On Saturday 19 December 2009, Richard Henderson wrote: > Changes from round 3: > > * Drop movcond for now. > * Only use movzbl and not xor in setcond.
I'm still catching up on mail backlog from this thread, but I'm concerned that we're exposing setcond to the target translation code if we're planning on implementing movcond later. My guess is that in a lot of cases we want a value other than 1, and I'd prefer to avoid proliferation of set+shift/mask sequences if we're going to get movcond anyway. I don't suppose you've tried something along the lines of #define tcg_gen_movcond_i32(cond...) gen_helper_mov##cond(...) And see how that compares in practice? Paul
