Re: 2.6.13-rc3 Battery times at 100/250/1000 Hz = Zero difference

From: Bill Davidsen
Date: Tue Jul 26 2005 - 09:49:16 EST


Brown, Len wrote:


Digging up this patch from last month regarding C2
on a AMD K7 implies
that the whole blame can be put on kernel acpi:

http://marc.theaimsgroup.com/?l=linux-kernel&m=111933745131301&w=2


The current Linus tree includes generic ACPI support
for deep C-states on SMP machines. (deep means higher than C1)

I don't have any problem with people having platform specific
modules to handle platform specific features. However, if
the system really intends to support SMP ACPI C-states deeper
than C1 and the generic ACPI code doesn't support it,
then it is either a Linux/ACPI bug or a BIOS bug -- file away:-)

I.e. The whole concept of ACPI is that you shoulud _not_ need
a platform specific driver to accomplish this.

Is anyone but me interested in low power states for servers? I have several groups of servers which are lightly utilized for at least 12 hours every day and on weekends. I currently use IDE drives so I can spin them down when idle, do logging either to a single drive or over the network whichever makes the most sense, and any drop in power use saves double, since I pay for the server power and the A/C power as well.

I haven't seen much discussion of this, but in many cases it would result in a saving, perhaps fairly large. Some environmental benefit as well, of course.

--
-bill davidsen (davidsen@xxxxxxx)
"The secret to procrastination is to put things off until the
last possible moment - but no longer" -me
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/