Subject: Re: CVS commit: src/sys/kern
To: John Hawkinson <jhawk@MIT.EDU>
From: Frank Kardel <email@example.com>
Date: 08/06/2006 09:13:34
John Hawkinson wrote:
>Christos Zoulas <firstname.lastname@example.org> wrote on Sun, 6 Aug 2006
>at 05:09:21 +0000 in <email@example.com>:
>>>I'm just afraid such not-so-meaningful digits might confuse
>>I agree with that assessment.
>You're kidding? Surely users will say "look, it's a big number," and
>go on with their lives. If the presence of big numbers confuses them,
>they probably have serious other problems we can't fix.
>I agree with Perry's assessment that it is nice to have the raw data.
>I'd much rather the OS not make decisions for me about what is
>significant and what is not.
AFAICS the raw data is still available in sysctl kern.timecounter.choice.
Given that debugging primarily starts after getting to single user or
further. Shouldn't that be enough?
Psychologically I agree that most users don't recognize the significance
of certain NTSC or whatever derived oscillator frequencies. For
debugging I do
use the precise numbers, though. So access to them should not go away