Telling the scheduler about thermal pressure
Telling the scheduler about thermal pressure
Posted May 17, 2019 15:29 UTC (Fri) by admalledd (subscriber, #95347)In reply to: Telling the scheduler about thermal pressure by ScienceMan
Parent article: Telling the scheduler about thermal pressure
My (naive!) understanding is that these specific thermal pressure events are very short lived, and can also be per-core. The scheduler already would take into account general frequency reduction over a longer term reasonably well, it is the rapid thermal reaction of cutting frequency drastically that is being added here specifically.
Posted Jun 4, 2019 17:06 UTC (Tue)
by oak (guest, #2786)
[Link]
Telling the scheduler about thermal pressure
TDP limits act are in my experience clearly faster than temperature limits. Heat takes some time to spread / dissipate whereas TDP limits are calculated and enforced (by firmware) based on HW counters of what HW is doing at given moment, it doesn’t need to wait for the effects of that activity.