61b616755a
The Cortex-A55 scheduling model is used for -mcpu=generic, meaning it can have a wider effect than just the A55. The changes to the A55 scheduling model seems to have caused performance regressions on Cortex-A510 device which have latencies closer to the original and different forwarding paths. This partially reverts the changes from D117003, at least until we can do something to improve Cortex-A510. According to my results, this improves the A510 results without altering the A55 very much. |
||
|---|---|---|
| .. | ||
| IPC | ||
| A53-carry-over.s | ||
| A55-add-sequence.s | ||
| A55-all-stats.s | ||
| A55-all-views.s | ||
| A55-basic-instructions.s | ||
| A55-in-order-retire.s | ||
| A55-load-readadv.s | ||
| A55-load-store-alias.s | ||
| A55-load-store-noalias.s | ||
| A55-neon-instructions.s | ||
| A55-out-of-order-retire.s | ||
| A55-store-readadv.s | ||
| direct-branch.s | ||
| forwarding-A57.s | ||
| in-order-bottleneck-analysis.s | ||
| shifted-register.s | ||