Re: [PATCH][RFC] Proposal for a new watchdog interface using sysfs

[Rusty Lynch]

On Wed, 2003-02-12 at 20:27, Daniel Pittman wrote:
> On 12 Feb 2003, Rusty Lynch wrote:
> > The following is a proposal for a new sysfs based watchdog interface
> > to be used as a replacement for the current char device w/ ioctl api
> > as described in Documentation/watchdog-api.txt.
> 
> [...]
> 
> > Where each these files to the following ==>
> > 
> > start (RO)
> >   - show: starts watchdog count
> 
> This would be much better as a store -- that way 'cat /.../watchdog0/*'
> will not activate the watchdog. A more deliberate action is safer for
> forgetful admins, such as me.
> 

Sounds logical.  My reasoning was more of a coin toss.

> [...]
> 
> > status (RO)
> >   - show: prints the current status value
> > 
> > bootstatus (RO)
> >   - show: same as 'status', but valid for just after the last reboot.
> 
> [...]
> 
> > enable (RW)
> >   - show: prints 0 or 1 to indicate if the wdt is enabled
> >   - store: expects 0 or 1 to disable or enable the wdt
> 
> Isn't this the same information as the 'status' and 'start' members?
> 
>       Daniel

Now that think about it, enable really is the same as start/stop, but
the status is still something different.  I didn't really talk about it,
but my intent was to provide the information currently available from
the WDIOC_GETSTATUS ioctl, which is a value composed of the following
flags:

#define	WDIOF_OVERHEAT		0x0001	/* Reset due to CPU overheat */
#define	WDIOF_FANFAULT		0x0002	/* Fan failed */
#define	WDIOF_EXTERN1		0x0004	/* External relay 1 */
#define	WDIOF_EXTERN2		0x0008	/* External relay 2 */
#define	WDIOF_POWERUNDER	0x0010	/* Power bad/power fault */
#define	WDIOF_CARDRESET		0x0020	/* Card previously reset the CPU */
#define WDIOF_POWEROVER		0x0040	/* Power over voltage */
#define WDIOF_SETTIMEOUT	0x0080  /* Set timeout (in seconds) */
#define WDIOF_MAGICCLOSE	0x0100	/* Supports magic close char */
#define	WDIOF_KEEPALIVEPING	0x8000	/* Keep alive ping reply */

I debated with myself if each of these flags should be broke out into
their own file, and I'm still not sure if it is better to keep the
status as a single file.

> 
> -- 
> A cathedral, a wave of a storm, a dancer's leap,
> never turn out to be as high as we had hoped.
>         -- Marcel Proust
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [e-mail]
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/