In package control mode, the performance of all cores depends on the
most recent value written to the request field. If the last write comes
from an E-core, all cores are forced to align with the E-core
performance level, resulting in significant performance degradation.
Therefore, package control is disabled on hybrid-core systems.
Details
Details
- Reviewers
olce cem - Commits
- rG7b26353a59d6: hwpstate_intel: Disable package control on hybrid CPU
Diff Detail
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Passed - Unit
No Test Coverage - Build Status
Buildable 73603 Build 70486: arc lint + arc unit
Event Timeline
Comment Actions
Fine like this.
Our discussion elsewhere, mentioning that small_core does not change after initialization, implied that you could even dispense with using the smp_rendezvous_cpus() and directly iterate on per-CPU structures and read the field there, but since this code is run only once on identify, changing what you already coded is not worth it. Maybe this function is going to be helpful in more settings, in which case we can factor it out later (which reminds me I should have a look to recent changes to the HMP revisions).