Re: [PATCH] badness() dramatically overcounts memory

[Balbir Singh <balbir@linux.vnet.ibm.com>]

Jeff Davis wrote:
> In oom_kill.c, one of the badness calculations is wildly inaccurate. If
> memory is shared among child processes, that same memory will be counted
> for each child, effectively multiplying the memory penalty by N, where N
> is the number of children.
> 
> This makes it almost certain that the parent will always be chosen as
> the victim of the OOM killer (assuming any substantial amount memory
> shared among the children), even if the parent and children are well
> behaved and have a reasonable and unchanging VM size.
> 
> Usually this does not actually alleviate the memory pressure because the
> truly bad process is completely unrelated; and the OOM killer must later
> kill the truly bad process.
> 
> This trivial patch corrects the calculation so that it does not count a
> child's shared memory against the parent.
> 

Hi, Jeff,

1. grep on the kernel source tells me that shared_vm is incremented only in
   vm_stat_account(), which is a NO-OP if CONFIG_PROC_FS is not defined.
2. How have you tested these patches? One way to do it would be to use the
   memory controller and set a small limit on the control group. A memory
   intensive application will soon see an OOM.

I do need to look at OOM kill sanity, my colleagues using the memory controller
have reported wrong actions taken by the OOM killer, but I am yet to analyze them.

The interesting thing is the use of total_vm and not the RSS which is used as
the basis by the OOM killer. I need to read/understand the code a bit more.

-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL