Yes, if it would double the performance in many cassis, then the size would have a significant benefit.

Here we are taking about large binaries without apparent benefit and a drive to keep binaries as lean as possible.

This is effectively what is happening btw; the crdb binary went from 80MB to 200MB in the same time it took to make it twice as fast. The % growth in size is not a problem on its own; it's more the % size attributed to the program vs. the % size attributed to unclear purposes, that's a problem.