[LAU] More recent kernel config options

[Brent Busby]

On Sun, 26 Jul 2009, Ken Restivo wrote:

>> I've also run into a handful of config options that either help or hurt
>> realtime performance, depending on the specific hardware:
>>
>> NO_HZ (seems to work properly on most hardware these days)
>
> Is that advised? I was told to use HZ_1000 in order to get MIDI to 
> work correctly. Should I be using NO_HZ instead? I've got several 
> machines to configure: a 32-bit Atom EEE, a 64-bit Intel Micro-ITX, 
> and a 64-bit Asus Intel Core2Duo

Oddly (and I don't really understand this), you can have *both*.  I've 
seen recommendations online from people who claim to have gotten good 
results with Jack even on laptops by having NO_HZ and HZ_1000 selected. 
I don't really know what the effect of this is.  Does this mean you 
generate no ticks except when you need them, but when you do, the jiffy 
rate is 1000Hz?  This naively seems to me like a better idea than just 
running a clock all the time, but I don't know how that pans out for 
recording and MIDI.

It seems like there ought to be more pooling of resources regarding 
complete kernel configs that work.  Everyone always talks about the 
patches themselves...but anyone who can do 'patch -p1' against a tree of 
source code can get that part right.

-- 
+ Brent A. Busby	 + "We've all heard that a million monkeys
+ UNIX Systems Admin	 +  banging on a million typewriters will
+ University of Chicago	 +  eventually reproduce the entire works of
+ Physical Sciences Div. +  Shakespeare.  Now, thanks to the Internet,
+ James Franck Institute +  we know this is not true." -Robert Wilensky