From d6990976af7c5d8f55903bfb4289b6fb030bf754 Mon Sep 17 00:00:00 2001 From: Eyal Birger Date: Thu, 7 Jun 2018 10:11:02 +0300 Subject: [PATCH 01/41] vti6: fix PMTU caching and reporting on xmit When setting the skb->dst before doing the MTU check, the route PMTU caching and reporting is done on the new dst which is about to be released. Instead, PMTU handling should be done using the original dst. This is aligned with IPv4 VTI. Fixes: ccd740cbc6 ("vti6: Add pmtu handling to vti6_xmit.") Signed-off-by: Eyal Birger Signed-off-by: Steffen Klassert --- net/ipv6/ip6_vti.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index b7f28deddaea..c72ae3a4fe09 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -480,10 +480,6 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl) goto tx_err_dst_release; } - skb_scrub_packet(skb, !net_eq(t->net, dev_net(dev))); - skb_dst_set(skb, dst); - skb->dev = skb_dst(skb)->dev; - mtu = dst_mtu(dst); if (!skb->ignore_df && skb->len > mtu) { skb_dst_update_pmtu(skb, mtu); @@ -498,9 +494,14 @@ vti6_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl) htonl(mtu)); } - return -EMSGSIZE; + err = -EMSGSIZE; + goto tx_err_dst_release; } + skb_scrub_packet(skb, !net_eq(t->net, dev_net(dev))); + skb_dst_set(skb, dst); + skb->dev = skb_dst(skb)->dev; + err = dst_output(t->net, skb->sk, skb); if (net_xmit_eval(err) == 0) { struct pcpu_sw_netstats *tstats = this_cpu_ptr(dev->tstats); From 45c180bc29babbedd6b8c01b975780ef44d9d09c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 18 Jun 2018 21:35:07 -0700 Subject: [PATCH 02/41] xfrm_user: prevent leaking 2 bytes of kernel memory struct xfrm_userpolicy_type has two holes, so we should not use C99 style initializer. KMSAN report: BUG: KMSAN: kernel-infoleak in copyout lib/iov_iter.c:140 [inline] BUG: KMSAN: kernel-infoleak in _copy_to_iter+0x1b14/0x2800 lib/iov_iter.c:571 CPU: 1 PID: 4520 Comm: syz-executor841 Not tainted 4.17.0+ #5 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x185/0x1d0 lib/dump_stack.c:113 kmsan_report+0x188/0x2a0 mm/kmsan/kmsan.c:1117 kmsan_internal_check_memory+0x138/0x1f0 mm/kmsan/kmsan.c:1211 kmsan_copy_to_user+0x7a/0x160 mm/kmsan/kmsan.c:1253 copyout lib/iov_iter.c:140 [inline] _copy_to_iter+0x1b14/0x2800 lib/iov_iter.c:571 copy_to_iter include/linux/uio.h:106 [inline] skb_copy_datagram_iter+0x422/0xfa0 net/core/datagram.c:431 skb_copy_datagram_msg include/linux/skbuff.h:3268 [inline] netlink_recvmsg+0x6f1/0x1900 net/netlink/af_netlink.c:1959 sock_recvmsg_nosec net/socket.c:802 [inline] sock_recvmsg+0x1d6/0x230 net/socket.c:809 ___sys_recvmsg+0x3fe/0x810 net/socket.c:2279 __sys_recvmmsg+0x58e/0xe30 net/socket.c:2391 do_sys_recvmmsg+0x2a6/0x3e0 net/socket.c:2472 __do_sys_recvmmsg net/socket.c:2485 [inline] __se_sys_recvmmsg net/socket.c:2481 [inline] __x64_sys_recvmmsg+0x15d/0x1c0 net/socket.c:2481 do_syscall_64+0x15b/0x230 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x446ce9 RSP: 002b:00007fc307918db8 EFLAGS: 00000293 ORIG_RAX: 000000000000012b RAX: ffffffffffffffda RBX: 00000000006dbc24 RCX: 0000000000446ce9 RDX: 000000000000000a RSI: 0000000020005040 RDI: 0000000000000003 RBP: 00000000006dbc20 R08: 0000000020004e40 R09: 0000000000000000 R10: 0000000040000000 R11: 0000000000000293 R12: 0000000000000000 R13: 00007ffc8d2df32f R14: 00007fc3079199c0 R15: 0000000000000001 Uninit was stored to memory at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:279 [inline] kmsan_save_stack mm/kmsan/kmsan.c:294 [inline] kmsan_internal_chain_origin+0x12b/0x210 mm/kmsan/kmsan.c:685 kmsan_memcpy_origins+0x11d/0x170 mm/kmsan/kmsan.c:527 __msan_memcpy+0x109/0x160 mm/kmsan/kmsan_instr.c:413 __nla_put lib/nlattr.c:569 [inline] nla_put+0x276/0x340 lib/nlattr.c:627 copy_to_user_policy_type net/xfrm/xfrm_user.c:1678 [inline] dump_one_policy+0xbe1/0x1090 net/xfrm/xfrm_user.c:1708 xfrm_policy_walk+0x45a/0xd00 net/xfrm/xfrm_policy.c:1013 xfrm_dump_policy+0x1c0/0x2a0 net/xfrm/xfrm_user.c:1749 netlink_dump+0x9b5/0x1550 net/netlink/af_netlink.c:2226 __netlink_dump_start+0x1131/0x1270 net/netlink/af_netlink.c:2323 netlink_dump_start include/linux/netlink.h:214 [inline] xfrm_user_rcv_msg+0x8a3/0x9b0 net/xfrm/xfrm_user.c:2577 netlink_rcv_skb+0x37e/0x600 net/netlink/af_netlink.c:2448 xfrm_netlink_rcv+0xb2/0xf0 net/xfrm/xfrm_user.c:2598 netlink_unicast_kernel net/netlink/af_netlink.c:1310 [inline] netlink_unicast+0x1680/0x1750 net/netlink/af_netlink.c:1336 netlink_sendmsg+0x104f/0x1350 net/netlink/af_netlink.c:1901 sock_sendmsg_nosec net/socket.c:629 [inline] sock_sendmsg net/socket.c:639 [inline] ___sys_sendmsg+0xec8/0x1320 net/socket.c:2117 __sys_sendmsg net/socket.c:2155 [inline] __do_sys_sendmsg net/socket.c:2164 [inline] __se_sys_sendmsg net/socket.c:2162 [inline] __x64_sys_sendmsg+0x331/0x460 net/socket.c:2162 do_syscall_64+0x15b/0x230 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Local variable description: ----upt.i@dump_one_policy Variable was created at: dump_one_policy+0x78/0x1090 net/xfrm/xfrm_user.c:1689 xfrm_policy_walk+0x45a/0xd00 net/xfrm/xfrm_policy.c:1013 Byte 130 of 137 is uninitialized Memory access starts at ffff88019550407f Fixes: c0144beaeca42 ("[XFRM] netlink: Use nla_put()/NLA_PUT() variantes") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Steffen Klassert Cc: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 080035f056d9..1e50b70ad668 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1671,9 +1671,11 @@ static inline unsigned int userpolicy_type_attrsize(void) #ifdef CONFIG_XFRM_SUB_POLICY static int copy_to_user_policy_type(u8 type, struct sk_buff *skb) { - struct xfrm_userpolicy_type upt = { - .type = type, - }; + struct xfrm_userpolicy_type upt; + + /* Sadly there are two holes in struct xfrm_userpolicy_type */ + memset(&upt, 0, sizeof(upt)); + upt.type = type; return nla_put(skb, XFRMA_POLICY_TYPE, sizeof(upt), &upt); } From 8cc88773855f988d6a3bbf102bbd9dd9c828eb81 Mon Sep 17 00:00:00 2001 From: Tommi Rantala Date: Thu, 21 Jun 2018 09:30:47 +0300 Subject: [PATCH 03/41] xfrm: fix missing dst_release() after policy blocking lbcast and multicast Fix missing dst_release() when local broadcast or multicast traffic is xfrm policy blocked. For IPv4 this results to dst leak: ip_route_output_flow() allocates dst_entry via __ip_route_output_key() and passes it to xfrm_lookup_route(). xfrm_lookup returns ERR_PTR(-EPERM) that is propagated. The dst that was allocated is never released. IPv4 local broadcast testcase: ping -b 192.168.1.255 & sleep 1 ip xfrm policy add src 0.0.0.0/0 dst 192.168.1.255/32 dir out action block IPv4 multicast testcase: ping 224.0.0.1 & sleep 1 ip xfrm policy add src 0.0.0.0/0 dst 224.0.0.1/32 dir out action block For IPv6 the missing dst_release() causes trouble e.g. when used in netns: ip netns add TEST ip netns exec TEST ip link set lo up ip link add dummy0 type dummy ip link set dev dummy0 netns TEST ip netns exec TEST ip addr add fd00::1111 dev dummy0 ip netns exec TEST ip link set dummy0 up ip netns exec TEST ping -6 -c 5 ff02::1%dummy0 & sleep 1 ip netns exec TEST ip xfrm policy add src ::/0 dst ff02::1 dir out action block wait ip netns del TEST After netns deletion we see: [ 258.239097] unregister_netdevice: waiting for lo to become free. Usage count = 2 [ 268.279061] unregister_netdevice: waiting for lo to become free. Usage count = 2 [ 278.367018] unregister_netdevice: waiting for lo to become free. Usage count = 2 [ 288.375259] unregister_netdevice: waiting for lo to become free. Usage count = 2 Fixes: ac37e2515c1a ("xfrm: release dst_orig in case of error in xfrm_lookup()") Signed-off-by: Tommi Rantala Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 5f48251c1319..7c5e8978aeaa 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2286,6 +2286,9 @@ struct dst_entry *xfrm_lookup_route(struct net *net, struct dst_entry *dst_orig, if (IS_ERR(dst) && PTR_ERR(dst) == -EREMOTE) return make_blackhole(net, dst_orig->ops->family, dst_orig); + if (IS_ERR(dst)) + dst_release(dst_orig); + return dst; } EXPORT_SYMBOL(xfrm_lookup_route); From 86126b77dcd551ce223e7293bb55854e3df05646 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 25 Jun 2018 14:00:07 +0200 Subject: [PATCH 04/41] xfrm: free skb if nlsk pointer is NULL nlmsg_multicast() always frees the skb, so in case we cannot call it we must do that ourselves. Fixes: 21ee543edc0dea ("xfrm: fix race between netns cleanup and state expire notification") Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 1e50b70ad668..33878e6e0d0a 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1025,10 +1025,12 @@ static inline int xfrm_nlmsg_multicast(struct net *net, struct sk_buff *skb, { struct sock *nlsk = rcu_dereference(net->xfrm.nlsk); - if (nlsk) - return nlmsg_multicast(nlsk, skb, pid, group, GFP_ATOMIC); - else - return -1; + if (!nlsk) { + kfree_skb(skb); + return -EPIPE; + } + + return nlmsg_multicast(nlsk, skb, pid, group, GFP_ATOMIC); } static inline unsigned int xfrm_spdinfo_msgsize(void) From 7284fdf39a912322ce97de2d30def3c6068a418c Mon Sep 17 00:00:00 2001 From: Zhen Lei Date: Wed, 27 Jun 2018 11:49:28 +0800 Subject: [PATCH 05/41] esp6: fix memleak on error path in esp6_input This ought to be an omission in e6194923237 ("esp: Fix memleaks on error paths."). The memleak on error path in esp6_input is similar to esp_input of esp4. Fixes: e6194923237 ("esp: Fix memleaks on error paths.") Fixes: 3f29770723f ("ipsec: check return value of skb_to_sgvec always") Signed-off-by: Zhen Lei Signed-off-by: Steffen Klassert --- net/ipv6/esp6.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 97513f35bcc5..88a7579c23bd 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -669,8 +669,10 @@ skip_cow: sg_init_table(sg, nfrags); ret = skb_to_sgvec(skb, sg, 0, skb->len); - if (unlikely(ret < 0)) + if (unlikely(ret < 0)) { + kfree(tmp); goto out; + } skb->ip_summed = CHECKSUM_NONE; From 6283fa38dc8744dc7c2bd2a03bb0478fe42f79fa Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Fri, 20 Jul 2018 17:38:37 -0700 Subject: [PATCH 06/41] bpf: btf: Ensure the member->offset is in the right order This patch ensures the member->offset of a struct is in the correct order (i.e the later member's offset cannot go backward). The current "pahole -J" BTF encoder does not generate something like this. However, checking this can ensure future encoder will not violate this. Fixes: 69b693f0aefa ("bpf: btf: Introduce BPF Type Format (BTF)") Signed-off-by: Martin KaFai Lau Acked-by: Yonghong Song Signed-off-by: Daniel Borkmann --- kernel/bpf/btf.c | 14 ++++++++++++- tools/testing/selftests/bpf/test_btf.c | 28 ++++++++++++++++++++++++++ 2 files changed, 41 insertions(+), 1 deletion(-) diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 9704934252b3..2590700237c1 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -1519,9 +1519,9 @@ static s32 btf_struct_check_meta(struct btf_verifier_env *env, { bool is_union = BTF_INFO_KIND(t->info) == BTF_KIND_UNION; const struct btf_member *member; + u32 meta_needed, last_offset; struct btf *btf = env->btf; u32 struct_size = t->size; - u32 meta_needed; u16 i; meta_needed = btf_type_vlen(t) * sizeof(*member); @@ -1534,6 +1534,7 @@ static s32 btf_struct_check_meta(struct btf_verifier_env *env, btf_verifier_log_type(env, t, NULL); + last_offset = 0; for_each_member(i, t, member) { if (!btf_name_offset_valid(btf, member->name_off)) { btf_verifier_log_member(env, t, member, @@ -1555,6 +1556,16 @@ static s32 btf_struct_check_meta(struct btf_verifier_env *env, return -EINVAL; } + /* + * ">" instead of ">=" because the last member could be + * "char a[0];" + */ + if (last_offset > member->offset) { + btf_verifier_log_member(env, t, member, + "Invalid member bits_offset"); + return -EINVAL; + } + if (BITS_ROUNDUP_BYTES(member->offset) > struct_size) { btf_verifier_log_member(env, t, member, "Memmber bits_offset exceeds its struct size"); @@ -1562,6 +1573,7 @@ static s32 btf_struct_check_meta(struct btf_verifier_env *env, } btf_verifier_log_member(env, t, member, NULL); + last_offset = member->offset; } return meta_needed; diff --git a/tools/testing/selftests/bpf/test_btf.c b/tools/testing/selftests/bpf/test_btf.c index 3619f3023088..402c0f7cc418 100644 --- a/tools/testing/selftests/bpf/test_btf.c +++ b/tools/testing/selftests/bpf/test_btf.c @@ -247,6 +247,34 @@ static struct btf_raw_test raw_tests[] = { .max_entries = 4, }, +{ + .descr = "struct test #3 Invalid member offset", + .raw_types = { + /* int */ /* [1] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), + /* int64 */ /* [2] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 64, 8), + + /* struct A { */ /* [3] */ + BTF_TYPE_ENC(NAME_TBD, BTF_INFO_ENC(BTF_KIND_STRUCT, 0, 2), 16), + BTF_MEMBER_ENC(NAME_TBD, 1, 64), /* int m; */ + BTF_MEMBER_ENC(NAME_TBD, 2, 0), /* int64 n; */ + /* } */ + BTF_END_RAW, + }, + .str_sec = "\0A\0m\0n\0", + .str_sec_size = sizeof("\0A\0m\0n\0"), + .map_type = BPF_MAP_TYPE_ARRAY, + .map_name = "struct_test3_map", + .key_size = sizeof(int), + .value_size = 16, + .key_type_id = 1, + .value_type_id = 3, + .max_entries = 4, + .btf_load_err = true, + .err_str = "Invalid member bits_offset", +}, + /* Test member exceeds the size of struct. * * struct A { From 64bb568488671048d25d7b3ada058bb6c7cb1d5d Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 24 Jul 2018 08:40:20 -0700 Subject: [PATCH 07/41] bpf: btf: Sync uapi btf.h to tools This patch sync the uapi btf.h to tools/ Fixes: 36fc3c8c282c bpf: btf: Clean up BTF_INT_BITS() in uapi btf.h Signed-off-by: Martin KaFai Lau Acked-by: Yonghong Song Signed-off-by: Daniel Borkmann --- tools/include/uapi/linux/btf.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/include/uapi/linux/btf.h b/tools/include/uapi/linux/btf.h index 0b5ddbe135a4..972265f32871 100644 --- a/tools/include/uapi/linux/btf.h +++ b/tools/include/uapi/linux/btf.h @@ -76,7 +76,7 @@ struct btf_type { */ #define BTF_INT_ENCODING(VAL) (((VAL) & 0x0f000000) >> 24) #define BTF_INT_OFFSET(VAL) (((VAL & 0x00ff0000)) >> 16) -#define BTF_INT_BITS(VAL) ((VAL) & 0x0000ffff) +#define BTF_INT_BITS(VAL) ((VAL) & 0x000000ff) /* Attributes stored in the BTF_INT_ENCODING */ #define BTF_INT_SIGNED (1 << 0) From 5b891af7fca14526b2a87c6f38b004e2df655ef4 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 24 Jul 2018 08:40:21 -0700 Subject: [PATCH 08/41] bpf: Replace [u]int32_t and [u]int64_t in libbpf This patch replaces [u]int32_t and [u]int64_t usage with __[su]32 and __[su]64. The same change goes for [u]int16_t and [u]int8_t. Fixes: 8a138aed4a80 ("bpf: btf: Add BTF support to libbpf") Signed-off-by: Martin KaFai Lau Acked-by: Yonghong Song Signed-off-by: Daniel Borkmann --- tools/lib/bpf/btf.c | 34 ++++++++++++++++------------------ tools/lib/bpf/btf.h | 8 ++++---- tools/lib/bpf/libbpf.c | 12 ++++++------ tools/lib/bpf/libbpf.h | 4 ++-- 4 files changed, 28 insertions(+), 30 deletions(-) diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index 8c54a4b6f187..b80de80b4584 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -2,7 +2,6 @@ /* Copyright (c) 2018 Facebook */ #include -#include #include #include #include @@ -27,13 +26,13 @@ struct btf { struct btf_type **types; const char *strings; void *nohdr_data; - uint32_t nr_types; - uint32_t types_size; - uint32_t data_size; + __u32 nr_types; + __u32 types_size; + __u32 data_size; int fd; }; -static const char *btf_name_by_offset(const struct btf *btf, uint32_t offset) +static const char *btf_name_by_offset(const struct btf *btf, __u32 offset) { if (offset < btf->hdr->str_len) return &btf->strings[offset]; @@ -45,7 +44,7 @@ static int btf_add_type(struct btf *btf, struct btf_type *t) { if (btf->types_size - btf->nr_types < 2) { struct btf_type **new_types; - u32 expand_by, new_size; + __u32 expand_by, new_size; if (btf->types_size == BTF_MAX_NR_TYPES) return -E2BIG; @@ -72,7 +71,7 @@ static int btf_add_type(struct btf *btf, struct btf_type *t) static int btf_parse_hdr(struct btf *btf, btf_print_fn_t err_log) { const struct btf_header *hdr = btf->hdr; - u32 meta_left; + __u32 meta_left; if (btf->data_size < sizeof(struct btf_header)) { elog("BTF header not found\n"); @@ -151,7 +150,7 @@ static int btf_parse_type_sec(struct btf *btf, btf_print_fn_t err_log) while (next_type < end_type) { struct btf_type *t = next_type; - uint16_t vlen = BTF_INFO_VLEN(t->info); + __u16 vlen = BTF_INFO_VLEN(t->info); int err; next_type += sizeof(*t); @@ -191,7 +190,7 @@ static int btf_parse_type_sec(struct btf *btf, btf_print_fn_t err_log) } static const struct btf_type *btf_type_by_id(const struct btf *btf, - uint32_t type_id) + __u32 type_id) { if (type_id > btf->nr_types) return NULL; @@ -209,7 +208,7 @@ static bool btf_type_is_void_or_null(const struct btf_type *t) return !t || btf_type_is_void(t); } -static int64_t btf_type_size(const struct btf_type *t) +static __s64 btf_type_size(const struct btf_type *t) { switch (BTF_INFO_KIND(t->info)) { case BTF_KIND_INT: @@ -226,12 +225,12 @@ static int64_t btf_type_size(const struct btf_type *t) #define MAX_RESOLVE_DEPTH 32 -int64_t btf__resolve_size(const struct btf *btf, uint32_t type_id) +__s64 btf__resolve_size(const struct btf *btf, __u32 type_id) { const struct btf_array *array; const struct btf_type *t; - uint32_t nelems = 1; - int64_t size = -1; + __u32 nelems = 1; + __s64 size = -1; int i; t = btf_type_by_id(btf, type_id); @@ -271,9 +270,9 @@ int64_t btf__resolve_size(const struct btf *btf, uint32_t type_id) return nelems * size; } -int32_t btf__find_by_name(const struct btf *btf, const char *type_name) +__s32 btf__find_by_name(const struct btf *btf, const char *type_name) { - uint32_t i; + __u32 i; if (!strcmp(type_name, "void")) return 0; @@ -302,10 +301,9 @@ void btf__free(struct btf *btf) free(btf); } -struct btf *btf__new(uint8_t *data, uint32_t size, - btf_print_fn_t err_log) +struct btf *btf__new(__u8 *data, __u32 size, btf_print_fn_t err_log) { - uint32_t log_buf_size = 0; + __u32 log_buf_size = 0; char *log_buf = NULL; struct btf *btf; int err; diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h index 74bb344035bb..ed3a84370ccc 100644 --- a/tools/lib/bpf/btf.h +++ b/tools/lib/bpf/btf.h @@ -4,7 +4,7 @@ #ifndef __BPF_BTF_H #define __BPF_BTF_H -#include +#include #define BTF_ELF_SEC ".BTF" @@ -14,9 +14,9 @@ typedef int (*btf_print_fn_t)(const char *, ...) __attribute__((format(printf, 1, 2))); void btf__free(struct btf *btf); -struct btf *btf__new(uint8_t *data, uint32_t size, btf_print_fn_t err_log); -int32_t btf__find_by_name(const struct btf *btf, const char *type_name); -int64_t btf__resolve_size(const struct btf *btf, uint32_t type_id); +struct btf *btf__new(__u8 *data, __u32 size, btf_print_fn_t err_log); +__s32 btf__find_by_name(const struct btf *btf, const char *type_name); +__s64 btf__resolve_size(const struct btf *btf, __u32 type_id); int btf__fd(const struct btf *btf); #endif diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index a1e96b5de5ff..6deb4fe4fffe 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -216,8 +216,8 @@ struct bpf_map { size_t offset; int map_ifindex; struct bpf_map_def def; - uint32_t btf_key_type_id; - uint32_t btf_value_type_id; + __u32 btf_key_type_id; + __u32 btf_value_type_id; void *priv; bpf_map_clear_priv_t clear_priv; }; @@ -1016,8 +1016,8 @@ static int bpf_map_find_btf_info(struct bpf_map *map, const struct btf *btf) { struct bpf_map_def *def = &map->def; const size_t max_name = 256; - int64_t key_size, value_size; - int32_t key_id, value_id; + __s64 key_size, value_size; + __s32 key_id, value_id; char name[max_name]; /* Find key type by name from BTF */ @@ -2089,12 +2089,12 @@ const char *bpf_map__name(struct bpf_map *map) return map ? map->name : NULL; } -uint32_t bpf_map__btf_key_type_id(const struct bpf_map *map) +__u32 bpf_map__btf_key_type_id(const struct bpf_map *map) { return map ? map->btf_key_type_id : 0; } -uint32_t bpf_map__btf_value_type_id(const struct bpf_map *map) +__u32 bpf_map__btf_value_type_id(const struct bpf_map *map) { return map ? map->btf_value_type_id : 0; } diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index 09976531aa74..b33ae02f7d0e 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -244,8 +244,8 @@ bpf_map__next(struct bpf_map *map, struct bpf_object *obj); int bpf_map__fd(struct bpf_map *map); const struct bpf_map_def *bpf_map__def(struct bpf_map *map); const char *bpf_map__name(struct bpf_map *map); -uint32_t bpf_map__btf_key_type_id(const struct bpf_map *map); -uint32_t bpf_map__btf_value_type_id(const struct bpf_map *map); +__u32 bpf_map__btf_key_type_id(const struct bpf_map *map); +__u32 bpf_map__btf_value_type_id(const struct bpf_map *map); typedef void (*bpf_map_clear_priv_t)(struct bpf_map *, void *); int bpf_map__set_priv(struct bpf_map *map, void *priv, From 38d5d3b3d5dbc0e0bb51fa6f7559d0d5a27916f6 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 24 Jul 2018 08:40:22 -0700 Subject: [PATCH 09/41] bpf: Introduce BPF_ANNOTATE_KV_PAIR This patch introduces BPF_ANNOTATE_KV_PAIR to signal the bpf loader about the btf key_type and value_type of a bpf map. Please refer to the changes in test_btf_haskv.c for its usage. Both iproute2 and libbpf loader will then have the same convention to find out the map's btf_key_type_id and btf_value_type_id from a map's name. Fixes: 8a138aed4a80 ("bpf: btf: Add BTF support to libbpf") Suggested-by: Daniel Borkmann Signed-off-by: Martin KaFai Lau Acked-by: Yonghong Song Signed-off-by: Daniel Borkmann --- tools/lib/bpf/btf.c | 7 +- tools/lib/bpf/btf.h | 2 + tools/lib/bpf/libbpf.c | 77 +++++++++++--------- tools/testing/selftests/bpf/bpf_helpers.h | 9 +++ tools/testing/selftests/bpf/test_btf_haskv.c | 7 +- 5 files changed, 57 insertions(+), 45 deletions(-) diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index b80de80b4584..2d270c560df3 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -189,8 +189,7 @@ static int btf_parse_type_sec(struct btf *btf, btf_print_fn_t err_log) return 0; } -static const struct btf_type *btf_type_by_id(const struct btf *btf, - __u32 type_id) +const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 type_id) { if (type_id > btf->nr_types) return NULL; @@ -233,7 +232,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id) __s64 size = -1; int i; - t = btf_type_by_id(btf, type_id); + t = btf__type_by_id(btf, type_id); for (i = 0; i < MAX_RESOLVE_DEPTH && !btf_type_is_void_or_null(t); i++) { size = btf_type_size(t); @@ -258,7 +257,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id) return -EINVAL; } - t = btf_type_by_id(btf, type_id); + t = btf__type_by_id(btf, type_id); } if (size < 0) diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h index ed3a84370ccc..e2a09a155f84 100644 --- a/tools/lib/bpf/btf.h +++ b/tools/lib/bpf/btf.h @@ -9,6 +9,7 @@ #define BTF_ELF_SEC ".BTF" struct btf; +struct btf_type; typedef int (*btf_print_fn_t)(const char *, ...) __attribute__((format(printf, 1, 2))); @@ -16,6 +17,7 @@ typedef int (*btf_print_fn_t)(const char *, ...) void btf__free(struct btf *btf); struct btf *btf__new(__u8 *data, __u32 size, btf_print_fn_t err_log); __s32 btf__find_by_name(const struct btf *btf, const char *type_name); +const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 id); __s64 btf__resolve_size(const struct btf *btf, __u32 type_id); int btf__fd(const struct btf *btf); diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 6deb4fe4fffe..d881d370616c 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -1014,68 +1015,72 @@ bpf_program__collect_reloc(struct bpf_program *prog, GElf_Shdr *shdr, static int bpf_map_find_btf_info(struct bpf_map *map, const struct btf *btf) { + const struct btf_type *container_type; + const struct btf_member *key, *value; struct bpf_map_def *def = &map->def; const size_t max_name = 256; + char container_name[max_name]; __s64 key_size, value_size; - __s32 key_id, value_id; - char name[max_name]; + __s32 container_id; - /* Find key type by name from BTF */ - if (snprintf(name, max_name, "%s_key", map->name) == max_name) { - pr_warning("map:%s length of BTF key_type:%s_key is too long\n", + if (snprintf(container_name, max_name, "____btf_map_%s", map->name) == + max_name) { + pr_warning("map:%s length of '____btf_map_%s' is too long\n", map->name, map->name); return -EINVAL; } - key_id = btf__find_by_name(btf, name); - if (key_id < 0) { - pr_debug("map:%s key_type:%s cannot be found in BTF\n", - map->name, name); - return key_id; + container_id = btf__find_by_name(btf, container_name); + if (container_id < 0) { + pr_debug("map:%s container_name:%s cannot be found in BTF. Missing BPF_ANNOTATE_KV_PAIR?\n", + map->name, container_name); + return container_id; } - key_size = btf__resolve_size(btf, key_id); + container_type = btf__type_by_id(btf, container_id); + if (!container_type) { + pr_warning("map:%s cannot find BTF type for container_id:%u\n", + map->name, container_id); + return -EINVAL; + } + + if (BTF_INFO_KIND(container_type->info) != BTF_KIND_STRUCT || + BTF_INFO_VLEN(container_type->info) < 2) { + pr_warning("map:%s container_name:%s is an invalid container struct\n", + map->name, container_name); + return -EINVAL; + } + + key = (struct btf_member *)(container_type + 1); + value = key + 1; + + key_size = btf__resolve_size(btf, key->type); if (key_size < 0) { - pr_warning("map:%s key_type:%s cannot get the BTF type_size\n", - map->name, name); + pr_warning("map:%s invalid BTF key_type_size\n", + map->name); return key_size; } if (def->key_size != key_size) { - pr_warning("map:%s key_type:%s has BTF type_size:%u != key_size:%u\n", - map->name, name, (unsigned int)key_size, def->key_size); + pr_warning("map:%s btf_key_type_size:%u != map_def_key_size:%u\n", + map->name, (__u32)key_size, def->key_size); return -EINVAL; } - /* Find value type from BTF */ - if (snprintf(name, max_name, "%s_value", map->name) == max_name) { - pr_warning("map:%s length of BTF value_type:%s_value is too long\n", - map->name, map->name); - return -EINVAL; - } - - value_id = btf__find_by_name(btf, name); - if (value_id < 0) { - pr_debug("map:%s value_type:%s cannot be found in BTF\n", - map->name, name); - return value_id; - } - - value_size = btf__resolve_size(btf, value_id); + value_size = btf__resolve_size(btf, value->type); if (value_size < 0) { - pr_warning("map:%s value_type:%s cannot get the BTF type_size\n", - map->name, name); + pr_warning("map:%s invalid BTF value_type_size\n", map->name); return value_size; } if (def->value_size != value_size) { - pr_warning("map:%s value_type:%s has BTF type_size:%u != value_size:%u\n", - map->name, name, (unsigned int)value_size, def->value_size); + pr_warning("map:%s btf_value_type_size:%u != map_def_value_size:%u\n", + map->name, (__u32)value_size, def->value_size); return -EINVAL; } - map->btf_key_type_id = key_id; - map->btf_value_type_id = value_id; + map->btf_key_type_id = key->type; + map->btf_value_type_id = value->type; return 0; } diff --git a/tools/testing/selftests/bpf/bpf_helpers.h b/tools/testing/selftests/bpf/bpf_helpers.h index f2f28b6c8915..810de20e8e26 100644 --- a/tools/testing/selftests/bpf/bpf_helpers.h +++ b/tools/testing/selftests/bpf/bpf_helpers.h @@ -158,6 +158,15 @@ struct bpf_map_def { unsigned int numa_node; }; +#define BPF_ANNOTATE_KV_PAIR(name, type_key, type_val) \ + struct ____btf_map_##name { \ + type_key key; \ + type_val value; \ + }; \ + struct ____btf_map_##name \ + __attribute__ ((section(".maps." #name), used)) \ + ____btf_map_##name = { } + static int (*bpf_skb_load_bytes)(void *ctx, int off, void *to, int len) = (void *) BPF_FUNC_skb_load_bytes; static int (*bpf_skb_store_bytes)(void *ctx, int off, void *from, int len, int flags) = diff --git a/tools/testing/selftests/bpf/test_btf_haskv.c b/tools/testing/selftests/bpf/test_btf_haskv.c index 8c7ca096ecf2..b21b876f475d 100644 --- a/tools/testing/selftests/bpf/test_btf_haskv.c +++ b/tools/testing/selftests/bpf/test_btf_haskv.c @@ -10,11 +10,6 @@ struct ipv_counts { unsigned int v6; }; -typedef int btf_map_key; -typedef struct ipv_counts btf_map_value; -btf_map_key dumm_key; -btf_map_value dummy_value; - struct bpf_map_def SEC("maps") btf_map = { .type = BPF_MAP_TYPE_ARRAY, .key_size = sizeof(int), @@ -22,6 +17,8 @@ struct bpf_map_def SEC("maps") btf_map = { .max_entries = 4, }; +BPF_ANNOTATE_KV_PAIR(btf_map, int, struct ipv_counts); + struct dummy_tracepoint_args { unsigned long long pad; struct sock *sock; From ecbc42ca5d665e9238a4cdb595024d2e6cf87f2d Mon Sep 17 00:00:00 2001 From: Toshiaki Makita Date: Mon, 23 Jul 2018 23:36:04 +0900 Subject: [PATCH 10/41] virtio_net: Fix incosistent received bytes counter When received packets are dropped in virtio_net driver, received packets counter is incremented but bytes counter is not. As a result, for instance if we drop all packets by XDP, only received is counted and bytes stays 0, which looks inconsistent. IMHO received packets/bytes should be counted if packets are produced by the hypervisor, like what common NICs on physical machines are doing. So fix the bytes counter. Signed-off-by: Toshiaki Makita Signed-off-by: David S. Miller --- drivers/net/virtio_net.c | 41 ++++++++++++++++++++++------------------ 1 file changed, 23 insertions(+), 18 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 53085c63277b..2b6ec927809e 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -586,7 +586,8 @@ static struct sk_buff *receive_small(struct net_device *dev, struct receive_queue *rq, void *buf, void *ctx, unsigned int len, - unsigned int *xdp_xmit) + unsigned int *xdp_xmit, + unsigned int *rbytes) { struct sk_buff *skb; struct bpf_prog *xdp_prog; @@ -601,6 +602,7 @@ static struct sk_buff *receive_small(struct net_device *dev, int err; len -= vi->hdr_len; + *rbytes += len; rcu_read_lock(); xdp_prog = rcu_dereference(rq->xdp_prog); @@ -705,11 +707,13 @@ static struct sk_buff *receive_big(struct net_device *dev, struct virtnet_info *vi, struct receive_queue *rq, void *buf, - unsigned int len) + unsigned int len, + unsigned int *rbytes) { struct page *page = buf; struct sk_buff *skb = page_to_skb(vi, rq, page, 0, len, PAGE_SIZE); + *rbytes += len - vi->hdr_len; if (unlikely(!skb)) goto err; @@ -727,7 +731,8 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, void *buf, void *ctx, unsigned int len, - unsigned int *xdp_xmit) + unsigned int *xdp_xmit, + unsigned int *rbytes) { struct virtio_net_hdr_mrg_rxbuf *hdr = buf; u16 num_buf = virtio16_to_cpu(vi->vdev, hdr->num_buffers); @@ -740,6 +745,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, int err; head_skb = NULL; + *rbytes += len - vi->hdr_len; rcu_read_lock(); xdp_prog = rcu_dereference(rq->xdp_prog); @@ -877,6 +883,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev, goto err_buf; } + *rbytes += len; page = virt_to_head_page(buf); truesize = mergeable_ctx_to_truesize(ctx); @@ -932,6 +939,7 @@ err_skb: dev->stats.rx_length_errors++; break; } + *rbytes += len; page = virt_to_head_page(buf); put_page(page); } @@ -942,14 +950,13 @@ xdp_xmit: return NULL; } -static int receive_buf(struct virtnet_info *vi, struct receive_queue *rq, - void *buf, unsigned int len, void **ctx, - unsigned int *xdp_xmit) +static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq, + void *buf, unsigned int len, void **ctx, + unsigned int *xdp_xmit, unsigned int *rbytes) { struct net_device *dev = vi->dev; struct sk_buff *skb; struct virtio_net_hdr_mrg_rxbuf *hdr; - int ret; if (unlikely(len < vi->hdr_len + ETH_HLEN)) { pr_debug("%s: short packet %i\n", dev->name, len); @@ -961,23 +968,22 @@ static int receive_buf(struct virtnet_info *vi, struct receive_queue *rq, } else { put_page(virt_to_head_page(buf)); } - return 0; + return; } if (vi->mergeable_rx_bufs) - skb = receive_mergeable(dev, vi, rq, buf, ctx, len, xdp_xmit); + skb = receive_mergeable(dev, vi, rq, buf, ctx, len, xdp_xmit, + rbytes); else if (vi->big_packets) - skb = receive_big(dev, vi, rq, buf, len); + skb = receive_big(dev, vi, rq, buf, len, rbytes); else - skb = receive_small(dev, vi, rq, buf, ctx, len, xdp_xmit); + skb = receive_small(dev, vi, rq, buf, ctx, len, xdp_xmit, rbytes); if (unlikely(!skb)) - return 0; + return; hdr = skb_vnet_hdr(skb); - ret = skb->len; - if (hdr->hdr.flags & VIRTIO_NET_HDR_F_DATA_VALID) skb->ip_summed = CHECKSUM_UNNECESSARY; @@ -994,12 +1000,11 @@ static int receive_buf(struct virtnet_info *vi, struct receive_queue *rq, ntohs(skb->protocol), skb->len, skb->pkt_type); napi_gro_receive(&rq->napi, skb); - return ret; + return; frame_err: dev->stats.rx_frame_errors++; dev_kfree_skb(skb); - return 0; } /* Unlike mergeable buffers, all buffers are allocated to the @@ -1249,13 +1254,13 @@ static int virtnet_receive(struct receive_queue *rq, int budget, while (received < budget && (buf = virtqueue_get_buf_ctx(rq->vq, &len, &ctx))) { - bytes += receive_buf(vi, rq, buf, len, ctx, xdp_xmit); + receive_buf(vi, rq, buf, len, ctx, xdp_xmit, &bytes); received++; } } else { while (received < budget && (buf = virtqueue_get_buf(rq->vq, &len)) != NULL) { - bytes += receive_buf(vi, rq, buf, len, NULL, xdp_xmit); + receive_buf(vi, rq, buf, len, NULL, xdp_xmit, &bytes); received++; } } From 7856e8616273098dc6c09a6e084afd98a283ff0d Mon Sep 17 00:00:00 2001 From: dann frazier Date: Mon, 23 Jul 2018 16:55:40 -0600 Subject: [PATCH 11/41] hinic: Link the logical network device to the pci device in sysfs Otherwise interfaces get exposed under /sys/devices/virtual, which doesn't give udev the context it needs for PCI-based predictable interface names. Signed-off-by: dann frazier Signed-off-by: David S. Miller --- drivers/net/ethernet/huawei/hinic/hinic_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index 5b122728dcb4..09e9da10b786 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -983,6 +983,7 @@ static int nic_dev_init(struct pci_dev *pdev) hinic_hwdev_cb_register(nic_dev->hwdev, HINIC_MGMT_MSG_CMD_LINK_STATUS, nic_dev, link_status_event_handler); + SET_NETDEV_DEV(netdev, &pdev->dev); err = register_netdev(netdev); if (err) { dev_err(&pdev->dev, "Failed to register netdev\n"); From 9aee40006190a3cda9a4d2dbae71e92617c8c362 Mon Sep 17 00:00:00 2001 From: Lawrence Brakmo Date: Mon, 23 Jul 2018 17:49:39 -0700 Subject: [PATCH 12/41] tcp: ack immediately when a cwr packet arrives We observed high 99 and 99.9% latencies when doing RPCs with DCTCP. The problem is triggered when the last packet of a request arrives CE marked. The reply will carry the ECE mark causing TCP to shrink its cwnd to 1 (because there are no packets in flight). When the 1st packet of the next request arrives, the ACK was sometimes delayed even though it is CWR marked, adding up to 40ms to the RPC latency. This patch insures that CWR marked data packets arriving will be acked immediately. Packetdrill script to reproduce the problem: 0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 0.000 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0 0.000 setsockopt(3, SOL_TCP, TCP_CONGESTION, "dctcp", 5) = 0 0.000 bind(3, ..., ...) = 0 0.000 listen(3, 1) = 0 0.100 < [ect0] SEW 0:0(0) win 32792 0.100 > SE. 0:0(0) ack 1 0.110 < [ect0] . 1:1(0) ack 1 win 257 0.200 accept(3, ..., ...) = 4 0.200 < [ect0] . 1:1001(1000) ack 1 win 257 0.200 > [ect01] . 1:1(0) ack 1001 0.200 write(4, ..., 1) = 1 0.200 > [ect01] P. 1:2(1) ack 1001 0.200 < [ect0] . 1001:2001(1000) ack 2 win 257 0.200 write(4, ..., 1) = 1 0.200 > [ect01] P. 2:3(1) ack 2001 0.200 < [ect0] . 2001:3001(1000) ack 3 win 257 0.200 < [ect0] . 3001:4001(1000) ack 3 win 257 0.200 > [ect01] . 3:3(0) ack 4001 0.210 < [ce] P. 4001:4501(500) ack 3 win 257 +0.001 read(4, ..., 4500) = 4500 +0 write(4, ..., 1) = 1 +0 > [ect01] PE. 3:4(1) ack 4501 +0.010 < [ect0] W. 4501:5501(1000) ack 4 win 257 // Previously the ACK sequence below would be 4501, causing a long RTO +0.040~+0.045 > [ect01] . 4:4(0) ack 5501 // delayed ack +0.311 < [ect0] . 5501:6501(1000) ack 4 win 257 // More data +0 > [ect01] . 4:4(0) ack 6501 // now acks everything +0.500 < F. 9501:9501(0) ack 4 win 257 Modified based on comments by Neal Cardwell Signed-off-by: Lawrence Brakmo Acked-by: Neal Cardwell Acked-by: Yuchung Cheng Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 3bcd30a2ba06..f9dcb29be12d 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -246,8 +246,15 @@ static void tcp_ecn_queue_cwr(struct tcp_sock *tp) static void tcp_ecn_accept_cwr(struct tcp_sock *tp, const struct sk_buff *skb) { - if (tcp_hdr(skb)->cwr) + if (tcp_hdr(skb)->cwr) { tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR; + + /* If the sender is telling us it has entered CWR, then its + * cwnd may be very low (even just 1 packet), so we should ACK + * immediately. + */ + tcp_enter_quickack_mode((struct sock *)tp, 2); + } } static void tcp_ecn_withdraw_cwr(struct tcp_sock *tp) From b87bac1012c483462e7776c7b7320b659dbb3295 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 25 Jul 2018 06:06:13 +0000 Subject: [PATCH 13/41] net: igmp: make function __ip_mc_inc_group() static Fixes the following sparse warnings: net/ipv4/igmp.c:1391:6: warning: symbol '__ip_mc_inc_group' was not declared. Should it be static? Fixes: 6e2059b53f98 ("ipv4/igmp: init group mode as INCLUDE when join source group") Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller --- net/ipv4/igmp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 28fef7d15959..75151be21413 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1387,7 +1387,8 @@ static void ip_mc_hash_remove(struct in_device *in_dev, /* * A socket has joined a multicast group on device dev. */ -void __ip_mc_inc_group(struct in_device *in_dev, __be32 addr, unsigned int mode) +static void __ip_mc_inc_group(struct in_device *in_dev, __be32 addr, + unsigned int mode) { struct ip_mc_list *im; #ifdef CONFIG_IP_MULTICAST From 9e4e5b5c86661e767f3074bedbbf2ca59f8fbd18 Mon Sep 17 00:00:00 2001 From: Wang YanQing Date: Thu, 26 Jul 2018 00:46:11 +0800 Subject: [PATCH 14/41] bpf, x32: Fix regression caused by commit 24dea04767e6 Commit 24dea04767e6 ("bpf, x32: remove ld_abs/ld_ind") removed the 4 /* Extra space for skb_copy_bits buffer */ from _STACK_SIZE, but it didn't fix the concerned code in emit_prologue and emit_epilogue, and this error will bring very strange kernel runtime errors. This patch fixes it. Fixes: 24dea04767e6 ("bpf, x32: remove ld_abs/ld_ind") Reported-by: Meelis Roos Bisected-by: Meelis Roos Signed-off-by: Wang YanQing Signed-off-by: Daniel Borkmann --- arch/x86/net/bpf_jit_comp32.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/net/bpf_jit_comp32.c b/arch/x86/net/bpf_jit_comp32.c index 55799873ebe5..8f6cc71e0848 100644 --- a/arch/x86/net/bpf_jit_comp32.c +++ b/arch/x86/net/bpf_jit_comp32.c @@ -1441,8 +1441,8 @@ static void emit_prologue(u8 **pprog, u32 stack_depth) /* sub esp,STACK_SIZE */ EMIT2_off32(0x81, 0xEC, STACK_SIZE); - /* sub ebp,SCRATCH_SIZE+4+12*/ - EMIT3(0x83, add_1reg(0xE8, IA32_EBP), SCRATCH_SIZE + 16); + /* sub ebp,SCRATCH_SIZE+12*/ + EMIT3(0x83, add_1reg(0xE8, IA32_EBP), SCRATCH_SIZE + 12); /* xor ebx,ebx */ EMIT2(0x31, add_2reg(0xC0, IA32_EBX, IA32_EBX)); @@ -1475,8 +1475,8 @@ static void emit_epilogue(u8 **pprog, u32 stack_depth) /* mov edx,dword ptr [ebp+off]*/ EMIT3(0x8B, add_2reg(0x40, IA32_EBP, IA32_EDX), STACK_VAR(r0[1])); - /* add ebp,SCRATCH_SIZE+4+12*/ - EMIT3(0x83, add_1reg(0xC0, IA32_EBP), SCRATCH_SIZE + 16); + /* add ebp,SCRATCH_SIZE+12*/ + EMIT3(0x83, add_1reg(0xC0, IA32_EBP), SCRATCH_SIZE + 12); /* mov ebx,dword ptr [ebp-12]*/ EMIT3(0x8B, add_2reg(0x40, IA32_EBP, IA32_EBX), -12); From d24458e43b103c7eb7b2fd57bcac392fd7750438 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B6rn=20T=C3=B6pel?= Date: Mon, 23 Jul 2018 11:43:03 +0200 Subject: [PATCH 15/41] xsk: fix poll/POLLIN premature returns MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Polling for the ingress queues relies on reading the producer/consumer pointers of the Rx queue. Prior this commit, a cached consumer pointer could be used, instead of the actual consumer pointer and therefore report POLLIN prematurely. This patch makes sure that the non-cached consumer pointer is used instead. Reported-by: Qi Zhang Tested-by: Qi Zhang Fixes: c497176cb2e4 ("xsk: add Rx receive functions and poll support") Signed-off-by: Björn Töpel Signed-off-by: Daniel Borkmann --- net/xdp/xsk_queue.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h index 52ecaf770642..8a64b150be54 100644 --- a/net/xdp/xsk_queue.h +++ b/net/xdp/xsk_queue.h @@ -250,7 +250,7 @@ static inline bool xskq_full_desc(struct xsk_queue *q) static inline bool xskq_empty_desc(struct xsk_queue *q) { - return xskq_nb_free(q, q->prod_tail, 1) == q->nentries; + return xskq_nb_free(q, q->prod_tail, q->nentries) == q->nentries; } void xskq_set_umem(struct xsk_queue *q, struct xdp_umem_props *umem_props); From c259b4fb33ee6e7667bf1d34bf0803b7c5fdbdce Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 25 Jul 2018 15:39:27 -0700 Subject: [PATCH 16/41] netdevsim: don't leak devlink resources Devlink resources registered with devlink_resource_register() have to be unregistered. Fixes: 37923ed6b8ce ("netdevsim: Add simple FIB resource controller via devlink") Signed-off-by: Jakub Kicinski Reviewed-by: Quentin Monnet Signed-off-by: David S. Miller --- drivers/net/netdevsim/devlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/netdevsim/devlink.c b/drivers/net/netdevsim/devlink.c index ba663e5af168..5135fc371f01 100644 --- a/drivers/net/netdevsim/devlink.c +++ b/drivers/net/netdevsim/devlink.c @@ -207,6 +207,7 @@ void nsim_devlink_teardown(struct netdevsim *ns) struct net *net = nsim_to_net(ns); bool *reg_devlink = net_generic(net, nsim_devlink_id); + devlink_resources_unregister(ns->devlink, NULL); devlink_unregister(ns->devlink); devlink_free(ns->devlink); ns->devlink = NULL; From 942a656f1f228f06a37adad0e6c347773cfe7bd6 Mon Sep 17 00:00:00 2001 From: Arjun Vynipadath Date: Wed, 25 Jul 2018 19:39:52 +0530 Subject: [PATCH 17/41] cxgb4: Added missing break in ndo_udp_tunnel_{add/del} Break statements were missing for Geneve case in ndo_udp_tunnel_{add/del}, thereby raw mac matchall entries were not getting added. Fixes: c746fc0e8b2d("cxgb4: add geneve offload support for T6") Signed-off-by: Arjun Vynipadath Signed-off-by: Ganesh Goudar Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index bc03c175a3cd..a8926e97935e 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -3072,6 +3072,7 @@ static void cxgb_del_udp_tunnel(struct net_device *netdev, adapter->geneve_port = 0; t4_write_reg(adapter, MPS_RX_GENEVE_TYPE_A, 0); + break; default: return; } @@ -3157,6 +3158,7 @@ static void cxgb_add_udp_tunnel(struct net_device *netdev, t4_write_reg(adapter, MPS_RX_GENEVE_TYPE_A, GENEVE_V(be16_to_cpu(ti->port)) | GENEVE_EN_F); + break; default: return; } From 7f3fc7ddf719cd6faaf787722c511f6918ac6aab Mon Sep 17 00:00:00 2001 From: tangpengpeng Date: Thu, 26 Jul 2018 14:45:16 +0800 Subject: [PATCH 18/41] net: fix amd-xgbe flow-control issue If we enable or disable xgbe flow-control by ethtool , it does't work.Because the parameter is not properly assigned,so we need to adjust the assignment order of the parameters. Fixes: c1ce2f77366b ("amd-xgbe: Fix flow control setting logic") Signed-off-by: tangpengpeng Acked-by: Tom Lendacky Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/xgbe/xgbe-mdio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c index 4b5d625de8f0..8a3a60bb2688 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c @@ -1111,14 +1111,14 @@ static void xgbe_phy_adjust_link(struct xgbe_prv_data *pdata) if (pdata->tx_pause != pdata->phy.tx_pause) { new_state = 1; - pdata->hw_if.config_tx_flow_control(pdata); pdata->tx_pause = pdata->phy.tx_pause; + pdata->hw_if.config_tx_flow_control(pdata); } if (pdata->rx_pause != pdata->phy.rx_pause) { new_state = 1; - pdata->hw_if.config_rx_flow_control(pdata); pdata->rx_pause = pdata->phy.rx_pause; + pdata->hw_if.config_rx_flow_control(pdata); } /* Speed support */ From 7effaf06c3cdef6855e127886c7405b9ab62f90d Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Tue, 24 Jul 2018 14:12:20 +0300 Subject: [PATCH 19/41] net: rollback orig value on failure of dev_qdisc_change_tx_queue_len Fix dev_change_tx_queue_len so it rolls back original value upon a failure in dev_qdisc_change_tx_queue_len. This is already done for notifirers' failures, share the code. In case of failure in dev_qdisc_change_tx_queue_len, some tx queues would still be of the new length, while they should be reverted. Currently, the revert is not done, and is marked with a TODO label in dev_qdisc_change_tx_queue_len, and should find some nice solution to do it. Yet it is still better to not apply the newly requested value. Fixes: 48bfd55e7e41 ("net_sched: plug in qdisc ops change_tx_queue_len") Signed-off-by: Tariq Toukan Reviewed-by: Eran Ben Elisha Reported-by: Ran Rozenstein Cc: Cong Wang Signed-off-by: David S. Miller --- net/core/dev.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index a5aa1c7444e6..559a91271f82 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -7149,16 +7149,19 @@ int dev_change_tx_queue_len(struct net_device *dev, unsigned long new_len) dev->tx_queue_len = new_len; res = call_netdevice_notifiers(NETDEV_CHANGE_TX_QUEUE_LEN, dev); res = notifier_to_errno(res); - if (res) { - netdev_err(dev, - "refused to change device tx_queue_len\n"); - dev->tx_queue_len = orig_len; - return res; - } - return dev_qdisc_change_tx_queue_len(dev); + if (res) + goto err_rollback; + res = dev_qdisc_change_tx_queue_len(dev); + if (res) + goto err_rollback; } return 0; + +err_rollback: + netdev_err(dev, "refused to change device tx_queue_len\n"); + dev->tx_queue_len = orig_len; + return res; } /** From 9e630bcb7701f94dbd729fe57d37c089c763ad9f Mon Sep 17 00:00:00 2001 From: Avinash Repaka Date: Tue, 24 Jul 2018 20:31:58 -0700 Subject: [PATCH 20/41] RDS: RDMA: Fix the NULL-ptr deref in rds_ib_get_mr Registration of a memory region(MR) through FRMR/fastreg(unlike FMR) needs a connection/qp. With a proxy qp, this dependency on connection will be removed, but that needs more infrastructure patches, which is a work in progress. As an intermediate fix, the get_mr returns EOPNOTSUPP when connection details are not populated. The MR registration through sendmsg() will continue to work even with fast registration, since connection in this case is formed upfront. This patch fixes the following crash: kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] SMP KASAN Modules linked in: CPU: 1 PID: 4244 Comm: syzkaller468044 Not tainted 4.16.0-rc6+ #361 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:rds_ib_get_mr+0x5c/0x230 net/rds/ib_rdma.c:544 RSP: 0018:ffff8801b059f890 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: ffff8801b07e1300 RCX: ffffffff8562d96e RDX: 000000000000000d RSI: 0000000000000001 RDI: 0000000000000068 RBP: ffff8801b059f8b8 R08: ffffed0036274244 R09: ffff8801b13a1200 R10: 0000000000000004 R11: ffffed0036274243 R12: ffff8801b13a1200 R13: 0000000000000001 R14: ffff8801ca09fa9c R15: 0000000000000000 FS: 00007f4d050af700(0000) GS:ffff8801db300000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f4d050aee78 CR3: 00000001b0d9b006 CR4: 00000000001606e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: __rds_rdma_map+0x710/0x1050 net/rds/rdma.c:271 rds_get_mr_for_dest+0x1d4/0x2c0 net/rds/rdma.c:357 rds_setsockopt+0x6cc/0x980 net/rds/af_rds.c:347 SYSC_setsockopt net/socket.c:1849 [inline] SyS_setsockopt+0x189/0x360 net/socket.c:1828 do_syscall_64+0x281/0x940 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x42/0xb7 RIP: 0033:0x4456d9 RSP: 002b:00007f4d050aedb8 EFLAGS: 00000246 ORIG_RAX: 0000000000000036 RAX: ffffffffffffffda RBX: 00000000006dac3c RCX: 00000000004456d9 RDX: 0000000000000007 RSI: 0000000000000114 RDI: 0000000000000004 RBP: 00000000006dac38 R08: 00000000000000a0 R09: 0000000000000000 R10: 0000000020000380 R11: 0000000000000246 R12: 0000000000000000 R13: 00007fffbfb36d6f R14: 00007f4d050af9c0 R15: 0000000000000005 Code: fa 48 c1 ea 03 80 3c 02 00 0f 85 cc 01 00 00 4c 8b bb 80 04 00 00 48 b8 00 00 00 00 00 fc ff df 49 8d 7f 68 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 9c 01 00 00 4d 8b 7f 68 48 b8 00 00 00 00 00 RIP: rds_ib_get_mr+0x5c/0x230 net/rds/ib_rdma.c:544 RSP: ffff8801b059f890 ---[ end trace 7e1cea13b85473b0 ]--- Reported-by: syzbot+b51c77ef956678a65834@syzkaller.appspotmail.com Signed-off-by: Santosh Shilimkar Signed-off-by: Avinash Repaka Signed-off-by: David S. Miller --- net/rds/ib_frmr.c | 5 +++++ net/rds/ib_mr.h | 3 ++- net/rds/ib_rdma.c | 21 +++++++++++++-------- net/rds/rdma.c | 13 ++++++++----- net/rds/rds.h | 5 ++++- net/rds/send.c | 12 +++++++----- 6 files changed, 39 insertions(+), 20 deletions(-) diff --git a/net/rds/ib_frmr.c b/net/rds/ib_frmr.c index 48332a6ed738..d152e48ea371 100644 --- a/net/rds/ib_frmr.c +++ b/net/rds/ib_frmr.c @@ -344,6 +344,11 @@ struct rds_ib_mr *rds_ib_reg_frmr(struct rds_ib_device *rds_ibdev, struct rds_ib_frmr *frmr; int ret; + if (!ic) { + /* TODO: Add FRWR support for RDS_GET_MR using proxy qp*/ + return ERR_PTR(-EOPNOTSUPP); + } + do { if (ibmr) rds_ib_free_frmr(ibmr, true); diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h index 0ea4ab017a8c..655f01d427fe 100644 --- a/net/rds/ib_mr.h +++ b/net/rds/ib_mr.h @@ -115,7 +115,8 @@ void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_connection *iinfo); void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *); void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, - struct rds_sock *rs, u32 *key_ret); + struct rds_sock *rs, u32 *key_ret, + struct rds_connection *conn); void rds_ib_sync_mr(void *trans_private, int dir); void rds_ib_free_mr(void *trans_private, int invalidate); void rds_ib_flush_mrs(void); diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c index e678699268a2..2e49a40a5e11 100644 --- a/net/rds/ib_rdma.c +++ b/net/rds/ib_rdma.c @@ -537,11 +537,12 @@ void rds_ib_flush_mrs(void) } void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, - struct rds_sock *rs, u32 *key_ret) + struct rds_sock *rs, u32 *key_ret, + struct rds_connection *conn) { struct rds_ib_device *rds_ibdev; struct rds_ib_mr *ibmr = NULL; - struct rds_ib_connection *ic = rs->rs_conn->c_transport_data; + struct rds_ib_connection *ic = NULL; int ret; rds_ibdev = rds_ib_get_device(rs->rs_bound_addr); @@ -550,6 +551,9 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, goto out; } + if (conn) + ic = conn->c_transport_data; + if (!rds_ibdev->mr_8k_pool || !rds_ibdev->mr_1m_pool) { ret = -ENODEV; goto out; @@ -559,17 +563,18 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents, ibmr = rds_ib_reg_frmr(rds_ibdev, ic, sg, nents, key_ret); else ibmr = rds_ib_reg_fmr(rds_ibdev, sg, nents, key_ret); - if (ibmr) - rds_ibdev = NULL; + if (IS_ERR(ibmr)) { + ret = PTR_ERR(ibmr); + pr_warn("RDS/IB: rds_ib_get_mr failed (errno=%d)\n", ret); + } else { + return ibmr; + } out: - if (!ibmr) - pr_warn("RDS/IB: rds_ib_get_mr failed (errno=%d)\n", ret); - if (rds_ibdev) rds_ib_dev_put(rds_ibdev); - return ibmr; + return ERR_PTR(ret); } void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *pool) diff --git a/net/rds/rdma.c b/net/rds/rdma.c index 634cfcb7bba6..80920e47f2c7 100644 --- a/net/rds/rdma.c +++ b/net/rds/rdma.c @@ -170,7 +170,8 @@ static int rds_pin_pages(unsigned long user_addr, unsigned int nr_pages, } static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args, - u64 *cookie_ret, struct rds_mr **mr_ret) + u64 *cookie_ret, struct rds_mr **mr_ret, + struct rds_conn_path *cp) { struct rds_mr *mr = NULL, *found; unsigned int nr_pages; @@ -269,7 +270,8 @@ static int __rds_rdma_map(struct rds_sock *rs, struct rds_get_mr_args *args, * Note that dma_map() implies that pending writes are * flushed to RAM, so no dma_sync is needed here. */ trans_private = rs->rs_transport->get_mr(sg, nents, rs, - &mr->r_key); + &mr->r_key, + cp ? cp->cp_conn : NULL); if (IS_ERR(trans_private)) { for (i = 0 ; i < nents; i++) @@ -330,7 +332,7 @@ int rds_get_mr(struct rds_sock *rs, char __user *optval, int optlen) sizeof(struct rds_get_mr_args))) return -EFAULT; - return __rds_rdma_map(rs, &args, NULL, NULL); + return __rds_rdma_map(rs, &args, NULL, NULL, NULL); } int rds_get_mr_for_dest(struct rds_sock *rs, char __user *optval, int optlen) @@ -354,7 +356,7 @@ int rds_get_mr_for_dest(struct rds_sock *rs, char __user *optval, int optlen) new_args.cookie_addr = args.cookie_addr; new_args.flags = args.flags; - return __rds_rdma_map(rs, &new_args, NULL, NULL); + return __rds_rdma_map(rs, &new_args, NULL, NULL, NULL); } /* @@ -782,7 +784,8 @@ int rds_cmsg_rdma_map(struct rds_sock *rs, struct rds_message *rm, rm->m_rdma_cookie != 0) return -EINVAL; - return __rds_rdma_map(rs, CMSG_DATA(cmsg), &rm->m_rdma_cookie, &rm->rdma.op_rdma_mr); + return __rds_rdma_map(rs, CMSG_DATA(cmsg), &rm->m_rdma_cookie, + &rm->rdma.op_rdma_mr, rm->m_conn_path); } /* diff --git a/net/rds/rds.h b/net/rds/rds.h index f2272fb8cd45..60b3b787fbdb 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -464,6 +464,8 @@ struct rds_message { struct scatterlist *op_sg; } data; }; + + struct rds_conn_path *m_conn_path; }; /* @@ -544,7 +546,8 @@ struct rds_transport { unsigned int avail); void (*exit)(void); void *(*get_mr)(struct scatterlist *sg, unsigned long nr_sg, - struct rds_sock *rs, u32 *key_ret); + struct rds_sock *rs, u32 *key_ret, + struct rds_connection *conn); void (*sync_mr)(void *trans_private, int direction); void (*free_mr)(void *trans_private, int invalidate); void (*flush_mrs)(void); diff --git a/net/rds/send.c b/net/rds/send.c index 94c7f74909be..59f17a2335f4 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -1169,6 +1169,13 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) rs->rs_conn = conn; } + if (conn->c_trans->t_mp_capable) + cpath = &conn->c_path[rds_send_mprds_hash(rs, conn)]; + else + cpath = &conn->c_path[0]; + + rm->m_conn_path = cpath; + /* Parse any control messages the user may have included. */ ret = rds_cmsg_send(rs, rm, msg, &allocated_mr); if (ret) { @@ -1192,11 +1199,6 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) goto out; } - if (conn->c_trans->t_mp_capable) - cpath = &conn->c_path[rds_send_mprds_hash(rs, conn)]; - else - cpath = &conn->c_path[0]; - if (rds_destroy_pending(conn)) { ret = -EAGAIN; goto out; From 36e0f12bbfd3016f495904b35e41c5711707509f Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Thu, 26 Jul 2018 23:17:03 +0900 Subject: [PATCH 21/41] xdp: add NULL pointer check in __xdp_return() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit rhashtable_lookup() can return NULL. so that NULL pointer check routine should be added. Fixes: 02b55e5657c3 ("xdp: add MEM_TYPE_ZERO_COPY") Signed-off-by: Taehee Yoo Acked-by: Martin KaFai Lau Acked-by: Björn Töpel Signed-off-by: Daniel Borkmann --- net/core/xdp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/core/xdp.c b/net/core/xdp.c index 9d1f22072d5d..6771f1855b96 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -345,7 +345,8 @@ static void __xdp_return(void *data, struct xdp_mem_info *mem, bool napi_direct, rcu_read_lock(); /* mem->id is valid, checked in xdp_rxq_info_reg_mem_model() */ xa = rhashtable_lookup(mem_id_ht, &mem->id, mem_id_rht_params); - xa->zc_alloc->free(xa->zc_alloc, handle); + if (!WARN_ON_ONCE(!xa)) + xa->zc_alloc->free(xa->zc_alloc, handle); rcu_read_unlock(); default: /* Not possible, checked in xdp_rxq_info_reg_mem_model() */ From 5f300e8004cb80182a24c0fa488218a4a43e6aac Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 26 Jul 2018 09:57:59 -0700 Subject: [PATCH 22/41] bpf: btf: Use exact btf value_size match in map_check_btf() The current map_check_btf() in BPF_MAP_TYPE_ARRAY rejects '> map->value_size' to ensure map_seq_show_elem() will not access things beyond an array element. Yonghong suggested that using '!=' is a more correct check. The 8 bytes round_up on value_size is stored in array->elem_size. Hence, using '!=' on map->value_size is a proper check. This patch also adds new tests to check the btf array key type and value type. Two of these new tests verify the btf's value_size (the change in this patch). It also fixes two existing tests that wrongly encoded a btf's type size (pprint_test) and the value_type_id (in one of the raw_tests[]). However, that do not affect these two BTF verification tests before or after this test changes. These two tests mainly failed at array creation time after this patch. Fixes: a26ca7c982cb ("bpf: btf: Add pretty print support to the basic arraymap") Suggested-by: Yonghong Song Acked-by: Yonghong Song Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann --- kernel/bpf/arraymap.c | 2 +- tools/testing/selftests/bpf/test_btf.c | 86 +++++++++++++++++++++++++- 2 files changed, 85 insertions(+), 3 deletions(-) diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 544e58f5f642..2aa55d030c77 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -378,7 +378,7 @@ static int array_map_check_btf(const struct bpf_map *map, const struct btf *btf, return -EINVAL; value_type = btf_type_id_size(btf, &btf_value_id, &value_size); - if (!value_type || value_size > map->value_size) + if (!value_type || value_size != map->value_size) return -EINVAL; return 0; diff --git a/tools/testing/selftests/bpf/test_btf.c b/tools/testing/selftests/bpf/test_btf.c index 402c0f7cc418..ffdd27737c9e 100644 --- a/tools/testing/selftests/bpf/test_btf.c +++ b/tools/testing/selftests/bpf/test_btf.c @@ -507,7 +507,7 @@ static struct btf_raw_test raw_tests[] = { .key_size = sizeof(int), .value_size = sizeof(void *) * 4, .key_type_id = 1, - .value_type_id = 4, + .value_type_id = 5, .max_entries = 4, }, @@ -1292,6 +1292,88 @@ static struct btf_raw_test raw_tests[] = { .err_str = "type != 0", }, +{ + .descr = "arraymap invalid btf key (a bit field)", + .raw_types = { + /* int */ /* [1] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), + /* 32 bit int with 32 bit offset */ /* [2] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 32, 32, 8), + BTF_END_RAW, + }, + .str_sec = "", + .str_sec_size = sizeof(""), + .map_type = BPF_MAP_TYPE_ARRAY, + .map_name = "array_map_check_btf", + .key_size = sizeof(int), + .value_size = sizeof(int), + .key_type_id = 2, + .value_type_id = 1, + .max_entries = 4, + .map_create_err = true, +}, + +{ + .descr = "arraymap invalid btf key (!= 32 bits)", + .raw_types = { + /* int */ /* [1] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), + /* 16 bit int with 0 bit offset */ /* [2] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 16, 2), + BTF_END_RAW, + }, + .str_sec = "", + .str_sec_size = sizeof(""), + .map_type = BPF_MAP_TYPE_ARRAY, + .map_name = "array_map_check_btf", + .key_size = sizeof(int), + .value_size = sizeof(int), + .key_type_id = 2, + .value_type_id = 1, + .max_entries = 4, + .map_create_err = true, +}, + +{ + .descr = "arraymap invalid btf value (too small)", + .raw_types = { + /* int */ /* [1] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), + BTF_END_RAW, + }, + .str_sec = "", + .str_sec_size = sizeof(""), + .map_type = BPF_MAP_TYPE_ARRAY, + .map_name = "array_map_check_btf", + .key_size = sizeof(int), + /* btf_value_size < map->value_size */ + .value_size = sizeof(__u64), + .key_type_id = 1, + .value_type_id = 1, + .max_entries = 4, + .map_create_err = true, +}, + +{ + .descr = "arraymap invalid btf value (too big)", + .raw_types = { + /* int */ /* [1] */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), + BTF_END_RAW, + }, + .str_sec = "", + .str_sec_size = sizeof(""), + .map_type = BPF_MAP_TYPE_ARRAY, + .map_name = "array_map_check_btf", + .key_size = sizeof(int), + /* btf_value_size > map->value_size */ + .value_size = sizeof(__u16), + .key_type_id = 1, + .value_type_id = 1, + .max_entries = 4, + .map_create_err = true, +}, + }; /* struct btf_raw_test raw_tests[] */ static const char *get_next_str(const char *start, const char *end) @@ -2051,7 +2133,7 @@ static struct btf_raw_test pprint_test = { BTF_ENUM_ENC(NAME_TBD, 2), BTF_ENUM_ENC(NAME_TBD, 3), /* struct pprint_mapv */ /* [16] */ - BTF_TYPE_ENC(NAME_TBD, BTF_INFO_ENC(BTF_KIND_STRUCT, 0, 8), 28), + BTF_TYPE_ENC(NAME_TBD, BTF_INFO_ENC(BTF_KIND_STRUCT, 0, 8), 32), BTF_MEMBER_ENC(NAME_TBD, 11, 0), /* uint32_t ui32 */ BTF_MEMBER_ENC(NAME_TBD, 10, 32), /* uint16_t ui16 */ BTF_MEMBER_ENC(NAME_TBD, 12, 64), /* int32_t si32 */ From 101f0cd4f2216d32f1b8a75a2154cf3997484ee2 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 26 Jul 2018 23:40:33 +0300 Subject: [PATCH 23/41] net: ena: Fix use of uninitialized DMA address bits field UBSAN triggers the following undefined behaviour warnings: [...] [ 13.236124] UBSAN: Undefined behaviour in drivers/net/ethernet/amazon/ena/ena_eth_com.c:468:22 [ 13.240043] shift exponent 64 is too large for 64-bit type 'long long unsigned int' [...] [ 13.744769] UBSAN: Undefined behaviour in drivers/net/ethernet/amazon/ena/ena_eth_com.c:373:4 [ 13.748694] shift exponent 64 is too large for 64-bit type 'long long unsigned int' [...] When splitting the address to high and low, GENMASK_ULL is used to generate a bitmask with dma_addr_bits field from io_sq (in ena_com_prepare_tx and ena_com_add_single_rx_desc). The problem is that dma_addr_bits is not initialized with a proper value (besides being cleared in ena_com_create_io_queue). Assign dma_addr_bits the correct value that is stored in ena_dev when initializing the SQ. Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)") Signed-off-by: Gal Pressman Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_com.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c index 1b9d3130af4d..17f12c18d225 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_com.c @@ -333,6 +333,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev, memset(&io_sq->desc_addr, 0x0, sizeof(io_sq->desc_addr)); + io_sq->dma_addr_bits = ena_dev->dma_addr_bits; io_sq->desc_entry_size = (io_sq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) ? sizeof(struct ena_eth_io_tx_desc) : From b611da43b68193dcb7e632adb44d506374a5d3ef Mon Sep 17 00:00:00 2001 From: Thomas Richter Date: Fri, 27 Jul 2018 10:21:26 +0200 Subject: [PATCH 24/41] perf build: Build error in libbpf missing initialization MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In linux-next tree compiling the perf tool with additional make flags EXTRA_CFLAGS="-Wp,-D_FORTIFY_SOURCE=2 -O2" causes a compiler error. It is the warning 'variable may be used uninitialized' which is treated as error: I compile it using a FEDORA 28 installation, my gcc compiler version: gcc (GCC) 8.0.1 20180324 (Red Hat 8.0.1-0.20). The file that causes the error is tools/lib/bpf/libbpf.c. [root@p23lp27] # make V=1 EXTRA_CFLAGS="-Wp,-D_FORTIFY_SOURCE=2 -O2" [...] Makefile.config:849: No openjdk development package found, please install JDK package, e.g. openjdk-8-jdk, java-1.8.0-openjdk-devel Warning: Kernel ABI header at 'tools/include/uapi/linux/if_link.h' differs from latest version at 'include/uapi/linux/if_link.h' CC libbpf.o libbpf.c: In function ‘bpf_perf_event_read_simple’: libbpf.c:2342:6: error: ‘ret’ may be used uninitialized in this function [-Werror=maybe-uninitialized] int ret; ^ cc1: all warnings being treated as errors mv: cannot stat './.libbpf.o.tmp': No such file or directory /home6/tmricht/linux-next/tools/build/Makefile.build:96: recipe for target 'libbpf.o' failed Suggested-by: Jakub Kicinski Signed-off-by: Thomas Richter Signed-off-by: Daniel Borkmann --- tools/lib/bpf/libbpf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index d881d370616c..1aafdbe827fe 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -2273,8 +2273,8 @@ bpf_perf_event_read_simple(void *mem, unsigned long size, volatile struct perf_event_mmap_page *header = mem; __u64 data_tail = header->data_tail; __u64 data_head = header->data_head; + int ret = LIBBPF_PERF_EVENT_ERROR; void *base, *begin, *end; - int ret; asm volatile("" ::: "memory"); /* in real code it should be smp_rmb() */ if (data_head == data_tail) From 3eee1f75f2b9c107d4a097e8b640553376a5b171 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Sat, 28 Jul 2018 00:17:56 +0200 Subject: [PATCH 25/41] bpf: fix bpf_skb_load_bytes_relative pkt length check The len > skb_headlen(skb) cannot be used as a maximum upper bound for the packet length since it does not have any relation to the full linear packet length when filtering is used from upper layers (e.g. in case of reuseport BPF programs) as by then skb->data, skb->len already got mangled through __skb_pull() and others. Fixes: 4e1ec56cdc59 ("bpf: add skb_load_bytes_relative helper") Signed-off-by: Daniel Borkmann Acked-by: Martin KaFai Lau --- net/core/filter.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/net/core/filter.c b/net/core/filter.c index 06da770f543f..9dfd145eedcc 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1712,24 +1712,26 @@ static const struct bpf_func_proto bpf_skb_load_bytes_proto = { BPF_CALL_5(bpf_skb_load_bytes_relative, const struct sk_buff *, skb, u32, offset, void *, to, u32, len, u32, start_header) { + u8 *end = skb_tail_pointer(skb); + u8 *net = skb_network_header(skb); + u8 *mac = skb_mac_header(skb); u8 *ptr; - if (unlikely(offset > 0xffff || len > skb_headlen(skb))) + if (unlikely(offset > 0xffff || len > (end - mac))) goto err_clear; switch (start_header) { case BPF_HDR_START_MAC: - ptr = skb_mac_header(skb) + offset; + ptr = mac + offset; break; case BPF_HDR_START_NET: - ptr = skb_network_header(skb) + offset; + ptr = net + offset; break; default: goto err_clear; } - if (likely(ptr >= skb_mac_header(skb) && - ptr + len <= skb_tail_pointer(skb))) { + if (likely(ptr >= mac && ptr + len <= end)) { memcpy(to, ptr, len); return 0; } From 71eb5255f55bdb484d35ff7c9a1803f453dfbf82 Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Sun, 29 Jul 2018 00:28:31 +0900 Subject: [PATCH 26/41] bpf: use GFP_ATOMIC instead of GFP_KERNEL in bpf_parse_prog() bpf_parse_prog() is protected by rcu_read_lock(). so that GFP_KERNEL is not allowed in the bpf_parse_prog(). [51015.579396] ============================= [51015.579418] WARNING: suspicious RCU usage [51015.579444] 4.18.0-rc6+ #208 Not tainted [51015.579464] ----------------------------- [51015.579488] ./include/linux/rcupdate.h:303 Illegal context switch in RCU read-side critical section! [51015.579510] other info that might help us debug this: [51015.579532] rcu_scheduler_active = 2, debug_locks = 1 [51015.579556] 2 locks held by ip/1861: [51015.579577] #0: 00000000a8c12fd1 (rtnl_mutex){+.+.}, at: rtnetlink_rcv_msg+0x2e0/0x910 [51015.579711] #1: 00000000bf815f8e (rcu_read_lock){....}, at: lwtunnel_build_state+0x96/0x390 [51015.579842] stack backtrace: [51015.579869] CPU: 0 PID: 1861 Comm: ip Not tainted 4.18.0-rc6+ #208 [51015.579891] Hardware name: To be filled by O.E.M. To be filled by O.E.M./Aptio CRB, BIOS 5.6.5 07/08/2015 [51015.579911] Call Trace: [51015.579950] dump_stack+0x74/0xbb [51015.580000] ___might_sleep+0x16b/0x3a0 [51015.580047] __kmalloc_track_caller+0x220/0x380 [51015.580077] kmemdup+0x1c/0x40 [51015.580077] bpf_parse_prog+0x10e/0x230 [51015.580164] ? kasan_kmalloc+0xa0/0xd0 [51015.580164] ? bpf_destroy_state+0x30/0x30 [51015.580164] ? bpf_build_state+0xe2/0x3e0 [51015.580164] bpf_build_state+0x1bb/0x3e0 [51015.580164] ? bpf_parse_prog+0x230/0x230 [51015.580164] ? lock_is_held_type+0x123/0x1a0 [51015.580164] lwtunnel_build_state+0x1aa/0x390 [51015.580164] fib_create_info+0x1579/0x33d0 [51015.580164] ? sched_clock_local+0xe2/0x150 [51015.580164] ? fib_info_update_nh_saddr+0x1f0/0x1f0 [51015.580164] ? sched_clock_local+0xe2/0x150 [51015.580164] fib_table_insert+0x201/0x1990 [51015.580164] ? lock_downgrade+0x610/0x610 [51015.580164] ? fib_table_lookup+0x1920/0x1920 [51015.580164] ? lwtunnel_valid_encap_type.part.6+0xcb/0x3a0 [51015.580164] ? rtm_to_fib_config+0x637/0xbd0 [51015.580164] inet_rtm_newroute+0xed/0x1b0 [51015.580164] ? rtm_to_fib_config+0xbd0/0xbd0 [51015.580164] rtnetlink_rcv_msg+0x331/0x910 [ ... ] Fixes: 3a0af8fd61f9 ("bpf: BPF for lightweight tunnel infrastructure") Signed-off-by: Taehee Yoo Signed-off-by: Daniel Borkmann --- net/core/lwt_bpf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/lwt_bpf.c b/net/core/lwt_bpf.c index e7e626fb87bb..e45098593dc0 100644 --- a/net/core/lwt_bpf.c +++ b/net/core/lwt_bpf.c @@ -217,7 +217,7 @@ static int bpf_parse_prog(struct nlattr *attr, struct bpf_lwt_prog *prog, if (!tb[LWT_BPF_PROG_FD] || !tb[LWT_BPF_PROG_NAME]) return -EINVAL; - prog->name = nla_memdup(tb[LWT_BPF_PROG_NAME], GFP_KERNEL); + prog->name = nla_memdup(tb[LWT_BPF_PROG_NAME], GFP_ATOMIC); if (!prog->name) return -ENOMEM; From ee614c871014045b45fae149b7245fc22a0bbdd8 Mon Sep 17 00:00:00 2001 From: John Hurley Date: Fri, 27 Jul 2018 20:56:52 -0700 Subject: [PATCH 27/41] nfp: flower: fix port metadata conversion bug Function nfp_flower_repr_get_type_and_port expects an enum nfp_repr_type return value but, if the repr type is unknown, returns a value of type enum nfp_flower_cmsg_port_type. This means that if FW encodes the port ID in a way the driver does not understand instead of dropping the frame driver may attribute it to a physical port (uplink) provided the port number is less than physical port count. Fix this and ensure a net_device of NULL is returned if the repr can not be determined. Fixes: 1025351a88a4 ("nfp: add flower app") Signed-off-by: John Hurley Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- drivers/net/ethernet/netronome/nfp/flower/main.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.c b/drivers/net/ethernet/netronome/nfp/flower/main.c index 1decf3a1cad3..e57d23746585 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/main.c +++ b/drivers/net/ethernet/netronome/nfp/flower/main.c @@ -80,7 +80,7 @@ nfp_flower_repr_get_type_and_port(struct nfp_app *app, u32 port_id, u8 *port) return NFP_REPR_TYPE_VF; } - return NFP_FLOWER_CMSG_PORT_TYPE_UNSPEC; + return __NFP_REPR_TYPE_MAX; } static struct net_device * @@ -91,6 +91,8 @@ nfp_flower_repr_get(struct nfp_app *app, u32 port_id) u8 port = 0; repr_type = nfp_flower_repr_get_type_and_port(app, port_id, &port); + if (repr_type > NFP_REPR_TYPE_MAX) + return NULL; reprs = rcu_dereference(app->reprs[repr_type]); if (!reprs) From 136f55f660192ce04af091642efc75d85e017364 Mon Sep 17 00:00:00 2001 From: Stefan Wahren Date: Sat, 28 Jul 2018 09:52:10 +0200 Subject: [PATCH 28/41] net: lan78xx: fix rx handling before first packet is send As long the bh tasklet isn't scheduled once, no packet from the rx path will be handled. Since the tx path also schedule the same tasklet this situation only persits until the first packet transmission. So fix this issue by scheduling the tasklet after link reset. Link: https://github.com/raspberrypi/linux/issues/2617 Fixes: 55d7de9de6c3 ("Microchip's LAN7800 family USB 2/3 to 10/100/1000 Ethernet") Suggested-by: Floris Bos Signed-off-by: Stefan Wahren Signed-off-by: David S. Miller --- drivers/net/usb/lan78xx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index ed10d49eb5e0..aeca484a75b8 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -1242,6 +1242,8 @@ static int lan78xx_link_reset(struct lan78xx_net *dev) mod_timer(&dev->stat_monitor, jiffies + STAT_UPDATE_TIMER); } + + tasklet_schedule(&dev->bh); } return ret; From ab123fe071c9aa9680ecd62eb080eb26cff4892c Mon Sep 17 00:00:00 2001 From: Govindarajulu Varadarajan Date: Fri, 27 Jul 2018 11:19:29 -0700 Subject: [PATCH 29/41] enic: handle mtu change for vf properly When driver gets notification for mtu change, driver does not handle it for all RQs. It handles only RQ[0]. Fix is to use enic_change_mtu() interface to change mtu for vf. Signed-off-by: Govindarajulu Varadarajan Signed-off-by: David S. Miller --- drivers/net/ethernet/cisco/enic/enic_main.c | 78 +++++++-------------- 1 file changed, 27 insertions(+), 51 deletions(-) diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index 90c645b8538e..6b0376123cde 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -2047,28 +2047,42 @@ static int enic_stop(struct net_device *netdev) return 0; } +static int _enic_change_mtu(struct net_device *netdev, int new_mtu) +{ + bool running = netif_running(netdev); + int err = 0; + + ASSERT_RTNL(); + if (running) { + err = enic_stop(netdev); + if (err) + return err; + } + + netdev->mtu = new_mtu; + + if (running) { + err = enic_open(netdev); + if (err) + return err; + } + + return 0; +} + static int enic_change_mtu(struct net_device *netdev, int new_mtu) { struct enic *enic = netdev_priv(netdev); - int running = netif_running(netdev); if (enic_is_dynamic(enic) || enic_is_sriov_vf(enic)) return -EOPNOTSUPP; - if (running) - enic_stop(netdev); - - netdev->mtu = new_mtu; - if (netdev->mtu > enic->port_mtu) netdev_warn(netdev, - "interface MTU (%d) set higher than port MTU (%d)\n", - netdev->mtu, enic->port_mtu); + "interface MTU (%d) set higher than port MTU (%d)\n", + netdev->mtu, enic->port_mtu); - if (running) - enic_open(netdev); - - return 0; + return _enic_change_mtu(netdev, new_mtu); } static void enic_change_mtu_work(struct work_struct *work) @@ -2076,47 +2090,9 @@ static void enic_change_mtu_work(struct work_struct *work) struct enic *enic = container_of(work, struct enic, change_mtu_work); struct net_device *netdev = enic->netdev; int new_mtu = vnic_dev_mtu(enic->vdev); - int err; - unsigned int i; - - new_mtu = max_t(int, ENIC_MIN_MTU, min_t(int, ENIC_MAX_MTU, new_mtu)); rtnl_lock(); - - /* Stop RQ */ - del_timer_sync(&enic->notify_timer); - - for (i = 0; i < enic->rq_count; i++) - napi_disable(&enic->napi[i]); - - vnic_intr_mask(&enic->intr[0]); - enic_synchronize_irqs(enic); - err = vnic_rq_disable(&enic->rq[0]); - if (err) { - rtnl_unlock(); - netdev_err(netdev, "Unable to disable RQ.\n"); - return; - } - vnic_rq_clean(&enic->rq[0], enic_free_rq_buf); - vnic_cq_clean(&enic->cq[0]); - vnic_intr_clean(&enic->intr[0]); - - /* Fill RQ with new_mtu-sized buffers */ - netdev->mtu = new_mtu; - vnic_rq_fill(&enic->rq[0], enic_rq_alloc_buf); - /* Need at least one buffer on ring to get going */ - if (vnic_rq_desc_used(&enic->rq[0]) == 0) { - rtnl_unlock(); - netdev_err(netdev, "Unable to alloc receive buffers.\n"); - return; - } - - /* Start RQ */ - vnic_rq_enable(&enic->rq[0]); - napi_enable(&enic->napi[0]); - vnic_intr_unmask(&enic->intr[0]); - enic_notify_timer_start(enic); - + (void)_enic_change_mtu(netdev, new_mtu); rtnl_unlock(); netdev_info(netdev, "interface MTU set as %d\n", netdev->mtu); From 9fc12023d6f51551d6ca9ed7e02ecc19d79caf17 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Fri, 27 Jul 2018 18:15:46 +0200 Subject: [PATCH 30/41] ipv4: remove BUG_ON() from fib_compute_spec_dst Remove BUG_ON() from fib_compute_spec_dst routine and check in_dev pointer during flowi4 data structure initialization. fib_compute_spec_dst routine can be run concurrently with device removal where ip_ptr net_device pointer is set to NULL. This can happen if userspace enables pkt info on UDP rx socket and the device is removed while traffic is flowing Fixes: 35ebf65e851c ("ipv4: Create and use fib_compute_spec_dst() helper") Signed-off-by: Lorenzo Bianconi Signed-off-by: David S. Miller --- net/ipv4/fib_frontend.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index e46cdd310e5f..2998b0e47d4b 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -292,19 +292,19 @@ __be32 fib_compute_spec_dst(struct sk_buff *skb) return ip_hdr(skb)->daddr; in_dev = __in_dev_get_rcu(dev); - BUG_ON(!in_dev); net = dev_net(dev); scope = RT_SCOPE_UNIVERSE; if (!ipv4_is_zeronet(ip_hdr(skb)->saddr)) { + bool vmark = in_dev && IN_DEV_SRC_VMARK(in_dev); struct flowi4 fl4 = { .flowi4_iif = LOOPBACK_IFINDEX, .flowi4_oif = l3mdev_master_ifindex_rcu(dev), .daddr = ip_hdr(skb)->saddr, .flowi4_tos = RT_TOS(ip_hdr(skb)->tos), .flowi4_scope = scope, - .flowi4_mark = IN_DEV_SRC_VMARK(in_dev) ? skb->mark : 0, + .flowi4_mark = vmark ? skb->mark : 0, }; if (!fib_lookup(net, &fl4, &res, 0)) return FIB_RES_PREFSRC(net, res); From b0753408aadf32c7ece9e6b765017881e54af833 Mon Sep 17 00:00:00 2001 From: Anton Vasilyev Date: Fri, 27 Jul 2018 18:57:47 +0300 Subject: [PATCH 31/41] net: mdio-mux: bcm-iproc: fix wrong getter and setter pair mdio_mux_iproc_probe() uses platform_set_drvdata() to store md pointer in device, whereas mdio_mux_iproc_remove() restores md pointer by dev_get_platdata(&pdev->dev). This leads to wrong resources release. The patch replaces getter to platform_get_drvdata. Fixes: 98bc865a1ec8 ("net: mdio-mux: Add MDIO mux driver for iProc SoCs") Signed-off-by: Anton Vasilyev Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/phy/mdio-mux-bcm-iproc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/mdio-mux-bcm-iproc.c b/drivers/net/phy/mdio-mux-bcm-iproc.c index 0831b7142df7..0c5b68e7da51 100644 --- a/drivers/net/phy/mdio-mux-bcm-iproc.c +++ b/drivers/net/phy/mdio-mux-bcm-iproc.c @@ -218,7 +218,7 @@ out: static int mdio_mux_iproc_remove(struct platform_device *pdev) { - struct iproc_mdiomux_desc *md = dev_get_platdata(&pdev->dev); + struct iproc_mdiomux_desc *md = platform_get_drvdata(pdev); mdio_mux_uninit(md->mux_handle); mdiobus_unregister(md->mii_bus); From c8e8cd579bb4265651df8223730105341e61a2d1 Mon Sep 17 00:00:00 2001 From: Jeremy Cline Date: Fri, 27 Jul 2018 22:43:01 +0000 Subject: [PATCH 32/41] net: socket: fix potential spectre v1 gadget in socketcall 'call' is a user-controlled value, so sanitize the array index after the bounds check to avoid speculating past the bounds of the 'nargs' array. Found with the help of Smatch: net/socket.c:2508 __do_sys_socketcall() warn: potential spectre issue 'nargs' [r] (local cap) Cc: Josh Poimboeuf Cc: stable@vger.kernel.org Signed-off-by: Jeremy Cline Signed-off-by: David S. Miller --- net/socket.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/socket.c b/net/socket.c index 85633622c94d..4ac3b834cce9 100644 --- a/net/socket.c +++ b/net/socket.c @@ -89,6 +89,7 @@ #include #include #include +#include #include #include @@ -2522,6 +2523,7 @@ SYSCALL_DEFINE2(socketcall, int, call, unsigned long __user *, args) if (call < 1 || call > SYS_SENDMMSG) return -EINVAL; + call = array_index_nospec(call, SYS_SENDMMSG + 1); len = nargs[call]; if (len > sizeof(a)) From e978de7a6d382ec378830ca2cf38e902df0b6d84 Mon Sep 17 00:00:00 2001 From: Jeremy Cline Date: Fri, 27 Jul 2018 22:43:02 +0000 Subject: [PATCH 33/41] net: socket: Fix potential spectre v1 gadget in sock_is_registered 'family' can be a user-controlled value, so sanitize it after the bounds check to avoid speculative out-of-bounds access. Cc: Josh Poimboeuf Cc: stable@vger.kernel.org Signed-off-by: Jeremy Cline Signed-off-by: David S. Miller --- net/socket.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/socket.c b/net/socket.c index 4ac3b834cce9..8c24d5dc4bc8 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2690,7 +2690,8 @@ EXPORT_SYMBOL(sock_unregister); bool sock_is_registered(int family) { - return family < NPROTO && rcu_access_pointer(net_families[family]); + return family < NPROTO && + rcu_access_pointer(net_families[array_index_nospec(family, NPROTO)]); } static int __init sock_init(void) From 383d470936c05554219094a4d364d964cb324827 Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Fri, 27 Jul 2018 17:19:12 -0400 Subject: [PATCH 34/41] tcp_bbr: fix bw probing to raise in-flight data for very small BDPs For some very small BDPs (with just a few packets) there was a quantization effect where the target number of packets in flight during the super-unity-gain (1.25x) phase of gain cycling was implicitly truncated to a number of packets no larger than the normal unity-gain (1.0x) phase of gain cycling. This meant that in multi-flow scenarios some flows could get stuck with a lower bandwidth, because they did not push enough packets inflight to discover that there was more bandwidth available. This was really only an issue in multi-flow LAN scenarios, where RTTs and BDPs are low enough for this to be an issue. This fix ensures that gain cycling can raise inflight for small BDPs by ensuring that in PROBE_BW mode target inflight values with a super-unity gain are always greater than inflight values with a gain <= 1. Importantly, this applies whether the inflight value is calculated for use as a cwnd value, or as a target inflight value for the end of the super-unity phase in bbr_is_next_cycle_phase() (both need to be bigger to ensure we can probe with more packets in flight reliably). This is a candidate fix for stable releases. Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control") Signed-off-by: Neal Cardwell Acked-by: Yuchung Cheng Acked-by: Soheil Hassas Yeganeh Acked-by: Priyaranjan Jha Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_bbr.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index 58e2f479ffb4..4bfff3c87e8e 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -354,6 +354,10 @@ static u32 bbr_target_cwnd(struct sock *sk, u32 bw, int gain) /* Reduce delayed ACKs by rounding up cwnd to the next even number. */ cwnd = (cwnd + 1) & ~1U; + /* Ensure gain cycling gets inflight above BDP even for small BDPs. */ + if (bbr->mode == BBR_PROBE_BW && gain > BBR_UNIT) + cwnd += 2; + return cwnd; } From 9939a46d90c6c76f4533d534dbadfa7b39dc6acc Mon Sep 17 00:00:00 2001 From: Eugeniy Paltsev Date: Thu, 26 Jul 2018 15:05:37 +0300 Subject: [PATCH 35/41] NET: stmmac: align DMA stuff to largest cache line length As for today STMMAC_ALIGN macro (which is used to align DMA stuff) relies on L1 line length (L1_CACHE_BYTES). This isn't correct in case of system with several cache levels which might have L1 cache line length smaller than L2 line. This can lead to sharing one cache line between DMA buffer and other data, so we can lose this data while invalidate DMA buffer before DMA transaction. Fix that by using SMP_CACHE_BYTES instead of L1_CACHE_BYTES for aligning. Signed-off-by: Eugeniy Paltsev Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 60f59abab009..ef6a8d39db2f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -53,7 +53,7 @@ #include "dwmac1000.h" #include "hwif.h" -#define STMMAC_ALIGN(x) L1_CACHE_ALIGN(x) +#define STMMAC_ALIGN(x) __ALIGN_KERNEL(x, SMP_CACHE_BYTES) #define TSO_MAX_BUFF_SIZE (SZ_16K - 1) /* Module parameters */ From 7acf9d4237c46894e0fa0492dd96314a41742e84 Mon Sep 17 00:00:00 2001 From: Dmitry Safonov Date: Fri, 27 Jul 2018 16:54:44 +0100 Subject: [PATCH 36/41] netlink: Do not subscribe to non-existent groups Make ABI more strict about subscribing to group > ngroups. Code doesn't check for that and it looks bogus. (one can subscribe to non-existing group) Still, it's possible to bind() to all possible groups with (-1) Cc: "David S. Miller" Cc: Herbert Xu Cc: Steffen Klassert Cc: netdev@vger.kernel.org Signed-off-by: Dmitry Safonov Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 393573a99a5a..ac805caed2e2 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1008,6 +1008,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, if (err) return err; } + groups &= (1UL << nlk->ngroups) - 1; bound = nlk->bound; if (bound) { From 25432eba9cd8f2ef5afef55be811b010a004b5fa Mon Sep 17 00:00:00 2001 From: Justin Pettit Date: Sat, 28 Jul 2018 15:26:01 -0700 Subject: [PATCH 37/41] openvswitch: meter: Fix setting meter id for new entries The meter code would create an entry for each new meter. However, it would not set the meter id in the new entry, so every meter would appear to have a meter id of zero. This commit properly sets the meter id when adding the entry. Fixes: 96fbc13d7e77 ("openvswitch: Add meter infrastructure") Signed-off-by: Justin Pettit Cc: Andy Zhou Signed-off-by: David S. Miller --- net/openvswitch/meter.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/net/openvswitch/meter.c b/net/openvswitch/meter.c index b891a91577f8..c038e021a591 100644 --- a/net/openvswitch/meter.c +++ b/net/openvswitch/meter.c @@ -211,6 +211,7 @@ static struct dp_meter *dp_meter_create(struct nlattr **a) if (!meter) return ERR_PTR(-ENOMEM); + meter->id = nla_get_u32(a[OVS_METER_ATTR_ID]); meter->used = div_u64(ktime_get_ns(), 1000 * 1000); meter->kbps = a[OVS_METER_ATTR_KBPS] ? 1 : 0; meter->keep_stats = !a[OVS_METER_ATTR_CLEAR]; @@ -280,6 +281,10 @@ static int ovs_meter_cmd_set(struct sk_buff *skb, struct genl_info *info) u32 meter_id; bool failed; + if (!a[OVS_METER_ATTR_ID]) { + return -ENODEV; + } + meter = dp_meter_create(a); if (IS_ERR_OR_NULL(meter)) return PTR_ERR(meter); @@ -298,11 +303,6 @@ static int ovs_meter_cmd_set(struct sk_buff *skb, struct genl_info *info) goto exit_unlock; } - if (!a[OVS_METER_ATTR_ID]) { - err = -ENODEV; - goto exit_unlock; - } - meter_id = nla_get_u32(a[OVS_METER_ATTR_ID]); /* Cannot fail after this. */ From 72c05f32f4a5055c9c8fe889bb6903ec959c0aad Mon Sep 17 00:00:00 2001 From: Anton Vasilyev Date: Fri, 27 Jul 2018 18:50:42 +0300 Subject: [PATCH 38/41] can: ems_usb: Fix memory leak on ems_usb_disconnect() ems_usb_probe() allocates memory for dev->tx_msg_buffer, but there is no its deallocation in ems_usb_disconnect(). Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Anton Vasilyev Cc: Signed-off-by: Marc Kleine-Budde --- drivers/net/can/usb/ems_usb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index 12ff0020ecd6..b7dfd4109d24 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -1072,6 +1072,7 @@ static void ems_usb_disconnect(struct usb_interface *intf) usb_free_urb(dev->intr_urb); kfree(dev->intr_in_buffer); + kfree(dev->tx_msg_buffer); } } From 822fb18a82abaf4ee7058793d95d340f5dab7bfc Mon Sep 17 00:00:00 2001 From: Xiao Liang Date: Fri, 27 Jul 2018 17:56:08 +0800 Subject: [PATCH 39/41] xen-netfront: wait xenbus state change when load module manually When loading module manually, after call xenbus_switch_state to initializes the state of the netfront device, the driver state did not change so fast that may lead no dev created in latest kernel. This patch adds wait to make sure xenbus knows the driver is not in closed/unknown state. Current state: [vm]# ethtool eth0 Settings for eth0: Link detected: yes [vm]# modprobe -r xen_netfront [vm]# modprobe xen_netfront [vm]# ethtool eth0 Settings for eth0: Cannot get device settings: No such device Cannot get wake-on-lan settings: No such device Cannot get message level: No such device Cannot get link status: No such device No data available With the patch installed. [vm]# ethtool eth0 Settings for eth0: Link detected: yes [vm]# modprobe -r xen_netfront [vm]# modprobe xen_netfront [vm]# ethtool eth0 Settings for eth0: Link detected: yes Signed-off-by: Xiao Liang Signed-off-by: David S. Miller --- drivers/net/xen-netfront.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index a57daecf1d57..2d8812dd1534 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -87,6 +87,7 @@ struct netfront_cb { /* IRQ name is queue name with "-tx" or "-rx" appended */ #define IRQ_NAME_SIZE (QUEUE_NAME_SIZE + 3) +static DECLARE_WAIT_QUEUE_HEAD(module_load_q); static DECLARE_WAIT_QUEUE_HEAD(module_unload_q); struct netfront_stats { @@ -1330,6 +1331,11 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev) netif_carrier_off(netdev); xenbus_switch_state(dev, XenbusStateInitialising); + wait_event(module_load_q, + xenbus_read_driver_state(dev->otherend) != + XenbusStateClosed && + xenbus_read_driver_state(dev->otherend) != + XenbusStateUnknown); return netdev; exit: From df18b50448fab1dff093731dfd0e25e77e1afcd1 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 30 Jul 2018 16:23:10 +0200 Subject: [PATCH 40/41] net/ipv6: fix metrics leak Since commit d4ead6b34b67 ("net/ipv6: move metrics from dst to rt6_info"), ipv6 metrics are shared and refcounted. rt6_set_from() assigns the rt->from pointer and increases the refcount on from's metrics. This reference is never released. Introduce the fib6_metrics_release() helper and use it to release the metrics. Fixes: d4ead6b34b67 ("net/ipv6: move metrics from dst to rt6_info") Signed-off-by: Sabrina Dubroca Signed-off-by: David S. Miller --- net/ipv6/ip6_fib.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index d212738e9d10..211a2d437b56 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -167,11 +167,22 @@ struct fib6_info *fib6_info_alloc(gfp_t gfp_flags) return f6i; } +static void fib6_metrics_release(struct fib6_info *f6i) +{ + struct dst_metrics *m; + + if (!f6i) + return; + + m = f6i->fib6_metrics; + if (m != &dst_default_metrics && refcount_dec_and_test(&m->refcnt)) + kfree(m); +} + void fib6_info_destroy_rcu(struct rcu_head *head) { struct fib6_info *f6i = container_of(head, struct fib6_info, rcu); struct rt6_exception_bucket *bucket; - struct dst_metrics *m; WARN_ON(f6i->fib6_node); @@ -201,9 +212,7 @@ void fib6_info_destroy_rcu(struct rcu_head *head) if (f6i->fib6_nh.nh_dev) dev_put(f6i->fib6_nh.nh_dev); - m = f6i->fib6_metrics; - if (m != &dst_default_metrics && refcount_dec_and_test(&m->refcnt)) - kfree(m); + fib6_metrics_release(f6i); kfree(f6i); } @@ -887,6 +896,7 @@ static void fib6_drop_pcpu_from(struct fib6_info *f6i, from = rcu_dereference_protected(pcpu_rt->from, lockdep_is_held(&table->tb6_lock)); + fib6_metrics_release(from); rcu_assign_pointer(pcpu_rt->from, NULL); fib6_info_release(from); } From 61f4b23769f0cc72ae62c9a81cf08f0397d40da8 Mon Sep 17 00:00:00 2001 From: Dmitry Safonov Date: Mon, 30 Jul 2018 18:32:36 +0100 Subject: [PATCH 41/41] netlink: Don't shift with UB on nlk->ngroups On i386 nlk->ngroups might be 32 or 0. Which leads to UB, resulting in hang during boot. Check for 0 ngroups and use (unsigned long long) as a type to shift. Fixes: 7acf9d4237c4 ("netlink: Do not subscribe to non-existent groups"). Reported-by: kernel test robot Signed-off-by: Dmitry Safonov Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index ac805caed2e2..7d860a22e5fb 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1008,7 +1008,11 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, if (err) return err; } - groups &= (1UL << nlk->ngroups) - 1; + + if (nlk->ngroups == 0) + groups = 0; + else + groups &= (1ULL << nlk->ngroups) - 1; bound = nlk->bound; if (bound) {