Pardon me if I am a bit un-informed or something

Paul King pking123-rieW9WUcm8FFJ04o6PK0Fg at public.gmane.org
Tue Jan 10 23:21:13 UTC 2006


On Tue, 2006-01-10 at 09:44 -0500, Lennart Sorensen wrote:
> On Tue, Jan 10, 2006 at 12:05:21AM -0500, Paul King wrote:
> > I mean *really*, what masochist would choose "randconfig" as the config
> > target? Can someone enlighten me on what purpose this option serves?
> > Same goes for "allyesconfig". I can see a use for the rest of these.
> 
> randconfig is great for testing builds.  Make a random config, see if it
> even builds.  Catches syntax errors and other build problems in less
> used areas of the kernel, and catches some config dependancy issues that
> cause build failures.

I am still not sure why I would not just choose my options and test my
own build. Wouldn't the error messages generated thereby give me more
useful info? It is entirely possible that randconfig would mostly select
options that I wouldn't wish to use anyway, and would generate error
messages on files I would never compile in to begin with. You also don't
know if a random config will really auto-select the less-used options if
that is what you are checking.

It sounds to me that the only useful option left would be for kernel
programmers to test out their new source code to see if anything blows
up on them. Or for ordinary Linux users to test kernel versions in the
"odd" series (2.1, 2.3, 2.5, etc) to see if they are safe to use. Since
mine is 2.6, there would be less of a need for that.

> 
> The all* options are probably also mostly for testing.

Thanks to everyone for your replies.

Paul


--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml





More information about the Legacy mailing list