In message <[EMAIL PROTECTED]>
[EMAIL PROTECTED] wrote:
> Author: alex
> Date: 2007-12-15 21:26:38 +0000 (Sat, 15 Dec 2007)
> New Revision: 2967
>
> Modified:
> trunk/gcc4/porting-scripts/ro-config
> Log:
> Disable ,e1f changes to configure scripts, as the compiler doesn't add
> the extension.
I have to revert this. The intention is good, but I think you're
focussing too much on the behaviour of an individual package over the
whole lot. This breaks a number of assumptions that we make - for
example, in the the orbit build, and I suspect, elsewhere.
We know that the current case isn't ideal, but having it the way above
in the end means more work - more patches to account for the missing
extension, and more work in the packaging stages. Adding the extension
lets us try and differentiate between native and cross binaries -
assuming that the extension variable is carried all the way through the
build - which is true in general, although sadly not all the time.
I'll work with John to try and get the ,e1f stuff added to the current
GCC. Unfortunately, this doesn't build for me at present. A sadly
ongoing theme is the fragility of the build for that.
--
Peter Naulls - [EMAIL PROTECTED] | http://www.chocky.org/
----------------------------------------------------------------------------
RISC OS Community Wiki - add your own content | http://www.riscos.info/
_______________________________________________
GCCSDK mailing list [email protected]
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
