Alpha Timekeeping - Hardware Questions

Bruce Elliott (bde@nwlink.com)
Thu, 13 May 1999 03:07:16 -0700


On May 12, 2:01pm, Tom Holroyd wrote:
} Subject: 2.2.8 gettimeofday() still doesn't work on Alpha
} Just booted 2.2.8 on my PC164 and the timetest program (included below)
} produces negative deltas with high frequency.

[ ... ]

I believe that I have read that (some) SMP Alphas have independent CPU
clocks (i.e. the oscillators that drive the "cycle counters"). Is this
true? If so, any cycle-counter-based timekeeping will be difficult, as
it requires drift-correction (something like ntp) at timer interrupt
level.

If only some Alphas have separate CPU clocks, are these older models?

For recent and current Alphas, is it true that the various clocks (PIT,
RT, CYCLES) are separate, or are they (or a subset) derived (pll?) from
a common source?

I am sending this to several lists and individuals. I would appreciate
that any definitive answers or references either be sent to the linux-
kernel list, or to me for forwarding to the list.

-- 
B. D. Elliott   bde@nwlink.com   (Seattle)

- 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/