[Devel] Re: [PATCH v3 3/6] expose fine-grained per-cpu data for cpuacct stats

Glauber Costa glommer at parallels.com
Wed May 30 06:26:12 PDT 2012


On 05/30/2012 04:48 PM, Paul Turner wrote:
>> >  Now, if you guys are okay with a file per-cpu, I can do it as well.
>> >  It pollutes the filesystem, but at least protects against the fact that this
>> >  is kmalloc-backed.
>> >
> As I prefaced, I'm not sure there's much that can be trivially done
> about it.  This is really a fundamental limitation of how read_map()
> works.
>
> What we really need is a proper seq_file exposed through cftypes.

Tejun, would you be okay with an interface that exports somehow the raw 
seq_file in cgroups ?

This way we could call s_show for each cpu and get away with the memory 
usage problem, I presume




More information about the Devel mailing list