Re: gcc done, eglibc problem

[Finn Thain <fthain@telegraphics.com.au>]

On Fri, 22 Oct 2010, Thorsten Glaser wrote:

> Finn Thain dixit:
> 
> >I'm not aware of any differences between native and cross gcc that affect 
> 
> There are code generation differences, disabled optimisation passes, etc.

I'm not aware of code generation differences. Can you be more specific?
 
> >In fact, since embedded Linux devices vastly outnumber all other 
> >production Linux systems, I'd be pretty confident that cross-compiling 
> >Linux kernels is the rule and not the exception.
> 
> That’s like saying because MS Outlook is most widely used it’s good ☺

No, it is like saying that MS Outlook works for most people (or they 
simply wouldn't pay for it). Linux cross-compilers are commercial 
products.

You are perhaps too young to remember EGCS, but I do. If it weren't for 
Cygnus and later CodeSourcery, the GNU toolchain would certainly not be 
what it is today.

> I don’t dispute cross compiling being practical, but I agree with the 
> OpenBSD people (wow!) that (only) cross compiling kills architectures. 
> Similar, although to a lesser extent , can be said about emulating.

I have no problem with emulating. I have no problem with cross-compiling. 
All native compilers, cross compilers, emulators and silicon have bugs.

As for the OpenBSD attitude, probably that comes from never working close 
enough to the silicon to realise that all architectures started life from 
emulators and cross-compilers. Probably they are also unaware of Plan 9.

> I think I remember you protesting that too :)

You are twisting my words. I said "I am not content with yet another 
virtual linux."

This does not argue against emulation, it is argues against ONLY emulating 
(the context was the lack of a working physical buildd).

> (but Ingo already offered some “bored” real hardware to run stuff on – 
> that we will!)

Cool.

Finn

> 
> bye,
> //mirabilos
>