Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-17 Thread Michal Hocko
On Tue 17-01-17 09:06:04, Huang, Ying wrote: > Michal Hocko writes: > > > Hi, > > I am seeing a lot of preempt unsafe warnings with the current mmotm and > > I assume that this patchset has introduced the issue. I haven't checked > > more closely but get_swap_page didn't use

Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-17 Thread Michal Hocko
On Tue 17-01-17 09:06:04, Huang, Ying wrote: > Michal Hocko writes: > > > Hi, > > I am seeing a lot of preempt unsafe warnings with the current mmotm and > > I assume that this patchset has introduced the issue. I haven't checked > > more closely but get_swap_page didn't use this_cpu_ptr before

Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-16 Thread Huang, Ying
Michal Hocko writes: > Hi, > I am seeing a lot of preempt unsafe warnings with the current mmotm and > I assume that this patchset has introduced the issue. I haven't checked > more closely but get_swap_page didn't use this_cpu_ptr before "mm/swap: > add cache for swap slots

Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-16 Thread Huang, Ying
Michal Hocko writes: > Hi, > I am seeing a lot of preempt unsafe warnings with the current mmotm and > I assume that this patchset has introduced the issue. I haven't checked > more closely but get_swap_page didn't use this_cpu_ptr before "mm/swap: > add cache for swap slots allocation" > > [

Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-16 Thread Michal Hocko
Hi, I am seeing a lot of preempt unsafe warnings with the current mmotm and I assume that this patchset has introduced the issue. I haven't checked more closely but get_swap_page didn't use this_cpu_ptr before "mm/swap: add cache for swap slots allocation" [ 57.812314] BUG: using

Re: [PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-16 Thread Michal Hocko
Hi, I am seeing a lot of preempt unsafe warnings with the current mmotm and I assume that this patchset has introduced the issue. I haven't checked more closely but get_swap_page didn't use this_cpu_ptr before "mm/swap: add cache for swap slots allocation" [ 57.812314] BUG: using

[PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-11 Thread Tim Chen
Change Log: v5: 1. Rebase patch series on 4.10-rc3 kernel. Update patch series to remove usage of obsoleted hot plug functions: cpu_notifier_register_begin(), cpu_notifier_register_done(), and __register_hotcpu_notifier() 2. Fix a bug returning uninitialized swap slot when we run out of swap

[PATCH v5 0/9] mm/swap: Regular page swap optimizations

2017-01-11 Thread Tim Chen
Change Log: v5: 1. Rebase patch series on 4.10-rc3 kernel. Update patch series to remove usage of obsoleted hot plug functions: cpu_notifier_register_begin(), cpu_notifier_register_done(), and __register_hotcpu_notifier() 2. Fix a bug returning uninitialized swap slot when we run out of swap