Thorsten Glaser wrote:
Since gcc 4.0 will be out of the door not in less than two months from now, and given that gcc 4.1, which ought to solve at least the basic problems with tree-ssa etc. is not even in sight yet, I propose that there will be a gcc-3.5 release with support for Pascal, based upon gcc-3.4, current gpc and Waldek's patches.
This would be great since we already had the new directory layout (what I would greatly like to see for Ada and Pascal is that their RTS moves from gcc/p/ to libgpc/ ASAP), and the gcc-3.4/3.5 specific diffs integrated into the core, while the backend patches for gcc 3.3 and older are still available (I don't think Frank is going to throw them away after the integration),
After the integration as I plan it (i.e., after the next major GPC release), I probably will "throw them away". Of course, they'll be still available in older archives, but if they're not maintained anymore, they'll probably become useless sooner or later.
Nobody said to stop integration work. It's about two things. To get stuff in at all and in the correct directories, and to adapt the code (esp. for gcc4). Start with the easier.
I don't quite agree. There's been much talk of those directories, but I don't see it as a big issue. It will involve moving some files (trivial) and adjusting the RTS build (which is partly broken and outdated anyway). All in all, it can probably done easily in a day by someone who's familiar with the issues (I'm only partly, currently).
Since early integration causes some "friction", as I tried to explain, I'd favour doing the hard stuff first (i.e., the necessary changes for newer backends which have to be done sometime anyway), and do the last (easy) step when we can be confident that we won't have to go back.
Marcel Cox dixit:
- I think doing major GPC integration work now based on GCC 3.x is
just a waste of resources
There is not much to do. I integrated gpc into a BSD in-tree gcc once, and besides having to rewrite some stuff for BSD make and applying the diffs correctly, there was next to no work involved.
Yes, that's the easy part as Waldek noted. Adjusting to the backend and infrastructural changes (such as TreeSSA) is what takes most of the work.
Frank