+ mm-fix-ever-decreasing-swap-priority.patch added to -mm tree

+ mm-fix-ever-decreasing-swap-priority.patch added to -mm tree

[akpm]

The patch titled
     mm: fix ever-decreasing swap priority
has been added to the -mm tree.  Its filename is
     mm-fix-ever-decreasing-swap-priority.patch

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/SubmitChecklist when testing your code ***

See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this

The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/

------------------------------------------------------
Subject: mm: fix ever-decreasing swap priority
From: Hugh Dickins <hugh@veritas.com>

Vegard Nossum has noticed the ever-decreasing negative priority in a
swapon /swapoff loop, which eventually would misprioritize when int wraps
positive.  Not worth spending much code on, but probably better fixed.

It's easy to handle the swapping on and off of just one area, but there's
not much point if a pair or more still misbehave.  To handle the general
case, swapoff should compact negative priorities, keeping them always from
-1 to -MAX_SWAPFILES.  That's a change, but should cause no regression,
since these negative (unspecified) priorities are disjoint from the the
positive specified priorities 0 to 32767.

One small functional difference, which seems appropriate: when swapoff
fails to free all swap from a negative priority area, that area is now
reinserted at lowest priority, rather than at its original priority.

In moving down swapon's setting of priority, I notice that an area is
visible to /proc/swaps when it has swap_map set, yet that was being set
before all the visible fields were properly filled in: corrected.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reported-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/swapfile.c |   49 ++++++++++++++++++++++++------------------------
 1 file changed, 25 insertions(+), 24 deletions(-)

diff -puN mm/swapfile.c~mm-fix-ever-decreasing-swap-priority mm/swapfile.c
--- a/mm/swapfile.c~mm-fix-ever-decreasing-swap-priority
+++ a/mm/swapfile.c
@@ -37,6 +37,7 @@ DEFINE_SPINLOCK(swap_lock);
 unsigned int nr_swapfiles;
 long total_swap_pages;
 static int swap_overflow;
+static int least_priority;
 
 static const char Bad_file[] = "Bad swap file entry ";
 static const char Unused_file[] = "Unused swap file entry ";
@@ -1260,6 +1261,11 @@ asmlinkage long sys_swapoff(const char _
 		/* just pick something that's safe... */
 		swap_list.next = swap_list.head;
 	}
+	if (p->prio < 0) {
+		for (i = p->next; i >= 0; i = swap_info[i].next)
+			swap_info[i].prio = p->prio--;
+		least_priority++;
+	}
 	nr_swap_pages -= p->pages;
 	total_swap_pages -= p->pages;
 	p->flags &= ~SWP_WRITEOK;
@@ -1272,9 +1278,14 @@ asmlinkage long sys_swapoff(const char _
 	if (err) {
 		/* re-insert swap space back into swap_list */
 		spin_lock(&swap_lock);
-		for (prev = -1, i = swap_list.head; i >= 0; prev = i, i = swap_info[i].next)
+		if (p->prio < 0)
+			p->prio = --least_priority;
+		prev = -1;
+		for (i = swap_list.head; i >= 0; i = swap_info[i].next) {
 			if (p->prio >= swap_info[i].prio)
 				break;
+			prev = i;
+		}
 		p->next = i;
 		if (prev < 0)
 			swap_list.head = swap_list.next = p - swap_info;
@@ -1447,7 +1458,6 @@ asmlinkage long sys_swapon(const char __
 	unsigned int type;
 	int i, prev;
 	int error;
-	static int least_priority;
 	union swap_header *swap_header = NULL;
 	int swap_header_version;
 	unsigned int nr_good_pages = 0;
@@ -1455,7 +1465,7 @@ asmlinkage long sys_swapon(const char __
 	sector_t span;
 	unsigned long maxpages = 1;
 	int swapfilesize;
-	unsigned short *swap_map;
+	unsigned short *swap_map = NULL;
 	struct page *page = NULL;
 	struct inode *inode = NULL;
 	int did_down = 0;
@@ -1474,22 +1484,10 @@ asmlinkage long sys_swapon(const char __
 	}
 	if (type >= nr_swapfiles)
 		nr_swapfiles = type+1;
+	memset(p, 0, sizeof(*p));
 	INIT_LIST_HEAD(&p->extent_list);
 	p->flags = SWP_USED;
-	p->swap_file = NULL;
-	p->old_block_size = 0;
-	p->swap_map = NULL;
-	p->lowest_bit = 0;
-	p->highest_bit = 0;
-	p->cluster_nr = 0;
-	p->inuse_pages = 0;
 	p->next = -1;
-	if (swap_flags & SWAP_FLAG_PREFER) {
-		p->prio =
-		  (swap_flags & SWAP_FLAG_PRIO_MASK)>>SWAP_FLAG_PRIO_SHIFT;
-	} else {
-		p->prio = --least_priority;
-	}
 	spin_unlock(&swap_lock);
 	name = getname(specialfile);
 	error = PTR_ERR(name);
@@ -1632,19 +1630,20 @@ asmlinkage long sys_swapon(const char __
 			goto bad_swap;
 
 		/* OK, set up the swap map and apply the bad block list */
-		if (!(p->swap_map = vmalloc(maxpages * sizeof(short)))) {
+		swap_map = vmalloc(maxpages * sizeof(short));
+		if (!swap_map) {
 			error = -ENOMEM;
 			goto bad_swap;
 		}
 
 		error = 0;
-		memset(p->swap_map, 0, maxpages * sizeof(short));
+		memset(swap_map, 0, maxpages * sizeof(short));
 		for (i = 0; i < swap_header->info.nr_badpages; i++) {
 			int page_nr = swap_header->info.badpages[i];
 			if (page_nr <= 0 || page_nr >= swap_header->info.last_page)
 				error = -EINVAL;
 			else
-				p->swap_map[page_nr] = SWAP_MAP_BAD;
+				swap_map[page_nr] = SWAP_MAP_BAD;
 		}
 		nr_good_pages = swap_header->info.last_page -
 				swap_header->info.nr_badpages -
@@ -1654,7 +1653,7 @@ asmlinkage long sys_swapon(const char __
 	}
 
 	if (nr_good_pages) {
-		p->swap_map[0] = SWAP_MAP_BAD;
+		swap_map[0] = SWAP_MAP_BAD;
 		p->max = maxpages;
 		p->pages = nr_good_pages;
 		nr_extents = setup_swap_extents(p, &span);
@@ -1672,6 +1671,12 @@ asmlinkage long sys_swapon(const char __
 
 	mutex_lock(&swapon_mutex);
 	spin_lock(&swap_lock);
+	if (swap_flags & SWAP_FLAG_PREFER)
+		p->prio =
+		  (swap_flags & SWAP_FLAG_PRIO_MASK) >> SWAP_FLAG_PRIO_SHIFT;
+	else
+		p->prio = --least_priority;
+	p->swap_map = swap_map;
 	p->flags = SWP_ACTIVE;
 	nr_swap_pages += nr_good_pages;
 	total_swap_pages += nr_good_pages;
@@ -1707,12 +1712,8 @@ bad_swap:
 	destroy_swap_extents(p);
 bad_swap_2:
 	spin_lock(&swap_lock);
-	swap_map = p->swap_map;
 	p->swap_file = NULL;
-	p->swap_map = NULL;
 	p->flags = 0;
-	if (!(swap_flags & SWAP_FLAG_PREFER))
-		++least_priority;
 	spin_unlock(&swap_lock);
 	vfree(swap_map);
 	if (swap_file)
_

Patches currently in -mm which might be from hugh@veritas.com are

origin.patch
ahci-use-correct-type-for-cpu-flags.patch
git-unionfs.patch
unionfs-fix-memory-leak.patch
fsstack-fsstack_copy_inode_size-locking.patch
mm-remove-nopfn-fix.patch
access_process_vm-device-memory-infrastructure.patch
use-generic_access_phys-for-dev-mem-mappings.patch
use-generic_access_phys-for-dev-mem-mappings-fix.patch
use-generic_access_phys-for-pci-mmap-on-x86.patch
powerpc-ioremap_prot.patch
spufs-use-the-new-vm_ops-access.patch
spufs-use-the-new-vm_ops-access-fix.patch
mm-remove-double-indirection-on-tlb-parameter-to-free_pgd_range-co.patch
hugetlb-move-hugetlb_acct_memory.patch
hugetlb-reserve-huge-pages-for-reliable-map_private-hugetlbfs-mappings-until-fork.patch
hugetlb-guarantee-that-cow-faults-for-a-process-that-called-mmapmap_private-on-hugetlbfs-will-succeed.patch
hugetlb-guarantee-that-cow-faults-for-a-process-that-called-mmapmap_private-on-hugetlbfs-will-succeed-fix.patch
hugetlb-guarantee-that-cow-faults-for-a-process-that-called-mmapmap_private-on-hugetlbfs-will-succeed-build-fix.patch
hugetlb-guarantee-that-cow-faults-for-a-process-that-called-mmapmap_private-on-hugetlbfs-will-succeed-align-faulting-address-to-a-hugepage-boundary-before-unmapping.patch
huge-page-private-reservation-review-cleanups.patch
huge-page-private-reservation-review-cleanups-fix.patch
mm-record-map_noreserve-status-on-vmas-and-fix-small-page-mprotect-reservations.patch
hugetlb-move-reservation-region-support-earlier.patch
hugetlb-allow-huge-page-mappings-to-be-created-without-reservations.patch
hugetlb-allow-huge-page-mappings-to-be-created-without-reservations-cleanups.patch
hugetlb-reservations-fix-hugetlb-map_private-reservations-across-vma-splits-v2.patch
vma-page-offset-has-no-callees-drop-it.patch
generic_file_aio_read-cleanups.patch
tmpfs-support-aio.patch
hugetlb-fix-race-when-reading-proc-meminfo.patch
hugetlb-quota-is-not-freed-for-unused-reserved-private-huge-pages.patch
mmu-notifiers-add-list_del_init_rcu.patch
mmu-notifiers-add-mm_take_all_locks-operation.patch
mmu-notifiers-add-mm_take_all_locks-operation-checkpatch-fixes.patch
mmu-notifier-core.patch
mmu-notifier-core-fix.patch
mmu-notifier-core-fix-2.patch
mm-fix-ever-decreasing-swap-priority.patch
security-remove-unused-forwards.patch
exec-remove-some-includes.patch
exec-remove-some-includes-fix.patch
memcg-better-migration-handling.patch
memcg-remove-refcnt-from-page_cgroup.patch
memcg-remove-refcnt-from-page_cgroup-fix.patch
memcg-remove-refcnt-from-page_cgroup-fix-2.patch
memcg-remove-refcnt-from-page_cgroup-fix-memcg-fix-mem_cgroup_end_migration-race.patch
memcg-remove-refcnt-from-page_cgroup-memcg-fix-shmem_unuse_inode-charging.patch
memcg-helper-function-for-relcaim-from-shmem.patch
memcg-helper-function-for-relcaim-from-shmem-memcg-shmem_getpage-release-page-sooner.patch
memcg-helper-function-for-relcaim-from-shmem-memcg-mem_cgroup_shrink_usage-css_put.patch
memcg-add-hints-for-branch.patch
memcg-remove-a-redundant-check.patch
memcg-clean-up-checking-of-the-disabled-flag-memcg-further-checking-of-disabled-flag.patch
memrlimit-cgroup-mm-owner-callback-changes-to-add-task-info-memrlimit-fix-mmap_sem-deadlock.patch
memrlimit-cgroup-mm-owner-callback-changes-to-add-task-info-memrlimit-fix-sleep-inside-sleeplock-in-mm_update_next_owner.patch
memrlimit-add-memrlimit-controller-accounting-and-control-memrlimit-improve-fork-and-error-handling.patch
memrlimit-improve-error-handling.patch
memrlimit-improve-error-handling-update.patch
memrlimit-handle-attach_task-failure-add-can_attach-callback.patch
memrlimit-handle-attach_task-failure-add-can_attach-callback-update.patch
full-conversion-to-early_initcall-interface-remove-old-interface-fix-fix-fix.patch
mm-add-zap_vma_ptes-a-library-function-to-unmap-driver-ptes.patch
gru-driver-v3-resource-management-unmap-driver-ptes-gru.patch
x86-implement-pte_special-fix.patch
mm-readahead-scan-lockless.patch
radix-tree-add-gang_lookup_slot-gang_lookup_slot_tag.patch
mm-speculative-page-references.patch
mm-speculative-page-references-fix.patch
mm-speculative-page-references-fix-fix.patch
mm-speculative-page-references-fix-add_to_page_cache.patch
mm-lockless-pagecache.patch
mm-spinlock-tree_lock.patch
powerpc-implement-pte_special.patch
powerpc-implement-pte_special-update.patch
powerpc-lockless-get_user_pages.patch
define-page_file_cache-function-fix-splitlru-shmem_getpage-setpageswapbacked-sooner.patch
vmscan-split-lru-lists-into-anon-file-sets-splitlru-memcg-swapbacked-pages-active.patch
vmscan-split-lru-lists-into-anon-file-sets-splitlru-bdi_cap_swap_backed.patch
mlock-mlocked-pages-are-unevictable-fix-4.patch
introduce-__get_user_pages.patch
split-lru-munlock-rework.patch
revert-to-unevictable-lru-infrastructure-kconfig-fixpatch.patch
vmstat-mlocked-pages-statistics-fix-incorrect-mlocked-field-of-proc-meminfo.patch
vmscan-unevictable-lru-scan-sysctl-add-sys_device-parameter.patch
prio_tree-debugging-patch.patch