Timeline for Should small overhead count towards an upper memory bound or rather be ignored?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 5, 2022 at 10:09 | audit | Suggested edits | |||
| Jan 5, 2022 at 12:06 | |||||
| Dec 17, 2021 at 9:35 | comment | added | Mecki | This is one of the things that is hard to say. The class is part of a framework that helps implementing network protocols in user space. When being used on a normal PC or a server, memory is not such an issue as these devices have plenty of it and they also have swap space. Yet when being used on an embedded device with limited RAM and no swap space, memory usage is far more critical, I guess. On the other hand, being able to know how much memory you can get out of the pool can be critical as well for some protocols. | |
| Dec 17, 2021 at 6:45 | comment | added | Bart van Ingen Schenau | What will be the typical system where your allocator will get used? Would the designer of such a system accept an unknown, unbounded overhead in memory usage in order to be guaranteed they can allocate 512 kiB from your allocator in arbitrary large/small chunks? Or would they want to have an upper bound on the total memory used by your allocator and accept that some of that will be used for internal administration? | |
| Dec 16, 2021 at 19:29 | answer | added | candied_orange | timeline score: 3 | |
| Dec 16, 2021 at 18:24 | history | asked | Mecki | CC BY-SA 4.0 |