-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ulrich,
>> On 10/29/2012 08:33 PM, Ulrich Pegelow wrote: >>> For some reason blendop.cl can not be built (compiled+linked) >>> by clBuildProgram(). The error code is -30 (CL_INVALID_VALUE) >>> which would indicate we called the routine with wrong >>> parameters. This is a bit surprising as all other kernels are >>> successfully built. So there must be something special with >>> bilateral.cl. >> >> I don't think I can follow you here. What does bilateral.cl have >> to do with blendop.cl not being compiled? >> > Oops, sorry. Mixed things up, when working on a bilateral.cl issue > in parallel :) Forget my proposal. I feel a lot less stupid now :-) > As to commenting out one of the kernels: all possible things might > happen here. We have not designed the code to survive such a case. Hmm.. ok. But is there perhaps another way to isolate the problem? - From one of the combinations I tried it seemed like dt worked without issues with only few of the openCL kernels working. Is there maybe another more systematic way of telling dt to skip some kernels while using others? At least we could find out if it is indeed a problem with one or more kernels, or rather something else... > What is optirun supposed to do btw? optirun is the main binary from the Bumblebee project (see http://www.bumblebee-project.org/) On NVidia "Optimus" systems the laptop display is only driven by the internal Intel (Ivy Bridge) graphics adapter. To still get acceleration from the NVidia card (without using a dedicated external screen) optirun starts a new headless X server using the nvidia driver, intercepts OpenGL (and OpenCL it seems) calls, and has those executed on the nvidia card. The results are then memcopyed into the video memory of the intel card so you get the output on the laptop screen. (This is how I understand the whole thing after being forced to use it for two days now). Unfortunately, the whole bumblebee project, related tools and tweaks are pretty much bleeding edge I guess - and it might very well turn out that darktable is just exposing a bug in some other component involved.. Jens -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iEYEARECAAYFAlCO28wACgkQbyTZFOIr/d1MvACghELc+MJ0nHHGxZO0qS5Ibo61 5dMAoIH4mfqRN7E/shmheW0Huv8rejoR =smgk -----END PGP SIGNATURE----- ------------------------------------------------------------------------------ The Windows 8 Center - In partnership with Sourceforge Your idea - your app - 30 days. Get started! http://windows8center.sourceforge.net/ what-html-developers-need-to-know-about-coding-windows-8-metro-style-apps/ _______________________________________________ darktable-devel mailing list darktable-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/darktable-devel
