Re: [PATCH] proc: task_mmu: show page size in /proc/<pid>/numa_maps

[David Rientjes <rientjes@google.com>]

On Sat, 20 Dec 2014, Dave Hansen wrote:

> I sometimes wonder what 'numa_maps' purpose is any if we should have
> _some_ kind of policy about what goes in there vs. smaps.  numa_maps
> seems to be turning in to smaps, minus the \n. :)
> 

It seems like an interface, similar to the one proposed by Ulrich, that 
described the NUMA topology would have obsoleted numa_maps and you could 
only rely on smaps to determine locality.  There's existing userspace 
dependencies on numa_maps already, though, so owell.

I had to fix numa_maps output when autonuma was merged for a much more 
subtle difference at 
http://marc.info/?l=git-commits-head&m=139113691614467 so I know some 
people actually parse this and care quite a bit about its accuracy, it's a 
shame it can't be deprecated in favor of adding the necessary information 
to smaps.

--
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>