TSC instead of ACPI: powerd doesn't work anymore
(to be expected?)
Scott Long
scottl at samsco.org
Sun Oct 30 06:30:41 PST 2005
David O'Brien wrote:
> On Sun, Oct 30, 2005 at 02:39:40PM +0800, David Xu wrote:
>
>>David O'Brien wrote:
>>
>>>On Fri, Oct 28, 2005 at 03:34:57PM +0200, Alexander Leidinger wrote:
>>>
>>>>I don't have the message at hand. I just had time to write the mail, but I
>>>>don't have my laptop with me to reproduce the message. But it's easy to
>>>>reproduce, just take a PC which is able to make use of powerd and switch
>>>>to
>>>>using TSC as the timecounter.
>>>
>>>What is the motivation to use the TSC as a timecounter?
>>
>>TSC is faster than any others, on many systems, so-called ACPI-fast
>>timer is really a slow chip,
>
>
> Correct, but why is it felt the latency of the ACPI timer is an issue?
> Of course we all want things to as fast as possible, but is that just an
> abstract desire, or a real issue was run into?
>
ACPI-fast requires an ioport read which takes about 1us (according to
Google). Do that 1000 times a second and you have each CPU spending
1% of its time doing nothing but reading the clock. Yikes.
Scott
More information about the freebsd-current
mailing list