P86 doubled the unified TLB from four entries to eight and reran the P84 BusyBox shell workload.
Result:
| metric | P84 | P86 | delta |
|---|---|---|---|
| post-load cycles | 239,533,716 | 223,777,049 | -6.58% |
| CPI | 2.6615 | 2.5615 | -3.76% |
| fetch walks | 2,263,038 | 1,117,037 | -50.64% |
| load walks | 2,267,672 | 973,288 | -57.08% |
| store walks | 601,266 | 199,592 | -66.80% |
| memory handshakes | 39,642,301 | 33,111,189 | -16.48% |
Shell phase notes: ls improved 7.68%, cat improved 44.69%, and the
8x ash loop improved 29.02%. The /tmp file phase got 20.44% slower in
this run, so the next round should avoid over-reading one aggregate
number.
Status: PASS for RTL simulation and shell benchmark. LibreLane
hardening is NOT RUN.