Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

More coding style clean-up's.
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This brings the watchdog drivers into line with coding style.
This patch takes cares of the indentation as described in chapter 1.
Main changes:
* Re-structure the ioctl switch call for all drivers as follows:
	switch (cmd) {
	case WDIOC_GETSUPPORT:
	case WDIOC_GETSTATUS:
	case WDIOC_GETBOOTSTATUS:
	case WDIOC_GETTEMP:
	case WDIOC_SETOPTIONS:
	case WDIOC_KEEPALIVE:
	case WDIOC_SETTIMEOUT:
	case WDIOC_GETTIMEOUT:
	case WDIOC_GETTIMELEFT:
	default:
	}

This to make the migration from the drivers to the uniform watchdog
device driver easier in the future.
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Review and switch to unlocked_ioctl
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Pádraig Brady requested the possibility of not disabling the watchdog
at module load time or kernel boot time if it had been previously enabled
in the bios. It may help rebooting the machine if it freezes before the
userland daemon kicks in.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Cc: Pádraig Brady <P@draigBrady.com>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Some non-exported functions always returned 0. Mark them void instead.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This reverts commit 3ff6eb4a.

the !found check in the for loop allready made sure that only one
device was found.
Signed-Off-By: NPádraig Brady <P@draigBrady.com>
Signed-Off-By: NWim Van Sebroeck <wim@iguana.be>

If no address is given for the W83697HF/HG watchdog IO port, stop looping
through possible locations when a watchdog device has been found.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Some watchdog drivers initialize global spinlocks in module's init function
which is tolerable, but some do it in PCI probe function.  So, switch to
static initialization to fix theoretical bugs and, more importantly, stop
giving people bad examples.
Signed-off-by: NAlexey Dobriyan <adobriyan@sw.ru>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>

* Convert files to UTF-8.

  * Also correct some people's names
    (one example is Eißfeldt, which was found in a source file.
    Given that the author used an ß at all in a source file
    indicates that the real name has in fact a 'ß' and not an 'ss',
    which is commonly used as a substitute for 'ß' when limited to
    7bit.)

  * Correct town names (Goettingen -> Göttingen)

  * Update Eberhard Mönkeberg's address (http://lkml.org/lkml/2007/1/8/313)
Signed-off-by: NJan Engelhardt <jengelh@gmx.de>
Signed-off-by: NAdrian Bunk <bunk@kernel.org>

move watchdog tree from drivers/char/watchdog to drivers/watchdog.
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Many struct file_operations in the kernel can be "const".  Marking them const
moves these to the .rodata section, which avoids false sharing with potential
dirty data.  In addition it'll catch accidental writes at compile time to
these shared resources.
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

change default=CONFIG_WATCHDOG_NOWAYOUT in the module parameter
for nowayout by it's real value (0 or 1) by using:
__MODULE_STRING(WATCHDOG_NOWAYOUT)
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Change the autodetect code so that it is more generic.
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 16 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Add copyright notice for Samuel Tardieu also.

This is the last patch in this series. 

The original description for Samuel's driver was:
driver for the Winbond W83697HF/W83697HG watchdog timer

The Winbond SuperIO W83697HF/HG includes a watchdog that can count from
1 to 255 seconds (or minutes). This drivers allows the seconds mode to
be used. It exposes a standard /dev/watchdog interface. This chip is
currently being used on some motherboards designed by VIA.

By default, the module looks for a chip at I/O port 0x2e. The chip can
be configured to be at 0x4e on some motherboards, the address can be
chosen using the wdt_io module parameter. Using 0 will try to autodetect
the address.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 15 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Clean-up initialization code - part 2: 
   * the line reading "set second mode & disable keyboard ..."
     is plain wrong, the register being manipulated (CRF4) is
     the counter itself, not the control byte (CRF3) -- looks
     like it has been copied from another driver.
   * I think garbage is being written in CRF3 (the control word)
     as the timeout value is being stored in this register (such
     as 60 for 60 seconds).
   * We only want to set pin 119 to WDTO# mode and leave the rest
     of CR29 like it is.
   * Set count mode to seconds and not minutes.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 14 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Clean-up initialization code (part 1: remove
   w83697hf_select_wd_register() and
   w83697hf_unselect_wd_register() functions).
 - Make sure that the watchdog device is stopped
   as soon as we found it.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 13 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Remove wdt_ctrl (it has been replaced with the
   w83697hf_write_timeout() function) and redo/clean-up
   the start/stop/ping code.
 - Make sure that the watchdog is enabled or disabled
   When starting or stoping the device (with a call
   to w83697hf_set_reg(0x30, ?); ).
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 12 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Add w83697hf_write_timeout() to set the
   watchdog's timeout value.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 11 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Add w83697hf_select_wdt() and w83697hf_deselect_wdt()
   so that the start/stop/ping code can directly talk to
   the watchdog.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 10 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - check whether the device is really present
   (we *can* probe for the device now).
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 9 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - add w83697hf_get_reg() and w83697hf_set_reg()
   functions.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 8 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - add w83697hf_lock function to leave the
   chipsets extended function mode.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 7 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - add w83697hf_unlock function to enter the
   chipsets extended function mode.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 6 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - The driver works for both the w83697hf
   and the w83697hg chipset's.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 5 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Rename the Extended Function Registers to the names
   used in the data-sheet.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 4 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - limits the watchdog timeout to 1-63 while this
   device accepts 1-255.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 3 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - Fix identation.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 2 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - wdt_io is 2 bytes long. We should do a
   request_region for 2 bytes instead of 1.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

This is patch 1 in the series of patches that converts
Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's
w83697hf/hg watchdog driver.

This patch contains following changes:
 - the note concerning tyan motherboards has been copied from
   another driver, This doesn't apply here.
 - the comments concerning CRF6 are wrong as CRF3 is manipulated
   and CRF6 is never read nor written.
 - the comments concerning CRF5 are wrong as CRF4 is manipulated
   and CRF5 is never read nor written.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Return ENOTTY instead of ENOIOCTLCMD in user-visible ioctl() results

The watchdog drivers used to return ENOIOCTLCMD for bad ioctl() commands.
ENOIOCTLCMD should not be visible by the user, so use ENOTTY instead.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>
Acked-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>

Add io spinlocks to prevent possible race
conditions between start and stop operations
that are issued from different child processes
where the master process opened /dev/watchdog.
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

New watchdog driver for the Winbond W83697HF chipset.
Signed-off-by: NMarcus Junker <junker@anduras.de>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Return ENOTTY instead of ENOIOCTLCMD in user-visible ioctl() results

The watchdog drivers used to return ENOIOCTLCMD for bad ioctl() commands.
ENOIOCTLCMD should not be visible by the user, so use ENOTTY instead.
Signed-off-by: NSamuel Tardieu <sam@rfc1149.net>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>
Acked-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>

Mark the static struct file_operations in drivers/char as const.  Making
them const prevents accidental bugs, and moves them to the .rodata section
so that they no longer do any false sharing; in addition with the proper
debug option they are then protected against corruption..

[akpm@osdl.org: build fix]
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The most trivial typo fix in the world.
Signed-off-by: NPozsar Balazs <pozsy@uhulinux.hu>
Signed-off-by: NPdraig Brady <P@draigBrady.com>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Attached is a small update to the w83627hf watchdog driver
to initialise appropriately if it was already initialised
in the BIOS. On tyan motherboards for e.g. you can init
the watchdog to 4 mins, then when the driver is loaded it
sets the watchdog to "seconds" mode, and then machine will
reboot within 4 seconds. So this patch resets the timeout
to the configured value if the watchdog is already running.
Signed-off-by: NP@draig Brady <P@draigBrady.com>
Signed-off-by: NWim Van Sebroeck <wim@iguana.be>

Attached patch removes #ifdef CONFIG_WATCHDOG_NOWAYOUT mess duplicated in
almost every watchdog driver and replaces it with common define in
linux/watchdog.h.
Signed-off-by: NAndrey Panin <pazke@donpac.ru>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!