Re: [PATCH] slub: Fix calculation of cpu slabs

2013-12-29 Thread Pekka Enberg
On Sat, Oct 5, 2013 at 10:48 AM, Wanpeng Li wrote: > On Tue, Sep 10, 2013 at 11:43:37AM +0800, Li Zefan wrote: >> /sys/kernel/slab/:t-048 # cat cpu_slabs >> 231 N0=16 N1=215 >> /sys/kernel/slab/:t-048 # cat slabs >> 145 N0=36 N1=109 >> >>See, the number of slabs is smaller than that

Re: [PATCH] slub: Fix calculation of cpu slabs

2013-12-29 Thread Pekka Enberg
On Sat, Oct 5, 2013 at 10:48 AM, Wanpeng Li liw...@linux.vnet.ibm.com wrote: On Tue, Sep 10, 2013 at 11:43:37AM +0800, Li Zefan wrote: /sys/kernel/slab/:t-048 # cat cpu_slabs 231 N0=16 N1=215 /sys/kernel/slab/:t-048 # cat slabs 145 N0=36 N1=109 See, the number of slabs is smaller

Re: [PATCH] slub: Fix calculation of cpu slabs

2013-10-14 Thread Li Zefan
Hi Pekka, could you pick up this patch? On 2013/9/11 5:06, Christoph Lameter wrote: > On Tue, 10 Sep 2013, Li Zefan wrote: > >> We should use page->pages instead of page->pobjects when calculating >> the number of cpu partial slabs. This also fixes the mapping of slabs >> and nodes. > > True.

Re: [PATCH] slub: Fix calculation of cpu slabs

2013-10-14 Thread Li Zefan
Hi Pekka, could you pick up this patch? On 2013/9/11 5:06, Christoph Lameter wrote: On Tue, 10 Sep 2013, Li Zefan wrote: We should use page-pages instead of page-pobjects when calculating the number of cpu partial slabs. This also fixes the mapping of slabs and nodes. True. As

Re: [PATCH] slub: Fix calculation of cpu slabs

2013-09-10 Thread Christoph Lameter
On Tue, 10 Sep 2013, Li Zefan wrote: > We should use page->pages instead of page->pobjects when calculating > the number of cpu partial slabs. This also fixes the mapping of slabs > and nodes. True. > As there's no variable storing the number of total/active objects in > cpu partial slabs, and

Re: [PATCH] slub: Fix calculation of cpu slabs

2013-09-10 Thread Christoph Lameter
On Tue, 10 Sep 2013, Li Zefan wrote: We should use page-pages instead of page-pobjects when calculating the number of cpu partial slabs. This also fixes the mapping of slabs and nodes. True. As there's no variable storing the number of total/active objects in cpu partial slabs, and we

[PATCH] slub: Fix calculation of cpu slabs

2013-09-09 Thread Li Zefan
/sys/kernel/slab/:t-048 # cat cpu_slabs 231 N0=16 N1=215 /sys/kernel/slab/:t-048 # cat slabs 145 N0=36 N1=109 See, the number of slabs is smaller than that of cpu slabs. The bug was introduced by commit 49e2258586b423684f03c278149ab46d8f8b6700 ("slub: per cpu cache for partial

[PATCH] slub: Fix calculation of cpu slabs

2013-09-09 Thread Li Zefan
/sys/kernel/slab/:t-048 # cat cpu_slabs 231 N0=16 N1=215 /sys/kernel/slab/:t-048 # cat slabs 145 N0=36 N1=109 See, the number of slabs is smaller than that of cpu slabs. The bug was introduced by commit 49e2258586b423684f03c278149ab46d8f8b6700 (slub: per cpu cache for partial