4.2.5-stable review

This is the start of the stable review cycle for the 4.2.5 release. There are 54 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 Sun Oct 25 17:45:07 UTC 2015. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.2.5-rc1.gz

and the diffstat can be found below.

thanks,

greg k-h

-------------
Pseudo-Shortlog of commits:

Greg Kroah-Hartman <[email protected]>

    Linux 4.2.5-rc1

Steve Wise <[email protected]>
svcrdma: handle rdma read with a non-zero initial page offset

Steve Capper <[email protected]>
arm64: Fix THP protection change logic

Uwe Kleine-König <[email protected]>
pinctrl: imx25: ensure that a pin with id i is at position i in the info array

Thomas Gleixner <[email protected]>
timekeeping: Increment clock_was_set_seq in timekeeping_init()

Ilya Dryomov <[email protected]>

    rbd: fix double free on rbd_dev->header_name

Mike Snitzer <[email protected]>
dm thin: fix missing pool reference count decrement in pool_ctr error path

Michel Dänzer <[email protected]>
drm/amdgpu: Keep the pflip interrupts always enabled v7

Alex Deucher <[email protected]>
drm/amdgpu: check before checking pci bridge registers

Alex Deucher <[email protected]>
drm/amdgpu: fix num_crtc on CZ

Alex Deucher <[email protected]>
drm/amdgpu: add pm sysfs files late

Alex Deucher <[email protected]>
drm/radeon: add pm sysfs files late

Dave Airlie <[email protected]>
drm/radeon: attach tile property to mst connector

Dave Airlie <[email protected]>
drm/dp/mst: make mst i2c transfer code more robust.

Ben Skeggs <[email protected]>
drm/nouveau/fbcon: take runpm reference when userspace has an open fd

Gerd Hoffmann <[email protected]>
drm/qxl: fix framebuffer dirty rectangle tracking.

Paolo Bonzini <[email protected]>
KVM: x86: fix RSM into 64-bit protected mode

Paolo Bonzini <[email protected]>
KVM: x86: fix SMI to halted VCPU

Paolo Bonzini <[email protected]>
KVM: x86: clean up kvm_arch_vcpu_runnable

Shaohua Li <[email protected]>
workqueue: make sure delayed work run in local cpu

Wolfram Sang <[email protected]>
i2c: designware-platdrv: enable RuntimePM before registering to the core

Mika Westerberg <[email protected]>
i2c: designware: Do not use parameters from ACPI on Dell Inspiron 7348

Wolfram Sang <[email protected]>
i2c: s3c2410: enable RuntimePM before registering to the core

Wolfram Sang <[email protected]>
i2c: rcar: enable RuntimePM before registering to the core

Javier Martinez Canillas <[email protected]>

    mfd: max77843: Fix max77843_chg_init() return on error

Christoph Hellwig <[email protected]>
nfsd/blocklayout: accept any minlength

Linus Walleij <[email protected]>
ARM: ux500: simplify secondary CPU boot

Will Deacon <[email protected]>
arm64: errata: use KBUILD_CFLAGS_MODULE for erratum #843419

Chris Mason <[email protected]>
btrfs: fix use after free iterating extrefs

David Sterba <[email protected]>
btrfs: check unsupported filters in balance arguments

Shaohua Li <[email protected]>
memcg: convert threshold to bytes

Russell King <[email protected]>
crypto: ahash - ensure statesize is non-zero

Ben Hutchings <[email protected]>
crypto: camellia_aesni_avx - Fix CPU feature checks

Dave Kleikamp <[email protected]>
crypto: sparc - initialize blkcipher.ivsize

Daniel Vetter <[email protected]>
drm: Fix locking for sysfs dpms file

Arad, Ronen <[email protected]>
netlink: Trim skb to alloc size to avoid MSG_TRUNC

Jon Paul Maloy <[email protected]>
tipc: move fragment importance field to new header position

Joe Perches <[email protected]>
ethtool: Use kcalloc instead of kmalloc for ethtool_get_strings

Eric W. Biederman <[email protected]>
ipv6: Don't call with rt6_uncached_list_flush_dev

Alexei Starovoitov <[email protected]>
bpf: clear sender_cpu before xmit

WANG Cong <[email protected]>
act_mirred: clear sender cpu before sending to tx

Konstantin Khlebnikov <[email protected]>
ovs: do not allocate memory from offline numa node

Daniel Borkmann <[email protected]>
bpf: fix panic in SO_GET_FILTER with native ebpf programs

Eric Dumazet <[email protected]>
inet: fix race in reqsk_queue_unlink()

Guillaume Nault <[email protected]>

    ppp: don't override sk->sk_state in pppoe_flush_dev()

Eric Dumazet <[email protected]>
net: add pfmemalloc check in sk_add_backlog()

Eric Dumazet <[email protected]>
inet: fix races in reqsk_queue_hash_req()

Vivien Didelot <[email protected]>
net: dsa: fix preparation of a port STP update

Pravin B Shelar <[email protected]>
skbuff: Fix skb checksum partial check.

Pravin B Shelar <[email protected]>
skbuff: Fix skb checksum flag on skb pull

Andrey Vagin <[email protected]>
net/unix: fix logic about sk_peek_offset

Aaron Conole <[email protected]>
af_unix: return data from multiple SKBs on recv() with MSG_PEEK flag

Aaron Conole <[email protected]>
af_unix: Convert the unix_sk macro to an inline function for type safety

Alexander Couzens <[email protected]>

    l2tp: protect tunnel->del_work by ref_count

Ivan Mikhaylov <[email protected]>
net/ibm/emac: bump version numbers for correct work with ethtool

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

Diffstat:

 Makefile                                    |   4 +-
 arch/arm/mach-ux500/Makefile                |   2 +-
 arch/arm/mach-ux500/cpu-db8500.c            |   1 -
 arch/arm/mach-ux500/headsmp.S               |  37 --------
 arch/arm/mach-ux500/platsmp.c               | 132 ++++++++--------------------
 arch/arm/mach-ux500/setup.h                 |   1 -
 arch/arm64/Makefile                         |   2 +-
 arch/arm64/include/asm/pgtable.h            |   4 +-
 arch/sparc/crypto/aes_glue.c                |   2 +
 arch/sparc/crypto/camellia_glue.c           |   1 +
 arch/sparc/crypto/des_glue.c                |   2 +
 arch/x86/crypto/camellia_aesni_avx_glue.c   |   5 ++
 arch/x86/kvm/emulate.c                      |  10 ++-
 arch/x86/kvm/x86.c                          |  42 ++++++---
 crypto/ahash.c                              |   3 +-
 drivers/block/rbd.c                         |   1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_display.c |   2 -
 drivers/gpu/drm/amd/amdgpu/ci_dpm.c         |   8 +-
 drivers/gpu/drm/amd/amdgpu/cik.c            |   3 +
 drivers/gpu/drm/amd/amdgpu/cz_dpm.c         |  10 ++-
 drivers/gpu/drm/amd/amdgpu/dce_v10_0.c      |  30 ++++++-
 drivers/gpu/drm/amd/amdgpu/dce_v11_0.c      |  32 ++++++-
 drivers/gpu/drm/amd/amdgpu/dce_v8_0.c       |  30 ++++++-
 drivers/gpu/drm/amd/amdgpu/kv_dpm.c         |   9 +-
 drivers/gpu/drm/amd/amdgpu/vi.c             |   3 +
 drivers/gpu/drm/drm_dp_mst_topology.c       |   3 +-
 drivers/gpu/drm/drm_sysfs.c                 |  12 +--
 drivers/gpu/drm/nouveau/nouveau_fbcon.c     |  24 +++++
 drivers/gpu/drm/qxl/qxl_fb.c                |  19 ++--
 drivers/gpu/drm/radeon/radeon_display.c     |  14 +--
 drivers/gpu/drm/radeon/radeon_dp_mst.c      |   1 +
 drivers/gpu/drm/radeon/radeon_pm.c          |  63 +++++++------
 drivers/i2c/busses/i2c-designware-platdrv.c |  33 +++++--
 drivers/i2c/busses/i2c-rcar.c               |   7 +-
 drivers/i2c/busses/i2c-s3c2410.c            |   8 +-
 drivers/md/dm-thin.c                        |   2 +-
 drivers/mfd/max77843.c                      |   2 +-
 drivers/net/ethernet/ibm/emac/core.h        |   6 +-
 drivers/net/ppp/pppoe.c                     |   1 -
 drivers/pinctrl/freescale/pinctrl-imx25.c   |   4 +-
 fs/btrfs/backref.c                          |   8 +-
 fs/btrfs/ioctl.c                            |   5 ++
 fs/btrfs/volumes.h                          |   8 ++
 fs/nfsd/blocklayout.c                       |   8 --
 include/drm/drm_dp_mst_helper.h             |   3 +-
 include/linux/skbuff.h                      |   3 +
 include/net/af_unix.h                       |   6 +-
 include/net/sock.h                          |   8 ++
 kernel/time/timekeeping.c                   |   2 +-
 kernel/workqueue.c                          |   8 +-
 mm/memcontrol.c                             |   1 +
 net/core/ethtool.c                          |   2 +-
 net/core/filter.c                           |   7 +-
 net/core/skbuff.c                           |   9 +-
 net/dsa/slave.c                             |  11 ++-
 net/ipv4/inet_connection_sock.c             |  27 +++---
 net/ipv6/route.c                            |  12 +--
 net/l2tp/l2tp_core.c                        |  11 ++-
 net/netlink/af_netlink.c                    |  34 ++++---
 net/openvswitch/flow_table.c                |   3 +-
 net/sched/act_mirred.c                      |   1 +
 net/sunrpc/xprtrdma/svc_rdma_recvfrom.c     |   6 +-
 net/tipc/msg.h                              |   4 +-
 net/unix/af_unix.c                          |  17 +++-
 64 files changed, 466 insertions(+), 313 deletions(-)

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

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

From: Ivan Mikhaylov <[email protected]>

[ Upstream commit 661dfc65f7981481ba2e31aaa702371e82336e56 ]

The size of the MAC register dump used to be the size specified by the reg property in the device tree. Userland has no good way of finding out that size, and it was not specified consistently for each MAC type, so ethtool would end up printing junk at the end of the register dump if the device tree didn't match the size it assumed.

Using the new version numbers indicates unambiguously that the size of the MAC register dump is dependent only on the MAC type.

---
 drivers/net/ethernet/ibm/emac/core.h |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

@@ -460,8 +460,8 @@ struct emac_ethtool_regs_subhdr {
 	u32 index;
 };
 
-#define EMAC_ETHTOOL_REGS_VER		0
-#define EMAC4_ETHTOOL_REGS_VER		1
-#define EMAC4SYNC_ETHTOOL_REGS_VER	2
+#define EMAC_ETHTOOL_REGS_VER		3
+#define EMAC4_ETHTOOL_REGS_VER		4
+#define EMAC4SYNC_ETHTOOL_REGS_VER	5
 
 #endif /* __IBM_NEWEMAC_CORE_H */

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

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

From: Alexander Couzens <[email protected]>

[ Upstream commit 06a15f51cf3618e32a73871ee6a547ef7fd902b5 ]

There is a small chance that tunnel_free() is called before tunnel->del_work scheduled resulting in a zero pointer dereference.

---
 net/l2tp/l2tp_core.c |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

@@ -1319,7 +1319,7 @@ static void l2tp_tunnel_del_work(struct
 	tunnel = container_of(work, struct l2tp_tunnel, del_work);
 	sk = l2tp_tunnel_sock_lookup(tunnel);
 	if (!sk)
-		return;
+		goto out;
 
 	sock = sk->sk_socket;
 

@@ -1341,6 +1341,8 @@ static void l2tp_tunnel_del_work(struct
 	}
 
 	l2tp_tunnel_sock_put(sk);
+out:
+	l2tp_tunnel_dec_refcount(tunnel);
 }
 
 /* Create a socket for the tunnel, if one isn't set up by

@@ -1636,8 +1638,13 @@ EXPORT_SYMBOL_GPL(l2tp_tunnel_create);
  */
 int l2tp_tunnel_delete(struct l2tp_tunnel *tunnel)
 {
+	l2tp_tunnel_inc_refcount(tunnel);
 	l2tp_tunnel_closeall(tunnel);
-	return (false == queue_work(l2tp_wq, &tunnel->del_work));
+	if (false == queue_work(l2tp_wq, &tunnel->del_work)) {
+		l2tp_tunnel_dec_refcount(tunnel);
+		return 1;
+	}
+	return 0;
 }
 EXPORT_SYMBOL_GPL(l2tp_tunnel_delete);
 

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

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

From: Aaron Conole <[email protected]>

[ Upstream commit 4613012db1d911f80897f9446a49de817b2c4c47 ]

As suggested by Eric Dumazet this change replaces the #define with a static inline function to enjoy complaints by the compiler when misusing the API.

---
 include/net/af_unix.h |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -63,7 +63,11 @@ struct unix_sock {
 #define UNIX_GC_MAYBE_CYCLE	1
 	struct socket_wq	peer_wq;
 };
-#define unix_sk(__sk) ((struct unix_sock *)__sk)
+
+static inline struct unix_sock *unix_sk(struct sock *sk)
+{
+	return (struct unix_sock *)sk;
+}
 
 #define peer_wait peer_wq.wait
 

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

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

From: Aaron Conole <[email protected]>

[ Upstream commit 9f389e35674f5b086edd70ed524ca0f287259725 ]

AF_UNIX sockets now return multiple skbs from recv() when MSG_PEEK flag is set.

This is referenced in kernel bugzilla #12323 @
https://bugzilla.kernel.org/show_bug.cgi?id=12323

As described both in the BZ and lkml thread @
http://lkml.org/lkml/2008/1/8/444 calling recv() with MSG_PEEK on an AF_UNIX socket only reads a single skb, where the desired effect is to return as much skb data has been queued, until hitting the recv buffer size (whichever comes first).

The modified MSG_PEEK path will now move to the next skb in the tree and jump to the again: label, rather than following the natural loop structure. This requires duplicating some of the loop head actions.

This was tested using the python socketpair python code attached to the bugzilla issue.

---
 net/unix/af_unix.c |   15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

@@ -2179,8 +2179,21 @@ unlock:
 			if (UNIXCB(skb).fp)
 				scm.fp = scm_fp_dup(UNIXCB(skb).fp);
 
-			sk_peek_offset_fwd(sk, chunk);
+			if (skip) {
+				sk_peek_offset_fwd(sk, chunk);
+				skip -= chunk;
+			}
 
+			if (UNIXCB(skb).fp)
+				break;
+
+			last = skb;
+			last_len = skb->len;
+			unix_state_lock(sk);
+			skb = skb_peek_next(skb, &sk->sk_receive_queue);
+			if (skb)
+				goto again;
+			unix_state_unlock(sk);
 			break;
 		}
 	} while (size);

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

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

From: Andrey Vagin <[email protected]>

[ Upstream commit e9193d60d363e4dff75ff6d43a48f22be26d59c7 ]

Now send with MSG_PEEK can return data from multiple SKBs.

Unfortunately we take into account the peek offset for each skb, that is wrong. We need to apply the peek offset only once.

In addition, the peek offset should be used only if MSG_PEEK is set.

---
 net/unix/af_unix.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

@@ -2064,6 +2064,11 @@ static int unix_stream_read_generic(stru
 		goto out;
 	}
 
+	if (flags & MSG_PEEK)
+		skip = sk_peek_offset(sk, flags);
+	else
+		skip = 0;
+
 	do {
 		int chunk;
 		struct sk_buff *skb, *last;

@@ -2112,7 +2117,6 @@ unlock:
 			break;
 		}
 
-		skip = sk_peek_offset(sk, flags);
 		while (skip >= unix_skb_len(skb)) {
 			skip -= unix_skb_len(skb);
 			last = skb;

@@ -2179,14 +2183,12 @@ unlock:
 			if (UNIXCB(skb).fp)
 				scm.fp = scm_fp_dup(UNIXCB(skb).fp);
 
-			if (skip) {
-				sk_peek_offset_fwd(sk, chunk);
-				skip -= chunk;
-			}
+			sk_peek_offset_fwd(sk, chunk);
 
 			if (UNIXCB(skb).fp)
 				break;
 
+			skip = 0;
 			last = skb;
 			last_len = skb->len;
 			unix_state_lock(sk);

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

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

From: Pravin B Shelar <[email protected]>

[ Upstream commit 6ae459bdaaeebc632b16e54dcbabb490c6931d61 ]

VXLAN device can receive skb with checksum partial. But the checksum offset could be in outer header which is pulled on receive. This results in negative checksum offset for the skb. Such skb can cause the assert failure in skb_checksum_help(). Following patch fixes the bug by setting checksum-none while pulling outer header.

Following is the kernel panic msg from old kernel hitting the bug.

------------[ cut here ]------------
kernel BUG at net/core/dev.c:1906!
RIP: 0010:[<ffffffff81518034>] skb_checksum_help+0x144/0x150
Call Trace:
<IRQ>
[<ffffffffa0164c28>] queue_userspace_packet+0x408/0x470 [openvswitch]
[<ffffffffa016614d>] ovs_dp_upcall+0x5d/0x60 [openvswitch]
[<ffffffffa0166236>] ovs_dp_process_packet_with_key+0xe6/0x100 [openvswitch]
[<ffffffffa016629b>] ovs_dp_process_received_packet+0x4b/0x80 [openvswitch]
[<ffffffffa016c51a>] ovs_vport_receive+0x2a/0x30 [openvswitch]
[<ffffffffa0171383>] vxlan_rcv+0x53/0x60 [openvswitch]
[<ffffffffa01734cb>] vxlan_udp_encap_recv+0x8b/0xf0 [openvswitch]
[<ffffffff8157addc>] udp_queue_rcv_skb+0x2dc/0x3b0
[<ffffffff8157b56f>] __udp4_lib_rcv+0x1cf/0x6c0
[<ffffffff8157ba7a>] udp_rcv+0x1a/0x20
[<ffffffff8154fdbd>] ip_local_deliver_finish+0xdd/0x280
[<ffffffff81550128>] ip_local_deliver+0x88/0x90
[<ffffffff8154fa7d>] ip_rcv_finish+0x10d/0x370
[<ffffffff81550365>] ip_rcv+0x235/0x300
[<ffffffff8151ba1d>] __netif_receive_skb+0x55d/0x620
[<ffffffff8151c360>] netif_receive_skb+0x80/0x90
[<ffffffff81459935>] virtnet_poll+0x555/0x6f0
[<ffffffff8151cd04>] net_rx_action+0x134/0x290
[<ffffffff810683d8>] __do_softirq+0xa8/0x210
[<ffffffff8162fe6c>] call_softirq+0x1c/0x30
[<ffffffff810161a5>] do_softirq+0x65/0xa0
[<ffffffff810687be>] irq_exit+0x8e/0xb0
[<ffffffff81630733>] do_IRQ+0x63/0xe0
[<ffffffff81625f2e>] common_interrupt+0x6e/0x6e

---
 include/linux/skbuff.h |    3 +++
 1 file changed, 3 insertions(+)

@@ -2601,6 +2601,9 @@ static inline void skb_postpull_rcsum(st
 {
 	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
+	else if (skb->ip_summed == CHECKSUM_PARTIAL &&
+		 skb_checksum_start_offset(skb) <= len)
+		skb->ip_summed = CHECKSUM_NONE;
 }
 
 unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len);

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

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

From: Pravin B Shelar <[email protected]>

[ Upstream commit 31b33dfb0a144469dd805514c9e63f4993729a48 ]

Earlier patch 6ae459bda tried to detect void ckecksum partial skb by comparing pull length to checksum offset. But it does not work for all cases since checksum-offset depends on updates to skb->data.

Following patch fixes it by validating checksum start offset after skb-data pointer is updated. Negative value of checksum offset start means there is no need to checksum.

---
 include/linux/skbuff.h |    2 +-
 net/core/skbuff.c      |    9 +++++----
 2 files changed, 6 insertions(+), 5 deletions(-)

@@ -2602,7 +2602,7 @@ static inline void skb_postpull_rcsum(st
 	if (skb->ip_summed == CHECKSUM_COMPLETE)
 		skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0));
 	else if (skb->ip_summed == CHECKSUM_PARTIAL &&
-		 skb_checksum_start_offset(skb) <= len)
+		 skb_checksum_start_offset(skb) < 0)
 		skb->ip_summed = CHECKSUM_NONE;
 }
 

@@ -2958,11 +2958,12 @@ EXPORT_SYMBOL_GPL(skb_append_pagefrags);
  */
 unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len)
 {
+	unsigned char *data = skb->data;
+
 	BUG_ON(len > skb->len);
-	skb->len -= len;
-	BUG_ON(skb->len < skb->data_len);
-	skb_postpull_rcsum(skb, skb->data, len);
-	return skb->data += len;
+	__skb_pull(skb, len);
+	skb_postpull_rcsum(skb, data, len);
+	return skb->data;
 }
 EXPORT_SYMBOL_GPL(skb_pull_rcsum);
 

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

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

From: Vivien Didelot <[email protected]>

[ Upstream commit 57a47532c4312159935c98b7f1cf0e62296b9171 ]

Because of the default 0 value of ret in dsa_slave_port_attr_set, a driver may return -EOPNOTSUPP from the commit phase of a STP state, which triggers a WARN() from switchdev.

This happened on a 6185 switch which does not support hardware bridging.

---
 net/dsa/slave.c |   11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

@@ -348,12 +348,17 @@ static int dsa_slave_stp_update(struct n
 static int dsa_slave_port_attr_set(struct net_device *dev,
 				   struct switchdev_attr *attr)
 {
-	int ret = 0;
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct dsa_switch *ds = p->parent;
+	int ret;
 
 	switch (attr->id) {
 	case SWITCHDEV_ATTR_PORT_STP_STATE:
-		if (attr->trans == SWITCHDEV_TRANS_COMMIT)
-			ret = dsa_slave_stp_update(dev, attr->u.stp_state);
+		if (attr->trans == SWITCHDEV_TRANS_PREPARE)
+			ret = ds->drv->port_stp_update ? 0 : -EOPNOTSUPP;
+		else
+			ret = ds->drv->port_stp_update(ds, p->port,
+						       attr->u.stp_state);
 		break;
 	default:
 		ret = -EOPNOTSUPP;

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

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

From: Eric Dumazet <[email protected]>

[ Upstream commit 29c6852602e259d2c1882f320b29d5c3fec0de04 ]

Before allowing lockless LISTEN processing, we need to make sure to arm the SYN_RECV timer before the req socket is visible in hash tables.

Also, req->rsk_hash should be written before we set rsk_refcnt to a non zero value.

---
 net/ipv4/inet_connection_sock.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -685,20 +685,20 @@ void reqsk_queue_hash_req(struct request
 	req->num_timeout = 0;
 	req->sk = NULL;
 
+	setup_timer(&req->rsk_timer, reqsk_timer_handler, (unsigned long)req);
+	mod_timer_pinned(&req->rsk_timer, jiffies + timeout);
+	req->rsk_hash = hash;
+
 	/* before letting lookups find us, make sure all req fields
 	 * are committed to memory and refcnt initialized.
 	 */
 	smp_wmb();
 	atomic_set(&req->rsk_refcnt, 2);
-	setup_timer(&req->rsk_timer, reqsk_timer_handler, (unsigned long)req);
-	req->rsk_hash = hash;
 
 	spin_lock(&queue->syn_wait_lock);
 	req->dl_next = lopt->syn_table[hash];
 	lopt->syn_table[hash] = req;
 	spin_unlock(&queue->syn_wait_lock);
-
-	mod_timer_pinned(&req->rsk_timer, jiffies + timeout);
 }
 EXPORT_SYMBOL(reqsk_queue_hash_req);
 

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

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

From: Eric Dumazet <[email protected]>

[ Upstream commit c7c49b8fde26b74277188bdc6c9dca38db6fa35b ]

Greg reported crashes hitting the following check in __sk_backlog_rcv()

	BUG_ON(!sock_flag(sk, SOCK_MEMALLOC));

The pfmemalloc bit is currently checked in sk_filter().

This works correctly for TCP, because sk_filter() is ran in tcp_v[46]_rcv() before hitting the prequeue or backlog checks.

For UDP or other protocols, this does not work, because the sk_filter() is ran from sock_queue_rcv_skb(), which might be called _after_ backlog queuing if socket is owned by user by the time packet is processed by softirq handler.

---
 include/net/sock.h |    8 ++++++++
 1 file changed, 8 insertions(+)

@@ -826,6 +826,14 @@ static inline __must_check int sk_add_ba
 	if (sk_rcvqueues_full(sk, limit))
 		return -ENOBUFS;
 
+	/*
+	 * If the skb was allocated from pfmemalloc reserves, only
+	 * allow SOCK_MEMALLOC sockets to use it as this socket is
+	 * helping free memory
+	 */
+	if (skb_pfmemalloc(skb) && !sock_flag(sk, SOCK_MEMALLOC))
+		return -ENOMEM;
+
 	__sk_add_backlog(sk, skb);
 	sk->sk_backlog.len += skb->truesize;
 	return 0;

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

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

From: Guillaume Nault <[email protected]>

[ Upstream commit e6740165b8f7f06d8caee0fceab3fb9d790a6fed ]

Since commit 2b018d57ff18 ("pppoe: drop PPPOX_ZOMBIEs in pppoe_release"), pppoe_release() calls dev_put(po->pppoe_dev) if sk is in the PPPOX_ZOMBIE state. But pppoe_flush_dev() can set sk->sk_state to PPPOX_ZOMBIE _and_ reset po->pppoe_dev to NULL. This leads to the following oops:

[  570.140800] BUG: unable to handle kernel NULL pointer dereference at 00000000000004e0
[  570.142931] IP: [<ffffffffa018c701>] pppoe_release+0x50/0x101 [pppoe]
[  570.144601] PGD 3d119067 PUD 3dbc1067 PMD 0
[  570.144601] Oops: 0000 [#1] SMP
[  570.144601] Modules linked in: l2tp_ppp l2tp_netlink l2tp_core ip6_udp_tunnel udp_tunnel pppoe pppox ppp_generic slhc loop crc32c_intel ghash_clmulni_intel jitterentropy_rng sha256_generic hmac drbg ansi_cprng aesni_intel aes_x86_64 ablk_helper cryptd lrw gf128mul glue_helper acpi_cpufreq evdev serio_raw processor button ext4 crc16 mbcache jbd2 virtio_net virtio_blk virtio_pci virtio_ring virtio
[  570.144601] CPU: 1 PID: 15738 Comm: ppp-apitest Not tainted 4.2.0 #1
[  570.144601] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Debian-1.8.2-1 04/01/2014
[  570.144601] task: ffff88003d30d600 ti: ffff880036b60000 task.ti: ffff880036b60000
[  570.144601] RIP: 0010:[<ffffffffa018c701>]  [<ffffffffa018c701>] pppoe_release+0x50/0x101 [pppoe]
[  570.144601] RSP: 0018:ffff880036b63e08  EFLAGS: 00010202
[  570.144601] RAX: 0000000000000000 RBX: ffff880034340000 RCX: 0000000000000206
[  570.144601] RDX: 0000000000000006 RSI: ffff88003d30dd20 RDI: ffff88003d30dd20
[  570.144601] RBP: ffff880036b63e28 R08: 0000000000000001 R09: 0000000000000000
[  570.144601] R10: 00007ffee9b50420 R11: ffff880034340078 R12: ffff8800387ec780
[  570.144601] R13: ffff8800387ec7b0 R14: ffff88003e222aa0 R15: ffff8800387ec7b0
[  570.144601] FS:  00007f5672f48700(0000) GS:ffff88003fc80000(0000) knlGS:0000000000000000
[  570.144601] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  570.144601] CR2: 00000000000004e0 CR3: 0000000037f7e000 CR4: 00000000000406a0
[  570.144601] Stack:
[  570.144601]  ffffffffa018f240 ffff8800387ec780 ffffffffa018f240 ffff8800387ec7b0
[  570.144601]  ffff880036b63e48 ffffffff812caabe ffff880039e4e000 0000000000000008
[  570.144601]  ffff880036b63e58 ffffffff812cabad ffff880036b63ea8 ffffffff811347f5
[  570.144601] Call Trace:
[  570.144601]  [<ffffffff812caabe>] sock_release+0x1a/0x75
[  570.144601]  [<ffffffff812cabad>] sock_close+0xd/0x11
[  570.144601]  [<ffffffff811347f5>] __fput+0xff/0x1a5
[  570.144601]  [<ffffffff811348cb>] ____fput+0x9/0xb
[  570.144601]  [<ffffffff81056682>] task_work_run+0x66/0x90
[  570.144601]  [<ffffffff8100189e>] prepare_exit_to_usermode+0x8c/0xa7
[  570.144601]  [<ffffffff81001a26>] syscall_return_slowpath+0x16d/0x19b
[  570.144601]  [<ffffffff813babb1>] int_ret_from_sys_call+0x25/0x9f
[  570.144601] Code: 48 8b 83 c8 01 00 00 a8 01 74 12 48 89 df e8 8b 27 14 e1 b8 f7 ff ff ff e9 b7 00 00 00 8a 43 12 a8 0b 74 1c 48 8b 83 a8 04 00 00 <48> 8b 80 e0 04 00 00 65 ff 08 48 c7 83 a8 04 00 00 00 00 00 00
[  570.144601] RIP  [<ffffffffa018c701>] pppoe_release+0x50/0x101 [pppoe]
[  570.144601]  RSP <ffff880036b63e08>
[  570.144601] CR2: 00000000000004e0
[  570.200518] ---[ end trace 46956baf17349563 ]---

pppoe_flush_dev() has no reason to override sk->sk_state with PPPOX_ZOMBIE. pppox_unbind_sock() already sets sk->sk_state to PPPOX_DEAD, which is the correct state given that sk is unbound and po->pppoe_dev is NULL.

---
 drivers/net/ppp/pppoe.c |    1 -
 1 file changed, 1 deletion(-)

@@ -313,7 +313,6 @@ static void pppoe_flush_dev(struct net_d
 			if (po->pppoe_dev == dev &&
 			    sk->sk_state & (PPPOX_CONNECTED | PPPOX_BOUND | PPPOX_ZOMBIE)) {
 				pppox_unbind_sock(sk);
-				sk->sk_state = PPPOX_ZOMBIE;
 				sk->sk_state_change(sk);
 				po->pppoe_dev = NULL;
 				dev_put(dev);

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

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

From: Eric Dumazet <[email protected]>

[ Upstream commit 2306c704ce280c97a60d1f45333b822b40281dea ]

reqsk_timer_handler() tests if icsk_accept_queue.listen_opt is NULL at its beginning.

By the time it calls inet_csk_reqsk_queue_drop() and reqsk_queue_unlink(), listener might have been closed and inet_csk_listen_stop() had called reqsk_queue_yank_acceptq() which sets icsk_accept_queue.listen_opt to NULL

We therefore need to correctly check listen_opt being NULL after holding syn_wait_lock for proper synchronization.

---
 net/ipv4/inet_connection_sock.c |   19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

@@ -577,21 +577,22 @@ EXPORT_SYMBOL(inet_rtx_syn_ack);
 static bool reqsk_queue_unlink(struct request_sock_queue *queue,
 			       struct request_sock *req)
 {
-	struct listen_sock *lopt = queue->listen_opt;
 	struct request_sock **prev;
+	struct listen_sock *lopt;
 	bool found = false;
 
 	spin_lock(&queue->syn_wait_lock);
-
-	for (prev = &lopt->syn_table[req->rsk_hash]; *prev != NULL;
-	     prev = &(*prev)->dl_next) {
-		if (*prev == req) {
-			*prev = req->dl_next;
-			found = true;
-			break;
+	lopt = queue->listen_opt;
+	if (lopt) {
+		for (prev = &lopt->syn_table[req->rsk_hash]; *prev != NULL;
+		     prev = &(*prev)->dl_next) {
+			if (*prev == req) {
+				*prev = req->dl_next;
+				found = true;
+				break;
+			}
 		}
 	}
-
 	spin_unlock(&queue->syn_wait_lock);
 	if (timer_pending(&req->rsk_timer) && del_timer_sync(&req->rsk_timer))
 		reqsk_put(req);

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

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

From: Daniel Borkmann <[email protected]>

[ Upstream commit 93d08b6966cf730ea669d4d98f43627597077153 ]

When sockets have a native eBPF program attached through setsockopt(sk, SOL_SOCKET, SO_ATTACH_BPF, ...), and then try to dump these over getsockopt(sk, SOL_SOCKET, SO_GET_FILTER, ...), the following panic appears:

  [49904.178642] BUG: unable to handle kernel NULL pointer dereference at (null)
  [49904.178762] IP: [<ffffffff81610fd9>] sk_get_filter+0x39/0x90
  [49904.182000] PGD 86fc9067 PUD 531a1067 PMD 0
  [49904.185196] Oops: 0000 [#1] SMP
  [...]
  [49904.224677] Call Trace:
  [49904.226090]  [<ffffffff815e3d49>] sock_getsockopt+0x319/0x740
  [49904.227535]  [<ffffffff812f59e3>] ? sock_has_perm+0x63/0x70
  [49904.228953]  [<ffffffff815e2fc8>] ? release_sock+0x108/0x150
  [49904.230380]  [<ffffffff812f5a43>] ? selinux_socket_getsockopt+0x23/0x30
  [49904.231788]  [<ffffffff815dff36>] SyS_getsockopt+0xa6/0xc0
  [49904.233267]  [<ffffffff8171b9ae>] entry_SYSCALL_64_fastpath+0x12/0x71

The underlying issue is the very same as in commit b382c0865600 ("sock, diag: fix panic in sock_diag_put_filterinfo"), that is, native eBPF programs don't store an original program since this is only needed in cBPF ones.

However, sk_get_filter() wasn't updated to test for this at the time when eBPF could be attached. Just throw an error to the user to indicate that eBPF cannot be dumped over this interface. That way, it can also be known that a program _is_ attached (as opposed to just return 0), and a different (future) method needs to be consulted for a dump.

---
 net/core/filter.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -1701,9 +1701,13 @@ int sk_get_filter(struct sock *sk, struc
 		goto out;
 
 	/* We're copying the filter that has been originally attached,
-	 * so no conversion/decode needed anymore.
+	 * so no conversion/decode needed anymore. eBPF programs that
+	 * have no original program cannot be dumped through this.
 	 */
+	ret = -EACCES;
 	fprog = filter->prog->orig_prog;
+	if (!fprog)
+		goto out;
 
 	ret = fprog->len;
 	if (!len)

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

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

From: Konstantin Khlebnikov <[email protected]>

[ Upstream commit 598c12d0ba6de9060f04999746eb1e015774044b ]

When openvswitch tries allocate memory from offline numa node 0:
stats = kmem_cache_alloc_node(flow_stats_cache, GFP_KERNEL | __GFP_ZERO, 0) It catches VM_BUG_ON(nid < 0 || nid >= MAX_NUMNODES || !node_online(nid)) [ replaced with VM_WARN_ON(!node_online(nid)) recently ] in linux/gfp.h This patch disables numa affinity in this case.

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

@@ -92,7 +92,8 @@ struct sw_flow *ovs_flow_alloc(void)
 
 	/* Initialize the default stat node. */
 	stats = kmem_cache_alloc_node(flow_stats_cache,
-				      GFP_KERNEL | __GFP_ZERO, 0);
+				      GFP_KERNEL | __GFP_ZERO,
+				      node_online(0) ? 0 : NUMA_NO_NODE);
 	if (!stats)
 		goto err;
 

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

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

From: WANG Cong <[email protected]>

[ Upstream commit d40496a56430eac0d330378816954619899fe303 ]

Similar to commit c29390c6dfee ("xps: must clear sender_cpu before forwarding") the skb->sender_cpu needs to be cleared when moving from Rx
Tx, otherwise kernel could crash.

---
 net/sched/act_mirred.c |    1 +
 1 file changed, 1 insertion(+)

@@ -168,6 +168,7 @@ static int tcf_mirred(struct sk_buff *sk
 
 	skb2->skb_iif = skb->dev->ifindex;
 	skb2->dev = dev;
+	skb_sender_cpu_clear(skb2);
 	err = dev_queue_xmit(skb2);
 
 out:

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

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

From: Alexei Starovoitov <[email protected]>

[ Upstream commit 6bf0577374cfb6c2301dbf4934a4f23ad3d72763 ]

Similar to commit c29390c6dfee ("xps: must clear sender_cpu before forwarding") the skb->sender_cpu needs to be cleared before xmit.

---
 net/core/filter.c |    1 +
 1 file changed, 1 insertion(+)

@@ -1412,6 +1412,7 @@ static u64 bpf_clone_redirect(u64 r1, u6
 		return dev_forward_skb(dev, skb2);
 
 	skb2->dev = dev;
+	skb_sender_cpu_clear(skb2);
 	return dev_queue_xmit(skb2);
 }
 

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

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

From: "Eric W. Biederman" <[email protected]>

[ Upstream commit e332bc67cf5e5e5b71a1aec9750d0791aac65183 ]

As originally written rt6_uncached_list_flush_dev makes no sense when called with dev == NULL as it attempts to flush all uncached routes regardless of network namespace when dev == NULL. Which is simply incorrect behavior.

Furthermore at the point rt6_ifdown is called with dev == NULL no more network devices exist in the network namespace so even if the code in rt6_uncached_list_flush_dev were to attempt something sensible it would be meaningless.

Therefore remove support in rt6_uncached_list_flush_dev for handling network devices where dev == NULL, and only call rt6_uncached_list_flush_dev
when rt6_ifdown is called with a network device.

---
 net/ipv6/route.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

@@ -139,6 +139,9 @@ static void rt6_uncached_list_flush_dev(
 	struct net_device *loopback_dev = net->loopback_dev;
 	int cpu;
 
+	if (dev == loopback_dev)
+		return;
+
 	for_each_possible_cpu(cpu) {
 		struct uncached_list *ul = per_cpu_ptr(&rt6_uncached_list, cpu);
 		struct rt6_info *rt;

@@ -148,14 +151,12 @@ static void rt6_uncached_list_flush_dev(
 			struct inet6_dev *rt_idev = rt->rt6i_idev;
 			struct net_device *rt_dev = rt->dst.dev;
 
-			if (rt_idev && (rt_idev->dev == dev || !dev) &&
-			    rt_idev->dev != loopback_dev) {
+			if (rt_idev->dev == dev) {
 				rt->rt6i_idev = in6_dev_get(loopback_dev);
 				in6_dev_put(rt_idev);
 			}
 
-			if (rt_dev && (rt_dev == dev || !dev) &&
-			    rt_dev != loopback_dev) {
+			if (rt_dev == dev) {
 				rt->dst.dev = loopback_dev;
 				dev_hold(rt->dst.dev);
 				dev_put(rt_dev);

@@ -2577,7 +2578,8 @@ void rt6_ifdown(struct net *net, struct
 
 	fib6_clean_all(net, fib6_ifdown, &adn);
 	icmp6_clean_all(fib6_ifdown, &adn);
-	rt6_uncached_list_flush_dev(net, dev);
+	if (dev)
+		rt6_uncached_list_flush_dev(net, dev);
 }
 
 struct rt6_mtu_change_arg {

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

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

From: Joe Perches <[email protected]>

[ Upstream commit 077cb37fcf6f00a45f375161200b5ee0cd4e937b ]

It seems that kernel memory can leak into userspace by a kmalloc, ethtool_get_strings, then copy_to_user sequence.

Avoid this by using kcalloc to zero fill the copied buffer.

---
 net/core/ethtool.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1284,7 +1284,7 @@ static int ethtool_get_strings(struct ne
 
 	gstrings.len = ret;
 
-	data = kmalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER);
+	data = kcalloc(gstrings.len, ETH_GSTRING_LEN, GFP_USER);
 	if (!data)
 		return -ENOMEM;
 

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

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

From: Jon Paul Maloy <[email protected]>

[ Upstream commit dde4b5ae65de659b9ec64bafdde0430459fcb495 ]

In commit e3eea1eb47a ("tipc: clean up handling of message priorities") we introduced a field in the packet header for keeping track of the priority of fragments, since this value is not present in the specified protocol header. Since the value so far only is used at the transmitting end of the link, we have not yet officially defined it as part of the protocol.

Unfortunately, the field we use for keeping this value, bits 13-15 in in word 5, has turned out to be a poor choice; it is already used by the broadcast protocol for carrying the 'network id' field of the sending node. Since packet fragments also need to be transported across the broadcast protocol, the risk of conflict is obvious, and we see this happen when we use network identities larger than 2^13-1. This has escaped our testing because we have so far only been using small network id values.

We now move this field to bits 0-2 in word 9, a field that is guaranteed to be unused by all involved protocols.

---
 net/tipc/msg.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -357,7 +357,7 @@ static inline u32 msg_importance(struct
 	if (likely((usr <= TIPC_CRITICAL_IMPORTANCE) && !msg_errcode(m)))
 		return usr;
 	if ((usr == MSG_FRAGMENTER) || (usr == MSG_BUNDLER))
-		return msg_bits(m, 5, 13, 0x7);
+		return msg_bits(m, 9, 0, 0x7);
 	return TIPC_SYSTEM_IMPORTANCE;
 }
 

@@ -366,7 +366,7 @@ static inline void msg_set_importance(st
 	int usr = msg_user(m);
 
 	if (likely((usr == MSG_FRAGMENTER) || (usr == MSG_BUNDLER)))
-		msg_set_bits(m, 5, 13, 0x7, i);
+		msg_set_bits(m, 9, 0, 0x7, i);
 	else if (i < TIPC_SYSTEM_IMPORTANCE)
 		msg_set_user(m, i);
 	else

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

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

From: "Arad, Ronen" <[email protected]>

[ Upstream commit db65a3aaf29ecce2e34271d52e8d2336b97bd9fe ]

netlink_dump() allocates skb based on the calculated min_dump_alloc or a per socket max_recvmsg_len.
min_alloc_size is maximum space required for any single netdev attributes as calculated by rtnl_calcit().
max_recvmsg_len tracks the user provided buffer to netlink_recvmsg. It is capped at 16KiB.
The intention is to avoid small allocations and to minimize the number of calls required to obtain dump information for all net devices.

netlink_dump packs as many small messages as could fit within an skb that was sized for the largest single netdev information. The actual space available within an skb is larger than what is requested. It could be much larger and up to near 2x with align to next power of 2 approach.

Allowing netlink_dump to use all the space available within the allocated skb increases the buffer size a user has to provide to avoid truncaion (i.e. MSG_TRUNG flag set).

It was observed that with many VLANs configured on at least one netdev, a larger buffer of near 64KiB was necessary to avoid "Message truncated" error in "ip link" or "bridge [-c[ompressvlans]] vlan show" when min_alloc_size was only little over 32KiB.

This patch trims skb to allocated size in order to allow the user to avoid truncation with more reasonable buffer size.

---
 net/netlink/af_netlink.c |   34 ++++++++++++++++++++++------------
 1 file changed, 22 insertions(+), 12 deletions(-)

@@ -2750,6 +2750,7 @@ static int netlink_dump(struct sock *sk)
 	struct sk_buff *skb = NULL;
 	struct nlmsghdr *nlh;
 	int len, err = -ENOBUFS;
+	int alloc_min_size;
 	int alloc_size;
 
 	mutex_lock(nlk->cb_mutex);

@@ -2758,9 +2759,6 @@ static int netlink_dump(struct sock *sk)
 		goto errout_skb;
 	}
 
-	cb = &nlk->cb;
-	alloc_size = max_t(int, cb->min_dump_alloc, NLMSG_GOODSIZE);
-
 	if (!netlink_rx_is_mmaped(sk) &&
 	    atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
 		goto errout_skb;

@@ -2770,23 +2768,35 @@ static int netlink_dump(struct sock *sk)
 	 * to reduce number of system calls on dump operations, if user
 	 * ever provided a big enough buffer.
 	 */
-	if (alloc_size < nlk->max_recvmsg_len) {
-		skb = netlink_alloc_skb(sk,
-					nlk->max_recvmsg_len,
-					nlk->portid,
+	cb = &nlk->cb;
+	alloc_min_size = max_t(int, cb->min_dump_alloc, NLMSG_GOODSIZE);
+
+	if (alloc_min_size < nlk->max_recvmsg_len) {
+		alloc_size = nlk->max_recvmsg_len;
+		skb = netlink_alloc_skb(sk, alloc_size, nlk->portid,
 					GFP_KERNEL |
 					__GFP_NOWARN |
 					__GFP_NORETRY);
-		/* available room should be exact amount to avoid MSG_TRUNC */
-		if (skb)
-			skb_reserve(skb, skb_tailroom(skb) -
-					 nlk->max_recvmsg_len);
 	}
-	if (!skb)
+	if (!skb) {
+		alloc_size = alloc_min_size;
 		skb = netlink_alloc_skb(sk, alloc_size, nlk->portid,
 					GFP_KERNEL);
+	}
 	if (!skb)
 		goto errout_skb;
+
+	/* Trim skb to allocated size. User is expected to provide buffer as
+	 * large as max(min_dump_alloc, 16KiB (mac_recvmsg_len capped at
+	 * netlink_recvmsg())). dump will pack as many smaller messages as
+	 * could fit within the allocated skb. skb is typically allocated
+	 * with larger space than required (could be as much as near 2x the
+	 * requested size with align to next power of 2 approach). Allowing
+	 * dump to use the excess space makes it difficult for a user to have a
+	 * reasonable static buffer based on the expected largest dump of a
+	 * single netdev. The outcome is MSG_TRUNC error.
+	 */
+	skb_reserve(skb, skb_tailroom(skb) - alloc_size);
 	netlink_skb_set_owner_r(skb, sk);
 
 	len = cb->dump(skb, cb);

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

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

From: Daniel Vetter <[email protected]>

commit 621bd0f6982badd6483acb191eb7b6226a578328 upstream.

With atomic drivers we need to make sure that (at least in general) property reads hold the right locks. But the legacy dpms property is special and can be read locklessly. Since userspace loves to just randomly look at that all the time (like with "status") do that.

To make it clear that we play tricks use the READ_ONCE compiler barrier (and also for paranoia).

Note that there's not really anything bad going on since even with the new atomic paths we eventually end up not chasing any pointers (and hence possibly freed memory and other fun stuff). The locking WARNING has been added in

commit 88a48e297b3a3bac6022c03babfb038f1a886cea

Author: Rob Clark <[email protected]>
Date:   Thu Dec 18 16:01:50 2014 -0500

drm: add atomic properties

but since drivers are converting not everyone will have seen this from the start.

Jens reported this and submitted a patch to just grab the mode_config.connection_mutex, but we can do a bit better.

v2: Remove unused variables I failed to git add for real.

Reference: http://mid.gmane.org/[email protected]

---
 drivers/gpu/drm/drm_sysfs.c |   12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

@@ -235,18 +235,12 @@ static ssize_t dpms_show(struct device *
 			   char *buf)
 {
 	struct drm_connector *connector = to_drm_connector(device);
-	struct drm_device *dev = connector->dev;
-	uint64_t dpms_status;
-	int ret;
+	int dpms;
 
-	ret = drm_object_property_get_value(&connector->base,
-					    dev->mode_config.dpms_property,
-					    &dpms_status);
-	if (ret)
-		return 0;
+	dpms = READ_ONCE(connector->dpms);
 
 	return snprintf(buf, PAGE_SIZE, "%s\n",
-			drm_get_dpms_name((int)dpms_status));
+			drm_get_dpms_name(dpms));
 }
 
 static ssize_t enabled_show(struct device *device,

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

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

From: Dave Kleikamp <[email protected]>

commit a66d7f724a96d6fd279bfbd2ee488def6b081bea upstream.

Some of the crypto algorithms write to the initialization vector, but no space has been allocated for it. This clobbers adjacent memory.

---
 arch/sparc/crypto/aes_glue.c      |    2 ++
 arch/sparc/crypto/camellia_glue.c |    1 +
 arch/sparc/crypto/des_glue.c      |    2 ++
 3 files changed, 5 insertions(+)

@@ -433,6 +433,7 @@ static struct crypto_alg algs[] = { {
 		.blkcipher = {
 			.min_keysize	= AES_MIN_KEY_SIZE,
 			.max_keysize	= AES_MAX_KEY_SIZE,
+			.ivsize		= AES_BLOCK_SIZE,
 			.setkey		= aes_set_key,
 			.encrypt	= cbc_encrypt,
 			.decrypt	= cbc_decrypt,

@@ -452,6 +453,7 @@ static struct crypto_alg algs[] = { {
 		.blkcipher = {
 			.min_keysize	= AES_MIN_KEY_SIZE,
 			.max_keysize	= AES_MAX_KEY_SIZE,
+			.ivsize		= AES_BLOCK_SIZE,
 			.setkey		= aes_set_key,
 			.encrypt	= ctr_crypt,
 			.decrypt	= ctr_crypt,

@@ -274,6 +274,7 @@ static struct crypto_alg algs[] = { {
 		.blkcipher = {
 			.min_keysize	= CAMELLIA_MIN_KEY_SIZE,
 			.max_keysize	= CAMELLIA_MAX_KEY_SIZE,
+			.ivsize		= CAMELLIA_BLOCK_SIZE,
 			.setkey		= camellia_set_key,
 			.encrypt	= cbc_encrypt,
 			.decrypt	= cbc_decrypt,

@@ -429,6 +429,7 @@ static struct crypto_alg algs[] = { {
 		.blkcipher = {
 			.min_keysize	= DES_KEY_SIZE,
 			.max_keysize	= DES_KEY_SIZE,
+			.ivsize		= DES_BLOCK_SIZE,
 			.setkey		= des_set_key,
 			.encrypt	= cbc_encrypt,
 			.decrypt	= cbc_decrypt,

@@ -485,6 +486,7 @@ static struct crypto_alg algs[] = { {
 		.blkcipher = {
 			.min_keysize	= DES3_EDE_KEY_SIZE,
 			.max_keysize	= DES3_EDE_KEY_SIZE,
+			.ivsize		= DES3_EDE_BLOCK_SIZE,
 			.setkey		= des3_ede_set_key,
 			.encrypt	= cbc3_encrypt,
 			.decrypt	= cbc3_decrypt,

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

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

From: Russell King <[email protected]>

commit 8996eafdcbad149ac0f772fb1649fbb75c482a6a upstream.

Unlike shash algorithms, ahash drivers must implement export and import as their descriptors may contain hardware state and cannot be exported as is. Unfortunately some ahash drivers did not provide them and end up causing crashes with algif_hash.

This patch adds a check to prevent these drivers from registering ahash algorithms until they are fixed.

---
 crypto/ahash.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -544,7 +544,8 @@ static int ahash_prepare_alg(struct ahas
 	struct crypto_alg *base = &alg->halg.base;
 
 	if (alg->halg.digestsize > PAGE_SIZE / 8 ||
-	    alg->halg.statesize > PAGE_SIZE / 8)
+	    alg->halg.statesize > PAGE_SIZE / 8 ||
+	    alg->halg.statesize == 0)
 		return -EINVAL;
 
 	base->cra_type = &crypto_ahash_type;

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

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

From: Shaohua Li <[email protected]>

commit 424cdc14138088ada1b0e407a2195b2783c6e5ef upstream.

page_counter_memparse() returns pages for the threshold, while mem_cgroup_usage() returns bytes for memory usage. Convert the threshold to bytes.

---
 mm/memcontrol.c |    1 +
 1 file changed, 1 insertion(+)

@@ -3687,6 +3687,7 @@ static int __mem_cgroup_usage_register_e
 	ret = page_counter_memparse(args, "-1", &threshold);
 	if (ret)
 		return ret;
+	threshold <<= PAGE_SHIFT;
 
 	mutex_lock(&memcg->thresholds_lock);
 

mem_cgroup_usage() returns a u64 and I think that the types of threshold and mem_cgroup_threshold::threshold also need be changed to u64 to avoid overflow on large 32-bit systems.

Ben.

Ben Hutchings
Beware of bugs in the above code;
I have only proved it correct, not tried it. - Donald Knuth

On Sat 24-10-15 14:46:58, Ben Hutchings wrote:
[...]

You are absolutely right! I have compltely missed that 3e32cb2e0a12 has changed the type as well. Should have noticed that during the review.

The whole thing is just way too confusing. All the tracking is done in page units yet tresholds are in bytes. This only calls for troubles. The patch below simply turns thresholds to page units as well. I hope I haven't screwed anything, I didn't get to more than compile test it.
---

From: Michal Hocko <[email protected]>
Date: Sun, 25 Oct 2015 17:12:44 +0100
Subject: [PATCH] memcg: Fix thresholds for 32b architectures.

424cdc141380 ("memcg: convert threshold to bytes") has fixed a regression introduced by 3e32cb2e0a12 ("mm: memcontrol: lockless page counters") where thresholds were silently converted to use page units rather than bytes when interpreting the user input.

The fix is not complete, though, as properly pointed out by Ben Hutchings during stable backport review. The page count is converted to bytes but unsigned long is used to hold the value which would be obviously not sufficient for 32b systems with more than 4G thresholds. The same applies to usage as taken from mem_cgroup_usage which might overflow.

Let's remove this bytes vs. pages internal tracking differences and handle thresholds in page units internally. Chage mem_cgroup_usage() to return the value in page units and revert 424cdc141380 because this should be sufficient for the consistent handling. mem_cgroup_read_u64 as the only users of mem_cgroup_usage outside of the threshold handling code is converted to give the proper in bytes result. It is doing that already for page_counter output so this is more consistent as well.

The value presented to the userspace is still in bytes units.

---
 mm/memcontrol.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

index f3cc594ffa2d..2823cafc269e 100644
@@ -2802,7 +2802,7 @@ static unsigned long tree_stat(struct mem_cgroup *memcg,
 	return val;
 }
 
-static inline u64 mem_cgroup_usage(struct mem_cgroup *memcg, bool swap)
+static inline unsigned long mem_cgroup_usage(struct mem_cgroup *memcg, bool swap)
 {
 	u64 val;
 

@@ -2817,7 +2817,7 @@ static inline u64 mem_cgroup_usage(struct mem_cgroup *memcg, bool swap)
 		else
 			val = page_counter_read(&memcg->memsw);
 	}
-	return val << PAGE_SHIFT;
+	return val;
 }
 
 enum {

@@ -2851,9 +2851,9 @@ static u64 mem_cgroup_read_u64(struct cgroup_subsys_state *css,
 	switch (MEMFILE_ATTR(cft->private)) {
 	case RES_USAGE:
 		if (counter == &memcg->memory)
-			return mem_cgroup_usage(memcg, false);
+			return (u64)mem_cgroup_usage(memcg, false) * PAGE_SIZE;
 		if (counter == &memcg->memsw)
-			return mem_cgroup_usage(memcg, true);
+			return (u64)mem_cgroup_usage(memcg, true) * PAGE_SIZE;
 		return (u64)page_counter_read(counter) * PAGE_SIZE;
 	case RES_LIMIT:
 		return (u64)counter->limit * PAGE_SIZE;

@@ -3353,7 +3353,6 @@ static int __mem_cgroup_usage_register_event(struct mem_cgroup *memcg,
 	ret = page_counter_memparse(args, "-1", &threshold);
 	if (ret)
 		return ret;
-	threshold <<= PAGE_SHIFT;
 
 	mutex_lock(&memcg->thresholds_lock);
 

2.6.1

-- 
Michal Hocko
SUSE Labs

OK, I've tested it and it seems working as expected. I will repost to the linux-mm mailing list.

Michal Hocko
SUSE Labs

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

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

From: David Sterba <[email protected]>

commit 8eb934591f8bf584969454a658f629cd06e59f3a upstream.

We don't verify that all the balance filter arguments supplemented by the flags are actually known to the kernel. Thus we let it silently pass and do nothing.

At the moment this means only the 'limit' filter, but we're going to add a few more soon so it's better to have that fixed. Also in older stable kernels so that it works with newer userspace tools.

---
 fs/btrfs/ioctl.c   |    5 +++++
 fs/btrfs/volumes.h |    8 ++++++++
 2 files changed, 13 insertions(+)

@@ -4647,6 +4647,11 @@ locked:
 		bctl->flags |= BTRFS_BALANCE_TYPE_MASK;
 	}
 
+	if (bctl->flags & ~(BTRFS_BALANCE_ARGS_MASK | BTRFS_BALANCE_TYPE_MASK)) {
+		ret = -EINVAL;
+		goto out_bargs;
+	}
+
 do_balance:
 	/*
 	 * Ownership of bctl and mutually_exclusive_operation_running

@@ -376,6 +376,14 @@ struct map_lookup {
 #define BTRFS_BALANCE_ARGS_VRANGE	(1ULL << 4)
 #define BTRFS_BALANCE_ARGS_LIMIT	(1ULL << 5)
 
+#define BTRFS_BALANCE_ARGS_MASK			\
+	(BTRFS_BALANCE_ARGS_PROFILES |		\
+	 BTRFS_BALANCE_ARGS_USAGE |		\
+	 BTRFS_BALANCE_ARGS_DEVID | 		\
+	 BTRFS_BALANCE_ARGS_DRANGE |		\
+	 BTRFS_BALANCE_ARGS_VRANGE |		\
+	 BTRFS_BALANCE_ARGS_LIMIT)
+
 /*
  * Profile changing flags.  When SOFT is set we won't relocate chunk if
  * it already has the target profile (even though it may be

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

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

From: Chris Mason <[email protected]>

commit dc6c5fb3b514221f2e9d21ee626a9d95d3418dff upstream.

The code for btrfs inode-resolve has never worked properly for files with enough hard links to trigger extrefs. It was trying to get the leaf out of a path after freeing the path:

	btrfs_release_path(path);
	leaf = path->nodes[0];
	item_size = btrfs_item_size_nr(leaf, slot);

The fix here is to use the extent buffer we cloned just a little higher up to avoid deadlocks caused by using the leaf in the path.

---
 fs/btrfs/backref.c |    8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

@@ -1809,7 +1809,6 @@ static int iterate_inode_extrefs(u64 inu
 	int found = 0;
 	struct extent_buffer *eb;
 	struct btrfs_inode_extref *extref;
-	struct extent_buffer *leaf;
 	u32 item_size;
 	u32 cur_offset;
 	unsigned long ptr;

@@ -1837,9 +1836,8 @@ static int iterate_inode_extrefs(u64 inu
 		btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
 		btrfs_release_path(path);
 
-		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, slot);
-		ptr = btrfs_item_ptr_offset(leaf, slot);
+		item_size = btrfs_item_size_nr(eb, slot);
+		ptr = btrfs_item_ptr_offset(eb, slot);
 		cur_offset = 0;
 
 		while (cur_offset < item_size) {

@@ -1853,7 +1851,7 @@ static int iterate_inode_extrefs(u64 inu
 			if (ret)
 				break;
 
-			cur_offset += btrfs_inode_extref_name_len(leaf, extref);
+			cur_offset += btrfs_inode_extref_name_len(eb, extref);
 			cur_offset += sizeof(*extref);
 		}
 		btrfs_tree_read_unlock_blocking(eb);

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

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

From: Will Deacon <[email protected]>

commit b6dd8e0719c0d2d01429639a11b7bc2677de240c upstream.

Commit df057cc7b4fa ("arm64: errata: add module build workaround for erratum #843419") sets CFLAGS_MODULE to ensure that the large memory model is used by the compiler when building kernel modules.

However, CFLAGS_MODULE is an environment variable and intended to be overridden on the command line, which appears to be the case with the Ubuntu kernel packaging system, so use KBUILD_CFLAGS_MODULE instead.

---
 arch/arm64/Makefile |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -31,7 +31,7 @@ endif
 CHECKFLAGS	+= -D__aarch64__
 
 ifeq ($(CONFIG_ARM64_ERRATUM_843419), y)
-CFLAGS_MODULE	+= -mcmodel=large
+KBUILD_CFLAGS_MODULE	+= -mcmodel=large
 endif
 
 # Default value

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

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

From: Linus Walleij <[email protected]>

commit c00def71efd919e8ae835a25f4f4c80a4b2d36d3 upstream.

This removes a lot of ancient cruft from the Ux500 SMP boot. Instead of the pen grab/release, just point the ROM to secondary_boot() and start the second CPU there, then send the IPI.

Use our own SMP enable method. This enables us to remove the last static mapping and get both CPUs booting properly.

Tested this and it just works.

---
 arch/arm/mach-ux500/Makefile     |    2 
 arch/arm/mach-ux500/cpu-db8500.c |    1 
 arch/arm/mach-ux500/headsmp.S    |   37 ----------
 arch/arm/mach-ux500/platsmp.c    |  132 +++++++++++----------------------------
 arch/arm/mach-ux500/setup.h      |    1 
 5 files changed, 40 insertions(+), 133 deletions(-)

@@ -7,7 +7,7 @@ obj-$(CONFIG_CACHE_L2X0)	+= cache-l2x0.o
 obj-$(CONFIG_UX500_SOC_DB8500)	+= cpu-db8500.o
 obj-$(CONFIG_MACH_MOP500)	+= board-mop500-regulators.o \
 				board-mop500-audio.o
-obj-$(CONFIG_SMP)		+= platsmp.o headsmp.o
+obj-$(CONFIG_SMP)		+= platsmp.o
 obj-$(CONFIG_HOTPLUG_CPU)	+= hotplug.o
 obj-$(CONFIG_PM_GENERIC_DOMAINS) += pm_domains.o
 

@@ -154,7 +154,6 @@ static const char * stericsson_dt_platfo
 };
 
 DT_MACHINE_START(U8500_DT, "ST-Ericsson Ux5x0 platform (Device Tree Support)")
-	.smp            = smp_ops(ux500_smp_ops),
 	.map_io		= u8500_map_io,
 	.init_irq	= ux500_init_irq,
 	/* we re-use nomadik timer here */

@@ -1,37 +0,0 @@
-/*
- *  Copyright (c) 2009 ST-Ericsson
- *	This file is based  ARM Realview platform
- *  Copyright (c) 2003 ARM Limited
- *  All Rights Reserved
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <linux/linkage.h>
-#include <linux/init.h>
-
-/*
- * U8500 specific entry point for secondary CPUs.
- */
-ENTRY(u8500_secondary_startup)
-	mrc	p15, 0, r0, c0, c0, 5
-	and	r0, r0, #15
-	adr	r4, 1f
-	ldmia	r4, {r5, r6}
-	sub	r4, r4, r5
-	add	r6, r6, r4
-pen:	ldr	r7, [r6]
-	cmp	r7, r0
-	bne	pen
-
-	/*
-	 * we've been released from the holding pen: secondary_stack
-	 * should now contain the SVC stack for this core
-	 */
-	b	secondary_startup
-ENDPROC(u8500_secondary_startup)
-
-	.align 2
-1:	.long	.
-	.long	pen_release

@@ -26,7 +26,6 @@ extern struct device *ux500_soc_device_i
 
 extern void ux500_timer_init(void);
 
-extern struct smp_operations ux500_smp_ops;
 extern void ux500_cpu_die(unsigned int cpu);
 
 #endif /*  __ASM_ARCH_SETUP_H */

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

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

From: Christoph Hellwig <[email protected]>

commit 8c3ad9cb7343dc5f61b8cf3cdbe1016c5e7c2c8b upstream.

Recent Linux clients have started to send GETLAYOUT requests with minlength less than blocksize.

Servers aren't really allowed to impose this kind of restriction on layouts; see RFC 5661 section 18.43.3 for details.

This has been observed to cause indefinite hangs on fsx runs on some clients.

---
 fs/nfsd/blocklayout.c |    8 --------
 1 file changed, 8 deletions(-)

@@ -56,14 +56,6 @@ nfsd4_block_proc_layoutget(struct inode
 	u32 device_generation = 0;
 	int error;
 
-	/*
-	 * We do not attempt to support I/O smaller than the fs block size,
-	 * or not aligned to it.
-	 */
-	if (args->lg_minlength < block_size) {
-		dprintk("pnfsd: I/O too small\n");
-		goto out_layoutunavailable;
-	}
 	if (seg->offset & (block_size - 1)) {
 		dprintk("pnfsd: I/O misaligned\n");
 		goto out_layoutunavailable;

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

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

From: Javier Martinez Canillas <[email protected]>

commit 1b52e50f2a402a266f1ba2281f0a57e87637a047 upstream.

If i2c_new_dummy() fails in max77843_chg_init(), an PTR_ERR(NULL) is returned which is 0. So the function was wrongly returning a success value instead of an error code.

---
 drivers/mfd/max77843.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -79,7 +79,7 @@ static int max77843_chg_init(struct max7
 	if (!max77843->i2c_chg) {
 		dev_err(&max77843->i2c->dev,
 				"Cannot allocate I2C device for Charger\n");
-		return PTR_ERR(max77843->i2c_chg);
+		return -ENODEV;
 	}
 	i2c_set_clientdata(max77843->i2c_chg, max77843);
 

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

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

From: Wolfram Sang <[email protected]>

commit 4f7effddf4549d57114289f273710f077c4c330a upstream.

The core may register clients attached to this master which may use funtionality from the master. So, RuntimePM must be enabled before, otherwise this will fail. While here, move drvdata, too.

---
 drivers/i2c/busses/i2c-rcar.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

@@ -690,15 +690,16 @@ static int rcar_i2c_probe(struct platfor
 		return ret;
 	}
 
+	pm_runtime_enable(dev);
+	platform_set_drvdata(pdev, priv);
+
 	ret = i2c_add_numbered_adapter(adap);
 	if (ret < 0) {
 		dev_err(dev, "reg adap failed: %d\n", ret);
+		pm_runtime_disable(dev);
 		return ret;
 	}
 
-	pm_runtime_enable(dev);
-	platform_set_drvdata(pdev, priv);
-
 	dev_info(dev, "probed\n");
 
 	return 0;

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

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

From: Wolfram Sang <[email protected]>

commit eadd709f5d2e8aebb1b7bf49460e97a68d81a9b0 upstream.

The core may register clients attached to this master which may use funtionality from the master. So, RuntimePM must be enabled before, otherwise this will fail. While here, move drvdata, too.

---
 drivers/i2c/busses/i2c-s3c2410.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

@@ -1243,17 +1243,19 @@ static int s3c24xx_i2c_probe(struct plat
 	i2c->adap.nr = i2c->pdata->bus_num;
 	i2c->adap.dev.of_node = pdev->dev.of_node;
 
+	platform_set_drvdata(pdev, i2c);
+
+	pm_runtime_enable(&pdev->dev);
+
 	ret = i2c_add_numbered_adapter(&i2c->adap);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "failed to add bus to i2c core\n");
+		pm_runtime_disable(&pdev->dev);
 		s3c24xx_i2c_deregister_cpufreq(i2c);
 		clk_unprepare(i2c->clk);
 		return ret;
 	}
 
-	platform_set_drvdata(pdev, i2c);
-
-	pm_runtime_enable(&pdev->dev);
 	pm_runtime_enable(&i2c->adap.dev);
 
 	dev_info(&pdev->dev, "%s: S3C I2C adapter\n", dev_name(&i2c->adap.dev));

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

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

From: Mika Westerberg <[email protected]>

commit 56d4b8a24cef5d66f0d10ac778a520d3c2c68a48 upstream.

ACPI SSCN/FMCN methods were originally added because then the platform can provide the most accurate HCNT/LCNT values to the driver. However, this seems not to be true for Dell Inspiron 7348 where using these causes the touchpad to fail in boot:

i2c_hid i2c-DLL0675:00: failed to retrieve report from device. i2c_designware INT3433:00: i2c_dw_handle_tx_abort: lost arbitration i2c_hid i2c-DLL0675:00: failed to retrieve report from device. i2c_designware INT3433:00: controller timed out

The values received from ACPI are (in fast mode):

HCNT: 72
LCNT: 160

this translates to following timings (input clock is 100MHz on Broadwell):

tHIGH: 720 ns (spec min 600 ns)
tLOW: 1600 ns (spec min 1300 ns)
Bus period: 2920 ns (assuming 300 ns tf and tr)
Bus speed: 342.5 kHz

Both tHIGH and tLOW are within the I2C specification.

The calculated values when ACPI parameters are not used are (in fast mode):

HCNT: 87
LCNT: 159

which translates to:

tHIGH: 870 ns (spec min 600 ns)
tLOW: 1590 ns (spec min 1300 ns)
Bus period 3060 ns (assuming 300 ns tf and tr)
Bus speed 326.8 kHz

These values are also within the I2C specification.

Since both ACPI and calculated values meet the I2C specification timing requirements it is hard to say why the touchpad does not function properly with the ACPI values except that the bus speed is higher in this case (but still well below the max 400kHz).

Solve this by adding DMI quirk to the driver that disables using ACPI parameters on this particulare machine.

---
 drivers/i2c/busses/i2c-designware-platdrv.c |   20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

@@ -24,6 +24,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/delay.h>
+#include <linux/dmi.h>
 #include <linux/i2c.h>
 #include <linux/clk.h>
 #include <linux/clk-provider.h>

@@ -51,6 +52,22 @@ static u32 i2c_dw_get_clk_rate_khz(struc
 }
 
 #ifdef CONFIG_ACPI
+/*
+ * The HCNT/LCNT information coming from ACPI should be the most accurate
+ * for given platform. However, some systems get it wrong. On such systems
+ * we get better results by calculating those based on the input clock.
+ */
+static const struct dmi_system_id dw_i2c_no_acpi_params[] = {
+	{
+		.ident = "Dell Inspiron 7348",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Inspiron 7348"),
+		},
+	},
+	{ }
+};
+
 static void dw_i2c_acpi_params(struct platform_device *pdev, char method[],
 			       u16 *hcnt, u16 *lcnt, u32 *sda_hold)
 {

@@ -58,6 +75,9 @@ static void dw_i2c_acpi_params(struct pl
 	acpi_handle handle = ACPI_HANDLE(&pdev->dev);
 	union acpi_object *obj;
 
+	if (dmi_check_system(dw_i2c_no_acpi_params))
+		return;
+
 	if (ACPI_FAILURE(acpi_evaluate_object(handle, method, NULL, &buf)))
 		return;
 

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

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

From: Wolfram Sang <[email protected]>

commit 36d48fb5766aee9717e429f772046696b215282d upstream.

The core may register clients attached to this master which may use funtionality from the master. So, RuntimePM must be enabled before, otherwise this will fail.

---
 drivers/i2c/busses/i2c-designware-platdrv.c |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

@@ -273,12 +273,6 @@ static int dw_i2c_probe(struct platform_
 	adap->dev.parent = &pdev->dev;
 	adap->dev.of_node = pdev->dev.of_node;
 
-	r = i2c_add_numbered_adapter(adap);
-	if (r) {
-		dev_err(&pdev->dev, "failure adding adapter\n");
-		return r;
-	}
-
 	if (dev->pm_runtime_disabled) {
 		pm_runtime_forbid(&pdev->dev);
 	} else {

@@ -288,6 +282,13 @@ static int dw_i2c_probe(struct platform_
 		pm_runtime_enable(&pdev->dev);
 	}
 
+	r = i2c_add_numbered_adapter(adap);
+	if (r) {
+		dev_err(&pdev->dev, "failure adding adapter\n");
+		pm_runtime_disable(&pdev->dev);
+		return r;
+	}
+
 	return 0;
 }
 

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

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

From: Shaohua Li <[email protected]>

commit 874bbfe600a660cba9c776b3957b1ce393151b76 upstream.

My system keeps crashing with below message. vmstat_update() schedules a delayed work in current cpu and expects the work runs in the cpu. schedule_delayed_work() is expected to make delayed work run in local cpu. The problem is timer can be migrated with NO_HZ. __queue_work() queues work in timer handler, which could run in a different cpu other than where the delayed work is scheduled. The end result is the delayed work runs in different cpu. The patch makes __queue_delayed_work records local cpu earlier. Where the timer runs doesn't change where the work runs with the change.

[   28.010131] ------------[ cut here ]------------
[   28.010609] kernel BUG at ../mm/vmstat.c:1392!
[   28.011099] invalid opcode: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC KASAN
[   28.011860] Modules linked in:
[   28.012245] CPU: 0 PID: 289 Comm: kworker/0:3 Tainted: G        W4.3.0-rc3+ #634
[   28.013065] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140709_153802- 04/01/2014
[   28.014160] Workqueue: events vmstat_update
[   28.014571] task: ffff880117682580 ti: ffff8800ba428000 task.ti: ffff8800ba428000
[   28.015445] RIP: 0010:[<ffffffff8115f921>]  [<ffffffff8115f921>]vmstat_update+0x31/0x80
[   28.016282] RSP: 0018:ffff8800ba42fd80  EFLAGS: 00010297
[   28.016812] RAX: 0000000000000000 RBX: ffff88011a858dc0 RCX:0000000000000000
[   28.017585] RDX: ffff880117682580 RSI: ffffffff81f14d8c RDI:ffffffff81f4df8d
[   28.018366] RBP: ffff8800ba42fd90 R08: 0000000000000001 R09:0000000000000000
[   28.019169] R10: 0000000000000000 R11: 0000000000000121 R12:ffff8800baa9f640
[   28.019947] R13: ffff88011a81e340 R14: ffff88011a823700 R15:0000000000000000
[   28.020071] FS:  0000000000000000(0000) GS:ffff88011a800000(0000)knlGS:0000000000000000
[   28.020071] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[   28.020071] CR2: 00007ff6144b01d0 CR3: 00000000b8e93000 CR4:00000000000006f0
[   28.020071] Stack:
[   28.020071]  ffff88011a858dc0 ffff8800baa9f640 ffff8800ba42fe00ffffffff8106bd88
[   28.020071]  ffffffff8106bd0b 0000000000000096 0000000000000000ffffffff82f9b1e8
[   28.020071]  ffffffff829f0b10 0000000000000000 ffffffff81f18460ffff88011a81e340
[   28.020071] Call Trace:
[   28.020071]  [<ffffffff8106bd88>] process_one_work+0x1c8/0x540
[   28.020071]  [<ffffffff8106bd0b>] ? process_one_work+0x14b/0x540
[   28.020071]  [<ffffffff8106c214>] worker_thread+0x114/0x460
[   28.020071]  [<ffffffff8106c100>] ? process_one_work+0x540/0x540
[   28.020071]  [<ffffffff81071bf8>] kthread+0xf8/0x110
[   28.020071]  [<ffffffff81071b00>] ?kthread_create_on_node+0x200/0x200
[   28.020071]  [<ffffffff81a6522f>] ret_from_fork+0x3f/0x70
[   28.020071]  [<ffffffff81071b00>] ?kthread_create_on_node+0x200/0x200

---
 kernel/workqueue.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -1458,13 +1458,13 @@ static void __queue_delayed_work(int cpu
 	timer_stats_timer_set_start_info(&dwork->timer);
 
 	dwork->wq = wq;
+	/* timer isn't guaranteed to run in this cpu, record earlier */
+	if (cpu == WORK_CPU_UNBOUND)
+		cpu = raw_smp_processor_id();
 	dwork->cpu = cpu;
 	timer->expires = jiffies + delay;
 
-	if (unlikely(cpu != WORK_CPU_UNBOUND))
-		add_timer_on(timer, cpu);
-	else
-		add_timer(timer);
+	add_timer_on(timer, cpu);
 }
 
 /**

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

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

From: Paolo Bonzini <[email protected]>

commit 73917739334c6509833b0403b81d4a04a8784bdf upstream.

An SMI to a halted VCPU must wake it up, hence a VCPU with a pending SMI must be considered runnable.

---
 arch/x86/kvm/x86.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -8030,6 +8030,9 @@ static inline bool kvm_vcpu_has_events(s
 	if (atomic_read(&vcpu->arch.nmi_queued))
 		return true;
 
+	if (test_bit(KVM_REQ_SMI, &vcpu->requests))
+		return true;
+
 	if (kvm_arch_interrupt_allowed(vcpu) &&
 	    kvm_cpu_has_interrupt(vcpu))
 		return true;

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

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

From: Paolo Bonzini <[email protected]>

commit b10d92a54dac25a6152f1aa1ffc95c12908035ce upstream.

In order to get into 64-bit protected mode, you need to enable paging while EFER.LMA=1. For this to work, CS.L must be 0. Currently, we load the segments before CR0 and CR4, which means that if RSM returns into 64-bit protected mode CS.L is already 1 and everything breaks.

Luckily, CS.L=0 is always the case when executing RSM, because it is forbidden to execute RSM from 64-bit protected mode. Hence it is enough to load CR0 and CR4 first, and only then the segments.

---
 arch/x86/kvm/emulate.c |   10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

@@ -2418,7 +2418,7 @@ static int rsm_load_state_64(struct x86_
 	u64 val, cr0, cr4;
 	u32 base3;
 	u16 selector;
-	int i;
+	int i, r;
 
 	for (i = 0; i < 16; i++)
 		*reg_write(ctxt, i) = GET_SMSTATE(u64, smbase, 0x7ff8 - i * 8);

@@ -2460,13 +2460,17 @@ static int rsm_load_state_64(struct x86_
 	dt.address =                GET_SMSTATE(u64, smbase, 0x7e68);
 	ctxt->ops->set_gdt(ctxt, &dt);
 
+	r = rsm_enter_protected_mode(ctxt, cr0, cr4);
+	if (r != X86EMUL_CONTINUE)
+		return r;
+
 	for (i = 0; i < 6; i++) {
-		int r = rsm_load_seg_64(ctxt, smbase, i);
+		r = rsm_load_seg_64(ctxt, smbase, i);
 		if (r != X86EMUL_CONTINUE)
 			return r;
 	}
 
-	return rsm_enter_protected_mode(ctxt, cr0, cr4);
+	return X86EMUL_CONTINUE;
 }
 
 static int em_rsm(struct x86_emulate_ctxt *ctxt)

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

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

From: Gerd Hoffmann <[email protected]>

commit 60906529bec506a4bde93f04eb5e7a9dd9c42a3b upstream.

Commit "c0fe07a drm/qxl: rewrite framebuffer support" has a bug in the dirty rectangle tracking: Instead of ignoring an empty dirty rectangle when adding a new dirty region the dirty region gets extended to the upper left corner. Fix it.

---
 drivers/gpu/drm/qxl/qxl_fb.c |   19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

@@ -144,14 +144,17 @@ static void qxl_dirty_update(struct qxl_
 
 	spin_lock_irqsave(&qfbdev->dirty.lock, flags);
 
-	if (qfbdev->dirty.y1 < y)
-		y = qfbdev->dirty.y1;
-	if (qfbdev->dirty.y2 > y2)
-		y2 = qfbdev->dirty.y2;
-	if (qfbdev->dirty.x1 < x)
-		x = qfbdev->dirty.x1;
-	if (qfbdev->dirty.x2 > x2)
-		x2 = qfbdev->dirty.x2;
+	if ((qfbdev->dirty.y2 - qfbdev->dirty.y1) &&
+	    (qfbdev->dirty.x2 - qfbdev->dirty.x1)) {
+		if (qfbdev->dirty.y1 < y)
+			y = qfbdev->dirty.y1;
+		if (qfbdev->dirty.y2 > y2)
+			y2 = qfbdev->dirty.y2;
+		if (qfbdev->dirty.x1 < x)
+			x = qfbdev->dirty.x1;
+		if (qfbdev->dirty.x2 > x2)
+			x2 = qfbdev->dirty.x2;
+	}
 
 	qfbdev->dirty.x1 = x;
 	qfbdev->dirty.x2 = x2;

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

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

From: Ben Skeggs <[email protected]>

commit f231976c2e8964ceaa9250e57d27c35ff03825c2 upstream.

We need to do this in order to prevent accesses to the device while it's powered down. Userspace may have an mmap of the fb, and there's no good way (that I know of) to prevent it from touching the device otherwise.

This fixes some nasty races between runpm and plymouth on some systems, which result in the GPU getting very upset and hanging the boot.

---
 drivers/gpu/drm/nouveau/nouveau_fbcon.c |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

@@ -178,8 +178,30 @@ nouveau_fbcon_sync(struct fb_info *info)
 	return 0;
 }
 
+static int
+nouveau_fbcon_open(struct fb_info *info, int user)
+{
+	struct nouveau_fbdev *fbcon = info->par;
+	struct nouveau_drm *drm = nouveau_drm(fbcon->dev);
+	int ret = pm_runtime_get_sync(drm->dev->dev);
+	if (ret < 0 && ret != -EACCES)
+		return ret;
+	return 0;
+}
+
+static int
+nouveau_fbcon_release(struct fb_info *info, int user)
+{
+	struct nouveau_fbdev *fbcon = info->par;
+	struct nouveau_drm *drm = nouveau_drm(fbcon->dev);
+	pm_runtime_put(drm->dev->dev);
+	return 0;
+}
+
 static struct fb_ops nouveau_fbcon_ops = {
 	.owner = THIS_MODULE,
+	.fb_open = nouveau_fbcon_open,
+	.fb_release = nouveau_fbcon_release,
 	.fb_check_var = drm_fb_helper_check_var,
 	.fb_set_par = drm_fb_helper_set_par,
 	.fb_fillrect = nouveau_fbcon_fillrect,

@@ -195,6 +217,8 @@ static struct fb_ops nouveau_fbcon_ops =
 
 static struct fb_ops nouveau_fbcon_sw_ops = {
 	.owner = THIS_MODULE,
+	.fb_open = nouveau_fbcon_open,
+	.fb_release = nouveau_fbcon_release,
 	.fb_check_var = drm_fb_helper_check_var,
 	.fb_set_par = drm_fb_helper_set_par,
 	.fb_fillrect = cfb_fillrect,

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

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

From: Dave Airlie <[email protected]>

commit ae491542cbbbcca0ec8938c37d4079a985e58440 upstream.

This zeroes the msg so no random stack data ends up getting sent, it also limits the function to not accepting > 4 i2c msgs.

---
 drivers/gpu/drm/drm_dp_mst_topology.c |    3 ++-
 include/drm/drm_dp_mst_helper.h       |    3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

@@ -2789,12 +2789,13 @@ static int drm_dp_mst_i2c_xfer(struct i2
 	if (msgs[num - 1].flags & I2C_M_RD)
 		reading = true;
 
-	if (!reading) {
+	if (!reading || (num - 1 > DP_REMOTE_I2C_READ_MAX_TRANSACTIONS)) {
 		DRM_DEBUG_KMS("Unsupported I2C transaction for MST device\n");
 		ret = -EIO;
 		goto out;
 	}
 
+	memset(&msg, 0, sizeof(msg));
 	msg.req_type = DP_REMOTE_I2C_READ;
 	msg.u.i2c_read.num_transactions = num - 1;
 	msg.u.i2c_read.port_number = port->port_num;

@@ -253,6 +253,7 @@ struct drm_dp_remote_dpcd_write {
 	u8 *bytes;
 };
 
+#define DP_REMOTE_I2C_READ_MAX_TRANSACTIONS 4
 struct drm_dp_remote_i2c_read {
 	u8 num_transactions;
 	u8 port_number;

@@ -262,7 +263,7 @@ struct drm_dp_remote_i2c_read {
 		u8 *bytes;
 		u8 no_stop_bit;
 		u8 i2c_transaction_delay;
-	} transactions[4];
+	} transactions[DP_REMOTE_I2C_READ_MAX_TRANSACTIONS];
 	u8 read_i2c_device_id;
 	u8 num_bytes_read;
 };

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

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

From: Dave Airlie <[email protected]>

commit bc8c131ccdd62d4ed4f33c6b50f92907e7c32dee upstream.

This allows tiled monitors to work with radeon once mst is enabled.

---
 drivers/gpu/drm/radeon/radeon_dp_mst.c |    1 +
 1 file changed, 1 insertion(+)

@@ -283,6 +283,7 @@ static struct drm_connector *radeon_dp_a
 	radeon_connector->mst_encoder = radeon_dp_create_fake_mst_encoder(master);
 
 	drm_object_attach_property(&connector->base, dev->mode_config.path_property, 0);
+	drm_object_attach_property(&connector->base, dev->mode_config.tile_property, 0);
 	drm_mode_connector_set_path_property(connector, pathprop);
 	drm_reinit_primary_mode_group(dev);
 

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

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

From: Alex Deucher <[email protected]>

commit 51a4726b04e880fdd9b4e0e58b13f70b0a68a7f5 upstream.

They were added relatively early in the driver init process which meant that in some cases the driver was not finished initializing before external tools tried to use them which could result in a crash depending on the timing.

---
 drivers/gpu/drm/radeon/radeon_display.c |   14 +------
 drivers/gpu/drm/radeon/radeon_pm.c      |   63 +++++++++++++++++++-------------
 2 files changed, 40 insertions(+), 37 deletions(-)

@@ -1633,18 +1633,8 @@ int radeon_modeset_init(struct radeon_de
 	radeon_fbdev_init(rdev);
 	drm_kms_helper_poll_init(rdev->ddev);
 
-	if (rdev->pm.dpm_enabled) {
-		/* do dpm late init */
-		ret = radeon_pm_late_init(rdev);
-		if (ret) {
-			rdev->pm.dpm_enabled = false;
-			DRM_ERROR("radeon_pm_late_init failed, disabling dpm\n");
-		}
-		/* set the dpm state for PX since there won't be
-		 * a modeset to call this.
-		 */
-		radeon_pm_compute_clocks(rdev);
-	}
+	/* do pm late init */
+	ret = radeon_pm_late_init(rdev);
 
 	return 0;
 }

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

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

From: Alex Deucher <[email protected]>

commit fa022a9b65d2886486a022fd66b20c823cd76ad9 upstream.

They were added relatively early in the driver init process which meant that in some cases the driver was not finished initializing before external tools tried to use them which could result in a crash depending on the timing.

---
 drivers/gpu/drm/amd/amdgpu/ci_dpm.c |    8 +++++---
 drivers/gpu/drm/amd/amdgpu/cz_dpm.c |   10 ++++++----
 drivers/gpu/drm/amd/amdgpu/kv_dpm.c |    9 ++++++---
 3 files changed, 17 insertions(+), 10 deletions(-)

@@ -6185,6 +6185,11 @@ static int ci_dpm_late_init(void *handle
 	if (!amdgpu_dpm)
 		return 0;
 
+	/* init the sysfs and debugfs files late */
+	ret = amdgpu_pm_sysfs_init(adev);
+	if (ret)
+		return ret;
+
 	ret = ci_set_temperature_range(adev);
 	if (ret)
 		return ret;

@@ -6232,9 +6237,6 @@ static int ci_dpm_sw_init(void *handle)
 	adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps;
 	if (amdgpu_dpm == 1)
 		amdgpu_pm_print_power_states(adev);
-	ret = amdgpu_pm_sysfs_init(adev);
-	if (ret)
-		goto dpm_failed;
 	mutex_unlock(&adev->pm.mutex);
 	DRM_INFO("amdgpu: dpm initialized\n");
 

@@ -596,6 +596,12 @@ static int cz_dpm_late_init(void *handle
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	if (amdgpu_dpm) {
+		int ret;
+		/* init the sysfs and debugfs files late */
+		ret = amdgpu_pm_sysfs_init(adev);
+		if (ret)
+			return ret;
+
 		/* powerdown unused blocks for now */
 		cz_dpm_powergate_uvd(adev, true);
 		cz_dpm_powergate_vce(adev, true);

@@ -632,10 +638,6 @@ static int cz_dpm_sw_init(void *handle)
 	if (amdgpu_dpm == 1)
 		amdgpu_pm_print_power_states(adev);
 
-	ret = amdgpu_pm_sysfs_init(adev);
-	if (ret)
-		goto dpm_init_failed;
-
 	mutex_unlock(&adev->pm.mutex);
 	DRM_INFO("amdgpu: dpm initialized\n");
 

@@ -2995,6 +2995,12 @@ static int kv_dpm_late_init(void *handle
 {
 	/* powerdown unused blocks for now */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
+
+	/* init the sysfs and debugfs files late */
+	ret = amdgpu_pm_sysfs_init(adev);
+	if (ret)
+		return ret;
 
 	kv_dpm_powergate_acp(adev, true);
 	kv_dpm_powergate_samu(adev, true);

@@ -3038,9 +3044,6 @@ static int kv_dpm_sw_init(void *handle)
 	adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps;
 	if (amdgpu_dpm == 1)
 		amdgpu_pm_print_power_states(adev);
-	ret = amdgpu_pm_sysfs_init(adev);
-	if (ret)
-		goto dpm_failed;
 	mutex_unlock(&adev->pm.mutex);
 	DRM_INFO("amdgpu: dpm initialized\n");
 

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

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

From: Alex Deucher <[email protected]>

commit 26d0c21db1fd1679fcf07863741c13ba5ce37a65 upstream.

Hw only has 3 crtcs. copy paste typo.

---
 drivers/gpu/drm/amd/amdgpu/dce_v11_0.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2888,7 +2888,7 @@ static int dce_v11_0_early_init(void *ha
 
 	switch (adev->asic_type) {
 	case CHIP_CARRIZO:
-		adev->mode_info.num_crtc = 4;
+		adev->mode_info.num_crtc = 3;
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 9;
 		break;

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

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

From: Mike Snitzer <[email protected]>

commit ba30670f4d5292c4e7f7980bbd5071f7c4794cdd upstream.

---
 drivers/md/dm-thin.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -3255,7 +3255,7 @@ static int pool_ctr(struct dm_target *ti
 						metadata_low_callback,
 						pool);
 	if (r)
-		goto out_free_pt;
+		goto out_flags_changed;
 
 	pt->callbacks.congested_fn = pool_is_congested;
 	dm_table_add_target_callbacks(ti->table, &pt->callbacks);

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

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

From: Ilya Dryomov <[email protected]>

commit 3ebe138ac642a195c7f2efdb918f464734421fd6 upstream.

If rbd_dev_image_probe() in rbd_dev_probe_parent() fails, header_name is freed twice: once in rbd_dev_probe_parent() and then in its caller rbd_dev_image_probe() (rbd_dev_image_probe() is called recursively to handle parent images).

rbd_dev_probe_parent() is responsible for probing the parent, so it shouldn't muck with clone's fields.

---
 drivers/block/rbd.c |    1 -
 1 file changed, 1 deletion(-)

@@ -5201,7 +5201,6 @@ static int rbd_dev_probe_parent(struct r
 out_err:
 	if (parent) {
 		rbd_dev_unparent(rbd_dev);
-		kfree(rbd_dev->header_name);
 		rbd_dev_destroy(parent);
 	} else {
 		rbd_put_client(rbdc);

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

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

From: Thomas Gleixner <[email protected]>

commit 56fd16cabac9cd8f15e2902898a9d0cc96e2fa70 upstream.

timekeeping_init() can set the wall time offset, so we need to increment the clock_was_set_seq counter. That way hrtimers will pick up the early offset immediately. Otherwise on a machine which does not set wall time later in the boot process the hrtimer offset is stale at 0 and wall time timers are going to expire with a delay of 45 years.

---
 kernel/time/timekeeping.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1244,7 +1244,7 @@ void __init timekeeping_init(void)
 	set_normalized_timespec64(&tmp, -boot.tv_sec, -boot.tv_nsec);
 	tk_set_wall_to_mono(tk, tmp);
 
-	timekeeping_update(tk, TK_MIRROR);
+	timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET);
 
 	write_seqcount_end(&tk_core.seq);
 	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);

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

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

From: Steve Capper <[email protected]>

commit 1a541b4e3cd6f5795022514114854b3e1345f24e upstream.

6910fa1 ("arm64: enable PTE type bit in the mask for pte_modify") fixes a problem whereby a large block of PROT_NONE mapped memory is incorrectly mapped as block descriptors when mprotect is called.

Unfortunately, a subtle bug was introduced by this fix to the THP logic.

If one mmaps a large block of memory, then faults it such that it is collapsed into THPs; resulting calls to mprotect on this area of memory will lead to incorrect table descriptors being written instead of block descriptors. This is because pmd_modify calls pte_modify which is now allowed to modify the type of the page table entry.

This patch reverts commit 6910fa16dbe142f6a0fd0fd7c249f9883ff7fc8a, and fixes the problem it was trying to address by adjusting PAGE_NONE to represent a table entry. Thus no change in pte type is required when moving from PROT_NONE to a different protection.

[SteveC: backported 1a541b4e3cd6f5795022514114854b3e1345f24e to 4.1 and
4.2 stable. Just one minor fix to second part to allow patch to apply
cleanly, no logic changed.]

---
 arch/arm64/include/asm/pgtable.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -80,7 +80,7 @@ extern void __pgd_error(const char *file
 #define PAGE_S2			__pgprot(PROT_DEFAULT | PTE_S2_MEMATTR(MT_S2_NORMAL) | PTE_S2_RDONLY)
 #define PAGE_S2_DEVICE		__pgprot(PROT_DEFAULT | PTE_S2_MEMATTR(MT_S2_DEVICE_nGnRE) | PTE_S2_RDONLY | PTE_UXN)
 
-#define PAGE_NONE		__pgprot(((_PAGE_DEFAULT) & ~PTE_TYPE_MASK) | PTE_PROT_NONE | PTE_PXN | PTE_UXN)
+#define PAGE_NONE		__pgprot(((_PAGE_DEFAULT) & ~PTE_VALID) | PTE_PROT_NONE | PTE_PXN | PTE_UXN)
 #define PAGE_SHARED		__pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_UXN | PTE_WRITE)
 #define PAGE_SHARED_EXEC	__pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_WRITE)
 #define PAGE_COPY		__pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_UXN)

@@ -460,7 +460,7 @@ static inline pud_t *pud_offset(pgd_t *p
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 {
 	const pteval_t mask = PTE_USER | PTE_PXN | PTE_UXN | PTE_RDONLY |
-			      PTE_PROT_NONE | PTE_WRITE | PTE_TYPE_MASK;
+			      PTE_PROT_NONE | PTE_VALID | PTE_WRITE;
 	pte_val(pte) = (pte_val(pte) & ~mask) | (pgprot_val(newprot) & mask);
 	return pte;
 }

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

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

From: Steve Wise <[email protected]>

commit c91aed9896946721bb30705ea2904edb3725dd61 upstream.

The server rdma_read_chunk_lcl() and rdma_read_chunk_frmr() functions were not taking into account the initial page_offset when determining the rdma read length. This resulted in a read who's starting address and length exceeded the base/bounds of the frmr.

The server gets an async error from the rdma device and kills the connection, and the client then reconnects and resends. This repeats indefinitely, and the application hangs.

Most work loads don't tickle this bug apparently, but one test hit it every time: building the linux kernel on a 16 core node with 'make -j 16 O=/mnt/0' where /mnt/0 is a ramdisk mounted via NFSRDMA.

This bug seems to only be tripped with devices having small fastreg page list depths. I didn't see it with mlx4, for instance.

---
 net/sunrpc/xprtrdma/svc_rdma_recvfrom.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

@@ -146,7 +146,8 @@ int rdma_read_chunk_lcl(struct svcxprt_r
 	ctxt->read_hdr = head;
 	pages_needed =
 		min_t(int, pages_needed, rdma_read_max_sge(xprt, pages_needed));
-	read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
+	read = min_t(int, (pages_needed << PAGE_SHIFT) - *page_offset,
+		     rs_length);
 
 	for (pno = 0; pno < pages_needed; pno++) {
 		int len = min_t(int, rs_length, PAGE_SIZE - pg_off);

@@ -245,7 +246,8 @@ int rdma_read_chunk_frmr(struct svcxprt_
 	ctxt->direction = DMA_FROM_DEVICE;
 	ctxt->frmr = frmr;
 	pages_needed = min_t(int, pages_needed, xprt->sc_frmr_pg_list_len);
-	read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
+	read = min_t(int, (pages_needed << PAGE_SHIFT) - *page_offset,
+		     rs_length);
 
 	frmr->kva = page_address(rqstp->rq_arg.pages[pg_no]);
 	frmr->direction = DMA_FROM_DEVICE;

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah

Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
[email protected] | (970) 217-8978

Thanks for testing all of these and letting me know.

greg k-h

Build results:

	total: 144 pass: 144 fail: 0
Qemu test results:
	total: 93 pass: 93 fail: 0

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

Guenter

Thanks for testing all of these and letting me know.

greg k-h

Don't worry about this one either. It doesn't support mainline until v4.3, so we need to blacklist it.

Kevin