/*
* 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_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();
}
/* Sharing code of page_mkwrite method for rhel5 and rhel6 */
static int ll_page_mkwrite0(struct vm_area_struct *vma, struct page *vmpage,
bool *retry)
{
struct lu_env *env;
struct cl_io *io;
struct vvp_io *vio;
struct cl_env_nest nest;
int result;
sigset_t set;
struct inode *inode;
struct ll_inode_info *lli;
io = ll_fault_io_init(vma, &env, &nest, vmpage->index, NULL);
if (IS_ERR(io)) {
result = PTR_ERR(io);
goto out;
}
result = io->ci_result;
if (result < 0)
goto out_io;
io->u.ci_fault.ft_mkwrite = 1;
io->u.ci_fault.ft_writable = 1;
vio = vvp_env_io(env);
vio->u.fault.ft_vma = vma;
vio->u.fault.ft_vmpage = vmpage;
set = cfs_block_sigsinv(sigmask(SIGKILL) | sigmask(SIGTERM));
inode = vvp_object_inode(io->ci_obj);
lli = ll_i2info(inode);
result = cl_io_loop(env, io);
cfs_restore_sigs(set);
if (result == 0) {
struct inode *inode = file_inode(vma->vm_file);
struct ll_inode_info *lli = ll_i2info(inode);
lock_page(vmpage);
if (!vmpage->mapping) {
unlock_page(vmpage);
/* page was truncated and lock was cancelled, return
* ENODATA so that VM_FAULT_NOPAGE will be returned
* to handle_mm_fault().
*/
if (result == 0)
result = -ENODATA;
} else if (!PageDirty(vmpage)) {
/* race, the page has been cleaned by ptlrpcd after
* it was unlocked, it has to be added into dirty
* cache again otherwise this soon-to-dirty page won't
* consume any grants, even worse if this page is being
* transferred because it will break RPC checksum.
*/
unlock_page(vmpage);
CDEBUG(D_MMAP, "Race on page_mkwrite %p/%lu, page has been written out, retry.\n",
vmpage, vmpage->index);
*retry = true;
result = -EAGAIN;
}
if (result == 0) {
spin_lock(&lli->lli_lock);
lli->lli_flags |= LLIF_DATA_MODIFIED;
spin_unlock(&lli->lli_lock);
}
}
out_io:
cl_io_fini(env, io);
cl_env_nested_put(&nest, env);
out:
CDEBUG(D_MMAP, "%s mkwrite with %d\n", current->comm, result);
LASSERT(ergo(result == 0, PageLocked(vmpage)));
return result;
}
/*
* For address_spaces which do not use buffers nor write back.
*/
int yramfs_file_set_page_dirty_no_writeback(struct page *page)
{
if (!PageDirty(page))
return !TestSetPageDirty(page);
return 0;
}
/**
* write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
* @mapping: address space structure to write
* @wbc: subtract the number of written pages from *@wbc->nr_to_write
* @writepage: function called for each page
* @data: data passed to writepage function
*
* If a page is already under I/O, write_cache_pages() skips it, even
* if it's dirty. This is desirable behaviour for memory-cleaning writeback,
* but it is INCORRECT for data-integrity system calls such as fsync(). fsync()
* and msync() need to guarantee that all the data which was dirty at the time
* the call was made get new I/O started against them. If wbc->sync_mode is
* WB_SYNC_ALL then we were called for data integrity and we must wait for
* existing IO to complete.
*
* To avoid livelocks (when other process dirties new pages), we first tag
* pages which should be written back with TOWRITE tag and only then start
* writing them. For data-integrity sync we have to be careful so that we do
* not miss some pages (e.g., because some other process has cleared TOWRITE
* tag we set). The rule we follow is that TOWRITE tag can be cleared only
* by the process clearing the DIRTY tag (and submitting the page for IO).
*/
int write_cache_pages(struct address_space *mapping,
struct writeback_control *wbc, writepage_t writepage,
void *data)
{
int ret = 0;
int done = 0;
struct pagevec pvec;
int nr_pages;
pgoff_t uninitialized_var(writeback_index);
pgoff_t index;
pgoff_t end; /* Inclusive */
pgoff_t done_index;
int cycled;
int range_whole = 0;
int tag;
pagevec_init(&pvec, 0);
if (wbc->range_cyclic) {
writeback_index = mapping->writeback_index; /* prev offset */
index = writeback_index;
if (index == 0)
cycled = 1;
else
cycled = 0;
end = -1;
} else {
index = wbc->range_start >> PAGE_CACHE_SHIFT;
end = wbc->range_end >> PAGE_CACHE_SHIFT;
if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
range_whole = 1;
cycled = 1; /* ignore range_cyclic tests */
}
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag = PAGECACHE_TAG_TOWRITE;
else
tag = PAGECACHE_TAG_DIRTY;
retry:
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag_pages_for_writeback(mapping, index, end);
done_index = index;
while (!done && (index <= end)) {
int i;
nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, tag,
min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1);
if (nr_pages == 0)
break;
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
/*
* At this point, the page may be truncated or
* invalidated (changing page->mapping to NULL), or
* even swizzled back from swapper_space to tmpfs file
* mapping. However, page->index will not change
* because we have a reference on the page.
*/
if (page->index > end) {
/*
* can't be range_cyclic (1st pass) because
* end == -1 in that case.
*/
done = 1;
break;
}
done_index = page->index;
lock_page(page);
/*
* Page truncated or invalidated. We can freely skip it
* then, even for data integrity operations: the page
* has disappeared concurrently, so there could be no
* real expectation of this data interity operation
* even if there is now a new, dirty page at the same
* pagecache address.
*/
if (unlikely(page->mapping != mapping)) {
continue_unlock:
unlock_page(page);
continue;
}
if (!PageDirty(page)) {
/* someone wrote it for us */
goto continue_unlock;
}
if (PageWriteback(page)) {
if (wbc->sync_mode != WB_SYNC_NONE)
wait_on_page_writeback(page);
else
goto continue_unlock;
}
BUG_ON(PageWriteback(page));
if (!clear_page_dirty_for_io(page))
goto continue_unlock;
trace_wbc_writepage(wbc, mapping->backing_dev_info);
ret = (*writepage)(page, wbc, data);
if (unlikely(ret)) {
if (ret == AOP_WRITEPAGE_ACTIVATE) {
unlock_page(page);
ret = 0;
} else {
/*
* done_index is set past this page,
* so media errors will not choke
* background writeout for the entire
* file. This has consequences for
* range_cyclic semantics (ie. it may
* not be suitable for data integrity
* writeout).
*/
done_index = page->index + 1;
done = 1;
break;
}
}
/*
* We stop writing back only if we are not doing
* integrity sync. In case of integrity sync we have to
* keep going until we have written all the pages
* we tagged for writeback prior to entering this loop.
*/
if (--wbc->nr_to_write <= 0 &&
wbc->sync_mode == WB_SYNC_NONE) {
done = 1;
break;
}
}
pagevec_release(&pvec);
cond_resched();
}
if (!cycled && !done) {
/*
* range_cyclic:
* We hit the last page and there is more work to be done: wrap
* back to the start of the file
*/
cycled = 1;
index = 0;
end = writeback_index - 1;
goto retry;
}
if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
mapping->writeback_index = done_index;
return ret;
}
/**
* nilfs_copy_back_pages -- copy back pages to original cache from shadow cache
* @dmap: destination page cache
* @smap: source page cache
*
* No pages must no be added to the cache during this process.
* This must be ensured by the caller.
*/
void nilfs_copy_back_pages(struct address_space *dmap,
struct address_space *smap)
{
struct pagevec pvec;
unsigned int i, n;
pgoff_t index = 0;
int err;
pagevec_init(&pvec, 0);
repeat:
n = pagevec_lookup(&pvec, smap, index, PAGEVEC_SIZE);
if (!n)
return;
index = pvec.pages[n - 1]->index + 1;
for (i = 0; i < pagevec_count(&pvec); i++) {
struct page *page = pvec.pages[i], *dpage;
pgoff_t offset = page->index;
lock_page(page);
dpage = find_lock_page(dmap, offset);
if (dpage) {
/* override existing page on the destination cache */
WARN_ON(PageDirty(dpage));
nilfs_copy_page(dpage, page, 0);
unlock_page(dpage);
page_cache_release(dpage);
} else {
struct page *page2;
/* move the page to the destination cache */
spin_lock_irq(&smap->tree_lock);
page2 = radix_tree_delete(&smap->page_tree, offset);
WARN_ON(page2 != page);
smap->nrpages--;
spin_unlock_irq(&smap->tree_lock);
spin_lock_irq(&dmap->tree_lock);
err = radix_tree_insert(&dmap->page_tree, offset, page);
if (unlikely(err < 0)) {
WARN_ON(err == -EEXIST);
page->mapping = NULL;
page_cache_release(page); /* for cache */
} else {
page->mapping = dmap;
dmap->nrpages++;
if (PageDirty(page))
radix_tree_tag_set(&dmap->page_tree,
offset,
PAGECACHE_TAG_DIRTY);
}
spin_unlock_irq(&dmap->tree_lock);
}
unlock_page(page);
}
pagevec_release(&pvec);
cond_resched();
goto repeat;
}
/*
* 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.
*/
swcount = *swap_map;
if ((swcount > 0) != PageSwapCache(page))
BUG();
if ((swcount > 1) && PageDirty(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();
}
/*
* Synchronously write back the locked page and any subsequent non-locked dirty
* pages.
*/
static int afs_write_back_from_locked_page(struct address_space *mapping,
struct writeback_control *wbc,
struct page *primary_page,
pgoff_t final_page)
{
struct afs_vnode *vnode = AFS_FS_I(mapping->host);
struct page *pages[8], *page;
unsigned long count, priv;
unsigned n, offset, to, f, t;
pgoff_t start, first, last;
int loop, ret;
_enter(",%lx", primary_page->index);
count = 1;
if (test_set_page_writeback(primary_page))
BUG();
/* Find all consecutive lockable dirty pages that have contiguous
* written regions, stopping when we find a page that is not
* immediately lockable, is not dirty or is missing, or we reach the
* end of the range.
*/
start = primary_page->index;
priv = page_private(primary_page);
offset = priv & AFS_PRIV_MAX;
to = priv >> AFS_PRIV_SHIFT;
trace_afs_page_dirty(vnode, tracepoint_string("store"),
primary_page->index, priv);
WARN_ON(offset == to);
if (offset == to)
trace_afs_page_dirty(vnode, tracepoint_string("WARN"),
primary_page->index, priv);
if (start >= final_page ||
(to < PAGE_SIZE && !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)))
goto no_more;
start++;
do {
_debug("more %lx [%lx]", start, count);
n = final_page - start + 1;
if (n > ARRAY_SIZE(pages))
n = ARRAY_SIZE(pages);
n = find_get_pages_contig(mapping, start, ARRAY_SIZE(pages), pages);
_debug("fgpc %u", n);
if (n == 0)
goto no_more;
if (pages[0]->index != start) {
do {
put_page(pages[--n]);
} while (n > 0);
goto no_more;
}
for (loop = 0; loop < n; loop++) {
page = pages[loop];
if (to != PAGE_SIZE &&
!test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags))
break;
if (page->index > final_page)
break;
if (!trylock_page(page))
break;
if (!PageDirty(page) || PageWriteback(page)) {
unlock_page(page);
break;
}
priv = page_private(page);
f = priv & AFS_PRIV_MAX;
t = priv >> AFS_PRIV_SHIFT;
if (f != 0 &&
!test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)) {
unlock_page(page);
break;
}
to = t;
trace_afs_page_dirty(vnode, tracepoint_string("store+"),
page->index, priv);
if (!clear_page_dirty_for_io(page))
BUG();
if (test_set_page_writeback(page))
BUG();
unlock_page(page);
put_page(page);
}
count += loop;
if (loop < n) {
for (; loop < n; loop++)
put_page(pages[loop]);
goto no_more;
}
start += loop;
} while (start <= final_page && count < 65536);
no_more:
/* We now have a contiguous set of dirty pages, each with writeback
* set; the first page is still locked at this point, but all the rest
* have been unlocked.
*/
unlock_page(primary_page);
first = primary_page->index;
last = first + count - 1;
_debug("write back %lx[%u..] to %lx[..%u]", first, offset, last, to);
ret = afs_store_data(mapping, first, last, offset, to);
switch (ret) {
case 0:
ret = count;
break;
default:
pr_notice("kAFS: Unexpected error from FS.StoreData %d\n", ret);
/* Fall through */
case -EACCES:
case -EPERM:
case -ENOKEY:
case -EKEYEXPIRED:
case -EKEYREJECTED:
case -EKEYREVOKED:
afs_redirty_pages(wbc, mapping, first, last);
mapping_set_error(mapping, ret);
break;
case -EDQUOT:
case -ENOSPC:
afs_redirty_pages(wbc, mapping, first, last);
mapping_set_error(mapping, -ENOSPC);
break;
case -EROFS:
case -EIO:
case -EREMOTEIO:
case -EFBIG:
case -ENOENT:
case -ENOMEDIUM:
case -ENXIO:
afs_kill_pages(mapping, first, last);
mapping_set_error(mapping, ret);
break;
}
_leave(" = %d", ret);
return ret;
}
static struct page *follow_page_pte(struct vm_area_struct *vma,
unsigned long address, pmd_t *pmd, unsigned int flags)
{
struct mm_struct *mm = vma->vm_mm;
struct page *page;
spinlock_t *ptl;
pte_t *ptep, pte;
retry:
if (unlikely(pmd_bad(*pmd)))
return no_page_table(vma, flags);
ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
pte = *ptep;
if (!pte_present(pte)) {
swp_entry_t entry;
/*
* KSM's break_ksm() relies upon recognizing a ksm page
* even while it is being migrated, so for that case we
* need migration_entry_wait().
*/
if (likely(!(flags & FOLL_MIGRATION)))
goto no_page;
if (pte_none(pte) || pte_file(pte))
goto no_page;
entry = pte_to_swp_entry(pte);
if (!is_migration_entry(entry))
goto no_page;
pte_unmap_unlock(ptep, ptl);
migration_entry_wait(mm, pmd, address);
goto retry;
}
if ((flags & FOLL_NUMA) && pte_numa(pte))
goto no_page;
if ((flags & FOLL_WRITE) && !pte_write(pte)) {
pte_unmap_unlock(ptep, ptl);
return NULL;
}
page = vm_normal_page(vma, address, pte);
if (unlikely(!page)) {
if ((flags & FOLL_DUMP) ||
!is_zero_pfn(pte_pfn(pte)))
goto bad_page;
page = pte_page(pte);
}
if (flags & FOLL_GET)
get_page_foll(page);
if (flags & FOLL_TOUCH) {
if ((flags & FOLL_WRITE) &&
!pte_dirty(pte) && !PageDirty(page))
set_page_dirty(page);
/*
* pte_mkyoung() would be more correct here, but atomic care
* is needed to avoid losing the dirty bit: it is easier to use
* mark_page_accessed().
*/
mark_page_accessed(page);
}
if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
/*
* The preliminary mapping check is mainly to avoid the
* pointless overhead of lock_page on the ZERO_PAGE
* which might bounce very badly if there is contention.
*
* If the page is already locked, we don't need to
* handle it now - vmscan will handle it later if and
* when it attempts to reclaim the page.
*/
if (page->mapping && trylock_page(page)) {
lru_add_drain(); /* push cached pages to LRU */
/*
* Because we lock page here, and migration is
* blocked by the pte's page reference, and we
* know the page is still mapped, we don't even
* need to check for file-cache page truncation.
*/
mlock_vma_page(page);
unlock_page(page);
}
}
pte_unmap_unlock(ptep, ptl);
return page;
bad_page:
pte_unmap_unlock(ptep, ptl);
return ERR_PTR(-EFAULT);
no_page:
pte_unmap_unlock(ptep, ptl);
if (!pte_none(pte))
return NULL;
return no_page_table(vma, flags);
}
static int tagfs_set_page_dirty(struct page *page)
{
if (!PageDirty(page))
return !TestSetPageDirty(page);
return 0;
}
// check_swap - check the correctness of swap & page replacement algorithm
static void
check_swap(void) {
size_t nr_free_pages_store = nr_free_pages();
size_t slab_allocated_store = slab_allocated();
size_t offset;
for (offset = 2; offset < max_swap_offset; offset ++) {
mem_map[offset] = 1;
}
struct mm_struct *mm = mm_create();
assert(mm != NULL);
extern struct mm_struct *check_mm_struct;
assert(check_mm_struct == NULL);
check_mm_struct = mm;
pgd_t *pgdir = mm->pgdir = boot_pgdir;
assert(pgdir[0] == 0);
struct vma_struct *vma = vma_create(0, PTSIZE, VM_WRITE | VM_READ);
assert(vma != NULL);
insert_vma_struct(mm, vma);
struct Page *rp0 = alloc_page(), *rp1 = alloc_page();
assert(rp0 != NULL && rp1 != NULL);
uint32_t perm = PTE_U | PTE_W;
int ret = page_insert(pgdir, rp1, 0, perm);
assert(ret == 0 && page_ref(rp1) == 1);
page_ref_inc(rp1);
ret = page_insert(pgdir, rp0, 0, perm);
assert(ret == 0 && page_ref(rp1) == 1 && page_ref(rp0) == 1);
// check try_alloc_swap_entry
swap_entry_t entry = try_alloc_swap_entry();
assert(swap_offset(entry) == 1);
mem_map[1] = 1;
assert(try_alloc_swap_entry() == 0);
// set rp1, Swap, Active, add to hash_list, active_list
swap_page_add(rp1, entry);
swap_active_list_add(rp1);
assert(PageSwap(rp1));
mem_map[1] = 0;
entry = try_alloc_swap_entry();
assert(swap_offset(entry) == 1);
assert(!PageSwap(rp1));
// check swap_remove_entry
assert(swap_hash_find(entry) == NULL);
mem_map[1] = 2;
swap_remove_entry(entry);
assert(mem_map[1] == 1);
swap_page_add(rp1, entry);
swap_inactive_list_add(rp1);
swap_remove_entry(entry);
assert(PageSwap(rp1));
assert(rp1->index == entry && mem_map[1] == 0);
// check page_launder, move page from inactive_list to active_list
assert(page_ref(rp1) == 1);
assert(nr_active_pages == 0 && nr_inactive_pages == 1);
assert(list_next(&(inactive_list.swap_list)) == &(rp1->swap_link));
page_launder();
assert(nr_active_pages == 1 && nr_inactive_pages == 0);
assert(PageSwap(rp1) && PageActive(rp1));
entry = try_alloc_swap_entry();
assert(swap_offset(entry) == 1);
assert(!PageSwap(rp1) && nr_active_pages == 0);
assert(list_empty(&(active_list.swap_list)));
// set rp1 inactive again
assert(page_ref(rp1) == 1);
swap_page_add(rp1, 0);
assert(PageSwap(rp1) && swap_offset(rp1->index) == 1);
swap_inactive_list_add(rp1);
mem_map[1] = 1;
assert(nr_inactive_pages == 1);
page_ref_dec(rp1);
size_t count = nr_free_pages();
swap_remove_entry(entry);
assert(nr_inactive_pages == 0 && nr_free_pages() == count + 1);
// check swap_out_mm
pte_t *ptep0 = get_pte(pgdir, 0, 0), *ptep1;
assert(ptep0 != NULL && pte2page(*ptep0) == rp0);
ret = swap_out_mm(mm, 0);
assert(ret == 0);
ret = swap_out_mm(mm, 10);
assert(ret == 1 && mm->swap_address == PGSIZE);
ret = swap_out_mm(mm, 10);
assert(ret == 0 && *ptep0 == entry && mem_map[1] == 1);
assert(PageDirty(rp0) && PageActive(rp0) && page_ref(rp0) == 0);
assert(nr_active_pages == 1 && list_next(&(active_list.swap_list)) == &(rp0->swap_link));
// check refill_inactive_scan()
refill_inactive_scan();
assert(!PageActive(rp0) && page_ref(rp0) == 0);
assert(nr_inactive_pages == 1 && list_next(&(inactive_list.swap_list)) == &(rp0->swap_link));
page_ref_inc(rp0);
page_launder();
assert(PageActive(rp0) && page_ref(rp0) == 1);
assert(nr_active_pages == 1 && list_next(&(active_list.swap_list)) == &(rp0->swap_link));
page_ref_dec(rp0);
refill_inactive_scan();
assert(!PageActive(rp0));
// save data in rp0
int i;
for (i = 0; i < PGSIZE; i ++) {
((char *)page2kva(rp0))[i] = (char)i;
}
page_launder();
assert(nr_inactive_pages == 0 && list_empty(&(inactive_list.swap_list)));
assert(mem_map[1] == 1);
rp1 = alloc_page();
assert(rp1 != NULL);
ret = swapfs_read(entry, rp1);
assert(ret == 0);
for (i = 0; i < PGSIZE; i ++) {
assert(((char *)page2kva(rp1))[i] == (char)i);
}
// page fault now
*(char *)0 = 0xEF;
rp0 = pte2page(*ptep0);
assert(page_ref(rp0) == 1);
assert(PageSwap(rp0) && PageActive(rp0));
entry = try_alloc_swap_entry();
assert(swap_offset(entry) == 1 && mem_map[1] == SWAP_UNUSED);
assert(!PageSwap(rp0) && nr_active_pages == 0 && nr_inactive_pages == 0);
// clear accessed flag
assert(rp0 == pte2page(*ptep0));
assert(!PageSwap(rp0));
ret = swap_out_mm(mm, 10);
assert(ret == 0);
assert(!PageSwap(rp0) && (*ptep0 & PTE_P));
// change page table
ret = swap_out_mm(mm, 10);
assert(ret == 1);
assert(*ptep0 == entry && page_ref(rp0) == 0 && mem_map[1] == 1);
count = nr_free_pages();
refill_inactive_scan();
page_launder();
assert(count + 1 == nr_free_pages());
ret = swapfs_read(entry, rp1);
assert(ret == 0 && *(char *)(page2kva(rp1)) == (char)0xEF);
free_page(rp1);
// duplictate *ptep0
ptep1 = get_pte(pgdir, PGSIZE, 0);
assert(ptep1 != NULL && *ptep1 == 0);
swap_duplicate(*ptep0);
*ptep1 = *ptep0;
// page fault again
// update for copy on write
*(char *)1 = 0x88;
*(char *)(PGSIZE) = 0x8F;
*(char *)(PGSIZE + 1) = 0xFF;
assert(pte2page(*ptep0) != pte2page(*ptep1));
assert(*(char *)0 == (char)0xEF);
assert(*(char *)1 == (char)0x88);
assert(*(char *)(PGSIZE) == (char)0x8F);
assert(*(char *)(PGSIZE + 1) == (char)0xFF);
rp0 = pte2page(*ptep0);
rp1 = pte2page(*ptep1);
assert(!PageSwap(rp0) && PageSwap(rp1) && PageActive(rp1));
entry = try_alloc_swap_entry();
assert(!PageSwap(rp0) && !PageSwap(rp1));
assert(swap_offset(entry) == 1 && mem_map[1] == SWAP_UNUSED);
assert(list_empty(&(active_list.swap_list)));
assert(list_empty(&(inactive_list.swap_list)));
page_insert(pgdir, rp0, PGSIZE, perm | PTE_A);
// check swap_out_mm
*(char *)0 = *(char *)PGSIZE = 0xEE;
mm->swap_address = PGSIZE * 2;
ret = swap_out_mm(mm, 2);
assert(ret == 0);
assert((*ptep0 & PTE_P) && !(*ptep0 & PTE_A));
assert((*ptep1 & PTE_P) && !(*ptep1 & PTE_A));
ret = swap_out_mm(mm, 2);
assert(ret == 2);
assert(mem_map[1] == 2 && page_ref(rp0) == 0);
refill_inactive_scan();
page_launder();
assert(mem_map[1] == 2 && swap_hash_find(entry) == NULL);
// check copy entry
swap_remove_entry(entry);
*ptep1 = 0;
assert(mem_map[1] == 1);
swap_entry_t store;
ret = swap_copy_entry(entry, &store);
assert(ret == -E_NO_MEM);
mem_map[2] = SWAP_UNUSED;
ret = swap_copy_entry(entry, &store);
assert(ret == 0 && swap_offset(store) == 2 && mem_map[2] == 0);
mem_map[2] = 1;
*ptep1 = store;
assert(*(char *)PGSIZE == (char)0xEE && *(char *)(PGSIZE + 1)== (char)0x88);
*(char *)PGSIZE = 1, *(char *)(PGSIZE + 1) = 2;
assert(*(char *)0 == (char)0xEE && *(char *)1 == (char)0x88);
ret = swap_in_page(entry, &rp0);
assert(ret == 0);
ret = swap_in_page(store, &rp1);
assert(ret == 0);
assert(rp1 != rp0);
// free memory
swap_list_del(rp0), swap_list_del(rp1);
swap_page_del(rp0), swap_page_del(rp1);
assert(page_ref(rp0) == 1 && page_ref(rp1) == 1);
assert(nr_active_pages == 0 && list_empty(&(active_list.swap_list)));
assert(nr_inactive_pages == 0 && list_empty(&(inactive_list.swap_list)));
for (i = 0; i < HASH_LIST_SIZE; i ++) {
assert(list_empty(hash_list + i));
}
page_remove(pgdir, 0);
page_remove(pgdir, PGSIZE);
free_page(pa2page(PMD_ADDR(*get_pmd(pgdir, 0, 0))));
free_page(pa2page(PUD_ADDR(*get_pud(pgdir, 0, 0))));
free_page(pa2page(PGD_ADDR(*get_pgd(pgdir, 0, 0))));
pgdir[0] = 0;
mm->pgdir = NULL;
mm_destroy(mm);
check_mm_struct = NULL;
assert(nr_active_pages == 0 && nr_inactive_pages == 0);
for (offset = 0; offset < max_swap_offset; offset ++) {
mem_map[offset] = SWAP_UNUSED;
}
assert(nr_free_pages_store == nr_free_pages());
assert(slab_allocated_store == slab_allocated());
cprintf("check_swap() succeeded.\n");
}
static int gfs2_write_jdata_pagevec(struct address_space *mapping,
struct writeback_control *wbc,
struct pagevec *pvec,
int nr_pages, pgoff_t end,
pgoff_t *done_index)
{
struct inode *inode = mapping->host;
struct gfs2_sbd *sdp = GFS2_SB(inode);
unsigned nrblocks = nr_pages * (PAGE_CACHE_SIZE/inode->i_sb->s_blocksize);
int i;
int ret;
ret = gfs2_trans_begin(sdp, nrblocks, nrblocks);
if (ret < 0)
return ret;
for(i = 0; i < nr_pages; i++) {
struct page *page = pvec->pages[i];
/*
* At this point, the page may be truncated or
* invalidated (changing page->mapping to NULL), or
* even swizzled back from swapper_space to tmpfs file
* mapping. However, page->index will not change
* because we have a reference on the page.
*/
if (page->index > end) {
/*
* can't be range_cyclic (1st pass) because
* end == -1 in that case.
*/
ret = 1;
break;
}
*done_index = page->index;
lock_page(page);
if (unlikely(page->mapping != mapping)) {
continue_unlock:
unlock_page(page);
continue;
}
if (!PageDirty(page)) {
/* someone wrote it for us */
goto continue_unlock;
}
if (PageWriteback(page)) {
if (wbc->sync_mode != WB_SYNC_NONE)
wait_on_page_writeback(page);
else
goto continue_unlock;
}
BUG_ON(PageWriteback(page));
if (!clear_page_dirty_for_io(page))
goto continue_unlock;
trace_wbc_writepage(wbc, mapping->backing_dev_info);
ret = __gfs2_jdata_writepage(page, wbc);
if (unlikely(ret)) {
if (ret == AOP_WRITEPAGE_ACTIVATE) {
unlock_page(page);
ret = 0;
} else {
/*
* done_index is set past this page,
* so media errors will not choke
* background writeout for the entire
* file. This has consequences for
* range_cyclic semantics (ie. it may
* not be suitable for data integrity
* writeout).
*/
*done_index = page->index + 1;
ret = 1;
break;
}
}
/*
* We stop writing back only if we are not doing
* integrity sync. In case of integrity sync we have to
* keep going until we have written all the pages
* we tagged for writeback prior to entering this loop.
*/
if (--wbc->nr_to_write <= 0 && wbc->sync_mode == WB_SYNC_NONE) {
ret = 1;
break;
}
}
gfs2_trans_end(sdp);
return ret;
}
static int f2fs_write_data_page(struct page *page,
struct writeback_control *wbc)
{
struct inode *inode = page->mapping->host;
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
loff_t i_size = i_size_read(inode);
const pgoff_t end_index = ((unsigned long long) i_size)
>> PAGE_CACHE_SHIFT;
unsigned offset = 0;
bool need_balance_fs = false;
int err = 0;
struct f2fs_io_info fio = {
.sbi = sbi,
.type = DATA,
.rw = (wbc->sync_mode == WB_SYNC_ALL) ? WRITE_SYNC : WRITE,
.page = page,
.encrypted_page = NULL,
};
trace_f2fs_writepage(page, DATA);
if (page->index < end_index)
goto write;
/*
* If the offset is out-of-range of file size,
* this page does not have to be written to disk.
*/
offset = i_size & (PAGE_CACHE_SIZE - 1);
if ((page->index >= end_index + 1) || !offset)
goto out;
zero_user_segment(page, offset, PAGE_CACHE_SIZE);
write:
if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
goto redirty_out;
if (f2fs_is_drop_cache(inode))
goto out;
if (f2fs_is_volatile_file(inode) && !wbc->for_reclaim &&
available_free_memory(sbi, BASE_CHECK))
goto redirty_out;
/* Dentry blocks are controlled by checkpoint */
if (S_ISDIR(inode->i_mode)) {
if (unlikely(f2fs_cp_error(sbi)))
goto redirty_out;
err = do_write_data_page(&fio);
goto done;
}
/* we should bypass data pages to proceed the kworkder jobs */
if (unlikely(f2fs_cp_error(sbi))) {
SetPageError(page);
goto out;
}
if (!wbc->for_reclaim)
need_balance_fs = true;
else if (has_not_enough_free_secs(sbi, 0))
goto redirty_out;
err = -EAGAIN;
f2fs_lock_op(sbi);
if (f2fs_has_inline_data(inode))
err = f2fs_write_inline_data(inode, page);
if (err == -EAGAIN)
err = do_write_data_page(&fio);
f2fs_unlock_op(sbi);
done:
if (err && err != -ENOENT)
goto redirty_out;
clear_cold_data(page);
out:
inode_dec_dirty_pages(inode);
if (err)
ClearPageUptodate(page);
unlock_page(page);
if (need_balance_fs)
f2fs_balance_fs(sbi);
if (wbc->for_reclaim)
f2fs_submit_merged_bio(sbi, DATA, WRITE);
return 0;
redirty_out:
redirty_page_for_writepage(wbc, page);
return AOP_WRITEPAGE_ACTIVATE;
}
static int __f2fs_writepage(struct page *page, struct writeback_control *wbc,
void *data)
{
struct address_space *mapping = data;
int ret = mapping->a_ops->writepage(page, wbc);
mapping_set_error(mapping, ret);
return ret;
}
/*
* This function was copied from write_cche_pages from mm/page-writeback.c.
* The major change is making write step of cold data page separately from
* warm/hot data page.
*/
static int f2fs_write_cache_pages(struct address_space *mapping,
struct writeback_control *wbc, writepage_t writepage,
void *data)
{
int ret = 0;
int done = 0;
struct pagevec pvec;
int nr_pages;
pgoff_t uninitialized_var(writeback_index);
pgoff_t index;
pgoff_t end; /* Inclusive */
pgoff_t done_index;
int cycled;
int range_whole = 0;
int tag;
int step = 0;
pagevec_init(&pvec, 0);
next:
if (wbc->range_cyclic) {
writeback_index = mapping->writeback_index; /* prev offset */
index = writeback_index;
if (index == 0)
cycled = 1;
else
cycled = 0;
end = -1;
} else {
index = wbc->range_start >> PAGE_CACHE_SHIFT;
end = wbc->range_end >> PAGE_CACHE_SHIFT;
if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
range_whole = 1;
cycled = 1; /* ignore range_cyclic tests */
}
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag = PAGECACHE_TAG_TOWRITE;
else
tag = PAGECACHE_TAG_DIRTY;
retry:
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag_pages_for_writeback(mapping, index, end);
done_index = index;
while (!done && (index <= end)) {
int i;
nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, tag,
min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1);
if (nr_pages == 0)
break;
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
if (page->index > end) {
done = 1;
break;
}
done_index = page->index;
lock_page(page);
if (unlikely(page->mapping != mapping)) {
continue_unlock:
unlock_page(page);
continue;
}
if (!PageDirty(page)) {
/* someone wrote it for us */
goto continue_unlock;
}
if (step == is_cold_data(page))
goto continue_unlock;
if (PageWriteback(page)) {
if (wbc->sync_mode != WB_SYNC_NONE)
f2fs_wait_on_page_writeback(page, DATA);
else
goto continue_unlock;
}
BUG_ON(PageWriteback(page));
if (!clear_page_dirty_for_io(page))
goto continue_unlock;
ret = (*writepage)(page, wbc, data);
if (unlikely(ret)) {
if (ret == AOP_WRITEPAGE_ACTIVATE) {
unlock_page(page);
ret = 0;
} else {
done_index = page->index + 1;
done = 1;
break;
}
}
if (--wbc->nr_to_write <= 0 &&
wbc->sync_mode == WB_SYNC_NONE) {
done = 1;
break;
}
}
pagevec_release(&pvec);
cond_resched();
}
if (step < 1) {
step++;
goto next;
}
if (!cycled && !done) {
cycled = 1;
index = 0;
end = writeback_index - 1;
goto retry;
}
if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
mapping->writeback_index = done_index;
return ret;
}
void __set_page_owner(struct page *page, unsigned int order, gfp_t gfp_mask)
{
struct page_ext *page_ext = lookup_page_ext(page);
struct stack_trace trace = {
.nr_entries = 0,
.max_entries = ARRAY_SIZE(page_ext->trace_entries),
.entries = &page_ext->trace_entries[0],
.skip = 3,
};
save_stack_trace(&trace);
page_ext->order = order;
page_ext->gfp_mask = gfp_mask;
page_ext->nr_entries = trace.nr_entries;
__set_bit(PAGE_EXT_OWNER, &page_ext->flags);
}
static ssize_t
print_page_owner(char __user *buf, size_t count, unsigned long pfn,
struct page *page, struct page_ext *page_ext)
{
int ret;
int pageblock_mt, page_mt;
char *kbuf;
struct stack_trace trace = {
.nr_entries = page_ext->nr_entries,
.entries = &page_ext->trace_entries[0],
};
kbuf = kmalloc(count, GFP_KERNEL);
if (!kbuf)
return -ENOMEM;
ret = snprintf(kbuf, count,
"Page allocated via order %u, mask 0x%x\n",
page_ext->order, page_ext->gfp_mask);
if (ret >= count)
goto err;
/* Print information relevant to grouping pages by mobility */
pageblock_mt = get_pfnblock_migratetype(page, pfn);
page_mt = gfpflags_to_migratetype(page_ext->gfp_mask);
ret += snprintf(kbuf + ret, count - ret,
"PFN %lu Block %lu type %d %s Flags %s%s%s%s%s%s%s%s%s%s%s%s\n",
pfn,
pfn >> pageblock_order,
pageblock_mt,
pageblock_mt != page_mt ? "Fallback" : " ",
PageLocked(page) ? "K" : " ",
PageError(page) ? "E" : " ",
PageReferenced(page) ? "R" : " ",
PageUptodate(page) ? "U" : " ",
PageDirty(page) ? "D" : " ",
PageLRU(page) ? "L" : " ",
PageActive(page) ? "A" : " ",
PageSlab(page) ? "S" : " ",
PageWriteback(page) ? "W" : " ",
PageCompound(page) ? "C" : " ",
PageSwapCache(page) ? "B" : " ",
PageMappedToDisk(page) ? "M" : " ");
if (ret >= count)
goto err;
ret += snprint_stack_trace(kbuf + ret, count - ret, &trace, 0);
if (ret >= count)
goto err;
ret += snprintf(kbuf + ret, count - ret, "\n");
if (ret >= count)
goto err;
if (copy_to_user(buf, kbuf, ret))
ret = -EFAULT;
kfree(kbuf);
return ret;
err:
kfree(kbuf);
return -ENOMEM;
}
static ssize_t
read_page_owner(struct file *file, char __user *buf, size_t count, loff_t *ppos)
{
unsigned long pfn;
struct page *page;
struct page_ext *page_ext;
if (!page_owner_inited)
return -EINVAL;
page = NULL;
pfn = min_low_pfn + *ppos;
/* Find a valid PFN or the start of a MAX_ORDER_NR_PAGES area */
while (!pfn_valid(pfn) && (pfn & (MAX_ORDER_NR_PAGES - 1)) != 0)
pfn++;
drain_all_pages(NULL);
/* Find an allocated page */
for (; pfn < max_pfn; pfn++) {
/*
* If the new page is in a new MAX_ORDER_NR_PAGES area,
* validate the area as existing, skip it if not
*/
if ((pfn & (MAX_ORDER_NR_PAGES - 1)) == 0 && !pfn_valid(pfn)) {
pfn += MAX_ORDER_NR_PAGES - 1;
continue;
}
/* Check for holes within a MAX_ORDER area */
if (!pfn_valid_within(pfn))
continue;
page = pfn_to_page(pfn);
if (PageBuddy(page)) {
unsigned long freepage_order = page_order_unsafe(page);
if (freepage_order < MAX_ORDER)
pfn += (1UL << freepage_order) - 1;
continue;
}
page_ext = lookup_page_ext(page);
/*
* Some pages could be missed by concurrent allocation or free,
* because we don't hold the zone lock.
*/
if (!test_bit(PAGE_EXT_OWNER, &page_ext->flags))
continue;
/* Record the next PFN to read in the file offset */
*ppos = (pfn - min_low_pfn) + 1;
return print_page_owner(buf, count, pfn, page, page_ext);
}
return 0;
}
/*
* write a region of pages back to the server
*/
static int afs_writepages_region(struct address_space *mapping,
struct writeback_control *wbc,
pgoff_t index, pgoff_t end, pgoff_t *_next)
{
struct page *page;
int ret, n;
_enter(",,%lx,%lx,", index, end);
do {
n = find_get_pages_range_tag(mapping, &index, end,
PAGECACHE_TAG_DIRTY, 1, &page);
if (!n)
break;
_debug("wback %lx", page->index);
/*
* at this point we hold neither the i_pages lock nor the
* page lock: the page may be truncated or invalidated
* (changing page->mapping to NULL), or even swizzled
* back from swapper_space to tmpfs file mapping
*/
ret = lock_page_killable(page);
if (ret < 0) {
put_page(page);
_leave(" = %d", ret);
return ret;
}
if (page->mapping != mapping || !PageDirty(page)) {
unlock_page(page);
put_page(page);
continue;
}
if (PageWriteback(page)) {
unlock_page(page);
if (wbc->sync_mode != WB_SYNC_NONE)
wait_on_page_writeback(page);
put_page(page);
continue;
}
if (!clear_page_dirty_for_io(page))
BUG();
ret = afs_write_back_from_locked_page(mapping, wbc, page, end);
put_page(page);
if (ret < 0) {
_leave(" = %d", ret);
return ret;
}
wbc->nr_to_write -= ret;
cond_resched();
} while (index < end && wbc->nr_to_write > 0);
*_next = index;
_leave(" = 0 [%lx]", *_next);
return 0;
}
static unsigned int shrink_pages(struct mm_struct *mm, struct zone **zone_id_0,
struct list_head *zone0_page_list,
struct zone **zone_id_1,
struct list_head *zone1_page_list,
unsigned int num_to_scan)
{
unsigned long addr;
unsigned int isolate_pages_countter = 0;
struct vm_area_struct *vma = mm->mmap;
while (vma != NULL) {
for (addr = vma->vm_start; addr < vma->vm_end;
addr += PAGE_SIZE) {
struct page *page;
/*get the page address from virtual memory address */
page = follow_page(vma, addr, FOLL_GET);
if (page && !IS_ERR(page)) {
put_page(page);
/* only moveable, anonymous and not dirty pages can be swapped */
if ((!PageUnevictable(page))
&& (!PageDirty(page)) && ((PageAnon(page)))
&& (0 == page_is_file_cache(page))) {
switch (page_zone_id(page)) {
case 0:
if (!isolate_lru_page(page)) {
/* isolate page from LRU and add to temp list */
*zone_id_0 =
page_zone(page);
/*create new page list, it will be used in shrink_page_list */
spin_lock_irq(&(*zone_id_0)->lru_lock);
list_add_tail(&page->lru, zone0_page_list);
spin_unlock_irq(&(*zone_id_0)->lru_lock);
isolate_pages_countter++;
}
break;
case 1:
if (!isolate_lru_page(page)) {
/* isolate page from LRU and add to temp list */
*zone_id_1 =
page_zone(page);
/*create new page list, it will be used in shrink_page_list */
spin_lock_irq(&(*zone_id_1)->lru_lock);
list_add_tail(&page->lru, zone1_page_list);
spin_unlock_irq(&(*zone_id_1)->lru_lock);
isolate_pages_countter++;
}
break;
default:
break;
}
}
}
if (isolate_pages_countter >= num_to_scan) {
return isolate_pages_countter;
}
}
vma = vma->vm_next;
}
return isolate_pages_countter;
}
static struct page *follow_page_pte(struct vm_area_struct *vma,
unsigned long address, pmd_t *pmd, unsigned int flags)
{
struct mm_struct *mm = vma->vm_mm;
struct dev_pagemap *pgmap = NULL;
struct page *page;
spinlock_t *ptl;
pte_t *ptep, pte;
retry:
if (unlikely(pmd_bad(*pmd)))
return no_page_table(vma, flags);
ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
pte = *ptep;
if (!pte_present(pte)) {
swp_entry_t entry;
/*
* KSM's break_ksm() relies upon recognizing a ksm page
* even while it is being migrated, so for that case we
* need migration_entry_wait().
*/
if (likely(!(flags & FOLL_MIGRATION)))
goto no_page;
if (pte_none(pte))
goto no_page;
entry = pte_to_swp_entry(pte);
if (!is_migration_entry(entry))
goto no_page;
pte_unmap_unlock(ptep, ptl);
migration_entry_wait(mm, pmd, address);
goto retry;
}
if ((flags & FOLL_NUMA) && pte_protnone(pte))
goto no_page;
if ((flags & FOLL_WRITE) && !pte_write(pte)) {
pte_unmap_unlock(ptep, ptl);
return NULL;
}
page = vm_normal_page(vma, address, pte);
if (!page && pte_devmap(pte) && (flags & FOLL_GET)) {
/*
* Only return device mapping pages in the FOLL_GET case since
* they are only valid while holding the pgmap reference.
*/
pgmap = get_dev_pagemap(pte_pfn(pte), NULL);
if (pgmap)
page = pte_page(pte);
else
goto no_page;
} else if (unlikely(!page)) {
if (flags & FOLL_DUMP) {
/* Avoid special (like zero) pages in core dumps */
page = ERR_PTR(-EFAULT);
goto out;
}
if (is_zero_pfn(pte_pfn(pte))) {
page = pte_page(pte);
} else {
int ret;
ret = follow_pfn_pte(vma, address, ptep, flags);
page = ERR_PTR(ret);
goto out;
}
}
if (flags & FOLL_SPLIT && PageTransCompound(page)) {
int ret;
get_page(page);
pte_unmap_unlock(ptep, ptl);
lock_page(page);
ret = split_huge_page(page);
unlock_page(page);
put_page(page);
if (ret)
return ERR_PTR(ret);
goto retry;
}
if (flags & FOLL_GET) {
get_page(page);
/* drop the pgmap reference now that we hold the page */
if (pgmap) {
put_dev_pagemap(pgmap);
pgmap = NULL;
}
}
if (flags & FOLL_TOUCH) {
if ((flags & FOLL_WRITE) &&
!pte_dirty(pte) && !PageDirty(page))
set_page_dirty(page);
/*
* pte_mkyoung() would be more correct here, but atomic care
* is needed to avoid losing the dirty bit: it is easier to use
* mark_page_accessed().
*/
mark_page_accessed(page);
}
if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) {
/* Do not mlock pte-mapped THP */
if (PageTransCompound(page))
goto out;
/*
* The preliminary mapping check is mainly to avoid the
* pointless overhead of lock_page on the ZERO_PAGE
* which might bounce very badly if there is contention.
*
* If the page is already locked, we don't need to
* handle it now - vmscan will handle it later if and
* when it attempts to reclaim the page.
*/
if (page->mapping && trylock_page(page)) {
lru_add_drain(); /* push cached pages to LRU */
/*
* Because we lock page here, and migration is
* blocked by the pte's page reference, and we
* know the page is still mapped, we don't even
* need to check for file-cache page truncation.
*/
mlock_vma_page(page);
unlock_page(page);
}
}
out:
pte_unmap_unlock(ptep, ptl);
return page;
no_page:
pte_unmap_unlock(ptep, ptl);
if (!pte_none(pte))
return NULL;
return no_page_table(vma, flags);
}
/**
* write_one_page - write out a single page and optionally wait on I/O
* @page: the page to write
* @wait: if true, wait on writeout
*
* The page must be locked by the caller and will be unlocked upon return.
*
* write_one_page() returns a negative error code if I/O failed.
*/
int write_one_page(struct page *page, int wait)
{
struct address_space *mapping = page->mapping;
int ret = 0;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_ALL,
.nr_to_write = 1,
};
BUG_ON(!PageLocked(page));
if (wait)
wait_on_page_writeback(page);
if (clear_page_dirty_for_io(page)) {
page_cache_get(page);
ret = mapping->a_ops->writepage(page, &wbc);
if (ret == 0 && wait) {
wait_on_page_writeback(page);
if (PageError(page))
ret = -EIO;
}
page_cache_release(page);
} else {
unlock_page(page);
}
return ret;
}
EXPORT_SYMBOL(write_one_page);
/*
* For address_spaces which do not use buffers nor write back.
*/
int __set_page_dirty_no_writeback(struct page *page)
{
if (!PageDirty(page))
SetPageDirty(page);
return 0;
}
/*
* Helper function for set_page_dirty family.
* NOTE: This relies on being atomic wrt interrupts.
*/
void account_page_dirtied(struct page *page, struct address_space *mapping)
{
if (mapping_cap_account_dirty(mapping)) {
__inc_zone_page_state(page, NR_FILE_DIRTY);
__inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE);
task_dirty_inc(current);
task_io_account_write(PAGE_CACHE_SIZE);
}
}
/*
* For address_spaces which do not use buffers. Just tag the page as dirty in
* its radix tree.
*
* This is also used when a single buffer is being dirtied: we want to set the
* page dirty in that case, but not all the buffers. This is a "bottom-up"
* dirtying, whereas __set_page_dirty_buffers() is a "top-down" dirtying.
*
* Most callers have locked the page, which pins the address_space in memory.
* But zap_pte_range() does not lock the page, however in that case the
* mapping is pinned by the vma's ->vm_file reference.
*
* We take care to handle the case where the page was truncated from the
* mapping by re-checking page_mapping() inside tree_lock.
*/
int __set_page_dirty_nobuffers(struct page *page)
{
if (!TestSetPageDirty(page)) {
struct address_space *mapping = page_mapping(page);
struct address_space *mapping2;
if (!mapping)
return 1;
spin_lock_irq(&mapping->tree_lock);
mapping2 = page_mapping(page);
if (mapping2) { /* Race with truncate? */
BUG_ON(mapping2 != mapping);
WARN_ON_ONCE(!PagePrivate(page) && !PageUptodate(page));
account_page_dirtied(page, mapping);
radix_tree_tag_set(&mapping->page_tree,
page_index(page), PAGECACHE_TAG_DIRTY);
}
spin_unlock_irq(&mapping->tree_lock);
if (mapping->host) {
/* !PageAnon && !swapper_space */
__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
}
return 1;
}
return 0;
}
EXPORT_SYMBOL(__set_page_dirty_nobuffers);
/*
* When a writepage implementation decides that it doesn't want to write this
* page for some reason, it should redirty the locked page via
* redirty_page_for_writepage() and it should then unlock the page and return 0
*/
int redirty_page_for_writepage(struct writeback_control *wbc, struct page *page)
{
wbc->pages_skipped++;
return __set_page_dirty_nobuffers(page);
}
void ub_io_save_context(struct page *page, size_t bytes_dirtied)
{
struct user_beancounter *ub;
struct page_beancounter *pb, *mapped_pb, *io_pb;
if (unlikely(in_interrupt())) {
WARN_ON_ONCE(1);
return;
}
/*
* FIXME - this can happen from atomic context and
* it's probably not that good to loose some requests
*/
pb = io_pb_alloc();
io_pb = NULL;
spin_lock(&pb_lock);
if (io_debug_precheck_save(page))
goto out_unlock;
mapped_pb = page_pbc(page);
io_pb = iopb_to_pb(mapped_pb);
if (io_pb != NULL) {
/*
* this page has an IO - release it and force a new one
* We could also race with page cleaning - see below
*/
mapped_pb = io_pb->page_pb_list;
put_page_io(page, io_pb);
}
/*
* If the page is mapped we must save the context
* it maps to. If the page isn't mapped we use current
* context as this is a regular write.
*/
if (mapped_pb != NULL)
ub = top_beancounter(mapped_pb->ub);
else
ub = get_io_ub();
if (!PageDirty(page)) {
/*
* race with clear_page_dirty(_for_io) - account
* writes for ub_io_release_context()
*/
if (io_pb != NULL)
io_pb->ub->bytes_wrote += PAGE_CACHE_SIZE;
if (pb != NULL)
io_pb_free(pb);
goto out_unlock;
}
if (pb == NULL) {
ub->bytes_dirty_missed += bytes_dirtied;
goto out_unlock;
}
/*
* the page may become clean here, but the context will be seen
* in ub_io_release_context()
*/
pb->ub = get_beancounter(ub);
pb->page_pb_list = mapped_pb;
ub->bytes_dirtied += bytes_dirtied;
set_page_io(page, pb, mapped_pb);
out_unlock:
spin_unlock(&pb_lock);
if (io_pb != NULL) {
put_beancounter(io_pb->ub);
io_pb_free(io_pb);
}
}
/**
* write_one_page - write out a single page and optionally wait on I/O
* @page: the page to write
* @wait: if true, wait on writeout
*
* The page must be locked by the caller and will be unlocked upon return.
*
* write_one_page() returns a negative error code if I/O failed.
*/
int write_one_page(struct page *page, int wait)
{
struct address_space *mapping = page->mapping;
int ret = 0;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_ALL,
.nr_to_write = 1,
};
BUG_ON(!PageLocked(page));
if (wait)
wait_on_page_writeback(page);
if (clear_page_dirty_for_io(page)) {
page_cache_get(page);
ret = mapping->a_ops->writepage(page, &wbc);
if (ret == 0 && wait) {
wait_on_page_writeback(page);
if (PageError(page))
ret = -EIO;
}
page_cache_release(page);
} else {
unlock_page(page);
}
return ret;
}
EXPORT_SYMBOL(write_one_page);
/*
* For address_spaces which do not use buffers nor write back.
*/
int __set_page_dirty_no_writeback(struct page *page)
{
if (!PageDirty(page))
return !TestSetPageDirty(page);
return 0;
}
/*
* Helper function for set_page_dirty family.
* NOTE: This relies on being atomic wrt interrupts.
*/
void account_page_dirtied(struct page *page, struct address_space *mapping)
{
if (mapping_cap_account_dirty(mapping)) {
__inc_zone_page_state(page, NR_FILE_DIRTY);
__inc_zone_page_state(page, NR_DIRTIED);
__inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE);
task_dirty_inc(current);
task_io_account_write(PAGE_CACHE_SIZE);
}
}
EXPORT_SYMBOL(account_page_dirtied);
/*
* Helper function for set_page_writeback family.
* NOTE: Unlike account_page_dirtied this does not rely on being atomic
* wrt interrupts.
*/
void account_page_writeback(struct page *page)
{
inc_zone_page_state(page, NR_WRITEBACK);
inc_zone_page_state(page, NR_WRITTEN);
}
int write_one_page(struct page *page, int wait)
{
struct address_space *mapping = page->mapping;
int ret = 0;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_ALL,
.nr_to_write = 1,
};
BUG_ON(!PageLocked(page));
if (wait)
wait_on_page_writeback(page);
if (clear_page_dirty_for_io(page)) {
page_cache_get(page);
ret = mapping->a_ops->writepage(page, &wbc);
if (ret == 0 && wait) {
wait_on_page_writeback(page);
if (PageError(page))
ret = -EIO;
}
page_cache_release(page);
} else {
unlock_page(page);
}
return ret;
}
EXPORT_SYMBOL(write_one_page);
int __set_page_dirty_no_writeback(struct page *page)
{
if (!PageDirty(page))
SetPageDirty(page);
return 0;
}
void account_page_dirtied(struct page *page, struct address_space *mapping)
{
if (mapping_cap_account_dirty(mapping)) {
__inc_zone_page_state(page, NR_FILE_DIRTY);
__inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE);
task_dirty_inc(current);
task_io_account_write(PAGE_CACHE_SIZE);
}
}
int __set_page_dirty_nobuffers(struct page *page)
{
if (!TestSetPageDirty(page)) {
struct address_space *mapping = page_mapping(page);
struct address_space *mapping2;
if (!mapping)
return 1;
spin_lock_irq(&mapping->tree_lock);
mapping2 = page_mapping(page);
if (mapping2) { /* Race with truncate? */
BUG_ON(mapping2 != mapping);
WARN_ON_ONCE(!PagePrivate(page) && !PageUptodate(page));
account_page_dirtied(page, mapping);
radix_tree_tag_set(&mapping->page_tree,
page_index(page), PAGECACHE_TAG_DIRTY);
}
spin_unlock_irq(&mapping->tree_lock);
if (mapping->host) {
/* !PageAnon && !swapper_space */
__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
}
return 1;
}
return 0;
}
EXPORT_SYMBOL(__set_page_dirty_nobuffers);
int redirty_page_for_writepage(struct writeback_control *wbc, struct page *page)
{
wbc->pages_skipped++;
return __set_page_dirty_nobuffers(page);
}
