openwrt/staging/blogic.git
9 years agounix/caif: sk_socket can disappear when state is unlocked
Mark Salyzyn [Tue, 26 May 2015 15:22:19 +0000 (08:22 -0700)]
unix/caif: sk_socket can disappear when state is unlocked

got a rare NULL pointer dereference in clear_bit

Signed-off-by: Mark Salyzyn <salyzyn@android.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
----
v2: switch to sock_flag(sk, SOCK_DEAD) and added net/caif/caif_socket.c
v3: return -ECONNRESET in upstream caller of wait function for SOCK_DEAD
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoamd-xgbe-phy: Fix initial mode when autoneg is disabled
Lendacky, Thomas [Tue, 26 May 2015 14:51:49 +0000 (09:51 -0500)]
amd-xgbe-phy: Fix initial mode when autoneg is disabled

When the ethtool command is used to set the speed of the device while
the device is down, the check to set the initial mode may fail when
the device is brought up, causing failure to bring the device up.

Update the code to set the initial mode based on the desired speed if
auto-negotiation is disabled.

This patch fixes a bug introduced by:
d9663c8c2149 ("amd-xgbe-phy: Use phydev advertising field vs supported")

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge tag 'mac80211-for-davem-2015-05-26' of git://git.kernel.org/pub/scm/linux/kerne...
David S. Miller [Tue, 26 May 2015 23:38:53 +0000 (19:38 -0400)]
Merge tag 'mac80211-for-davem-2015-05-26' of git://git./linux/kernel/git/jberg/mac80211

Johannes Berg says:

====================
We have three more fixes:
 * AP_VLAN tailroom calculation fix, the bug leads to warnings
   along with dropped packets
 * NAPI context issue, calling napi_gro_receive() from a timer
   (obviously) can lead to crashes
 * remain-on-channel combining leads to dropped requests and not
   being able to finish certain operations, so remove it
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'phyter-bug-fixes'
David S. Miller [Mon, 25 May 2015 22:22:00 +0000 (18:22 -0400)]
Merge branch 'phyter-bug-fixes'

Richard Cochran says:

====================
phyter bug fixes

While working on a project using the phyter, I noticed some bugs that
have crept in over time.  This series fixes those bugs.  These patches
are also meant for stable.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: dp83640: fix improper double spin locking.
Richard Cochran [Mon, 25 May 2015 09:55:45 +0000 (11:55 +0200)]
net: dp83640: fix improper double spin locking.

A pair of nested spin locks was introduced in commit 63502b8d0
"dp83640: Fix receive timestamp race condition".

Unfortunately the 'flags' parameter was reused for the inner lock,
clobbering the originally saved IRQ state.  This patch fixes the issue
by changing the inner lock to plain spin_lock without irqsave.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: dp83640: reinforce locking rules.
Richard Cochran [Mon, 25 May 2015 09:55:44 +0000 (11:55 +0200)]
net: dp83640: reinforce locking rules.

Callers of the ext_write function are supposed to hold a mutex that
protects the state of the dialed page, but one caller was missing the
lock from the very start, and over time the code has been changed
without following the rule.  This patch cleans up the call sites in
violation of the rule.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: dp83640: fix broken calibration routine.
Richard Cochran [Mon, 25 May 2015 09:55:43 +0000 (11:55 +0200)]
net: dp83640: fix broken calibration routine.

Currently, the calibration function that corrects the initial offsets
among multiple devices only works the first time.  If the function is
called more than once, the calibration fails and bogus offsets will be
programmed into the devices.

In a well hidden spot, the device documentation tells that trigger indexes
0 and 1 are special in allowing the TRIG_IF_LATE flag to actually work.

This patch fixes the issue by using one of the special triggers during the
recalibration method.

Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: stmmac: create one debugfs dir per net-device
Mathieu Olivari [Sat, 23 May 2015 02:03:29 +0000 (19:03 -0700)]
net: stmmac: create one debugfs dir per net-device

stmmac DebugFS entries are currently global to the driver. As a result,
having more than one stmmac device in the system creates the following
error:
* ERROR stmmaceth, debugfs create directory failed
* stmmac_hw_setup: failed debugFS registration

This also results in being able to access the debugfs information for
the first registered device only.

This patch changes the debugfs structure to have one sub-directory per
net-device. Files under "/sys/kernel/debug/stmmaceth" will now show-up
under /sys/kernel/debug/stmmaceth/ethN/.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet/ibm/emac: fix size of emac dump memory areas
Ivan Mikhaylov [Thu, 21 May 2015 15:11:02 +0000 (19:11 +0400)]
net/ibm/emac: fix size of emac dump memory areas

Fix in send of emac regs dump to ethtool which
causing in wrong data interpretation on ethtool
layer for MII and EMAC.

Signed-off-by: Ivan Mikhaylov <ivan@ru.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge tag 'wireless-drivers-for-davem-2015-05-25' of git://git.kernel.org/pub/scm...
David S. Miller [Mon, 25 May 2015 17:08:59 +0000 (13:08 -0400)]
Merge tag 'wireless-drivers-for-davem-2015-05-25' of git://git./linux/kernel/git/kvalo/wireless-drivers

Kalle Valo says:

====================
iwlwifi:

* fix firmware name and other things to enable 3165
* fix bad APMG configuration for 8000 (no AMPG on these devices)
* fix MAC address assignment for 8000
* fix firmware debugging triggers (MLME)
* fix several bugs in low power states code (net-detect, d0i3)

ssb:

* fix reboot after device reset for WRT350N v1
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agox86: bpf_jit: fix compilation of large bpf programs
Alexei Starovoitov [Fri, 22 May 2015 22:42:55 +0000 (15:42 -0700)]
x86: bpf_jit: fix compilation of large bpf programs

x86 has variable length encoding. x86 JIT compiler is trying
to pick the shortest encoding for given bpf instruction.
While doing so the jump targets are changing, so JIT is doing
multiple passes over the program. Typical program needs 3 passes.
Some very short programs converge with 2 passes. Large programs
may need 4 or 5. But specially crafted bpf programs may hit the
pass limit and if the program converges on the last iteration
the JIT compiler will be producing an image full of 'int 3' insns.
Fix this corner case by doing final iteration over bpf program.

Fixes: 0a14842f5a3c ("net: filter: Just In Time compiler for x86-64")
Reported-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Tested-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: phy: bcm7xxx: Fix 7425 PHY ID and flags
Florian Fainelli [Fri, 22 May 2015 21:07:30 +0000 (14:07 -0700)]
net: phy: bcm7xxx: Fix 7425 PHY ID and flags

While adding support for 7425 PHY in the 7xxx PHY driver, the ID that
was used was actually coming from an external PHY: a BCM5461x. Fix this
by using the proper ID for the internal 7425 PHY and set the
PHY_IS_INTERNAL flag, otherwise consumers of this PHY driver would not
be able to properly identify it as such.

Fixes: d068b02cfdfc2 ("net: phy: add BCM7425 and BCM7429 PHYs")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Petri Gynther <pgynther@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Sat, 23 May 2015 00:34:24 +0000 (17:34 -0700)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "Radeon has two displayport fixes, one for a regression.

  i915 regression flicker fix needed so 4.0 can get fixed.

  A bunch of msm fixes and a bunch of exynos fixes, these two are
  probably a bit larger than I'd like, but most of them seems pretty
  good"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (29 commits)
  drm/radeon: fix error flag checking in native aux path
  drm/radeon: retry dcpd fetch
  drm/msm/mdp5: fix incorrect parameter for msm_framebuffer_iova()
  drm/exynos: dp: Lower level of EDID read success message
  drm/exynos: cleanup exynos_drm_plane
  drm/exynos: 'win' is always unsigned
  drm/exynos: mixer: don't dump registers under spinlock
  drm/exynos: Consolidate return statements in fimd_bind()
  drm/exynos: Constify exynos_drm_crtc_ops
  drm/exynos: Fix build breakage on !DRM_EXYNOS_FIMD
  drm/exynos: mixer: Constify platform_device_id
  drm/exynos: mixer: cleanup pixelformat handling
  drm/exynos: mixer: also allow NV21 for the video processor
  drm/exynos: mixer: remove buffer count handling in vp_video_buffer()
  drm/exynos: plane: honor buffer offset for dma_addr
  drm/exynos: fb: use drm_format_num_planes to get buffer count
  drm/i915: fix screen flickering
  drm/msm: fix locking inconsistencies in gpu->destroy()
  drm/msm/dsi: Simplify the code to get the number of read byte
  drm/msm: Attach assigned encoder to eDP and DSI connectors
  ...

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Fri, 22 May 2015 22:44:50 +0000 (15:44 -0700)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Don't leak ipvs->sysctl_tbl, from Tommi Rentala.

 2) Fix neighbour table entry leak in rocker driver, from Ying Xue.

 3) Do not emit bonding notifications for unregistered interfaces, from
    Nicolas Dichtel.

 4) Set ipv6 flow label properly when in TIME_WAIT state, from Florent
    Fourcot.

 5) Fix regression in ipv6 multicast filter test, from Henning Rogge.

 6) do_replace() in various footables netfilter modules is missing a
    check for 0 counters in the datastructure provided by the user.  Fix
    from Dave Jones, and found with trinity.

 7) Fix RCU bug in packet scheduler classifier module unloads, from
    Daniel Borkmann.

 8) Avoid deadlock in tcp_get_info() by using u64_sync.  From Eric
    Dumzaet.

 9) Input packet processing can race with inetdev_destroy() teardown,
    fix potential OOPS in ip_error() by explicitly testing whether the
    inetdev is still attached.  From Eric W Biederman.

10) MLDv2 parser in bridge multicast code breaks too early while
    parsing.  Fix from Thadeu Lima de Souza Cascardo.

11) Asking for settings on non-zero PHYID doesn't work because we do not
    import the command structure from the user and use the PHYID
    provided there.  Fix from Arun Parameswaran.

12) Fix UDP checksums with IPV6 RAW sockets, from Vlad Yasevich.

13) Missing NF_TABLES depends for TPROXY etc can cause build failures,
    fix from Florian Westphal.

14) Fix netfilter conntrack to handle RFC5961 challenge ACKs properly,
    from Jesper Dangaard Brouer.

15) If netlink autobind retry fails, we have to reset the sockets portid
    back to zero.  From Herbert Xu.

16) VXLAN netns exit code unregisters using wrong device, from John W
    Linville.

17) Add some USB device IDs to ath3k and btusb bluetooth drivers, from
    Dmitry Tunin and Wen-chien Jesse Sung.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (44 commits)
  bridge: fix lockdep splat
  net: core: 'ethtool' issue with querying phy settings
  bridge: fix parsing of MLDv2 reports
  ARM: zynq: DT: Use the zynq binding with macb
  net: macb: Disable half duplex gigabit on Zynq
  net: macb: Document zynq gem dt binding
  ipv4: fill in table id when replacing a route
  cdc_ncm: Fix tx_bytes statistics
  ipv4: Avoid crashing in ip_error
  tcp: fix a potential deadlock in tcp_get_info()
  net: sched: fix call_rcu() race on classifier module unloads
  net: phy: Make sure phy_start() always re-enables the phy interrupts
  ipv6: fix ECMP route replacement
  ipv6: do not delete previously existing ECMP routes if add fails
  Revert "netfilter: bridge: query conntrack about skb dnat"
  netfilter: ensure number of counters is >0 in do_replace()
  netfilter: nfnetlink_{log,queue}: Register pernet in first place
  tcp: don't over-send F-RTO probes
  tcp: only undo on partial ACKs in CA_Loss
  net/ipv6/udp: Fix ipv6 multicast socket filter regression
  ...

9 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-block
Linus Torvalds [Fri, 22 May 2015 22:15:30 +0000 (15:15 -0700)]
Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "Three small fixes that have been picked up the last few weeks.
  Specifically:

   - Fix a memory corruption issue in NVMe with malignant user
     constructed request.  From Christoph.

   - Kill (now) unused blk_queue_bio(), dm was changed to not need this
     anymore.  From Mike Snitzer.

   - Always use blk_schedule_flush_plug() from the io_schedule() path
     when flushing a plug, fixing a !TASK_RUNNING warning with md.  From
     Shaohua"

* 'for-linus' of git://git.kernel.dk/linux-block:
  sched: always use blk_schedule_flush_plug in io_schedule_out
  nvme: fix kernel memory corruption with short INQUIRY buffers
  block: remove export for blk_queue_bio

9 years agoMerge tag 'md/4.1-rc4-fixes' of git://neil.brown.name/md
Linus Torvalds [Fri, 22 May 2015 22:10:07 +0000 (15:10 -0700)]
Merge tag 'md/4.1-rc4-fixes' of git://neil.brown.name/md

Pull md bugfixes from Neil Brown:
 "I have a few more raid5 bugfixes pending, but I want them to get a bit
  more review first.  In the meantime:

   - one serious RAID0 data corruption - caused by recent bugfix that
     wasn't reviewed properly.

   - one raid5 fix in new code (a couple more of those to come).

   - one little fix to stop static analysis complaining about silly rcu
     annotation"

* tag 'md/4.1-rc4-fixes' of git://neil.brown.name/md:
  md/bitmap: remove rcu annotation from pointer arithmetic.
  md/raid0: fix restore to sector variable in raid0_make_request
  raid5: fix broken async operation chain

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Fri, 22 May 2015 21:49:55 +0000 (14:49 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

Pull input fixes from Dmitry Torokhov:
 "Updates for the input subsystem.

  The main change is that we tell joydev not to touch "absolute mice",
  such as VMware virtual mouse, as that produced bad result (cursor
  stuck in upper right corner) with games"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: smtpe-ts - wait 50mS until polling for pen-up
  Input: smtpe-ts - use msecs_to_jiffies() instead of HZ
  Input: joydev - don't classify the vmmouse as a joystick
  Input: vmmouse - do not reference non-existing version of X driver
  Input: alps - fix finger jumps on lifting 2 fingers on v7 touchpad
  Input: elantech - fix semi-mt protocol for v3 HW
  Input: sx8654 - fix memory allocation check

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Linus Torvalds [Fri, 22 May 2015 21:26:36 +0000 (14:26 -0700)]
Merge git://git./linux/kernel/git/herbert/crypto-2.6

Pull another crypto fix from Herbert Xu:
 "Fix ICV corruption in s390/ghash when the same tfm is used by more
  than one thread"

* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
  crypto: s390/ghash - Fix incorrect ghash icv buffer handling.

9 years agobridge: fix lockdep splat
Eric Dumazet [Thu, 21 May 2015 20:28:29 +0000 (13:28 -0700)]
bridge: fix lockdep splat

Following lockdep splat was reported :

[   29.382286] ===============================
[   29.382315] [ INFO: suspicious RCU usage. ]
[   29.382344] 4.1.0-0.rc0.git11.1.fc23.x86_64 #1 Not tainted
[   29.382380] -------------------------------
[   29.382409] net/bridge/br_private.h:626 suspicious
rcu_dereference_check() usage!
[   29.382455]
               other info that might help us debug this:

[   29.382507]
               rcu_scheduler_active = 1, debug_locks = 0
[   29.382549] 2 locks held by swapper/0/0:
[   29.382576]  #0:  (((&p->forward_delay_timer))){+.-...}, at:
[<ffffffff81139f75>] call_timer_fn+0x5/0x4f0
[   29.382660]  #1:  (&(&br->lock)->rlock){+.-...}, at:
[<ffffffffa0450dc1>] br_forward_delay_timer_expired+0x31/0x140
[bridge]
[   29.382754]
               stack backtrace:
[   29.382787] CPU: 0 PID: 0 Comm: swapper/0 Not tainted
4.1.0-0.rc0.git11.1.fc23.x86_64 #1
[   29.382838] Hardware name: LENOVO 422916G/LENOVO, BIOS A1KT53AUS 04/07/2015
[   29.382882]  0000000000000000 3ebfc20364115825 ffff880666603c48
ffffffff81892d4b
[   29.382943]  0000000000000000 ffffffff81e124e0 ffff880666603c78
ffffffff8110bcd7
[   29.383004]  ffff8800785c9d00 ffff88065485ac58 ffff880c62002800
ffff880c5fc88ac0
[   29.383065] Call Trace:
[   29.383084]  <IRQ>  [<ffffffff81892d4b>] dump_stack+0x4c/0x65
[   29.383130]  [<ffffffff8110bcd7>] lockdep_rcu_suspicious+0xe7/0x120
[   29.383178]  [<ffffffffa04520f9>] br_fill_ifinfo+0x4a9/0x6a0 [bridge]
[   29.383225]  [<ffffffffa045266b>] br_ifinfo_notify+0x11b/0x4b0 [bridge]
[   29.383271]  [<ffffffffa0450d90>] ? br_hold_timer_expired+0x70/0x70 [bridge]
[   29.383320]  [<ffffffffa0450de8>]
br_forward_delay_timer_expired+0x58/0x140 [bridge]
[   29.383371]  [<ffffffffa0450d90>] ? br_hold_timer_expired+0x70/0x70 [bridge]
[   29.383416]  [<ffffffff8113a033>] call_timer_fn+0xc3/0x4f0
[   29.383454]  [<ffffffff81139f75>] ? call_timer_fn+0x5/0x4f0
[   29.383493]  [<ffffffff8110a90f>] ? lock_release_holdtime.part.29+0xf/0x200
[   29.383541]  [<ffffffffa0450d90>] ? br_hold_timer_expired+0x70/0x70 [bridge]
[   29.383587]  [<ffffffff8113a6a4>] run_timer_softirq+0x244/0x490
[   29.383629]  [<ffffffff810b68cc>] __do_softirq+0xec/0x670
[   29.383666]  [<ffffffff810b70d5>] irq_exit+0x145/0x150
[   29.383703]  [<ffffffff8189f506>] smp_apic_timer_interrupt+0x46/0x60
[   29.383744]  [<ffffffff8189d523>] apic_timer_interrupt+0x73/0x80
[   29.383782]  <EOI>  [<ffffffff816f131f>] ? cpuidle_enter_state+0x5f/0x2f0
[   29.383832]  [<ffffffff816f131b>] ? cpuidle_enter_state+0x5b/0x2f0

Problem here is that br_forward_delay_timer_expired() is a timer
handler, calling br_ifinfo_notify() which assumes either rcu_read_lock()
or RTNL are held.

Simplest fix seems to add rcu read lock section.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Josh Boyer <jwboyer@fedoraproject.org>
Reported-by: Dominick Grift <dac.override@gmail.com>
Cc: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: core: 'ethtool' issue with querying phy settings
Arun Parameswaran [Wed, 20 May 2015 21:35:30 +0000 (14:35 -0700)]
net: core: 'ethtool' issue with querying phy settings

When trying to configure the settings for PHY1, using commands
like 'ethtool -s eth0 phyad 1 speed 100', the 'ethtool' seems to
modify other settings apart from the speed of the PHY1, in the
above case.

The ethtool seems to query the settings for PHY0, and use this
as the base to apply the new settings to the PHY1. This is
causing the other settings of the PHY 1 to be wrongly
configured.

The issue is caused by the '_ethtool_get_settings()' API, which
gets called because of the 'ETHTOOL_GSET' command, is clearing
the 'cmd' pointer (of type 'struct ethtool_cmd') by calling
memset. This clears all the parameters (if any) passed for the
'ETHTOOL_GSET' cmd. So the driver's callback is always invoked
with 'cmd->phy_address' as '0'.

The '_ethtool_get_settings()' is called from other files in the
'net/core'. So the fix is applied to the 'ethtool_get_settings()'
which is only called in the context of the 'ethtool'.

Signed-off-by: Arun Parameswaran <aparames@broadcom.com>
Reviewed-by: Ray Jui <rjui@broadcom.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agobridge: fix parsing of MLDv2 reports
Thadeu Lima de Souza Cascardo [Fri, 22 May 2015 15:18:59 +0000 (12:18 -0300)]
bridge: fix parsing of MLDv2 reports

When more than a multicast address is present in a MLDv2 report, all but
the first address is ignored, because the code breaks out of the loop if
there has not been an error adding that address.

This has caused failures when two guests connected through the bridge
tried to communicate using IPv6. Neighbor discoveries would not be
transmitted to the other guest when both used a link-local address and a
static address.

This only happens when there is a MLDv2 querier in the network.

The fix will only break out of the loop when there is a failure adding a
multicast address.

The mdb before the patch:

dev ovirtmgmt port vnet0 grp ff02::1:ff7d:6603 temp
dev ovirtmgmt port vnet1 grp ff02::1:ff7d:6604 temp
dev ovirtmgmt port bond0.86 grp ff02::2 temp

After the patch:

dev ovirtmgmt port vnet0 grp ff02::1:ff7d:6603 temp
dev ovirtmgmt port vnet1 grp ff02::1:ff7d:6604 temp
dev ovirtmgmt port bond0.86 grp ff02::fb temp
dev ovirtmgmt port bond0.86 grp ff02::2 temp
dev ovirtmgmt port bond0.86 grp ff02::d temp
dev ovirtmgmt port vnet0 grp ff02::1:ff00:76 temp
dev ovirtmgmt port bond0.86 grp ff02::16 temp
dev ovirtmgmt port vnet1 grp ff02::1:ff00:77 temp
dev ovirtmgmt port bond0.86 grp ff02::1:ff00:def temp
dev ovirtmgmt port bond0.86 grp ff02::1:ffa1:40bf temp

Fixes: 08b202b67264 ("bridge br_multicast: IPv6 MLD support.")
Reported-by: Rik Theys <Rik.Theys@esat.kuleuven.be>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@redhat.com>
Tested-by: Rik Theys <Rik.Theys@esat.kuleuven.be>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoARM: zynq: DT: Use the zynq binding with macb
Nathan Sullivan [Fri, 22 May 2015 14:22:11 +0000 (09:22 -0500)]
ARM: zynq: DT: Use the zynq binding with macb

Use the new zynq binding for macb ethernet, since it will disable half
duplex gigabit like the Zynq TRM says to do.

Signed-off-by: Nathan Sullivan <nathan.sullivan@ni.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: macb: Disable half duplex gigabit on Zynq
Nathan Sullivan [Fri, 22 May 2015 14:22:10 +0000 (09:22 -0500)]
net: macb: Disable half duplex gigabit on Zynq

According to the Zynq TRM, gigabit half duplex is not supported.  Add a
new cap and compatible string so Zynq can avoid advertising that mode.

Signed-off-by: Nathan Sullivan <nathan.sullivan@ni.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet: macb: Document zynq gem dt binding
Nathan Sullivan [Fri, 22 May 2015 14:22:09 +0000 (09:22 -0500)]
net: macb: Document zynq gem dt binding

Signed-off-by: Nathan Sullivan <nathan.sullivan@ni.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoipv4: fill in table id when replacing a route
Michal Kubeček [Fri, 22 May 2015 11:40:09 +0000 (13:40 +0200)]
ipv4: fill in table id when replacing a route

When replacing an IPv4 route, tb_id member of the new fib_alias
structure is not set in the replace code path so that the new route is
ignored.

Fixes: 0ddcf43d5d4a ("ipv4: FIB Local/MAIN table collapse")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Alexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agocdc_ncm: Fix tx_bytes statistics
Bjørn Mork [Fri, 22 May 2015 11:15:22 +0000 (13:15 +0200)]
cdc_ncm: Fix tx_bytes statistics

The tx_curr_frame_payload field is u32. When we try to calculate a
small negative delta based on it, we end up with a positive integer
close to 2^32 instead.  So the tx_bytes pointer increases by about
2^32 for every transmitted frame.

Fix by calculating the delta as a signed long.

Cc: Ben Hutchings <ben.hutchings@codethink.co.uk>
Reported-by: Florian Bruhin <me@the-compiler.org>
Fixes: 7a1e890e2168 ("usbnet: Fix tx_bytes statistic running backward in cdc_ncm")
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf
David S. Miller [Fri, 22 May 2015 18:25:45 +0000 (14:25 -0400)]
Merge git://git./pub/scm/linux/kernel/git/pablo/nf

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contain Netfilter fixes for your net tree, they are:

1) Fix a race in nfnetlink_log and nfnetlink_queue that can lead to a crash.
   This problem is due to wrong order in the per-net registration and netlink
   socket events. Patch from Francesco Ruggeri.

2) Make sure that counters that userspace pass us are higher than 0 in all the
   x_tables frontends. Discovered via Trinity, patch from Dave Jones.

3) Revert a patch for br_netfilter to rely on the conntrack status bits. This
   breaks stateless IPv6 NAT transformations. Patch from Florian Westphal.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoipv4: Avoid crashing in ip_error
Eric W. Biederman [Fri, 22 May 2015 09:58:12 +0000 (04:58 -0500)]
ipv4: Avoid crashing in ip_error

ip_error does not check if in_dev is NULL before dereferencing it.

IThe following sequence of calls is possible:
CPU A                          CPU B
ip_rcv_finish
    ip_route_input_noref()
        ip_route_input_slow()
                               inetdev_destroy()
    dst_input()

With the result that a network device can be destroyed while processing
an input packet.

A crash was triggered with only unicast packets in flight, and
forwarding enabled on the only network device.   The error condition
was created by the removal of the network device.

As such it is likely the that error code was -EHOSTUNREACH, and the
action taken by ip_error (if in_dev had been accessible) would have
been to not increment any counters and to have tried and likely failed
to send an icmp error as the network device is going away.

Therefore handle this weird case by just dropping the packet if
!in_dev.  It will result in dropping the packet sooner, and will not
result in an actual change of behavior.

Fixes: 251da4130115b ("ipv4: Cache ip_error() routes even when not forwarding.")
Reported-by: Vittorio Gambaletta <linuxbugs@vittgam.net>
Tested-by: Vittorio Gambaletta <linuxbugs@vittgam.net>
Signed-off-by: Vittorio Gambaletta <linuxbugs@vittgam.net>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: fix a potential deadlock in tcp_get_info()
Eric Dumazet [Fri, 22 May 2015 04:51:19 +0000 (21:51 -0700)]
tcp: fix a potential deadlock in tcp_get_info()

Taking socket spinlock in tcp_get_info() can deadlock, as
inet_diag_dump_icsk() holds the &hashinfo->ehash_locks[i],
while packet processing can use the reverse locking order.

We could avoid this locking for TCP_LISTEN states, but lockdep would
certainly get confused as all TCP sockets share same lockdep classes.

[  523.722504] ======================================================
[  523.728706] [ INFO: possible circular locking dependency detected ]
[  523.734990] 4.1.0-dbg-DEV #1676 Not tainted
[  523.739202] -------------------------------------------------------
[  523.745474] ss/18032 is trying to acquire lock:
[  523.750002]  (slock-AF_INET){+.-...}, at: [<ffffffff81669d44>] tcp_get_info+0x2c4/0x360
[  523.758129]
[  523.758129] but task is already holding lock:
[  523.763968]  (&(&hashinfo->ehash_locks[i])->rlock){+.-...}, at: [<ffffffff816bcb75>] inet_diag_dump_icsk+0x1d5/0x6c0
[  523.774661]
[  523.774661] which lock already depends on the new lock.
[  523.774661]
[  523.782850]
[  523.782850] the existing dependency chain (in reverse order) is:
[  523.790326]
-> #1 (&(&hashinfo->ehash_locks[i])->rlock){+.-...}:
[  523.796599]        [<ffffffff811126bb>] lock_acquire+0xbb/0x270
[  523.802565]        [<ffffffff816f5868>] _raw_spin_lock+0x38/0x50
[  523.808628]        [<ffffffff81665af8>] __inet_hash_nolisten+0x78/0x110
[  523.815273]        [<ffffffff816819db>] tcp_v4_syn_recv_sock+0x24b/0x350
[  523.822067]        [<ffffffff81684d41>] tcp_check_req+0x3c1/0x500
[  523.828199]        [<ffffffff81682d09>] tcp_v4_do_rcv+0x239/0x3d0
[  523.834331]        [<ffffffff816842fe>] tcp_v4_rcv+0xa8e/0xc10
[  523.840202]        [<ffffffff81658fa3>] ip_local_deliver_finish+0x133/0x3e0
[  523.847214]        [<ffffffff81659a9a>] ip_local_deliver+0xaa/0xc0
[  523.853440]        [<ffffffff816593b8>] ip_rcv_finish+0x168/0x5c0
[  523.859624]        [<ffffffff81659db7>] ip_rcv+0x307/0x420

Lets use u64_sync infrastructure instead. As a bonus, 64bit
arches get optimized, as these are nop for them.

Fixes: 0df48c26d841 ("tcp: add tcpi_bytes_acked to tcp_info")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge tag 'iwlwifi-for-kalle-2015-05-21' of https://git.kernel.org/pub/scm/linux...
Kalle Valo [Fri, 22 May 2015 07:47:02 +0000 (10:47 +0300)]
Merge tag 'iwlwifi-for-kalle-2015-05-21' of https://git./linux/kernel/git/iwlwifi/iwlwifi-fixes

* fix firmware name and other things to enable 3165
* fix bad APMG configuration for 8000 (no AMPG on these devices)
* fix MAC address assignment for 8000
* fix firmware debugging triggers (MLME)
* fix several bugs in low power states code (net-detect,
  d0i3)

9 years agoMerge branch 'drm-fixes-4.1' of git://people.freedesktop.org/~agd5f/linux into drm...
Dave Airlie [Fri, 22 May 2015 03:31:54 +0000 (13:31 +1000)]
Merge branch 'drm-fixes-4.1' of git://people.freedesktop.org/~agd5f/linux into drm-fixes

Just two small DP fixes for 4.1

* 'drm-fixes-4.1' of git://people.freedesktop.org/~agd5f/linux:
  drm/radeon: fix error flag checking in native aux path
  drm/radeon: retry dcpd fetch

9 years agoMerge tag 'drm-intel-fixes-2015-05-21' of git://anongit.freedesktop.org/drm-intel...
Dave Airlie [Fri, 22 May 2015 03:31:03 +0000 (13:31 +1000)]
Merge tag 'drm-intel-fixes-2015-05-21' of git://anongit.freedesktop.org/drm-intel into drm-fixes

There's a stable backport from Ander [1] that combines this and a few
other commits to fix the flickering on v4.0, reported in [2] among
others. Having this upstream is obviously a requirement for stable.

* tag 'drm-intel-fixes-2015-05-21' of git://anongit.freedesktop.org/drm-intel:
  drm/i915: fix screen flickering

9 years agocrypto: s390/ghash - Fix incorrect ghash icv buffer handling.
Harald Freudenberger [Thu, 21 May 2015 08:01:11 +0000 (10:01 +0200)]
crypto: s390/ghash - Fix incorrect ghash icv buffer handling.

Multitheaded tests showed that the icv buffer in the current ghash
implementation is not handled correctly. A move of this working ghash
buffer value to the descriptor context fixed this. Code is tested and
verified with an multithreaded application via af_alg interface.

Cc: stable@vger.kernel.org
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Gerald Schaefer <geraldsc@linux.vnet.ibm.com>
Reported-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
9 years agoMerge tag 'for-linus-4.1b-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 22 May 2015 03:19:38 +0000 (20:19 -0700)]
Merge tag 'for-linus-4.1b-rc4-tag' of git://git./linux/kernel/git/xen/tip

Pull two xen bugfixes from David Vrabel:

 - fix ARM build regression.

 - fix VIRQ_CONSOLE related oops.

* tag 'for-linus-4.1b-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/events: don't bind non-percpu VIRQs with percpu chip
  xen/arm: Define xen_arch_suspend()

9 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Fri, 22 May 2015 03:15:16 +0000 (20:15 -0700)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull KVM fixes from Paolo Bonzini:
 "This includes a fix for two oopses, one on PPC and on x86.

  The rest is fixes for bugs with newer Intel processors"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  kvm/fpu: Enable eager restore kvm FPU for MPX
  Revert "KVM: x86: drop fpu_activate hook"
  kvm: fix crash in kvm_vcpu_reload_apic_access_page
  KVM: MMU: fix SMAP virtualization
  KVM: MMU: fix CR4.SMEP=1, CR0.WP=0 with shadow pages
  KVM: MMU: fix smap permission check
  KVM: PPC: Book3S HV: Fix list traversal in error case

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Linus Torvalds [Fri, 22 May 2015 02:54:50 +0000 (19:54 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/s390/linux

Pull s390 fixes from Martin Schwidefsky:
 "Bug fixes.

  Three for our crypto code, two for eBPF, and one memory management fix
  to get machines with memory > 8TB working"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390/mm: correct return value of pmd_pfn
  s390/crypto: fix stckf loop
  s390/zcrypt: Fix invalid domain handling during ap module unload
  s390/bpf: Fix gcov stack space problem
  s390/zcrypt: fixed ap poll timer behavior
  s390/bpf: Adjust ALU64_DIV/MOD to match interpreter change

9 years agoMerge tag 'sound-4.1-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 22 May 2015 00:54:00 +0000 (17:54 -0700)]
Merge tag 'sound-4.1-rc5' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "This batch became slightly large, just because I've been on vacation
  for the last two weeks.  Nothing to scare much here, all
  device-specific fixes, mostly small patches.

  Majority of patches are for HD-audio, especially Dell machines.  The
  rest are small ASoC fixes for various codecs, and a USB-audio quirk.

  One PCM fix is included to ease the faulty condition checks in the
  case of two periods PCM buffers"

* tag 'sound-4.1-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda - Disable widget power-saving for ALC292 & co
  ALSA: hda - Reduce verbs by node power-saves
  ALSA: sound/atmel/ac97c.c: remove unused variable
  ALSA: usb-audio: Add quirk for MS LifeCam Studio
  ALSA: pcm: Modify double acknowledged interrupts check condition
  ALSA: hda/realtek - ALC292 dock fix for Thinkpad L450
  ALSA: hda - Add Conexant codecs CX20721, CX20722, CX20723 and CX20724
  ALSA: hda - Fix headset mic and mic-in for a Dell desktop
  ASoC: wm8994: correct BCLK DIV 348 to 384
  ASoC: wm8960: fix "RINPUT3" audio route error
  ASoC: dapm: Modify widget stream name according to prefix
  ALSA: hda - Add headset mic quirk for Dell Inspiron 5548
  ASoC: rt5645: Fix mask for setting RT5645_DMIC_2_DP_GPIO12 bit
  ASoC: rt5645: Add ACPI match ID
  ALSA: hda/realtek - Add ALC298 alias name for Dell
  ASoC: uda1380: Avoid accessing i2c bus when codec is disabled
  ALSA: hda/realtek - Fix typo for ALC286/ALC288
  ASoC: mc13783: Fix wrong mask value used in mc13xxx_reg_rmw() calls
  ALSA: hda - Add headphone quirk for Lifebook E752
  ASoC: davinci-mcasp: Correct pm status check in suspend callback

9 years agoMerge branch 'for-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux
Linus Torvalds [Fri, 22 May 2015 00:42:54 +0000 (17:42 -0700)]
Merge branch 'for-rc' of git://git./linux/kernel/git/rzhang/linux

Pull thermal fixes from Zhang Rui:
 "Three fixes for Armada (380) and TI (dra7 and OMAP5) thermal soc
  drivers"

* 'for-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux:
  thermal: armada: Update Armada 380 thermal sensor coefficients
  thermal: ti-soc-thermal: OMAP5: Implement Workaround for Errata i813
  thermal: ti-soc-thermal: dra7: Implement Workaround for Errata i814

9 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma
Linus Torvalds [Fri, 22 May 2015 00:36:23 +0000 (17:36 -0700)]
Merge tag 'for-linus' of git://git./linux/kernel/git/dledford/rdma

Pull infiniband/rdma fixes from Doug Ledford:
 "This should hopefully be the last request for 4.1-rc for the RDMA
  stack.  It contains some late ocrdma fixes that I'm including because
  they are small and self contained.  It also contains two bug fixes
  that are simple and easily verified.

  Summary:

   - a number of small, well contained bug fixes for ocrdma driver

   - a simple fix for the connection negotiation sequence on IB

   - fix for broken AF_IB address on UD queue pair support"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma:
  IB/cma: Fix broken AF_IB UD support
  ib/cm: Change reject message type when destroying cm_id
  RDMA/ocrdma: Update ocrdma version number
  RDMA/ocrdma: Fail connection for MTU lesser than 512
  RDMA/ocrdma: Fix dmac resolution for link local address
  RDMA/ocrdma: Prevent allocation of DPP PDs if FW doesnt support it
  RDMA/ocrdma: Fix the request length for RDMA_QUERY_QP mailbox command to FW.
  RDMA/ocrdma: Use VID 0 if PFC is enabled and vlan is not configured
  RDMA/ocrdma: Fix QP state transition in destroy_qp
  RDMA/ocrdma: Report EQ full fatal error
  RDMA/ocrdma: Fix EQ destroy failure during driver unload

9 years agoMerge tag 'mmc-4.1-rc3' of git://git.linaro.org/people/ulf.hansson/mmc
Linus Torvalds [Fri, 22 May 2015 00:31:31 +0000 (17:31 -0700)]
Merge tag 'mmc-4.1-rc3' of git://git.linaro.org/people/ulf.hansson/mmc

Pull MMC fix from Ulf Hansson:
 "One more mmc fix intended for v4.1 rc5:

  MMC host:
   - atmel-mci: fix bad variable type for clkdiv"

* tag 'mmc-4.1-rc3' of git://git.linaro.org/people/ulf.hansson/mmc:
  mmc: atmel-mci: fix bad variable type for clkdiv

9 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
Linus Torvalds [Fri, 22 May 2015 00:23:11 +0000 (17:23 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jikos/hid

Pull HID fixes from Jiri Kosina:
 "Bugfixes for HID subsystem that should go in 4.1.  Important
  highlights:

   - the patch that extended support for HID++ protocol for TK820
     touchpad turns out to be causing regressions due to firmware
     issues; patch reverting back to basic support from Benjamin
     Tissoires

   - Wacom driver can oops for devices that report non-touch data on
     touch interfaces.  Fix from Ping Cheng

   - gpiolib is not mandatory for i2c-hid, so the driver shouldn't fail
     if gpiolib is not enabled.  Fix from Mika Westerberg"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
  HID: wacom: fix an Oops caused by wacom_wac_finger_count_touches
  HID: usbhid: Add HID_QUIRK_NOGET for Aten DVI KVM switch
  HID: hid-sensor-hub: Fix debug lock warning
  Revert "HID: logitech-hidpp: support combo keyboard touchpad TK820"
  HID: i2c-hid: Do not fail probing if gpiolib is not enabled

9 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Linus Torvalds [Fri, 22 May 2015 00:16:49 +0000 (17:16 -0700)]
Merge git://git./linux/kernel/git/herbert/crypto-2.6

Pull crypto fix from Herbert Xu:
 "This fixes a the crash in the newly added algif_aead interface when it
  tries to link SG lists"

* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
  crypto: algif_aead - fix invalid sgl linking

9 years agoInput: smtpe-ts - wait 50mS until polling for pen-up
Marek Vasut [Fri, 8 May 2015 05:25:49 +0000 (22:25 -0700)]
Input: smtpe-ts - wait 50mS until polling for pen-up

Wait a little bit longer, 50mS instead of 20mS, until the driver starts
polling for pen-up. The problematic behavior before this patch is applied
is as follows. The behavior was observed on the STMPE610QTR controller.

Upon a physical pen-down event, the touchscreen reports one set of x-y-p
coordinates and a pen-down event. After that, the pen-up polling is
triggered and since the controller is not ready yet, the polling mistakenly
detects a pen-up event while the physical state is still such that the pen
is down on the touch surface.

The pen-up handling flushes the controller FIFO, so after that, all the
samples in the controller are discarded. The controller becomes ready
shortly after this bogus pen-up handling and does generate again a pen-down
interrupt. This time, the controller contains x-y-p samples which all read
as zero. Since pressure value is zero, this set of samples is effectively
ignored by userland.

In the end, the driver just bounces between pen-down and bogus pen-up
handling, generating no useful results. Fix this by giving the controller a
bit more time before polling it for pen-up.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
9 years agoInput: smtpe-ts - use msecs_to_jiffies() instead of HZ
Marek Vasut [Fri, 8 May 2015 05:24:58 +0000 (22:24 -0700)]
Input: smtpe-ts - use msecs_to_jiffies() instead of HZ

Use msecs_to_jiffies(20) instead of plain (HZ / 50), as the former is much
more explicit about it's behavior. We want to schedule the task 20 mS from
now, so make it explicit in the code.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
9 years agoMerge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 21 May 2015 23:57:50 +0000 (16:57 -0700)]
Merge tag 'clk-fixes-for-linus' of git://git./linux/kernel/git/clk/linux

Pull clk fixes from Michael Turquette:
 "The first set of clk fixes for 4.1 are all driver bugs, with the
  exception of a single locking fix in the core code.

  All driver fixes are for code that was merged recently.  The Samsung
  stuff is mostly fixes around suspend/resume, the Qualcomm fixes are
  for invalid hardware configuration data and the Silicon Labs patches
  are fixes following their move away from platform_data to Device Tree"

* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
  clk: si5351: Do not pass struct clk in platform_data
  clk: si5351: Mention clock-names in the binding documentation
  clk: add missing lock when call clk_core_enable in clk_set_parent
  clk: exynos5420: Restore GATE_BUS_TOP on suspend
  clk: qcom: Fix MSM8916 gfx3d_clk_src configuration
  clk: qcom: Fix MSM8916 venus divider value
  clk: exynos5433: Fix wrong PMS value of exynos5433_pll_rates
  clk: exynos5433: Fix wrong parent clock of sclk_apollo clock
  clk: exynos5433: Fix CLK_PCLK_MONOTONIC_CNT clk register assignment
  clk: exynos5433: Fix wrong offset of PCLK_MSCL_SECURE_SMMU_JPEG
  clk: Use CONFIG_ARCH_EXYNOS instead of CONFIG_ARCH_EXYNOS5433

9 years agoMerge tag 'hwmon-for-linus-v4.1-rc5' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 21 May 2015 23:38:14 +0000 (16:38 -0700)]
Merge tag 'hwmon-for-linus-v4.1-rc5' of git://git./linux/kernel/git/groeck/linux-staging

Pull hwmon fixlet from Guenter Roeck:
 "Update location of Jean Delvare's hwmon quilt tree"

* tag 'hwmon-for-linus-v4.1-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
  hwmon: Update the location of my quilt tree

9 years agoInput: joydev - don't classify the vmmouse as a joystick
Thomas Hellstrom [Wed, 20 May 2015 21:50:30 +0000 (14:50 -0700)]
Input: joydev - don't classify the vmmouse as a joystick

Joydev is currently thinking some absolute mice are joystick, and that
messes up games in VMware guests, as the cursor typically gets stuck in
the top left corner.

Try to detect the event signature of a VMmouse input device and back off
for such devices. We're still incorrectly detecting, for example, the
VMware absolute USB mouse as a joystick, but adding an event signature
matching also that device would be considerably more risky, so defer that
to a later merge window.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
9 years agonet: sched: fix call_rcu() race on classifier module unloads
Daniel Borkmann [Wed, 20 May 2015 15:13:33 +0000 (17:13 +0200)]
net: sched: fix call_rcu() race on classifier module unloads

Vijay reported that a loop as simple as ...

  while true; do
    tc qdisc add dev foo root handle 1: prio
    tc filter add dev foo parent 1: u32 match u32 0 0  flowid 1
    tc qdisc del dev foo root
    rmmod cls_u32
  done

... will panic the kernel. Moreover, he bisected the change
apparently introducing it to 78fd1d0ab072 ("netlink: Re-add
locking to netlink_lookup() and seq walker").

The removal of synchronize_net() from the netlink socket
triggering the qdisc to be removed, seems to have uncovered
an RCU resp. module reference count race from the tc API.
Given that RCU conversion was done after e341694e3eb5 ("netlink:
Convert netlink_lookup() to use RCU protected hash table")
which added the synchronize_net() originally, occasion of
hitting the bug was less likely (not impossible though):

When qdiscs that i) support attaching classifiers and,
ii) have at least one of them attached, get deleted, they
invoke tcf_destroy_chain(), and thus call into ->destroy()
handler from a classifier module.

After RCU conversion, all classifier that have an internal
prio list, unlink them and initiate freeing via call_rcu()
deferral.

Meanhile, tcf_destroy() releases already reference to the
tp->ops->owner module before the queued RCU callback handler
has been invoked.

Subsequent rmmod on the classifier module is then not prevented
since all module references are already dropped.

By the time, the kernel invokes the RCU callback handler from
the module, that function address is then invalid.

One way to fix it would be to add an rcu_barrier() to
unregister_tcf_proto_ops() to wait for all pending call_rcu()s
to complete.

synchronize_rcu() is not appropriate as under heavy RCU
callback load, registered call_rcu()s could be deferred
longer than a grace period. In case we don't have any pending
call_rcu()s, the barrier is allowed to return immediately.

Since we came here via unregister_tcf_proto_ops(), there
are no users of a given classifier anymore. Further nested
call_rcu()s pointing into the module space are not being
done anywhere.

Only cls_bpf_delete_prog() may schedule a work item, to
unlock pages eventually, but that is not in the range/context
of cls_bpf anymore.

Fixes: 25d8c0d55f24 ("net: rcu-ify tcf_proto")
Fixes: 9888faefe132 ("net: sched: cls_basic use RCU")
Reported-by: Vijay Subramanian <subramanian.vijay@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: John Fastabend <john.r.fastabend@intel.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Alexei Starovoitov <ast@plumgrid.com>
Tested-by: Vijay Subramanian <subramanian.vijay@gmail.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoiwlwifi: mvm: avoid use-after-free on iwl_mvm_d0i3_enable_tx()
Eliad Peller [Tue, 14 Apr 2015 08:36:23 +0000 (11:36 +0300)]
iwlwifi: mvm: avoid use-after-free on iwl_mvm_d0i3_enable_tx()

qos_seq points (to a struct) inside the command response data.

Make sure to free the response only after qos_seq is not
needed anymore.

Reported-by: Heng Luo <heng.luo@intel.com>
Signed-off-by: Eliad Peller <eliadx.peller@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
9 years agoiwlwifi: mvm: clean net-detect info if device was reset during suspend
Luciano Coelho [Mon, 4 May 2015 14:03:17 +0000 (17:03 +0300)]
iwlwifi: mvm: clean net-detect info if device was reset during suspend

If the device is reset during suspend with net-detect enabled, we
leave the net-detect information dangling and this causes the next
suspend to fail with a warning:

[21795.351010] WARNING: at /root/iwlwifi/iwlwifi-stack-dev/drivers/net/wireless/iwlwifi/mvm/d3.c:989 __iwl_mvm_suspend.isra.6+0x2be/0x460 [iwlmvm]()
[21795.353253] Modules linked in: iwlmvm(O) iwlwifi(O) mac80211(O) cfg80211(O) compat(O) [...]
[21795.366168] CPU: 1 PID: 3645 Comm: bash Tainted: G           O 3.10.29-dev #1
[21795.368785] Hardware name: Dell Inc. Latitude E6430/0CPWYR, BIOS A09 12/13/2012
[21795.371441]  f8ec6748 f8ec6748 e51f3ce8 c168aa62 e51f3d10 c103a824 c1871238 f8ec6748
[21795.374228]  000003dd f8eb982e f8eb982e 00000000 c3408ed4 c41edbbc e51f3d20 c103a862
[21795.377006]  00000009 00000000 e51f3da8 f8eb982e c41ee3dc 00000004 e7970000 e51f3d74
[21795.379792] Call Trace:
[21795.382461]  [<c168aa62>] dump_stack+0x16/0x18
[21795.385133]  [<c103a824>] warn_slowpath_common+0x64/0x80
[21795.387803]  [<f8eb982e>] ? __iwl_mvm_suspend.isra.6+0x2be/0x460 [iwlmvm]
[21795.390485]  [<f8eb982e>] ? __iwl_mvm_suspend.isra.6+0x2be/0x460 [iwlmvm]
[21795.393124]  [<c103a862>] warn_slowpath_null+0x22/0x30
[21795.395787]  [<f8eb982e>] __iwl_mvm_suspend.isra.6+0x2be/0x460 [iwlmvm]
[21795.398464]  [<f8eb9d7c>] iwl_mvm_suspend+0xec/0x140 [iwlmvm]
[21795.401127]  [<c104be11>] ? del_timer_sync+0xa1/0xc0
[21795.403800]  [<f8d4107e>] __ieee80211_suspend+0x1de/0xff0 [mac80211]
[21795.406459]  [<c168e43d>] ? mutex_lock_nested+0x25d/0x350
[21795.409084]  [<c1586b64>] ? rtnl_lock+0x14/0x20
[21795.411685]  [<f8cf0076>] ieee80211_suspend+0x16/0x20 [mac80211]
[21795.414318]  [<f8c4e014>] wiphy_suspend+0x74/0x710 [cfg80211]
[21795.416916]  [<c141e612>] __device_suspend+0x1e2/0x220
[21795.419521]  [<f8c4dfa0>] ? addresses_show+0xa0/0xa0 [cfg80211]
[21795.422097]  [<c141f997>] dpm_suspend+0x67/0x210
[21795.424661]  [<c141fd6f>] dpm_suspend_start+0x4f/0x60
[21795.427219]  [<c108d8e0>] suspend_devices_and_enter+0x60/0x480
[21795.429768]  [<c168646a>] ? printk+0x4d/0x4f
[21795.432295]  [<c108de76>] pm_suspend+0x176/0x210
[21795.434830]  [<c108ca5d>] state_store+0x5d/0xb0
[21795.437410]  [<c108ca00>] ? wakeup_count_show+0x50/0x50
[21795.439961]  [<c13208db>] kobj_attr_store+0x1b/0x30
[21795.442514]  [<c11e3a4b>] sysfs_write_file+0xab/0x100
[21795.445088]  [<c11e39a0>] ? sysfs_poll+0xa0/0xa0
[21795.447659]  [<c1179655>] vfs_write+0xa5/0x1c0
[21795.450212]  [<c1179af7>] SyS_write+0x57/0xa0
[21795.452699]  [<c1699ec1>] sysenter_do_call+0x12/0x32
[21795.455146] ---[ end trace faf5321baba2bfdb ]---

To fix this, call the iwl_mvm_free_nd() function in case of any error
during resume.  Additionally, rename the "out_unlock" label to err to
make it clearer that it's only called in error conditions.

Cc: stable@vger.kernel.org [3.19+]
Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
9 years agoiwlwifi: mvm: take the UCODE_DOWN reference when resuming
Luciano Coelho [Tue, 28 Apr 2015 05:41:55 +0000 (08:41 +0300)]
iwlwifi: mvm: take the UCODE_DOWN reference when resuming

The __iwl_mvm_resume() function always returns 1, which causes
mac80211 to do a reconfig with IEEE80211_RECONFIG_TYPE_RESTART.  This
type of reconfig calls iwl_mvm_restart_complete(), where we unref the
IWL_MVM_REF_UCODE_DOWN, so we should always take the reference in this
case.

This prevents this kind of warning from happening:

[40026.103025] WARNING: at /root/iwlwifi/iwlwifi-stack-dev/drivers/net/wireless/iwlwifi/mvm/mac80211.c:236 iwl_mvm_unref+0xc9/0xd0 [iwlmvm]()
[40026.105145] Modules linked in: iwlmvm(O) iwlwifi(O) mac80211(O) cfg80211(O) compat(O) ctr ccm arc4 autofs4 snd_hda_codec_hdmi snd_hda_codec_idt joydev coretemp kvm_intel kvm aesni_intel ablk_helper cryptd lrw aes_i586 snd_hda_intel xts snd_hda_codec gf128mul snd_hwdep snd_pcm snd_seq_midi dell_wmi snd_rawmidi sparse_keymap snd_seq_midi_event snd_seq uvcvideo dell_laptop videobuf2_core dcdbas microcode videodev psmouse snd_timer videobuf2_vmalloc videobuf2_memops serio_raw snd_seq_device btusb i915 snd bluetooth lpc_ich drm_kms_helper soundcore snd_page_alloc drm i2c_algo_bit wmi parport_pc ppdev video binfmt_misc rpcsec_gss_krb5 nfsd mac_hid nfs_acl nfsv4 auth_rpcgss nfs fscache lockd sunrpc msdos lp parport sdhci_pci sdhci ahci libahci e1000e mmc_core ptp pps_core [last unloaded: compat]
[40026.117640] CPU: 2 PID: 3827 Comm: bash Tainted: G        W  O 3.10.29-dev #1
[40026.120216] Hardware name: Dell Inc. Latitude E6430/0CPWYR, BIOS A09 12/13/2012
[40026.122815]  f8effd18 f8effd18 e740fd18 c168aa62 e740fd40 c103a824 c1871238 f8effd18
[40026.125527]  000000ec f8ec79c9 f8ec79c9 d5d29ba4 d5d2a20c 00000000 e740fd50 c103a862
[40026.128209]  00000009 00000000 e740fd7c f8ec79c9 f1c591c4 00000400 00000000 f8efb490
[40026.130886] Call Trace:
[40026.133506]  [<c168aa62>] dump_stack+0x16/0x18
[40026.136115]  [<c103a824>] warn_slowpath_common+0x64/0x80
[40026.138727]  [<f8ec79c9>] ? iwl_mvm_unref+0xc9/0xd0 [iwlmvm]
[40026.141319]  [<f8ec79c9>] ? iwl_mvm_unref+0xc9/0xd0 [iwlmvm]
[40026.143881]  [<c103a862>] warn_slowpath_null+0x22/0x30
[40026.146453]  [<f8ec79c9>] iwl_mvm_unref+0xc9/0xd0 [iwlmvm]
[40026.149030]  [<f8ec7a4d>] iwl_mvm_mac_reconfig_complete+0x7d/0x210 [iwlmvm]
[40026.151645]  [<f8b74b20>] ? ftrace_raw_event_drv_reconfig_complete+0xc0/0xe0 [mac80211]
[40026.154291]  [<f8b6769e>] ieee80211_reconfig+0x28e/0x2620 [mac80211]
[40026.156920]  [<c10ef0ea>] ? ring_buffer_unlock_commit+0xba/0x100
[40026.159585]  [<f8b4a04d>] ieee80211_resume+0x6d/0x80 [mac80211]
[40026.162206]  [<f8a79722>] wiphy_resume+0x72/0x260 [cfg80211]
[40026.164799]  [<c141e2e7>] ? device_resume+0x57/0x150
[40026.167425]  [<f8a796b0>] ? wiphy_suspend+0x710/0x710 [cfg80211]
[40026.170075]  [<c141e26e>] dpm_run_callback+0x2e/0x50
[40026.172695]  [<c141e321>] device_resume+0x91/0x150
[40026.175334]  [<c141f636>] dpm_resume+0xf6/0x200
[40026.177922]  [<c141f920>] dpm_resume_end+0x10/0x20
[40026.180489]  [<c108d9f7>] suspend_devices_and_enter+0x177/0x480
[40026.183037]  [<c168646a>] ? printk+0x4d/0x4f
[40026.185559]  [<c108de76>] pm_suspend+0x176/0x210
[40026.188065]  [<c108ca5d>] state_store+0x5d/0xb0
[40026.190581]  [<c108ca00>] ? wakeup_count_show+0x50/0x50
[40026.193052]  [<c13208db>] kobj_attr_store+0x1b/0x30
[40026.195608]  [<c11e3a4b>] sysfs_write_file+0xab/0x100
[40026.198055]  [<c11e39a0>] ? sysfs_poll+0xa0/0xa0
[40026.200469]  [<c1179655>] vfs_write+0xa5/0x1c0
[40026.202893]  [<c1179af7>] SyS_write+0x57/0xa0
[40026.205245]  [<c1699ec1>] sysenter_do_call+0x12/0x32
[40026.207619] ---[ end trace db1d5a72a0381b0a ]---

Signed-off-by: Luciano Coelho <luciano.coelho@intel.com>
Reviewed-by: EliadX Peller <eliad@wizery.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
9 years agoiwlwifi: mvm: BT Coex - duplicate the command if sent ASYNC
Emmanuel Grumbach [Mon, 4 May 2015 08:20:52 +0000 (11:20 +0300)]
iwlwifi: mvm: BT Coex - duplicate the command if sent ASYNC

There are buses that can't handle ASYNC command without
copying them. Duplicate the host command instead.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
9 years agoiwlwifi: nvm: force mac from otp in case nvm mac is reserved
Liad Kaufman [Wed, 20 May 2015 12:50:07 +0000 (15:50 +0300)]
iwlwifi: nvm: force mac from otp in case nvm mac is reserved

Take the MAC address from the OTP even if one is present in
the NVM, if that MAC address happens to be a reserved one.

Signed-off-by: Liad Kaufman <liad.kaufman@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
9 years agoiwlwifi: mvm: Free fw_status after use to avoid memory leak
Haim Dreyfuss [Wed, 20 May 2015 05:10:43 +0000 (08:10 +0300)]
iwlwifi: mvm: Free fw_status after use to avoid memory leak

fw_status is the only pointer pointing to a block of memory
allocated above and should be freed after use.
Note: this come from Klockwork static analyzer.

Cc: stable@vger.kernel.org [3.19+]
Fixes: 2021a89d7b8a ("iwlwifi: mvm: treat netdetect wake up separately")
Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
9 years agodrm/radeon: fix error flag checking in native aux path
Alex Deucher [Wed, 20 May 2015 21:58:49 +0000 (17:58 -0400)]
drm/radeon: fix error flag checking in native aux path

That atom table does not check these bits.  Fixes aux
regressions on some boards.

Reported-by: Malte Schröder <malte@tnxip.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
9 years agodrm/radeon: retry dcpd fetch
Alex Deucher [Mon, 18 May 2015 14:38:25 +0000 (10:38 -0400)]
drm/radeon: retry dcpd fetch

Retry the dpcd fetch several times.  Some eDP panels
fail several times before the fetch is successful.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=73530

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
9 years agodrm/msm/mdp5: fix incorrect parameter for msm_framebuffer_iova()
Stephane Viau [Wed, 20 May 2015 14:57:27 +0000 (10:57 -0400)]
drm/msm/mdp5: fix incorrect parameter for msm_framebuffer_iova()

The index of ->planes[] array (3rd parameter) cannot be equal to MAX_PLANE.
This looks like a typo that is now fixed.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Acked-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
9 years agomd/bitmap: remove rcu annotation from pointer arithmetic.
NeilBrown [Wed, 20 May 2015 05:05:09 +0000 (15:05 +1000)]
md/bitmap: remove rcu annotation from pointer arithmetic.

Evaluating  "&mddev->disks" is simple pointer arithmetic, so
it does not need 'rcu' annotations - no dereferencing is happening.

Also enhance the comment to explain that 'rdev' in that case
is not actually a pointer to an rdev.

Reported-by: Patrick Marlier <patrick.marlier@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
9 years agomd/raid0: fix restore to sector variable in raid0_make_request
Eric Work [Tue, 19 May 2015 06:26:23 +0000 (23:26 -0700)]
md/raid0: fix restore to sector variable in raid0_make_request

The variable "sector" in "raid0_make_request()" was improperly updated
by a call to "sector_div()" which modifies its first argument in place.
Commit 47d68979cc968535cb87f3e5f2e6a3533ea48fbd restored this variable
after the call for later re-use.  Unfortunetly the restore was done after
the referenced variable "bio" was advanced.  This lead to the original
value and the restored value being different.  Here we move this line to
the proper place.

One observed side effect of this bug was discarding a file though
unlinking would cause an unrelated file's contents to be discarded.

Signed-off-by: NeilBrown <neilb@suse.de>
Fixes: 47d68979cc96 ("md/raid0: fix bug with chunksize not a power of 2.")
Cc: stable@vger.kernel.org (any that received above backport)
URL: https://bugzilla.kernel.org/show_bug.cgi?id=98501

9 years agoraid5: fix broken async operation chain
Shaohua Li [Wed, 13 May 2015 16:30:08 +0000 (09:30 -0700)]
raid5: fix broken async operation chain

ops_run_reconstruct6() doesn't correctly chain asyn operations. The tx returned
by async_gen_syndrome should be added as the dependent tx of next stripe.

The issue is introduced by commit 59fc630b8b5f9f21c8ce3ba153341c107dce1b0c
    RAID5: batch adjacent full stripe write

Reported-and-tested-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: NeilBrown <neilb@suse.de>
9 years agoInput: vmmouse - do not reference non-existing version of X driver
Thomas Hellstrom [Wed, 20 May 2015 21:46:14 +0000 (14:46 -0700)]
Input: vmmouse - do not reference non-existing version of X driver

The vmmouse Kconfig help text was referring to an incorrect user-space
driver version. Fix this.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
9 years agoInput: alps - fix finger jumps on lifting 2 fingers on v7 touchpad
Hans de Goede [Wed, 20 May 2015 21:37:41 +0000 (14:37 -0700)]
Input: alps - fix finger jumps on lifting 2 fingers on v7 touchpad

On v7 touchpads sometimes when 2 fingers are moved down on the touchpad
until they "fall of" the touchpad, the second touch will report 0 for y
(max y really since the y axis is inverted) and max x as coordinates,
rather then reporting 0, 0 as is expected for a non touching finger.

This commit detects this and treats these touches as non touching.

See the evemu-recording here:
https://bugzilla.redhat.com/attachment.cgi?id=1025058

BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1221200
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
9 years agoIB/cma: Fix broken AF_IB UD support
Matthew Finlay [Tue, 19 May 2015 07:11:48 +0000 (00:11 -0700)]
IB/cma: Fix broken AF_IB UD support

Support for using UD and AF_IB is currently broken.  The
IB_CM_SIDR_REQ_RECEIVED message is not handled properly in
cma_save_net_info() and we end up falling into code that will try and
process the request as ipv4/ipv6, which will end up failing.

The resolution is to add a check for the SIDR_REQ and call
cma_save_ib_info() with a NULL path record.  Change cma_save_ib_info()
to copy the src sib info from the listen_id when the path record is NULL.

Reported-by: Hari Shankar <Hari.Shankar@netapp.com>
Signed-off-by: Matt Finlay <matt@mellanox.com>
Acked-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
9 years agoib/cm: Change reject message type when destroying cm_id
Ted Kim [Thu, 14 May 2015 19:49:01 +0000 (12:49 -0700)]
ib/cm: Change reject message type when destroying cm_id

Problem reported by: Ted Kim <ted.h.kim@oracle.com>:

We have a case where a Linux system and a non-Linux system are
trying to interoperate.  The Linux host is the active side and
starts the connection establishment, but later decides to not go
through with the connection setup and does rdma_destroy_id().

The rdma_destroy_id() eventually works its way down to cm_destroy_id()
in core/cm.c, where a REJ is sent. The non-Linux system
has some trouble recognizing the REJ because of:

A. CM states which can't receive the REJ
B. Some issues about REJ formatting (missing comm ID)

ISSUE A: That part of the spec says, a Consumer Reject REJ can be
sent for a connection abort, but it goes further
and says: can send a REJ message with a "Consumer Reject"
Reason code if they are in a CM state (i.e. REP
Rcvd, MRA(REP) Sent, REQ Rcvd, MRA Sent) that allows
a REJ to be sent (lines 35-38).

Of the states listed there in that sentence, it would
seem to limit the active side to using the Consumer Reject
(for the abort case) in just the REP-Rcvd and MRA-REP-Sent
states. That is basically only after the active side
sees a REP (or alternatively goes down the state transitions
to timeout in which case a Timeout REJ is sent).

As a fix, in cm-destroy-id() move the IB-CM-MRA-REQ-RCVD case
to the same as REQ-SENT.  Essentially, make a REJ sent after
getting an MRA on active side a timeout rather than Consumer-
Reject, which is arguably more correct with the CM state
diagrams previous to getting a REP.

Signed-off-by: Ted Kim <ted.h.kim@oracle.com>
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
9 years agonet: phy: Make sure phy_start() always re-enables the phy interrupts
Tim Beale [Mon, 18 May 2015 03:38:38 +0000 (15:38 +1200)]
net: phy: Make sure phy_start() always re-enables the phy interrupts

This is an alternative way of fixing:
 commit db9683fb412d ("net: phy: Make sure PHY_RESUMING state change
                      is always processed")

When the PHY state transitions from PHY_HALTED to PHY_RESUMING, there are
two things we need to do:
1). Re-enable interrupts (and power up the physical link, if powered down)
2). Update the PHY state and net-device based on the link status.

There's no strict reason why #1 has to be done from within the main
phy_state_machine() function. There is a risk that other changes to the
PHY (e.g. setting speed/duplex, which calls phy_start_aneg()) could cause
a subsequent state transition before phy_state_machine() has processed
the PHY_RESUMING state change. This would leave the PHY with interrupts
disabled and/or still in the BMCR_PDOWN/low-power mode.

Moving enabling the interrupts and phy_resume() into phy_start() will
guarantee this work always gets done. As the PHY is already in the HALTED
state and interrupts are disabled, it shouldn't conflict with any work
being done in phy_state_machine(). The downside of this change is that if
the PHY_RESUMING state is ever entered from anywhere else, it'll also have
to repeat this work.

Signed-off-by: Tim Beale <tim.beale@alliedtelesis.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoMerge branch 'ipv6_ecmp_fixes'
David S. Miller [Wed, 20 May 2015 16:02:26 +0000 (12:02 -0400)]
Merge branch 'ipv6_ecmp_fixes'

Michal Kubecek says:

====================
IPv6 ECMP route add/replace fixes

(1) When adding a nexthop of a multipath route fails (e.g. because of a
conflict with an existing route), we are supposed to delete nexthops
already added. However, currently we try to also delete all nexthops we
haven't even tried to add yet so that a "ip route add" command can
actually remove pre-existing routes if it fails.

(2) Attempt to replace a multipath route results in a broken siblings
linked list. Following commands (like "ip route del") can then either
follow a link into freed memory or end in an infinite loop (if the slab
object has been reused).

v2: fix an omission in first patch

v3: change the semantics of replace operation to better match IPv4
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoipv6: fix ECMP route replacement
Michal Kubeček [Mon, 18 May 2015 18:54:00 +0000 (20:54 +0200)]
ipv6: fix ECMP route replacement

When replacing an IPv6 multipath route with "ip route replace", i.e.
NLM_F_CREATE | NLM_F_REPLACE, fib6_add_rt2node() replaces only first
matching route without fixing its siblings, resulting in corrupted
siblings linked list; removing one of the siblings can then end in an
infinite loop.

IPv6 ECMP implementation is a bit different from IPv4 so that route
replacement cannot work in exactly the same way. This should be a
reasonable approximation:

1. If the new route is ECMP-able and there is a matching ECMP-able one
already, replace it and all its siblings (if any).

2. If the new route is ECMP-able and no matching ECMP-able route exists,
replace first matching non-ECMP-able (if any) or just add the new one.

3. If the new route is not ECMP-able, replace first matching
non-ECMP-able route (if any) or add the new route.

We also need to remove the NLM_F_REPLACE flag after replacing old
route(s) by first nexthop of an ECMP route so that each subsequent
nexthop does not replace previous one.

Fixes: 51ebd3181572 ("ipv6: add support of equal cost multipath (ECMP)")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoipv6: do not delete previously existing ECMP routes if add fails
Michal Kubeček [Mon, 18 May 2015 18:53:55 +0000 (20:53 +0200)]
ipv6: do not delete previously existing ECMP routes if add fails

If adding a nexthop of an IPv6 multipath route fails, comment in
ip6_route_multipath() says we are going to delete all nexthops already
added. However, current implementation deletes even the routes it
hasn't even tried to add yet. For example, running

  ip route add 1234:5678::/64 \
      nexthop via fe80::aa dev dummy1 \
      nexthop via fe80::bb dev dummy1 \
      nexthop via fe80::cc dev dummy1

twice results in removing all routes first command added.

Limit the second (delete) run to nexthops that succeeded in the first
(add) run.

Fixes: 51ebd3181572 ("ipv6: add support of equal cost multipath (ECMP)")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agossb: extend fix for PCI related silent reboots to all chipsets
Rafał Miłecki [Wed, 13 May 2015 05:36:38 +0000 (07:36 +0200)]
ssb: extend fix for PCI related silent reboots to all chipsets

Recent fix for BCM4704 reboots has to be extended as the same problem
affects Linksys WRT350N v1 (BCM4705).

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Reported-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
9 years agomac80211: fix AP_VLAN crypto tailroom calculation
Michal Kazior [Wed, 13 May 2015 09:16:48 +0000 (09:16 +0000)]
mac80211: fix AP_VLAN crypto tailroom calculation

Some splats I was seeing:

 (a) WARNING: CPU: 1 PID: 0 at /devel/src/linux/net/mac80211/wep.c:102 ieee80211_wep_add_iv
 (b) WARNING: CPU: 1 PID: 0 at /devel/src/linux/net/mac80211/wpa.c:73 ieee80211_tx_h_michael_mic_add
 (c) WARNING: CPU: 3 PID: 0 at /devel/src/linux/net/mac80211/wpa.c:433 ieee80211_crypto_ccmp_encrypt

I've seen (a) and (b) with ath9k hw crypto and (c)
with ath9k sw crypto. All of them were related to
insufficient skb tailroom and I was able to
trigger these with ping6 program.

AP_VLANs may inherit crypto keys from parent AP.
This wasn't considered and yielded problems in
some setups resulting in inability to transmit
data because mac80211 wouldn't resize skbs when
necessary and subsequently drop some packets due
to insufficient tailroom.

For efficiency purposes don't inspect both AP_VLAN
and AP sdata looking for tailroom counter. Instead
update AP_VLAN tailroom counters whenever their
master AP tailroom counter changes.

Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
9 years agomac80211: don't split remain-on-channel for coalescing
Johannes Berg [Tue, 19 May 2015 11:36:38 +0000 (13:36 +0200)]
mac80211: don't split remain-on-channel for coalescing

Due to remain-on-channel scheduling delays, when we split an ROC
while coalescing, we'll usually get a picture like this:

existing ROC:  |------------------|
current time:              ^
new ROC:                   |------|              |-------|

If the expected response frames are then transmitted by the peer
in the hole between the two fragments of the new ROC, we miss
them and the process (e.g. ANQP query) fails.

mac80211 expects that the window to miss something is small:

existing ROC:  |------------------|
new ROC:                   |------||-------|

but that's normally not the case.

To avoid this problem, coalesce only if the new ROC's duration
is <= the remaining time on the existing one:

existing ROC:  |------------------|
new ROC:                   |-----|

and never split a new one but schedule it afterwards instead:

existing ROC:  |------------------|
new ROC:                                       |-------------|

type=bugfix
bug=not-tracked
fixes=unknown

Reported-by: Matti Gottlieb <matti.gottlieb@intel.com>
Reviewed-by: EliadX Peller <eliad@wizery.com>
Reviewed-by: Matti Gottlieb <matti.gottlieb@intel.com>
Tested-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
9 years agoRevert "netfilter: bridge: query conntrack about skb dnat"
Florian Westphal [Wed, 20 May 2015 11:42:25 +0000 (13:42 +0200)]
Revert "netfilter: bridge: query conntrack about skb dnat"

This reverts commit c055d5b03bb4cb69d349d787c9787c0383abd8b2.

There are two issues:
'dnat_took_place' made me think that this is related to
-j DNAT/MASQUERADE.

But thats only one part of the story.  This is also relevant for SNAT
when we undo snat translation in reverse/reply direction.

Furthermore, I originally wanted to do this mainly to avoid
storing ipv6 addresses once we make DNAT/REDIRECT work
for ipv6 on bridges.

However, I forgot about SNPT/DNPT which is stateless.

So we can't escape storing address for ipv6 anyway. Might as
well do it for ipv4 too.

Reported-and-tested-by: Bernhard Thaler <bernhard.thaler@wvnet.at>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
9 years agonetfilter: ensure number of counters is >0 in do_replace()
Dave Jones [Wed, 20 May 2015 00:55:17 +0000 (20:55 -0400)]
netfilter: ensure number of counters is >0 in do_replace()

After improving setsockopt() coverage in trinity, I started triggering
vmalloc failures pretty reliably from this code path:

warn_alloc_failed+0xe9/0x140
__vmalloc_node_range+0x1be/0x270
vzalloc+0x4b/0x50
__do_replace+0x52/0x260 [ip_tables]
do_ipt_set_ctl+0x15d/0x1d0 [ip_tables]
nf_setsockopt+0x65/0x90
ip_setsockopt+0x61/0xa0
raw_setsockopt+0x16/0x60
sock_common_setsockopt+0x14/0x20
SyS_setsockopt+0x71/0xd0

It turns out we don't validate that the num_counters field in the
struct we pass in from userspace is initialized.

The same problem also exists in ebtables, arptables, ipv6, and the
compat variants.

Signed-off-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
9 years agonetfilter: nfnetlink_{log,queue}: Register pernet in first place
Francesco Ruggeri [Sun, 17 May 2015 21:30:31 +0000 (14:30 -0700)]
netfilter: nfnetlink_{log,queue}: Register pernet in first place

nfnetlink_{log,queue}_init() register the netlink callback nf*_rcv_nl_event
before registering the pernet_subsys, but the callback relies on data
structures allocated by pernet init functions.

When nfnetlink_{log,queue} is loaded, if a netlink message is received after
the netlink callback is registered but before the pernet_subsys is registered,
the kernel will panic in the sequence

nfulnl_rcv_nl_event
  nfnl_log_pernet
    net_generic
      BUG_ON(id == 0)  where id is nfnl_log_net_id.

The panic can be easily reproduced in 4.0.3 by:

while true ;do modprobe nfnetlink_log ; rmmod nfnetlink_log ; done &
while true ;do ip netns add dummy ; ip netns del dummy ; done &

This patch moves register_pernet_subsys to earlier in nfnetlink_log_init.

Notice that the BUG_ON hit in 4.0.3 was recently removed in 2591ffd308
["netns: remove BUG_ONs from net_generic()"].

Signed-off-by: Francesco Ruggeri <fruggeri@arista.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
9 years agokvm/fpu: Enable eager restore kvm FPU for MPX
Liang Li [Wed, 20 May 2015 20:41:25 +0000 (04:41 +0800)]
kvm/fpu: Enable eager restore kvm FPU for MPX

The MPX feature requires eager KVM FPU restore support. We have verified
that MPX cannot work correctly with the current lazy KVM FPU restore
mechanism. Eager KVM FPU restore should be enabled if the MPX feature is
exposed to VM.

Signed-off-by: Yang Zhang <yang.z.zhang@intel.com>
Signed-off-by: Liang Li <liang.z.li@intel.com>
[Also activate the FPU on AMD processors. - Paolo]
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoRevert "KVM: x86: drop fpu_activate hook"
Paolo Bonzini [Wed, 20 May 2015 09:33:43 +0000 (11:33 +0200)]
Revert "KVM: x86: drop fpu_activate hook"

This reverts commit 4473b570a7ebb502f63f292ccfba7df622e5fdd3.  We'll
use the hook again.

Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agokvm: fix crash in kvm_vcpu_reload_apic_access_page
Andrea Arcangeli [Fri, 8 May 2015 12:32:56 +0000 (14:32 +0200)]
kvm: fix crash in kvm_vcpu_reload_apic_access_page

memslot->userfault_addr is set by the kernel with a mmap executed
from the kernel but the userland can still munmap it and lead to the
below oops after memslot->userfault_addr points to a host virtual
address that has no vma or mapping.

[  327.538306] BUG: unable to handle kernel paging request at fffffffffffffffe
[  327.538407] IP: [<ffffffff811a7b55>] put_page+0x5/0x50
[  327.538474] PGD 1a01067 PUD 1a03067 PMD 0
[  327.538529] Oops: 0000 [#1] SMP
[  327.538574] Modules linked in: macvtap macvlan xt_CHECKSUM iptable_mangle ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_nat_ipv4 nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ipt_REJECT iptable_filter ip_tables tun bridge stp llc rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache xprtrdma ib_isert iscsi_target_mod ib_iser libiscsi scsi_transport_iscsi ib_srpt target_core_mod ib_srp scsi_transport_srp scsi_tgt ib_ipoib rdma_ucm ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm ipmi_devintf iTCO_wdt iTCO_vendor_support intel_powerclamp coretemp dcdbas intel_rapl kvm_intel kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd pcspkr sb_edac edac_core ipmi_si ipmi_msghandler acpi_pad wmi acpi_power_meter lpc_ich mfd_core mei_me
[  327.539488]  mei shpchp nfsd auth_rpcgss nfs_acl lockd grace sunrpc mlx4_ib ib_sa ib_mad ib_core mlx4_en vxlan ib_addr ip_tunnel xfs libcrc32c sd_mod crc_t10dif crct10dif_common crc32c_intel mgag200 syscopyarea sysfillrect sysimgblt i2c_algo_bit drm_kms_helper ttm drm ahci i2c_core libahci mlx4_core libata tg3 ptp pps_core megaraid_sas ntb dm_mirror dm_region_hash dm_log dm_mod
[  327.539956] CPU: 3 PID: 3161 Comm: qemu-kvm Not tainted 3.10.0-240.el7.userfault19.4ca4011.x86_64.debug #1
[  327.540045] Hardware name: Dell Inc. PowerEdge R420/0CN7CM, BIOS 2.1.2 01/20/2014
[  327.540115] task: ffff8803280ccf00 ti: ffff880317c58000 task.ti: ffff880317c58000
[  327.540184] RIP: 0010:[<ffffffff811a7b55>]  [<ffffffff811a7b55>] put_page+0x5/0x50
[  327.540261] RSP: 0018:ffff880317c5bcf8  EFLAGS: 00010246
[  327.540313] RAX: 00057ffffffff000 RBX: ffff880616a20000 RCX: 0000000000000000
[  327.540379] RDX: 0000000000002014 RSI: 00057ffffffff000 RDI: fffffffffffffffe
[  327.540445] RBP: ffff880317c5bd10 R08: 0000000000000103 R09: 0000000000000000
[  327.540511] R10: 0000000000000000 R11: 0000000000000000 R12: fffffffffffffffe
[  327.540576] R13: 0000000000000000 R14: ffff880317c5bd70 R15: ffff880317c5bd50
[  327.540643] FS:  00007fd230b7f700(0000) GS:ffff880630800000(0000) knlGS:0000000000000000
[  327.540717] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  327.540771] CR2: fffffffffffffffe CR3: 000000062a2c3000 CR4: 00000000000427e0
[  327.540837] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  327.540904] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[  327.540974] Stack:
[  327.541008]  ffffffffa05d6d0c ffff880616a20000 0000000000000000 ffff880317c5bdc0
[  327.541093]  ffffffffa05ddaa2 0000000000000000 00000000002191bf 00000042f3feab2d
[  327.541177]  00000042f3feab2d 0000000000000002 0000000000000001 0321000000000000
[  327.541261] Call Trace:
[  327.541321]  [<ffffffffa05d6d0c>] ? kvm_vcpu_reload_apic_access_page+0x6c/0x80 [kvm]
[  327.543615]  [<ffffffffa05ddaa2>] vcpu_enter_guest+0x3f2/0x10f0 [kvm]
[  327.545918]  [<ffffffffa05e2f10>] kvm_arch_vcpu_ioctl_run+0x2b0/0x5a0 [kvm]
[  327.548211]  [<ffffffffa05e2d02>] ? kvm_arch_vcpu_ioctl_run+0xa2/0x5a0 [kvm]
[  327.550500]  [<ffffffffa05ca845>] kvm_vcpu_ioctl+0x2b5/0x680 [kvm]
[  327.552768]  [<ffffffff810b8d12>] ? creds_are_invalid.part.1+0x12/0x50
[  327.555069]  [<ffffffff810b8d71>] ? creds_are_invalid+0x21/0x30
[  327.557373]  [<ffffffff812d6066>] ? inode_has_perm.isra.49.constprop.65+0x26/0x80
[  327.559663]  [<ffffffff8122d985>] do_vfs_ioctl+0x305/0x530
[  327.561917]  [<ffffffff8122dc51>] SyS_ioctl+0xa1/0xc0
[  327.564185]  [<ffffffff816de829>] system_call_fastpath+0x16/0x1b
[  327.566480] Code: 0b 31 f6 4c 89 e7 e8 4b 7f ff ff 0f 0b e8 24 fd ff ff e9 a9 fd ff ff 66 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 66 66 66 66 90 <48> f7 07 00 c0 00 00 55 48 89 e5 75 2a 8b 47 1c 85 c0 74 1e f0

Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agoHID: wacom: fix an Oops caused by wacom_wac_finger_count_touches
Ping Cheng [Wed, 20 May 2015 00:42:02 +0000 (17:42 -0700)]
HID: wacom: fix an Oops caused by wacom_wac_finger_count_touches

We assumed all touch interfaces report touch data. But, Bamboo
and Intuos non-touch devices report express keys on touch
interface. We need to check touch_max before counting touches.

Reported-by: Tasos Sahanidis <tasos@tasossah.com>
Signed-off-by: Ping Cheng <pingc@wacom.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
9 years agoALSA: hda - Disable widget power-saving for ALC292 & co
Takashi Iwai [Wed, 20 May 2015 04:56:23 +0000 (06:56 +0200)]
ALSA: hda - Disable widget power-saving for ALC292 & co

We've got reports that ALC3226 (a Dell variant of ALC292) gives click
noises at transition from D3 to D0 when the widget power-saving is
enabled.  Further debugging session showed that avoiding it isn't
trivial, unfortunately, since paths are basically activated
dynamically while the pins have been already enabled.

This patch disables the widget power-saving for such codecs.

Reported-by: Jonathan McDowell <noodles@earth.li>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
9 years agoALSA: hda - Reduce verbs by node power-saves
Takashi Iwai [Wed, 20 May 2015 04:49:37 +0000 (06:49 +0200)]
ALSA: hda - Reduce verbs by node power-saves

The widget (node) power-saves restore the widget states at each
transition from D3 to D0 on each node.  This was added in the commit
[d545a57c5f84:ALSA: hda - Sync node attributes at resume from widget
power saving].  However, the test was rater false-positive; this
wasn't needed for any codecs.

Since the resync may take significant number of additional verbs to be
executed, it's better to reduce it.  Let's disable it for now again.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
9 years agoALSA: sound/atmel/ac97c.c: remove unused variable
Arnd Bergmann [Tue, 19 May 2015 15:20:36 +0000 (17:20 +0200)]
ALSA: sound/atmel/ac97c.c: remove unused variable

The recently added DT support for the ac97 driver is causing
a gcc warning:

sound/atmel/ac97c.c: In function 'atmel_ac97c_probe_dt':
sound/atmel/ac97c.c:919:29: warning: unused variable 'match' [-Wunused-variable]
  const struct of_device_id *match;

The variable is clearly unused, so we can remove it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Alexander Stein <alexanders83@web.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
9 years agoMerge branch 'exynos-drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git...
Dave Airlie [Tue, 19 May 2015 23:22:00 +0000 (09:22 +1000)]
Merge branch 'exynos-drm-fixes' of git://git./linux/kernel/git/daeinki/drm-exynos into drm-fixes

Summary:
- Use generic function to get buffer count instead of specific one.
  In case of Exynos DRM, There was a special case which decides pixel
  format of a given buffer according to planer types, which is NV12M and NV12.
  However, NV12M doesn't exist in drm fourcc so it removes
  exynos_drm_format_num_buffers() specific to Exynos DRM and use a generic function,
  drm_format_num_planes() instead.
- Allow mixer driver to support NV21 format for Video processor.
  This format was already supported but we just missed DRM_FORMAT_NV21 case
  so this patch considers the case so that Mixer driver can handle it correctly.
- Add regression fix and some code cleanups.

* 'exynos-drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos:
  drm/exynos: dp: Lower level of EDID read success message
  drm/exynos: cleanup exynos_drm_plane
  drm/exynos: 'win' is always unsigned
  drm/exynos: mixer: don't dump registers under spinlock
  drm/exynos: Consolidate return statements in fimd_bind()
  drm/exynos: Constify exynos_drm_crtc_ops
  drm/exynos: Fix build breakage on !DRM_EXYNOS_FIMD
  drm/exynos: mixer: Constify platform_device_id
  drm/exynos: mixer: cleanup pixelformat handling
  drm/exynos: mixer: also allow NV21 for the video processor
  drm/exynos: mixer: remove buffer count handling in vp_video_buffer()
  drm/exynos: plane: honor buffer offset for dma_addr
  drm/exynos: fb: use drm_format_num_planes to get buffer count

9 years agoMerge tag 'mac80211-for-davem-2015-05-19' of git://git.kernel.org/pub/scm/linux/kerne...
David S. Miller [Tue, 19 May 2015 20:44:25 +0000 (16:44 -0400)]
Merge tag 'mac80211-for-davem-2015-05-19' of git://git./linux/kernel/git/jberg/mac80211

Johannes Berg says:

====================
This has just a single fix, for a WEP tailroom check
problem that leads to dropped frames.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: don't over-send F-RTO probes
Yuchung Cheng [Mon, 18 May 2015 19:31:45 +0000 (12:31 -0700)]
tcp: don't over-send F-RTO probes

After sending the new data packets to probe (step 2), F-RTO may
incorrectly send more probes if the next ACK advances SND_UNA and
does not sack new packet. However F-RTO RFC 5682 probes at most
once. This bug may cause sender to always send new data instead of
repairing holes, inducing longer HoL blocking on the receiver for
the application.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agotcp: only undo on partial ACKs in CA_Loss
Yuchung Cheng [Mon, 18 May 2015 19:31:44 +0000 (12:31 -0700)]
tcp: only undo on partial ACKs in CA_Loss

Undo based on TCP timestamps should only happen on ACKs that advance
SND_UNA, according to the Eifel algorithm in RFC 3522:

Section 3.2:

  (4) If the value of the Timestamp Echo Reply field of the
      acceptable ACK's Timestamps option is smaller than the
      value of RetransmitTS, then proceed to step (5),

Section Terminology:
   We use the term 'acceptable ACK' as defined in [RFC793].  That is an
   ACK that acknowledges previously unacknowledged data.

This is because upon receiving an out-of-order packet, the receiver
returns the last timestamp that advances RCV_NXT, not the current
timestamp of the packet in the DUPACK. Without checking the flag,
the DUPACK will cause tcp_packet_delayed() to return true and
tcp_try_undo_loss() will revert cwnd reduction.

Note that we check the condition in CA_Recovery already by only
calling tcp_try_undo_partial() if FLAG_SND_UNA_ADVANCED is set or
tcp_try_undo_recovery() if snd_una crosses high_seq.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agonet/ipv6/udp: Fix ipv6 multicast socket filter regression
Henning Rogge [Mon, 18 May 2015 19:08:49 +0000 (21:08 +0200)]
net/ipv6/udp: Fix ipv6 multicast socket filter regression

Commit <5cf3d46192fc> ("udp: Simplify__udp*_lib_mcast_deliver")
simplified the filter for incoming IPv6 multicast but removed
the check of the local socket address and the UDP destination
address.

This patch restores the filter to prevent sockets bound to a IPv6
multicast IP to receive other UDP traffic link unicast.

Signed-off-by: Henning Rogge <hrogge@gmail.com>
Fixes: 5cf3d46192fc ("udp: Simplify__udp*_lib_mcast_deliver")
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agoxen/events: don't bind non-percpu VIRQs with percpu chip
David Vrabel [Tue, 19 May 2015 17:40:49 +0000 (18:40 +0100)]
xen/events: don't bind non-percpu VIRQs with percpu chip

A non-percpu VIRQ (e.g., VIRQ_CONSOLE) may be freed on a different
VCPU than it is bound to.  This can result in a race between
handle_percpu_irq() and removing the action in __free_irq() because
handle_percpu_irq() does not take desc->lock.  The interrupt handler
sees a NULL action and oopses.

Only use the percpu chip/handler for per-CPU VIRQs (like VIRQ_TIMER).

  # cat /proc/interrupts | grep virq
   40:      87246          0  xen-percpu-virq      timer0
   44:          0          0  xen-percpu-virq      debug0
   47:          0      20995  xen-percpu-virq      timer1
   51:          0          0  xen-percpu-virq      debug1
   69:          0          0   xen-dyn-virq      xen-pcpu
   74:          0          0   xen-dyn-virq      mce
   75:         29          0   xen-dyn-virq      hvc_console

Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Cc: <stable@vger.kernel.org>
9 years agoMerge tag 'nfs-for-4.1-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Linus Torvalds [Tue, 19 May 2015 18:20:48 +0000 (11:20 -0700)]
Merge tag 'nfs-for-4.1-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull two NFS client bugfixes from Trond Myklebust:
 "Highlights include:

   - fix a Linux-4.1 regression affecting stat()

   - take an extra reference to fl->fl_file when running a setlk"

* tag 'nfs-for-4.1-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
  nfs: take extra reference to fl->fl_file when running a setlk
  nfs: stat(2) fails during cthon04 basic test5 on NFSv4.0

9 years agoMerge tag 'powerpc-4.1-4' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux
Linus Torvalds [Tue, 19 May 2015 18:19:49 +0000 (11:19 -0700)]
Merge tag 'powerpc-4.1-4' of git://git./linux/kernel/git/mpe/linux

Pull powerpc fixes from Michael Ellerman:

 - THP/hugetlb fixes from Aneesh.

 - MCE fix from Daniel.

 - TOC fix from Anton.

* tag 'powerpc-4.1-4' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux:
  powerpc: Align TOC to 256 bytes
  powerpc/mce: fix off by one errors in mce event handling
  powerpc/mm: Return NULL for not present hugetlb page
  powerpc/thp: Serialize pmd clear against a linux page table walk.

9 years agoMerge tag 'pwm/for-4.1-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry...
Linus Torvalds [Tue, 19 May 2015 18:18:14 +0000 (11:18 -0700)]
Merge tag 'pwm/for-4.1-rc5' of git://git./linux/kernel/git/thierry.reding/linux-pwm

Pull pwm fix from Thierry Reding:
 "A single fix to make the Pistachio driver respect the limits imposed
  by hardware"

* tag 'pwm/for-4.1-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm:
  pwm: img: Impose upper and lower timebase steps value

9 years agowatchdog: fix double lock in watchdog_nmi_enable_all
Michal Hocko [Tue, 19 May 2015 07:07:27 +0000 (09:07 +0200)]
watchdog: fix double lock in watchdog_nmi_enable_all

Commit ab992dc38f9a ("watchdog: Fix merge 'conflict'") has introduced an
obvious deadlock because of a typo.  watchdog_proc_mutex should be
unlocked on exit.

Thanks to Miroslav Benes who was staring at the code with me and noticed
this.

Signed-off-by: Michal Hocko <mhocko@suse.cz>
Duh-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoinet: properly align icsk_ca_priv
Eric Dumazet [Tue, 19 May 2015 00:06:14 +0000 (17:06 -0700)]
inet: properly align icsk_ca_priv

tcp_illinois and upcoming tcp_cdg require 64bit alignment of
icsk_ca_priv

x86 does not care, but other architectures might.

Fixes: 05cbc0db03e82 ("ipv4: Create probe timer for tcp PMTU as per RFC4821")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Fan Du <fan.du@intel.com>
Acked-by: Fan Du <fan.du@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
9 years agopwm: img: Impose upper and lower timebase steps value
Naidu Tellapati [Fri, 8 May 2015 21:47:31 +0000 (18:47 -0300)]
pwm: img: Impose upper and lower timebase steps value

The PWM hardware on Pistachio platform has a maximum timebase steps
value to 255. To fix it, let's introduce a compatible-specific
data structure to contain the SoC-specific details and use it to
specify a maximum timebase.

Also, let's limit the minimum timebase to 16 steps, to allow a sane
range of duty cycle steps.

Fixes: 277bb6a29e00 ("pwm: Imagination Technologies PWM DAC driver")
Signed-off-by: Naidu Tellapati <naidu.tellapati@imgtec.com>
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@imgtec.com>
Signed-off-by: Thierry Reding <thierry.reding@gmail.com>
9 years agodrm/exynos: dp: Lower level of EDID read success message
Krzysztof Kozlowski [Thu, 14 May 2015 00:03:06 +0000 (09:03 +0900)]
drm/exynos: dp: Lower level of EDID read success message

Don't pollute the dmesg with EDID read success message as an error.
Printing as debug should be fine.

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: cleanup exynos_drm_plane
Tobias Jakobi [Sat, 25 Apr 2015 18:06:54 +0000 (20:06 +0200)]
drm/exynos: cleanup exynos_drm_plane

Remove the unused fields of struct exynos_drm_plane.

v2: Remove index_color as well, also unused (thanks Joonyoung).

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: 'win' is always unsigned
Tobias Jakobi [Wed, 6 May 2015 12:10:22 +0000 (14:10 +0200)]
drm/exynos: 'win' is always unsigned

The index for the hardware layer is always >=0. Previous
code that also used -1 as special index is now gone.

Also apply this to 'ch_enabled' (decon/fimd), since the
variable is on the same line (and is again always unsigned).

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: mixer: don't dump registers under spinlock
Tobias Jakobi [Wed, 6 May 2015 12:10:21 +0000 (14:10 +0200)]
drm/exynos: mixer: don't dump registers under spinlock

mixer_regs_dump() was called in mixer_run(), which was called
under the register spinlock in mixer_graph_buffer() and
vp_video_buffer().

This would trigger a sysmmu pagefault with drm.debug=0xff because
of the large delay caused by the register dumping.

To keep consistency also move register dumping out of mixer_stop(),
which is the counterpart to mixer_run().

Kernel dump:
[  131.296529] [drm:mixer_win_commit] win: 2
[  131.300693] [drm:mixer_regs_dump] MXR_STATUS = 00000081
[  131.305888] [drm:mixer_regs_dump] MXR_CFG = 000007d5
[  131.310835] [drm:mixer_regs_dump] MXR_INT_EN = 00000000
[  131.316043] [drm:mixer_regs_dump] MXR_INT_STATUS = 00000900
[  131.321598] [drm:mixer_regs_dump] MXR_LAYER_CFG = 00000321
[  131.327066] [drm:mixer_regs_dump] MXR_VIDEO_CFG = 00000000
[  131.332535] [drm:mixer_regs_dump] MXR_GRAPHIC0_CFG = 00310700
[  131.338263] [drm:mixer_regs_dump] MXR_GRAPHIC0_BASE = 20c00000
[  131.344079] [drm:mixer_regs_dump] MXR_GRAPHIC0_SPAN = 00000780
[  131.349895] [drm:mixer_regs_dump] MXR_GRAPHIC0_WH = 07800438
[  131.355537] [drm:mixer_regs_dump] MXR_GRAPHIC0_SXY = 00000000
[  131.361265] [drm:mixer_regs_dump] MXR_GRAPHIC0_DXY = 00000000
[  131.366994] [drm:mixer_regs_dump] MXR_GRAPHIC1_CFG = 00000000
[  131.372723] [drm:mixer_regs_dump] MXR_GRAPHIC1_BASE = 00000000
[  131.378539] [drm:mixer_regs_dump] MXR_GRAPHIC1_SPAN = 00000000
[  131.384354] [drm:mixer_regs_dump] MXR_GRAPHIC1_WH = 00000000
[  131.389996] [drm:mixer_regs_dump] MXR_GRAPHIC1_SXY = 00000000
[  131.395725] [drm:mixer_regs_dump] MXR_GRAPHIC1_DXY = 00000000
[  131.401486] PAGE FAULT occurred at 0x0 by 12e20000.sysmmu(Page table base: 0x6d990000)
[  131.409353]  Lv1 entry: 0x6e0f2401
[  131.412753] ------------[ cut here ]------------
[  131.417339] kernel BUG at drivers/iommu/exynos-iommu.c:358!
[  131.422894] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
[  131.428709] Modules linked in: ecb bridge stp llc bnep btrfs xor xor_neon zlib_inflate zlib_deflate raid6_pq btusb bluetooth usb_storage s5p_jpeg
videobuf2_dma_contig videobuf2_memops v4l2_mem2mem videobuf2_core
[  131.447461] CPU: 0 PID: 2418 Comm: lt-modetest Tainted: G        W       4.0.1-debug+ #3
[  131.455530] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  131.461607] task: ee194100 ti: ec4fe000 task.ti: ec4fe000
[  131.466995] PC is at exynos_sysmmu_irq+0x2a0/0x2a8
[  131.471766] LR is at vprintk_emit+0x268/0x594
[  131.476103] pc : [<c02781a4>]    lr : [<c00650d0>]    psr: a00001d3
[  131.476103] sp : ec4ff9d8  ip : 00000000  fp : ec4ffa14
[  131.487559] r10: ffffffda  r9 : ee206e28  r8 : ee2d1a10
[  131.492767] r7 : 00000000  r6 : 00000000  r5 : 00000000  r4 : ee206e10
[  131.499277] r3 : c06fca20  r2 : 00000000  r1 : 00000000  r0 : ee28be00
[  131.505788] Flags: NzCv  IRQs off  FIQs off  Mode SVC_32  ISA ARM  Segment user
[  131.513079] Control: 10c5387d  Table: 6c72404a  DAC: 00000015
[  131.518808] Process lt-modetest (pid: 2418, stack limit = 0xec4fe218)
[  131.525231] Stack: (0xec4ff9d8 to 0xec500000)
[  131.529571] f9c0:                                                       ec4ff9e4 c03a0c40
[  131.537732] f9e0: bbfa6e35 6d990000 6d161c3d ee20a900 ee04a7e0 00000028 ee007000 00000000
[  131.545891] fa00: 00000000 c06fb1fc ec4ffa5c ec4ffa18 c0066a34 c0277f10 ee257664 0000000b
[  131.554050] fa20: ec4ffa5c c06fafbb ee04a780 c06fb1e8 00000000 ee04a780 ee04a7e0 ee20a900
[  131.562209] fa40: ee007000 00000015 ec4ffb48 ee008000 ec4ffa7c ec4ffa60 c0066c90 c00669e0
[  131.570369] fa60: 00020000 ee04a780 ee04a7e0 00001000 ec4ffa94 ec4ffa80 c0069c6c c0066c58
[  131.578528] fa80: 00000028 ee004450 ec4ffaac ec4ffa98 c0066028 c0069bac 000000a0 c06e19b4
[  131.586687] faa0: ec4ffad4 ec4ffab0 c0223678 c0066000 c02235dc 00000015 00000000 00000015
[  131.594846] fac0: ec4ffc80 00000001 ec4ffaec ec4ffad8 c0066028 c02235e8 00000089 c06bfc54
[  131.603005] fae0: ec4ffb1c ec4ffaf0 c006633c c0066000 ec4ffb48 f002000c 00000025 00000015
[  131.611165] fb00: c06c680c ec4ffb48 f0020000 ee008000 ec4ffb44 ec4ffb20 c000867c c00662c4
[  131.619324] fb20: c02046ac 60000153 ffffffff ec4ffb7c 00000000 00000101 ec4ffbb4 ec4ffb48
[  131.627483] fb40: c0013240 c0008650 00000001 ee257508 00000002 00000001 ee257504 ee257508
[  131.635642] fb60: 00000000 c06bf27c 00000000 00000101 ee008000 ec4ffbb4 00000000 ec4ffb90
[  131.643802] fb80: c002e124 c02046ac 60000153 ffffffff c002e09c 00000000 c06c6080 00000283
[  131.651960] fba0: 00000001 c06fb1ac ec4ffc0c ec4ffbb8 c002d690 c002e0a8 ee78d080 ee008000
[  131.660120] fbc0: 00400000 c04eb3b0 ffff7c44 c06c6100 c06fdac0 0000000a c06bf2f0 c06c6080
[  131.668279] fbe0: c06bfc54 c06bfc54 00000000 00000025 00000000 00000001 ec4ffc80 ee008000
[  131.676438] fc00: ec4ffc24 ec4ffc10 c002dbb8 c002d564 00000089 c06bfc54 ec4ffc54 ec4ffc28
[  131.684597] fc20: c0066340 c002dafc ec4ffc80 f002000c 0000001c 0000000c c06c680c ec4ffc80
[  131.692757] fc40: f0020000 00000080 ec4ffc7c ec4ffc58 c000867c c00662c4 c04e6624 60000053
[  131.700916] fc60: ffffffff ec4ffcb4 c072df54 ee22d010 ec4ffcdc ec4ffc80 c0013240 c0008650
[  131.709075] fc80: ee22d664 ee194100 00000000 ec4fe000 60000053 00000400 00000002 ee22d420
[  131.717234] fca0: c072df54 ee22d010 00000080 ec4ffcdc ec4ffcc8 ec4ffcc8 c04e6620 c04e6624
[  131.725393] fcc0: 60000053 ffffffff ec4fe000 c072df54 ec4ffd34 ec4ffce0 c02b64d0 c04e6618
[  131.733552] fce0: ec4ffcf8 00000000 00000000 60000053 00010000 00010000 00000000 200cb000
[  131.741712] fd00: 20080000 ee22d664 00000001 ee256000 ee261400 ee22d420 00000080 00000080
[  131.749871] fd20: ee256000 00000280 ec4ffd74 ec4ffd38 c02a8844 c02b5fec 00000080 00000280
[  131.758030] fd40: 000001e0 00000000 00000000 00000280 000001e0 ee22d220 01e00000 00000002
[  131.766189] fd60: ee22d420 ee261400 ec4ffdbc ec4ffd78 c0293cbc c02a87a4 00000080 00000280
[  131.774348] fd80: 000001e0 00000000 00000000 02800000 01e00000 ee261400 ee22d460 ee261400
[  131.782508] fda0: ee22d420 00000000 01e00000 000001e0 ec4ffe24 ec4ffdc0 c0297800 c0293b24
[  131.790667] fdc0: 00000080 00000280 000001e0 00000000 00000000 02800000 01e00000 ec4ffdf8
[  131.798826] fde0: c028db00 00000080 00000080 ee256000 02800000 00000000 ec4ffe24 c06c6448
[  131.806985] fe00: c072df54 000000b7 ee013800 ec4ffe54 edbf7300 ec4ffe54 ec4fff04 ec4ffe28
[  131.815145] fe20: c028a848 c029768c 00000001 c06195d8 ec4ffe5c ec4ffe40 c0297680 c0521f6c
[  131.823304] fe40: 00000030 bed45d38 00000030 c03064b7 ec4ffe8c 00000011 00000015 00000022
[  131.831463] fe60: 00000000 00000080 00000080 00000280 000001e0 00000000 00000000 01e00000
[  131.839622] fe80: 02800000 00000000 00000000 0004b000 00000000 00000000 c00121e4 c0011080
[  131.847781] fea0: c00110a4 00000000 00000000 00000000 ec4ffeec ec4ffec0 c00110f0 c00121cc
[  131.855940] fec0: 00000000 c00e7fec ec4ffeec ec4ffed8 c004af2c dc8ba201 edae4fc0 edbf7000
[  131.864100] fee0: edbf7000 00000003 bed45d38 00000003 bed45d38 ee3f2040 ec4fff7c ec4fff08
[  131.872259] ff00: c010b62c c028a684 edae4fc0 00000000 00000000 b6666000 ec40d108 edae4fc4
[  131.880418] ff20: ec4fff6c ec4fff30 c00e7fec c02207b0 000001f9 00000000 edae5008 ec40d110
[  131.888577] ff40: 00070800 edae5008 edae4fc0 00070800 b6666000 edbf7000 edbf7000 c03064b7
[  131.896736] ff60: bed45d38 00000003 ec4fe000 00000000 ec4fffa4 ec4fff80 c010b84c c010b208
[  131.904896] ff80: 00000022 00000000 bed45d38 c03064b7 00000036 c000ede4 00000000 ec4fffa8
[  131.913055] ffa0: c000ec40 c010b81c 00000000 bed45d38 00000003 c03064b7 bed45d38 00000022
[  131.921214] ffc0: 00000000 bed45d38 c03064b7 00000036 00000080 00000080 00000000 000001e0
[  131.929373] ffe0: b6da4064 bed45d1c b6d98968 b6e8082c 60000050 00000003 00000000 00000000
[  131.937529] Backtrace:
[  131.939967] [<c0277f04>] (exynos_sysmmu_irq) from [<c0066a34>] (handle_irq_event_percpu+0x60/0x278)
[  131.948988]  r10:c06fb1fc r9:00000000 r8:00000000 r7:ee007000 r6:00000028 r5:ee04a7e0
[  131.956799]  r4:ee20a900
[  131.959320] [<c00669d4>] (handle_irq_event_percpu) from [<c0066c90>] (handle_irq_event+0x44/0x64)
[  131.968170]  r10:ee008000 r9:ec4ffb48 r8:00000015 r7:ee007000 r6:ee20a900 r5:ee04a7e0
[  131.975982]  r4:ee04a780
[  131.978504] [<c0066c4c>] (handle_irq_event) from [<c0069c6c>] (handle_level_irq+0xcc/0x144)
[  131.986832]  r6:00001000 r5:ee04a7e0 r4:ee04a780 r3:00020000
[  131.992478] [<c0069ba0>] (handle_level_irq) from [<c0066028>] (generic_handle_irq+0x34/0x44)
[  132.000894]  r5:ee004450 r4:00000028
[  132.004459] [<c0065ff4>] (generic_handle_irq) from [<c0223678>] (combiner_handle_cascade_irq+0x9c/0x108)
[  132.013914]  r4:c06e19b4 r3:000000a0
[  132.017476] [<c02235dc>] (combiner_handle_cascade_irq) from [<c0066028>] (generic_handle_irq+0x34/0x44)
[  132.026847]  r8:00000001 r7:ec4ffc80 r6:00000015 r5:00000000 r4:00000015 r3:c02235dc
[  132.034576] [<c0065ff4>] (generic_handle_irq) from [<c006633c>] (__handle_domain_irq+0x84/0xf0)
[  132.043252]  r4:c06bfc54 r3:00000089
[  132.046815] [<c00662b8>] (__handle_domain_irq) from [<c000867c>] (gic_handle_irq+0x38/0x70)
[  132.055144]  r10:ee008000 r9:f0020000 r8:ec4ffb48 r7:c06c680c r6:00000015 r5:00000025
[  132.062956]  r4:f002000c r3:ec4ffb48
[  132.066520] [<c0008644>] (gic_handle_irq) from [<c0013240>] (__irq_svc+0x40/0x74)
[  132.073980] Exception stack(0xec4ffb48 to 0xec4ffb90)
[  132.079016] fb40:                   00000001 ee257508 00000002 00000001 ee257504 ee257508
[  132.087176] fb60: 00000000 c06bf27c 00000000 00000101 ee008000 ec4ffbb4 00000000 ec4ffb90
[  132.095333] fb80: c002e124 c02046ac 60000153 ffffffff
[  132.100367]  r9:00000101 r8:00000000 r7:ec4ffb7c r6:ffffffff r5:60000153 r4:c02046ac
[  132.108098] [<c002e09c>] (tasklet_hi_action) from [<c002d690>] (__do_softirq+0x138/0x38c)
[  132.116251]  r8:c06fb1ac r7:00000001 r6:00000283 r5:c06c6080 r4:00000000 r3:c002e09c
[  132.123980] [<c002d558>] (__do_softirq) from [<c002dbb8>] (irq_exit+0xc8/0x104)
[  132.131268]  r10:ee008000 r9:ec4ffc80 r8:00000001 r7:00000000 r6:00000025 r5:00000000
[  132.139080]  r4:c06bfc54
[  132.141600] [<c002daf0>] (irq_exit) from [<c0066340>] (__handle_domain_irq+0x88/0xf0)
[  132.149409]  r4:c06bfc54 r3:00000089
[  132.152971] [<c00662b8>] (__handle_domain_irq) from [<c000867c>] (gic_handle_irq+0x38/0x70)
[  132.161300]  r10:00000080 r9:f0020000 r8:ec4ffc80 r7:c06c680c r6:0000000c r5:0000001c
[  132.169112]  r4:f002000c r3:ec4ffc80
[  132.172675] [<c0008644>] (gic_handle_irq) from [<c0013240>] (__irq_svc+0x40/0x74)
[  132.180137] Exception stack(0xec4ffc80 to 0xec4ffcc8)
[  132.185173] fc80: ee22d664 ee194100 00000000 ec4fe000 60000053 00000400 00000002 ee22d420
[  132.193332] fca0: c072df54 ee22d010 00000080 ec4ffcdc ec4ffcc8 ec4ffcc8 c04e6620 c04e6624
[  132.201489] fcc0: 60000053 ffffffff
[  132.204961]  r9:ee22d010 r8:c072df54 r7:ec4ffcb4 r6:ffffffff r5:60000053 r4:c04e6624
[  132.212694] [<c04e660c>] (_raw_spin_unlock_irqrestore) from [<c02b64d0>] (mixer_win_commit+0x4f0/0xcc8)
[  132.222060]  r4:c072df54 r3:ec4fe000
[  132.225625] [<c02b5fe0>] (mixer_win_commit) from [<c02a8844>] (exynos_update_plane+0xac/0xb8)
[  132.234126]  r10:00000280 r9:ee256000 r8:00000080 r7:00000080 r6:ee22d420 r5:ee261400
[  132.241937]  r4:ee256000
[  132.244461] [<c02a8798>] (exynos_update_plane) from [<c0293cbc>] (__setplane_internal+0x1a4/0x2c0)
[  132.253395]  r7:ee261400 r6:ee22d420 r5:00000002 r4:01e00000
[  132.259041] [<c0293b18>] (__setplane_internal) from [<c0297800>] (drm_mode_setplane+0x180/0x244)
[  132.267804]  r9:000001e0 r8:01e00000 r7:00000000 r6:ee22d420 r5:ee261400 r4:ee22d460
[  132.275535] [<c0297680>] (drm_mode_setplane) from [<c028a848>] (drm_ioctl+0x1d0/0x58c)
[  132.283428]  r10:ec4ffe54 r9:edbf7300 r8:ec4ffe54 r7:ee013800 r6:000000b7 r5:c072df54
[  132.291240]  r4:c06c6448
[  132.293763] [<c028a678>] (drm_ioctl) from [<c010b62c>] (do_vfs_ioctl+0x430/0x614)
[  132.301222]  r10:ee3f2040 r9:bed45d38 r8:00000003 r7:bed45d38 r6:00000003 r5:edbf7000
[  132.309034]  r4:edbf7000
[  132.311555] [<c010b1fc>] (do_vfs_ioctl) from [<c010b84c>] (SyS_ioctl+0x3c/0x64)
[  132.318842]  r10:00000000 r9:ec4fe000 r8:00000003 r7:bed45d38 r6:c03064b7 r5:edbf7000
[  132.326654]  r4:edbf7000
[  132.329176] [<c010b810>] (SyS_ioctl) from [<c000ec40>] (ret_fast_syscall+0x0/0x34)
[  132.336723]  r8:c000ede4 r7:00000036 r6:c03064b7 r5:bed45d38 r4:00000000 r3:00000022
[  132.344451] Code: e3130002 0affffaf eb09a67d eaffffad (e7f001f2)
[  132.350528] ---[ end trace d428689b94df895c ]---
[  132.355126] Kernel panic - not syncing: Fatal exception in interrupt
[  132.361465] CPU2: stopping
[  132.364155] CPU: 2 PID: 0 Comm: swapper/2 Tainted: G      D W       4.0.1-debug+ #3
[  132.371791] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  132.377866] Backtrace:
[  132.380304] [<c0012484>] (dump_backtrace) from [<c001269c>] (show_stack+0x18/0x1c)
[  132.387849]  r6:c06e158c r5:ffffffff r4:00000000 r3:dc8ba201
[  132.393497] [<c0012684>] (show_stack) from [<c04dfb94>] (dump_stack+0x88/0xc8)
[  132.400698] [<c04dfb0c>] (dump_stack) from [<c0014894>] (handle_IPI+0x1c8/0x2c4)
[  132.408073]  r6:c06bfc54 r5:c06bfc54 r4:00000005 r3:ee0b0000
[  132.413718] [<c00146cc>] (handle_IPI) from [<c00086b0>] (gic_handle_irq+0x6c/0x70)
[  132.421267]  r9:f0028000 r8:ee0b1f48 r7:c06c680c r6:fffffff5 r5:00000005 r4:f002800c
[  132.428995] [<c0008644>] (gic_handle_irq) from [<c0013240>] (__irq_svc+0x40/0x74)
[  132.436457] Exception stack(0xee0b1f48 to 0xee0b1f90)
[  132.441493] 1f40:                   00000001 00000000 00000000 c00206c0 c06c6518 c04eb3a4
[  132.449653] 1f60: 00000000 00000000 c06c0dc0 00000001 c06fb774 ee0b1f9c ee0b1fa0 ee0b1f90
[  132.457811] 1f80: c000f82c c000f830 600f0053 ffffffff
[  132.462844]  r9:00000001 r8:c06c0dc0 r7:ee0b1f7c r6:ffffffff r5:600f0053 r4:c000f830
[  132.470575] [<c000f7f0>] (arch_cpu_idle) from [<c005b6e8>] (cpu_startup_entry+0x318/0x4ec)
[  132.478818] [<c005b3d0>] (cpu_startup_entry) from [<c00144d0>] (secondary_start_kernel+0xf4/0x100)
[  132.487755]  r7:c06fd440
[  132.490279] [<c00143dc>] (secondary_start_kernel) from [<40008744>] (0x40008744)
[  132.497651]  r4:6e09006a r3:c000872c
[  132.501210] CPU3: stopping
[  132.503904] CPU: 3 PID: 0 Comm: swapper/3 Tainted: G      D W       4.0.1-debug+ #3
[  132.511539] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  132.517614] Backtrace:
[  132.520051] [<c0012484>] (dump_backtrace) from [<c001269c>] (show_stack+0x18/0x1c)
[  132.527597]  r6:c06e158c r5:ffffffff r4:00000000 r3:dc8ba201
[  132.533243] [<c0012684>] (show_stack) from [<c04dfb94>] (dump_stack+0x88/0xc8)
[  132.540446] [<c04dfb0c>] (dump_stack) from [<c0014894>] (handle_IPI+0x1c8/0x2c4)
[  132.547820]  r6:c06bfc54 r5:c06bfc54 r4:00000005 r3:ee0b2000
[  132.553466] [<c00146cc>] (handle_IPI) from [<c00086b0>] (gic_handle_irq+0x6c/0x70)
[  132.561014]  r9:f002c000 r8:ee0b3f48 r7:c06c680c r6:fffffff5 r5:00000005 r4:f002c00c
[  132.568743] [<c0008644>] (gic_handle_irq) from [<c0013240>] (__irq_svc+0x40/0x74)
[  132.576205] Exception stack(0xee0b3f48 to 0xee0b3f90)
[  132.581241] 3f40:                   00000001 00000000 00000000 c00206c0 c06c6518 c04eb3a4
[  132.589401] 3f60: 00000000 00000000 c06c0dc0 00000001 c06fb774 ee0b3f9c ee0b3fa0 ee0b3f90
[  132.597558] 3f80: c000f82c c000f830 600f0053 ffffffff
[  132.602591]  r9:00000001 r8:c06c0dc0 r7:ee0b3f7c r6:ffffffff r5:600f0053 r4:c000f830
[  132.610321] [<c000f7f0>] (arch_cpu_idle) from [<c005b6e8>] (cpu_startup_entry+0x318/0x4ec)
[  132.618566] [<c005b3d0>] (cpu_startup_entry) from [<c00144d0>] (secondary_start_kernel+0xf4/0x100)
[  132.627503]  r7:c06fd440
[  132.630023] [<c00143dc>] (secondary_start_kernel) from [<40008744>] (0x40008744)
[  132.637399]  r4:6e09006a r3:c000872c
[  132.640958] CPU1: stopping
[  132.643651] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G      D W       4.0.1-debug+ #3
[  132.651287] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[  132.657362] Backtrace:
[  132.659799] [<c0012484>] (dump_backtrace) from [<c001269c>] (show_stack+0x18/0x1c)
[  132.667344]  r6:c06e158c r5:ffffffff r4:00000000 r3:dc8ba201
[  132.672991] [<c0012684>] (show_stack) from [<c04dfb94>] (dump_stack+0x88/0xc8)
[  132.680194] [<c04dfb0c>] (dump_stack) from [<c0014894>] (handle_IPI+0x1c8/0x2c4)
[  132.687569]  r6:c06bfc54 r5:c06bfc54 r4:00000005 r3:ee0ae000
[  132.693214] [<c00146cc>] (handle_IPI) from [<c00086b0>] (gic_handle_irq+0x6c/0x70)
[  132.700762]  r9:f0024000 r8:ee0aff48 r7:c06c680c r6:fffffff5 r5:00000005 r4:f002400c
[  132.708491] [<c0008644>] (gic_handle_irq) from [<c0013240>] (__irq_svc+0x40/0x74)
[  132.715953] Exception stack(0xee0aff48 to 0xee0aff90)
[  132.720989] ff40:                   00000001 00000000 00000000 c00206c0 c06c6518 c04eb3a4
[  132.729149] ff60: 00000000 00000000 c06c0dc0 00000001 c06fb774 ee0aff9c ee0affa0 ee0aff90
[  132.737306] ff80: c000f82c c000f830 60070053 ffffffff
[  132.742339]  r9:00000001 r8:c06c0dc0 r7:ee0aff7c r6:ffffffff r5:60070053 r4:c000f830
[  132.750069] [<c000f7f0>] (arch_cpu_idle) from [<c005b6e8>] (cpu_startup_entry+0x318/0x4ec)
[  132.758314] [<c005b3d0>] (cpu_startup_entry) from [<c00144d0>] (secondary_start_kernel+0xf4/0x100)
[  132.767251]  r7:c06fd440
[  132.769772] [<c00143dc>] (secondary_start_kernel) from [<40008744>] (0x40008744)
[  132.777146]  r4:6e09006a r3:c000872c
[  132.780709] ---[ end Kernel panic - not syncing: Fatal exception in interrupt

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: Consolidate return statements in fimd_bind()
Krzysztof Kozlowski [Thu, 7 May 2015 00:04:46 +0000 (09:04 +0900)]
drm/exynos: Consolidate return statements in fimd_bind()

Simplify the code and remove superfluous return statement. Just return
the result of fimd_iommu_attach_devices().

Signed-off-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Reviewed-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Tested-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: Constify exynos_drm_crtc_ops
Krzysztof Kozlowski [Thu, 7 May 2015 00:04:45 +0000 (09:04 +0900)]
drm/exynos: Constify exynos_drm_crtc_ops

The Exynos DRM code does not modify the ops provided by CRTC driver in
exynos_drm_crtc_create() call.

Signed-off-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Reviewed-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Tested-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
9 years agodrm/exynos: Fix build breakage on !DRM_EXYNOS_FIMD
Krzysztof Kozlowski [Thu, 7 May 2015 00:04:44 +0000 (09:04 +0900)]
drm/exynos: Fix build breakage on !DRM_EXYNOS_FIMD

Disabling the CONFIG_DRM_EXYNOS_FIMD (e.g. by enabling of CONFIG_FB_S3C)
leads to build error:

drivers/built-in.o: In function `exynos_dp_dpms':
binder.c:(.text+0xd6a840): undefined reference to `fimd_dp_clock_enable'
binder.c:(.text+0xd6ab54): undefined reference to `fimd_dp_clock_enable'

Fix this by changing direct call to fimd_dp_clock_enable() into optional
call to exynos_drm_crtc_ops->clock_enable(). Only the DRM_EXYNOS_FIMD
implements this op.

Signed-off-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Reviewed-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Tested-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>