The big software unification driving this has been chemists moving to the (post-)DFT methods the materials people had to use, even for some of their small molecule work. Oversimplifying wildly, a lot of small molecule chemistry is driven by medicine and they started caring a lot about, for example, ab-initio protein models…
As many other people have said, there is no physics or chemistry in solid-state materials; most universities will have multiple groups in this area across three to five different departments (physics, chemistry, materials science, materials engineering, earth sciences).
(The reason materials people were on DFT is that you need to simulate more atoms to model a material. Standard DFT methods scale as O(n^3) with system size, and there are linear-scaling methods like SIESTA and ONETEP; many computational chemistry methods like Configuration Interaction are much worse.)
