Re: [PATCH] mm/slab: re-implement pfmemalloc support

2016-02-05 Thread Joonsoo Kim
2016-02-05 19:33 GMT+09:00 Mel Gorman : > On Thu, Feb 04, 2016 at 04:40:12PM +0900, Joonsoo Kim wrote: >> Current implementation of pfmemalloc handling in SLAB has some problems. >> >> 1) pfmemalloc_active is set to true when there is just one or more >> pfmemalloc slabs in the system, but it is

Re: [PATCH] mm/slab: re-implement pfmemalloc support

2016-02-05 Thread Mel Gorman
On Thu, Feb 04, 2016 at 04:40:12PM +0900, Joonsoo Kim wrote: > Current implementation of pfmemalloc handling in SLAB has some problems. > > 1) pfmemalloc_active is set to true when there is just one or more > pfmemalloc slabs in the system, but it is cleared when there is > no pfmemalloc slab in

Re: [PATCH] mm/slab: re-implement pfmemalloc support

2016-02-05 Thread Joonsoo Kim
2016-02-05 19:33 GMT+09:00 Mel Gorman : > On Thu, Feb 04, 2016 at 04:40:12PM +0900, Joonsoo Kim wrote: >> Current implementation of pfmemalloc handling in SLAB has some problems. >> >> 1) pfmemalloc_active is set to true when there is just one or more >> pfmemalloc

Re: [PATCH] mm/slab: re-implement pfmemalloc support

2016-02-05 Thread Mel Gorman
On Thu, Feb 04, 2016 at 04:40:12PM +0900, Joonsoo Kim wrote: > Current implementation of pfmemalloc handling in SLAB has some problems. > > 1) pfmemalloc_active is set to true when there is just one or more > pfmemalloc slabs in the system, but it is cleared when there is > no pfmemalloc slab in

[PATCH] mm/slab: re-implement pfmemalloc support

2016-02-03 Thread Joonsoo Kim
Current implementation of pfmemalloc handling in SLAB has some problems. 1) pfmemalloc_active is set to true when there is just one or more pfmemalloc slabs in the system, but it is cleared when there is no pfmemalloc slab in one arbitrary kmem_cache. So, pfmemalloc_active could be wrongly

[PATCH] mm/slab: re-implement pfmemalloc support

2016-02-03 Thread Joonsoo Kim
Current implementation of pfmemalloc handling in SLAB has some problems. 1) pfmemalloc_active is set to true when there is just one or more pfmemalloc slabs in the system, but it is cleared when there is no pfmemalloc slab in one arbitrary kmem_cache. So, pfmemalloc_active could be wrongly