3.4.77-stable review

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/ethernet/calxeda/xgmac.c              |    2 +-
 drivers/net/ethernet/ibm/ehea/ehea_main.c         |    2 +-
 drivers/net/ethernet/tehuti/tehuti.c              |    1 -
 drivers/net/ethernet/xilinx/ll_temac_main.c       |    2 +-
 drivers/net/ethernet/xilinx/xilinx_axienet_main.c |    2 +-
 5 files changed, 4 insertions(+), 5 deletions(-)

drivers/net/ethernet/calxeda/xgmac.c +1 -1

@@ -1776,7 +1776,7 @@ static int xgmac_probe(struct platform_d
 	if (device_can_wakeup(priv->device))
 		priv->wolopts = WAKE_MAGIC;	/* Magic Frame as default */
 
-	ndev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA;
+	ndev->hw_features = NETIF_F_SG | NETIF_F_HIGHDMA;
 	if (readl(priv->base + XGMAC_DMA_HW_FEATURE) & DMA_HW_FEAT_TXCOESEL)
 		ndev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 				     NETIF_F_RXCSUM;

drivers/net/ethernet/ibm/ehea/ehea_main.c +1 -1

@@ -3029,7 +3029,7 @@ static struct ehea_port *ehea_setup_sing
 
 	dev->hw_features = NETIF_F_SG | NETIF_F_TSO
 		      | NETIF_F_IP_CSUM | NETIF_F_HW_VLAN_TX | NETIF_F_LRO;
-	dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_TSO
+	dev->features = NETIF_F_SG | NETIF_F_TSO
 		      | NETIF_F_HIGHDMA | NETIF_F_IP_CSUM | NETIF_F_HW_VLAN_TX
 		      | NETIF_F_HW_VLAN_RX | NETIF_F_HW_VLAN_FILTER
 		      | NETIF_F_RXCSUM;

drivers/net/ethernet/tehuti/tehuti.c +0 -1

@@ -1995,7 +1995,6 @@ bdx_probe(struct pci_dev *pdev, const st
 		ndev->features = NETIF_F_IP_CSUM | NETIF_F_SG | NETIF_F_TSO
 		    | NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX |
 		    NETIF_F_HW_VLAN_FILTER | NETIF_F_RXCSUM
-		    /*| NETIF_F_FRAGLIST */
 		    ;
 		ndev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
 			NETIF_F_TSO | NETIF_F_HW_VLAN_TX;

drivers/net/ethernet/xilinx/ll_temac_main.c +1 -1

@@ -1026,7 +1026,7 @@ static int __devinit temac_of_probe(stru
 	dev_set_drvdata(&op->dev, ndev);
 	SET_NETDEV_DEV(ndev, &op->dev);
 	ndev->flags &= ~IFF_MULTICAST;  /* clear multicast */
-	ndev->features = NETIF_F_SG | NETIF_F_FRAGLIST;
+	ndev->features = NETIF_F_SG;
 	ndev->netdev_ops = &temac_netdev_ops;
 	ndev->ethtool_ops = &temac_ethtool_ops;
 #if 0

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1 -1

@@ -1494,7 +1494,7 @@ static int __devinit axienet_of_probe(st
 
 	SET_NETDEV_DEV(ndev, &op->dev);
 	ndev->flags &= ~IFF_MULTICAST;  /* clear multicast */
-	ndev->features = NETIF_F_SG | NETIF_F_FRAGLIST;
+	ndev->features = NETIF_F_SG;
 	ndev->netdev_ops = &axienet_netdev_ops;
 	ndev->ethtool_ops = &axienet_ethtool_ops;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #2 lore raw

PATCH 2/27 rds: prevent BUG_ON triggered on congestion update to loopback

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Venkat Venkatsubra <[email protected]>

[ Upstream commit 18fc25c94eadc52a42c025125af24657a93638c0 ]

After congestion update on a local connection, when rds_ib_xmit returns less bytes than that are there in the message, rds_send_xmit calls back rds_ib_xmit with an offset that causes BUG_ON(off & RDS_FRAG_SIZE) to trigger.

For a 4Kb PAGE_SIZE rds_ib_xmit returns min(8240,4096)=4096 when actually the message contains 8240 bytes. rds_send_xmit thinks there is more to send and calls rds_ib_xmit again with a data offset "off" of 4096-48(rds header) =4048 bytes thus hitting the BUG_ON(off & RDS_FRAG_SIZE) [RDS_FRAG_SIZE=4k].

The commit 6094628bfd94323fc1cea05ec2c6affd98c18f7f
"rds: prevent BUG_ON triggering on congestion map updates" introduced this regression. That change was addressing the triggering of a different BUG_ON in rds_send_xmit() on PowerPC architecture with 64Kbytes PAGE_SIZE:
BUG_ON(ret != 0 &&

    		 conn->c_xmit_sg == rm->data.op_nents);

This was the sequence it was going through:
(rds_ib_xmit)
/* Do not send cong updates to IB loopback */ if (conn->c_loopback
&& rm->m_inc.i_hdr.h_flags & RDS_FLAG_CONG_BITMAP) {

  	rds_cong_map_updated(conn->c_fcong, ~(u64) 0);
    	return sizeof(struct rds_header) + RDS_CONG_MAP_BYTES;
}

rds_ib_xmit returns 8240
rds_send_xmit:
c_xmit_data_off = 0 + 8240 - 48 (rds header accounted only the first time)
= 8192
c_xmit_data_off < 65536 (sg->length), so calls rds_ib_xmit again
rds_ib_xmit returns 8240
rds_send_xmit:
c_xmit_data_off = 8192 + 8240 = 16432, calls rds_ib_xmit again and so on (c_xmit_data_off 24672,32912,41152,49392,57632)
rds_ib_xmit returns 8240
On this iteration this sequence causes the BUG_ON in rds_send_xmit:

    while (ret) {
    	tmp = min_t(int, ret, sg->length - conn->c_xmit_data_off);
    	[tmp = 65536 - 57632 = 7904]
    	conn->c_xmit_data_off += tmp;
    	[c_xmit_data_off = 57632 + 7904 = 65536]
    	ret -= tmp;
    	[ret = 8240 - 7904 = 336]
    	if (conn->c_xmit_data_off == sg->length) {
    		conn->c_xmit_data_off = 0;
    		sg++;
    		conn->c_xmit_sg++;
    		BUG_ON(ret != 0 &&
    			conn->c_xmit_sg == rm->data.op_nents);
    		[c_xmit_sg = 1, rm->data.op_nents = 1]

What the current fix does:
Since the congestion update over loopback is not actually transmitted as a message, all that rds_ib_xmit needs to do is let the caller think the full message has been transmitted and not return partial bytes. It will return 8240 (RDS_CONG_MAP_BYTES+48) when PAGE_SIZE is 4Kb. And 64Kb+48 when page size is 64Kb.

Reported-by Josh Hunt <[email protected]>

Tested-by Honggang Li <[email protected]>

Acked-by Bang Nguyen <[email protected]>

Signed-off-by Venkat Venkatsubra <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/rds/ib_send.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

net/rds/ib_send.c +2 -3

@@ -552,9 +552,8 @@ int rds_ib_xmit(struct rds_connection *c
 	    && rm->m_inc.i_hdr.h_flags & RDS_FLAG_CONG_BITMAP) {
 		rds_cong_map_updated(conn->c_fcong, ~(u64) 0);
 		scat = &rm->data.op_sg[sg];
-		ret = sizeof(struct rds_header) + RDS_CONG_MAP_BYTES;
-		ret = min_t(int, ret, scat->length - conn->c_xmit_data_off);
-		return ret;
+		ret = max_t(int, RDS_CONG_MAP_BYTES, scat->length);
+		return sizeof(struct rds_header) + ret;
 	}
 
 	/* FIXME we may overallocate here */

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #3 lore raw

PATCH 3/27 macvtap: Do not double-count received packets

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vlad Yasevich <[email protected]>

[ Upstream commit 006da7b07bc4d3a7ffabad17cf639eec6849c9dc ]

Currently macvlan will count received packets after calling each
vlans receive handler. Macvtap attempts to count the packet
yet again when the user reads the packet from the tap socket. This code doesn't do this consistently either. Remove the counting from macvtap and let only macvlan count received packets.

Signed-off-by Vlad Yasevich <[email protected]>

Acked-by Michael S. Tsirkin <[email protected]>

Acked-by Jason Wang <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/macvtap.c |    7 -------
 1 file changed, 7 deletions(-)

drivers/net/macvtap.c +0 -7

@@ -797,7 +797,6 @@ static ssize_t macvtap_put_user(struct m
 				const struct sk_buff *skb,
 				const struct iovec *iv, int len)
 {
-	struct macvlan_dev *vlan;
 	int ret;
 	int vnet_hdr_len = 0;
 	int vlan_offset = 0;

@@ -851,12 +850,6 @@ static ssize_t macvtap_put_user(struct m
 	copied += len;
 
 done:
-	rcu_read_lock_bh();
-	vlan = rcu_dereference_bh(q->vlan);
-	if (vlan)
-		macvlan_count_rx(vlan, copied - vnet_hdr_len, ret == 0, 0);
-	rcu_read_unlock_bh();
-
 	return ret ? ret : copied;
 }

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #4 lore raw

PATCH 4/27 macvtap: update file current position

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhi Yong Wu <[email protected]>

[ Upstream commit e6ebc7f16ca1434a334647aa56399c546be4e64b ]

Signed-off-by Zhi Yong Wu <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/macvtap.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/net/macvtap.c +2 -0

@@ -905,6 +905,8 @@ static ssize_t macvtap_aio_read(struct k
 
 	ret = macvtap_do_read(q, iocb, iv, len, file->f_flags & O_NONBLOCK);
 	ret = min_t(ssize_t, ret, len); /* XXX copied from tun.c. Why? */
+	if (ret > 0)
+		iocb->ki_pos = ret;
 out:
 	return ret;
 }

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #5 lore raw

PATCH 5/27 tun: update file current position

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhi Yong Wu <[email protected]>

[ Upstream commit d0b7da8afa079ffe018ab3e92879b7138977fc8f ]

Signed-off-by Zhi Yong Wu <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/tun.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/net/tun.c +2 -0

@@ -903,6 +903,8 @@ static ssize_t tun_chr_aio_read(struct k
 
 	ret = tun_do_read(tun, iocb, iv, len, file->f_flags & O_NONBLOCK);
 	ret = min_t(ssize_t, ret, len);
+	if (ret > 0)
+		iocb->ki_pos = ret;
 out:
 	tun_put(tun);
 	return ret;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #6 lore raw

PATCH 6/27 macvtap: signal truncated packets

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jason Wang <[email protected]>

[ Upstream commit ce232ce01d61b184202bb185103d119820e1260c ]

macvtap_put_user() never return a value grater than iov length, this in fact bypasses the truncated checking in macvtap_recvmsg(). Fix this by always returning the size of packet plus the possible vlan header to let the trunca checking work.

Cc Vlad Yasevich <[email protected]>

Cc Zhi Yong Wu <[email protected]>

Cc Michael S. Tsirkin <[email protected]>

Signed-off-by Jason Wang <[email protected]>

Acked-by Vlad Yasevich <[email protected]>

Acked-by Michael S. Tsirkin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/macvtap.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

drivers/net/macvtap.c +6 -5

@@ -800,7 +800,7 @@ static ssize_t macvtap_put_user(struct m
 	int ret;
 	int vnet_hdr_len = 0;
 	int vlan_offset = 0;
-	int copied;
+	int copied, total;
 
 	if (q->flags & IFF_VNET_HDR) {
 		struct virtio_net_hdr vnet_hdr;

@@ -815,7 +815,8 @@ static ssize_t macvtap_put_user(struct m
 		if (memcpy_toiovecend(iv, (void *)&vnet_hdr, 0, sizeof(vnet_hdr)))
 			return -EFAULT;
 	}
-	copied = vnet_hdr_len;
+	total = copied = vnet_hdr_len;
+	total += skb->len;
 
 	if (!vlan_tx_tag_present(skb))
 		len = min_t(int, skb->len, len);

@@ -830,6 +831,7 @@ static ssize_t macvtap_put_user(struct m
 
 		vlan_offset = offsetof(struct vlan_ethhdr, h_vlan_proto);
 		len = min_t(int, skb->len + VLAN_HLEN, len);
+		total += VLAN_HLEN;
 
 		copy = min_t(int, vlan_offset, len);
 		ret = skb_copy_datagram_const_iovec(skb, 0, iv, copied, copy);

@@ -847,10 +849,9 @@ static ssize_t macvtap_put_user(struct m
 	}
 
 	ret = skb_copy_datagram_const_iovec(skb, vlan_offset, iv, copied, len);
-	copied += len;
 
 done:
-	return ret ? ret : copied;
+	return ret ? ret : total;
 }
 
 static ssize_t macvtap_do_read(struct macvtap_queue *q, struct kiocb *iocb,

@@ -904,7 +905,7 @@ static ssize_t macvtap_aio_read(struct k
 	}
 
 	ret = macvtap_do_read(q, iocb, iv, len, file->f_flags & O_NONBLOCK);
-	ret = min_t(ssize_t, ret, len); /* XXX copied from tun.c. Why? */
+	ret = min_t(ssize_t, ret, len);
 	if (ret > 0)
 		iocb->ki_pos = ret;
 out:

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #7 lore raw

PATCH 7/27 ipv6: dont count addrconf generated routes against gc limit

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hannes Frederic Sowa <[email protected]>

[ Upstream commit a3300ef4bbb1f1e33ff0400e1e6cf7733d988f4f ]

Brett Ciphery reported that new ipv6 addresses failed to get installed because the addrconf generated dsts where counted against the dst gc limit. We don't need to count those routes like we currently don't count administratively added routes.

Because the max_addresses check enforces a limit on unbounded address generation first in case someone plays with router advertisments, we are still safe here.

Reported-by Brett Ciphery <[email protected]>

Signed-off-by Hannes Frederic Sowa <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/ipv6/route.c |    8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

net/ipv6/route.c +2 -6

@@ -2114,15 +2114,11 @@ struct rt6_info *addrconf_dst_alloc(stru
 {
 	struct net *net = dev_net(idev->dev);
 	struct rt6_info *rt = ip6_dst_alloc(&net->ipv6.ip6_dst_ops,
-					    net->loopback_dev, 0);
+					    net->loopback_dev, DST_NOCOUNT);
 	int err;
 
-	if (!rt) {
-		if (net_ratelimit())
-			pr_warning("IPv6:  Maximum number of routes reached,"
-				   " consider increasing route/max_size.\n");
+	if (!rt)
 		return ERR_PTR(-ENOMEM);
-	}
 
 	in6_dev_hold(idev);

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #8 lore raw

PATCH 8/27 net: drop_monitor: fix the value of maxattr

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Changli Gao <[email protected]>

[ Upstream commit d323e92cc3f4edd943610557c9ea1bb4bb5056e8 ]

maxattr in genl_family should be used to save the max attribute type, but not the max command type. Drop monitor doesn't support any attributes, so we should leave it as zero.

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/core/drop_monitor.c |    1 -
 1 file changed, 1 deletion(-)

net/core/drop_monitor.c +0 -1

@@ -61,7 +61,6 @@ static struct genl_family net_drop_monit
 	.hdrsize        = 0,
 	.name           = "NET_DM",
 	.version        = 2,
-	.maxattr        = NET_DM_CMD_MAX,
 };
 
 static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_cpu_data);

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #9 lore raw

PATCH 9/27 net: unix: allow set_peek_off to fail

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sasha Levin <[email protected]>

[ Upstream commit 12663bfc97c8b3fdb292428105dd92d563164050 ]

unix_dgram_recvmsg() will hold the readlock of the socket until recv is complete.

In the same time, we may try to setsockopt(SO_PEEK_OFF) which will hang until unix_dgram_recvmsg() will complete (which can take a while) without allowing us to break out of it, triggering a hung task spew.

Instead, allow set_peek_off to fail, this way userspace will not hang.

Signed-off-by Sasha Levin <[email protected]>

Acked-by Pavel Emelyanov <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 include/linux/net.h |    2 +-
 net/core/sock.c     |    2 +-
 net/unix/af_unix.c  |    8 ++++++--
 3 files changed, 8 insertions(+), 4 deletions(-)

include/linux/net.h +1 -1

@@ -215,7 +215,7 @@ struct proto_ops {
 				      int offset, size_t size, int flags);
 	ssize_t 	(*splice_read)(struct socket *sock,  loff_t *ppos,
 				       struct pipe_inode_info *pipe, size_t len, unsigned int flags);
-	void		(*set_peek_off)(struct sock *sk, int val);
+	int		(*set_peek_off)(struct sock *sk, int val);
 };
 
 #define DECLARE_SOCKADDR(type, dst, src)	\

net/core/sock.c +1 -1

@@ -795,7 +795,7 @@ set_rcvbuf:
 
 	case SO_PEEK_OFF:
 		if (sock->ops->set_peek_off)
-			sock->ops->set_peek_off(sk, val);
+			ret = sock->ops->set_peek_off(sk, val);
 		else
 			ret = -EOPNOTSUPP;
 		break;

net/unix/af_unix.c +6 -2

@@ -524,13 +524,17 @@ static int unix_seqpacket_sendmsg(struct
 static int unix_seqpacket_recvmsg(struct kiocb *, struct socket *,
 				  struct msghdr *, size_t, int);
 
-static void unix_set_peek_off(struct sock *sk, int val)
+static int unix_set_peek_off(struct sock *sk, int val)
 {
 	struct unix_sock *u = unix_sk(sk);
 
-	mutex_lock(&u->readlock);
+	if (mutex_lock_interruptible(&u->readlock))
+		return -EINTR;
+
 	sk->sk_peek_off = val;
 	mutex_unlock(&u->readlock);
+
+	return 0;
 }

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #10 lore raw

PATCH 10/27 tg3: Initialize REG_BASE_ADDR at PCI config offset 120 to 0

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nat Gurumoorthy <[email protected]>

[ Upstream commit 388d3335575f4c056dcf7138a30f1454e2145cd8 ]

The new tg3 driver leaves REG_BASE_ADDR (PCI config offset 120) uninitialized. From power on reset this register may have garbage in it. The Register Base Address register defines the device local address of a register. The data pointed to by this location is read or written using the Register Data register (PCI config offset 128). When REG_BASE_ADDR has garbage any read or write of Register Data Register (PCI 128) will cause the PCI bus to lock up. The TCO watchdog will fire and bring down the system.

Signed-off-by Nat Gurumoorthy <[email protected]>

Acked-by Michael Chan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/ethernet/broadcom/tg3.c |    3 +++
 1 file changed, 3 insertions(+)

drivers/net/ethernet/broadcom/tg3.c +3 -0

@@ -14671,6 +14671,9 @@ static int __devinit tg3_get_invariants(
 	/* Clear this out for sanity. */
 	tw32(TG3PCI_MEM_WIN_BASE_ADDR, 0);
 
+	/* Clear TG3PCI_REG_BASE_ADDR to prevent hangs. */
+	tw32(TG3PCI_REG_BASE_ADDR, 0);
+
 	pci_read_config_dword(tp->pdev, TG3PCI_PCISTATE,
 			      &pci_state_reg);
 	if ((pci_state_reg & PCISTATE_CONV_PCI_MODE) == 0 &&

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #11 lore raw

PATCH 11/27 netvsc: dont flush peers notifying work during setting mtu

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jason Wang <[email protected]>

[ Upstream commit 50dc875f2e6e2e04aed3b3033eb0ac99192d6d02 ]

There's a possible deadlock if we flush the peers notifying work during setting mtu:

[   22.991149] ======================================================
[   22.991173] [ INFO: possible circular locking dependency detected ]
[   22.991198] 3.10.0-54.0.1.el7.x86_64.debug #1 Not tainted
[   22.991219] -------------------------------------------------------
[   22.991243] ip/974 is trying to acquire lock:
[   22.991261]  ((&(&net_device_ctx->dwork)->work)){+.+.+.}, at: [<ffffffff8108af95>] flush_work+0x5/0x2e0
[   22.991307]
but task is already holding lock:
[   22.991330]  (rtnl_mutex){+.+.+.}, at: [<ffffffff81539deb>] rtnetlink_rcv+0x1b/0x40
[   22.991367]

which lock already depends on the new lock.

[   22.991398]
the existing dependency chain (in reverse order) is:
[   22.991426]
-> #1 (rtnl_mutex){+.+.+.}:
[   22.991449]        [<ffffffff810dfdd9>] __lock_acquire+0xb19/0x1260
[   22.991477]        [<ffffffff810e0d12>] lock_acquire+0xa2/0x1f0
[   22.991501]        [<ffffffff81673659>] mutex_lock_nested+0x89/0x4f0
[   22.991529]        [<ffffffff815392b7>] rtnl_lock+0x17/0x20
[   22.991552]        [<ffffffff815230b2>] netdev_notify_peers+0x12/0x30
[   22.991579]        [<ffffffffa0340212>] netvsc_send_garp+0x22/0x30 [hv_netvsc]
[   22.991610]        [<ffffffff8108d251>] process_one_work+0x211/0x6e0
[   22.991637]        [<ffffffff8108d83b>] worker_thread+0x11b/0x3a0
[   22.991663]        [<ffffffff81095e5d>] kthread+0xed/0x100
[   22.991686]        [<ffffffff81681c6c>] ret_from_fork+0x7c/0xb0
[   22.991715]
-> #0 ((&(&net_device_ctx->dwork)->work)){+.+.+.}:
[   22.991715]        [<ffffffff810de817>] check_prevs_add+0x967/0x970
[   22.991715]        [<ffffffff810dfdd9>] __lock_acquire+0xb19/0x1260
[   22.991715]        [<ffffffff810e0d12>] lock_acquire+0xa2/0x1f0
[   22.991715]        [<ffffffff8108afde>] flush_work+0x4e/0x2e0
[   22.991715]        [<ffffffff8108e1b5>] __cancel_work_timer+0x95/0x130
[   22.991715]        [<ffffffff8108e303>] cancel_delayed_work_sync+0x13/0x20
[   22.991715]        [<ffffffffa03404e4>] netvsc_change_mtu+0x84/0x200 [hv_netvsc]
[   22.991715]        [<ffffffff815233d4>] dev_set_mtu+0x34/0x80
[   22.991715]        [<ffffffff8153bc2a>] do_setlink+0x23a/0xa00
[   22.991715]        [<ffffffff8153d054>] rtnl_newlink+0x394/0x5e0
[   22.991715]        [<ffffffff81539eac>] rtnetlink_rcv_msg+0x9c/0x260
[   22.991715]        [<ffffffff8155cdd9>] netlink_rcv_skb+0xa9/0xc0
[   22.991715]        [<ffffffff81539dfa>] rtnetlink_rcv+0x2a/0x40
[   22.991715]        [<ffffffff8155c41d>] netlink_unicast+0xdd/0x190
[   22.991715]        [<ffffffff8155c807>] netlink_sendmsg+0x337/0x750
[   22.991715]        [<ffffffff8150d219>] sock_sendmsg+0x99/0xd0
[   22.991715]        [<ffffffff8150d63e>] ___sys_sendmsg+0x39e/0x3b0
[   22.991715]        [<ffffffff8150eba2>] __sys_sendmsg+0x42/0x80
[   22.991715]        [<ffffffff8150ebf2>] SyS_sendmsg+0x12/0x20
[   22.991715]        [<ffffffff81681d19>] system_call_fastpath+0x16/0x1b

This is because we hold the rtnl_lock() before ndo_change_mtu() and try to flush the work in netvsc_change_mtu(), in the mean time, netdev_notify_peers() may be called from worker and also trying to hold the rtnl_lock. This will lead the flush won't succeed forever. Solve this by not canceling and flushing the work, this is safe because the transmission done by NETDEV_NOTIFY_PEERS was synchronized with the netif_tx_disable() called by netvsc_change_mtu().

Reported-by Yaju Cao <[email protected]>

Tested-by Yaju Cao <[email protected]>

Cc K. Y. Srinivasan <[email protected]>

Cc Haiyang Zhang <[email protected]>

Signed-off-by Jason Wang <[email protected]>

Acked-by Haiyang Zhang <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/hyperv/netvsc_drv.c |    1 -
 1 file changed, 1 deletion(-)

drivers/net/hyperv/netvsc_drv.c +0 -1

@@ -321,7 +321,6 @@ static int netvsc_change_mtu(struct net_
 		return -EINVAL;
 
 	nvdev->start_remove = true;
-	cancel_delayed_work_sync(&ndevctx->dwork);
 	cancel_work_sync(&ndevctx->work);
 	netif_tx_disable(ndev);
 	rndis_filter_device_remove(hdev);

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #12 lore raw

PATCH 12/27 net: unix: allow bind to fail on mutex lock

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sasha Levin <[email protected]>

[ Upstream commit 37ab4fa7844a044dc21fde45e2a0fc2f3c3b6490 ]

This is similar to the set_peek_off patch where calling bind while the socket is stuck in unix_dgram_recvmsg() will block and cause a hung task spew after a while.

This is also the last place that did a straightforward mutex_lock(), so there shouldn't be any more of these patches.

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/unix/af_unix.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

net/unix/af_unix.c +6 -2

@@ -712,7 +712,9 @@ static int unix_autobind(struct socket *
 	int err;
 	unsigned int retries = 0;
 
-	mutex_lock(&u->readlock);
+	err = mutex_lock_interruptible(&u->readlock);
+	if (err)
+		return err;
 
 	err = 0;
 	if (u->addr)

@@ -845,7 +847,9 @@ static int unix_bind(struct socket *sock
 		goto out;
 	addr_len = err;
 
-	mutex_lock(&u->readlock);
+	err = mutex_lock_interruptible(&u->readlock);
+	if (err)
+		goto out;
 
 	err = -EINVAL;
 	if (u->addr)

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #13 lore raw

PATCH 13/27 net: inet_diag: zero out uninitialized idiag_{src,dst} fields

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Borkmann <[email protected]>

[ Upstream commit b1aac815c0891fe4a55a6b0b715910142227700f ]

Jakub reported while working with nlmon netlink sniffer that parts of the inet_diag_sockid are not initialized when r->idiag_family != AF_INET6. That is, fields of r->id.idiag_src[1 ... 3], r->id.idiag_dst[1 ... 3].

In fact, it seems that we can leak 6 * sizeof(u32) byte of kernel [slab] memory through this. At least, in udp_dump_one(), we allocate a skb in ...

  rep = nlmsg_new(sizeof(struct inet_diag_msg) + ..., GFP_KERNEL);

... and then pass that to inet_sk_diag_fill() that puts the whole struct inet_diag_msg into the skb, where we only fill out r->id.idiag_src[0], r->id.idiag_dst[0] and leave the rest untouched:

  r->id.idiag_src[0] = inet->inet_rcv_saddr;
  r->id.idiag_dst[0] = inet->inet_daddr;

struct inet_diag_msg embeds struct inet_diag_sockid that is correctly / fully filled out in IPv6 case, but for IPv4 not.

So just zero them out by using plain memset (for this little amount of bytes it's probably not worth the extra check for idiag_family == AF_INET).

Similarly, fix also other places where we fill that out.

Reported-by Jakub Zawadzki <[email protected]>

Signed-off-by Daniel Borkmann <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/ipv4/inet_diag.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)

net/ipv4/inet_diag.c +16 -0

@@ -110,6 +110,10 @@ int inet_sk_diag_fill(struct sock *sk, s
 
 	r->id.idiag_sport = inet->inet_sport;
 	r->id.idiag_dport = inet->inet_dport;
+
+	memset(&r->id.idiag_src, 0, sizeof(r->id.idiag_src));
+	memset(&r->id.idiag_dst, 0, sizeof(r->id.idiag_dst));
+
 	r->id.idiag_src[0] = inet->inet_rcv_saddr;
 	r->id.idiag_dst[0] = inet->inet_daddr;

@@ -227,12 +231,19 @@ static int inet_twsk_diag_fill(struct in
 
 	r->idiag_family	      = tw->tw_family;
 	r->idiag_retrans      = 0;
+
 	r->id.idiag_if	      = tw->tw_bound_dev_if;
 	sock_diag_save_cookie(tw, r->id.idiag_cookie);
+
 	r->id.idiag_sport     = tw->tw_sport;
 	r->id.idiag_dport     = tw->tw_dport;
+
+	memset(&r->id.idiag_src, 0, sizeof(r->id.idiag_src));
+	memset(&r->id.idiag_dst, 0, sizeof(r->id.idiag_dst));
+
 	r->id.idiag_src[0]    = tw->tw_rcv_saddr;
 	r->id.idiag_dst[0]    = tw->tw_daddr;
+
 	r->idiag_state	      = tw->tw_substate;
 	r->idiag_timer	      = 3;
 	r->idiag_expires      = DIV_ROUND_UP(tmo * 1000, HZ);

@@ -714,8 +725,13 @@ static int inet_diag_fill_req(struct sk_
 
 	r->id.idiag_sport = inet->inet_sport;
 	r->id.idiag_dport = ireq->rmt_port;
+
+	memset(&r->id.idiag_src, 0, sizeof(r->id.idiag_src));
+	memset(&r->id.idiag_dst, 0, sizeof(r->id.idiag_dst));
+
 	r->id.idiag_src[0] = ireq->loc_addr;
 	r->id.idiag_dst[0] = ireq->rmt_addr;
+
 	r->idiag_expires = jiffies_to_msecs(tmo);
 	r->idiag_rqueue = 0;
 	r->idiag_wqueue = 0;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #14 lore raw

PATCH 14/27 drivers/net/hamradio: Integer overflow in hdlcdrv_ioctl()

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wenliang Fan <[email protected]>

[ Upstream commit e9db5c21d3646a6454fcd04938dd215ac3ab620a ]

The local variable 'bi' comes from userspace. If userspace passed a large number to 'bi.data.calibrate', there would be an integer overflow in the following line:

	s->hdlctx.calibrate = bi.data.calibrate * s->par.bitrate / 16;

Signed-off-by Wenliang Fan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/net/hamradio/hdlcdrv.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/net/hamradio/hdlcdrv.c +2 -0

@@ -571,6 +571,8 @@ static int hdlcdrv_ioctl(struct net_devi
 	case HDLCDRVCTL_CALIBRATE:
 		if(!capable(CAP_SYS_RAWIO))
 			return -EPERM;
+		if (bi.data.calibrate > INT_MAX / s->par.bitrate)
+			return -EINVAL;
 		s->hdlctx.calibrate = bi.data.calibrate * s->par.bitrate / 16;
 		return 0;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #15 lore raw

PATCH 16/27 rds: prevent dereference of a NULL device

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sasha Levin <[email protected]>

[ Upstream commit c2349758acf1874e4c2b93fe41d072336f1a31d0 ]

Binding might result in a NULL device, which is dereferenced causing this BUG:

[ 1317.260548] BUG: unable to handle kernel NULL pointer dereference at 000000000000097
4
[ 1317.261847] IP: [<ffffffff84225f52>] rds_ib_laddr_check+0x82/0x110
[ 1317.263315] PGD 418bcb067 PUD 3ceb21067 PMD 0
[ 1317.263502] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
[ 1317.264179] Dumping ftrace buffer:
[ 1317.264774]    (ftrace buffer empty)
[ 1317.265220] Modules linked in:
[ 1317.265824] CPU: 4 PID: 836 Comm: trinity-child46 Tainted: G        W    3.13.0-rc4-
next-20131218-sasha-00013-g2cebb9b-dirty #4159
[ 1317.267415] task: ffff8803ddf33000 ti: ffff8803cd31a000 task.ti: ffff8803cd31a000
[ 1317.268399] RIP: 0010:[<ffffffff84225f52>]  [<ffffffff84225f52>] rds_ib_laddr_check+
0x82/0x110
[ 1317.269670] RSP: 0000:ffff8803cd31bdf8  EFLAGS: 00010246
[ 1317.270230] RAX: 0000000000000000 RBX: ffff88020b0dd388 RCX: 0000000000000000
[ 1317.270230] RDX: ffffffff8439822e RSI: 00000000000c000a RDI: 0000000000000286
[ 1317.270230] RBP: ffff8803cd31be38 R08: 0000000000000000 R09: 0000000000000000
[ 1317.270230] R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000
[ 1317.270230] R13: 0000000054086700 R14: 0000000000a25de0 R15: 0000000000000031
[ 1317.270230] FS:  00007ff40251d700(0000) GS:ffff88022e200000(0000) knlGS:000000000000
0000
[ 1317.270230] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 1317.270230] CR2: 0000000000000974 CR3: 00000003cd478000 CR4: 00000000000006e0
[ 1317.270230] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1317.270230] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000090602
[ 1317.270230] Stack:
[ 1317.270230]  0000000054086700 5408670000a25de0 5408670000000002 0000000000000000
[ 1317.270230]  ffffffff84223542 00000000ea54c767 0000000000000000 ffffffff86d26160
[ 1317.270230]  ffff8803cd31be68 ffffffff84223556 ffff8803cd31beb8 ffff8800c6765280
[ 1317.270230] Call Trace:
[ 1317.270230]  [<ffffffff84223542>] ? rds_trans_get_preferred+0x42/0xa0
[ 1317.270230]  [<ffffffff84223556>] rds_trans_get_preferred+0x56/0xa0
[ 1317.270230]  [<ffffffff8421c9c3>] rds_bind+0x73/0xf0
[ 1317.270230]  [<ffffffff83e4ce62>] SYSC_bind+0x92/0xf0
[ 1317.270230]  [<ffffffff812493f8>] ? context_tracking_user_exit+0xb8/0x1d0
[ 1317.270230]  [<ffffffff8119313d>] ? trace_hardirqs_on+0xd/0x10
[ 1317.270230]  [<ffffffff8107a852>] ? syscall_trace_enter+0x32/0x290
[ 1317.270230]  [<ffffffff83e4cece>] SyS_bind+0xe/0x10
[ 1317.270230]  [<ffffffff843a6ad0>] tracesys+0xdd/0xe2
[ 1317.270230] Code: 00 8b 45 cc 48 8d 75 d0 48 c7 45 d8 00 00 00 00 66 c7 45 d0 02 00
89 45 d4 48 89 df e8 78 49 76 ff 41 89 c4 85 c0 75 0c 48 8b 03 <80> b8 74 09 00 00 01 7
4 06 41 bc 9d ff ff ff f6 05 2a b6 c2 02
[ 1317.270230] RIP  [<ffffffff84225f52>] rds_ib_laddr_check+0x82/0x110
[ 1317.270230]  RSP <ffff8803cd31bdf8>
[ 1317.270230] CR2: 0000000000000974

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/rds/ib.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

net/rds/ib.c +2 -1

@@ -338,7 +338,8 @@ static int rds_ib_laddr_check(__be32 add
 	ret = rdma_bind_addr(cm_id, (struct sockaddr *)&sin);
 	/* due to this, we will claim to support iWARP devices unless we
 	   check node_type. */
-	if (ret || cm_id->device->node_type != RDMA_NODE_IB_CA)
+	if (ret || !cm_id->device ||
+	    cm_id->device->node_type != RDMA_NODE_IB_CA)
 		ret = -EADDRNOTAVAIL;
 
 	rdsdebug("addr %pI4 ret %d node type %d\n",

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #16 lore raw

PATCH 17/27 net: rose: restore old recvmsg behavior

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit f81152e35001e91997ec74a7b4e040e6ab0acccf ]

recvmsg handler in net/rose/af_rose.c performs size-check ->msg_namelen.

After commit f3d3342602f8bcbf37d7c46641cb9bca7618eb1c (net: rework recvmsg handler msg_name and msg_namelen logic), we now always take the else branch due to namelen being initialized to 0.

Digging in netdev-vger-cvs git repo shows that msg_namelen was initialized with a fixed-size since at least 1995, so the else branch was never taken.

Compile tested only.

Signed-off-by Florian Westphal <[email protected]>

Acked-by Hannes Frederic Sowa <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/rose/af_rose.c |   16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

net/rose/af_rose.c +4 ~~-12~~

@@ -1257,6 +1257,7 @@ static int rose_recvmsg(struct kiocb *io
 
 	if (msg->msg_name) {
 		struct sockaddr_rose *srose;
+		struct full_sockaddr_rose *full_srose = msg->msg_name;
 
 		memset(msg->msg_name, 0, sizeof(struct full_sockaddr_rose));
 		srose = msg->msg_name;

@@ -1264,18 +1265,9 @@ static int rose_recvmsg(struct kiocb *io
 		srose->srose_addr   = rose->dest_addr;
 		srose->srose_call   = rose->dest_call;
 		srose->srose_ndigis = rose->dest_ndigis;
-		if (msg->msg_namelen >= sizeof(struct full_sockaddr_rose)) {
-			struct full_sockaddr_rose *full_srose = (struct full_sockaddr_rose *)msg->msg_name;
-			for (n = 0 ; n < rose->dest_ndigis ; n++)
-				full_srose->srose_digis[n] = rose->dest_digis[n];
-			msg->msg_namelen = sizeof(struct full_sockaddr_rose);
-		} else {
-			if (rose->dest_ndigis >= 1) {
-				srose->srose_ndigis = 1;
-				srose->srose_digi = rose->dest_digis[0];
-			}
-			msg->msg_namelen = sizeof(struct sockaddr_rose);
-		}
+		for (n = 0 ; n < rose->dest_ndigis ; n++)
+			full_srose->srose_digis[n] = rose->dest_digis[n];
+		msg->msg_namelen = sizeof(struct full_sockaddr_rose);
 	}
 
 	skb_free_datagram(sk, skb);

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #17 lore raw

PATCH 18/27 vlan: Fix header ops passthru when doing TX VLAN offload.

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: "David S. Miller" <[email protected]>

[ Upstream commit 2205369a314e12fcec4781cc73ac9c08fc2b47de ]

When the vlan code detects that the real device can do TX VLAN offloads in hardware, it tries to arrange for the real device's header_ops to be invoked directly.

But it does so illegally, by simply hooking the real device's header_ops up to the VLAN device.

This doesn't work because we will end up invoking a set of header_ops routines which expect a device type which matches the real device, but will see a VLAN device instead.

Fix this by providing a pass-thru set of header_ops which will arrange to pass the proper real device instead.

To facilitate this add a dev_rebuild_header(). There are implementations which provide a ->cache and ->create but not a ->rebuild (f.e. PLIP). So we need a helper function just like dev_hard_header() to avoid crashes.

Use this helper in the one existing place where the header_ops->rebuild was being invoked, the neighbour code.

With lots of help from Florian Westphal.

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 include/linux/netdevice.h |    9 +++++++++
 net/8021q/vlan_dev.c      |   19 ++++++++++++++++++-
 2 files changed, 27 insertions(+), 1 deletion(-)

include/linux/netdevice.h +9 -0

@@ -1702,6 +1702,15 @@ static inline int dev_parse_header(const
 	return dev->header_ops->parse(skb, haddr);
 }
 
+static inline int dev_rebuild_header(struct sk_buff *skb)
+{
+	const struct net_device *dev = skb->dev;
+
+	if (!dev->header_ops || !dev->header_ops->rebuild)
+		return 0;
+	return dev->header_ops->rebuild(skb);
+}
+
 typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
 extern int		register_gifconf(unsigned int family, gifconf_func_t * gifconf);
 static inline int unregister_gifconf(unsigned int family)

net/8021q/vlan_dev.c +18 -1

@@ -525,6 +525,23 @@ static const struct header_ops vlan_head
 	.parse	 = eth_header_parse,
 };
 
+static int vlan_passthru_hard_header(struct sk_buff *skb, struct net_device *dev,
+				     unsigned short type,
+				     const void *daddr, const void *saddr,
+				     unsigned int len)
+{
+	struct vlan_dev_priv *vlan = vlan_dev_priv(dev);
+	struct net_device *real_dev = vlan->real_dev;
+
+	return dev_hard_header(skb, real_dev, type, daddr, saddr, len);
+}
+
+static const struct header_ops vlan_passthru_header_ops = {
+	.create	 = vlan_passthru_hard_header,
+	.rebuild = dev_rebuild_header,
+	.parse	 = eth_header_parse,
+};
+
 static const struct net_device_ops vlan_netdev_ops;
 
 static int vlan_dev_init(struct net_device *dev)

@@ -564,7 +581,7 @@ static int vlan_dev_init(struct net_devi
 
 	dev->needed_headroom = real_dev->needed_headroom;
 	if (real_dev->features & NETIF_F_HW_VLAN_TX) {
-		dev->header_ops      = real_dev->header_ops;
+		dev->header_ops      = &vlan_passthru_header_ops;
 		dev->hard_header_len = real_dev->hard_header_len;
 	} else {
 		dev->header_ops      = &vlan_header_ops;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #18 lore raw

PATCH 19/27 net: llc: fix use after free in llc_ui_recvmsg

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Borkmann <[email protected]>

[ Upstream commit 4d231b76eef6c4a6bd9c96769e191517765942cb ]

While commit 30a584d944fb fixes datagram interface in LLC, a use after free bug has been introduced for SOCK_STREAM sockets that do not make use of MSG_PEEK.

The flow is as follow ...

  if (!(flags & MSG_PEEK)) {
    ...
    sk_eat_skb(sk, skb, false);
    ...
  }
  ...
  if (used + offset < skb->len)
    continue;

... where sk_eat_skb() calls __kfree_skb(). Therefore, cache original length and work on skb_len to check partial reads.

Fixes 30a584d944fb ("[LLX]: SOCK_DGRAM interface fixes")

Signed-off-by Daniel Borkmann <[email protected]>

Cc Stephen Hemminger <[email protected]>

Cc Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/llc/af_llc.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

net/llc/af_llc.c +3 -2

@@ -716,7 +716,7 @@ static int llc_ui_recvmsg(struct kiocb *
 	unsigned long cpu_flags;
 	size_t copied = 0;
 	u32 peek_seq = 0;
-	u32 *seq;
+	u32 *seq, skb_len;
 	unsigned long used;
 	int target;	/* Read at least this many bytes */
 	long timeo;

@@ -814,6 +814,7 @@ static int llc_ui_recvmsg(struct kiocb *
 		}
 		continue;
 	found_ok_skb:
+		skb_len = skb->len;
 		/* Ok so how much can we use? */
 		used = skb->len - offset;
 		if (len < used)

@@ -846,7 +847,7 @@ static int llc_ui_recvmsg(struct kiocb *
 		}
 
 		/* Partial read */
-		if (used + offset < skb->len)
+		if (used + offset < skb_len)
 			continue;
 	} while (len > 0);

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #19 lore raw

PATCH 20/27 bridge: use spin_lock_bh() in br_multicast_set_hash_max

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Curt Brune <[email protected]>

[ Upstream commit fe0d692bbc645786bce1a98439e548ae619269f5 ]

br_multicast_set_hash_max() is called from process context in net/bridge/br_sysfs_br.c by the sysfs store_hash_max() function.

br_multicast_set_hash_max() calls spin_lock(&br->multicast_lock), which can deadlock the CPU if a softirq that also tries to take the same lock interrupts br_multicast_set_hash_max() while the lock is held . This can happen quite easily when any of the bridge multicast timers expire, which try to take the same lock.

The fix here is to use spin_lock_bh(), preventing other softirqs from executing on this CPU.

Steps to reproduce:

1. Create a bridge with several interfaces (I used 4).
2. Set the "multicast query interval" to a low number, like 2.
3. Enable the bridge as a multicast querier.
4. Repeatedly set the bridge hash_max parameter via sysfs.

  # brctl addbr br0
  # brctl addif br0 eth1 eth2 eth3 eth4
  # brctl setmcqi br0 2
  # brctl setmcquerier br0 1

  # while true ; do echo 4096 > /sys/class/net/br0/bridge/hash_max; done

Signed-off-by Curt Brune <[email protected]>

Signed-off-by Scott Feldman <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/bridge/br_multicast.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

net/bridge/br_multicast.c +2 -2

@@ -1744,7 +1744,7 @@ int br_multicast_set_hash_max(struct net
 	u32 old;
 	struct net_bridge_mdb_htable *mdb;
 
-	spin_lock(&br->multicast_lock);
+	spin_lock_bh(&br->multicast_lock);
 	if (!netif_running(br->dev))
 		goto unlock;

@@ -1776,7 +1776,7 @@ rollback:
 	}
 
 unlock:
-	spin_unlock(&br->multicast_lock);
+	spin_unlock_bh(&br->multicast_lock);
 
 	return err;
 }

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #20 lore raw

PATCH 21/27 ARM: fix "bad mode in ... handler" message for undefined instructions

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Russell King <[email protected]>

commit 29c350bf28da333e41e30497b649fe335712a2ab upstream.

The array was missing the final entry for the undefined instruction exception handler; this commit adds it.

Signed-off-by Russell King <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/arm/kernel/traps.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

arch/arm/kernel/traps.c +7 -1

@@ -37,7 +37,13 @@
 
 #include "signal.h"
 
-static const char *handler[]= { "prefetch abort", "data abort", "address exception", "interrupt" };
+static const char *handler[]= {
+	"prefetch abort",
+	"data abort",
+	"address exception",
+	"interrupt",
+	"undefined instruction",
+};
 
 void *vectors_page;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #21 lore raw

PATCH 22/27 ARM: shmobile: mackerel: Fix coherent DMA mask

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Laurent Pinchart <[email protected]>

commit b6328a6b7ba57fc84c38248f6f0e387e1170f1a8 upstream.

Commit 4dcfa60071b3d23f0181f27d8519f12e37cefbb9 ("ARM: DMA-API: better handing of DMA masks for coherent allocations") added an additional check to the coherent DMA mask that results in an error when the mask is larger than what dma_addr_t can address.

Set the LCDC coherent DMA mask to DMA_BIT_MASK(32) instead of ~0 to fix the problem.

Signed-off-by Laurent Pinchart <[email protected]>

Signed-off-by Simon Horman <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/arm/mach-shmobile/board-mackerel.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

arch/arm/mach-shmobile/board-mackerel.c +2 -2

@@ -422,7 +422,7 @@ static struct platform_device lcdc_devic
 	.resource	= lcdc_resources,
 	.dev	= {
 		.platform_data	= &lcdc_info,
-		.coherent_dma_mask = ~0,
+		.coherent_dma_mask = DMA_BIT_MASK(32),
 	},
 };

@@ -498,7 +498,7 @@ static struct platform_device hdmi_lcdc_
 	.id		= 1,
 	.dev	= {
 		.platform_data	= &hdmi_lcdc_info,
-		.coherent_dma_mask = ~0,
+		.coherent_dma_mask = DMA_BIT_MASK(32),
 	},
 };

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #22 lore raw

PATCH 23/27 x86, fpu, amd: Clear exceptions in AMD FXSAVE workaround

Link http://lkml.kernel.org/r/CA%2B55aFxQnY_PCG_n4=0w-VG=YLXL-yr7oMxyy0WU2gCBAf3ydg@mail.gmail.com open here

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Linus Torvalds <[email protected]>

commit 26bef1318adc1b3a530ecc807ef99346db2aa8b0 upstream.

Before we do an EMMS in the AMD FXSAVE information leak workaround we need to clear any pending exceptions, otherwise we trap with a floating-point exception inside this code.

Reported-by halfdog <[email protected]>

Tested-by Borislav Petkov <[email protected]>

Signed-off-by H. Peter Anvin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/include/asm/fpu-internal.h |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

arch/x86/include/asm/fpu-internal.h +7 -6

@@ -266,12 +266,13 @@ static inline int restore_fpu_checking(s
 	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
 	   is pending.  Clear the x87 state here by setting it to fixed
 	   values. "m" is a random variable that should be in L1 */
-	alternative_input(
-		ASM_NOP8 ASM_NOP2,
-		"emms\n\t"		/* clear stack tags */
-		"fildl %P[addr]",	/* set F?P to defined value */
-		X86_FEATURE_FXSAVE_LEAK,
-		[addr] "m" (tsk->thread.fpu.has_fpu));
+	if (unlikely(static_cpu_has(X86_FEATURE_FXSAVE_LEAK))) {
+		asm volatile(
+			"fnclex\n\t"
+			"emms\n\t"
+			"fildl %P[addr]"	/* set F?P to defined value */
+			: : [addr] "m" (tsk->thread.fpu.has_fpu));
+	}
 
 	return fpu_restore_checking(&tsk->thread.fpu);
 }

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #23 lore raw

PATCH 24/27 sched: Fix race on toggling cfs_bandwidth_used

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ben Segall <[email protected]>

commit 1ee14e6c8cddeeb8a490d7b54cd9016e4bb900b4 upstream.

When we transition cfs_bandwidth_used to false, any currently throttled groups will incorrectly return false from cfs_rq_throttled. While tg_set_cfs_bandwidth will unthrottle them eventually, currently running code (including at least dequeue_task_fair and distribute_cfs_runtime) will cause errors.

Fix this by turning off cfs_bandwidth_used only after unthrottling all cfs_rqs.

Tested: toggle bandwidth back and forth on a loaded cgroup. Caused crashes in minutes without the patch, hasn't crashed with it.

Signed-off-by Ben Segall <[email protected]>

Link http://lkml.kernel.org/r/20131016181611.22647.80365.stgit@sword-of-the-dawn.mtv.corp.google.com open here

Cc [email protected]

Signed-off-by Ingo Molnar <[email protected]>

Cc Chris J Arges <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 kernel/sched/core.c  |    9 ++++++++-
 kernel/sched/fair.c  |   16 +++++++++-------
 kernel/sched/sched.h |    3 ++-
 3 files changed, 19 insertions(+), 9 deletions(-)

kernel/sched/core.c +8 -1

@@ -7906,7 +7906,12 @@ static int tg_set_cfs_bandwidth(struct t
 
 	runtime_enabled = quota != RUNTIME_INF;
 	runtime_was_enabled = cfs_b->quota != RUNTIME_INF;
-	account_cfs_bandwidth_used(runtime_enabled, runtime_was_enabled);
+	/*
+	 * If we need to toggle cfs_bandwidth_used, off->on must occur
+	 * before making related changes, and on->off must occur afterwards
+	 */
+	if (runtime_enabled && !runtime_was_enabled)
+		cfs_bandwidth_usage_inc();
 	raw_spin_lock_irq(&cfs_b->lock);
 	cfs_b->period = ns_to_ktime(period);
 	cfs_b->quota = quota;

@@ -7932,6 +7937,8 @@ static int tg_set_cfs_bandwidth(struct t
 			unthrottle_cfs_rq(cfs_rq);
 		raw_spin_unlock_irq(&rq->lock);
 	}
+	if (runtime_was_enabled && !runtime_enabled)
+		cfs_bandwidth_usage_dec();
 out_unlock:
 	mutex_unlock(&cfs_constraints_mutex);

kernel/sched/fair.c +9 -7

@@ -1393,13 +1393,14 @@ static inline bool cfs_bandwidth_used(vo
 	return static_key_false(&__cfs_bandwidth_used);
 }
 
-void account_cfs_bandwidth_used(int enabled, int was_enabled)
+void cfs_bandwidth_usage_inc(void)
 {
-	/* only need to count groups transitioning between enabled/!enabled */
-	if (enabled && !was_enabled)
-		static_key_slow_inc(&__cfs_bandwidth_used);
-	else if (!enabled && was_enabled)
-		static_key_slow_dec(&__cfs_bandwidth_used);
+	static_key_slow_inc(&__cfs_bandwidth_used);
+}
+
+void cfs_bandwidth_usage_dec(void)
+{
+	static_key_slow_dec(&__cfs_bandwidth_used);
 }
 #else /* HAVE_JUMP_LABEL */
 static bool cfs_bandwidth_used(void)

@@ -1407,7 +1408,8 @@ static bool cfs_bandwidth_used(void)
 	return true;
 }
 
-void account_cfs_bandwidth_used(int enabled, int was_enabled) {}
+void cfs_bandwidth_usage_inc(void) {}
+void cfs_bandwidth_usage_dec(void) {}
 #endif /* HAVE_JUMP_LABEL */
 
 /*

kernel/sched/sched.h +2 -1

@@ -1140,7 +1140,8 @@ extern void init_cfs_rq(struct cfs_rq *c
 extern void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq);
 extern void unthrottle_offline_cfs_rqs(struct rq *rq);
 
-extern void account_cfs_bandwidth_used(int enabled, int was_enabled);
+extern void cfs_bandwidth_usage_inc(void);
+extern void cfs_bandwidth_usage_dec(void);
 
 #ifdef CONFIG_NO_HZ
 enum rq_nohz_flag_bits {

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #24 lore raw

PATCH 25/27 sched: Fix cfs_bandwidth misuse of hrtimer_expires_remaining

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ben Segall <[email protected]>

commit db06e78cc13d70f10877e0557becc88ab3ad2be8 upstream.

hrtimer_expires_remaining does not take internal hrtimer locks and thus must be guarded against concurrent __hrtimer_start_range_ns (but returning HRTIMER_RESTART is safe). Use cfs_b->lock to make it safe.

Signed-off-by Ben Segall <[email protected]>

Link http://lkml.kernel.org/r/20131016181617.22647.73829.stgit@sword-of-the-dawn.mtv.corp.google.com open here

Cc [email protected]

Signed-off-by Ingo Molnar <[email protected]>

Cc Chris J Arges <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 kernel/sched/fair.c |   14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

kernel/sched/fair.c +11 -3

@@ -1831,7 +1831,13 @@ static const u64 min_bandwidth_expiratio
 /* how long we wait to gather additional slack before distributing */
 static const u64 cfs_bandwidth_slack_period = 5 * NSEC_PER_MSEC;
 
-/* are we near the end of the current quota period? */
+/*
+ * Are we near the end of the current quota period?
+ *
+ * Requires cfs_b->lock for hrtimer_expires_remaining to be safe against the
+ * hrtimer base being cleared by __hrtimer_start_range_ns. In the case of
+ * migrate_hrtimers, base is never cleared, so we are fine.
+ */
 static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire)
 {
 	struct hrtimer *refresh_timer = &cfs_b->period_timer;

@@ -1907,10 +1913,12 @@ static void do_sched_cfs_slack_timer(str
 	u64 expires;
 
 	/* confirm we're still not at a refresh boundary */
-	if (runtime_refresh_within(cfs_b, min_bandwidth_expiration))
+	raw_spin_lock(&cfs_b->lock);
+	if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) {
+		raw_spin_unlock(&cfs_b->lock);
 		return;
+	}
 
-	raw_spin_lock(&cfs_b->lock);
 	if (cfs_b->quota != RUNTIME_INF && cfs_b->runtime > slice) {
 		runtime = cfs_b->runtime;
 		cfs_b->runtime = 0;

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #25 lore raw

PATCH 26/27 sched: Fix hrtimer_cancel()/rq->lock deadlock

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ben Segall <[email protected]>

commit 927b54fccbf04207ec92f669dce6806848cbec7d upstream.

__start_cfs_bandwidth calls hrtimer_cancel while holding rq->lock, waiting for the hrtimer to finish. However, if sched_cfs_period_timer runs for another loop iteration, the hrtimer can attempt to take rq->lock, resulting in deadlock.

Fix this by ensuring that cfs_b->timer_active is cleared only if the _latest_ call to do_sched_cfs_period_timer is returning as idle. Then __start_cfs_bandwidth can just call hrtimer_try_to_cancel and wait for that to succeed or timer_active == 1.

Signed-off-by Ben Segall <[email protected]>

Link http://lkml.kernel.org/r/20131016181622.22647.16643.stgit@sword-of-the-dawn.mtv.corp.google.com open here

Cc [email protected]

Signed-off-by Ingo Molnar <[email protected]>

Cc Chris J Arges <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 kernel/sched/fair.c |   15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

kernel/sched/fair.c +11 -4

@@ -1771,6 +1771,13 @@ static int do_sched_cfs_period_timer(str
 	if (idle)
 		goto out_unlock;
 
+	/*
+	 * if we have relooped after returning idle once, we need to update our
+	 * status as actually running, so that other cpus doing
+	 * __start_cfs_bandwidth will stop trying to cancel us.
+	 */
+	cfs_b->timer_active = 1;
+
 	__refill_cfs_bandwidth_runtime(cfs_b);
 
 	if (!throttled) {

@@ -2043,11 +2050,11 @@ void __start_cfs_bandwidth(struct cfs_ba
 	 * (timer_active==0 becomes visible before the hrtimer call-back
 	 * terminates).  In either case we ensure that it's re-programmed
 	 */
-	while (unlikely(hrtimer_active(&cfs_b->period_timer))) {
+	while (unlikely(hrtimer_active(&cfs_b->period_timer)) &&
+	       hrtimer_try_to_cancel(&cfs_b->period_timer) < 0) {
+		/* bounce the lock to allow do_sched_cfs_period_timer to run */
 		raw_spin_unlock(&cfs_b->lock);
-		/* ensure cfs_b->lock is available while we wait */
-		hrtimer_cancel(&cfs_b->period_timer);
-
+		cpu_relax();
 		raw_spin_lock(&cfs_b->lock);
 		/* if someone else restarted the timer then we're done */
 		if (cfs_b->timer_active)

Greg Kroah-Hartman

author maintainer

2014-01-14 00:26 UTC #26 lore raw

PATCH 27/27 sched: Guarantee new group-entities always have weight

3.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paul Turner <[email protected]>

commit 0ac9b1c21874d2490331233b3242085f8151e166 upstream.

Currently, group entity load-weights are initialized to zero. This admits some races with respect to the first time they are re-weighted in earlty use. ( Let g[x] denote the se for "g" on cpu "x". )

Suppose that we have root->a and that a enters a throttled state, immediately followed by a[0]->t1 (the only task running on cpu[0]) blocking:

put_prev_task(group_cfs_rq(a[0]), t1)
put_prev_entity(..., t1)
check_cfs_rq_runtime(group_cfs_rq(a[0]))
throttle_cfs_rq(group_cfs_rq(a[0]))

Then, before unthrottling occurs, let a[0]->b[0]->t2 wake for the first time:

enqueue_task_fair(rq[0], t2) enqueue_entity(group_cfs_rq(b[0]), t2) enqueue_entity_load_avg(group_cfs_rq(b[0]), t2) account_entity_enqueue(group_cfs_ra(b[0]), t2) update_cfs_shares(group_cfs_rq(b[0]))
< skipped because b is part of a throttled hierarchy > enqueue_entity(group_cfs_rq(a[0]), b[0])
...

We now have b[0] enqueued, yet group_cfs_rq(a[0])->load.weight == 0 which violates invariants in several code-paths. Eliminate the possibility of this by initializing group entity weight.

Signed-off-by Paul Turner <[email protected]>

Link http://lkml.kernel.org/r/20131016181627.22647.47543.stgit@sword-of-the-dawn.mtv.corp.google.com open here

Signed-off-by Ingo Molnar <[email protected]>

Cc Chris J Arges <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 kernel/sched/fair.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

kernel/sched/fair.c +2 -1

@@ -5470,7 +5470,8 @@ void init_tg_cfs_entry(struct task_group
 		se->cfs_rq = parent->my_q;
 
 	se->my_q = cfs_rq;
-	update_load_set(&se->load, 0);
+	/* guarantee group entities always have weight */
+	update_load_set(&se->load, NICE_0_LOAD);
 	se->parent = parent;
 }

Guenter Roeck

2014-01-14 02:59 UTC #27 lore raw

Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 3.4.77 release. There are 27 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Thu Jan 16 00:26:11 UTC 2014. Anything received after that time might be too late.

Build results:

	total: 119 pass: 97 skipped: 18 fail: 4

qemu tests all passed.

Results are as expected. Note that I dropped arm:allmodconfig and replaced it with several individual builds.

Details are available as always at http://server.roeck-us.net:8010/builders.

Guenter

Greg Kroah-Hartman

author maintainer

2014-01-14 03:03 UTC #28 lore raw

Quoted 10 lines · Guenter Roeck

Quoted 8 lines · Greg Kroah-Hartman
This is the start of the stable review cycle for the 3.4.77 release. There are 27 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Thu Jan 16 00:26:11 UTC 2014. Anything received after that time might be too late.

Quoted 8 lines

Build results:
	total: 119 pass: 97 skipped: 18 fail: 4
qemu tests all passed.
Results are as expected. Note that I dropped arm:allmodconfig and replaced it with several individual builds.
Details are available as always at http://server.roeck-us.net:8010/builders.

Great, thanks for testing.

greg k-h

Shuah Khan

2014-01-14 19:29 UTC #29 lore raw