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

[Dave Hansen <dave.hansen@linux.intel.com>]

On 12/20/2014 05:54 AM, Rafael Aquini wrote:
> This patch introduces 'pagesize' line element to /proc/<pid>/numa_maps
> report file in order to help disambiguating the size of pages that are
> backing memory areas mapped by a task. When the VMA backing page size
> is observed different from kernel's default PAGE_SIZE, the new element 
> is printed out to complement report output. This is specially useful to
> help differentiating between HUGE and GIGANTIC page VMAs.

Heh, I completely forgot about this.  Thanks for picking it back up.

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. :)

But that isn't the case for this patch.  The "anon=50 dirty=50 N0=50"
output of numa_maps is wholly *useless* without either this patch or
some other mechanism to find out of hugetbfs memory is present.  I think
that needs to make it in to the description.

I'm fine with the code, though.  Feel free to add my acked-by.

--
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>