"New" Real Time Clock vs "Old" RTC in 2.6.24 (Hardy 8.04)

[Colin Ian King]

Tim,

Good point. How are deprecated features handled normally in a LTS disto?
Surely we must have seen a lot of /proc features being deprecated and
their replacements appear in /sys before now.

I suppose the /dev/rtc --> /dev/rtc0 is not too problematic as one
could, for example, symlink around the problem as an ugly hack.
The /dev/rtc0 interface should be backwardly compatible.

As for /proc/acpi/alarm, this is a more complex issue - I am unsure if
we can retain it for 5 years if the new RTC mechanism becomes the
de-facto way of doing things.

Colin.


On Fri, 2008-02-15 at 08:41 -0700, Tim Gardner wrote:
<snip snip> 8< --------------
stuff delete
<snip snip> 8< -------------

> Since there does not seem to be a backwards compatible solution, my
> opinion is that this is a Hardy+1 feature. Especially since the Hardy
> feature freeze started today (or yesterday according to
> https://wiki.ubuntu.com/HardyReleaseSchedule?highlight=%28hardyrelease%29 ).
> 
> On the other hand, we also have to think about supporting this release
> for the next 5 years as it is targeted as an LTS. Is there no way to
> preserve /dev/rtc and /proc/acpi/alarm ?
> 
> rtg