Added a comment

Author: Przemog1

src/nbl/video/utilities/ImageRegionIterator.cpp

@@ -67,7 +67,7 @@ ImageRegionIterator::ImageRegionIterator(
         asset::ICPUImage::SCreationParams inCPUImageParams = dstImageParams;
         inCPUImageParams.flags = asset::IImage::ECF_NONE; // Because we may want to write to first few layers of CUBEMAP (<6) but it's not valid to create an Cube ICPUImage with less that 6 layers.
         inCPUImageParams.format = srcImageFormat;
-        inCPUImageParams.viewFormats.reset();
+        inCPUImageParams.viewFormats.reset(); // Resetting viewFormats to 0 since all flags were reset earlier, setting flags to ECF_NONE and not resetting viewFormats may cause validation errors during ICPUImage creation
         inCPUImageParams.extent = region.imageExtent;
         inCPUImageParams.arrayLayers = region.imageSubresource.layerCount;
         inCPUImageParams.mipLevels = 1u; // since we copy one mip at a time to our dst image, it doesn't matter at the stage when we copy from cpu memory to staging memory
@@ -105,7 +105,7 @@ ImageRegionIterator::ImageRegionIterator(
 
         asset::ICPUImage::SCreationParams outCPUImageParams = dstImageParams;
         outCPUImageParams.flags = asset::IImage::ECF_NONE; // Because we may want to write to first few layers of CUBEMAP (<6) but it's not valid to create an Cube ICPUImage with less that 6 layers.
-        outCPUImageParams.viewFormats.reset();
+        outCPUImageParams.viewFormats.reset(); // Resetting viewFormats to 0 since all flags were reset earlier, setting flags to ECF_NONE and not resetting viewFormats may cause validation errors during ICPUImage creation
         outCPUImageParams.extent = region.imageExtent;
         outCPUImageParams.arrayLayers = region.imageSubresource.layerCount;
         outCPUImageParams.mipLevels = 1u;