Hi,
I CC'ed the list so outside people see interesting discussion going along.
On Saturday 11 November 2006 19:12, Luca wrote:
> The rational is nothing really special, simply to compile first a
> cross-compiler which support only c, non shared and non threaded to
> compile the cross-tools and then recompile it (the last step of the
> cross-toolchain) with support for c and other languages, shared and with
> threads so to make the cross-compiled binutils, glibc, etc... sure to
> get linked and compiled only with the cross-compiled compiler and not
> with the host (not cross-compiled) compiler and that nothings gets
> possibly linked against host headers and libraries (like glibc which
> links to the host system libgd, libpng, libz, etc... for the
> memusagestat program unless a sed or "--without-gd" option is passed).
In my opinion our setup is pretty fine right now. we bootstrap a fresh and new
cross compiler (0-gcc) even in the native build case. This is used to build
everything for the target system (1-*).
If we do not properly disable glibc's libgd usage your setup would change
nothing for that matter - does it ?
I think we would need to properly disable gd usage in glibc in stage 1 in
either case and should do so now and another gcc intermediate step is not
necessary as 0-gcc decouples the host and the new sandbox already - unless I
overlook some detail right now.
Yours,
--
René Rebe - ExactCODE - Berlin (Europe / Germany)
http://exactcode.de | http://t2-project.org | http://rene.rebe.name
+49 (0)30 / 255 897 45
-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
lists@xxxxxxxxxxxxxx with a subject of: unsubscribe t2
|
