Re: crosstool build matrix ambiguity

["Robert P. J. Day" <rpjday at mindspring dot com>]

On Mon, 25 Apr 2005, Allan Clark wrote:

> How much different should the resulting toolchain be for different
> host compilers?

possibly not at all, *provided* you can build with different host
compilers.  (my memory may be failing me, but i recall being the one
who kind of whined and moaned about sanitized headers, because it
turns out i needed them to solve a build problem i was having.  could
it have been using the experimental gcc-4.0.0 as the host compiler?
damn, it seems so long ago ...)

anyway, it may be that the end result is the same, but getting the
build to work might require different patches.  so, no, it's not clear
whether you'd want to distinguish based on the host compiler, but i
can't say for sure.

> I would agree that "pxa271-2.4.19" would probably be the toolchain I
> am building.  It makes sense to give a logical name.

what about a simple patch so that, if TOOLCOMBO is already defined in
the environment, that's used instead of building the current string?
and if that's done, a better name for that variable would be, perhaps,
CHAINID or something, since it would just contain some unique string
that only has to be meaningful to the builder.

just a thought.

rday

p.s.  has anyone submitted a patch to allow a selection of host
compiler?  so i can live dangerously and pick gcc4 if i want?

------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com