introduce sys_membarrier(): process-wide memory barrier (v8)
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest3 patches 37 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/3 | Create spin lock/spin unlock with distinct memory barrierarch/ arch/ include/ +2 more | +65 | 6 replies | |
| 2/3 | scheduler: add full memory barriers upon task switch at runqueue lock/unlockkernel/ | +20 | 31 replies | |
| 3/3 | introduce sys_membarrier(): process-wide memory barrier (v8)arch/ arch/ arch/ +4 more | +180 | 0 replies |
Reviews trailers per patch
2 of 3 patches have a Reviewed-by or Acked-by.
| Patch | KOSAKI Motohiro | Steven Rostedt |
|---|---|---|
| 1/3 | ||
| 2/3 | A-b | |
| 3/3 | A-b | A-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [patch 0/3] introduce sys_membarrier(): process-wide memory barrier (v8)
Hi,
Here is an updated version of the sys_membarrier system call. It implements a system call that allows userspace to distribute the overhead of memory barriers asymmetrically. It is particularly useful for Userspace RCU (liburcu), to allow the RCU read-side to only issue compiler barriers, matched by calls to sys_membarrier() in the userspace synchronize_rcu().
Even though sys_membarrier should be portable to other architecture as-is without arch-specific modifications, I propose to incrementally reserve the system call IDs as we test it on each architecture.
Here is a version that I think should be ready for merging. It is based on the 2.6.33-rc6 kernel. I would appreciate if people who gave their Acked-by in the RFC rounds could confirm their Acked-by following the patch set updates.
Thanks,
Mathieu
Signature · 2 lines
Mathieu Desnoyers OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68