Skip to contentSkip to conversation

xfrm: policy: Fix out-of-bound array accesses in __xfrm_policy_unlink

PATCH New Sasha Levin 1 participant 25 messages May 16, 2019

Subsystems (12)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/25 xfrm: policy: Fix out-of-bound array accesses in __xfrm_policy_unlink

From: YueHaibing <[email protected]>

[ Upstream commit b805d78d300bcf2c83d6df7da0c818b0fee41427 ]

UBSAN report this:

UBSAN: Undefined behaviour in net/xfrm/xfrm_policy.c:1289:24
index 6 is out of range for type 'unsigned int [6]'
CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.4.162-514.55.6.9.x86_64+ #13
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014
 0000000000000000 1466cf39b41b23c9 ffff8801f6b07a58 ffffffff81cb35f4
 0000000041b58ab3 ffffffff83230f9c ffffffff81cb34e0 ffff8801f6b07a80
 ffff8801f6b07a20 1466cf39b41b23c9 ffffffff851706e0 ffff8801f6b07ae8
Call Trace:
 <IRQ>  [<ffffffff81cb35f4>] __dump_stack lib/dump_stack.c:15 [inline]
 <IRQ>  [<ffffffff81cb35f4>] dump_stack+0x114/0x1a0 lib/dump_stack.c:51
 [<ffffffff81d94225>] ubsan_epilogue+0x12/0x8f lib/ubsan.c:164
 [<ffffffff81d954db>] __ubsan_handle_out_of_bounds+0x16e/0x1b2 lib/ubsan.c:382
 [<ffffffff82a25acd>] __xfrm_policy_unlink+0x3dd/0x5b0 net/xfrm/xfrm_policy.c:1289
 [<ffffffff82a2e572>] xfrm_policy_delete+0x52/0xb0 net/xfrm/xfrm_policy.c:1309
 [<ffffffff82a3319b>] xfrm_policy_timer+0x30b/0x590 net/xfrm/xfrm_policy.c:243
 [<ffffffff813d3927>] call_timer_fn+0x237/0x990 kernel/time/timer.c:1144
 [<ffffffff813d8e7e>] __run_timers kernel/time/timer.c:1218 [inline]
 [<ffffffff813d8e7e>] run_timer_softirq+0x6ce/0xb80 kernel/time/timer.c:1401
 [<ffffffff8120d6f9>] __do_softirq+0x299/0xe10 kernel/softirq.c:273
 [<ffffffff8120e676>] invoke_softirq kernel/softirq.c:350 [inline]
 [<ffffffff8120e676>] irq_exit+0x216/0x2c0 kernel/softirq.c:391
 [<ffffffff82c5edab>] exiting_irq arch/x86/include/asm/apic.h:652 [inline]
 [<ffffffff82c5edab>] smp_apic_timer_interrupt+0x8b/0xc0 arch/x86/kernel/apic/apic.c:926
 [<ffffffff82c5c985>] apic_timer_interrupt+0xa5/0xb0 arch/x86/entry/entry_64.S:735
 <EOI>  [<ffffffff81188096>] ? native_safe_halt+0x6/0x10 arch/x86/include/asm/irqflags.h:52
 [<ffffffff810834d7>] arch_safe_halt arch/x86/include/asm/paravirt.h:111 [inline]
 [<ffffffff810834d7>] default_idle+0x27/0x430 arch/x86/kernel/process.c:446
 [<ffffffff81085f05>] arch_cpu_idle+0x15/0x20 arch/x86/kernel/process.c:437
 [<ffffffff8132abc3>] default_idle_call+0x53/0x90 kernel/sched/idle.c:92
 [<ffffffff8132b32d>] cpuidle_idle_call kernel/sched/idle.c:156 [inline]
 [<ffffffff8132b32d>] cpu_idle_loop kernel/sched/idle.c:251 [inline]
 [<ffffffff8132b32d>] cpu_startup_entry+0x60d/0x9a0 kernel/sched/idle.c:299
 [<ffffffff8113e119>] start_secondary+0x3c9/0x560 arch/x86/kernel/smpboot.c:245

The issue is triggered as this:

xfrm_add_policy

    -->verify_newpolicy_info  //check the index provided by user with XFRM_POLICY_MAX
			      //In my case, the index is 0x6E6BB6, so it pass the check.
    -->xfrm_policy_construct  //copy the user's policy and set xfrm_policy_timer
    -->xfrm_policy_insert
	--> __xfrm_policy_link //use the orgin dir, in my case is 2
	--> xfrm_gen_index   //generate policy index, there is 0x6E6BB6

then xfrm_policy_timer be fired

xfrm_policy_timer
--> xfrm_policy_id2dir //get dir from (policy index & 7), in my case is 6 --> xfrm_policy_delete

      --> __xfrm_policy_unlink //access policy_count[dir], trigger out of range access

Add xfrm_policy_id2dir check in verify_newpolicy_info, make sure the computed dir is valid, to fix the issue.

Fixes e682adf021be ("xfrm: Try to honor policy index if it's supplied by user")
---
 net/xfrm/xfrm_user.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/xfrm/xfrm_user.c +1 -1
index 7e4904b930041..060afc4ffd958 100644
@@ -1424,7 +1424,7 @@ static int verify_newpolicy_info(struct xfrm_userpolicy_info *p)
 	ret = verify_policy_dir(p->dir);
 	if (ret)
 		return ret;
-	if (p->index && ((p->index & XFRM_POLICY_MAX) != p->dir))
+	if (p->index && (xfrm_policy_id2dir(p->index) != p->dir))
 		return -EINVAL;
 
 	return 0;
Signature · 1 line
2.20.1
author maintainer
#1 lore raw

PATCH 2/25 xfrm6_tunnel: Fix potential panic when unloading xfrm6_tunnel module

From: Su Yanjun <[email protected]>

[ Upstream commit 6ee02a54ef990a71bf542b6f0a4e3321de9d9c66 ]

When unloading xfrm6_tunnel module, xfrm6_tunnel_fini directly frees the xfrm6_tunnel_spi_kmem. Maybe someone has gotten the xfrm6_tunnel_spi, so need to wait it.

Fixes 91cc3bb0b04ff("xfrm6_tunnel: RCU conversion")
---
 net/ipv6/xfrm6_tunnel.c | 4 ++++
 1 file changed, 4 insertions(+)
net/ipv6/xfrm6_tunnel.c +4 -0
index bc65db782bfb1..12cb3aa990af4 100644
@@ -402,6 +402,10 @@ static void __exit xfrm6_tunnel_fini(void)
 	xfrm6_tunnel_deregister(&xfrm6_tunnel_handler, AF_INET6);
 	xfrm_unregister_type(&xfrm6_tunnel_type, AF_INET6);
 	unregister_pernet_subsys(&xfrm6_tunnel_net_ops);
+	/* Someone maybe has gotten the xfrm6_tunnel_spi.
+	 * So need to wait it.
+	 */
+	rcu_barrier();
 	kmem_cache_destroy(xfrm6_tunnel_spi_kmem);
 }
 
Signature · 1 line
2.20.1
author maintainer
#2 lore raw

PATCH 3/25 vti4: ipip tunnel deregistration fixes.

In reply to Sasha Levin

From: Jeremy Sowden <[email protected]>

[ Upstream commit 5483844c3fc18474de29f5d6733003526e0a9f78 ]

If tunnel registration failed during module initialization, the module would fail to deregister the IPPROTO_COMP protocol and would attempt to deregister the tunnel.

The tunnel was not deregistered during module-exit.

Fixes dd9ee3444014e ("vti4: Fix a ipip packet processing bug in 'IPCOMP' virtual tunnel")
---
 net/ipv4/ip_vti.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv4/ip_vti.c +3 -2
index 40a7cd56e0087..808f8d15c5197 100644
@@ -659,9 +659,9 @@ static int __init vti_init(void)
 	return err;
 
 rtnl_link_failed:
-	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
-xfrm_tunnel_failed:
 	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
+xfrm_tunnel_failed:
+	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
 xfrm_proto_comp_failed:
 	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
 xfrm_proto_ah_failed:
@@ -676,6 +676,7 @@ static int __init vti_init(void)
 static void __exit vti_fini(void)
 {
 	rtnl_link_unregister(&vti_link_ops);
+	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
 	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
 	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
 	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
Signature · 1 line
2.20.1
author maintainer
#3 lore raw

PATCH 4/25 xfrm: clean up xfrm protocol checks

In reply to Sasha Levin

From: Cong Wang <[email protected]>

[ Upstream commit dbb2483b2a46fbaf833cfb5deb5ed9cace9c7399 ]

In commit 6a53b7593233 ("xfrm: check id proto in validate_tmpl()") I introduced a check for xfrm protocol, but according to Herbert IPSEC_PROTO_ANY should only be used as a wildcard for lookup, so it should be removed from validate_tmpl().

And, IPSEC_PROTO_ANY is expected to only match 3 IPSec-specific protocols, this is why xfrm_state_flush() could still miss IPPROTO_ROUTING, which leads that those entries are left in net->xfrm.state_all before exit net. Fix this by replacing IPSEC_PROTO_ANY with zero.

This patch also extracts the check from validate_tmpl() to xfrm_id_proto_valid() and uses it in parse_ipsecrequest(). With this, no other protocols should be added into xfrm.

Fixes 6a53b7593233 ("xfrm: check id proto in validate_tmpl()")
Reported-by [email protected]
Cc Steffen Klassert <[email protected]>
Cc Herbert Xu <[email protected]>
---
 include/net/xfrm.h      | 17 +++++++++++++++++
 net/ipv6/xfrm6_tunnel.c |  2 +-
 net/key/af_key.c        |  4 +++-
 net/xfrm/xfrm_state.c   |  2 +-
 net/xfrm/xfrm_user.c    | 14 +-------------
 5 files changed, 23 insertions(+), 16 deletions(-)
include/net/xfrm.h +17 -0
index 5e3daf53b3d1e..3e966c632f3b2 100644
@@ -1430,6 +1430,23 @@ static inline int xfrm_state_kern(const struct xfrm_state *x)
 	return atomic_read(&x->tunnel_users);
 }
 
+static inline bool xfrm_id_proto_valid(u8 proto)
+{
+	switch (proto) {
+	case IPPROTO_AH:
+	case IPPROTO_ESP:
+	case IPPROTO_COMP:
+#if IS_ENABLED(CONFIG_IPV6)
+	case IPPROTO_ROUTING:
+	case IPPROTO_DSTOPTS:
+#endif
+		return true;
+	default:
+		return false;
+	}
+}
+
+/* IPSEC_PROTO_ANY only matches 3 IPsec protocols, 0 could match all. */
 static inline int xfrm_id_proto_match(u8 proto, u8 userproto)
 {
 	return (!userproto || proto == userproto ||
net/ipv6/xfrm6_tunnel.c +1 -1
index 12cb3aa990af4..d9e5f6808811a 100644
@@ -345,7 +345,7 @@ static void __net_exit xfrm6_tunnel_net_exit(struct net *net)
 	unsigned int i;
 
 	xfrm_flush_gc();
-	xfrm_state_flush(net, IPSEC_PROTO_ANY, false, true);
+	xfrm_state_flush(net, 0, false, true);
 
 	for (i = 0; i < XFRM6_TUNNEL_SPI_BYADDR_HSIZE; i++)
 		WARN_ON_ONCE(!hlist_empty(&xfrm6_tn->spi_byaddr[i]));
net/key/af_key.c +3 -1
index 7d4bed9550605..0b79c9aa8eb1f 100644
@@ -1951,8 +1951,10 @@ parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq)
 
 	if (rq->sadb_x_ipsecrequest_mode == 0)
 		return -EINVAL;
+	if (!xfrm_id_proto_valid(rq->sadb_x_ipsecrequest_proto))
+		return -EINVAL;
 
-	t->id.proto = rq->sadb_x_ipsecrequest_proto; /* XXX check proto */
+	t->id.proto = rq->sadb_x_ipsecrequest_proto;
 	if ((mode = pfkey_mode_to_xfrm(rq->sadb_x_ipsecrequest_mode)) < 0)
 		return -EINVAL;
 	t->mode = mode;
net/xfrm/xfrm_state.c +1 -1
index 3f729cd512aff..11e09eb138d60 100644
@@ -2386,7 +2386,7 @@ void xfrm_state_fini(struct net *net)
 
 	flush_work(&net->xfrm.state_hash_work);
 	flush_work(&xfrm_state_gc_work);
-	xfrm_state_flush(net, IPSEC_PROTO_ANY, false, true);
+	xfrm_state_flush(net, 0, false, true);
 
 	WARN_ON(!list_empty(&net->xfrm.state_all));
 
net/xfrm/xfrm_user.c +1 -13
index 060afc4ffd958..2122f89f61555 100644
@@ -1513,20 +1513,8 @@ static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family)
 			return -EINVAL;
 		}
 
-		switch (ut[i].id.proto) {
-		case IPPROTO_AH:
-		case IPPROTO_ESP:
-		case IPPROTO_COMP:
-#if IS_ENABLED(CONFIG_IPV6)
-		case IPPROTO_ROUTING:
-		case IPPROTO_DSTOPTS:
-#endif
-		case IPSEC_PROTO_ANY:
-			break;
-		default:
+		if (!xfrm_id_proto_valid(ut[i].id.proto))
 			return -EINVAL;
-		}
-
 	}
 
 	return 0;
Signature · 1 line
2.20.1
author maintainer
#4 lore raw

PATCH 5/25 esp4: add length check for UDP encapsulation

In reply to Sasha Levin

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 8dfb4eba4100e7cdd161a8baef2d8d61b7a7e62e ]

esp_output_udp_encap can produce a length that doesn't fit in the 16 bits of a UDP header's length field. In that case, we'll send a fragmented packet whose length is larger than IP_MAX_MTU (resulting in "Oversized IP packet" warnings on receive) and with a bogus UDP length.

To prevent this, add a length check to esp_output_udp_encap and return
-EMSGSIZE on failure.

This seems to be older than git history.

---
 net/ipv4/esp4.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)
net/ipv4/esp4.c +15 -5
index 12a43a5369a54..114f9def1ec54 100644
@@ -223,7 +223,7 @@ static void esp_output_fill_trailer(u8 *tail, int tfclen, int plen, __u8 proto)
 	tail[plen - 1] = proto;
 }
 
-static void esp_output_udp_encap(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *esp)
+static int esp_output_udp_encap(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *esp)
 {
 	int encap_type;
 	struct udphdr *uh;
@@ -231,6 +231,7 @@ static void esp_output_udp_encap(struct xfrm_state *x, struct sk_buff *skb, stru
 	__be16 sport, dport;
 	struct xfrm_encap_tmpl *encap = x->encap;
 	struct ip_esp_hdr *esph = esp->esph;
+	unsigned int len;
 
 	spin_lock_bh(&x->lock);
 	sport = encap->encap_sport;
@@ -238,11 +239,14 @@ static void esp_output_udp_encap(struct xfrm_state *x, struct sk_buff *skb, stru
 	encap_type = encap->encap_type;
 	spin_unlock_bh(&x->lock);
 
+	len = skb->len + esp->tailen - skb_transport_offset(skb);
+	if (len + sizeof(struct iphdr) >= IP_MAX_MTU)
+		return -EMSGSIZE;
+
 	uh = (struct udphdr *)esph;
 	uh->source = sport;
 	uh->dest = dport;
-	uh->len = htons(skb->len + esp->tailen
-		  - skb_transport_offset(skb));
+	uh->len = htons(len);
 	uh->check = 0;
 
 	switch (encap_type) {
@@ -259,6 +263,8 @@ static void esp_output_udp_encap(struct xfrm_state *x, struct sk_buff *skb, stru
 
 	*skb_mac_header(skb) = IPPROTO_UDP;
 	esp->esph = esph;
+
+	return 0;
 }
 
 int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *esp)
@@ -272,8 +278,12 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info *
 	int tailen = esp->tailen;
 
 	/* this is non-NULL only with UDP Encapsulation */
-	if (x->encap)
-		esp_output_udp_encap(x, skb, esp);
+	if (x->encap) {
+		int err = esp_output_udp_encap(x, skb, esp);
+
+		if (err < 0)
+			return err;
+	}
 
 	if (!skb_cloned(skb)) {
 		if (tailen <= skb_tailroom(skb)) {
Signature · 1 line
2.20.1
author maintainer
#5 lore raw

PATCH 6/25 xfrm: Honor original L3 slave device in xfrmi policy lookup

In reply to Sasha Levin

From: Martin Willi <[email protected]>

[ Upstream commit 025c65e119bf58b610549ca359c9ecc5dee6a8d2 ]

If an xfrmi is associated to a vrf layer 3 master device, xfrm_policy_check() fails after traffic decapsulation. The input interface is replaced by the layer 3 master device, and hence xfrmi_decode_session() can't match the xfrmi anymore to satisfy policy checking.

Extend ingress xfrmi lookup to honor the original layer 3 slave device, allowing xfrm interfaces to operate within a vrf domain.

Fixes f203b76d7809 ("xfrm: Add virtual xfrm interfaces")
---
 include/net/xfrm.h        |  3 ++-
 net/xfrm/xfrm_interface.c | 17 ++++++++++++++---
 net/xfrm/xfrm_policy.c    |  2 +-
 3 files changed, 17 insertions(+), 5 deletions(-)
include/net/xfrm.h +2 -1
index 3e966c632f3b2..4ddd2b13ac8d6 100644
@@ -295,7 +295,8 @@ struct xfrm_replay {
 };
 
 struct xfrm_if_cb {
-	struct xfrm_if	*(*decode_session)(struct sk_buff *skb);
+	struct xfrm_if	*(*decode_session)(struct sk_buff *skb,
+					   unsigned short family);
 };
 
 void xfrm_if_register_cb(const struct xfrm_if_cb *ifcb);
net/xfrm/xfrm_interface.c +14 -3
index 82723ef44db3e..555ee2aca6c01 100644
@@ -70,17 +70,28 @@ static struct xfrm_if *xfrmi_lookup(struct net *net, struct xfrm_state *x)
 	return NULL;
 }
 
-static struct xfrm_if *xfrmi_decode_session(struct sk_buff *skb)
+static struct xfrm_if *xfrmi_decode_session(struct sk_buff *skb,
+					    unsigned short family)
 {
 	struct xfrmi_net *xfrmn;
-	int ifindex;
 	struct xfrm_if *xi;
+	int ifindex = 0;
 
 	if (!secpath_exists(skb) || !skb->dev)
 		return NULL;
 
+	switch (family) {
+	case AF_INET6:
+		ifindex = inet6_sdif(skb);
+		break;
+	case AF_INET:
+		ifindex = inet_sdif(skb);
+		break;
+	}
+	if (!ifindex)
+		ifindex = skb->dev->ifindex;
+
 	xfrmn = net_generic(xs_net(xfrm_input_state(skb)), xfrmi_net_id);
-	ifindex = skb->dev->ifindex;
 
 	for_each_xfrmi_rcu(xfrmn->xfrmi[0], xi) {
 		if (ifindex == xi->dev->ifindex &&
net/xfrm/xfrm_policy.c +1 -1
index bf5d59270f79d..ce1b262ce9646 100644
@@ -2339,7 +2339,7 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 	ifcb = xfrm_if_get_cb();
 
 	if (ifcb) {
-		xi = ifcb->decode_session(skb);
+		xi = ifcb->decode_session(skb, family);
 		if (xi) {
 			if_id = xi->p.if_id;
 			net = xi->net;
Signature · 1 line
2.20.1
author maintainer
#6 lore raw

PATCH 7/25 xfrm4: Fix uninitialized memory read in _decode_session4

In reply to Sasha Levin

From: Steffen Klassert <[email protected]>

[ Upstream commit 8742dc86d0c7a9628117a989c11f04a9b6b898f3 ]

We currently don't reload pointers pointing into skb header after doing pskb_may_pull() in _decode_session4(). So in case pskb_may_pull() changed the pointers, we read from random memory. Fix this by putting all the needed infos on the stack, so that we don't need to access the header pointers after doing pskb_may_pull().

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 net/ipv4/xfrm4_policy.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)
net/ipv4/xfrm4_policy.c +13 -11
index d73a6d6652f60..2b144b92ae46a 100644
@@ -111,7 +111,8 @@ static void
 _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	u8 *xprth = skb_network_header(skb) + iph->ihl * 4;
+	int ihl = iph->ihl;
+	u8 *xprth = skb_network_header(skb) + ihl * 4;
 	struct flowi4 *fl4 = &fl->u.ip4;
 	int oif = 0;
 
@@ -122,6 +123,11 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 	fl4->flowi4_mark = skb->mark;
 	fl4->flowi4_oif = reverse ? skb->skb_iif : oif;
 
+	fl4->flowi4_proto = iph->protocol;
+	fl4->daddr = reverse ? iph->saddr : iph->daddr;
+	fl4->saddr = reverse ? iph->daddr : iph->saddr;
+	fl4->flowi4_tos = iph->tos;
+
 	if (!ip_is_fragment(iph)) {
 		switch (iph->protocol) {
 		case IPPROTO_UDP:
@@ -133,7 +139,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 			    pskb_may_pull(skb, xprth + 4 - skb->data)) {
 				__be16 *ports;
 
-				xprth = skb_network_header(skb) + iph->ihl * 4;
+				xprth = skb_network_header(skb) + ihl * 4;
 				ports = (__be16 *)xprth;
 
 				fl4->fl4_sport = ports[!!reverse];
@@ -146,7 +152,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 			    pskb_may_pull(skb, xprth + 2 - skb->data)) {
 				u8 *icmp;
 
-				xprth = skb_network_header(skb) + iph->ihl * 4;
+				xprth = skb_network_header(skb) + ihl * 4;
 				icmp = xprth;
 
 				fl4->fl4_icmp_type = icmp[0];
@@ -159,7 +165,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 			    pskb_may_pull(skb, xprth + 4 - skb->data)) {
 				__be32 *ehdr;
 
-				xprth = skb_network_header(skb) + iph->ihl * 4;
+				xprth = skb_network_header(skb) + ihl * 4;
 				ehdr = (__be32 *)xprth;
 
 				fl4->fl4_ipsec_spi = ehdr[0];
@@ -171,7 +177,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 			    pskb_may_pull(skb, xprth + 8 - skb->data)) {
 				__be32 *ah_hdr;
 
-				xprth = skb_network_header(skb) + iph->ihl * 4;
+				xprth = skb_network_header(skb) + ihl * 4;
 				ah_hdr = (__be32 *)xprth;
 
 				fl4->fl4_ipsec_spi = ah_hdr[1];
@@ -183,7 +189,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 			    pskb_may_pull(skb, xprth + 4 - skb->data)) {
 				__be16 *ipcomp_hdr;
 
-				xprth = skb_network_header(skb) + iph->ihl * 4;
+				xprth = skb_network_header(skb) + ihl * 4;
 				ipcomp_hdr = (__be16 *)xprth;
 
 				fl4->fl4_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
@@ -196,7 +202,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 				__be16 *greflags;
 				__be32 *gre_hdr;
 
-				xprth = skb_network_header(skb) + iph->ihl * 4;
+				xprth = skb_network_header(skb) + ihl * 4;
 				greflags = (__be16 *)xprth;
 				gre_hdr = (__be32 *)xprth;
 
@@ -213,10 +219,6 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
 			break;
 		}
 	}
-	fl4->flowi4_proto = iph->protocol;
-	fl4->daddr = reverse ? iph->saddr : iph->daddr;
-	fl4->saddr = reverse ? iph->daddr : iph->saddr;
-	fl4->flowi4_tos = iph->tos;
 }
 
 static void xfrm4_update_pmtu(struct dst_entry *dst, struct sock *sk,
Signature · 1 line
2.20.1
author maintainer
#7 lore raw

PATCH 8/25 clk: sunxi-ng: nkmp: Avoid GENMASK(-1, 0)

In reply to Sasha Levin

From: Jernej Skrabec <[email protected]>

[ Upstream commit 2abc330c514fe56c570bb1a6318b054b06a4f72e ]

Sometimes one of the nkmp factors is unused. This means that one of the factors shift and width values are set to 0. Current nkmp clock code generates a mask for each factor with GENMASK(width + shift - 1, shift). For unused factor this translates to GENMASK(-1, 0). This code is further expanded by C preprocessor to final version:
(((~0UL) - (1UL << (0)) + 1) & (~0UL >> (BITS_PER_LONG - 1 - (-1)))) or a bit simplified:
(~0UL & (~0UL >> BITS_PER_LONG))

It turns out that result of the second part (~0UL >> BITS_PER_LONG) is actually undefined by C standard, which clearly specifies:

"If the value of the right operand is negative or is greater than or equal to the width of the promoted left operand, the behavior is undefined."

Additionally, compiling kernel with aarch64-linux-gnu-gcc 8.3.0 gave different results whether literals or variables with same values as literals were used. GENMASK with literals -1 and 0 gives zero and with variables gives 0xFFFFFFFFFFFFFFF (~0UL). Because nkmp driver uses GENMASK with variables as parameter, expression calculates mask as ~0UL instead of 0. This has further consequences that LSB in register is always set to 1 (1 is neutral value for a factor and shift is 0).

For example, H6 pll-de clock is set to 600 MHz by sun4i-drm driver, but due to this bug ends up being 300 MHz. Additionally, 300 MHz seems to be too low because following warning can be found in dmesg:

[    1.752763] WARNING: CPU: 2 PID: 41 at drivers/clk/sunxi-ng/ccu_common.c:41 ccu_helper_wait_for_lock.part.0+0x6c/0x90
[    1.763378] Modules linked in:
[    1.766441] CPU: 2 PID: 41 Comm: kworker/2:1 Not tainted 5.1.0-rc2-next-20190401 #138
[    1.774269] Hardware name: Pine H64 (DT)
[    1.778200] Workqueue: events deferred_probe_work_func
[    1.783341] pstate: 40000005 (nZcv daif -PAN -UAO)
[    1.788135] pc : ccu_helper_wait_for_lock.part.0+0x6c/0x90
[    1.793623] lr : ccu_helper_wait_for_lock.part.0+0x48/0x90
[    1.799107] sp : ffff000010f93840
[    1.802422] x29: ffff000010f93840 x28: 0000000000000000
[    1.807735] x27: ffff800073ce9d80 x26: ffff000010afd1b8
[    1.813049] x25: ffffffffffffffff x24: 00000000ffffffff
[    1.818362] x23: 0000000000000001 x22: ffff000010abd5c8
[    1.823675] x21: 0000000010000000 x20: 00000000685f367e
[    1.828987] x19: 0000000000001801 x18: 0000000000000001
[    1.834300] x17: 0000000000000001 x16: 0000000000000000
[    1.839613] x15: 0000000000000000 x14: ffff000010789858
[    1.844926] x13: 0000000000000000 x12: 0000000000000001
[    1.850239] x11: 0000000000000000 x10: 0000000000000970
[    1.855551] x9 : ffff000010f936c0 x8 : ffff800074cec0d0
[    1.860864] x7 : 0000800067117000 x6 : 0000000115c30b41
[    1.866177] x5 : 00ffffffffffffff x4 : 002c959300bfe500
[    1.871490] x3 : 0000000000000018 x2 : 0000000029aaaaab
[    1.876802] x1 : 00000000000002e6 x0 : 00000000686072bc
[    1.882114] Call trace:
[    1.884565]  ccu_helper_wait_for_lock.part.0+0x6c/0x90
[    1.889705]  ccu_helper_wait_for_lock+0x10/0x20
[    1.894236]  ccu_nkmp_set_rate+0x244/0x2a8
[    1.898334]  clk_change_rate+0x144/0x290
[    1.902258]  clk_core_set_rate_nolock+0x180/0x1b8
[    1.906963]  clk_set_rate+0x34/0xa0
[    1.910455]  sun8i_mixer_bind+0x484/0x558
[    1.914466]  component_bind_all+0x10c/0x230
[    1.918651]  sun4i_drv_bind+0xc4/0x1a0
[    1.922401]  try_to_bring_up_master+0x164/0x1c0
[    1.926932]  __component_add+0xa0/0x168
[    1.930769]  component_add+0x10/0x18
[    1.934346]  sun8i_dw_hdmi_probe+0x18/0x20
[    1.938443]  platform_drv_probe+0x50/0xa0
[    1.942455]  really_probe+0xcc/0x280
[    1.946032]  driver_probe_device+0x54/0xe8
[    1.950130]  __device_attach_driver+0x80/0xb8
[    1.954488]  bus_for_each_drv+0x78/0xc8
[    1.958326]  __device_attach+0xd4/0x130
[    1.962163]  device_initial_probe+0x10/0x18
[    1.966348]  bus_probe_device+0x90/0x98
[    1.970185]  deferred_probe_work_func+0x6c/0xa0
[    1.974720]  process_one_work+0x1e0/0x320
[    1.978732]  worker_thread+0x228/0x428
[    1.982484]  kthread+0x120/0x128
[    1.985714]  ret_from_fork+0x10/0x18
[    1.989290] ---[ end trace 9babd42e1ca4b84f ]---

This commit solves the issue by first checking value of the factor width. If it is equal to 0 (unused factor), mask is set to 0, otherwise GENMASK() macro is used as before.

Fixes d897ef56faf9 ("clk: sunxi-ng: Mask nkmp factors when setting register")
---
 drivers/clk/sunxi-ng/ccu_nkmp.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/clk/sunxi-ng/ccu_nkmp.c +13 -5
index ebd9436d2c7cd..1ad53d1016a3e 100644
@@ -160,7 +160,7 @@ static int ccu_nkmp_set_rate(struct clk_hw *hw, unsigned long rate,
 			   unsigned long parent_rate)
 {
 	struct ccu_nkmp *nkmp = hw_to_ccu_nkmp(hw);
-	u32 n_mask, k_mask, m_mask, p_mask;
+	u32 n_mask = 0, k_mask = 0, m_mask = 0, p_mask = 0;
 	struct _ccu_nkmp _nkmp;
 	unsigned long flags;
 	u32 reg;
@@ -179,10 +179,18 @@ static int ccu_nkmp_set_rate(struct clk_hw *hw, unsigned long rate,
 
 	ccu_nkmp_find_best(parent_rate, rate, &_nkmp);
 
-	n_mask = GENMASK(nkmp->n.width + nkmp->n.shift - 1, nkmp->n.shift);
-	k_mask = GENMASK(nkmp->k.width + nkmp->k.shift - 1, nkmp->k.shift);
-	m_mask = GENMASK(nkmp->m.width + nkmp->m.shift - 1, nkmp->m.shift);
-	p_mask = GENMASK(nkmp->p.width + nkmp->p.shift - 1, nkmp->p.shift);
+	if (nkmp->n.width)
+		n_mask = GENMASK(nkmp->n.width + nkmp->n.shift - 1,
+				 nkmp->n.shift);
+	if (nkmp->k.width)
+		k_mask = GENMASK(nkmp->k.width + nkmp->k.shift - 1,
+				 nkmp->k.shift);
+	if (nkmp->m.width)
+		m_mask = GENMASK(nkmp->m.width + nkmp->m.shift - 1,
+				 nkmp->m.shift);
+	if (nkmp->p.width)
+		p_mask = GENMASK(nkmp->p.width + nkmp->p.shift - 1,
+				 nkmp->p.shift);
 
 	spin_lock_irqsave(nkmp->common.lock, flags);
 
Signature · 1 line
2.20.1
author maintainer
#8 lore raw

PATCH 9/25 power: supply: cpcap-battery: Fix division by zero

In reply to Sasha Levin

From: Tony Lindgren <[email protected]>

[ Upstream commit dbe7208c6c4aec083571f2ec742870a0d0edbea3 ]

If called fast enough so samples do not increment, we can get division by zero in kernel:

__div0
cpcap_battery_cc_raw_div
cpcap_battery_get_property
power_supply_get_property.part.1
power_supply_get_property
power_supply_show_property
power_supply_uevent

Fixes 874b2adbed12 ("power: supply: cpcap-battery: Add a battery driver")
---
 drivers/power/supply/cpcap-battery.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/power/supply/cpcap-battery.c +3 -0
index 98ba07869c3b0..3bae02380bb22 100644
@@ -221,6 +221,9 @@ static int cpcap_battery_cc_raw_div(struct cpcap_battery_ddata *ddata,
 	int avg_current;
 	u32 cc_lsb;
 
+	if (!divider)
+		return 0;
+
 	sample &= 0xffffff;		/* 24-bits, unsigned */
 	offset &= 0x7ff;		/* 10-bits, signed */
 
Signature · 1 line
2.20.1
author maintainer
#9 lore raw

PATCH 10/25 securityfs: fix use-after-free on symlink traversal

In reply to Sasha Levin

From: Al Viro <[email protected]>

[ Upstream commit 46c874419652bbefdfed17420fd6e88d8a31d9ec ]

symlink body shouldn't be freed without an RCU delay. Switch securityfs to ->destroy_inode() and use of call_rcu(); free both the inode and symlink body in the callback.

---
 security/inode.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
security/inode.c +9 -4
index 8dd9ca8848e43..829f15672e01f 100644
@@ -26,17 +26,22 @@
 static struct vfsmount *mount;
 static int mount_count;
 
-static void securityfs_evict_inode(struct inode *inode)
+static void securityfs_i_callback(struct rcu_head *head)
 {
-	truncate_inode_pages_final(&inode->i_data);
-	clear_inode(inode);
+	struct inode *inode = container_of(head, struct inode, i_rcu);
 	if (S_ISLNK(inode->i_mode))
 		kfree(inode->i_link);
+	free_inode_nonrcu(inode);
+}
+
+static void securityfs_destroy_inode(struct inode *inode)
+{
+	call_rcu(&inode->i_rcu, securityfs_i_callback);
 }
 
 static const struct super_operations securityfs_super_operations = {
 	.statfs		= simple_statfs,
-	.evict_inode	= securityfs_evict_inode,
+	.destroy_inode	= securityfs_destroy_inode,
 };
 
 static int fill_super(struct super_block *sb, void *data, int silent)
Signature · 1 line
2.20.1
author maintainer
#10 lore raw

PATCH 11/25 apparmorfs: fix use-after-free on symlink traversal

In reply to Sasha Levin

From: Al Viro <[email protected]>

[ Upstream commit f51dcd0f621caac5380ce90fbbeafc32ce4517ae ]

symlink body shouldn't be freed without an RCU delay. Switch apparmorfs to ->destroy_inode() and use of call_rcu(); free both the inode and symlink body in the callback.

---
 security/apparmor/apparmorfs.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
security/apparmor/apparmorfs.c +9 -4
index e09fe4d7307cd..40e3a098f6fb5 100644
@@ -123,17 +123,22 @@ static int aafs_show_path(struct seq_file *seq, struct dentry *dentry)
 	return 0;
 }
 
-static void aafs_evict_inode(struct inode *inode)
+static void aafs_i_callback(struct rcu_head *head)
 {
-	truncate_inode_pages_final(&inode->i_data);
-	clear_inode(inode);
+	struct inode *inode = container_of(head, struct inode, i_rcu);
 	if (S_ISLNK(inode->i_mode))
 		kfree(inode->i_link);
+	free_inode_nonrcu(inode);
+}
+
+static void aafs_destroy_inode(struct inode *inode)
+{
+	call_rcu(&inode->i_rcu, aafs_i_callback);
 }
 
 static const struct super_operations aafs_super_ops = {
 	.statfs = simple_statfs,
-	.evict_inode = aafs_evict_inode,
+	.destroy_inode = aafs_destroy_inode,
 	.show_path = aafs_show_path,
 };
 
Signature · 1 line
2.20.1
author maintainer
#11 lore raw

PATCH 12/25 PCI: Fix issue with "pci=disable_acs_redir" parameter being ignored

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit d5bc73f34cc97c4b4b9202cc93182c2515076edf ]

In most cases, kmalloc() will not be available early in boot when pci_setup() is called. Thus, the kstrdup() call that was added to fix the __initdata bug with the disable_acs_redir parameter usually returns NULL, so the parameter is discarded and has no effect.

To fix this, store the string that's in initdata until an initcall function can allocate the memory appropriately. This way we don't need any additional static memory.

Fixes d2fd6e81912a ("PCI: Fix __initdata issue with "pci=disable_acs_redir" parameter")
---
 drivers/pci/pci.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)
drivers/pci/pci.c +17 -2
index 30649addc6252..61f2ef28ea1c7 100644
@@ -6135,8 +6135,7 @@ static int __init pci_setup(char *str)
 			} else if (!strncmp(str, "pcie_scan_all", 13)) {
 				pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);
 			} else if (!strncmp(str, "disable_acs_redir=", 18)) {
-				disable_acs_redir_param =
-					kstrdup(str + 18, GFP_KERNEL);
+				disable_acs_redir_param = str + 18;
 			} else {
 				printk(KERN_ERR "PCI: Unknown option `%s'\n",
 						str);
@@ -6147,3 +6146,19 @@ static int __init pci_setup(char *str)
 	return 0;
 }
 early_param("pci", pci_setup);
+
+/*
+ * 'disable_acs_redir_param' is initialized in pci_setup(), above, to point
+ * to data in the __initdata section which will be freed after the init
+ * sequence is complete. We can't allocate memory in pci_setup() because some
+ * architectures do not have any memory allocation service available during
+ * an early_param() call. So we allocate memory and copy the variable here
+ * before the init section is freed.
+ */
+static int __init pci_realloc_setup_params(void)
+{
+	disable_acs_redir_param = kstrdup(disable_acs_redir_param, GFP_KERNEL);
+
+	return 0;
+}
+pure_initcall(pci_realloc_setup_params);
Signature · 1 line
2.20.1
author maintainer
#12 lore raw

PATCH 13/25 x86: kvm: hyper-v: deal with buggy TLB flush requests from WS2012

In reply to Sasha Levin

From: Vitaly Kuznetsov <[email protected]>

[ Upstream commit da66761c2d93a46270d69001abb5692717495a68 ]

It was reported that with some special Multi Processor Group configuration, e.g:
bcdedit.exe /set groupsize 1
bcdedit.exe /set maxgroup on
bcdedit.exe /set groupaware on
for a 16-vCPU guest WS2012 shows BSOD on boot when PV TLB flush mechanism is in use.

Tracing kvm_hv_flush_tlb immediately reveals the issue:

kvm_hv_flush_tlb: processor_mask 0x0 address_space 0x0 flags 0x2

The only flag set in this request is HV_FLUSH_ALL_VIRTUAL_ADDRESS_SPACES, however, processor_mask is 0x0 and no HV_FLUSH_ALL_PROCESSORS is specified. We don't flush anything and apparently it's not what Windows expects.

TLFS doesn't say anything about such requests and newer Windows versions seem to be unaffected. This all feels like a WS2012 bug, which is, however, easy to workaround in KVM: let's flush everything when we see an empty flush request, over-flushing doesn't hurt.

---
 arch/x86/kvm/hyperv.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
arch/x86/kvm/hyperv.c +10 -1
index 01d209ab5481b..229d996051653 100644
@@ -1291,7 +1291,16 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *current_vcpu, u64 ingpa,
 				       flush.address_space, flush.flags);
 
 		sparse_banks[0] = flush.processor_mask;
-		all_cpus = flush.flags & HV_FLUSH_ALL_PROCESSORS;
+
+		/*
+		 * Work around possible WS2012 bug: it sends hypercalls
+		 * with processor_mask = 0x0 and HV_FLUSH_ALL_PROCESSORS clear,
+		 * while also expecting us to flush something and crashing if
+		 * we don't. Let's treat processor_mask == 0 same as
+		 * HV_FLUSH_ALL_PROCESSORS.
+		 */
+		all_cpus = (flush.flags & HV_FLUSH_ALL_PROCESSORS) ||
+			flush.processor_mask == 0;
 	} else {
 		if (unlikely(kvm_read_guest(kvm, ingpa, &flush_ex,
 					    sizeof(flush_ex))))
Signature · 1 line
2.20.1
author maintainer
#13 lore raw

PATCH 14/25 mac80211: Fix kernel panic due to use of txq after free

In reply to Sasha Levin

From: Bhagavathi Perumal S <[email protected]>

[ Upstream commit f1267cf3c01b12e0f843fb6a7450a7f0b2efab8a ]

The txq of vif is added to active_txqs list for ATF TXQ scheduling in the function ieee80211_queue_skb(), but it was not properly removed before freeing the txq object. It was causing use after free of the txq objects from the active_txqs list, result was kernel panic due to invalid memory access.

Fix kernel invalid memory access by properly removing txq object from active_txqs list before free the object.

---
 net/mac80211/iface.c | 3 +++
 1 file changed, 3 insertions(+)
net/mac80211/iface.c +3 -0
index 3a0171a65db32..152d4365f9616 100644
@@ -1910,6 +1910,9 @@ void ieee80211_if_remove(struct ieee80211_sub_if_data *sdata)
 	list_del_rcu(&sdata->list);
 	mutex_unlock(&sdata->local->iflist_mtx);
 
+	if (sdata->vif.txq)
+		ieee80211_txq_purge(sdata->local, to_txq_info(sdata->vif.txq));
+
 	synchronize_rcu();
 
 	if (sdata->dev) {
Signature · 1 line
2.20.1
author maintainer
#14 lore raw

PATCH 15/25 net: ieee802154: fix missing checks for regmap_update_bits

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit 22e8860cf8f777fbf6a83f2fb7127f682a8e9de4 ]

regmap_update_bits could fail and deserves a check.

The patch adds the checks and if it fails, returns its error code upstream.

---
 drivers/net/ieee802154/mcr20a.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/net/ieee802154/mcr20a.c +6 -0
index 04891429a5542..fe4057fca83d8 100644
@@ -539,6 +539,8 @@ mcr20a_start(struct ieee802154_hw *hw)
 	dev_dbg(printdev(lp), "no slotted operation\n");
 	ret = regmap_update_bits(lp->regmap_dar, DAR_PHY_CTRL1,
 				 DAR_PHY_CTRL1_SLOTTED, 0x0);
+	if (ret < 0)
+		return ret;
 
 	/* enable irq */
 	enable_irq(lp->spi->irq);
@@ -546,11 +548,15 @@ mcr20a_start(struct ieee802154_hw *hw)
 	/* Unmask SEQ interrupt */
 	ret = regmap_update_bits(lp->regmap_dar, DAR_PHY_CTRL2,
 				 DAR_PHY_CTRL2_SEQMSK, 0x0);
+	if (ret < 0)
+		return ret;
 
 	/* Start the RX sequence */
 	dev_dbg(printdev(lp), "start the RX sequence\n");
 	ret = regmap_update_bits(lp->regmap_dar, DAR_PHY_CTRL1,
 				 DAR_PHY_CTRL1_XCVSEQ_MASK, MCR20A_XCVSEQ_RX);
+	if (ret < 0)
+		return ret;
 
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#15 lore raw

PATCH 16/25 KVM: arm/arm64: Ensure vcpu target is unset on reset failure

In reply to Sasha Levin

From: Andrew Jones <[email protected]>

[ Upstream commit 811328fc3222f7b55846de0cd0404339e2e1e6d7 ]

A failed KVM_ARM_VCPU_INIT should not set the vcpu target, as the vcpu target is used by kvm_vcpu_initialized() to determine if other vcpu ioctls may proceed. We need to set the target before calling kvm_reset_vcpu(), but if that call fails, we should then unset it and clear the feature bitmap while we're at it.

[maz: Simplified patch, completed commit message]
---
 virt/kvm/arm/arm.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)
virt/kvm/arm/arm.c +8 -3
index 1415e36fed3db..fef3527af3bd7 100644
@@ -949,7 +949,7 @@ int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_level,
 static int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 			       const struct kvm_vcpu_init *init)
 {
-	unsigned int i;
+	unsigned int i, ret;
 	int phys_target = kvm_target_cpu();
 
 	if (init->target != phys_target)
@@ -984,9 +984,14 @@ static int kvm_vcpu_set_target(struct kvm_vcpu *vcpu,
 	vcpu->arch.target = phys_target;
 
 	/* Now we know what it is, we can reset it. */
-	return kvm_reset_vcpu(vcpu);
-}
+	ret = kvm_reset_vcpu(vcpu);
+	if (ret) {
+		vcpu->arch.target = -1;
+		bitmap_zero(vcpu->arch.features, KVM_VCPU_MAX_FEATURES);
+	}
 
+	return ret;
+}
 
 static int kvm_arch_vcpu_ioctl_vcpu_init(struct kvm_vcpu *vcpu,
 					 struct kvm_vcpu_init *init)
Signature · 6 lines
2.20.1

_______________________________________________
kvmarm mailing list
[email protected]
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
author maintainer
#16 lore raw

PATCH 17/25 power: supply: sysfs: prevent endless uevent loop with CONFIG_POWER_SUPPLY_DEBUG

In reply to Sasha Levin

From: Andrey Smirnov <[email protected]>

[ Upstream commit 349ced9984ff540ce74ca8a0b2e9b03dc434b9dd ]

Fix a similar endless event loop as was done in commit 8dcf32175b4e ("i2c: prevent endless uevent loop with CONFIG_I2C_DEBUG_CORE"):

The culprit is the dev_dbg printk in the i2c uevent handler. If this is activated (for instance by CONFIG_I2C_DEBUG_CORE) it results in an endless loop with systemd-journald.

This happens if user-space scans the system log and reads the uevent file to get information about a newly created device, which seems fair use to me. Unfortunately reading the "uevent" file uses the same function that runs for creating the uevent for a new device, generating the next syslog entry

Both CONFIG_I2C_DEBUG_CORE and CONFIG_POWER_SUPPLY_DEBUG were reported in https://bugs.freedesktop.org/show_bug.cgi?id=76886 but only former seems to have been fixed. Drop debug prints as it was done in I2C subsystem to resolve the issue.

---
 drivers/power/supply/power_supply_sysfs.c | 6 ------
 1 file changed, 6 deletions(-)
drivers/power/supply/power_supply_sysfs.c +0 -6
index 6170ed8b6854b..5a2757a7f4088 100644
@@ -382,15 +382,11 @@ int power_supply_uevent(struct device *dev, struct kobj_uevent_env *env)
 	char *prop_buf;
 	char *attrname;
 
-	dev_dbg(dev, "uevent\n");
-
 	if (!psy || !psy->desc) {
 		dev_dbg(dev, "No power supply yet\n");
 		return ret;
 	}
 
-	dev_dbg(dev, "POWER_SUPPLY_NAME=%s\n", psy->desc->name);
-
 	ret = add_uevent_var(env, "POWER_SUPPLY_NAME=%s", psy->desc->name);
 	if (ret)
 		return ret;
@@ -426,8 +422,6 @@ int power_supply_uevent(struct device *dev, struct kobj_uevent_env *env)
 			goto out;
 		}
 
-		dev_dbg(dev, "prop %s=%s\n", attrname, prop_buf);
-
 		ret = add_uevent_var(env, "POWER_SUPPLY_%s=%s", attrname, prop_buf);
 		kfree(attrname);
 		if (ret)
Signature · 1 line
2.20.1
author maintainer
#17 lore raw

PATCH 18/25 bpf: Fix preempt_enable_no_resched() abuse

In reply to Sasha Levin

From: Peter Zijlstra <[email protected]>

[ Upstream commit 0edd6b64d1939e9e9168ff27947995bb7751db5d ]

Unless the very next line is schedule(), or implies it, one must not use preempt_enable_no_resched(). It can cause a preemption to go missing and thereby cause arbitrary delays, breaking the PREEMPT=y invariant.

---
 include/linux/bpf.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/bpf.h +1 -1
index 523481a3471b5..fd95f2efe5f32 100644
@@ -400,7 +400,7 @@ int bpf_prog_array_copy(struct bpf_prog_array __rcu *old_array,
 		}					\
 _out:							\
 		rcu_read_unlock();			\
-		preempt_enable_no_resched();		\
+		preempt_enable();			\
 		_ret;					\
 	 })
 
Signature · 1 line
2.20.1
author maintainer
#18 lore raw

PATCH 19/25 qmi_wwan: new Wistron, ZTE and D-Link devices

In reply to Sasha Levin

From: Bjørn Mork <[email protected]>

[ Upstream commit 88ef66a28391ea7b624bfb7508a5b015c13b28f3 ]

Adding device entries found in vendor modified versions of this driver. Function maps for some of the devices follow:

WNC D16Q1, D16Q5, D18Q1 LTE CAT3 module (1435:0918)

MI_00 Qualcomm HS-USB Diagnostics
MI_01 Android Debug interface
MI_02 Qualcomm HS-USB Modem
MI_03 Qualcomm Wireless HS-USB Ethernet Adapter
MI_04 Qualcomm Wireless HS-USB Ethernet Adapter
MI_05 Qualcomm Wireless HS-USB Ethernet Adapter
MI_06 USB Mass Storage Device

 T:  Bus=02 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  2 Spd=480 MxCh= 0
 D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
 P:  Vendor=1435 ProdID=0918 Rev= 2.32
 S:  Manufacturer=Android
 S:  Product=Android
 S:  SerialNumber=0123456789ABCDEF
 C:* #Ifs= 7 Cfg#= 1 Atr=80 MxPwr=500mA
 I:* If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
 E:  Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 E:  Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 I:* If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=42 Prot=01 Driver=(none)
 E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 I:* If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
 E:  Ad=84(I) Atr=03(Int.) MxPS=  64 Ivl=32ms
 E:  Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 E:  Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 I:* If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=qmi_wwan
 E:  Ad=86(I) Atr=03(Int.) MxPS=  64 Ivl=32ms
 E:  Ad=85(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 E:  Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 I:* If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=qmi_wwan
 E:  Ad=88(I) Atr=03(Int.) MxPS=  64 Ivl=32ms
 E:  Ad=87(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 I:* If#= 5 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=qmi_wwan
 E:  Ad=8a(I) Atr=03(Int.) MxPS=  64 Ivl=32ms
 E:  Ad=89(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
 E:  Ad=06(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms

WNC D18 LTE CAT3 module (1435:d182)

MI_00 Qualcomm HS-USB Diagnostics
MI_01 Androd Debug interface
MI_02 Qualcomm HS-USB Modem
MI_03 Qualcomm HS-USB NMEA
MI_04 Qualcomm Wireless HS-USB Ethernet Adapter
MI_05 Qualcomm Wireless HS-USB Ethernet Adapter
MI_06 USB Mass Storage Device

ZM8510/ZM8620/ME3960 (19d2:0396)

MI_00 ZTE Mobile Broadband Diagnostics Port
MI_01 ZTE Mobile Broadband AT Port
MI_02 ZTE Mobile Broadband Modem
MI_03 ZTE Mobile Broadband NDIS Port (qmi_wwan)
MI_04 ZTE Mobile Broadband ADB Port

ME3620_X (19d2:1432)

MI_00 ZTE Diagnostics Device
MI_01 ZTE UI AT Interface
MI_02 ZTE Modem Device
MI_03 ZTE Mobile Broadband Network Adapter
MI_04 ZTE Composite ADB Interface

---
 drivers/net/usb/qmi_wwan.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/net/usb/qmi_wwan.c +10 -0
index 9195f3476b1d7..679e404a5224f 100644
@@ -1122,9 +1122,16 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x0846, 0x68d3, 8)},	/* Netgear Aircard 779S */
 	{QMI_FIXED_INTF(0x12d1, 0x140c, 1)},	/* Huawei E173 */
 	{QMI_FIXED_INTF(0x12d1, 0x14ac, 1)},	/* Huawei E1820 */
+	{QMI_FIXED_INTF(0x1435, 0x0918, 3)},	/* Wistron NeWeb D16Q1 */
+	{QMI_FIXED_INTF(0x1435, 0x0918, 4)},	/* Wistron NeWeb D16Q1 */
+	{QMI_FIXED_INTF(0x1435, 0x0918, 5)},	/* Wistron NeWeb D16Q1 */
+	{QMI_FIXED_INTF(0x1435, 0x3185, 4)},	/* Wistron NeWeb M18Q5 */
+	{QMI_FIXED_INTF(0x1435, 0xd111, 4)},	/* M9615A DM11-1 D51QC */
 	{QMI_FIXED_INTF(0x1435, 0xd181, 3)},	/* Wistron NeWeb D18Q1 */
 	{QMI_FIXED_INTF(0x1435, 0xd181, 4)},	/* Wistron NeWeb D18Q1 */
 	{QMI_FIXED_INTF(0x1435, 0xd181, 5)},	/* Wistron NeWeb D18Q1 */
+	{QMI_FIXED_INTF(0x1435, 0xd182, 4)},	/* Wistron NeWeb D18 */
+	{QMI_FIXED_INTF(0x1435, 0xd182, 5)},	/* Wistron NeWeb D18 */
 	{QMI_FIXED_INTF(0x1435, 0xd191, 4)},	/* Wistron NeWeb D19Q1 */
 	{QMI_QUIRK_SET_DTR(0x1508, 0x1001, 4)},	/* Fibocom NL668 series */
 	{QMI_FIXED_INTF(0x16d8, 0x6003, 0)},	/* CMOTech 6003 */
@@ -1180,6 +1187,7 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x19d2, 0x0265, 4)},	/* ONDA MT8205 4G LTE */
 	{QMI_FIXED_INTF(0x19d2, 0x0284, 4)},	/* ZTE MF880 */
 	{QMI_FIXED_INTF(0x19d2, 0x0326, 4)},	/* ZTE MF821D */
+	{QMI_FIXED_INTF(0x19d2, 0x0396, 3)},	/* ZTE ZM8620 */
 	{QMI_FIXED_INTF(0x19d2, 0x0412, 4)},	/* Telewell TW-LTE 4G */
 	{QMI_FIXED_INTF(0x19d2, 0x1008, 4)},	/* ZTE (Vodafone) K3570-Z */
 	{QMI_FIXED_INTF(0x19d2, 0x1010, 4)},	/* ZTE (Vodafone) K3571-Z */
@@ -1200,7 +1208,9 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x19d2, 0x1425, 2)},
 	{QMI_FIXED_INTF(0x19d2, 0x1426, 2)},	/* ZTE MF91 */
 	{QMI_FIXED_INTF(0x19d2, 0x1428, 2)},	/* Telewell TW-LTE 4G v2 */
+	{QMI_FIXED_INTF(0x19d2, 0x1432, 3)},	/* ZTE ME3620 */
 	{QMI_FIXED_INTF(0x19d2, 0x2002, 4)},	/* ZTE (Vodafone) K3765-Z */
+	{QMI_FIXED_INTF(0x2001, 0x7e16, 3)},	/* D-Link DWM-221 */
 	{QMI_FIXED_INTF(0x2001, 0x7e19, 4)},	/* D-Link DWM-221 B1 */
 	{QMI_FIXED_INTF(0x2001, 0x7e35, 4)},	/* D-Link DWM-222 */
 	{QMI_FIXED_INTF(0x2020, 0x2031, 4)},	/* Olicard 600 */
Signature · 1 line
2.20.1
author maintainer
#19 lore raw

PATCH 20/25 iwlwifi: mvm: check for length correctness in iwl_mvm_create_skb()

In reply to Sasha Levin

From: Luca Coelho <[email protected]>

[ Upstream commit de1887c064b9996ac03120d90d0a909a3f678f98 ]

We don't check for the validity of the lengths in the packet received from the firmware. If the MPDU length received in the rx descriptor is too short to contain the header length and the crypt length together, we may end up trying to copy a negative number of bytes (headlen - hdrlen < 0) which will underflow and cause us to try to copy a huge amount of data. This causes oopses such as this one:

BUG: unable to handle kernel paging request at ffff896be2970000
PGD 5e201067 P4D 5e201067 PUD 5e205067 PMD 16110d063 PTE 8000000162970161
Oops: 0003 [#1] PREEMPT SMP NOPTI
CPU: 2 PID: 1824 Comm: irq/134-iwlwifi Not tainted 4.19.33-04308-geea41cf4930f #1
Hardware name: [...]
RIP: 0010:memcpy_erms+0x6/0x10
Code: 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3
 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe
RSP: 0018:ffffa4630196fc60 EFLAGS: 00010287
RAX: ffff896be2924618 RBX: ffff896bc8ecc600 RCX: 00000000fffb4610
RDX: 00000000fffffff8 RSI: ffff896a835e2a38 RDI: ffff896be2970000
RBP: ffffa4630196fd30 R08: ffff896bc8ecc600 R09: ffff896a83597000
R10: ffff896bd6998400 R11: 000000000200407f R12: ffff896a83597050
R13: 00000000fffffff8 R14: 0000000000000010 R15: ffff896a83597038
FS:  0000000000000000(0000) GS:ffff896be8280000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffff896be2970000 CR3: 000000005dc12002 CR4: 00000000003606e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 iwl_mvm_rx_mpdu_mq+0xb51/0x121b [iwlmvm]
 iwl_pcie_rx_handle+0x58c/0xa89 [iwlwifi]
 iwl_pcie_irq_rx_msix_handler+0xd9/0x12a [iwlwifi]
 irq_thread_fn+0x24/0x49
 irq_thread+0xb0/0x122
 kthread+0x138/0x140
 ret_from_fork+0x1f/0x40

Fix that by checking the lengths for correctness and trigger a warning to show that we have received wrong data.

---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 28 ++++++++++++++++---
 1 file changed, 24 insertions(+), 4 deletions(-)
drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +24 -4
index b53148f972a4a..036d1d82d93e7 100644
@@ -143,9 +143,9 @@ static inline int iwl_mvm_check_pn(struct iwl_mvm *mvm, struct sk_buff *skb,
 }
 
 /* iwl_mvm_create_skb Adds the rxb to a new skb */
-static void iwl_mvm_create_skb(struct sk_buff *skb, struct ieee80211_hdr *hdr,
-			       u16 len, u8 crypt_len,
-			       struct iwl_rx_cmd_buffer *rxb)
+static int iwl_mvm_create_skb(struct iwl_mvm *mvm, struct sk_buff *skb,
+			      struct ieee80211_hdr *hdr, u16 len, u8 crypt_len,
+			      struct iwl_rx_cmd_buffer *rxb)
 {
 	struct iwl_rx_packet *pkt = rxb_addr(rxb);
 	struct iwl_rx_mpdu_desc *desc = (void *)pkt->data;
@@ -178,6 +178,20 @@ static void iwl_mvm_create_skb(struct sk_buff *skb, struct ieee80211_hdr *hdr,
 	 * present before copying packet data.
 	 */
 	hdrlen += crypt_len;
+
+	if (WARN_ONCE(headlen < hdrlen,
+		      "invalid packet lengths (hdrlen=%d, len=%d, crypt_len=%d)\n",
+		      hdrlen, len, crypt_len)) {
+		/*
+		 * We warn and trace because we want to be able to see
+		 * it in trace-cmd as well.
+		 */
+		IWL_DEBUG_RX(mvm,
+			     "invalid packet lengths (hdrlen=%d, len=%d, crypt_len=%d)\n",
+			     hdrlen, len, crypt_len);
+		return -EINVAL;
+	}
+
 	skb_put_data(skb, hdr, hdrlen);
 	skb_put_data(skb, (u8 *)hdr + hdrlen + pad_len, headlen - hdrlen);
 
@@ -190,6 +204,8 @@ static void iwl_mvm_create_skb(struct sk_buff *skb, struct ieee80211_hdr *hdr,
 		skb_add_rx_frag(skb, 0, rxb_steal_page(rxb), offset,
 				fraglen, rxb->truesize);
 	}
+
+	return 0;
 }
 
 /* iwl_mvm_pass_packet_to_mac80211 - passes the packet for mac80211 */
@@ -1425,7 +1441,11 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 			rx_status->boottime_ns = ktime_get_boot_ns();
 	}
 
-	iwl_mvm_create_skb(skb, hdr, len, crypt_len, rxb);
+	if (iwl_mvm_create_skb(mvm, skb, hdr, len, crypt_len, rxb)) {
+		kfree_skb(skb);
+		goto out;
+	}
+
 	if (!iwl_mvm_reorder(mvm, napi, queue, sta, skb, desc))
 		iwl_mvm_pass_packet_to_mac80211(mvm, napi, skb, queue, sta);
 out:
Signature · 1 line
2.20.1
author maintainer
#20 lore raw

PATCH 21/25 sched/cpufreq: Fix kobject memleak

In reply to Sasha Levin

From: "Tobin C. Harding" <[email protected]>

[ Upstream commit 9a4f26cc98d81b67ecc23b890c28e2df324e29f3 ]

Currently the error return path from kobject_init_and_add() is not followed by a call to kobject_put() - which means we are leaking the kobject.

Fix it by adding a call to kobject_put() in the error path of kobject_init_and_add().

Cc Greg Kroah-Hartman <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Tobin C. Harding <[email protected]>
Cc Vincent Guittot <[email protected]>
Cc Viresh Kumar <[email protected]>
---
 kernel/sched/cpufreq_schedutil.c | 1 +
 1 file changed, 1 insertion(+)
kernel/sched/cpufreq_schedutil.c +1 -0
index 217f81ecae176..4e3625109b28d 100644
@@ -751,6 +751,7 @@ static int sugov_init(struct cpufreq_policy *policy)
 	return 0;
 
 fail:
+	kobject_put(&tunables->attr_set.kobj);
 	policy->governor_data = NULL;
 	sugov_tunables_free(tunables);
 
Signature · 1 line
2.20.1
author maintainer
#21 lore raw

PATCH 22/25 x86/mm/mem_encrypt: Disable all instrumentation for early SME setup

In reply to Sasha Levin

From: Gary Hook <[email protected]>

[ Upstream commit b51ce3744f115850166f3d6c292b9c8cb849ad4f ]

Enablement of AMD's Secure Memory Encryption feature is determined very early after start_kernel() is entered. Part of this procedure involves scanning the command line for the parameter 'mem_encrypt'.

To determine intended state, the function sme_enable() uses library functions cmdline_find_option() and strncmp(). Their use occurs early enough such that it cannot be assumed that any instrumentation subsystem is initialized.

For example, making calls to a KASAN-instrumented function before KASAN is set up will result in the use of uninitialized memory and a boot failure.

When AMD's SME support is enabled, conditionally disable instrumentation of these dependent functions in lib/string.c and arch/x86/lib/cmdline.c.

[ bp: Get rid of intermediary nostackp var and cleanup whitespace. ]

Fixes aca20d546214 ("x86/mm: Add support to make use of Secure Memory Encryption")
Cc Alexander Shishkin <[email protected]>
Cc Andrew Morton <[email protected]>
Cc Andy Shevchenko <[email protected]>
Cc Boris Brezillon <[email protected]>
Cc Coly Li <[email protected]>
Cc "H. Peter Anvin" <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Kees Cook <[email protected]>
Cc Kent Overstreet <[email protected]>
Cc Masahiro Yamada <[email protected]>
Cc Matthew Wilcox <[email protected]>
Cc Sebastian Andrzej Siewior <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc x86-ml <[email protected]>
---
 arch/x86/lib/Makefile | 12 ++++++++++++
 lib/Makefile          | 11 +++++++++++
 2 files changed, 23 insertions(+)
arch/x86/lib/Makefile +12 -0
index 25a972c61b0ae..3c19d60316a88 100644
@@ -6,6 +6,18 @@
 # Produces uninteresting flaky coverage.
 KCOV_INSTRUMENT_delay.o	:= n
 
+# Early boot use of cmdline; don't instrument it
+ifdef CONFIG_AMD_MEM_ENCRYPT
+KCOV_INSTRUMENT_cmdline.o := n
+KASAN_SANITIZE_cmdline.o  := n
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_cmdline.o = -pg
+endif
+
+CFLAGS_cmdline.o := $(call cc-option, -fno-stack-protector)
+endif
+
 inat_tables_script = $(srctree)/arch/x86/tools/gen-insn-attr-x86.awk
 inat_tables_maps = $(srctree)/arch/x86/lib/x86-opcode-map.txt
 quiet_cmd_inat_tables = GEN     $@
lib/Makefile +11 -0
index 4238764468109..0ab808318202c 100644
@@ -17,6 +17,17 @@ KCOV_INSTRUMENT_list_debug.o := n
 KCOV_INSTRUMENT_debugobjects.o := n
 KCOV_INSTRUMENT_dynamic_debug.o := n
 
+# Early boot use of cmdline, don't instrument it
+ifdef CONFIG_AMD_MEM_ENCRYPT
+KASAN_SANITIZE_string.o := n
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_string.o = -pg
+endif
+
+CFLAGS_string.o := $(call cc-option, -fno-stack-protector)
+endif
+
 lib-y := ctype.o string.o vsprintf.o cmdline.o \
 	 rbtree.o radix-tree.o timerqueue.o\
 	 idr.o int_sqrt.o extable.o \
Signature · 1 line
2.20.1
author maintainer
#22 lore raw

PATCH 23/25 ufs: fix braino in ufs_get_inode_gid() for solaris UFS flavour

In reply to Sasha Levin

From: Al Viro <[email protected]>

[ Upstream commit 4e9036042fedaffcd868d7f7aa948756c48c637d ]

To choose whether to pick the GID from the old (16bit) or new (32bit) field, we should check if the old gid field is set to 0xffff. Mainline checks the old *UID* field instead - cut'n'paste from the corresponding code in ufs_get_inode_uid().

Fixes 252e211e90ce
---
 fs/ufs/util.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/ufs/util.h +1 -1
index 1fd3011ea6236..7fd4802222b8c 100644
@@ -229,7 +229,7 @@ ufs_get_inode_gid(struct super_block *sb, struct ufs_inode *inode)
 	case UFS_UID_44BSD:
 		return fs32_to_cpu(sb, inode->ui_u3.ui_44.ui_gid);
 	case UFS_UID_EFT:
-		if (inode->ui_u1.oldids.ui_suid == 0xFFFF)
+		if (inode->ui_u1.oldids.ui_sgid == 0xFFFF)
 			return fs32_to_cpu(sb, inode->ui_u3.ui_sun.ui_gid);
 		/* Fall through */
 	default:
Signature · 1 line
2.20.1
author maintainer
#23 lore raw

PATCH 24/25 perf bench numa: Add define for RUSAGE_THREAD if not present

In reply to Sasha Levin

From: Arnaldo Carvalho de Melo <[email protected]>

[ Upstream commit bf561d3c13423fc54daa19b5d49dc15fafdb7acc ]

While cross building perf to the ARC architecture on a fedora 30 host, we were failing with:

CC /tmp/build/perf/bench/numa.o
bench/numa.c: In function ?worker_thread?:
bench/numa.c:1261:12: error: ?RUSAGE_THREAD? undeclared (first use in this function); did you mean ?SIGEV_THREAD??

    getrusage(RUSAGE_THREAD, &rusage);
              ^~~~~~~~~~~~~
              SIGEV_THREAD

bench/numa.c:1261:12: note: each undeclared identifier is reported only once for each function it appears in

[perfbuilder at 60d5802468f6 perf]$ /arc_gnu_2019.03-rc1_prebuilt_uclibc_le_archs_linux_install/bin/arc-linux-gcc --version | head -1 arc-linux-gcc (ARCv2 ISA Linux uClibc toolchain 2019.03-rc1) 8.3.1 20190225
[perfbuilder at 60d5802468f6 perf]$

Trying to reproduce a report by Vineet, I noticed that, with just cross-built zlib and numactl libraries, I ended up with the above failure.

So, since RUSAGE_THREAD is available as a define, check for that and numactl libraries, I ended up with the above failure.

So, since RUSAGE_THREAD is available as a define in the system headers, check if it is defined in the 'perf bench numa' sources and define it if not.

Now it builds and I have to figure out if the problem reported by Vineet only takes place if we have libelf or some other library available.

Cc Arnd Bergmann <arnd at arndb.de>
Cc Jiri Olsa <jolsa at kernel.org>
Cc linux-snps-arc at lists.infradead.org
Cc Namhyung Kim <namhyung at kernel.org>
Cc Vineet Gupta <Vineet.Gupta1 at synopsys.com>
Signed-off-by Arnaldo Carvalho de Melo <acme at redhat.com>
Signed-off-by Sasha Levin <sashal at kernel.org>
---
 tools/perf/bench/numa.c | 4 ++++
 1 file changed, 4 insertions(+)
tools/perf/bench/numa.c +4 -0
index 44195514b19e6..fa56fde6e8d80 100644
@@ -38,6 +38,10 @@
 #include <numa.h>
 #include <numaif.h>
 
+#ifndef RUSAGE_THREAD
+# define RUSAGE_THREAD 1
+#endif
+
 /*
  * Regular printout to the terminal, supressed if -q is specified:
  */
Signature · 1 line
2.20.1
author maintainer
#24 lore raw

PATCH 25/25 perf/x86/intel: Fix race in intel_pmu_disable_event()

In reply to Sasha Levin

From: Jiri Olsa <[email protected]>

[ Upstream commit 6f55967ad9d9752813e36de6d5fdbd19741adfc7 ]

New race in x86_pmu_stop() was introduced by replacing the atomic __test_and_clear_bit() of cpuc->active_mask by separate test_bit() and __clear_bit() calls in the following commit:

3966c3feca3f ("x86/perf/amd: Remove need to check "running" bit in NMI handler")

The race causes panic for PEBS events with enabled callchains:

  BUG: unable to handle kernel NULL pointer dereference at 0000000000000000
  ...
  RIP: 0010:perf_prepare_sample+0x8c/0x530
  Call Trace:
   <NMI>
   perf_event_output_forward+0x2a/0x80
   __perf_event_overflow+0x51/0xe0
   handle_pmi_common+0x19e/0x240
   intel_pmu_handle_irq+0xad/0x170
   perf_event_nmi_handler+0x2e/0x50
   nmi_handle+0x69/0x110
   default_do_nmi+0x3e/0x100
   do_nmi+0x11a/0x180
   end_repeat_nmi+0x16/0x1a
  RIP: 0010:native_write_msr+0x6/0x20
  ...
   </NMI>
   intel_pmu_disable_event+0x98/0xf0
   x86_pmu_stop+0x6e/0xb0
   x86_pmu_del+0x46/0x140
   event_sched_out.isra.97+0x7e/0x160
  ...

The event is configured to make samples from PEBS drain code, but when it's disabled, we'll go through NMI path instead, where data->callchain will not get allocated and we'll crash:

          x86_pmu_stop
            test_bit(hwc->idx, cpuc->active_mask)
            intel_pmu_disable_event(event)
            {
              ...
              intel_pmu_pebs_disable(event);
              ...

EVENT OVERFLOW -> <NMI>
intel_pmu_handle_irq
handle_pmi_common
TEST PASSES -> test_bit(bit, cpuc->active_mask))

                           perf_event_overflow
                             perf_prepare_sample
                             {
                               ...
                               if (!(sample_type & __PERF_SAMPLE_CALLCHAIN_EARLY))
                                     data->callchain = perf_callchain(event, regs);

         CRASH ->              size += data->callchain->nr;
                             }
                   </NMI>
              ...
              x86_pmu_disable_event(event)
            }

            __clear_bit(hwc->idx, cpuc->active_mask);

Fixing this by disabling the event itself before setting off the PEBS bit.

Cc Alexander Shishkin <[email protected]>
Cc Arnaldo Carvalho de Melo <[email protected]>
Cc David Arcari <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Lendacky Thomas <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Stephane Eranian <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Vince Weaver <[email protected]>
Fixes 3966c3feca3f ("x86/perf/amd: Remove need to check "running" bit in NMI handler")
---
 arch/x86/events/intel/core.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
arch/x86/events/intel/core.c +7 -3
index a759e59990fbd..09c53bcbd497d 100644
@@ -2074,15 +2074,19 @@ static void intel_pmu_disable_event(struct perf_event *event)
 	cpuc->intel_ctrl_host_mask &= ~(1ull << hwc->idx);
 	cpuc->intel_cp_status &= ~(1ull << hwc->idx);
 
-	if (unlikely(event->attr.precise_ip))
-		intel_pmu_pebs_disable(event);
-
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_disable_fixed(hwc);
 		return;
 	}
 
 	x86_pmu_disable_event(event);
+
+	/*
+	 * Needs to be called after x86_pmu_disable_event,
+	 * so we don't trigger the event without PEBS bit set.
+	 */
+	if (unlikely(event->attr.precise_ip))
+		intel_pmu_pebs_disable(event);
 }
 
 static void intel_pmu_del_event(struct perf_event *event)
Signature · 1 line
2.20.1