linux-toradex.git/drivers/watchdog, branch v4.9.70

watchdog: kempld: fix gcc-4.3 build

2017-10-21T15:21:33+00:00

[ Upstream commit 3736d4eb6af37492aeded7fec0072dedd959c842 ]

gcc-4.3 can't decide whether the constant value in
kempld_prescaler[PRESCALER_21] is built-time constant or
not, and gets confused by the logic in do_div():

drivers/watchdog/kempld_wdt.o: In function `kempld_wdt_set_stage_timeout':
kempld_wdt.c:(.text.kempld_wdt_set_stage_timeout+0x130): undefined reference to `__aeabi_uldivmod'

This adds a call to ACCESS_ONCE() to force it to not consider
it to be constant, and leaves the more efficient normal case
in place for modern compilers, using an #ifdef to annotate
why we do this hack.

Signed-off-by: Arnd Bergmann 
Signed-off-by: Guenter Roeck 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

watchdog: bcm281xx: Fix use of uninitialized spinlock.

2017-07-05T12:40:28+00:00

commit fedf266f9955d9a019643cde199a2fd9a0259f6f upstream.

The bcm_kona_wdt_set_resolution_reg() call takes the spinlock, so
initialize it earlier.  Fixes a warning at boot with lock debugging
enabled.

Fixes: 6adb730dc208 ("watchdog: bcm281xx: Watchdog Driver")
Signed-off-by: Eric Anholt 
Reviewed-by: Florian Fainelli 
Reviewed-by: Guenter Roeck 
Signed-off-by: Guenter Roeck 
Signed-off-by: Wim Van Sebroeck 
Signed-off-by: Greg Kroah-Hartman

watchdog: pcwd_usb: fix NULL-deref at probe

2017-05-25T13:44:30+00:00

commit 46c319b848268dab3f0e7c4a5b6e9146d3bca8a4 upstream.

Make sure to check the number of endpoints to avoid dereferencing a
NULL-pointer should a malicious device lack endpoints.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Johan Hovold 
Reviewed-by: Guenter Roeck 
Signed-off-by: Guenter Roeck 
Signed-off-by: Wim Van Sebroeck 
Signed-off-by: Greg Kroah-Hartman

watchdog: s3c2410: Fix infinite interrupt in soft mode

2017-04-12T10:41:22+00:00

[ Upstream commit 0b445549ea6f91ffea78a976fe89b932db6e077a ]

In soft (no-reboot) mode, the driver self-pings watchdog upon expiration
of an interrupt.  However the interrupt itself was not cleared thus on
first hit, the system enters infinite interrupt handling loop.

On Odroid U3 (Exynos4412), when booted with s3c2410_wdt.soft_noboot=1
argument the console is flooded:
	# killall -9 watchdog
	[   60.523760] s3c2410-wdt 10060000.watchdog: watchdog timer expired (irq)
	[   60.536744] s3c2410-wdt 10060000.watchdog: watchdog timer expired (irq)

Fix this by writing something to the WTCLRINT register to clear the
interrupt.  The register WTCLRINT however appeared in S3C6410 so a new
watchdog quirk and flavor are needed.

Signed-off-by: Krzysztof Kozlowski 
Reviewed-by: Guenter Roeck 
Signed-off-by: Guenter Roeck 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

watchdog: qcom: fix kernel panic due to external abort on non-linefetch

2017-01-06T09:40:16+00:00

commit f06f35c66fdbd5ac38901a3305ce763a0cd59375 upstream.

This patch fixes a off-by-one in the "watchdog: qcom: add option for
standalone watchdog not in timer block" patch that causes the
following panic on boot:

> Unhandled fault: external abort on non-linefetch (0x1008) at 0xc8874002
> pgd = c0204000
> [c8874002] *pgd=87806811, *pte=0b017653, *ppte=0b017453
> Internal error: : 1008 [#1] SMP ARM
> CPU: 2 PID: 1 Comm: swapper/0 Not tainted 4.8.6 #0
> Hardware name: Generic DT based system
> PC is at 0xc02222f4
> LR is at 0x1
> pc : []    lr : [<00000001>]    psr: 00000113
> sp : c782fc98  ip : 00000003  fp : 00000000
> r10: 00000004  r9 : c782e000  r8 : c04ab98c
> r7 : 00000001  r6 : c8874002  r5 : c782fe00  r4 : 00000002
> r3 : 00000000  r2 : c782fe00  r1 : 00100000  r0 : c8874002
> Flags: nzcv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
> Control: 10c5387d  Table: 8020406a  DAC: 00000051
> Process swapper/0 (pid: 1, stack limit = 0xc782e210)
> Stack: (0xc782fc98 to 0xc7830000)
> [...]

The WDT_STS (status) needs to be translated via wdt_addr as well.

fixes: f0d9d0f4b44a ("watchdog: qcom: add option for standalone watchdog not in timer block")
Signed-off-by: Christian Lamparter 
Reviewed-by: Guenter Roeck 
Signed-off-by: Guenter Roeck 
Signed-off-by: Greg Kroah-Hartman

watchdog: mei_wdt: request stop on reboot to prevent false positive event

2017-01-06T09:40:16+00:00

commit 9eff1140a82db8c5520f76e51c21827b4af670b3 upstream.

Systemd on reboot enables shutdown watchdog that leaves the watchdog
device open to ensure that even if power down process get stuck the
platform reboots nonetheless.
The iamt_wdt is an alarm-only watchdog and can't reboot system, but the
FW will generate an alarm event reboot was completed in time, as the
watchdog is not automatically disabled during power cycle.
So we should request stop watchdog on reboot to eliminate wrong alarm
from the FW.

Signed-off-by: Alexander Usyskin 
Signed-off-by: Tomas Winkler 
Reviewed-by: Guenter Roeck 
Signed-off-by: Guenter Roeck 
Signed-off-by: Greg Kroah-Hartman

Merge branches 'acpi-sleep-fixes' and 'acpi-wdat-fixes'

2016-11-25T21:24:07+00:00

* acpi-sleep-fixes:
  Revert "ACPI: Execute _PTS before system reboot"

* acpi-wdat-fixes:
  watchdog: wdat_wdt: Select WATCHDOG_CORE

watchdog: wdat_wdt: Select WATCHDOG_CORE

2016-11-21T21:49:17+00:00

The WDAT watchdog driver uses functionality provided by the watchdog timer
core but it did not select it explicitly. This results following linker
error when only WDAT_WDT is enabled in Kconfig:

  drivers/built-in.o: In function `wdat_wdt_probe':
  drivers/watchdog/wdat_wdt.c:444: undefined reference to `devm_watchdog_register_device'

Fix this by explicitly selecting WATCHDOG_CORE when WDAT watchdog driver is
enabled.

Fixes: 058dfc767008 (ACPI / watchdog: Add support for WDAT hardware watchdog)
Reported-by: Vegard Nossum 
Signed-off-by: Mika Westerberg 
Reviewed-by: Guenter Roeck 
Signed-off-by: Rafael J. Wysocki

Merge branches 'acpi-wdat' and 'acpi-cppc'

2016-10-21T20:24:23+00:00

* acpi-wdat:
  watchdog: wdat_wdt: Ping the watchdog on resume

* acpi-cppc:
  mailbox: PCC: Fix return value of pcc_mbox_request_channel()

watchdog: wdat_wdt: Ping the watchdog on resume

2016-10-20T23:28:33+00:00

It turns out we need to ping the watchdog hardware on resume when we
re-program it. Otherwise this causes inadvertent reset to trigger
right after the resume is complete.

Fixes: 058dfc767008 (ACPI / watchdog: Add support for WDAT hardware watchdog)
Signed-off-by: Mika Westerberg 
Acked-by: Guenter Roeck 
Signed-off-by: Rafael J. Wysocki