header_hard(skb, hlen)) { - th = skb_gro_header_slow(skb, hlen, off); - if (unlikely(!th)) - goto out; - } + th = skb_gro_header(skb, hlen, off); + if (unlikely(!th)) + goto out; thlen = th->doff * 4; if (thlen < sizeof(*th)) @@ -257,7 +255,15 @@ found: mss = skb_shinfo(p)->gso_size; - flush |= (len - 1) >= mss; + /* If skb is a GRO packet, make sure its gso_size matches prior packet mss. + * If it is a single frame, do not aggregate it if its length + * is bigger than our mss. + */ + if (unlikely(skb_is_gso(skb))) + flush |= (mss != skb_shinfo(skb)->gso_size); + else + flush |= (len - 1) >= mss; + flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq); #ifdef CONFIG_TLS_DEVICE flush |= p->decrypted ^ skb->decrypted; @@ -271,7 +277,12 @@ found: tcp_flag_word(th2) |= flags & (TCP_FLAG_FIN | TCP_FLAG_PSH); out_check_final: - flush = len < mss; + /* Force a flush if last segment is smaller than mss. */ + if (unlikely(skb_is_gso(skb))) + flush = len != NAPI_GRO_CB(skb)->count * skb_shinfo(skb)->gso_size; + else + flush = len < mss; + flush |= (__force int)(flags & (TCP_FLAG_URG | TCP_FLAG_PSH | TCP_FLAG_RST | TCP_FLAG_SYN | TCP_FLAG_FIN)); @@ -298,6 +309,9 @@ int tcp_gro_complete(struct sk_buff *skb) if (th->cwr) skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN; + if (skb->encapsulation) + skb->inner_transport_header = skb->transport_header; + return 0; } EXPORT_SYMBOL(tcp_gro_complete); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 306e25d743e8..c69f4d966024 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -82,6 +82,7 @@ static void tcp_event_new_data_sent(struct sock *sk, struct sk_buff *skb) NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPORIGDATASENT, tcp_skb_pcount(skb)); + tcp_check_space(sk); } /* SND.NXT, if window was not shrunk or the amount of shrunk was less than one @@ -142,7 +143,7 @@ void tcp_cwnd_restart(struct sock *sk, s32 delta) { struct tcp_sock *tp = tcp_sk(sk); u32 restart_cwnd = tcp_init_cwnd(tp, __sk_dst_get(sk)); - u32 cwnd = tp->snd_cwnd; + u32 cwnd = tcp_snd_cwnd(tp); tcp_ca_event(sk, CA_EVENT_CWND_RESTART); @@ -151,7 +152,7 @@ void tcp_cwnd_restart(struct sock *sk, s32 delta) while ((delta -= inet_csk(sk)->icsk_rto) > 0 && cwnd > restart_cwnd) cwnd >>= 1; - tp->snd_cwnd = max(cwnd, restart_cwnd); + tcp_snd_cwnd_set(tp, max(cwnd, restart_cwnd)); tp->snd_cwnd_stamp = tcp_jiffies32; tp->snd_cwnd_used = 0; } @@ -166,16 +167,13 @@ static void tcp_event_data_sent(struct tcp_sock *tp, if (tcp_packets_in_flight(tp) == 0) tcp_ca_event(sk, CA_EVENT_TX_START); - /* If this is the first data packet sent in response to the - * previous received data, - * and it is a reply for ato after last received packet, - * increase pingpong count. - */ - if (before(tp->lsndtime, icsk->icsk_ack.lrcvtime) && - (u32)(now - icsk->icsk_ack.lrcvtime) < icsk->icsk_ack.ato) - inet_csk_inc_pingpong_cnt(sk); - tp->lsndtime = now; + + /* If it is a reply for ato after last received + * packet, enter pingpong mode. + */ + if ((u32)(now - icsk->icsk_ack.lrcvtime) < icsk->icsk_ack.ato) + inet_csk_enter_pingpong_mode(sk); } /* Account for an ACK we sent. */ @@ -184,10 +182,10 @@ static inline void tcp_event_ack_sent(struct sock *sk, unsigned int pkts, { struct tcp_sock *tp = tcp_sk(sk); - if (unlikely(tp->compressed_ack > TCP_FASTRETRANS_THRESH)) { + if (unlikely(tp->compressed_ack)) { NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPACKCOMPRESSED, - tp->compressed_ack - TCP_FASTRETRANS_THRESH); - tp->compressed_ack = TCP_FASTRETRANS_THRESH; + tp->compressed_ack); + tp->compressed_ack = 0; if (hrtimer_try_to_cancel(&tp->compressed_ack_timer) == 1) __sock_put(sk); } @@ -229,7 +227,7 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, * which we interpret as a sign the remote TCP is not * misinterpreting the window field as a signed quantity. */ - if (sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows) + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows)) (*rcv_wnd) = min(space, MAX_TCP_WINDOW); else (*rcv_wnd) = min_t(u32, space, U16_MAX); @@ -240,8 +238,8 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, *rcv_wscale = 0; if (wscale_ok) { /* Set window scaling on max possible window */ - space = max_t(u32, space, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); - space = max_t(u32, space, sysctl_rmem_max); + space = max_t(u32, space, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])); + space = max_t(u32, space, READ_ONCE(sysctl_rmem_max)); space = min_t(u32, space, *window_clamp); *rcv_wscale = clamp_t(int, ilog2(space) - 15, 0, TCP_MAX_WSCALE); @@ -284,7 +282,7 @@ static u16 tcp_select_window(struct sock *sk) * scaled window. */ if (!tp->rx_opt.rcv_wscale && - sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows) + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows)) new_win = min(new_win, MAX_TCP_WINDOW); else new_win = min(new_win, (65535U << tp->rx_opt.rcv_wscale)); @@ -323,7 +321,7 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); bool bpf_needs_ecn = tcp_bpf_ca_needs_ecn(sk); - bool use_ecn = sock_net(sk)->ipv4.sysctl_tcp_ecn == 1 || + bool use_ecn = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn) == 1 || tcp_ca_needs_ecn(sk) || bpf_needs_ecn; if (!use_ecn) { @@ -345,7 +343,7 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb) static void tcp_ecn_clear_syn(struct sock *sk, struct sk_buff *skb) { - if (sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback) + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback)) /* tp->ecn_flags are cleared at a later point in time when * SYN ACK is ultimatively being received. */ @@ -394,7 +392,6 @@ static void tcp_init_nondata_skb(struct sk_buff *skb, u32 seq, u8 flags) skb->ip_summed = CHECKSUM_PARTIAL; TCP_SKB_CB(skb)->tcp_flags = flags; - TCP_SKB_CB(skb)->sacked = 0; tcp_skb_pcount_set(skb, 1); @@ -409,13 +406,13 @@ static inline bool tcp_urg_mode(const struct tcp_sock *tp) return tp->snd_una != tp->snd_up; } -#define OPTION_SACK_ADVERTISE (1 << 0) -#define OPTION_TS (1 << 1) -#define OPTION_MD5 (1 << 2) -#define OPTION_WSCALE (1 << 3) -#define OPTION_FAST_OPEN_COOKIE (1 << 8) -#define OPTION_SMC (1 << 9) -#define OPTION_MPTCP (1 << 10) +#define OPTION_SACK_ADVERTISE BIT(0) +#define OPTION_TS BIT(1) +#define OPTION_MD5 BIT(2) +#define OPTION_WSCALE BIT(3) +#define OPTION_FAST_OPEN_COOKIE BIT(8) +#define OPTION_SMC BIT(9) +#define OPTION_MPTCP BIT(10) static void smc_options_write(__be32 *ptr, u16 *options) { @@ -438,20 +435,162 @@ struct tcp_out_options { u8 ws; /* window scale, 0 to disable */ u8 num_sack_blocks; /* number of SACK blocks to include */ u8 hash_size; /* bytes in hash_location */ + u8 bpf_opt_len; /* length of BPF hdr option */ __u8 *hash_location; /* temporary pointer, overloaded */ __u32 tsval, tsecr; /* need to include OPTION_TS */ struct tcp_fastopen_cookie *fastopen_cookie; /* Fast open cookie */ struct mptcp_out_options mptcp; }; -static void mptcp_options_write(__be32 *ptr, struct tcp_out_options *opts) +static void mptcp_options_write(struct tcphdr *th, __be32 *ptr, + struct tcp_sock *tp, + struct tcp_out_options *opts) { #if IS_ENABLED(CONFIG_MPTCP) if (unlikely(OPTION_MPTCP & opts->options)) - mptcp_write_options(ptr, &opts->mptcp); + mptcp_write_options(th, ptr, tp, &opts->mptcp); #endif } +#ifdef CONFIG_CGROUP_BPF +static int bpf_skops_write_hdr_opt_arg0(struct sk_buff *skb, + enum tcp_synack_type synack_type) +{ + if (unlikely(!skb)) + return BPF_WRITE_HDR_TCP_CURRENT_MSS; + + if (unlikely(synack_type == TCP_SYNACK_COOKIE)) + return BPF_WRITE_HDR_TCP_SYNACK_COOKIE; + + return 0; +} + +/* req, syn_skb and synack_type are used when writing synack */ +static void bpf_skops_hdr_opt_len(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct sk_buff *syn_skb, + enum tcp_synack_type synack_type, + struct tcp_out_options *opts, + unsigned int *remaining) +{ + struct bpf_sock_ops_kern sock_ops; + int err; + + if (likely(!BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk), + BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG)) || + !*remaining) + return; + + /* *remaining has already been aligned to 4 bytes, so *remaining >= 4 */ + + /* init sock_ops */ + memset(&sock_ops, 0, offsetof(struct bpf_sock_ops_kern, temp)); + + sock_ops.op = BPF_SOCK_OPS_HDR_OPT_LEN_CB; + + if (req) { + /* The listen "sk" cannot be passed here because + * it is not locked. It would not make too much + * sense to do bpf_setsockopt(listen_sk) based + * on individual connection request also. + * + * Thus, "req" is passed here and the cgroup-bpf-progs + * of the listen "sk" will be run. + * + * "req" is also used here for fastopen even the "sk" here is + * a fullsock "child" sk. It is to keep the behavior + * consistent between fastopen and non-fastopen on + * the bpf programming side. + */ + sock_ops.sk = (struct sock *)req; + sock_ops.syn_skb = syn_skb; + } else { + sock_owned_by_me(sk); + + sock_ops.is_fullsock = 1; + sock_ops.sk = sk; + } + + sock_ops.args[0] = bpf_skops_write_hdr_opt_arg0(skb, synack_type); + sock_ops.remaining_opt_len = *remaining; + /* tcp_current_mss() does not pass a skb */ + if (skb) + bpf_skops_init_skb(&sock_ops, skb, 0); + + err = BPF_CGROUP_RUN_PROG_SOCK_OPS_SK(&sock_ops, sk); + + if (err || sock_ops.remaining_opt_len == *remaining) + return; + + opts->bpf_opt_len = *remaining - sock_ops.remaining_opt_len; + /* round up to 4 bytes */ + opts->bpf_opt_len = (opts->bpf_opt_len + 3) & ~3; + + *remaining -= opts->bpf_opt_len; +} + +static void bpf_skops_write_hdr_opt(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct sk_buff *syn_skb, + enum tcp_synack_type synack_type, + struct tcp_out_options *opts) +{ + u8 first_opt_off, nr_written, max_opt_len = opts->bpf_opt_len; + struct bpf_sock_ops_kern sock_ops; + int err; + + if (likely(!max_opt_len)) + return; + + memset(&sock_ops, 0, offsetof(struct bpf_sock_ops_kern, temp)); + + sock_ops.op = BPF_SOCK_OPS_WRITE_HDR_OPT_CB; + + if (req) { + sock_ops.sk = (struct sock *)req; + sock_ops.syn_skb = syn_skb; + } else { + sock_owned_by_me(sk); + + sock_ops.is_fullsock = 1; + sock_ops.sk = sk; + } + + sock_ops.args[0] = bpf_skops_write_hdr_opt_arg0(skb, synack_type); + sock_ops.remaining_opt_len = max_opt_len; + first_opt_off = tcp_hdrlen(skb) - max_opt_len; + bpf_skops_init_skb(&sock_ops, skb, first_opt_off); + + err = BPF_CGROUP_RUN_PROG_SOCK_OPS_SK(&sock_ops, sk); + + if (err) + nr_written = 0; + else + nr_written = max_opt_len - sock_ops.remaining_opt_len; + + if (nr_written < max_opt_len) + memset(skb->data + first_opt_off + nr_written, TCPOPT_NOP, + max_opt_len - nr_written); +} +#else +static void bpf_skops_hdr_opt_len(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct sk_buff *syn_skb, + enum tcp_synack_type synack_type, + struct tcp_out_options *opts, + unsigned int *remaining) +{ +} + +static void bpf_skops_write_hdr_opt(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct sk_buff *syn_skb, + enum tcp_synack_type synack_type, + struct tcp_out_options *opts) +{ +} +#endif + /* Write previously computed TCP options to the packet. * * Beware: Something in the Internet is very sensitive to the ordering of @@ -465,9 +604,10 @@ static void mptcp_options_write(__be32 *ptr, struct tcp_out_options *opts) * At least SACK_PERM as the first option is known to lead to a disaster * (but it may well be that other scenarios fail similarly). */ -static void tcp_options_write(__be32 *ptr, struct tcp_sock *tp, +static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp, struct tcp_out_options *opts) { + __be32 *ptr = (__be32 *)(th + 1); u16 options = opts->options; /* mungable copy */ if (unlikely(OPTION_MD5 & options)) { @@ -561,7 +701,7 @@ static void tcp_options_write(__be32 *ptr, struct tcp_sock *tp, smc_options_write(ptr, &options); - mptcp_options_write(ptr, opts); + mptcp_options_write(th, ptr, tp, opts); } static void smc_set_option(const struct tcp_sock *tp, @@ -648,18 +788,18 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb, opts->mss = tcp_advertise_mss(sk); remaining -= TCPOLEN_MSS_ALIGNED; - if (likely(sock_net(sk)->ipv4.sysctl_tcp_timestamps && !*md5)) { + if (likely(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_timestamps) && !*md5)) { opts->options |= OPTION_TS; opts->tsval = tcp_skb_timestamp(skb) + tp->tsoffset; opts->tsecr = tp->rx_opt.ts_recent; remaining -= TCPOLEN_TSTAMP_ALIGNED; } - if (likely(sock_net(sk)->ipv4.sysctl_tcp_window_scaling)) { + if (likely(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_window_scaling))) { opts->ws = tp->rx_opt.rcv_wscale; opts->options |= OPTION_WSCALE; remaining -= TCPOLEN_WSCALE_ALIGNED; } - if (likely(sock_net(sk)->ipv4.sysctl_tcp_sack)) { + if (likely(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_sack))) { opts->options |= OPTION_SACK_ADVERTISE; if (unlikely(!(OPTION_TS & opts->options))) remaining -= TCPOLEN_SACKPERM_ALIGNED; @@ -691,6 +831,8 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb, } } + bpf_skops_hdr_opt_len(sk, skb, NULL, NULL, 0, opts, &remaining); + return MAX_TCP_OPTION_SPACE - remaining; } @@ -700,7 +842,9 @@ static unsigned int tcp_synack_options(const struct sock *sk, unsigned int mss, struct sk_buff *skb, struct tcp_out_options *opts, const struct tcp_md5sig_key *md5, - struct tcp_fastopen_cookie *foc) + struct tcp_fastopen_cookie *foc, + enum tcp_synack_type synack_type, + struct sk_buff *syn_skb) { struct inet_request_sock *ireq = inet_rsk(req); unsigned int remaining = MAX_TCP_OPTION_SPACE; @@ -715,7 +859,8 @@ static unsigned int tcp_synack_options(const struct sock *sk, * rather than TS in order to fit in better with old, * buggy kernels, but that was deemed to be unnecessary. */ - ireq->tstamp_ok &= !ireq->sack_ok; + if (synack_type != TCP_SYNACK_COOKIE) + ireq->tstamp_ok &= !ireq->sack_ok; } #endif @@ -756,6 +901,9 @@ static unsigned int tcp_synack_options(const struct sock *sk, smc_set_option_cond(tcp_sk(sk), ireq, opts, &remaining); + bpf_skops_hdr_opt_len((struct sock *)sk, skb, req, syn_skb, + synack_type, opts, &remaining); + return MAX_TCP_OPTION_SPACE - remaining; } @@ -824,6 +972,15 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK; } + if (unlikely(BPF_SOCK_OPS_TEST_FLAG(tp, + BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG))) { + unsigned int remaining = MAX_TCP_OPTION_SPACE - size; + + bpf_skops_hdr_opt_len(sk, skb, NULL, NULL, 0, opts, &remaining); + + size = MAX_TCP_OPTION_SPACE - remaining; + } + return size; } @@ -856,7 +1013,7 @@ static void tcp_tsq_write(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); if (tp->lost_out > tp->retrans_out && - tp->snd_cwnd > tcp_packets_in_flight(tp)) { + tcp_snd_cwnd(tp) > tcp_packets_in_flight(tp)) { tcp_mstamp_refresh(tp); tcp_xmit_retransmit_queue(sk); } @@ -881,9 +1038,9 @@ static void tcp_tsq_handler(struct sock *sk) * transferring tsq->head because tcp_wfree() might * interrupt us (non NAPI drivers) */ -static void tcp_tasklet_func(unsigned long data) +static void tcp_tasklet_func(struct tasklet_struct *t) { - struct tsq_tasklet *tsq = (struct tsq_tasklet *)data; + struct tsq_tasklet *tsq = from_tasklet(tsq, t, tasklet); LIST_HEAD(list); unsigned long flags; struct list_head *q, *n; @@ -968,9 +1125,7 @@ void __init tcp_tasklet_init(void) struct tsq_tasklet *tsq = &per_cpu(tsq_tasklet, i); INIT_LIST_HEAD(&tsq->head); - tasklet_init(&tsq->tasklet, - tcp_tasklet_func, - (unsigned long)tsq); + tasklet_setup(&tsq->tasklet, tcp_tasklet_func); } } @@ -1064,6 +1219,10 @@ static void tcp_update_skb_after_send(struct sock *sk, struct sk_buff *skb, list_move_tail(&skb->tcp_tsorted_anchor, &tp->tsorted_sent_queue); } +INDIRECT_CALLABLE_DECLARE(int ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl)); +INDIRECT_CALLABLE_DECLARE(int inet6_csk_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl)); +INDIRECT_CALLABLE_DECLARE(void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb)); + /* This routine actually transmits TCP packets queued in by * tcp_do_sendmsg(). This is used by both the initial * transmission and possible later retransmissions. @@ -1094,10 +1253,8 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, tp = tcp_sk(sk); prior_wstamp = tp->tcp_wstamp_ns; tp->tcp_wstamp_ns = max(tp->tcp_wstamp_ns, tp->tcp_clock_cache); - skb->skb_mstamp_ns = tp->tcp_wstamp_ns; + skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true); if (clone_it) { - TCP_SKB_CB(skb)->tx.in_flight = TCP_SKB_CB(skb)->end_seq - - tp->snd_una; oskb = skb; tcp_skb_tsorted_save(oskb) { @@ -1109,6 +1266,10 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, if (unlikely(!skb)) return -ENOBUFS; + /* retransmit skbs might have a non zero value in skb->dev + * because skb->dev is aliased with skb->rbnode.rb_left + */ + skb->dev = NULL; } inet = inet_sk(sk); @@ -1155,7 +1316,6 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, skb_orphan(skb); skb->sk = sk; skb->destructor = skb_is_tcp_pure_ack(skb) ? __sock_wfree : tcp_wfree; - skb_set_hash_from_sk(skb, sk); refcount_add(skb->truesize, &sk->sk_wmem_alloc); skb_set_dst_pending_confirm(skb, sk->sk_dst_pending_confirm); @@ -1183,7 +1343,6 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, } } - tcp_options_write((__be32 *)(th + 1), tp, &opts); skb_shinfo(skb)->gso_type = sk->sk_gso_type; if (likely(!(tcb->tcp_flags & TCPHDR_SYN))) { th->window = htons(tcp_select_window(sk)); @@ -1194,16 +1353,24 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, */ th->window = htons(min(tp->rcv_wnd, 65535U)); } + + tcp_options_write(th, tp, &opts); + #ifdef CONFIG_TCP_MD5SIG /* Calculate the MD5 hash, as we have all we need now */ if (md5) { - sk_nocaps_add(sk, NETIF_F_GSO_MASK); + sk_gso_disable(sk); tp->af_specific->calc_md5_hash(opts.hash_location, md5, sk, skb); } #endif - icsk->icsk_af_ops->send_check(sk, skb); + /* BPF prog is the last one writing header option */ + bpf_skops_write_hdr_opt(sk, skb, NULL, NULL, 0, &opts); + + INDIRECT_CALL_INET(icsk->icsk_af_ops->send_check, + tcp_v6_send_check, tcp_v4_send_check, + sk, skb); if (likely(tcb->tcp_flags & TCPHDR_ACK)) tcp_event_ack_sent(sk, tcp_skb_pcount(skb), rcv_nxt); @@ -1219,6 +1386,7 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, tcp_skb_pcount(skb)); tp->segs_out += tcp_skb_pcount(skb); + skb_set_hash_from_sk(skb, sk); /* OK, its time to fill skb_shinfo(skb)->gso_{segs|size} */ skb_shinfo(skb)->gso_segs = tcp_skb_pcount(skb); skb_shinfo(skb)->gso_size = tcp_skb_mss(skb); @@ -1231,7 +1399,9 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, tcp_add_tx_delay(skb, tp); - err = icsk->icsk_af_ops->queue_xmit(sk, skb, &inet->cork.fl); + err = INDIRECT_CALL_INET(icsk->icsk_af_ops->queue_xmit, + inet6_csk_xmit, ip_queue_xmit, + sk, skb, &inet->cork.fl); if (unlikely(err > 0)) { tcp_enter_cwr(sk); @@ -1380,7 +1550,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, * SO_SNDBUF values. * Also allow first and last skb in retransmit queue to be split. */ - limit = sk->sk_sndbuf + 2 * SKB_TRUESIZE(GSO_MAX_SIZE); + limit = sk->sk_sndbuf + 2 * SKB_TRUESIZE(GSO_LEGACY_MAX_SIZE); if (unlikely((sk->sk_wmem_queued >> 1) > limit && tcp_queue != TCP_FRAG_IN_WRITE_QUEUE && skb != tcp_rtx_queue_head(sk) && @@ -1389,14 +1559,15 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, return -ENOMEM; } - if (skb_unclone(skb, gfp)) + if (skb_unclone_keeptruesize(skb, gfp)) return -ENOMEM; /* Get a new skb... force flag on. */ - buff = sk_stream_alloc_skb(sk, nsize, gfp, true); + buff = tcp_stream_alloc_skb(sk, nsize, gfp, true); if (!buff) return -ENOMEM; /* We'll just try again later. */ skb_copy_decrypted(buff, skb); + mptcp_skb_ext_copy(buff, skb); sk_wmem_queued_add(sk, buff->truesize); sk_mem_charge(sk, buff->truesize); @@ -1418,9 +1589,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, skb_split(skb, buff, len); - buff->ip_summed = CHECKSUM_PARTIAL; - - buff->tstamp = skb->tstamp; + skb_set_delivery_time(buff, skb->tstamp, true); tcp_fragment_tstamp(skb, buff); old_factor = tcp_skb_pcount(skb); @@ -1498,19 +1667,18 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len) { u32 delta_truesize; - if (skb_unclone(skb, GFP_ATOMIC)) + if (skb_unclone_keeptruesize(skb, GFP_ATOMIC)) return -ENOMEM; delta_truesize = __pskb_trim_head(skb, len); TCP_SKB_CB(skb)->seq += len; - skb->ip_summed = CHECKSUM_PARTIAL; if (delta_truesize) { skb->truesize -= delta_truesize; sk_wmem_queued_add(sk, -delta_truesize); - sk_mem_uncharge(sk, delta_truesize); - sock_set_flag(sk, SOCK_QUEUE_SHRUNK); + if (!skb_zcopy_pure(skb)) + sk_mem_uncharge(sk, delta_truesize); } /* Any change of skb->len requires recalculation of tso factor. */ @@ -1548,7 +1716,8 @@ static inline int __tcp_mtu_to_mss(struct sock *sk, int pmtu) mss_now -= icsk->icsk_ext_hdr_len; /* Then reserve room for full set of TCP options and 8 bytes of data */ - mss_now = max(mss_now, sock_net(sk)->ipv4.sysctl_tcp_min_snd_mss); + mss_now = max(mss_now, + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_min_snd_mss)); return mss_now; } @@ -1559,6 +1728,7 @@ int tcp_mtu_to_mss(struct sock *sk, int pmtu) return __tcp_mtu_to_mss(sk, pmtu) - (tcp_sk(sk)->tcp_header_len - sizeof(struct tcphdr)); } +EXPORT_SYMBOL(tcp_mtu_to_mss); /* Inverse of above */ int tcp_mss_to_mtu(struct sock *sk, int mss) @@ -1590,10 +1760,10 @@ void tcp_mtup_init(struct sock *sk) struct inet_connection_sock *icsk = inet_csk(sk); struct net *net = sock_net(sk); - icsk->icsk_mtup.enabled = net->ipv4.sysctl_tcp_mtu_probing > 1; + icsk->icsk_mtup.enabled = READ_ONCE(net->ipv4.sysctl_tcp_mtu_probing) > 1; icsk->icsk_mtup.search_high = tp->rx_opt.mss_clamp + sizeof(struct tcphdr) + icsk->icsk_af_ops->net_header_len; - icsk->icsk_mtup.search_low = tcp_mss_to_mtu(sk, net->ipv4.sysctl_tcp_base_mss); + icsk->icsk_mtup.search_low = tcp_mss_to_mtu(sk, READ_ONCE(net->ipv4.sysctl_tcp_base_mss)); icsk->icsk_mtup.probe_size = 0; if (icsk->icsk_mtup.enabled) icsk->icsk_mtup.probe_timestamp = tcp_jiffies32; @@ -1691,9 +1861,9 @@ static void tcp_cwnd_application_limited(struct sock *sk) /* Limited by application or receiver window. */ u32 init_win = tcp_init_cwnd(tp, __sk_dst_get(sk)); u32 win_used = max(tp->snd_cwnd_used, init_win); - if (win_used < tp->snd_cwnd) { + if (win_used < tcp_snd_cwnd(tp)) { tp->snd_ssthresh = tcp_current_ssthresh(sk); - tp->snd_cwnd = (tp->snd_cwnd + win_used) >> 1; + tcp_snd_cwnd_set(tp, (tcp_snd_cwnd(tp) + win_used) >> 1); } tp->snd_cwnd_used = 0; } @@ -1705,14 +1875,20 @@ static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited) const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops; struct tcp_sock *tp = tcp_sk(sk); - /* Track the maximum number of outstanding packets in each - * window, and remember whether we were cwnd-limited then. + /* Track the strongest available signal of the degree to which the cwnd + * is fully utilized. If cwnd-limited then remember that fact for the + * current window. If not cwnd-limited then track the maximum number of + * outstanding packets in the current window. (If cwnd-limited then we + * chose to not update tp->max_packets_out to avoid an extra else + * clause with no functional impact.) */ - if (!before(tp->snd_una, tp->max_packets_seq) || - tp->packets_out > tp->max_packets_out) { - tp->max_packets_out = tp->packets_out; - tp->max_packets_seq = tp->snd_nxt; + if (!before(tp->snd_una, tp->cwnd_usage_seq) || + is_cwnd_limited || + (!tp->is_cwnd_limited && + tp->packets_out > tp->max_packets_out)) { tp->is_cwnd_limited = is_cwnd_limited; + tp->max_packets_out = tp->packets_out; + tp->cwnd_usage_seq = tp->snd_nxt; } if (tcp_is_cwnd_limited(sk)) { @@ -1724,7 +1900,7 @@ static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited) if (tp->packets_out > tp->snd_cwnd_used) tp->snd_cwnd_used = tp->packets_out; - if (sock_net(sk)->ipv4.sysctl_tcp_slow_start_after_idle && + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_slow_start_after_idle) && (s32)(tcp_jiffies32 - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto && !ca_ops->cong_control) tcp_cwnd_application_limited(sk); @@ -1781,25 +1957,34 @@ static bool tcp_nagle_check(bool partial, const struct tcp_sock *tp, } /* Return how many segs we'd like on a TSO packet, - * to send one TSO packet per ms + * depending on current pacing rate, and how close the peer is. + * + * Rationale is: + * - For close peers, we rather send bigger packets to reduce + * cpu costs, because occasional losses will be repaired fast. + * - For long distance/rtt flows, we would like to get ACK clocking + * with 1 ACK per ms. + * + * Use min_rtt to help adapt TSO burst size, with smaller min_rtt resulting + * in bigger TSO bursts. We we cut the RTT-based allowance in half + * for every 2^9 usec (aka 512 us) of RTT, so that the RTT-based allowance + * is below 1500 bytes after 6 * ~500 usec = 3ms. */ static u32 tcp_tso_autosize(const struct sock *sk, unsigned int mss_now, int min_tso_segs) { - u32 bytes, segs; + unsigned long bytes; + u32 r; - bytes = min_t(unsigned long, - sk->sk_pacing_rate >> READ_ONCE(sk->sk_pacing_shift), - sk->sk_gso_max_size - 1 - MAX_TCP_HEADER); + bytes = sk->sk_pacing_rate >> READ_ONCE(sk->sk_pacing_shift); - /* Goal is to send at least one packet per ms, - * not one big TSO packet every 100 ms. - * This preserves ACK clocking and is consistent - * with tcp_tso_should_defer() heuristic. - */ - segs = max_t(u32, bytes / mss_now, min_tso_segs); + r = tcp_min_rtt(tcp_sk(sk)) >> READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_tso_rtt_log); + if (r < BITS_PER_TYPE(sk->sk_gso_max_size)) + bytes += sk->sk_gso_max_size >> r; + + bytes = min_t(unsigned long, bytes, sk->sk_gso_max_size); - return segs; + return max_t(u32, bytes / mss_now, min_tso_segs); } /* Return the number of segments we want in the skb we are transmitting. @@ -1812,7 +1997,7 @@ static u32 tcp_tso_segs(struct sock *sk, unsigned int mss_now) min_tso = ca_ops->min_tso_segs ? ca_ops->min_tso_segs(sk) : - sock_net(sk)->ipv4.sysctl_tcp_min_tso_segs; + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_min_tso_segs); tso_segs = tcp_tso_autosize(sk, mss_now, min_tso); return min_t(u32, tso_segs, sk->sk_gso_max_segs); @@ -1864,7 +2049,7 @@ static inline unsigned int tcp_cwnd_test(const struct tcp_sock *tp, return 1; in_flight = tcp_packets_in_flight(tp); - cwnd = tp->snd_cwnd; + cwnd = tcp_snd_cwnd(tp); if (in_flight >= cwnd) return 0; @@ -1948,10 +2133,11 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, return tcp_fragment(sk, TCP_FRAG_IN_WRITE_QUEUE, skb, len, mss_now, gfp); - buff = sk_stream_alloc_skb(sk, 0, gfp, true); + buff = tcp_stream_alloc_skb(sk, 0, gfp, true); if (unlikely(!buff)) return -ENOMEM; skb_copy_decrypted(buff, skb); + mptcp_skb_ext_copy(buff, skb); sk_wmem_queued_add(sk, buff->truesize); sk_mem_charge(sk, buff->truesize); @@ -1968,12 +2154,8 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, TCP_SKB_CB(skb)->tcp_flags = flags & ~(TCPHDR_FIN | TCPHDR_PSH); TCP_SKB_CB(buff)->tcp_flags = flags; - /* This packet was never sent out yet, so no SACK bits. */ - TCP_SKB_CB(buff)->sacked = 0; - tcp_skb_fragment_eor(skb, buff); - buff->ip_summed = CHECKSUM_PARTIAL; skb_split(skb, buff, len); tcp_fragment_tstamp(skb, buff); @@ -2020,12 +2202,12 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb, in_flight = tcp_packets_in_flight(tp); BUG_ON(tcp_skb_pcount(skb) <= 1); - BUG_ON(tp->snd_cwnd <= in_flight); + BUG_ON(tcp_snd_cwnd(tp) <= in_flight); send_win = tcp_wnd_end(tp) - TCP_SKB_CB(skb)->seq; /* From in_flight test above, we know that cwnd > in_flight. */ - cong_win = (tp->snd_cwnd - in_flight) * tp->mss_cache; + cong_win = (tcp_snd_cwnd(tp) - in_flight) * tp->mss_cache; limit = min(send_win, cong_win); @@ -2039,7 +2221,7 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb, win_divisor = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_tso_win_divisor); if (win_divisor) { - u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache); + u32 chunk = min(tp->snd_wnd, tcp_snd_cwnd(tp) * tp->mss_cache); /* If at least some fraction of a window is available, * just use it. @@ -2103,7 +2285,7 @@ static inline void tcp_mtu_check_reprobe(struct sock *sk) u32 interval; s32 delta; - interval = net->ipv4.sysctl_tcp_probe_interval; + interval = READ_ONCE(net->ipv4.sysctl_tcp_probe_interval); delta = tcp_jiffies32 - icsk->icsk_mtup.probe_timestamp; if (unlikely(delta >= interval * HZ)) { int mss = tcp_current_mss(sk); @@ -2129,7 +2311,9 @@ static bool tcp_can_coalesce_send_queue_head(struct sock *sk, int len) if (len <= skb->len) break; - if (unlikely(TCP_SKB_CB(skb)->eor) || tcp_has_tx_tstamp(skb)) + if (unlikely(TCP_SKB_CB(skb)->eor) || + tcp_has_tx_tstamp(skb) || + !skb_pure_zcopy_same(skb, next)) return false; len -= skb->len; @@ -2167,7 +2351,7 @@ static int tcp_mtu_probe(struct sock *sk) if (likely(!icsk->icsk_mtup.enabled || icsk->icsk_mtup.probe_size || inet_csk(sk)->icsk_ca_state != TCP_CA_Open || - tp->snd_cwnd < 11 || + tcp_snd_cwnd(tp) < 11 || tp->rx_opt.num_sacks || tp->rx_opt.dsack)) return -1; @@ -2185,7 +2369,7 @@ static int tcp_mtu_probe(struct sock *sk) * probing process by not resetting search range to its orignal. */ if (probe_size > tcp_mtu_to_mss(sk, icsk->icsk_mtup.search_high) || - interval < net->ipv4.sysctl_tcp_probe_threshold) { + interval < READ_ONCE(net->ipv4.sysctl_tcp_probe_threshold)) { /* Check whether enough time has elaplased for * another round of probing. */ @@ -2203,7 +2387,7 @@ static int tcp_mtu_probe(struct sock *sk) return 0; /* Do we need to wait to drain cwnd? With none in flight, don't stall */ - if (tcp_packets_in_flight(tp) + 2 > tp->snd_cwnd) { + if (tcp_packets_in_flight(tp) + 2 > tcp_snd_cwnd(tp)) { if (!tcp_packets_in_flight(tp)) return -1; else @@ -2214,7 +2398,7 @@ static int tcp_mtu_probe(struct sock *sk) return -1; /* We're allowed to probe. Build it now. */ - nskb = sk_stream_alloc_skb(sk, probe_size, GFP_ATOMIC, false); + nskb = tcp_stream_alloc_skb(sk, probe_size, GFP_ATOMIC, false); if (!nskb) return -1; sk_wmem_queued_add(sk, nskb->truesize); @@ -2222,13 +2406,11 @@ static int tcp_mtu_probe(struct sock *sk) skb = tcp_send_head(sk); skb_copy_decrypted(nskb, skb); + mptcp_skb_ext_copy(nskb, skb); TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(skb)->seq; TCP_SKB_CB(nskb)->end_seq = TCP_SKB_CB(skb)->seq + probe_size; TCP_SKB_CB(nskb)->tcp_flags = TCPHDR_ACK; - TCP_SKB_CB(nskb)->sacked = 0; - nskb->csum = 0; - nskb->ip_summed = CHECKSUM_PARTIAL; tcp_insert_write_queue_before(nskb, skb, sk); tcp_highest_sack_replace(sk, skb, nskb); @@ -2248,7 +2430,7 @@ static int tcp_mtu_probe(struct sock *sk) TCP_SKB_CB(nskb)->eor = TCP_SKB_CB(skb)->eor; tcp_skb_collapse_tstamp(nskb, skb); tcp_unlink_write_queue(skb, sk); - sk_wmem_free_skb(sk, skb); + tcp_wmem_free_skb(sk, skb); } else { TCP_SKB_CB(nskb)->tcp_flags |= TCP_SKB_CB(skb)->tcp_flags & ~(TCPHDR_FIN|TCPHDR_PSH); @@ -2274,7 +2456,7 @@ static int tcp_mtu_probe(struct sock *sk) if (!tcp_transmit_skb(sk, nskb, 1, GFP_ATOMIC)) { /* Decrement cwnd here because we are sending * effectively two packets. */ - tp->snd_cwnd--; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) - 1); tcp_event_new_data_sent(sk, nskb); icsk->icsk_mtup.probe_size = tcp_mss_to_mtu(sk, nskb->len); @@ -2327,7 +2509,7 @@ static bool tcp_small_queue_check(struct sock *sk, const struct sk_buff *skb, sk->sk_pacing_rate >> READ_ONCE(sk->sk_pacing_shift)); if (sk->sk_pacing_status == SK_PACING_NONE) limit = min_t(unsigned long, limit, - sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes)); limit <<= factor; if (static_branch_unlikely(&tcp_tx_delay_enabled) && @@ -2449,7 +2631,8 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE) { /* "skb_mstamp_ns" is used as a start point for the retransmit timer */ - skb->skb_mstamp_ns = tp->tcp_wstamp_ns = tp->tcp_clock_cache; + tp->tcp_wstamp_ns = tp->tcp_clock_cache; + skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true); list_move_tail(&skb->tcp_tsorted_anchor, &tp->tsorted_sent_queue); tcp_init_tso_segs(skb, mss_now); goto repair; /* Skip network transmission */ @@ -2531,6 +2714,10 @@ repair: else tcp_chrono_stop(sk, TCP_CHRONO_RWND_LIMITED); + is_cwnd_limited |= (tcp_packets_in_flight(tp) >= tcp_snd_cwnd(tp)); + if (likely(sent_pkts || is_cwnd_limited)) + tcp_cwnd_validate(sk, is_cwnd_limited); + if (likely(sent_pkts)) { if (tcp_in_cwnd_reduction(sk)) tp->prr_out += sent_pkts; @@ -2538,8 +2725,6 @@ repair: /* Send one loss probe per tail loss episode. */ if (push_one != 2) tcp_schedule_loss_probe(sk, false); - is_cwnd_limited |= (tcp_packets_in_flight(tp) >= tp->snd_cwnd); - tcp_cwnd_validate(sk, is_cwnd_limited); return false; } return !tp->packets_out && !tcp_write_queue_empty(sk); @@ -2558,7 +2743,7 @@ bool tcp_schedule_loss_probe(struct sock *sk, bool advancing_rto) if (rcu_access_pointer(tp->fastopen_rsk)) return false; - early_retrans = sock_net(sk)->ipv4.sysctl_tcp_early_retrans; + early_retrans = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_early_retrans); /* Schedule a loss probe in 2*RTT for SACK capable connections * not in loss recovery, that are either limited by cwnd or application. */ @@ -2589,8 +2774,7 @@ bool tcp_schedule_loss_probe(struct sock *sk, bool advancing_rto) if (rto_delta_us > 0) timeout = min_t(u32, timeout, usecs_to_jiffies(rto_delta_us)); - tcp_reset_xmit_timer(sk, ICSK_TIME_LOSS_PROBE, timeout, - TCP_RTO_MAX, NULL); + tcp_reset_xmit_timer(sk, ICSK_TIME_LOSS_PROBE, timeout, TCP_RTO_MAX); return true; } @@ -2598,13 +2782,17 @@ bool tcp_schedule_loss_probe(struct sock *sk, bool advancing_rto) * a packet is still in a qdisc or driver queue. * In this case, there is very little point doing a retransmit ! */ -static bool skb_still_in_host_queue(const struct sock *sk, +static bool skb_still_in_host_queue(struct sock *sk, const struct sk_buff *skb) { if (unlikely(skb_fclone_busy(sk, skb))) { - NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPSPURIOUS_RTX_HOSTQUEUES); - return true; + set_bit(TSQ_THROTTLED, &sk->sk_tsq_flags); + smp_mb__after_atomic(); + if (skb_fclone_busy(sk, skb)) { + NET_INC_STATS(sock_net(sk), + LINUX_MIB_TCPSPURIOUS_RTX_HOSTQUEUES); + return true; + } } return false; } @@ -2619,6 +2807,11 @@ void tcp_send_loss_probe(struct sock *sk) int pcount; int mss = tcp_current_mss(sk); + /* At most one outstanding TLP */ + if (tp->tlp_high_seq) + goto rearm_timer; + + tp->tlp_retrans = 0; skb = tcp_send_head(sk); if (skb && tcp_snd_wnd_test(tp, skb, mss)) { pcount = tp->packets_out; @@ -2631,15 +2824,11 @@ void tcp_send_loss_probe(struct sock *sk) if (unlikely(!skb)) { WARN_ONCE(tp->packets_out, "invalid inflight: %u state %u cwnd %u mss %d\n", - tp->packets_out, sk->sk_state, tp->snd_cwnd, mss); + tp->packets_out, sk->sk_state, tcp_snd_cwnd(tp), mss); inet_csk(sk)->icsk_pending = 0; return; } - /* At most one outstanding TLP retransmission. */ - if (tp->tlp_high_seq) - goto rearm_timer; - if (skb_still_in_host_queue(sk, skb)) goto rearm_timer; @@ -2661,10 +2850,12 @@ void tcp_send_loss_probe(struct sock *sk) if (__tcp_retransmit_skb(sk, skb, 1)) goto rearm_timer; + tp->tlp_retrans = 1; + +probe_sent: /* Record snd_nxt for loss detection. */ tp->tlp_high_seq = tp->snd_nxt; -probe_sent: NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPLOSSPROBES); /* Reset s.t. tcp_rearm_rto will restart timer from now */ inet_csk(sk)->icsk_pending = 0; @@ -2768,8 +2959,12 @@ u32 __tcp_select_window(struct sock *sk) int mss = icsk->icsk_ack.rcv_mss; int free_space = tcp_space(sk); int allowed_space = tcp_full_space(sk); - int full_space = min_t(int, tp->window_clamp, allowed_space); - int window; + int full_space, window; + + if (sk_is_mptcp(sk)) + mptcp_space(sk, &free_space, &allowed_space); + + full_space = min_t(int, tp->window_clamp, allowed_space); if (unlikely(mss > full_space)) { mss = full_space; @@ -2780,8 +2975,7 @@ u32 __tcp_select_window(struct sock *sk) icsk->icsk_ack.quick = 0; if (tcp_under_memory_pressure(sk)) - tp->rcv_ssthresh = min(tp->rcv_ssthresh, - 4U * tp->advmss); + tcp_adjust_rcv_ssthresh(sk); /* free_space might become our new window, make sure we don't * increase it due to wscale. @@ -2859,13 +3053,9 @@ static bool tcp_collapse_retrans(struct sock *sk, struct sk_buff *skb) BUG_ON(tcp_skb_pcount(skb) != 1 || tcp_skb_pcount(next_skb) != 1); - if (next_skb_size) { - if (next_skb_size <= skb_availroom(skb)) - skb_copy_bits(next_skb, 0, skb_put(skb, next_skb_size), - next_skb_size); - else if (!tcp_skb_shift(skb, next_skb, 1, next_skb_size)) - return false; - } + if (next_skb_size && !tcp_skb_shift(skb, next_skb, 1, next_skb_size)) + return false; + tcp_highest_sack_replace(sk, next_skb, skb); /* Update sequence range on original skb. */ @@ -2917,7 +3107,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *to, struct sk_buff *skb = to, *tmp; bool first = true; - if (!sock_net(sk)->ipv4.sysctl_tcp_retrans_collapse) + if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_retrans_collapse)) return; if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN) return; @@ -2957,20 +3147,12 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) struct tcp_sock *tp = tcp_sk(sk); unsigned int cur_mss; int diff, len, err; - + int avail_wnd; /* Inconclusive MTU probe */ if (icsk->icsk_mtup.probe_size) icsk->icsk_mtup.probe_size = 0; - /* Do not sent more than we queued. 1/4 is reserved for possible - * copying overhead: fragmentation, tunneling, mangling etc. - */ - if (refcount_read(&sk->sk_wmem_alloc) > - min_t(u32, sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), - sk->sk_sndbuf)) - return -EAGAIN; - if (skb_still_in_host_queue(sk, skb)) return -EBUSY; @@ -2987,23 +3169,31 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) return -EHOSTUNREACH; /* Routing failure or similar. */ cur_mss = tcp_current_mss(sk); + avail_wnd = tcp_wnd_end(tp) - TCP_SKB_CB(skb)->seq; /* If receiver has shrunk his window, and skb is out of * new window, do not retransmit it. The exception is the * case, when window is shrunk to zero. In this case - * our retransmit serves as a zero window probe. + * our retransmit of one segment serves as a zero window probe. */ - if (!before(TCP_SKB_CB(skb)->seq, tcp_wnd_end(tp)) && - TCP_SKB_CB(skb)->seq != tp->snd_una) - return -EAGAIN; + if (avail_wnd <= 0) { + if (TCP_SKB_CB(skb)->seq != tp->snd_una) + return -EAGAIN; + avail_wnd = cur_mss; + } len = cur_mss * segs; + if (len > avail_wnd) { + len = rounddown(avail_wnd, cur_mss); + if (!len) + len = avail_wnd; + } if (skb->len > len) { if (tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, len, cur_mss, GFP_ATOMIC)) return -ENOMEM; /* We'll try again later. */ } else { - if (skb_unclone(skb, GFP_ATOMIC)) + if (skb_unclone_keeptruesize(skb, GFP_ATOMIC)) return -ENOMEM; diff = tcp_skb_pcount(skb); @@ -3011,8 +3201,9 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) diff -= tcp_skb_pcount(skb); if (diff) tcp_adjust_pcount(sk, skb, diff); - if (skb->len < cur_mss) - tcp_retrans_try_collapse(sk, skb, cur_mss); + avail_wnd = min_t(int, avail_wnd, cur_mss); + if (skb->len < avail_wnd) + tcp_retrans_try_collapse(sk, skb, avail_wnd); } /* RFC3168, section 6.1.1.1. ECN fallback */ @@ -3037,8 +3228,12 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) tcp_skb_tsorted_save(skb) { nskb = __pskb_copy(skb, MAX_TCP_HEADER, GFP_ATOMIC); - err = nskb ? tcp_transmit_skb(sk, nskb, 0, GFP_ATOMIC) : - -ENOBUFS; + if (nskb) { + nskb->dev = NULL; + err = tcp_transmit_skb(sk, nskb, 0, GFP_ATOMIC); + } else { + err = -ENOBUFS; + } } tcp_skb_tsorted_restore(skb); if (!err) { @@ -3101,6 +3296,7 @@ void tcp_xmit_retransmit_queue(struct sock *sk) const struct inet_connection_sock *icsk = inet_csk(sk); struct sk_buff *skb, *rtx_head, *hole = NULL; struct tcp_sock *tp = tcp_sk(sk); + bool rearm_timer = false; u32 max_segs; int mib_idx; @@ -3121,9 +3317,9 @@ void tcp_xmit_retransmit_queue(struct sock *sk) if (!hole) tp->retransmit_skb_hint = skb; - segs = tp->snd_cwnd - tcp_packets_in_flight(tp); + segs = tcp_snd_cwnd(tp) - tcp_packets_in_flight(tp); if (segs <= 0) - return; + break; sacked = TCP_SKB_CB(skb)->sacked; /* In case tcp_shift_skb_data() have aggregated large skbs, * we need to make sure not sending too bigs TSO packets @@ -3148,10 +3344,10 @@ void tcp_xmit_retransmit_queue(struct sock *sk) continue; if (tcp_small_queue_check(sk, skb, 1)) - return; + break; if (tcp_retransmit_skb(sk, skb, segs)) - return; + break; NET_ADD_STATS(sock_net(sk), mib_idx, tcp_skb_pcount(skb)); @@ -3160,11 +3356,13 @@ void tcp_xmit_retransmit_queue(struct sock *sk) if (skb == rtx_head && icsk->icsk_pending != ICSK_TIME_REO_TIMEOUT) - tcp_reset_xmit_timer(sk, ICSK_TIME_RETRANS, - inet_csk(sk)->icsk_rto, - TCP_RTO_MAX, - skb); + rearm_timer = true; + } + if (rearm_timer) + tcp_reset_xmit_timer(sk, ICSK_TIME_RETRANS, + inet_csk(sk)->icsk_rto, + TCP_RTO_MAX); } /* We allow to exceed memory limits for FIN packets to expedite @@ -3176,16 +3374,18 @@ void tcp_xmit_retransmit_queue(struct sock *sk) */ void sk_forced_mem_schedule(struct sock *sk, int size) { - int amt; + int delta, amt; - if (size <= sk->sk_forward_alloc) + delta = size - sk->sk_forward_alloc; + if (delta <= 0) return; - amt = sk_mem_pages(size); - sk->sk_forward_alloc += amt * SK_MEM_QUANTUM; + amt = sk_mem_pages(delta); + sk->sk_forward_alloc += amt << PAGE_SHIFT; sk_memory_allocated_add(sk, amt); if (mem_cgroup_sockets_enabled && sk->sk_memcg) - mem_cgroup_charge_skmem(sk->sk_memcg, amt); + mem_cgroup_charge_skmem(sk->sk_memcg, amt, + gfp_memcg_charge() | __GFP_NOFAIL); } /* Send a FIN. The caller locks the socket for us. @@ -3309,18 +3509,20 @@ int tcp_send_synack(struct sock *sk) } /** - * tcp_make_synack - Prepare a SYN-ACK. - * sk: listener socket - * dst: dst entry attached to the SYNACK - * req: request_sock pointer - * - * Allocate one skb and build a SYNACK packet. - * @dst is consumed : Caller should not use it again. + * tcp_make_synack - Allocate one skb and build a SYNACK packet. + * @sk: listener socket + * @dst: dst entry attached to the SYNACK. It is consumed and caller + * should not use it again. + * @req: request_sock pointer + * @foc: cookie for tcp fast open + * @synack_type: Type of synack to prepare + * @syn_skb: SYN packet just received. It could be NULL for rtx case. */ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, struct request_sock *req, struct tcp_fastopen_cookie *foc, - enum tcp_synack_type synack_type) + enum tcp_synack_type synack_type, + struct sk_buff *syn_skb) { struct inet_request_sock *ireq = inet_rsk(req); const struct tcp_sock *tp = tcp_sk(sk); @@ -3364,12 +3566,13 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, memset(&opts, 0, sizeof(opts)); now = tcp_clock_ns(); #ifdef CONFIG_SYN_COOKIES - if (unlikely(req->cookie_ts)) - skb->skb_mstamp_ns = cookie_init_timestamp(req, now); + if (unlikely(synack_type == TCP_SYNACK_COOKIE && ireq->tstamp_ok)) + skb_set_delivery_time(skb, cookie_init_timestamp(req, now), + true); else #endif { - skb->skb_mstamp_ns = now; + skb_set_delivery_time(skb, now, true); if (!tcp_rsk(req)->snt_synack) /* Timestamp first SYNACK */ tcp_rsk(req)->snt_synack = tcp_skb_timestamp_us(skb); } @@ -3379,8 +3582,11 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, md5 = tcp_rsk(req)->af_specific->req_md5_lookup(sk, req_to_sk(req)); #endif skb_set_hash(skb, tcp_rsk(req)->txhash, PKT_HASH_TYPE_L4); + /* bpf program will be interested in the tcp_flags */ + TCP_SKB_CB(skb)->tcp_flags = TCPHDR_SYN | TCPHDR_ACK; tcp_header_size = tcp_synack_options(sk, req, mss, skb, &opts, md5, - foc) + sizeof(*th); + foc, synack_type, + syn_skb) + sizeof(*th); skb_push(skb, tcp_header_size); skb_reset_transport_header(skb); @@ -3400,7 +3606,7 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, /* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */ th->window = htons(min(req->rsk_rcv_wnd, 65535U)); - tcp_options_write((__be32 *)(th + 1), NULL, &opts); + tcp_options_write(th, NULL, &opts); th->doff = (tcp_header_size >> 2); __TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTSEGS); @@ -3412,7 +3618,10 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, rcu_read_unlock(); #endif - skb->skb_mstamp_ns = now; + bpf_skops_write_hdr_opt((struct sock *)sk, skb, req, syn_skb, + synack_type, &opts); + + skb_set_delivery_time(skb, now, true); tcp_add_tx_delay(skb, tp); return skb; @@ -3450,7 +3659,7 @@ static void tcp_connect_init(struct sock *sk) * See tcp_input.c:tcp_rcv_state_process case TCP_SYN_SENT. */ tp->tcp_header_len = sizeof(struct tcphdr); - if (sock_net(sk)->ipv4.sysctl_tcp_timestamps) + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_timestamps)) tp->tcp_header_len += TCPOLEN_TSTAMP_ALIGNED; #ifdef CONFIG_TCP_MD5SIG @@ -3486,7 +3695,7 @@ static void tcp_connect_init(struct sock *sk) tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0), &tp->rcv_wnd, &tp->window_clamp, - sock_net(sk)->ipv4.sysctl_tcp_window_scaling, + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_window_scaling), &rcv_wscale, rcv_wnd); @@ -3537,6 +3746,7 @@ static void tcp_connect_queue_skb(struct sock *sk, struct sk_buff *skb) */ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn) { + struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct tcp_fastopen_request *fo = tp->fastopen_req; int space, err = 0; @@ -3551,8 +3761,10 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn) * private TCP options. The cost is reduced data space in SYN :( */ tp->rx_opt.mss_clamp = tcp_mss_clamp(tp, tp->rx_opt.mss_clamp); + /* Sync mss_cache after updating the mss_clamp */ + tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); - space = __tcp_mtu_to_mss(sk, inet_csk(sk)->icsk_pmtu_cookie) - + space = __tcp_mtu_to_mss(sk, icsk->icsk_pmtu_cookie) - MAX_TCP_OPTION_SPACE; space = min_t(size_t, space, fo->size); @@ -3560,10 +3772,9 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn) /* limit to order-0 allocations */ space = min_t(size_t, space, SKB_MAX_HEAD(MAX_TCP_HEADER)); - syn_data = sk_stream_alloc_skb(sk, space, sk->sk_allocation, false); + syn_data = tcp_stream_alloc_skb(sk, space, sk->sk_allocation, false); if (!syn_data) goto fallback; - syn_data->ip_summed = CHECKSUM_PARTIAL; memcpy(syn_data->cb, syn->cb, sizeof(syn->cb)); if (space) { int copied = copy_from_iter(skb_put(syn_data, space), space, @@ -3590,7 +3801,7 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn) err = tcp_transmit_skb(sk, syn_data, 1, sk->sk_allocation); - syn->skb_mstamp_ns = syn_data->skb_mstamp_ns; + skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, true); /* Now full SYN+DATA was cloned and sent (or not), * remove the SYN from the original skb (syn_data) @@ -3641,7 +3852,7 @@ int tcp_connect(struct sock *sk) return 0; } - buff = sk_stream_alloc_skb(sk, 0, sk->sk_allocation, true); + buff = tcp_stream_alloc_skb(sk, 0, sk->sk_allocation, true); if (unlikely(!buff)) return -ENOBUFS; @@ -3712,16 +3923,15 @@ void tcp_send_delayed_ack(struct sock *sk) ato = min(ato, max_ato); } + ato = min_t(u32, ato, inet_csk(sk)->icsk_delack_max); + /* Stay within the limit we were given */ timeout = jiffies + ato; /* Use new timeout only if there wasn't a older one earlier. */ if (icsk->icsk_ack.pending & ICSK_ACK_TIMER) { - /* If delack timer was blocked or is about to expire, - * send ACK now. - */ - if (icsk->icsk_ack.blocked || - time_before_eq(icsk->icsk_ack.timeout, jiffies + (ato >> 2))) { + /* If delack timer is about to expire, send ACK now. */ + if (time_before_eq(icsk->icsk_ack.timeout, jiffies + (ato >> 2))) { tcp_send_ack(sk); return; } @@ -3750,10 +3960,15 @@ void __tcp_send_ack(struct sock *sk, u32 rcv_nxt) buff = alloc_skb(MAX_TCP_HEADER, sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN)); if (unlikely(!buff)) { + struct inet_connection_sock *icsk = inet_csk(sk); + unsigned long delay; + + delay = TCP_DELACK_MAX << icsk->icsk_ack.retry; + if (delay < TCP_RTO_MAX) + icsk->icsk_ack.retry++; inet_csk_schedule_ack(sk); - inet_csk(sk)->icsk_ack.ato = TCP_ATO_MIN; - inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, - TCP_DELACK_MAX, TCP_RTO_MAX); + icsk->icsk_ack.ato = TCP_ATO_MIN; + inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, delay, TCP_RTO_MAX); return; } @@ -3881,12 +4096,13 @@ void tcp_send_probe0(struct sock *sk) /* Cancel probe timer, if it is not required. */ icsk->icsk_probes_out = 0; icsk->icsk_backoff = 0; + icsk->icsk_probes_tstamp = 0; return; } icsk->icsk_probes_out++; if (err <= 0) { - if (icsk->icsk_backoff < net->ipv4.sysctl_tcp_retries2) + if (icsk->icsk_backoff < READ_ONCE(net->ipv4.sysctl_tcp_retries2)) icsk->icsk_backoff++; timeout = tcp_probe0_when(sk, TCP_RTO_MAX); } else { @@ -3895,7 +4111,9 @@ void tcp_send_probe0(struct sock *sk) */ timeout = TCP_RESOURCE_PROBE_INTERVAL; } - tcp_reset_xmit_timer(sk, ICSK_TIME_PROBE0, timeout, TCP_RTO_MAX, NULL); + + timeout = tcp_clamp_probe0_to_user_timeout(sk, timeout); + tcp_reset_xmit_timer(sk, ICSK_TIME_PROBE0, timeout, TCP_RTO_MAX); } int tcp_rtx_synack(const struct sock *sk, struct request_sock *req) @@ -3904,11 +4122,14 @@ int tcp_rtx_synack(const struct sock *sk, struct request_sock *req) struct flowi fl; int res; - tcp_rsk(req)->txhash = net_tx_rndhash(); - res = af_ops->send_synack(sk, NULL, &fl, req, NULL, TCP_SYNACK_NORMAL); + /* Paired with WRITE_ONCE() in sock_setsockopt() */ + if (READ_ONCE(sk->sk_txrehash) == SOCK_TXREHASH_ENABLED) + tcp_rsk(req)->txhash = net_tx_rndhash(); + res = af_ops->send_synack(sk, NULL, &fl, req, NULL, TCP_SYNACK_NORMAL, + NULL); if (!res) { - __TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS); - __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNRETRANS); + TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS); + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNRETRANS); if (unlikely(tcp_passive_fastopen(sk))) tcp_sk(sk)->total_retrans++; trace_tcp_retransmit_synack(sk, req); diff --git a/net/ipv4/tcp_rate.c b/net/ipv4/tcp_rate.c index 0de693565963..a8f6d9d06f2e 100644 --- a/net/ipv4/tcp_rate.c +++ b/net/ipv4/tcp_rate.c @@ -65,6 +65,7 @@ void tcp_rate_skb_sent(struct sock *sk, struct sk_buff *skb) TCP_SKB_CB(skb)->tx.first_tx_mstamp = tp->first_tx_mstamp; TCP_SKB_CB(skb)->tx.delivered_mstamp = tp->delivered_mstamp; TCP_SKB_CB(skb)->tx.delivered = tp->delivered; + TCP_SKB_CB(skb)->tx.delivered_ce = tp->delivered_ce; TCP_SKB_CB(skb)->tx.is_app_limited = tp->app_limited ? 1 : 0; } @@ -73,26 +74,32 @@ void tcp_rate_skb_sent(struct sock *sk, struct sk_buff *skb) * * If an ACK (s)acks multiple skbs (e.g., stretched-acks), this function is * called multiple times. We favor the information from the most recently - * sent skb, i.e., the skb with the highest prior_delivered count. + * sent skb, i.e., the skb with the most recently sent time and the highest + * sequence. */ void tcp_rate_skb_delivered(struct sock *sk, struct sk_buff *skb, struct rate_sample *rs) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_skb_cb *scb = TCP_SKB_CB(skb); + u64 tx_tstamp; if (!scb->tx.delivered_mstamp) return; + tx_tstamp = tcp_skb_timestamp_us(skb); if (!rs->prior_delivered || - after(scb->tx.delivered, rs->prior_delivered)) { + tcp_skb_sent_after(tx_tstamp, tp->first_tx_mstamp, + scb->end_seq, rs->last_end_seq)) { + rs->prior_delivered_ce = scb->tx.delivered_ce; rs->prior_delivered = scb->tx.delivered; rs->prior_mstamp = scb->tx.delivered_mstamp; rs->is_app_limited = scb->tx.is_app_limited; rs->is_retrans = scb->sacked & TCPCB_RETRANS; + rs->last_end_seq = scb->end_seq; /* Record send time of most recently ACKed packet: */ - tp->first_tx_mstamp = tcp_skb_timestamp_us(skb); + tp->first_tx_mstamp = tx_tstamp; /* Find the duration of the "send phase" of this window: */ rs->interval_us = tcp_stamp_us_delta(tp->first_tx_mstamp, scb->tx.first_tx_mstamp); @@ -138,6 +145,10 @@ void tcp_rate_gen(struct sock *sk, u32 delivered, u32 lost, } rs->delivered = tp->delivered - rs->prior_delivered; + rs->delivered_ce = tp->delivered_ce - rs->prior_delivered_ce; + /* delivered_ce occupies less than 32 bits in the skb control block */ + rs->delivered_ce &= TCPCB_DELIVERED_CE_MASK; + /* Model sending data and receiving ACKs as separate pipeline phases * for a window. Usually the ACK phase is longer, but with ACK * compression the send phase can be longer. To be safe we use the @@ -189,7 +200,7 @@ void tcp_rate_check_app_limited(struct sock *sk) /* Nothing in sending host's qdisc queues or NIC tx queue. */ sk_wmem_alloc_get(sk) < SKB_TRUESIZE(1) && /* We are not limited by CWND. */ - tcp_packets_in_flight(tp) < tp->snd_cwnd && + tcp_packets_in_flight(tp) < tcp_snd_cwnd(tp) && /* All lost packets have been retransmitted. */ tp->lost_out <= tp->retrans_out) tp->app_limited = diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c index fdb715bdd2d1..50abaa941387 100644 --- a/net/ipv4/tcp_recovery.c +++ b/net/ipv4/tcp_recovery.c @@ -2,25 +2,6 @@ #include <linux/tcp.h> #include <net/tcp.h> -void tcp_mark_skb_lost(struct sock *sk, struct sk_buff *skb) -{ - struct tcp_sock *tp = tcp_sk(sk); - - tcp_skb_mark_lost_uncond_verify(tp, skb); - if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) { - /* Account for retransmits that are lost again */ - TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; - tp->retrans_out -= tcp_skb_pcount(skb); - NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT, - tcp_skb_pcount(skb)); - } -} - -static bool tcp_rack_sent_after(u64 t1, u64 t2, u32 seq1, u32 seq2) -{ - return t1 > t2 || (t1 == t2 && after(seq1, seq2)); -} - static u32 tcp_rack_reo_wnd(const struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); @@ -33,7 +14,8 @@ static u32 tcp_rack_reo_wnd(const struct sock *sk) return 0; if (tp->sacked_out >= tp->reordering && - !(sock_net(sk)->ipv4.sysctl_tcp_recovery & TCP_RACK_NO_DUPTHRESH)) + !(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_recovery) & + TCP_RACK_NO_DUPTHRESH)) return 0; } @@ -91,9 +73,9 @@ static void tcp_rack_detect_loss(struct sock *sk, u32 *reo_timeout) !(scb->sacked & TCPCB_SACKED_RETRANS)) continue; - if (!tcp_rack_sent_after(tp->rack.mstamp, - tcp_skb_timestamp_us(skb), - tp->rack.end_seq, scb->end_seq)) + if (!tcp_skb_sent_after(tp->rack.mstamp, + tcp_skb_timestamp_us(skb), + tp->rack.end_seq, scb->end_seq)) break; /* A packet is lost if it has not been s/acked beyond @@ -110,13 +92,13 @@ static void tcp_rack_detect_loss(struct sock *sk, u32 *reo_timeout) } } -void tcp_rack_mark_lost(struct sock *sk) +bool tcp_rack_mark_lost(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); u32 timeout; if (!tp->rack.advanced) - return; + return false; /* Reset the advanced flag to avoid unnecessary queue scanning */ tp->rack.advanced = 0; @@ -126,6 +108,7 @@ void tcp_rack_mark_lost(struct sock *sk) inet_csk_reset_xmit_timer(sk, ICSK_TIME_REO_TIMEOUT, timeout, inet_csk(sk)->icsk_rto); } + return !!timeout; } /* Record the most recently (re)sent time among the (s)acked packets @@ -153,8 +136,8 @@ void tcp_rack_advance(struct tcp_sock *tp, u8 sacked, u32 end_seq, } tp->rack.advanced = 1; tp->rack.rtt_us = rtt_us; - if (tcp_rack_sent_after(xmit_time, tp->rack.mstamp, - end_seq, tp->rack.end_seq)) { + if (tcp_skb_sent_after(xmit_time, tp->rack.mstamp, + end_seq, tp->rack.end_seq)) { tp->rack.mstamp = xmit_time; tp->rack.end_seq = end_seq; } @@ -167,6 +150,7 @@ void tcp_rack_reo_timeout(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); u32 timeout, prior_inflight; + u32 lost = tp->lost; prior_inflight = tcp_packets_in_flight(tp); tcp_rack_detect_loss(sk, &timeout); @@ -174,7 +158,7 @@ void tcp_rack_reo_timeout(struct sock *sk) if (inet_csk(sk)->icsk_ca_state != TCP_CA_Recovery) { tcp_enter_recovery(sk, false); if (!inet_csk(sk)->icsk_ca_ops->cong_control) - tcp_cwnd_reduction(sk, 1, 0); + tcp_cwnd_reduction(sk, 1, tp->lost - lost, 0); } tcp_xmit_retransmit_queue(sk); } @@ -184,7 +168,8 @@ void tcp_rack_reo_timeout(struct sock *sk) /* Updates the RACK's reo_wnd based on DSACK and no. of recoveries. * - * If DSACK is received, increment reo_wnd by min_rtt/4 (upper bounded + * If a DSACK is received that seems like it may have been due to reordering + * triggering fast recovery, increment reo_wnd by min_rtt/4 (upper bounded * by srtt), since there is possibility that spurious retransmission was * due to reordering delay longer than reo_wnd. * @@ -203,7 +188,8 @@ void tcp_rack_update_reo_wnd(struct sock *sk, struct rate_sample *rs) { struct tcp_sock *tp = tcp_sk(sk); - if (sock_net(sk)->ipv4.sysctl_tcp_recovery & TCP_RACK_STATIC_REO_WND || + if ((READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_recovery) & + TCP_RACK_STATIC_REO_WND) || !rs->prior_delivered) return; @@ -246,6 +232,6 @@ void tcp_newreno_mark_lost(struct sock *sk, bool snd_una_advanced) tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, mss, mss, GFP_ATOMIC); - tcp_skb_mark_lost_uncond_verify(tp, skb); + tcp_mark_skb_lost(sk, skb); } } diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c index 471571e1ab26..862b96248a92 100644 --- a/net/ipv4/tcp_scalable.c +++ b/net/ipv4/tcp_scalable.c @@ -10,10 +10,9 @@ #include <net/tcp.h> /* These factors derived from the recommended values in the aer: - * .01 and and 7/8. We use 50 instead of 100 to account for - * delayed ack. + * .01 and 7/8. */ -#define TCP_SCALABLE_AI_CNT 50U +#define TCP_SCALABLE_AI_CNT 100U #define TCP_SCALABLE_MD_SCALE 3 static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked) @@ -23,18 +22,20 @@ static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked) if (!tcp_is_cwnd_limited(sk)) return; - if (tcp_in_slow_start(tp)) - tcp_slow_start(tp, acked); - else - tcp_cong_avoid_ai(tp, min(tp->snd_cwnd, TCP_SCALABLE_AI_CNT), - 1); + if (tcp_in_slow_start(tp)) { + acked = tcp_slow_start(tp, acked); + if (!acked) + return; + } + tcp_cong_avoid_ai(tp, min(tcp_snd_cwnd(tp), TCP_SCALABLE_AI_CNT), + acked); } static u32 tcp_scalable_ssthresh(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); - return max(tp->snd_cwnd - (tp->snd_cwnd>>TCP_SCALABLE_MD_SCALE), 2U); + return max(tcp_snd_cwnd(tp) - (tcp_snd_cwnd(tp)>>TCP_SCALABLE_MD_SCALE), 2U); } static struct tcp_congestion_ops tcp_scalable __read_mostly = { diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index c3f26dcd6704..cb79127f45c3 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -40,6 +40,24 @@ static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk) return min_t(u32, icsk->icsk_rto, msecs_to_jiffies(remaining)); } +u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + u32 remaining; + s32 elapsed; + + if (!icsk->icsk_user_timeout || !icsk->icsk_probes_tstamp) + return when; + + elapsed = tcp_jiffies32 - icsk->icsk_probes_tstamp; + if (unlikely(elapsed < 0)) + elapsed = 0; + remaining = msecs_to_jiffies(icsk->icsk_user_timeout) - elapsed; + remaining = max_t(u32, remaining, TCP_TIMEOUT_MIN); + + return min_t(u32, remaining, when); +} + /** * tcp_write_err() - close socket and save error info * @sk: The socket the error has appeared on. @@ -50,7 +68,7 @@ static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk) static void tcp_write_err(struct sock *sk) { sk->sk_err = sk->sk_err_soft ? : ETIMEDOUT; - sk->sk_error_report(sk); + sk_error_report(sk); tcp_write_queue_purge(sk); tcp_done(sk); @@ -125,7 +143,7 @@ static int tcp_out_of_resources(struct sock *sk, bool do_reset) */ static int tcp_orphan_retries(struct sock *sk, bool alive) { - int retries = sock_net(sk)->ipv4.sysctl_tcp_orphan_retries; /* May be zero. */ + int retries = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_orphan_retries); /* May be zero. */ /* We know from an ICMP that something is wrong. */ if (sk->sk_err_soft && !alive) @@ -145,7 +163,7 @@ static void tcp_mtu_probing(struct inet_connection_sock *icsk, struct sock *sk) int mss; /* Black hole detection */ - if (!net->ipv4.sysctl_tcp_mtu_probing) + if (!READ_ONCE(net->ipv4.sysctl_tcp_mtu_probing)) return; if (!icsk->icsk_mtup.enabled) { @@ -153,9 +171,9 @@ static void tcp_mtu_probing(struct inet_connection_sock *icsk, struct sock *sk) icsk->icsk_mtup.probe_timestamp = tcp_jiffies32; } else { mss = tcp_mtu_to_mss(sk, icsk->icsk_mtup.search_low) >> 1; - mss = min(net->ipv4.sysctl_tcp_base_mss, mss); - mss = max(mss, net->ipv4.sysctl_tcp_mtu_probe_floor); - mss = max(mss, net->ipv4.sysctl_tcp_min_snd_mss); + mss = min(READ_ONCE(net->ipv4.sysctl_tcp_base_mss), mss); + mss = max(mss, READ_ONCE(net->ipv4.sysctl_tcp_mtu_probe_floor)); + mss = max(mss, READ_ONCE(net->ipv4.sysctl_tcp_min_snd_mss)); icsk->icsk_mtup.search_low = tcp_mss_to_mtu(sk, mss); } tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); @@ -219,30 +237,20 @@ static int tcp_write_timeout(struct sock *sk) int retry_until; if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) { - if (icsk->icsk_retransmits) { - dst_negative_advice(sk); - } else { - sk_rethink_txhash(sk); - tp->timeout_rehash++; - __NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPTIMEOUTREHASH); - } - retry_until = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries; + if (icsk->icsk_retransmits) + __dst_negative_advice(sk); + retry_until = icsk->icsk_syn_retries ? : + READ_ONCE(net->ipv4.sysctl_tcp_syn_retries); expired = icsk->icsk_retransmits >= retry_until; } else { - if (retransmits_timed_out(sk, net->ipv4.sysctl_tcp_retries1, 0)) { + if (retransmits_timed_out(sk, READ_ONCE(net->ipv4.sysctl_tcp_retries1), 0)) { /* Black hole detection */ tcp_mtu_probing(icsk, sk); - dst_negative_advice(sk); - } else { - sk_rethink_txhash(sk); - tp->timeout_rehash++; - __NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPTIMEOUTREHASH); + __dst_negative_advice(sk); } - retry_until = net->ipv4.sysctl_tcp_retries2; + retry_until = READ_ONCE(net->ipv4.sysctl_tcp_retries2); if (sock_flag(sk, SOCK_DEAD)) { const bool alive = icsk->icsk_rto < TCP_RTO_MAX; @@ -270,6 +278,11 @@ static int tcp_write_timeout(struct sock *sk) return 1; } + if (sk_rethink_txhash(sk)) { + tp->timeout_rehash++; + __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPTIMEOUTREHASH); + } + return 0; } @@ -278,15 +291,13 @@ void tcp_delack_timer_handler(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); - sk_mem_reclaim_partial(sk); - if (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) || !(icsk->icsk_ack.pending & ICSK_ACK_TIMER)) - goto out; + return; if (time_after(icsk->icsk_ack.timeout, jiffies)) { sk_reset_timer(sk, &icsk->icsk_delack_timer, icsk->icsk_ack.timeout); - goto out; + return; } icsk->icsk_ack.pending &= ~ICSK_ACK_TIMER; @@ -305,16 +316,12 @@ void tcp_delack_timer_handler(struct sock *sk) tcp_send_ack(sk); __NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKS); } - -out: - if (tcp_under_memory_pressure(sk)) - sk_mem_reclaim(sk); } /** * tcp_delack_timer() - The TCP delayed ACK timeout handler - * @data: Pointer to the current socket. (gets casted to struct sock *) + * @t: Pointer to the timer. (gets casted to struct sock *) * * This function gets (indirectly) called when the kernel timer for a TCP packet * of this socket expires. Calls tcp_delack_timer_handler() to do the actual work. @@ -331,7 +338,6 @@ static void tcp_delack_timer(struct timer_list *t) if (!sock_owned_by_user(sk)) { tcp_delack_timer_handler(sk); } else { - icsk->icsk_ack.blocked = 1; __NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOCKED); /* deleguate our work to tcp_release_cb() */ if (!test_and_set_bit(TCP_DELACK_TIMER_DEFERRED, &sk->sk_tsq_flags)) @@ -350,6 +356,7 @@ static void tcp_probe_timer(struct sock *sk) if (tp->packets_out || !skb) { icsk->icsk_probes_out = 0; + icsk->icsk_probes_tstamp = 0; return; } @@ -361,15 +368,14 @@ static void tcp_probe_timer(struct sock *sk) * corresponding system limit. We also implement similar policy when * we use RTO to probe window in tcp_retransmit_timer(). */ - if (icsk->icsk_user_timeout) { - u32 elapsed = tcp_model_timeout(sk, icsk->icsk_probes_out, - tcp_probe0_base(sk)); - - if (elapsed >= icsk->icsk_user_timeout) - goto abort; - } - - max_probes = sock_net(sk)->ipv4.sysctl_tcp_retries2; + if (!icsk->icsk_probes_tstamp) + icsk->icsk_probes_tstamp = tcp_jiffies32; + else if (icsk->icsk_user_timeout && + (s32)(tcp_jiffies32 - icsk->icsk_probes_tstamp) >= + msecs_to_jiffies(icsk->icsk_user_timeout)) + goto abort; + + max_probes = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_retries2); if (sock_flag(sk, SOCK_DEAD)) { const bool alive = inet_csk_rto_backoff(icsk, TCP_RTO_MAX) < TCP_RTO_MAX; @@ -395,12 +401,15 @@ abort: tcp_write_err(sk); static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) { struct inet_connection_sock *icsk = inet_csk(sk); - int max_retries = icsk->icsk_syn_retries ? : - sock_net(sk)->ipv4.sysctl_tcp_synack_retries + 1; /* add one more retry for fastopen */ struct tcp_sock *tp = tcp_sk(sk); + int max_retries; req->rsk_ops->syn_ack_timeout(req); + /* add one more retry for fastopen */ + max_retries = icsk->icsk_syn_retries ? : + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_synack_retries) + 1; + if (req->num_timeout >= max_retries) { tcp_write_err(sk); return; @@ -419,7 +428,7 @@ static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) if (!tp->retrans_stamp) tp->retrans_stamp = tcp_time_stamp(tp); inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, - TCP_TIMEOUT_INIT << req->num_timeout, TCP_RTO_MAX); + req->timeout << req->num_timeout, TCP_RTO_MAX); } @@ -430,7 +439,7 @@ static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) * This function gets called when the kernel timer for a TCP packet * of this socket expires. * - * It handles retransmission, timer adjustment and other necesarry measures. + * It handles retransmission, timer adjustment and other necessary measures. * * Returns: Nothing (void) */ @@ -563,7 +572,7 @@ out_reset_timer: * linear-timeout retransmissions into a black hole */ if (sk->sk_state == TCP_ESTABLISHED && - (tp->thin_lto || net->ipv4.sysctl_tcp_thin_linear_timeouts) && + (tp->thin_lto || READ_ONCE(net->ipv4.sysctl_tcp_thin_linear_timeouts)) && tcp_stream_is_thin(tp) && icsk->icsk_retransmits <= TCP_THIN_LINEAR_RETRIES) { icsk->icsk_backoff = 0; @@ -574,7 +583,7 @@ out_reset_timer: } inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, tcp_clamp_rto_to_user_timeout(sk), TCP_RTO_MAX); - if (retransmits_timed_out(sk, net->ipv4.sysctl_tcp_retries1 + 1, 0)) + if (retransmits_timed_out(sk, READ_ONCE(net->ipv4.sysctl_tcp_retries1) + 1, 0)) __sk_dst_reset(sk); out:; @@ -589,11 +598,11 @@ void tcp_write_timer_handler(struct sock *sk) if (((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) || !icsk->icsk_pending) - goto out; + return; if (time_after(icsk->icsk_timeout, jiffies)) { sk_reset_timer(sk, &icsk->icsk_retransmit_timer, icsk->icsk_timeout); - goto out; + return; } tcp_mstamp_refresh(tcp_sk(sk)); @@ -615,9 +624,6 @@ void tcp_write_timer_handler(struct sock *sk) tcp_probe_timer(sk); break; } - -out: - sk_mem_reclaim(sk); } static void tcp_write_timer(struct timer_list *t) @@ -732,8 +738,6 @@ static void tcp_keepalive_timer (struct timer_list *t) elapsed = keepalive_time_when(tp) - elapsed; } - sk_mem_reclaim(sk); - resched: inet_csk_reset_keepalive_timer (sk, elapsed); goto out; @@ -753,8 +757,14 @@ static enum hrtimer_restart tcp_compressed_ack_kick(struct hrtimer *timer) bh_lock_sock(sk); if (!sock_owned_by_user(sk)) { - if (tp->compressed_ack > TCP_FASTRETRANS_THRESH) + if (tp->compressed_ack) { + /* Since we have to send one ack finally, + * subtract one from tp->compressed_ack to keep + * LINUX_MIB_TCPACKCOMPRESSED accurate. + */ + tp->compressed_ack--; tcp_send_ack(sk); + } } else { if (!test_and_set_bit(TCP_DELACK_TIMER_DEFERRED, &sk->sk_tsq_flags)) diff --git a/net/ipv4/tcp_ulp.c b/net/ipv4/tcp_ulp.c index 38d3ad141161..9ae50b1bd844 100644 --- a/net/ipv4/tcp_ulp.c +++ b/net/ipv4/tcp_ulp.c @@ -22,7 +22,8 @@ static struct tcp_ulp_ops *tcp_ulp_find(const char *name) { struct tcp_ulp_ops *e; - list_for_each_entry_rcu(e, &tcp_ulp_list, list) { + list_for_each_entry_rcu(e, &tcp_ulp_list, list, + lockdep_is_held(&tcp_ulp_list_lock)) { if (strcmp(e->name, name) == 0) return e; } @@ -104,12 +105,6 @@ void tcp_update_ulp(struct sock *sk, struct proto *proto, { struct inet_connection_sock *icsk = inet_csk(sk); - if (!icsk->icsk_ulp_ops) { - sk->sk_write_space = write_space; - sk->sk_prot = proto; - return; - } - if (icsk->icsk_ulp_ops->update) icsk->icsk_ulp_ops->update(sk, proto, write_space); } @@ -141,6 +136,9 @@ static int __tcp_set_ulp(struct sock *sk, const struct tcp_ulp_ops *ulp_ops) if (icsk->icsk_ulp_ops) goto out_err; + if (sk->sk_socket) + clear_bit(SOCK_SUPPORT_ZC, &sk->sk_socket->flags); + err = ulp_ops->init(sk); if (err) goto out_err; diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index 3f51e781562a..786848ad37ea 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -159,7 +159,7 @@ EXPORT_SYMBOL_GPL(tcp_vegas_cwnd_event); static inline u32 tcp_vegas_ssthresh(struct tcp_sock *tp) { - return min(tp->snd_ssthresh, tp->snd_cwnd); + return min(tp->snd_ssthresh, tcp_snd_cwnd(tp)); } static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) @@ -217,14 +217,14 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) * This is: * (actual rate in segments) * baseRTT */ - target_cwnd = (u64)tp->snd_cwnd * vegas->baseRTT; + target_cwnd = (u64)tcp_snd_cwnd(tp) * vegas->baseRTT; do_div(target_cwnd, rtt); /* Calculate the difference between the window we had, * and the window we would like to have. This quantity * is the "Diff" from the Arizona Vegas papers. */ - diff = tp->snd_cwnd * (rtt-vegas->baseRTT) / vegas->baseRTT; + diff = tcp_snd_cwnd(tp) * (rtt-vegas->baseRTT) / vegas->baseRTT; if (diff > gamma && tcp_in_slow_start(tp)) { /* Going too fast. Time to slow down @@ -238,7 +238,8 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) * truncation robs us of full link * utilization. */ - tp->snd_cwnd = min(tp->snd_cwnd, (u32)target_cwnd+1); + tcp_snd_cwnd_set(tp, min(tcp_snd_cwnd(tp), + (u32)target_cwnd + 1)); tp->snd_ssthresh = tcp_vegas_ssthresh(tp); } else if (tcp_in_slow_start(tp)) { @@ -254,14 +255,14 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) /* The old window was too fast, so * we slow down. */ - tp->snd_cwnd--; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) - 1); tp->snd_ssthresh = tcp_vegas_ssthresh(tp); } else if (diff < alpha) { /* We don't have enough extra packets * in the network, so speed up. */ - tp->snd_cwnd++; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); } else { /* Sending just as fast as we * should be. @@ -269,10 +270,10 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) } } - if (tp->snd_cwnd < 2) - tp->snd_cwnd = 2; - else if (tp->snd_cwnd > tp->snd_cwnd_clamp) - tp->snd_cwnd = tp->snd_cwnd_clamp; + if (tcp_snd_cwnd(tp) < 2) + tcp_snd_cwnd_set(tp, 2); + else if (tcp_snd_cwnd(tp) > tp->snd_cwnd_clamp) + tcp_snd_cwnd_set(tp, tp->snd_cwnd_clamp); tp->snd_ssthresh = tcp_current_ssthresh(sk); } @@ -293,10 +294,10 @@ size_t tcp_vegas_get_info(struct sock *sk, u32 ext, int *attr, const struct vegas *ca = inet_csk_ca(sk); if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) { - info->vegas.tcpv_enabled = ca->doing_vegas_now, - info->vegas.tcpv_rttcnt = ca->cntRTT, - info->vegas.tcpv_rtt = ca->baseRTT, - info->vegas.tcpv_minrtt = ca->minRTT, + info->vegas.tcpv_enabled = ca->doing_vegas_now; + info->vegas.tcpv_rttcnt = ca->cntRTT; + info->vegas.tcpv_rtt = ca->baseRTT; + info->vegas.tcpv_minrtt = ca->minRTT; *attr = INET_DIAG_VEGASINFO; return sizeof(struct tcpvegas_info); diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp_veno.c index 3b36bb1a0dda..366ff6f214b2 100644 --- a/net/ipv4/tcp_veno.c +++ b/net/ipv4/tcp_veno.c @@ -7,7 +7,7 @@ * "TCP Veno: TCP Enhancement for Transmission over Wireless Access Networks." * IEEE Journal on Selected Areas in Communication, * Feb. 2003. - * See http://www.ie.cuhk.edu.hk/fileadmin/staff_upload/soung/Journal/J3.pdf + * See https://www.ie.cuhk.edu.hk/fileadmin/staff_upload/soung/Journal/J3.pdf */ #include <linux/mm.h> @@ -146,42 +146,45 @@ static void tcp_veno_cong_avoid(struct sock *sk, u32 ack, u32 acked) rtt = veno->minrtt; - target_cwnd = (u64)tp->snd_cwnd * veno->basertt; + target_cwnd = (u64)tcp_snd_cwnd(tp) * veno->basertt; target_cwnd <<= V_PARAM_SHIFT; do_div(target_cwnd, rtt); - veno->diff = (tp->snd_cwnd << V_PARAM_SHIFT) - target_cwnd; + veno->diff = (tcp_snd_cwnd(tp) << V_PARAM_SHIFT) - target_cwnd; if (tcp_in_slow_start(tp)) { - /* Slow start. */ - tcp_slow_start(tp, acked); + /* Slow start. */ + acked = tcp_slow_start(tp, acked); + if (!acked) + goto done; + } + + /* Congestion avoidance. */ + if (veno->diff < beta) { + /* In the "non-congestive state", increase cwnd + * every rtt. + */ + tcp_cong_avoid_ai(tp, tcp_snd_cwnd(tp), acked); } else { - /* Congestion avoidance. */ - if (veno->diff < beta) { - /* In the "non-congestive state", increase cwnd - * every rtt. - */ - tcp_cong_avoid_ai(tp, tp->snd_cwnd, 1); - } else { - /* In the "congestive state", increase cwnd - * every other rtt. - */ - if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { - if (veno->inc && - tp->snd_cwnd < tp->snd_cwnd_clamp) { - tp->snd_cwnd++; - veno->inc = 0; - } else - veno->inc = 1; - tp->snd_cwnd_cnt = 0; + /* In the "congestive state", increase cwnd + * every other rtt. + */ + if (tp->snd_cwnd_cnt >= tcp_snd_cwnd(tp)) { + if (veno->inc && + tcp_snd_cwnd(tp) < tp->snd_cwnd_clamp) { + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) + 1); + veno->inc = 0; } else - tp->snd_cwnd_cnt++; - } + veno->inc = 1; + tp->snd_cwnd_cnt = 0; + } else + tp->snd_cwnd_cnt += acked; } - if (tp->snd_cwnd < 2) - tp->snd_cwnd = 2; - else if (tp->snd_cwnd > tp->snd_cwnd_clamp) - tp->snd_cwnd = tp->snd_cwnd_clamp; +done: + if (tcp_snd_cwnd(tp) < 2) + tcp_snd_cwnd_set(tp, 2); + else if (tcp_snd_cwnd(tp) > tp->snd_cwnd_clamp) + tcp_snd_cwnd_set(tp, tp->snd_cwnd_clamp); } /* Wipe the slate clean for the next rtt. */ /* veno->cntrtt = 0; */ @@ -196,10 +199,10 @@ static u32 tcp_veno_ssthresh(struct sock *sk) if (veno->diff < beta) /* in "non-congestive state", cut cwnd by 1/5 */ - return max(tp->snd_cwnd * 4 / 5, 2U); + return max(tcp_snd_cwnd(tp) * 4 / 5, 2U); else /* in "congestive state", cut cwnd by 1/2 */ - return max(tp->snd_cwnd >> 1U, 2U); + return max(tcp_snd_cwnd(tp) >> 1U, 2U); } static struct tcp_congestion_ops tcp_veno __read_mostly = { diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c index b2e05c4cea00..c6e97141eef2 100644 --- a/net/ipv4/tcp_westwood.c +++ b/net/ipv4/tcp_westwood.c @@ -244,7 +244,8 @@ static void tcp_westwood_event(struct sock *sk, enum tcp_ca_event event) switch (event) { case CA_EVENT_COMPLETE_CWR: - tp->snd_cwnd = tp->snd_ssthresh = tcp_westwood_bw_rttmin(sk); + tp->snd_ssthresh = tcp_westwood_bw_rttmin(sk); + tcp_snd_cwnd_set(tp, tp->snd_ssthresh); break; case CA_EVENT_LOSS: tp->snd_ssthresh = tcp_westwood_bw_rttmin(sk); diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp_yeah.c index e00570dd0a69..18b07ff5d20e 100644 --- a/net/ipv4/tcp_yeah.c +++ b/net/ipv4/tcp_yeah.c @@ -36,8 +36,6 @@ struct yeah { u32 reno_count; u32 fast_count; - - u32 pkts_acked; }; static void tcp_yeah_init(struct sock *sk) @@ -57,18 +55,6 @@ static void tcp_yeah_init(struct sock *sk) tp->snd_cwnd_clamp = min_t(u32, tp->snd_cwnd_clamp, 0xffffffff/128); } -static void tcp_yeah_pkts_acked(struct sock *sk, - const struct ack_sample *sample) -{ - const struct inet_connection_sock *icsk = inet_csk(sk); - struct yeah *yeah = inet_csk_ca(sk); - - if (icsk->icsk_ca_state == TCP_CA_Open) - yeah->pkts_acked = sample->pkts_acked; - - tcp_vegas_pkts_acked(sk, sample); -} - static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); @@ -77,24 +63,19 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) if (!tcp_is_cwnd_limited(sk)) return; - if (tcp_in_slow_start(tp)) - tcp_slow_start(tp, acked); + if (tcp_in_slow_start(tp)) { + acked = tcp_slow_start(tp, acked); + if (!acked) + goto do_vegas; + } - else if (!yeah->doing_reno_now) { + if (!yeah->doing_reno_now) { /* Scalable */ - - tp->snd_cwnd_cnt += yeah->pkts_acked; - if (tp->snd_cwnd_cnt > min(tp->snd_cwnd, TCP_SCALABLE_AI_CNT)) { - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - tp->snd_cwnd_cnt = 0; - } - - yeah->pkts_acked = 1; - + tcp_cong_avoid_ai(tp, min(tcp_snd_cwnd(tp), TCP_SCALABLE_AI_CNT), + acked); } else { /* Reno */ - tcp_cong_avoid_ai(tp, tp->snd_cwnd, 1); + tcp_cong_avoid_ai(tp, tcp_snd_cwnd(tp), acked); } /* The key players are v_vegas.beg_snd_una and v_beg_snd_nxt. @@ -118,7 +99,7 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) * of bytes we send in an RTT is often less than our cwnd will allow. * So we keep track of our cwnd separately, in v_beg_snd_cwnd. */ - +do_vegas: if (after(ack, yeah->vegas.beg_snd_nxt)) { /* We do the Vegas calculations only if we got enough RTT * samples that we can be reasonably sure that we got @@ -149,7 +130,7 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) /* Compute excess number of packets above bandwidth * Avoid doing full 64 bit divide. */ - bw = tp->snd_cwnd; + bw = tcp_snd_cwnd(tp); bw *= rtt - yeah->vegas.baseRTT; do_div(bw, rtt); queue = bw; @@ -157,20 +138,20 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) if (queue > TCP_YEAH_ALPHA || rtt - yeah->vegas.baseRTT > (yeah->vegas.baseRTT / TCP_YEAH_PHY)) { if (queue > TCP_YEAH_ALPHA && - tp->snd_cwnd > yeah->reno_count) { + tcp_snd_cwnd(tp) > yeah->reno_count) { u32 reduction = min(queue / TCP_YEAH_GAMMA , - tp->snd_cwnd >> TCP_YEAH_EPSILON); + tcp_snd_cwnd(tp) >> TCP_YEAH_EPSILON); - tp->snd_cwnd -= reduction; + tcp_snd_cwnd_set(tp, tcp_snd_cwnd(tp) - reduction); - tp->snd_cwnd = max(tp->snd_cwnd, - yeah->reno_count); + tcp_snd_cwnd_set(tp, max(tcp_snd_cwnd(tp), + yeah->reno_count)); - tp->snd_ssthresh = tp->snd_cwnd; + tp->snd_ssthresh = tcp_snd_cwnd(tp); } if (yeah->reno_count <= 2) - yeah->reno_count = max(tp->snd_cwnd>>1, 2U); + yeah->reno_count = max(tcp_snd_cwnd(tp)>>1, 2U); else yeah->reno_count++; @@ -195,7 +176,7 @@ static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) */ yeah->vegas.beg_snd_una = yeah->vegas.beg_snd_nxt; yeah->vegas.beg_snd_nxt = tp->snd_nxt; - yeah->vegas.beg_snd_cwnd = tp->snd_cwnd; + yeah->vegas.beg_snd_cwnd = tcp_snd_cwnd(tp); /* Wipe the slate clean for the next RTT. */ yeah->vegas.cntRTT = 0; @@ -212,16 +193,16 @@ static u32 tcp_yeah_ssthresh(struct sock *sk) if (yeah->doing_reno_now < TCP_YEAH_RHO) { reduction = yeah->lastQ; - reduction = min(reduction, max(tp->snd_cwnd>>1, 2U)); + reduction = min(reduction, max(tcp_snd_cwnd(tp)>>1, 2U)); - reduction = max(reduction, tp->snd_cwnd >> TCP_YEAH_DELTA); + reduction = max(reduction, tcp_snd_cwnd(tp) >> TCP_YEAH_DELTA); } else - reduction = max(tp->snd_cwnd>>1, 2U); + reduction = max(tcp_snd_cwnd(tp)>>1, 2U); yeah->fast_count = 0; yeah->reno_count = max(yeah->reno_count>>1, 2U); - return max_t(int, tp->snd_cwnd - reduction, 2); + return max_t(int, tcp_snd_cwnd(tp) - reduction, 2); } static struct tcp_congestion_ops tcp_yeah __read_mostly = { @@ -232,7 +213,7 @@ static struct tcp_congestion_ops tcp_yeah __read_mostly = { .set_state = tcp_vegas_state, .cwnd_event = tcp_vegas_cwnd_event, .get_info = tcp_vegas_get_info, - .pkts_acked = tcp_yeah_pkts_acked, + .pkts_acked = tcp_vegas_pkts_acked, .owner = THIS_MODULE, .name = "yeah", @@ -240,7 +221,7 @@ static struct tcp_congestion_ops tcp_yeah __read_mostly = { static int __init tcp_yeah_register(void) { - BUG_ON(sizeof(struct yeah) > ICSK_CA_PRIV_SIZE); + BUILD_BUG_ON(sizeof(struct yeah) > ICSK_CA_PRIV_SIZE); tcp_register_congestion_control(&tcp_yeah); return 0; } diff --git a/net/ipv4/tunnel4.c b/net/ipv4/tunnel4.c index c4b2ccbeba04..5048c47c79b2 100644 --- a/net/ipv4/tunnel4.c +++ b/net/ipv4/tunnel4.c @@ -110,6 +110,33 @@ drop: return 0; } +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) +static int tunnel4_rcv_cb(struct sk_buff *skb, u8 proto, int err) +{ + struct xfrm_tunnel __rcu *head; + struct xfrm_tunnel *handler; + int ret; + + head = (proto == IPPROTO_IPIP) ? tunnel4_handlers : tunnel64_handlers; + + for_each_tunnel_rcu(head, handler) { + if (handler->cb_handler) { + ret = handler->cb_handler(skb, err); + if (ret <= 0) + return ret; + } + } + + return 0; +} + +static const struct xfrm_input_afinfo tunnel4_input_afinfo = { + .family = AF_INET, + .is_ipip = true, + .callback = tunnel4_rcv_cb, +}; +#endif + #if IS_ENABLED(CONFIG_IPV6) static int tunnel64_rcv(struct sk_buff *skb) { @@ -191,7 +218,6 @@ static const struct net_protocol tunnel4_protocol = { .handler = tunnel4_rcv, .err_handler = tunnel4_err, .no_policy = 1, - .netns_ok = 1, }; #if IS_ENABLED(CONFIG_IPV6) @@ -199,7 +225,6 @@ static const struct net_protocol tunnel64_protocol = { .handler = tunnel64_rcv, .err_handler = tunnel64_err, .no_policy = 1, - .netns_ok = 1, }; #endif @@ -208,7 +233,6 @@ static const struct net_protocol tunnelmpls4_protocol = { .handler = tunnelmpls4_rcv, .err_handler = tunnelmpls4_err, .no_policy = 1, - .netns_ok = 1, }; #endif @@ -231,6 +255,18 @@ static int __init tunnel4_init(void) goto err; } #endif +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) + if (xfrm_input_register_afinfo(&tunnel4_input_afinfo)) { + inet_del_protocol(&tunnel4_protocol, IPPROTO_IPIP); +#if IS_ENABLED(CONFIG_IPV6) + inet_del_protocol(&tunnel64_protocol, IPPROTO_IPV6); +#endif +#if IS_ENABLED(CONFIG_MPLS) + inet_del_protocol(&tunnelmpls4_protocol, IPPROTO_MPLS); +#endif + goto err; + } +#endif return 0; err: @@ -240,6 +276,10 @@ err: static void __exit tunnel4_fini(void) { +#if IS_ENABLED(CONFIG_INET_XFRM_TUNNEL) + if (xfrm_input_unregister_afinfo(&tunnel4_input_afinfo)) + pr_err("tunnel4 close: can't remove input afinfo\n"); +#endif #if IS_ENABLED(CONFIG_MPLS) if (inet_del_protocol(&tunnelmpls4_protocol, IPPROTO_MPLS)) pr_err("tunnelmpls4 close: can't remove protocol\n"); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 08a41f1e1cd2..6a320a614e54 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -74,11 +74,11 @@ #define pr_fmt(fmt) "UDP: " fmt +#include <linux/bpf-cgroup.h> #include <linux/uaccess.h> #include <asm/ioctls.h> #include <linux/memblock.h> #include <linux/highmem.h> -#include <linux/swap.h> #include <linux/types.h> #include <linux/fcntl.h> #include <linux/module.h> @@ -106,12 +106,16 @@ #include <net/xfrm.h> #include <trace/events/udp.h> #include <linux/static_key.h> +#include <linux/btf_ids.h> #include <trace/events/skb.h> #include <net/busy_poll.h> #include "udp_impl.h" #include <net/sock_reuseport.h> #include <net/addrconf.h> #include <net/udp_tunnel.h> +#if IS_ENABLED(CONFIG_IPV6) +#include <net/ipv6_stubs.h> +#endif struct udp_table udp_table __read_mostly; EXPORT_SYMBOL(udp_table); @@ -119,8 +123,10 @@ EXPORT_SYMBOL(udp_table); long sysctl_udp_mem[3] __read_mostly; EXPORT_SYMBOL(sysctl_udp_mem); -atomic_long_t udp_memory_allocated; +atomic_long_t udp_memory_allocated ____cacheline_aligned_in_smp; EXPORT_SYMBOL(udp_memory_allocated); +DEFINE_PER_CPU(int, udp_memory_per_cpu_fw_alloc); +EXPORT_PER_CPU_SYMBOL_GPL(udp_memory_per_cpu_fw_alloc); #define MAX_UDP_PORTS 65536 #define PORTS_PER_CHAIN (MAX_UDP_PORTS / UDP_HTABLE_SIZE_MIN) @@ -240,7 +246,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum, inet_get_local_port_range(net, &low, &high); remaining = (high - low) + 1; - rand = prandom_u32(); + rand = get_random_u32(); first = reciprocal_scale(rand, remaining) + low; /* * force rand to be an odd multiple of UDP_HTABLE_SIZE @@ -386,7 +392,8 @@ static int compute_score(struct sock *sk, struct net *net, dif, sdif); if (!dev_match) return -1; - score += 4; + if (sk->sk_bound_dev_if) + score += 4; if (READ_ONCE(sk->sk_incoming_cpu) == raw_smp_processor_id()) score++; @@ -405,6 +412,22 @@ static u32 udp_ehashfn(const struct net *net, const __be32 laddr, udp_ehash_secret + net_hash_mix(net)); } +static struct sock *lookup_reuseport(struct net *net, struct sock *sk, + struct sk_buff *skb, + __be32 saddr, __be16 sport, + __be32 daddr, unsigned short hnum) +{ + struct sock *reuse_sk = NULL; + u32 hash; + + if (sk->sk_reuseport && sk->sk_state != TCP_ESTABLISHED) { + hash = udp_ehashfn(net, daddr, hnum, saddr, sport); + reuse_sk = reuseport_select_sock(sk, hash, skb, + sizeof(struct udphdr)); + } + return reuse_sk; +} + /* called with rcu_read_lock() */ static struct sock *udp4_lib_lookup2(struct net *net, __be32 saddr, __be16 sport, @@ -415,7 +438,6 @@ static struct sock *udp4_lib_lookup2(struct net *net, { struct sock *sk, *result; int score, badness; - u32 hash = 0; result = NULL; badness = 0; @@ -423,22 +445,42 @@ static struct sock *udp4_lib_lookup2(struct net *net, score = compute_score(sk, net, saddr, sport, daddr, hnum, dif, sdif); if (score > badness) { - if (sk->sk_reuseport && - sk->sk_state != TCP_ESTABLISHED) { - hash = udp_ehashfn(net, daddr, hnum, - saddr, sport); - result = reuseport_select_sock(sk, hash, skb, - sizeof(struct udphdr)); - if (result && !reuseport_has_conns(sk, false)) - return result; - } + result = lookup_reuseport(net, sk, skb, + saddr, sport, daddr, hnum); + /* Fall back to scoring if group has connections */ + if (result && !reuseport_has_conns(sk)) + return result; + + result = result ? : sk; badness = score; - result = sk; } } return result; } +static struct sock *udp4_lookup_run_bpf(struct net *net, + struct udp_table *udptable, + struct sk_buff *skb, + __be32 saddr, __be16 sport, + __be32 daddr, u16 hnum, const int dif) +{ + struct sock *sk, *reuse_sk; + bool no_reuseport; + + if (udptable != &udp_table) + return NULL; /* only UDP is supported */ + + no_reuseport = bpf_sk_lookup_run_v4(net, IPPROTO_UDP, saddr, sport, + daddr, hnum, dif, &sk); + if (no_reuseport || IS_ERR_OR_NULL(sk)) + return sk; + + reuse_sk = lookup_reuseport(net, sk, skb, saddr, sport, daddr, hnum); + if (reuse_sk) + sk = reuse_sk; + return sk; +} + /* UDP is nearly always wildcards out the wazoo, it makes no sense to try * harder than this. -DaveM */ @@ -446,27 +488,45 @@ struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport, __be32 daddr, __be16 dport, int dif, int sdif, struct udp_table *udptable, struct sk_buff *skb) { - struct sock *result; unsigned short hnum = ntohs(dport); unsigned int hash2, slot2; struct udp_hslot *hslot2; + struct sock *result, *sk; hash2 = ipv4_portaddr_hash(net, daddr, hnum); slot2 = hash2 & udptable->mask; hslot2 = &udptable->hash2[slot2]; + /* Lookup connected or non-wildcard socket */ result = udp4_lib_lookup2(net, saddr, sport, daddr, hnum, dif, sdif, hslot2, skb); - if (!result) { - hash2 = ipv4_portaddr_hash(net, htonl(INADDR_ANY), hnum); - slot2 = hash2 & udptable->mask; - hslot2 = &udptable->hash2[slot2]; - - result = udp4_lib_lookup2(net, saddr, sport, - htonl(INADDR_ANY), hnum, dif, sdif, - hslot2, skb); + if (!IS_ERR_OR_NULL(result) && result->sk_state == TCP_ESTABLISHED) + goto done; + + /* Lookup redirect from BPF */ + if (static_branch_unlikely(&bpf_sk_lookup_enabled)) { + sk = udp4_lookup_run_bpf(net, udptable, skb, + saddr, sport, daddr, hnum, dif); + if (sk) { + result = sk; + goto done; + } } + + /* Got non-wildcard socket or error on first lookup */ + if (result) + goto done; + + /* Lookup wildcard sockets */ + hash2 = ipv4_portaddr_hash(net, htonl(INADDR_ANY), hnum); + slot2 = hash2 & udptable->mask; + hslot2 = &udptable->hash2[slot2]; + + result = udp4_lib_lookup2(net, saddr, sport, + htonl(INADDR_ANY), hnum, dif, sdif, + hslot2, skb); +done: if (IS_ERR(result)) return NULL; return result; @@ -484,7 +544,7 @@ static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb, inet_sdif(skb), udptable, skb); } -struct sock *udp4_lib_lookup_skb(struct sk_buff *skb, +struct sock *udp4_lib_lookup_skb(const struct sk_buff *skb, __be16 sport, __be16 dport) { const struct iphdr *iph = ip_hdr(skb); @@ -493,7 +553,6 @@ struct sock *udp4_lib_lookup_skb(struct sk_buff *skb, iph->daddr, dport, inet_iif(skb), inet_sdif(skb), &udp_table, NULL); } -EXPORT_SYMBOL_GPL(udp4_lib_lookup_skb); /* Must be called under rcu_read_lock(). * Does increment socket refcount. @@ -540,6 +599,12 @@ void udp_encap_enable(void) } EXPORT_SYMBOL(udp_encap_enable); +void udp_encap_disable(void) +{ + static_branch_dec(&udp_encap_needed_key); +} +EXPORT_SYMBOL(udp_encap_disable); + /* Handler for tunnels with arbitrary destination ports: no socket lookup, go * through error handlers in encapsulations looking for a match. */ @@ -583,10 +648,12 @@ static struct sock *__udp4_lib_err_encap(struct net *net, const struct iphdr *iph, struct udphdr *uh, struct udp_table *udptable, + struct sock *sk, struct sk_buff *skb, u32 info) { + int (*lookup)(struct sock *sk, struct sk_buff *skb); int network_offset, transport_offset; - struct sock *sk; + struct udp_sock *up; network_offset = skb_network_offset(skb); transport_offset = skb_transport_offset(skb); @@ -597,18 +664,28 @@ static struct sock *__udp4_lib_err_encap(struct net *net, /* Transport header needs to point to the UDP header */ skb_set_transport_header(skb, iph->ihl << 2); + if (sk) { + up = udp_sk(sk); + + lookup = READ_ONCE(up->encap_err_lookup); + if (lookup && lookup(sk, skb)) + sk = NULL; + + goto out; + } + sk = __udp4_lib_lookup(net, iph->daddr, uh->source, iph->saddr, uh->dest, skb->dev->ifindex, 0, udptable, NULL); if (sk) { - int (*lookup)(struct sock *sk, struct sk_buff *skb); - struct udp_sock *up = udp_sk(sk); + up = udp_sk(sk); lookup = READ_ONCE(up->encap_err_lookup); if (!lookup || lookup(sk, skb)) sk = NULL; } +out: if (!sk) sk = ERR_PTR(__udp4_lib_err_encap_no_sk(skb, info)); @@ -645,15 +722,16 @@ int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable) sk = __udp4_lib_lookup(net, iph->daddr, uh->dest, iph->saddr, uh->source, skb->dev->ifindex, inet_sdif(skb), udptable, NULL); - if (!sk) { + + if (!sk || udp_sk(sk)->encap_type) { /* No socket for error: try tunnels before discarding */ - sk = ERR_PTR(-ENOENT); if (static_branch_unlikely(&udp_encap_needed_key)) { - sk = __udp4_lib_err_encap(net, iph, uh, udptable, skb, + sk = __udp4_lib_err_encap(net, iph, uh, udptable, sk, skb, info); if (!sk) return 0; - } + } else + sk = ERR_PTR(-ENOENT); if (IS_ERR(sk)) { __ICMP_INC_STATS(net, ICMP_MIB_INERRORS); @@ -705,6 +783,8 @@ int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable) */ if (tunnel) { /* ...not for tunnels though: we don't have a sending socket */ + if (udp_sk(sk)->encap_err_rcv) + udp_sk(sk)->encap_err_rcv(sk, skb, iph->ihl << 2); goto out; } if (!inet->recverr) { @@ -714,7 +794,7 @@ int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable) ip_icmp_error(sk, skb, err, uh->dest, info, (u8 *)(uh+1)); sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); out: return 0; } @@ -817,7 +897,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4, struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(sk); struct udphdr *uh; - int err = 0; + int err; int is_udplite = IS_UDPLITE(sk); int offset = skb_transport_offset(skb); int len = skb->len - offset; @@ -841,7 +921,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4, kfree_skb(skb); return -EINVAL; } - if (skb->len > cork->gso_size * UDP_MAX_SEGMENTS) { + if (datalen > cork->gso_size * UDP_MAX_SEGMENTS) { kfree_skb(skb); return -EINVAL; } @@ -978,7 +1058,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) __be16 dport; u8 tos; int err, is_udplite = IS_UDPLITE(sk); - int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; + int corkreq = READ_ONCE(up->corkflag) || msg->msg_flags&MSG_MORE; int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); struct sk_buff *skb; struct ip_options_data opt_copy; @@ -1040,7 +1120,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } ipcm_init_sk(&ipc, inet); - ipc.gso_size = up->gso_size; + ipc.gso_size = READ_ONCE(up->gso_size); if (msg->msg_controllen) { err = udp_cmsg_send(sk, msg, &ipc.gso_size); @@ -1068,7 +1148,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) rcu_read_unlock(); } - if (cgroup_bpf_enabled && !connected) { + if (cgroup_bpf_enabled(CGROUP_UDP4_SENDMSG) && !connected) { err = BPF_CGROUP_RUN_PROG_UDP4_SENDMSG_LOCK(sk, (struct sockaddr *)usin, &ipc.addr); if (err) @@ -1113,7 +1193,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) ipc.oif = inet->uc_index; } else if (ipv4_is_lbcast(daddr) && inet->uc_index) { /* oif is set, packet is to local broadcast and - * and uc_index is set. oif is most likely set + * uc_index is set. oif is most likely set * by sk_bound_dev_if. If uc_index != oif check if the * oif is an L3 master and uc_index is an L3 slave. * If so, we want to allow the send using the uc_index. @@ -1140,7 +1220,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) faddr, saddr, dport, inet->inet_sport, sk->sk_uid); - security_sk_classify_flow(sk, flowi4_to_flowi(fl4)); + security_sk_classify_flow(sk, flowi4_to_flowi_common(fl4)); rt = ip_route_output_flow(net, fl4, sk); if (IS_ERR(rt)) { err = PTR_ERR(rt); @@ -1286,7 +1366,7 @@ int udp_sendpage(struct sock *sk, struct page *page, int offset, } up->len += size; - if (!(up->corkflag || (flags&MSG_MORE))) + if (!(READ_ONCE(up->corkflag) || (flags&MSG_MORE))) ret = udp_push_pending_frames(sk); if (!ret) ret = size; @@ -1385,11 +1465,11 @@ static void udp_rmem_release(struct sock *sk, int size, int partial, sk->sk_forward_alloc += size; - amt = (sk->sk_forward_alloc - partial) & ~(SK_MEM_QUANTUM - 1); + amt = (sk->sk_forward_alloc - partial) & ~(PAGE_SIZE - 1); sk->sk_forward_alloc -= amt; if (amt) - __sk_mem_reduce_allocated(sk, amt >> SK_MEM_QUANTUM_SHIFT); + __sk_mem_reduce_allocated(sk, amt >> PAGE_SHIFT); atomic_sub(size, &sk->sk_rmem_alloc); @@ -1482,7 +1562,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) spin_lock(&list->lock); if (size >= sk->sk_forward_alloc) { amt = sk_mem_pages(size); - delta = amt << SK_MEM_QUANTUM_SHIFT; + delta = amt << PAGE_SHIFT; if (!__sk_mem_raise_allocated(sk, delta, amt, SK_MEM_RECV)) { err = -ENOBUFS; spin_unlock(&list->lock); @@ -1518,7 +1598,7 @@ drop: } EXPORT_SYMBOL_GPL(__udp_enqueue_schedule_skb); -void udp_destruct_sock(struct sock *sk) +void udp_destruct_common(struct sock *sk) { /* reclaim completely the forward allocated memory */ struct udp_sock *up = udp_sk(sk); @@ -1531,18 +1611,22 @@ void udp_destruct_sock(struct sock *sk) kfree_skb(skb); } udp_rmem_release(sk, total, 0, true); +} +EXPORT_SYMBOL_GPL(udp_destruct_common); +static void udp_destruct_sock(struct sock *sk) +{ + udp_destruct_common(sk); inet_sock_destruct(sk); } -EXPORT_SYMBOL_GPL(udp_destruct_sock); int udp_init_sock(struct sock *sk) { skb_queue_head_init(&udp_sk(sk)->reader_queue); sk->sk_destruct = udp_destruct_sock; + set_bit(SOCK_SUPPORT_ZC, &sk->sk_socket->flags); return 0; } -EXPORT_SYMBOL_GPL(udp_init_sock); void skb_consume_udp(struct sock *sk, struct sk_buff *skb, int len) { @@ -1650,7 +1734,7 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) EXPORT_SYMBOL(udp_ioctl); struct sk_buff *__skb_recv_udp(struct sock *sk, unsigned int flags, - int noblock, int *off, int *err) + int *off, int *err) { struct sk_buff_head *sk_queue = &sk->sk_receive_queue; struct sk_buff_head *queue; @@ -1659,7 +1743,6 @@ struct sk_buff *__skb_recv_udp(struct sock *sk, unsigned int flags, int error; queue = &udp_sk(sk)->reader_queue; - flags |= noblock ? MSG_DONTWAIT : 0; timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); do { struct sk_buff *skb; @@ -1671,10 +1754,11 @@ struct sk_buff *__skb_recv_udp(struct sock *sk, unsigned int flags, error = -EAGAIN; do { spin_lock_bh(&queue->lock); - skb = __skb_try_recv_from_queue(sk, queue, flags, - udp_skb_destructor, - off, err, &last); + skb = __skb_try_recv_from_queue(sk, queue, flags, off, + err, &last); if (skb) { + if (!(flags & MSG_PEEK)) + udp_skb_destructor(sk, skb); spin_unlock_bh(&queue->lock); return skb; } @@ -1692,9 +1776,10 @@ struct sk_buff *__skb_recv_udp(struct sock *sk, unsigned int flags, spin_lock(&sk_queue->lock); skb_queue_splice_tail_init(sk_queue, queue); - skb = __skb_try_recv_from_queue(sk, queue, flags, - udp_skb_dtor_locked, - off, err, &last); + skb = __skb_try_recv_from_queue(sk, queue, flags, off, + err, &last); + if (skb && !(flags & MSG_PEEK)) + udp_skb_dtor_locked(sk, skb); spin_unlock(&sk_queue->lock); spin_unlock_bh(&queue->lock); if (skb) @@ -1718,13 +1803,42 @@ busy_check: } EXPORT_SYMBOL(__skb_recv_udp); +int udp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) +{ + struct sk_buff *skb; + int err, copied; + +try_again: + skb = skb_recv_udp(sk, MSG_DONTWAIT, &err); + if (!skb) + return err; + + if (udp_lib_checksum_complete(skb)) { + int is_udplite = IS_UDPLITE(sk); + struct net *net = sock_net(sk); + + __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS, is_udplite); + __UDP_INC_STATS(net, UDP_MIB_INERRORS, is_udplite); + atomic_inc(&sk->sk_drops); + kfree_skb(skb); + goto try_again; + } + + WARN_ON_ONCE(!skb_set_owner_sk_safe(skb, sk)); + copied = recv_actor(sk, skb); + kfree_skb(skb); + + return copied; +} +EXPORT_SYMBOL(udp_read_skb); + /* * This should be easy, if there is something there we * return it, otherwise we block. */ -int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, - int flags, int *addr_len) +int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags, + int *addr_len) { struct inet_sock *inet = inet_sk(sk); DECLARE_SOCKADDR(struct sockaddr_in *, sin, msg->msg_name); @@ -1739,7 +1853,7 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, try_again: off = sk_peek_offset(sk, flags); - skb = __skb_recv_udp(sk, flags, noblock, &off, &err); + skb = __skb_recv_udp(sk, flags, &off, &err); if (!skb) return err; @@ -1790,7 +1904,7 @@ try_again: UDP_INC_STATS(sock_net(sk), UDP_MIB_INDATAGRAMS, is_udplite); - sock_recv_ts_and_drops(msg, sk, skb); + sock_recv_cmsgs(msg, sk, skb); /* Copy the address. */ if (sin) { @@ -1800,9 +1914,8 @@ try_again: memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); *addr_len = sizeof(*sin); - if (cgroup_bpf_enabled) - BPF_CGROUP_RUN_PROG_UDP4_RECVMSG_LOCK(sk, - (struct sockaddr *)sin); + BPF_CGROUP_RUN_PROG_UDP4_RECVMSG_LOCK(sk, + (struct sockaddr *)sin); } if (udp_sk(sk)->gro_enabled) @@ -1974,13 +2087,20 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) rc = __udp_enqueue_schedule_skb(sk, skb); if (rc < 0) { int is_udplite = IS_UDPLITE(sk); + int drop_reason; /* Note that an ENOMEM error is charged twice */ - if (rc == -ENOMEM) + if (rc == -ENOMEM) { UDP_INC_STATS(sock_net(sk), UDP_MIB_RCVBUFERRORS, is_udplite); + drop_reason = SKB_DROP_REASON_SOCKET_RCVBUFF; + } else { + UDP_INC_STATS(sock_net(sk), UDP_MIB_MEMERRORS, + is_udplite); + drop_reason = SKB_DROP_REASON_PROTO_MEM; + } UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite); - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); trace_udp_fail_queue_rcv_skb(rc, sk); return -1; } @@ -1998,14 +2118,17 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) */ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb) { + int drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; struct udp_sock *up = udp_sk(sk); int is_udplite = IS_UDPLITE(sk); /* * Charge it to the socket, dropping if the queue is full. */ - if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) { + drop_reason = SKB_DROP_REASON_XFRM_POLICY; goto drop; + } nf_reset_ct(skb); if (static_branch_unlikely(&udp_encap_needed_key) && up->encap_type) { @@ -2046,7 +2169,7 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb) /* * UDP-Lite specific tests, ignored on UDP sockets */ - if ((is_udplite & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + if ((up->pcflag & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { /* * MIB statistics other than incrementing the error count are @@ -2082,8 +2205,10 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb) udp_lib_checksum_complete(skb)) goto csum_error; - if (sk_filter_trim_cap(sk, skb, sizeof(struct udphdr))) + if (sk_filter_trim_cap(sk, skb, sizeof(struct udphdr))) { + drop_reason = SKB_DROP_REASON_SOCKET_FILTER; goto drop; + } udp_csum_pull_header(skb); @@ -2091,11 +2216,12 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb) return __udp_queue_rcv_skb(sk, skb); csum_error: + drop_reason = SKB_DROP_REASON_UDP_CSUM; __UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS, is_udplite); drop: __UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite); atomic_inc(&sk->sk_drops); - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); return -1; } @@ -2107,14 +2233,16 @@ static int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) if (likely(!udp_unexpected_gso(sk, skb))) return udp_queue_rcv_one_skb(sk, skb); - BUILD_BUG_ON(sizeof(struct udp_skb_cb) > SKB_SGO_CB_OFFSET); + BUILD_BUG_ON(sizeof(struct udp_skb_cb) > SKB_GSO_CB_OFFSET); __skb_push(skb, -skb_mac_offset(skb)); segs = udp_rcv_segment(sk, skb, true); skb_list_walk_safe(segs, skb, next) { __skb_pull(skb, skb_transport_offset(skb)); + + udp_post_segment_fix_csum(skb); ret = udp_queue_rcv_one_skb(sk, skb); if (ret > 0) - ip_protocol_deliver_rcu(dev_net(skb->dev), skb, -ret); + ip_protocol_deliver_rcu(dev_net(skb->dev), skb, ret); } return 0; } @@ -2127,7 +2255,7 @@ bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) struct dst_entry *old; if (dst_hold_safe(dst)) { - old = xchg(&sk->sk_rx_dst, dst); + old = xchg((__force struct dst_entry **)&sk->sk_rx_dst, dst); dst_release(old); return old != dst; } @@ -2286,6 +2414,10 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, struct rtable *rt = skb_rtable(skb); __be32 saddr, daddr; struct net *net = dev_net(skb->dev); + bool refcounted; + int drop_reason; + + drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; /* * Validate the packet. @@ -2311,16 +2443,17 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (udp4_csum_init(skb, uh, proto)) goto csum_error; - sk = skb_steal_sock(skb); + sk = skb_steal_sock(skb, &refcounted); if (sk) { struct dst_entry *dst = skb_dst(skb); int ret; - if (unlikely(sk->sk_rx_dst != dst)) + if (unlikely(rcu_dereference(sk->sk_rx_dst) != dst)) udp_sk_rx_dst_set(sk, dst); ret = udp_unicast_rcv_skb(sk, skb, uh); - sock_put(sk); + if (refcounted) + sock_put(sk); return ret; } @@ -2340,6 +2473,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (udp_lib_checksum_complete(skb)) goto csum_error; + drop_reason = SKB_DROP_REASON_NO_SOCKET; __UDP_INC_STATS(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); @@ -2347,10 +2481,11 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, * Hmm. We got an UDP packet to a port to which we * don't wanna listen. Ignore it. */ - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); return 0; short_packet: + drop_reason = SKB_DROP_REASON_PKT_TOO_SMALL; net_dbg_ratelimited("UDP%s: short packet: From %pI4:%u %d/%d to %pI4:%u\n", proto == IPPROTO_UDPLITE ? "Lite" : "", &saddr, ntohs(uh->source), @@ -2363,6 +2498,7 @@ csum_error: * RFC1122: OK. Discards the bad packet silently (as far as * the network is concerned, anyway) as per 4.1.3.4 (MUST). */ + drop_reason = SKB_DROP_REASON_UDP_CSUM; net_dbg_ratelimited("UDP%s: bad checksum. From %pI4:%u to %pI4:%u ulen %d\n", proto == IPPROTO_UDPLITE ? "Lite" : "", &saddr, ntohs(uh->source), &daddr, ntohs(uh->dest), @@ -2370,7 +2506,7 @@ csum_error: __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS, proto == IPPROTO_UDPLITE); drop: __UDP_INC_STATS(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE); - kfree_skb(skb); + kfree_skb_reason(skb, drop_reason); return 0; } @@ -2422,8 +2558,7 @@ static struct sock *__udp4_lib_demux_lookup(struct net *net, struct sock *sk; udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { - if (INET_MATCH(sk, net, acookie, rmt_addr, - loc_addr, ports, dif, sdif)) + if (inet_match(net, sk, acookie, ports, dif, sdif)) return sk; /* Only check first socket in chain */ break; @@ -2474,7 +2609,7 @@ int udp_v4_early_demux(struct sk_buff *skb) skb->sk = sk; skb->destructor = sock_efree; - dst = READ_ONCE(sk->sk_rx_dst); + dst = rcu_dereference(sk->sk_rx_dst); if (dst) dst = dst_check(dst, 0); @@ -2492,7 +2627,8 @@ int udp_v4_early_demux(struct sk_buff *skb) */ if (!inet_sk(sk)->inet_daddr && in_dev) return ip_mc_validate_source(skb, iph->daddr, - iph->saddr, iph->tos, + iph->saddr, + iph->tos & IPTOS_RT_MASK, skb->dev, in_dev, &itag); } return 0; @@ -2507,6 +2643,9 @@ void udp_destroy_sock(struct sock *sk) { struct udp_sock *up = udp_sk(sk); bool slow = lock_sock_fast(sk); + + /* protects from races with udp_abort() */ + sock_set_flag(sk, SOCK_DEAD); udp_flush_pending_frames(sk); unlock_sock_fast(sk, slow); if (static_branch_unlikely(&udp_encap_needed_key)) { @@ -2525,7 +2664,7 @@ void udp_destroy_sock(struct sock *sk) * Socket option code for UDP */ int udp_lib_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen, + sockptr_t optval, unsigned int optlen, int (*push_pending_frames)(struct sock *)) { struct udp_sock *up = udp_sk(sk); @@ -2536,7 +2675,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, if (optlen < sizeof(int)) return -EINVAL; - if (get_user(val, (int __user *)optval)) + if (copy_from_sockptr(&val, optval, sizeof(val))) return -EFAULT; valbool = val ? 1 : 0; @@ -2544,9 +2683,9 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, switch (optname) { case UDP_CORK: if (val != 0) { - up->corkflag = 1; + WRITE_ONCE(up->corkflag, 1); } else { - up->corkflag = 0; + WRITE_ONCE(up->corkflag, 0); lock_sock(sk); push_pending_frames(sk); release_sock(sk); @@ -2559,9 +2698,14 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, #ifdef CONFIG_XFRM case UDP_ENCAP_ESPINUDP: case UDP_ENCAP_ESPINUDP_NON_IKE: - up->encap_rcv = xfrm4_udp_encap_rcv; +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) + up->encap_rcv = ipv6_stub->xfrm6_udp_encap_rcv; + else +#endif + up->encap_rcv = xfrm4_udp_encap_rcv; #endif - /* FALLTHROUGH */ + fallthrough; case UDP_ENCAP_L2TPINUDP: up->encap_type = val; lock_sock(sk); @@ -2585,14 +2729,17 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, case UDP_SEGMENT: if (val < 0 || val > USHRT_MAX) return -EINVAL; - up->gso_size = val; + WRITE_ONCE(up->gso_size, val); break; case UDP_GRO: lock_sock(sk); + + /* when enabling GRO, accept the related GSO packet type */ if (valbool) udp_tunnel_encap_enable(sk->sk_socket); up->gro_enabled = valbool; + up->accept_udp_l4 = valbool; release_sock(sk); break; @@ -2635,26 +2782,16 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, } EXPORT_SYMBOL(udp_lib_setsockopt); -int udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) +int udp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval, + unsigned int optlen) { if (level == SOL_UDP || level == SOL_UDPLITE) - return udp_lib_setsockopt(sk, level, optname, optval, optlen, + return udp_lib_setsockopt(sk, level, optname, + optval, optlen, udp_push_pending_frames); return ip_setsockopt(sk, level, optname, optval, optlen); } -#ifdef CONFIG_COMPAT -int compat_udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen) -{ - if (level == SOL_UDP || level == SOL_UDPLITE) - return udp_lib_setsockopt(sk, level, optname, optval, optlen, - udp_push_pending_frames); - return compat_ip_setsockopt(sk, level, optname, optval, optlen); -} -#endif - int udp_lib_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -2671,7 +2808,7 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case UDP_CORK: - val = up->corkflag; + val = READ_ONCE(up->corkflag); break; case UDP_ENCAP: @@ -2687,7 +2824,11 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname, break; case UDP_SEGMENT: - val = up->gso_size; + val = READ_ONCE(up->gso_size); + break; + + case UDP_GRO: + val = up->gro_enabled; break; /* The following two cannot be changed on UDP sockets, the return is @@ -2720,20 +2861,11 @@ int udp_getsockopt(struct sock *sk, int level, int optname, return ip_getsockopt(sk, level, optname, optval, optlen); } -#ifdef CONFIG_COMPAT -int compat_udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (level == SOL_UDP || level == SOL_UDPLITE) - return udp_lib_getsockopt(sk, level, optname, optval, optlen); - return compat_ip_getsockopt(sk, level, optname, optval, optlen); -} -#endif /** * udp_poll - wait for a UDP event. - * @file - file struct - * @sock - socket - * @wait - poll table + * @file: - file struct + * @sock: - socket + * @wait: - poll table * * This is same as datagram poll, except for the special case of * blocking sockets. If application is using a blocking fd @@ -2755,6 +2887,9 @@ __poll_t udp_poll(struct file *file, struct socket *sock, poll_table *wait) !(sk->sk_shutdown & RCV_SHUTDOWN) && first_packet_length(sk) == -1) mask &= ~(EPOLLIN | EPOLLRDNORM); + /* psock ingress_msg queue should not contain any bad checksum frames */ + if (sk_is_readable(sk)) + mask |= EPOLLIN | EPOLLRDNORM; return mask; } @@ -2764,10 +2899,17 @@ int udp_abort(struct sock *sk, int err) { lock_sock(sk); + /* udp{v6}_destroy_sock() sets it under the sk lock, avoid racing + * with close() + */ + if (sock_flag(sk, SOCK_DEAD)) + goto out; + sk->sk_err = err; - sk->sk_error_report(sk); + sk_error_report(sk); __udp_disconnect(sk, 0); +out: release_sock(sk); return 0; @@ -2794,16 +2936,18 @@ struct proto udp_prot = { .unhash = udp_lib_unhash, .rehash = udp_v4_rehash, .get_port = udp_v4_get_port, + .put_port = udp_lib_unhash, +#ifdef CONFIG_BPF_SYSCALL + .psock_update_sk_prot = udp_bpf_update_proto, +#endif .memory_allocated = &udp_memory_allocated, + .per_cpu_fw_alloc = &udp_memory_per_cpu_fw_alloc, + .sysctl_mem = sysctl_udp_mem, .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min), .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min), .obj_size = sizeof(struct udp_sock), .h.udp_table = &udp_table, -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_udp_setsockopt, - .compat_getsockopt = compat_udp_getsockopt, -#endif .diag_destroy = udp_abort, }; EXPORT_SYMBOL(udp_prot); @@ -2814,10 +2958,15 @@ EXPORT_SYMBOL(udp_prot); static struct sock *udp_get_first(struct seq_file *seq, int start) { struct sock *sk; - struct udp_seq_afinfo *afinfo = PDE_DATA(file_inode(seq->file)); + struct udp_seq_afinfo *afinfo; struct udp_iter_state *state = seq->private; struct net *net = seq_file_net(seq); + if (state->bpf_seq_afinfo) + afinfo = state->bpf_seq_afinfo; + else + afinfo = pde_data(file_inode(seq->file)); + for (state->bucket = start; state->bucket <= afinfo->udp_table->mask; ++state->bucket) { struct udp_hslot *hslot = &afinfo->udp_table->hash[state->bucket]; @@ -2829,7 +2978,8 @@ static struct sock *udp_get_first(struct seq_file *seq, int start) sk_for_each(sk, &hslot->head) { if (!net_eq(sock_net(sk), net)) continue; - if (sk->sk_family == afinfo->family) + if (afinfo->family == AF_UNSPEC || + sk->sk_family == afinfo->family) goto found; } spin_unlock_bh(&hslot->lock); @@ -2841,13 +2991,20 @@ found: static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk) { - struct udp_seq_afinfo *afinfo = PDE_DATA(file_inode(seq->file)); + struct udp_seq_afinfo *afinfo; struct udp_iter_state *state = seq->private; struct net *net = seq_file_net(seq); + if (state->bpf_seq_afinfo) + afinfo = state->bpf_seq_afinfo; + else + afinfo = pde_data(file_inode(seq->file)); + do { sk = sk_next(sk); - } while (sk && (!net_eq(sock_net(sk), net) || sk->sk_family != afinfo->family)); + } while (sk && (!net_eq(sock_net(sk), net) || + (afinfo->family != AF_UNSPEC && + sk->sk_family != afinfo->family))); if (!sk) { if (state->bucket <= afinfo->udp_table->mask) @@ -2892,9 +3049,14 @@ EXPORT_SYMBOL(udp_seq_next); void udp_seq_stop(struct seq_file *seq, void *v) { - struct udp_seq_afinfo *afinfo = PDE_DATA(file_inode(seq->file)); + struct udp_seq_afinfo *afinfo; struct udp_iter_state *state = seq->private; + if (state->bpf_seq_afinfo) + afinfo = state->bpf_seq_afinfo; + else + afinfo = pde_data(file_inode(seq->file)); + if (state->bucket <= afinfo->udp_table->mask) spin_unlock_bh(&afinfo->udp_table->hash[state->bucket].lock); } @@ -2926,7 +3088,7 @@ int udp4_seq_show(struct seq_file *seq, void *v) { seq_setwidth(seq, 127); if (v == SEQ_START_TOKEN) - seq_puts(seq, " sl local_address rem_address st tx_queue " + seq_puts(seq, " sl local_address rem_address st tx_queue " "rx_queue tr tm->when retrnsmt uid timeout " "inode ref pointer drops"); else { @@ -2938,6 +3100,67 @@ int udp4_seq_show(struct seq_file *seq, void *v) return 0; } +#ifdef CONFIG_BPF_SYSCALL +struct bpf_iter__udp { + __bpf_md_ptr(struct bpf_iter_meta *, meta); + __bpf_md_ptr(struct udp_sock *, udp_sk); + uid_t uid __aligned(8); + int bucket __aligned(8); +}; + +static int udp_prog_seq_show(struct bpf_prog *prog, struct bpf_iter_meta *meta, + struct udp_sock *udp_sk, uid_t uid, int bucket) +{ + struct bpf_iter__udp ctx; + + meta->seq_num--; /* skip SEQ_START_TOKEN */ + ctx.meta = meta; + ctx.udp_sk = udp_sk; + ctx.uid = uid; + ctx.bucket = bucket; + return bpf_iter_run_prog(prog, &ctx); +} + +static int bpf_iter_udp_seq_show(struct seq_file *seq, void *v) +{ + struct udp_iter_state *state = seq->private; + struct bpf_iter_meta meta; + struct bpf_prog *prog; + struct sock *sk = v; + uid_t uid; + + if (v == SEQ_START_TOKEN) + return 0; + + uid = from_kuid_munged(seq_user_ns(seq), sock_i_uid(sk)); + meta.seq = seq; + prog = bpf_iter_get_info(&meta, false); + return udp_prog_seq_show(prog, &meta, v, uid, state->bucket); +} + +static void bpf_iter_udp_seq_stop(struct seq_file *seq, void *v) +{ + struct bpf_iter_meta meta; + struct bpf_prog *prog; + + if (!v) { + meta.seq = seq; + prog = bpf_iter_get_info(&meta, true); + if (prog) + (void)udp_prog_seq_show(prog, &meta, v, 0, 0); + } + + udp_seq_stop(seq, v); +} + +static const struct seq_operations bpf_iter_udp_seq_ops = { + .start = udp_seq_start, + .next = udp_seq_next, + .stop = bpf_iter_udp_seq_stop, + .show = bpf_iter_udp_seq_show, +}; +#endif + const struct seq_operations udp_seq_ops = { .start = udp_seq_start, .next = udp_seq_next, @@ -3035,19 +3258,15 @@ u32 udp_flow_hashrnd(void) } EXPORT_SYMBOL(udp_flow_hashrnd); -static void __udp_sysctl_init(struct net *net) +static int __net_init udp_sysctl_init(struct net *net) { - net->ipv4.sysctl_udp_rmem_min = SK_MEM_QUANTUM; - net->ipv4.sysctl_udp_wmem_min = SK_MEM_QUANTUM; + net->ipv4.sysctl_udp_rmem_min = PAGE_SIZE; + net->ipv4.sysctl_udp_wmem_min = PAGE_SIZE; #ifdef CONFIG_NET_L3_MASTER_DEV net->ipv4.sysctl_udp_l3mdev_accept = 0; #endif -} -static int __net_init udp_sysctl_init(struct net *net) -{ - __udp_sysctl_init(net); return 0; } @@ -3055,6 +3274,62 @@ static struct pernet_operations __net_initdata udp_sysctl_ops = { .init = udp_sysctl_init, }; +#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) +DEFINE_BPF_ITER_FUNC(udp, struct bpf_iter_meta *meta, + struct udp_sock *udp_sk, uid_t uid, int bucket) + +static int bpf_iter_init_udp(void *priv_data, struct bpf_iter_aux_info *aux) +{ + struct udp_iter_state *st = priv_data; + struct udp_seq_afinfo *afinfo; + int ret; + + afinfo = kmalloc(sizeof(*afinfo), GFP_USER | __GFP_NOWARN); + if (!afinfo) + return -ENOMEM; + + afinfo->family = AF_UNSPEC; + afinfo->udp_table = &udp_table; + st->bpf_seq_afinfo = afinfo; + ret = bpf_iter_init_seq_net(priv_data, aux); + if (ret) + kfree(afinfo); + return ret; +} + +static void bpf_iter_fini_udp(void *priv_data) +{ + struct udp_iter_state *st = priv_data; + + kfree(st->bpf_seq_afinfo); + bpf_iter_fini_seq_net(priv_data); +} + +static const struct bpf_iter_seq_info udp_seq_info = { + .seq_ops = &bpf_iter_udp_seq_ops, + .init_seq_private = bpf_iter_init_udp, + .fini_seq_private = bpf_iter_fini_udp, + .seq_priv_size = sizeof(struct udp_iter_state), +}; + +static struct bpf_iter_reg udp_reg_info = { + .target = "udp", + .ctx_arg_info_size = 1, + .ctx_arg_info = { + { offsetof(struct bpf_iter__udp, udp_sk), + PTR_TO_BTF_ID_OR_NULL }, + }, + .seq_info = &udp_seq_info, +}; + +static void __init bpf_iter_register(void) +{ + udp_reg_info.ctx_arg_info[0].btf_id = btf_sock_ids[BTF_SOCK_TYPE_UDP]; + if (bpf_iter_reg_target(&udp_reg_info)) + pr_warn("Warning: could not register bpf iterator udp\n"); +} +#endif + void __init udp_init(void) { unsigned long limit; @@ -3067,8 +3342,6 @@ void __init udp_init(void) sysctl_udp_mem[1] = limit; sysctl_udp_mem[2] = sysctl_udp_mem[0] * 2; - __udp_sysctl_init(&init_net); - /* 16 spinlocks per cpu */ udp_busylocks_log = ilog2(nr_cpu_ids) + 4; udp_busylocks = kmalloc(sizeof(spinlock_t) << udp_busylocks_log, @@ -3080,4 +3353,8 @@ void __init udp_init(void) if (register_pernet_subsys(&udp_sysctl_ops)) panic("UDP: failed to init sysctl parameters.\n"); + +#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) + bpf_iter_register(); +#endif } diff --git a/net/ipv4/udp_bpf.c b/net/ipv4/udp_bpf.c new file mode 100644 index 000000000000..e5dc91d0e079 --- /dev/null +++ b/net/ipv4/udp_bpf.c @@ -0,0 +1,154 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Cloudflare Ltd https://cloudflare.com */ + +#include <linux/skmsg.h> +#include <net/sock.h> +#include <net/udp.h> +#include <net/inet_common.h> + +#include "udp_impl.h" + +static struct proto *udpv6_prot_saved __read_mostly; + +static int sk_udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, + int flags, int *addr_len) +{ +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) + return udpv6_prot_saved->recvmsg(sk, msg, len, flags, addr_len); +#endif + return udp_prot.recvmsg(sk, msg, len, flags, addr_len); +} + +static bool udp_sk_has_data(struct sock *sk) +{ + return !skb_queue_empty(&udp_sk(sk)->reader_queue) || + !skb_queue_empty(&sk->sk_receive_queue); +} + +static bool psock_has_data(struct sk_psock *psock) +{ + return !skb_queue_empty(&psock->ingress_skb) || + !sk_psock_queue_empty(psock); +} + +#define udp_msg_has_data(__sk, __psock) \ + ({ udp_sk_has_data(__sk) || psock_has_data(__psock); }) + +static int udp_msg_wait_data(struct sock *sk, struct sk_psock *psock, + long timeo) +{ + DEFINE_WAIT_FUNC(wait, woken_wake_function); + int ret = 0; + + if (sk->sk_shutdown & RCV_SHUTDOWN) + return 1; + + if (!timeo) + return ret; + + add_wait_queue(sk_sleep(sk), &wait); + sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); + ret = udp_msg_has_data(sk, psock); + if (!ret) { + wait_woken(&wait, TASK_INTERRUPTIBLE, timeo); + ret = udp_msg_has_data(sk, psock); + } + sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); + remove_wait_queue(sk_sleep(sk), &wait); + return ret; +} + +static int udp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, + int flags, int *addr_len) +{ + struct sk_psock *psock; + int copied, ret; + + if (unlikely(flags & MSG_ERRQUEUE)) + return inet_recv_error(sk, msg, len, addr_len); + + psock = sk_psock_get(sk); + if (unlikely(!psock)) + return sk_udp_recvmsg(sk, msg, len, flags, addr_len); + + if (!psock_has_data(psock)) { + ret = sk_udp_recvmsg(sk, msg, len, flags, addr_len); + goto out; + } + +msg_bytes_ready: + copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + if (!copied) { + long timeo; + int data; + + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + data = udp_msg_wait_data(sk, psock, timeo); + if (data) { + if (psock_has_data(psock)) + goto msg_bytes_ready; + ret = sk_udp_recvmsg(sk, msg, len, flags, addr_len); + goto out; + } + copied = -EAGAIN; + } + ret = copied; +out: + sk_psock_put(sk, psock); + return ret; +} + +enum { + UDP_BPF_IPV4, + UDP_BPF_IPV6, + UDP_BPF_NUM_PROTS, +}; + +static DEFINE_SPINLOCK(udpv6_prot_lock); +static struct proto udp_bpf_prots[UDP_BPF_NUM_PROTS]; + +static void udp_bpf_rebuild_protos(struct proto *prot, const struct proto *base) +{ + *prot = *base; + prot->close = sock_map_close; + prot->recvmsg = udp_bpf_recvmsg; + prot->sock_is_readable = sk_msg_is_readable; +} + +static void udp_bpf_check_v6_needs_rebuild(struct proto *ops) +{ + if (unlikely(ops != smp_load_acquire(&udpv6_prot_saved))) { + spin_lock_bh(&udpv6_prot_lock); + if (likely(ops != udpv6_prot_saved)) { + udp_bpf_rebuild_protos(&udp_bpf_prots[UDP_BPF_IPV6], ops); + smp_store_release(&udpv6_prot_saved, ops); + } + spin_unlock_bh(&udpv6_prot_lock); + } +} + +static int __init udp_bpf_v4_build_proto(void) +{ + udp_bpf_rebuild_protos(&udp_bpf_prots[UDP_BPF_IPV4], &udp_prot); + return 0; +} +late_initcall(udp_bpf_v4_build_proto); + +int udp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore) +{ + int family = sk->sk_family == AF_INET ? UDP_BPF_IPV4 : UDP_BPF_IPV6; + + if (restore) { + sk->sk_write_space = psock->saved_write_space; + sock_replace_proto(sk, psock->sk_proto); + return 0; + } + + if (sk->sk_family == AF_INET6) + udp_bpf_check_v6_needs_rebuild(psock->sk_proto); + + sock_replace_proto(sk, &udp_bpf_prots[family]); + return 0; +} +EXPORT_SYMBOL_GPL(udp_bpf_update_proto); diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c index dccd2286bc28..1ed8c4d78e5c 100644 --- a/net/ipv4/udp_diag.c +++ b/net/ipv4/udp_diag.c @@ -21,17 +21,16 @@ static int sk_diag_dump(struct sock *sk, struct sk_buff *skb, if (!inet_diag_bc_sk(bc, sk)) return 0; - return inet_sk_diag_fill(sk, NULL, skb, req, - sk_user_ns(NETLINK_CB(cb->skb).sk), - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, NLM_F_MULTI, cb->nlh, net_admin); + return inet_sk_diag_fill(sk, NULL, skb, cb, req, NLM_F_MULTI, + net_admin); } -static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb, - const struct nlmsghdr *nlh, +static int udp_dump_one(struct udp_table *tbl, + struct netlink_callback *cb, const struct inet_diag_req_v2 *req) { - int err = -EINVAL; + struct sk_buff *in_skb = cb->skb; + int err; struct sock *sk = NULL; struct sk_buff *rep; struct net *net = sock_net(in_skb->sk); @@ -71,20 +70,15 @@ static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb, if (!rep) goto out; - err = inet_sk_diag_fill(sk, NULL, rep, req, - sk_user_ns(NETLINK_CB(in_skb).sk), - NETLINK_CB(in_skb).portid, - nlh->nlmsg_seq, 0, nlh, - netlink_net_capable(in_skb, CAP_NET_ADMIN)); + err = inet_sk_diag_fill(sk, NULL, rep, cb, req, 0, + netlink_net_capable(in_skb, CAP_NET_ADMIN)); if (err < 0) { WARN_ON(err == -EMSGSIZE); kfree_skb(rep); goto out; } - err = netlink_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid, - MSG_DONTWAIT); - if (err > 0) - err = 0; + err = nlmsg_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid); + out: if (sk) sock_put(sk); @@ -94,12 +88,16 @@ out_nosk: static void udp_dump(struct udp_table *table, struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, struct nlattr *bc) + const struct inet_diag_req_v2 *r) { bool net_admin = netlink_net_capable(cb->skb, CAP_NET_ADMIN); struct net *net = sock_net(skb->sk); + struct inet_diag_dump_data *cb_data; int num, s_num, slot, s_slot; + struct nlattr *bc; + cb_data = cb->data; + bc = cb_data->inet_diag_nla_bc; s_slot = cb->args[0]; num = s_num = cb->args[1]; @@ -147,15 +145,15 @@ done: } static void udp_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, struct nlattr *bc) + const struct inet_diag_req_v2 *r) { - udp_dump(&udp_table, skb, cb, r, bc); + udp_dump(&udp_table, skb, cb, r); } -static int udp_diag_dump_one(struct sk_buff *in_skb, const struct nlmsghdr *nlh, +static int udp_diag_dump_one(struct netlink_callback *cb, const struct inet_diag_req_v2 *req) { - return udp_dump_one(&udp_table, in_skb, nlh, req); + return udp_dump_one(&udp_table, cb, req); } static void udp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, @@ -250,16 +248,15 @@ static const struct inet_diag_handler udp_diag_handler = { }; static void udplite_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, - const struct inet_diag_req_v2 *r, - struct nlattr *bc) + const struct inet_diag_req_v2 *r) { - udp_dump(&udplite_table, skb, cb, r, bc); + udp_dump(&udplite_table, skb, cb, r); } -static int udplite_diag_dump_one(struct sk_buff *in_skb, const struct nlmsghdr *nlh, +static int udplite_diag_dump_one(struct netlink_callback *cb, const struct inet_diag_req_v2 *req) { - return udp_dump_one(&udplite_table, in_skb, nlh, req); + return udp_dump_one(&udplite_table, cb, req); } static const struct inet_diag_handler udplite_diag_handler = { diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h index 6b2fa77eeb1c..4ba7a88a1b1d 100644 --- a/net/ipv4/udp_impl.h +++ b/net/ipv4/udp_impl.h @@ -12,19 +12,13 @@ int __udp4_lib_err(struct sk_buff *, u32, struct udp_table *); int udp_v4_get_port(struct sock *sk, unsigned short snum); void udp_v4_rehash(struct sock *sk); -int udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen); +int udp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval, + unsigned int optlen); int udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen); -#ifdef CONFIG_COMPAT -int compat_udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, unsigned int optlen); -int compat_udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen); -#endif -int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, - int flags, int *addr_len); +int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags, + int *addr_len); int udp_sendpage(struct sock *sk, struct page *page, int offset, size_t size, int flags); void udp_destroy_sock(struct sock *sk); diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 1a98583a79f4..6d1a4bec2614 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -7,6 +7,7 @@ */ #include <linux/skbuff.h> +#include <net/gro.h> #include <net/udp.h> #include <net/protocol.h> #include <net/inet_common.h> @@ -49,6 +50,7 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, __skb_pull(skb, tnl_hlen); skb_reset_mac_header(skb); skb_set_network_header(skb, skb_inner_network_offset(skb)); + skb_set_transport_header(skb, skb_inner_transport_offset(skb)); skb->mac_len = skb_inner_network_offset(skb); skb->protocol = new_protocol; @@ -67,6 +69,8 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb, (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM)))); features &= skb->dev->hw_enc_features; + if (need_csum) + features &= ~NETIF_F_SCTP_CRC; /* The only checksum offload we care about from here on out is the * outer one so strip the existing checksum feature flags and @@ -149,8 +153,8 @@ struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb, netdev_features_t features, bool is_ipv6) { + const struct net_offload __rcu **offloads; __be16 protocol = skb->protocol; - const struct net_offload **offloads; const struct net_offload *ops; struct sk_buff *segs = ERR_PTR(-EINVAL); struct sk_buff *(*gso_inner_segment)(struct sk_buff *skb, @@ -184,8 +188,67 @@ out_unlock: } EXPORT_SYMBOL(skb_udp_tunnel_segment); +static void __udpv4_gso_segment_csum(struct sk_buff *seg, + __be32 *oldip, __be32 *newip, + __be16 *oldport, __be16 *newport) +{ + struct udphdr *uh; + struct iphdr *iph; + + if (*oldip == *newip && *oldport == *newport) + return; + + uh = udp_hdr(seg); + iph = ip_hdr(seg); + + if (uh->check) { + inet_proto_csum_replace4(&uh->check, seg, *oldip, *newip, + true); + inet_proto_csum_replace2(&uh->check, seg, *oldport, *newport, + false); + if (!uh->check) + uh->check = CSUM_MANGLED_0; + } + *oldport = *newport; + + csum_replace4(&iph->check, *oldip, *newip); + *oldip = *newip; +} + +static struct sk_buff *__udpv4_gso_segment_list_csum(struct sk_buff *segs) +{ + struct sk_buff *seg; + struct udphdr *uh, *uh2; + struct iphdr *iph, *iph2; + + seg = segs; + uh = udp_hdr(seg); + iph = ip_hdr(seg); + + if ((udp_hdr(seg)->dest == udp_hdr(seg->next)->dest) && + (udp_hdr(seg)->source == udp_hdr(seg->next)->source) && + (ip_hdr(seg)->daddr == ip_hdr(seg->next)->daddr) && + (ip_hdr(seg)->saddr == ip_hdr(seg->next)->saddr)) + return segs; + + while ((seg = seg->next)) { + uh2 = udp_hdr(seg); + iph2 = ip_hdr(seg); + + __udpv4_gso_segment_csum(seg, + &iph2->saddr, &iph->saddr, + &uh2->source, &uh->source); + __udpv4_gso_segment_csum(seg, + &iph2->daddr, &iph->daddr, + &uh2->dest, &uh->dest); + } + + return segs; +} + static struct sk_buff *__udp_gso_segment_list(struct sk_buff *skb, - netdev_features_t features) + netdev_features_t features, + bool is_ipv6) { unsigned int mss = skb_shinfo(skb)->gso_size; @@ -195,11 +258,11 @@ static struct sk_buff *__udp_gso_segment_list(struct sk_buff *skb, udp_hdr(skb)->len = htons(sizeof(struct udphdr) + mss); - return skb; + return is_ipv6 ? skb : __udpv4_gso_segment_list_csum(skb); } struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, - netdev_features_t features) + netdev_features_t features, bool is_ipv6) { struct sock *sk = gso_skb->sk; unsigned int sum_truesize = 0; @@ -211,7 +274,7 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, __be16 newlen; if (skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST) - return __udp_gso_segment_list(gso_skb, features); + return __udp_gso_segment_list(gso_skb, features, is_ipv6); mss = skb_shinfo(gso_skb)->gso_size; if (gso_skb->len <= sizeof(*uh) + mss) @@ -325,7 +388,7 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, goto out; if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) - return __udp_gso_segment(skb, features); + return __udp_gso_segment(skb, features, false); mss = skb_shinfo(skb)->gso_size; if (unlikely(skb->len <= mss)) @@ -362,11 +425,38 @@ out: return segs; } +static int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb) +{ + if (unlikely(p->len + skb->len >= 65536)) + return -E2BIG; + + if (NAPI_GRO_CB(p)->last == p) + skb_shinfo(p)->frag_list = skb; + else + NAPI_GRO_CB(p)->last->next = skb; + + skb_pull(skb, skb_gro_offset(skb)); + + NAPI_GRO_CB(p)->last = skb; + NAPI_GRO_CB(p)->count++; + p->data_len += skb->len; + + /* sk owenrship - if any - completely transferred to the aggregated packet */ + skb->destructor = NULL; + p->truesize += skb->truesize; + p->len += skb->len; + + NAPI_GRO_CB(skb)->same_flow = 1; + + return 0; +} + + #define UDP_GRO_CNT_MAX 64 static struct sk_buff *udp_gro_receive_segment(struct list_head *head, struct sk_buff *skb) { - struct udphdr *uh = udp_hdr(skb); + struct udphdr *uh = udp_gro_udphdr(skb); struct sk_buff *pp = NULL; struct udphdr *uh2; struct sk_buff *p; @@ -453,19 +543,26 @@ struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb, unsigned int off = skb_gro_offset(skb); int flush = 1; - if (skb->dev->features & NETIF_F_GRO_FRAGLIST) - NAPI_GRO_CB(skb)->is_flist = sk ? !udp_sk(sk)->gro_enabled: 1; + /* we can do L4 aggregation only if the packet can't land in a tunnel + * otherwise we could corrupt the inner stream + */ + NAPI_GRO_CB(skb)->is_flist = 0; + if (!sk || !udp_sk(sk)->gro_receive) { + if (skb->dev->features & NETIF_F_GRO_FRAGLIST) + NAPI_GRO_CB(skb)->is_flist = sk ? !udp_sk(sk)->gro_enabled : 1; - if ((sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist) { - pp = call_gro_receive(udp_gro_receive_segment, head, skb); - return pp; + if ((!sk && (skb->dev->features & NETIF_F_GRO_UDP_FWD)) || + (sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist) + return call_gro_receive(udp_gro_receive_segment, head, skb); + + /* no GRO, be sure flush the current packet */ + goto out; } - if (!sk || NAPI_GRO_CB(skb)->encap_mark || - (skb->ip_summed != CHECKSUM_PARTIAL && + if (NAPI_GRO_CB(skb)->encap_mark || + (uh->check && skb->ip_summed != CHECKSUM_PARTIAL && NAPI_GRO_CB(skb)->csum_cnt == 0 && - !NAPI_GRO_CB(skb)->csum_valid) || - !udp_sk(sk)->gro_receive) + !NAPI_GRO_CB(skb)->csum_valid)) goto out; /* mark that this skb passed once through the tunnel gro layer */ @@ -499,12 +596,22 @@ out: } EXPORT_SYMBOL(udp_gro_receive); +static struct sock *udp4_gro_lookup_skb(struct sk_buff *skb, __be16 sport, + __be16 dport) +{ + const struct iphdr *iph = skb_gro_network_header(skb); + + return __udp4_lib_lookup(dev_net(skb->dev), iph->saddr, sport, + iph->daddr, dport, inet_iif(skb), + inet_sdif(skb), &udp_table, NULL); +} + INDIRECT_CALLABLE_SCOPE struct sk_buff *udp4_gro_receive(struct list_head *head, struct sk_buff *skb) { struct udphdr *uh = udp_gro_udphdr(skb); + struct sock *sk = NULL; struct sk_buff *pp; - struct sock *sk; if (unlikely(!uh)) goto flush; @@ -521,10 +628,11 @@ struct sk_buff *udp4_gro_receive(struct list_head *head, struct sk_buff *skb) inet_gro_compute_pseudo); skip: NAPI_GRO_CB(skb)->is_ipv6 = 0; - rcu_read_lock(); - sk = static_branch_unlikely(&udp_encap_needed_key) ? udp4_lib_lookup_skb(skb, uh->source, uh->dest) : NULL; + + if (static_branch_unlikely(&udp_encap_needed_key)) + sk = udp4_gro_lookup_skb(skb, uh->source, uh->dest); + pp = udp_gro_receive(head, skb, uh, sk); - rcu_read_unlock(); return pp; flush: @@ -542,6 +650,10 @@ static int udp_gro_complete_segment(struct sk_buff *skb) skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count; skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_L4; + + if (skb->encapsulation) + skb->inner_transport_header = skb->transport_header; + return 0; } @@ -550,18 +662,22 @@ int udp_gro_complete(struct sk_buff *skb, int nhoff, { __be16 newlen = htons(skb->len - nhoff); struct udphdr *uh = (struct udphdr *)(skb->data + nhoff); - int err = -ENOSYS; struct sock *sk; + int err; uh->len = newlen; - rcu_read_lock(); sk = INDIRECT_CALL_INET(lookup, udp6_lib_lookup_skb, udp4_lib_lookup_skb, skb, uh->source, uh->dest); if (sk && udp_sk(sk)->gro_complete) { skb_shinfo(skb)->gso_type = uh->check ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL; + /* clear the encap mark, so that inner frag_list gro_complete + * can take place + */ + NAPI_GRO_CB(skb)->encap_mark = 0; + /* Set encapsulation before calling into inner gro_complete() * functions to make them set up the inner offsets. */ @@ -571,7 +687,6 @@ int udp_gro_complete(struct sk_buff *skb, int nhoff, } else { err = udp_gro_complete_segment(skb); } - rcu_read_unlock(); if (skb->remcsum_offload) skb_shinfo(skb)->gso_type |= SKB_GSO_TUNNEL_REMCSUM; @@ -585,7 +700,8 @@ INDIRECT_CALLABLE_SCOPE int udp4_gro_complete(struct sk_buff *skb, int nhoff) const struct iphdr *iph = ip_hdr(skb); struct udphdr *uh = (struct udphdr *)(skb->data + nhoff); - if (NAPI_GRO_CB(skb)->is_flist) { + /* do fraglist only if there is no outer UDP encap (or we already processed it) */ + if (NAPI_GRO_CB(skb)->is_flist && !NAPI_GRO_CB(skb)->encap_mark) { uh->len = htons(skb->len - nhoff); skb_shinfo(skb)->gso_type |= (SKB_GSO_FRAGLIST|SKB_GSO_UDP_L4); diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel_core.c index 150e6f0fdbf5..8242c8947340 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel_core.c @@ -2,11 +2,8 @@ #include <linux/module.h> #include <linux/errno.h> #include <linux/socket.h> -#include <linux/udp.h> -#include <linux/types.h> #include <linux/kernel.h> #include <net/dst_metadata.h> -#include <net/net_namespace.h> #include <net/udp.h> #include <net/udp_tunnel.h> @@ -22,9 +19,7 @@ int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg, goto error; if (cfg->bind_ifindex) { - err = kernel_setsockopt(sock, SOL_SOCKET, SO_BINDTOIFINDEX, - (void *)&cfg->bind_ifindex, - sizeof(cfg->bind_ifindex)); + err = sock_bindtoindex(sock->sk, cfg->bind_ifindex, true); if (err < 0) goto error; } @@ -77,6 +72,7 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, udp_sk(sk)->encap_type = cfg->encap_type; udp_sk(sk)->encap_rcv = cfg->encap_rcv; + udp_sk(sk)->encap_err_rcv = cfg->encap_err_rcv; udp_sk(sk)->encap_err_lookup = cfg->encap_err_lookup; udp_sk(sk)->encap_destroy = cfg->encap_destroy; udp_sk(sk)->gro_receive = cfg->gro_receive; @@ -92,15 +88,11 @@ void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, struct sock *sk = sock->sk; struct udp_tunnel_info ti; - if (!dev->netdev_ops->ndo_udp_tunnel_add || - !(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) - return; - ti.type = type; ti.sa_family = sk->sk_family; ti.port = inet_sk(sk)->inet_sport; - dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); + udp_tunnel_nic_add_port(dev, &ti); } EXPORT_SYMBOL_GPL(udp_tunnel_push_rx_port); @@ -110,15 +102,11 @@ void udp_tunnel_drop_rx_port(struct net_device *dev, struct socket *sock, struct sock *sk = sock->sk; struct udp_tunnel_info ti; - if (!dev->netdev_ops->ndo_udp_tunnel_del || - !(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) - return; - ti.type = type; ti.sa_family = sk->sk_family; ti.port = inet_sk(sk)->inet_sport; - dev->netdev_ops->ndo_udp_tunnel_del(dev, &ti); + udp_tunnel_nic_del_port(dev, &ti); } EXPORT_SYMBOL_GPL(udp_tunnel_drop_rx_port); @@ -136,11 +124,7 @@ void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type) rcu_read_lock(); for_each_netdev_rcu(net, dev) { - if (!dev->netdev_ops->ndo_udp_tunnel_add) - continue; - if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) - continue; - dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); + udp_tunnel_nic_add_port(dev, &ti); } rcu_read_unlock(); } @@ -160,11 +144,7 @@ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) rcu_read_lock(); for_each_netdev_rcu(net, dev) { - if (!dev->netdev_ops->ndo_udp_tunnel_del) - continue; - if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT)) - continue; - dev->netdev_ops->ndo_udp_tunnel_del(dev, &ti); + udp_tunnel_nic_del_port(dev, &ti); } rcu_read_unlock(); } diff --git a/net/ipv4/udp_tunnel_nic.c b/net/ipv4/udp_tunnel_nic.c new file mode 100644 index 000000000000..bc3a043a5d5c --- /dev/null +++ b/net/ipv4/udp_tunnel_nic.c @@ -0,0 +1,973 @@ +// SPDX-License-Identifier: GPL-2.0-only +// Copyright (c) 2020 Facebook Inc. + +#include <linux/ethtool_netlink.h> +#include <linux/netdevice.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/workqueue.h> +#include <net/udp_tunnel.h> +#include <net/vxlan.h> + +enum udp_tunnel_nic_table_entry_flags { + UDP_TUNNEL_NIC_ENTRY_ADD = BIT(0), + UDP_TUNNEL_NIC_ENTRY_DEL = BIT(1), + UDP_TUNNEL_NIC_ENTRY_OP_FAIL = BIT(2), + UDP_TUNNEL_NIC_ENTRY_FROZEN = BIT(3), +}; + +struct udp_tunnel_nic_table_entry { + __be16 port; + u8 type; + u8 flags; + u16 use_cnt; +#define UDP_TUNNEL_NIC_USE_CNT_MAX U16_MAX + u8 hw_priv; +}; + +/** + * struct udp_tunnel_nic - UDP tunnel port offload state + * @work: async work for talking to hardware from process context + * @dev: netdev pointer + * @need_sync: at least one port start changed + * @need_replay: space was freed, we need a replay of all ports + * @work_pending: @work is currently scheduled + * @n_tables: number of tables under @entries + * @missed: bitmap of tables which overflown + * @entries: table of tables of ports currently offloaded + */ +struct udp_tunnel_nic { + struct work_struct work; + + struct net_device *dev; + + u8 need_sync:1; + u8 need_replay:1; + u8 work_pending:1; + + unsigned int n_tables; + unsigned long missed; + struct udp_tunnel_nic_table_entry **entries; +}; + +/* We ensure all work structs are done using driver state, but not the code. + * We need a workqueue we can flush before module gets removed. + */ +static struct workqueue_struct *udp_tunnel_nic_workqueue; + +static const char *udp_tunnel_nic_tunnel_type_name(unsigned int type) +{ + switch (type) { + case UDP_TUNNEL_TYPE_VXLAN: + return "vxlan"; + case UDP_TUNNEL_TYPE_GENEVE: + return "geneve"; + case UDP_TUNNEL_TYPE_VXLAN_GPE: + return "vxlan-gpe"; + default: + return "unknown"; + } +} + +static bool +udp_tunnel_nic_entry_is_free(struct udp_tunnel_nic_table_entry *entry) +{ + return entry->use_cnt == 0 && !entry->flags; +} + +static bool +udp_tunnel_nic_entry_is_present(struct udp_tunnel_nic_table_entry *entry) +{ + return entry->use_cnt && !(entry->flags & ~UDP_TUNNEL_NIC_ENTRY_FROZEN); +} + +static bool +udp_tunnel_nic_entry_is_frozen(struct udp_tunnel_nic_table_entry *entry) +{ + return entry->flags & UDP_TUNNEL_NIC_ENTRY_FROZEN; +} + +static void +udp_tunnel_nic_entry_freeze_used(struct udp_tunnel_nic_table_entry *entry) +{ + if (!udp_tunnel_nic_entry_is_free(entry)) + entry->flags |= UDP_TUNNEL_NIC_ENTRY_FROZEN; +} + +static void +udp_tunnel_nic_entry_unfreeze(struct udp_tunnel_nic_table_entry *entry) +{ + entry->flags &= ~UDP_TUNNEL_NIC_ENTRY_FROZEN; +} + +static bool +udp_tunnel_nic_entry_is_queued(struct udp_tunnel_nic_table_entry *entry) +{ + return entry->flags & (UDP_TUNNEL_NIC_ENTRY_ADD | + UDP_TUNNEL_NIC_ENTRY_DEL); +} + +static void +udp_tunnel_nic_entry_queue(struct udp_tunnel_nic *utn, + struct udp_tunnel_nic_table_entry *entry, + unsigned int flag) +{ + entry->flags |= flag; + utn->need_sync = 1; +} + +static void +udp_tunnel_nic_ti_from_entry(struct udp_tunnel_nic_table_entry *entry, + struct udp_tunnel_info *ti) +{ + memset(ti, 0, sizeof(*ti)); + ti->port = entry->port; + ti->type = entry->type; + ti->hw_priv = entry->hw_priv; +} + +static bool +udp_tunnel_nic_is_empty(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + unsigned int i, j; + + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) + if (!udp_tunnel_nic_entry_is_free(&utn->entries[i][j])) + return false; + return true; +} + +static bool +udp_tunnel_nic_should_replay(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_table_info *table; + unsigned int i, j; + + if (!utn->missed) + return false; + + for (i = 0; i < utn->n_tables; i++) { + table = &dev->udp_tunnel_nic_info->tables[i]; + if (!test_bit(i, &utn->missed)) + continue; + + for (j = 0; j < table->n_entries; j++) + if (udp_tunnel_nic_entry_is_free(&utn->entries[i][j])) + return true; + } + + return false; +} + +static void +__udp_tunnel_nic_get_port(struct net_device *dev, unsigned int table, + unsigned int idx, struct udp_tunnel_info *ti) +{ + struct udp_tunnel_nic_table_entry *entry; + struct udp_tunnel_nic *utn; + + utn = dev->udp_tunnel_nic; + entry = &utn->entries[table][idx]; + + if (entry->use_cnt) + udp_tunnel_nic_ti_from_entry(entry, ti); +} + +static void +__udp_tunnel_nic_set_port_priv(struct net_device *dev, unsigned int table, + unsigned int idx, u8 priv) +{ + dev->udp_tunnel_nic->entries[table][idx].hw_priv = priv; +} + +static void +udp_tunnel_nic_entry_update_done(struct udp_tunnel_nic_table_entry *entry, + int err) +{ + bool dodgy = entry->flags & UDP_TUNNEL_NIC_ENTRY_OP_FAIL; + + WARN_ON_ONCE(entry->flags & UDP_TUNNEL_NIC_ENTRY_ADD && + entry->flags & UDP_TUNNEL_NIC_ENTRY_DEL); + + if (entry->flags & UDP_TUNNEL_NIC_ENTRY_ADD && + (!err || (err == -EEXIST && dodgy))) + entry->flags &= ~UDP_TUNNEL_NIC_ENTRY_ADD; + + if (entry->flags & UDP_TUNNEL_NIC_ENTRY_DEL && + (!err || (err == -ENOENT && dodgy))) + entry->flags &= ~UDP_TUNNEL_NIC_ENTRY_DEL; + + if (!err) + entry->flags &= ~UDP_TUNNEL_NIC_ENTRY_OP_FAIL; + else + entry->flags |= UDP_TUNNEL_NIC_ENTRY_OP_FAIL; +} + +static void +udp_tunnel_nic_device_sync_one(struct net_device *dev, + struct udp_tunnel_nic *utn, + unsigned int table, unsigned int idx) +{ + struct udp_tunnel_nic_table_entry *entry; + struct udp_tunnel_info ti; + int err; + + entry = &utn->entries[table][idx]; + if (!udp_tunnel_nic_entry_is_queued(entry)) + return; + + udp_tunnel_nic_ti_from_entry(entry, &ti); + if (entry->flags & UDP_TUNNEL_NIC_ENTRY_ADD) + err = dev->udp_tunnel_nic_info->set_port(dev, table, idx, &ti); + else + err = dev->udp_tunnel_nic_info->unset_port(dev, table, idx, + &ti); + udp_tunnel_nic_entry_update_done(entry, err); + + if (err) + netdev_warn(dev, + "UDP tunnel port sync failed port %d type %s: %d\n", + be16_to_cpu(entry->port), + udp_tunnel_nic_tunnel_type_name(entry->type), + err); +} + +static void +udp_tunnel_nic_device_sync_by_port(struct net_device *dev, + struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + unsigned int i, j; + + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) + udp_tunnel_nic_device_sync_one(dev, utn, i, j); +} + +static void +udp_tunnel_nic_device_sync_by_table(struct net_device *dev, + struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + unsigned int i, j; + int err; + + for (i = 0; i < utn->n_tables; i++) { + /* Find something that needs sync in this table */ + for (j = 0; j < info->tables[i].n_entries; j++) + if (udp_tunnel_nic_entry_is_queued(&utn->entries[i][j])) + break; + if (j == info->tables[i].n_entries) + continue; + + err = info->sync_table(dev, i); + if (err) + netdev_warn(dev, "UDP tunnel port sync failed for table %d: %d\n", + i, err); + + for (j = 0; j < info->tables[i].n_entries; j++) { + struct udp_tunnel_nic_table_entry *entry; + + entry = &utn->entries[i][j]; + if (udp_tunnel_nic_entry_is_queued(entry)) + udp_tunnel_nic_entry_update_done(entry, err); + } + } +} + +static void +__udp_tunnel_nic_device_sync(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + if (!utn->need_sync) + return; + + if (dev->udp_tunnel_nic_info->sync_table) + udp_tunnel_nic_device_sync_by_table(dev, utn); + else + udp_tunnel_nic_device_sync_by_port(dev, utn); + + utn->need_sync = 0; + /* Can't replay directly here, in case we come from the tunnel driver's + * notification - trying to replay may deadlock inside tunnel driver. + */ + utn->need_replay = udp_tunnel_nic_should_replay(dev, utn); +} + +static void +udp_tunnel_nic_device_sync(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + bool may_sleep; + + if (!utn->need_sync) + return; + + /* Drivers which sleep in the callback need to update from + * the workqueue, if we come from the tunnel driver's notification. + */ + may_sleep = info->flags & UDP_TUNNEL_NIC_INFO_MAY_SLEEP; + if (!may_sleep) + __udp_tunnel_nic_device_sync(dev, utn); + if (may_sleep || utn->need_replay) { + queue_work(udp_tunnel_nic_workqueue, &utn->work); + utn->work_pending = 1; + } +} + +static bool +udp_tunnel_nic_table_is_capable(const struct udp_tunnel_nic_table_info *table, + struct udp_tunnel_info *ti) +{ + return table->tunnel_types & ti->type; +} + +static bool +udp_tunnel_nic_is_capable(struct net_device *dev, struct udp_tunnel_nic *utn, + struct udp_tunnel_info *ti) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + unsigned int i; + + /* Special case IPv4-only NICs */ + if (info->flags & UDP_TUNNEL_NIC_INFO_IPV4_ONLY && + ti->sa_family != AF_INET) + return false; + + for (i = 0; i < utn->n_tables; i++) + if (udp_tunnel_nic_table_is_capable(&info->tables[i], ti)) + return true; + return false; +} + +static int +udp_tunnel_nic_has_collision(struct net_device *dev, struct udp_tunnel_nic *utn, + struct udp_tunnel_info *ti) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic_table_entry *entry; + unsigned int i, j; + + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) { + entry = &utn->entries[i][j]; + + if (!udp_tunnel_nic_entry_is_free(entry) && + entry->port == ti->port && + entry->type != ti->type) { + __set_bit(i, &utn->missed); + return true; + } + } + return false; +} + +static void +udp_tunnel_nic_entry_adj(struct udp_tunnel_nic *utn, + unsigned int table, unsigned int idx, int use_cnt_adj) +{ + struct udp_tunnel_nic_table_entry *entry = &utn->entries[table][idx]; + bool dodgy = entry->flags & UDP_TUNNEL_NIC_ENTRY_OP_FAIL; + unsigned int from, to; + + WARN_ON(entry->use_cnt + (u32)use_cnt_adj > U16_MAX); + + /* If not going from used to unused or vice versa - all done. + * For dodgy entries make sure we try to sync again (queue the entry). + */ + entry->use_cnt += use_cnt_adj; + if (!dodgy && !entry->use_cnt == !(entry->use_cnt - use_cnt_adj)) + return; + + /* Cancel the op before it was sent to the device, if possible, + * otherwise we'd need to take special care to issue commands + * in the same order the ports arrived. + */ + if (use_cnt_adj < 0) { + from = UDP_TUNNEL_NIC_ENTRY_ADD; + to = UDP_TUNNEL_NIC_ENTRY_DEL; + } else { + from = UDP_TUNNEL_NIC_ENTRY_DEL; + to = UDP_TUNNEL_NIC_ENTRY_ADD; + } + + if (entry->flags & from) { + entry->flags &= ~from; + if (!dodgy) + return; + } + + udp_tunnel_nic_entry_queue(utn, entry, to); +} + +static bool +udp_tunnel_nic_entry_try_adj(struct udp_tunnel_nic *utn, + unsigned int table, unsigned int idx, + struct udp_tunnel_info *ti, int use_cnt_adj) +{ + struct udp_tunnel_nic_table_entry *entry = &utn->entries[table][idx]; + + if (udp_tunnel_nic_entry_is_free(entry) || + entry->port != ti->port || + entry->type != ti->type) + return false; + + if (udp_tunnel_nic_entry_is_frozen(entry)) + return true; + + udp_tunnel_nic_entry_adj(utn, table, idx, use_cnt_adj); + return true; +} + +/* Try to find existing matching entry and adjust its use count, instead of + * adding a new one. Returns true if entry was found. In case of delete the + * entry may have gotten removed in the process, in which case it will be + * queued for removal. + */ +static bool +udp_tunnel_nic_try_existing(struct net_device *dev, struct udp_tunnel_nic *utn, + struct udp_tunnel_info *ti, int use_cnt_adj) +{ + const struct udp_tunnel_nic_table_info *table; + unsigned int i, j; + + for (i = 0; i < utn->n_tables; i++) { + table = &dev->udp_tunnel_nic_info->tables[i]; + if (!udp_tunnel_nic_table_is_capable(table, ti)) + continue; + + for (j = 0; j < table->n_entries; j++) + if (udp_tunnel_nic_entry_try_adj(utn, i, j, ti, + use_cnt_adj)) + return true; + } + + return false; +} + +static bool +udp_tunnel_nic_add_existing(struct net_device *dev, struct udp_tunnel_nic *utn, + struct udp_tunnel_info *ti) +{ + return udp_tunnel_nic_try_existing(dev, utn, ti, +1); +} + +static bool +udp_tunnel_nic_del_existing(struct net_device *dev, struct udp_tunnel_nic *utn, + struct udp_tunnel_info *ti) +{ + return udp_tunnel_nic_try_existing(dev, utn, ti, -1); +} + +static bool +udp_tunnel_nic_add_new(struct net_device *dev, struct udp_tunnel_nic *utn, + struct udp_tunnel_info *ti) +{ + const struct udp_tunnel_nic_table_info *table; + unsigned int i, j; + + for (i = 0; i < utn->n_tables; i++) { + table = &dev->udp_tunnel_nic_info->tables[i]; + if (!udp_tunnel_nic_table_is_capable(table, ti)) + continue; + + for (j = 0; j < table->n_entries; j++) { + struct udp_tunnel_nic_table_entry *entry; + + entry = &utn->entries[i][j]; + if (!udp_tunnel_nic_entry_is_free(entry)) + continue; + + entry->port = ti->port; + entry->type = ti->type; + entry->use_cnt = 1; + udp_tunnel_nic_entry_queue(utn, entry, + UDP_TUNNEL_NIC_ENTRY_ADD); + return true; + } + + /* The different table may still fit this port in, but there + * are no devices currently which have multiple tables accepting + * the same tunnel type, and false positives are okay. + */ + __set_bit(i, &utn->missed); + } + + return false; +} + +static void +__udp_tunnel_nic_add_port(struct net_device *dev, struct udp_tunnel_info *ti) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic *utn; + + utn = dev->udp_tunnel_nic; + if (!utn) + return; + if (!netif_running(dev) && info->flags & UDP_TUNNEL_NIC_INFO_OPEN_ONLY) + return; + if (info->flags & UDP_TUNNEL_NIC_INFO_STATIC_IANA_VXLAN && + ti->port == htons(IANA_VXLAN_UDP_PORT)) { + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + netdev_warn(dev, "device assumes port 4789 will be used by vxlan tunnels\n"); + return; + } + + if (!udp_tunnel_nic_is_capable(dev, utn, ti)) + return; + + /* It may happen that a tunnel of one type is removed and different + * tunnel type tries to reuse its port before the device was informed. + * Rely on utn->missed to re-add this port later. + */ + if (udp_tunnel_nic_has_collision(dev, utn, ti)) + return; + + if (!udp_tunnel_nic_add_existing(dev, utn, ti)) + udp_tunnel_nic_add_new(dev, utn, ti); + + udp_tunnel_nic_device_sync(dev, utn); +} + +static void +__udp_tunnel_nic_del_port(struct net_device *dev, struct udp_tunnel_info *ti) +{ + struct udp_tunnel_nic *utn; + + utn = dev->udp_tunnel_nic; + if (!utn) + return; + + if (!udp_tunnel_nic_is_capable(dev, utn, ti)) + return; + + udp_tunnel_nic_del_existing(dev, utn, ti); + + udp_tunnel_nic_device_sync(dev, utn); +} + +static void __udp_tunnel_nic_reset_ntf(struct net_device *dev) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic *utn; + unsigned int i, j; + + ASSERT_RTNL(); + + utn = dev->udp_tunnel_nic; + if (!utn) + return; + + utn->need_sync = false; + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) { + struct udp_tunnel_nic_table_entry *entry; + + entry = &utn->entries[i][j]; + + entry->flags &= ~(UDP_TUNNEL_NIC_ENTRY_DEL | + UDP_TUNNEL_NIC_ENTRY_OP_FAIL); + /* We don't release rtnl across ops */ + WARN_ON(entry->flags & UDP_TUNNEL_NIC_ENTRY_FROZEN); + if (!entry->use_cnt) + continue; + + udp_tunnel_nic_entry_queue(utn, entry, + UDP_TUNNEL_NIC_ENTRY_ADD); + } + + __udp_tunnel_nic_device_sync(dev, utn); +} + +static size_t +__udp_tunnel_nic_dump_size(struct net_device *dev, unsigned int table) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic *utn; + unsigned int j; + size_t size; + + utn = dev->udp_tunnel_nic; + if (!utn) + return 0; + + size = 0; + for (j = 0; j < info->tables[table].n_entries; j++) { + if (!udp_tunnel_nic_entry_is_present(&utn->entries[table][j])) + continue; + + size += nla_total_size(0) + /* _TABLE_ENTRY */ + nla_total_size(sizeof(__be16)) + /* _ENTRY_PORT */ + nla_total_size(sizeof(u32)); /* _ENTRY_TYPE */ + } + + return size; +} + +static int +__udp_tunnel_nic_dump_write(struct net_device *dev, unsigned int table, + struct sk_buff *skb) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic *utn; + struct nlattr *nest; + unsigned int j; + + utn = dev->udp_tunnel_nic; + if (!utn) + return 0; + + for (j = 0; j < info->tables[table].n_entries; j++) { + if (!udp_tunnel_nic_entry_is_present(&utn->entries[table][j])) + continue; + + nest = nla_nest_start(skb, ETHTOOL_A_TUNNEL_UDP_TABLE_ENTRY); + + if (nla_put_be16(skb, ETHTOOL_A_TUNNEL_UDP_ENTRY_PORT, + utn->entries[table][j].port) || + nla_put_u32(skb, ETHTOOL_A_TUNNEL_UDP_ENTRY_TYPE, + ilog2(utn->entries[table][j].type))) + goto err_cancel; + + nla_nest_end(skb, nest); + } + + return 0; + +err_cancel: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static const struct udp_tunnel_nic_ops __udp_tunnel_nic_ops = { + .get_port = __udp_tunnel_nic_get_port, + .set_port_priv = __udp_tunnel_nic_set_port_priv, + .add_port = __udp_tunnel_nic_add_port, + .del_port = __udp_tunnel_nic_del_port, + .reset_ntf = __udp_tunnel_nic_reset_ntf, + .dump_size = __udp_tunnel_nic_dump_size, + .dump_write = __udp_tunnel_nic_dump_write, +}; + +static void +udp_tunnel_nic_flush(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + unsigned int i, j; + + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) { + int adj_cnt = -utn->entries[i][j].use_cnt; + + if (adj_cnt) + udp_tunnel_nic_entry_adj(utn, i, j, adj_cnt); + } + + __udp_tunnel_nic_device_sync(dev, utn); + + for (i = 0; i < utn->n_tables; i++) + memset(utn->entries[i], 0, array_size(info->tables[i].n_entries, + sizeof(**utn->entries))); + WARN_ON(utn->need_sync); + utn->need_replay = 0; +} + +static void +udp_tunnel_nic_replay(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic_shared_node *node; + unsigned int i, j; + + /* Freeze all the ports we are already tracking so that the replay + * does not double up the refcount. + */ + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) + udp_tunnel_nic_entry_freeze_used(&utn->entries[i][j]); + utn->missed = 0; + utn->need_replay = 0; + + if (!info->shared) { + udp_tunnel_get_rx_info(dev); + } else { + list_for_each_entry(node, &info->shared->devices, list) + udp_tunnel_get_rx_info(node->dev); + } + + for (i = 0; i < utn->n_tables; i++) + for (j = 0; j < info->tables[i].n_entries; j++) + udp_tunnel_nic_entry_unfreeze(&utn->entries[i][j]); +} + +static void udp_tunnel_nic_device_sync_work(struct work_struct *work) +{ + struct udp_tunnel_nic *utn = + container_of(work, struct udp_tunnel_nic, work); + + rtnl_lock(); + utn->work_pending = 0; + __udp_tunnel_nic_device_sync(utn->dev, utn); + + if (utn->need_replay) + udp_tunnel_nic_replay(utn->dev, utn); + rtnl_unlock(); +} + +static struct udp_tunnel_nic * +udp_tunnel_nic_alloc(const struct udp_tunnel_nic_info *info, + unsigned int n_tables) +{ + struct udp_tunnel_nic *utn; + unsigned int i; + + utn = kzalloc(sizeof(*utn), GFP_KERNEL); + if (!utn) + return NULL; + utn->n_tables = n_tables; + INIT_WORK(&utn->work, udp_tunnel_nic_device_sync_work); + + utn->entries = kmalloc_array(n_tables, sizeof(void *), GFP_KERNEL); + if (!utn->entries) + goto err_free_utn; + + for (i = 0; i < n_tables; i++) { + utn->entries[i] = kcalloc(info->tables[i].n_entries, + sizeof(*utn->entries[i]), GFP_KERNEL); + if (!utn->entries[i]) + goto err_free_prev_entries; + } + + return utn; + +err_free_prev_entries: + while (i--) + kfree(utn->entries[i]); + kfree(utn->entries); +err_free_utn: + kfree(utn); + return NULL; +} + +static void udp_tunnel_nic_free(struct udp_tunnel_nic *utn) +{ + unsigned int i; + + for (i = 0; i < utn->n_tables; i++) + kfree(utn->entries[i]); + kfree(utn->entries); + kfree(utn); +} + +static int udp_tunnel_nic_register(struct net_device *dev) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + struct udp_tunnel_nic_shared_node *node = NULL; + struct udp_tunnel_nic *utn; + unsigned int n_tables, i; + + BUILD_BUG_ON(sizeof(utn->missed) * BITS_PER_BYTE < + UDP_TUNNEL_NIC_MAX_TABLES); + /* Expect use count of at most 2 (IPv4, IPv6) per device */ + BUILD_BUG_ON(UDP_TUNNEL_NIC_USE_CNT_MAX < + UDP_TUNNEL_NIC_MAX_SHARING_DEVICES * 2); + + /* Check that the driver info is sane */ + if (WARN_ON(!info->set_port != !info->unset_port) || + WARN_ON(!info->set_port == !info->sync_table) || + WARN_ON(!info->tables[0].n_entries)) + return -EINVAL; + + if (WARN_ON(info->shared && + info->flags & UDP_TUNNEL_NIC_INFO_OPEN_ONLY)) + return -EINVAL; + + n_tables = 1; + for (i = 1; i < UDP_TUNNEL_NIC_MAX_TABLES; i++) { + if (!info->tables[i].n_entries) + continue; + + n_tables++; + if (WARN_ON(!info->tables[i - 1].n_entries)) + return -EINVAL; + } + + /* Create UDP tunnel state structures */ + if (info->shared) { + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + + node->dev = dev; + } + + if (info->shared && info->shared->udp_tunnel_nic_info) { + utn = info->shared->udp_tunnel_nic_info; + } else { + utn = udp_tunnel_nic_alloc(info, n_tables); + if (!utn) { + kfree(node); + return -ENOMEM; + } + } + + if (info->shared) { + if (!info->shared->udp_tunnel_nic_info) { + INIT_LIST_HEAD(&info->shared->devices); + info->shared->udp_tunnel_nic_info = utn; + } + + list_add_tail(&node->list, &info->shared->devices); + } + + utn->dev = dev; + dev_hold(dev); + dev->udp_tunnel_nic = utn; + + if (!(info->flags & UDP_TUNNEL_NIC_INFO_OPEN_ONLY)) + udp_tunnel_get_rx_info(dev); + + return 0; +} + +static void +udp_tunnel_nic_unregister(struct net_device *dev, struct udp_tunnel_nic *utn) +{ + const struct udp_tunnel_nic_info *info = dev->udp_tunnel_nic_info; + + /* For a shared table remove this dev from the list of sharing devices + * and if there are other devices just detach. + */ + if (info->shared) { + struct udp_tunnel_nic_shared_node *node, *first; + + list_for_each_entry(node, &info->shared->devices, list) + if (node->dev == dev) + break; + if (list_entry_is_head(node, &info->shared->devices, list)) + return; + + list_del(&node->list); + kfree(node); + + first = list_first_entry_or_null(&info->shared->devices, + typeof(*first), list); + if (first) { + udp_tunnel_drop_rx_info(dev); + utn->dev = first->dev; + goto release_dev; + } + + info->shared->udp_tunnel_nic_info = NULL; + } + + /* Flush before we check work, so we don't waste time adding entries + * from the work which we will boot immediately. + */ + udp_tunnel_nic_flush(dev, utn); + + /* Wait for the work to be done using the state, netdev core will + * retry unregister until we give up our reference on this device. + */ + if (utn->work_pending) + return; + + udp_tunnel_nic_free(utn); +release_dev: + dev->udp_tunnel_nic = NULL; + dev_put(dev); +} + +static int +udp_tunnel_nic_netdevice_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + const struct udp_tunnel_nic_info *info; + struct udp_tunnel_nic *utn; + + info = dev->udp_tunnel_nic_info; + if (!info) + return NOTIFY_DONE; + + if (event == NETDEV_REGISTER) { + int err; + + err = udp_tunnel_nic_register(dev); + if (err) + netdev_WARN(dev, "failed to register for UDP tunnel offloads: %d", err); + return notifier_from_errno(err); + } + /* All other events will need the udp_tunnel_nic state */ + utn = dev->udp_tunnel_nic; + if (!utn) + return NOTIFY_DONE; + + if (event == NETDEV_UNREGISTER) { + udp_tunnel_nic_unregister(dev, utn); + return NOTIFY_OK; + } + + /* All other events only matter if NIC has to be programmed open */ + if (!(info->flags & UDP_TUNNEL_NIC_INFO_OPEN_ONLY)) + return NOTIFY_DONE; + + if (event == NETDEV_UP) { + WARN_ON(!udp_tunnel_nic_is_empty(dev, utn)); + udp_tunnel_get_rx_info(dev); + return NOTIFY_OK; + } + if (event == NETDEV_GOING_DOWN) { + udp_tunnel_nic_flush(dev, utn); + return NOTIFY_OK; + } + + return NOTIFY_DONE; +} + +static struct notifier_block udp_tunnel_nic_notifier_block __read_mostly = { + .notifier_call = udp_tunnel_nic_netdevice_event, +}; + +static int __init udp_tunnel_nic_init_module(void) +{ + int err; + + udp_tunnel_nic_workqueue = alloc_ordered_workqueue("udp_tunnel_nic", 0); + if (!udp_tunnel_nic_workqueue) + return -ENOMEM; + + rtnl_lock(); + udp_tunnel_nic_ops = &__udp_tunnel_nic_ops; + rtnl_unlock(); + + err = register_netdevice_notifier(&udp_tunnel_nic_notifier_block); + if (err) + goto err_unset_ops; + + return 0; + +err_unset_ops: + rtnl_lock(); + udp_tunnel_nic_ops = NULL; + rtnl_unlock(); + destroy_workqueue(udp_tunnel_nic_workqueue); + return err; +} +late_initcall(udp_tunnel_nic_init_module); + +static void __exit udp_tunnel_nic_cleanup_module(void) +{ + unregister_netdevice_notifier(&udp_tunnel_nic_notifier_block); + + rtnl_lock(); + udp_tunnel_nic_ops = NULL; + rtnl_unlock(); + + destroy_workqueue(udp_tunnel_nic_workqueue); +} +module_exit(udp_tunnel_nic_cleanup_module); + +MODULE_LICENSE("GPL"); diff --git a/net/ipv4/udp_tunnel_stub.c b/net/ipv4/udp_tunnel_stub.c new file mode 100644 index 000000000000..c4b2888f5fef --- /dev/null +++ b/net/ipv4/udp_tunnel_stub.c @@ -0,0 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0-only +// Copyright (c) 2020 Facebook Inc. + +#include <net/udp_tunnel.h> + +const struct udp_tunnel_nic_ops *udp_tunnel_nic_ops; +EXPORT_SYMBOL_GPL(udp_tunnel_nic_ops); diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 5936d66d1ce2..e0c9cc39b81e 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -17,6 +17,14 @@ struct udp_table udplite_table __read_mostly; EXPORT_SYMBOL(udplite_table); +/* Designate sk as UDP-Lite socket */ +static int udplite_sk_init(struct sock *sk) +{ + udp_init_sock(sk); + udp_sk(sk)->pcflag = UDPLITE_BIT; + return 0; +} + static int udplite_rcv(struct sk_buff *skb) { return __udp4_lib_rcv(skb, &udplite_table, IPPROTO_UDPLITE); @@ -31,7 +39,6 @@ static const struct net_protocol udplite_protocol = { .handler = udplite_rcv, .err_handler = udplite_err, .no_policy = 1, - .netns_ok = 1, }; struct proto udplite_prot = { @@ -52,14 +59,13 @@ struct proto udplite_prot = { .unhash = udp_lib_unhash, .rehash = udp_v4_rehash, .get_port = udp_v4_get_port, + .memory_allocated = &udp_memory_allocated, + .per_cpu_fw_alloc = &udp_memory_per_cpu_fw_alloc, + .sysctl_mem = sysctl_udp_mem, .obj_size = sizeof(struct udp_sock), .h.udp_table = &udplite_table, -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_udp_setsockopt, - .compat_getsockopt = compat_udp_getsockopt, -#endif }; EXPORT_SYMBOL(udplite_prot); diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c index f8de2482a529..ad2afeef4f10 100644 --- a/net/ipv4/xfrm4_input.c +++ b/net/ipv4/xfrm4_input.c @@ -18,11 +18,6 @@ #include <net/ip.h> #include <net/xfrm.h> -int xfrm4_extract_input(struct xfrm_state *x, struct sk_buff *skb) -{ - return xfrm4_extract_header(skb); -} - static int xfrm4_rcv_encap_finish2(struct net *net, struct sock *sk, struct sk_buff *skb) { diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index 89ba7c87de5d..3cff51ba72bb 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -14,79 +14,18 @@ #include <net/xfrm.h> #include <net/icmp.h> -static int xfrm4_tunnel_check_size(struct sk_buff *skb) -{ - int mtu, ret = 0; - - if (IPCB(skb)->flags & IPSKB_XFRM_TUNNEL_SIZE) - goto out; - - if (!(ip_hdr(skb)->frag_off & htons(IP_DF)) || skb->ignore_df) - goto out; - - mtu = dst_mtu(skb_dst(skb)); - if ((!skb_is_gso(skb) && skb->len > mtu) || - (skb_is_gso(skb) && - !skb_gso_validate_network_len(skb, ip_skb_dst_mtu(skb->sk, skb)))) { - skb->protocol = htons(ETH_P_IP); - - if (skb->sk) - xfrm_local_error(skb, mtu); - else - icmp_send(skb, ICMP_DEST_UNREACH, - ICMP_FRAG_NEEDED, htonl(mtu)); - ret = -EMSGSIZE; - } -out: - return ret; -} - -int xfrm4_extract_output(struct xfrm_state *x, struct sk_buff *skb) -{ - int err; - - err = xfrm4_tunnel_check_size(skb); - if (err) - return err; - - XFRM_MODE_SKB_CB(skb)->protocol = ip_hdr(skb)->protocol; - - return xfrm4_extract_header(skb); -} - -int xfrm4_output_finish(struct sock *sk, struct sk_buff *skb) -{ - memset(IPCB(skb), 0, sizeof(*IPCB(skb))); - -#ifdef CONFIG_NETFILTER - IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED; -#endif - - return xfrm_output(sk, skb); -} - static int __xfrm4_output(struct net *net, struct sock *sk, struct sk_buff *skb) { +#ifdef CONFIG_NETFILTER struct xfrm_state *x = skb_dst(skb)->xfrm; - const struct xfrm_state_afinfo *afinfo; - int ret = -EAFNOSUPPORT; -#ifdef CONFIG_NETFILTER if (!x) { IPCB(skb)->flags |= IPSKB_REROUTED; return dst_output(net, sk, skb); } #endif - rcu_read_lock(); - afinfo = xfrm_state_afinfo_get_rcu(x->outer_mode.family); - if (likely(afinfo)) - ret = afinfo->output_finish(sk, skb); - else - kfree_skb(skb); - rcu_read_unlock(); - - return ret; + return xfrm_output(sk, skb); } int xfrm4_output(struct net *net, struct sock *sk, struct sk_buff *skb) diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 9ebd54752e03..3d0dfa6cf9f9 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -28,13 +28,11 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4, memset(fl4, 0, sizeof(*fl4)); fl4->daddr = daddr->a4; fl4->flowi4_tos = tos; - fl4->flowi4_oif = l3mdev_master_ifindex_by_index(net, oif); + fl4->flowi4_l3mdev = l3mdev_master_ifindex_by_index(net, oif); fl4->flowi4_mark = mark; if (saddr) fl4->saddr = saddr->a4; - fl4->flowi4_flags = FLOWI_FLAG_SKIP_NH_OIF; - rt = __ip_route_output_key(net, fl4); if (!IS_ERR(rt)) return &rt->dst; @@ -77,7 +75,7 @@ static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, xdst->u.rt.rt_iif = fl4->flowi4_iif; xdst->u.dst.dev = dev; - dev_hold(dev); + netdev_hold(dev, &xdst->u.dst.dev_tracker, GFP_ATOMIC); /* Sheit... I remember I did this right. Apparently, * it was magically lost, so this code needs audit */ diff --git a/net/ipv4/xfrm4_protocol.c b/net/ipv4/xfrm4_protocol.c index ea595c8549c7..b146ce88c5d0 100644 --- a/net/ipv4/xfrm4_protocol.c +++ b/net/ipv4/xfrm4_protocol.c @@ -181,21 +181,18 @@ static const struct net_protocol esp4_protocol = { .handler = xfrm4_esp_rcv, .err_handler = xfrm4_esp_err, .no_policy = 1, - .netns_ok = 1, }; static const struct net_protocol ah4_protocol = { .handler = xfrm4_ah_rcv, .err_handler = xfrm4_ah_err, .no_policy = 1, - .netns_ok = 1, }; static const struct net_protocol ipcomp4_protocol = { .handler = xfrm4_ipcomp_rcv, .err_handler = xfrm4_ipcomp_err, .no_policy = 1, - .netns_ok = 1, }; static const struct xfrm_input_afinfo xfrm4_input_afinfo = { @@ -307,4 +304,3 @@ void __init xfrm4_protocol_init(void) { xfrm_input_register_afinfo(&xfrm4_input_afinfo); } -EXPORT_SYMBOL(xfrm4_protocol_init); diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c index f8ed3c3bb928..87d4db591488 100644 --- a/net/ipv4/xfrm4_state.c +++ b/net/ipv4/xfrm4_state.c @@ -8,36 +8,12 @@ * */ -#include <net/ip.h> #include <net/xfrm.h> -#include <linux/pfkeyv2.h> -#include <linux/ipsec.h> -#include <linux/netfilter_ipv4.h> -#include <linux/export.h> - -int xfrm4_extract_header(struct sk_buff *skb) -{ - const struct iphdr *iph = ip_hdr(skb); - - XFRM_MODE_SKB_CB(skb)->ihl = sizeof(*iph); - XFRM_MODE_SKB_CB(skb)->id = iph->id; - XFRM_MODE_SKB_CB(skb)->frag_off = iph->frag_off; - XFRM_MODE_SKB_CB(skb)->tos = iph->tos; - XFRM_MODE_SKB_CB(skb)->ttl = iph->ttl; - XFRM_MODE_SKB_CB(skb)->optlen = iph->ihl * 4 - sizeof(*iph); - memset(XFRM_MODE_SKB_CB(skb)->flow_lbl, 0, - sizeof(XFRM_MODE_SKB_CB(skb)->flow_lbl)); - - return 0; -} static struct xfrm_state_afinfo xfrm4_state_afinfo = { .family = AF_INET, .proto = IPPROTO_IPIP, .output = xfrm4_output, - .output_finish = xfrm4_output_finish, - .extract_input = xfrm4_extract_input, - .extract_output = xfrm4_extract_output, .transport_finish = xfrm4_transport_finish, .local_error = xfrm4_local_error, }; diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c index dc19aff7c2e0..8489fa106583 100644 --- a/net/ipv4/xfrm4_tunnel.c +++ b/net/ipv4/xfrm4_tunnel.c @@ -8,9 +8,7 @@ #include <linux/skbuff.h> #include <linux/module.h> -#include <linux/mutex.h> #include <net/xfrm.h> -#include <net/ip.h> #include <net/protocol.h> static int ipip_output(struct xfrm_state *x, struct sk_buff *skb) @@ -24,13 +22,17 @@ static int ipip_xfrm_rcv(struct xfrm_state *x, struct sk_buff *skb) return ip_hdr(skb)->protocol; } -static int ipip_init_state(struct xfrm_state *x) +static int ipip_init_state(struct xfrm_state *x, struct netlink_ext_ack *extack) { - if (x->props.mode != XFRM_MODE_TUNNEL) + if (x->props.mode != XFRM_MODE_TUNNEL) { + NL_SET_ERR_MSG(extack, "IPv4 tunnel can only be used with tunnel mode"); return -EINVAL; + } - if (x->encap) + if (x->encap) { + NL_SET_ERR_MSG(extack, "IPv4 tunnel is not compatible with encapsulation"); return -EINVAL; + } x->props.header_len = sizeof(struct iphdr); @@ -42,7 +44,6 @@ static void ipip_destroy(struct xfrm_state *x) } static const struct xfrm_type ipip_type = { - .description = "IPIP", .owner = THIS_MODULE, .proto = IPPROTO_IPIP, .init_state = ipip_init_state, @@ -64,14 +65,14 @@ static int xfrm_tunnel_err(struct sk_buff *skb, u32 info) static struct xfrm_tunnel xfrm_tunnel_handler __read_mostly = { .handler = xfrm_tunnel_rcv, .err_handler = xfrm_tunnel_err, - .priority = 3, + .priority = 4, }; #if IS_ENABLED(CONFIG_IPV6) static struct xfrm_tunnel xfrm64_tunnel_handler __read_mostly = { .handler = xfrm_tunnel_rcv, .err_handler = xfrm_tunnel_err, - .priority = 2, + .priority = 3, }; #endif |