This is an idea proposed in 2011 as a good starter project, and has been completed by William Morland. It was supervised by Robert M Watson and Anil Madhavapeddy as part of my Unikernels project.
In the summer of 2011, we hosted William Morland to do an internship in the Computer Lab just as the CTSRD/CHERI project kicked off. I was interested in MIPS as a potential target for MirageOS, and Robert M Watson in using it for the future CHERI processor.
William Morland hacked on the gxemul MIPS simulator, validating (and often creating) the CHERI test suite against the gxemul simulator. He then shifted gears into the (then experimental) LLVM/MIPS backend, filling in missing instructions and finding bugs via exercising the test suite. His LLVM repository is up at GitHub, along with the discussions from back in 2011 on the llvm-dev lists. There's also a nice poster of this work from the 2011 CTSRD project meeting!