On Wed, Sep 30, 2015 at 12:51:18PM -0700, Greg Thelen wrote:
>
> Vladimir Davydov wrote:
...
> > diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
> > index 416509e26d6d..a190719c2f46 100644
> > --- a/include/linux/page-flags.h
> > +++ b/include/linux/page-flags.h
> > @@ -594,6 +594,28 @@ static inline void __ClearPageBalloon(struct page *page)
> > }
> >
> > /*
> > + * PageKmem() returns true if the page was allocated with alloc_kmem_pages().
> > + */
> > +#define PAGE_KMEM_MAPCOUNT_VALUE (-512)
> > +
> > +static inline int PageKmem(struct page *page)
> > +{
> > + return atomic_read(&page->_mapcount) == PAGE_KMEM_MAPCOUNT_VALUE;
> > +}
> > +
> > +static inline void __SetPageKmem(struct page *page)
> > +{
> > + VM_BUG_ON_PAGE(atomic_read(&page->_mapcount) != -1, page);
> > + atomic_set(&page->_mapcount, PAGE_KMEM_MAPCOUNT_VALUE);
> > +}
>
> What do you think about several special mapcount values for various
> types of kmem?
>
> It's helps user and administrators break down memory usage.
>
> A nice equation is:
> memory.usage_in_bytes = memory.stat[file + anon + unevictable + kmem]
>
> Next, it's helpful to be able to breakdown kmem into:
> kmem = stack + pgtable + slab + ...
>
> On one hand (and the kernel I use internally) we can use separate per
> memcg counters for each kmem type. Then reconstitute memory.kmem as
> needed by adding them together. But using keeping a single kernel kmem
> counter is workable if there is a way to breakdown the memory charge to
> a container (e.g. by walking /proc/kpageflags-ish or per memcg
> memory.kpageflags-ish file).
I don't think that storing information about kmem type on the page
struct just to report it via /proc/kpageflags is a good idea, because
the number of unused bits left on the page struct is limited so we'd
better (ab)use them carefully, only when it's really difficult to get
along w/o them.
OTOH I do agree that some extra info showing what "kmem" is actually
used for could be helpful. To accumulate this info we can always use
per-cpu counters, which are pretty cheap and won't degrade performance,
and then report it via memory.stat. Furthermore, it will be more
convenient for administrators to read this info in human-readable format
than parsing /proc/kpageflags, which in addition takes long on systems
with a lot of RAM.
Thanks,
Vladimir
--
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>