1 ====================
2 Percpu rw semaphores
3 ====================
4
5 Percpu rw semaphores is a new read-write semap
6 optimized for locking for reading.
7
8 The problem with traditional read-write semaph
9 cores take the lock for reading, the cache lin
10 is bouncing between L1 caches of the cores, ca
11 degradation.
12
13 Locking for reading is very fast, it uses RCU
14 instruction in the lock and unlock path. On th
15 writing is very expensive, it calls synchroniz
16 hundreds of milliseconds.
17
18 The lock is declared with "struct percpu_rw_se
19 The lock is initialized percpu_init_rwsem, it
20 -ENOMEM on allocation failure.
21 The lock must be freed with percpu_free_rwsem
22
23 The lock is locked for read with percpu_down_r
24 for write with percpu_down_write, percpu_up_wr
25
26 The idea of using RCU for optimized rw-lock wa
27 Eric Dumazet <eric.dumazet@gmail.com>.
28 The code was written by Mikulas Patocka <mpatoc
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.