
On 06/14/2012 01:11 AM, Uwe F. Mayer wrote:
From: Ronny Pfannschmidt<Ronny.Pfannschmidt@gmx.de> I think we should encourage contributors to use tools that help the core developers/reviewers whenever possible.
Ronny: I certainly don't mind generating diffs in a format that works for the dev team. It would be nice if that format or desired tools were described somewhere easily found on pypy.org (maybe it's there, but I didn't see it). I can foresee that I may stay involved with pypy, so if I understand correctly, you'd prefer the output of "diff -cru" instead of "diff -cr"? --Uwe
basically use mercurial, i think we need a more detailed howto for setting contributors up for sending patches and sending pull requests (our branching is quite bit more complex than the common mercurial project)
On 06/13/2012 01:20 PM, Armin Rigo wrote:
I have checked in a slightly edited version of
http://www.tux.org/~mayer/cygwin/pypy/pypy-pypy-ac392fb76904.patch (hg revision c7dff5469611). Please tell me if I missed something.
Armin: You did not check in the change below (you may have left that one out on purpose, I don't know). If I recall correctly, having write_barrier_failing_case_ptr uninitialized caused an error when compiling with --gc=hybrid --opt=jit. I doubt this is Cygwin specific. On any case, without the patch below "python translate.py -Ojit" succeeds straight out of the repository under Cygwin with pypy source revision: 1613ac9e0cbd, tag: tip, date: Wed Jun 13 16:16:58 2012 +0200. Great!
--Uwe *** pypy/pypy/rpython/memory/gctransform/framework.py Sun Jun 3 11:01:06 2012 --- ../pypy-work/pypy/pypy/rpython/memory/gctransform/framework.py Tue Jun *************** *** 448,453 **** --- 448,454 ----
self.write_barrier_ptr = None self.write_barrier_from_array_ptr = None + self.write_barrier_failing_case_ptr = None if GCClass.needs_write_barrier: self.write_barrier_ptr = getfn(GCClass.write_barrier.im_func, [s_gc,