Let m=f(1); by step 5, m>0. If m≤1, apply the full form of (∗) with x=1−m,y=1: f(1−m+f(1))=f(1−m)+f(1)+2f(1−m) becomes f(1)=f(1−m)+f(1)+2f(1−m), so f(1−m)=−2f(1−m)≤0; combined with f(1−m)≥0, f(1−m)=0, so 1−m=0, i.e. m=1. If m≥1, then by step 1, f(m)=f(f(1))=f(1)=m, so f(m2)=f(f(m))=f(m)=m; apply the full form of (∗) with x=m2−m,y=1: f(m2−m+f(1))=f(m2−m)+f(1)+2f(m2−m) becomes f(m2)=f(m2−m)+m+2f(m2−m), i.e. m=f(m2−m)+m+2f(m2−m), so f(m2−m)=−2f(m2−m)≤0; combined with f(m2−m)≥0, f(m2−m)=0, so m2−m=0, and since m≥1, m=1. Either way, f(1)=1.