Using the pixel quality layers of the Landsat SR products to create cloud and water masks in R

I tried to use the RStoolbox package function classifyQA to generate information on quality of the layers of interest, namely – cloud, cloud shadows, and water. I am unable to understand how to mask clouds and water using this function.

#One of the attempts

cloudMask(img,crop_pixelqaband_1992) # trying to use the pixel QA band provided by SR products to create a mask. I understand right now the code is not specific in pulling out only the cloud and cloud shadow data from the QA band layer.

Return error

Error in .local(x, ...) : invalid layer names # the function only seems to take blue and thermal bands

May be I am doing this wrong? Is there another way to do cloud and water masking using Landsat surface reflectance products that do not provide thermal bands in R? Can someone please suggest how one can go about carrying out masking for clouds and water in R using the pixel quality band of SR products?