Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 0056fb13 authored by Chao Yu's avatar Chao Yu Committed by Jaegeuk Kim
Browse files

f2fs: add app/fs io stat 



commit b0af6d491a6b5f5622fa91ac75f34f3640f862c4 upstream.

This patch enables inner app/fs io stats and introduces below virtual fs
nodes for exposing stats info:
/sys/fs/f2fs/<dev>/iostat_enable
/proc/fs/f2fs/<dev>/iostat_info

Signed-off-by: default avatarChao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: fix wrong stat assignment]
Signed-off-by: default avatarJaegeuk Kim <jaegeuk@kernel.org>
parent 28530214

fs/f2fs/checkpoint.c

+24 −10
Original line number Diff line number Diff line
@@ -231,8 +231,9 @@ void ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index) 
		ra_meta_pages(sbi, index, BIO_MAX_PAGES, META_POR, true);

}



static int f2fs_write_meta_page(struct page *page,

				struct writeback_control *wbc)

static int __f2fs_write_meta_page(struct page *page,

				struct writeback_control *wbc,

				enum iostat_type io_type)

{

	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
@@ -245,7 +246,7 @@ static int f2fs_write_meta_page(struct page *page, 
	if (unlikely(f2fs_cp_error(sbi)))

		goto redirty_out;



	write_meta_page(sbi, page);

	write_meta_page(sbi, page, io_type);

	dec_page_count(sbi, F2FS_DIRTY_META);



	if (wbc->for_reclaim)
@@ -264,6 +265,12 @@ redirty_out: 
	return AOP_WRITEPAGE_ACTIVATE;

}



static int f2fs_write_meta_page(struct page *page,

				struct writeback_control *wbc)

{

	return __f2fs_write_meta_page(page, wbc, FS_META_IO);

}



static int f2fs_write_meta_pages(struct address_space *mapping,

				struct writeback_control *wbc)

{
@@ -284,7 +291,7 @@ static int f2fs_write_meta_pages(struct address_space *mapping, 


	trace_f2fs_writepages(mapping->host, wbc, META);

	diff = nr_pages_to_write(sbi, META, wbc);

	written = sync_meta_pages(sbi, META, wbc->nr_to_write);

	written = sync_meta_pages(sbi, META, wbc->nr_to_write, FS_META_IO);

	mutex_unlock(&sbi->cp_mutex);

	wbc->nr_to_write = max((long)0, wbc->nr_to_write - written - diff);

	return 0;
@@ -296,7 +303,7 @@ skip_write: 
}



long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,

						long nr_to_write)

				long nr_to_write, enum iostat_type io_type)

{

	struct address_space *mapping = META_MAPPING(sbi);

	pgoff_t index = 0, end = ULONG_MAX, prev = ULONG_MAX;
@@ -347,7 +354,7 @@ continue_unlock: 
			if (!clear_page_dirty_for_io(page))

				goto continue_unlock;



			if (mapping->a_ops->writepage(page, &wbc)) {

			if (__f2fs_write_meta_page(page, &wbc, io_type)) {

				unlock_page(page);

				break;

			}
@@ -905,7 +912,14 @@ retry: 
	if (inode) {

		unsigned long cur_ino = inode->i_ino;



		if (is_dir)

			F2FS_I(inode)->cp_task = current;



		filemap_fdatawrite(inode->i_mapping);



		if (is_dir)

			F2FS_I(inode)->cp_task = NULL;



		iput(inode);

		/* We need to give cpu to another writers. */

		if (ino == cur_ino) {
@@ -1018,7 +1032,7 @@ retry_flush_nodes: 


	if (get_pages(sbi, F2FS_DIRTY_NODES)) {

		up_write(&sbi->node_write);

		err = sync_node_pages(sbi, &wbc, false);

		err = sync_node_pages(sbi, &wbc, false, FS_CP_NODE_IO);

		if (err) {

			up_write(&sbi->node_change);

			f2fs_unlock_all(sbi);
@@ -1116,7 +1130,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc) 


	/* Flush all the NAT/SIT pages */

	while (get_pages(sbi, F2FS_DIRTY_META)) {

		sync_meta_pages(sbi, META, LONG_MAX);

		sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);

		if (unlikely(f2fs_cp_error(sbi)))

			return -EIO;

	}
@@ -1195,7 +1209,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc) 


		/* Flush all the NAT BITS pages */

		while (get_pages(sbi, F2FS_DIRTY_META)) {

			sync_meta_pages(sbi, META, LONG_MAX);

			sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);

			if (unlikely(f2fs_cp_error(sbi)))

				return -EIO;

		}
@@ -1250,7 +1264,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc) 
	percpu_counter_set(&sbi->alloc_valid_block_count, 0);



	/* Here, we only have one bio having CP pack */

	sync_meta_pages(sbi, META_FLUSH, LONG_MAX);

	sync_meta_pages(sbi, META_FLUSH, LONG_MAX, FS_CP_META_IO);



	/* wait for previous submitted meta pages writeback */

	wait_on_all_pages_writeback(sbi);

fs/f2fs/data.c

+26 −9
Original line number Diff line number Diff line
@@ -1472,7 +1472,8 @@ out: 
}



static int __write_data_page(struct page *page, bool *submitted,

				struct writeback_control *wbc)

				struct writeback_control *wbc,

				enum iostat_type io_type)

{

	struct inode *inode = page->mapping->host;

	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
@@ -1493,6 +1494,7 @@ static int __write_data_page(struct page *page, bool *submitted, 
		.encrypted_page = NULL,

		.submitted = false,

		.need_lock = LOCK_RETRY,

		.io_type = io_type,

	};



	trace_f2fs_writepage(page, DATA);
@@ -1599,7 +1601,7 @@ redirty_out: 
static int f2fs_write_data_page(struct page *page,

					struct writeback_control *wbc)

{

	return __write_data_page(page, NULL, wbc);

	return __write_data_page(page, NULL, wbc, FS_DATA_IO);

}



/*
@@ -1608,7 +1610,8 @@ static int f2fs_write_data_page(struct page *page, 
 * warm/hot data page.

 */

static int f2fs_write_cache_pages(struct address_space *mapping,

					struct writeback_control *wbc)

					struct writeback_control *wbc,

					enum iostat_type io_type)

{

	int ret = 0;

	int done = 0;
@@ -1698,7 +1701,7 @@ continue_unlock: 
			if (!clear_page_dirty_for_io(page))

				goto continue_unlock;



			ret = __write_data_page(page, &submitted, wbc);

			ret = __write_data_page(page, &submitted, wbc, io_type);

			if (unlikely(ret)) {

				/*

				 * keep nr_to_write, since vfs uses this to
@@ -1753,8 +1756,9 @@ continue_unlock: 
	return ret;

}



static int f2fs_write_data_pages(struct address_space *mapping,

			    struct writeback_control *wbc)

int __f2fs_write_data_pages(struct address_space *mapping,

						struct writeback_control *wbc,

						enum iostat_type io_type)

{

	struct inode *inode = mapping->host;

	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
@@ -1791,7 +1795,7 @@ static int f2fs_write_data_pages(struct address_space *mapping, 
		goto skip_write;



	blk_start_plug(&plug);

	ret = f2fs_write_cache_pages(mapping, wbc);

	ret = f2fs_write_cache_pages(mapping, wbc, io_type);

	blk_finish_plug(&plug);



	if (wbc->sync_mode == WB_SYNC_ALL)
@@ -1810,6 +1814,16 @@ skip_write: 
	return 0;

}



static int f2fs_write_data_pages(struct address_space *mapping,

			    struct writeback_control *wbc)

{

	struct inode *inode = mapping->host;



	return __f2fs_write_data_pages(mapping, wbc,

			F2FS_I(inode)->cp_task == current ?

			FS_CP_DATA_IO : FS_DATA_IO);

}



static void f2fs_write_failed(struct address_space *mapping, loff_t to)

{

	struct inode *inode = mapping->host;
@@ -2105,11 +2119,14 @@ static ssize_t f2fs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, 
	up_read(&F2FS_I(inode)->dio_rwsem[rw]);



	if (rw & WRITE) {

		if (err > 0)

		if (err > 0) {

			f2fs_update_iostat(F2FS_I_SB(inode), APP_DIRECT_IO,

									err);

			set_inode_flag(inode, FI_UPDATE_WRITE);

		else if (err < 0)

		} else if (err < 0) {

			f2fs_write_failed(mapping, offset + count);

		}

	}



	trace_f2fs_direct_IO_exit(inode, offset, count, rw, err);

fs/f2fs/f2fs.h

+56 −3
Original line number Diff line number Diff line
@@ -665,6 +665,7 @@ struct f2fs_inode_info { 
	f2fs_hash_t chash;		/* hash value of given file name */

	unsigned int clevel;		/* maximum level of given file name */

	struct task_struct *task;	/* lookup and create consistency */

	struct task_struct *cp_task;	/* separate cp/wb IO stats*/

	nid_t i_xattr_nid;		/* node id that contains xattrs */

	loff_t	last_disk_size;		/* lastly written file size */
@@ -969,6 +970,23 @@ enum need_lock_type { 
	LOCK_RETRY,

};



enum iostat_type {

	APP_DIRECT_IO,			/* app direct IOs */

	APP_BUFFERED_IO,		/* app buffered IOs */

	APP_WRITE_IO,			/* app write IOs */

	APP_MAPPED_IO,			/* app mapped IOs */

	FS_DATA_IO,			/* data IOs from kworker/fsync/reclaimer */

	FS_NODE_IO,			/* node IOs from kworker/fsync/reclaimer */

	FS_META_IO,			/* meta IOs from kworker/reclaimer */

	FS_GC_DATA_IO,			/* data IOs from forground gc */

	FS_GC_NODE_IO,			/* node IOs from forground gc */

	FS_CP_DATA_IO,			/* data IOs from checkpoint */

	FS_CP_NODE_IO,			/* node IOs from checkpoint */

	FS_CP_META_IO,			/* meta IOs from checkpoint */

	FS_DISCARD,			/* discard */

	NR_IO_TYPE,

};



struct f2fs_io_info {

	struct f2fs_sb_info *sbi;	/* f2fs_sb_info pointer */

	enum page_type type;	/* contains DATA/NODE/META/META_FLUSH */
@@ -983,6 +1001,7 @@ struct f2fs_io_info { 
	bool submitted;		/* indicate IO submission */

	int need_lock;		/* indicate we need to lock cp_rwsem */

	bool in_list;		/* indicate fio is in io_list */

	enum iostat_type io_type;	/* io type */

};



#define is_read_io(rw) ((rw) == READ)
@@ -1178,6 +1197,11 @@ struct f2fs_sb_info { 
#endif

	spinlock_t stat_lock;			/* lock for stat operations */



	/* For app/fs IO statistics */

	spinlock_t iostat_lock;

	unsigned long long write_iostat[NR_IO_TYPE];

	bool iostat_enable;



	/* For sysfs suppport */

	struct kobject s_kobj;

	struct completion s_kobj_unregister;
@@ -2417,6 +2441,31 @@ static inline int get_extra_isize(struct inode *inode) 
		sizeof((f2fs_inode)->field))			\

		<= (F2FS_OLD_ATTRIBUTE_SIZE + extra_isize))	\



static inline void f2fs_reset_iostat(struct f2fs_sb_info *sbi)

{

	int i;



	spin_lock(&sbi->iostat_lock);

	for (i = 0; i < NR_IO_TYPE; i++)

		sbi->write_iostat[i] = 0;

	spin_unlock(&sbi->iostat_lock);

}



static inline void f2fs_update_iostat(struct f2fs_sb_info *sbi,

			enum iostat_type type, unsigned long long io_bytes)

{

	if (!sbi->iostat_enable)

		return;

	spin_lock(&sbi->iostat_lock);

	sbi->write_iostat[type] += io_bytes;



	if (type == APP_WRITE_IO || type == APP_DIRECT_IO)

		sbi->write_iostat[APP_BUFFERED_IO] =

			sbi->write_iostat[APP_WRITE_IO] -

			sbi->write_iostat[APP_DIRECT_IO];

	spin_unlock(&sbi->iostat_lock);

}



/*

 * file.c

 */
@@ -2544,7 +2593,7 @@ void move_node_page(struct page *node_page, int gc_type); 
int fsync_node_pages(struct f2fs_sb_info *sbi, struct inode *inode,

			struct writeback_control *wbc, bool atomic);

int sync_node_pages(struct f2fs_sb_info *sbi, struct writeback_control *wbc,

			bool do_balance);

			bool do_balance, enum iostat_type io_type);

void build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount);

bool alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid);

void alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid);
@@ -2587,7 +2636,8 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range); 
bool exist_trim_candidates(struct f2fs_sb_info *sbi, struct cp_control *cpc);

struct page *get_sum_page(struct f2fs_sb_info *sbi, unsigned int segno);

void update_meta_page(struct f2fs_sb_info *sbi, void *src, block_t blk_addr);

void write_meta_page(struct f2fs_sb_info *sbi, struct page *page);

void write_meta_page(struct f2fs_sb_info *sbi, struct page *page,

						enum iostat_type io_type);

void write_node_page(unsigned int nid, struct f2fs_io_info *fio);

void write_data_page(struct dnode_of_data *dn, struct f2fs_io_info *fio);

int rewrite_data_page(struct f2fs_io_info *fio);
@@ -2628,7 +2678,7 @@ int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, 
			int type, bool sync);

void ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index);

long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,

			long nr_to_write);

			long nr_to_write, enum iostat_type io_type);

void add_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type);

void remove_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type);

void release_ino_entry(struct f2fs_sb_info *sbi, bool all);
@@ -2682,6 +2732,9 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, 
int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,

			u64 start, u64 len);

void f2fs_set_page_dirty_nobuffers(struct page *page);

int __f2fs_write_data_pages(struct address_space *mapping,

						struct writeback_control *wbc,

						enum iostat_type io_type);

void f2fs_invalidate_page(struct page *page, unsigned int offset,

			unsigned int length);

int f2fs_release_page(struct page *page, gfp_t wait);

fs/f2fs/file.c

+6 −1
Original line number Diff line number Diff line
@@ -101,6 +101,8 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma, 
	if (!PageUptodate(page))

		SetPageUptodate(page);



	f2fs_update_iostat(sbi, APP_MAPPED_IO, F2FS_BLKSIZE);



	trace_f2fs_vm_page_mkwrite(page, DATA);

mapped:

	/* fill the page */
@@ -1799,7 +1801,7 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg) 
		f2fs_stop_checkpoint(sbi, false);

		break;

	case F2FS_GOING_DOWN_METAFLUSH:

		sync_meta_pages(sbi, META, LONG_MAX);

		sync_meta_pages(sbi, META, LONG_MAX, FS_META_IO);

		f2fs_stop_checkpoint(sbi, false);

		break;

	default:
@@ -2477,6 +2479,9 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) 
		blk_start_plug(&plug);

		ret = __generic_file_write_iter(iocb, from);

		blk_finish_plug(&plug);



		if (ret > 0)

			f2fs_update_iostat(F2FS_I_SB(inode), APP_WRITE_IO, ret);

	}

	inode_unlock(inode);

fs/f2fs/gc.c

+3 −0
Original line number Diff line number Diff line
@@ -689,6 +689,8 @@ static void move_encrypted_block(struct inode *inode, block_t bidx, 
	fio.new_blkaddr = newaddr;

	f2fs_submit_page_write(&fio);



	f2fs_update_iostat(fio.sbi, FS_GC_DATA_IO, F2FS_BLKSIZE);



	f2fs_update_data_blkaddr(&dn, newaddr);

	set_inode_flag(inode, FI_APPEND_WRITE);

	if (page->index == 0)
@@ -736,6 +738,7 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type, 
			.page = page,

			.encrypted_page = NULL,

			.need_lock = LOCK_REQ,

			.io_type = FS_GC_DATA_IO,

		};

		bool is_dirty = PageDirty(page);

		int err;