Re: [patch 2/2] mm: memcontrol: default hierarchy interface for memory

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



I have overlooked the `none' setting...

On Thu 08-01-15 23:15:04, Johannes Weiner wrote:
[...]
> +static int memory_low_show(struct seq_file *m, void *v)
> +{
> +	struct mem_cgroup *memcg = mem_cgroup_from_css(seq_css(m));
> +	unsigned long low = ACCESS_ONCE(memcg->low);
> +
> +	if (low == 0)
> +		seq_printf(m, "none\n");
> +	else
> +		seq_printf(m, "%llu\n", (u64)low * PAGE_SIZE);
> +
> +	return 0;
> +}

This is really confusing. What if somebody wants to protect a group
from being reclaimed? One possible and natural way would by copying
memory.max value but then `none' means something else completely.

Besides that why to call 0, which has a clear meaning, any other name?

Now that I think about the naming `none' doesn't sound that great for
max resp. high either. If for nothing else then for the above copy
example (who knows what shows up later). Sure, a huge number is bad
as well for reasons you have mentioned in other email. `resource_max'
sounds like a better fit to me. But I am lame at naming.
-- 
Michal Hocko
SUSE Labs

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



[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]