Reinstate ZERO_PAGE optimization in 'get_user_pages()' and fix XIP

KAMEZAWA Hiroyuki and Oleg Nesterov point out that since the commit
557ed1f ("remove ZERO_PAGE") removed
the ZERO_PAGE from the VM mappings, any users of get_user_pages() will
generally now populate the VM with real empty pages needlessly.

We used to get the ZERO_PAGE when we did the "handle_mm_fault()", but
since fault handling no longer uses ZERO_PAGE for new anonymous pages,
we now need to handle that special case in follow_page() instead.

In particular, the removal of ZERO_PAGE effectively removed the core
file writing optimization where we would skip writing pages that had not
been populated at all, and increased memory pressure a lot by allocating
all those useless newly zeroed pages.

This reinstates the optimization by making the unmapped PTE case the
same as for a non-existent page table, which already did this correctly.

While at it, this also fixes the XIP case for follow_page(), where the
caller could not differentiate between the case of a page that simply
could not be used (because it had no "struct page" associated with it)
and a page that just wasn't mapped.

We do that by simply returning an error pointer for pages that could not
be turned into a "struct page *".  The error is arbitrarily picked to be
EFAULT, since that was what get_user_pages() already used for the
equivalent IO-mapped page case.

[ Also removed an impossible test for pte_offset_map_lock() failing:
  that's not how that function works ]

Acked-by: Oleg Nesterov <[email protected]>
Acked-by: Nick Piggin <[email protected]>
Cc: KAMEZAWA Hiroyuki <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Roland McGrath <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

arch/powerpc/kernel/vdso.c

@@ -142,7 +142,7 @@ static void dump_one_vdso_page(struct page *pg, struct page *upg)
 	printk("kpg: %p (c:%d,f:%08lx)", __va(page_to_pfn(pg) << PAGE_SHIFT),
 	       page_count(pg),
 	       pg->flags);
-	if (upg/* && pg != upg*/) {
+	if (upg && !IS_ERR(upg) /* && pg != upg*/) {
 		printk(" upg: %p (c:%d,f:%08lx)", __va(page_to_pfn(upg)
 						       << PAGE_SHIFT),
 		       page_count(upg),

mm/memory.c

@@ -999,17 +999,15 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
 		goto no_page_table;
 
 	ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
-	if (!ptep)
-		goto out;
 
 	pte = *ptep;
 	if (!pte_present(pte))
-		goto unlock;
+		goto no_page;
 	if ((flags & FOLL_WRITE) && !pte_write(pte))
 		goto unlock;
 	page = vm_normal_page(vma, address, pte);
 	if (unlikely(!page))
-		goto unlock;
+		goto bad_page;
 
 	if (flags & FOLL_GET)
 		get_page(page);
@@ -1024,6 +1022,15 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
 out:
 	return page;
 
+bad_page:
+	pte_unmap_unlock(ptep, ptl);
+	return ERR_PTR(-EFAULT);
+
+no_page:
+	pte_unmap_unlock(ptep, ptl);
+	if (!pte_none(pte))
+		return page;
+	/* Fall through to ZERO_PAGE handling */
 no_page_table:
 	/*
 	 * When core dumping an enormous anonymous area that nobody
@@ -1159,6 +1166,8 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 
 				cond_resched();
 			}
+			if (IS_ERR(page))
+				return i ? i : PTR_ERR(page);
 			if (pages) {
 				pages[i] = page;
 

mm/migrate.c

@@ -865,6 +865,11 @@ static int do_move_pages(struct mm_struct *mm, struct page_to_node *pm,
 			goto set_status;
 
 		page = follow_page(vma, pp->addr, FOLL_GET);
+
+		err = PTR_ERR(page);
+		if (IS_ERR(page))
+			goto set_status;
+
 		err = -ENOENT;
 		if (!page)
 			goto set_status;
@@ -928,6 +933,11 @@ static int do_pages_stat(struct mm_struct *mm, struct page_to_node *pm)
 			goto set_status;
 
 		page = follow_page(vma, pm->addr, 0);
+
+		err = PTR_ERR(page);
+		if (IS_ERR(page))
+			goto set_status;
+
 		err = -ENOENT;
 		/* Use PageReserved to check for zero page */
 		if (!page || PageReserved(page))