[PATCH 3/4] f2fs: use wait_for_stable_page to avoid contention

From: Jaegeuk Kim
Date: Sat Jan 23 2016 - 15:17:21 EST


In write_begin, if storage supports stable_page, we don't need to wait for
writeback to update its contents.
This patch introduces to use wait_for_stable_page instead of
wait_on_page_writeback.

Signed-off-by: Jaegeuk Kim <jaegeuk@xxxxxxxxxx>
---
fs/f2fs/checkpoint.c | 4 ++--
fs/f2fs/data.c | 7 ++++---
fs/f2fs/dir.c | 8 ++++----
fs/f2fs/f2fs.h | 2 +-
fs/f2fs/file.c | 6 +++---
fs/f2fs/gc.c | 10 +++++-----
fs/f2fs/inline.c | 16 ++++++++--------
fs/f2fs/inode.c | 4 ++--
fs/f2fs/node.c | 10 +++++-----
fs/f2fs/node.h | 2 +-
fs/f2fs/recovery.c | 2 +-
fs/f2fs/segment.c | 12 ++++++++----
fs/f2fs/xattr.c | 6 +++---
13 files changed, 47 insertions(+), 42 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 3842af9..4b6f9c0 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -39,7 +39,7 @@ repeat:
cond_resched();
goto repeat;
}
- f2fs_wait_on_page_writeback(page, META);
+ f2fs_wait_on_page_writeback(page, META, true);
SetPageUptodate(page);
return page;
}
@@ -232,7 +232,7 @@ static int f2fs_write_meta_page(struct page *page,
if (unlikely(f2fs_cp_error(sbi)))
goto redirty_out;

- f2fs_wait_on_page_writeback(page, META);
+ f2fs_wait_on_page_writeback(page, META, true);
write_meta_page(sbi, page);
dec_page_count(sbi, F2FS_DIRTY_META);
unlock_page(page);
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 9a8edeb..7d9f1d9 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -256,7 +256,7 @@ void set_data_blkaddr(struct dnode_of_data *dn)
struct page *node_page = dn->node_page;
unsigned int ofs_in_node = dn->ofs_in_node;

- f2fs_wait_on_page_writeback(node_page, NODE);
+ f2fs_wait_on_page_writeback(node_page, NODE, true);

rn = F2FS_NODE(node_page);

@@ -1320,7 +1320,8 @@ continue_unlock:

if (PageWriteback(page)) {
if (wbc->sync_mode != WB_SYNC_NONE)
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page,
+ DATA, false);
else
goto continue_unlock;
}
@@ -1548,7 +1549,7 @@ repeat:
}
}

- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, false);

/* wait for GCed encrypted page writeback */
if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index faa7495..8950fc3 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -296,7 +296,7 @@ void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de,
{
enum page_type type = f2fs_has_inline_dentry(dir) ? NODE : DATA;
lock_page(page);
- f2fs_wait_on_page_writeback(page, type);
+ f2fs_wait_on_page_writeback(page, type, true);
de->ino = cpu_to_le32(inode->i_ino);
set_de_type(de, inode->i_mode);
f2fs_dentry_kunmap(dir, page);
@@ -311,7 +311,7 @@ static void init_dent_inode(const struct qstr *name, struct page *ipage)
{
struct f2fs_inode *ri;

- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);

/* copy name info. to this inode page */
ri = F2FS_INODE(ipage);
@@ -598,7 +598,7 @@ start:
++level;
goto start;
add_dentry:
- f2fs_wait_on_page_writeback(dentry_page, DATA);
+ f2fs_wait_on_page_writeback(dentry_page, DATA, true);

if (inode) {
down_write(&F2FS_I(inode)->i_sem);
@@ -709,7 +709,7 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
return f2fs_delete_inline_entry(dentry, page, dir, inode);

lock_page(page);
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);

dentry_blk = page_address(page);
bit_pos = dentry - dentry_blk->dentry;
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 6c08cb1..c4e723b 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1837,7 +1837,7 @@ void f2fs_replace_block(struct f2fs_sb_info *, struct dnode_of_data *,
block_t, block_t, unsigned char, bool);
void allocate_data_block(struct f2fs_sb_info *, struct page *,
block_t, block_t *, struct f2fs_summary *, int);
-void f2fs_wait_on_page_writeback(struct page *, enum page_type);
+void f2fs_wait_on_page_writeback(struct page *, enum page_type, bool);
void f2fs_wait_on_encrypted_page_writeback(struct f2fs_sb_info *, block_t);
void write_data_summaries(struct f2fs_sb_info *, block_t);
void write_node_summaries(struct f2fs_sb_info *, block_t);
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 18ddb1e..b19617e 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -86,7 +86,7 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
trace_f2fs_vm_page_mkwrite(page, DATA);
mapped:
/* fill the page */
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, false);

/* wait for GCed encrypted page writeback */
if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
@@ -521,7 +521,7 @@ static int truncate_partial_data_page(struct inode *inode, u64 from,
if (IS_ERR(page))
return 0;
truncate_out:
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);
zero_user(page, offset, PAGE_CACHE_SIZE - offset);
if (!cache_only || !f2fs_encrypted_inode(inode) || !S_ISREG(inode->i_mode))
set_page_dirty(page);
@@ -743,7 +743,7 @@ static int fill_zero(struct inode *inode, pgoff_t index,
if (IS_ERR(page))
return PTR_ERR(page);

- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);
zero_user(page, start, len);
set_page_dirty(page);
f2fs_put_page(page, 1);
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index f610c2a..0204433 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -446,7 +446,7 @@ next_step:

/* set page dirty and write it */
if (gc_type == FG_GC) {
- f2fs_wait_on_page_writeback(node_page, NODE);
+ f2fs_wait_on_page_writeback(node_page, NODE, true);
set_page_dirty(node_page);
} else {
if (!PageWriteback(node_page))
@@ -567,7 +567,7 @@ static void move_encrypted_block(struct inode *inode, block_t bidx)
* don't cache encrypted data into meta inode until previous dirty
* data were writebacked to avoid racing between GC and flush.
*/
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);

get_node_info(fio.sbi, dn.nid, &ni);
set_summary(&sum, dn.nid, dn.ofs_in_node, ni.version);
@@ -596,14 +596,14 @@ static void move_encrypted_block(struct inode *inode, block_t bidx)
goto put_page_out;

set_page_dirty(fio.encrypted_page);
- f2fs_wait_on_page_writeback(fio.encrypted_page, DATA);
+ f2fs_wait_on_page_writeback(fio.encrypted_page, DATA, true);
if (clear_page_dirty_for_io(fio.encrypted_page))
dec_page_count(fio.sbi, F2FS_DIRTY_META);

set_page_writeback(fio.encrypted_page);

/* allocate block address */
- f2fs_wait_on_page_writeback(dn.node_page, NODE);
+ f2fs_wait_on_page_writeback(dn.node_page, NODE, true);
allocate_data_block(fio.sbi, NULL, fio.blk_addr,
&fio.blk_addr, &sum, CURSEG_COLD_DATA);
fio.rw = WRITE_SYNC;
@@ -645,7 +645,7 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type)
.encrypted_page = NULL,
};
set_page_dirty(page);
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);
if (clear_page_dirty_for_io(page))
inode_dec_dirty_pages(inode);
set_cold_data(page);
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index c3f0b7d..8df13e5 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -71,7 +71,7 @@ bool truncate_inline_inode(struct page *ipage, u64 from)

addr = inline_data_addr(ipage);

- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);
memset(addr + from, 0, MAX_INLINE_DATA - from);

return true;
@@ -124,7 +124,7 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page)
if (err)
return err;

- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);

if (PageUptodate(page))
goto no_update;
@@ -150,7 +150,7 @@ no_update:
write_data_page(dn, &fio);
set_data_blkaddr(dn);
f2fs_update_extent_cache(dn);
- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);
if (dirty)
inode_dec_dirty_pages(dn->inode);

@@ -223,7 +223,7 @@ int f2fs_write_inline_data(struct inode *inode, struct page *page)

f2fs_bug_on(F2FS_I_SB(inode), page->index);

- f2fs_wait_on_page_writeback(dn.inode_page, NODE);
+ f2fs_wait_on_page_writeback(dn.inode_page, NODE, true);
src_addr = kmap_atomic(page);
dst_addr = inline_data_addr(dn.inode_page);
memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
@@ -261,7 +261,7 @@ process_inline:
ipage = get_node_page(sbi, inode->i_ino);
f2fs_bug_on(sbi, IS_ERR(ipage));

- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);

src_addr = inline_data_addr(npage);
dst_addr = inline_data_addr(ipage);
@@ -389,7 +389,7 @@ static int f2fs_convert_inline_dir(struct inode *dir, struct page *ipage,
if (err)
goto out;

- f2fs_wait_on_page_writeback(page, DATA);
+ f2fs_wait_on_page_writeback(page, DATA, true);
zero_user_segment(page, MAX_INLINE_DATA, PAGE_CACHE_SIZE);

dentry_blk = kmap_atomic(page);
@@ -469,7 +469,7 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *name,
}
}

- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);

name_hash = f2fs_dentry_hash(name);
make_dentry_ptr(NULL, &d, (void *)dentry_blk, 2);
@@ -507,7 +507,7 @@ void f2fs_delete_inline_entry(struct f2fs_dir_entry *dentry, struct page *page,
int i;

lock_page(page);
- f2fs_wait_on_page_writeback(page, NODE);
+ f2fs_wait_on_page_writeback(page, NODE, true);

inline_dentry = inline_data_addr(page);
bit_pos = dentry - inline_dentry->dentry;
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index 2adeff2..229e11b 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -83,7 +83,7 @@ static void __recover_inline_status(struct inode *inode, struct page *ipage)

while (start < end) {
if (*start++) {
- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);

set_inode_flag(F2FS_I(inode), FI_DATA_EXIST);
set_raw_inline(F2FS_I(inode), F2FS_INODE(ipage));
@@ -227,7 +227,7 @@ int update_inode(struct inode *inode, struct page *node_page)
{
struct f2fs_inode *ri;

- f2fs_wait_on_page_writeback(node_page, NODE);
+ f2fs_wait_on_page_writeback(node_page, NODE, true);

ri = F2FS_INODE(node_page);

diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index 6369281..23b800d 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -861,7 +861,7 @@ skip_partial:
f2fs_put_page(page, 1);
goto restart;
}
- f2fs_wait_on_page_writeback(page, NODE);
+ f2fs_wait_on_page_writeback(page, NODE, true);
ri->i_nid[offset[0] - NODE_DIR1_BLOCK] = 0;
set_page_dirty(page);
unlock_page(page);
@@ -976,7 +976,7 @@ struct page *new_node_page(struct dnode_of_data *dn,
new_ni.ino = dn->inode->i_ino;
set_node_addr(sbi, &new_ni, NEW_ADDR, false);

- f2fs_wait_on_page_writeback(page, NODE);
+ f2fs_wait_on_page_writeback(page, NODE, true);
fill_node_footer(page, dn->nid, dn->inode->i_ino, ofs, true);
set_cold_node(dn->inode, page);
SetPageUptodate(page);
@@ -1287,7 +1287,7 @@ int wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, nid_t ino)
continue;

if (ino && ino_of_node(page) == ino) {
- f2fs_wait_on_page_writeback(page, NODE);
+ f2fs_wait_on_page_writeback(page, NODE, true);
if (TestClearPageError(page))
ret = -EIO;
}
@@ -1326,7 +1326,7 @@ static int f2fs_write_node_page(struct page *page,
if (unlikely(f2fs_cp_error(sbi)))
goto redirty_out;

- f2fs_wait_on_page_writeback(page, NODE);
+ f2fs_wait_on_page_writeback(page, NODE, true);

/* get old block addr of this node page */
nid = nid_of_node(page);
@@ -1703,7 +1703,7 @@ void recover_inline_xattr(struct inode *inode, struct page *page)
src_addr = inline_xattr_addr(page);
inline_size = inline_xattr_size(inode);

- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);
memcpy(dst_addr, src_addr, inline_size);
update_inode:
update_inode(inode, ipage);
diff --git a/fs/f2fs/node.h b/fs/f2fs/node.h
index 56c4519..23bd992 100644
--- a/fs/f2fs/node.h
+++ b/fs/f2fs/node.h
@@ -330,7 +330,7 @@ static inline int set_nid(struct page *p, int off, nid_t nid, bool i)
{
struct f2fs_node *rn = F2FS_NODE(p);

- f2fs_wait_on_page_writeback(p, NODE);
+ f2fs_wait_on_page_writeback(p, NODE, true);

if (i)
rn->i.i_nid[off - NODE_DIR1_BLOCK] = cpu_to_le32(nid);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 589b20b..5a8fd4a 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -418,7 +418,7 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
if (err)
goto out;

- f2fs_wait_on_page_writeback(dn.node_page, NODE);
+ f2fs_wait_on_page_writeback(dn.node_page, NODE, true);

get_node_info(sbi, dn.nid, &ni);
f2fs_bug_on(sbi, ni.ino != ino_of_node(page));
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 53baf4a..e21c896 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -223,7 +223,8 @@ int commit_inmem_pages(struct inode *inode, bool abort)
if (!abort) {
if (cur->page->mapping == inode->i_mapping) {
set_page_dirty(cur->page);
- f2fs_wait_on_page_writeback(cur->page, DATA);
+ f2fs_wait_on_page_writeback(cur->page, DATA,
+ true);
if (clear_page_dirty_for_io(cur->page))
inode_dec_dirty_pages(inode);
trace_f2fs_commit_inmem_page(cur->page, INMEM);
@@ -1417,14 +1418,17 @@ void f2fs_replace_block(struct f2fs_sb_info *sbi, struct dnode_of_data *dn,
}

void f2fs_wait_on_page_writeback(struct page *page,
- enum page_type type)
+ enum page_type type, bool ordered)
{
if (PageWriteback(page)) {
struct f2fs_sb_info *sbi = F2FS_P_SB(page);

if (is_merged_page(sbi, page, type))
f2fs_submit_merged_bio(sbi, type, WRITE);
- wait_on_page_writeback(page);
+ if (ordered)
+ wait_on_page_writeback(page);
+ else
+ wait_for_stable_page(page);
}
}

@@ -1440,7 +1444,7 @@ void f2fs_wait_on_encrypted_page_writeback(struct f2fs_sb_info *sbi,

cpage = find_lock_page(META_MAPPING(sbi), blkaddr);
if (cpage) {
- f2fs_wait_on_page_writeback(cpage, DATA);
+ f2fs_wait_on_page_writeback(cpage, DATA, true);
f2fs_put_page(cpage, 1);
}
}
diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index 10f1e78..06a72dc 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -300,7 +300,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,

if (ipage) {
inline_addr = inline_xattr_addr(ipage);
- f2fs_wait_on_page_writeback(ipage, NODE);
+ f2fs_wait_on_page_writeback(ipage, NODE, true);
} else {
page = get_node_page(sbi, inode->i_ino);
if (IS_ERR(page)) {
@@ -308,7 +308,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
return PTR_ERR(page);
}
inline_addr = inline_xattr_addr(page);
- f2fs_wait_on_page_writeback(page, NODE);
+ f2fs_wait_on_page_writeback(page, NODE, true);
}
memcpy(inline_addr, txattr_addr, inline_size);
f2fs_put_page(page, 1);
@@ -329,7 +329,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
return PTR_ERR(xpage);
}
f2fs_bug_on(sbi, new_nid);
- f2fs_wait_on_page_writeback(xpage, NODE);
+ f2fs_wait_on_page_writeback(xpage, NODE, true);
} else {
struct dnode_of_data dn;
set_new_dnode(&dn, inode, NULL, NULL, new_nid);
--
2.6.3