On Sat, Jun 5, 2021 at 4:32 PM Matthew Wilcox <willy@xxxxxxxxxxxxx> wrote:
>
> On Sat, Jun 05, 2021 at 12:59:50AM +0200, Matteo Croce wrote:
> > On Fri, Jun 4, 2021 at 9:08 PM Matthew Wilcox <willy@xxxxxxxxxxxxx> wrote:
> > >
> > > On Fri, Jun 04, 2021 at 08:33:45PM +0200, Matteo Croce wrote:
> > > > @@ -130,7 +137,10 @@ struct page {
> > > > };
> > > > };
> > > > struct { /* Tail pages of compound page */
> > > > - unsigned long compound_head; /* Bit zero is set */
> > > > + /* Bit zero is set
> > > > + * Bit one if pfmemalloc page
> > > > + */
> > > > + unsigned long compound_head;
> > >
> > > I would drop this hunk. Bit 1 is not used for this purpose in tail
> > > pages; it's used for that purpose in head and base pages.
> > >
> > > I suppose we could do something like ...
> > >
> > > static inline void set_page_pfmemalloc(struct page *page)
> > > {
> > > - page->index = -1UL;
> > > + page->lru.next = (void *)2;
> > > }
> > >
> > > if it's causing confusion.
> > >
> >
And change all the *_pfmemalloc functions to use page->lru.next like this?
@@ -1668,10 +1668,12 @@ struct address_space *page_mapping(struct page *page);
static inline bool page_is_pfmemalloc(const struct page *page)
{
/*
- * Page index cannot be this large so this must be
- * a pfmemalloc page.
+ * This is not a tail page; compound_head of a head page is unused
+ * at return from the page allocator, and will be overwritten
+ * by callers who do not care whether the page came from the
+ * reserves.
*/
- return page->index == -1UL;
+ return (uintptr_t)page->lru.next & BIT(1);
}
/*
@@ -1680,12 +1682,12 @@ static inline bool page_is_pfmemalloc(const
struct page *page)
*/
static inline void set_page_pfmemalloc(struct page *page)
{
- page->index = -1UL;
+ page->lru.next = (void *)BIT(1);
}
static inline void clear_page_pfmemalloc(struct page *page)
{
- page->index = 0;
+ page->lru.next = NULL;
}
--
per aspera ad upstream
