[permalink] [raw]

Subject: [PATCH v2 16/23] f2fs: Convert f2fs_sync_meta_pages() to use filemap_get_folios_tag()

Convert function to use folios. This is in preparation for the removal
of find_get_pages_range_tag().

Initially the function was checking if the previous page index is truly the
previous page i.e. 1 index behind the current page. To convert to folios and
maintain this check we need to make the check
folio->index != prev + folio_nr_pages(previous folio) since we don't know
how many pages are in a folio.

At index i == 0 the check is guaranteed to succeed, so to workaround indexing
bounds we can simply ignore the check for that specific index. This makes the
initial assignment of prev trivial, so I removed that as well.

Also modified a comment in commit_checkpoint for consistency.

Signed-off-by: Vishal Moola (Oracle) <[email protected]>
---
fs/f2fs/checkpoint.c | 49 +++++++++++++++++++++++---------------------
1 file changed, 26 insertions(+), 23 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 8259e0fa97e1..9f6694f7d723 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -377,59 +377,62 @@ long f2fs_sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
{
struct address_space *mapping = META_MAPPING(sbi);
pgoff_t index = 0, prev = ULONG_MAX;
- struct pagevec pvec;
+ struct folio_batch fbatch;
long nwritten = 0;
- int nr_pages;
+ int nr_folios;
struct writeback_control wbc = {
.for_reclaim = 0,
};
struct blk_plug plug;

- pagevec_init(&pvec);
+ folio_batch_init(&fbatch);

blk_start_plug(&plug);

- while ((nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
- PAGECACHE_TAG_DIRTY))) {
+ while ((nr_folios = filemap_get_folios_tag(mapping, &index,
+ (pgoff_t)-1,
+ PAGECACHE_TAG_DIRTY, &fbatch))) {
int i;

- for (i = 0; i < nr_pages; i++) {
- struct page *page = pvec.pages[i];
+ for (i = 0; i < nr_folios; i++) {
+ struct folio *folio = fbatch.folios[i];

- if (prev == ULONG_MAX)
- prev = page->index - 1;
- if (nr_to_write != LONG_MAX && page->index != prev + 1) {
- pagevec_release(&pvec);
+ if (nr_to_write != LONG_MAX && i != 0 &&
+ folio->index != prev +
+ folio_nr_pages(fbatch.folios[i-1])) {
+ folio_batch_release(&fbatch);
goto stop;
}

- lock_page(page);
+ folio_lock(folio);

- if (unlikely(page->mapping != mapping)) {
+ if (unlikely(folio->mapping != mapping)) {
continue_unlock:
- unlock_page(page);
+ folio_unlock(folio);
continue;
}
- if (!PageDirty(page)) {
+ if (!folio_test_dirty(folio)) {
/* someone wrote it for us */
goto continue_unlock;
}

- f2fs_wait_on_page_writeback(page, META, true, true);
+ f2fs_wait_on_page_writeback(&folio->page, META,
+ true, true);

- if (!clear_page_dirty_for_io(page))
+ if (!folio_clear_dirty_for_io(folio))
goto continue_unlock;

- if (__f2fs_write_meta_page(page, &wbc, io_type)) {
- unlock_page(page);
+ if (__f2fs_write_meta_page(&folio->page, &wbc,
+ io_type)) {
+ folio_unlock(folio);
break;
}
- nwritten++;
- prev = page->index;
+ nwritten += folio_nr_pages(folio);
+ prev = folio->index;
if (unlikely(nwritten >= nr_to_write))
break;
}
- pagevec_release(&pvec);
+ folio_batch_release(&fbatch);
cond_resched();
}
stop:
@@ -1381,7 +1384,7 @@ static void commit_checkpoint(struct f2fs_sb_info *sbi,
};

/*
- * pagevec_lookup_tag and lock_page again will take
+ * filemap_get_folios_tag and lock_page again will take
* some extra time. Therefore, f2fs_update_meta_pages and
* f2fs_sync_meta_pages are combined in this function.
*/
--
2.36.1

2022-09-12 18:29:47

by Vishal Moola

[permalink] [raw]

Subject: [PATCH v2 15/23] f2fs: Convert last_fsync_dnode() to use filemap_get_folios_tag()

Convert to use folios. This is in preparation for the removal of
find_get_pages_range_tag().

Signed-off-by: Vishal Moola (Oracle) <[email protected]>
---
fs/f2fs/node.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index cf8665f04c0d..b993be76013e 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -1513,23 +1513,24 @@ static void flush_inline_data(struct f2fs_sb_info *sbi, nid_t ino)
static struct page *last_fsync_dnode(struct f2fs_sb_info *sbi, nid_t ino)
{
pgoff_t index;
- struct pagevec pvec;
+ struct folio_batch fbatch;
struct page *last_page = NULL;
- int nr_pages;
+ int nr_folios;

- pagevec_init(&pvec);
+ folio_batch_init(&fbatch);
index = 0;

- while ((nr_pages = pagevec_lookup_tag(&pvec, NODE_MAPPING(sbi), &index,
- PAGECACHE_TAG_DIRTY))) {
+ while ((nr_folios = filemap_get_folios_tag(NODE_MAPPING(sbi), &index,
+ (pgoff_t)-1, PAGECACHE_TAG_DIRTY,
+ &fbatch))) {
int i;

- for (i = 0; i < nr_pages; i++) {
- struct page *page = pvec.pages[i];
+ for (i = 0; i < nr_folios; i++) {
+ struct page *page = &fbatch.folios[i]->page;

if (unlikely(f2fs_cp_error(sbi))) {
f2fs_put_page(last_page, 0);
- pagevec_release(&pvec);
+ folio_batch_release(&fbatch);
return ERR_PTR(-EIO);
}

@@ -1560,7 +1561,7 @@ static struct page *last_fsync_dnode(struct f2fs_sb_info *sbi, nid_t ino)
last_page = page;
unlock_page(page);
}
- pagevec_release(&pvec);
+ folio_batch_release(&fbatch);
cond_resched();
}
return last_page;
--
2.36.1

2022-09-12 18:30:08

by Vishal Moola

[permalink] [raw]

On Fri, Sep 30, 2022 at 9:25 AM Jeff Layton <[email protected]> wrote:
>
> On Mon, 2022-09-12 at 11:22 -0700, Vishal Moola (Oracle) wrote:
> > Convert function to use folios throughout. This is in preparation for
> > the removal of find_get_pages_range_tag().
> >
> > This change does NOT support large folios. This shouldn't be an issue as
> > of now since ceph only utilizes folios of size 1 anyways, and there is a
> > lot of work to be done on ceph conversions to folios for later patches
> > at some point.
> >
> > Also some minor renaming for consistency.
> >
> > Signed-off-by: Vishal Moola (Oracle) <[email protected]>
> > ---
> > fs/ceph/addr.c | 138 +++++++++++++++++++++++++------------------------
> > 1 file changed, 70 insertions(+), 68 deletions(-)
> >
> > diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
> > index dcf701b05cc1..33dbe55b08be 100644
> > --- a/fs/ceph/addr.c
> > +++ b/fs/ceph/addr.c
> > @@ -792,7 +792,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > struct ceph_vino vino = ceph_vino(inode);
> > pgoff_t index, start_index, end = -1;
> > struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc;
> > - struct pagevec pvec;
> > + struct folio_batch fbatch;
> > int rc = 0;
> > unsigned int wsize = i_blocksize(inode);
> > struct ceph_osd_request *req = NULL;
> > @@ -821,7 +821,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > if (fsc->mount_options->wsize < wsize)
> > wsize = fsc->mount_options->wsize;
> >
> > - pagevec_init(&pvec);
> > + folio_batch_init(&fbatch);
> >
> > start_index = wbc->range_cyclic ? mapping->writeback_index : 0;
> > index = start_index;
> > @@ -869,9 +869,9 @@ static int ceph_writepages_start(struct address_space *mapping,
> >
> > while (!done && index <= end) {
> > int num_ops = 0, op_idx;
> > - unsigned i, pvec_pages, max_pages, locked_pages = 0;
> > + unsigned i, nr_folios, max_pages, locked_pages = 0;
> > struct page **pages = NULL, **data_pages;
> > - struct page *page;
> > + struct folio *folio;
> > pgoff_t strip_unit_end = 0;
> > u64 offset = 0, len = 0;
> > bool from_pool = false;
> > @@ -879,28 +879,28 @@ static int ceph_writepages_start(struct address_space *mapping,
> > max_pages = wsize >> PAGE_SHIFT;
> >
> > get_more_pages:
> > - pvec_pages = pagevec_lookup_range_tag(&pvec, mapping, &index,
> > - end, PAGECACHE_TAG_DIRTY);
> > - dout("pagevec_lookup_range_tag got %d\n", pvec_pages);
> > - if (!pvec_pages && !locked_pages)
> > + nr_folios = filemap_get_folios_tag(mapping, &index,
> > + end, PAGECACHE_TAG_DIRTY, &fbatch);
> > + dout("filemap_get_folios_tag got %d\n", nr_folios);
> > + if (!nr_folios && !locked_pages)
> > break;
> > - for (i = 0; i < pvec_pages && locked_pages < max_pages; i++) {
> > - page = pvec.pages[i];
> > - dout("? %p idx %lu\n", page, page->index);
> > + for (i = 0; i < nr_folios && locked_pages < max_pages; i++) {
> > + folio = fbatch.folios[i];
> > + dout("? %p idx %lu\n", folio, folio->index);
> > if (locked_pages == 0)
> > - lock_page(page); /* first page */
> > - else if (!trylock_page(page))
> > + folio_lock(folio); /* first folio */
> > + else if (!folio_trylock(folio))
> > break;
> >
> > /* only dirty pages, or our accounting breaks */
> > - if (unlikely(!PageDirty(page)) ||
> > - unlikely(page->mapping != mapping)) {
> > - dout("!dirty or !mapping %p\n", page);
> > - unlock_page(page);
> > + if (unlikely(!folio_test_dirty(folio)) ||
> > + unlikely(folio->mapping != mapping)) {
> > + dout("!dirty or !mapping %p\n", folio);
> > + folio_unlock(folio);
> > continue;
> > }
> > /* only if matching snap context */
> > - pgsnapc = page_snap_context(page);
> > + pgsnapc = page_snap_context(&folio->page);
> > if (pgsnapc != snapc) {
> > dout("page snapc %p %lld != oldest %p %lld\n",
> > pgsnapc, pgsnapc->seq, snapc, snapc->seq);
> > @@ -908,11 +908,10 @@ static int ceph_writepages_start(struct address_space *mapping,
> > !ceph_wbc.head_snapc &&
> > wbc->sync_mode != WB_SYNC_NONE)
> > should_loop = true;
> > - unlock_page(page);
> > + folio_unlock(folio);
> > continue;
> > }
> > - if (page_offset(page) >= ceph_wbc.i_size) {
> > - struct folio *folio = page_folio(page);
> > + if (folio_pos(folio) >= ceph_wbc.i_size) {
> >
> > dout("folio at %lu beyond eof %llu\n",
> > folio->index, ceph_wbc.i_size);
> > @@ -924,25 +923,26 @@ static int ceph_writepages_start(struct address_space *mapping,
> > folio_unlock(folio);
> > continue;
> > }
> > - if (strip_unit_end && (page->index > strip_unit_end)) {
> > - dout("end of strip unit %p\n", page);
> > - unlock_page(page);
> > + if (strip_unit_end && (folio->index > strip_unit_end)) {
> > + dout("end of strip unit %p\n", folio);
> > + folio_unlock(folio);
> > break;
> > }
> > - if (PageWriteback(page) || PageFsCache(page)) {
> > + if (folio_test_writeback(folio) ||
> > + folio_test_fscache(folio)) {
> > if (wbc->sync_mode == WB_SYNC_NONE) {
> > - dout("%p under writeback\n", page);
> > - unlock_page(page);
> > + dout("%p under writeback\n", folio);
> > + folio_unlock(folio);
> > continue;
> > }
> > - dout("waiting on writeback %p\n", page);
> > - wait_on_page_writeback(page);
> > - wait_on_page_fscache(page);
> > + dout("waiting on writeback %p\n", folio);
> > + folio_wait_writeback(folio);
> > + folio_wait_fscache(folio);
> > }
> >
> > - if (!clear_page_dirty_for_io(page)) {
> > - dout("%p !clear_page_dirty_for_io\n", page);
> > - unlock_page(page);
> > + if (!folio_clear_dirty_for_io(folio)) {
> > + dout("%p !clear_page_dirty_for_io\n", folio);
> > + folio_unlock(folio);
> > continue;
> > }
> >
> > @@ -958,7 +958,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > u32 xlen;
> >
> > /* prepare async write request */
> > - offset = (u64)page_offset(page);
> > + offset = (u64)folio_pos(folio);
> > ceph_calc_file_object_mapping(&ci->i_layout,
> > offset, wsize,
> > &objnum, &objoff,
> > @@ -966,7 +966,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > len = xlen;
> >
> > num_ops = 1;
> > - strip_unit_end = page->index +
> > + strip_unit_end = folio->index +
> > ((len - 1) >> PAGE_SHIFT);
> >
> > BUG_ON(pages);
> > @@ -981,54 +981,53 @@ static int ceph_writepages_start(struct address_space *mapping,
> > }
> >
> > len = 0;
> > - } else if (page->index !=
> > + } else if (folio->index !=
> > (offset + len) >> PAGE_SHIFT) {
> > if (num_ops >= (from_pool ? CEPH_OSD_SLAB_OPS :
> > CEPH_OSD_MAX_OPS)) {
> > - redirty_page_for_writepage(wbc, page);
> > - unlock_page(page);
> > + folio_redirty_for_writepage(wbc, folio);
> > + folio_unlock(folio);
> > break;
> > }
> >
> > num_ops++;
> > - offset = (u64)page_offset(page);
> > + offset = (u64)folio_pos(folio);
> > len = 0;
> > }
> >
> > - /* note position of first page in pvec */
> > + /* note position of first page in fbatch */
> > dout("%p will write page %p idx %lu\n",
> > - inode, page, page->index);
> > + inode, folio, folio->index);
> >
> > if (atomic_long_inc_return(&fsc->writeback_count) >
> > CONGESTION_ON_THRESH(
> > fsc->mount_options->congestion_kb))
> > fsc->write_congested = true;
> >
> > - pages[locked_pages++] = page;
> > - pvec.pages[i] = NULL;
> > + pages[locked_pages++] = &folio->page;
> > + fbatch.folios[i] = NULL;
> >
> > - len += thp_size(page);
> > + len += folio_size(folio);
> > }
> >
> > /* did we get anything? */
> > if (!locked_pages)
> > - goto release_pvec_pages;
> > + goto release_folio_batches;
> > if (i) {
> > unsigned j, n = 0;
> > - /* shift unused page to beginning of pvec */
> > - for (j = 0; j < pvec_pages; j++) {
> > - if (!pvec.pages[j])
> > + /* shift unused folio to the beginning of fbatch */
> > + for (j = 0; j < nr_folios; j++) {
> > + if (!fbatch.folios[j])
> > continue;
> > if (n < j)
> > - pvec.pages[n] = pvec.pages[j];
> > + fbatch.folios[n] = fbatch.folios[j];
> > n++;
> > }
> > - pvec.nr = n;
> > -
> > - if (pvec_pages && i == pvec_pages &&
> > + fbatch.nr = n;
> > + if (nr_folios && i == nr_folios &&
> > locked_pages < max_pages) {
> > - dout("reached end pvec, trying for more\n");
> > - pagevec_release(&pvec);
> > + dout("reached end of fbatch, trying for more\n");
> > + folio_batch_release(&fbatch);
> > goto get_more_pages;
> > }
> > }
> > @@ -1056,7 +1055,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > BUG_ON(IS_ERR(req));
> > }
> > BUG_ON(len < page_offset(pages[locked_pages - 1]) +
> > - thp_size(page) - offset);
> > + folio_size(folio) - offset);
> >
> > req->r_callback = writepages_finish;
> > req->r_inode = inode;
> > @@ -1098,7 +1097,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > set_page_writeback(pages[i]);
> > if (caching)
> > ceph_set_page_fscache(pages[i]);
> > - len += thp_size(page);
> > + len += folio_size(folio);
> > }
> > ceph_fscache_write_to_cache(inode, offset, len, caching);
> >
> > @@ -1108,7 +1107,7 @@ static int ceph_writepages_start(struct address_space *mapping,
> > /* writepages_finish() clears writeback pages
> > * according to the data length, so make sure
> > * data length covers all locked pages */
> > - u64 min_len = len + 1 - thp_size(page);
> > + u64 min_len = len + 1 - folio_size(folio);
> > len = get_writepages_data_length(inode, pages[i - 1],
> > offset);
> > len = max(len, min_len);
> > @@ -1164,10 +1163,10 @@ static int ceph_writepages_start(struct address_space *mapping,
> > if (wbc->nr_to_write <= 0 && wbc->sync_mode == WB_SYNC_NONE)
> > done = true;
> >
> > -release_pvec_pages:
> > - dout("pagevec_release on %d pages (%p)\n", (int)pvec.nr,
> > - pvec.nr ? pvec.pages[0] : NULL);
> > - pagevec_release(&pvec);
> > +release_folio_batches:
> > + dout("folio_batch_release on %d batches (%p)", (int) fbatch.nr,
> > + fbatch.nr ? fbatch.folios[0] : NULL);
> > + folio_batch_release(&fbatch);
> > }
> >
> > if (should_loop && !done) {
> > @@ -1180,19 +1179,22 @@ static int ceph_writepages_start(struct address_space *mapping,
> > if (wbc->sync_mode != WB_SYNC_NONE &&
> > start_index == 0 && /* all dirty pages were checked */
> > !ceph_wbc.head_snapc) {
> > - struct page *page;
> > + struct folio *folio;
> > unsigned i, nr;
> > index = 0;
> > while ((index <= end) &&
> > - (nr = pagevec_lookup_tag(&pvec, mapping, &index,
> > - PAGECACHE_TAG_WRITEBACK))) {
> > + (nr = filemap_get_folios_tag(mapping, &index,
> > + (pgoff_t)-1,
> > + PAGECACHE_TAG_WRITEBACK,
> > + &fbatch))) {
> > for (i = 0; i < nr; i++) {
> > - page = pvec.pages[i];
> > - if (page_snap_context(page) != snapc)
> > + folio = fbatch.folios[i];
> > + if (page_snap_context(&folio->page) !=
> > + snapc)
> > continue;
> > - wait_on_page_writeback(page);
> > + folio_wait_writeback(folio);
> > }
> > - pagevec_release(&pvec);
> > + folio_batch_release(&fbatch);
> > cond_resched();
> > }
> > }
>
>
> We have some work in progress to add write helpers to netfslib. Once we
> get those in place, we plan to convert ceph to use them. At that point
> ceph_writepages just goes away.
> I think it'd be best to just wait for that and to just ensure that
> netfslib uses filemap_get_folios_tag and the like where appropriate.
> --
> Jeff Layton <[email protected]>

Sounds good, let's do that. That will make the patch a lot simpler, and less
prone to errors. I'll strip this down to the necessary changes in v3.