Apple’s Chris Lattner, in a post to the LLVMdev mailing list:
I’m happy to announce that Apple is contributing some major
extensions to the LLVM ARM backend. The improvements include support
for ARM v4/v6, vfp support, soft float, pre/postinc support, load/
store multiple generation, constant pool entry motion (to support
large functions), and support for the darwin/arm ABI.
In other words, more evidence that the iPhone is using an ARM processor, and that Apple is using LLVM to get higher performance out of what is, compared to a Mac, the iPhone’s wimpy CPU and GPU. (John Siracusa had some good background pieces on LLVM a few months ago.) It’s worth pointing out that Lattner declined to comment on whether this was actually related to the iPhone, but, well, come on.
Wes Felter says VFM support implies that it’s not an Xscale CPU, which I think narrows it down to Samsung.
★ Wednesday, 31 January 2007