/**
* 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);
}
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)
{
#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);
}
/*
* 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...
*/
void swapin_readahead(swp_entry_t entry)
{
int i, num;
struct page *new_page;
unsigned long offset;
/*
* Get the number of handles we should do readahead io to. Also,
* grab temporary references on them, releasing them as io completes.
*/
num = valid_swaphandles(entry, &offset);
for (i = 0; i < num; offset++, i++) {
/* Don't block on I/O for read-ahead */
if (atomic_read(&nr_async_pages) >= pager_daemon.swap_cluster
* (1 << page_cluster)) {
while (i++ < num)
swap_free(SWP_ENTRY(SWP_TYPE(entry), offset++));
break;
}
/* Ok, do the async read-ahead now */
new_page = read_swap_cache_async(SWP_ENTRY(SWP_TYPE(entry), offset), 0);
if (new_page != NULL)
page_cache_release(new_page);
swap_free(SWP_ENTRY(SWP_TYPE(entry), offset));
}
return;
}
/**
* 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);
}
/*
* 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...
*/
void swapin_readahead(swp_entry_t entry)
{
int i, num;
struct page *new_page;
unsigned long offset;
/*
* Get the number of handles we should do readahead io to.
*/
num = valid_swaphandles(entry, &offset);
for (i = 0; i < num; offset++, i++) {
/* Ok, do the async read-ahead now */
new_page = read_swap_cache_async(SWP_ENTRY(SWP_TYPE(entry), offset));
if (!new_page)
break;
page_cache_release(new_page);
}
return;
}
/*
* 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;
int i = 0;
int retval = 0;
int reset_overflow = 0;
/*
* 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 (now preserved by swap_out()),
* which clusters forked address spaces together, most recent
* child immediately after parent. If we race with dup_mmap(),
* we very much want to resolve parent before child, otherwise
* we may miss some entries: using last mm would invert that.
*/
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:
* mmput() removes mm from mmlist before exit_mmap() and its
* zap_page_range(). That's not too bad, those entries are
* on their way out, and handled faster there than here.
* do_munmap() behaves similarly, taking the range out of mm's
* vma list before zap_page_range(). But unfortunately, when
* unmapping a part of a vma, it takes the whole out first,
* then reinserts what's left after (might even reschedule if
* open() method called) - so swap entries may be invisible
* to swapoff for a while, then reappear - but that is rare.
*/
while ((i = find_next_to_unuse(si, i))) {
/*
* 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);
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" 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(page);
lock_page(page);
/*
* Remove all references to entry, without blocking.
* Whenever we reach init_mm, there's no address space
* to search, but use it as a reminder to search shmem.
*/
swcount = *swap_map;
if (swcount > 1) {
flush_page_to_ram(page);
if (start_mm == &init_mm)
shmem_unuse(entry, page);
else
unuse_process(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 *mm;
spin_lock(&mmlist_lock);
while (*swap_map > 1 &&
(p = p->next) != &start_mm->mmlist) {
mm = list_entry(p, struct mm_struct, mmlist);
swcount = *swap_map;
if (mm == &init_mm) {
set_start_mm = 1;
shmem_unuse(entry, page);
} else
unuse_process(mm, entry, page);
if (set_start_mm && *swap_map < swcount) {
new_start_mm = mm;
set_start_mm = 0;
}
}
atomic_inc(&new_start_mm->mm_users);
spin_unlock(&mmlist_lock);
mmput(start_mm);
start_mm = new_start_mm;
}
/*
* 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) {
swap_list_lock();
swap_device_lock(si);
nr_swap_pages++;
*swap_map = 1;
swap_device_unlock(si);
swap_list_unlock();
reset_overflow = 1;
}
/*
* If a reference remains (rare), we would like to leave
* the page in the swap cache; but try_to_swap_out 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 its from swap cache.
*/
if ((*swap_map > 1) && PageDirty(page) && PageSwapCache(page)) {
rw_swap_page(WRITE, page);
lock_page(page);
}
if (PageSwapCache(page))
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 try_to_swap_out will preserve it.
*/
SetPageDirty(page);
UnlockPage(page);
page_cache_release(page);
/*
* Make sure that we aren't completely killing
* interactive performance. Interruptible check on
* signal_pending() would be nice, but changes the spec?
*/
if (current->need_resched)
schedule();
}
/*
* 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();
}
/*
* We hold the mm semaphore and the page_table_lock on entry and
* should release the pagetable lock on exit..
*/
static int do_swap_page(struct mm_struct * mm,
struct vm_area_struct * vma, unsigned long address,
pte_t * page_table, pte_t orig_pte, int write_access)
{
struct page *page;
swp_entry_t entry = pte_to_swp_entry(orig_pte);
pte_t pte;
int ret = 1;
spin_unlock(&mm->page_table_lock);
page = lookup_swap_cache(entry);
if (!page) {
swapin_readahead(entry);
page = read_swap_cache_async(entry);
if (!page) {
/*
* Back out if somebody else faulted in this pte while
* we released the page table lock.
*/
int retval;
spin_lock(&mm->page_table_lock);
retval = pte_same(*page_table, orig_pte) ? -1 : 1;
spin_unlock(&mm->page_table_lock);
return retval;
}
/* Had to read the page from swap area: Major fault */
ret = 2;
}
lock_page(page);
/*
* Back out if somebody else faulted in this pte while we
* released the page table lock.
*/
spin_lock(&mm->page_table_lock);
if (!pte_same(*page_table, orig_pte)) {
spin_unlock(&mm->page_table_lock);
unlock_page(page);
page_cache_release(page);
return 1;
}
/* The page isn't present yet, go ahead with the fault. */
swap_free(entry);
if (vm_swap_full())
remove_exclusive_swap_page(page);
mm->rss++;
pte = mk_pte(page, vma->vm_page_prot);
if (write_access && can_share_swap_page(page))
pte = pte_mkdirty(pte_mkwrite(pte));
unlock_page(page);
flush_page_to_ram(page);
flush_icache_page(vma, page);
set_pte(page_table, pte);
/* No need to invalidate - it was non-present before */
update_mmu_cache(vma, address, pte);
spin_unlock(&mm->page_table_lock);
return ret;
}