Looking at my branches: - numpypy-ctypes, numpy-record-type-pure-python: can be deleted - jit-tracehook: Seems like a good idea, let's you run the JIT even when there's a trace func -- looks like maybe some parts were landed and some weren't? - struct-double: I think all this work got done in a different way by someone else, can be deleted - dynamic-specialized-tuple: A really good idea; creates specialized tuples dynamically, I think it was basically merge ready except for one missing optimization, but I'm sure it's completely bitrotted now. - jit-sys-exc-info: no clue here - asm-backend-dupe-removal: no clue, maybe fijal remembers - unroll-virtual-dict-resize: probably still a good idea, will be tons of conflicts and I don't know if the heapcache issues that blocked it were ever fixed (needed interiorfield support there) - python-loop-unroll: very cool idea, looks like it was almost working (with huge hacks)
I'm not going to have time to finish any of my good idea branches :-) They can either be closed, or maybe someone else some day wants to look at them. Alex On Mon, Feb 11, 2019 at 11:28 AM Antonio Cuni <anto.c...@gmail.com> wrote: > On Mon, Feb 11, 2019 at 4:52 PM Carl Friedrich Bolz-Tereick <cfb...@gmx.de> > wrote: > >> Antonio Cuni <anto.c...@gmail.com> 2019-02-11 11:50 +0100 default >> > > let's close this one :) > _______________________________________________ > pypy-dev mailing list > pypy-dev@python.org > https://mail.python.org/mailman/listinfo/pypy-dev > -- All that is necessary for evil to succeed is for good people to do nothing.
_______________________________________________ pypy-dev mailing list pypy-dev@python.org https://mail.python.org/mailman/listinfo/pypy-dev
