On 8 April 2013 13:04, Sebastian Berg <sebast...@sipsolutions.net> wrote:
> On Mon, 2013-04-08 at 12:31 +0200, Dag Sverre Seljebotn wrote: > > On 04/08/2013 09:59 AM, Sebastian Berg wrote: > > > On Mon, 2013-04-08 at 08:42 +0200, Dag Sverre Seljebotn wrote: > > >> On 04/06/2013 04:19 PM, Nathaniel Smith wrote: > > >>> Hi all, > > >>> > > >>> If you build current numpy master with > > >>> NPY_RELAXED_STRIDES_CHECKING=1 python setup.py install > > >>> then Cython code using ndarrays starts blowing up, e.g.: > > >>> > > >>> # foo.pyx > > >>> def add_one(array): > > >>> cdef double[::1] a = array > > >>> a[0] += 1. > > >>> return array > > >>> > > >>>>>> foo.add_one(np.ascontiguousarray(np.arange(10.)[::100])) > > >>> Traceback (most recent call last): > > >>> File "<stdin>", line 1, in <module> > > >>> File "foo.pyx", line 2, in foo.add_one (foo.c:1210) > > >>> cdef double[::1] a = array > > >>> ValueError: Buffer and memoryview are not contiguous in the same > dimension. > > >>> > > >>> The problem (as discussed before) is that Cython has an unnecessarily > > >>> strict definition of "contiguous", so NPY_RELAXED_STRIDES_CHECKING=1 > > >>> pretty much breaks all existing compiled Cython modules. > > >>> > <snip> > > >> > > >> I guess you have changed your implementation of PEP 3118 too slightly > on > > >> the NumPy side? Is this undefined in the PEP or are you now not in > > >> strict adherence to it? > > >> > > > > > > Hi, > > > > > > maybe I will have a look at that, but not sure if I will manage. The > > > PEP 3118 is a point, but it does not seem to cover this (probably we > > > should try to get a clarification into 3118). I am still wondering > > > whether for buffers that are requested contiguous numpy should set the > > > strides again, since it cannot hurt. Would that make a difference for > > > Cython? I expected Cython just got any buffer and then checked the > > > strides instead of requesting the buffer contiguous and then double > > > checking. > > > > At least when I implemented > > > > cdef np.ndarray[double, mode='fortran'] arr > > > > that relies on PEP 3118 contiguous-flags and I did no checking myself. > > Lots of Cython code does this instead of memoryviews (I still write my > > own code that way). > > Yeah, though even if numpy "fixes" the buffer strides, `arr.strides` > points at the original array I believe, leaving it to the user. But that > just means users have to be a bit more careful with strides. > > I think I will have numpy "fix" the buffers stride anyway (when > requested contiguous through the buffer protocol). I cannot think of > any reason not to do it and it may work around bugs in extensions which > may not even be aware of numpy's existence. > > > > > The memory views OTOH does their own checking, but I also see plenty of > > references to PyBUF_C_CONTIGUOUS etc. inside > > Cython/Utility/MemoryView.pyx, so perhaps it does both. Mark would have > > the definitive answer here. > > > > After a quick check, it seems to me that those are there for exposing > the memoryviews buffer and not for getting a buffer from another object. Yeah, I think most of the checking is in MemoryView_C.c, when obtaining the memoryview. I don't know how well buffer producers check these flags, but maybe we can simply remove some of those checks, or amend them otherwise. I'll try to look into it one of these days. > > - Sebastian > > > Dag Sverre > > _______________________________________________ > > cython-devel mailing list > > cython-devel@python.org > > http://mail.python.org/mailman/listinfo/cython-devel > > > > > _______________________________________________ > cython-devel mailing list > cython-devel@python.org > http://mail.python.org/mailman/listinfo/cython-devel >
_______________________________________________ cython-devel mailing list cython-devel@python.org http://mail.python.org/mailman/listinfo/cython-devel
