Optimisation flags

I have been using GCC for my project for a while now and I am now in the process to migrating my project to use Clang/LLVM (ELLCC to be specific).

To optimise the performance of the project on the targte system I specify a number of flags to GCC. The flags are just enabling the use of SSE2 or SSE3, along other target specific instructions along with some definitions about the size of the cache.

I have managed to find a way to pass an equivalent set of flags to llc from clang for all of the instructions that are available on the target system, but I am so far unable to find a way to specify the cache sizes. Specifically, I am setting l1-cache-size, l1-cache-line-size, and l2-cache-size with GCC, does Clang/LLVM provide a means to specify the same details?