/**
* swapin_readahead - swap in pages in hope we need them soon
* @entry: swap entry of this memory
* @gfp_mask: memory allocation flags
* @vma: user vma this address belongs to
* @addr: target address for mempolicy
*
* Returns the struct page for entry and addr, after queueing swapin.
*
* Primitive swap readahead code. We simply read an aligned block of
* (1 << page_cluster) entries in the swap area. This method is chosen
* because it doesn't cost us any seek time. We also make sure to queue
* the 'original' request together with the readahead ones...
*
* This has been extended to use the NUMA policies from the mm triggering
* the readahead.
*
* Caller must hold down_read on the vma->vm_mm if vma is not NULL.
*/
struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma, unsigned long addr)
{
#ifdef CONFIG_SWAP_ENABLE_READAHEAD
struct page *page;
unsigned long offset = swp_offset(entry);
unsigned long start_offset, end_offset;
unsigned long mask = (1UL << page_cluster) - 1;
struct blk_plug plug;
/* Read a page_cluster sized and aligned cluster around offset. */
start_offset = offset & ~mask;
end_offset = offset | mask;
if (!start_offset) /* First page is swap header. */
start_offset++;
blk_start_plug(&plug);
for (offset = start_offset; offset <= end_offset ; offset++) {
/* Ok, do the async read-ahead now */
page = read_swap_cache_async(swp_entry(swp_type(entry), offset),
gfp_mask, vma, addr);
if (!page)
continue;
page_cache_release(page);
}
blk_finish_plug(&plug);
lru_add_drain(); /* Push any new pages onto the LRU now */
#endif /* CONFIG_SWAP_ENABLE_READAHEAD */
return read_swap_cache_async(entry, gfp_mask, vma, addr);
}
struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma, unsigned long addr)
{
struct page *page;
unsigned long offset = swp_offset(entry);
unsigned long start_offset, end_offset;
unsigned long mask = (1UL << page_cluster) - 1;
start_offset = offset & ~mask;
end_offset = offset | mask;
if (!start_offset)
start_offset++;
for (offset = start_offset; offset <= end_offset ; offset++) {
page = read_swap_cache_async(swp_entry(swp_type(entry), offset),
gfp_mask, vma, addr);
if (!page)
continue;
page_cache_release(page);
}
lru_add_drain();
return read_swap_cache_async(entry, gfp_mask, vma, addr);
}
/**
* swapin_readahead - swap in pages in hope we need them soon
* @entry: swap entry of this memory
* @gfp_mask: memory allocation flags
* @vma: user vma this address belongs to
* @addr: target address for mempolicy
*
* Returns the struct page for entry and addr, after queueing swapin.
*
* Primitive swap readahead code. We simply read an aligned block of
* (1 << page_cluster) entries in the swap area. This method is chosen
* because it doesn't cost us any seek time. We also make sure to queue
* the 'original' request together with the readahead ones...
*
* This has been extended to use the NUMA policies from the mm triggering
* the readahead.
*
* Caller must hold down_read on the vma->vm_mm if vma is not NULL.
*/
struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma, unsigned long addr)
{
struct page *page;
unsigned long offset = swp_offset(entry);
unsigned long start_offset, end_offset;
unsigned long mask = is_swap_fast(entry) ? 0 :
(1UL << page_cluster) - 1;
/* Read a page_cluster sized and aligned cluster around offset. */
start_offset = offset & ~mask;
end_offset = offset | mask;
if (!start_offset) /* First page is swap header. */
start_offset++;
for (offset = start_offset; offset <= end_offset ; offset++) {
/* Ok, do the async read-ahead now */
page = read_swap_cache_async(swp_entry(swp_type(entry), offset),
gfp_mask, vma, addr);
if (!page)
continue;
page_cache_release(page);
}
lru_add_drain(); /* Push any new pages onto the LRU now */
return read_swap_cache_async(entry, gfp_mask, vma, addr);
}
/**
* swapin_readahead - swap in pages in hope we need them soon
* @entry: swap entry of this memory
* @gfp_mask: memory allocation flags
* @vma: user vma this address belongs to
* @addr: target address for mempolicy
*
* Returns the struct page for entry and addr, after queueing swapin.
*
* Primitive swap readahead code. We simply read an aligned block of
* (1 << page_cluster) entries in the swap area. This method is chosen
* because it doesn't cost us any seek time. We also make sure to queue
* the 'original' request together with the readahead ones...
*
* This has been extended to use the NUMA policies from the mm triggering
* the readahead.
*
* Caller must hold down_read on the vma->vm_mm if vma is not NULL.
*/
struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma, unsigned long addr)
{
int nr_pages;
struct page *page;
unsigned long offset;
unsigned long end_offset;
/*
* Get starting offset for readaround, and number of pages to read.
* Adjust starting address by readbehind (for NUMA interleave case)?
* No, it's very unlikely that swap layout would follow vma layout,
* more likely that neighbouring swap pages came from the same node:
* so use the same "addr" to choose the same node for each swap read.
*/
nr_pages = valid_swaphandles(entry, &offset);
for (end_offset = offset + nr_pages; offset < end_offset; offset++) {
/* Ok, do the async read-ahead now */
page = read_swap_cache_async(swp_entry(swp_type(entry), offset),
gfp_mask, vma, addr);
if (!page)
break;
page_cache_release(page);
}
lru_add_drain(); /* Push any new pages onto the LRU now */
return read_swap_cache_async(entry, gfp_mask, vma, addr);
}
static int write_page(void *buf, unsigned long offset)
{
swp_entry_t entry;
int error = -ENOSPC;
if (offset) {
entry = swp_entry(root_swap, offset);
error = rw_swap_page_sync(WRITE, entry, virt_to_page(buf));
}
return error;
}
sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap)
{
unsigned long offset;
offset = swp_offset(get_swap_page_of_type(swap));
if (offset) {
if (bitmap_set(bitmap, offset))
swap_free(swp_entry(swap, offset));
else
return swapdev_block(swap, offset);
}
return 0;
}
static int mark_swapfiles(swp_entry_t start)
{
int error;
rw_swap_page_sync(READ,
swp_entry(root_swap, 0),
virt_to_page((unsigned long)&swsusp_header));
if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
!memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
swsusp_header.image = start;
error = rw_swap_page_sync(WRITE,
swp_entry(root_swap, 0),
virt_to_page((unsigned long)
&swsusp_header));
} else {
pr_debug("swsusp: Partition is not swap space.\n");
error = -ENODEV;
}
return error;
}
static int mark_swapfiles(swp_entry_t prev)
{
int error;
rw_swap_page_sync(READ,
swp_entry(root_swap, 0),
virt_to_page((unsigned long)&pmdisk_header));
if (!memcmp("SWAP-SPACE",pmdisk_header.sig,10) ||
!memcmp("SWAPSPACE2",pmdisk_header.sig,10)) {
memcpy(pmdisk_header.orig_sig,pmdisk_header.sig,10);
memcpy(pmdisk_header.sig,PMDISK_SIG,10);
pmdisk_header.pmdisk_info = prev;
error = rw_swap_page_sync(WRITE,
swp_entry(root_swap, 0),
virt_to_page((unsigned long)
&pmdisk_header));
} else {
pr_debug("pmdisk: Partition is not swap space.\n");
error = -ENODEV;
}
return error;
}
unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap)
{
unsigned long offset;
offset = swp_offset(get_swap_page_of_type(swap));
if (offset) {
if (bitmap_set(bitmap, offset)) {
swap_free(swp_entry(swap, offset));
offset = 0;
}
}
return offset;
}
void free_all_swap_pages(int swap, struct bitmap_page *bitmap)
{
unsigned int bit, n;
unsigned long test;
bit = 0;
while (bitmap) {
for (n = 0; n < BITMAP_PAGE_CHUNKS; n++)
for (test = 1UL; test; test <<= 1) {
if (bitmap->chunks[n] & test)
swap_free(swp_entry(swap, bit));
bit++;
}
bitmap = bitmap->next;
}
}
swp_entry_t get_swap_page_of_type(int type)
{
struct swap_info_struct *si;
pgoff_t offset;
spin_lock(&swap_lock);
si = swap_info + type;
if (si->flags & SWP_WRITEOK) {
nr_swap_pages--;
offset = scan_swap_map(si);
if (offset) {
spin_unlock(&swap_lock);
return swp_entry(type, offset);
}
nr_swap_pages++;
}
spin_unlock(&swap_lock);
return (swp_entry_t) {0};
}
swp_entry_t get_swap_page(void)
{
struct swap_info_struct *si;
pgoff_t offset;
int type, next;
int wrapped = 0;
spin_lock(&swap_lock);
if (nr_swap_pages <= 0)
goto noswap;
nr_swap_pages--;
for (type = swap_list.next; type >= 0 && wrapped < 2; type = next) {
si = swap_info + type;
next = si->next;
if (next < 0 ||
(!wrapped && si->prio != swap_info[next].prio)) {
next = swap_list.head;
wrapped++;
}
if (!si->highest_bit)
continue;
if (!(si->flags & SWP_WRITEOK))
continue;
swap_list.next = next;
offset = scan_swap_map(si);
if (offset) {
spin_unlock(&swap_lock);
return swp_entry(type, offset);
}
next = swap_list.next;
}
nr_swap_pages++;
noswap:
spin_unlock(&swap_lock);
return (swp_entry_t) {0};
}
int swsusp_write(void)
{
struct swap_map_handle handle;
struct snapshot_handle snapshot;
struct swsusp_info *header;
int error;
if ((error = swsusp_swap_check())) {
printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
return error;
}
memset(&snapshot, 0, sizeof(struct snapshot_handle));
error = snapshot_read_next(&snapshot, PAGE_SIZE);
if (error < PAGE_SIZE)
return error < 0 ? error : -EFAULT;
header = (struct swsusp_info *)data_of(snapshot);
if (!enough_swap(header->pages)) {
printk(KERN_ERR "swsusp: Not enough free swap\n");
return -ENOSPC;
}
error = get_swap_writer(&handle);
if (!error) {
unsigned long start = handle.cur_swap;
error = swap_write_page(&handle, header);
if (!error)
error = save_image(&handle, &snapshot,
header->pages - 1);
if (!error) {
flush_swap_writer(&handle);
printk("S");
error = mark_swapfiles(swp_entry(root_swap, start));
printk("|\n");
}
}
if (error)
free_all_swap_pages(root_swap, handle.bitmap);
release_swap_writer(&handle);
return error;
}
/**
* swapin_readahead - swap in pages in hope we need them soon
* @entry: swap entry of this memory
* @gfp_mask: memory allocation flags
* @vma: user vma this address belongs to
* @addr: target address for mempolicy
*
* Returns the struct page for entry and addr, after queueing swapin.
*
* Primitive swap readahead code. We simply read an aligned block of
* (1 << page_cluster) entries in the swap area. This method is chosen
* because it doesn't cost us any seek time. We also make sure to queue
* the 'original' request together with the readahead ones...
*
* This has been extended to use the NUMA policies from the mm triggering
* the readahead.
*
* Caller must hold down_read on the vma->vm_mm if vma is not NULL.
*/
struct page *swapin_readahead(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma, unsigned long addr)
{
struct page *page;
unsigned long entry_offset = swp_offset(entry);
unsigned long offset = entry_offset;
unsigned long start_offset, end_offset;
unsigned long mask;
struct blk_plug plug;
bool do_poll = true;
mask = swapin_nr_pages(offset) - 1;
if (!mask)
goto skip;
do_poll = false;
/* Read a page_cluster sized and aligned cluster around offset. */
start_offset = offset & ~mask;
end_offset = offset | mask;
if (!start_offset) /* First page is swap header. */
start_offset++;
blk_start_plug(&plug);
for (offset = start_offset; offset <= end_offset ; offset++) {
/* Ok, do the async read-ahead now */
page = read_swap_cache_async(swp_entry(swp_type(entry), offset),
gfp_mask, vma, addr, false);
if (!page)
continue;
if (offset != entry_offset && likely(!PageTransCompound(page)))
SetPageReadahead(page);
put_page(page);
}
blk_finish_plug(&plug);
lru_add_drain(); /* Push any new pages onto the LRU now */
skip:
return read_swap_cache_async(entry, gfp_mask, vma, addr, do_poll);
}
static void mark_swapfiles(swp_entry_t prev, int mode)
{
swp_entry_t entry;
union diskpage *cur;
struct page *page;
if (root_swap == 0xFFFF) /* ignored */
return;
page = alloc_page(GFP_ATOMIC);
if (!page)
panic("Out of memory in mark_swapfiles");
cur = page_address(page);
/* XXX: this is dirty hack to get first page of swap file */
entry = swp_entry(root_swap, 0);
rw_swap_page_sync(READ, entry, page);
if (mode == MARK_SWAP_RESUME) {
if (!memcmp("S1",cur->swh.magic.magic,2))
memcpy(cur->swh.magic.magic,"SWAP-SPACE",10);
else if (!memcmp("S2",cur->swh.magic.magic,2))
memcpy(cur->swh.magic.magic,"SWAPSPACE2",10);
else printk("%sUnable to find suspended-data signature (%.10s - misspelled?\n",
name_resume, cur->swh.magic.magic);
} else {
if ((!memcmp("SWAP-SPACE",cur->swh.magic.magic,10)))
memcpy(cur->swh.magic.magic,"S1SUSP....",10);
else if ((!memcmp("SWAPSPACE2",cur->swh.magic.magic,10)))
memcpy(cur->swh.magic.magic,"S2SUSP....",10);
else panic("\nSwapspace is not swapspace (%.10s)\n", cur->swh.magic.magic);
cur->link.next = prev; /* prev is the first/last swap page of the resume area */
/* link.next lies *no more* in last 4/8 bytes of magic */
}
rw_swap_page_sync(WRITE, entry, page);
__free_page(page);
}
/* attempts to compress and store an single page */
static int zswap_frontswap_store(unsigned type, pgoff_t offset,
struct page *page)
{
struct zswap_tree *tree = zswap_trees[type];
struct zswap_entry *entry, *dupentry;
int ret;
unsigned int dlen = PAGE_SIZE, len;
unsigned long handle;
char *buf;
u8 *src, *dst;
struct zswap_header *zhdr;
if (!tree) {
ret = -ENODEV;
goto reject;
}
/* reclaim space if needed */
if (zswap_is_full()) {
zswap_pool_limit_hit++;
if (zbud_reclaim_page(tree->pool, 8)) {
zswap_reject_reclaim_fail++;
ret = -ENOMEM;
goto reject;
}
}
/* allocate entry */
entry = zswap_entry_cache_alloc(GFP_KERNEL);
if (!entry) {
zswap_reject_kmemcache_fail++;
ret = -ENOMEM;
goto reject;
}
/* compress */
dst = get_cpu_var(zswap_dstmem);
src = kmap_atomic(page);
ret = zswap_comp_op(ZSWAP_COMPOP_COMPRESS, src, PAGE_SIZE, dst, &dlen);
kunmap_atomic(src);
if (ret) {
ret = -EINVAL;
goto freepage;
}
/* store */
len = dlen + sizeof(struct zswap_header);
ret = zbud_alloc(tree->pool, len, __GFP_NORETRY | __GFP_NOWARN,
&handle);
if (ret == -ENOSPC) {
zswap_reject_compress_poor++;
goto freepage;
}
if (ret) {
zswap_reject_alloc_fail++;
goto freepage;
}
zhdr = zbud_map(tree->pool, handle);
zhdr->swpentry = swp_entry(type, offset);
buf = (u8 *)(zhdr + 1);
memcpy(buf, dst, dlen);
zbud_unmap(tree->pool, handle);
put_cpu_var(zswap_dstmem);
/* populate entry */
entry->offset = offset;
entry->handle = handle;
entry->length = dlen;
/* map */
spin_lock(&tree->lock);
do {
ret = zswap_rb_insert(&tree->rbroot, entry, &dupentry);
if (ret == -EEXIST) {
zswap_duplicate_entry++;
/* remove from rbtree */
zswap_rb_erase(&tree->rbroot, dupentry);
zswap_entry_put(tree, dupentry);
}
} while (ret == -EEXIST);
spin_unlock(&tree->lock);
/* update stats */
atomic_inc(&zswap_stored_pages);
zswap_pool_pages = zbud_get_pool_size(tree->pool);
return 0;
freepage:
put_cpu_var(zswap_dstmem);
zswap_entry_cache_free(entry);
reject:
return ret;
}
/* attempts to compress and store an single page */
static int zswap_frontswap_store(unsigned type, pgoff_t offset,
struct page *page)
{
struct zswap_tree *tree = zswap_trees[type];
struct zswap_entry *entry, *dupentry;
int ret;
unsigned int dlen = PAGE_SIZE, len;
unsigned long handle;
char *buf;
u8 *src, *dst;
struct zswap_header *zhdr;
if (!tree) {
ret = -ENODEV;
goto reject;
}
/* if this page got EIO on pageout before, give up immediately */
if (PageError(page)) {
ret = -ENOMEM;
goto reject;
}
/* reclaim space if needed */
if (zswap_is_full()) {
zswap_pool_limit_hit++;
if (zpool_shrink(zswap_pool, 1, NULL)) {
zswap_reject_reclaim_fail++;
ret = -ENOMEM;
goto reject;
}
}
/* allocate entry */
entry = zswap_entry_cache_alloc(GFP_KERNEL);
if (!entry) {
zswap_reject_kmemcache_fail++;
ret = -ENOMEM;
goto reject;
}
/* compress */
src = kmap_atomic(page);
if (page_zero_filled(src)) {
atomic_inc(&zswap_zero_pages);
entry->zero_flag = 1;
kunmap_atomic(src);
handle = 0;
dlen = PAGE_SIZE;
goto zeropage_out;
}
dst = get_cpu_var(zswap_dstmem);
ret = zswap_comp_op(ZSWAP_COMPOP_COMPRESS, src, PAGE_SIZE, dst, &dlen);
kunmap_atomic(src);
if (ret) {
ret = -EINVAL;
goto freepage;
}
/* store */
len = dlen + sizeof(struct zswap_header);
ret = zpool_malloc(zswap_pool, len, __GFP_NORETRY | __GFP_NOWARN,
&handle);
if (ret == -ENOSPC) {
zswap_reject_compress_poor++;
goto freepage;
}
if (ret) {
zswap_reject_alloc_fail++;
goto freepage;
}
zhdr = zpool_map_handle(zswap_pool, handle, ZPOOL_MM_RW);
zhdr->swpentry = swp_entry(type, offset);
buf = (u8 *)(zhdr + 1);
memcpy(buf, dst, dlen);
zpool_unmap_handle(zswap_pool, handle);
put_cpu_var(zswap_dstmem);
zeropage_out:
/* populate entry */
entry->offset = offset;
entry->handle = handle;
entry->length = dlen;
/* map */
spin_lock(&tree->lock);
do {
ret = zswap_rb_insert(&tree->rbroot, entry, &dupentry);
if (ret == -EEXIST) {
zswap_duplicate_entry++;
/* remove from rbtree */
zswap_rb_erase(&tree->rbroot, dupentry);
zswap_entry_put(tree, dupentry);
}
} while (ret == -EEXIST);
spin_unlock(&tree->lock);
/* update stats */
atomic_inc(&zswap_stored_pages);
zswap_pool_total_size = zpool_get_total_size(zswap_pool);
zswap_pool_pages = zswap_pool_total_size >> PAGE_SHIFT;
return 0;
freepage:
put_cpu_var(zswap_dstmem);
zswap_entry_cache_free(entry);
reject:
return ret;
}
/*
* We completely avoid races by reading each swap page in advance,
* and then search for the process using it. All the necessary
* page table adjustments can then be made atomically.
*/
static int try_to_unuse(unsigned int type)
{
struct swap_info_struct * si = &swap_info[type];
struct mm_struct *start_mm;
unsigned short *swap_map;
unsigned short swcount;
struct page *page;
swp_entry_t entry;
unsigned int i = 0;
int retval = 0;
int reset_overflow = 0;
int shmem;
/*
* When searching mms for an entry, a good strategy is to
* start at the first mm we freed the previous entry from
* (though actually we don't notice whether we or coincidence
* freed the entry). Initialize this start_mm with a hold.
*
* A simpler strategy would be to start at the last mm we
* freed the previous entry from; but that would take less
* advantage of mmlist ordering, which clusters forked mms
* together, child after parent. If we race with dup_mmap(), we
* prefer to resolve parent before child, lest we miss entries
* duplicated after we scanned child: using last mm would invert
* that. Though it's only a serious concern when an overflowed
* swap count is reset from SWAP_MAP_MAX, preventing a rescan.
*/
start_mm = &init_mm;
atomic_inc(&init_mm.mm_users);
/*
* Keep on scanning until all entries have gone. Usually,
* one pass through swap_map is enough, but not necessarily:
* there are races when an instance of an entry might be missed.
*/
while ((i = find_next_to_unuse(si, i)) != 0) {
if (signal_pending(current)) {
retval = -EINTR;
break;
}
/*
* Get a page for the entry, using the existing swap
* cache page if there is one. Otherwise, get a clean
* page and read the swap into it.
*/
swap_map = &si->swap_map[i];
entry = swp_entry(type, i);
page = read_swap_cache_async(entry, NULL, 0);
if (!page) {
/*
* Either swap_duplicate() failed because entry
* has been freed independently, and will not be
* reused since sys_swapoff() already disabled
* allocation from here, or alloc_page() failed.
*/
if (!*swap_map)
continue;
retval = -ENOMEM;
break;
}
/*
* Don't hold on to start_mm if it looks like exiting.
*/
if (atomic_read(&start_mm->mm_users) == 1) {
mmput(start_mm);
start_mm = &init_mm;
atomic_inc(&init_mm.mm_users);
}
/*
* Wait for and lock page. When do_swap_page races with
* try_to_unuse, do_swap_page can handle the fault much
* faster than try_to_unuse can locate the entry. This
* apparently redundant "wait_on_page_locked" lets try_to_unuse
* defer to do_swap_page in such a case - in some tests,
* do_swap_page and try_to_unuse repeatedly compete.
*/
wait_on_page_locked(page);
wait_on_page_writeback(page);
lock_page(page);
wait_on_page_writeback(page);
/*
* Remove all references to entry.
* Whenever we reach init_mm, there's no address space
* to search, but use it as a reminder to search shmem.
*/
shmem = 0;
swcount = *swap_map;
if (swcount > 1) {
if (start_mm == &init_mm)
shmem = shmem_unuse(entry, page);
else
retval = unuse_mm(start_mm, entry, page);
}
if (*swap_map > 1) {
int set_start_mm = (*swap_map >= swcount);
struct list_head *p = &start_mm->mmlist;
struct mm_struct *new_start_mm = start_mm;
struct mm_struct *prev_mm = start_mm;
struct mm_struct *mm;
atomic_inc(&new_start_mm->mm_users);
atomic_inc(&prev_mm->mm_users);
spin_lock(&mmlist_lock);
while (*swap_map > 1 && !retval &&
(p = p->next) != &start_mm->mmlist) {
mm = list_entry(p, struct mm_struct, mmlist);
if (!atomic_inc_not_zero(&mm->mm_users))
continue;
spin_unlock(&mmlist_lock);
mmput(prev_mm);
prev_mm = mm;
cond_resched();
swcount = *swap_map;
if (swcount <= 1)
;
else if (mm == &init_mm) {
set_start_mm = 1;
shmem = shmem_unuse(entry, page);
} else
retval = unuse_mm(mm, entry, page);
if (set_start_mm && *swap_map < swcount) {
mmput(new_start_mm);
atomic_inc(&mm->mm_users);
new_start_mm = mm;
set_start_mm = 0;
}
spin_lock(&mmlist_lock);
}
spin_unlock(&mmlist_lock);
mmput(prev_mm);
mmput(start_mm);
start_mm = new_start_mm;
}
if (retval) {
unlock_page(page);
page_cache_release(page);
break;
}
/*
* How could swap count reach 0x7fff when the maximum
* pid is 0x7fff, and there's no way to repeat a swap
* page within an mm (except in shmem, where it's the
* shared object which takes the reference count)?
* We believe SWAP_MAP_MAX cannot occur in Linux 2.4.
*
* If that's wrong, then we should worry more about
* exit_mmap() and do_munmap() cases described above:
* we might be resetting SWAP_MAP_MAX too early here.
* We know "Undead"s can happen, they're okay, so don't
* report them; but do report if we reset SWAP_MAP_MAX.
*/
if (*swap_map == SWAP_MAP_MAX) {
spin_lock(&swap_lock);
*swap_map = 1;
spin_unlock(&swap_lock);
reset_overflow = 1;
}
/*
* If a reference remains (rare), we would like to leave
* the page in the swap cache; but try_to_unmap could
* then re-duplicate the entry once we drop page lock,
* so we might loop indefinitely; also, that page could
* not be swapped out to other storage meanwhile. So:
* delete from cache even if there's another reference,
* after ensuring that the data has been saved to disk -
* since if the reference remains (rarer), it will be
* read from disk into another page. Splitting into two
* pages would be incorrect if swap supported "shared
* private" pages, but they are handled by tmpfs files.
*
* Note shmem_unuse already deleted a swappage from
* the swap cache, unless the move to filepage failed:
* in which case it left swappage in cache, lowered its
* swap count to pass quickly through the loops above,
* and now we must reincrement count to try again later.
*/
if ((*swap_map > 1) && PageDirty(page) && PageSwapCache(page)) {
struct writeback_control wbc = {
.sync_mode = WB_SYNC_NONE,
};
swap_writepage(page, &wbc);
lock_page(page);
wait_on_page_writeback(page);
}
if (PageSwapCache(page)) {
if (shmem)
swap_duplicate(entry);
else
delete_from_swap_cache(page);
}
/*
* So we could skip searching mms once swap count went
* to 1, we did not mark any present ptes as dirty: must
* mark page dirty so shrink_page_list will preserve it.
*/
SetPageDirty(page);
unlock_page(page);
page_cache_release(page);
/*
* Make sure that we aren't completely killing
* interactive performance.
*/
cond_resched();
}
/*
* zcache_get_swap_cache_page
*
* This is an adaption of read_swap_cache_async()
*
* If success, page is returned in retpage
* Returns 0 if page was already in the swap cache, page is not locked
* Returns 1 if the new page needs to be populated, page is locked
*/
static int zcache_get_swap_cache_page(int type, pgoff_t offset,
struct page *new_page)
{
struct page *found_page;
swp_entry_t entry = swp_entry(type, offset);
int err;
BUG_ON(new_page == NULL);
do {
/*
* First check the swap cache. Since this is normally
* called after lookup_swap_cache() failed, re-calling
* that would confuse statistics.
*/
found_page = find_get_page(&swapper_space, entry.val);
if (found_page)
return 0;
/*
* call radix_tree_preload() while we can wait.
*/
err = radix_tree_preload(GFP_KERNEL);
if (err)
break;
/*
* Swap entry may have been freed since our caller observed it.
*/
err = swapcache_prepare(entry);
if (err == -EEXIST) { /* seems racy */
radix_tree_preload_end();
continue;
}
if (err) { /* swp entry is obsolete ? */
radix_tree_preload_end();
break;
}
/* May fail (-ENOMEM) if radix-tree node allocation failed. */
__set_page_locked(new_page);
SetPageSwapBacked(new_page);
err = __add_to_swap_cache(new_page, entry);
if (likely(!err)) {
radix_tree_preload_end();
lru_cache_add_anon(new_page);
return 1;
}
radix_tree_preload_end();
ClearPageSwapBacked(new_page);
__clear_page_locked(new_page);
/*
* add_to_swap_cache() doesn't return -EEXIST, so we can safely
* clear SWAP_HAS_CACHE flag.
*/
swapcache_free(entry, NULL);
/* FIXME: is it possible to get here without err==-ENOMEM?
* If not, we can dispense with the do loop, use goto retry */
} while (err != -ENOMEM);
return -ENOMEM;
}
