Merge branch 'dynamic_sg' into rdma.git for-next

From Maor Gottlieb says:

====================
This series extends __sg_alloc_table_from_pages to allow chaining of new
pages to an already initialized SG table.

This allows for drivers to utilize the optimization of merging contiguous
pages without a need to pre allocate all the pages and hold them in a very
large temporary buffer prior to the call to SG table initialization.

The last patch changes the Infiniband core to use the new API. It removes
duplicate functionality from the code and benefits from the optimization
of allocating dynamic SG table from pages.

In huge pages system of 2MB page size, without this change, the SG table
would contain x512 SG entries.
====================

* branch 'dynamic_sg':
  RDMA/umem: Move to allocate SG table from pages
  lib/scatterlist: Add support in dynamic allocation of SG table from pages
  tools/testing/scatterlist: Show errors in human readable form
  tools/testing/scatterlist: Rejuvenate bit-rotten test

1 files changed, 277 insertions, 88 deletions

diff --git a/mm/memory.c b/mm/memory.c
index 469af373ae76..fcfc4ca36eba 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -695,84 +695,218 @@ out:
  * covered by this vma.
  */
 
-static inline unsigned long
-copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+static unsigned long
+copy_nonpresent_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		pte_t *dst_pte, pte_t *src_pte, struct vm_area_struct *vma,
 		unsigned long addr, int *rss)
 {
 	unsigned long vm_flags = vma->vm_flags;
 	pte_t pte = *src_pte;
 	struct page *page;
+	swp_entry_t entry = pte_to_swp_entry(pte);
+
+	if (likely(!non_swap_entry(entry))) {
+		if (swap_duplicate(entry) < 0)
+			return entry.val;
+
+		/* make sure dst_mm is on swapoff's mmlist. */
+		if (unlikely(list_empty(&dst_mm->mmlist))) {
+			spin_lock(&mmlist_lock);
+			if (list_empty(&dst_mm->mmlist))
+				list_add(&dst_mm->mmlist,
+						&src_mm->mmlist);
+			spin_unlock(&mmlist_lock);
+		}
+		rss[MM_SWAPENTS]++;
+	} else if (is_migration_entry(entry)) {
+		page = migration_entry_to_page(entry);
 
-	/* pte contains position in swap or file, so copy. */
-	if (unlikely(!pte_present(pte))) {
-		swp_entry_t entry = pte_to_swp_entry(pte);
-
-		if (likely(!non_swap_entry(entry))) {
-			if (swap_duplicate(entry) < 0)
-				return entry.val;
-
-			/* make sure dst_mm is on swapoff's mmlist. */
-			if (unlikely(list_empty(&dst_mm->mmlist))) {
-				spin_lock(&mmlist_lock);
-				if (list_empty(&dst_mm->mmlist))
-					list_add(&dst_mm->mmlist,
-							&src_mm->mmlist);
-				spin_unlock(&mmlist_lock);
-			}
-			rss[MM_SWAPENTS]++;
-		} else if (is_migration_entry(entry)) {
-			page = migration_entry_to_page(entry);
-
-			rss[mm_counter(page)]++;
-
-			if (is_write_migration_entry(entry) &&
-					is_cow_mapping(vm_flags)) {
-				/*
-				 * COW mappings require pages in both
-				 * parent and child to be set to read.
-				 */
-				make_migration_entry_read(&entry);
-				pte = swp_entry_to_pte(entry);
-				if (pte_swp_soft_dirty(*src_pte))
-					pte = pte_swp_mksoft_dirty(pte);
-				if (pte_swp_uffd_wp(*src_pte))
-					pte = pte_swp_mkuffd_wp(pte);
-				set_pte_at(src_mm, addr, src_pte, pte);
-			}
-		} else if (is_device_private_entry(entry)) {
-			page = device_private_entry_to_page(entry);
+		rss[mm_counter(page)]++;
 
+		if (is_write_migration_entry(entry) &&
+				is_cow_mapping(vm_flags)) {
 			/*
-			 * Update rss count even for unaddressable pages, as
-			 * they should treated just like normal pages in this
-			 * respect.
-			 *
-			 * We will likely want to have some new rss counters
-			 * for unaddressable pages, at some point. But for now
-			 * keep things as they are.
+			 * COW mappings require pages in both
+			 * parent and child to be set to read.
 			 */
-			get_page(page);
-			rss[mm_counter(page)]++;
-			page_dup_rmap(page, false);
+			make_migration_entry_read(&entry);
+			pte = swp_entry_to_pte(entry);
+			if (pte_swp_soft_dirty(*src_pte))
+				pte = pte_swp_mksoft_dirty(pte);
+			if (pte_swp_uffd_wp(*src_pte))
+				pte = pte_swp_mkuffd_wp(pte);
+			set_pte_at(src_mm, addr, src_pte, pte);
+		}
+	} else if (is_device_private_entry(entry)) {
+		page = device_private_entry_to_page(entry);
 
-			/*
-			 * We do not preserve soft-dirty information, because so
-			 * far, checkpoint/restore is the only feature that
-			 * requires that. And checkpoint/restore does not work
-			 * when a device driver is involved (you cannot easily
-			 * save and restore device driver state).
-			 */
-			if (is_write_device_private_entry(entry) &&
-			    is_cow_mapping(vm_flags)) {
-				make_device_private_entry_read(&entry);
-				pte = swp_entry_to_pte(entry);
-				if (pte_swp_uffd_wp(*src_pte))
-					pte = pte_swp_mkuffd_wp(pte);
-				set_pte_at(src_mm, addr, src_pte, pte);
-			}
+		/*
+		 * Update rss count even for unaddressable pages, as
+		 * they should treated just like normal pages in this
+		 * respect.
+		 *
+		 * We will likely want to have some new rss counters
+		 * for unaddressable pages, at some point. But for now
+		 * keep things as they are.
+		 */
+		get_page(page);
+		rss[mm_counter(page)]++;
+		page_dup_rmap(page, false);
+
+		/*
+		 * We do not preserve soft-dirty information, because so
+		 * far, checkpoint/restore is the only feature that
+		 * requires that. And checkpoint/restore does not work
+		 * when a device driver is involved (you cannot easily
+		 * save and restore device driver state).
+		 */
+		if (is_write_device_private_entry(entry) &&
+		    is_cow_mapping(vm_flags)) {
+			make_device_private_entry_read(&entry);
+			pte = swp_entry_to_pte(entry);
+			if (pte_swp_uffd_wp(*src_pte))
+				pte = pte_swp_mkuffd_wp(pte);
+			set_pte_at(src_mm, addr, src_pte, pte);
 		}
-		goto out_set_pte;
+	}
+	set_pte_at(dst_mm, addr, dst_pte, pte);
+	return 0;
+}
+
+/*
+ * Copy a present and normal page if necessary.
+ *
+ * NOTE! The usual case is that this doesn't need to do
+ * anything, and can just return a positive value. That
+ * will let the caller know that it can just increase
+ * the page refcount and re-use the pte the traditional
+ * way.
+ *
+ * But _if_ we need to copy it because it needs to be
+ * pinned in the parent (and the child should get its own
+ * copy rather than just a reference to the same page),
+ * we'll do that here and return zero to let the caller
+ * know we're done.
+ *
+ * And if we need a pre-allocated page but don't yet have
+ * one, return a negative error to let the preallocation
+ * code know so that it can do so outside the page table
+ * lock.
+ */
+static inline int
+copy_present_page(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+		pte_t *dst_pte, pte_t *src_pte,
+		struct vm_area_struct *vma, struct vm_area_struct *new,
+		unsigned long addr, int *rss, struct page **prealloc,
+		pte_t pte, struct page *page)
+{
+	struct page *new_page;
+
+	if (!is_cow_mapping(vma->vm_flags))
+		return 1;
+
+	/*
+	 * The trick starts.
+	 *
+	 * What we want to do is to check whether this page may
+	 * have been pinned by the parent process.  If so,
+	 * instead of wrprotect the pte on both sides, we copy
+	 * the page immediately so that we'll always guarantee
+	 * the pinned page won't be randomly replaced in the
+	 * future.
+	 *
+	 * To achieve this, we do the following:
+	 *
+	 * 1. Write-protect the pte if it's writable.  This is
+	 *    to protect concurrent write fast-gup with
+	 *    FOLL_PIN, so that we'll fail the fast-gup with
+	 *    the write bit removed.
+	 *
+	 * 2. Check page_maybe_dma_pinned() to see whether this
+	 *    page may have been pinned.
+	 *
+	 * The order of these steps is important to serialize
+	 * against the fast-gup code (gup_pte_range()) on the
+	 * pte check and try_grab_compound_head(), so that
+	 * we'll make sure either we'll capture that fast-gup
+	 * so we'll copy the pinned page here, or we'll fail
+	 * that fast-gup.
+	 *
+	 * NOTE! Even if we don't end up copying the page,
+	 * we won't undo this wrprotect(), because the normal
+	 * reference copy will need it anyway.
+	 */
+	if (pte_write(pte))
+		ptep_set_wrprotect(src_mm, addr, src_pte);
+
+	/*
+	 * These are the "normally we can just copy by reference"
+	 * checks.
+	 */
+	if (likely(!atomic_read(&src_mm->has_pinned)))
+		return 1;
+	if (likely(!page_maybe_dma_pinned(page)))
+		return 1;
+
+	/*
+	 * Uhhuh. It looks like the page might be a pinned page,
+	 * and we actually need to copy it. Now we can set the
+	 * source pte back to being writable.
+	 */
+	if (pte_write(pte))
+		set_pte_at(src_mm, addr, src_pte, pte);
+
+	new_page = *prealloc;
+	if (!new_page)
+		return -EAGAIN;
+
+	/*
+	 * We have a prealloc page, all good!  Take it
+	 * over and copy the page & arm it.
+	 */
+	*prealloc = NULL;
+	copy_user_highpage(new_page, page, addr, vma);
+	__SetPageUptodate(new_page);
+	page_add_new_anon_rmap(new_page, new, addr, false);
+	lru_cache_add_inactive_or_unevictable(new_page, new);
+	rss[mm_counter(new_page)]++;
+
+	/* All done, just insert the new page copy in the child */
+	pte = mk_pte(new_page, new->vm_page_prot);
+	pte = maybe_mkwrite(pte_mkdirty(pte), new);
+	set_pte_at(dst_mm, addr, dst_pte, pte);
+	return 0;
+}
+
+/*
+ * Copy one pte.  Returns 0 if succeeded, or -EAGAIN if one preallocated page
+ * is required to copy this pte.
+ */
+static inline int
+copy_present_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+		pte_t *dst_pte, pte_t *src_pte, struct vm_area_struct *vma,
+		struct vm_area_struct *new,
+		unsigned long addr, int *rss, struct page **prealloc)
+{
+	unsigned long vm_flags = vma->vm_flags;
+	pte_t pte = *src_pte;
+	struct page *page;
+
+	page = vm_normal_page(vma, addr, pte);
+	if (page) {
+		int retval;
+
+		retval = copy_present_page(dst_mm, src_mm,
+			dst_pte, src_pte,
+			vma, new,
+			addr, rss, prealloc,
+			pte, page);
+		if (retval <= 0)
+			return retval;
+
+		get_page(page);
+		page_dup_rmap(page, false);
+		rss[mm_counter(page)]++;
 	}
 
 	/*
@@ -800,35 +934,51 @@ copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	if (!(vm_flags & VM_UFFD_WP))
 		pte = pte_clear_uffd_wp(pte);
 
-	page = vm_normal_page(vma, addr, pte);
-	if (page) {
-		get_page(page);
-		page_dup_rmap(page, false);
-		rss[mm_counter(page)]++;
-	}
-
-out_set_pte:
 	set_pte_at(dst_mm, addr, dst_pte, pte);
 	return 0;
 }
 
+static inline struct page *
+page_copy_prealloc(struct mm_struct *src_mm, struct vm_area_struct *vma,
+		   unsigned long addr)
+{
+	struct page *new_page;
+
+	new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, addr);
+	if (!new_page)
+		return NULL;
+
+	if (mem_cgroup_charge(new_page, src_mm, GFP_KERNEL)) {
+		put_page(new_page);
+		return NULL;
+	}
+	cgroup_throttle_swaprate(new_page, GFP_KERNEL);
+
+	return new_page;
+}
+
 static int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		   pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma,
+		   struct vm_area_struct *new,
 		   unsigned long addr, unsigned long end)
 {
 	pte_t *orig_src_pte, *orig_dst_pte;
 	pte_t *src_pte, *dst_pte;
 	spinlock_t *src_ptl, *dst_ptl;
-	int progress = 0;
+	int progress, ret = 0;
 	int rss[NR_MM_COUNTERS];
 	swp_entry_t entry = (swp_entry_t){0};
+	struct page *prealloc = NULL;
 
 again:
+	progress = 0;
 	init_rss_vec(rss);
 
 	dst_pte = pte_alloc_map_lock(dst_mm, dst_pmd, addr, &dst_ptl);
-	if (!dst_pte)
-		return -ENOMEM;
+	if (!dst_pte) {
+		ret = -ENOMEM;
+		goto out;
+	}
 	src_pte = pte_offset_map(src_pmd, addr);
 	src_ptl = pte_lockptr(src_mm, src_pmd);
 	spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
@@ -851,10 +1001,34 @@ again:
 			progress++;
 			continue;
 		}
-		entry.val = copy_one_pte(dst_mm, src_mm, dst_pte, src_pte,
+		if (unlikely(!pte_present(*src_pte))) {
+			entry.val = copy_nonpresent_pte(dst_mm, src_mm,
+							dst_pte, src_pte,
 							vma, addr, rss);
-		if (entry.val)
+			if (entry.val)
+				break;
+			progress += 8;
+			continue;
+		}
+		/* copy_present_pte() will clear `*prealloc' if consumed */
+		ret = copy_present_pte(dst_mm, src_mm, dst_pte, src_pte,
+				       vma, new, addr, rss, &prealloc);
+		/*
+		 * If we need a pre-allocated page for this pte, drop the
+		 * locks, allocate, and try again.
+		 */
+		if (unlikely(ret == -EAGAIN))
 			break;
+		if (unlikely(prealloc)) {
+			/*
+			 * pre-alloc page cannot be reused by next time so as
+			 * to strictly follow mempolicy (e.g., alloc_page_vma()
+			 * will allocate page according to address).  This
+			 * could only happen if one pinned pte changed.
+			 */
+			put_page(prealloc);
+			prealloc = NULL;
+		}
 		progress += 8;
 	} while (dst_pte++, src_pte++, addr += PAGE_SIZE, addr != end);
 
@@ -866,17 +1040,30 @@ again:
 	cond_resched();
 
 	if (entry.val) {
-		if (add_swap_count_continuation(entry, GFP_KERNEL) < 0)
+		if (add_swap_count_continuation(entry, GFP_KERNEL) < 0) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		entry.val = 0;
+	} else if (ret) {
+		WARN_ON_ONCE(ret != -EAGAIN);
+		prealloc = page_copy_prealloc(src_mm, vma, addr);
+		if (!prealloc)
 			return -ENOMEM;
-		progress = 0;
+		/* We've captured and resolved the error. Reset, try again. */
+		ret = 0;
 	}
 	if (addr != end)
 		goto again;
-	return 0;
+out:
+	if (unlikely(prealloc))
+		put_page(prealloc);
+	return ret;
 }
 
 static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		pud_t *dst_pud, pud_t *src_pud, struct vm_area_struct *vma,
+		struct vm_area_struct *new,
 		unsigned long addr, unsigned long end)
 {
 	pmd_t *src_pmd, *dst_pmd;
@@ -903,7 +1090,7 @@ static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src
 		if (pmd_none_or_clear_bad(src_pmd))
 			continue;
 		if (copy_pte_range(dst_mm, src_mm, dst_pmd, src_pmd,
-						vma, addr, next))
+				   vma, new, addr, next))
 			return -ENOMEM;
 	} while (dst_pmd++, src_pmd++, addr = next, addr != end);
 	return 0;
@@ -911,6 +1098,7 @@ static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src
 
 static inline int copy_pud_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		p4d_t *dst_p4d, p4d_t *src_p4d, struct vm_area_struct *vma,
+		struct vm_area_struct *new,
 		unsigned long addr, unsigned long end)
 {
 	pud_t *src_pud, *dst_pud;
@@ -937,7 +1125,7 @@ static inline int copy_pud_range(struct mm_struct *dst_mm, struct mm_struct *src
 		if (pud_none_or_clear_bad(src_pud))
 			continue;
 		if (copy_pmd_range(dst_mm, src_mm, dst_pud, src_pud,
-						vma, addr, next))
+				   vma, new, addr, next))
 			return -ENOMEM;
 	} while (dst_pud++, src_pud++, addr = next, addr != end);
 	return 0;
@@ -945,6 +1133,7 @@ static inline int copy_pud_range(struct mm_struct *dst_mm, struct mm_struct *src
 
 static inline int copy_p4d_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		pgd_t *dst_pgd, pgd_t *src_pgd, struct vm_area_struct *vma,
+		struct vm_area_struct *new,
 		unsigned long addr, unsigned long end)
 {
 	p4d_t *src_p4d, *dst_p4d;
@@ -959,14 +1148,14 @@ static inline int copy_p4d_range(struct mm_struct *dst_mm, struct mm_struct *src
 		if (p4d_none_or_clear_bad(src_p4d))
 			continue;
 		if (copy_pud_range(dst_mm, src_mm, dst_p4d, src_p4d,
-						vma, addr, next))
+				   vma, new, addr, next))
 			return -ENOMEM;
 	} while (dst_p4d++, src_p4d++, addr = next, addr != end);
 	return 0;
 }
 
 int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
-		struct vm_area_struct *vma)
+		    struct vm_area_struct *vma, struct vm_area_struct *new)
 {
 	pgd_t *src_pgd, *dst_pgd;
 	unsigned long next;
@@ -1021,7 +1210,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		if (pgd_none_or_clear_bad(src_pgd))
 			continue;
 		if (unlikely(copy_p4d_range(dst_mm, src_mm, dst_pgd, src_pgd,
-					    vma, addr, next))) {
+					    vma, new, addr, next))) {
 			ret = -ENOMEM;
 			break;
 		}
@@ -2955,8 +3144,8 @@ static vm_fault_t do_wp_page(struct vm_fault *vmf)
 		 * page count reference, and the page is locked,
 		 * it's dark out, and we're wearing sunglasses. Hit it.
 		 */
-		wp_page_reuse(vmf);
 		unlock_page(page);
+		wp_page_reuse(vmf);
 		return VM_FAULT_WRITE;
 	} else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
 					(VM_WRITE|VM_SHARED))) {