This makes two changes to trace projection:

• Handle a single backward jump anywhere in a trace, rather than just at the beginning.
• Rather than having a minimum trace length, just ensure that at least one instruction is fully translated. This prevents us from requiring the translation of several tiny instructions or "successfully" half-translating a large instruction.

Perf is neutral, but stats move in the right direction: 83% reduction in "inner loop found", 7% reduction in "trace too short", and 5% fewer optimization attempts overall. We execute 0.5% more uops and 0.5% fewer traces. 2% fewer tier one instructions executed overall.

• Issue: Increase fraction of code executed by tier 2. #118093