Created attachment 132595 [details] debugoutput testing amdovx on intel git beignet logoutput is with export OCL_OUTPUT_SEL_IR=1 export OCL_OUTPUT_GEN_IR=1 export OCL_OUTPUT_KERNEL_SOURCE=1 https://github.com/GPUOpen-ProfessionalCompute-Libraries/amdovx-core
beignet build info -- Building mode: RelWithDebInfo -- LLVM llvm-config found at: /usr/bin/llvm-config -- find unstable LLVM version 3.9.1 -- Looking for XLib - found -- Looking for DRM - found at /usr 2.4.81 -- Checking for module 'libdrm_intel>=2.4.52' -- Found libdrm_intel, version 2.4.81 -- Looking for DRM Intel - found at /usr 2.4.81 -- Enable userptr support -- Enable EU total query support -- Enable subslice total query support -- Enable pooled eu query support -- Enable min eu in pool query support -- Looking for Xext - found at /usr -- Looking for Xfixes - found at /usr -- Looking for OpenGL - found at /usr 17.0.5 -- Looking for EGL - found at /usr 17.0.5 -- Looking for OCL ICD header file - found -- Building with OpenCL 1.2. -- Building mode: RelWithDebInfo -- Configuring done -- Generating done
In kernel calc_lens_distortion_and_warp_ma, use double constant M_PI_2, but beignet don't support double. Use M_PI_2_F can fix this issue.
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/beignet/beignet/issues/64.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.