My understanding, from what I have been told in my thesis work, is that the literature review should demonstrate that you have read everything that is critical to your work, understood it, and can explain why it is relevant.
More practically, this means that you need to identify the core areas of your subject that relate to your thesis topic. They within these areas, identify the most critical texts, and read them. Then write sections that explore the topic, drawing heavily from the books with quotes. Make sure that you are arguing for What this area involves into Why this is relevant.
In my case, for HCI, I talk a little about introducing new technology, because this is a critical part of HCI design, and I refer to Zuboff and Bannon, referened below. These are the two key authors in the specific area focus that I need to address, and I quote them liberally talking around the focus of the topic that I require.
If your writing is just a litereature review, then it probably needs more detail, more personal exploration of the topic in more depth. But the principles are the same - read the most important literature on the specific topic, and demonstrate that you understand it.
Zuboff, S (1988) In the age of the smart machine – the future of work and power, Oxford: Heinemann Professional Publishing
Bannon, L (1991) From Human Factors to Human Actors. In Greenbaum, J and Kyng, M (Eds.)(1991) Design at work: Cooperative Design of Computer Systems. Hillsdale: Lawrence Erlbaum Associates pp. 25-44
Bannon, L (2011) Reimagining HCI: Towards a more Human-Centred Perspective. Interactions. New York: ACM, pp50-57