Missing case for T_FLOAT4 in AllocateTextureGPU
wouterraateland opened this issue · comments
The project throws an "invalid configuration" error when adding a float4
channel to a gvdb instance.
This seems to be due to the function Allocator::AllocateTextureGPU
, located in gvdb_allocator.cpp
.
Adding the following lines in the switch statements above the call to cuArray3DCreate
fixes the problem for me:
case T_FLOAT4:
desc.Format = CU_AD_FORMAT_FLOAT;
desc.NumChannels = 4;
break;
I'm just getting started with GVDB, so perhaps I'm making another mistake, but I thought It'd be good to bring this simple fix to you attention anyway.