I believe the ancestor of bird lungs first evolved in a branch of theropoda. I'm not sure what drove that adaptation but it proved useful fodder when much later

I believe the ancestor of bird lungs first evolved in a branch of theropoda. I'm not sure what drove that adaptation but it proved useful fodder when much later the avia branch of theropoda evolved flight. This also mirrors feathers which seem to have evolved long before flight. It is likely small lightweight avia already possessed feathers, light bones, and efficient lungs which allowed them to do a tiny bit of gliding in their jumps and that is what led to flight. Once flight exists there's really strong pressure to optimize to the limit given the metabolic cost of flying. Mammals descend from synapsids which branched off before sauropsids branched off into theropoda, reptiles, etc.