Re: [patch v2 2/3] mm, oom: organize oom context into struct

[David Rientjes <rientjes@google.com>]

On Thu, 2 Jul 2015, Michal Hocko wrote:

> On Wed 01-07-15 14:37:14, David Rientjes wrote:
> > The force_kill member of struct oom_control isn't needed if an order of
> > -1 is used instead.  This is the same as order == -1 in
> > struct compact_control which requires full memory compaction.
> > 
> > This patch introduces no functional change.
> 
> But it obscures the code and I really dislike this change as pointed out
> previously.
> 

The oom killer is often called at the end of a very lengthy stack since 
memory allocation itself can be called deep in the stack.  Thus, reducing 
the amount of memory, even for a small lil bool, is helpful.  This is 
especially true when other such structs, struct compact_control, does the 
exact same thing by using order == -1 to mean explicit compaction.

I'm personally tired of fixing stack overflows and you're arguing against 
"obscurity" that even occurs in other parts of the mm code.  
oc->force_kill has no reason to exist, and thus it's removed in this patch 
and for good reason.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>