From df7f160dfdc768d07374c52465c191cb0e36cf31 Mon Sep 17 00:00:00 2001 From: Kyle Zeng Date: Thu, 14 Sep 2023 22:12:57 -0700 Subject: [PATCH 01/13] UPSTREAM: ipv4: fix null-deref in ipv4_link_failure [ Upstream commit 0113d9c9d1ccc07f5a3710dac4aa24b6d711278c ] Currently, we assume the skb is associated with a device before calling __ip_options_compile, which is not always the case if it is re-routed by ipvs. When skb->dev is NULL, dev_net(skb->dev) will become null-dereference. This patch adds a check for the edge case and switch to use the net_device from the rtable when skb->dev is NULL. Bug: 304913674 Fixes: ed0de45a1008 ("ipv4: recompile ip options in ipv4_link_failure") Suggested-by: David Ahern Signed-off-by: Kyle Zeng Cc: Stephen Suryaputra Cc: Vadim Fedorenko Reviewed-by: David Ahern Signed-off-by: David S. Miller Signed-off-by: Sasha Levin (cherry picked from commit 2712545e535d7a2e4c53b9c9658a9c88c6055862) Signed-off-by: Lee Jones Change-Id: Ie840ff3351e487f7095c49fac4fdd1e81021a982 --- net/ipv4/route.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 7004e379c325..f82d456afd0e 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1221,6 +1221,7 @@ static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie) static void ipv4_send_dest_unreach(struct sk_buff *skb) { + struct net_device *dev; struct ip_options opt; int res; @@ -1238,7 +1239,8 @@ static void ipv4_send_dest_unreach(struct sk_buff *skb) opt.optlen = ip_hdr(skb)->ihl * 4 - sizeof(struct iphdr); rcu_read_lock(); - res = __ip_options_compile(dev_net(skb->dev), &opt, skb, NULL); + dev = skb->dev ? skb->dev : skb_rtable(skb)->dst.dev; + res = __ip_options_compile(dev_net(dev), &opt, skb, NULL); rcu_read_unlock(); if (res) From 72e1ceae4b346d808e93e46d09c70c3e5d43e14b Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Tue, 14 Feb 2023 08:49:15 -0500 Subject: [PATCH 02/13] UPSTREAM: net/sched: Retire rsvp classifier commit 265b4da82dbf5df04bee5a5d46b7474b1aaf326a upstream. The rsvp classifier has served us well for about a quarter of a century but has has not been getting much maintenance attention due to lack of known users. Bug: 304913975 Signed-off-by: Jamal Hadi Salim Acked-by: Jiri Pirko Signed-off-by: Paolo Abeni Signed-off-by: Kyle Zeng Signed-off-by: Greg Kroah-Hartman (cherry picked from commit 08569c92f7f339de21b7a68d43d6795fc0aa24f2) Signed-off-by: Lee Jones Change-Id: I4716954c5e8d5414a580eb34c699908028aa754b --- net/sched/Kconfig | 28 -- net/sched/Makefile | 2 - net/sched/cls_rsvp.c | 24 -- net/sched/cls_rsvp.h | 777 ------------------------------------------ net/sched/cls_rsvp6.c | 24 -- 5 files changed, 855 deletions(-) delete mode 100644 net/sched/cls_rsvp.c delete mode 100644 net/sched/cls_rsvp.h delete mode 100644 net/sched/cls_rsvp6.c diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 2e4604cdaf23..46f2847a071e 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -514,34 +514,6 @@ config CLS_U32_MARK ---help--- Say Y here to be able to use netfilter marks as u32 key. -config NET_CLS_RSVP - tristate "IPv4 Resource Reservation Protocol (RSVP)" - select NET_CLS - ---help--- - The Resource Reservation Protocol (RSVP) permits end systems to - request a minimum and maximum data flow rate for a connection; this - is important for real time data such as streaming sound or video. - - Say Y here if you want to be able to classify outgoing packets based - on their RSVP requests. - - To compile this code as a module, choose M here: the - module will be called cls_rsvp. - -config NET_CLS_RSVP6 - tristate "IPv6 Resource Reservation Protocol (RSVP6)" - select NET_CLS - ---help--- - The Resource Reservation Protocol (RSVP) permits end systems to - request a minimum and maximum data flow rate for a connection; this - is important for real time data such as streaming sound or video. - - Say Y here if you want to be able to classify outgoing packets based - on their RSVP requests and you are using the IPv6 protocol. - - To compile this code as a module, choose M here: the - module will be called cls_rsvp6. - config NET_CLS_FLOW tristate "Flow classifier" select NET_CLS diff --git a/net/sched/Makefile b/net/sched/Makefile index 9e1f904d0c18..fb2b90648a20 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -65,8 +65,6 @@ obj-$(CONFIG_NET_SCH_TAPRIO) += sch_taprio.o obj-$(CONFIG_NET_CLS_U32) += cls_u32.o obj-$(CONFIG_NET_CLS_ROUTE4) += cls_route.o obj-$(CONFIG_NET_CLS_FW) += cls_fw.o -obj-$(CONFIG_NET_CLS_RSVP) += cls_rsvp.o -obj-$(CONFIG_NET_CLS_RSVP6) += cls_rsvp6.o obj-$(CONFIG_NET_CLS_BASIC) += cls_basic.o obj-$(CONFIG_NET_CLS_FLOW) += cls_flow.o obj-$(CONFIG_NET_CLS_CGROUP) += cls_cgroup.o diff --git a/net/sched/cls_rsvp.c b/net/sched/cls_rsvp.c deleted file mode 100644 index de1c1d4da597..000000000000 --- a/net/sched/cls_rsvp.c +++ /dev/null @@ -1,24 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * net/sched/cls_rsvp.c Special RSVP packet classifier for IPv4. - * - * Authors: Alexey Kuznetsov, - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#define RSVP_DST_LEN 1 -#define RSVP_ID "rsvp" -#define RSVP_OPS cls_rsvp_ops - -#include "cls_rsvp.h" -MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h deleted file mode 100644 index d36949d9382c..000000000000 --- a/net/sched/cls_rsvp.h +++ /dev/null @@ -1,777 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-or-later */ -/* - * net/sched/cls_rsvp.h Template file for RSVPv[46] classifiers. - * - * Authors: Alexey Kuznetsov, - */ - -/* - Comparing to general packet classification problem, - RSVP needs only sevaral relatively simple rules: - - * (dst, protocol) are always specified, - so that we are able to hash them. - * src may be exact, or may be wildcard, so that - we can keep a hash table plus one wildcard entry. - * source port (or flow label) is important only if src is given. - - IMPLEMENTATION. - - We use a two level hash table: The top level is keyed by - destination address and protocol ID, every bucket contains a list - of "rsvp sessions", identified by destination address, protocol and - DPI(="Destination Port ID"): triple (key, mask, offset). - - Every bucket has a smaller hash table keyed by source address - (cf. RSVP flowspec) and one wildcard entry for wildcard reservations. - Every bucket is again a list of "RSVP flows", selected by - source address and SPI(="Source Port ID" here rather than - "security parameter index"): triple (key, mask, offset). - - - NOTE 1. All the packets with IPv6 extension headers (but AH and ESP) - and all fragmented packets go to the best-effort traffic class. - - - NOTE 2. Two "port id"'s seems to be redundant, rfc2207 requires - only one "Generalized Port Identifier". So that for classic - ah, esp (and udp,tcp) both *pi should coincide or one of them - should be wildcard. - - At first sight, this redundancy is just a waste of CPU - resources. But DPI and SPI add the possibility to assign different - priorities to GPIs. Look also at note 4 about tunnels below. - - - NOTE 3. One complication is the case of tunneled packets. - We implement it as following: if the first lookup - matches a special session with "tunnelhdr" value not zero, - flowid doesn't contain the true flow ID, but the tunnel ID (1...255). - In this case, we pull tunnelhdr bytes and restart lookup - with tunnel ID added to the list of keys. Simple and stupid 8)8) - It's enough for PIMREG and IPIP. - - - NOTE 4. Two GPIs make it possible to parse even GRE packets. - F.e. DPI can select ETH_P_IP (and necessary flags to make - tunnelhdr correct) in GRE protocol field and SPI matches - GRE key. Is it not nice? 8)8) - - - Well, as result, despite its simplicity, we get a pretty - powerful classification engine. */ - - -struct rsvp_head { - u32 tmap[256/32]; - u32 hgenerator; - u8 tgenerator; - struct rsvp_session __rcu *ht[256]; - struct rcu_head rcu; -}; - -struct rsvp_session { - struct rsvp_session __rcu *next; - __be32 dst[RSVP_DST_LEN]; - struct tc_rsvp_gpi dpi; - u8 protocol; - u8 tunnelid; - /* 16 (src,sport) hash slots, and one wildcard source slot */ - struct rsvp_filter __rcu *ht[16 + 1]; - struct rcu_head rcu; -}; - - -struct rsvp_filter { - struct rsvp_filter __rcu *next; - __be32 src[RSVP_DST_LEN]; - struct tc_rsvp_gpi spi; - u8 tunnelhdr; - - struct tcf_result res; - struct tcf_exts exts; - - u32 handle; - struct rsvp_session *sess; - struct rcu_work rwork; -}; - -static inline unsigned int hash_dst(__be32 *dst, u8 protocol, u8 tunnelid) -{ - unsigned int h = (__force __u32)dst[RSVP_DST_LEN - 1]; - - h ^= h>>16; - h ^= h>>8; - return (h ^ protocol ^ tunnelid) & 0xFF; -} - -static inline unsigned int hash_src(__be32 *src) -{ - unsigned int h = (__force __u32)src[RSVP_DST_LEN-1]; - - h ^= h>>16; - h ^= h>>8; - h ^= h>>4; - return h & 0xF; -} - -#define RSVP_APPLY_RESULT() \ -{ \ - int r = tcf_exts_exec(skb, &f->exts, res); \ - if (r < 0) \ - continue; \ - else if (r > 0) \ - return r; \ -} - -static int rsvp_classify(struct sk_buff *skb, const struct tcf_proto *tp, - struct tcf_result *res) -{ - struct rsvp_head *head = rcu_dereference_bh(tp->root); - struct rsvp_session *s; - struct rsvp_filter *f; - unsigned int h1, h2; - __be32 *dst, *src; - u8 protocol; - u8 tunnelid = 0; - u8 *xprt; -#if RSVP_DST_LEN == 4 - struct ipv6hdr *nhptr; - - if (!pskb_network_may_pull(skb, sizeof(*nhptr))) - return -1; - nhptr = ipv6_hdr(skb); -#else - struct iphdr *nhptr; - - if (!pskb_network_may_pull(skb, sizeof(*nhptr))) - return -1; - nhptr = ip_hdr(skb); -#endif -restart: - -#if RSVP_DST_LEN == 4 - src = &nhptr->saddr.s6_addr32[0]; - dst = &nhptr->daddr.s6_addr32[0]; - protocol = nhptr->nexthdr; - xprt = ((u8 *)nhptr) + sizeof(struct ipv6hdr); -#else - src = &nhptr->saddr; - dst = &nhptr->daddr; - protocol = nhptr->protocol; - xprt = ((u8 *)nhptr) + (nhptr->ihl<<2); - if (ip_is_fragment(nhptr)) - return -1; -#endif - - h1 = hash_dst(dst, protocol, tunnelid); - h2 = hash_src(src); - - for (s = rcu_dereference_bh(head->ht[h1]); s; - s = rcu_dereference_bh(s->next)) { - if (dst[RSVP_DST_LEN-1] == s->dst[RSVP_DST_LEN - 1] && - protocol == s->protocol && - !(s->dpi.mask & - (*(u32 *)(xprt + s->dpi.offset) ^ s->dpi.key)) && -#if RSVP_DST_LEN == 4 - dst[0] == s->dst[0] && - dst[1] == s->dst[1] && - dst[2] == s->dst[2] && -#endif - tunnelid == s->tunnelid) { - - for (f = rcu_dereference_bh(s->ht[h2]); f; - f = rcu_dereference_bh(f->next)) { - if (src[RSVP_DST_LEN-1] == f->src[RSVP_DST_LEN - 1] && - !(f->spi.mask & (*(u32 *)(xprt + f->spi.offset) ^ f->spi.key)) -#if RSVP_DST_LEN == 4 - && - src[0] == f->src[0] && - src[1] == f->src[1] && - src[2] == f->src[2] -#endif - ) { - *res = f->res; - RSVP_APPLY_RESULT(); - -matched: - if (f->tunnelhdr == 0) - return 0; - - tunnelid = f->res.classid; - nhptr = (void *)(xprt + f->tunnelhdr - sizeof(*nhptr)); - goto restart; - } - } - - /* And wildcard bucket... */ - for (f = rcu_dereference_bh(s->ht[16]); f; - f = rcu_dereference_bh(f->next)) { - *res = f->res; - RSVP_APPLY_RESULT(); - goto matched; - } - return -1; - } - } - return -1; -} - -static void rsvp_replace(struct tcf_proto *tp, struct rsvp_filter *n, u32 h) -{ - struct rsvp_head *head = rtnl_dereference(tp->root); - struct rsvp_session *s; - struct rsvp_filter __rcu **ins; - struct rsvp_filter *pins; - unsigned int h1 = h & 0xFF; - unsigned int h2 = (h >> 8) & 0xFF; - - for (s = rtnl_dereference(head->ht[h1]); s; - s = rtnl_dereference(s->next)) { - for (ins = &s->ht[h2], pins = rtnl_dereference(*ins); ; - ins = &pins->next, pins = rtnl_dereference(*ins)) { - if (pins->handle == h) { - RCU_INIT_POINTER(n->next, pins->next); - rcu_assign_pointer(*ins, n); - return; - } - } - } - - /* Something went wrong if we are trying to replace a non-existant - * node. Mind as well halt instead of silently failing. - */ - BUG_ON(1); -} - -static void *rsvp_get(struct tcf_proto *tp, u32 handle) -{ - struct rsvp_head *head = rtnl_dereference(tp->root); - struct rsvp_session *s; - struct rsvp_filter *f; - unsigned int h1 = handle & 0xFF; - unsigned int h2 = (handle >> 8) & 0xFF; - - if (h2 > 16) - return NULL; - - for (s = rtnl_dereference(head->ht[h1]); s; - s = rtnl_dereference(s->next)) { - for (f = rtnl_dereference(s->ht[h2]); f; - f = rtnl_dereference(f->next)) { - if (f->handle == handle) - return f; - } - } - return NULL; -} - -static int rsvp_init(struct tcf_proto *tp) -{ - struct rsvp_head *data; - - data = kzalloc(sizeof(struct rsvp_head), GFP_KERNEL); - if (data) { - rcu_assign_pointer(tp->root, data); - return 0; - } - return -ENOBUFS; -} - -static void __rsvp_delete_filter(struct rsvp_filter *f) -{ - tcf_exts_destroy(&f->exts); - tcf_exts_put_net(&f->exts); - kfree(f); -} - -static void rsvp_delete_filter_work(struct work_struct *work) -{ - struct rsvp_filter *f = container_of(to_rcu_work(work), - struct rsvp_filter, - rwork); - rtnl_lock(); - __rsvp_delete_filter(f); - rtnl_unlock(); -} - -static void rsvp_delete_filter(struct tcf_proto *tp, struct rsvp_filter *f) -{ - tcf_unbind_filter(tp, &f->res); - /* all classifiers are required to call tcf_exts_destroy() after rcu - * grace period, since converted-to-rcu actions are relying on that - * in cleanup() callback - */ - if (tcf_exts_get_net(&f->exts)) - tcf_queue_work(&f->rwork, rsvp_delete_filter_work); - else - __rsvp_delete_filter(f); -} - -static void rsvp_destroy(struct tcf_proto *tp, bool rtnl_held, - struct netlink_ext_ack *extack) -{ - struct rsvp_head *data = rtnl_dereference(tp->root); - int h1, h2; - - if (data == NULL) - return; - - for (h1 = 0; h1 < 256; h1++) { - struct rsvp_session *s; - - while ((s = rtnl_dereference(data->ht[h1])) != NULL) { - RCU_INIT_POINTER(data->ht[h1], s->next); - - for (h2 = 0; h2 <= 16; h2++) { - struct rsvp_filter *f; - - while ((f = rtnl_dereference(s->ht[h2])) != NULL) { - rcu_assign_pointer(s->ht[h2], f->next); - rsvp_delete_filter(tp, f); - } - } - kfree_rcu(s, rcu); - } - } - kfree_rcu(data, rcu); -} - -static int rsvp_delete(struct tcf_proto *tp, void *arg, bool *last, - bool rtnl_held, struct netlink_ext_ack *extack) -{ - struct rsvp_head *head = rtnl_dereference(tp->root); - struct rsvp_filter *nfp, *f = arg; - struct rsvp_filter __rcu **fp; - unsigned int h = f->handle; - struct rsvp_session __rcu **sp; - struct rsvp_session *nsp, *s = f->sess; - int i, h1; - - fp = &s->ht[(h >> 8) & 0xFF]; - for (nfp = rtnl_dereference(*fp); nfp; - fp = &nfp->next, nfp = rtnl_dereference(*fp)) { - if (nfp == f) { - RCU_INIT_POINTER(*fp, f->next); - rsvp_delete_filter(tp, f); - - /* Strip tree */ - - for (i = 0; i <= 16; i++) - if (s->ht[i]) - goto out; - - /* OK, session has no flows */ - sp = &head->ht[h & 0xFF]; - for (nsp = rtnl_dereference(*sp); nsp; - sp = &nsp->next, nsp = rtnl_dereference(*sp)) { - if (nsp == s) { - RCU_INIT_POINTER(*sp, s->next); - kfree_rcu(s, rcu); - goto out; - } - } - - break; - } - } - -out: - *last = true; - for (h1 = 0; h1 < 256; h1++) { - if (rcu_access_pointer(head->ht[h1])) { - *last = false; - break; - } - } - - return 0; -} - -static unsigned int gen_handle(struct tcf_proto *tp, unsigned salt) -{ - struct rsvp_head *data = rtnl_dereference(tp->root); - int i = 0xFFFF; - - while (i-- > 0) { - u32 h; - - if ((data->hgenerator += 0x10000) == 0) - data->hgenerator = 0x10000; - h = data->hgenerator|salt; - if (!rsvp_get(tp, h)) - return h; - } - return 0; -} - -static int tunnel_bts(struct rsvp_head *data) -{ - int n = data->tgenerator >> 5; - u32 b = 1 << (data->tgenerator & 0x1F); - - if (data->tmap[n] & b) - return 0; - data->tmap[n] |= b; - return 1; -} - -static void tunnel_recycle(struct rsvp_head *data) -{ - struct rsvp_session __rcu **sht = data->ht; - u32 tmap[256/32]; - int h1, h2; - - memset(tmap, 0, sizeof(tmap)); - - for (h1 = 0; h1 < 256; h1++) { - struct rsvp_session *s; - for (s = rtnl_dereference(sht[h1]); s; - s = rtnl_dereference(s->next)) { - for (h2 = 0; h2 <= 16; h2++) { - struct rsvp_filter *f; - - for (f = rtnl_dereference(s->ht[h2]); f; - f = rtnl_dereference(f->next)) { - if (f->tunnelhdr == 0) - continue; - data->tgenerator = f->res.classid; - tunnel_bts(data); - } - } - } - } - - memcpy(data->tmap, tmap, sizeof(tmap)); -} - -static u32 gen_tunnel(struct rsvp_head *data) -{ - int i, k; - - for (k = 0; k < 2; k++) { - for (i = 255; i > 0; i--) { - if (++data->tgenerator == 0) - data->tgenerator = 1; - if (tunnel_bts(data)) - return data->tgenerator; - } - tunnel_recycle(data); - } - return 0; -} - -static const struct nla_policy rsvp_policy[TCA_RSVP_MAX + 1] = { - [TCA_RSVP_CLASSID] = { .type = NLA_U32 }, - [TCA_RSVP_DST] = { .len = RSVP_DST_LEN * sizeof(u32) }, - [TCA_RSVP_SRC] = { .len = RSVP_DST_LEN * sizeof(u32) }, - [TCA_RSVP_PINFO] = { .len = sizeof(struct tc_rsvp_pinfo) }, -}; - -static int rsvp_change(struct net *net, struct sk_buff *in_skb, - struct tcf_proto *tp, unsigned long base, - u32 handle, - struct nlattr **tca, - void **arg, bool ovr, bool rtnl_held, - struct netlink_ext_ack *extack) -{ - struct rsvp_head *data = rtnl_dereference(tp->root); - struct rsvp_filter *f, *nfp; - struct rsvp_filter __rcu **fp; - struct rsvp_session *nsp, *s; - struct rsvp_session __rcu **sp; - struct tc_rsvp_pinfo *pinfo = NULL; - struct nlattr *opt = tca[TCA_OPTIONS]; - struct nlattr *tb[TCA_RSVP_MAX + 1]; - struct tcf_exts e; - unsigned int h1, h2; - __be32 *dst; - int err; - - if (opt == NULL) - return handle ? -EINVAL : 0; - - err = nla_parse_nested_deprecated(tb, TCA_RSVP_MAX, opt, rsvp_policy, - NULL); - if (err < 0) - return err; - - err = tcf_exts_init(&e, net, TCA_RSVP_ACT, TCA_RSVP_POLICE); - if (err < 0) - return err; - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr, true, - extack); - if (err < 0) - goto errout2; - - f = *arg; - if (f) { - /* Node exists: adjust only classid */ - struct rsvp_filter *n; - - if (f->handle != handle && handle) - goto errout2; - - n = kmemdup(f, sizeof(*f), GFP_KERNEL); - if (!n) { - err = -ENOMEM; - goto errout2; - } - - err = tcf_exts_init(&n->exts, net, TCA_RSVP_ACT, - TCA_RSVP_POLICE); - if (err < 0) { - kfree(n); - goto errout2; - } - - if (tb[TCA_RSVP_CLASSID]) { - n->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]); - tcf_bind_filter(tp, &n->res, base); - } - - tcf_exts_change(&n->exts, &e); - rsvp_replace(tp, n, handle); - return 0; - } - - /* Now more serious part... */ - err = -EINVAL; - if (handle) - goto errout2; - if (tb[TCA_RSVP_DST] == NULL) - goto errout2; - - err = -ENOBUFS; - f = kzalloc(sizeof(struct rsvp_filter), GFP_KERNEL); - if (f == NULL) - goto errout2; - - err = tcf_exts_init(&f->exts, net, TCA_RSVP_ACT, TCA_RSVP_POLICE); - if (err < 0) - goto errout; - h2 = 16; - if (tb[TCA_RSVP_SRC]) { - memcpy(f->src, nla_data(tb[TCA_RSVP_SRC]), sizeof(f->src)); - h2 = hash_src(f->src); - } - if (tb[TCA_RSVP_PINFO]) { - pinfo = nla_data(tb[TCA_RSVP_PINFO]); - f->spi = pinfo->spi; - f->tunnelhdr = pinfo->tunnelhdr; - } - if (tb[TCA_RSVP_CLASSID]) - f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]); - - dst = nla_data(tb[TCA_RSVP_DST]); - h1 = hash_dst(dst, pinfo ? pinfo->protocol : 0, pinfo ? pinfo->tunnelid : 0); - - err = -ENOMEM; - if ((f->handle = gen_handle(tp, h1 | (h2<<8))) == 0) - goto errout; - - if (f->tunnelhdr) { - err = -EINVAL; - if (f->res.classid > 255) - goto errout; - - err = -ENOMEM; - if (f->res.classid == 0 && - (f->res.classid = gen_tunnel(data)) == 0) - goto errout; - } - - for (sp = &data->ht[h1]; - (s = rtnl_dereference(*sp)) != NULL; - sp = &s->next) { - if (dst[RSVP_DST_LEN-1] == s->dst[RSVP_DST_LEN-1] && - pinfo && pinfo->protocol == s->protocol && - memcmp(&pinfo->dpi, &s->dpi, sizeof(s->dpi)) == 0 && -#if RSVP_DST_LEN == 4 - dst[0] == s->dst[0] && - dst[1] == s->dst[1] && - dst[2] == s->dst[2] && -#endif - pinfo->tunnelid == s->tunnelid) { - -insert: - /* OK, we found appropriate session */ - - fp = &s->ht[h2]; - - f->sess = s; - if (f->tunnelhdr == 0) - tcf_bind_filter(tp, &f->res, base); - - tcf_exts_change(&f->exts, &e); - - fp = &s->ht[h2]; - for (nfp = rtnl_dereference(*fp); nfp; - fp = &nfp->next, nfp = rtnl_dereference(*fp)) { - __u32 mask = nfp->spi.mask & f->spi.mask; - - if (mask != f->spi.mask) - break; - } - RCU_INIT_POINTER(f->next, nfp); - rcu_assign_pointer(*fp, f); - - *arg = f; - return 0; - } - } - - /* No session found. Create new one. */ - - err = -ENOBUFS; - s = kzalloc(sizeof(struct rsvp_session), GFP_KERNEL); - if (s == NULL) - goto errout; - memcpy(s->dst, dst, sizeof(s->dst)); - - if (pinfo) { - s->dpi = pinfo->dpi; - s->protocol = pinfo->protocol; - s->tunnelid = pinfo->tunnelid; - } - sp = &data->ht[h1]; - for (nsp = rtnl_dereference(*sp); nsp; - sp = &nsp->next, nsp = rtnl_dereference(*sp)) { - if ((nsp->dpi.mask & s->dpi.mask) != s->dpi.mask) - break; - } - RCU_INIT_POINTER(s->next, nsp); - rcu_assign_pointer(*sp, s); - - goto insert; - -errout: - tcf_exts_destroy(&f->exts); - kfree(f); -errout2: - tcf_exts_destroy(&e); - return err; -} - -static void rsvp_walk(struct tcf_proto *tp, struct tcf_walker *arg, - bool rtnl_held) -{ - struct rsvp_head *head = rtnl_dereference(tp->root); - unsigned int h, h1; - - if (arg->stop) - return; - - for (h = 0; h < 256; h++) { - struct rsvp_session *s; - - for (s = rtnl_dereference(head->ht[h]); s; - s = rtnl_dereference(s->next)) { - for (h1 = 0; h1 <= 16; h1++) { - struct rsvp_filter *f; - - for (f = rtnl_dereference(s->ht[h1]); f; - f = rtnl_dereference(f->next)) { - if (arg->count < arg->skip) { - arg->count++; - continue; - } - if (arg->fn(tp, f, arg) < 0) { - arg->stop = 1; - return; - } - arg->count++; - } - } - } - } -} - -static int rsvp_dump(struct net *net, struct tcf_proto *tp, void *fh, - struct sk_buff *skb, struct tcmsg *t, bool rtnl_held) -{ - struct rsvp_filter *f = fh; - struct rsvp_session *s; - struct nlattr *nest; - struct tc_rsvp_pinfo pinfo; - - if (f == NULL) - return skb->len; - s = f->sess; - - t->tcm_handle = f->handle; - - nest = nla_nest_start_noflag(skb, TCA_OPTIONS); - if (nest == NULL) - goto nla_put_failure; - - if (nla_put(skb, TCA_RSVP_DST, sizeof(s->dst), &s->dst)) - goto nla_put_failure; - pinfo.dpi = s->dpi; - pinfo.spi = f->spi; - pinfo.protocol = s->protocol; - pinfo.tunnelid = s->tunnelid; - pinfo.tunnelhdr = f->tunnelhdr; - pinfo.pad = 0; - if (nla_put(skb, TCA_RSVP_PINFO, sizeof(pinfo), &pinfo)) - goto nla_put_failure; - if (f->res.classid && - nla_put_u32(skb, TCA_RSVP_CLASSID, f->res.classid)) - goto nla_put_failure; - if (((f->handle >> 8) & 0xFF) != 16 && - nla_put(skb, TCA_RSVP_SRC, sizeof(f->src), f->src)) - goto nla_put_failure; - - if (tcf_exts_dump(skb, &f->exts) < 0) - goto nla_put_failure; - - nla_nest_end(skb, nest); - - if (tcf_exts_dump_stats(skb, &f->exts) < 0) - goto nla_put_failure; - return skb->len; - -nla_put_failure: - nla_nest_cancel(skb, nest); - return -1; -} - -static void rsvp_bind_class(void *fh, u32 classid, unsigned long cl, void *q, - unsigned long base) -{ - struct rsvp_filter *f = fh; - - if (f && f->res.classid == classid) { - if (cl) - __tcf_bind_filter(q, &f->res, base); - else - __tcf_unbind_filter(q, &f->res); - } -} - -static struct tcf_proto_ops RSVP_OPS __read_mostly = { - .kind = RSVP_ID, - .classify = rsvp_classify, - .init = rsvp_init, - .destroy = rsvp_destroy, - .get = rsvp_get, - .change = rsvp_change, - .delete = rsvp_delete, - .walk = rsvp_walk, - .dump = rsvp_dump, - .bind_class = rsvp_bind_class, - .owner = THIS_MODULE, -}; - -static int __init init_rsvp(void) -{ - return register_tcf_proto_ops(&RSVP_OPS); -} - -static void __exit exit_rsvp(void) -{ - unregister_tcf_proto_ops(&RSVP_OPS); -} - -module_init(init_rsvp) -module_exit(exit_rsvp) diff --git a/net/sched/cls_rsvp6.c b/net/sched/cls_rsvp6.c deleted file mode 100644 index 64078846000e..000000000000 --- a/net/sched/cls_rsvp6.c +++ /dev/null @@ -1,24 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * net/sched/cls_rsvp6.c Special RSVP packet classifier for IPv6. - * - * Authors: Alexey Kuznetsov, - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#define RSVP_DST_LEN 4 -#define RSVP_ID "rsvp6" -#define RSVP_OPS cls_rsvp6_ops - -#include "cls_rsvp.h" -MODULE_LICENSE("GPL"); From 1f18bb1187ae7df8b16001dcc4d99d045fa0831c Mon Sep 17 00:00:00 2001 From: Wander Lairson Costa Date: Fri, 1 Sep 2023 10:50:20 -0300 Subject: [PATCH 03/13] UPSTREAM: netfilter: nfnetlink_osf: avoid OOB read [ Upstream commit f4f8a7803119005e87b716874bec07c751efafec ] The opt_num field is controlled by user mode and is not currently validated inside the kernel. An attacker can take advantage of this to trigger an OOB read and potentially leak information. BUG: KASAN: slab-out-of-bounds in nf_osf_match_one+0xbed/0xd10 net/netfilter/nfnetlink_osf.c:88 Read of size 2 at addr ffff88804bc64272 by task poc/6431 CPU: 1 PID: 6431 Comm: poc Not tainted 6.0.0-rc4 #1 Call Trace: nf_osf_match_one+0xbed/0xd10 net/netfilter/nfnetlink_osf.c:88 nf_osf_find+0x186/0x2f0 net/netfilter/nfnetlink_osf.c:281 nft_osf_eval+0x37f/0x590 net/netfilter/nft_osf.c:47 expr_call_ops_eval net/netfilter/nf_tables_core.c:214 nft_do_chain+0x2b0/0x1490 net/netfilter/nf_tables_core.c:264 nft_do_chain_ipv4+0x17c/0x1f0 net/netfilter/nft_chain_filter.c:23 [..] Also add validation to genre, subtype and version fields. Bug: 304913642 Fixes: 11eeef41d5f6 ("netfilter: passive OS fingerprint xtables match") Reported-by: Lucas Leong Signed-off-by: Wander Lairson Costa Signed-off-by: Florian Westphal Signed-off-by: Sasha Levin (cherry picked from commit 7bb8d52b4271be7527b6e3120ae6ce4c6cdf6e34) Signed-off-by: Lee Jones Change-Id: If79c79e3f55de8c81b70c19661cb0084b02c3da2 --- net/netfilter/nfnetlink_osf.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/netfilter/nfnetlink_osf.c b/net/netfilter/nfnetlink_osf.c index 9dbaa5ce24e5..573a372e760f 100644 --- a/net/netfilter/nfnetlink_osf.c +++ b/net/netfilter/nfnetlink_osf.c @@ -316,6 +316,14 @@ static int nfnl_osf_add_callback(struct net *net, struct sock *ctnl, f = nla_data(osf_attrs[OSF_ATTR_FINGER]); + if (f->opt_num > ARRAY_SIZE(f->opt)) + return -EINVAL; + + if (!memchr(f->genre, 0, MAXGENRELEN) || + !memchr(f->subtype, 0, MAXGENRELEN) || + !memchr(f->version, 0, MAXGENRELEN)) + return -EINVAL; + kf = kmalloc(sizeof(struct nf_osf_finger), GFP_KERNEL); if (!kf) return -ENOMEM; From 16645e28d6737368479d027f0e53193cfc55e096 Mon Sep 17 00:00:00 2001 From: Wander Lairson Costa Date: Mon, 28 Aug 2023 10:21:07 -0300 Subject: [PATCH 04/13] UPSTREAM: netfilter: xt_u32: validate user space input commit 69c5d284f67089b4750d28ff6ac6f52ec224b330 upstream. The xt_u32 module doesn't validate the fields in the xt_u32 structure. An attacker may take advantage of this to trigger an OOB read by setting the size fields with a value beyond the arrays boundaries. Add a checkentry function to validate the structure. This was originally reported by the ZDI project (ZDI-CAN-18408). Bug: 304913716 Fixes: 1b50b8a371e9 ("[NETFILTER]: Add u32 match") Cc: stable@vger.kernel.org Signed-off-by: Wander Lairson Costa Signed-off-by: Pablo Neira Ayuso Signed-off-by: Greg Kroah-Hartman (cherry picked from commit 1c164c1e9e93b0a72a03a7edb754e3857d4e4302) Signed-off-by: Lee Jones Change-Id: Ic2ff70b303f55f9c3c5db24295bcb223ed7175a7 --- net/netfilter/xt_u32.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/net/netfilter/xt_u32.c b/net/netfilter/xt_u32.c index 177b40d08098..117d4615d668 100644 --- a/net/netfilter/xt_u32.c +++ b/net/netfilter/xt_u32.c @@ -96,11 +96,32 @@ static bool u32_mt(const struct sk_buff *skb, struct xt_action_param *par) return ret ^ data->invert; } +static int u32_mt_checkentry(const struct xt_mtchk_param *par) +{ + const struct xt_u32 *data = par->matchinfo; + const struct xt_u32_test *ct; + unsigned int i; + + if (data->ntests > ARRAY_SIZE(data->tests)) + return -EINVAL; + + for (i = 0; i < data->ntests; ++i) { + ct = &data->tests[i]; + + if (ct->nnums > ARRAY_SIZE(ct->location) || + ct->nvalues > ARRAY_SIZE(ct->value)) + return -EINVAL; + } + + return 0; +} + static struct xt_match xt_u32_mt_reg __read_mostly = { .name = "u32", .revision = 0, .family = NFPROTO_UNSPEC, .match = u32_mt, + .checkentry = u32_mt_checkentry, .matchsize = sizeof(struct xt_u32), .me = THIS_MODULE, }; From 5e99400d74d7ef9bfc7820db76dba30eeeafa3ab Mon Sep 17 00:00:00 2001 From: Wander Lairson Costa Date: Mon, 28 Aug 2023 19:12:55 -0300 Subject: [PATCH 05/13] UPSTREAM: netfilter: xt_sctp: validate the flag_info count commit e99476497687ef9e850748fe6d232264f30bc8f9 upstream. sctp_mt_check doesn't validate the flag_count field. An attacker can take advantage of that to trigger a OOB read and leak memory information. Add the field validation in the checkentry function. Bug: 304913898 Fixes: 2e4e6a17af35 ("[NETFILTER] x_tables: Abstraction layer for {ip,ip6,arp}_tables") Cc: stable@vger.kernel.org Reported-by: Lucas Leong Signed-off-by: Wander Lairson Costa Signed-off-by: Pablo Neira Ayuso Signed-off-by: Greg Kroah-Hartman (cherry picked from commit 4921f9349b66da7c5a2b6418fe45e9ae0ae72924) Signed-off-by: Lee Jones Change-Id: Ife4e69f6218fdaca2a8647b5ed00d875a5ed0d34 --- net/netfilter/xt_sctp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/netfilter/xt_sctp.c b/net/netfilter/xt_sctp.c index 680015ba7cb6..d4bf089c9e3f 100644 --- a/net/netfilter/xt_sctp.c +++ b/net/netfilter/xt_sctp.c @@ -150,6 +150,8 @@ static int sctp_mt_check(const struct xt_mtchk_param *par) { const struct xt_sctp_info *info = par->matchinfo; + if (info->flag_count > ARRAY_SIZE(info->flag_info)) + return -EINVAL; if (info->flags & ~XT_SCTP_VALID_FLAGS) return -EINVAL; if (info->invflags & ~XT_SCTP_VALID_FLAGS) From d5a866abc0566d3b97b4e6520c6eb96061bcac93 Mon Sep 17 00:00:00 2001 From: Radoslaw Tyl Date: Thu, 24 Aug 2023 13:46:19 -0700 Subject: [PATCH 06/13] UPSTREAM: igb: set max size RX buffer when store bad packet is enabled commit bb5ed01cd2428cd25b1c88a3a9cba87055eb289f upstream. Increase the RX buffer size to 3K when the SBP bit is on. The size of the RX buffer determines the number of pages allocated which may not be sufficient for receive frames larger than the set MTU size. Bug: 306483516 Cc: stable@vger.kernel.org Fixes: 89eaefb61dc9 ("igb: Support RX-ALL feature flag.") Reported-by: Manfred Rudigier Signed-off-by: Radoslaw Tyl Tested-by: Arpana Arland (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman (cherry picked from commit be7353af5b35c8f4a11573a38b997f74f8c2ae4b) Signed-off-by: Lee Jones Change-Id: I3f05b369cd9ac24643af3e37e2ce1d099c63bc02 --- drivers/net/ethernet/intel/igb/igb_main.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index 8c6c0d9c7f76..b8113235f281 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -4552,6 +4552,10 @@ void igb_configure_rx_ring(struct igb_adapter *adapter, static void igb_set_rx_buffer_len(struct igb_adapter *adapter, struct igb_ring *rx_ring) { +#if (PAGE_SIZE < 8192) + struct e1000_hw *hw = &adapter->hw; +#endif + /* set build_skb and buffer size flags */ clear_ring_build_skb_enabled(rx_ring); clear_ring_uses_large_buffer(rx_ring); @@ -4562,10 +4566,9 @@ static void igb_set_rx_buffer_len(struct igb_adapter *adapter, set_ring_build_skb_enabled(rx_ring); #if (PAGE_SIZE < 8192) - if (adapter->max_frame_size <= IGB_MAX_FRAME_BUILD_SKB) - return; - - set_ring_uses_large_buffer(rx_ring); + if (adapter->max_frame_size > IGB_MAX_FRAME_BUILD_SKB || + rd32(E1000_RCTL) & E1000_RCTL_SBP) + set_ring_uses_large_buffer(rx_ring); #endif } From 44ed16ddf6eac72a8c8745804eba804cb29f3ac9 Mon Sep 17 00:00:00 2001 From: Lin Ma Date: Tue, 27 Jun 2023 11:31:38 +0800 Subject: [PATCH 07/13] UPSTREAM: net: xfrm: Fix xfrm_address_filter OOB read [ Upstream commit dfa73c17d55b921e1d4e154976de35317e43a93a ] We found below OOB crash: [ 44.211730] ================================================================== [ 44.212045] BUG: KASAN: slab-out-of-bounds in memcmp+0x8b/0xb0 [ 44.212045] Read of size 8 at addr ffff88800870f320 by task poc.xfrm/97 [ 44.212045] [ 44.212045] CPU: 0 PID: 97 Comm: poc.xfrm Not tainted 6.4.0-rc7-00072-gdad9774deaf1-dirty #4 [ 44.212045] Call Trace: [ 44.212045] [ 44.212045] dump_stack_lvl+0x37/0x50 [ 44.212045] print_report+0xcc/0x620 [ 44.212045] ? __virt_addr_valid+0xf3/0x170 [ 44.212045] ? memcmp+0x8b/0xb0 [ 44.212045] kasan_report+0xb2/0xe0 [ 44.212045] ? memcmp+0x8b/0xb0 [ 44.212045] kasan_check_range+0x39/0x1c0 [ 44.212045] memcmp+0x8b/0xb0 [ 44.212045] xfrm_state_walk+0x21c/0x420 [ 44.212045] ? __pfx_dump_one_state+0x10/0x10 [ 44.212045] xfrm_dump_sa+0x1e2/0x290 [ 44.212045] ? __pfx_xfrm_dump_sa+0x10/0x10 [ 44.212045] ? __kernel_text_address+0xd/0x40 [ 44.212045] ? kasan_unpoison+0x27/0x60 [ 44.212045] ? mutex_lock+0x60/0xe0 [ 44.212045] ? __pfx_mutex_lock+0x10/0x10 [ 44.212045] ? kasan_save_stack+0x22/0x50 [ 44.212045] netlink_dump+0x322/0x6c0 [ 44.212045] ? __pfx_netlink_dump+0x10/0x10 [ 44.212045] ? mutex_unlock+0x7f/0xd0 [ 44.212045] ? __pfx_mutex_unlock+0x10/0x10 [ 44.212045] __netlink_dump_start+0x353/0x430 [ 44.212045] xfrm_user_rcv_msg+0x3a4/0x410 [ 44.212045] ? __pfx__raw_spin_lock_irqsave+0x10/0x10 [ 44.212045] ? __pfx_xfrm_user_rcv_msg+0x10/0x10 [ 44.212045] ? __pfx_xfrm_dump_sa+0x10/0x10 [ 44.212045] ? __pfx_xfrm_dump_sa_done+0x10/0x10 [ 44.212045] ? __stack_depot_save+0x382/0x4e0 [ 44.212045] ? filter_irq_stacks+0x1c/0x70 [ 44.212045] ? kasan_save_stack+0x32/0x50 [ 44.212045] ? kasan_save_stack+0x22/0x50 [ 44.212045] ? kasan_set_track+0x25/0x30 [ 44.212045] ? __kasan_slab_alloc+0x59/0x70 [ 44.212045] ? kmem_cache_alloc_node+0xf7/0x260 [ 44.212045] ? kmalloc_reserve+0xab/0x120 [ 44.212045] ? __alloc_skb+0xcf/0x210 [ 44.212045] ? netlink_sendmsg+0x509/0x700 [ 44.212045] ? sock_sendmsg+0xde/0xe0 [ 44.212045] ? __sys_sendto+0x18d/0x230 [ 44.212045] ? __x64_sys_sendto+0x71/0x90 [ 44.212045] ? do_syscall_64+0x3f/0x90 [ 44.212045] ? entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] ? netlink_sendmsg+0x509/0x700 [ 44.212045] ? sock_sendmsg+0xde/0xe0 [ 44.212045] ? __sys_sendto+0x18d/0x230 [ 44.212045] ? __x64_sys_sendto+0x71/0x90 [ 44.212045] ? do_syscall_64+0x3f/0x90 [ 44.212045] ? entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] ? kasan_save_stack+0x22/0x50 [ 44.212045] ? kasan_set_track+0x25/0x30 [ 44.212045] ? kasan_save_free_info+0x2e/0x50 [ 44.212045] ? __kasan_slab_free+0x10a/0x190 [ 44.212045] ? kmem_cache_free+0x9c/0x340 [ 44.212045] ? netlink_recvmsg+0x23c/0x660 [ 44.212045] ? sock_recvmsg+0xeb/0xf0 [ 44.212045] ? __sys_recvfrom+0x13c/0x1f0 [ 44.212045] ? __x64_sys_recvfrom+0x71/0x90 [ 44.212045] ? do_syscall_64+0x3f/0x90 [ 44.212045] ? entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] ? copyout+0x3e/0x50 [ 44.212045] netlink_rcv_skb+0xd6/0x210 [ 44.212045] ? __pfx_xfrm_user_rcv_msg+0x10/0x10 [ 44.212045] ? __pfx_netlink_rcv_skb+0x10/0x10 [ 44.212045] ? __pfx_sock_has_perm+0x10/0x10 [ 44.212045] ? mutex_lock+0x8d/0xe0 [ 44.212045] ? __pfx_mutex_lock+0x10/0x10 [ 44.212045] xfrm_netlink_rcv+0x44/0x50 [ 44.212045] netlink_unicast+0x36f/0x4c0 [ 44.212045] ? __pfx_netlink_unicast+0x10/0x10 [ 44.212045] ? netlink_recvmsg+0x500/0x660 [ 44.212045] netlink_sendmsg+0x3b7/0x700 [ 44.212045] ? __pfx_netlink_sendmsg+0x10/0x10 [ 44.212045] ? __pfx_netlink_sendmsg+0x10/0x10 [ 44.212045] sock_sendmsg+0xde/0xe0 [ 44.212045] __sys_sendto+0x18d/0x230 [ 44.212045] ? __pfx___sys_sendto+0x10/0x10 [ 44.212045] ? rcu_core+0x44a/0xe10 [ 44.212045] ? __rseq_handle_notify_resume+0x45b/0x740 [ 44.212045] ? _raw_spin_lock_irq+0x81/0xe0 [ 44.212045] ? __pfx___rseq_handle_notify_resume+0x10/0x10 [ 44.212045] ? __pfx_restore_fpregs_from_fpstate+0x10/0x10 [ 44.212045] ? __pfx_blkcg_maybe_throttle_current+0x10/0x10 [ 44.212045] ? __pfx_task_work_run+0x10/0x10 [ 44.212045] __x64_sys_sendto+0x71/0x90 [ 44.212045] do_syscall_64+0x3f/0x90 [ 44.212045] entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] RIP: 0033:0x44b7da [ 44.212045] RSP: 002b:00007ffdc8838548 EFLAGS: 00000246 ORIG_RAX: 000000000000002c [ 44.212045] RAX: ffffffffffffffda RBX: 00007ffdc8839978 RCX: 000000000044b7da [ 44.212045] RDX: 0000000000000038 RSI: 00007ffdc8838770 RDI: 0000000000000003 [ 44.212045] RBP: 00007ffdc88385b0 R08: 00007ffdc883858c R09: 000000000000000c [ 44.212045] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 [ 44.212045] R13: 00007ffdc8839968 R14: 00000000004c37d0 R15: 0000000000000001 [ 44.212045] [ 44.212045] [ 44.212045] Allocated by task 97: [ 44.212045] kasan_save_stack+0x22/0x50 [ 44.212045] kasan_set_track+0x25/0x30 [ 44.212045] __kasan_kmalloc+0x7f/0x90 [ 44.212045] __kmalloc_node_track_caller+0x5b/0x140 [ 44.212045] kmemdup+0x21/0x50 [ 44.212045] xfrm_dump_sa+0x17d/0x290 [ 44.212045] netlink_dump+0x322/0x6c0 [ 44.212045] __netlink_dump_start+0x353/0x430 [ 44.212045] xfrm_user_rcv_msg+0x3a4/0x410 [ 44.212045] netlink_rcv_skb+0xd6/0x210 [ 44.212045] xfrm_netlink_rcv+0x44/0x50 [ 44.212045] netlink_unicast+0x36f/0x4c0 [ 44.212045] netlink_sendmsg+0x3b7/0x700 [ 44.212045] sock_sendmsg+0xde/0xe0 [ 44.212045] __sys_sendto+0x18d/0x230 [ 44.212045] __x64_sys_sendto+0x71/0x90 [ 44.212045] do_syscall_64+0x3f/0x90 [ 44.212045] entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] [ 44.212045] The buggy address belongs to the object at ffff88800870f300 [ 44.212045] which belongs to the cache kmalloc-64 of size 64 [ 44.212045] The buggy address is located 32 bytes inside of [ 44.212045] allocated 36-byte region [ffff88800870f300, ffff88800870f324) [ 44.212045] [ 44.212045] The buggy address belongs to the physical page: [ 44.212045] page:00000000e4de16ee refcount:1 mapcount:0 mapping:000000000 ... [ 44.212045] flags: 0x100000000000200(slab|node=0|zone=1) [ 44.212045] page_type: 0xffffffff() [ 44.212045] raw: 0100000000000200 ffff888004c41640 dead000000000122 0000000000000000 [ 44.212045] raw: 0000000000000000 0000000080200020 00000001ffffffff 0000000000000000 [ 44.212045] page dumped because: kasan: bad access detected [ 44.212045] [ 44.212045] Memory state around the buggy address: [ 44.212045] ffff88800870f200: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc [ 44.212045] ffff88800870f280: 00 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] >ffff88800870f300: 00 00 00 00 04 fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] ^ [ 44.212045] ffff88800870f380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] ffff88800870f400: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] ================================================================== By investigating the code, we find the root cause of this OOB is the lack of checks in xfrm_dump_sa(). The buggy code allows a malicious user to pass arbitrary value of filter->splen/dplen. Hence, with crafted xfrm states, the attacker can achieve 8 bytes heap OOB read, which causes info leak. if (attrs[XFRMA_ADDRESS_FILTER]) { filter = kmemdup(nla_data(attrs[XFRMA_ADDRESS_FILTER]), sizeof(*filter), GFP_KERNEL); if (filter == NULL) return -ENOMEM; // NO MORE CHECKS HERE !!! } This patch fixes the OOB by adding necessary boundary checks, just like the code in pfkey_dump() function. Bug: 304913834 Fixes: d3623099d350 ("ipsec: add support of limited SA dump") Signed-off-by: Lin Ma Signed-off-by: Steffen Klassert Signed-off-by: Sasha Levin (cherry picked from commit 1960f468078b3471d1ee9aafa0cf06c8c34a505f) Signed-off-by: Lee Jones Change-Id: Iee18d8ee7e2b99ce5cf4e1ff7137f5b31531b890 --- net/xfrm/xfrm_user.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index b7f95fde8ac0..743d0465bdf7 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1049,6 +1049,15 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) sizeof(*filter), GFP_KERNEL); if (filter == NULL) return -ENOMEM; + + /* see addr_match(), (prefix length >> 5) << 2 + * will be used to compare xfrm_address_t + */ + if (filter->splen > (sizeof(xfrm_address_t) << 3) || + filter->dplen > (sizeof(xfrm_address_t) << 3)) { + kfree(filter); + return -EINVAL; + } } if (attrs[XFRMA_PROTO]) From fda8861893ecfbcb5775c1ac8ec20e41ce326e42 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 19 Sep 2023 20:04:45 +0200 Subject: [PATCH 08/13] UPSTREAM: netfilter: ipset: Fix race between IPSET_CMD_CREATE and IPSET_CMD_SWAP [ Upstream commit 7433b6d2afd512d04398c73aa984d1e285be125b ] Kyle Zeng reported that there is a race between IPSET_CMD_ADD and IPSET_CMD_SWAP in netfilter/ip_set, which can lead to the invocation of `__ip_set_put` on a wrong `set`, triggering the `BUG_ON(set->ref == 0);` check in it. The race is caused by using the wrong reference counter, i.e. the ref counter instead of ref_netlink. Bug: 303172721 Fixes: 24e227896bbf ("netfilter: ipset: Add schedule point in call_ad().") Reported-by: Kyle Zeng Closes: https://lore.kernel.org/netfilter-devel/ZPZqetxOmH+w%2Fmyc@westworld/#r Tested-by: Kyle Zeng Signed-off-by: Jozsef Kadlecsik Signed-off-by: Florian Westphal Signed-off-by: Sasha Levin (cherry picked from commit ea5a61d58886ae875f1b4a371999f2a8b58cf26d) Signed-off-by: Lee Jones Change-Id: I33a6a6234830c600a4ebd62ed1fee3a48876b98d --- net/netfilter/ipset/ip_set_core.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 1cf143f5df2e..d3be0d0b0bda 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -530,6 +530,14 @@ __ip_set_put(struct ip_set *set) /* set->ref can be swapped out by ip_set_swap, netlink events (like dump) need * a separate reference counter */ +static void +__ip_set_get_netlink(struct ip_set *set) +{ + write_lock_bh(&ip_set_ref_lock); + set->ref_netlink++; + write_unlock_bh(&ip_set_ref_lock); +} + static inline void __ip_set_put_netlink(struct ip_set *set) { @@ -1529,11 +1537,11 @@ call_ad(struct sock *ctnl, struct sk_buff *skb, struct ip_set *set, do { if (retried) { - __ip_set_get(set); + __ip_set_get_netlink(set); nfnl_unlock(NFNL_SUBSYS_IPSET); cond_resched(); nfnl_lock(NFNL_SUBSYS_IPSET); - __ip_set_put(set); + __ip_set_put_netlink(set); } ip_set_lock(set); From 0d7affc8ca0c20ad00409e3de06ab84dadbc5367 Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Thu, 5 Oct 2023 10:12:00 +0900 Subject: [PATCH 09/13] UPSTREAM: ravb: Fix up dma_free_coherent() call in ravb_remove() [ Upstream commit e6864af61493113558c502b5cd0d754c19b93277 ] In ravb_remove(), dma_free_coherent() should be call after unregister_netdev(). Otherwise, this controller is possible to use the freed buffer. Bug: 289003868 Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") Signed-off-by: Yoshihiro Shimoda Reviewed-by: Sergey Shtylyov Link: https://lore.kernel.org/r/20231005011201.14368-2-yoshihiro.shimoda.uh@renesas.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin (cherry picked from commit 3f9295ad7f9478e65debcef496da4e4eb83db5ea) Signed-off-by: Lee Jones Change-Id: I8e7da5816f715307c7d8bcd881a2a5ecb52439bb --- drivers/net/ethernet/renesas/ravb_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 231a1295c470..4698fbde7fb7 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2212,14 +2212,14 @@ static int ravb_remove(struct platform_device *pdev) if (priv->chip_id != RCAR_GEN2) ravb_ptp_stop(ndev); - dma_free_coherent(ndev->dev.parent, priv->desc_bat_size, priv->desc_bat, - priv->desc_bat_dma); /* Set reset mode */ ravb_write(ndev, CCC_OPC_RESET, CCC); unregister_netdev(ndev); netif_napi_del(&priv->napi[RAVB_NC]); netif_napi_del(&priv->napi[RAVB_BE]); ravb_mdio_release(priv); + dma_free_coherent(ndev->dev.parent, priv->desc_bat_size, priv->desc_bat, + priv->desc_bat_dma); pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); free_netdev(ndev); From 756378d5aa1c9098da83f8078a8820ca2302d849 Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Thu, 5 Oct 2023 10:12:01 +0900 Subject: [PATCH 10/13] BACKPORT: ravb: Fix use-after-free issue in ravb_tx_timeout_work() [ Upstream commit 3971442870713de527684398416970cf025b4f89 ] The ravb_stop() should call cancel_work_sync(). Otherwise, ravb_tx_timeout_work() is possible to use the freed priv after ravb_remove() was called like below: CPU0 CPU1 ravb_tx_timeout() ravb_remove() unregister_netdev() free_netdev(ndev) // free priv ravb_tx_timeout_work() // use priv unregister_netdev() will call .ndo_stop() so that ravb_stop() is called. And, after phy_stop() is called, netif_carrier_off() is also called. So that .ndo_tx_timeout() will not be called after phy_stop(). Bug: 289003868 Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") Reported-by: Zheng Wang Closes: https://lore.kernel.org/netdev/20230725030026.1664873-1-zyytlz.wz@163.com/ Signed-off-by: Yoshihiro Shimoda Reviewed-by: Sergey Shtylyov Link: https://lore.kernel.org/r/20231005011201.14368-3-yoshihiro.shimoda.uh@renesas.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin (cherry picked from commit 6f6fa8061f756aedb93af12a8a5d3cf659127965) Signed-off-by: Lee Jones Change-Id: I36fcd769d93817adaa04162cae0b54b1addbb9bf --- drivers/net/ethernet/renesas/ravb_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 4698fbde7fb7..367da05ddb1e 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1703,6 +1703,8 @@ static int ravb_close(struct net_device *ndev) of_phy_deregister_fixed_link(np); } + cancel_work_sync(&priv->work); + if (priv->chip_id != RCAR_GEN2) { free_irq(priv->tx_irqs[RAVB_NC], ndev); free_irq(priv->rx_irqs[RAVB_NC], ndev); From 49de253fb4d3abd4a9480701a7eb806eb876c17f Mon Sep 17 00:00:00 2001 From: Kyle Zeng Date: Tue, 5 Sep 2023 15:04:09 -0700 Subject: [PATCH 11/13] UPSTREAM: netfilter: ipset: add the missing IP_SET_HASH_WITH_NET0 macro for ip_set_hash_netportnet.c commit 050d91c03b28ca479df13dfb02bcd2c60dd6a878 upstream. The missing IP_SET_HASH_WITH_NET0 macro in ip_set_hash_netportnet can lead to the use of wrong `CIDR_POS(c)` for calculating array offsets, which can lead to integer underflow. As a result, it leads to slab out-of-bound access. This patch adds back the IP_SET_HASH_WITH_NET0 macro to ip_set_hash_netportnet to address the issue. Bug: 302199939 Fixes: 886503f34d63 ("netfilter: ipset: actually allow allowable CIDR 0 in hash:net,port,net") Suggested-by: Jozsef Kadlecsik Signed-off-by: Kyle Zeng Acked-by: Jozsef Kadlecsik Signed-off-by: Florian Westphal Signed-off-by: Greg Kroah-Hartman (cherry picked from commit a9e6142e5f8f6ac7d1bca45c1b2b13b084ea9e14) Signed-off-by: Lee Jones Change-Id: I11cc1650e7df9d54041164b6bdb01f3a0de46de4 --- net/netfilter/ipset/ip_set_hash_netportnet.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/ipset/ip_set_hash_netportnet.c b/net/netfilter/ipset/ip_set_hash_netportnet.c index a82b70e8b9a6..aed319815358 100644 --- a/net/netfilter/ipset/ip_set_hash_netportnet.c +++ b/net/netfilter/ipset/ip_set_hash_netportnet.c @@ -35,6 +35,7 @@ MODULE_ALIAS("ip_set_hash:net,port,net"); #define IP_SET_HASH_WITH_PROTO #define IP_SET_HASH_WITH_NETS #define IPSET_NET_COUNT 2 +#define IP_SET_HASH_WITH_NET0 /* IPv4 variant */ From 93e172c43edea5566f26bd3ddfc0009309f0054c Mon Sep 17 00:00:00 2001 From: Mukesh Ojha Date: Thu, 26 Oct 2023 19:57:38 +0530 Subject: [PATCH 12/13] UPSTREAM: firmware_loader: Refactor kill_pending_fw_fallback_reqs() Rename 'only_kill_custom' and refactor logic related to it to be more meaningful. Bug: 309378049 Change-Id: I119d2f8c29b9b624e6c1d8546c1533d76a2cc51d Signed-off-by: Mukesh Ojha Acked-by: Luis Chamberlain Link: https://lore.kernel.org/r/1698330459-31776-1-git-send-email-quic_mojha@quicinc.com Signed-off-by: Greg Kroah-Hartman (cherry picked from commit 87ffa98eeee8d62a56afdad80ea697e7a6e5c354) Signed-off-by: Srinivasarao Pathipati --- drivers/base/firmware_loader/fallback.c | 4 ++-- drivers/base/firmware_loader/fallback.h | 4 ++-- drivers/base/firmware_loader/main.c | 8 ++++---- 3 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/base/firmware_loader/fallback.c b/drivers/base/firmware_loader/fallback.c index 19d18afb3086..8c172f545681 100644 --- a/drivers/base/firmware_loader/fallback.c +++ b/drivers/base/firmware_loader/fallback.c @@ -103,7 +103,7 @@ static void fw_load_abort(struct fw_sysfs *fw_sysfs) static LIST_HEAD(pending_fw_head); -void kill_pending_fw_fallback_reqs(bool only_kill_custom) +void kill_pending_fw_fallback_reqs(bool kill_all) { struct fw_priv *fw_priv; struct fw_priv *next; @@ -111,7 +111,7 @@ void kill_pending_fw_fallback_reqs(bool only_kill_custom) mutex_lock(&fw_lock); list_for_each_entry_safe(fw_priv, next, &pending_fw_head, pending_list) { - if (!fw_priv->need_uevent || !only_kill_custom) + if (kill_all || !fw_priv->need_uevent) __fw_load_abort(fw_priv); } mutex_unlock(&fw_lock); diff --git a/drivers/base/firmware_loader/fallback.h b/drivers/base/firmware_loader/fallback.h index 21063503e4ea..dc4e83136005 100644 --- a/drivers/base/firmware_loader/fallback.h +++ b/drivers/base/firmware_loader/fallback.h @@ -35,7 +35,7 @@ int firmware_fallback_sysfs(struct firmware *fw, const char *name, struct device *device, enum fw_opt opt_flags, int ret); -void kill_pending_fw_fallback_reqs(bool only_kill_custom); +void kill_pending_fw_fallback_reqs(bool kill_all); void fw_fallback_set_cache_timeout(void); void fw_fallback_set_default_timeout(void); @@ -52,7 +52,7 @@ static inline int firmware_fallback_sysfs(struct firmware *fw, const char *name, return ret; } -static inline void kill_pending_fw_fallback_reqs(bool only_kill_custom) { } +static inline void kill_pending_fw_fallback_reqs(bool kill_all) { } static inline void fw_fallback_set_cache_timeout(void) { } static inline void fw_fallback_set_default_timeout(void) { } diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c index f30a824b6ad7..12dfbeeed558 100644 --- a/drivers/base/firmware_loader/main.c +++ b/drivers/base/firmware_loader/main.c @@ -1363,10 +1363,10 @@ static int fw_pm_notify(struct notifier_block *notify_block, case PM_SUSPEND_PREPARE: case PM_RESTORE_PREPARE: /* - * kill pending fallback requests with a custom fallback - * to avoid stalling suspend. + * Here, kill pending fallback requests will only kill + * non-uevent firmware request to avoid stalling suspend. */ - kill_pending_fw_fallback_reqs(true); + kill_pending_fw_fallback_reqs(false); device_cache_fw_images(); break; @@ -1452,7 +1452,7 @@ static int fw_shutdown_notify(struct notifier_block *unused1, * Kill all pending fallback requests to avoid both stalling shutdown, * and avoid a deadlock with the usermode_lock. */ - kill_pending_fw_fallback_reqs(false); + kill_pending_fw_fallback_reqs(true); return NOTIFY_DONE; } From d43ac48de222318c621dccb5b140e46676779736 Mon Sep 17 00:00:00 2001 From: Mukesh Ojha Date: Thu, 26 Oct 2023 19:57:39 +0530 Subject: [PATCH 13/13] BACKPORT: firmware_loader: Abort all upcoming firmware load request once reboot triggered There could be following scenario where there is a ongoing reboot is going from processA which tries to call all the reboot notifier callback and one of them is firmware reboot call which tries to abort all the ongoing firmware userspace request under fw_lock but there could be another processB which tries to do request firmware, which came just after abort done from ProcessA and ask for userspace to load the firmware and this can stop the ongoing reboot ProcessA to stall for next 60s(default timeout) which may not be expected behaviour everyone like to see, instead we should abort any firmware load request which came once firmware knows about the reboot through notification. ProcessA ProcessB kernel_restart_prepare blocking_notifier_call_chain fw_shutdown_notify kill_pending_fw_fallback_reqs __fw_load_abort fw_state_aborted request_firmware __fw_state_set firmware_fallback_sysfs ... fw_load_from_user_helper .. ... . .. usermodehelper_read_trylock fw_load_sysfs_fallback fw_sysfs_wait_timeout usermodehelper_disable __usermodehelper_disable down_write() Bug: 309378049 Change-Id: I61eb91f21a01460f340f890b25c60de7597a87ff Signed-off-by: Mukesh Ojha Acked-by: Luis Chamberlain Link: https://lore.kernel.org/r/1698330459-31776-2-git-send-email-quic_mojha@quicinc.com Signed-off-by: Greg Kroah-Hartman (cherry picked from commit effd7c70eaa0440688b60b9d419243695ede3c45) Signed-off-by: Srinivasarao Pathipati --- drivers/base/firmware_loader/fallback.c | 6 +++++- drivers/base/firmware_loader/firmware.h | 1 + drivers/base/firmware_loader/main.c | 1 + 3 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/base/firmware_loader/fallback.c b/drivers/base/firmware_loader/fallback.c index 8c172f545681..f84ada93452f 100644 --- a/drivers/base/firmware_loader/fallback.c +++ b/drivers/base/firmware_loader/fallback.c @@ -114,6 +114,10 @@ void kill_pending_fw_fallback_reqs(bool kill_all) if (kill_all || !fw_priv->need_uevent) __fw_load_abort(fw_priv); } + + if (kill_all) + fw_load_abort_all = true; + mutex_unlock(&fw_lock); } @@ -510,7 +514,7 @@ static int fw_load_sysfs_fallback(struct fw_sysfs *fw_sysfs, } mutex_lock(&fw_lock); - if (fw_state_is_aborted(fw_priv)) { + if (fw_load_abort_all || fw_state_is_aborted(fw_priv)) { mutex_unlock(&fw_lock); retval = -EINTR; goto out; diff --git a/drivers/base/firmware_loader/firmware.h b/drivers/base/firmware_loader/firmware.h index ffb2a2724cc5..d7c05fc20a08 100644 --- a/drivers/base/firmware_loader/firmware.h +++ b/drivers/base/firmware_loader/firmware.h @@ -78,6 +78,7 @@ struct fw_priv { }; extern struct mutex fw_lock; +extern bool fw_load_abort_all; static inline bool __fw_state_check(struct fw_priv *fw_priv, enum fw_status status) diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c index 12dfbeeed558..82f43b74a456 100644 --- a/drivers/base/firmware_loader/main.c +++ b/drivers/base/firmware_loader/main.c @@ -90,6 +90,7 @@ static inline struct fw_priv *to_fw_priv(struct kref *ref) DEFINE_MUTEX(fw_lock); static struct firmware_cache fw_cache; +bool fw_load_abort_all; /* Builtin firmware support */