Hi Richard, Thanks for the review, now committed.
> The new aarch64_split_compare_and_swap code looks a bit twisty. > The approach in lse.S seems more obvious. But I'm guessing you > didn't want to spend any time restructuring the pre-LSE > -mno-outline-atomics code, and I agree the patch in its current > form is more backportable. Indeed this code needs cleaning up - all the complex speculation stuff should be behind a simple interface. I was thinking of emitting CSEL here but it would require adding new TI mode patterns or manually splitting into low/high parts and emitting CSEL. > I suppose we might want to backport this after it has been in trunk > for a bit. Yes that was my plan. Cheers, Wilco