From 955d3411a17f590364238bd0d3329b61f20c1cd2 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 30 Dec 2018 12:46:01 +0100 Subject: batman-adv: Avoid WARN on net_device without parent in netns It is not allowed to use WARN* helpers on potential incorrect input from the user or transient problems because systems configured as panic_on_warn will reboot due to such a problem. A NULL return value of __dev_get_by_index can be caused by various problems which can either be related to the system configuration or problems (incorrectly returned network namespaces) in other (virtual) net_device drivers. batman-adv should not cause a (harmful) WARN in this situation and instead only report it via a simple message. Fixes: b7eddd0b3950 ("batman-adv: prevent using any virtual device created on batman-adv as hard-interface") Reported-by: syzbot+c764de0fcfadca9a8595@syzkaller.appspotmail.com Reported-by: Dmitry Vyukov Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/hard-interface.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index 508f4416dfc9..415d494cbe22 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -20,7 +20,6 @@ #include "main.h" #include -#include #include #include #include @@ -179,8 +178,10 @@ static bool batadv_is_on_batman_iface(const struct net_device *net_dev) parent_dev = __dev_get_by_index((struct net *)parent_net, dev_get_iflink(net_dev)); /* if we got a NULL parent_dev there is something broken.. */ - if (WARN(!parent_dev, "Cannot find parent device")) + if (!parent_dev) { + pr_err("Cannot find parent device\n"); return false; + } if (batadv_mutual_parents(net_dev, net, parent_dev, parent_net)) return false; -- cgit v1.2.3 From 9114daa825fc3f335f9bea3313ce667090187280 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Mon, 31 Dec 2018 22:31:01 +0100 Subject: batman-adv: Force mac header to start of data on xmit The caller of ndo_start_xmit may not already have called skb_reset_mac_header. The returned value of skb_mac_header/eth_hdr therefore can be in the wrong position and even outside the current skbuff. This for example happens when the user binds to the device using a PF_PACKET-SOCK_RAW with enabled qdisc-bypass: int opt = 4; setsockopt(sock, SOL_PACKET, PACKET_QDISC_BYPASS, &opt, sizeof(opt)); Since eth_hdr is used all over the codebase, the batadv_interface_tx function must always take care of resetting it. Fixes: c6c8fea29769 ("net: Add batman-adv meshing protocol") Reported-by: syzbot+9d7405c7faa390e60b4e@syzkaller.appspotmail.com Reported-by: syzbot+7d20bc3f1ddddc0f9079@syzkaller.appspotmail.com Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/soft-interface.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 5db5a0a4c959..b85ca809e509 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -221,6 +221,8 @@ static netdev_tx_t batadv_interface_tx(struct sk_buff *skb, netif_trans_update(soft_iface); vid = batadv_get_vid(skb, 0); + + skb_reset_mac_header(skb); ethhdr = eth_hdr(skb); switch (ntohs(ethhdr->h_proto)) { -- cgit v1.2.3 From c9e4576743eeda8d24dedc164d65b78877f9a98c Mon Sep 17 00:00:00 2001 From: Yafang Shao Date: Wed, 23 Jan 2019 12:37:19 +0800 Subject: bpf: sock recvbuff must be limited by rmem_max in bpf_setsockopt() When sock recvbuff is set by bpf_setsockopt(), the value must by limited by rmem_max. It is the same with sendbuff. Fixes: 8c4b4c7e9ff0 ("bpf: Add setsockopt helper function to bpf") Signed-off-by: Yafang Shao Acked-by: Martin KaFai Lau Acked-by: Lawrence Brakmo Signed-off-by: Daniel Borkmann --- net/core/filter.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 7559d6835ecb..7a54dc11ac2d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4112,10 +4112,12 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, /* Only some socketops are supported */ switch (optname) { case SO_RCVBUF: + val = min_t(u32, val, sysctl_rmem_max); sk->sk_userlocks |= SOCK_RCVBUF_LOCK; sk->sk_rcvbuf = max_t(int, val * 2, SOCK_MIN_RCVBUF); break; case SO_SNDBUF: + val = min_t(u32, val, sysctl_wmem_max); sk->sk_userlocks |= SOCK_SNDBUF_LOCK; sk->sk_sndbuf = max_t(int, val * 2, SOCK_MIN_SNDBUF); break; -- cgit v1.2.3 From 7d652669b61d702c6e62a39579d17f6881670ab6 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 25 Jan 2019 08:21:26 +0100 Subject: batman-adv: release station info tidstats With the addition of TXQ stats in the per-tid statistics the struct station_info grew significantly. This resulted in stack size warnings due to the structure itself being above the limit for the warnings. To work around this, the TID array was allocated dynamically. Also a function to free this content was introduced with commit 7ea3e110f2f8 ("cfg80211: release station info tidstats where needed") but the necessary changes were not provided for batman-adv's B.A.T.M.A.N. V implementation. Signed-off-by: Felix Fietkau Fixes: 8689c051a201 ("cfg80211: dynamically allocate per-tid stats for station info") [sven@narfation.org: add commit message] Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/bat_v_elp.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index e8090f099eb8..ef0dec20c7d8 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -104,6 +104,9 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) ret = cfg80211_get_station(real_netdev, neigh->addr, &sinfo); + /* free the TID stats immediately */ + cfg80211_sinfo_release_content(&sinfo); + dev_put(real_netdev); if (ret == -ENOENT) { /* Node is not associated anymore! It would be -- cgit v1.2.3 From 1d79895aef18fa05789995d86d523c9b2ee58a02 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Mon, 28 Jan 2019 10:13:35 +0100 Subject: sk_msg: Always cancel strp work before freeing the psock Despite having stopped the parser, we still need to deinitialize it by calling strp_done so that it cancels its work. Otherwise the worker thread can run after we have freed the parser, and attempt to access its workqueue resulting in a use-after-free: ================================================================== BUG: KASAN: use-after-free in pwq_activate_delayed_work+0x1b/0x1d0 Read of size 8 at addr ffff888069975240 by task kworker/u2:2/93 CPU: 0 PID: 93 Comm: kworker/u2:2 Not tainted 5.0.0-rc2-00335-g28f9d1a3d4fe-dirty #14 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-2.fc27 04/01/2014 Workqueue: (null) (kstrp) Call Trace: print_address_description+0x6e/0x2b0 ? pwq_activate_delayed_work+0x1b/0x1d0 kasan_report+0xfd/0x177 ? pwq_activate_delayed_work+0x1b/0x1d0 ? pwq_activate_delayed_work+0x1b/0x1d0 pwq_activate_delayed_work+0x1b/0x1d0 ? process_one_work+0x4aa/0x660 pwq_dec_nr_in_flight+0x9b/0x100 worker_thread+0x82/0x680 ? process_one_work+0x660/0x660 kthread+0x1b9/0x1e0 ? __kthread_create_on_node+0x250/0x250 ret_from_fork+0x1f/0x30 Allocated by task 111: sk_psock_init+0x3c/0x1b0 sock_map_link.isra.2+0x103/0x4b0 sock_map_update_common+0x94/0x270 sock_map_update_elem+0x145/0x160 __se_sys_bpf+0x152e/0x1e10 do_syscall_64+0xb2/0x3e0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Freed by task 112: kfree+0x7f/0x140 process_one_work+0x40b/0x660 worker_thread+0x82/0x680 kthread+0x1b9/0x1e0 ret_from_fork+0x1f/0x30 The buggy address belongs to the object at ffff888069975180 which belongs to the cache kmalloc-512 of size 512 The buggy address is located 192 bytes inside of 512-byte region [ffff888069975180, ffff888069975380) The buggy address belongs to the page: page:ffffea0001a65d00 count:1 mapcount:0 mapping:ffff88806d401280 index:0x0 compound_mapcount: 0 flags: 0x4000000000010200(slab|head) raw: 4000000000010200 dead000000000100 dead000000000200 ffff88806d401280 raw: 0000000000000000 00000000800c000c 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff888069975100: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff888069975180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff888069975200: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff888069975280: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff888069975300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== Reported-by: Marek Majkowski Signed-off-by: Jakub Sitnicki Link: https://lore.kernel.org/netdev/CAJPywTLwgXNEZ2dZVoa=udiZmtrWJ0q5SuBW64aYs0Y1khXX3A@mail.gmail.com Acked-by: Song Liu Signed-off-by: Daniel Borkmann --- net/core/skmsg.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/skmsg.c b/net/core/skmsg.c index d6d5c20d7044..8c826603bf36 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -545,8 +545,7 @@ static void sk_psock_destroy_deferred(struct work_struct *gc) struct sk_psock *psock = container_of(gc, struct sk_psock, gc); /* No sk_callback_lock since already detached. */ - if (psock->parser.enabled) - strp_done(&psock->parser.strp); + strp_done(&psock->parser.strp); cancel_work_sync(&psock->work); -- cgit v1.2.3 From 63ff03ab786ab1bc6cca01d48eacd22c95b9b3eb Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 25 Jan 2019 22:43:17 +0100 Subject: Revert "socket: fix struct ifreq size in compat ioctl" This reverts commit 1cebf8f143c2 ("socket: fix struct ifreq size in compat ioctl"), it's a bugfix for another commit that I'll revert next. This is not a 'perfect' revert, I'm keeping some coding style intact rather than revert to the state with indentation errors. Cc: stable@vger.kernel.org Fixes: 1cebf8f143c2 ("socket: fix struct ifreq size in compat ioctl") Signed-off-by: Johannes Berg Signed-off-by: David S. Miller --- net/socket.c | 22 ++++++++-------------- 1 file changed, 8 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/socket.c b/net/socket.c index e89884e2197b..63b53af7379b 100644 --- a/net/socket.c +++ b/net/socket.c @@ -941,8 +941,7 @@ void dlci_ioctl_set(int (*hook) (unsigned int, void __user *)) EXPORT_SYMBOL(dlci_ioctl_set); static long sock_do_ioctl(struct net *net, struct socket *sock, - unsigned int cmd, unsigned long arg, - unsigned int ifreq_size) + unsigned int cmd, unsigned long arg) { int err; void __user *argp = (void __user *)arg; @@ -968,11 +967,11 @@ static long sock_do_ioctl(struct net *net, struct socket *sock, } else { struct ifreq ifr; bool need_copyout; - if (copy_from_user(&ifr, argp, ifreq_size)) + if (copy_from_user(&ifr, argp, sizeof(struct ifreq))) return -EFAULT; err = dev_ioctl(net, cmd, &ifr, &need_copyout); if (!err && need_copyout) - if (copy_to_user(argp, &ifr, ifreq_size)) + if (copy_to_user(argp, &ifr, sizeof(struct ifreq))) return -EFAULT; } return err; @@ -1071,8 +1070,7 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) err = open_related_ns(&net->ns, get_net_ns); break; default: - err = sock_do_ioctl(net, sock, cmd, arg, - sizeof(struct ifreq)); + err = sock_do_ioctl(net, sock, cmd, arg); break; } return err; @@ -2780,8 +2778,7 @@ static int do_siocgstamp(struct net *net, struct socket *sock, int err; set_fs(KERNEL_DS); - err = sock_do_ioctl(net, sock, cmd, (unsigned long)&ktv, - sizeof(struct compat_ifreq)); + err = sock_do_ioctl(net, sock, cmd, (unsigned long)&ktv); set_fs(old_fs); if (!err) err = compat_put_timeval(&ktv, up); @@ -2797,8 +2794,7 @@ static int do_siocgstampns(struct net *net, struct socket *sock, int err; set_fs(KERNEL_DS); - err = sock_do_ioctl(net, sock, cmd, (unsigned long)&kts, - sizeof(struct compat_ifreq)); + err = sock_do_ioctl(net, sock, cmd, (unsigned long)&kts); set_fs(old_fs); if (!err) err = compat_put_timespec(&kts, up); @@ -3109,8 +3105,7 @@ static int routing_ioctl(struct net *net, struct socket *sock, } set_fs(KERNEL_DS); - ret = sock_do_ioctl(net, sock, cmd, (unsigned long) r, - sizeof(struct compat_ifreq)); + ret = sock_do_ioctl(net, sock, cmd, (unsigned long) r); set_fs(old_fs); out: @@ -3223,8 +3218,7 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCBONDSETHWADDR: case SIOCBONDCHANGEACTIVE: case SIOCGIFNAME: - return sock_do_ioctl(net, sock, cmd, arg, - sizeof(struct compat_ifreq)); + return sock_do_ioctl(net, sock, cmd, arg); } return -ENOIOCTLCMD; -- cgit v1.2.3 From 37ac39bdddc528c998a9f36db36937de923fdf2a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 25 Jan 2019 22:43:18 +0100 Subject: Revert "kill dev_ifsioc()" This reverts commit bf4405737f9f ("kill dev_ifsioc()"). This wasn't really unused as implied by the original commit, it still handles the copy to/from user differently, and the commit thus caused issues such as https://bugzilla.kernel.org/show_bug.cgi?id=199469 and https://bugzilla.kernel.org/show_bug.cgi?id=202273 However, deviating from a strict revert, rename dev_ifsioc() to compat_ifreq_ioctl() to be clearer as to its purpose and add a comment. Cc: stable@vger.kernel.org Fixes: bf4405737f9f ("kill dev_ifsioc()") Signed-off-by: Johannes Berg Signed-off-by: David S. Miller --- net/socket.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) (limited to 'net') diff --git a/net/socket.c b/net/socket.c index 63b53af7379b..fbf80f9fb057 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2990,6 +2990,53 @@ static int compat_ifr_data_ioctl(struct net *net, unsigned int cmd, return dev_ioctl(net, cmd, &ifreq, NULL); } +static int compat_ifreq_ioctl(struct net *net, struct socket *sock, + unsigned int cmd, + struct compat_ifreq __user *uifr32) +{ + struct ifreq __user *uifr; + int err; + + /* Handle the fact that while struct ifreq has the same *layout* on + * 32/64 for everything but ifreq::ifru_ifmap and ifreq::ifru_data, + * which are handled elsewhere, it still has different *size* due to + * ifreq::ifru_ifmap (which is 16 bytes on 32 bit, 24 bytes on 64-bit, + * resulting in struct ifreq being 32 and 40 bytes respectively). + * As a result, if the struct happens to be at the end of a page and + * the next page isn't readable/writable, we get a fault. To prevent + * that, copy back and forth to the full size. + */ + + uifr = compat_alloc_user_space(sizeof(*uifr)); + if (copy_in_user(uifr, uifr32, sizeof(*uifr32))) + return -EFAULT; + + err = sock_do_ioctl(net, sock, cmd, (unsigned long)uifr); + + if (!err) { + switch (cmd) { + case SIOCGIFFLAGS: + case SIOCGIFMETRIC: + case SIOCGIFMTU: + case SIOCGIFMEM: + case SIOCGIFHWADDR: + case SIOCGIFINDEX: + case SIOCGIFADDR: + case SIOCGIFBRDADDR: + case SIOCGIFDSTADDR: + case SIOCGIFNETMASK: + case SIOCGIFPFLAGS: + case SIOCGIFTXQLEN: + case SIOCGMIIPHY: + case SIOCGMIIREG: + if (copy_in_user(uifr32, uifr, sizeof(*uifr32))) + err = -EFAULT; + break; + } + } + return err; +} + static int compat_sioc_ifmap(struct net *net, unsigned int cmd, struct compat_ifreq __user *uifr32) { @@ -3209,6 +3256,8 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: + return compat_ifreq_ioctl(net, sock, cmd, argp); + case SIOCSARP: case SIOCGARP: case SIOCDARP: -- cgit v1.2.3 From c6c9fee35dc27362b7bac34b2fc9f5b8ace2e22c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 25 Jan 2019 22:43:19 +0100 Subject: net: socket: fix SIOCGIFNAME in compat As reported by Robert O'Callahan in https://bugzilla.kernel.org/show_bug.cgi?id=202273 reverting the previous changes in this area broke the SIOCGIFNAME ioctl in compat again (I'd previously fixed it after his previous report of breakage in https://bugzilla.kernel.org/show_bug.cgi?id=199469). This is obviously because I fixed SIOCGIFNAME more or less by accident. Fix it explicitly now by making it pass through the restored compat translation code. Cc: stable@vger.kernel.org Fixes: 4cf808e7ac32 ("kill dev_ifname32()") Reported-by: Robert O'Callahan Signed-off-by: Johannes Berg Signed-off-by: David S. Miller --- net/socket.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/socket.c b/net/socket.c index fbf80f9fb057..473ac8d7c54e 100644 --- a/net/socket.c +++ b/net/socket.c @@ -3029,6 +3029,7 @@ static int compat_ifreq_ioctl(struct net *net, struct socket *sock, case SIOCGIFTXQLEN: case SIOCGMIIPHY: case SIOCGMIIREG: + case SIOCGIFNAME: if (copy_in_user(uifr32, uifr, sizeof(*uifr32))) err = -EFAULT; break; @@ -3252,6 +3253,7 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCSIFTXQLEN: case SIOCBRADDIF: case SIOCBRDELIF: + case SIOCGIFNAME: case SIOCSIFNAME: case SIOCGMIIPHY: case SIOCGMIIREG: @@ -3266,7 +3268,6 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCBONDRELEASE: case SIOCBONDSETHWADDR: case SIOCBONDCHANGEACTIVE: - case SIOCGIFNAME: return sock_do_ioctl(net, sock, cmd, arg); } -- cgit v1.2.3 From 98406133dd9cb9f195676eab540c270dceca879a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 25 Jan 2019 22:43:20 +0100 Subject: net: socket: make bond ioctls go through compat_ifreq_ioctl() Same story as before, these use struct ifreq and thus need to be read with the shorter version to not cause faults. Cc: stable@vger.kernel.org Fixes: f92d4fc95341 ("kill bond_ioctl()") Signed-off-by: Johannes Berg Signed-off-by: David S. Miller --- net/socket.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/socket.c b/net/socket.c index 473ac8d7c54e..d80d87a395ea 100644 --- a/net/socket.c +++ b/net/socket.c @@ -3258,16 +3258,16 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: + case SIOCBONDENSLAVE: + case SIOCBONDRELEASE: + case SIOCBONDSETHWADDR: + case SIOCBONDCHANGEACTIVE: return compat_ifreq_ioctl(net, sock, cmd, argp); case SIOCSARP: case SIOCGARP: case SIOCDARP: case SIOCATMARK: - case SIOCBONDENSLAVE: - case SIOCBONDRELEASE: - case SIOCBONDSETHWADDR: - case SIOCBONDCHANGEACTIVE: return sock_do_ioctl(net, sock, cmd, arg); } -- cgit v1.2.3 From feaf5c796b3f0240f10d0d6d0b686715fd58a05b Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 28 Jan 2019 22:23:48 +0100 Subject: net: ip_gre: always reports o_key to userspace Erspan protocol (version 1 and 2) relies on o_key to configure session id header field. However TUNNEL_KEY bit is cleared in erspan_xmit since ERSPAN protocol does not set the key field of the external GRE header and so the configured o_key is not reported to userspace. The issue can be triggered with the following reproducer: $ip link add erspan1 type erspan local 192.168.0.1 remote 192.168.0.2 \ key 1 seq erspan_ver 1 $ip link set erspan1 up $ip -d link sh erspan1 erspan1@NONE: mtu 1450 qdisc pfifo_fast state UNKNOWN mode DEFAULT link/ether 52:aa:99:95:9a:b5 brd ff:ff:ff:ff:ff:ff promiscuity 0 minmtu 68 maxmtu 1500 erspan remote 192.168.0.2 local 192.168.0.1 ttl inherit ikey 0.0.0.1 iseq oseq erspan_index 0 Fix the issue adding TUNNEL_KEY bit to the o_flags parameter in ipgre_fill_info Fixes: 84e54fe0a5ea ("gre: introduce native tunnel support for ERSPAN") Signed-off-by: Lorenzo Bianconi Signed-off-by: David S. Miller --- net/ipv4/ip_gre.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 20a64fe6254b..3978f807fa8b 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -1455,12 +1455,17 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip_tunnel *t = netdev_priv(dev); struct ip_tunnel_parm *p = &t->parms; + __be16 o_flags = p->o_flags; + + if ((t->erspan_ver == 1 || t->erspan_ver == 2) && + !t->collect_md) + o_flags |= TUNNEL_KEY; if (nla_put_u32(skb, IFLA_GRE_LINK, p->link) || nla_put_be16(skb, IFLA_GRE_IFLAGS, gre_tnl_flags_to_gre_flags(p->i_flags)) || nla_put_be16(skb, IFLA_GRE_OFLAGS, - gre_tnl_flags_to_gre_flags(p->o_flags)) || + gre_tnl_flags_to_gre_flags(o_flags)) || nla_put_be32(skb, IFLA_GRE_IKEY, p->i_key) || nla_put_be32(skb, IFLA_GRE_OKEY, p->o_key) || nla_put_in_addr(skb, IFLA_GRE_LOCAL, p->iph.saddr) || -- cgit v1.2.3 From c706863bc8902d0c2d1a5a27ac8e1ead5d06b79d Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Mon, 28 Jan 2019 22:23:49 +0100 Subject: net: ip6_gre: always reports o_key to userspace As Erspan_v4, Erspan_v6 protocol relies on o_key to configure session id header field. However TUNNEL_KEY bit is cleared in ip6erspan_tunnel_xmit since ERSPAN protocol does not set the key field of the external GRE header and so the configured o_key is not reported to userspace. The issue can be triggered with the following reproducer: $ip link add ip6erspan1 type ip6erspan local 2000::1 remote 2000::2 \ key 1 seq erspan_ver 1 $ip link set ip6erspan1 up ip -d link sh ip6erspan1 ip6erspan1@NONE: mtu 1422 qdisc noop state DOWN mode DEFAULT link/ether ba:ff:09:24:c3:0e brd ff:ff:ff:ff:ff:ff promiscuity 0 minmtu 68 maxmtu 1500 ip6erspan remote 2000::2 local 2000::1 encaplimit 4 flowlabel 0x00000 ikey 0.0.0.1 iseq oseq Fix the issue adding TUNNEL_KEY bit to the o_flags parameter in ip6gre_fill_info Fixes: 5a963eb61b7c ("ip6_gre: Add ERSPAN native tunnel support") Signed-off-by: Lorenzo Bianconi Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 4416368dbd49..801a9a0c217e 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -2098,12 +2098,17 @@ static int ip6gre_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); struct __ip6_tnl_parm *p = &t->parms; + __be16 o_flags = p->o_flags; + + if ((p->erspan_ver == 1 || p->erspan_ver == 2) && + !p->collect_md) + o_flags |= TUNNEL_KEY; if (nla_put_u32(skb, IFLA_GRE_LINK, p->link) || nla_put_be16(skb, IFLA_GRE_IFLAGS, gre_tnl_flags_to_gre_flags(p->i_flags)) || nla_put_be16(skb, IFLA_GRE_OFLAGS, - gre_tnl_flags_to_gre_flags(p->o_flags)) || + gre_tnl_flags_to_gre_flags(o_flags)) || nla_put_be32(skb, IFLA_GRE_IKEY, p->i_key) || nla_put_be32(skb, IFLA_GRE_OKEY, p->o_key) || nla_put_in6_addr(skb, IFLA_GRE_LOCAL, &p->laddr) || -- cgit v1.2.3 From ef489749aae508e6f17886775c075f12ff919fb1 Mon Sep 17 00:00:00 2001 From: Yohei Kanemaru Date: Tue, 29 Jan 2019 15:52:34 +0900 Subject: ipv6: sr: clear IP6CB(skb) on SRH ip4ip6 encapsulation skb->cb may contain data from previous layers (in an observed case IPv4 with L3 Master Device). In the observed scenario, the data in IPCB(skb)->frags was misinterpreted as IP6CB(skb)->frag_max_size, eventually caused an unexpected IPv6 fragmentation in ip6_fragment() through ip6_finish_output(). This patch clears IP6CB(skb), which potentially contains garbage data, on the SRH ip4ip6 encapsulation. Fixes: 32d99d0b6702 ("ipv6: sr: add support for ip4ip6 encapsulation") Signed-off-by: Yohei Kanemaru Signed-off-by: David S. Miller --- net/ipv6/seg6_iptunnel.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 8181ee7e1e27..ee5403cbe655 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -146,6 +146,8 @@ int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) } else { ip6_flow_hdr(hdr, 0, flowlabel); hdr->hop_limit = ip6_dst_hoplimit(skb_dst(skb)); + + memset(IP6CB(skb), 0, sizeof(*IP6CB(skb))); } hdr->nexthdr = NEXTHDR_ROUTING; -- cgit v1.2.3 From 4522a70db7aa5e77526a4079628578599821b193 Mon Sep 17 00:00:00 2001 From: Jacob Wen Date: Wed, 30 Jan 2019 14:55:14 +0800 Subject: l2tp: fix reading optional fields of L2TPv3 Use pskb_may_pull() to make sure the optional fields are in skb linear parts, so we can safely read them later. It's easy to reproduce the issue with a net driver that supports paged skb data. Just create a L2TPv3 over IP tunnel and then generates some network traffic. Once reproduced, rx err in /sys/kernel/debug/l2tp/tunnels will increase. Changes in v4: 1. s/l2tp_v3_pull_opt/l2tp_v3_ensure_opt_in_linear/ 2. s/tunnel->version != L2TP_HDR_VER_2/tunnel->version == L2TP_HDR_VER_3/ 3. Add 'Fixes' in commit messages. Changes in v3: 1. To keep consistency, move the code out of l2tp_recv_common. 2. Use "net" instead of "net-next", since this is a bug fix. Changes in v2: 1. Only fix L2TPv3 to make code simple. To fix both L2TPv3 and L2TPv2, we'd better refactor l2tp_recv_common. It's complicated to do so. 2. Reloading pointers after pskb_may_pull Fixes: f7faffa3ff8e ("l2tp: Add L2TPv3 protocol support") Fixes: 0d76751fad77 ("l2tp: Add L2TPv3 IP encapsulation (no UDP) support") Fixes: a32e0eec7042 ("l2tp: introduce L2TPv3 IP encapsulation support for IPv6") Signed-off-by: Jacob Wen Acked-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 4 ++++ net/l2tp/l2tp_core.h | 20 ++++++++++++++++++++ net/l2tp/l2tp_ip.c | 3 +++ net/l2tp/l2tp_ip6.c | 3 +++ 4 files changed, 30 insertions(+) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 26f1d435696a..dd5ba0c11ab3 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -884,6 +884,10 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) goto error; } + if (tunnel->version == L2TP_HDR_VER_3 && + l2tp_v3_ensure_opt_in_linear(session, skb, &ptr, &optr)) + goto error; + l2tp_recv_common(session, skb, ptr, optr, hdrflags, length); l2tp_session_dec_refcount(session); diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h index 9c9afe94d389..b2ce90260c35 100644 --- a/net/l2tp/l2tp_core.h +++ b/net/l2tp/l2tp_core.h @@ -301,6 +301,26 @@ static inline bool l2tp_tunnel_uses_xfrm(const struct l2tp_tunnel *tunnel) } #endif +static inline int l2tp_v3_ensure_opt_in_linear(struct l2tp_session *session, struct sk_buff *skb, + unsigned char **ptr, unsigned char **optr) +{ + int opt_len = session->peer_cookie_len + l2tp_get_l2specific_len(session); + + if (opt_len > 0) { + int off = *ptr - *optr; + + if (!pskb_may_pull(skb, off + opt_len)) + return -1; + + if (skb->data != *optr) { + *optr = skb->data; + *ptr = skb->data + off; + } + } + + return 0; +} + #define l2tp_printk(ptr, type, func, fmt, ...) \ do { \ if (((ptr)->debug) & (type)) \ diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 35f6f86d4dcc..d4c60523c549 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -165,6 +165,9 @@ static int l2tp_ip_recv(struct sk_buff *skb) print_hex_dump_bytes("", DUMP_PREFIX_OFFSET, ptr, length); } + if (l2tp_v3_ensure_opt_in_linear(session, skb, &ptr, &optr)) + goto discard_sess; + l2tp_recv_common(session, skb, ptr, optr, 0, skb->len); l2tp_session_dec_refcount(session); diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index 237f1a4a0b0c..0ae6899edac0 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -178,6 +178,9 @@ static int l2tp_ip6_recv(struct sk_buff *skb) print_hex_dump_bytes("", DUMP_PREFIX_OFFSET, ptr, length); } + if (l2tp_v3_ensure_opt_in_linear(session, skb, &ptr, &optr)) + goto discard_sess; + l2tp_recv_common(session, skb, ptr, optr, 0, skb->len); l2tp_session_dec_refcount(session); -- cgit v1.2.3 From 91c524708de6207f59dd3512518d8a1c7b434ee3 Mon Sep 17 00:00:00 2001 From: Jacob Wen Date: Thu, 31 Jan 2019 15:18:56 +0800 Subject: l2tp: copy 4 more bytes to linear part if necessary The size of L2TPv2 header with all optional fields is 14 bytes. l2tp_udp_recv_core only moves 10 bytes to the linear part of a skb. This may lead to l2tp_recv_common read data outside of a skb. This patch make sure that there is at least 14 bytes in the linear part of a skb to meet the maximum need of l2tp_udp_recv_core and l2tp_recv_common. The minimum size of both PPP HDLC-like frame and Ethernet frame is larger than 14 bytes, so we are safe to do so. Also remove L2TP_HDR_SIZE_NOSEQ, it is unused now. Fixes: fd558d186df2 ("l2tp: Split pppol2tp patch into separate l2tp and ppp parts") Suggested-by: Guillaume Nault Signed-off-by: Jacob Wen Acked-by: Guillaume Nault Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index dd5ba0c11ab3..fed6becc5daf 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -83,8 +83,7 @@ #define L2TP_SLFLAG_S 0x40000000 #define L2TP_SL_SEQ_MASK 0x00ffffff -#define L2TP_HDR_SIZE_SEQ 10 -#define L2TP_HDR_SIZE_NOSEQ 6 +#define L2TP_HDR_SIZE_MAX 14 /* Default trace flags */ #define L2TP_DEFAULT_DEBUG_FLAGS 0 @@ -808,7 +807,7 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) __skb_pull(skb, sizeof(struct udphdr)); /* Short packet? */ - if (!pskb_may_pull(skb, L2TP_HDR_SIZE_SEQ)) { + if (!pskb_may_pull(skb, L2TP_HDR_SIZE_MAX)) { l2tp_info(tunnel, L2TP_MSG_DATA, "%s: recv short packet (len=%d)\n", tunnel->name, skb->len); -- cgit v1.2.3 From 6fa19f5637a6c22bc0999596bcc83bdcac8a4fa6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 31 Jan 2019 08:47:10 -0800 Subject: rds: fix refcount bug in rds_sock_addref syzbot was able to catch a bug in rds [1] The issue here is that the socket might be found in a hash table but that its refcount has already be set to 0 by another cpu. We need to use refcount_inc_not_zero() to be safe here. [1] refcount_t: increment on 0; use-after-free. WARNING: CPU: 1 PID: 23129 at lib/refcount.c:153 refcount_inc_checked lib/refcount.c:153 [inline] WARNING: CPU: 1 PID: 23129 at lib/refcount.c:153 refcount_inc_checked+0x61/0x70 lib/refcount.c:151 Kernel panic - not syncing: panic_on_warn set ... CPU: 1 PID: 23129 Comm: syz-executor3 Not tainted 5.0.0-rc4+ #53 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+0x1db/0x2d0 lib/dump_stack.c:113 panic+0x2cb/0x65c kernel/panic.c:214 __warn.cold+0x20/0x48 kernel/panic.c:571 report_bug+0x263/0x2b0 lib/bug.c:186 fixup_bug arch/x86/kernel/traps.c:178 [inline] fixup_bug arch/x86/kernel/traps.c:173 [inline] do_error_trap+0x11b/0x200 arch/x86/kernel/traps.c:271 do_invalid_op+0x37/0x50 arch/x86/kernel/traps.c:290 invalid_op+0x14/0x20 arch/x86/entry/entry_64.S:973 RIP: 0010:refcount_inc_checked lib/refcount.c:153 [inline] RIP: 0010:refcount_inc_checked+0x61/0x70 lib/refcount.c:151 Code: 1d 51 63 c8 06 31 ff 89 de e8 eb 1b f2 fd 84 db 75 dd e8 a2 1a f2 fd 48 c7 c7 60 9f 81 88 c6 05 31 63 c8 06 01 e8 af 65 bb fd <0f> 0b eb c1 90 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 54 49 RSP: 0018:ffff8880a0cbf1e8 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc90006113000 RDX: 000000000001047d RSI: ffffffff81685776 RDI: 0000000000000005 RBP: ffff8880a0cbf1f8 R08: ffff888097c9e100 R09: ffffed1015ce5021 R10: ffffed1015ce5020 R11: ffff8880ae728107 R12: ffff8880723c20c0 R13: ffff8880723c24b0 R14: dffffc0000000000 R15: ffffed1014197e64 sock_hold include/net/sock.h:647 [inline] rds_sock_addref+0x19/0x20 net/rds/af_rds.c:675 rds_find_bound+0x97c/0x1080 net/rds/bind.c:82 rds_recv_incoming+0x3be/0x1430 net/rds/recv.c:362 rds_loop_xmit+0xf3/0x2a0 net/rds/loop.c:96 rds_send_xmit+0x1355/0x2a10 net/rds/send.c:355 rds_sendmsg+0x323c/0x44e0 net/rds/send.c:1368 sock_sendmsg_nosec net/socket.c:621 [inline] sock_sendmsg+0xdd/0x130 net/socket.c:631 __sys_sendto+0x387/0x5f0 net/socket.c:1788 __do_sys_sendto net/socket.c:1800 [inline] __se_sys_sendto net/socket.c:1796 [inline] __x64_sys_sendto+0xe1/0x1a0 net/socket.c:1796 do_syscall_64+0x1a3/0x800 arch/x86/entry/common.c:290 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x458089 Code: 6d b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 3b b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007fc266df8c78 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 0000000000000006 RCX: 0000000000458089 RDX: 0000000000000000 RSI: 00000000204b3fff RDI: 0000000000000005 RBP: 000000000073bf00 R08: 00000000202b4000 R09: 0000000000000010 R10: 0000000000000000 R11: 0000000000000246 R12: 00007fc266df96d4 R13: 00000000004c56e4 R14: 00000000004d94a8 R15: 00000000ffffffff Fixes: cc4dfb7f70a3 ("rds: fix two RCU related problems") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Sowmini Varadhan Cc: Santosh Shilimkar Cc: rds-devel@oss.oracle.com Cc: Cong Wang Acked-by: Santosh Shilimkar Signed-off-by: David S. Miller --- net/rds/bind.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/rds/bind.c b/net/rds/bind.c index 762d2c6788a3..17c9d9f0c848 100644 --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -78,10 +78,10 @@ struct rds_sock *rds_find_bound(const struct in6_addr *addr, __be16 port, __rds_create_bind_key(key, addr, port, scope_id); rcu_read_lock(); rs = rhashtable_lookup(&bind_hash_table, key, ht_parms); - if (rs && !sock_flag(rds_rs_to_sk(rs), SOCK_DEAD)) - rds_sock_addref(rs); - else + if (rs && (sock_flag(rds_rs_to_sk(rs), SOCK_DEAD) || + !refcount_inc_not_zero(&rds_rs_to_sk(rs)->sk_refcnt))) rs = NULL; + rcu_read_unlock(); rdsdebug("returning rs %p for %pI6c:%u\n", rs, addr, -- cgit v1.2.3 From 9d0f50b80222dc273e67e4e14410fcfa4130a90c Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 29 Jan 2019 11:10:57 +0100 Subject: mac80211: ensure that mgmt tx skbs have tailroom for encryption Some drivers use IEEE80211_KEY_FLAG_SW_MGMT_TX to indicate that management frames need to be software encrypted. Since normal data packets are still encrypted by the hardware, crypto_tx_tailroom_needed_cnt gets decremented after key upload to hw. This can lead to passing skbs to ccmp_encrypt_skb, which don't have the necessary tailroom for software encryption. Change the code to add tailroom for encrypted management packets, even if crypto_tx_tailroom_needed_cnt is 0. Cc: stable@vger.kernel.org Signed-off-by: Felix Fietkau Signed-off-by: Johannes Berg --- net/mac80211/tx.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index f170d6c6629a..928f13a208b0 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1938,9 +1938,16 @@ static int ieee80211_skb_resize(struct ieee80211_sub_if_data *sdata, int head_need, bool may_encrypt) { struct ieee80211_local *local = sdata->local; + struct ieee80211_hdr *hdr; + bool enc_tailroom; int tail_need = 0; - if (may_encrypt && sdata->crypto_tx_tailroom_needed_cnt) { + hdr = (struct ieee80211_hdr *) skb->data; + enc_tailroom = may_encrypt && + (sdata->crypto_tx_tailroom_needed_cnt || + ieee80211_is_mgmt(hdr->frame_control)); + + if (enc_tailroom) { tail_need = IEEE80211_ENCRYPT_TAILROOM; tail_need -= skb_tailroom(skb); tail_need = max_t(int, tail_need, 0); @@ -1948,8 +1955,7 @@ static int ieee80211_skb_resize(struct ieee80211_sub_if_data *sdata, if (skb_cloned(skb) && (!ieee80211_hw_check(&local->hw, SUPPORTS_CLONED_SKBS) || - !skb_clone_writable(skb, ETH_HLEN) || - (may_encrypt && sdata->crypto_tx_tailroom_needed_cnt))) + !skb_clone_writable(skb, ETH_HLEN) || enc_tailroom)) I802_DEBUG_INC(local->tx_expand_skb_head_cloned); else if (head_need || tail_need) I802_DEBUG_INC(local->tx_expand_skb_head); -- cgit v1.2.3 From e005bd7ddea06784c1eb91ac5bb6b171a94f3b05 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 1 Feb 2019 11:09:54 +0100 Subject: cfg80211: call disconnect_wk when AP stops Since we now prevent regulatory restore during STA disconnect if concurrent AP interfaces are active, we need to reschedule this check when the AP state changes. This fixes never doing a restore when an AP is the last interface to stop. Or to put it another way: we need to re-check after anything we check here changes. Cc: stable@vger.kernel.org Fixes: 113f3aaa81bd ("cfg80211: Prevent regulatory restore during STA disconnect in concurrent interfaces") Signed-off-by: Johannes Berg --- net/wireless/ap.c | 2 ++ net/wireless/core.h | 2 ++ net/wireless/sme.c | 2 +- 3 files changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/ap.c b/net/wireless/ap.c index 882d97bdc6bf..550ac9d827fe 100644 --- a/net/wireless/ap.c +++ b/net/wireless/ap.c @@ -41,6 +41,8 @@ int __cfg80211_stop_ap(struct cfg80211_registered_device *rdev, cfg80211_sched_dfs_chan_update(rdev); } + schedule_work(&cfg80211_disconnect_work); + return err; } diff --git a/net/wireless/core.h b/net/wireless/core.h index c5d6f3418601..f6b40563dc63 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -445,6 +445,8 @@ void cfg80211_process_wdev_events(struct wireless_dev *wdev); bool cfg80211_does_bw_fit_range(const struct ieee80211_freq_range *freq_range, u32 center_freq_khz, u32 bw_khz); +extern struct work_struct cfg80211_disconnect_work; + /** * cfg80211_chandef_dfs_usable - checks if chandef is DFS usable * @wiphy: the wiphy to validate against diff --git a/net/wireless/sme.c b/net/wireless/sme.c index f741d8376a46..7d34cb884840 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -667,7 +667,7 @@ static void disconnect_work(struct work_struct *work) rtnl_unlock(); } -static DECLARE_WORK(cfg80211_disconnect_work, disconnect_work); +DECLARE_WORK(cfg80211_disconnect_work, disconnect_work); /* -- cgit v1.2.3 From ba59fb0273076637f0add4311faa990a5eec27c0 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 1 Feb 2019 15:15:22 +0100 Subject: sctp: walk the list of asoc safely In sctp_sendmesg(), when walking the list of endpoint associations, the association can be dropped from the list, making the list corrupt. Properly handle this by using list_for_each_entry_safe() Fixes: 4910280503f3 ("sctp: add support for snd flag SCTP_SENDALL process in sendmsg") Reported-by: Secunia Research Tested-by: Secunia Research Signed-off-by: Greg Kroah-Hartman Acked-by: Marcelo Ricardo Leitner Acked-by: Neil Horman Signed-off-by: David S. Miller --- net/sctp/socket.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index f93c3cf9e567..65d6d04546ae 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2027,7 +2027,7 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_transport *transport = NULL; struct sctp_sndrcvinfo _sinfo, *sinfo; - struct sctp_association *asoc; + struct sctp_association *asoc, *tmp; struct sctp_cmsgs cmsgs; union sctp_addr *daddr; bool new = false; @@ -2053,7 +2053,7 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) /* SCTP_SENDALL process */ if ((sflags & SCTP_SENDALL) && sctp_style(sk, UDP)) { - list_for_each_entry(asoc, &ep->asocs, asocs) { + list_for_each_entry_safe(asoc, tmp, &ep->asocs, asocs) { err = sctp_sendmsg_check_sflags(asoc, sflags, msg, msg_len); if (err == 0) -- cgit v1.2.3 From 14d22d4d61e40623a7c5816728bfe55c322e779a Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Wed, 30 Jan 2019 18:51:00 +0100 Subject: net/smc: fix another sizeof to int comparison Comparing an int to a size, which is unsigned, causes the int to become unsigned, giving the wrong result. kernel_sendmsg can return a negative error code. Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_clc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/smc_clc.c b/net/smc/smc_clc.c index 776e9dfc915d..d53fd588d1f5 100644 --- a/net/smc/smc_clc.c +++ b/net/smc/smc_clc.c @@ -378,7 +378,7 @@ int smc_clc_send_decline(struct smc_sock *smc, u32 peer_diag_info) vec.iov_len = sizeof(struct smc_clc_msg_decline); len = kernel_sendmsg(smc->clcsock, &msg, &vec, 1, sizeof(struct smc_clc_msg_decline)); - if (len < sizeof(struct smc_clc_msg_decline)) + if (len < 0 || len < sizeof(struct smc_clc_msg_decline)) len = -EPROTO; return len > 0 ? 0 : len; } -- cgit v1.2.3 From ca8dc1334a71d6081b09e18d55198a27e28fd44b Mon Sep 17 00:00:00 2001 From: Hans Wippel Date: Wed, 30 Jan 2019 18:51:01 +0100 Subject: net/smc: allow 16 byte pnetids in netlink policy Currently, users can only send pnetids with a maximum length of 15 bytes over the SMC netlink interface although the maximum pnetid length is 16 bytes. This patch changes the SMC netlink policy to accept 16 byte pnetids. Signed-off-by: Hans Wippel Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_pnet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/smc_pnet.c b/net/smc/smc_pnet.c index 7cb3e4f07c10..632c3109dee5 100644 --- a/net/smc/smc_pnet.c +++ b/net/smc/smc_pnet.c @@ -27,7 +27,7 @@ static struct nla_policy smc_pnet_policy[SMC_PNETID_MAX + 1] = { [SMC_PNETID_NAME] = { .type = NLA_NUL_STRING, - .len = SMC_MAX_PNETID_LEN - 1 + .len = SMC_MAX_PNETID_LEN }, [SMC_PNETID_ETHNAME] = { .type = NLA_NUL_STRING, -- cgit v1.2.3 From 77f838ace755d2f466536c44dac6c856f62cd901 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:02 +0100 Subject: net/smc: prevent races between smc_lgr_terminate() and smc_conn_free() To prevent races between smc_lgr_terminate() and smc_conn_free() add an extra check of the lgr field before accessing it, and cancel a delayed free_work when a new smc connection is created. This fixes the problem that free_work cleared the lgr variable but smc_lgr_terminate() or smc_conn_free() still access it in parallel. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_core.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 35c1cdc93e1c..097c798983ca 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -128,6 +128,8 @@ static void smc_lgr_unregister_conn(struct smc_connection *conn) { struct smc_link_group *lgr = conn->lgr; + if (!lgr) + return; write_lock_bh(&lgr->conns_lock); if (conn->alert_token_local) { __smc_lgr_unregister_conn(conn); @@ -628,6 +630,8 @@ int smc_conn_create(struct smc_sock *smc, bool is_smcd, int srv_first_contact, local_contact = SMC_REUSE_CONTACT; conn->lgr = lgr; smc_lgr_register_conn(conn); /* add smc conn to lgr */ + if (delayed_work_pending(&lgr->free_work)) + cancel_delayed_work(&lgr->free_work); write_unlock_bh(&lgr->conns_lock); break; } -- cgit v1.2.3 From 6889b36da78a21a312d8b462c1fa25a03c2ff192 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:03 +0100 Subject: net/smc: don't wait for send buffer space when data was already sent When there is no more send buffer space and at least 1 byte was already sent then return to user space. The wait is only done when no data was sent by the sendmsg() call. This fixes smc_tx_sendmsg() which tried to always send all user data and started to wait for free send buffer space when needed. During this wait the user space program was blocked in the sendmsg() call and hence not able to receive incoming data. When both sides were in such a situation then the connection stalled forever. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_tx.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index d8366ed51757..f99951f3f7fd 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -165,12 +165,11 @@ int smc_tx_sendmsg(struct smc_sock *smc, struct msghdr *msg, size_t len) conn->local_tx_ctrl.prod_flags.urg_data_pending = 1; if (!atomic_read(&conn->sndbuf_space) || conn->urg_tx_pend) { + if (send_done) + return send_done; rc = smc_tx_wait(smc, msg->msg_flags); - if (rc) { - if (send_done) - return send_done; + if (rc) goto out_err; - } continue; } -- cgit v1.2.3 From 51c5aba3b672c4285fca052817f34b22dc79dda7 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:04 +0100 Subject: net/smc: recvmsg and splice_read should return 0 after shutdown When a socket was connected and is now shut down for read, return 0 to indicate end of data in recvmsg and splice_read (like TCP) and do not return ENOTCONN. This behavior is required by the socket api. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/af_smc.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index c4e56602e0c6..b04a813fc865 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1505,6 +1505,11 @@ static int smc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, smc = smc_sk(sk); lock_sock(sk); + if (sk->sk_state == SMC_CLOSED && (sk->sk_shutdown & RCV_SHUTDOWN)) { + /* socket was connected before, no more data to read */ + rc = 0; + goto out; + } if ((sk->sk_state == SMC_INIT) || (sk->sk_state == SMC_LISTEN) || (sk->sk_state == SMC_CLOSED)) @@ -1840,7 +1845,11 @@ static ssize_t smc_splice_read(struct socket *sock, loff_t *ppos, smc = smc_sk(sk); lock_sock(sk); - + if (sk->sk_state == SMC_CLOSED && (sk->sk_shutdown & RCV_SHUTDOWN)) { + /* socket was connected before, no more data to read */ + rc = 0; + goto out; + } if (sk->sk_state == SMC_INIT || sk->sk_state == SMC_LISTEN || sk->sk_state == SMC_CLOSED) -- cgit v1.2.3 From 33f3fcc290671590821ff3c0c9396db1ec9b7d4c Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:05 +0100 Subject: net/smc: do not wait under send_lock smc_cdc_get_free_slot() might wait for free transfer buffers when using SMC-R. This wait should not be done under the send_lock, which is a spin_lock. This fixes a cpu loop in parallel threads waiting for the send_lock. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_tx.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index f99951f3f7fd..36af3de731b9 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -488,25 +488,23 @@ static int smcr_tx_sndbuf_nonempty(struct smc_connection *conn) struct smc_wr_buf *wr_buf; int rc; - spin_lock_bh(&conn->send_lock); rc = smc_cdc_get_free_slot(conn, &wr_buf, &pend); if (rc < 0) { if (rc == -EBUSY) { struct smc_sock *smc = container_of(conn, struct smc_sock, conn); - if (smc->sk.sk_err == ECONNABORTED) { - rc = sock_error(&smc->sk); - goto out_unlock; - } + if (smc->sk.sk_err == ECONNABORTED) + return sock_error(&smc->sk); rc = 0; if (conn->alert_token_local) /* connection healthy */ mod_delayed_work(system_wq, &conn->tx_work, SMC_TX_WORK_DELAY); } - goto out_unlock; + return rc; } + spin_lock_bh(&conn->send_lock); if (!conn->local_tx_ctrl.prod_flags.urg_data_present) { rc = smc_tx_rdma_writes(conn); if (rc) { -- cgit v1.2.3 From 2dee25af42f99b476d5916aeac5c994e4dcc910b Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:06 +0100 Subject: net/smc: call smc_cdc_msg_send() under send_lock Call smc_cdc_msg_send() under the connection send_lock to make sure all send operations for one connection are serialized. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_cdc.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c index db83332ac1c8..1c5333d494e9 100644 --- a/net/smc/smc_cdc.c +++ b/net/smc/smc_cdc.c @@ -125,7 +125,10 @@ static int smcr_cdc_get_slot_and_msg_send(struct smc_connection *conn) if (rc) return rc; - return smc_cdc_msg_send(conn, wr_buf, pend); + spin_lock_bh(&conn->send_lock); + rc = smc_cdc_msg_send(conn, wr_buf, pend); + spin_unlock_bh(&conn->send_lock); + return rc; } int smc_cdc_get_slot_and_msg_send(struct smc_connection *conn) -- cgit v1.2.3 From e5f3aa04dbfd92154f21edfeb7a1676a45918928 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:07 +0100 Subject: net/smc: use device link provided in qp_context The device field of the IB event structure does not always point to the SMC IB device. Load the pointer from the qp_context which is always provided to smc_ib_qp_event_handler() in the priv field. And for qp events the affected port is given in the qp structure of the ibevent, derive it from there. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_ib.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index e519ef29c0ff..76487a16934e 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -289,8 +289,8 @@ int smc_ib_create_protection_domain(struct smc_link *lnk) static void smc_ib_qp_event_handler(struct ib_event *ibevent, void *priv) { - struct smc_ib_device *smcibdev = - (struct smc_ib_device *)ibevent->device; + struct smc_link *lnk = (struct smc_link *)priv; + struct smc_ib_device *smcibdev = lnk->smcibdev; u8 port_idx; switch (ibevent->event) { @@ -298,7 +298,7 @@ static void smc_ib_qp_event_handler(struct ib_event *ibevent, void *priv) case IB_EVENT_GID_CHANGE: case IB_EVENT_PORT_ERR: case IB_EVENT_QP_ACCESS_ERR: - port_idx = ibevent->element.port_num - 1; + port_idx = ibevent->element.qp->port - 1; set_bit(port_idx, &smcibdev->port_event_mask); schedule_work(&smcibdev->port_event_work); break; -- cgit v1.2.3 From 46ad02229d6b4ee276eb295ca08f039993f323c8 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Wed, 30 Jan 2019 18:51:08 +0100 Subject: net/smc: fix use of variable in cleared area Do not use pend->idx as index for the arrays because its value is located in the cleared area. Use the existing local variable instead. Without this fix the wrong area might be cleared. Signed-off-by: Karsten Graul Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_wr.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index c2694750a6a8..1dc88c32d6bb 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -218,10 +218,10 @@ int smc_wr_tx_put_slot(struct smc_link *link, u32 idx = pend->idx; /* clear the full struct smc_wr_tx_pend including .priv */ - memset(&link->wr_tx_pends[pend->idx], 0, - sizeof(link->wr_tx_pends[pend->idx])); - memset(&link->wr_tx_bufs[pend->idx], 0, - sizeof(link->wr_tx_bufs[pend->idx])); + memset(&link->wr_tx_pends[idx], 0, + sizeof(link->wr_tx_pends[idx])); + memset(&link->wr_tx_bufs[idx], 0, + sizeof(link->wr_tx_bufs[idx])); test_and_clear_bit(idx, link->wr_tx_mask); return 1; } -- cgit v1.2.3 From 9b1f19d810e92d6cdc68455fbc22d9f961a58ce1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 30 Jan 2019 11:39:41 -0800 Subject: dccp: fool proof ccid_hc_[rt]x_parse_options() Similarly to commit 276bdb82dedb ("dccp: check ccid before dereferencing") it is wise to test for a NULL ccid. kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 1 PID: 16 Comm: ksoftirqd/1 Not tainted 5.0.0-rc3+ #37 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:ccid_hc_tx_parse_options net/dccp/ccid.h:205 [inline] RIP: 0010:dccp_parse_options+0x8d9/0x12b0 net/dccp/options.c:233 Code: c5 0f b6 75 b3 80 38 00 0f 85 d6 08 00 00 48 b9 00 00 00 00 00 fc ff df 48 8b 45 b8 4c 8b b8 f8 07 00 00 4c 89 f8 48 c1 e8 03 <80> 3c 08 00 0f 85 95 08 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8b kobject: 'loop5' (0000000080f78fc1): kobject_uevent_env RSP: 0018:ffff8880a94df0b8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff8880858ac723 RCX: dffffc0000000000 RDX: 0000000000000100 RSI: 0000000000000007 RDI: 0000000000000001 RBP: ffff8880a94df140 R08: 0000000000000001 R09: ffff888061b83a80 R10: ffffed100c370752 R11: ffff888061b83a97 R12: 0000000000000026 R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff8880ae700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f0defa33518 CR3: 000000008db5e000 CR4: 00000000001406e0 kobject: 'loop5' (0000000080f78fc1): fill_kobj_path: path = '/devices/virtual/block/loop5' DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: dccp_rcv_state_process+0x2b6/0x1af6 net/dccp/input.c:654 dccp_v4_do_rcv+0x100/0x190 net/dccp/ipv4.c:688 sk_backlog_rcv include/net/sock.h:936 [inline] __sk_receive_skb+0x3a9/0xea0 net/core/sock.c:473 dccp_v4_rcv+0x10cb/0x1f80 net/dccp/ipv4.c:880 ip_protocol_deliver_rcu+0xb6/0xa20 net/ipv4/ip_input.c:208 ip_local_deliver_finish+0x23b/0x390 net/ipv4/ip_input.c:234 NF_HOOK include/linux/netfilter.h:289 [inline] NF_HOOK include/linux/netfilter.h:283 [inline] ip_local_deliver+0x1f0/0x740 net/ipv4/ip_input.c:255 dst_input include/net/dst.h:450 [inline] ip_rcv_finish+0x1f4/0x2f0 net/ipv4/ip_input.c:414 NF_HOOK include/linux/netfilter.h:289 [inline] NF_HOOK include/linux/netfilter.h:283 [inline] ip_rcv+0xed/0x620 net/ipv4/ip_input.c:524 __netif_receive_skb_one_core+0x160/0x210 net/core/dev.c:4973 __netif_receive_skb+0x2c/0x1c0 net/core/dev.c:5083 process_backlog+0x206/0x750 net/core/dev.c:5923 napi_poll net/core/dev.c:6346 [inline] net_rx_action+0x76d/0x1930 net/core/dev.c:6412 __do_softirq+0x30b/0xb11 kernel/softirq.c:292 run_ksoftirqd kernel/softirq.c:654 [inline] run_ksoftirqd+0x8e/0x110 kernel/softirq.c:646 smpboot_thread_fn+0x6ab/0xa10 kernel/smpboot.c:164 kthread+0x357/0x430 kernel/kthread.c:246 ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:352 Modules linked in: ---[ end trace 58a0ba03bea2c376 ]--- RIP: 0010:ccid_hc_tx_parse_options net/dccp/ccid.h:205 [inline] RIP: 0010:dccp_parse_options+0x8d9/0x12b0 net/dccp/options.c:233 Code: c5 0f b6 75 b3 80 38 00 0f 85 d6 08 00 00 48 b9 00 00 00 00 00 fc ff df 48 8b 45 b8 4c 8b b8 f8 07 00 00 4c 89 f8 48 c1 e8 03 <80> 3c 08 00 0f 85 95 08 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8b RSP: 0018:ffff8880a94df0b8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff8880858ac723 RCX: dffffc0000000000 RDX: 0000000000000100 RSI: 0000000000000007 RDI: 0000000000000001 RBP: ffff8880a94df140 R08: 0000000000000001 R09: ffff888061b83a80 R10: ffffed100c370752 R11: ffff888061b83a97 R12: 0000000000000026 R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff8880ae700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f0defa33518 CR3: 0000000009871000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/ccid.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index 6eb837a47b5c..baaaeb2b2c42 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -202,7 +202,7 @@ static inline void ccid_hc_tx_packet_recv(struct ccid *ccid, struct sock *sk, static inline int ccid_hc_tx_parse_options(struct ccid *ccid, struct sock *sk, u8 pkt, u8 opt, u8 *val, u8 len) { - if (ccid->ccid_ops->ccid_hc_tx_parse_options == NULL) + if (!ccid || !ccid->ccid_ops->ccid_hc_tx_parse_options) return 0; return ccid->ccid_ops->ccid_hc_tx_parse_options(sk, pkt, opt, val, len); } @@ -214,7 +214,7 @@ static inline int ccid_hc_tx_parse_options(struct ccid *ccid, struct sock *sk, static inline int ccid_hc_rx_parse_options(struct ccid *ccid, struct sock *sk, u8 pkt, u8 opt, u8 *val, u8 len) { - if (ccid->ccid_ops->ccid_hc_rx_parse_options == NULL) + if (!ccid || !ccid->ccid_ops->ccid_hc_rx_parse_options) return 0; return ccid->ccid_ops->ccid_hc_rx_parse_options(sk, pkt, opt, val, len); } -- cgit v1.2.3 From 22b5c0b63f32568e130fa2df4ba23efce3eb495b Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Fri, 1 Feb 2019 12:42:06 +0100 Subject: vsock/virtio: fix kernel panic after device hot-unplug virtio_vsock_remove() invokes the vsock_core_exit() also if there are opened sockets for the AF_VSOCK protocol family. In this way the vsock "transport" pointer is set to NULL, triggering the kernel panic at the first socket activity. This patch move the vsock_core_init()/vsock_core_exit() in the virtio_vsock respectively in module_init and module_exit functions, that cannot be invoked until there are open sockets. Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1609699 Reported-by: Yan Fu Signed-off-by: Stefano Garzarella Acked-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- net/vmw_vsock/virtio_transport.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c index 5d3cce9e8744..9dae54698737 100644 --- a/net/vmw_vsock/virtio_transport.c +++ b/net/vmw_vsock/virtio_transport.c @@ -75,6 +75,9 @@ static u32 virtio_transport_get_local_cid(void) { struct virtio_vsock *vsock = virtio_vsock_get(); + if (!vsock) + return VMADDR_CID_ANY; + return vsock->guest_cid; } @@ -584,10 +587,6 @@ static int virtio_vsock_probe(struct virtio_device *vdev) virtio_vsock_update_guest_cid(vsock); - ret = vsock_core_init(&virtio_transport.transport); - if (ret < 0) - goto out_vqs; - vsock->rx_buf_nr = 0; vsock->rx_buf_max_nr = 0; atomic_set(&vsock->queued_replies, 0); @@ -618,8 +617,6 @@ static int virtio_vsock_probe(struct virtio_device *vdev) mutex_unlock(&the_virtio_vsock_mutex); return 0; -out_vqs: - vsock->vdev->config->del_vqs(vsock->vdev); out: kfree(vsock); mutex_unlock(&the_virtio_vsock_mutex); @@ -669,7 +666,6 @@ static void virtio_vsock_remove(struct virtio_device *vdev) mutex_lock(&the_virtio_vsock_mutex); the_virtio_vsock = NULL; - vsock_core_exit(); mutex_unlock(&the_virtio_vsock_mutex); vdev->config->del_vqs(vdev); @@ -702,14 +698,28 @@ static int __init virtio_vsock_init(void) virtio_vsock_workqueue = alloc_workqueue("virtio_vsock", 0, 0); if (!virtio_vsock_workqueue) return -ENOMEM; + ret = register_virtio_driver(&virtio_vsock_driver); if (ret) - destroy_workqueue(virtio_vsock_workqueue); + goto out_wq; + + ret = vsock_core_init(&virtio_transport.transport); + if (ret) + goto out_vdr; + + return 0; + +out_vdr: + unregister_virtio_driver(&virtio_vsock_driver); +out_wq: + destroy_workqueue(virtio_vsock_workqueue); return ret; + } static void __exit virtio_vsock_exit(void) { + vsock_core_exit(); unregister_virtio_driver(&virtio_vsock_driver); destroy_workqueue(virtio_vsock_workqueue); } -- cgit v1.2.3 From 85965487abc540368393a15491e6e7fcd230039d Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Fri, 1 Feb 2019 12:42:07 +0100 Subject: vsock/virtio: reset connected sockets on device removal When the virtio transport device disappear, we should reset all connected sockets in order to inform the users. Signed-off-by: Stefano Garzarella Reviewed-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- net/vmw_vsock/virtio_transport.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c index 9dae54698737..15eb5d3d4750 100644 --- a/net/vmw_vsock/virtio_transport.c +++ b/net/vmw_vsock/virtio_transport.c @@ -634,6 +634,9 @@ static void virtio_vsock_remove(struct virtio_device *vdev) flush_work(&vsock->event_work); flush_work(&vsock->send_pkt_work); + /* Reset all connected sockets when the device disappear */ + vsock_for_each_connected_socket(virtio_vsock_reset_sock); + vdev->config->reset(vdev); mutex_lock(&vsock->rx_lock); -- cgit v1.2.3 From cfe4bd7a257f6d6f81d3458d8c9d9ec4957539e6 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Mon, 4 Feb 2019 03:27:58 +0800 Subject: sctp: check and update stream->out_curr when allocating stream_out Now when using stream reconfig to add out streams, stream->out will get re-allocated, and all old streams' information will be copied to the new ones and the old ones will be freed. So without stream->out_curr updated, next time when trying to send from stream->out_curr stream, a panic would be caused. This patch is to check and update stream->out_curr when allocating stream_out. v1->v2: - define fa_index() to get elem index from stream->out_curr. v2->v3: - repost with no change. Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations") Reported-by: Ying Xu Reported-by: syzbot+e33a3a138267ca119c7d@syzkaller.appspotmail.com Signed-off-by: Xin Long Acked-by: Neil Horman Signed-off-by: David S. Miller --- net/sctp/stream.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'net') diff --git a/net/sctp/stream.c b/net/sctp/stream.c index 80e0ae5534ec..f24633114dfd 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -84,6 +84,19 @@ static void fa_zero(struct flex_array *fa, size_t index, size_t count) } } +static size_t fa_index(struct flex_array *fa, void *elem, size_t count) +{ + size_t index = 0; + + while (count--) { + if (elem == flex_array_get(fa, index)) + break; + index++; + } + + return index; +} + /* Migrates chunks from stream queues to new stream queues if needed, * but not across associations. Also, removes those chunks to streams * higher than the new max. @@ -147,6 +160,13 @@ static int sctp_stream_alloc_out(struct sctp_stream *stream, __u16 outcnt, if (stream->out) { fa_copy(out, stream->out, 0, min(outcnt, stream->outcnt)); + if (stream->out_curr) { + size_t index = fa_index(stream->out, stream->out_curr, + stream->outcnt); + + BUG_ON(index == stream->outcnt); + stream->out_curr = flex_array_get(out, index); + } fa_free(stream->out); } -- cgit v1.2.3 From 4e35c1cb9460240e983a01745b5f29fe3a4d8e39 Mon Sep 17 00:00:00 2001 From: Martynas Pumputis Date: Tue, 29 Jan 2019 15:51:42 +0100 Subject: netfilter: nf_nat: skip nat clash resolution for same-origin entries It is possible that two concurrent packets originating from the same socket of a connection-less protocol (e.g. UDP) can end up having different IP_CT_DIR_REPLY tuples which results in one of the packets being dropped. To illustrate this, consider the following simplified scenario: 1. Packet A and B are sent at the same time from two different threads by same UDP socket. No matching conntrack entry exists yet. Both packets cause allocation of a new conntrack entry. 2. get_unique_tuple gets called for A. No clashing entry found. conntrack entry for A is added to main conntrack table. 3. get_unique_tuple is called for B and will find that the reply tuple of B is already taken by A. It will allocate a new UDP source port for B to resolve the clash. 4. conntrack entry for B cannot be added to main conntrack table because its ORIGINAL direction is clashing with A and the REPLY directions of A and B are not the same anymore due to UDP source port reallocation done in step 3. This patch modifies nf_conntrack_tuple_taken so it doesn't consider colliding reply tuples if the IP_CT_DIR_ORIGINAL tuples are equal. [ Florian: simplify patch to not use .allow_clash setting and always ignore identical flows ] Signed-off-by: Martynas Pumputis Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_core.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 741b533148ba..db4d46332e86 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1007,6 +1007,22 @@ nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, } if (nf_ct_key_equal(h, tuple, zone, net)) { + /* Tuple is taken already, so caller will need to find + * a new source port to use. + * + * Only exception: + * If the *original tuples* are identical, then both + * conntracks refer to the same flow. + * This is a rare situation, it can occur e.g. when + * more than one UDP packet is sent from same socket + * in different threads. + * + * Let nf_ct_resolve_clash() deal with this later. + */ + if (nf_ct_tuple_equal(&ignored_conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple, + &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple)) + continue; + NF_CT_STAT_INC_ATOMIC(net, found); rcu_read_unlock(); return 1; -- cgit v1.2.3 From f6ac8585897684374a19863fff21186a05805286 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sat, 2 Feb 2019 10:49:13 +0100 Subject: netfilter: nf_tables: unbind set in rule from commit path Anonymous sets that are bound to rules from the same transaction trigger a kernel splat from the abort path due to double set list removal and double free. This patch updates the logic to search for the transaction that is responsible for creating the set and disable the set list removal and release, given the rule is now responsible for this. Lookup is reverse since the transaction that adds the set is likely to be at the tail of the list. Moreover, this patch adds the unbind step to deliver the event from the commit path. This should not be done from the worker thread, since we have no guarantees of in-order delivery to the listener. This patch removes the assumption that both activate and deactivate callbacks need to be provided. Fixes: cd5125d8f518 ("netfilter: nf_tables: split set destruction in deactivate and destroy phase") Reported-by: Mikhail Morfikov Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 17 ++++++-- net/netfilter/nf_tables_api.c | 85 +++++++++++++++++++-------------------- net/netfilter/nft_compat.c | 6 ++- net/netfilter/nft_dynset.c | 18 ++++----- net/netfilter/nft_immediate.c | 6 ++- net/netfilter/nft_lookup.c | 18 ++++----- net/netfilter/nft_objref.c | 18 ++++----- 7 files changed, 85 insertions(+), 83 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 841835a387e1..b4984bbbe157 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -469,9 +469,7 @@ struct nft_set_binding { int nf_tables_bind_set(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_binding *binding); void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, - struct nft_set_binding *binding); -void nf_tables_rebind_set(const struct nft_ctx *ctx, struct nft_set *set, - struct nft_set_binding *binding); + struct nft_set_binding *binding, bool commit); void nf_tables_destroy_set(const struct nft_ctx *ctx, struct nft_set *set); /** @@ -721,6 +719,13 @@ struct nft_expr_type { #define NFT_EXPR_STATEFUL 0x1 #define NFT_EXPR_GC 0x2 +enum nft_trans_phase { + NFT_TRANS_PREPARE, + NFT_TRANS_ABORT, + NFT_TRANS_COMMIT, + NFT_TRANS_RELEASE +}; + /** * struct nft_expr_ops - nf_tables expression operations * @@ -750,7 +755,8 @@ struct nft_expr_ops { void (*activate)(const struct nft_ctx *ctx, const struct nft_expr *expr); void (*deactivate)(const struct nft_ctx *ctx, - const struct nft_expr *expr); + const struct nft_expr *expr, + enum nft_trans_phase phase); void (*destroy)(const struct nft_ctx *ctx, const struct nft_expr *expr); void (*destroy_clone)(const struct nft_ctx *ctx, @@ -1323,12 +1329,15 @@ struct nft_trans_rule { struct nft_trans_set { struct nft_set *set; u32 set_id; + bool bound; }; #define nft_trans_set(trans) \ (((struct nft_trans_set *)trans->data)->set) #define nft_trans_set_id(trans) \ (((struct nft_trans_set *)trans->data)->set_id) +#define nft_trans_set_bound(trans) \ + (((struct nft_trans_set *)trans->data)->bound) struct nft_trans_chain { bool update; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index fb07f6cfc719..5a92f23f179f 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -116,6 +116,23 @@ static void nft_trans_destroy(struct nft_trans *trans) kfree(trans); } +static void nft_set_trans_bind(const struct nft_ctx *ctx, struct nft_set *set) +{ + struct net *net = ctx->net; + struct nft_trans *trans; + + if (!nft_set_is_anonymous(set)) + return; + + list_for_each_entry_reverse(trans, &net->nft.commit_list, list) { + if (trans->msg_type == NFT_MSG_NEWSET && + nft_trans_set(trans) == set) { + nft_trans_set_bound(trans) = true; + break; + } + } +} + static int nf_tables_register_hook(struct net *net, const struct nft_table *table, struct nft_chain *chain) @@ -211,18 +228,6 @@ static int nft_delchain(struct nft_ctx *ctx) return err; } -/* either expr ops provide both activate/deactivate, or neither */ -static bool nft_expr_check_ops(const struct nft_expr_ops *ops) -{ - if (!ops) - return true; - - if (WARN_ON_ONCE((!ops->activate ^ !ops->deactivate))) - return false; - - return true; -} - static void nft_rule_expr_activate(const struct nft_ctx *ctx, struct nft_rule *rule) { @@ -238,14 +243,15 @@ static void nft_rule_expr_activate(const struct nft_ctx *ctx, } static void nft_rule_expr_deactivate(const struct nft_ctx *ctx, - struct nft_rule *rule) + struct nft_rule *rule, + enum nft_trans_phase phase) { struct nft_expr *expr; expr = nft_expr_first(rule); while (expr != nft_expr_last(rule) && expr->ops) { if (expr->ops->deactivate) - expr->ops->deactivate(ctx, expr); + expr->ops->deactivate(ctx, expr, phase); expr = nft_expr_next(expr); } @@ -296,7 +302,7 @@ static int nft_delrule(struct nft_ctx *ctx, struct nft_rule *rule) nft_trans_destroy(trans); return err; } - nft_rule_expr_deactivate(ctx, rule); + nft_rule_expr_deactivate(ctx, rule, NFT_TRANS_PREPARE); return 0; } @@ -1929,9 +1935,6 @@ static int nf_tables_delchain(struct net *net, struct sock *nlsk, */ int nft_register_expr(struct nft_expr_type *type) { - if (!nft_expr_check_ops(type->ops)) - return -EINVAL; - nfnl_lock(NFNL_SUBSYS_NFTABLES); if (type->family == NFPROTO_UNSPEC) list_add_tail_rcu(&type->list, &nf_tables_expressions); @@ -2079,10 +2082,6 @@ static int nf_tables_expr_parse(const struct nft_ctx *ctx, err = PTR_ERR(ops); goto err1; } - if (!nft_expr_check_ops(ops)) { - err = -EINVAL; - goto err1; - } } else ops = type->ops; @@ -2511,7 +2510,7 @@ static void nf_tables_rule_destroy(const struct nft_ctx *ctx, static void nf_tables_rule_release(const struct nft_ctx *ctx, struct nft_rule *rule) { - nft_rule_expr_deactivate(ctx, rule); + nft_rule_expr_deactivate(ctx, rule, NFT_TRANS_RELEASE); nf_tables_rule_destroy(ctx, rule); } @@ -3708,39 +3707,30 @@ int nf_tables_bind_set(const struct nft_ctx *ctx, struct nft_set *set, bind: binding->chain = ctx->chain; list_add_tail_rcu(&binding->list, &set->bindings); + nft_set_trans_bind(ctx, set); + return 0; } EXPORT_SYMBOL_GPL(nf_tables_bind_set); -void nf_tables_rebind_set(const struct nft_ctx *ctx, struct nft_set *set, - struct nft_set_binding *binding) -{ - if (list_empty(&set->bindings) && nft_set_is_anonymous(set) && - nft_is_active(ctx->net, set)) - list_add_tail_rcu(&set->list, &ctx->table->sets); - - list_add_tail_rcu(&binding->list, &set->bindings); -} -EXPORT_SYMBOL_GPL(nf_tables_rebind_set); - void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, - struct nft_set_binding *binding) + struct nft_set_binding *binding, bool event) { list_del_rcu(&binding->list); - if (list_empty(&set->bindings) && nft_set_is_anonymous(set) && - nft_is_active(ctx->net, set)) + if (list_empty(&set->bindings) && nft_set_is_anonymous(set)) { list_del_rcu(&set->list); + if (event) + nf_tables_set_notify(ctx, set, NFT_MSG_DELSET, + GFP_KERNEL); + } } EXPORT_SYMBOL_GPL(nf_tables_unbind_set); void nf_tables_destroy_set(const struct nft_ctx *ctx, struct nft_set *set) { - if (list_empty(&set->bindings) && nft_set_is_anonymous(set) && - nft_is_active(ctx->net, set)) { - nf_tables_set_notify(ctx, set, NFT_MSG_DELSET, GFP_ATOMIC); + if (list_empty(&set->bindings) && nft_set_is_anonymous(set)) nft_set_destroy(set); - } } EXPORT_SYMBOL_GPL(nf_tables_destroy_set); @@ -6535,6 +6525,9 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nf_tables_rule_notify(&trans->ctx, nft_trans_rule(trans), NFT_MSG_DELRULE); + nft_rule_expr_deactivate(&trans->ctx, + nft_trans_rule(trans), + NFT_TRANS_COMMIT); break; case NFT_MSG_NEWSET: nft_clear(net, nft_trans_set(trans)); @@ -6621,7 +6614,8 @@ static void nf_tables_abort_release(struct nft_trans *trans) nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); break; case NFT_MSG_NEWSET: - nft_set_destroy(nft_trans_set(trans)); + if (!nft_trans_set_bound(trans)) + nft_set_destroy(nft_trans_set(trans)); break; case NFT_MSG_NEWSETELEM: nft_set_elem_destroy(nft_trans_elem_set(trans), @@ -6682,7 +6676,9 @@ static int __nf_tables_abort(struct net *net) case NFT_MSG_NEWRULE: trans->ctx.chain->use--; list_del_rcu(&nft_trans_rule(trans)->list); - nft_rule_expr_deactivate(&trans->ctx, nft_trans_rule(trans)); + nft_rule_expr_deactivate(&trans->ctx, + nft_trans_rule(trans), + NFT_TRANS_ABORT); break; case NFT_MSG_DELRULE: trans->ctx.chain->use++; @@ -6692,7 +6688,8 @@ static int __nf_tables_abort(struct net *net) break; case NFT_MSG_NEWSET: trans->ctx.table->use--; - list_del_rcu(&nft_trans_set(trans)->list); + if (!nft_trans_set_bound(trans)) + list_del_rcu(&nft_trans_set(trans)->list); break; case NFT_MSG_DELSET: trans->ctx.table->use++; diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 5eb269428832..0732a2fc697c 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -587,10 +587,14 @@ static void nft_compat_activate_tg(const struct nft_ctx *ctx, } static void nft_compat_deactivate(const struct nft_ctx *ctx, - const struct nft_expr *expr) + const struct nft_expr *expr, + enum nft_trans_phase phase) { struct nft_xt *xt = container_of(expr->ops, struct nft_xt, ops); + if (phase == NFT_TRANS_COMMIT) + return; + if (--xt->listcnt == 0) list_del_init(&xt->head); } diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c index 07d4efd3d851..f1172f99752b 100644 --- a/net/netfilter/nft_dynset.c +++ b/net/netfilter/nft_dynset.c @@ -235,20 +235,17 @@ err1: return err; } -static void nft_dynset_activate(const struct nft_ctx *ctx, - const struct nft_expr *expr) -{ - struct nft_dynset *priv = nft_expr_priv(expr); - - nf_tables_rebind_set(ctx, priv->set, &priv->binding); -} - static void nft_dynset_deactivate(const struct nft_ctx *ctx, - const struct nft_expr *expr) + const struct nft_expr *expr, + enum nft_trans_phase phase) { struct nft_dynset *priv = nft_expr_priv(expr); - nf_tables_unbind_set(ctx, priv->set, &priv->binding); + if (phase == NFT_TRANS_PREPARE) + return; + + nf_tables_unbind_set(ctx, priv->set, &priv->binding, + phase == NFT_TRANS_COMMIT); } static void nft_dynset_destroy(const struct nft_ctx *ctx, @@ -296,7 +293,6 @@ static const struct nft_expr_ops nft_dynset_ops = { .eval = nft_dynset_eval, .init = nft_dynset_init, .destroy = nft_dynset_destroy, - .activate = nft_dynset_activate, .deactivate = nft_dynset_deactivate, .dump = nft_dynset_dump, }; diff --git a/net/netfilter/nft_immediate.c b/net/netfilter/nft_immediate.c index 0777a93211e2..3f6d1d2a6281 100644 --- a/net/netfilter/nft_immediate.c +++ b/net/netfilter/nft_immediate.c @@ -72,10 +72,14 @@ static void nft_immediate_activate(const struct nft_ctx *ctx, } static void nft_immediate_deactivate(const struct nft_ctx *ctx, - const struct nft_expr *expr) + const struct nft_expr *expr, + enum nft_trans_phase phase) { const struct nft_immediate_expr *priv = nft_expr_priv(expr); + if (phase == NFT_TRANS_COMMIT) + return; + return nft_data_release(&priv->data, nft_dreg_to_type(priv->dreg)); } diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 227b2b15a19c..14496da5141d 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -121,20 +121,17 @@ static int nft_lookup_init(const struct nft_ctx *ctx, return 0; } -static void nft_lookup_activate(const struct nft_ctx *ctx, - const struct nft_expr *expr) -{ - struct nft_lookup *priv = nft_expr_priv(expr); - - nf_tables_rebind_set(ctx, priv->set, &priv->binding); -} - static void nft_lookup_deactivate(const struct nft_ctx *ctx, - const struct nft_expr *expr) + const struct nft_expr *expr, + enum nft_trans_phase phase) { struct nft_lookup *priv = nft_expr_priv(expr); - nf_tables_unbind_set(ctx, priv->set, &priv->binding); + if (phase == NFT_TRANS_PREPARE) + return; + + nf_tables_unbind_set(ctx, priv->set, &priv->binding, + phase == NFT_TRANS_COMMIT); } static void nft_lookup_destroy(const struct nft_ctx *ctx, @@ -225,7 +222,6 @@ static const struct nft_expr_ops nft_lookup_ops = { .size = NFT_EXPR_SIZE(sizeof(struct nft_lookup)), .eval = nft_lookup_eval, .init = nft_lookup_init, - .activate = nft_lookup_activate, .deactivate = nft_lookup_deactivate, .destroy = nft_lookup_destroy, .dump = nft_lookup_dump, diff --git a/net/netfilter/nft_objref.c b/net/netfilter/nft_objref.c index a3185ca2a3a9..ae178e914486 100644 --- a/net/netfilter/nft_objref.c +++ b/net/netfilter/nft_objref.c @@ -155,20 +155,17 @@ nla_put_failure: return -1; } -static void nft_objref_map_activate(const struct nft_ctx *ctx, - const struct nft_expr *expr) -{ - struct nft_objref_map *priv = nft_expr_priv(expr); - - nf_tables_rebind_set(ctx, priv->set, &priv->binding); -} - static void nft_objref_map_deactivate(const struct nft_ctx *ctx, - const struct nft_expr *expr) + const struct nft_expr *expr, + enum nft_trans_phase phase) { struct nft_objref_map *priv = nft_expr_priv(expr); - nf_tables_unbind_set(ctx, priv->set, &priv->binding); + if (phase == NFT_TRANS_PREPARE) + return; + + nf_tables_unbind_set(ctx, priv->set, &priv->binding, + phase == NFT_TRANS_COMMIT); } static void nft_objref_map_destroy(const struct nft_ctx *ctx, @@ -185,7 +182,6 @@ static const struct nft_expr_ops nft_objref_map_ops = { .size = NFT_EXPR_SIZE(sizeof(struct nft_objref_map)), .eval = nft_objref_map_eval, .init = nft_objref_map_init, - .activate = nft_objref_map_activate, .deactivate = nft_objref_map_deactivate, .destroy = nft_objref_map_destroy, .dump = nft_objref_map_dump, -- cgit v1.2.3 From ad6f317f720f4a3121756c23831a43dda9b095e5 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Mon, 4 Feb 2019 13:44:44 +0100 Subject: net/smc: preallocated memory for rdma work requests The work requests for rdma writes are built in local variables within function smc_tx_rdma_write(). This violates the rule that the work request storage has to stay till the work request is confirmed by a completion queue response. This patch introduces preallocated memory for these work requests. The storage is allocated, once a link (and thus a queue pair) is established. Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_cdc.c | 11 +++-------- net/smc/smc_cdc.h | 8 +++++++- net/smc/smc_core.h | 20 ++++++++++++++++++++ net/smc/smc_llc.c | 3 ++- net/smc/smc_tx.c | 44 ++++++++++++++++++++++---------------------- net/smc/smc_wr.c | 38 +++++++++++++++++++++++++++++++++++++- net/smc/smc_wr.h | 1 + 7 files changed, 92 insertions(+), 33 deletions(-) (limited to 'net') diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c index 1c5333d494e9..b80ef104ab4e 100644 --- a/net/smc/smc_cdc.c +++ b/net/smc/smc_cdc.c @@ -21,13 +21,6 @@ /********************************** send *************************************/ -struct smc_cdc_tx_pend { - struct smc_connection *conn; /* socket connection */ - union smc_host_cursor cursor; /* tx sndbuf cursor sent */ - union smc_host_cursor p_cursor; /* rx RMBE cursor produced */ - u16 ctrl_seq; /* conn. tx sequence # */ -}; - /* handler for send/transmission completion of a CDC msg */ static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd, struct smc_link *link, @@ -61,12 +54,14 @@ static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd, int smc_cdc_get_free_slot(struct smc_connection *conn, struct smc_wr_buf **wr_buf, + struct smc_rdma_wr **wr_rdma_buf, struct smc_cdc_tx_pend **pend) { struct smc_link *link = &conn->lgr->lnk[SMC_SINGLE_LINK]; int rc; rc = smc_wr_tx_get_free_slot(link, smc_cdc_tx_handler, wr_buf, + wr_rdma_buf, (struct smc_wr_tx_pend_priv **)pend); if (!conn->alert_token_local) /* abnormal termination */ @@ -121,7 +116,7 @@ static int smcr_cdc_get_slot_and_msg_send(struct smc_connection *conn) struct smc_wr_buf *wr_buf; int rc; - rc = smc_cdc_get_free_slot(conn, &wr_buf, &pend); + rc = smc_cdc_get_free_slot(conn, &wr_buf, NULL, &pend); if (rc) return rc; diff --git a/net/smc/smc_cdc.h b/net/smc/smc_cdc.h index b5bfe38c7f9b..2148da7a26b1 100644 --- a/net/smc/smc_cdc.h +++ b/net/smc/smc_cdc.h @@ -270,10 +270,16 @@ static inline void smc_cdc_msg_to_host(struct smc_host_cdc_msg *local, smcr_cdc_msg_to_host(local, peer, conn); } -struct smc_cdc_tx_pend; +struct smc_cdc_tx_pend { + struct smc_connection *conn; /* socket connection */ + union smc_host_cursor cursor; /* tx sndbuf cursor sent */ + union smc_host_cursor p_cursor; /* rx RMBE cursor produced */ + u16 ctrl_seq; /* conn. tx sequence # */ +}; int smc_cdc_get_free_slot(struct smc_connection *conn, struct smc_wr_buf **wr_buf, + struct smc_rdma_wr **wr_rdma_buf, struct smc_cdc_tx_pend **pend); void smc_cdc_tx_dismiss_slots(struct smc_connection *conn); int smc_cdc_msg_send(struct smc_connection *conn, struct smc_wr_buf *wr_buf, diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index b00287989a3d..8806d2afa6ed 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -52,6 +52,24 @@ enum smc_wr_reg_state { FAILED /* ib_wr_reg_mr response: failure */ }; +struct smc_rdma_sge { /* sges for RDMA writes */ + struct ib_sge wr_tx_rdma_sge[SMC_IB_MAX_SEND_SGE]; +}; + +#define SMC_MAX_RDMA_WRITES 2 /* max. # of RDMA writes per + * message send + */ + +struct smc_rdma_sges { /* sges per message send */ + struct smc_rdma_sge tx_rdma_sge[SMC_MAX_RDMA_WRITES]; +}; + +struct smc_rdma_wr { /* work requests per message + * send + */ + struct ib_rdma_wr wr_tx_rdma[SMC_MAX_RDMA_WRITES]; +}; + struct smc_link { struct smc_ib_device *smcibdev; /* ib-device */ u8 ibport; /* port - values 1 | 2 */ @@ -64,6 +82,8 @@ struct smc_link { struct smc_wr_buf *wr_tx_bufs; /* WR send payload buffers */ struct ib_send_wr *wr_tx_ibs; /* WR send meta data */ struct ib_sge *wr_tx_sges; /* WR send gather meta data */ + struct smc_rdma_sges *wr_tx_rdma_sges;/*RDMA WRITE gather meta data*/ + struct smc_rdma_wr *wr_tx_rdmas; /* WR RDMA WRITE */ struct smc_wr_tx_pend *wr_tx_pends; /* WR send waiting for CQE */ /* above four vectors have wr_tx_cnt elements and use the same index */ dma_addr_t wr_tx_dma_addr; /* DMA address of wr_tx_bufs */ diff --git a/net/smc/smc_llc.c b/net/smc/smc_llc.c index a6d3623d06f4..4fd60c522802 100644 --- a/net/smc/smc_llc.c +++ b/net/smc/smc_llc.c @@ -166,7 +166,8 @@ static int smc_llc_add_pending_send(struct smc_link *link, { int rc; - rc = smc_wr_tx_get_free_slot(link, smc_llc_tx_handler, wr_buf, pend); + rc = smc_wr_tx_get_free_slot(link, smc_llc_tx_handler, wr_buf, NULL, + pend); if (rc < 0) return rc; BUILD_BUG_ON_MSG( diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index 36af3de731b9..2fdfaff60cf9 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -266,27 +266,23 @@ int smcd_tx_ism_write(struct smc_connection *conn, void *data, size_t len, /* sndbuf consumer: actual data transfer of one target chunk with RDMA write */ static int smc_tx_rdma_write(struct smc_connection *conn, int peer_rmbe_offset, - int num_sges, struct ib_sge sges[]) + int num_sges, struct ib_rdma_wr *rdma_wr) { struct smc_link_group *lgr = conn->lgr; - struct ib_rdma_wr rdma_wr; struct smc_link *link; int rc; - memset(&rdma_wr, 0, sizeof(rdma_wr)); link = &lgr->lnk[SMC_SINGLE_LINK]; - rdma_wr.wr.wr_id = smc_wr_tx_get_next_wr_id(link); - rdma_wr.wr.sg_list = sges; - rdma_wr.wr.num_sge = num_sges; - rdma_wr.wr.opcode = IB_WR_RDMA_WRITE; - rdma_wr.remote_addr = + rdma_wr->wr.wr_id = smc_wr_tx_get_next_wr_id(link); + rdma_wr->wr.num_sge = num_sges; + rdma_wr->remote_addr = lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].dma_addr + /* RMBE within RMB */ conn->tx_off + /* offset within RMBE */ peer_rmbe_offset; - rdma_wr.rkey = lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].rkey; - rc = ib_post_send(link->roce_qp, &rdma_wr.wr, NULL); + rdma_wr->rkey = lgr->rtokens[conn->rtoken_idx][SMC_SINGLE_LINK].rkey; + rc = ib_post_send(link->roce_qp, &rdma_wr->wr, NULL); if (rc) { conn->local_tx_ctrl.conn_state_flags.peer_conn_abort = 1; smc_lgr_terminate(lgr); @@ -313,24 +309,25 @@ static inline void smc_tx_advance_cursors(struct smc_connection *conn, /* SMC-R helper for smc_tx_rdma_writes() */ static int smcr_tx_rdma_writes(struct smc_connection *conn, size_t len, size_t src_off, size_t src_len, - size_t dst_off, size_t dst_len) + size_t dst_off, size_t dst_len, + struct smc_rdma_wr *wr_rdma_buf) { dma_addr_t dma_addr = sg_dma_address(conn->sndbuf_desc->sgt[SMC_SINGLE_LINK].sgl); - struct smc_link *link = &conn->lgr->lnk[SMC_SINGLE_LINK]; int src_len_sum = src_len, dst_len_sum = dst_len; - struct ib_sge sges[SMC_IB_MAX_SEND_SGE]; int sent_count = src_off; int srcchunk, dstchunk; int num_sges; int rc; for (dstchunk = 0; dstchunk < 2; dstchunk++) { + struct ib_sge *sge = + wr_rdma_buf->wr_tx_rdma[dstchunk].wr.sg_list; + num_sges = 0; for (srcchunk = 0; srcchunk < 2; srcchunk++) { - sges[srcchunk].addr = dma_addr + src_off; - sges[srcchunk].length = src_len; - sges[srcchunk].lkey = link->roce_pd->local_dma_lkey; + sge[srcchunk].addr = dma_addr + src_off; + sge[srcchunk].length = src_len; num_sges++; src_off += src_len; @@ -343,7 +340,8 @@ static int smcr_tx_rdma_writes(struct smc_connection *conn, size_t len, src_len = dst_len - src_len; /* remainder */ src_len_sum += src_len; } - rc = smc_tx_rdma_write(conn, dst_off, num_sges, sges); + rc = smc_tx_rdma_write(conn, dst_off, num_sges, + &wr_rdma_buf->wr_tx_rdma[dstchunk]); if (rc) return rc; if (dst_len_sum == len) @@ -402,7 +400,8 @@ static int smcd_tx_rdma_writes(struct smc_connection *conn, size_t len, /* sndbuf consumer: prepare all necessary (src&dst) chunks of data transmit; * usable snd_wnd as max transmit */ -static int smc_tx_rdma_writes(struct smc_connection *conn) +static int smc_tx_rdma_writes(struct smc_connection *conn, + struct smc_rdma_wr *wr_rdma_buf) { size_t len, src_len, dst_off, dst_len; /* current chunk values */ union smc_host_cursor sent, prep, prod, cons; @@ -463,7 +462,7 @@ static int smc_tx_rdma_writes(struct smc_connection *conn) dst_off, dst_len); else rc = smcr_tx_rdma_writes(conn, len, sent.count, src_len, - dst_off, dst_len); + dst_off, dst_len, wr_rdma_buf); if (rc) return rc; @@ -484,11 +483,12 @@ static int smc_tx_rdma_writes(struct smc_connection *conn) static int smcr_tx_sndbuf_nonempty(struct smc_connection *conn) { struct smc_cdc_producer_flags *pflags; + struct smc_rdma_wr *wr_rdma_buf; struct smc_cdc_tx_pend *pend; struct smc_wr_buf *wr_buf; int rc; - rc = smc_cdc_get_free_slot(conn, &wr_buf, &pend); + rc = smc_cdc_get_free_slot(conn, &wr_buf, &wr_rdma_buf, &pend); if (rc < 0) { if (rc == -EBUSY) { struct smc_sock *smc = @@ -506,7 +506,7 @@ static int smcr_tx_sndbuf_nonempty(struct smc_connection *conn) spin_lock_bh(&conn->send_lock); if (!conn->local_tx_ctrl.prod_flags.urg_data_present) { - rc = smc_tx_rdma_writes(conn); + rc = smc_tx_rdma_writes(conn, wr_rdma_buf); if (rc) { smc_wr_tx_put_slot(&conn->lgr->lnk[SMC_SINGLE_LINK], (struct smc_wr_tx_pend_priv *)pend); @@ -533,7 +533,7 @@ static int smcd_tx_sndbuf_nonempty(struct smc_connection *conn) spin_lock_bh(&conn->send_lock); if (!pflags->urg_data_present) - rc = smc_tx_rdma_writes(conn); + rc = smc_tx_rdma_writes(conn, NULL); if (!rc) rc = smcd_cdc_msg_send(conn); diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index 1dc88c32d6bb..253aa75dc2b6 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -160,6 +160,7 @@ static inline int smc_wr_tx_get_free_slot_index(struct smc_link *link, u32 *idx) * @link: Pointer to smc_link used to later send the message. * @handler: Send completion handler function pointer. * @wr_buf: Out value returns pointer to message buffer. + * @wr_rdma_buf: Out value returns pointer to rdma work request. * @wr_pend_priv: Out value returns pointer serving as handler context. * * Return: 0 on success, or -errno on error. @@ -167,6 +168,7 @@ static inline int smc_wr_tx_get_free_slot_index(struct smc_link *link, u32 *idx) int smc_wr_tx_get_free_slot(struct smc_link *link, smc_wr_tx_handler handler, struct smc_wr_buf **wr_buf, + struct smc_rdma_wr **wr_rdma_buf, struct smc_wr_tx_pend_priv **wr_pend_priv) { struct smc_wr_tx_pend *wr_pend; @@ -204,6 +206,8 @@ int smc_wr_tx_get_free_slot(struct smc_link *link, wr_ib = &link->wr_tx_ibs[idx]; wr_ib->wr_id = wr_id; *wr_buf = &link->wr_tx_bufs[idx]; + if (wr_rdma_buf) + *wr_rdma_buf = &link->wr_tx_rdmas[idx]; *wr_pend_priv = &wr_pend->priv; return 0; } @@ -465,12 +469,26 @@ static void smc_wr_init_sge(struct smc_link *lnk) lnk->wr_tx_dma_addr + i * SMC_WR_BUF_SIZE; lnk->wr_tx_sges[i].length = SMC_WR_TX_SIZE; lnk->wr_tx_sges[i].lkey = lnk->roce_pd->local_dma_lkey; + lnk->wr_tx_rdma_sges[i].tx_rdma_sge[0].wr_tx_rdma_sge[0].lkey = + lnk->roce_pd->local_dma_lkey; + lnk->wr_tx_rdma_sges[i].tx_rdma_sge[0].wr_tx_rdma_sge[1].lkey = + lnk->roce_pd->local_dma_lkey; + lnk->wr_tx_rdma_sges[i].tx_rdma_sge[1].wr_tx_rdma_sge[0].lkey = + lnk->roce_pd->local_dma_lkey; + lnk->wr_tx_rdma_sges[i].tx_rdma_sge[1].wr_tx_rdma_sge[1].lkey = + lnk->roce_pd->local_dma_lkey; lnk->wr_tx_ibs[i].next = NULL; lnk->wr_tx_ibs[i].sg_list = &lnk->wr_tx_sges[i]; lnk->wr_tx_ibs[i].num_sge = 1; lnk->wr_tx_ibs[i].opcode = IB_WR_SEND; lnk->wr_tx_ibs[i].send_flags = IB_SEND_SIGNALED | IB_SEND_SOLICITED; + lnk->wr_tx_rdmas[i].wr_tx_rdma[0].wr.opcode = IB_WR_RDMA_WRITE; + lnk->wr_tx_rdmas[i].wr_tx_rdma[1].wr.opcode = IB_WR_RDMA_WRITE; + lnk->wr_tx_rdmas[i].wr_tx_rdma[0].wr.sg_list = + lnk->wr_tx_rdma_sges[i].tx_rdma_sge[0].wr_tx_rdma_sge; + lnk->wr_tx_rdmas[i].wr_tx_rdma[1].wr.sg_list = + lnk->wr_tx_rdma_sges[i].tx_rdma_sge[1].wr_tx_rdma_sge; } for (i = 0; i < lnk->wr_rx_cnt; i++) { lnk->wr_rx_sges[i].addr = @@ -521,8 +539,12 @@ void smc_wr_free_link_mem(struct smc_link *lnk) lnk->wr_tx_mask = NULL; kfree(lnk->wr_tx_sges); lnk->wr_tx_sges = NULL; + kfree(lnk->wr_tx_rdma_sges); + lnk->wr_tx_rdma_sges = NULL; kfree(lnk->wr_rx_sges); lnk->wr_rx_sges = NULL; + kfree(lnk->wr_tx_rdmas); + lnk->wr_tx_rdmas = NULL; kfree(lnk->wr_rx_ibs); lnk->wr_rx_ibs = NULL; kfree(lnk->wr_tx_ibs); @@ -552,10 +574,20 @@ int smc_wr_alloc_link_mem(struct smc_link *link) GFP_KERNEL); if (!link->wr_rx_ibs) goto no_mem_wr_tx_ibs; + link->wr_tx_rdmas = kcalloc(SMC_WR_BUF_CNT, + sizeof(link->wr_tx_rdmas[0]), + GFP_KERNEL); + if (!link->wr_tx_rdmas) + goto no_mem_wr_rx_ibs; + link->wr_tx_rdma_sges = kcalloc(SMC_WR_BUF_CNT, + sizeof(link->wr_tx_rdma_sges[0]), + GFP_KERNEL); + if (!link->wr_tx_rdma_sges) + goto no_mem_wr_tx_rdmas; link->wr_tx_sges = kcalloc(SMC_WR_BUF_CNT, sizeof(link->wr_tx_sges[0]), GFP_KERNEL); if (!link->wr_tx_sges) - goto no_mem_wr_rx_ibs; + goto no_mem_wr_tx_rdma_sges; link->wr_rx_sges = kcalloc(SMC_WR_BUF_CNT * 3, sizeof(link->wr_rx_sges[0]), GFP_KERNEL); @@ -579,6 +611,10 @@ no_mem_wr_rx_sges: kfree(link->wr_rx_sges); no_mem_wr_tx_sges: kfree(link->wr_tx_sges); +no_mem_wr_tx_rdma_sges: + kfree(link->wr_tx_rdma_sges); +no_mem_wr_tx_rdmas: + kfree(link->wr_tx_rdmas); no_mem_wr_rx_ibs: kfree(link->wr_rx_ibs); no_mem_wr_tx_ibs: diff --git a/net/smc/smc_wr.h b/net/smc/smc_wr.h index 1d85bb14fd6f..09bf32fd3959 100644 --- a/net/smc/smc_wr.h +++ b/net/smc/smc_wr.h @@ -85,6 +85,7 @@ void smc_wr_add_dev(struct smc_ib_device *smcibdev); int smc_wr_tx_get_free_slot(struct smc_link *link, smc_wr_tx_handler handler, struct smc_wr_buf **wr_buf, + struct smc_rdma_wr **wrs, struct smc_wr_tx_pend_priv **wr_pend_priv); int smc_wr_tx_put_slot(struct smc_link *link, struct smc_wr_tx_pend_priv *wr_pend_priv); -- cgit v1.2.3 From b8649efad879c69c7ab1f19ce8814fcabef1f72b Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Mon, 4 Feb 2019 13:44:45 +0100 Subject: net/smc: fix sender_free computation In some scenarios a separate consumer cursor update is necessary. The decision is made in smc_tx_consumer_cursor_update(). The sender_free computation could be wrong: The rx confirmed cursor is always smaller than or equal to the rx producer cursor. The parameters in the smc_curs_diff() call have to be exchanged, otherwise sender_free might even be negative. And if more data arrives local_rx_ctrl.prod might be updated, enabling a cursor difference between local_rx_ctrl.prod and rx confirmed cursor larger than the RMB size. This case is not covered by smc_curs_diff(). Thus function smc_curs_diff_large() is introduced here. If a recvmsg() is processed in parallel, local_tx_ctrl.cons might change during smc_cdc_msg_send. Make sure rx_curs_confirmed is updated with the actually sent local_tx_ctrl.cons value. Fixes: e82f2e31f559 ("net/smc: optimize consumer cursor updates") Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_cdc.c | 5 +++-- net/smc/smc_cdc.h | 26 +++++++++++++++++++++++++- net/smc/smc_tx.c | 3 ++- 3 files changed, 30 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c index b80ef104ab4e..a712c9f8699b 100644 --- a/net/smc/smc_cdc.c +++ b/net/smc/smc_cdc.c @@ -91,6 +91,7 @@ int smc_cdc_msg_send(struct smc_connection *conn, struct smc_wr_buf *wr_buf, struct smc_cdc_tx_pend *pend) { + union smc_host_cursor cfed; struct smc_link *link; int rc; @@ -102,10 +103,10 @@ int smc_cdc_msg_send(struct smc_connection *conn, conn->local_tx_ctrl.seqno = conn->tx_cdc_seq; smc_host_msg_to_cdc((struct smc_cdc_msg *)wr_buf, &conn->local_tx_ctrl, conn); + smc_curs_copy(&cfed, &((struct smc_host_cdc_msg *)wr_buf)->cons, conn); rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend); if (!rc) - smc_curs_copy(&conn->rx_curs_confirmed, - &conn->local_tx_ctrl.cons, conn); + smc_curs_copy(&conn->rx_curs_confirmed, &cfed, conn); return rc; } diff --git a/net/smc/smc_cdc.h b/net/smc/smc_cdc.h index 2148da7a26b1..271e2524dc8f 100644 --- a/net/smc/smc_cdc.h +++ b/net/smc/smc_cdc.h @@ -160,7 +160,9 @@ static inline void smcd_curs_copy(union smcd_cdc_cursor *tgt, #endif } -/* calculate cursor difference between old and new, where old <= new */ +/* calculate cursor difference between old and new, where old <= new and + * difference cannot exceed size + */ static inline int smc_curs_diff(unsigned int size, union smc_host_cursor *old, union smc_host_cursor *new) @@ -185,6 +187,28 @@ static inline int smc_curs_comp(unsigned int size, return smc_curs_diff(size, old, new); } +/* calculate cursor difference between old and new, where old <= new and + * difference may exceed size + */ +static inline int smc_curs_diff_large(unsigned int size, + union smc_host_cursor *old, + union smc_host_cursor *new) +{ + if (old->wrap < new->wrap) + return min_t(int, + (size - old->count) + new->count + + (new->wrap - old->wrap - 1) * size, + size); + + if (old->wrap > new->wrap) /* wrap has switched from 0xffff to 0x0000 */ + return min_t(int, + (size - old->count) + new->count + + (new->wrap + 0xffff - old->wrap) * size, + size); + + return max_t(int, 0, (new->count - old->count)); +} + static inline void smc_host_cursor_to_cdc(union smc_cdc_cursor *peer, union smc_host_cursor *local, struct smc_connection *conn) diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index 2fdfaff60cf9..f93f3580c100 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -595,7 +595,8 @@ void smc_tx_consumer_update(struct smc_connection *conn, bool force) if (to_confirm > conn->rmbe_update_limit) { smc_curs_copy(&prod, &conn->local_rx_ctrl.prod, conn); sender_free = conn->rmb_desc->len - - smc_curs_diff(conn->rmb_desc->len, &prod, &cfed); + smc_curs_diff_large(conn->rmb_desc->len, + &cfed, &prod); } if (conn->local_rx_ctrl.prod_flags.cons_curs_upd_req || -- cgit v1.2.3 From a5e04318c83a31925300af1ce358dbc1a708b732 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Mon, 4 Feb 2019 13:44:46 +0100 Subject: net/smc: delete rkey first before switching to unused Once RMBs are flagged as unused they are candidates for reuse. Thus the LLC DELETE RKEY operaton should be made before flagging the RMB as unused. Fixes: c7674c001b11 ("net/smc: unregister rkeys of unused buffer") Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 097c798983ca..aa1c551cee81 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -302,13 +302,13 @@ static void smc_buf_unuse(struct smc_connection *conn, conn->sndbuf_desc->used = 0; if (conn->rmb_desc) { if (!conn->rmb_desc->regerr) { - conn->rmb_desc->used = 0; if (!lgr->is_smcd) { /* unregister rmb with peer */ smc_llc_do_delete_rkey( &lgr->lnk[SMC_SINGLE_LINK], conn->rmb_desc); } + conn->rmb_desc->used = 0; } else { /* buf registration failed, reuse not possible */ write_lock_bh(&lgr->rmbs_lock); -- cgit v1.2.3 From 84b799a292ebc03de388f8573a169de6eb12c340 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Mon, 4 Feb 2019 13:44:47 +0100 Subject: net/smc: correct state change for peer closing If some kind of closing is received from the peer while still in state SMC_INIT, it means the peer has had an active connection and closed the socket quickly before listen_work finished. This should not result in a shortcut from state SMC_INIT to state SMC_CLOSED. This patch adds the socket to the accept queue in state SMC_APPCLOSEWAIT1. The socket reaches state SMC_CLOSED once being accepted and closed with smc_release(). Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/smc_close.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'net') diff --git a/net/smc/smc_close.c b/net/smc/smc_close.c index ea2b87f29469..e39cadda1bf5 100644 --- a/net/smc/smc_close.c +++ b/net/smc/smc_close.c @@ -345,14 +345,7 @@ static void smc_close_passive_work(struct work_struct *work) switch (sk->sk_state) { case SMC_INIT: - if (atomic_read(&conn->bytes_to_rcv) || - (rxflags->peer_done_writing && - !smc_cdc_rxed_any_close(conn))) { - sk->sk_state = SMC_APPCLOSEWAIT1; - } else { - sk->sk_state = SMC_CLOSED; - sock_put(sk); /* passive closing */ - } + sk->sk_state = SMC_APPCLOSEWAIT1; break; case SMC_ACTIVE: sk->sk_state = SMC_APPCLOSEWAIT1; -- cgit v1.2.3 From c1f7e02979edd7a3a3e69fe04be60b1d650dc8a7 Mon Sep 17 00:00:00 2001 From: Petr Machata Date: Mon, 4 Feb 2019 14:50:38 +0000 Subject: net: cls_flower: Remove filter from mask before freeing it In fl_change(), when adding a new rule (i.e. fold == NULL), a driver may reject the new rule, for example due to resource exhaustion. By that point, the new rule was already assigned a mask, and it was added to that mask's hash table. The clean-up path that's invoked as a result of the rejection however neglects to undo the hash table addition, and proceeds to free the new rule, thus leaving a dangling pointer in the hash table. Fix by removing fnew from the mask's hash table before it is freed. Fixes: 35cc3cefc4de ("net/sched: cls_flower: Reject duplicated rules also under skip_sw") Signed-off-by: Petr Machata Acked-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index f6aa57fbbbaf..12ca9d13db83 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -1371,7 +1371,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, if (!tc_skip_hw(fnew->flags)) { err = fl_hw_replace_filter(tp, fnew, extack); if (err) - goto errout_mask; + goto errout_mask_ht; } if (!tc_in_hw(fnew->flags)) @@ -1401,6 +1401,10 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, kfree(mask); return 0; +errout_mask_ht: + rhashtable_remove_fast(&fnew->mask->ht, &fnew->ht_node, + fnew->mask->filter_ht_params); + errout_mask: fl_mask_put(head, fnew->mask, false); -- cgit v1.2.3 From 17ab4f61b8cd6f9c38e9d0b935d86d73b5d0d2b5 Mon Sep 17 00:00:00 2001 From: Rundong Ge Date: Sat, 2 Feb 2019 14:29:35 +0000 Subject: net: dsa: slave: Don't propagate flag changes on down slave interfaces The unbalance of master's promiscuity or allmulti will happen after ifdown and ifup a slave interface which is in a bridge. When we ifdown a slave interface , both the 'dsa_slave_close' and 'dsa_slave_change_rx_flags' will clear the master's flags. The flags of master will be decrease twice. In the other hand, if we ifup the slave interface again, since the slave's flags were cleared the 'dsa_slave_open' won't set the master's flag, only 'dsa_slave_change_rx_flags' that triggered by 'br_add_if' will set the master's flags. The flags of master is increase once. Only propagating flag changes when a slave interface is up makes sure this does not happen. The 'vlan_dev_change_rx_flags' had the same problem and was fixed, and changes here follows that fix. Fixes: 91da11f870f0 ("net: Distributed Switch Architecture protocol support") Signed-off-by: Rundong Ge Signed-off-by: David S. Miller --- net/dsa/slave.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/dsa/slave.c b/net/dsa/slave.c index a3fcc1d01615..b5e44825d173 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -140,11 +140,14 @@ static int dsa_slave_close(struct net_device *dev) static void dsa_slave_change_rx_flags(struct net_device *dev, int change) { struct net_device *master = dsa_slave_to_master(dev); - - if (change & IFF_ALLMULTI) - dev_set_allmulti(master, dev->flags & IFF_ALLMULTI ? 1 : -1); - if (change & IFF_PROMISC) - dev_set_promiscuity(master, dev->flags & IFF_PROMISC ? 1 : -1); + if (dev->flags & IFF_UP) { + if (change & IFF_ALLMULTI) + dev_set_allmulti(master, + dev->flags & IFF_ALLMULTI ? 1 : -1); + if (change & IFF_PROMISC) + dev_set_promiscuity(master, + dev->flags & IFF_PROMISC ? 1 : -1); + } } static void dsa_slave_set_rx_mode(struct net_device *dev) -- cgit v1.2.3 From c8101f7729daee251f4f6505f9d135ec08e1342f Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Sat, 2 Feb 2019 17:53:29 +0000 Subject: net: dsa: Fix lockdep false positive splat Creating a macvtap on a DSA-backed interface results in the following splat when lockdep is enabled: [ 19.638080] IPv6: ADDRCONF(NETDEV_CHANGE): lan0: link becomes ready [ 23.041198] device lan0 entered promiscuous mode [ 23.043445] device eth0 entered promiscuous mode [ 23.049255] [ 23.049557] ============================================ [ 23.055021] WARNING: possible recursive locking detected [ 23.060490] 5.0.0-rc3-00013-g56c857a1b8d3 #118 Not tainted [ 23.066132] -------------------------------------------- [ 23.071598] ip/2861 is trying to acquire lock: [ 23.076171] 00000000f61990cb (_xmit_ETHER){+...}, at: dev_set_rx_mode+0x1c/0x38 [ 23.083693] [ 23.083693] but task is already holding lock: [ 23.089696] 00000000ecf0c3b4 (_xmit_ETHER){+...}, at: dev_uc_add+0x24/0x70 [ 23.096774] [ 23.096774] other info that might help us debug this: [ 23.103494] Possible unsafe locking scenario: [ 23.103494] [ 23.109584] CPU0 [ 23.112093] ---- [ 23.114601] lock(_xmit_ETHER); [ 23.117917] lock(_xmit_ETHER); [ 23.121233] [ 23.121233] *** DEADLOCK *** [ 23.121233] [ 23.127325] May be due to missing lock nesting notation [ 23.127325] [ 23.134315] 2 locks held by ip/2861: [ 23.137987] #0: 000000003b766c72 (rtnl_mutex){+.+.}, at: rtnetlink_rcv_msg+0x338/0x4e0 [ 23.146231] #1: 00000000ecf0c3b4 (_xmit_ETHER){+...}, at: dev_uc_add+0x24/0x70 [ 23.153757] [ 23.153757] stack backtrace: [ 23.158243] CPU: 0 PID: 2861 Comm: ip Not tainted 5.0.0-rc3-00013-g56c857a1b8d3 #118 [ 23.166212] Hardware name: Globalscale Marvell ESPRESSOBin Board (DT) [ 23.172843] Call trace: [ 23.175358] dump_backtrace+0x0/0x188 [ 23.179116] show_stack+0x14/0x20 [ 23.182524] dump_stack+0xb4/0xec [ 23.185928] __lock_acquire+0x123c/0x1860 [ 23.190048] lock_acquire+0xc8/0x248 [ 23.193724] _raw_spin_lock_bh+0x40/0x58 [ 23.197755] dev_set_rx_mode+0x1c/0x38 [ 23.201607] dev_set_promiscuity+0x3c/0x50 [ 23.205820] dsa_slave_change_rx_flags+0x5c/0x70 [ 23.210567] __dev_set_promiscuity+0x148/0x1e0 [ 23.215136] __dev_set_rx_mode+0x74/0x98 [ 23.219167] dev_uc_add+0x54/0x70 [ 23.222575] macvlan_open+0x170/0x1d0 [ 23.226336] __dev_open+0xe0/0x160 [ 23.229830] __dev_change_flags+0x16c/0x1b8 [ 23.234132] dev_change_flags+0x20/0x60 [ 23.238074] do_setlink+0x2d0/0xc50 [ 23.241658] __rtnl_newlink+0x5f8/0x6e8 [ 23.245601] rtnl_newlink+0x50/0x78 [ 23.249184] rtnetlink_rcv_msg+0x360/0x4e0 [ 23.253397] netlink_rcv_skb+0xe8/0x130 [ 23.257338] rtnetlink_rcv+0x14/0x20 [ 23.261012] netlink_unicast+0x190/0x210 [ 23.265043] netlink_sendmsg+0x288/0x350 [ 23.269075] sock_sendmsg+0x18/0x30 [ 23.272659] ___sys_sendmsg+0x29c/0x2c8 [ 23.276602] __sys_sendmsg+0x60/0xb8 [ 23.280276] __arm64_sys_sendmsg+0x1c/0x28 [ 23.284488] el0_svc_common+0xd8/0x138 [ 23.288340] el0_svc_handler+0x24/0x80 [ 23.292192] el0_svc+0x8/0xc This looks fairly harmless (no actual deadlock occurs), and is fixed in a similar way to c6894dec8ea9 ("bridge: fix lockdep addr_list_lock false positive splat") by putting the addr_list_lock in its own lockdep class. Signed-off-by: Marc Zyngier Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/master.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/dsa/master.c b/net/dsa/master.c index 71bb15f491c8..54f5551fb799 100644 --- a/net/dsa/master.c +++ b/net/dsa/master.c @@ -205,6 +205,8 @@ static void dsa_master_reset_mtu(struct net_device *dev) rtnl_unlock(); } +static struct lock_class_key dsa_master_addr_list_lock_key; + int dsa_master_setup(struct net_device *dev, struct dsa_port *cpu_dp) { int ret; @@ -218,6 +220,8 @@ int dsa_master_setup(struct net_device *dev, struct dsa_port *cpu_dp) wmb(); dev->dsa_ptr = cpu_dp; + lockdep_set_class(&dev->addr_list_lock, + &dsa_master_addr_list_lock_key); ret = dsa_master_ethtool_setup(dev); if (ret) -- cgit v1.2.3 From 15df03c661cb362366ecfc3a21820cb934f3e4ca Mon Sep 17 00:00:00 2001 From: Eli Cooper Date: Mon, 21 Jan 2019 18:45:27 +0800 Subject: netfilter: ipv6: Don't preserve original oif for loopback address Commit 508b09046c0f ("netfilter: ipv6: Preserve link scope traffic original oif") made ip6_route_me_harder() keep the original oif for link-local and multicast packets. However, it also affected packets for the loopback address because it used rt6_need_strict(). REDIRECT rules in the OUTPUT chain rewrite the destination to loopback address; thus its oif should not be preserved. This commit fixes the bug that redirected local packets are being dropped. Actually the packet was not exactly dropped; Instead it was sent out to the original oif rather than lo. When a packet with daddr ::1 is sent to the router, it is effectively dropped. Fixes: 508b09046c0f ("netfilter: ipv6: Preserve link scope traffic original oif") Signed-off-by: Eli Cooper Signed-off-by: Pablo Neira Ayuso --- net/ipv6/netfilter.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 8b075f0bc351..6d0b1f3e927b 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -23,9 +23,11 @@ int ip6_route_me_harder(struct net *net, struct sk_buff *skb) struct sock *sk = sk_to_full_sk(skb->sk); unsigned int hh_len; struct dst_entry *dst; + int strict = (ipv6_addr_type(&iph->daddr) & + (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL)); struct flowi6 fl6 = { .flowi6_oif = sk && sk->sk_bound_dev_if ? sk->sk_bound_dev_if : - rt6_need_strict(&iph->daddr) ? skb_dst(skb)->dev->ifindex : 0, + strict ? skb_dst(skb)->dev->ifindex : 0, .flowi6_mark = skb->mark, .flowi6_uid = sock_net_uid(net, sk), .daddr = iph->daddr, -- cgit v1.2.3 From 947e492c0fc2132ae5fca081a9c2952ccaab0404 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 5 Feb 2019 12:16:18 +0100 Subject: netfilter: nft_compat: don't use refcount_inc on newly allocated entry When I moved the refcount to refcount_t type I missed the fact that refcount_inc() will result in use-after-free warning with CONFIG_REFCOUNT_FULL=y builds. The correct fix would be to init the reference count to 1 at allocation time, but, unfortunately we cannot do this, as we can't undo that in case something else fails later in the batch. So only solution I see is to special-case the 'new entry' condition and replace refcount_inc() with a "delayed" refcount_set(1) in this case, as done here. The .activate callback can be removed to simplify things, we only need to make sure that deactivate() decrements/unlinks the entry from the list at end of transaction phase (commit or abort). Fixes: 12c44aba6618 ("netfilter: nft_compat: use refcnt_t type for nft_xt reference count") Reported-by: Jordan Glover Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_compat.c | 62 +++++++++++++++++----------------------------- 1 file changed, 23 insertions(+), 39 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 0732a2fc697c..fe64df848365 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -61,6 +61,21 @@ static struct nft_compat_net *nft_compat_pernet(struct net *net) return net_generic(net, nft_compat_net_id); } +static void nft_xt_get(struct nft_xt *xt) +{ + /* refcount_inc() warns on 0 -> 1 transition, but we can't + * init the reference count to 1 in .select_ops -- we can't + * undo such an increase when another expression inside the same + * rule fails afterwards. + */ + if (xt->listcnt == 0) + refcount_set(&xt->refcnt, 1); + else + refcount_inc(&xt->refcnt); + + xt->listcnt++; +} + static bool nft_xt_put(struct nft_xt *xt) { if (refcount_dec_and_test(&xt->refcnt)) { @@ -291,7 +306,7 @@ nft_target_init(const struct nft_ctx *ctx, const struct nft_expr *expr, return -EINVAL; nft_xt = container_of(expr->ops, struct nft_xt, ops); - refcount_inc(&nft_xt->refcnt); + nft_xt_get(nft_xt); return 0; } @@ -504,7 +519,7 @@ __nft_match_init(const struct nft_ctx *ctx, const struct nft_expr *expr, return ret; nft_xt = container_of(expr->ops, struct nft_xt, ops); - refcount_inc(&nft_xt->refcnt); + nft_xt_get(nft_xt); return 0; } @@ -558,45 +573,16 @@ nft_match_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) __nft_match_destroy(ctx, expr, nft_expr_priv(expr)); } -static void nft_compat_activate(const struct nft_ctx *ctx, - const struct nft_expr *expr, - struct list_head *h) -{ - struct nft_xt *xt = container_of(expr->ops, struct nft_xt, ops); - - if (xt->listcnt == 0) - list_add(&xt->head, h); - - xt->listcnt++; -} - -static void nft_compat_activate_mt(const struct nft_ctx *ctx, - const struct nft_expr *expr) -{ - struct nft_compat_net *cn = nft_compat_pernet(ctx->net); - - nft_compat_activate(ctx, expr, &cn->nft_match_list); -} - -static void nft_compat_activate_tg(const struct nft_ctx *ctx, - const struct nft_expr *expr) -{ - struct nft_compat_net *cn = nft_compat_pernet(ctx->net); - - nft_compat_activate(ctx, expr, &cn->nft_target_list); -} - static void nft_compat_deactivate(const struct nft_ctx *ctx, const struct nft_expr *expr, enum nft_trans_phase phase) { struct nft_xt *xt = container_of(expr->ops, struct nft_xt, ops); - if (phase == NFT_TRANS_COMMIT) - return; - - if (--xt->listcnt == 0) - list_del_init(&xt->head); + if (phase == NFT_TRANS_ABORT || phase == NFT_TRANS_COMMIT) { + if (--xt->listcnt == 0) + list_del_init(&xt->head); + } } static void @@ -852,7 +838,6 @@ nft_match_select_ops(const struct nft_ctx *ctx, nft_match->ops.eval = nft_match_eval; nft_match->ops.init = nft_match_init; nft_match->ops.destroy = nft_match_destroy; - nft_match->ops.activate = nft_compat_activate_mt; nft_match->ops.deactivate = nft_compat_deactivate; nft_match->ops.dump = nft_match_dump; nft_match->ops.validate = nft_match_validate; @@ -870,7 +855,7 @@ nft_match_select_ops(const struct nft_ctx *ctx, nft_match->ops.size = matchsize; - nft_match->listcnt = 1; + nft_match->listcnt = 0; list_add(&nft_match->head, &cn->nft_match_list); return &nft_match->ops; @@ -957,7 +942,6 @@ nft_target_select_ops(const struct nft_ctx *ctx, nft_target->ops.size = NFT_EXPR_SIZE(XT_ALIGN(target->targetsize)); nft_target->ops.init = nft_target_init; nft_target->ops.destroy = nft_target_destroy; - nft_target->ops.activate = nft_compat_activate_tg; nft_target->ops.deactivate = nft_compat_deactivate; nft_target->ops.dump = nft_target_dump; nft_target->ops.validate = nft_target_validate; @@ -968,7 +952,7 @@ nft_target_select_ops(const struct nft_ctx *ctx, else nft_target->ops.eval = nft_target_eval_xt; - nft_target->listcnt = 1; + nft_target->listcnt = 0; list_add(&nft_target->head, &cn->nft_target_list); return &nft_target->ops; -- cgit v1.2.3 From 6dce3c20ac429e7a651d728e375853370c796e8d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 4 Feb 2019 08:36:06 -0800 Subject: rxrpc: bad unlock balance in rxrpc_recvmsg When either "goto wait_interrupted;" or "goto wait_error;" paths are taken, socket lock has already been released. This patch fixes following syzbot splat : WARNING: bad unlock balance detected! 5.0.0-rc4+ #59 Not tainted ------------------------------------- syz-executor223/8256 is trying to release lock (sk_lock-AF_RXRPC) at: [] rxrpc_recvmsg+0x6d3/0x3099 net/rxrpc/recvmsg.c:598 but there are no more locks to release! other info that might help us debug this: 1 lock held by syz-executor223/8256: #0: 00000000fa9ed0f4 (slock-AF_RXRPC){+...}, at: spin_lock_bh include/linux/spinlock.h:334 [inline] #0: 00000000fa9ed0f4 (slock-AF_RXRPC){+...}, at: release_sock+0x20/0x1c0 net/core/sock.c:2798 stack backtrace: CPU: 1 PID: 8256 Comm: syz-executor223 Not tainted 5.0.0-rc4+ #59 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+0x172/0x1f0 lib/dump_stack.c:113 print_unlock_imbalance_bug kernel/locking/lockdep.c:3391 [inline] print_unlock_imbalance_bug.cold+0x114/0x123 kernel/locking/lockdep.c:3368 __lock_release kernel/locking/lockdep.c:3601 [inline] lock_release+0x67e/0xa00 kernel/locking/lockdep.c:3860 sock_release_ownership include/net/sock.h:1471 [inline] release_sock+0x183/0x1c0 net/core/sock.c:2808 rxrpc_recvmsg+0x6d3/0x3099 net/rxrpc/recvmsg.c:598 sock_recvmsg_nosec net/socket.c:794 [inline] sock_recvmsg net/socket.c:801 [inline] sock_recvmsg+0xd0/0x110 net/socket.c:797 __sys_recvfrom+0x1ff/0x350 net/socket.c:1845 __do_sys_recvfrom net/socket.c:1863 [inline] __se_sys_recvfrom net/socket.c:1859 [inline] __x64_sys_recvfrom+0xe1/0x1a0 net/socket.c:1859 do_syscall_64+0x103/0x610 arch/x86/entry/common.c:290 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x446379 Code: e8 2c b3 02 00 48 83 c4 18 c3 0f 1f 80 00 00 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 2b 09 fc ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007fe5da89fd98 EFLAGS: 00000246 ORIG_RAX: 000000000000002d RAX: ffffffffffffffda RBX: 00000000006dbc28 RCX: 0000000000446379 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003 RBP: 00000000006dbc20 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000006dbc2c R13: 0000000000000000 R14: 0000000000000000 R15: 20c49ba5e353f7cf Fixes: 248f219cb8bc ("rxrpc: Rewrite the data and ack handling code") Signed-off-by: Eric Dumazet Cc: David Howells Reported-by: syzbot Signed-off-by: David S. Miller --- net/rxrpc/recvmsg.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index eaf19ebaa964..3f7bb11f3290 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -596,6 +596,7 @@ error_requeue_call: } error_no_call: release_sock(&rx->sk); +error_trace: trace_rxrpc_recvmsg(call, rxrpc_recvmsg_return, 0, 0, 0, ret); return ret; @@ -604,7 +605,7 @@ wait_interrupted: wait_error: finish_wait(sk_sleep(&rx->sk), &wait); call = NULL; - goto error_no_call; + goto error_trace; } /** -- cgit v1.2.3 From e248aa7be86e8179f20ac0931774ecd746f3f5bf Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 25 Jan 2019 16:54:54 -0500 Subject: svcrdma: Remove max_sge check at connect time Two and a half years ago, the client was changed to use gathered Send for larger inline messages, in commit 655fec6987b ("xprtrdma: Use gathered Send for large inline messages"). Several fixes were required because there are a few in-kernel device drivers whose max_sge is 3, and these were broken by the change. Apparently my memory is going, because some time later, I submitted commit 25fd86eca11c ("svcrdma: Don't overrun the SGE array in svc_rdma_send_ctxt"), and after that, commit f3c1fd0ee294 ("svcrdma: Reduce max_send_sges"). These too incorrectly assumed in-kernel device drivers would have more than a few Send SGEs available. The fix for the server side is not the same. This is because the fundamental problem on the server is that, whether or not the client has provisioned a chunk for the RPC reply, the server must squeeze even the most complex RPC replies into a single RDMA Send. Failing in the send path because of Send SGE exhaustion should never be an option. Therefore, instead of failing when the send path runs out of SGEs, switch to using a bounce buffer mechanism to handle RPC replies that are too complex for the device to send directly. That allows us to remove the max_sge check to enable drivers with small max_sge to work again. Reported-by: Don Dutile Fixes: 25fd86eca11c ("svcrdma: Don't overrun the SGE array in ...") Cc: stable@vger.kernel.org Signed-off-by: Chuck Lever Signed-off-by: J. Bruce Fields --- net/sunrpc/xprtrdma/svc_rdma_sendto.c | 105 +++++++++++++++++++++++++++++-- net/sunrpc/xprtrdma/svc_rdma_transport.c | 9 +-- 2 files changed, 102 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c index cf51b8f9b15f..1f200119268c 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c +++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c @@ -537,6 +537,99 @@ void svc_rdma_sync_reply_hdr(struct svcxprt_rdma *rdma, DMA_TO_DEVICE); } +/* If the xdr_buf has more elements than the device can + * transmit in a single RDMA Send, then the reply will + * have to be copied into a bounce buffer. + */ +static bool svc_rdma_pull_up_needed(struct svcxprt_rdma *rdma, + struct xdr_buf *xdr, + __be32 *wr_lst) +{ + int elements; + + /* xdr->head */ + elements = 1; + + /* xdr->pages */ + if (!wr_lst) { + unsigned int remaining; + unsigned long pageoff; + + pageoff = xdr->page_base & ~PAGE_MASK; + remaining = xdr->page_len; + while (remaining) { + ++elements; + remaining -= min_t(u32, PAGE_SIZE - pageoff, + remaining); + pageoff = 0; + } + } + + /* xdr->tail */ + if (xdr->tail[0].iov_len) + ++elements; + + /* assume 1 SGE is needed for the transport header */ + return elements >= rdma->sc_max_send_sges; +} + +/* The device is not capable of sending the reply directly. + * Assemble the elements of @xdr into the transport header + * buffer. + */ +static int svc_rdma_pull_up_reply_msg(struct svcxprt_rdma *rdma, + struct svc_rdma_send_ctxt *ctxt, + struct xdr_buf *xdr, __be32 *wr_lst) +{ + unsigned char *dst, *tailbase; + unsigned int taillen; + + dst = ctxt->sc_xprt_buf; + dst += ctxt->sc_sges[0].length; + + memcpy(dst, xdr->head[0].iov_base, xdr->head[0].iov_len); + dst += xdr->head[0].iov_len; + + tailbase = xdr->tail[0].iov_base; + taillen = xdr->tail[0].iov_len; + if (wr_lst) { + u32 xdrpad; + + xdrpad = xdr_padsize(xdr->page_len); + if (taillen && xdrpad) { + tailbase += xdrpad; + taillen -= xdrpad; + } + } else { + unsigned int len, remaining; + unsigned long pageoff; + struct page **ppages; + + ppages = xdr->pages + (xdr->page_base >> PAGE_SHIFT); + pageoff = xdr->page_base & ~PAGE_MASK; + remaining = xdr->page_len; + while (remaining) { + len = min_t(u32, PAGE_SIZE - pageoff, remaining); + + memcpy(dst, page_address(*ppages), len); + remaining -= len; + dst += len; + pageoff = 0; + } + } + + if (taillen) + memcpy(dst, tailbase, taillen); + + ctxt->sc_sges[0].length += xdr->len; + ib_dma_sync_single_for_device(rdma->sc_pd->device, + ctxt->sc_sges[0].addr, + ctxt->sc_sges[0].length, + DMA_TO_DEVICE); + + return 0; +} + /* svc_rdma_map_reply_msg - Map the buffer holding RPC message * @rdma: controlling transport * @ctxt: send_ctxt for the Send WR @@ -559,8 +652,10 @@ int svc_rdma_map_reply_msg(struct svcxprt_rdma *rdma, u32 xdr_pad; int ret; - if (++ctxt->sc_cur_sge_no >= rdma->sc_max_send_sges) - return -EIO; + if (svc_rdma_pull_up_needed(rdma, xdr, wr_lst)) + return svc_rdma_pull_up_reply_msg(rdma, ctxt, xdr, wr_lst); + + ++ctxt->sc_cur_sge_no; ret = svc_rdma_dma_map_buf(rdma, ctxt, xdr->head[0].iov_base, xdr->head[0].iov_len); @@ -591,8 +686,7 @@ int svc_rdma_map_reply_msg(struct svcxprt_rdma *rdma, while (remaining) { len = min_t(u32, PAGE_SIZE - page_off, remaining); - if (++ctxt->sc_cur_sge_no >= rdma->sc_max_send_sges) - return -EIO; + ++ctxt->sc_cur_sge_no; ret = svc_rdma_dma_map_page(rdma, ctxt, *ppages++, page_off, len); if (ret < 0) @@ -606,8 +700,7 @@ int svc_rdma_map_reply_msg(struct svcxprt_rdma *rdma, len = xdr->tail[0].iov_len; tail: if (len) { - if (++ctxt->sc_cur_sge_no >= rdma->sc_max_send_sges) - return -EIO; + ++ctxt->sc_cur_sge_no; ret = svc_rdma_dma_map_buf(rdma, ctxt, base, len); if (ret < 0) return ret; diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c index 924c17d46903..57f86c63a463 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c @@ -419,12 +419,9 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt) /* Transport header, head iovec, tail iovec */ newxprt->sc_max_send_sges = 3; /* Add one SGE per page list entry */ - newxprt->sc_max_send_sges += svcrdma_max_req_size / PAGE_SIZE; - if (newxprt->sc_max_send_sges > dev->attrs.max_send_sge) { - pr_err("svcrdma: too few Send SGEs available (%d needed)\n", - newxprt->sc_max_send_sges); - goto errout; - } + newxprt->sc_max_send_sges += (svcrdma_max_req_size / PAGE_SIZE) + 1; + if (newxprt->sc_max_send_sges > dev->attrs.max_send_sge) + newxprt->sc_max_send_sges = dev->attrs.max_send_sge; newxprt->sc_max_req_size = svcrdma_max_req_size; newxprt->sc_max_requests = svcrdma_max_requests; newxprt->sc_max_bc_requests = svcrdma_max_bc_requests; -- cgit v1.2.3 From 00670cb8a73b10b10d3c40f045c15411715e4465 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 6 Feb 2019 18:35:15 +0300 Subject: net: dsa: Fix NULL checking in dsa_slave_set_eee() This function can't succeed if dp->pl is NULL. It will Oops inside the call to return phylink_ethtool_get_eee(dp->pl, e); Fixes: 1be52e97ed3e ("dsa: slave: eee: Allow ports to use phylink") Signed-off-by: Dan Carpenter Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller --- net/dsa/slave.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/dsa/slave.c b/net/dsa/slave.c index b5e44825d173..a1c9fe155057 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -642,7 +642,7 @@ static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e) int ret; /* Port's PHY and MAC both need to be EEE capable */ - if (!dev->phydev && !dp->pl) + if (!dev->phydev || !dp->pl) return -ENODEV; if (!ds->ops->set_mac_eee) @@ -662,7 +662,7 @@ static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e) int ret; /* Port's PHY and MAC both need to be EEE capable */ - if (!dev->phydev && !dp->pl) + if (!dev->phydev || !dp->pl) return -ENODEV; if (!ds->ops->get_mac_eee) -- cgit v1.2.3 From 173656accaf583698bac3f9e269884ba60d51ef4 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Thu, 7 Feb 2019 18:36:11 +0800 Subject: sit: check if IPv6 enabled before calling ip6_err_gen_icmpv6_unreach() If we disabled IPv6 from the kernel command line (ipv6.disable=1), we should not call ip6_err_gen_icmpv6_unreach(). This: ip link add sit1 type sit local 192.0.2.1 remote 192.0.2.2 ttl 1 ip link set sit1 up ip addr add 198.51.100.1/24 dev sit1 ping 198.51.100.2 if IPv6 is disabled at boot time, will crash the kernel. v2: there's no need to use in6_dev_get(), use __in6_dev_get() instead, as we only need to check that idev exists and we are under rcu_read_lock() (from netif_receive_skb_internal()). Reported-by: Jianlin Shi Fixes: ca15a078bd90 ("sit: generate icmpv6 error when receiving icmpv4 error") Cc: Oussama Ghorbel Signed-off-by: Hangbin Liu Reviewed-by: Stefano Brivio Signed-off-by: David S. Miller --- net/ipv6/sit.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 1e03305c0549..e8a1dabef803 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -546,7 +546,8 @@ static int ipip6_err(struct sk_buff *skb, u32 info) } err = 0; - if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type, data_len)) + if (__in6_dev_get(skb->dev) && + !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type, data_len)) goto out; if (t->parms.iph.daddr == 0) -- cgit v1.2.3