shmem: replace page if mapping excludes its zone

The GMA500 GPU driver uses GEM shmem objects, but with a new twist: the
backing RAM has to be below 4GB.  Not a problem while the boards
supported only 4GB: but now Intel's D2700MUD boards support 8GB, and
their GMA3600 is managed by the GMA500 driver.

shmem/tmpfs has never pretended to support hardware restrictions on the
backing memory, but it might have appeared to do so before v3.1, and
even now it works fine until a page is swapped out then back in.  When
read_cache_page_gfp() supplied a freshly allocated page for copy, that
compensated for whatever choice might have been made by earlier swapin
readahead; but swapoff was likely to destroy the illusion.

We'd like to continue to support GMA500, so now add a new
shmem_should_replace_page() check on the zone when about to move a page
from swapcache to filecache (in swapin and swapoff cases), with
shmem_replace_page() to allocate and substitute a suitable page (given
gma500/gem.c's mapping_set_gfp_mask GFP_KERNEL | __GFP_DMA32).

This does involve a minor extension to mem_cgroup_replace_page_cache()
(the page may or may not have already been charged); and I've removed a
comment and call to mem_cgroup_uncharge_cache_page(), which in fact is
always a no-op while PageSwapCache.

Also removed optimization of an unlikely path in shmem_getpage_gfp(),
now that we need to check PageSwapCache more carefully (a racing caller
might already have made the copy).  And at one point shmem_unuse_inode()
needs to use the hitherto private page_swapcount(), to guard against
racing with inode eviction.

It would make sense to extend shmem_should_replace_page(), to cover
cpuset and NUMA mempolicy restrictions too, but set that aside for now:
needs a cleanup of shmem mempolicy handling, and more testing, and ought
to handle swap faults in do_swap_page() as well as shmem.

Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Christoph Hellwig <hch@infradead.org>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Stephane Marchesin <marcheu@chromium.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Airlie <airlied@gmail.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

4 files changed, 142 insertions, 24 deletions

diff --git a/include/linux/swap.h b/include/linux/swap.h
index bc3073ce95cc..d965c4bfab3a 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -351,6 +351,7 @@ extern int swap_type_of(dev_t, sector_t, struct block_device **);
 extern unsigned int count_swap_pages(int, int);
 extern sector_t map_swap_page(struct page *, struct block_device **);
 extern sector_t swapdev_block(int, pgoff_t);
+extern int page_swapcount(struct page *);
 extern int reuse_swap_page(struct page *);
 extern int try_to_free_swap(struct page *);
 struct backing_dev_info;
@@ -445,6 +446,11 @@ static inline void delete_from_swap_cache(struct page *page)
 {
 }
 
+static inline int page_swapcount(struct page *page)
+{
+        return 0;
+}
+
 #define reuse_swap_page(page)   (page_mapcount(page) == 1)
 
 static inline int try_to_free_swap(struct page *page)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 4f71219cc53e..d7ce417cae7c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3373,7 +3373,7 @@ void mem_cgroup_end_migration(struct mem_cgroup *memcg,
 void mem_cgroup_replace_page_cache(struct page *oldpage,
                                   struct page *newpage)
 {
-        struct mem_cgroup *memcg;
+        struct mem_cgroup *memcg = NULL;
         struct page_cgroup *pc;
         enum charge_type type = MEM_CGROUP_CHARGE_TYPE_CACHE;
 
@@ -3383,11 +3383,20 @@ void mem_cgroup_replace_page_cache(struct page *oldpage,
         pc = lookup_page_cgroup(oldpage);
         /* fix accounting on old pages */
         lock_page_cgroup(pc);
-        memcg = pc->mem_cgroup;
-        mem_cgroup_charge_statistics(memcg, false, -1);
-        ClearPageCgroupUsed(pc);
+        if (PageCgroupUsed(pc)) {
+                memcg = pc->mem_cgroup;
+                mem_cgroup_charge_statistics(memcg, false, -1);
+                ClearPageCgroupUsed(pc);
+        }
         unlock_page_cgroup(pc);
 
+        /*
+         * When called from shmem_replace_page(), in some cases the
+         * oldpage has already been charged, and in some cases not.
+         */
+        if (!memcg)
+                return;
+
         if (PageSwapBacked(oldpage))
                 type = MEM_CGROUP_CHARGE_TYPE_SHMEM;
 
diff --git a/mm/shmem.c b/mm/shmem.c
index be5af34a070d..db72d8e44ec6 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -103,6 +103,9 @@ static unsigned long shmem_default_max_inodes(void)
 }
 #endif
 
+static bool shmem_should_replace_page(struct page *page, gfp_t gfp);
+static int shmem_replace_page(struct page **pagep, gfp_t gfp,
+                                struct shmem_inode_info *info, pgoff_t index);
 static int shmem_getpage_gfp(struct inode *inode, pgoff_t index,
         struct page **pagep, enum sgp_type sgp, gfp_t gfp, int *fault_type);
 
@@ -604,12 +607,13 @@ static void shmem_evict_inode(struct inode *inode)
  * If swap found in inode, free it and move page from swapcache to filecache.
  */
 static int shmem_unuse_inode(struct shmem_inode_info *info,
-                             swp_entry_t swap, struct page *page)
+                             swp_entry_t swap, struct page **pagep)
 {
         struct address_space *mapping = info->vfs_inode.i_mapping;
         void *radswap;
         pgoff_t index;
-        int error;
+        gfp_t gfp;
+        int error = 0;
 
         radswap = swp_to_radix_entry(swap);
         index = radix_tree_locate_item(&mapping->page_tree, radswap);
@@ -625,22 +629,37 @@ static int shmem_unuse_inode(struct shmem_inode_info *info,
         if (shmem_swaplist.next != &info->swaplist)
                 list_move_tail(&shmem_swaplist, &info->swaplist);
 
+        gfp = mapping_gfp_mask(mapping);
+        if (shmem_should_replace_page(*pagep, gfp)) {
+                mutex_unlock(&shmem_swaplist_mutex);
+                error = shmem_replace_page(pagep, gfp, info, index);
+                mutex_lock(&shmem_swaplist_mutex);
+                /*
+                 * We needed to drop mutex to make that restrictive page
+                 * allocation; but the inode might already be freed by now,
+                 * and we cannot refer to inode or mapping or info to check.
+                 * However, we do hold page lock on the PageSwapCache page,
+                 * so can check if that still has our reference remaining.
+                 */
+                if (!page_swapcount(*pagep))
+                        error = -ENOENT;
+        }
+
         /*
          * We rely on shmem_swaplist_mutex, not only to protect the swaplist,
          * but also to hold up shmem_evict_inode(): so inode cannot be freed
          * beneath us (pagelock doesn't help until the page is in pagecache).
          */
-        error = shmem_add_to_page_cache(page, mapping, index,
+        if (!error)
+                error = shmem_add_to_page_cache(*pagep, mapping, index,
                                                 GFP_NOWAIT, radswap);
-        /* which does mem_cgroup_uncharge_cache_page on error */
-
         if (error != -ENOMEM) {
                 /*
                  * Truncation and eviction use free_swap_and_cache(), which
                  * only does trylock page: if we raced, best clean up here.
                  */
-                delete_from_swap_cache(page);
-                set_page_dirty(page);
+                delete_from_swap_cache(*pagep);
+                set_page_dirty(*pagep);
                 if (!error) {
                         spin_lock(&info->lock);
                         info->swapped--;
@@ -660,7 +679,14 @@ int shmem_unuse(swp_entry_t swap, struct page *page)
         struct list_head *this, *next;
         struct shmem_inode_info *info;
         int found = 0;
-        int error;
+        int error = 0;
+
+        /*
+         * There's a faint possibility that swap page was replaced before
+         * caller locked it: it will come back later with the right page.
+         */
+        if (unlikely(!PageSwapCache(page)))
+                goto out;
 
         /*
          * Charge page using GFP_KERNEL while we can wait, before taking
@@ -676,7 +702,7 @@ int shmem_unuse(swp_entry_t swap, struct page *page)
         list_for_each_safe(this, next, &shmem_swaplist) {
                 info = list_entry(this, struct shmem_inode_info, swaplist);
                 if (info->swapped)
-                        found = shmem_unuse_inode(info, swap, page);
+                        found = shmem_unuse_inode(info, swap, &page);
                 else
                         list_del_init(&info->swaplist);
                 cond_resched();
@@ -685,8 +711,6 @@ int shmem_unuse(swp_entry_t swap, struct page *page)
         }
         mutex_unlock(&shmem_swaplist_mutex);
 
-        if (!found)
-                mem_cgroup_uncharge_cache_page(page);