On 03/19/2013 11:40 AM, Zhouping Liu wrote:
> changed log from v1 to v2:
>  1. added a new is_numa() function instead of ISNUMA global variable;
>  2. added new patch to remove the prefix underscroe '_' of the
>     internal function name;
>  3. used SAFE_FILE_SCANF() instead of read_file().
>  4. some cleanup...
> 
> changed log from v2 to v3:
>  1. as Jan's suggestion, pass 'nodes=NULL' to get_allowed_nodes_arr()
>     in is_numa(), and free 'nodes' variable when it passed a non-NULL
>     value inside testoom().
> 
> changed log from v3 to v4:
>  1. fixed some patch style error
> 
> The patch set updated all OOM testcases, and added a new testcase.
> 
> Inside old version, OOM only test MPOL_MBIND mempolicy in NUMA
> system, and the combination of CPUSET, MEMCG & NUMA is complex,
> and not clear. Also CPUSET is only used in NUMA system.
> 
> These patch set adjust them, now we can coverage the below tests:
>  1) OOM under normal - oom01
>  2) OOM under all mempolicy - oom02
>  3) OOM under MEMCG (+ mempolicy) - oom03
>  4) OOM under CPUSET (+ mempolicy)  - oom04
>  5) OOM under CPUSET & MEMCG - oom05


Applied, thank you.

Wanlong Gao


------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Reply via email to