On 30 April 2011 01:26, Bruno Cardoso Lopes <[email protected]> wrote: > I know in some cases we prefer avoiding the usage of intrinsics as > much as possible and I'm not aware what it's the policy regarding the > addition of new ARM builtins and intrinsics. However, this patch > implements builtins and intrinsics for some coprocessor instructions, > they have a clean interface and directly match the real instructions.
Hi Bruno, I'm not sure about the intrinsic policy in this case, too, but the change is most welcomed. I think the patch looks nice, just would be better to have some tests on the arguments, too. I know most LLVM tests don't do it, but in this case, we need to make sure the options are going to the right place. cheers, --renato _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
