[HN Gopher] The 5-year journey to bring restartable sequences to...
___________________________________________________________________
The 5-year journey to bring restartable sequences to Linux (2019)
Author : gbrown_
Score : 25 points
Date : 2022-03-25 09:38 UTC (1 days ago)
(HTM) web link (www.efficios.com)
(TXT) w3m dump (www.efficios.com)
| throwaway81523 wrote:
| I hadn't heard about this. Cool. Two obvious questions:
|
| 1) You are supposed to check the cpu number at the start and end
| of your critical section, to make sure it hasn't changed. But
| what if it has changed twice? Traditionally this is handled by a
| CAS instruction, but they are trying to avoid that.
|
| 2) Does the single instruction commit not itself need a lock
| prefix or something similar? Futex has to do something like that
| unless I'm mistaken.
|
| Actually, 3) if your cpu supports something like TSX (did Intel
| ever get that working right?), would you want to use that in
| preference to rseq?
___________________________________________________________________
(page generated 2022-03-26 23:01 UTC)