RE: PATCH: /dev/nvram cleanups.

Jones D (djones2@glam.ac.uk)
Thu, 29 Jul 1999 12:31:13 +0100


> >> Some RTCs implement only 64 bytes of NVRAM, with bytes 64-127
> >> mapping back to 0-63. That way, you could not only write the
> >> NVRAM, but make the RTC generate interrupts (which we use as
> >> timer interrupts on some Alphas), set the system time or cause
> >> alarms as soon as you'd have write access to the NVRAM. AFAIK
> >> there is no way to autodetect the size of the NVRAM.
> > OK, I've seen a 64-byte CMOS myself. But 64 is still better than
> > 50 :) (no, I don't have real need to access these bytes yet)
> I thought your arithmetic was better than that...
>
> Q> Bytes Description
> Q> ~~~~~ ~~~~~~~~~~~
> Q> 14 RTC registers
> Q> 50 CMOS RAM
> Q> ~~~~~
> Q> 64 Total

Yes, that's correct. The current nvram driver decodes up to 64,
using an offset of 14 for each access. At least that's what it does
when it builds the /proc/nvram interface. I'm not sure if the /dev/nvram
interface allows access to the RTC registers too.

regards,

d.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/