[yvdriess]

What do you mean by 'controlled by hardware'? The registers the compiler chooses for you are an abstraction themselves, one of the first things a CPU does is register renaming. Same for main memory, you are presented mostly an abstraction of main memory, with a ton of layers in-between (load/store buffers, write-combine buffers, coherent caches, etc)

Turning the argument the other way, you as a programmer can control a lot about caches: you can prefetch cache lines, invalidate them and use streaming instructions to bypass them.

[vagab0nd]

What you said makes sense. The question is more from the perspective of the compiler/assembler. At these layers you explicitly say, move this number to this register, or read a number from this address. But you very rarely are able to say, copy this chunk of data into this cache line. Sure there are exceptions (like the GPU), and there are cases where you can hack it to do what you want. But in general you don't get to control the cache in a very specific way.