4.4.81-stable review

This is the start of the stable review cycle for the 4.4.81 release. There are 58 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 Fri Aug 11 19:41:25 UTC 2017. 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-4.4.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.4.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 4.4.81-rc1

Michal Kubeček <[email protected]>
net: account for current skb length when deciding about UFO

zheng li <[email protected]>
ipv4: Should use consistent conditional judgement for ip fragment in __ip_append_data and ip_finish_output

Ard Biesheuvel <[email protected]>
mm: don't dereference struct page fields of invalid pages

Jamie Iles <[email protected]>
signal: protect SIGNAL_UNKILLABLE from unintentional clearing.

Sudip Mukherjee <[email protected]>
lib/Kconfig.debug: fix frv build failure

Michal Hocko <[email protected]>
mm, slab: make sure that KMALLOC_MAX_SIZE will fit into MAX_ORDER

Rabin Vincent <[email protected]>
ARM: 8632/1: ftrace: fix syscall name matching

Omar Sandoval <[email protected]>
virtio_blk: fix panic in initialization error path

Gerd Hoffmann <[email protected]>
drm/virtio: fix framebuffer sparse warning

Milan P. Gandhi <[email protected]>
scsi: qla2xxx: Get mutex lock before checking optrom_state

Zefir Kurtisi <[email protected]>
phy state machine: failsafe leave invalid RUNNING state

Nicholas Mc Guire <[email protected]>
x86/boot: Add missing declaration of string functions

Michael Chan <[email protected]>
tg3: Fix race condition in tg3_get_stats64().

Grygorii Strashko <[email protected]>
net: phy: dp83867: fix irq generation

Sergei Shtylyov <[email protected]>
sh_eth: R8A7740 supports packet shecksumming

Arnd Bergmann <[email protected]>
wext: handle NULL extra data in iwe_stream_add_point better

Rob Gardner <[email protected]>
sparc64: Prevent perf from running during super critical sections

Jane Chu <[email protected]>
sparc64: Measure receiver forward progress to avoid send mondo timeout

Wei Liu <[email protected]>
xen-netback: correctly schedule rate-limited queues

Florian Fainelli <[email protected]>
net: phy: Correctly process PHY_HALTED in phy_stop_machine()

Moshe Shemesh <[email protected]>
net/mlx5: Fix command bad flow on command entry allocation failure

Xin Long <[email protected]>
sctp: fix the check for _sctp_walk_params and _sctp_walk_errors

Alexander Potapenko <[email protected]>
sctp: don't dereference ptr before leaving _sctp_walk_{params, errors}()

Xin Long <[email protected]>
dccp: fix a memleak for dccp_feat_init err process

Xin Long <[email protected]>
dccp: fix a memleak that dccp_ipv4 doesn't put reqsk properly

Xin Long <[email protected]>
dccp: fix a memleak that dccp_ipv6 doesn't put reqsk properly

Marc Gonzalez <[email protected]>
net: ethernet: nb8800: Handle all 4 RGMII modes identically

Stefano Brivio <[email protected]>
ipv6: Don't increase IPSTATS_MIB_FRAGFAILS twice in ip6_fragment()

WANG Cong <[email protected]>
packet: fix use-after-free in prb_retire_rx_blk_timer_expired()

Liping Zhang <[email protected]>
openvswitch: fix potential out of bound access in parse_ct

Thomas Jarosch <[email protected]>
mcs7780: Fix initialization when CONFIG_VMAP_STACK is enabled

WANG Cong <[email protected]>
rtnetlink: allocate more memory for dev_set_mac_address()

Mahesh Bandewar <[email protected]>
ipv4: initialize fib_trie prior to register_netdev_notifier call.

Sabrina Dubroca <[email protected]>
ipv6: avoid overflow of offset in ip6_find_1stfragopt

David S. Miller <[email protected]>
net: Zero terminate ifr_name in dev_ifname().

Alexander Potapenko <[email protected]>

    ipv4: ipv6: initialize treq->txhash in cookie_v[46]_check()

Steven Toth <[email protected]>
saa7164: fix double fetch PCIe access condition

Greg Kroah-Hartman <[email protected]>
drm: rcar-du: fix backport bug

Jin Qian <[email protected]>
f2fs: sanity check checkpoint segno and blkoff

Sean Young <[email protected]>
media: lirc: LIRC_GET_REC_RESOLUTION should return microseconds

Mel Gorman <[email protected]>
mm, mprotect: flush TLB if potentially racing with a parallel reclaim leaving stale TLB entries

Nicholas Bellinger <[email protected]>

    iser-target: Avoid isert_conn->cm_id dereference in isert_login_recv_done

Nicholas Bellinger <[email protected]>
iscsi-target: Fix delayed logout processing greater than SECONDS_FOR_LOGOUT_COMP

Nicholas Bellinger <[email protected]>
iscsi-target: Fix initial login PDU asynchronous socket close OOPs

Nicholas Bellinger <[email protected]>
iscsi-target: Fix early sk_data_ready LOGIN_FLAGS_READY race

Jiang Yi <[email protected]>
iscsi-target: Always wait for kthread_should_stop() before kthread exit

Nicholas Bellinger <[email protected]>
target: Avoid mappedlun symlink creation during lun shutdown

Prabhakar Lad <[email protected]>
media: platform: davinci: return -EINVAL for VPFE_CMD_S_CCDC_RAW_PARAMS ioctl

Gregory CLEMENT <[email protected]>
ARM: dts: armada-38x: Fix irq type for pca955

Jerry Lee <[email protected]>
ext4: fix overflow caused by missing cast in ext4_resize_fs()

Jan Kara <[email protected]>
ext4: fix SEEK_HOLE/SEEK_DATA for blocksize < pagesize

Josh Poimboeuf <[email protected]>
mm/page_alloc: Remove kernel address exposure in free_reserved_area()

Wanpeng Li <[email protected]>
KVM: async_pf: make rcu irq exit if not triggered from idle task

Banajit Goswami <[email protected]>
ASoC: do not close shared backend dailink

Sergei A. Trusov <[email protected]>
ALSA: hda - Fix speaker output from VAIO VPCL14M1R

Tejun Heo <[email protected]>

    workqueue: restore WQ_UNBOUND/max_active==1 to be ordered

Dan Carpenter <[email protected]>
libata: array underflow in ata_find_dev()

Helge Deller <[email protected]>
parisc: Increase thread and stack size to 32kb

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

Diffstat:

 Makefile                                      |   4 +-
 arch/arm/boot/dts/armada-388-gp.dts           |   4 +-
 arch/arm/include/asm/ftrace.h                 |  18 +++
 arch/parisc/include/asm/thread_info.h         |   2 +-
 arch/parisc/kernel/irq.c                      |   2 +-
 arch/sparc/include/asm/mmu_context_64.h       |  12 +-
 arch/sparc/include/asm/trap_block.h           |   1 +
 arch/sparc/kernel/smp_64.c                    | 185 ++++++++++++++---------
 arch/sparc/kernel/sun4v_ivec.S                |  15 ++
 arch/sparc/kernel/traps_64.c                  |   1 +
 arch/sparc/kernel/tsb.S                       |  12 ++
 arch/sparc/power/hibernate.c                  |   3 +-
 arch/x86/boot/string.c                        |   1 +
 arch/x86/boot/string.h                        |   9 ++
 arch/x86/kernel/kvm.c                         |   6 +-
 drivers/ata/libata-scsi.c                     |   6 +-
 drivers/block/virtio_blk.c                    |   3 +-
 drivers/gpu/drm/rcar-du/rcar_du_drv.c         |   2 +-
 drivers/gpu/drm/virtio/virtgpu_fb.c           |   2 +-
 drivers/infiniband/ulp/isert/ib_isert.c       |   2 +-
 drivers/media/pci/saa7164/saa7164-bus.c       |  13 +-
 drivers/media/platform/davinci/vpfe_capture.c |  22 +--
 drivers/media/rc/ir-lirc-codec.c              |   2 +-
 drivers/net/ethernet/aurora/nb8800.c          |   9 +-
 drivers/net/ethernet/broadcom/tg3.c           |   3 +
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c |  19 ++-
 drivers/net/ethernet/renesas/sh_eth.c         |   1 +
 drivers/net/irda/mcs7780.c                    |  16 +-
 drivers/net/phy/dp83867.c                     |  10 ++
 drivers/net/phy/phy.c                         |  12 ++
 drivers/net/xen-netback/common.h              |   1 +
 drivers/net/xen-netback/interface.c           |   6 +-
 drivers/net/xen-netback/netback.c             |   6 +-
 drivers/scsi/qla2xxx/qla_attr.c               |  18 ++-
 drivers/target/iscsi/iscsi_target.c           |  38 ++++-
 drivers/target/iscsi/iscsi_target_erl0.c      |   6 +-
 drivers/target/iscsi/iscsi_target_erl0.h      |   2 +-
 drivers/target/iscsi/iscsi_target_login.c     |   4 +
 drivers/target/iscsi/iscsi_target_nego.c      | 208 +++++++++++++++++---------
 drivers/target/target_core_fabric_configfs.c  |   5 +
 drivers/target/target_core_tpg.c              |   4 +
 fs/ext4/file.c                                |   3 +
 fs/ext4/resize.c                              |   3 +-
 fs/f2fs/super.c                               |  16 ++
 include/linux/mm_types.h                      |   4 +
 include/linux/sched.h                         |  10 ++
 include/linux/slab.h                          |   4 +-
 include/net/iw_handler.h                      |   3 +-
 include/net/sctp/sctp.h                       |   4 +
 include/target/iscsi/iscsi_target_core.h      |   1 +
 include/target/target_core_base.h             |   1 +
 kernel/signal.c                               |   4 +-
 kernel/workqueue.c                            |  10 ++
 lib/Kconfig.debug                             |   2 +-
 mm/internal.h                                 |   5 +-
 mm/memory.c                                   |   1 +
 mm/mprotect.c                                 |   1 +
 mm/mremap.c                                   |   1 +
 mm/page_alloc.c                               |  10 +-
 mm/rmap.c                                     |  36 +++++
 net/core/dev_ioctl.c                          |   1 +
 net/core/rtnetlink.c                          |   3 +-
 net/dccp/feat.c                               |   7 +-
 net/dccp/ipv4.c                               |   1 +
 net/dccp/ipv6.c                               |   1 +
 net/ipv4/fib_frontend.c                       |   9 +-
 net/ipv4/ip_output.c                          |   3 +-
 net/ipv4/syncookies.c                         |   1 +
 net/ipv6/ip6_output.c                         |   6 +-
 net/ipv6/output_core.c                        |   8 +-
 net/ipv6/syncookies.c                         |   1 +
 net/openvswitch/conntrack.c                   |   7 +-
 net/packet/af_packet.c                        |   2 +-
 sound/pci/hda/patch_realtek.c                 |   1 +
 sound/soc/soc-pcm.c                           |   4 +
 75 files changed, 618 insertions(+), 251 deletions(-)

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

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

From: Helge Deller <[email protected]>

commit 8f8201dfed91a43ac38c899c82f81eef3d36afd9 upstream.

Since kernel 4.11 the thread and irq stacks on parisc randomly overflow the default size of 16k. The reason why stack usage suddenly grew is yet unknown.

---
 arch/parisc/include/asm/thread_info.h |    2 +-
 arch/parisc/kernel/irq.c              |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

@@ -34,7 +34,7 @@ struct thread_info {
 
 /* thread information allocation */
 
-#define THREAD_SIZE_ORDER	2 /* PA-RISC requires at least 16k stack */
+#define THREAD_SIZE_ORDER	3 /* PA-RISC requires at least 32k stack */
 /* Be sure to hunt all references to this down when you change the size of
  * the kernel stack */
 #define THREAD_SIZE             (PAGE_SIZE << THREAD_SIZE_ORDER)

@@ -380,7 +380,7 @@ static inline int eirr_to_irq(unsigned l
 /*
  * IRQ STACK - used for irq handler
  */
-#define IRQ_STACK_SIZE      (4096 << 2) /* 16k irq stack size */
+#define IRQ_STACK_SIZE      (4096 << 3) /* 32k irq stack size */
 
 union irq_stack_union {
 	unsigned long stack[IRQ_STACK_SIZE/sizeof(unsigned long)];

So we don't need this for 4.4.

Ben.

Ben Hutchings
Software Developer, Codethink Ltd.

Correct.

I had Cc: [email protected] # 4.11+
in the commit itself:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=8f8201dfed91a43ac38c899c82f81eef3d36afd9

Helge

Yes, my fault, I shouldn't have applied it there, now dropped.

thanks,

greg k-h

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

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

From: Dan Carpenter <[email protected]>

commit 59a5e266c3f5c1567508888dd61a45b86daed0fa upstream.

My static checker complains that "devno" can be negative, meaning that we read before the start of the loop. I've looked at the code, and I think the warning is right. This come from /proc so it's root only or it would be quite a quite a serious bug. The call tree looks like this:

proc_scsi_write() <- gets id and channel from simple_strtoul() -> scsi_add_single_device() <- calls shost->transportt->user_scan()
-> ata_scsi_user_scan()

      -> ata_find_dev()

---
 drivers/ata/libata-scsi.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

@@ -2832,10 +2832,12 @@ static unsigned int atapi_xlat(struct at
 static struct ata_device *ata_find_dev(struct ata_port *ap, int devno)
 {
 	if (!sata_pmp_attached(ap)) {
-		if (likely(devno < ata_link_max_devices(&ap->link)))
+		if (likely(devno >= 0 &&
+			   devno < ata_link_max_devices(&ap->link)))
 			return &ap->link.device[devno];
 	} else {
-		if (likely(devno < ap->nr_pmp_links))
+		if (likely(devno >= 0 &&
+			   devno < ap->nr_pmp_links))
 			return &ap->pmp_link[devno].device[0];
 	}
 

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

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

From: Tejun Heo <[email protected]>

commit 5c0338c68706be53b3dc472e4308961c36e4ece1 upstream.

The combination of WQ_UNBOUND and max_active == 1 used to imply ordered execution. After NUMA affinity 4c16bd327c74 ("workqueue:
implement NUMA affinity for unbound workqueues"), this is no longer true due to per-node worker pools.

While the right way to create an ordered workqueue is alloc_ordered_workqueue(), the documentation has been misleading for a long time and people do use WQ_UNBOUND and max_active == 1 for ordered workqueues which can lead to subtle bugs which are very difficult to trigger.

It's unlikely that we'd see noticeable performance impact by enforcing ordering on WQ_UNBOUND / max_active == 1 workqueues. Let's automatically set __WQ_ORDERED for those workqueues.

---
 kernel/workqueue.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

@@ -3834,6 +3834,16 @@ struct workqueue_struct *__alloc_workque
 	struct workqueue_struct *wq;
 	struct pool_workqueue *pwq;
 
+	/*
+	 * Unbound && max_active == 1 used to imply ordered, which is no
+	 * longer the case on NUMA machines due to per-node pools.  While
+	 * alloc_ordered_workqueue() is the right way to create an ordered
+	 * workqueue, keep the previous behavior to avoid subtle breakages
+	 * on NUMA.
+	 */
+	if ((flags & WQ_UNBOUND) && max_active == 1)
+		flags |= __WQ_ORDERED;
+
 	/* see the comment above the definition of WQ_POWER_EFFICIENT */
 	if ((flags & WQ_POWER_EFFICIENT) && wq_power_efficient)
 		flags |= WQ_UNBOUND;

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

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

From: Sergei A. Trusov <[email protected]>

commit 3f3c371421e601fa93b6cb7fb52da9ad59ec90b4 upstream.

Sony VAIO VPCL14M1R needs the quirk to make the speaker working properly.

---
 sound/pci/hda/patch_realtek.c |    1 +
 1 file changed, 1 insertion(+)

@@ -2233,6 +2233,7 @@ static const struct snd_pci_quirk alc882
 	SND_PCI_QUIRK(0x1043, 0x8691, "ASUS ROG Ranger VIII", ALC882_FIXUP_GPIO3),
 	SND_PCI_QUIRK(0x104d, 0x9047, "Sony Vaio TT", ALC889_FIXUP_VAIO_TT),
 	SND_PCI_QUIRK(0x104d, 0x905a, "Sony Vaio Z", ALC882_FIXUP_NO_PRIMARY_HP),
+	SND_PCI_QUIRK(0x104d, 0x9060, "Sony Vaio VPCL14M1R", ALC882_FIXUP_NO_PRIMARY_HP),
 	SND_PCI_QUIRK(0x104d, 0x9043, "Sony Vaio VGC-LN51JGB", ALC882_FIXUP_NO_PRIMARY_HP),
 	SND_PCI_QUIRK(0x104d, 0x9044, "Sony VAIO AiO", ALC882_FIXUP_NO_PRIMARY_HP),
 

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

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

From: Banajit Goswami <[email protected]>

commit b1cd2e34c69a2f3988786af451b6e17967c293a0 upstream.

Multiple frontend dailinks may be connected to a backend dailink at the same time. When one of frontend dailinks is closed, the associated backend dailink should not be closed if it is connected to other active frontend dailinks. Change ensures that backend dailink is closed only after all connected frontend dailinks are closed.

---
 sound/soc/soc-pcm.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -181,6 +181,10 @@ int dpcm_dapm_stream_event(struct snd_so
 		dev_dbg(be->dev, "ASoC: BE %s event %d dir %d\n",
 				be->dai_link->name, event, dir);
 
+		if ((event == SND_SOC_DAPM_STREAM_STOP) &&
+		    (be->dpcm[dir].users >= 1))
+			continue;
+
 		snd_soc_dapm_stream_event(be, dir, event);
 	}
 

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

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

From: Josh Poimboeuf <[email protected]>

commit adb1fe9ae2ee6ef6bc10f3d5a588020e7664dfa7 upstream.

Linus suggested we try to remove some of the low-hanging fruit related to kernel address exposure in dmesg. The only leaks I see on my local system are:

Freeing SMP alternatives memory: 32K (ffffffff9e309000 - ffffffff9e311000) Freeing initrd memory: 10588K (ffffa0b736b42000 - ffffa0b737599000) Freeing unused kernel memory: 3592K (ffffffff9df87000 - ffffffff9e309000) Freeing unused kernel memory: 1352K (ffffa0b7288ae000 - ffffa0b728a00000) Freeing unused kernel memory: 632K (ffffa0b728d62000 - ffffa0b728e00000)

Linus says:

"I suspect we should just remove [the addresses in the 'Freeing'
messages]. I'm sure they are useful in theory, but I suspect they were more useful back when the whole "free init memory" was originally done.

These days, if we have a use-after-free, I suspect the init-mem situation is the easiest situation by far. Compared to all the dynamic allocations which are much more likely to show it anyway. So having debug output for that case is likely not all that productive."

With this patch the freeing messages now look like this:

Freeing SMP alternatives memory: 32K
Freeing initrd memory: 10588K
Freeing unused kernel memory: 3592K
Freeing unused kernel memory: 1352K
Freeing unused kernel memory: 632K

---
 mm/page_alloc.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -5847,8 +5847,8 @@ unsigned long free_reserved_area(void *s
 	}
 
 	if (pages && s)
-		pr_info("Freeing %s memory: %ldK (%p - %p)\n",
-			s, pages << (PAGE_SHIFT - 10), start, end);
+		pr_info("Freeing %s memory: %ldK\n",
+			s, pages << (PAGE_SHIFT - 10));
 
 	return pages;
 }

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

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

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

From: Jan Kara <[email protected]>

commit fcf5ea10992fbac3c7473a1db33d56a139333cd1 upstream.

ext4_find_unwritten_pgoff() does not properly handle a situation when starting index is in the middle of a page and blocksize < pagesize. The following command shows the bug on filesystem with 1k blocksize:

  xfs_io -f -c "falloc 0 4k" \
            -c "pwrite 1k 1k" \
            -c "pwrite 3k 1k" \
            -c "seek -a -r 0" foo

In this example, neither lseek(fd, 1024, SEEK_HOLE) nor lseek(fd, 2048, SEEK_DATA) will return the correct result.

Fix the problem by neglecting buffers in a page before starting offset.

---
 fs/ext4/file.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -500,6 +500,8 @@ static int ext4_find_unwritten_pgoff(str
 				lastoff = page_offset(page);
 				bh = head = page_buffers(page);
 				do {
+					if (lastoff + bh->b_size <= startoff)
+						goto next;
 					if (buffer_uptodate(bh) ||
 					    buffer_unwritten(bh)) {
 						if (whence == SEEK_DATA)

@@ -514,6 +516,7 @@ static int ext4_find_unwritten_pgoff(str
 						unlock_page(page);
 						goto out;
 					}
+next:
 					lastoff += bh->b_size;
 					bh = bh->b_this_page;
 				} while (bh != head);

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

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

From: Jerry Lee <[email protected]>

commit aec51758ce10a9c847a62a48a168f8c804c6e053 upstream.

On a 32-bit platform, the value of n_blcoks_count may be wrong during the file system is resized to size larger than 2^32 blocks. This may caused the superblock being corrupted with zero blocks count.

---
 fs/ext4/resize.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -1926,7 +1926,8 @@ retry:
 			n_desc_blocks = o_desc_blocks +
 				le16_to_cpu(es->s_reserved_gdt_blocks);
 			n_group = n_desc_blocks * EXT4_DESC_PER_BLOCK(sb);
-			n_blocks_count = n_group * EXT4_BLOCKS_PER_GROUP(sb);
+			n_blocks_count = (ext4_fsblk_t)n_group *
+				EXT4_BLOCKS_PER_GROUP(sb);
 			n_group--; /* set to last group number */
 		}
 

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

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

From: Gregory CLEMENT <[email protected]>

commit 8d4514173211586c6238629b1ef1e071927735f5 upstream.

As written in the datasheet the PCA955 can only handle low level irq and not edge irq.

Without this fix the interrupt is not usable for pca955: the gpio-pca953x driver already set the irq type as low level which is incompatible with edge type, then the kernel prevents using the interrupt:

"irq: type mismatch, failed to map hwirq-18 for
/soc/internal-regs/gpio@18100!"

Development Board support")

---
 arch/arm/boot/dts/armada-388-gp.dts |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -89,7 +89,7 @@
 					pinctrl-names = "default";
 					pinctrl-0 = <&pca0_pins>;
 					interrupt-parent = <&gpio0>;
-					interrupts = <18 IRQ_TYPE_EDGE_FALLING>;
+					interrupts = <18 IRQ_TYPE_LEVEL_LOW>;
 					gpio-controller;
 					#gpio-cells = <2>;
 					interrupt-controller;

@@ -101,7 +101,7 @@
 					compatible = "nxp,pca9555";
 					pinctrl-names = "default";
 					interrupt-parent = <&gpio0>;
-					interrupts = <18 IRQ_TYPE_EDGE_FALLING>;
+					interrupts = <18 IRQ_TYPE_LEVEL_LOW>;
 					gpio-controller;
 					#gpio-cells = <2>;
 					interrupt-controller;

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

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

From: Prabhakar Lad <[email protected]>

commit da05d52d2f0f6bd61094a0cd045fed94bf7d673a upstream.

this patch makes sure VPFE_CMD_S_CCDC_RAW_PARAMS ioctl no longer works for vpfe_capture driver with a minimal patch suitable for backporting.

- This ioctl was never in public api and was only defined in kernel header.
- The function set_params constantly mixes up pointers and phys_addr_t numbers.
- This is part of a 'VPFE_CMD_S_CCDC_RAW_PARAMS' ioctl command that is described as an 'experimental ioctl that will change in future kernels'.
- The code to allocate the table never gets called after we copy_from_user the user input over the kernel settings, and then compare them for inequality.
- We then go on to use an address provided by user space as both the __user pointer for input and pass it through phys_to_virt to come up with a kernel pointer to copy the data to. This looks like a trivially exploitable root hole.

Due to these reasons we make sure this ioctl now returns -EINVAL and backport this patch as far as possible.

---
 drivers/media/platform/davinci/vpfe_capture.c |   22 ++--------------------
 1 file changed, 2 insertions(+), 20 deletions(-)

@@ -1709,27 +1709,9 @@ static long vpfe_param_handler(struct fi
 
 	switch (cmd) {
 	case VPFE_CMD_S_CCDC_RAW_PARAMS:
+		ret = -EINVAL;
 		v4l2_warn(&vpfe_dev->v4l2_dev,
-			  "VPFE_CMD_S_CCDC_RAW_PARAMS: experimental ioctl\n");
-		if (ccdc_dev->hw_ops.set_params) {
-			ret = ccdc_dev->hw_ops.set_params(param);
-			if (ret) {
-				v4l2_dbg(1, debug, &vpfe_dev->v4l2_dev,
-					"Error setting parameters in CCDC\n");
-				goto unlock_out;
-			}
-			ret = vpfe_get_ccdc_image_format(vpfe_dev,
-							 &vpfe_dev->fmt);
-			if (ret < 0) {
-				v4l2_dbg(1, debug, &vpfe_dev->v4l2_dev,
-					"Invalid image format at CCDC\n");
-				goto unlock_out;
-			}
-		} else {
-			ret = -EINVAL;
-			v4l2_dbg(1, debug, &vpfe_dev->v4l2_dev,
-				"VPFE_CMD_S_CCDC_RAW_PARAMS not supported\n");
-		}
+			"VPFE_CMD_S_CCDC_RAW_PARAMS not supported\n");
 		break;
 	default:
 		ret = -ENOTTY;

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

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

From: Nicholas Bellinger <[email protected]>

commit 49cb77e297dc611a1b795cfeb79452b3002bd331 upstream.

This patch closes a race between se_lun deletion during configfs unlink in target_fabric_port_unlink() -> core_dev_del_lun() -> core_tpg_remove_lun(), when transport_clear_lun_ref() blocks waiting for percpu_ref RCU grace period to finish, but a new NodeACL mappedlun is added before the RCU grace period has completed.

This can happen in target_fabric_mappedlun_link() because it only checks for se_lun->lun_se_dev, which is not cleared until after transport_clear_lun_ref() percpu_ref RCU grace period finishes.

This bug originally manifested as NULL pointer dereference OOPsen in target_stat_scsi_att_intr_port_show_attr_dev() on v4.1.y code, because it dereferences lun->lun_se_dev without a explicit NULL pointer check.

In post v4.1 code with target-core RCU conversion, the code in target_stat_scsi_att_intr_port_show_attr_dev() no longer uses se_lun->lun_se_dev, but the same race still exists.

To address the bug, go ahead and set se_lun>lun_shutdown as early as possible in core_tpg_remove_lun(), and ensure new NodeACL mappedlun creation in target_fabric_mappedlun_link() fails during se_lun shutdown.

---
 drivers/target/target_core_fabric_configfs.c |    5 +++++
 drivers/target/target_core_tpg.c             |    4 ++++
 include/target/target_core_base.h            |    1 +
 3 files changed, 10 insertions(+)

@@ -92,6 +92,11 @@ static int target_fabric_mappedlun_link(
 		pr_err("Source se_lun->lun_se_dev does not exist\n");
 		return -EINVAL;
 	}
+	if (lun->lun_shutdown) {
+		pr_err("Unable to create mappedlun symlink because"
+			" lun->lun_shutdown=true\n");
+		return -EINVAL;
+	}
 	se_tpg = lun->lun_tpg;
 
 	nacl_ci = &lun_acl_ci->ci_parent->ci_group->cg_item;

@@ -673,6 +673,8 @@ void core_tpg_remove_lun(
 	 */
 	struct se_device *dev = rcu_dereference_raw(lun->lun_se_dev);
 
+	lun->lun_shutdown = true;
+
 	core_clear_lun_from_tpg(lun, tpg);
 	/*
 	 * Wait for any active I/O references to percpu se_lun->lun_ref to

@@ -694,6 +696,8 @@ void core_tpg_remove_lun(
 	}
 	if (!(dev->se_hba->hba_flags & HBA_FLAGS_INTERNAL_USE))
 		hlist_del_rcu(&lun->link);
+
+	lun->lun_shutdown = false;
 	mutex_unlock(&tpg->tpg_lun_mutex);
 
 	percpu_ref_exit(&lun->lun_ref);

@@ -714,6 +714,7 @@ struct se_lun {
 #define SE_LUN_LINK_MAGIC			0xffff7771
 	u32			lun_link_magic;
 	u32			lun_access;
+	bool			lun_shutdown;
 	u32			lun_index;
 
 	/* RELATIVE TARGET PORT IDENTIFER */

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

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

From: Jiang Yi <[email protected]>

commit 5e0cf5e6c43b9e19fc0284f69e5cd2b4a47523b0 upstream.

There are three timing problems in the kthread usages of iscsi_target_mod:

- np_thread of struct iscsi_np
- rx_thread and tx_thread of struct iscsi_conn

In iscsit_close_connection(), it calls

send_sig(SIGINT, conn->tx_thread, 1);
kthread_stop(conn->tx_thread);

In conn->tx_thread, which is iscsi_target_tx_thread(), when it receive SIGINT the kthread will exit without checking the return value of kthread_should_stop().

So if iscsi_target_tx_thread() exit right between send_sig(SIGINT...) and kthread_stop(...), the kthread_stop() will try to stop an already stopped kthread.

This is invalid according to the documentation of kthread_stop().

(Fix -ECONNRESET logout handling in iscsi_target_tx_thread and
early iscsi_target_rx_thread failure case - nab)

---
 drivers/target/iscsi/iscsi_target.c       |   28 ++++++++++++++++++++++------
 drivers/target/iscsi/iscsi_target_erl0.c  |    6 +++++-
 drivers/target/iscsi/iscsi_target_erl0.h  |    2 +-
 drivers/target/iscsi/iscsi_target_login.c |    4 ++++
 4 files changed, 32 insertions(+), 8 deletions(-)

@@ -3965,6 +3965,8 @@ int iscsi_target_tx_thread(void *arg)
 {
 	int ret = 0;
 	struct iscsi_conn *conn = arg;
+	bool conn_freed = false;
+
 	/*
 	 * Allow ourselves to be interrupted by SIGINT so that a
 	 * connection recovery / failure event can be triggered externally.

@@ -3990,12 +3992,14 @@ get_immediate:
 			goto transport_err;
 
 		ret = iscsit_handle_response_queue(conn);
-		if (ret == 1)
+		if (ret == 1) {
 			goto get_immediate;
-		else if (ret == -ECONNRESET)
+		} else if (ret == -ECONNRESET) {
+			conn_freed = true;
 			goto out;
-		else if (ret < 0)
+		} else if (ret < 0) {
 			goto transport_err;
+		}
 	}
 
 transport_err:

@@ -4005,8 +4009,13 @@ transport_err:
 	 * responsible for cleaning up the early connection failure.
 	 */
 	if (conn->conn_state != TARG_CONN_STATE_IN_LOGIN)
-		iscsit_take_action_for_connection_exit(conn);
+		iscsit_take_action_for_connection_exit(conn, &conn_freed);
 out:
+	if (!conn_freed) {
+		while (!kthread_should_stop()) {
+			msleep(100);
+		}
+	}
 	return 0;
 }
 

@@ -4105,6 +4114,7 @@ int iscsi_target_rx_thread(void *arg)
 	u32 checksum = 0, digest = 0;
 	struct iscsi_conn *conn = arg;
 	struct kvec iov;
+	bool conn_freed = false;
 	/*
 	 * Allow ourselves to be interrupted by SIGINT so that a
 	 * connection recovery / failure event can be triggered externally.

@@ -4116,7 +4126,7 @@ int iscsi_target_rx_thread(void *arg)
 	 */
 	rc = wait_for_completion_interruptible(&conn->rx_login_comp);
 	if (rc < 0 || iscsi_target_check_conn_state(conn))
-		return 0;
+		goto out;
 
 	if (conn->conn_transport->transport_type == ISCSI_INFINIBAND) {
 		struct completion comp;

@@ -4201,7 +4211,13 @@ int iscsi_target_rx_thread(void *arg)
 transport_err:
 	if (!signal_pending(current))
 		atomic_set(&conn->transport_failed, 1);
-	iscsit_take_action_for_connection_exit(conn);
+	iscsit_take_action_for_connection_exit(conn, &conn_freed);
+out:
+	if (!conn_freed) {
+		while (!kthread_should_stop()) {
+			msleep(100);
+		}
+	}
 	return 0;
 }
 

@@ -930,8 +930,10 @@ static void iscsit_handle_connection_cle
 	}
 }
 
-void iscsit_take_action_for_connection_exit(struct iscsi_conn *conn)
+void iscsit_take_action_for_connection_exit(struct iscsi_conn *conn, bool *conn_freed)
 {
+	*conn_freed = false;
+
 	spin_lock_bh(&conn->state_lock);
 	if (atomic_read(&conn->connection_exit)) {
 		spin_unlock_bh(&conn->state_lock);

@@ -942,6 +944,7 @@ void iscsit_take_action_for_connection_e
 	if (conn->conn_state == TARG_CONN_STATE_IN_LOGOUT) {
 		spin_unlock_bh(&conn->state_lock);
 		iscsit_close_connection(conn);
+		*conn_freed = true;
 		return;
 	}
 

@@ -955,4 +958,5 @@ void iscsit_take_action_for_connection_e
 	spin_unlock_bh(&conn->state_lock);
 
 	iscsit_handle_connection_cleanup(conn);
+	*conn_freed = true;
 }

@@ -9,6 +9,6 @@ extern int iscsit_stop_time2retain_timer
 extern void iscsit_connection_reinstatement_rcfr(struct iscsi_conn *);
 extern void iscsit_cause_connection_reinstatement(struct iscsi_conn *, int);
 extern void iscsit_fall_back_to_erl0(struct iscsi_session *);
-extern void iscsit_take_action_for_connection_exit(struct iscsi_conn *);
+extern void iscsit_take_action_for_connection_exit(struct iscsi_conn *, bool *);
 
 #endif   /*** ISCSI_TARGET_ERL0_H ***/

@@ -1436,5 +1436,9 @@ int iscsi_target_login_thread(void *arg)
 			break;
 	}
 
+	while (!kthread_should_stop()) {
+		msleep(100);
+	}
+
 	return 0;
 }

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

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

From: Nicholas Bellinger <[email protected]>

commit 8f0dfb3d8b1120c61f6e2cc3729290db10772b2d upstream.

There is a iscsi-target/tcp login race in LOGIN_FLAGS_READY state assignment that can result in frequent errors during iscsi discovery:

"iSCSI Login negotiation failed."

To address this bug, move the initial LOGIN_FLAGS_READY assignment ahead of iscsi_target_do_login() when handling the initial iscsi_target_start_negotiation() request PDU during connection login.

As iscsi_target_do_login_rx() work_struct callback is clearing LOGIN_FLAGS_READ_ACTIVE after subsequent calls to iscsi_target_do_login(), the early sk_data_ready ahead of the first iscsi_target_do_login() expects LOGIN_FLAGS_READY to also be set for the initial login request PDU.

As reported by Maged, this was first obsered using an MSFT initiator running across multiple VMWare host virtual machines with iscsi-target/tcp.

---
 drivers/target/iscsi/iscsi_target_nego.c |   18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

@@ -1248,16 +1248,16 @@ int iscsi_target_start_negotiation(
 {
 	int ret;
 
-	ret = iscsi_target_do_login(conn, login);
-	if (!ret) {
-		if (conn->sock) {
-			struct sock *sk = conn->sock->sk;
+       if (conn->sock) {
+               struct sock *sk = conn->sock->sk;
 
-			write_lock_bh(&sk->sk_callback_lock);
-			set_bit(LOGIN_FLAGS_READY, &conn->login_flags);
-			write_unlock_bh(&sk->sk_callback_lock);
-		}
-	} else if (ret < 0) {
+               write_lock_bh(&sk->sk_callback_lock);
+               set_bit(LOGIN_FLAGS_READY, &conn->login_flags);
+               write_unlock_bh(&sk->sk_callback_lock);
+       }
+
+       ret = iscsi_target_do_login(conn, login);
+       if (ret < 0) {
 		cancel_delayed_work_sync(&conn->login_work);
 		cancel_delayed_work_sync(&conn->login_cleanup_work);
 		iscsi_target_restore_sock_callbacks(conn);

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

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

From: Nicholas Bellinger <[email protected]>

commit 25cdda95fda78d22d44157da15aa7ea34be3c804 upstream.

This patch fixes a OOPs originally introduced by:

commit bb048357dad6d604520c91586334c9c230366a14
Author: Nicholas Bellinger <[email protected]>
Date: Thu Sep 5 14:54:04 2013 -0700

iscsi-target: Add sk->sk_state_change to cleanup after TCP failure

which would trigger a NULL pointer dereference when a TCP connection was closed asynchronously via iscsi_target_sk_state_change(), but only when the initial PDU processing in iscsi_target_do_login() from iscsi_np process context was blocked waiting for backend I/O to complete.

To address this issue, this patch makes the following changes.

First, it introduces some common helper functions used for checking socket closing state, checking login_flags, and atomically checking socket closing state + setting login_flags.

Second, it introduces a LOGIN_FLAGS_INITIAL_PDU bit to know when a TCP connection has dropped via iscsi_target_sk_state_change(), but the initial PDU processing within iscsi_target_do_login() in iscsi_np context is still running. For this case, it sets LOGIN_FLAGS_CLOSED, but doesn't invoke schedule_delayed_work().

The original NULL pointer dereference case reported by MNC is now handled by iscsi_target_do_login() doing a iscsi_target_sk_check_close() before transitioning to FFP to determine when the socket has already closed, or iscsi_target_start_negotiation() if the login needs to exchange more PDUs (eg: iscsi_target_do_login returned 0) but the socket has closed. For both of these cases, the cleanup up of remaining connection resources will occur in iscsi_target_start_negotiation() from iscsi_np process context once the failure is detected.

Finally, to handle to case where iscsi_target_sk_state_change() is called after the initial PDU procesing is complete, it now invokes conn->login_work -> iscsi_target_do_login_rx() to perform cleanup once existing iscsi_target_sk_check_close() checks detect connection failure. For this case, the cleanup of remaining connection resources will occur in iscsi_target_do_login_rx() from delayed workqueue process context once the failure is detected.

---
 drivers/target/iscsi/iscsi_target_nego.c |  204 ++++++++++++++++++++-----------
 include/target/iscsi/iscsi_target_core.h |    1 
 2 files changed, 138 insertions(+), 67 deletions(-)

@@ -489,14 +489,60 @@ static void iscsi_target_restore_sock_ca
 
 static int iscsi_target_do_login(struct iscsi_conn *, struct iscsi_login *);
 
-static bool iscsi_target_sk_state_check(struct sock *sk)
+static bool __iscsi_target_sk_check_close(struct sock *sk)
 {
 	if (sk->sk_state == TCP_CLOSE_WAIT || sk->sk_state == TCP_CLOSE) {
-		pr_debug("iscsi_target_sk_state_check: TCP_CLOSE_WAIT|TCP_CLOSE,"
+		pr_debug("__iscsi_target_sk_check_close: TCP_CLOSE_WAIT|TCP_CLOSE,"
 			"returning FALSE\n");
-		return false;
+		return true;
 	}
-	return true;
+	return false;
+}
+
+static bool iscsi_target_sk_check_close(struct iscsi_conn *conn)
+{
+	bool state = false;
+
+	if (conn->sock) {
+		struct sock *sk = conn->sock->sk;
+
+		read_lock_bh(&sk->sk_callback_lock);
+		state = (__iscsi_target_sk_check_close(sk) ||
+			 test_bit(LOGIN_FLAGS_CLOSED, &conn->login_flags));
+		read_unlock_bh(&sk->sk_callback_lock);
+	}
+	return state;
+}
+
+static bool iscsi_target_sk_check_flag(struct iscsi_conn *conn, unsigned int flag)
+{
+	bool state = false;
+
+	if (conn->sock) {
+		struct sock *sk = conn->sock->sk;
+
+		read_lock_bh(&sk->sk_callback_lock);
+		state = test_bit(flag, &conn->login_flags);
+		read_unlock_bh(&sk->sk_callback_lock);
+	}
+	return state;
+}
+
+static bool iscsi_target_sk_check_and_clear(struct iscsi_conn *conn, unsigned int flag)
+{
+	bool state = false;
+
+	if (conn->sock) {
+		struct sock *sk = conn->sock->sk;
+
+		write_lock_bh(&sk->sk_callback_lock);
+		state = (__iscsi_target_sk_check_close(sk) ||
+			 test_bit(LOGIN_FLAGS_CLOSED, &conn->login_flags));
+		if (!state)
+			clear_bit(flag, &conn->login_flags);
+		write_unlock_bh(&sk->sk_callback_lock);
+	}
+	return state;
 }
 
 static void iscsi_target_login_drop(struct iscsi_conn *conn, struct iscsi_login *login)

@@ -536,6 +582,20 @@ static void iscsi_target_do_login_rx(str
 
 	pr_debug("entering iscsi_target_do_login_rx, conn: %p, %s:%d\n",
 			conn, current->comm, current->pid);
+	/*
+	 * If iscsi_target_do_login_rx() has been invoked by ->sk_data_ready()
+	 * before initial PDU processing in iscsi_target_start_negotiation()
+	 * has completed, go ahead and retry until it's cleared.
+	 *
+	 * Otherwise if the TCP connection drops while this is occuring,
+	 * iscsi_target_start_negotiation() will detect the failure, call
+	 * cancel_delayed_work_sync(&conn->login_work), and cleanup the
+	 * remaining iscsi connection resources from iscsi_np process context.
+	 */
+	if (iscsi_target_sk_check_flag(conn, LOGIN_FLAGS_INITIAL_PDU)) {
+		schedule_delayed_work(&conn->login_work, msecs_to_jiffies(10));
+		return;
+	}
 
 	spin_lock(&tpg->tpg_state_lock);
 	state = (tpg->tpg_state == TPG_STATE_ACTIVE);

@@ -543,26 +603,12 @@ static void iscsi_target_do_login_rx(str
 
 	if (!state) {
 		pr_debug("iscsi_target_do_login_rx: tpg_state != TPG_STATE_ACTIVE\n");
-		iscsi_target_restore_sock_callbacks(conn);
-		iscsi_target_login_drop(conn, login);
-		iscsit_deaccess_np(np, tpg, tpg_np);
-		return;
+		goto err;
 	}
 
-	if (conn->sock) {
-		struct sock *sk = conn->sock->sk;
-
-		read_lock_bh(&sk->sk_callback_lock);
-		state = iscsi_target_sk_state_check(sk);
-		read_unlock_bh(&sk->sk_callback_lock);
-
-		if (!state) {
-			pr_debug("iscsi_target_do_login_rx, TCP state CLOSE\n");
-			iscsi_target_restore_sock_callbacks(conn);
-			iscsi_target_login_drop(conn, login);
-			iscsit_deaccess_np(np, tpg, tpg_np);
-			return;
-		}
+	if (iscsi_target_sk_check_close(conn)) {
+		pr_debug("iscsi_target_do_login_rx, TCP state CLOSE\n");
+		goto err;
 	}
 
 	conn->login_kworker = current;

@@ -580,34 +626,29 @@ static void iscsi_target_do_login_rx(str
 	flush_signals(current);
 	conn->login_kworker = NULL;
 
-	if (rc < 0) {
-		iscsi_target_restore_sock_callbacks(conn);
-		iscsi_target_login_drop(conn, login);
-		iscsit_deaccess_np(np, tpg, tpg_np);
-		return;
-	}
+	if (rc < 0)
+		goto err;
 
 	pr_debug("iscsi_target_do_login_rx after rx_login_io, %p, %s:%d\n",
 			conn, current->comm, current->pid);
 
 	rc = iscsi_target_do_login(conn, login);
 	if (rc < 0) {
-		iscsi_target_restore_sock_callbacks(conn);
-		iscsi_target_login_drop(conn, login);
-		iscsit_deaccess_np(np, tpg, tpg_np);
+		goto err;
 	} else if (!rc) {
-		if (conn->sock) {
-			struct sock *sk = conn->sock->sk;
-
-			write_lock_bh(&sk->sk_callback_lock);
-			clear_bit(LOGIN_FLAGS_READ_ACTIVE, &conn->login_flags);
-			write_unlock_bh(&sk->sk_callback_lock);
-		}
+		if (iscsi_target_sk_check_and_clear(conn, LOGIN_FLAGS_READ_ACTIVE))
+			goto err;
 	} else if (rc == 1) {
 		iscsi_target_nego_release(conn);
 		iscsi_post_login_handler(np, conn, zero_tsih);
 		iscsit_deaccess_np(np, tpg, tpg_np);
 	}
+	return;
+
+err:
+	iscsi_target_restore_sock_callbacks(conn);
+	iscsi_target_login_drop(conn, login);
+	iscsit_deaccess_np(np, tpg, tpg_np);
 }
 
 static void iscsi_target_do_cleanup(struct work_struct *work)

@@ -655,31 +696,54 @@ static void iscsi_target_sk_state_change
 		orig_state_change(sk);
 		return;
 	}
+	state = __iscsi_target_sk_check_close(sk);
+	pr_debug("__iscsi_target_sk_close_change: state: %d\n", state);
+
 	if (test_bit(LOGIN_FLAGS_READ_ACTIVE, &conn->login_flags)) {
 		pr_debug("Got LOGIN_FLAGS_READ_ACTIVE=1 sk_state_change"
 			 " conn: %p\n", conn);
+		if (state)
+			set_bit(LOGIN_FLAGS_CLOSED, &conn->login_flags);
 		write_unlock_bh(&sk->sk_callback_lock);
 		orig_state_change(sk);
 		return;
 	}
-	if (test_and_set_bit(LOGIN_FLAGS_CLOSED, &conn->login_flags)) {
+	if (test_bit(LOGIN_FLAGS_CLOSED, &conn->login_flags)) {
 		pr_debug("Got LOGIN_FLAGS_CLOSED=1 sk_state_change conn: %p\n",
 			 conn);
 		write_unlock_bh(&sk->sk_callback_lock);
 		orig_state_change(sk);
 		return;
 	}
+	/*
+	 * If the TCP connection has dropped, go ahead and set LOGIN_FLAGS_CLOSED,
+	 * but only queue conn->login_work -> iscsi_target_do_login_rx()
+	 * processing if LOGIN_FLAGS_INITIAL_PDU has already been cleared.
+	 *
+	 * When iscsi_target_do_login_rx() runs, iscsi_target_sk_check_close()
+	 * will detect the dropped TCP connection from delayed workqueue context.
+	 *
+	 * If LOGIN_FLAGS_INITIAL_PDU is still set, which means the initial
+	 * iscsi_target_start_negotiation() is running, iscsi_target_do_login()
+	 * via iscsi_target_sk_check_close() or iscsi_target_start_negotiation()
+	 * via iscsi_target_sk_check_and_clear() is responsible for detecting the
+	 * dropped TCP connection in iscsi_np process context, and cleaning up
+	 * the remaining iscsi connection resources.
+	 */
+	if (state) {
+		pr_debug("iscsi_target_sk_state_change got failed state\n");
+		set_bit(LOGIN_FLAGS_CLOSED, &conn->login_flags);
+		state = test_bit(LOGIN_FLAGS_INITIAL_PDU, &conn->login_flags);
+		write_unlock_bh(&sk->sk_callback_lock);
 
-	state = iscsi_target_sk_state_check(sk);
-	write_unlock_bh(&sk->sk_callback_lock);
-
-	pr_debug("iscsi_target_sk_state_change: state: %d\n", state);
+		orig_state_change(sk);
 
-	if (!state) {
-		pr_debug("iscsi_target_sk_state_change got failed state\n");
-		schedule_delayed_work(&conn->login_cleanup_work, 0);
+		if (!state)
+			schedule_delayed_work(&conn->login_work, 0);
 		return;
 	}
+	write_unlock_bh(&sk->sk_callback_lock);
+
 	orig_state_change(sk);
 }
 

@@ -944,6 +1008,15 @@ static int iscsi_target_do_login(struct
 			if (iscsi_target_handle_csg_one(conn, login) < 0)
 				return -1;
 			if (login_rsp->flags & ISCSI_FLAG_LOGIN_TRANSIT) {
+				/*
+				 * Check to make sure the TCP connection has not
+				 * dropped asynchronously while session reinstatement
+				 * was occuring in this kthread context, before
+				 * transitioning to full feature phase operation.
+				 */
+				if (iscsi_target_sk_check_close(conn))
+					return -1;
+
 				login->tsih = conn->sess->tsih;
 				login->login_complete = 1;
 				iscsi_target_restore_sock_callbacks(conn);

@@ -970,21 +1043,6 @@ static int iscsi_target_do_login(struct
 		break;
 	}
 
-	if (conn->sock) {
-		struct sock *sk = conn->sock->sk;
-		bool state;
-
-		read_lock_bh(&sk->sk_callback_lock);
-		state = iscsi_target_sk_state_check(sk);
-		read_unlock_bh(&sk->sk_callback_lock);
-
-		if (!state) {
-			pr_debug("iscsi_target_do_login() failed state for"
-				 " conn: %p\n", conn);
-			return -1;
-		}
-	}
-
 	return 0;
 }
 

@@ -1251,13 +1309,25 @@ int iscsi_target_start_negotiation(
        if (conn->sock) {
                struct sock *sk = conn->sock->sk;
 
-               write_lock_bh(&sk->sk_callback_lock);
-               set_bit(LOGIN_FLAGS_READY, &conn->login_flags);
-               write_unlock_bh(&sk->sk_callback_lock);
-       }
+		write_lock_bh(&sk->sk_callback_lock);
+		set_bit(LOGIN_FLAGS_READY, &conn->login_flags);
+		set_bit(LOGIN_FLAGS_INITIAL_PDU, &conn->login_flags);
+		write_unlock_bh(&sk->sk_callback_lock);
+	}
+	/*
+	 * If iscsi_target_do_login returns zero to signal more PDU
+	 * exchanges are required to complete the login, go ahead and
+	 * clear LOGIN_FLAGS_INITIAL_PDU but only if the TCP connection
+	 * is still active.
+	 *
+	 * Otherwise if TCP connection dropped asynchronously, go ahead
+	 * and perform connection cleanup now.
+	 */
+	ret = iscsi_target_do_login(conn, login);
+	if (!ret && iscsi_target_sk_check_and_clear(conn, LOGIN_FLAGS_INITIAL_PDU))
+		ret = -1;
 
-       ret = iscsi_target_do_login(conn, login);
-       if (ret < 0) {
+	if (ret < 0) {
 		cancel_delayed_work_sync(&conn->login_work);
 		cancel_delayed_work_sync(&conn->login_cleanup_work);
 		iscsi_target_restore_sock_callbacks(conn);

@@ -562,6 +562,7 @@ struct iscsi_conn {
 #define LOGIN_FLAGS_READ_ACTIVE		1
 #define LOGIN_FLAGS_CLOSED		2
 #define LOGIN_FLAGS_READY		4
+#define LOGIN_FLAGS_INITIAL_PDU		8
 	unsigned long		login_flags;
 	struct delayed_work	login_work;
 	struct delayed_work	login_cleanup_work;

[...]

Ben Hutchings
Software Developer, Codethink Ltd.

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

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

From: Nicholas Bellinger <[email protected]>

commit 105fa2f44e504c830697b0c794822112d79808dc upstream.

This patch fixes a BUG() in iscsit_close_session() that could be triggered when iscsit_logout_post_handler() execution from within tx thread context was not run for more than SECONDS_FOR_LOGOUT_COMP (15 seconds), and the TCP connection didn't already close before then forcing tx thread context to automatically exit.

This would manifest itself during explicit logout as:

[33206.974254] 1 connection(s) still exist for iSCSI session to iqn.1993-08.org.debian:01:3f5523242179
[33206.980184] INFO: NMI handler (kgdb_nmi_handler) took too long to run: 2100.772 msecs
[33209.078643] ------------[ cut here ]------------
[33209.078646] kernel BUG at drivers/target/iscsi/iscsi_target.c:4346!

Normally when explicit logout attempt fails, the tx thread context exits and iscsit_close_connection() from rx thread context does the extra cleanup once it detects conn->conn_logout_remove has not been cleared by the logout type specific post handlers.

To address this special case, if the logout post handler in tx thread context detects conn->tx_thread_active has already been cleared, simply return and exit in order for existing iscsit_close_connection() logic from rx thread context do failed logout cleanup.

---
 drivers/target/iscsi/iscsi_target.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

@@ -4591,8 +4591,11 @@ static void iscsit_logout_post_handler_c
 	 * always sleep waiting for RX/TX thread shutdown to complete
 	 * within iscsit_close_connection().
 	 */
-	if (conn->conn_transport->transport_type == ISCSI_TCP)
+	if (conn->conn_transport->transport_type == ISCSI_TCP) {
 		sleep = cmpxchg(&conn->tx_thread_active, true, false);
+		if (!sleep)
+			return;
+	}
 
 	atomic_set(&conn->conn_logout_remove, 0);
 	complete(&conn->conn_logout_comp);

@@ -4608,8 +4611,11 @@ static void iscsit_logout_post_handler_s
 {
 	int sleep = 1;
 
-	if (conn->conn_transport->transport_type == ISCSI_TCP)
+	if (conn->conn_transport->transport_type == ISCSI_TCP) {
 		sleep = cmpxchg(&conn->tx_thread_active, true, false);
+		if (!sleep)
+			return;
+	}
 
 	atomic_set(&conn->conn_logout_remove, 0);
 	complete(&conn->conn_logout_comp);

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

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

From: Nicholas Bellinger <[email protected]>

commit fce50a2fa4e9c6e103915c351b6d4a98661341d6 upstream.

This patch fixes a NULL pointer dereference in isert_login_recv_done() of isert_conn->cm_id due to isert_cma_handler() -> isert_connect_error() resetting isert_conn->cm_id = NULL during a failed login attempt.

As per Sagi, we will always see the completion of all recv wrs posted on the qp (given that we assigned a ->done handler), this is a FLUSH error completion, we just don't get to verify that because we deref NULL before.

The issue here, was the assumption that dereferencing the connection cm_id is always safe, which is not true since:

commit 4a579da2586bd3b79b025947ea24ede2bbfede62
Author: Sagi Grimberg <[email protected]>
Date: Sun Mar 29 15:52:04 2015 +0300

iser-target: Fix possible deadlock in RDMA_CM connection error

As I see it, we have a direct reference to the isert_device from isert_conn which is the one-liner fix that we actually need like we do in isert_rdma_read_done() and isert_rdma_write_done().

---
 drivers/infiniband/ulp/isert/ib_isert.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1581,7 +1581,7 @@ isert_rcv_completion(struct iser_rx_desc
 		     struct isert_conn *isert_conn,
 		     u32 xfer_len)
 {
-	struct ib_device *ib_dev = isert_conn->cm_id->device;
+	struct ib_device *ib_dev = isert_conn->device->ib_device;
 	struct iscsi_hdr *hdr;
 	u64 rx_dma;
 	int rx_buflen;

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

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

From: Mel Gorman <[email protected]>

commit 3ea277194daaeaa84ce75180ec7c7a2075027a68 upstream.

Stable note for 4.4: The upstream patch patches madvise(MADV_FREE) but 4.4
does not have support for that feature. The changelog is left as-is but the hunk related to madvise is omitted from the backport.

Nadav Amit identified a theoritical race between page reclaim and mprotect due to TLB flushes being batched outside of the PTL being held.

He described the race as follows:

        CPU0                            CPU1
        ----                            ----
                                        user accesses memory using RW PTE
                                        [PTE now cached in TLB]
        try_to_unmap_one()
        ==> ptep_get_and_clear()
        ==> set_tlb_ubc_flush_pending()
                                        mprotect(addr, PROT_READ)
                                        ==> change_pte_range()
                                        ==> [ PTE non-present - no flush ]

                                        user writes using cached RW PTE
        ...

        try_to_unmap_flush()

The same type of race exists for reads when protecting for PROT_NONE and also exists for operations that can leave an old TLB entry behind such as munmap, mremap and madvise.

For some operations like mprotect, it's not necessarily a data integrity issue but it is a correctness issue as there is a window where an mprotect that limits access still allows access. For munmap, it's potentially a data integrity issue although the race is massive as an munmap, mmap and return to userspace must all complete between the window when reclaim drops the PTL and flushes the TLB. However, it's theoritically possible so handle this issue by flushing the mm if reclaim is potentially currently batching TLB flushes.

Other instances where a flush is required for a present pte should be ok as either the page lock is held preventing parallel reclaim or a page reference count is elevated preventing a parallel free leading to corruption. In the case of page_mkclean there isn't an obvious path that userspace could take advantage of without using the operations that are guarded by this patch. Other users such as gup as a race with reclaim looks just at PTEs. huge page variants should be ok as they don't race with reclaim. mincore only looks at PTEs. userfault also should be ok as if a parallel reclaim takes place, it will either fault the page back in or read some of the data before the flush occurs triggering a fault.

Note that a variant of this patch was acked by Andy Lutomirski but this was for the x86 parts on top of his PCID work which didn't make the 4.13 merge window as expected. His ack is dropped from this version and there will be a follow-on patch on top of PCID that will include his ack.

---
 include/linux/mm_types.h |    4 ++++
 mm/internal.h            |    5 ++++-
 mm/memory.c              |    1 +
 mm/mprotect.c            |    1 +
 mm/mremap.c              |    1 +
 mm/rmap.c                |   36 ++++++++++++++++++++++++++++++++++++
 6 files changed, 47 insertions(+), 1 deletion(-)

@@ -504,6 +504,10 @@ struct mm_struct {
 	 */
 	bool tlb_flush_pending;
 #endif
+#ifdef CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
+	/* See flush_tlb_batched_pending() */
+	bool tlb_flush_batched;
+#endif
 	struct uprobes_state uprobes_state;
 #ifdef CONFIG_X86_INTEL_MPX
 	/* address of the bounds directory */

@@ -453,6 +453,7 @@ struct tlbflush_unmap_batch;
 #ifdef CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
 void try_to_unmap_flush(void);
 void try_to_unmap_flush_dirty(void);
+void flush_tlb_batched_pending(struct mm_struct *mm);
 #else
 static inline void try_to_unmap_flush(void)
 {

@@ -460,6 +461,8 @@ static inline void try_to_unmap_flush(vo
 static inline void try_to_unmap_flush_dirty(void)
 {
 }
-
+static inline void flush_tlb_batched_pending(struct mm_struct *mm)
+{
+}
 #endif /* CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH */
 #endif	/* __MM_INTERNAL_H */

@@ -1127,6 +1127,7 @@ again:
 	init_rss_vec(rss);
 	start_pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
 	pte = start_pte;
+	flush_tlb_batched_pending(mm);
 	arch_enter_lazy_mmu_mode();
 	do {
 		pte_t ptent = *pte;

@@ -72,6 +72,7 @@ static unsigned long change_pte_range(st
 	if (!pte)
 		return 0;
 
+	flush_tlb_batched_pending(vma->vm_mm);
 	arch_enter_lazy_mmu_mode();
 	do {
 		oldpte = *pte;

@@ -135,6 +135,7 @@ static void move_ptes(struct vm_area_str
 	new_ptl = pte_lockptr(mm, new_pmd);
 	if (new_ptl != old_ptl)
 		spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
+	flush_tlb_batched_pending(vma->vm_mm);
 	arch_enter_lazy_mmu_mode();
 
 	for (; old_addr < old_end; old_pte++, old_addr += PAGE_SIZE,

@@ -649,6 +649,13 @@ static void set_tlb_ubc_flush_pending(st
 	tlb_ubc->flush_required = true;
 
 	/*
+	 * Ensure compiler does not re-order the setting of tlb_flush_batched
+	 * before the PTE is cleared.
+	 */
+	barrier();
+	mm->tlb_flush_batched = true;
+
+	/*
 	 * If the PTE was dirty then it's best to assume it's writable. The
 	 * caller must use try_to_unmap_flush_dirty() or try_to_unmap_flush()
 	 * before the page is queued for IO.

@@ -675,6 +682,35 @@ static bool should_defer_flush(struct mm
 
 	return should_defer;
 }
+
+/*
+ * Reclaim unmaps pages under the PTL but do not flush the TLB prior to
+ * releasing the PTL if TLB flushes are batched. It's possible for a parallel
+ * operation such as mprotect or munmap to race between reclaim unmapping
+ * the page and flushing the page. If this race occurs, it potentially allows
+ * access to data via a stale TLB entry. Tracking all mm's that have TLB
+ * batching in flight would be expensive during reclaim so instead track
+ * whether TLB batching occurred in the past and if so then do a flush here
+ * if required. This will cost one additional flush per reclaim cycle paid
+ * by the first operation at risk such as mprotect and mumap.
+ *
+ * This must be called under the PTL so that an access to tlb_flush_batched
+ * that is potentially a "reclaim vs mprotect/munmap/etc" race will synchronise
+ * via the PTL.
+ */
+void flush_tlb_batched_pending(struct mm_struct *mm)
+{
+	if (mm->tlb_flush_batched) {
+		flush_tlb_mm(mm);
+
+		/*
+		 * Do not allow the compiler to re-order the clearing of
+		 * tlb_flush_batched before the tlb is flushed.
+		 */
+		barrier();
+		mm->tlb_flush_batched = false;
+	}
+}
 #else
 static void set_tlb_ubc_flush_pending(struct mm_struct *mm,
 		struct page *page, bool writable)

[...]

What about USE_SPLIT_PTE_PTLOCKS? I don't see how you can use "the PTL" to synchronise access to a per-mm flag.

Ben.

Ben Hutchings
Software Developer, Codethink Ltd.

Although it is a per-mm flag, the only situations we care about it are those in which “the PTL” (i.e. the same PTL) is accessed by both the reclaimer (which batches the flushes) and mprotect/munmap/etc.

Nadav

Is there anything that presents this sequence?

P0                              P1                             P2
--                              --                             --

change_pte_range() [ptl=X]
-> flush_tlb_batch_pending()
-> flush_tlb_mm()

                                try_to_unmap_one() [ptl=Y]
                                -> set_tlb_ubc_flush_pending()
                                   -> tlb_flush_batched = true

-> tlb_flush_batched = false

change_pte_range() [ptl=Y]

                                                               ->
flush_tlb_batch_pending()
                                                                  (nop)

Ben.

Ben Hutchings
Software Developer, Codethink Ltd.

I think (but not sure) that you regard a similar concern I raised before. Mel gave an answer [1], but I cannot say I feel very comfortable with it.

[1] http://www.spinics.net/lists/linux-mm/msg131265.html

Nadav

In this context, the primary concern is a race with clearing and checking PTEs at the location protected by a single PTL lock. While the flag in question is a per-mm flag, the ordering only matters when a race can potentially occur.

Mel Gorman
SUSE Labs

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

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

From: Sean Young <[email protected]>

commit 9f5039ba440e499d85c29b1ddbc3cbc9dc90e44b upstream.

Since commit e8f4818895b3 ("[media] lirc: advertise LIRC_CAN_GET_REC_RESOLUTION and improve") lircd uses the ioctl LIRC_GET_REC_RESOLUTION to determine the shortest pulse or space that the hardware can detect. This breaks decoding in lirc because lircd expects the answer in microseconds, but nanoseconds is returned.

---
 drivers/media/rc/ir-lirc-codec.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -254,7 +254,7 @@ static long ir_lirc_ioctl(struct file *f
 		return 0;
 
 	case LIRC_GET_REC_RESOLUTION:
-		val = dev->rx_resolution;
+		val = dev->rx_resolution / 1000;
 		break;
 
 	case LIRC_SET_WIDEBAND_RECEIVER:

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

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

From: Jin Qian <[email protected]>

commit 15d3042a937c13f5d9244241c7a9c8416ff6e82a upstream.

Make sure segno and blkoff read from raw image are valid.

---
 fs/f2fs/super.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)

@@ -1078,6 +1078,8 @@ static int sanity_check_ckpt(struct f2fs
 	unsigned int total, fsmeta;
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
+	unsigned int main_segs, blocks_per_seg;
+	int i;
 
 	total = le32_to_cpu(raw_super->segment_count);
 	fsmeta = le32_to_cpu(raw_super->segment_count_ckpt);

@@ -1089,6 +1091,20 @@ static int sanity_check_ckpt(struct f2fs
 	if (unlikely(fsmeta >= total))
 		return 1;
 
+	main_segs = le32_to_cpu(raw_super->segment_count_main);
+	blocks_per_seg = sbi->blocks_per_seg;
+
+	for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) {
+		if (le32_to_cpu(ckpt->cur_node_segno[i]) >= main_segs ||
+			le16_to_cpu(ckpt->cur_node_blkoff[i]) >= blocks_per_seg)
+			return 1;
+	}
+	for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) {
+		if (le32_to_cpu(ckpt->cur_data_segno[i]) >= main_segs ||
+			le16_to_cpu(ckpt->cur_data_blkoff[i]) >= blocks_per_seg)
+			return 1;
+	}
+
 	if (unlikely(f2fs_cp_error(sbi))) {
 		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
 		return 1;

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

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

From: Greg Kroah-Hartman <[email protected]>

In the backport of commit 4f7b0d263833 ("drm: rcar-du: Simplify and fix probe error handling"), which is commit 8255d26322a3 in this tree, the error handling path was incorrect. This patch fixes it up.

---
 drivers/gpu/drm/rcar-du/rcar_du_drv.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -296,7 +296,7 @@ static int rcar_du_probe(struct platform
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	rcdu->mmio = devm_ioremap_resource(&pdev->dev, mem);
 	if (IS_ERR(rcdu->mmio))
-		ret = PTR_ERR(rcdu->mmio);
+		return PTR_ERR(rcdu->mmio);
 
 	/* DRM/KMS objects */
 	ddev = drm_dev_alloc(&rcar_du_driver, &pdev->dev);

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

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

From: Steven Toth <[email protected]>

commit 6fb05e0dd32e566facb96ea61a48c7488daa5ac3 upstream.

Avoid a double fetch by reusing the values from the prior transfer.

Originally reported via https://bugzilla.kernel.org/show_bug.cgi?id=195559

Thanks to Pengfei Wang <[email protected]> for reporting.

---
 drivers/media/pci/saa7164/saa7164-bus.c |   13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

@@ -393,11 +393,11 @@ int saa7164_bus_get(struct saa7164_dev *
 	msg_tmp.size = le16_to_cpu((__force __le16)msg_tmp.size);
 	msg_tmp.command = le32_to_cpu((__force __le32)msg_tmp.command);
 	msg_tmp.controlselector = le16_to_cpu((__force __le16)msg_tmp.controlselector);
+	memcpy(msg, &msg_tmp, sizeof(*msg));
 
 	/* No need to update the read positions, because this was a peek */
 	/* If the caller specifically want to peek, return */
 	if (peekonly) {
-		memcpy(msg, &msg_tmp, sizeof(*msg));
 		goto peekout;
 	}
 

@@ -442,21 +442,15 @@ int saa7164_bus_get(struct saa7164_dev *
 		space_rem = bus->m_dwSizeGetRing - curr_grp;
 
 		if (space_rem < sizeof(*msg)) {
-			/* msg wraps around the ring */
-			memcpy_fromio(msg, bus->m_pdwGetRing + curr_grp, space_rem);
-			memcpy_fromio((u8 *)msg + space_rem, bus->m_pdwGetRing,
-				sizeof(*msg) - space_rem);
 			if (buf)
 				memcpy_fromio(buf, bus->m_pdwGetRing + sizeof(*msg) -
 					space_rem, buf_size);
 
 		} else if (space_rem == sizeof(*msg)) {
-			memcpy_fromio(msg, bus->m_pdwGetRing + curr_grp, sizeof(*msg));
 			if (buf)
 				memcpy_fromio(buf, bus->m_pdwGetRing, buf_size);
 		} else {
 			/* Additional data wraps around the ring */
-			memcpy_fromio(msg, bus->m_pdwGetRing + curr_grp, sizeof(*msg));
 			if (buf) {
 				memcpy_fromio(buf, bus->m_pdwGetRing + curr_grp +
 					sizeof(*msg), space_rem - sizeof(*msg));

@@ -469,15 +463,10 @@ int saa7164_bus_get(struct saa7164_dev *
 
 	} else {
 		/* No wrapping */
-		memcpy_fromio(msg, bus->m_pdwGetRing + curr_grp, sizeof(*msg));
 		if (buf)
 			memcpy_fromio(buf, bus->m_pdwGetRing + curr_grp + sizeof(*msg),
 				buf_size);
 	}
-	/* Convert from little endian to CPU */
-	msg->size = le16_to_cpu((__force __le16)msg->size);
-	msg->command = le32_to_cpu((__force __le32)msg->command);
-	msg->controlselector = le16_to_cpu((__force __le16)msg->controlselector);
 
 	/* Update the read positions, adjusting the ring */
 	saa7164_writel(bus->m_dwGetReadPos, new_grp);

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

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

From: Alexander Potapenko <[email protected]>

[ Upstream commit 18bcf2907df935981266532e1e0d052aff2e6fae ]

KMSAN reported use of uninitialized memory in skb_set_hash_from_sk(), which originated from the TCP request socket created in cookie_v6_check():

==================================================================

 BUG: KMSAN: use of uninitialized memory in tcp_transmit_skb+0xf77/0x3ec0
 CPU: 1 PID: 2949 Comm: syz-execprog Not tainted 4.11.0-rc5+ #2931
 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
 TCP: request_sock_TCPv6: Possible SYN flooding on port 20028. Sending cookies.  Check SNMP counters.
 Call Trace:
  <IRQ>
  __dump_stack lib/dump_stack.c:16
  dump_stack+0x172/0x1c0 lib/dump_stack.c:52
  kmsan_report+0x12a/0x180 mm/kmsan/kmsan.c:927
  __msan_warning_32+0x61/0xb0 mm/kmsan/kmsan_instr.c:469
  skb_set_hash_from_sk ./include/net/sock.h:2011
  tcp_transmit_skb+0xf77/0x3ec0 net/ipv4/tcp_output.c:983
  tcp_send_ack+0x75b/0x830 net/ipv4/tcp_output.c:3493
  tcp_delack_timer_handler+0x9a6/0xb90 net/ipv4/tcp_timer.c:284
  tcp_delack_timer+0x1b0/0x310 net/ipv4/tcp_timer.c:309
  call_timer_fn+0x240/0x520 kernel/time/timer.c:1268
  expire_timers kernel/time/timer.c:1307
  __run_timers+0xc13/0xf10 kernel/time/timer.c:1601
  run_timer_softirq+0x36/0xa0 kernel/time/timer.c:1614
  __do_softirq+0x485/0x942 kernel/softirq.c:284
  invoke_softirq kernel/softirq.c:364
  irq_exit+0x1fa/0x230 kernel/softirq.c:405
  exiting_irq+0xe/0x10 ./arch/x86/include/asm/apic.h:657
  smp_apic_timer_interrupt+0x5a/0x80 arch/x86/kernel/apic/apic.c:966
  apic_timer_interrupt+0x86/0x90 arch/x86/entry/entry_64.S:489
 RIP: 0010:native_restore_fl ./arch/x86/include/asm/irqflags.h:36
 RIP: 0010:arch_local_irq_restore ./arch/x86/include/asm/irqflags.h:77
 RIP: 0010:__msan_poison_alloca+0xed/0x120 mm/kmsan/kmsan_instr.c:440
 RSP: 0018:ffff880024917cd8 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff10
 RAX: 0000000000000246 RBX: ffff8800224c0000 RCX: 0000000000000005
 RDX: 0000000000000004 RSI: ffff880000000000 RDI: ffffea0000b6d770
 RBP: ffff880024917d58 R08: 0000000000000dd8 R09: 0000000000000004
 R10: 0000160000000000 R11: 0000000000000000 R12: ffffffff85abf810
 R13: ffff880024917dd8 R14: 0000000000000010 R15: ffffffff81cabde4
  </IRQ>
  poll_select_copy_remaining+0xac/0x6b0 fs/select.c:293
  SYSC_select+0x4b4/0x4e0 fs/select.c:653
  SyS_select+0x76/0xa0 fs/select.c:634
  entry_SYSCALL_64_fastpath+0x13/0x94 arch/x86/entry/entry_64.S:204
 RIP: 0033:0x4597e7
 RSP: 002b:000000c420037ee0 EFLAGS: 00000246 ORIG_RAX: 0000000000000017
 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00000000004597e7
 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
 RBP: 000000c420037ef0 R08: 000000c420037ee0 R09: 0000000000000059
 R10: 0000000000000000 R11: 0000000000000246 R12: 000000000042dc20
 R13: 00000000000000f3 R14: 0000000000000030 R15: 0000000000000003
 chained origin:
  save_stack_trace+0x37/0x40 arch/x86/kernel/stacktrace.c:59
  kmsan_save_stack_with_flags mm/kmsan/kmsan.c:302
  kmsan_save_stack mm/kmsan/kmsan.c:317
  kmsan_internal_chain_origin+0x12a/0x1f0 mm/kmsan/kmsan.c:547
  __msan_store_shadow_origin_4+0xac/0x110 mm/kmsan/kmsan_instr.c:259
  tcp_create_openreq_child+0x709/0x1ae0 net/ipv4/tcp_minisocks.c:472
  tcp_v6_syn_recv_sock+0x7eb/0x2a30 net/ipv6/tcp_ipv6.c:1103
  tcp_get_cookie_sock+0x136/0x5f0 net/ipv4/syncookies.c:212
  cookie_v6_check+0x17a9/0x1b50 net/ipv6/syncookies.c:245
  tcp_v6_cookie_check net/ipv6/tcp_ipv6.c:989
  tcp_v6_do_rcv+0xdd8/0x1c60 net/ipv6/tcp_ipv6.c:1298
  tcp_v6_rcv+0x41a3/0x4f00 net/ipv6/tcp_ipv6.c:1487
  ip6_input_finish+0x82f/0x1ee0 net/ipv6/ip6_input.c:279
  NF_HOOK ./include/linux/netfilter.h:257
  ip6_input+0x239/0x290 net/ipv6/ip6_input.c:322
  dst_input ./include/net/dst.h:492
  ip6_rcv_finish net/ipv6/ip6_input.c:69
  NF_HOOK ./include/linux/netfilter.h:257
  ipv6_rcv+0x1dbd/0x22e0 net/ipv6/ip6_input.c:203
  __netif_receive_skb_core+0x2f6f/0x3a20 net/core/dev.c:4208
  __netif_receive_skb net/core/dev.c:4246
  process_backlog+0x667/0xba0 net/core/dev.c:4866
  napi_poll net/core/dev.c:5268
  net_rx_action+0xc95/0x1590 net/core/dev.c:5333
  __do_softirq+0x485/0x942 kernel/softirq.c:284
 origin:
  save_stack_trace+0x37/0x40 arch/x86/kernel/stacktrace.c:59
  kmsan_save_stack_with_flags mm/kmsan/kmsan.c:302
  kmsan_internal_poison_shadow+0xb1/0x1a0 mm/kmsan/kmsan.c:198
  kmsan_kmalloc+0x7f/0xe0 mm/kmsan/kmsan.c:337
  kmem_cache_alloc+0x1c2/0x1e0 mm/slub.c:2766
  reqsk_alloc ./include/net/request_sock.h:87
  inet_reqsk_alloc+0xa4/0x5b0 net/ipv4/tcp_input.c:6200
  cookie_v6_check+0x4f4/0x1b50 net/ipv6/syncookies.c:169
  tcp_v6_cookie_check net/ipv6/tcp_ipv6.c:989
  tcp_v6_do_rcv+0xdd8/0x1c60 net/ipv6/tcp_ipv6.c:1298
  tcp_v6_rcv+0x41a3/0x4f00 net/ipv6/tcp_ipv6.c:1487
  ip6_input_finish+0x82f/0x1ee0 net/ipv6/ip6_input.c:279
  NF_HOOK ./include/linux/netfilter.h:257
  ip6_input+0x239/0x290 net/ipv6/ip6_input.c:322
  dst_input ./include/net/dst.h:492
  ip6_rcv_finish net/ipv6/ip6_input.c:69
  NF_HOOK ./include/linux/netfilter.h:257
  ipv6_rcv+0x1dbd/0x22e0 net/ipv6/ip6_input.c:203
  __netif_receive_skb_core+0x2f6f/0x3a20 net/core/dev.c:4208
  __netif_receive_skb net/core/dev.c:4246
  process_backlog+0x667/0xba0 net/core/dev.c:4866
  napi_poll net/core/dev.c:5268
  net_rx_action+0xc95/0x1590 net/core/dev.c:5333
  __do_softirq+0x485/0x942 kernel/softirq.c:284
 ==================================================================

Similar error is reported for cookie_v4_check().

---
 net/ipv4/syncookies.c |    1 +
 net/ipv6/syncookies.c |    1 +
 2 files changed, 2 insertions(+)

@@ -337,6 +337,7 @@ struct sock *cookie_v4_check(struct sock
 	treq = tcp_rsk(req);
 	treq->rcv_isn		= ntohl(th->seq) - 1;
 	treq->snt_isn		= cookie;
+	treq->txhash		= net_tx_rndhash();
 	req->mss		= mss;
 	ireq->ir_num		= ntohs(th->dest);
 	ireq->ir_rmt_port	= th->source;

@@ -210,6 +210,7 @@ struct sock *cookie_v6_check(struct sock
 	treq->snt_synack.v64	= 0;
 	treq->rcv_isn = ntohl(th->seq) - 1;
 	treq->snt_isn = cookie;
+	treq->txhash = net_tx_rndhash();
 
 	/*
 	 * We need to lookup the dst_entry to get the correct window size.

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

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

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

[ Upstream commit 63679112c536289826fec61c917621de95ba2ade ]

The ifr.ifr_name is passed around and assumed to be NULL terminated.

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

@@ -28,6 +28,7 @@ static int dev_ifname(struct net *net, s
 
 	if (copy_from_user(&ifr, arg, sizeof(struct ifreq)))
 		return -EFAULT;
+	ifr.ifr_name[IFNAMSIZ-1] = 0;
 
 	error = netdev_get_name(net, ifr.ifr_name, ifr.ifr_ifindex);
 	if (error)

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

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

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 6399f1fae4ec29fab5ec76070435555e256ca3a6 ]

In some cases, offset can overflow and can cause an infinite loop in ip6_find_1stfragopt(). Make it unsigned int to prevent the overflow, and cap it at IPV6_MAXPLEN, since packets larger than that should be invalid.

This problem has been here since before the beginning of git history.

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

@@ -78,7 +78,7 @@ EXPORT_SYMBOL(ipv6_select_ident);
 
 int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr)
 {
-	u16 offset = sizeof(struct ipv6hdr);
+	unsigned int offset = sizeof(struct ipv6hdr);
 	unsigned int packet_len = skb_tail_pointer(skb) -
 		skb_network_header(skb);
 	int found_rhdr = 0;

@@ -86,6 +86,7 @@ int ip6_find_1stfragopt(struct sk_buff *
 
 	while (offset <= packet_len) {
 		struct ipv6_opt_hdr *exthdr;
+		unsigned int len;
 
 		switch (**nexthdr) {
 

@@ -111,7 +112,10 @@ int ip6_find_1stfragopt(struct sk_buff *
 
 		exthdr = (struct ipv6_opt_hdr *)(skb_network_header(skb) +
 						 offset);
-		offset += ipv6_optlen(exthdr);
+		len = ipv6_optlen(exthdr);
+		if (len + offset >= IPV6_MAXPLEN)
+			return -EINVAL;
+		offset += len;
 		*nexthdr = &exthdr->nexthdr;
 	}
 

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

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

From: Mahesh Bandewar <[email protected]>

[ Upstream commit 8799a221f5944a7d74516ecf46d58c28ec1d1f75 ]

Net stack initialization currently initializes fib-trie after the first call to netdevice_notifier() call. In fact fib_trie initialization needs to happen before first rtnl_register(). It does not cause any problem since there are no devices UP at this moment, but trying to bring 'lo' UP at initialization would make this assumption wrong and exposes the issue.

Fixes following crash

 Call Trace:
  ? alternate_node_alloc+0x76/0xa0
  fib_table_insert+0x1b7/0x4b0
  fib_magic.isra.17+0xea/0x120
  fib_add_ifaddr+0x7b/0x190
  fib_netdev_event+0xc0/0x130
  register_netdevice_notifier+0x1c1/0x1d0
  ip_fib_init+0x72/0x85
  ip_rt_init+0x187/0x1e9
  ip_init+0xe/0x1a
  inet_init+0x171/0x26c
  ? ipv4_offload_init+0x66/0x66
  do_one_initcall+0x43/0x160
  kernel_init_freeable+0x191/0x219
  ? rest_init+0x80/0x80
  kernel_init+0xe/0x150
  ret_from_fork+0x22/0x30
 Code: f6 46 23 04 74 86 4c 89 f7 e8 ae 45 01 00 49 89 c7 4d 85 ff 0f 85 7b ff ff ff 31 db eb 08 4c 89 ff e8 16 47 01 00 48 8b 44 24 38 <45> 8b 6e 14 4d 63 76 74 48 89 04 24 0f 1f 44 00 00 48 83 c4 08
 RIP: kmem_cache_alloc+0xcf/0x1c0 RSP: ffff9b1500017c28
 CR2: 0000000000000014

---
 net/ipv4/fib_frontend.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

@@ -1319,13 +1319,14 @@ static struct pernet_operations fib_net_
 
 void __init ip_fib_init(void)
 {
-	rtnl_register(PF_INET, RTM_NEWROUTE, inet_rtm_newroute, NULL, NULL);
-	rtnl_register(PF_INET, RTM_DELROUTE, inet_rtm_delroute, NULL, NULL);
-	rtnl_register(PF_INET, RTM_GETROUTE, NULL, inet_dump_fib, NULL);
+	fib_trie_init();
 
 	register_pernet_subsys(&fib_net_ops);
+
 	register_netdevice_notifier(&fib_netdev_notifier);
 	register_inetaddr_notifier(&fib_inetaddr_notifier);
 
-	fib_trie_init();
+	rtnl_register(PF_INET, RTM_NEWROUTE, inet_rtm_newroute, NULL, NULL);
+	rtnl_register(PF_INET, RTM_DELROUTE, inet_rtm_delroute, NULL, NULL);
+	rtnl_register(PF_INET, RTM_GETROUTE, NULL, inet_dump_fib, NULL);
 }

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

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

From: WANG Cong <[email protected]>

[ Upstream commit 153711f9421be5dbc973dc57a4109dc9d54c89b1 ]

virtnet_set_mac_address() interprets mac address as struct sockaddr, but upper layer only allocates dev->addr_len which is ETH_ALEN + sizeof(sa_family_t) in this case.

We lack a unified definition for mac address, so just fix the upper layer, this also allows drivers to interpret it to struct sockaddr freely.

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

@@ -1742,7 +1742,8 @@ static int do_setlink(const struct sk_bu
 		struct sockaddr *sa;
 		int len;
 
-		len = sizeof(sa_family_t) + dev->addr_len;
+		len = sizeof(sa_family_t) + max_t(size_t, dev->addr_len,
+						  sizeof(*sa));
 		sa = kmalloc(len, GFP_KERNEL);
 		if (!sa) {
 			err = -ENOMEM;

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

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

From: Thomas Jarosch <[email protected]>

[ Upstream commit 9476d393667968b4a02afbe9d35a3558482b943e ]

DMA transfers are not allowed to buffers that are on the stack. Therefore allocate a buffer to store the result of usb_control_message().

Fixes these bugreports:
https://bugzilla.kernel.org/show_bug.cgi?id=195217

https://bugzilla.redhat.com/show_bug.cgi?id=1421387
https://bugzilla.redhat.com/show_bug.cgi?id=1427398

Shortened kernel backtrace from 4.11.9-200.fc25.x86_64:
kernel: ------------[ cut here ]------------
kernel: WARNING: CPU: 3 PID: 2957 at drivers/usb/core/hcd.c:1587
kernel: transfer buffer not dma capable
kernel: Call Trace:
kernel: dump_stack+0x63/0x86
kernel: __warn+0xcb/0xf0
kernel: warn_slowpath_fmt+0x5a/0x80
kernel: usb_hcd_map_urb_for_dma+0x37f/0x570
kernel: ? try_to_del_timer_sync+0x53/0x80
kernel: usb_hcd_submit_urb+0x34e/0xb90
kernel: ? schedule_timeout+0x17e/0x300
kernel: ? del_timer_sync+0x50/0x50
kernel: ? __slab_free+0xa9/0x300
kernel: usb_submit_urb+0x2f4/0x560
kernel: ? urb_destroy+0x24/0x30
kernel: usb_start_wait_urb+0x6e/0x170
kernel: usb_control_msg+0xdc/0x120
kernel: mcs_get_reg+0x36/0x40 [mcs7780]
kernel: mcs_net_open+0xb5/0x5c0 [mcs7780]
...

Regression goes back to 4.9, so it's a good candidate for -stable. Though it's the decision of the maintainer.

Thanks to Dan Williams for adding the "transfer buffer not dma capable" warning in the first place. It instantly pointed me in the right direction.

Patch has been tested with transferring data from a Polar watch.

---
 drivers/net/irda/mcs7780.c |   16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

@@ -141,9 +141,19 @@ static int mcs_set_reg(struct mcs_cb *mc
 static int mcs_get_reg(struct mcs_cb *mcs, __u16 reg, __u16 * val)
 {
 	struct usb_device *dev = mcs->usbdev;
-	int ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0), MCS_RDREQ,
-				  MCS_RD_RTYPE, 0, reg, val, 2,
-				  msecs_to_jiffies(MCS_CTRL_TIMEOUT));
+	void *dmabuf;
+	int ret;
+
+	dmabuf = kmalloc(sizeof(__u16), GFP_KERNEL);
+	if (!dmabuf)
+		return -ENOMEM;
+
+	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0), MCS_RDREQ,
+			      MCS_RD_RTYPE, 0, reg, dmabuf, 2,
+			      msecs_to_jiffies(MCS_CTRL_TIMEOUT));
+
+	memcpy(val, dmabuf, sizeof(__u16));
+	kfree(dmabuf);
 
 	return ret;
 }

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

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

From: Liping Zhang <[email protected]>

[ Upstream commit 69ec932e364b1ba9c3a2085fe96b76c8a3f71e7c ]

Before the 'type' is validated, we shouldn't use it to fetch the ovs_ct_attr_lens's minlen and maxlen, else, out of bound access may happen.

---
 net/openvswitch/conntrack.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

@@ -577,8 +577,8 @@ static int parse_ct(const struct nlattr
 
 	nla_for_each_nested(a, attr, rem) {
 		int type = nla_type(a);
-		int maxlen = ovs_ct_attr_lens[type].maxlen;
-		int minlen = ovs_ct_attr_lens[type].minlen;
+		int maxlen;
+		int minlen;
 
 		if (type > OVS_CT_ATTR_MAX) {
 			OVS_NLERR(log,

@@ -586,6 +586,9 @@ static int parse_ct(const struct nlattr
 				  type, OVS_CT_ATTR_MAX);
 			return -EINVAL;
 		}
+
+		maxlen = ovs_ct_attr_lens[type].maxlen;
+		minlen = ovs_ct_attr_lens[type].minlen;
 		if (nla_len(a) < minlen || nla_len(a) > maxlen) {
 			OVS_NLERR(log,
 				  "Conntrack attr type has unexpected length (type=%d, length=%d, expected=%d)",

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

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

From: WANG Cong <[email protected]>

[ Upstream commit c800aaf8d869f2b9b47b10c5c312fe19f0a94042 ]

There are multiple reports showing we have a use-after-free in the timer prb_retire_rx_blk_timer_expired(), where we use struct tpacket_kbdq_core::pkbdq, a pg_vec, after it gets freed by free_pg_vec().

The interesting part is it is not freed via packet_release() but via packet_setsockopt(), which means we are not closing the socket. Looking into the big and fat function packet_set_ring(), this could happen if we satisfy the following conditions:

1. closing == 0, not on packet_release() path
2. req->tp_block_nr == 0, we don't allocate a new pg_vec
3. rx_ring->pg_vec is already set as V3, which means we already called packet_set_ring() wtih req->tp_block_nr > 0 previously
4. req->tp_frame_nr == 0, pass sanity check
5. po->mapped == 0, never called mmap()

In this scenario we are clearing the old rx_ring->pg_vec, so we need to free this pg_vec, but we don't stop the timer on this path because of closing==0.

The timer has to be stopped as long as we need to free pg_vec, therefore the check on closing!=0 is wrong, we should check pg_vec!=NULL instead.

Thanks to liujian for testing different fixes.

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

@@ -4225,7 +4225,7 @@ static int packet_set_ring(struct sock *
 		register_prot_hook(sk);
 	}
 	spin_unlock(&po->bind_lock);
-	if (closing && (po->tp_version > TPACKET_V2)) {
+	if (pg_vec && (po->tp_version > TPACKET_V2)) {
 		/* Because we don't support block-based V3 on tx-ring */
 		if (!tx_ring)
 			prb_shutdown_retire_blk_timer(po, rb_queue);

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

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

From: Stefano Brivio <[email protected]>

[ Upstream commit afce615aaabfbaad02550e75c0bec106dafa1adf ]

RFC 2465 defines ipv6IfStatsOutFragFails as:

"The number of IPv6 datagrams that have been discarded
because they needed to be fragmented at this output interface but could not be."

The existing implementation, instead, would increase the counter twice in case we fail to allocate room for single fragments:
once for the fragment, once for the datagram.

This didn't look intentional though. In one of the two affected affected failure paths, the double increase was simply a result of a new 'goto fail' statement, introduced to avoid a skb leak. The other path appears to be affected since at least 2.6.12-rc2.

---
 net/ipv6/ip6_output.c |    4 ----
 1 file changed, 4 deletions(-)

@@ -647,8 +647,6 @@ int ip6_fragment(struct net *net, struct
 		*prevhdr = NEXTHDR_FRAGMENT;
 		tmp_hdr = kmemdup(skb_network_header(skb), hlen, GFP_ATOMIC);
 		if (!tmp_hdr) {
-			IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)),
-				      IPSTATS_MIB_FRAGFAILS);
 			err = -ENOMEM;
 			goto fail;
 		}

@@ -767,8 +765,6 @@ slow_path:
 		frag = alloc_skb(len + hlen + sizeof(struct frag_hdr) +
 				 hroom + troom, GFP_ATOMIC);
 		if (!frag) {
-			IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)),
-				      IPSTATS_MIB_FRAGFAILS);
 			err = -ENOMEM;
 			goto fail;
 		}

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

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

From: Marc Gonzalez <[email protected]>

[ Upstream commit 4813497b537c6208c90d6cbecac5072d347de900 ]

Before commit bf8f6952a233 ("Add blurb about RGMII") it was unclear whose responsibility it was to insert the required clock skew, and in hindsight, some PHY drivers got it wrong. The solution forward is to introduce a new property, explicitly requiring skew from the node to which it is attached. In the interim, this driver will handle all 4 RGMII modes identically (no skew).

---
 drivers/net/ethernet/aurora/nb8800.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

@@ -608,7 +608,7 @@ static void nb8800_mac_config(struct net
 		mac_mode |= HALF_DUPLEX;
 
 	if (gigabit) {
-		if (priv->phy_mode == PHY_INTERFACE_MODE_RGMII)
+		if (phy_interface_is_rgmii(dev->phydev))
 			mac_mode |= RGMII_MODE;
 
 		mac_mode |= GMAC_MODE;

@@ -1295,11 +1295,10 @@ static int nb8800_tangox_init(struct net
 		break;
 
 	case PHY_INTERFACE_MODE_RGMII:
-		pad_mode = PAD_MODE_RGMII;
-		break;
-
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
 	case PHY_INTERFACE_MODE_RGMII_TXID:
-		pad_mode = PAD_MODE_RGMII | PAD_MODE_GTX_CLK_DELAY;
+		pad_mode = PAD_MODE_RGMII;
 		break;
 
 	default:

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

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

From: Xin Long <[email protected]>

[ Upstream commit 0c2232b0a71db0ac1d22f751aa1ac0cadb950fd2 ]

In dccp_v6_conn_request, after reqsk gets alloced and hashed into ehash table, reqsk's refcnt is set 3. one is for req->rsk_timer, one is for hlist, and the other one is for current using.

The problem is when dccp_v6_conn_request returns and finishes using reqsk, it doesn't put reqsk. This will cause reqsk refcnt leaks and reqsk obj never gets freed.

Jianlin found this issue when running dccp_memleak.c in a loop, the system memory would run out.

dccp_memleak.c:
int s1 = socket(PF_INET6, 6, IPPROTO_IP);
bind(s1, &sa1, 0x20);
listen(s1, 0x9);
int s2 = socket(PF_INET6, 6, IPPROTO_IP);
connect(s2, &sa1, 0x20);
close(s1);
close(s2);

This patch is to put the reqsk before dccp_v6_conn_request returns, just as what tcp_conn_request does.

---
 net/dccp/ipv6.c |    1 +
 1 file changed, 1 insertion(+)

@@ -376,6 +376,7 @@ static int dccp_v6_conn_request(struct s
 		goto drop_and_free;
 
 	inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
+	reqsk_put(req);
 	return 0;
 
 drop_and_free:

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

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

From: Xin Long <[email protected]>

[ Upstream commit b7953d3c0e30a5fc944f6b7bd0bcceb0794bcd85 ]

The patch "dccp: fix a memleak that dccp_ipv6 doesn't put reqsk properly" fixed reqsk refcnt leak for dccp_ipv6. The same issue exists on dccp_ipv4.

This patch is to fix it for dccp_ipv4.

---
 net/dccp/ipv4.c |    1 +
 1 file changed, 1 insertion(+)

@@ -635,6 +635,7 @@ int dccp_v4_conn_request(struct sock *sk
 		goto drop_and_free;
 
 	inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
+	reqsk_put(req);
 	return 0;
 
 drop_and_free:

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

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

From: Xin Long <[email protected]>

[ Upstream commit e90ce2fc27cad7e7b1e72b9e66201a7a4c124c2b ]

In dccp_feat_init, when ccid_get_builtin_ccids failsto alloc memory for rx.val, it should free tx.val before returning an error.

---
 net/dccp/feat.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

@@ -1471,9 +1471,12 @@ int dccp_feat_init(struct sock *sk)
 	 * singleton values (which always leads to failure).
 	 * These settings can still (later) be overridden via sockopts.
 	 */
-	if (ccid_get_builtin_ccids(&tx.val, &tx.len) ||
-	    ccid_get_builtin_ccids(&rx.val, &rx.len))
+	if (ccid_get_builtin_ccids(&tx.val, &tx.len))
 		return -ENOBUFS;
+	if (ccid_get_builtin_ccids(&rx.val, &rx.len)) {
+		kfree(tx.val);
+		return -ENOBUFS;
+	}
 
 	if (!dccp_feat_prefer(sysctl_dccp_tx_ccid, tx.val, tx.len) ||
 	    !dccp_feat_prefer(sysctl_dccp_rx_ccid, rx.val, rx.len))

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

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

From: Alexander Potapenko <[email protected]>

[ Upstream commit b1f5bfc27a19f214006b9b4db7b9126df2dfdf5a ]

If the length field of the iterator (|pos.p| or |err|) is past the end of the chunk, we shouldn't access it.

This bug has been detected by KMSAN. For the following pair of system calls:

socket(PF_INET6, SOCK_STREAM, 0x84 /* IPPROTO_??? */) = 3 sendto(3, "A", 1, MSG_OOB, {sa_family=AF_INET6, sin6_port=htons(0),

         inet_pton(AF_INET6, "::1", &sin6_addr), sin6_flowinfo=0,
         sin6_scope_id=0}, 28) = 1

the tool has reported a use of uninitialized memory:

==================================================================

  BUG: KMSAN: use of uninitialized memory in sctp_rcv+0x17b8/0x43b0
  CPU: 1 PID: 2940 Comm: probe Not tainted 4.11.0-rc5+ #2926
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs
  01/01/2011
  Call Trace:
   <IRQ>
   __dump_stack lib/dump_stack.c:16
   dump_stack+0x172/0x1c0 lib/dump_stack.c:52
   kmsan_report+0x12a/0x180 mm/kmsan/kmsan.c:927
   __msan_warning_32+0x61/0xb0 mm/kmsan/kmsan_instr.c:469
   __sctp_rcv_init_lookup net/sctp/input.c:1074
   __sctp_rcv_lookup_harder net/sctp/input.c:1233
   __sctp_rcv_lookup net/sctp/input.c:1255
   sctp_rcv+0x17b8/0x43b0 net/sctp/input.c:170
   sctp6_rcv+0x32/0x70 net/sctp/ipv6.c:984
   ip6_input_finish+0x82f/0x1ee0 net/ipv6/ip6_input.c:279
   NF_HOOK ./include/linux/netfilter.h:257
   ip6_input+0x239/0x290 net/ipv6/ip6_input.c:322
   dst_input ./include/net/dst.h:492
   ip6_rcv_finish net/ipv6/ip6_input.c:69
   NF_HOOK ./include/linux/netfilter.h:257
   ipv6_rcv+0x1dbd/0x22e0 net/ipv6/ip6_input.c:203
   __netif_receive_skb_core+0x2f6f/0x3a20 net/core/dev.c:4208
   __netif_receive_skb net/core/dev.c:4246
   process_backlog+0x667/0xba0 net/core/dev.c:4866
   napi_poll net/core/dev.c:5268
   net_rx_action+0xc95/0x1590 net/core/dev.c:5333
   __do_softirq+0x485/0x942 kernel/softirq.c:284
   do_softirq_own_stack+0x1c/0x30 arch/x86/entry/entry_64.S:902
   </IRQ>
   do_softirq kernel/softirq.c:328
   __local_bh_enable_ip+0x25b/0x290 kernel/softirq.c:181
   local_bh_enable+0x37/0x40 ./include/linux/bottom_half.h:31
   rcu_read_unlock_bh ./include/linux/rcupdate.h:931
   ip6_finish_output2+0x19b2/0x1cf0 net/ipv6/ip6_output.c:124
   ip6_finish_output+0x764/0x970 net/ipv6/ip6_output.c:149
   NF_HOOK_COND ./include/linux/netfilter.h:246
   ip6_output+0x456/0x520 net/ipv6/ip6_output.c:163
   dst_output ./include/net/dst.h:486
   NF_HOOK ./include/linux/netfilter.h:257
   ip6_xmit+0x1841/0x1c00 net/ipv6/ip6_output.c:261
   sctp_v6_xmit+0x3b7/0x470 net/sctp/ipv6.c:225
   sctp_packet_transmit+0x38cb/0x3a20 net/sctp/output.c:632
   sctp_outq_flush+0xeb3/0x46e0 net/sctp/outqueue.c:885
   sctp_outq_uncork+0xb2/0xd0 net/sctp/outqueue.c:750
   sctp_side_effects net/sctp/sm_sideeffect.c:1773
   sctp_do_sm+0x6962/0x6ec0 net/sctp/sm_sideeffect.c:1147
   sctp_primitive_ASSOCIATE+0x12c/0x160 net/sctp/primitive.c:88
   sctp_sendmsg+0x43e5/0x4f90 net/sctp/socket.c:1954
   inet_sendmsg+0x498/0x670 net/ipv4/af_inet.c:762
   sock_sendmsg_nosec net/socket.c:633
   sock_sendmsg net/socket.c:643
   SYSC_sendto+0x608/0x710 net/socket.c:1696
   SyS_sendto+0x8a/0xb0 net/socket.c:1664
   do_syscall_64+0xe6/0x130 arch/x86/entry/common.c:285
   entry_SYSCALL64_slow_path+0x25/0x25 arch/x86/entry/entry_64.S:246
  RIP: 0033:0x401133
  RSP: 002b:00007fff6d99cd38 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
  RAX: ffffffffffffffda RBX: 00000000004002b0 RCX: 0000000000401133
  RDX: 0000000000000001 RSI: 0000000000494088 RDI: 0000000000000003
  RBP: 00007fff6d99cd90 R08: 00007fff6d99cd50 R09: 000000000000001c
  R10: 0000000000000001 R11: 0000000000000246 R12: 0000000000000000
  R13: 00000000004063d0 R14: 0000000000406460 R15: 0000000000000000
  origin:
   save_stack_trace+0x37/0x40 arch/x86/kernel/stacktrace.c:59
   kmsan_save_stack_with_flags mm/kmsan/kmsan.c:302
   kmsan_internal_poison_shadow+0xb1/0x1a0 mm/kmsan/kmsan.c:198
   kmsan_poison_shadow+0x6d/0xc0 mm/kmsan/kmsan.c:211
   slab_alloc_node mm/slub.c:2743
   __kmalloc_node_track_caller+0x200/0x360 mm/slub.c:4351
   __kmalloc_reserve net/core/skbuff.c:138
   __alloc_skb+0x26b/0x840 net/core/skbuff.c:231
   alloc_skb ./include/linux/skbuff.h:933
   sctp_packet_transmit+0x31e/0x3a20 net/sctp/output.c:570
   sctp_outq_flush+0xeb3/0x46e0 net/sctp/outqueue.c:885
   sctp_outq_uncork+0xb2/0xd0 net/sctp/outqueue.c:750
   sctp_side_effects net/sctp/sm_sideeffect.c:1773
   sctp_do_sm+0x6962/0x6ec0 net/sctp/sm_sideeffect.c:1147
   sctp_primitive_ASSOCIATE+0x12c/0x160 net/sctp/primitive.c:88
   sctp_sendmsg+0x43e5/0x4f90 net/sctp/socket.c:1954
   inet_sendmsg+0x498/0x670 net/ipv4/af_inet.c:762
   sock_sendmsg_nosec net/socket.c:633
   sock_sendmsg net/socket.c:643
   SYSC_sendto+0x608/0x710 net/socket.c:1696
   SyS_sendto+0x8a/0xb0 net/socket.c:1664
   do_syscall_64+0xe6/0x130 arch/x86/entry/common.c:285
   return_from_SYSCALL_64+0x0/0x6a arch/x86/entry/entry_64.S:246
  ==================================================================

---
 include/net/sctp/sctp.h |    4 ++++
 1 file changed, 4 insertions(+)

@@ -444,6 +444,8 @@ _sctp_walk_params((pos), (chunk), ntohs(
 
 #define _sctp_walk_params(pos, chunk, end, member)\
 for (pos.v = chunk->member;\
+     (pos.v + offsetof(struct sctp_paramhdr, length) + sizeof(pos.p->length) <\
+      (void *)chunk + end) &&\
      pos.v <= (void *)chunk + end - ntohs(pos.p->length) &&\
      ntohs(pos.p->length) >= sizeof(sctp_paramhdr_t);\
      pos.v += WORD_ROUND(ntohs(pos.p->length)))

@@ -454,6 +456,8 @@ _sctp_walk_errors((err), (chunk_hdr), nt
 #define _sctp_walk_errors(err, chunk_hdr, end)\
 for (err = (sctp_errhdr_t *)((void *)chunk_hdr + \
 	    sizeof(sctp_chunkhdr_t));\
+     ((void *)err + offsetof(sctp_errhdr_t, length) + sizeof(err->length) <\
+      (void *)chunk_hdr + end) &&\
      (void *)err <= (void *)chunk_hdr + end - ntohs(err->length) &&\
      ntohs(err->length) >= sizeof(sctp_errhdr_t); \
      err = (sctp_errhdr_t *)((void *)err + WORD_ROUND(ntohs(err->length))))

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

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

From: Xin Long <[email protected]>

[ Upstream commit 6b84202c946cd3da3a8daa92c682510e9ed80321 ]

Commit b1f5bfc27a19 ("sctp: don't dereference ptr before leaving _sctp_walk_{params, errors}()") tried to fix the issue that it may overstep the chunk end for _sctp_walk_{params, errors} with 'chunk_end > offset(length) + sizeof(length)'.

But it introduced a side effect: When processing INIT, it verifies the chunks with 'param.v == chunk_end' after iterating all params by sctp_walk_params(). With the check 'chunk_end > offset(length)
+ sizeof(length)', it would return when the last param is not yet accessed. Because the last param usually is fwdtsn supported param whose size is 4 and 'chunk_end == offset(length) + sizeof(length)'

This is a badly issue even causing sctp couldn't process 4-shakes. Client would always get abort when connecting to server, due to the failure of INIT chunk verification on server.

The patch is to use 'chunk_end <= offset(length) + sizeof(length)' instead of 'chunk_end < offset(length) + sizeof(length)' for both _sctp_walk_params and _sctp_walk_errors.

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

@@ -444,7 +444,7 @@ _sctp_walk_params((pos), (chunk), ntohs(
 
 #define _sctp_walk_params(pos, chunk, end, member)\
 for (pos.v = chunk->member;\
-     (pos.v + offsetof(struct sctp_paramhdr, length) + sizeof(pos.p->length) <\
+     (pos.v + offsetof(struct sctp_paramhdr, length) + sizeof(pos.p->length) <=\
       (void *)chunk + end) &&\
      pos.v <= (void *)chunk + end - ntohs(pos.p->length) &&\
      ntohs(pos.p->length) >= sizeof(sctp_paramhdr_t);\

@@ -456,7 +456,7 @@ _sctp_walk_errors((err), (chunk_hdr), nt
 #define _sctp_walk_errors(err, chunk_hdr, end)\
 for (err = (sctp_errhdr_t *)((void *)chunk_hdr + \
 	    sizeof(sctp_chunkhdr_t));\
-     ((void *)err + offsetof(sctp_errhdr_t, length) + sizeof(err->length) <\
+     ((void *)err + offsetof(sctp_errhdr_t, length) + sizeof(err->length) <=\
       (void *)chunk_hdr + end) &&\
      (void *)err <= (void *)chunk_hdr + end - ntohs(err->length) &&\
      ntohs(err->length) >= sizeof(sctp_errhdr_t); \

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

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

From: Moshe Shemesh <[email protected]>

[ Upstream commit 219c81f7d1d5a89656cb3b53d3b4e11e93608d80 ]

When driver fail to allocate an entry to send command to FW, it must notify the calling function and release the memory allocated for this command.

---
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c |   19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

@@ -630,6 +630,10 @@ static void dump_command(struct mlx5_cor
 		pr_debug("\n");
 }
 
+static void free_msg(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *msg);
+static void mlx5_free_cmd_msg(struct mlx5_core_dev *dev,
+			      struct mlx5_cmd_msg *msg);
+
 static void cmd_work_handler(struct work_struct *work)
 {
 	struct mlx5_cmd_work_ent *ent = container_of(work, struct mlx5_cmd_work_ent, work);

@@ -638,16 +642,27 @@ static void cmd_work_handler(struct work
 	struct mlx5_cmd_layout *lay;
 	struct semaphore *sem;
 	unsigned long flags;
+	int alloc_ret;
 
 	sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem;
 	down(sem);
 	if (!ent->page_queue) {
-		ent->idx = alloc_ent(cmd);
-		if (ent->idx < 0) {
+		alloc_ret = alloc_ent(cmd);
+		if (alloc_ret < 0) {
+			if (ent->callback) {
+				ent->callback(-EAGAIN, ent->context);
+				mlx5_free_cmd_msg(dev, ent->out);
+				free_msg(dev, ent->in);
+				free_cmd(ent);
+			} else {
+				ent->ret = -EAGAIN;
+				complete(&ent->done);
+			}
 			mlx5_core_err(dev, "failed to allocate command entry\n");
 			up(sem);
 			return;
 		}
+		ent->idx = alloc_ret;
 	} else {
 		ent->idx = cmd->max_reg_cmds;
 		spin_lock_irqsave(&cmd->alloc_lock, flags);

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

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

From: Florian Fainelli <[email protected]>

[ Upstream commit 7ad813f208533cebfcc32d3d7474dc1677d1b09a ]

Marc reported that he was not getting the PHY library adjust_link() callback function to run when calling phy_stop() + phy_disconnect() which does not indeed happen because we set the state machine to PHY_HALTED but we don't get to run it to process this state past that point.

Fix this with a synchronous call to phy_state_machine() in order to have the state machine actually act on PHY_HALTED, set the PHY device's link down, turn the network device's carrier off and finally call the adjust_link() function.

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

@@ -541,6 +541,9 @@ void phy_stop_machine(struct phy_device
 	if (phydev->state > PHY_UP && phydev->state != PHY_HALTED)
 		phydev->state = PHY_UP;
 	mutex_unlock(&phydev->lock);
+
+	/* Now we can run the state machine synchronously */
+	phy_state_machine(&phydev->state_queue.work);
 }
 
 /**

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

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

From: Wei Liu <[email protected]>

[ Upstream commit dfa523ae9f2542bee4cddaea37b3be3e157f6e6b ]

Add a flag to indicate if a queue is rate-limited. Test the flag in NAPI poll handler and avoid rescheduling the queue if true, otherwise we risk locking up the host. The rescheduling will be done in the timer callback function.

---
 drivers/net/xen-netback/common.h    |    1 +
 drivers/net/xen-netback/interface.c |    6 +++++-
 drivers/net/xen-netback/netback.c   |    6 +++++-
 3 files changed, 11 insertions(+), 2 deletions(-)

@@ -201,6 +201,7 @@ struct xenvif_queue { /* Per-queue data
 	unsigned long   remaining_credit;
 	struct timer_list credit_timeout;
 	u64 credit_window_start;
+	bool rate_limited;
 
 	/* Statistics */
 	struct xenvif_stats stats;

@@ -105,7 +105,11 @@ static int xenvif_poll(struct napi_struc
 
 	if (work_done < budget) {
 		napi_complete(napi);
-		xenvif_napi_schedule_or_enable_events(queue);
+		/* If the queue is rate-limited, it shall be
+		 * rescheduled in the timer callback.
+		 */
+		if (likely(!queue->rate_limited))
+			xenvif_napi_schedule_or_enable_events(queue);
 	}
 
 	return work_done;

@@ -687,6 +687,7 @@ static void tx_add_credit(struct xenvif_
 		max_credit = ULONG_MAX; /* wrapped: clamp to ULONG_MAX */
 
 	queue->remaining_credit = min(max_credit, max_burst);
+	queue->rate_limited = false;
 }
 
 void xenvif_tx_credit_callback(unsigned long data)

@@ -1184,8 +1185,10 @@ static bool tx_credit_exceeded(struct xe
 		msecs_to_jiffies(queue->credit_usec / 1000);
 
 	/* Timer could already be pending in rare cases. */
-	if (timer_pending(&queue->credit_timeout))
+	if (timer_pending(&queue->credit_timeout)) {
+		queue->rate_limited = true;
 		return true;
+	}
 
 	/* Passed the point where we can replenish credit? */
 	if (time_after_eq64(now, next_credit)) {

@@ -1200,6 +1203,7 @@ static bool tx_credit_exceeded(struct xe
 		mod_timer(&queue->credit_timeout,
 			  next_credit);
 		queue->credit_window_start = next_credit;
+		queue->rate_limited = true;
 
 		return true;
 	}

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

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

From: Jane Chu <[email protected]>

[ Upstream commit 9d53caec84c7c5700e7c1ed744ea584fff55f9ac ]

A large sun4v SPARC system may have moments of intensive xcall activities, usually caused by unmapping many pages on many CPUs concurrently. This can flood receivers with CPU mondo interrupts for an extended period, causing some unlucky senders to hit send-mondo timeout. This problem gets worse as cpu count increases because sometimes mappings must be invalidated on all CPUs, and sometimes all CPUs may gang up on a single CPU.

But a busy system is not a broken system. In the above scenario, as long as the receiver is making forward progress processing mondo interrupts, the sender should continue to retry.

This patch implements the receiver's forward progress meter by introducing a per cpu counter 'cpu_mondo_counter[cpu]' where 'cpu' is in the range of 0..NR_CPUS. The receiver increments its counter as soon as it receives a mondo and the sender tracks the receiver's counter. If the receiver has stopped making forward progress when the retry limit is reached, the sender declares send-mondo-timeout and panic; otherwise, the receiver is allowed to keep making forward progress.

In addition, it's been observed that PCIe hotplug events generate Correctable Errors that are handled by hypervisor and then OS. Hypervisor 'borrows' a guest cpu strand briefly to provide the service. If the cpu strand is simultaneously the only cpu targeted by a mondo, it may not be available for the mondo in 20msec, causing SUN4V mondo timeout. It appears that 1 second is the agreed wait time between hypervisor and guest OS, this patch makes the adjustment.

Orabug: 25476541
Orabug: 26417466

---
 arch/sparc/include/asm/trap_block.h |    1 
 arch/sparc/kernel/smp_64.c          |  189 ++++++++++++++++++++++--------------
 arch/sparc/kernel/sun4v_ivec.S      |   15 ++
 arch/sparc/kernel/traps_64.c        |    1 
 4 files changed, 134 insertions(+), 72 deletions(-)

@@ -54,6 +54,7 @@ extern struct trap_per_cpu trap_block[NR
 void init_cur_cpu_trap(struct thread_info *);
 void setup_tba(void);
 extern int ncpus_probed;
+extern u64 cpu_mondo_counter[NR_CPUS];
 
 unsigned long real_hard_smp_processor_id(void);
 

@@ -26,6 +26,21 @@ sun4v_cpu_mondo:
 	ldxa	[%g0] ASI_SCRATCHPAD, %g4
 	sub	%g4, TRAP_PER_CPU_FAULT_INFO, %g4
 
+	/* Get smp_processor_id() into %g3 */
+	sethi	%hi(trap_block), %g5
+	or	%g5, %lo(trap_block), %g5
+	sub	%g4, %g5, %g3
+	srlx	%g3, TRAP_BLOCK_SZ_SHIFT, %g3
+
+	/* Increment cpu_mondo_counter[smp_processor_id()] */
+	sethi	%hi(cpu_mondo_counter), %g5
+	or	%g5, %lo(cpu_mondo_counter), %g5
+	sllx	%g3, 3, %g3
+	add	%g5, %g3, %g5
+	ldx	[%g5], %g3
+	add	%g3, 1, %g3
+	stx	%g3, [%g5]
+
 	/* Get CPU mondo queue base phys address into %g7.  */
 	ldx	[%g4 + TRAP_PER_CPU_CPU_MONDO_PA], %g7
 

@@ -2659,6 +2659,7 @@ void do_getpsr(struct pt_regs *regs)
 	}
 }
 
+u64 cpu_mondo_counter[NR_CPUS] = {0};
 struct trap_per_cpu trap_block[NR_CPUS];
 EXPORT_SYMBOL(trap_block);
 

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

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

From: Rob Gardner <[email protected]>

[ Upstream commit fc290a114fc6034b0f6a5a46e2fb7d54976cf87a ]

This fixes another cause of random segfaults and bus errors that may occur while running perf with the callgraph option.

Critical sections beginning with spin_lock_irqsave() raise the interrupt level to PIL_NORMAL_MAX (14) and intentionally do not block performance counter interrupts, which arrive at PIL_NMI (15).

But some sections of code are "super critical" with respect to perf because the perf_callchain_user() path accesses user space and may cause TLB activity as well as faults as it unwinds the user stack.

One particular critical section occurs in switch_mm:

        spin_lock_irqsave(&mm->context.lock, flags);
        ...
        load_secondary_context(mm);
        tsb_context_switch(mm);
        ...
        spin_unlock_irqrestore(&mm->context.lock, flags);

If a perf interrupt arrives in between load_secondary_context() and tsb_context_switch(), then perf_callchain_user() could execute with the context ID of one process, but with an active TSB for a different process. When the user stack is accessed, it is very likely to incur a TLB miss, since the h/w context ID has been changed. The TLB will then be reloaded with a translation from the TSB for one process, but using a context ID for another process. This exposes memory from one process to another, and since it is a mapping for stack memory, this usually causes the new process to crash quickly.

This super critical section needs more protection than is provided by spin_lock_irqsave() since perf interrupts must not be allowed in.

Since __tsb_context_switch already goes through the trouble of disabling interrupts completely, we fix this by moving the secondary context load down into this better protected region.

Orabug: 25577560

---
 arch/sparc/include/asm/mmu_context_64.h |   12 +++++++-----
 arch/sparc/kernel/tsb.S                 |   12 ++++++++++++
 arch/sparc/power/hibernate.c            |    3 +--
 3 files changed, 20 insertions(+), 7 deletions(-)

@@ -25,9 +25,11 @@ void destroy_context(struct mm_struct *m
 void __tsb_context_switch(unsigned long pgd_pa,
 			  struct tsb_config *tsb_base,
 			  struct tsb_config *tsb_huge,
-			  unsigned long tsb_descr_pa);
+			  unsigned long tsb_descr_pa,
+			  unsigned long secondary_ctx);
 
-static inline void tsb_context_switch(struct mm_struct *mm)
+static inline void tsb_context_switch_ctx(struct mm_struct *mm,
+					  unsigned long ctx)
 {
 	__tsb_context_switch(__pa(mm->pgd),
 			     &mm->context.tsb_block[0],

@@ -38,7 +40,8 @@ static inline void tsb_context_switch(st
 #else
 			     NULL
 #endif
-			     , __pa(&mm->context.tsb_descr[0]));
+			     , __pa(&mm->context.tsb_descr[0]),
+			     ctx);
 }
 
 void tsb_grow(struct mm_struct *mm,

@@ -110,8 +113,7 @@ static inline void switch_mm(struct mm_s
 	 * cpu0 to update it's TSB because at that point the cpu_vm_mask
 	 * only had cpu1 set in it.
 	 */
-	load_secondary_context(mm);
-	tsb_context_switch(mm);
+	tsb_context_switch_ctx(mm, CTX_HWBITS(mm->context));
 
 	/* Any time a processor runs a context on an address space
 	 * for the first time, we must flush that context out of the

@@ -375,6 +375,7 @@ tsb_flush:
 	 * %o1:	TSB base config pointer
 	 * %o2:	TSB huge config pointer, or NULL if none
 	 * %o3:	Hypervisor TSB descriptor physical address
+	 * %o4: Secondary context to load, if non-zero
 	 *
 	 * We have to run this whole thing with interrupts
 	 * disabled so that the current cpu doesn't change

@@ -387,6 +388,17 @@ __tsb_context_switch:
 	rdpr	%pstate, %g1
 	wrpr	%g1, PSTATE_IE, %pstate
 
+	brz,pn	%o4, 1f
+	 mov	SECONDARY_CONTEXT, %o5
+
+661:	stxa	%o4, [%o5] ASI_DMMU
+	.section .sun4v_1insn_patch, "ax"
+	.word	661b
+	stxa	%o4, [%o5] ASI_MMU
+	.previous
+	flush	%g6
+
+1:
 	TRAP_LOAD_TRAP_BLOCK(%g2, %g3)
 
 	stx	%o0, [%g2 + TRAP_PER_CPU_PGD_PADDR]

@@ -35,6 +35,5 @@ void restore_processor_state(void)
 {
 	struct mm_struct *mm = current->active_mm;
 
-	load_secondary_context(mm);
-	tsb_context_switch(mm);
+	tsb_context_switch_ctx(mm, CTX_HWBITS(mm->context));
 }

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

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

From: Arnd Bergmann <[email protected]>

commit 93be2b74279c15c2844684b1a027fdc71dd5d9bf upstream.

gcc-7 complains that wl3501_cs passes NULL into a function that then uses the argument as the input for memcpy:

drivers/net/wireless/wl3501_cs.c: In function 'wl3501_get_scan':
include/net/iw_handler.h:559:3: error: argument 2 null where non-null expected [-Werror=nonnull]
memcpy(stream + point_len, extra, iwe->u.data.length);

This works fine here because iwe->u.data.length is guaranteed to be 0 and the memcpy doesn't actually have an effect.

Making the length check explicit avoids the warning and should have no other effect here.

Also check the pointer itself, since otherwise we get warnings elsewhere in the code.

---
 include/net/iw_handler.h |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -556,7 +556,8 @@ iwe_stream_add_point(struct iw_request_i
 		memcpy(stream + lcp_len,
 		       ((char *) &iwe->u) + IW_EV_POINT_OFF,
 		       IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN);
-		memcpy(stream + point_len, extra, iwe->u.data.length);
+		if (iwe->u.data.length && extra)
+			memcpy(stream + point_len, extra, iwe->u.data.length);
 		stream += event_len;
 	}
 	return stream;

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

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

From: Sergei Shtylyov <[email protected]>

[ Upstream commit 0f1f9cbc04dbb3cc310f70a11cba0cf1f2109d9c ]

The R8A7740 GEther controller supports the packet checksum offloading but the 'hw_crc' (bad name, I'll fix it) flag isn't set in the R8A7740 data, thus CSMR isn't cleared...

---
 drivers/net/ethernet/renesas/sh_eth.c |    1 +
 1 file changed, 1 insertion(+)

@@ -819,6 +819,7 @@ static struct sh_eth_cpu_data r8a7740_da
 	.rpadir_value   = 2 << 16,
 	.no_trimd	= 1,
 	.no_ade		= 1,
+	.hw_crc		= 1,
 	.tsu		= 1,
 	.select_mii	= 1,
 	.shift_rd0	= 1,

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

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

From: Grygorii Strashko <[email protected]>

[ Upstream commit 5ca7d1ca77dc23934504b95a96d2660d345f83c2 ]

For proper IRQ generation by DP83867 phy the INT/PWDN pin has to be programmed as an interrupt output instead of a Powerdown input in Configuration Register 3 (CFG3), Address 0x001E, bit 7 INT_OE = 1. The current driver doesn't do this and as result IRQs will not be generated by DP83867 phy even if they are properly configured in DT.

Hence, fix IRQ generation by properly configuring CFG3.INT_OE bit and ensure that Link Status Change (LINK_STATUS_CHNG_INT) and Auto-Negotiation Complete (AUTONEG_COMP_INT) interrupt are enabled. After this the DP83867 driver will work properly in interrupt enabled mode.

---
 drivers/net/phy/dp83867.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

@@ -29,6 +29,7 @@
 #define MII_DP83867_MICR	0x12
 #define MII_DP83867_ISR		0x13
 #define DP83867_CTRL		0x1f
+#define DP83867_CFG3		0x1e
 
 /* Extended Registers */
 #define DP83867_RGMIICTL	0x0032

@@ -89,6 +90,8 @@ static int dp83867_config_intr(struct ph
 		micr_status |=
 			(MII_DP83867_MICR_AN_ERR_INT_EN |
 			MII_DP83867_MICR_SPEED_CHNG_INT_EN |
+			MII_DP83867_MICR_AUTONEG_COMP_INT_EN |
+			MII_DP83867_MICR_LINK_STS_CHNG_INT_EN |
 			MII_DP83867_MICR_DUP_MODE_CHNG_INT_EN |
 			MII_DP83867_MICR_SLEEP_MODE_CHNG_INT_EN);
 

@@ -184,6 +187,13 @@ static int dp83867_config_init(struct ph
 				       DP83867_DEVADDR, phydev->addr, delay);
 	}
 
+	/* Enable Interrupt output INT_OE in CFG3 register */
+	if (phy_interrupt_is_valid(phydev)) {
+		val = phy_read(phydev, DP83867_CFG3);
+		val |= BIT(7);
+		phy_write(phydev, DP83867_CFG3, val);
+	}
+
 	return 0;
 }
 

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

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

From: Michael Chan <[email protected]>

[ Upstream commit f5992b72ebe0dde488fa8f706b887194020c66fc ]

The driver's ndo_get_stats64() method is not always called under RTNL. So it can race with driver close or ethtool reconfigurations. Fix the race condition by taking tp->lock spinlock in tg3_free_consistent() when freeing the tp->hw_stats memory block. tg3_get_stats64() is already taking tp->lock.

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

@@ -8722,11 +8722,14 @@ static void tg3_free_consistent(struct t
 	tg3_mem_rx_release(tp);
 	tg3_mem_tx_release(tp);
 
+	/* Protect tg3_get_stats64() from reading freed tp->hw_stats. */
+	tg3_full_lock(tp, 0);
 	if (tp->hw_stats) {
 		dma_free_coherent(&tp->pdev->dev, sizeof(struct tg3_hw_stats),
 				  tp->hw_stats, tp->stats_mapping);
 		tp->hw_stats = NULL;
 	}
+	tg3_full_unlock(tp);
 }
 
 /*

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

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

From: Nicholas Mc Guire <[email protected]>

[ Upstream commit fac69d0efad08fc15e4dbfc116830782acc0dc9a ]

Add the missing declarations of basic string functions to string.h to allow a clean build.

---
 arch/x86/boot/string.c |    1 +
 arch/x86/boot/string.h |    9 +++++++++
 2 files changed, 10 insertions(+)

@@ -14,6 +14,7 @@
 
 #include <linux/types.h>
 #include "ctype.h"
+#include "string.h"
 
 int memcmp(const void *s1, const void *s2, size_t len)
 {

@@ -18,4 +18,13 @@ int memcmp(const void *s1, const void *s
 #define memset(d,c,l) __builtin_memset(d,c,l)
 #define memcmp	__builtin_memcmp
 
+extern int strcmp(const char *str1, const char *str2);
+extern int strncmp(const char *cs, const char *ct, size_t count);
+extern size_t strlen(const char *s);
+extern char *strstr(const char *s1, const char *s2);
+extern size_t strnlen(const char *s, size_t maxlen);
+extern unsigned int atou(const char *s);
+extern unsigned long long simple_strtoull(const char *cp, char **endp,
+					  unsigned int base);
+
 #endif /* BOOT_STRING_H */

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

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

From: Zefir Kurtisi <[email protected]>

[ Upstream commit 811a919135b980bac8009d042acdccf10dc1ef5e ]

While in RUNNING state, phy_state_machine() checks for link changes by comparing phydev->link before and after calling phy_read_status(). This works as long as it is guaranteed that phydev->link is never changed outside the phy_state_machine().

If in some setups this happens, it causes the state machine to miss a link loss and remain RUNNING despite phydev->link being 0.

This has been observed running a dsa setup with a process continuously polling the link states over ethtool each second (SNMPD RFC-1213 agent). Disconnecting the link on a phy followed by a ETHTOOL_GSET causes dsa_slave_get_settings() / dsa_slave_get_link_ksettings() to call phy_read_status() and with that modify the link status - and with that bricking the phy state machine.

This patch adds a fail-safe check while in RUNNING, which causes to move to CHANGELINK when the link is gone and we are still RUNNING.

---
 drivers/net/phy/phy.c |    9 +++++++++
 1 file changed, 9 insertions(+)

@@ -921,6 +921,15 @@ void phy_state_machine(struct work_struc
 			if (old_link != phydev->link)
 				phydev->state = PHY_CHANGELINK;
 		}
+		/*
+		 * Failsafe: check that nobody set phydev->link=0 between two
+		 * poll cycles, otherwise we won't leave RUNNING state as long
+		 * as link remains down.
+		 */
+		if (!phydev->link && phydev->state == PHY_RUNNING) {
+			phydev->state = PHY_CHANGELINK;
+			dev_err(&phydev->dev, "no link in PHY_RUNNING\n");
+		}
 		break;
 	case PHY_CHANGELINK:
 		err = phy_read_status(phydev);

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

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

From: "Milan P. Gandhi" <[email protected]>

[ Upstream commit c7702b8c22712a06080e10f1d2dee1a133ec8809 ]

There is a race condition with qla2xxx optrom functions where one thread might modify optrom buffer, optrom_state while other thread is still reading from it.

In couple of crashes, it was found that we had successfully passed the following 'if' check where we confirm optrom_state to be QLA_SREADING. But by the time we acquired mutex lock to proceed with memory_read_from_buffer function, some other thread/process had already modified that option rom buffer and optrom_state from QLA_SREADING to QLA_SWAITING. Then we got ha->optrom_buffer 0x0 and crashed the system:

        if (ha->optrom_state != QLA_SREADING)
                return 0;

        mutex_lock(&ha->optrom_mutex);
        rval = memory_read_from_buffer(buf, count, &off, ha->optrom_buffer,
            ha->optrom_region_size);
        mutex_unlock(&ha->optrom_mutex);

With current optrom function we get following crash due to a race condition:

[ 1479.466679] BUG: unable to handle kernel NULL pointer dereference at           (null)
[ 1479.466707] IP: [<ffffffff81326756>] memcpy+0x6/0x110
[...]
[ 1479.473673] Call Trace:
[ 1479.474296]  [<ffffffff81225cbc>] ? memory_read_from_buffer+0x3c/0x60
[ 1479.474941]  [<ffffffffa01574dc>] qla2x00_sysfs_read_optrom+0x9c/0xc0 [qla2xxx]
[ 1479.475571]  [<ffffffff8127e76b>] read+0xdb/0x1f0
[ 1479.476206]  [<ffffffff811fdf9e>] vfs_read+0x9e/0x170
[ 1479.476839]  [<ffffffff811feb6f>] SyS_read+0x7f/0xe0
[ 1479.477466]  [<ffffffff816964c9>] system_call_fastpath+0x16/0x1b

Below patch modifies qla2x00_sysfs_read_optrom, qla2x00_sysfs_write_optrom functions to get the mutex_lock before checking ha->optrom_state to avoid similar crashes.

The patch was applied and tested and same crashes were no longer observed again.

---
 drivers/scsi/qla2xxx/qla_attr.c |   18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

@@ -329,12 +329,15 @@ qla2x00_sysfs_read_optrom(struct file *f
 	struct qla_hw_data *ha = vha->hw;
 	ssize_t rval = 0;
 
+	mutex_lock(&ha->optrom_mutex);
+
 	if (ha->optrom_state != QLA_SREADING)
-		return 0;
+		goto out;
 
-	mutex_lock(&ha->optrom_mutex);
 	rval = memory_read_from_buffer(buf, count, &off, ha->optrom_buffer,
 	    ha->optrom_region_size);
+
+out:
 	mutex_unlock(&ha->optrom_mutex);
 
 	return rval;

@@ -349,14 +352,19 @@ qla2x00_sysfs_write_optrom(struct file *
 	    struct device, kobj)));
 	struct qla_hw_data *ha = vha->hw;
 
-	if (ha->optrom_state != QLA_SWRITING)
+	mutex_lock(&ha->optrom_mutex);
+
+	if (ha->optrom_state != QLA_SWRITING) {
+		mutex_unlock(&ha->optrom_mutex);
 		return -EINVAL;
-	if (off > ha->optrom_region_size)
+	}
+	if (off > ha->optrom_region_size) {
+		mutex_unlock(&ha->optrom_mutex);
 		return -ERANGE;
+	}
 	if (off + count > ha->optrom_region_size)
 		count = ha->optrom_region_size - off;
 
-	mutex_lock(&ha->optrom_mutex);
 	memcpy(&ha->optrom_buffer[off], buf, count);
 	mutex_unlock(&ha->optrom_mutex);
 

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

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

From: Gerd Hoffmann <[email protected]>

[ Upstream commit 71d3f6ef7f5af38dea2975ec5715c88bae92e92d ]

virtio uses normal ram as backing storage for the framebuffer, so we should assign the address to new screen_buffer (added by commit 17a7b0b4d9749f80d365d7baff5dec2f54b0e992) instead of screen_base.

---
 drivers/gpu/drm/virtio/virtgpu_fb.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -338,7 +338,7 @@ static int virtio_gpufb_create(struct dr
 	info->fbops = &virtio_gpufb_ops;
 	info->pixmap.flags = FB_PIXMAP_SYSTEM;
 
-	info->screen_base = obj->vmap;
+	info->screen_buffer = obj->vmap;
 	info->screen_size = obj->gem_base.size;
 	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(info, &vfbdev->helper,

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

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

From: Omar Sandoval <[email protected]>

[ Upstream commit 6bf6b0aa3da84a3d9126919a94c49c0fb7ee2fb3 ]

If blk_mq_init_queue() returns an error, it gets assigned to vblk->disk->queue. Then, when we call put_disk(), we end up calling blk_put_queue() with the ERR_PTR, causing a bad dereference. Fix it by only assigning to vblk->disk->queue on success.

---
 drivers/block/virtio_blk.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -641,11 +641,12 @@ static int virtblk_probe(struct virtio_d
 	if (err)
 		goto out_put_disk;
 
-	q = vblk->disk->queue = blk_mq_init_queue(&vblk->tag_set);
+	q = blk_mq_init_queue(&vblk->tag_set);
 	if (IS_ERR(q)) {
 		err = -ENOMEM;
 		goto out_free_tags;
 	}
+	vblk->disk->queue = q;
 
 	q->queuedata = vblk;
 

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

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

From: Rabin Vincent <[email protected]>

[ Upstream commit 270c8cf1cacc69cb8d99dea812f06067a45e4609 ]

ARM has a few system calls (most notably mmap) for which the names of the functions which are referenced in the syscall table do not match the names of the syscall tracepoints. As a consequence of this, these tracepoints are not made available. Implement arch_syscall_match_sym_name to fix this and allow tracing even these system calls.