Commit graph

708,694 commits

Author SHA1 Message Date
Alan Stern
63aea0dbab USB: xhci: fix lock-inversion problem
With threaded interrupts, bottom-half handlers are called with
interrupts enabled.  Therefore they can't safely use spin_lock(); they
have to use spin_lock_irqsave().  Lockdep warns about a violation
occurring in xhci_irq():

=========================================================
[ INFO: possible irq lock inversion dependency detected ]
4.11.0-rc8-dbg+ #1 Not tainted
---------------------------------------------------------
swapper/7/0 just changed the state of lock:
 (&(&ehci->lock)->rlock){-.-...}, at: [<ffffffffa0130a69>]
ehci_hrtimer_func+0x29/0xc0 [ehci_hcd]
but this lock took another, HARDIRQ-unsafe lock in the past:
 (hcd_urb_list_lock){+.....}

and interrupts could create inverse lock ordering between them.

other info that might help us debug this:
 Possible interrupt unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(hcd_urb_list_lock);
                               local_irq_disable();
                               lock(&(&ehci->lock)->rlock);
                               lock(hcd_urb_list_lock);
  <Interrupt>
    lock(&(&ehci->lock)->rlock);
 *** DEADLOCK ***

no locks held by swapper/7/0.
the shortest dependencies between 2nd lock and 1st lock:
 -> (hcd_urb_list_lock){+.....} ops: 252 {
    HARDIRQ-ON-W at:
                      __lock_acquire+0x602/0x1280
                      lock_acquire+0xd5/0x1c0
                      _raw_spin_lock+0x2f/0x40
                      usb_hcd_unlink_urb_from_ep+0x1b/0x60 [usbcore]
                      xhci_giveback_urb_in_irq.isra.45+0x70/0x1b0 [xhci_hcd]
                      finish_td.constprop.60+0x1d8/0x2e0 [xhci_hcd]
                      xhci_irq+0xdd6/0x1fa0 [xhci_hcd]
                      usb_hcd_irq+0x26/0x40 [usbcore]
                      irq_forced_thread_fn+0x2f/0x70
                      irq_thread+0x149/0x1d0
                      kthread+0x113/0x150
                      ret_from_fork+0x2e/0x40

This patch fixes the problem.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Reported-and-tested-by: Bart Van Assche <bart.vanassche@sandisk.com>
CC: <stable@vger.kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-05-18 15:19:42 +02:00
Peter Chen
6a29beef9d usb: host: xhci-ring: don't need to clear interrupt pending for MSI enabled hcd
According to xHCI spec Figure 30: Interrupt Throttle Flow Diagram

	If PCI Message Signaled Interrupts (MSI or MSI-X) are enabled,
       	then the assertion of the Interrupt Pending (IP) flag in Figure 30
       	generates a PCI Dword write. The IP flag is automatically cleared
       	by the completion of the PCI write.

the MSI enabled HCs don't need to clear interrupt pending bit, but
hcd->irq = 0 doesn't equal to MSI enabled HCD. At some Dual-role
controller software designs, it sets hcd->irq as 0 to avoid HCD
requesting interrupt, and they want to decide when to call usb_hcd_irq
by software.

Signed-off-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-05-18 15:19:41 +02:00
Peter Chen
7480d912d5 usb: host: xhci-mem: allocate zeroed Scratchpad Buffer
According to xHCI ch4.20 Scratchpad Buffers, the Scratchpad
Buffer needs to be zeroed.

	...
	The following operations take place to allocate
       	Scratchpad Buffers to the xHC:
	...
		b. Software clears the Scratchpad Buffer to '0'

Cc: stable <stable@vger.kernel.org>
Signed-off-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-05-18 15:19:41 +02:00
Mathias Nyman
a0c16630d3 xhci: apply PME_STUCK_QUIRK and MISSING_CAS quirk for Denverton
Intel Denverton microserver is Atom based and need the PME and CAS quirks
as well.

Cc: stable <stable@vger.kernel.org>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-05-18 15:19:41 +02:00
Mathias Nyman
7bc5d5aff3 usb: xhci: trace URB before giving it back instead of after
Don't access any members of a URB after giving it back.
URB might be freed by then already.

Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-05-18 15:19:41 +02:00
Dan Carpenter
d3e7dec054 KVM: Silence underflow warning in avic_get_physical_id_entry()
Smatch complains that we check cap the upper bound of "index" but don't
check for negatives.  It's a false positive because "index" is never
negative.  But it's also simple enough to make it unsigned which makes
the code easier to audit.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
2017-05-18 14:53:54 +02:00
Hans de Goede
d8db7ae4ee drm/i915: Fix new -Wint-in-bool-context gcc compiler warning
This commit fixes the following compiler warning:

drivers/gpu/drm/i915/intel_dsi.c: In function ‘intel_dsi_prepare’:
drivers/gpu/drm/i915/intel_dsi.c:1487:23: warning:
    ?: using integer constants in boolean context [-Wint-in-bool-context]
       PORT_A ? PORT_C : PORT_A),

Fixes: f4c3a88e5f ("drm/i915: Tighten mmio arrays for MIPI_PORT")
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170518110644.9902-1-hdegoede@redhat.com
(cherry picked from commit 0ad4dc887d)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2017-05-18 15:42:28 +03:00
Radim Krčmář
55c315ee09 KVM/ARM Fixes for v4.12-rc2.
Includes:
  - A fix for a build failure introduced in -rc1 when tracepoints are
    enabled on 32-bit ARM.
  - Disabling use of stack pointer protection in the hyp code which can
    cause panics.
  - A handful of VGIC fixes.
  - A fix to the init of the redistributors on GICv3 systems that
    prevented boot with kvmtool on GICv3 systems introduced in -rc1.
  - A number of race conditions fixed in our MMU handling code.
  - A fix for the guest being able to program the debug extensions for
    the host on the 32-bit side.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJZHWzeAAoJEEtpOizt6ddyFO0H/jpgdDvmu8mL+sk4xdjCxsqE
 sl0uH8vnlziTmIFtRlXKtB1ecL+waj22YEoVLIA3lXz92uZXF3RDoFqVW3NGnpbO
 pFyUs8ZHfhyo3nHI7DZqT+/SeButwwX+cw02tRpaEOPhlun7BlEZEco26r2y/2xR
 WZdTDYYkAjTtNtY1dJ7xzNrhSJZXpf54rvQshYSbqn+gVdGVHaosQygMPohU8tOF
 V0AX3gQ3pnR3hEgE3Cz+F3TGhORg9buOADS28CdvDCx6ekJdy55Snf6AHFoZ9cdT
 n4YZpon2HO1ZIxpMLvg1Ud+M3bQxeYLBWj2av3PAHsMFoR23x4xTH6YVA53sfFY=
 =SJeq
 -----END PGP SIGNATURE-----

Merge tag 'kvm-arm-for-v4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm

KVM/ARM Fixes for v4.12-rc2.

Includes:
 - A fix for a build failure introduced in -rc1 when tracepoints are
   enabled on 32-bit ARM.
 - Disabling use of stack pointer protection in the hyp code which can
   cause panics.
 - A handful of VGIC fixes.
 - A fix to the init of the redistributors on GICv3 systems that
   prevented boot with kvmtool on GICv3 systems introduced in -rc1.
 - A number of race conditions fixed in our MMU handling code.
 - A fix for the guest being able to program the debug extensions for
   the host on the 32-bit side.
2017-05-18 14:40:32 +02:00
Matthew Auld
171d8b9363 drm/i915: use vma->size for appgtt allocate_va_range
For the aliasing ppgtt we clear the va range up to vma->size, but seem
to allocate up to vma->node.size, which is a little inconsistent given
that vma->node.size >= vma->size. Not that is really matters all that
much since we preallocate anyway, but for consistency just use
vma->size.

Fixes: ff685975d9 ("drm/i915: Move allocate_va_range to GTT")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170516085514.5853-1-matthew.auld@intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
(cherry picked from commit d567232cbd)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2017-05-18 15:37:47 +03:00
Colin Ian King
52d9d38c18 drm/sti:fix spelling mistake: "compoment" -> "component"
Trivial fix to spelling mistake in DRM_ERROR message

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@linaro.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20170513224150.19955-1-colin.king@canonical.com
2017-05-18 14:12:52 +02:00
Tobias Regnery
c42c88e6c8 Bluetooth: hci_nokia: select BT_HCIUART_H4
We see the following build failure with CONFIG_BT_HCIUART_NOKIA=y and
CONFIG_BT_HCIUART_H4=n:

drivers/bluetooth/hci_nokia.c: In function 'nokia_recv':
drivers/bluetooth/hci_nokia.c:644:18: error: implicit declaration of function 'h4_recv_buf' [-Werror=implicit-function-declaration]
...

Fix this by selecting the BT_HCIUART_H4 symbol like all the other users
of the protocoll.

Fixes: 7bb318680e ("Bluetooth: add nokia driver")
Signed-off-by: Tobias Regnery <tobias.regnery@gmail.com>
Reviewed-by: Sebastian Reichel <sre@kernel.org>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2017-05-18 13:59:05 +02:00
Dean Jenkins
dec2c92880 Bluetooth: hci_ldisc: Use rwlocking to avoid closing proto races
When HCI_UART_PROTO_READY is in the set state, the Data Link protocol
layer (proto) is bound to the HCI UART driver. This state allows the
registered proto function pointers to be used by the HCI UART driver.

When unbinding (closing) the Data Link protocol layer, the proto
function pointers much be prevented from being used immediately before
running the proto close function pointer. Otherwise, there is a risk
that a proto non-close function pointer is used during or after the
proto close function pointer is used. The consequences are likely to
be a kernel crash because the proto close function pointer will free
resources used in the Data Link protocol layer.

Therefore, add a reader writer lock (rwlock) solution to prevent the
close proto function pointer from running by using write_lock_irqsave()
whilst the other proto function pointers are protected using
read_lock(). This means HCI_UART_PROTO_READY can safely be cleared
in the knowledge that no proto function pointers are running.

When flag HCI_UART_PROTO_READY is put into the clear state,
proto close function pointer can safely be run. Note
flag HCI_UART_PROTO_SET being in the set state prevents the proto
open function pointer from being run so there is no race condition
between proto open and close function pointers.

Signed-off-by: Dean Jenkins <Dean_Jenkins@mentor.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2017-05-18 13:52:50 +02:00
Marcel Holtmann
b56c7b2548 Bluetooth: Skip vendor diagnostic configuration for HCI User Channel
When the HCI User Channel access is requested, then do not try to
undermine it with vendor diagnostic configuration. The exclusive user
is required to configure its own vendor diagnostic in that case and
can not rely on the host stack support.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2017-05-18 13:52:49 +02:00
Tobias Regnery
76c4969fec Bluetooth: hci_uart: fix kconfig dependency
We see the following link error with CONFIG_BT_HCIUART=y,
CONFIG_BT_HCIUART_LL=y and CONFIG_SERIAL_DEV_BUS=m:

drivers/built-in.o: In function 'll_close':
supp.c:(.text+0x55add4): undefined reference to 'serdev_device_close'
supp.c:(.text+0x55add4): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol 'serdev_device_close'
drivers/built-in.o: In function 'll_open':
supp.c:(.text+0x55aed0): undefined reference to 'serdev_device_open'
supp.c:(.text+0x55aed0): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol 'serdev_device_open'
drivers/built-in.o: In function `hci_ti_probe':
supp.c:(.text+0x55b00c): undefined reference to 'hci_uart_register_device'
supp.c:(.text+0x55b00c): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol 'hci_uart_register_device'
drivers/built-in.o: In function `ll_setup':
supp.c:(.text+0x55b08c): undefined reference to 'serdev_device_set_flow_control'
supp.c:(.text+0x55b08c): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol 'serdev_device_set_flow_control'
supp.c:(.text+0x55b324): undefined reference to 'serdev_device_set_baudrate'
supp.c:(.text+0x55b324): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol 'serdev_device_set_baudrate'
drivers/built-in.o: In function 'll_init':
supp.c:(.init.text+0x1b508): undefined reference to '__serdev_device_driver_register'
supp.c:(.init.text+0x1b508): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol '__serdev_device_driver_register'

Fix this by dependig BT_HCIUART_LL on the BT_HCIUART_SERDEV symbol.
This implies a dependency on BT_HCIUART and hci_ll.c is only compiled in
if SERIAl_DEV_BUS is built in or SERIAL_DEV_BUS and BT_HCIUART are
modules.

Fixes: 371805522f ("bluetooth: hci_uart: add LL protocol serdev driver support")
Signed-off-by: Tobias Regnery <tobias.regnery@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2017-05-18 13:52:49 +02:00
Marcel Holtmann
de2ba3039c Bluetooth: Set LE Default PHY preferences
If the LE Set Default PHY command is supported, the indicate to the
controller that the host has no preferences for transmitter PHY or
receiver PHY selection.

Issuing this command gives the controller a clear indication that other
PHY can be selected if available.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2017-05-18 13:52:49 +02:00
Marcel Holtmann
27bbca4402 Bluetooth: Enable LE PHY Update Complete event
If either LE Set Default PHY command or LE Set PHY commands is
supported, then enable the LE PHY Update Complete event.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2017-05-18 13:52:49 +02:00
Marcel Holtmann
9756d33b85 Bluetooth: Enable LE Channel Selection Algorithm event
If the Channel Selection Algorithm #2 feature is supported, then enable
the new LE Channel Selection Algorithm event.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2017-05-18 13:52:49 +02:00
Marcel Holtmann
122048752e Bluetooth: Set LE Suggested Default Data Length to maximum
When LE Data Packet Length Extension is supported, then actually
increase the suggested default data length to the maximum to enable
higher througput.

< HCI Command: LE Read Maximum Data Length (0x08|0x002f) plen 0
> HCI Event: Command Complete (0x0e) plen 12
      LE Read Maximum Data Length (0x08|0x002f) ncmd 1
        Status: Success (0x00)
        Max TX octets: 251
        Max TX time: 2120
        Max RX octets: 251
        Max RX time: 2120

< HCI Command: LE Read Suggested Default Data Length (0x08|0x0023) plen 0
> HCI Event: Command Complete (0x0e) plen 8
      LE Read Suggested Default Data Length (0x08|0x0023) ncmd 1
        Status: Success (0x00)
        TX octets: 27
        TX time: 328

< HCI Command: LE Write Suggested Default Data Length (0x08|0x0024) plen 4
        TX octets: 251
        TX time: 2120
> HCI Event: Command Complete (0x0e) plen 4
      LE Write Suggested Default Data Length (0x08|0x0024) ncmd 1
        Status: Success (0x00)

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2017-05-18 13:52:49 +02:00
Tedd Ho-Jeong An
86a6129ae2 Bluetooth: Add support for Intel Bluetooth device 9460/9560 [8087:0aaa]
This patch adds support for Intel Bluetooth device 9460/9560 also known
as Jefferson Peak (JfP). The firmware downloading mechanism is same as
previous generation. So include the new USB product identifier and
whitelist the hardware variant.

T:  Bus=01 Lev=01 Prnt=01 Port=09 Cnt=04 Dev#=  5 Spd=12   MxCh= 0
D:  Ver= 2.01 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=8087 ProdID=0aaa Rev= 0.02
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  64 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:  If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  63 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  63 Ivl=1ms

Bootloader version:
< HCI Command: Intel Read Version (0x3f|0x0005) plen 0
> HCI Event: Command Complete (0x0e) plen 13
      Intel Read Version (0x3f|0x0005) ncmd 32
        Status: Success (0x00)
        Hardware platform: 0x37
        Hardware variant: 0x11
        Hardware revision: 0.0
        Firmware variant: 0x06
        Firmware revision: 0.1
        Firmware build: 42-52.2015
        Firmware patch: 0

Signed-off-by: Tedd Ho-Jeong An <tedd.an@intel.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
2017-05-18 13:52:49 +02:00
Hans de Goede
0ad4dc887d drm/i915: Fix new -Wint-in-bool-context gcc compiler warning
This commit fixes the following compiler warning:

drivers/gpu/drm/i915/intel_dsi.c: In function ‘intel_dsi_prepare’:
drivers/gpu/drm/i915/intel_dsi.c:1487:23: warning:
    ?: using integer constants in boolean context [-Wint-in-bool-context]
       PORT_A ? PORT_C : PORT_A),

Fixes: f4c3a88e5f ("drm/i915: Tighten mmio arrays for MIPI_PORT")
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170518110644.9902-1-hdegoede@redhat.com
2017-05-18 14:32:23 +03:00
Joonas Lahtinen
4681ee21d6 drm/i915: Do not sync RCU during shrinking
Due to the complex dependencies between workqueues and RCU, which
are not easily detected by lockdep, do not synchronize RCU during
shrinking.

On low-on-memory systems (mem=1G for example), the RCU sync leads
to all system workqueus freezing and unrelated lockdep splats are
displayed according to reports. GIT bisecting done by J. R.
Okajima points to the commit where RCU syncing was extended.

RCU sync gains us very little benefit in real life scenarios
where the amount of memory used by object backing storage is
dominant over the metadata under RCU, so drop it altogether.

 " Yeeeaah, if core could just, go ahead and reclaim RCU
   queues, that'd be great. "

  - Chris Wilson, 2016 (0eafec6d32)

v2: More information to commit message.
v3: Remove "grep _rcu_" escapee from i915_gem_shrink_all (Andrea)

Fixes: c053b5a506 ("drm/i915: Don't call synchronize_rcu_expedited under struct_mutex")
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Reported-by: J. R. Okajima <hooanon05g@gmail.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Tested-by: Hugh Dickins <hughd@google.com>
Tested-by: Andrea Arcangeli <aarcange@redhat.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: J. R. Okajima <hooanon05g@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: <stable@vger.kernel.org> # v4.11+
(cherry picked from commit 73cc0b9aa9)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1495097379-573-1-git-send-email-joonas.lahtinen@linux.intel.com
2017-05-18 14:15:20 +03:00
Willem de Bruijn
751a9c7638 netfilter: xtables: fix build failure from COMPAT_XT_ALIGN outside CONFIG_COMPAT
The patch in the Fixes references COMPAT_XT_ALIGN in the definition
of XT_DATA_TO_USER, outside an #ifdef CONFIG_COMPAT block.

Split XT_DATA_TO_USER into separate compat and non compat variants and
define the first inside an CONFIG_COMPAT block.

This simplifies both variants by removing branches inside the macro.

Fixes: 324318f024 ("netfilter: xtables: zero padding in data_to_user")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
2017-05-18 13:10:03 +02:00
Philipp Zabel
f2c61f98e0 [media] tc358743: fix register i2c_rd/wr function fix
The below mentioned fix contains a small but severe bug,
fix it to make the driver work again.

Fixes: 3538aa6ecf ("[media] tc358743: fix register i2c_rd/wr functions")

Cc: Hans Verkuil <hans.verkuil@cisco.com>
Cc: Mauro Carvalho Chehab <mchehab@s-opensource.com>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hans Verkuil <hansverk@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 08:00:56 -03:00
Arnd Bergmann
79eddc992e [media] cec-notifier.h: handle unreachable CONFIG_CEC_CORE
Fix a link error in this specific combination of config options:

CONFIG_MEDIA_CEC_SUPPORT=y
CONFIG_CEC_CORE=m
CONFIG_MEDIA_CEC_NOTIFIER=y
CONFIG_VIDEO_STI_HDMI_CEC=m
CONFIG_DRM_STI=y

drivers/gpu/drm/sti/sti_hdmi.o: In function `sti_hdmi_remove':
sti_hdmi.c:(.text.sti_hdmi_remove+0x10): undefined reference to
`cec_notifier_set_phys_addr'
sti_hdmi.c:(.text.sti_hdmi_remove+0x34): undefined reference to
`cec_notifier_put'
drivers/gpu/drm/sti/sti_hdmi.o: In function `sti_hdmi_connector_get_modes':
sti_hdmi.c:(.text.sti_hdmi_connector_get_modes+0x4a): undefined
reference to `cec_notifier_set_phys_addr_from_edid'
drivers/gpu/drm/sti/sti_hdmi.o: In function `sti_hdmi_probe':
sti_hdmi.c:(.text.sti_hdmi_probe+0x204): undefined reference to
`cec_notifier_get'
drivers/gpu/drm/sti/sti_hdmi.o: In function `sti_hdmi_connector_detect':
sti_hdmi.c:(.text.sti_hdmi_connector_detect+0x36): undefined reference
to `cec_notifier_set_phys_addr'
drivers/gpu/drm/sti/sti_hdmi.o: In function `sti_hdmi_disable':
sti_hdmi.c:(.text.sti_hdmi_disable+0xc0): undefined reference to
`cec_notifier_set_phys_addr'

The version below seems to work, though I don't particularly
like the IS_REACHABLE() addition since that can be confusing
to users.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:33:31 -03:00
Christophe JAILLET
ef6ff8f472 [media] vb2: Fix an off by one error in 'vb2_plane_vaddr'
We should ensure that 'plane_no' is '< vb->num_planes' as done in
'vb2_plane_cookie' just a few lines below.

Fixes: e23ccc0ad9 ("[media] v4l: add videobuf2 Video for Linux 2 driver framework")

Cc: stable@vger.kernel.org
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:33:04 -03:00
Wei Yongjun
df0e1fb999 [media] s5p-cec: remove unused including <linux/version.h>
Remove including <linux/version.h> that is not needed.

Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:32:40 -03:00
Wei Yongjun
be321a82bf [media] rainshadow-cec: Fix missing spin_lock_init()
The driver allocates the spinlock but not initialize it.
Use spin_lock_init() on it to initialize it correctly.

This is detected by Coccinelle semantic patch.

Fixes: 0f314f6c2e ("[media] rainshadow-cec: new RainShadow Tech HDMI
CEC driver")

Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:30:31 -03:00
Petr Cvek
e3b4d10cc0 [media] pxa_camera: fix module remove codepath for v4l2 clock
The conversion from soc_camera omitted a correct handling of the clock
gating for a sensor. When the pxa_camera driver module was removed it
tried to unregister clk, but this caused a similar warning:

  WARNING: CPU: 0 PID: 6740 at drivers/media/v4l2-core/v4l2-clk.c:278
  v4l2_clk_unregister(): Refusing to unregister ref-counted 0-0030 clock!

The clock was at time still refcounted by the sensor driver. Before
the removing of the pxa_camera the clock must be dropped by the sensor
driver. This should be triggered by v4l2_async_notifier_unregister() call
which removes sensor driver module too, calls unbind() function and then
tries to probe sensor driver again. Inside unbind() we can safely
unregister the v4l2 clock as the sensor driver got removed. The original
v4l2_clk_unregister() should be put inside test as the clock can be
already unregistered from unbind(). If there was not any bound sensor
the clock is still present.

The codepath is practically a copy from the old soc_camera. The bug was
tested with a pxa_camera+ov9640 combination during the conversion
of the ov9640 from the soc_camera.

Signed-off-by: Petr Cvek <petr.cvek@tul.cz>
Tested-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:29:57 -03:00
Arnd Bergmann
71e22ed1e2 [media] cec: improve MEDIA_CEC_RC dependencies
Changing the IS_REACHABLE() into a plain #ifdef broke the case of
CONFIG_MEDIA_RC=m && CONFIG_MEDIA_CEC=y:

drivers/media/cec/cec-core.o: In function `cec_unregister_adapter':
cec-core.c:(.text.cec_unregister_adapter+0x18): undefined reference to `rc_unregister_device'
drivers/media/cec/cec-core.o: In function `cec_delete_adapter':
cec-core.c:(.text.cec_delete_adapter+0x54): undefined reference to `rc_free_device'
drivers/media/cec/cec-core.o: In function `cec_register_adapter':
cec-core.c:(.text.cec_register_adapter+0x94): undefined reference to `rc_register_device'
cec-core.c:(.text.cec_register_adapter+0xa4): undefined reference to `rc_free_device'
cec-core.c:(.text.cec_register_adapter+0x110): undefined reference to `rc_unregister_device'
drivers/media/cec/cec-core.o: In function `cec_allocate_adapter':
cec-core.c:(.text.cec_allocate_adapter+0x234): undefined reference to `rc_allocate_device'
drivers/media/cec/cec-adap.o: In function `cec_received_msg':
cec-adap.c:(.text.cec_received_msg+0x734): undefined reference to `rc_keydown'
cec-adap.c:(.text.cec_received_msg+0x768): undefined reference to `rc_keyup'

This adds an additional dependency to explicitly forbid this combination.

Fixes: 5f2c467c54 ("[media] cec: add MEDIA_CEC_RC config option")

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:29:09 -03:00
Arnd Bergmann
ea6a69defd [media] rainshadow-cec: avoid -Wmaybe-uninitialized warning
The barrier implied by spin_unlock() in rain_irq_work_handler makes it hard
for gcc to figure out the state of the variables, leading to a false-positive
warning:

drivers/media/usb/rainshadow-cec/rainshadow-cec.c: In function 'rain_irq_work_handler':
drivers/media/usb/rainshadow-cec/rainshadow-cec.c:171:31: error: 'data' may be used uninitialized in this function [-Werror=maybe-uninitialized]

Slightly rearranging the code makes it easier for the compiler to see that the
code is correct, and gets rid of the warning.

Fixes: 0f314f6c2e ("[media] rainshadow-cec: new RainShadow Tech HDMI CEC driver")

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:28:38 -03:00
Arnd Bergmann
5c621744b4 [media] rainshadow-cec: use strlcat instead of strncat
gcc warns about an obviously incorrect use of strncat():

drivers/media/usb/rainshadow-cec/rainshadow-cec.c: In function 'rain_cec_adap_transmit':
drivers/media/usb/rainshadow-cec/rainshadow-cec.c:299:4: error: specified bound 48 equals the size of the destination [-Werror=stringop-overflow=]

It seems that strlcat was intended here, and using that makes the
code correct.

Fixes: 0f314f6c2e ("[media] rainshadow-cec: new RainShadow Tech HDMI CEC driver")

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 07:24:46 -03:00
David Härdeman
0f7c4063f8 [media] ir-lirc-codec: let lirc_dev handle the lirc_buffer
ir_lirc_register() currently creates its own lirc_buffer before
passing the lirc_driver to lirc_register_driver().

When a module is later unloaded, ir_lirc_unregister() gets called
which performs a call to lirc_unregister_driver() and then free():s
the lirc_buffer.

The problem is that:

a) there can still be a userspace app holding an open lirc fd
   when lirc_unregister_driver() returns; and

b) the lirc_buffer contains "wait_queue_head_t wait_poll" which
   is potentially used as long as any userspace app is still around.

The result is an oops which can be triggered quite easily by a
userspace app monitoring its lirc fd using epoll() and not closing
the fd promptly on device removal.

The minimalistic fix is to let lirc_dev create the lirc_buffer since
lirc_dev will then also free the buffer once it believes it is safe to
do so.

Signed-off-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 06:19:29 -03:00
David Härdeman
b2aceb739b [media] rc-core: fix input repeat handling
The call to input_register_device() needs to take place
before the repeat parameters are set or the input subsystem
repeat handling will be disabled (as was already noted in
the comments in that function).

Cc: stable <stable@vger.kernel.org> # v4.11
Signed-off-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 06:18:21 -03:00
Christoffer Dall
fa472fa91a KVM: arm/arm64: Hold slots_lock when unregistering kvm io bus devices
We were not holding the kvm->slots_lock as required when calling
kvm_io_bus_unregister_dev() as required.

This only affects the error path, but still, let's do our due
diligence.

Reported by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
2017-05-18 11:18:16 +02:00
Christoffer Dall
552c9f47f8 KVM: arm/arm64: Fix bug when registering redist iodevs
If userspace creates the VCPUs after initializing the VGIC, then we end
up in a situation where we trigger a bug in kvm_vcpu_get_idx(), because
it is called prior to adding the VCPU into the vcpus array on the VM.

There is no tight coupling between the VCPU index and the area of the
redistributor region used for the VCPU, so we can simply ensure that all
creations of redistributors are serialized per VM, and increment an
offset when we successfully add a redistributor.

The vgic_register_redist_iodev() function can be called from two paths:
vgic_redister_all_redist_iodev() which is called via the kvm_vgic_addr()
device attribute handler.  This patch already holds the kvm->lock mutex.

The other path is via kvm_vgic_vcpu_init, which is called through a
longer chain from kvm_vm_ioctl_create_vcpu(), which releases the
kvm->lock mutex just before calling kvm_arch_vcpu_create(), so we can
simply take this mutex again later for our purposes.

Fixes: ab6f468c10 ("KVM: arm/arm64: Register iodevs when setting redist base and creating VCPUs")
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Tested-by: Jean-Philippe Brucker <jean-philippe.brucker@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
2017-05-18 11:18:12 +02:00
Sean Young
592ddc9f7d [media] sir_ir: infinite loop in interrupt handler
Since this driver does no detection of hardware, it might be used with
a non-sir port. Escape out if we are spinning.

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 06:16:41 -03:00
Mauro Carvalho Chehab
dd8245f445 [media] atomisp: don't treat warnings as errors
Several atomisp files use:
	 ccflags-y += -Werror

As, on media, our usual procedure is to use W=1, and atomisp
has *a lot* of warnings with such flag enabled,like:

./drivers/staging/media/atomisp/pci/atomisp2/css2400/hive_isp_css_common/host/system_local.h:62:26: warning: 'DDR_BASE' defined but not used [-Wunused-const-variable=]

At the end, it causes our build to fail, impacting our workflow.

So, remove this crap. If one wants to force -Werror, he
can still build with it enabled by passing a parameter to
make.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
2017-05-18 05:44:00 -03:00
Quentin Schulz
d14ccea0e7 can: m_can: add deep Suspend/Resume support
This adds Power Management deep Suspend/Resume support for Bosch M_CAN
chip.

When entering deep sleep, the clocks are gated, the interrupts are
disabled. When resuming from deep sleep, the chip needs to be
reinitialized, the clocks ungated and the interrupts enabled.

Signed-off-by: Quentin Schulz <quentin.schulz@free-electrons.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2017-05-18 10:32:48 +02:00
Quentin Schulz
ef7b8aa8ca can: m_can: factorize clock gating and ungating
This creates a function to ungate M_CAN clocks and another to gate the
same clocks, then swaps all gating/ungating code with their respective
function.

Signed-off-by: Quentin Schulz <quentin.schulz@free-electrons.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2017-05-18 10:32:48 +02:00
Quentin Schulz
8a3f3f24a8 can: m_can: make m_can_start and m_can_stop symmetric
This moves clocks gating outside of the m_can_stop function as the
m_can_start function does not (and cannot, at least in current
implementation) ungate clocks. This way, both functions can now be used
symmetrically.

Signed-off-by: Quentin Schulz <quentin.schulz@free-electrons.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2017-05-18 10:32:48 +02:00
Quentin Schulz
6a792e8158 can: m_can: move Message RAM initialization to function
To avoid possible ECC/parity checksum errors when reading an
uninitialized buffer, the entire Message RAM is initialized when probing
the driver. This initialization is done in the same function reading the
Device Tree properties.

This patch moves the RAM initialization to a separate function so it can
be called separately from device initialization from Device Tree.

Signed-off-by: Quentin Schulz <quentin.schulz@free-electrons.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
2017-05-18 10:32:48 +02:00
Icenowy Zheng
2a451bfa98 ARM: sun8i: v3s: enable SPI
Allwinner V3s SoC has a SPI controller, muxed with the MMC2 controller
at PC bank. The controller itself is identical to the one in H3 SoC.

Add device tree node and the only pinmux node for it.

Tested with a Winbond W25Q128FV SPI NOR soldered on the Lichee Pi
early sample.

Signed-off-by: Icenowy Zheng <icenowy@aosc.io>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-05-18 09:41:46 +02:00
Icenowy Zheng
87ac8e18ee ARM: sun8i: v3s: add support for Lichee Pi Zero w/ dock board
The Lichee Pi Zero board has a "dock board" which needs to be soldered
with the 1.27mm stamp holes on a Lichee Pi Zero board.

It features:
- Onboard MIC and headphone jack (not supported yet)
- Ethernet port (not supported yet)
- An extra MicroSD slot connected to MMC1 controller
- four keys connected to the LRADC.

As it needs to be soldered with the main board to use, add a stand-alone
device tree for it.

Signed-off-by: Icenowy Zheng <icenowy@aosc.io>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-05-18 09:41:45 +02:00
Arnd Bergmann
f36afd38c5 clk: sunxi-ng: enable SUNXI_CCU_MP for PRCM
The newly added PRCM CCU driver uses SUNXI_CCU_MP_WITH_MUX_GATE, which causes
a link error when no other driver enables SUNXI_CCU_MP:

drivers/clk/built-in.o:(.data+0x5c8c8): undefined reference to `ccu_mp_ops'

This adds an explicit 'select' statement for it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
2017-05-18 08:49:18 +02:00
Gabriel Krisman Bertazi
5e851c9de7 drm: trivial documentation fix to drm_for_each_connector_iter
While reading drm_for_each_connector_iter, I noticed a mention to
drm_connector_begin which doesn't exist. It should be
drm_connector_get.

Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170421003819.17685-1-krisman@collabora.co.uk
2017-05-18 07:57:32 +02:00
Sabrina Dubroca
cce2ea8d90 crypto: aesni - add generic gcm(aes)
Now that the asm side of things can support all the valid lengths of ICV
and all lengths of associated data, provide the glue code to expose a
generic gcm(aes) crypto algorithm.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-05-18 13:19:55 +08:00
Sabrina Dubroca
2df7e81362 crypto: aesni - make AVX2 AES-GCM work with all valid auth_tag_len
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-05-18 13:19:55 +08:00
Sabrina Dubroca
27352c45c5 crypto: aesni - make AVX2 AES-GCM work with any aadlen
This is the first step to make the aesni AES-GCM implementation
generic. The current code was written for rfc4106, so it handles only
some specific sizes of associated data.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-05-18 13:19:54 +08:00
Sabrina Dubroca
0120af7795 crypto: aesni - make AVX AES-GCM work with all valid auth_tag_len
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-05-18 13:19:54 +08:00
Sabrina Dubroca
e10f9cb223 crypto: aesni - make AVX AES-GCM work with any aadlen
This is the first step to make the aesni AES-GCM implementation
generic. The current code was written for rfc4106, so it handles
only some specific sizes of associated data.

Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-05-18 13:19:53 +08:00