I486_CPU and I586_CPU removed from GENERIC kernel [was Re: svn commit: r205307 - head/sys/i386/conf]

Bernd Walter ticso at cicely7.cicely.de
Sat Mar 27 11:08:46 UTC 2010


On Fri, Mar 19, 2010 at 05:28:06PM +0100, Ivan Voras wrote:
> On 19 March 2010 17:22, Valentin Nechayev <netch at netch.kiev.ua> wrote:
> >  Fri, Mar 19, 2010 at 17:13:00, ivoras wrote about "Re: I486_CPU and I586_CPU removed from GENERIC kernel [was Re: svn commit: r205307 - head/sys/i386/conf]":
> >
> >> SSE in the userland you mean? Regardless, I don't think there is now
> >> reason for compiling everything as for i386. E.g. why not add at least
> >> -mtune=generic or even also -march=i686 to default gcc options?
> >>
> >> http://gcc.gnu.org/onlinedocs/gcc/i386-and-x86_002d64-Options.html
> >
> > Having userland compiled with i686 will give the same effect as i686-only
> > kernel: it won't boot on machines which doesn't conform to. If it is
> > supposed to boot on i486 and higher, no more than -march=i486 can be used.
> 
> Yes, this is how I read the change - the move from "i386" to "i686". I
> apologize if I got it wrong :)
> 
> As it was pointed out earlier - small systems users and designers
> probably have special install procedures because of the nature of the
> business.

The assumption is wrong to my knowledge.
I think it is a fact that an old 586 usually can't boot our CDs.
It is also true that many embedded boards don't have CD boot support,
like Soekris systems.
But there are also some miniATX systems build on 586 class CPUs, which
have recent BIOS code and can boot from CD.
It is also quite common for unexeperienced users to binary install the
HDD in a modern system.
This is not business, because many home users run such small systems as
storage servers, gateways, dhcp, ... - there are many unexperienced
users with such systems.

-- 
B.Walter <bernd at bwct.de> http://www.bwct.de
Modbus/TCP Ethernet I/O Baugruppen, ARM basierte FreeBSD Rechner uvm.


More information about the svn-src-all mailing list