Hi!
/* XXX this driverfs stuff should probably go elsewhere later -john
*/
static struct sys_device device_i8253 = {
- .name = "rtc",
.id = 0,
- .dev = {
- .name = "i8253 Real Time Clock",
- },
+ .cls = &rtc_sysclass,
};
...but i8253 is *not* real time clock. Its clock since
bootup. Realtime clock is near battery-backed CMOS RAM, its driver is
linux/drivers/char/rtc.c...
Pavel
--
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
At 4:23pm +0200 6/16/03, Pavel Machek wrote:
> /* XXX this driverfs stuff should probably go elsewhere later -john
>*/
> static struct sys_device device_i8253 = {
>- .name = "rtc",
> .id = 0,
>- .dev = {
>- .name = "i8253 Real Time Clock",
>- },
>+ .cls = &rtc_sysclass,
> };
>
>...but i8253 is *not* real time clock. Its clock since
>bootup. Realtime clock is near battery-backed CMOS RAM, its driver is
>linux/drivers/char/rtc.c...
FWIW, the legacy Intel databooks call it a "Programmable Interval
Timer" (PIT). Likewise the 8254, a superset of the 8253. I don't know
that Intel ever used the 'i' prefix on these parts.
--
/Jonathan Lundell.
On Mon, 16 Jun 2003, Pavel Machek wrote:
> Hi!
>
> /* XXX this driverfs stuff should probably go elsewhere later -john
> */
> static struct sys_device device_i8253 = {
> - .name = "rtc",
> .id = 0,
> - .dev = {
> - .name = "i8253 Real Time Clock",
> - },
> + .cls = &rtc_sysclass,
> };
>
> ...but i8253 is *not* real time clock.
D'oh. Noted and changed to 'pit'.
-pat