Re: [PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-10-19 Thread Pekka Enberg
On Thu, 16 Aug 2012, Joonsoo Kim wrote: > When we try to free object, there is some of case that we need > to take a node lock. This is the necessary step for preventing a race. > After taking a lock, then we try to cmpxchg_double_slab(). > But, there is a possible scenario that

Re: [PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-10-19 Thread Pekka Enberg
On Thu, 16 Aug 2012, Joonsoo Kim wrote: When we try to free object, there is some of case that we need to take a node lock. This is the necessary step for preventing a race. After taking a lock, then we try to cmpxchg_double_slab(). But, there is a possible scenario that cmpxchg_double_slab()

Re: [PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-09-06 Thread JoonSoo Kim
2012/8/25 JoonSoo Kim : > 2012/8/16 Joonsoo Kim : >> When we try to free object, there is some of case that we need >> to take a node lock. This is the necessary step for preventing a race. >> After taking a lock, then we try to cmpxchg_double_slab(). >> But, there is a possible scenario that

Re: [PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-09-06 Thread JoonSoo Kim
2012/8/25 JoonSoo Kim js1...@gmail.com: 2012/8/16 Joonsoo Kim js1...@gmail.com: When we try to free object, there is some of case that we need to take a node lock. This is the necessary step for preventing a race. After taking a lock, then we try to cmpxchg_double_slab(). But, there is a

Re: [PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-08-24 Thread JoonSoo Kim
2012/8/16 Joonsoo Kim : > When we try to free object, there is some of case that we need > to take a node lock. This is the necessary step for preventing a race. > After taking a lock, then we try to cmpxchg_double_slab(). > But, there is a possible scenario that cmpxchg_double_slab() is failed >

Re: [PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-08-24 Thread JoonSoo Kim
2012/8/16 Joonsoo Kim js1...@gmail.com: When we try to free object, there is some of case that we need to take a node lock. This is the necessary step for preventing a race. After taking a lock, then we try to cmpxchg_double_slab(). But, there is a possible scenario that cmpxchg_double_slab()

[PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-08-15 Thread Joonsoo Kim
When we try to free object, there is some of case that we need to take a node lock. This is the necessary step for preventing a race. After taking a lock, then we try to cmpxchg_double_slab(). But, there is a possible scenario that cmpxchg_double_slab() is failed with taking a lock. Following

[PATCH 2/2] slub: remove one code path and reduce lock contention in __slab_free()

2012-08-15 Thread Joonsoo Kim
When we try to free object, there is some of case that we need to take a node lock. This is the necessary step for preventing a race. After taking a lock, then we try to cmpxchg_double_slab(). But, there is a possible scenario that cmpxchg_double_slab() is failed with taking a lock. Following