In case you havent gotten to the very rootcause of this symptom yet, some addt'l means for debugging might come in handy.
Obviously, debugging OpenCL kernels encapsulated in python host apps isnt much of a milk run for sure but, well, tricking around never is forbidden :) ==> May want to check out Andreas' site: http://wiki.tiker.net/PyOpenCL A new category "Debugging" exists, sharing some information on a patricular approach that helps to debug opencl kernel beasts inside a python wilderness. Text sheds light on what is needed to get installed i.t.o. tools. Attachment to that page (PDF) contains a screenshot and illustrates a functioning configuration of em. Look it up. Not that trailblazing piece of theory but, hey, need work smarter not harder and if that simple fix does the job efficiently we're fat and happy :-) Enjoy, good luck ! P.S. To be sure - once to that site pls. note the statement on removal of printf's from kernels ! On Tue, Sep 15, 2015 at 11:30 PM, Joe Haywood <[email protected]> wrote: > So I wrote an OpenCL interface using scipy.Weave.inline for the section of > code I referred to in the previous post. When I run the scipy.weave code I > get the expected result, not the result I get with Pyopencl. So I am at a > loss. Does that spark any ideas for anyone? > > Reese > > > > Confidentiality Notice: > This e-mail, including any attachments is the property of Trinity Health > and is intended for the sole use of the intended recipient(s). It may > contain information that is privileged and confidential. Any unauthorized > review, use, disclosure, or distribution is prohibited. If you are not the > intended recipient, please delete this message, and reply to the sender > regarding the error in a separate email. > _______________________________________________ > PyOpenCL mailing list > [email protected] > http://lists.tiker.net/listinfo/pyopencl >
_______________________________________________ PyOpenCL mailing list [email protected] http://lists.tiker.net/listinfo/pyopencl
