remzi-arpacidusseau / ostep-typos

50 stars 44 forks source link

Possible error in figure 8.4? #87

Open marcoaffolter opened 9 months ago

marcoaffolter commented 9 months ago

Sorry if this is just me misunderstanding (which it probably is), but in the right part of the figure 8.4, shouldn't the black job be deprioritized after being scheduled for the first time after the initial priority boost? It looks as if it has used it's time slice and I would therefore expect it to be deprioritized into Q1, according to the MLFQ rule 4a, before recieving another priority boost at around 200ms.

huoenter commented 7 months ago

I have the same question. Found #13 when the boost time was still 50 ms. Changing it to 100 ms solves the question in #13 but also makes the right half of the (right) figure inconsistent. Still feels it's better to keep 50ms boost time and fix the left half of the (right) figure, like in the two slides below:

~Chen Huo (Shippensburg University)