> I try change you make, with "unsigned" but I have same error as few of my > attempt as well, there is problem with computation not being accurate: > number of registers used: 9 > Traceback (most recent call last): > File "DemoMetaMatrixmulCheetah.py", line 150, in <module> > assert error <= 1e-2, error > AssertionError: 64.2192 > I am use most recent pycuda with most recent cuda (3.2) with OSX 10.6. > when you are saying "what compiler" I am not sure exactly your meaning?
I've had similar problems with OS X, writing custom CUDA C code. Mine were fixed with an #include <stdint.h> -josh > On Tue, Feb 8, 2011 at 6:11 PM, Andreas Kloeckner <li...@informa.tiker.net> > wrote: >> >> On Tue, 8 Feb 2011 16:30:19 -0500, Tobjan Brejicz <toba...@gmail.com> >> wrote: >> > Hello again, >> > >> > Thank you for quick reply before. This was very helping, I now run >> > things >> > correctly on that example. >> > >> > But here is one new question I have. I am trying example >> > "DemoMetaMatrixmulCheetah.py". >> > >> > So the error is: >> > >> > error: identifier "uint" is undefined >> >> Weird, works for me. What version of CUDA? What compiler? In any case, >> replacing those by 'unsigned' should work. I've done that for the file >> on the wiki. The new version works for me, too. >> >> Andreas > > > _______________________________________________ > PyCUDA mailing list > PyCUDA@tiker.net > http://lists.tiker.net/listinfo/pycuda > > _______________________________________________ PyCUDA mailing list PyCUDA@tiker.net http://lists.tiker.net/listinfo/pycuda