Oh cool! I always thought of a branch as a conditional jump - I guess I was wrong. It appears that even unconditional branches are still branches!

"A branch is an instruction in a computer program that may, when executed by a computer, cause the computer to begin execution of a different instruction sequence. Branch (or branching, branched) may also refer to the act of beginning execution of a different instruction sequence due to executing a branch instruction. A branch instruction can be either an unconditional branch, which always results in branching, or a conditional branch, which may or may not cause branching depending on some condition."

I wouldn't be surprised if there was some place that used "branch" to indicate conditional-only. Assembly language terminology seems to be wonderfully inconsistent. But typical usage these days seems to be as you quoted.

Intel uses Jcc (Jump conditional) while Motorola uses Bcc (Branch conditional). Additionally, the Motorola 68k has both JMP and BRA (branch unconditionally) and the Motorola 6809 has JMP, BRA and BRN (branch never---useful for a two or four byte NOP, or for changing a conditional branch when patching code).

But generally being branch-free is in the context of avoiding a mispredict and pipeline stall. The processor is perfectly capable of pipelining an unconditional jump