When I run my app and select the GPU OpenCL device, the feeder thread stalls inside a blocking call to clEnqueueMapBuffer().
I've some problem with Shared Virtual Memory and its buffers.
I've got structure like this:
From the issue: [NOT SOLVED] ioc64: simple kernel killed.
Guide to reporting issues: https://software.intel.com/en-us/articles/ocl-issue-reporting-guide
I use OpenCL SPIR for two algorithms, each one has templated functions for uchar, ushort, uint, float and double (templates are implemented via defines/includes).
It seems to me that GEN might benefit more from detecting "subgroup uniform" values than other architectures because of its
I see that Kaby Lake IGPs have OpenCL 2.1 support.
Does this imply that SPIR-V is supported on that platform?
Is there any plan to bring 2.1 support to GEN9 and GEN8?