On Tue, Mar 2, 2021 at 2:11 AM Shakeel Butt <[email protected]> wrote:
>
> On Sun, Feb 28, 2021 at 10:25 PM Muchun Song <[email protected]> wrote:
> >
> > We want to reuse the obj_cgroup APIs to reparent the kmem pages when
> > the memcg offlined. If we do this, we should store an object cgroup
> > pointer to page->memcg_data for the kmem pages.
> >
> > Finally, page->memcg_data can have 3 different meanings.
> >
> > 1) For the slab pages, page->memcg_data points to an object cgroups
> > vector.
> >
> > 2) For the kmem pages (exclude the slab pages), page->memcg_data
> > points to an object cgroup.
> >
> > 3) For the user pages (e.g. the LRU pages), page->memcg_data points
> > to a memory cgroup.
> >
> > Currently we always get the memcg associated with a page via page_memcg
> > or page_memcg_rcu. page_memcg_check is special, it has to be used in
> > cases when it's not known if a page has an associated memory cgroup
> > pointer or an object cgroups vector. Because the page->memcg_data of
> > the kmem page is not pointing to a memory cgroup in the later patch,
> > the page_memcg and page_memcg_rcu cannot be applicable for the kmem
> > pages. In this patch, we introduce page_memcg_kmem to get the memcg
> > associated with the kmem pages. And make page_memcg and page_memcg_rcu
> > no longer apply to the kmem pages.
> >
> > In the end, there are 4 helpers to get the memcg associated with a
> > page. The usage is as follows.
> >
> > 1) Get the memory cgroup associated with a non-kmem page (e.g. the LRU
> > pages).
> >
> > - page_memcg()
> > - page_memcg_rcu()
>
> Can you rename these to page_memcg_lru[_rcu] to make them explicitly
> for LRU pages?
Yes. Will do. Thanks.
>
> >
> > 2) Get the memory cgroup associated with a kmem page (exclude the slab
> > pages).
> >
> > - page_memcg_kmem()
> >
> > 3) Get the memory cgroup associated with a page. It has to be used in
> > cases when it's not known if a page has an associated memory cgroup
> > pointer or an object cgroups vector. Returns NULL for slab pages or
> > uncharged pages, otherwise, returns memory cgroup for charged pages
> > (e.g. kmem pages, LRU pages).
> >
> > - page_memcg_check()
> >
> > In some place, we use page_memcg to check whether the page is charged.
> > Now we introduce page_memcg_charged helper to do this.
> >
> > This is a preparation for reparenting the kmem pages. To support reparent
> > kmem pages, we just need to adjust page_memcg_kmem and page_memcg_check in
> > the later patch.
> >
> > Signed-off-by: Muchun Song <[email protected]>
> > ---
> [snip]
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -855,10 +855,11 @@ void __mod_lruvec_page_state(struct page *page, enum node_stat_item idx,
> > int val)
> > {
> > struct page *head = compound_head(page); /* rmap on tail pages */
> > - struct mem_cgroup *memcg = page_memcg(head);
> > + struct mem_cgroup *memcg;
> > pg_data_t *pgdat = page_pgdat(page);
> > struct lruvec *lruvec;
> >
> > + memcg = PageMemcgKmem(head) ? page_memcg_kmem(head) : page_memcg(head);
>
> Should page_memcg_check() be used here?
Yeah. page_memcg_check() can be used here.
But on the inside of the page_memcg_check(),
there is a READ_ONCE(). Actually, we do not
need READ_ONCE() here. So I use page_memcg
or page_memcg_kmem directly. Thanks.
>
> > /* Untracked pages have no memcg, no lruvec. Update only the node */
> > if (!memcg) {
> > __mod_node_page_state(pgdat, idx, val);
> > @@ -3170,12 +3171,13 @@ int __memcg_kmem_charge_page(struct page *page, gfp_t gfp, int order)
> > */
> > void __memcg_kmem_uncharge_page(struct page *page, int order)
> > {
> > - struct mem_cgroup *memcg = page_memcg(page);
> > + struct mem_cgroup *memcg;
> > unsigned int nr_pages = 1 << order;
> >
> > - if (!memcg)
> > + if (!page_memcg_charged(page))
> > return;
> >
> > + memcg = page_memcg_kmem(page);
> > VM_BUG_ON_PAGE(mem_cgroup_is_root(memcg), page);
> > __memcg_kmem_uncharge(memcg, nr_pages);
> > page->memcg_data = 0;
> > @@ -6831,24 +6833,25 @@ static void uncharge_batch(const struct uncharge_gather *ug)
> > static void uncharge_page(struct page *page, struct uncharge_gather *ug)
> > {
> > unsigned long nr_pages;
> > + struct mem_cgroup *memcg;
> >
> > VM_BUG_ON_PAGE(PageLRU(page), page);
> >
> > - if (!page_memcg(page))
> > + if (!page_memcg_charged(page))
> > return;
> >
> > /*
> > * Nobody should be changing or seriously looking at
> > - * page_memcg(page) at this point, we have fully
> > - * exclusive access to the page.
> > + * page memcg at this point, we have fully exclusive
> > + * access to the page.
> > */
> > -
> > - if (ug->memcg != page_memcg(page)) {
> > + memcg = PageMemcgKmem(page) ? page_memcg_kmem(page) : page_memcg(page);
>
> Same, should page_memcg_check() be used here?
Same as above.
On Mon, Mar 1, 2021 at 7:03 PM Muchun Song <[email protected]> wrote:
>
> On Tue, Mar 2, 2021 at 2:11 AM Shakeel Butt <[email protected]> wrote:
> >
> > On Sun, Feb 28, 2021 at 10:25 PM Muchun Song <[email protected]> wrote:
> > >
> > > We want to reuse the obj_cgroup APIs to reparent the kmem pages when
> > > the memcg offlined. If we do this, we should store an object cgroup
> > > pointer to page->memcg_data for the kmem pages.
> > >
> > > Finally, page->memcg_data can have 3 different meanings.
> > >
> > > 1) For the slab pages, page->memcg_data points to an object cgroups
> > > vector.
> > >
> > > 2) For the kmem pages (exclude the slab pages), page->memcg_data
> > > points to an object cgroup.
> > >
> > > 3) For the user pages (e.g. the LRU pages), page->memcg_data points
> > > to a memory cgroup.
> > >
> > > Currently we always get the memcg associated with a page via page_memcg
> > > or page_memcg_rcu. page_memcg_check is special, it has to be used in
> > > cases when it's not known if a page has an associated memory cgroup
> > > pointer or an object cgroups vector. Because the page->memcg_data of
> > > the kmem page is not pointing to a memory cgroup in the later patch,
> > > the page_memcg and page_memcg_rcu cannot be applicable for the kmem
> > > pages. In this patch, we introduce page_memcg_kmem to get the memcg
> > > associated with the kmem pages. And make page_memcg and page_memcg_rcu
> > > no longer apply to the kmem pages.
> > >
> > > In the end, there are 4 helpers to get the memcg associated with a
> > > page. The usage is as follows.
> > >
> > > 1) Get the memory cgroup associated with a non-kmem page (e.g. the LRU
> > > pages).
> > >
> > > - page_memcg()
> > > - page_memcg_rcu()
> >
> > Can you rename these to page_memcg_lru[_rcu] to make them explicitly
> > for LRU pages?
>
> Yes. Will do. Thanks.
>
Please follow Johannes' suggestion regarding page_memcg_kmem() and
then no need to rename these.
On Tue, Mar 2, 2021 at 11:36 AM Shakeel Butt <[email protected]> wrote:
>
> On Mon, Mar 1, 2021 at 7:03 PM Muchun Song <[email protected]> wrote:
> >
> > On Tue, Mar 2, 2021 at 2:11 AM Shakeel Butt <[email protected]> wrote:
> > >
> > > On Sun, Feb 28, 2021 at 10:25 PM Muchun Song <[email protected]> wrote:
> > > >
> > > > We want to reuse the obj_cgroup APIs to reparent the kmem pages when
> > > > the memcg offlined. If we do this, we should store an object cgroup
> > > > pointer to page->memcg_data for the kmem pages.
> > > >
> > > > Finally, page->memcg_data can have 3 different meanings.
> > > >
> > > > 1) For the slab pages, page->memcg_data points to an object cgroups
> > > > vector.
> > > >
> > > > 2) For the kmem pages (exclude the slab pages), page->memcg_data
> > > > points to an object cgroup.
> > > >
> > > > 3) For the user pages (e.g. the LRU pages), page->memcg_data points
> > > > to a memory cgroup.
> > > >
> > > > Currently we always get the memcg associated with a page via page_memcg
> > > > or page_memcg_rcu. page_memcg_check is special, it has to be used in
> > > > cases when it's not known if a page has an associated memory cgroup
> > > > pointer or an object cgroups vector. Because the page->memcg_data of
> > > > the kmem page is not pointing to a memory cgroup in the later patch,
> > > > the page_memcg and page_memcg_rcu cannot be applicable for the kmem
> > > > pages. In this patch, we introduce page_memcg_kmem to get the memcg
> > > > associated with the kmem pages. And make page_memcg and page_memcg_rcu
> > > > no longer apply to the kmem pages.
> > > >
> > > > In the end, there are 4 helpers to get the memcg associated with a
> > > > page. The usage is as follows.
> > > >
> > > > 1) Get the memory cgroup associated with a non-kmem page (e.g. the LRU
> > > > pages).
> > > >
> > > > - page_memcg()
> > > > - page_memcg_rcu()
> > >
> > > Can you rename these to page_memcg_lru[_rcu] to make them explicitly
> > > for LRU pages?
> >
> > Yes. Will do. Thanks.
> >
>
> Please follow Johannes' suggestion regarding page_memcg_kmem() and
> then no need to rename these.
OK.