6 Feb
2023
6 Feb
'23
4:22 p.m.
On Mon Feb 6 16:19:06 2023 +0000, Jinoh Kang wrote:
Hm, the approach might not be viable unless we roll our own SList, since the RTL SList operations do not perform atomic access on `entry->Next`.
Fwiw I think this also creates another problem with when fully empty groups are released. Right now as we wait for a group to go trough a fully used -> fully freed cycle before deciding what to do, we have the option of releasing the group entirely to the subheap.
Putting groups into the linked lists earlier removes that option and I think it becomes harder to act on fully freed groups.