On Tue, Nov 06, 2012 at 03:13:10PM +0100, Lluís Batlle i Rossell wrote: > On Tue, Nov 06, 2012 at 12:43:52PM +0000, Brian Drummond wrote: > > Remaining problems are possibly due to removing the --disable-bootstrap > > configure option, which appears to affect which compiler (C or C++) is > > used for parts of the build. > > > > As GCC is moving gradually towards more C++ use, this is worth resolving > > before it becomes another big obstacle. Stay tuned for more news... > > > > Thanks to Lluis for all the work he is putting into this! > > Hello, > > I'm still working on it. Although I've solved some of the c++ problems, > I see more troubles in the 'ghdl' driver. The 'ghdl' driver build depends on > 'ggc-none.o' among others, and it is not set as a dependency for the 'ghdl' > target. And in my case, it isn't built before 'ghdl'. > > There are clearly broken Makefile rules overall... I don't know how we could > fix all this. > > There the build command for 'ghdl' is full of '../..' kind of references, > which > don't work very well in the recursive makefiles.
Ah I think I understood how to fix the Makefiles, finally. I'll test that in the next hours. On the other hand, I know that Brian has seen some documentation that recommends not building a bootstrapped compiler, but in Nixpkgs we have been providing ghdl 0.29 for gcc 4.3 always with a bootstrapped C compiler (not a bootstrapped gnat). _______________________________________________ Ghdl-discuss mailing list [email protected] https://mail.gna.org/listinfo/ghdl-discuss
