From f5980d099295d4113635ded88425834a38d17125 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Thu, 24 Jul 2014 18:54:47 +0300 Subject: [PATCH 01/25] bnx2x: fix crash during TSO tunneling [ Upstream commit fe26566d8a05151ba1dce75081f6270f73ec4ae1 ] When TSO packet is transmitted additional BD w/o mapping is used to describe the packed. The BD needs special handling in tx completion. kernel: Call Trace: kernel: [] dump_stack+0x19/0x1b kernel: [] warn_slowpath_common+0x61/0x80 kernel: [] warn_slowpath_fmt+0x5c/0x80 kernel: [] ? find_iova+0x4d/0x90 kernel: [] intel_unmap_page.part.36+0x142/0x160 kernel: [] intel_unmap_page+0x26/0x30 kernel: [] bnx2x_free_tx_pkt+0x157/0x2b0 [bnx2x] kernel: [] bnx2x_tx_int+0xac/0x220 [bnx2x] kernel: [] ? read_tsc+0x9/0x20 kernel: [] bnx2x_poll+0xbb/0x3c0 [bnx2x] kernel: [] net_rx_action+0x15a/0x250 kernel: [] __do_softirq+0xf7/0x290 kernel: [] call_softirq+0x1c/0x30 kernel: [] do_softirq+0x55/0x90 kernel: [] irq_exit+0x115/0x120 kernel: [] do_IRQ+0x58/0xf0 kernel: [] common_interrupt+0x6d/0x6d kernel: [] ? clockevents_notify+0x127/0x140 kernel: [] ? cpuidle_enter_state+0x4f/0xc0 kernel: [] cpuidle_idle_call+0xc5/0x200 kernel: [] arch_cpu_idle+0xe/0x30 kernel: [] cpu_startup_entry+0xf5/0x290 kernel: [] start_secondary+0x265/0x27b kernel: ---[ end trace 11aa7726f18d7e80 ]--- Fixes: a848ade408b ("bnx2x: add CSUM and TSO support for encapsulation protocols") Reported-by: Yulong Pei Cc: Michal Schmidt Signed-off-by: Dmitry Kravkov Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 + drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 9 +++++++++ 2 files changed, 10 insertions(+) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 3dba2a70a00e..ec86177be1df 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -312,6 +312,7 @@ struct sw_tx_bd { u8 flags; /* Set on the first BD descriptor when there is a split BD */ #define BNX2X_TSO_SPLIT_BD (1<<0) +#define BNX2X_HAS_SECOND_PBD (1<<1) }; struct sw_rx_page { diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index b04f7f128f49..372a7557e1fa 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -180,6 +180,12 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fp_txdata *txdata, --nbd; bd_idx = TX_BD(NEXT_TX_IDX(bd_idx)); + if (tx_buf->flags & BNX2X_HAS_SECOND_PBD) { + /* Skip second parse bd... */ + --nbd; + bd_idx = TX_BD(NEXT_TX_IDX(bd_idx)); + } + /* TSO headers+data bds share a common mapping. See bnx2x_tx_split() */ if (tx_buf->flags & BNX2X_TSO_SPLIT_BD) { tx_data_bd = &txdata->tx_desc_ring[bd_idx].reg_bd; @@ -3755,6 +3761,9 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev) /* set encapsulation flag in start BD */ SET_FLAG(tx_start_bd->general_data, ETH_TX_START_BD_TUNNEL_EXIST, 1); + + tx_buf->flags |= BNX2X_HAS_SECOND_PBD; + nbd++; } else if (xmit_type & XMIT_CSUM) { /* Set PBD in checksum offload case w/o encapsulation */ From ff1f69a89a613223c57c13190a6c9be928ac4b9d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 2 Jun 2014 05:26:03 -0700 Subject: [PATCH 02/25] inetpeer: get rid of ip_id_count [ Upstream commit 73f156a6e8c1074ac6327e0abd1169e95eb66463 ] Ideally, we would need to generate IP ID using a per destination IP generator. linux kernels used inet_peer cache for this purpose, but this had a huge cost on servers disabling MTU discovery. 1) each inet_peer struct consumes 192 bytes 2) inetpeer cache uses a binary tree of inet_peer structs, with a nominal size of ~66000 elements under load. 3) lookups in this tree are hitting a lot of cache lines, as tree depth is about 20. 4) If server deals with many tcp flows, we have a high probability of not finding the inet_peer, allocating a fresh one, inserting it in the tree with same initial ip_id_count, (cf secure_ip_id()) 5) We garbage collect inet_peer aggressively. IP ID generation do not have to be 'perfect' Goal is trying to avoid duplicates in a short period of time, so that reassembly units have a chance to complete reassembly of fragments belonging to one message before receiving other fragments with a recycled ID. We simply use an array of generators, and a Jenkin hash using the dst IP as a key. ipv6_select_ident() is put back into net/ipv6/ip6_output.c where it belongs (it is only used from this file) secure_ip_id() and secure_ipv6_id() no longer are needed. Rename ip_select_ident_more() to ip_select_ident_segs() to avoid unnecessary decrement/increment of the number of segments. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/ppp/pptp.c | 2 +- drivers/net/vxlan.c | 2 +- include/net/inetpeer.h | 16 ++--------- include/net/ip.h | 40 +++++++++++++++------------ include/net/ipv6.h | 11 +++++--- include/net/secure_seq.h | 2 -- net/core/secure_seq.c | 25 ----------------- net/ipv4/igmp.c | 4 +-- net/ipv4/inetpeer.c | 18 ------------ net/ipv4/ip_output.c | 7 ++--- net/ipv4/ip_tunnel.c | 2 +- net/ipv4/ipmr.c | 2 +- net/ipv4/raw.c | 2 +- net/ipv4/route.c | 49 ++++++++++++++------------------- net/ipv4/xfrm4_mode_tunnel.c | 2 +- net/ipv6/ip6_output.c | 15 ++++++++++ net/ipv6/output_core.c | 23 ---------------- net/ipv6/sit.c | 2 +- net/netfilter/ipvs/ip_vs_xmit.c | 2 +- 19 files changed, 81 insertions(+), 145 deletions(-) diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c index 7f10588fe668..8161c3f066a3 100644 --- a/drivers/net/ppp/pptp.c +++ b/drivers/net/ppp/pptp.c @@ -281,7 +281,7 @@ static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb) nf_reset(skb); skb->ip_summed = CHECKSUM_NONE; - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); ip_send_check(iph); ip_local_out(skb); diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index fcbd4eee52cc..a1dc186c6f66 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -1093,7 +1093,7 @@ static netdev_tx_t vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, iph->daddr = dst; iph->saddr = fl4.saddr; iph->ttl = ttl ? : ip4_dst_hoplimit(&rt->dst); - __ip_select_ident(iph, &rt->dst, (skb_shinfo(skb)->gso_segs ?: 1) - 1); + __ip_select_ident(iph, skb_shinfo(skb)->gso_segs ?: 1); nf_reset(skb); diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h index 6ca347a0717e..bb06fd26a7bd 100644 --- a/include/net/inetpeer.h +++ b/include/net/inetpeer.h @@ -41,14 +41,13 @@ struct inet_peer { struct rcu_head gc_rcu; }; /* - * Once inet_peer is queued for deletion (refcnt == -1), following fields - * are not available: rid, ip_id_count + * Once inet_peer is queued for deletion (refcnt == -1), following field + * is not available: rid * We can share memory with rcu_head to help keep inet_peer small. */ union { struct { atomic_t rid; /* Frag reception counter */ - atomic_t ip_id_count; /* IP ID for the next packet */ }; struct rcu_head rcu; struct inet_peer *gc_next; @@ -166,7 +165,7 @@ extern void inetpeer_invalidate_tree(struct inet_peer_base *); extern void inetpeer_invalidate_family(int family); /* - * temporary check to make sure we dont access rid, ip_id_count, tcp_ts, + * temporary check to make sure we dont access rid, tcp_ts, * tcp_ts_stamp if no refcount is taken on inet_peer */ static inline void inet_peer_refcheck(const struct inet_peer *p) @@ -174,13 +173,4 @@ static inline void inet_peer_refcheck(const struct inet_peer *p) WARN_ON_ONCE(atomic_read(&p->refcnt) <= 0); } - -/* can be called with or without local BH being disabled */ -static inline int inet_getid(struct inet_peer *p, int more) -{ - more++; - inet_peer_refcheck(p); - return atomic_add_return(more, &p->ip_id_count) - more; -} - #endif /* _NET_INETPEER_H */ diff --git a/include/net/ip.h b/include/net/ip.h index 788f1d8a796f..dd72c8f93797 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -252,9 +252,19 @@ int ip_dont_fragment(struct sock *sk, struct dst_entry *dst) !(dst_metric_locked(dst, RTAX_MTU))); } -extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more); +#define IP_IDENTS_SZ 2048u +extern atomic_t *ip_idents; -static inline void ip_select_ident(struct sk_buff *skb, struct dst_entry *dst, struct sock *sk) +static inline u32 ip_idents_reserve(u32 hash, int segs) +{ + atomic_t *id_ptr = ip_idents + hash % IP_IDENTS_SZ; + + return atomic_add_return(segs, id_ptr) - segs; +} + +void __ip_select_ident(struct iphdr *iph, int segs); + +static inline void ip_select_ident_segs(struct sk_buff *skb, struct sock *sk, int segs) { struct iphdr *iph = ip_hdr(skb); @@ -264,24 +274,20 @@ static inline void ip_select_ident(struct sk_buff *skb, struct dst_entry *dst, s * does not change, they drop every other packet in * a TCP stream using header compression. */ - iph->id = (sk && inet_sk(sk)->inet_daddr) ? - htons(inet_sk(sk)->inet_id++) : 0; - } else - __ip_select_ident(iph, dst, 0); -} - -static inline void ip_select_ident_more(struct sk_buff *skb, struct dst_entry *dst, struct sock *sk, int more) -{ - struct iphdr *iph = ip_hdr(skb); - - if ((iph->frag_off & htons(IP_DF)) && !skb->local_df) { if (sk && inet_sk(sk)->inet_daddr) { iph->id = htons(inet_sk(sk)->inet_id); - inet_sk(sk)->inet_id += 1 + more; - } else + inet_sk(sk)->inet_id += segs; + } else { iph->id = 0; - } else - __ip_select_ident(iph, dst, more); + } + } else { + __ip_select_ident(iph, segs); + } +} + +static inline void ip_select_ident(struct sk_buff *skb, struct sock *sk) +{ + ip_select_ident_segs(skb, sk, 1); } /* diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 9e093fc33dab..087370ff05f1 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -530,14 +530,19 @@ static inline u32 ipv6_addr_hash(const struct in6_addr *a) } /* more secured version of ipv6_addr_hash() */ -static inline u32 ipv6_addr_jhash(const struct in6_addr *a) +static inline u32 __ipv6_addr_jhash(const struct in6_addr *a, const u32 initval) { u32 v = (__force u32)a->s6_addr32[0] ^ (__force u32)a->s6_addr32[1]; return jhash_3words(v, (__force u32)a->s6_addr32[2], (__force u32)a->s6_addr32[3], - ipv6_hash_secret); + initval); +} + +static inline u32 ipv6_addr_jhash(const struct in6_addr *a) +{ + return __ipv6_addr_jhash(a, ipv6_hash_secret); } static inline bool ipv6_addr_loopback(const struct in6_addr *a) @@ -649,8 +654,6 @@ static inline int ipv6_addr_diff(const struct in6_addr *a1, const struct in6_add return __ipv6_addr_diff(a1, a2, sizeof(struct in6_addr)); } -extern void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt); - /* * Header manipulation */ diff --git a/include/net/secure_seq.h b/include/net/secure_seq.h index c2e542b27a5a..b1c3d1c63c4e 100644 --- a/include/net/secure_seq.h +++ b/include/net/secure_seq.h @@ -3,8 +3,6 @@ #include -extern __u32 secure_ip_id(__be32 daddr); -extern __u32 secure_ipv6_id(const __be32 daddr[4]); extern u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport); extern u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, __be16 dport); diff --git a/net/core/secure_seq.c b/net/core/secure_seq.c index 8d9d05edd2eb..d0afc322b961 100644 --- a/net/core/secure_seq.c +++ b/net/core/secure_seq.c @@ -95,31 +95,6 @@ EXPORT_SYMBOL(secure_ipv6_port_ephemeral); #endif #ifdef CONFIG_INET -__u32 secure_ip_id(__be32 daddr) -{ - u32 hash[MD5_DIGEST_WORDS]; - - net_secret_init(); - hash[0] = (__force __u32) daddr; - hash[1] = net_secret[13]; - hash[2] = net_secret[14]; - hash[3] = net_secret[15]; - - md5_transform(hash, net_secret); - - return hash[0]; -} - -__u32 secure_ipv6_id(const __be32 daddr[4]) -{ - __u32 hash[4]; - - net_secret_init(); - memcpy(hash, daddr, 16); - md5_transform(hash, net_secret); - - return hash[0]; -} __u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 38d63ca8a6b5..155adf8729c2 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -343,7 +343,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size) pip->saddr = fl4.saddr; pip->protocol = IPPROTO_IGMP; pip->tot_len = 0; /* filled in later */ - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); ((u8 *)&pip[1])[0] = IPOPT_RA; ((u8 *)&pip[1])[1] = 4; ((u8 *)&pip[1])[2] = 0; @@ -687,7 +687,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, iph->daddr = dst; iph->saddr = fl4.saddr; iph->protocol = IPPROTO_IGMP; - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); ((u8 *)&iph[1])[0] = IPOPT_RA; ((u8 *)&iph[1])[1] = 4; ((u8 *)&iph[1])[2] = 0; diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index 33d5537881ed..67140efc15fd 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -26,20 +26,7 @@ * Theory of operations. * We keep one entry for each peer IP address. The nodes contains long-living * information about the peer which doesn't depend on routes. - * At this moment this information consists only of ID field for the next - * outgoing IP packet. This field is incremented with each packet as encoded - * in inet_getid() function (include/net/inetpeer.h). - * At the moment of writing this notes identifier of IP packets is generated - * to be unpredictable using this code only for packets subjected - * (actually or potentially) to defragmentation. I.e. DF packets less than - * PMTU in size when local fragmentation is disabled use a constant ID and do - * not use this code (see ip_select_ident() in include/net/ip.h). * - * Route cache entries hold references to our nodes. - * New cache entries get references via lookup by destination IP address in - * the avl tree. The reference is grabbed only when it's needed i.e. only - * when we try to output IP packet which needs an unpredictable ID (see - * __ip_select_ident() in net/ipv4/route.c). * Nodes are removed only when reference counter goes to 0. * When it's happened the node may be removed when a sufficient amount of * time has been passed since its last use. The less-recently-used entry can @@ -62,7 +49,6 @@ * refcnt: atomically against modifications on other CPU; * usually under some other lock to prevent node disappearing * daddr: unchangeable - * ip_id_count: atomic value (no lock needed) */ static struct kmem_cache *peer_cachep __read_mostly; @@ -504,10 +490,6 @@ relookup: p->daddr = *daddr; atomic_set(&p->refcnt, 1); atomic_set(&p->rid, 0); - atomic_set(&p->ip_id_count, - (daddr->family == AF_INET) ? - secure_ip_id(daddr->addr.a4) : - secure_ipv6_id(daddr->addr.a6)); p->metrics[RTAX_LOCK-1] = INETPEER_METRICS_NEW; p->rate_tokens = 0; /* 60*HZ is arbitrary, but chosen enough high so that the first diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 6ca5873d6175..5afbbbe03b0e 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -148,7 +148,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, iph->daddr = (opt && opt->opt.srr ? opt->opt.faddr : daddr); iph->saddr = saddr; iph->protocol = sk->sk_protocol; - ip_select_ident(skb, &rt->dst, sk); + ip_select_ident(skb, sk); if (opt && opt->opt.optlen) { iph->ihl += opt->opt.optlen>>2; @@ -394,8 +394,7 @@ packet_routed: ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt, 0); } - ip_select_ident_more(skb, &rt->dst, sk, - (skb_shinfo(skb)->gso_segs ?: 1) - 1); + ip_select_ident_segs(skb, sk, skb_shinfo(skb)->gso_segs ?: 1); skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; @@ -1332,7 +1331,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk, iph->ttl = ttl; iph->protocol = sk->sk_protocol; ip_copy_addrs(iph, fl4); - ip_select_ident(skb, &rt->dst, sk); + ip_select_ident(skb, sk); if (opt) { iph->ihl += opt->optlen>>2; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 5642374cb751..84aa69caee59 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -691,7 +691,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, iph->daddr = fl4.daddr; iph->saddr = fl4.saddr; iph->ttl = ttl; - __ip_select_ident(iph, &rt->dst, (skb_shinfo(skb)->gso_segs ?: 1) - 1); + __ip_select_ident(iph, skb_shinfo(skb)->gso_segs ?: 1); iptunnel_xmit(skb, dev); return; diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 49797ed0917c..56d079b63ad3 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1661,7 +1661,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) iph->protocol = IPPROTO_IPIP; iph->ihl = 5; iph->tot_len = htons(skb->len); - ip_select_ident(skb, skb_dst(skb), NULL); + ip_select_ident(skb, NULL); ip_send_check(iph); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 402870fdfa0e..b4a1c42a627f 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -387,7 +387,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, iph->check = 0; iph->tot_len = htons(length); if (!iph->id) - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 2b9887becb5c..da2d4efc0bf7 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -89,6 +89,7 @@ #include #include #include +#include #include #include #include @@ -464,39 +465,23 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst, return neigh_create(&arp_tbl, pkey, dev); } -/* - * Peer allocation may fail only in serious out-of-memory conditions. However - * we still can generate some output. - * Random ID selection looks a bit dangerous because we have no chances to - * select ID being unique in a reasonable period of time. - * But broken packet identifier may be better than no packet at all. - */ -static void ip_select_fb_ident(struct iphdr *iph) +atomic_t *ip_idents __read_mostly; +EXPORT_SYMBOL(ip_idents); + +void __ip_select_ident(struct iphdr *iph, int segs) { - static DEFINE_SPINLOCK(ip_fb_id_lock); - static u32 ip_fallback_id; - u32 salt; + static u32 ip_idents_hashrnd __read_mostly; + static bool hashrnd_initialized = false; + u32 hash, id; - spin_lock_bh(&ip_fb_id_lock); - salt = secure_ip_id((__force __be32)ip_fallback_id ^ iph->daddr); - iph->id = htons(salt & 0xFFFF); - ip_fallback_id = salt; - spin_unlock_bh(&ip_fb_id_lock); -} - -void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more) -{ - struct net *net = dev_net(dst->dev); - struct inet_peer *peer; - - peer = inet_getpeer_v4(net->ipv4.peers, iph->daddr, 1); - if (peer) { - iph->id = htons(inet_getid(peer, more)); - inet_putpeer(peer); - return; + if (unlikely(!hashrnd_initialized)) { + hashrnd_initialized = true; + get_random_bytes(&ip_idents_hashrnd, sizeof(ip_idents_hashrnd)); } - ip_select_fb_ident(iph); + hash = jhash_1word((__force u32)iph->daddr, ip_idents_hashrnd); + id = ip_idents_reserve(hash, segs); + iph->id = htons(id); } EXPORT_SYMBOL(__ip_select_ident); @@ -2656,6 +2641,12 @@ int __init ip_rt_init(void) { int rc = 0; + ip_idents = kmalloc(IP_IDENTS_SZ * sizeof(*ip_idents), GFP_KERNEL); + if (!ip_idents) + panic("IP: failed to allocate ip_idents\n"); + + prandom_bytes(ip_idents, IP_IDENTS_SZ * sizeof(*ip_idents)); + #ifdef CONFIG_IP_ROUTE_CLASSID ip_rt_acct = __alloc_percpu(256 * sizeof(struct ip_rt_acct), __alignof__(struct ip_rt_acct)); if (!ip_rt_acct) diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c index b5663c37f089..e3f64831bc36 100644 --- a/net/ipv4/xfrm4_mode_tunnel.c +++ b/net/ipv4/xfrm4_mode_tunnel.c @@ -117,12 +117,12 @@ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ? 0 : (XFRM_MODE_SKB_CB(skb)->frag_off & htons(IP_DF)); - ip_select_ident(skb, dst->child, NULL); top_iph->ttl = ip4_dst_hoplimit(dst->child); top_iph->saddr = x->props.saddr.a4; top_iph->daddr = x->id.daddr.a4; + ip_select_ident(skb, NULL); return 0; } diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index ffa8d295c56c..a425052ac274 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -540,6 +540,21 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) skb_copy_secmark(to, from); } +static void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt) +{ + static u32 ip6_idents_hashrnd __read_mostly; + static bool hashrnd_initialized = false; + u32 hash, id; + + if (unlikely(!hashrnd_initialized)) { + hashrnd_initialized = true; + get_random_bytes(&ip6_idents_hashrnd, sizeof(ip6_idents_hashrnd)); + } + hash = __ipv6_addr_jhash(&rt->rt6i_dst.addr, ip6_idents_hashrnd); + id = ip_idents_reserve(hash, 1); + fhdr->identification = htonl(id); +} + int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) { struct sk_buff *frag; diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index 3d2c81a66d6a..a5d465105b69 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -6,29 +6,6 @@ #include #include -void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt) -{ - static atomic_t ipv6_fragmentation_id; - int ident; - -#if IS_ENABLED(CONFIG_IPV6) - if (rt && !(rt->dst.flags & DST_NOPEER)) { - struct inet_peer *peer; - struct net *net; - - net = dev_net(rt->dst.dev); - peer = inet_getpeer_v6(net->ipv6.peers, &rt->rt6i_dst.addr, 1); - if (peer) { - fhdr->identification = htonl(inet_getid(peer, 0)); - inet_putpeer(peer); - return; - } - } -#endif - ident = atomic_inc_return(&ipv6_fragmentation_id); - fhdr->identification = htonl(ident); -} -EXPORT_SYMBOL(ipv6_select_ident); int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) { diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 540d58921007..8d22460a811b 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -919,7 +919,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, iph->ttl = iph6->hop_limit; skb->ip_summed = CHECKSUM_NONE; - ip_select_ident(skb, skb_dst(skb), NULL); + ip_select_ident(skb, NULL); iptunnel_xmit(skb, dev); return NETDEV_TX_OK; diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index c47444e4cf8c..7f0e1cf2d7e8 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -883,7 +883,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, iph->daddr = cp->daddr.ip; iph->saddr = saddr; iph->ttl = old_iph->ttl; - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); /* Another hack: avoid icmp_send in ip_fragment */ skb->local_df = 1; From 4176df017abbe847f0c09e5b839feb40e04b25ba Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 26 Jul 2014 08:58:10 +0200 Subject: [PATCH 03/25] ip: make IP identifiers less predictable [ Upstream commit 04ca6973f7c1a0d8537f2d9906a0cf8e69886d75 ] In "Counting Packets Sent Between Arbitrary Internet Hosts", Jeffrey and Jedidiah describe ways exploiting linux IP identifier generation to infer whether two machines are exchanging packets. With commit 73f156a6e8c1 ("inetpeer: get rid of ip_id_count"), we changed IP id generation, but this does not really prevent this side-channel technique. This patch adds a random amount of perturbation so that IP identifiers for a given destination [1] are no longer monotonically increasing after an idle period. Note that prandom_u32_max(1) returns 0, so if generator is used at most once per jiffy, this patch inserts no hole in the ID suite and do not increase collision probability. This is jiffies based, so in the worst case (HZ=1000), the id can rollover after ~65 seconds of idle time, which should be fine. We also change the hash used in __ip_select_ident() to not only hash on daddr, but also saddr and protocol, so that ICMP probes can not be used to infer information for other protocols. For IPv6, adds saddr into the hash as well, but not nexthdr. If I ping the patched target, we can see ID are now hard to predict. 21:57:11.008086 IP (...) A > target: ICMP echo request, seq 1, length 64 21:57:11.010752 IP (... id 2081 ...) target > A: ICMP echo reply, seq 1, length 64 21:57:12.013133 IP (...) A > target: ICMP echo request, seq 2, length 64 21:57:12.015737 IP (... id 3039 ...) target > A: ICMP echo reply, seq 2, length 64 21:57:13.016580 IP (...) A > target: ICMP echo request, seq 3, length 64 21:57:13.019251 IP (... id 3437 ...) target > A: ICMP echo reply, seq 3, length 64 [1] TCP sessions uses a per flow ID generator not changed by this patch. Signed-off-by: Eric Dumazet Reported-by: Jeffrey Knockel Reported-by: Jedidiah R. Crandall Cc: Willy Tarreau Cc: Hannes Frederic Sowa Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- include/net/ip.h | 11 +---------- net/ipv4/route.c | 36 +++++++++++++++++++++++++++++++++--- net/ipv6/ip6_output.c | 2 ++ 3 files changed, 36 insertions(+), 13 deletions(-) diff --git a/include/net/ip.h b/include/net/ip.h index dd72c8f93797..8695359982d1 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -252,16 +252,7 @@ int ip_dont_fragment(struct sock *sk, struct dst_entry *dst) !(dst_metric_locked(dst, RTAX_MTU))); } -#define IP_IDENTS_SZ 2048u -extern atomic_t *ip_idents; - -static inline u32 ip_idents_reserve(u32 hash, int segs) -{ - atomic_t *id_ptr = ip_idents + hash % IP_IDENTS_SZ; - - return atomic_add_return(segs, id_ptr) - segs; -} - +u32 ip_idents_reserve(u32 hash, int segs); void __ip_select_ident(struct iphdr *iph, int segs); static inline void ip_select_ident_segs(struct sk_buff *skb, struct sock *sk, int segs) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index da2d4efc0bf7..d4d162eac4df 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -465,8 +465,35 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst, return neigh_create(&arp_tbl, pkey, dev); } -atomic_t *ip_idents __read_mostly; -EXPORT_SYMBOL(ip_idents); +#define IP_IDENTS_SZ 2048u +struct ip_ident_bucket { + atomic_t id; + u32 stamp32; +}; + +static struct ip_ident_bucket *ip_idents __read_mostly; + +/* In order to protect privacy, we add a perturbation to identifiers + * if one generator is seldom used. This makes hard for an attacker + * to infer how many packets were sent between two points in time. + */ +u32 ip_idents_reserve(u32 hash, int segs) +{ + struct ip_ident_bucket *bucket = ip_idents + hash % IP_IDENTS_SZ; + u32 old = ACCESS_ONCE(bucket->stamp32); + u32 now = (u32)jiffies; + u32 delta = 0; + + if (old != now && cmpxchg(&bucket->stamp32, old, now) == old) { + u64 x = prandom_u32(); + + x *= (now - old); + delta = (u32)(x >> 32); + } + + return atomic_add_return(segs + delta, &bucket->id) - segs; +} +EXPORT_SYMBOL(ip_idents_reserve); void __ip_select_ident(struct iphdr *iph, int segs) { @@ -479,7 +506,10 @@ void __ip_select_ident(struct iphdr *iph, int segs) get_random_bytes(&ip_idents_hashrnd, sizeof(ip_idents_hashrnd)); } - hash = jhash_1word((__force u32)iph->daddr, ip_idents_hashrnd); + hash = jhash_3words((__force u32)iph->daddr, + (__force u32)iph->saddr, + iph->protocol, + ip_idents_hashrnd); id = ip_idents_reserve(hash, segs); iph->id = htons(id); } diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a425052ac274..071edcba4158 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -551,6 +551,8 @@ static void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt) get_random_bytes(&ip6_idents_hashrnd, sizeof(ip6_idents_hashrnd)); } hash = __ipv6_addr_jhash(&rt->rt6i_dst.addr, ip6_idents_hashrnd); + hash = __ipv6_addr_jhash(&rt->rt6i_src.addr, hash); + id = ip_idents_reserve(hash, 1); fhdr->identification = htonl(id); } From 15229fa9d4588b2d0e91ee81954c3a4f3c30dcb8 Mon Sep 17 00:00:00 2001 From: Andrey Ryabinin Date: Sat, 26 Jul 2014 21:26:58 +0400 Subject: [PATCH 04/25] net: sendmsg: fix NULL pointer dereference [ Upstream commit 40eea803c6b2cfaab092f053248cbeab3f368412 ] Sasha's report: > While fuzzing with trinity inside a KVM tools guest running the latest -next > kernel with the KASAN patchset, I've stumbled on the following spew: > > [ 4448.949424] ================================================================== > [ 4448.951737] AddressSanitizer: user-memory-access on address 0 > [ 4448.952988] Read of size 2 by thread T19638: > [ 4448.954510] CPU: 28 PID: 19638 Comm: trinity-c76 Not tainted 3.16.0-rc4-next-20140711-sasha-00046-g07d3099-dirty #813 > [ 4448.956823] ffff88046d86ca40 0000000000000000 ffff880082f37e78 ffff880082f37a40 > [ 4448.958233] ffffffffb6e47068 ffff880082f37a68 ffff880082f37a58 ffffffffb242708d > [ 4448.959552] 0000000000000000 ffff880082f37a88 ffffffffb24255b1 0000000000000000 > [ 4448.961266] Call Trace: > [ 4448.963158] dump_stack (lib/dump_stack.c:52) > [ 4448.964244] kasan_report_user_access (mm/kasan/report.c:184) > [ 4448.965507] __asan_load2 (mm/kasan/kasan.c:352) > [ 4448.966482] ? netlink_sendmsg (net/netlink/af_netlink.c:2339) > [ 4448.967541] netlink_sendmsg (net/netlink/af_netlink.c:2339) > [ 4448.968537] ? get_parent_ip (kernel/sched/core.c:2555) > [ 4448.970103] sock_sendmsg (net/socket.c:654) > [ 4448.971584] ? might_fault (mm/memory.c:3741) > [ 4448.972526] ? might_fault (./arch/x86/include/asm/current.h:14 mm/memory.c:3740) > [ 4448.973596] ? verify_iovec (net/core/iovec.c:64) > [ 4448.974522] ___sys_sendmsg (net/socket.c:2096) > [ 4448.975797] ? put_lock_stats.isra.13 (./arch/x86/include/asm/preempt.h:98 kernel/locking/lockdep.c:254) > [ 4448.977030] ? lock_release_holdtime (kernel/locking/lockdep.c:273) > [ 4448.978197] ? lock_release_non_nested (kernel/locking/lockdep.c:3434 (discriminator 1)) > [ 4448.979346] ? check_chain_key (kernel/locking/lockdep.c:2188) > [ 4448.980535] __sys_sendmmsg (net/socket.c:2181) > [ 4448.981592] ? trace_hardirqs_on_caller (kernel/locking/lockdep.c:2600) > [ 4448.982773] ? trace_hardirqs_on (kernel/locking/lockdep.c:2607) > [ 4448.984458] ? syscall_trace_enter (arch/x86/kernel/ptrace.c:1500 (discriminator 2)) > [ 4448.985621] ? trace_hardirqs_on_caller (kernel/locking/lockdep.c:2600) > [ 4448.986754] SyS_sendmmsg (net/socket.c:2201) > [ 4448.987708] tracesys (arch/x86/kernel/entry_64.S:542) > [ 4448.988929] ================================================================== This reports means that we've come to netlink_sendmsg() with msg->msg_name == NULL and msg->msg_namelen > 0. After this report there was no usual "Unable to handle kernel NULL pointer dereference" and this gave me a clue that address 0 is mapped and contains valid socket address structure in it. This bug was introduced in f3d3342602f8bcbf37d7c46641cb9bca7618eb1c (net: rework recvmsg handler msg_name and msg_namelen logic). Commit message states that: "Set msg->msg_name = NULL if user specified a NULL in msg_name but had a non-null msg_namelen in verify_iovec/verify_compat_iovec. This doesn't affect sendto as it would bail out earlier while trying to copy-in the address." But in fact this affects sendto when address 0 is mapped and contains socket address structure in it. In such case copy-in address will succeed, verify_iovec() function will successfully exit with msg->msg_namelen > 0 and msg->msg_name == NULL. This patch fixes it by setting msg_namelen to 0 if msg_name == NULL. Cc: Hannes Frederic Sowa Cc: Eric Dumazet Cc: Reported-by: Sasha Levin Signed-off-by: Andrey Ryabinin Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/compat.c | 9 +++++---- net/core/iovec.c | 6 +++--- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/net/compat.c b/net/compat.c index f50161fb812e..cbc1a2a26587 100644 --- a/net/compat.c +++ b/net/compat.c @@ -85,7 +85,7 @@ int verify_compat_iovec(struct msghdr *kern_msg, struct iovec *kern_iov, { int tot_len; - if (kern_msg->msg_namelen) { + if (kern_msg->msg_name && kern_msg->msg_namelen) { if (mode == VERIFY_READ) { int err = move_addr_to_kernel(kern_msg->msg_name, kern_msg->msg_namelen, @@ -93,10 +93,11 @@ int verify_compat_iovec(struct msghdr *kern_msg, struct iovec *kern_iov, if (err < 0) return err; } - if (kern_msg->msg_name) - kern_msg->msg_name = kern_address; - } else + kern_msg->msg_name = kern_address; + } else { kern_msg->msg_name = NULL; + kern_msg->msg_namelen = 0; + } tot_len = iov_from_user_compat_to_kern(kern_iov, (struct compat_iovec __user *)kern_msg->msg_iov, diff --git a/net/core/iovec.c b/net/core/iovec.c index 9a31515fb8e3..2145b7150beb 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -39,7 +39,7 @@ int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr_storage *a { int size, ct, err; - if (m->msg_namelen) { + if (m->msg_name && m->msg_namelen) { if (mode == VERIFY_READ) { void __user *namep; namep = (void __user __force *) m->msg_name; @@ -48,10 +48,10 @@ int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr_storage *a if (err < 0) return err; } - if (m->msg_name) - m->msg_name = address; + m->msg_name = address; } else { m->msg_name = NULL; + m->msg_namelen = 0; } size = m->msg_iovlen * sizeof(struct iovec); From 822fc8dd9dec94a3505998c5febdcb41729f7d57 Mon Sep 17 00:00:00 2001 From: Christoph Paasch Date: Tue, 29 Jul 2014 12:07:27 +0200 Subject: [PATCH 05/25] tcp: Fix integer-overflows in TCP veno [ Upstream commit 45a07695bc64b3ab5d6d2215f9677e5b8c05a7d0 ] In veno we do a multiplication of the cwnd and the rtt. This may overflow and thus their result is stored in a u64. However, we first need to cast the cwnd so that actually 64-bit arithmetic is done. A first attempt at fixing 76f1017757aa0 ([TCP]: TCP Veno congestion control) was made by 159131149c2 (tcp: Overflow bug in Vegas), but it failed to add the required cast in tcp_veno_cong_avoid(). Fixes: 76f1017757aa0 ([TCP]: TCP Veno congestion control) Signed-off-by: Christoph Paasch Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/ipv4/tcp_veno.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp_veno.c index ac43cd747bce..b4d1858be550 100644 --- a/net/ipv4/tcp_veno.c +++ b/net/ipv4/tcp_veno.c @@ -144,7 +144,7 @@ static void tcp_veno_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) rtt = veno->minrtt; - target_cwnd = (tp->snd_cwnd * veno->basertt); + target_cwnd = (u64)tp->snd_cwnd * veno->basertt; target_cwnd <<= V_PARAM_SHIFT; do_div(target_cwnd, rtt); From 5191776cd6a25948c6a6622d40e6fe4c2442794a Mon Sep 17 00:00:00 2001 From: Christoph Paasch Date: Tue, 29 Jul 2014 13:40:57 +0200 Subject: [PATCH 06/25] tcp: Fix integer-overflow in TCP vegas [ Upstream commit 1f74e613ded11517db90b2bd57e9464d9e0fb161 ] In vegas we do a multiplication of the cwnd and the rtt. This may overflow and thus their result is stored in a u64. However, we first need to cast the cwnd so that actually 64-bit arithmetic is done. Then, we need to do do_div to allow this to be used on 32-bit arches. Cc: Stephen Hemminger Cc: Neal Cardwell Cc: Eric Dumazet Cc: David Laight Cc: Doug Leith Fixes: 8d3a564da34e (tcp: tcp_vegas cong avoid fix) Signed-off-by: Christoph Paasch Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/ipv4/tcp_vegas.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index 80fa2bfd7ede..c042e529a11e 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -218,7 +218,8 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) * This is: * (actual rate in segments) * baseRTT */ - target_cwnd = tp->snd_cwnd * vegas->baseRTT / rtt; + target_cwnd = (u64)tp->snd_cwnd * vegas->baseRTT; + do_div(target_cwnd, rtt); /* Calculate the difference between the window we had, * and the window we would like to have. This quantity From 495d049f3e499227d28c89800bca27cc726b3bb2 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 22 Jul 2014 15:22:45 +0200 Subject: [PATCH 07/25] net: sctp: inherit auth_capable on INIT collisions [ Upstream commit 1be9a950c646c9092fb3618197f7b6bfb50e82aa ] Jason reported an oops caused by SCTP on his ARM machine with SCTP authentication enabled: Internal error: Oops: 17 [#1] ARM CPU: 0 PID: 104 Comm: sctp-test Not tainted 3.13.0-68744-g3632f30c9b20-dirty #1 task: c6eefa40 ti: c6f52000 task.ti: c6f52000 PC is at sctp_auth_calculate_hmac+0xc4/0x10c LR is at sg_init_table+0x20/0x38 pc : [] lr : [] psr: 40000013 sp : c6f538e8 ip : 00000000 fp : c6f53924 r10: c6f50d80 r9 : 00000000 r8 : 00010000 r7 : 00000000 r6 : c7be4000 r5 : 00000000 r4 : c6f56254 r3 : c00c8170 r2 : 00000001 r1 : 00000008 r0 : c6f1e660 Flags: nZcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 0005397f Table: 06f28000 DAC: 00000015 Process sctp-test (pid: 104, stack limit = 0xc6f521c0) Stack: (0xc6f538e8 to 0xc6f54000) [...] Backtrace: [] (sctp_auth_calculate_hmac+0x0/0x10c) from [] (sctp_packet_transmit+0x33c/0x5c8) [] (sctp_packet_transmit+0x0/0x5c8) from [] (sctp_outq_flush+0x7fc/0x844) [] (sctp_outq_flush+0x0/0x844) from [] (sctp_outq_uncork+0x24/0x28) [] (sctp_outq_uncork+0x0/0x28) from [] (sctp_side_effects+0x1134/0x1220) [] (sctp_side_effects+0x0/0x1220) from [] (sctp_do_sm+0xac/0xd4) [] (sctp_do_sm+0x0/0xd4) from [] (sctp_assoc_bh_rcv+0x118/0x160) [] (sctp_assoc_bh_rcv+0x0/0x160) from [] (sctp_inq_push+0x6c/0x74) [] (sctp_inq_push+0x0/0x74) from [] (sctp_rcv+0x7d8/0x888) While we already had various kind of bugs in that area ec0223ec48a9 ("net: sctp: fix sctp_sf_do_5_1D_ce to verify if we/peer is AUTH capable") and b14878ccb7fa ("net: sctp: cache auth_enable per endpoint"), this one is a bit of a different kind. Giving a bit more background on why SCTP authentication is needed can be found in RFC4895: SCTP uses 32-bit verification tags to protect itself against blind attackers. These values are not changed during the lifetime of an SCTP association. Looking at new SCTP extensions, there is the need to have a method of proving that an SCTP chunk(s) was really sent by the original peer that started the association and not by a malicious attacker. To cause this bug, we're triggering an INIT collision between peers; normal SCTP handshake where both sides intent to authenticate packets contains RANDOM; CHUNKS; HMAC-ALGO parameters that are being negotiated among peers: ---------- INIT[RANDOM; CHUNKS; HMAC-ALGO] ----------> <------- INIT-ACK[RANDOM; CHUNKS; HMAC-ALGO] --------- -------------------- COOKIE-ECHO --------------------> <-------------------- COOKIE-ACK --------------------- RFC4895 says that each endpoint therefore knows its own random number and the peer's random number *after* the association has been established. The local and peer's random number along with the shared key are then part of the secret used for calculating the HMAC in the AUTH chunk. Now, in our scenario, we have 2 threads with 1 non-blocking SEQ_PACKET socket each, setting up common shared SCTP_AUTH_KEY and SCTP_AUTH_ACTIVE_KEY properly, and each of them calling sctp_bindx(3), listen(2) and connect(2) against each other, thus the handshake looks similar to this, e.g.: ---------- INIT[RANDOM; CHUNKS; HMAC-ALGO] ----------> <------- INIT-ACK[RANDOM; CHUNKS; HMAC-ALGO] --------- <--------- INIT[RANDOM; CHUNKS; HMAC-ALGO] ----------- -------- INIT-ACK[RANDOM; CHUNKS; HMAC-ALGO] --------> ... Since such collisions can also happen with verification tags, the RFC4895 for AUTH rather vaguely says under section 6.1: In case of INIT collision, the rules governing the handling of this Random Number follow the same pattern as those for the Verification Tag, as explained in Section 5.2.4 of RFC 2960 [5]. Therefore, each endpoint knows its own Random Number and the peer's Random Number after the association has been established. In RFC2960, section 5.2.4, we're eventually hitting Action B: B) In this case, both sides may be attempting to start an association at about the same time but the peer endpoint started its INIT after responding to the local endpoint's INIT. Thus it may have picked a new Verification Tag not being aware of the previous Tag it had sent this endpoint. The endpoint should stay in or enter the ESTABLISHED state but it MUST update its peer's Verification Tag from the State Cookie, stop any init or cookie timers that may running and send a COOKIE ACK. In other words, the handling of the Random parameter is the same as behavior for the Verification Tag as described in Action B of section 5.2.4. Looking at the code, we exactly hit the sctp_sf_do_dupcook_b() case which triggers an SCTP_CMD_UPDATE_ASSOC command to the side effect interpreter, and in fact it properly copies over peer_{random, hmacs, chunks} parameters from the newly created association to update the existing one. Also, the old asoc_shared_key is being released and based on the new params, sctp_auth_asoc_init_active_key() updated. However, the issue observed in this case is that the previous asoc->peer.auth_capable was 0, and has *not* been updated, so that instead of creating a new secret, we're doing an early return from the function sctp_auth_asoc_init_active_key() leaving asoc->asoc_shared_key as NULL. However, we now have to authenticate chunks from the updated chunk list (e.g. COOKIE-ACK). That in fact causes the server side when responding with ... <------------------ AUTH; COOKIE-ACK ----------------- ... to trigger a NULL pointer dereference, since in sctp_packet_transmit(), it discovers that an AUTH chunk is being queued for xmit, and thus it calls sctp_auth_calculate_hmac(). Since the asoc->active_key_id is still inherited from the endpoint, and the same as encoded into the chunk, it uses asoc->asoc_shared_key, which is still NULL, as an asoc_key and dereferences it in ... crypto_hash_setkey(desc.tfm, &asoc_key->data[0], asoc_key->len) ... causing an oops. All this happens because sctp_make_cookie_ack() called with the *new* association has the peer.auth_capable=1 and therefore marks the chunk with auth=1 after checking sctp_auth_send_cid(), but it is *actually* sent later on over the then *updated* association's transport that didn't initialize its shared key due to peer.auth_capable=0. Since control chunks in that case are not sent by the temporary association which are scheduled for deletion, they are issued for xmit via SCTP_CMD_REPLY in the interpreter with the context of the *updated* association. peer.auth_capable was 0 in the updated association (which went from COOKIE_WAIT into ESTABLISHED state), since all previous processing that performed sctp_process_init() was being done on temporary associations, that we eventually throw away each time. The correct fix is to update to the new peer.auth_capable value as well in the collision case via sctp_assoc_update(), so that in case the collision migrated from 0 -> 1, sctp_auth_asoc_init_active_key() can properly recalculate the secret. This therefore fixes the observed server panic. Fixes: 730fc3d05cd4 ("[SCTP]: Implete SCTP-AUTH parameter processing") Reported-by: Jason Gunthorpe Signed-off-by: Daniel Borkmann Tested-by: Jason Gunthorpe Cc: Vlad Yasevich Acked-by: Vlad Yasevich Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/sctp/associola.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 229b3c3fb6c9..62e86d98bc36 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -1213,6 +1213,7 @@ void sctp_assoc_update(struct sctp_association *asoc, asoc->c = new->c; asoc->peer.rwnd = new->peer.rwnd; asoc->peer.sack_needed = new->peer.sack_needed; + asoc->peer.auth_capable = new->peer.auth_capable; asoc->peer.i = new->peer.i; sctp_tsnmap_init(&asoc->peer.tsn_map, SCTP_TSN_MAP_INITIAL, asoc->peer.i.initial_tsn, GFP_ATOMIC); From d09fdc66ecb543136f9e3304c1b6071d3dae9792 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Thu, 31 Jul 2014 10:30:25 -0400 Subject: [PATCH 08/25] macvlan: Initialize vlan_features to turn on offload support. [ Upstream commit 081e83a78db9b0ae1f5eabc2dedecc865f509b98 ] Macvlan devices do not initialize vlan_features. As a result, any vlan devices configured on top of macvlans perform very poorly. Initialize vlan_features based on the vlan features of the lower-level device. Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/macvlan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 155ef4bbde91..9be91cb4f4a3 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -500,6 +500,7 @@ static int macvlan_init(struct net_device *dev) (lowerdev->state & MACVLAN_STATE_MASK); dev->features = lowerdev->features & MACVLAN_FEATURES; dev->features |= NETIF_F_LLTX; + dev->vlan_features = lowerdev->vlan_features & MACVLAN_FEATURES; dev->gso_max_size = lowerdev->gso_max_size; dev->iflink = lowerdev->ifindex; dev->hard_header_len = lowerdev->hard_header_len; From c290a4ef1fe38bd8d29573468a189e49dbaa0681 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Thu, 31 Jul 2014 10:33:06 -0400 Subject: [PATCH 09/25] net: Correctly set segment mac_len in skb_segment(). [ Upstream commit fcdfe3a7fa4cb74391d42b6a26dc07c20dab1d82 ] When performing segmentation, the mac_len value is copied right out of the original skb. However, this value is not always set correctly (like when the packet is VLAN-tagged) and we'll end up copying a bad value. One way to demonstrate this is to configure a VM which tags packets internally and turn off VLAN acceleration on the forwarding bridge port. The packets show up corrupt like this: 16:18:24.985548 52:54:00:ab:be:25 > 52:54:00:26:ce:a3, ethertype 802.1Q (0x8100), length 1518: vlan 100, p 0, ethertype 0x05e0, 0x0000: 8cdb 1c7c 8cdb 0064 4006 b59d 0a00 6402 ...|...d@.....d. 0x0010: 0a00 6401 9e0d b441 0a5e 64ec 0330 14fa ..d....A.^d..0.. 0x0020: 29e3 01c9 f871 0000 0101 080a 000a e833)....q.........3 0x0030: 000f 8c75 6e65 7470 6572 6600 6e65 7470 ...unetperf.netp 0x0040: 6572 6600 6e65 7470 6572 6600 6e65 7470 erf.netperf.netp 0x0050: 6572 6600 6e65 7470 6572 6600 6e65 7470 erf.netperf.netp 0x0060: 6572 6600 6e65 7470 6572 6600 6e65 7470 erf.netperf.netp ... This also leads to awful throughput as GSO packets are dropped and cause retransmissions. The solution is to set the mac_len using the values already available in then new skb. We've already adjusted all of the header offset, so we might as well correctly figure out the mac_len using skb_reset_mac_len(). After this change, packets are segmented correctly and performance is restored. CC: Eric Dumazet Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 9f84a5f7404d..6148716884ae 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2810,7 +2810,6 @@ struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features) tail = nskb; __copy_skb_header(nskb, skb); - nskb->mac_len = skb->mac_len; /* nskb and skb might have different headroom */ if (nskb->ip_summed == CHECKSUM_PARTIAL) @@ -2820,6 +2819,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features) skb_set_network_header(nskb, skb->mac_len); nskb->transport_header = (nskb->network_header + skb_network_header_len(skb)); + skb_reset_mac_len(nskb); skb_copy_from_linear_data_offset(skb, -tnl_hlen, nskb->data - tnl_hlen, From 9d868b94a6f52a17a2a94b34e544800d741b7852 Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Thu, 31 Jul 2014 23:00:35 -0400 Subject: [PATCH 10/25] iovec: make sure the caller actually wants anything in memcpy_fromiovecend [ Upstream commit 06ebb06d49486676272a3c030bfeef4bd969a8e6 ] Check for cases when the caller requests 0 bytes instead of running off and dereferencing potentially invalid iovecs. Signed-off-by: Sasha Levin Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/core/iovec.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/core/iovec.c b/net/core/iovec.c index 2145b7150beb..1117a26a8548 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -107,6 +107,10 @@ EXPORT_SYMBOL(memcpy_toiovecend); int memcpy_fromiovecend(unsigned char *kdata, const struct iovec *iov, int offset, int len) { + /* No data? Done! */ + if (len == 0) + return 0; + /* Skip over the finished iovecs */ while (offset >= iov->iov_len) { offset -= iov->iov_len; From 6e5f6266d635809c560f7fb48a710701a1d54139 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 5 Aug 2014 16:49:52 +0200 Subject: [PATCH 11/25] sctp: fix possible seqlock seadlock in sctp_packet_transmit() [ Upstream commit 757efd32d5ce31f67193cc0e6a56e4dffcc42fb1 ] Dave reported following splat, caused by improper use of IP_INC_STATS_BH() in process context. BUG: using __this_cpu_add() in preemptible [00000000] code: trinity-c117/14551 caller is __this_cpu_preempt_check+0x13/0x20 CPU: 3 PID: 14551 Comm: trinity-c117 Not tainted 3.16.0+ #33 ffffffff9ec898f0 0000000047ea7e23 ffff88022d32f7f0 ffffffff9e7ee207 0000000000000003 ffff88022d32f818 ffffffff9e397eaa ffff88023ee70b40 ffff88022d32f970 ffff8801c026d580 ffff88022d32f828 ffffffff9e397ee3 Call Trace: [] dump_stack+0x4e/0x7a [] check_preemption_disabled+0xfa/0x100 [] __this_cpu_preempt_check+0x13/0x20 [] sctp_packet_transmit+0x692/0x710 [sctp] [] sctp_outq_flush+0x2a2/0xc30 [sctp] [] ? mark_held_locks+0x7c/0xb0 [] ? _raw_spin_unlock_irqrestore+0x5d/0x80 [] sctp_outq_uncork+0x1a/0x20 [sctp] [] sctp_cmd_interpreter.isra.23+0x1142/0x13f0 [sctp] [] sctp_do_sm+0xdb/0x330 [sctp] [] ? preempt_count_sub+0xab/0x100 [] ? sctp_cname+0x70/0x70 [sctp] [] sctp_primitive_ASSOCIATE+0x3a/0x50 [sctp] [] sctp_sendmsg+0x88f/0xe30 [sctp] [] ? lock_release_holdtime.part.28+0x9a/0x160 [] ? put_lock_stats.isra.27+0xe/0x30 [] inet_sendmsg+0x104/0x220 [] ? inet_sendmsg+0x5/0x220 [] sock_sendmsg+0x9e/0xe0 [] ? might_fault+0xb9/0xc0 [] ? might_fault+0x5e/0xc0 [] SYSC_sendto+0x124/0x1c0 [] ? syscall_trace_enter+0x250/0x330 [] SyS_sendto+0xe/0x10 [] tracesys+0xdd/0xe2 This is a followup of commits f1d8cba61c3c4b ("inet: fix possible seqlock deadlocks") and 7f88c6b23afbd315 ("ipv6: fix possible seqlock deadlock in ip6_finish_output2") Signed-off-by: Eric Dumazet Cc: Hannes Frederic Sowa Reported-by: Dave Jones Acked-by: Neil Horman Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/sctp/output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/output.c b/net/sctp/output.c index 0beb2f9c8a7c..b6f5fc3127b9 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -618,7 +618,7 @@ out: return err; no_route: kfree_skb(nskb); - IP_INC_STATS_BH(sock_net(asoc->base.sk), IPSTATS_MIB_OUTNOROUTES); + IP_INC_STATS(sock_net(asoc->base.sk), IPSTATS_MIB_OUTNOROUTES); /* FIXME: Returning the 'err' will effect all the associations * associated with a socket, although only one of the paths of the From af25a08645b730f1d327110235d480266fa18acb Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 30 Apr 2014 19:37:48 -0700 Subject: [PATCH 12/25] sparc64: Fix argument sign extension for compat_sys_futex(). [ Upstream commit aa3449ee9c87d9b7660dd1493248abcc57769e31 ] Only the second argument, 'op', is signed. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/kernel/sys32.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index f7c72b6efc27..d066eb18650c 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -44,7 +44,7 @@ SIGN1(sys32_timer_settime, compat_sys_timer_settime, %o1) SIGN1(sys32_io_submit, compat_sys_io_submit, %o1) SIGN1(sys32_mq_open, compat_sys_mq_open, %o1) SIGN1(sys32_select, compat_sys_select, %o0) -SIGN3(sys32_futex, compat_sys_futex, %o1, %o2, %o5) +SIGN1(sys32_futex, compat_sys_futex, %o1) SIGN1(sys32_recvfrom, compat_sys_recvfrom, %o0) SIGN1(sys32_recvmsg, compat_sys_recvmsg, %o0) SIGN1(sys32_sendmsg, compat_sys_sendmsg, %o0) From f07030c1e2eadec2ee9863e1b7e06036a2166886 Mon Sep 17 00:00:00 2001 From: Kirill Tkhai Date: Thu, 17 Apr 2014 00:45:24 +0400 Subject: [PATCH 13/25] sparc64: Make itc_sync_lock raw [ Upstream commit 49b6c01f4c1de3b5e5427ac5aba80f9f6d27837a ] One more place where we must not be able to be preempted or to be interrupted in RT. Always actually disable interrupts during synchronization cycle. Signed-off-by: Kirill Tkhai Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/kernel/smp_64.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index 77539eda928c..8565ecd7d48a 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -150,7 +150,7 @@ void cpu_panic(void) #define NUM_ROUNDS 64 /* magic value */ #define NUM_ITERS 5 /* likewise */ -static DEFINE_SPINLOCK(itc_sync_lock); +static DEFINE_RAW_SPINLOCK(itc_sync_lock); static unsigned long go[SLAVE + 1]; #define DEBUG_TICK_SYNC 0 @@ -258,7 +258,7 @@ static void smp_synchronize_one_tick(int cpu) go[MASTER] = 0; membar_safe("#StoreLoad"); - spin_lock_irqsave(&itc_sync_lock, flags); + raw_spin_lock_irqsave(&itc_sync_lock, flags); { for (i = 0; i < NUM_ROUNDS*NUM_ITERS; i++) { while (!go[MASTER]) @@ -269,7 +269,7 @@ static void smp_synchronize_one_tick(int cpu) membar_safe("#StoreLoad"); } } - spin_unlock_irqrestore(&itc_sync_lock, flags); + raw_spin_unlock_irqrestore(&itc_sync_lock, flags); } #if defined(CONFIG_SUN_LDOMS) && defined(CONFIG_HOTPLUG_CPU) From ff156a9ba7c160c8fc1c8d4bb0fb935771e105dc Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Apr 2014 23:50:08 -0700 Subject: [PATCH 14/25] sparc64: Handle 32-bit tasks properly in compute_effective_address(). [ Upstream commit d037d16372bbe4d580342bebbb8826821ad9edf0 ] If we have a 32-bit task we must chop off the top 32-bits of the 64-bit value just as the cpu would. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/kernel/unaligned_64.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c index 8201c25e7669..4db8898199f7 100644 --- a/arch/sparc/kernel/unaligned_64.c +++ b/arch/sparc/kernel/unaligned_64.c @@ -163,17 +163,23 @@ static unsigned long *fetch_reg_addr(unsigned int reg, struct pt_regs *regs) unsigned long compute_effective_address(struct pt_regs *regs, unsigned int insn, unsigned int rd) { + int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; unsigned int rs1 = (insn >> 14) & 0x1f; unsigned int rs2 = insn & 0x1f; - int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; + unsigned long addr; if (insn & 0x2000) { maybe_flush_windows(rs1, 0, rd, from_kernel); - return (fetch_reg(rs1, regs) + sign_extend_imm13(insn)); + addr = (fetch_reg(rs1, regs) + sign_extend_imm13(insn)); } else { maybe_flush_windows(rs1, rs2, rd, from_kernel); - return (fetch_reg(rs1, regs) + fetch_reg(rs2, regs)); + addr = (fetch_reg(rs1, regs) + fetch_reg(rs2, regs)); } + + if (!from_kernel && test_thread_flag(TIF_32BIT)) + addr &= 0xffffffff; + + return addr; } /* This is just to make gcc think die_if_kernel does return... */ From f2701cd605156eee1cd5edbc709f989a8bbe218a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Apr 2014 23:52:11 -0700 Subject: [PATCH 15/25] sparc64: Fix top-level fault handling bugs. [ Upstream commit 70ffc6ebaead783ac8dafb1e87df0039bb043596 ] Make get_user_insn() able to cope with huge PMDs. Next, make do_fault_siginfo() more robust when get_user_insn() can't actually fetch the instruction. In particular, use the MMU announced fault address when that happens, instead of calling compute_effective_address() and computing garbage. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/mm/fault_64.c | 84 +++++++++++++++++++++++++--------------- 1 file changed, 53 insertions(+), 31 deletions(-) diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c index 5062ff389e83..1992fa04095f 100644 --- a/arch/sparc/mm/fault_64.c +++ b/arch/sparc/mm/fault_64.c @@ -95,38 +95,51 @@ static unsigned int get_user_insn(unsigned long tpc) pte_t *ptep, pte; unsigned long pa; u32 insn = 0; - unsigned long pstate; - if (pgd_none(*pgdp)) - goto outret; + if (pgd_none(*pgdp) || unlikely(pgd_bad(*pgdp))) + goto out; pudp = pud_offset(pgdp, tpc); - if (pud_none(*pudp)) - goto outret; - pmdp = pmd_offset(pudp, tpc); - if (pmd_none(*pmdp)) - goto outret; - - /* This disables preemption for us as well. */ - __asm__ __volatile__("rdpr %%pstate, %0" : "=r" (pstate)); - __asm__ __volatile__("wrpr %0, %1, %%pstate" - : : "r" (pstate), "i" (PSTATE_IE)); - ptep = pte_offset_map(pmdp, tpc); - pte = *ptep; - if (!pte_present(pte)) + if (pud_none(*pudp) || unlikely(pud_bad(*pudp))) goto out; - pa = (pte_pfn(pte) << PAGE_SHIFT); - pa += (tpc & ~PAGE_MASK); + /* This disables preemption for us as well. */ + local_irq_disable(); - /* Use phys bypass so we don't pollute dtlb/dcache. */ - __asm__ __volatile__("lduwa [%1] %2, %0" - : "=r" (insn) - : "r" (pa), "i" (ASI_PHYS_USE_EC)); + pmdp = pmd_offset(pudp, tpc); + if (pmd_none(*pmdp) || unlikely(pmd_bad(*pmdp))) + goto out_irq_enable; +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + if (pmd_trans_huge(*pmdp)) { + if (pmd_trans_splitting(*pmdp)) + goto out_irq_enable; + + pa = pmd_pfn(*pmdp) << PAGE_SHIFT; + pa += tpc & ~HPAGE_MASK; + + /* Use phys bypass so we don't pollute dtlb/dcache. */ + __asm__ __volatile__("lduwa [%1] %2, %0" + : "=r" (insn) + : "r" (pa), "i" (ASI_PHYS_USE_EC)); + } else +#endif + { + ptep = pte_offset_map(pmdp, tpc); + pte = *ptep; + if (pte_present(pte)) { + pa = (pte_pfn(pte) << PAGE_SHIFT); + pa += (tpc & ~PAGE_MASK); + + /* Use phys bypass so we don't pollute dtlb/dcache. */ + __asm__ __volatile__("lduwa [%1] %2, %0" + : "=r" (insn) + : "r" (pa), "i" (ASI_PHYS_USE_EC)); + } + pte_unmap(ptep); + } +out_irq_enable: + local_irq_enable(); out: - pte_unmap(ptep); - __asm__ __volatile__("wrpr %0, 0x0, %%pstate" : : "r" (pstate)); -outret: return insn; } @@ -152,7 +165,8 @@ show_signal_msg(struct pt_regs *regs, int sig, int code, } static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, - unsigned int insn, int fault_code) + unsigned long fault_addr, unsigned int insn, + int fault_code) { unsigned long addr; siginfo_t info; @@ -160,10 +174,18 @@ static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, info.si_code = code; info.si_signo = sig; info.si_errno = 0; - if (fault_code & FAULT_CODE_ITLB) + if (fault_code & FAULT_CODE_ITLB) { addr = regs->tpc; - else - addr = compute_effective_address(regs, insn, 0); + } else { + /* If we were able to probe the faulting instruction, use it + * to compute a precise fault address. Otherwise use the fault + * time provided address which may only have page granularity. + */ + if (insn) + addr = compute_effective_address(regs, insn, 0); + else + addr = fault_addr; + } info.si_addr = (void __user *) addr; info.si_trapno = 0; @@ -238,7 +260,7 @@ static void __kprobes do_kernel_fault(struct pt_regs *regs, int si_code, /* The si_code was set to make clear whether * this was a SEGV_MAPERR or SEGV_ACCERR fault. */ - do_fault_siginfo(si_code, SIGSEGV, regs, insn, fault_code); + do_fault_siginfo(si_code, SIGSEGV, regs, address, insn, fault_code); return; } @@ -519,7 +541,7 @@ do_sigbus: * Send a sigbus, regardless of whether we were in kernel * or user mode. */ - do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, insn, fault_code); + do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, address, insn, fault_code); /* Kernel mode? Handle exceptions or die */ if (regs->tstate & TSTATE_PRIV) From e5bd62b99772098f5123bff888247dbc8e68f602 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 6 May 2014 21:27:37 -0700 Subject: [PATCH 16/25] sparc64: Don't bark so loudly about 32-bit tasks generating 64-bit fault addresses. [ Upstream commit e5c460f46ae7ee94831cb55cb980f942aa9e5a85 ] This was found using Dave Jone's trinity tool. When a user process which is 32-bit performs a load or a store, the cpu chops off the top 32-bits of the effective address before translating it. This is because we run 32-bit tasks with the PSTATE_AM (address masking) bit set. We can't run the kernel with that bit set, so when the kernel accesses userspace no address masking occurs. Since a 32-bit process will have no mappings in that region we will properly fault, so we don't try to handle this using access_ok(), which can safely just be a NOP on sparc64. Real faults from 32-bit processes should never generate such addresses so a bug check was added long ago, and it barks in the logs if this happens. But it also barks when a kernel user access causes this condition, and that _can_ happen. For example, if a pointer passed into a system call is "0xfffffffc" and the kernel access 4 bytes offset from that pointer. Just handle such faults normally via the exception entries. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/mm/fault_64.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c index 1992fa04095f..ea83f82464da 100644 --- a/arch/sparc/mm/fault_64.c +++ b/arch/sparc/mm/fault_64.c @@ -280,18 +280,6 @@ static void noinline __kprobes bogus_32bit_fault_tpc(struct pt_regs *regs) show_regs(regs); } -static void noinline __kprobes bogus_32bit_fault_address(struct pt_regs *regs, - unsigned long addr) -{ - static int times; - - if (times++ < 10) - printk(KERN_ERR "FAULT[%s:%d]: 32-bit process " - "reports 64-bit fault address [%lx]\n", - current->comm, current->pid, addr); - show_regs(regs); -} - asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) { struct mm_struct *mm = current->mm; @@ -320,10 +308,8 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) goto intr_or_no_mm; } } - if (unlikely((address >> 32) != 0)) { - bogus_32bit_fault_address(regs, address); + if (unlikely((address >> 32) != 0)) goto intr_or_no_mm; - } } if (regs->tstate & TSTATE_PRIV) { From 720910a60cc2509160ec043459081f274c12ccce Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 7 May 2014 14:07:32 -0700 Subject: [PATCH 17/25] sparc64: Fix huge TSB mapping on pre-UltraSPARC-III cpus. [ Upstream commit b18eb2d779240631a098626cb6841ee2dd34fda0 ] Access to the TSB hash tables during TLB misses requires that there be an atomic 128-bit quad load available so that we fetch a matching TAG and DATA field at the same time. On cpus prior to UltraSPARC-III only virtual address based quad loads are available. UltraSPARC-III and later provide physical address based variants which are easier to use. When we only have virtual address based quad loads available this means that we have to lock the TSB into the TLB at a fixed virtual address on each cpu when it runs that process. We can't just access the PAGE_OFFSET based aliased mapping of these TSBs because we cannot take a recursive TLB miss inside of the TLB miss handler without risking running out of hardware trap levels (some trap combinations can be deep, such as those generated by register window spill and fill traps). Without huge pages it's working perfectly fine, but when the huge TSB got added another chunk of fixed virtual address space was not allocated for this second TSB mapping. So we were mapping both the 8K and 4MB TSBs to the same exact virtual address, causing multiple TLB matches which gives undefined behavior. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/include/asm/pgtable_64.h | 6 ++++-- arch/sparc/mm/tsb.c | 14 +++++++++++++- 2 files changed, 17 insertions(+), 3 deletions(-) diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index dfb0019bf05b..6663604a902a 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -24,7 +24,8 @@ /* The kernel image occupies 0x4000000 to 0x6000000 (4MB --> 96MB). * The page copy blockops can use 0x6000000 to 0x8000000. - * The TSB is mapped in the 0x8000000 to 0xa000000 range. + * The 8K TSB is mapped in the 0x8000000 to 0x8400000 range. + * The 4M TSB is mapped in the 0x8400000 to 0x8800000 range. * The PROM resides in an area spanning 0xf0000000 to 0x100000000. * The vmalloc area spans 0x100000000 to 0x200000000. * Since modules need to be in the lowest 32-bits of the address space, @@ -33,7 +34,8 @@ * 0x400000000. */ #define TLBTEMP_BASE _AC(0x0000000006000000,UL) -#define TSBMAP_BASE _AC(0x0000000008000000,UL) +#define TSBMAP_8K_BASE _AC(0x0000000008000000,UL) +#define TSBMAP_4M_BASE _AC(0x0000000008400000,UL) #define MODULES_VADDR _AC(0x0000000010000000,UL) #define MODULES_LEN _AC(0x00000000e0000000,UL) #define MODULES_END _AC(0x00000000f0000000,UL) diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c index 2cc3bce5ee91..71d99a6c75a7 100644 --- a/arch/sparc/mm/tsb.c +++ b/arch/sparc/mm/tsb.c @@ -133,7 +133,19 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsign mm->context.tsb_block[tsb_idx].tsb_nentries = tsb_bytes / sizeof(struct tsb); - base = TSBMAP_BASE; + switch (tsb_idx) { + case MM_TSB_BASE: + base = TSBMAP_8K_BASE; + break; +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) + case MM_TSB_HUGE: + base = TSBMAP_4M_BASE; + break; +#endif + default: + BUG(); + } + tte = pgprot_val(PAGE_KERNEL_LOCKED); tsb_paddr = __pa(mm->context.tsb_block[tsb_idx].tsb); BUG_ON(tsb_paddr & (tsb_bytes - 1UL)); From 921df8ed4c86310d215b62579703cc89741d6361 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 17 May 2014 11:28:05 -0700 Subject: [PATCH 18/25] sparc64: Add membar to Niagara2 memcpy code. [ Upstream commit 5aa4ecfd0ddb1e6dcd1c886e6c49677550f581aa ] This is the prevent previous stores from overlapping the block stores done by the memcpy loop. Based upon a glibc patch by Jose E. Marchesi Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/lib/NG2memcpy.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/sparc/lib/NG2memcpy.S b/arch/sparc/lib/NG2memcpy.S index 2c20ad63ddbf..30eee6e8a81b 100644 --- a/arch/sparc/lib/NG2memcpy.S +++ b/arch/sparc/lib/NG2memcpy.S @@ -236,6 +236,7 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ */ VISEntryHalf + membar #Sync alignaddr %o1, %g0, %g0 add %o1, (64 - 1), %o4 From 5b36a75bc06a00ede43bf8d9e921052357384072 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 4 Aug 2014 16:34:01 -0700 Subject: [PATCH 19/25] sparc64: Do not insert non-valid PTEs into the TSB hash table. [ Upstream commit 18f38132528c3e603c66ea464727b29e9bbcb91b ] The assumption was that update_mmu_cache() (and the equivalent for PMDs) would only be called when the PTE being installed will be accessible by the user. This is not true for code paths originating from remove_migration_pte(). There are dire consequences for placing a non-valid PTE into the TSB. The TLB miss frramework assumes thatwhen a TSB entry matches we can just load it into the TLB and return from the TLB miss trap. So if a non-valid PTE is in there, we will deadlock taking the TLB miss over and over, never satisfying the miss. Just exit early from update_mmu_cache() and friends in this situation. Based upon a report and patch from Christopher Alexander Tobias Schulze. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/mm/init_64.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index 04fd55a6e461..d6c1c85ff1c6 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -350,6 +350,10 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t * mm = vma->vm_mm; + /* Don't insert a non-valid PTE into the TSB, we'll deadlock. */ + if (!pte_accessible(mm, pte)) + return; + spin_lock_irqsave(&mm->context.lock, flags); #if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) From 38f4577f05042a9151ae920d0107404fdb21e59a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 4 Aug 2014 20:07:37 -0700 Subject: [PATCH 20/25] sparc64: Guard against flushing openfirmware mappings. [ Upstream commit 4ca9a23765da3260058db3431faf5b4efd8cf926 ] Based almost entirely upon a patch by Christopher Alexander Tobias Schulze. In commit db64fe02258f1507e13fe5212a989922323685ce ("mm: rewrite vmap layer") lazy VMAP tlb flushing was added to the vmalloc layer. This causes problems on sparc64. Sparc64 has two VMAP mapped regions and they are not contiguous with eachother. First we have the malloc mapping area, then another unrelated region, then the vmalloc region. This "another unrelated region" is where the firmware is mapped. If the lazy TLB flushing logic in the vmalloc code triggers after we've had both a module unload and a vfree or similar, it will pass an address range that goes from somewhere inside the malloc region to somewhere inside the vmalloc region, and thus covering the openfirmware area entirely. The sparc64 kernel learns about openfirmware's dynamic mappings in this region early in the boot, and then services TLB misses in this area. But openfirmware has some locked TLB entries which are not mentioned in those dynamic mappings and we should thus not disturb them. These huge lazy TLB flush ranges causes those openfirmware locked TLB entries to be removed, resulting in all kinds of problems including hard hangs and crashes during reboot/reset. Besides causing problems like this, such huge TLB flush ranges are also incredibly inefficient. A plea has been made with the author of the VMAP lazy TLB flushing code, but for now we'll put a safety guard into our flush_tlb_kernel_range() implementation. Since the implementation has become non-trivial, stop defining it as a macro and instead make it a function in a C source file. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/include/asm/tlbflush_64.h | 12 ++---------- arch/sparc/mm/init_64.c | 23 +++++++++++++++++++++++ 2 files changed, 25 insertions(+), 10 deletions(-) diff --git a/arch/sparc/include/asm/tlbflush_64.h b/arch/sparc/include/asm/tlbflush_64.h index f0d6a9700f4c..1a4bb971e06d 100644 --- a/arch/sparc/include/asm/tlbflush_64.h +++ b/arch/sparc/include/asm/tlbflush_64.h @@ -35,6 +35,8 @@ static inline void flush_tlb_range(struct vm_area_struct *vma, { } +void flush_tlb_kernel_range(unsigned long start, unsigned long end); + #define __HAVE_ARCH_ENTER_LAZY_MMU_MODE extern void flush_tlb_pending(void); @@ -49,11 +51,6 @@ extern void __flush_tlb_kernel_range(unsigned long start, unsigned long end); #ifndef CONFIG_SMP -#define flush_tlb_kernel_range(start,end) \ -do { flush_tsb_kernel_range(start,end); \ - __flush_tlb_kernel_range(start,end); \ -} while (0) - static inline void global_flush_tlb_page(struct mm_struct *mm, unsigned long vaddr) { __flush_tlb_page(CTX_HWBITS(mm->context), vaddr); @@ -64,11 +61,6 @@ static inline void global_flush_tlb_page(struct mm_struct *mm, unsigned long vad extern void smp_flush_tlb_kernel_range(unsigned long start, unsigned long end); extern void smp_flush_tlb_page(struct mm_struct *mm, unsigned long vaddr); -#define flush_tlb_kernel_range(start, end) \ -do { flush_tsb_kernel_range(start,end); \ - smp_flush_tlb_kernel_range(start, end); \ -} while (0) - #define global_flush_tlb_page(mm, vaddr) \ smp_flush_tlb_page(mm, vaddr) diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index d6c1c85ff1c6..a751023dbdcd 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -2768,3 +2768,26 @@ void hugetlb_setup(struct pt_regs *regs) } } #endif + +#ifdef CONFIG_SMP +#define do_flush_tlb_kernel_range smp_flush_tlb_kernel_range +#else +#define do_flush_tlb_kernel_range __flush_tlb_kernel_range +#endif + +void flush_tlb_kernel_range(unsigned long start, unsigned long end) +{ + if (start < HI_OBP_ADDRESS && end > LOW_OBP_ADDRESS) { + if (start < LOW_OBP_ADDRESS) { + flush_tsb_kernel_range(start, LOW_OBP_ADDRESS); + do_flush_tlb_kernel_range(start, LOW_OBP_ADDRESS); + } + if (end > HI_OBP_ADDRESS) { + flush_tsb_kernel_range(end, HI_OBP_ADDRESS); + do_flush_tlb_kernel_range(end, HI_OBP_ADDRESS); + } + } else { + flush_tsb_kernel_range(start, end); + do_flush_tlb_kernel_range(start, end); + } +} From c1853d6752caa7843dfd51121df1f95697514499 Mon Sep 17 00:00:00 2001 From: Christopher Alexander Tobias Schulze Date: Sun, 3 Aug 2014 15:44:52 +0200 Subject: [PATCH 21/25] bbc-i2c: Fix BBC I2C envctrl on SunBlade 2000 [ Upstream commit 5cdceab3d5e02eb69ea0f5d8fa9181800baf6f77 ] Fix regression in bbc i2c temperature and fan control on some Sun systems that causes the driver to refuse to load due to the bbc_i2c_bussel resource not being present on the (second) i2c bus where the temperature sensors and fan control are located. (The check for the number of resources was removed when the driver was ported to a pure OF driver in mid 2008.) Signed-off-by: Christopher Alexander Tobias Schulze Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/sbus/char/bbc_envctrl.c | 6 ++++++ drivers/sbus/char/bbc_i2c.c | 11 ++++++++--- 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/drivers/sbus/char/bbc_envctrl.c b/drivers/sbus/char/bbc_envctrl.c index 160e7510aca6..0787b9756165 100644 --- a/drivers/sbus/char/bbc_envctrl.c +++ b/drivers/sbus/char/bbc_envctrl.c @@ -452,6 +452,9 @@ static void attach_one_temp(struct bbc_i2c_bus *bp, struct platform_device *op, if (!tp) return; + INIT_LIST_HEAD(&tp->bp_list); + INIT_LIST_HEAD(&tp->glob_list); + tp->client = bbc_i2c_attach(bp, op); if (!tp->client) { kfree(tp); @@ -497,6 +500,9 @@ static void attach_one_fan(struct bbc_i2c_bus *bp, struct platform_device *op, if (!fp) return; + INIT_LIST_HEAD(&fp->bp_list); + INIT_LIST_HEAD(&fp->glob_list); + fp->client = bbc_i2c_attach(bp, op); if (!fp->client) { kfree(fp); diff --git a/drivers/sbus/char/bbc_i2c.c b/drivers/sbus/char/bbc_i2c.c index c1441ed282eb..e0e6cd605cca 100644 --- a/drivers/sbus/char/bbc_i2c.c +++ b/drivers/sbus/char/bbc_i2c.c @@ -301,13 +301,18 @@ static struct bbc_i2c_bus * attach_one_i2c(struct platform_device *op, int index if (!bp) return NULL; + INIT_LIST_HEAD(&bp->temps); + INIT_LIST_HEAD(&bp->fans); + bp->i2c_control_regs = of_ioremap(&op->resource[0], 0, 0x2, "bbc_i2c_regs"); if (!bp->i2c_control_regs) goto fail; - bp->i2c_bussel_reg = of_ioremap(&op->resource[1], 0, 0x1, "bbc_i2c_bussel"); - if (!bp->i2c_bussel_reg) - goto fail; + if (op->num_resources == 2) { + bp->i2c_bussel_reg = of_ioremap(&op->resource[1], 0, 0x1, "bbc_i2c_bussel"); + if (!bp->i2c_bussel_reg) + goto fail; + } bp->waiting = 0; init_waitqueue_head(&bp->wq); From ff0adae5df0ff94592f03f84c952e624f0945a7c Mon Sep 17 00:00:00 2001 From: Christopher Alexander Tobias Schulze Date: Sun, 3 Aug 2014 16:01:53 +0200 Subject: [PATCH 22/25] sunsab: Fix detection of BREAK on sunsab serial console [ Upstream commit fe418231b195c205701c0cc550a03f6c9758fd9e ] Fix detection of BREAK on sunsab serial console: BREAK detection was only performed when there were also serial characters received simultaneously. To handle all BREAKs correctly, the check for BREAK and the corresponding call to uart_handle_break() must also be done if count == 0, therefore duplicate this code fragment and pull it out of the loop over the received characters. Patch applies to 3.16-rc6. Signed-off-by: Christopher Alexander Tobias Schulze Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/sunsab.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/tty/serial/sunsab.c b/drivers/tty/serial/sunsab.c index a422c8b55a47..aa53fee1df63 100644 --- a/drivers/tty/serial/sunsab.c +++ b/drivers/tty/serial/sunsab.c @@ -157,6 +157,15 @@ receive_chars(struct uart_sunsab_port *up, (up->port.line == up->port.cons->index)) saw_console_brk = 1; + if (count == 0) { + if (unlikely(stat->sreg.isr1 & SAB82532_ISR1_BRK)) { + stat->sreg.isr0 &= ~(SAB82532_ISR0_PERR | + SAB82532_ISR0_FERR); + up->port.icount.brk++; + uart_handle_break(&up->port); + } + } + for (i = 0; i < count; i++) { unsigned char ch = buf[i], flag; From 87c64964c0f424ec6b992f4d81516d698eb0a413 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Fri, 1 Aug 2014 09:50:40 -0400 Subject: [PATCH 23/25] sparc64: ldc_connect() should not return EINVAL when handshake is in progress. [ Upstream commit 4ec1b01029b4facb651b8ef70bc20a4be4cebc63 ] The LDC handshake could have been asynchronously triggered after ldc_bind() enables the ldc_rx() receive interrupt-handler (and thus intercepts incoming control packets) and before vio_port_up() calls ldc_connect(). If that is the case, ldc_connect() should return 0 and let the state-machine progress. Signed-off-by: Sowmini Varadhan Acked-by: Karl Volz Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/kernel/ldc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/sparc/kernel/ldc.c b/arch/sparc/kernel/ldc.c index 54df554b82d9..fa4c900a0d1f 100644 --- a/arch/sparc/kernel/ldc.c +++ b/arch/sparc/kernel/ldc.c @@ -1336,7 +1336,7 @@ int ldc_connect(struct ldc_channel *lp) if (!(lp->flags & LDC_FLAG_ALLOCED_QUEUES) || !(lp->flags & LDC_FLAG_REGISTERED_QUEUES) || lp->hs_state != LDC_HS_OPEN) - err = -EINVAL; + err = ((lp->hs_state > LDC_HS_OPEN) ? 0 : -EINVAL); else err = start_handshake(lp); From df6023479d431628fff9bd425ecc05ee593561c1 Mon Sep 17 00:00:00 2001 From: Andrey Utkin Date: Mon, 4 Aug 2014 23:47:41 +0300 Subject: [PATCH 24/25] arch/sparc/math-emu/math_32.c: drop stray break operator [ Upstream commit 093758e3daede29cb4ce6aedb111becf9d4bfc57 ] This commit is a guesswork, but it seems to make sense to drop this break, as otherwise the following line is never executed and becomes dead code. And that following line actually saves the result of local calculation by the pointer given in function argument. So the proposed change makes sense if this code in the whole makes sense (but I am unable to analyze it in the whole). Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=81641 Reported-by: David Binderman Signed-off-by: Andrey Utkin Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- arch/sparc/math-emu/math_32.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/sparc/math-emu/math_32.c b/arch/sparc/math-emu/math_32.c index aa4d55b0bdf0..5ce8f2f64604 100644 --- a/arch/sparc/math-emu/math_32.c +++ b/arch/sparc/math-emu/math_32.c @@ -499,7 +499,7 @@ static int do_one_mathemu(u32 insn, unsigned long *pfsr, unsigned long *fregs) case 0: fsr = *pfsr; if (IR == -1) IR = 2; /* fcc is always fcc0 */ - fsr &= ~0xc00; fsr |= (IR << 10); break; + fsr &= ~0xc00; fsr |= (IR << 10); *pfsr = fsr; break; case 1: rd->s = IR; break; From 7f363d2d04aa06a58619ffb5f22a84ae4f362c17 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 14 Aug 2014 09:24:29 +0800 Subject: [PATCH 25/25] Linux 3.10.53 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index b94f00938acc..2ac415a7e937 100644 --- a/Makefile +++ b/Makefile @@ -1,6 +1,6 @@ VERSION = 3 PATCHLEVEL = 10 -SUBLEVEL = 52 +SUBLEVEL = 53 EXTRAVERSION = NAME = TOSSUG Baby Fish