Re: CONFIG_SMP patch available for 2.1.54

Pavel Machek (pavel@Elf.mj.gts.cz)
Wed, 10 Sep 1997 22:50:46 +0200

> From: alan@lxorguk.ukuu.org.uk (Alan Cox)
> Date: Tue, 9 Sep 1997 22:10:07 +0100 (BST)
>
> Using SMP kernels on an Uniprocessor throws away performance - on some
> architectures quite a lot.
>
> I had always thought that it wasn't a bad idea for kernel hackers with
> only uniprocessors to occasionally trying compiling one of their kernels
> SMP, just to test for bugs and make sure their drivers, et. al all
> compiled in SMP operation. Is this really worth the effort, or should I
> not bother?

Some races, etc. will probably still show only in real-SMP cases.

BTW it would be _very_ nice to print warning somewhere that this is
SMP kernel - I just build SMP kernel on my 486, and during boot it
said _nothing_ about being SMP.

What about changing this
Linux version 2.1.55 (root@Elf) (gcc version 2.7.2.1) #1 Wed Sep 10
to
Linux version 2.1.55/SMP (root@Elf) (gcc version 2.7.2.1) #1 Wed Sep
?

I think that it is worth it. (I do not submit patch. My kernel is too
modified, and linus would ignore it from me, anyway.)
Pavel

-- 
I'm really pavel@atrey.karlin.mff.cuni.cz. 	   Pavel
Look at http://atrey.karlin.mff.cuni.cz/~pavel/ ;-).