Timeline for Operating systems - whose responsibility is it to coordinate process I/O requests?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 5, 2022 at 9:27 | vote | accept | Stefan Rendevski | ||
| Oct 3, 2022 at 13:25 | comment | added | Kilian Foth | No, by calling a blocking system call, you implicitly accept that you might never (in this round) wake up again :-) | |
| Oct 3, 2022 at 13:22 | comment | added | Stefan Rendevski | Is the scheduler able to force a process to yield in that case, or must this be handled explicitly by the process's thread? | |
| Oct 3, 2022 at 12:51 | comment | added | Kilian Foth | Yes, if your control flow cannot proceed without the external data, then you effectively waste the rest of your time slice. (Yielding your time slice early might earn you credit from the scheduler to be treated favorably on further wakeups, but either way, I/O waiting massively increases your wallclock time.) | |
| Oct 3, 2022 at 12:12 | comment | added | Stefan Rendevski | if I understand the first paragraph correctly, assuming requested data does not arrive within the process' allocated time slot, the remainder of that time slot will effectively be wasted? | |
| Oct 3, 2022 at 12:06 | history | answered | Kilian Foth | CC BY-SA 4.0 |