From 9cacf81f8161111db25f98e78a7a0e32ae142b3f Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Fri, 15 Jan 2021 08:34:59 -0800 Subject: [PATCH] bpf: Remove extra lock_sock for TCP_ZEROCOPY_RECEIVE Add custom implementation of getsockopt hook for TCP_ZEROCOPY_RECEIVE. We skip generic hooks for TCP_ZEROCOPY_RECEIVE and have a custom call in do_tcp_getsockopt using the on-stack data. This removes 3% overhead for locking/unlocking the socket. Without this patch: 3.38% 0.07% tcp_mmap [kernel.kallsyms] [k] __cgroup_bpf_run_filter_getsockopt | --3.30%--__cgroup_bpf_run_filter_getsockopt | --0.81%--__kmalloc With the patch applied: 0.52% 0.12% tcp_mmap [kernel.kallsyms] [k] __cgroup_bpf_run_filter_getsockopt_kern Note, exporting uapi/tcp.h requires removing netinet/tcp.h from test_progs.h because those headers have confliciting definitions. Signed-off-by: Stanislav Fomichev Signed-off-by: Alexei Starovoitov Acked-by: Martin KaFai Lau Link: https://lore.kernel.org/bpf/20210115163501.805133-2-sdf@google.com --- include/linux/bpf-cgroup.h | 27 +- include/linux/indirect_call_wrapper.h | 6 + include/net/sock.h | 2 + include/net/tcp.h | 1 + kernel/bpf/cgroup.c | 46 +++ net/ipv4/tcp.c | 14 + net/ipv4/tcp_ipv4.c | 1 + net/ipv6/tcp_ipv6.c | 1 + net/socket.c | 3 + tools/include/uapi/linux/tcp.h | 357 ++++++++++++++++++ .../selftests/bpf/prog_tests/bpf_tcp_ca.c | 1 + .../selftests/bpf/prog_tests/cls_redirect.c | 1 + .../selftests/bpf/prog_tests/sockmap_basic.c | 1 + .../selftests/bpf/prog_tests/sockopt_sk.c | 28 ++ .../testing/selftests/bpf/progs/sockopt_sk.c | 23 +- tools/testing/selftests/bpf/test_progs.h | 1 - 16 files changed, 506 insertions(+), 7 deletions(-) create mode 100644 tools/include/uapi/linux/tcp.h diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h index 72e69a0e1e8c..bcb2915e6124 100644 --- a/include/linux/bpf-cgroup.h +++ b/include/linux/bpf-cgroup.h @@ -147,6 +147,10 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level, int __user *optlen, int max_optlen, int retval); +int __cgroup_bpf_run_filter_getsockopt_kern(struct sock *sk, int level, + int optname, void *optval, + int *optlen, int retval); + static inline enum bpf_cgroup_storage_type cgroup_storage_type( struct bpf_map *map) { @@ -364,10 +368,23 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key, ({ \ int __ret = retval; \ if (cgroup_bpf_enabled) \ - __ret = __cgroup_bpf_run_filter_getsockopt(sock, level, \ - optname, optval, \ - optlen, max_optlen, \ - retval); \ + if (!(sock)->sk_prot->bpf_bypass_getsockopt || \ + !INDIRECT_CALL_INET_1((sock)->sk_prot->bpf_bypass_getsockopt, \ + tcp_bpf_bypass_getsockopt, \ + level, optname)) \ + __ret = __cgroup_bpf_run_filter_getsockopt( \ + sock, level, optname, optval, optlen, \ + max_optlen, retval); \ + __ret; \ +}) + +#define BPF_CGROUP_RUN_PROG_GETSOCKOPT_KERN(sock, level, optname, optval, \ + optlen, retval) \ +({ \ + int __ret = retval; \ + if (cgroup_bpf_enabled) \ + __ret = __cgroup_bpf_run_filter_getsockopt_kern( \ + sock, level, optname, optval, optlen, retval); \ __ret; \ }) @@ -452,6 +469,8 @@ static inline int bpf_percpu_cgroup_storage_update(struct bpf_map *map, #define BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN(optlen) ({ 0; }) #define BPF_CGROUP_RUN_PROG_GETSOCKOPT(sock, level, optname, optval, \ optlen, max_optlen, retval) ({ retval; }) +#define BPF_CGROUP_RUN_PROG_GETSOCKOPT_KERN(sock, level, optname, optval, \ + optlen, retval) ({ retval; }) #define BPF_CGROUP_RUN_PROG_SETSOCKOPT(sock, level, optname, optval, optlen, \ kernel_optval) ({ 0; }) diff --git a/include/linux/indirect_call_wrapper.h b/include/linux/indirect_call_wrapper.h index 54c02c84906a..cfcfef37b2f1 100644 --- a/include/linux/indirect_call_wrapper.h +++ b/include/linux/indirect_call_wrapper.h @@ -60,4 +60,10 @@ #define INDIRECT_CALL_INET(f, f2, f1, ...) f(__VA_ARGS__) #endif +#if IS_ENABLED(CONFIG_INET) +#define INDIRECT_CALL_INET_1(f, f1, ...) INDIRECT_CALL_1(f, f1, __VA_ARGS__) +#else +#define INDIRECT_CALL_INET_1(f, f1, ...) f(__VA_ARGS__) +#endif + #endif diff --git a/include/net/sock.h b/include/net/sock.h index 129d200bccb4..7644ea64a376 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1174,6 +1174,8 @@ struct proto { int (*backlog_rcv) (struct sock *sk, struct sk_buff *skb); + bool (*bpf_bypass_getsockopt)(int level, + int optname); void (*release_cb)(struct sock *sk); diff --git a/include/net/tcp.h b/include/net/tcp.h index 78d13c88720f..4bb42fb19711 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -403,6 +403,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, struct poll_table_struct *wait); int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen); +bool tcp_bpf_bypass_getsockopt(int level, int optname); int tcp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval, unsigned int optlen); void tcp_set_keepalive(struct sock *sk, int val); diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c index 96555a8a2c54..416e7738981b 100644 --- a/kernel/bpf/cgroup.c +++ b/kernel/bpf/cgroup.c @@ -1486,6 +1486,52 @@ out: sockopt_free_buf(&ctx); return ret; } + +int __cgroup_bpf_run_filter_getsockopt_kern(struct sock *sk, int level, + int optname, void *optval, + int *optlen, int retval) +{ + struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data); + struct bpf_sockopt_kern ctx = { + .sk = sk, + .level = level, + .optname = optname, + .retval = retval, + .optlen = *optlen, + .optval = optval, + .optval_end = optval + *optlen, + }; + int ret; + + /* Note that __cgroup_bpf_run_filter_getsockopt doesn't copy + * user data back into BPF buffer when reval != 0. This is + * done as an optimization to avoid extra copy, assuming + * kernel won't populate the data in case of an error. + * Here we always pass the data and memset() should + * be called if that data shouldn't be "exported". + */ + + ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[BPF_CGROUP_GETSOCKOPT], + &ctx, BPF_PROG_RUN); + if (!ret) + return -EPERM; + + if (ctx.optlen > *optlen) + return -EFAULT; + + /* BPF programs only allowed to set retval to 0, not some + * arbitrary value. + */ + if (ctx.retval != 0 && ctx.retval != retval) + return -EFAULT; + + /* BPF programs can shrink the buffer, export the modifications. + */ + if (ctx.optlen != 0) + *optlen = ctx.optlen; + + return ctx.retval; +} #endif static ssize_t sysctl_cpy_dir(const struct ctl_dir *dir, char **bufp, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 856ae516ac18..26aa923cf522 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -4099,6 +4099,8 @@ static int do_tcp_getsockopt(struct sock *sk, int level, return -EFAULT; lock_sock(sk); err = tcp_zerocopy_receive(sk, &zc); + err = BPF_CGROUP_RUN_PROG_GETSOCKOPT_KERN(sk, level, optname, + &zc, &len, err); release_sock(sk); if (len >= offsetofend(struct tcp_zerocopy_receive, err)) goto zerocopy_rcv_sk_err; @@ -4133,6 +4135,18 @@ zerocopy_rcv_out: 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) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 777306b5bc22..62b6fd385a47 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2793,6 +2793,7 @@ 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, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0e1509b02cb3..8539715ff035 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2121,6 +2121,7 @@ struct proto tcpv6_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, diff --git a/net/socket.c b/net/socket.c index 33e8b6c4e1d3..7f0617ab5437 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2126,6 +2126,9 @@ SYSCALL_DEFINE5(setsockopt, int, fd, int, level, int, optname, return __sys_setsockopt(fd, level, optname, optval, optlen); } +INDIRECT_CALLABLE_DECLARE(bool tcp_bpf_bypass_getsockopt(int level, + int optname)); + /* * Get a socket option. Because we don't know the option lengths we have * to pass a user mode parameter for the protocols to sort out. diff --git a/tools/include/uapi/linux/tcp.h b/tools/include/uapi/linux/tcp.h new file mode 100644 index 000000000000..13ceeb395eb8 --- /dev/null +++ b/tools/include/uapi/linux/tcp.h @@ -0,0 +1,357 @@ +/* SPDX-License-Identifier: GPL-2.0+ WITH Linux-syscall-note */ +/* + * INET An implementation of the TCP/IP protocol suite for the LINUX + * operating system. INET is implemented using the BSD Socket + * interface as the means of communication with the user level. + * + * Definitions for the TCP protocol. + * + * Version: @(#)tcp.h 1.0.2 04/28/93 + * + * Author: Fred N. van Kempen, + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#ifndef _UAPI_LINUX_TCP_H +#define _UAPI_LINUX_TCP_H + +#include +#include +#include + +struct tcphdr { + __be16 source; + __be16 dest; + __be32 seq; + __be32 ack_seq; +#if defined(__LITTLE_ENDIAN_BITFIELD) + __u16 res1:4, + doff:4, + fin:1, + syn:1, + rst:1, + psh:1, + ack:1, + urg:1, + ece:1, + cwr:1; +#elif defined(__BIG_ENDIAN_BITFIELD) + __u16 doff:4, + res1:4, + cwr:1, + ece:1, + urg:1, + ack:1, + psh:1, + rst:1, + syn:1, + fin:1; +#else +#error "Adjust your defines" +#endif + __be16 window; + __sum16 check; + __be16 urg_ptr; +}; + +/* + * The union cast uses a gcc extension to avoid aliasing problems + * (union is compatible to any of its members) + * This means this part of the code is -fstrict-aliasing safe now. + */ +union tcp_word_hdr { + struct tcphdr hdr; + __be32 words[5]; +}; + +#define tcp_flag_word(tp) ( ((union tcp_word_hdr *)(tp))->words [3]) + +enum { + TCP_FLAG_CWR = __constant_cpu_to_be32(0x00800000), + TCP_FLAG_ECE = __constant_cpu_to_be32(0x00400000), + TCP_FLAG_URG = __constant_cpu_to_be32(0x00200000), + TCP_FLAG_ACK = __constant_cpu_to_be32(0x00100000), + TCP_FLAG_PSH = __constant_cpu_to_be32(0x00080000), + TCP_FLAG_RST = __constant_cpu_to_be32(0x00040000), + TCP_FLAG_SYN = __constant_cpu_to_be32(0x00020000), + TCP_FLAG_FIN = __constant_cpu_to_be32(0x00010000), + TCP_RESERVED_BITS = __constant_cpu_to_be32(0x0F000000), + TCP_DATA_OFFSET = __constant_cpu_to_be32(0xF0000000) +}; + +/* + * TCP general constants + */ +#define TCP_MSS_DEFAULT 536U /* IPv4 (RFC1122, RFC2581) */ +#define TCP_MSS_DESIRED 1220U /* IPv6 (tunneled), EDNS0 (RFC3226) */ + +/* TCP socket options */ +#define TCP_NODELAY 1 /* Turn off Nagle's algorithm. */ +#define TCP_MAXSEG 2 /* Limit MSS */ +#define TCP_CORK 3 /* Never send partially complete segments */ +#define TCP_KEEPIDLE 4 /* Start keeplives after this period */ +#define TCP_KEEPINTVL 5 /* Interval between keepalives */ +#define TCP_KEEPCNT 6 /* Number of keepalives before death */ +#define TCP_SYNCNT 7 /* Number of SYN retransmits */ +#define TCP_LINGER2 8 /* Life time of orphaned FIN-WAIT-2 state */ +#define TCP_DEFER_ACCEPT 9 /* Wake up listener only when data arrive */ +#define TCP_WINDOW_CLAMP 10 /* Bound advertised window */ +#define TCP_INFO 11 /* Information about this connection. */ +#define TCP_QUICKACK 12 /* Block/reenable quick acks */ +#define TCP_CONGESTION 13 /* Congestion control algorithm */ +#define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */ +#define TCP_THIN_LINEAR_TIMEOUTS 16 /* Use linear timeouts for thin streams*/ +#define TCP_THIN_DUPACK 17 /* Fast retrans. after 1 dupack */ +#define TCP_USER_TIMEOUT 18 /* How long for loss retry before timeout */ +#define TCP_REPAIR 19 /* TCP sock is under repair right now */ +#define TCP_REPAIR_QUEUE 20 +#define TCP_QUEUE_SEQ 21 +#define TCP_REPAIR_OPTIONS 22 +#define TCP_FASTOPEN 23 /* Enable FastOpen on listeners */ +#define TCP_TIMESTAMP 24 +#define TCP_NOTSENT_LOWAT 25 /* limit number of unsent bytes in write queue */ +#define TCP_CC_INFO 26 /* Get Congestion Control (optional) info */ +#define TCP_SAVE_SYN 27 /* Record SYN headers for new connections */ +#define TCP_SAVED_SYN 28 /* Get SYN headers recorded for connection */ +#define TCP_REPAIR_WINDOW 29 /* Get/set window parameters */ +#define TCP_FASTOPEN_CONNECT 30 /* Attempt FastOpen with connect */ +#define TCP_ULP 31 /* Attach a ULP to a TCP connection */ +#define TCP_MD5SIG_EXT 32 /* TCP MD5 Signature with extensions */ +#define TCP_FASTOPEN_KEY 33 /* Set the key for Fast Open (cookie) */ +#define TCP_FASTOPEN_NO_COOKIE 34 /* Enable TFO without a TFO cookie */ +#define TCP_ZEROCOPY_RECEIVE 35 +#define TCP_INQ 36 /* Notify bytes available to read as a cmsg on read */ + +#define TCP_CM_INQ TCP_INQ + +#define TCP_TX_DELAY 37 /* delay outgoing packets by XX usec */ + + +#define TCP_REPAIR_ON 1 +#define TCP_REPAIR_OFF 0 +#define TCP_REPAIR_OFF_NO_WP -1 /* Turn off without window probes */ + +struct tcp_repair_opt { + __u32 opt_code; + __u32 opt_val; +}; + +struct tcp_repair_window { + __u32 snd_wl1; + __u32 snd_wnd; + __u32 max_window; + + __u32 rcv_wnd; + __u32 rcv_wup; +}; + +enum { + TCP_NO_QUEUE, + TCP_RECV_QUEUE, + TCP_SEND_QUEUE, + TCP_QUEUES_NR, +}; + +/* why fastopen failed from client perspective */ +enum tcp_fastopen_client_fail { + TFO_STATUS_UNSPEC, /* catch-all */ + TFO_COOKIE_UNAVAILABLE, /* if not in TFO_CLIENT_NO_COOKIE mode */ + TFO_DATA_NOT_ACKED, /* SYN-ACK did not ack SYN data */ + TFO_SYN_RETRANSMITTED, /* SYN-ACK did not ack SYN data after timeout */ +}; + +/* for TCP_INFO socket option */ +#define TCPI_OPT_TIMESTAMPS 1 +#define TCPI_OPT_SACK 2 +#define TCPI_OPT_WSCALE 4 +#define TCPI_OPT_ECN 8 /* ECN was negociated at TCP session init */ +#define TCPI_OPT_ECN_SEEN 16 /* we received at least one packet with ECT */ +#define TCPI_OPT_SYN_DATA 32 /* SYN-ACK acked data in SYN sent or rcvd */ + +/* + * Sender's congestion state indicating normal or abnormal situations + * in the last round of packets sent. The state is driven by the ACK + * information and timer events. + */ +enum tcp_ca_state { + /* + * Nothing bad has been observed recently. + * No apparent reordering, packet loss, or ECN marks. + */ + TCP_CA_Open = 0, +#define TCPF_CA_Open (1< +#include #include #include "bpf_dctcp.skel.h" #include "bpf_cubic.skel.h" diff --git a/tools/testing/selftests/bpf/prog_tests/cls_redirect.c b/tools/testing/selftests/bpf/prog_tests/cls_redirect.c index 9781d85cb223..e075d03ab630 100644 --- a/tools/testing/selftests/bpf/prog_tests/cls_redirect.c +++ b/tools/testing/selftests/bpf/prog_tests/cls_redirect.c @@ -7,6 +7,7 @@ #include #include +#include #include diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c index 85f73261fab0..b8b48cac2ac3 100644 --- a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c +++ b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 // Copyright (c) 2020 Cloudflare #include +#include #include "test_progs.h" #include "test_skmsg_load_helpers.skel.h" diff --git a/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c b/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c index b25c9c45c148..d5b44b135c00 100644 --- a/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c +++ b/tools/testing/selftests/bpf/prog_tests/sockopt_sk.c @@ -2,6 +2,12 @@ #include #include "cgroup_helpers.h" +#include + +#ifndef SOL_TCP +#define SOL_TCP IPPROTO_TCP +#endif + #define SOL_CUSTOM 0xdeadbeef static int getsetsockopt(void) @@ -11,6 +17,7 @@ static int getsetsockopt(void) char u8[4]; __u32 u32; char cc[16]; /* TCP_CA_NAME_MAX */ + struct tcp_zerocopy_receive zc; } buf = {}; socklen_t optlen; char *big_buf = NULL; @@ -154,6 +161,27 @@ static int getsetsockopt(void) goto err; } + /* TCP_ZEROCOPY_RECEIVE triggers */ + memset(&buf, 0, sizeof(buf)); + optlen = sizeof(buf.zc); + err = getsockopt(fd, SOL_TCP, TCP_ZEROCOPY_RECEIVE, &buf, &optlen); + if (err) { + log_err("Unexpected getsockopt(TCP_ZEROCOPY_RECEIVE) err=%d errno=%d", + err, errno); + goto err; + } + + memset(&buf, 0, sizeof(buf)); + buf.zc.address = 12345; /* rejected by BPF */ + optlen = sizeof(buf.zc); + errno = 0; + err = getsockopt(fd, SOL_TCP, TCP_ZEROCOPY_RECEIVE, &buf, &optlen); + if (errno != EPERM) { + log_err("Unexpected getsockopt(TCP_ZEROCOPY_RECEIVE) err=%d errno=%d", + err, errno); + goto err; + } + free(big_buf); close(fd); return 0; diff --git a/tools/testing/selftests/bpf/progs/sockopt_sk.c b/tools/testing/selftests/bpf/progs/sockopt_sk.c index 712df7b49cb1..d3597f81e6e9 100644 --- a/tools/testing/selftests/bpf/progs/sockopt_sk.c +++ b/tools/testing/selftests/bpf/progs/sockopt_sk.c @@ -1,8 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 #include -#include -#include +#include #include +#include #include char _license[] SEC("license") = "GPL"; @@ -12,6 +12,10 @@ __u32 _version SEC("version") = 1; #define PAGE_SIZE 4096 #endif +#ifndef SOL_TCP +#define SOL_TCP IPPROTO_TCP +#endif + #define SOL_CUSTOM 0xdeadbeef struct sockopt_sk { @@ -57,6 +61,21 @@ int _getsockopt(struct bpf_sockopt *ctx) return 1; } + if (ctx->level == SOL_TCP && ctx->optname == TCP_ZEROCOPY_RECEIVE) { + /* Verify that TCP_ZEROCOPY_RECEIVE triggers. + * It has a custom implementation for performance + * reasons. + */ + + if (optval + sizeof(struct tcp_zerocopy_receive) > optval_end) + return 0; /* EPERM, bounds check */ + + if (((struct tcp_zerocopy_receive *)optval)->address != 0) + return 0; /* EPERM, unexpected data */ + + return 1; + } + if (ctx->level == SOL_IP && ctx->optname == IP_FREEBIND) { if (optval + 1 > optval_end) return 0; /* EPERM, bounds check */ diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h index e49e2fdde942..f7c2fd89d01a 100644 --- a/tools/testing/selftests/bpf/test_progs.h +++ b/tools/testing/selftests/bpf/test_progs.h @@ -16,7 +16,6 @@ typedef __u16 __sum16; #include #include #include -#include #include #include #include