RE: [f2fs-dev] [PATCH v2] f2fs: refactor bio-related operations

[Chao Yu <chao2.yu@samsung.com>]

> -----Original Message-----
> From: Jaegeuk Kim [mailto:jaegeuk.kim@samsung.com]
> Sent: Monday, December 02, 2013 4:27 PM
> To: linux-fsdevel@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org; linux-f2fs-devel@lists.sourceforge.net
> Subject: Re: [f2fs-dev] [PATCH v2] f2fs: refactor bio-related operations
> 
> Change log from v1:
>  o remove redundant codes
> 
> >From a480dfc915490f4bca7275f6fbb44fa34aa00eaa Mon Sep 17 00:00:00 2001
> From: Jaegeuk Kim <jaegeuk.kim@samsung.com>
> Date: Sat, 30 Nov 2013 12:51:14 +0900
> Subject: [PATCH] f2fs: refactor bio-related operations
> Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net
> 
> This patch integrates redundant bio operations on read and write IOs.
> 
> 1. Move bio-related codes to the top of data.c.
> 2. Replace f2fs_submit_bio with f2fs_submit_merged_bio, which handles read
>    bios additionally.
> 3. Introduce __submit_merged_bio to submit the merged bio.
> 4. Change f2fs_readpage to f2fs_submit_page_bio.
> 5. Introduce f2fs_submit_page_mbio to integrate previous submit_read_page and
>    submit_write_page.

Good job!

> 
> Reviewed-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
> Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>

Reviewed-by: Chao Yu <chao2.yu@samsung.com >

> ---
>  fs/f2fs/checkpoint.c        |  14 +-
>  fs/f2fs/data.c              | 316 +++++++++++++++++++++++++++++---------------
>  fs/f2fs/f2fs.h              |  12 +-
>  fs/f2fs/gc.c                |   2 +-
>  fs/f2fs/node.c              |  14 +-
>  fs/f2fs/recovery.c          |   4 +-
>  fs/f2fs/segment.c           | 164 +++--------------------
>  include/trace/events/f2fs.h |  30 ++---
>  8 files changed, 257 insertions(+), 299 deletions(-)
> 
> diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
> index 40eea42..38f4a224 100644
> --- a/fs/f2fs/checkpoint.c
> +++ b/fs/f2fs/checkpoint.c
> @@ -61,7 +61,8 @@ repeat:
>  	if (PageUptodate(page))
>  		goto out;
> 
> -	if (f2fs_readpage(sbi, page, index, READ_SYNC | REQ_META | REQ_PRIO))
> +	if (f2fs_submit_page_bio(sbi, page, index,
> +				READ_SYNC | REQ_META | REQ_PRIO))
>  		goto repeat;
> 
>  	lock_page(page);
> @@ -157,7 +158,8 @@ long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
>  	}
> 
>  	if (nwritten)
> -		f2fs_submit_bio(sbi, type, nr_to_write == LONG_MAX);
> +		f2fs_submit_merged_bio(sbi, type, nr_to_write == LONG_MAX,
> +								WRITE);
> 
>  	return nwritten;
>  }
> @@ -590,7 +592,7 @@ retry:
>  		 * We should submit bio, since it exists several
>  		 * wribacking dentry pages in the freeing inode.
>  		 */
> -		f2fs_submit_bio(sbi, DATA, true);
> +		f2fs_submit_merged_bio(sbi, DATA, true, WRITE);
>  	}
>  	goto retry;
>  }
> @@ -796,9 +798,9 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
> 
>  	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish block_ops");
> 
> -	f2fs_submit_bio(sbi, DATA, true);
> -	f2fs_submit_bio(sbi, NODE, true);
> -	f2fs_submit_bio(sbi, META, true);
> +	f2fs_submit_merged_bio(sbi, DATA, true, WRITE);
> +	f2fs_submit_merged_bio(sbi, NODE, true, WRITE);
> +	f2fs_submit_merged_bio(sbi, META, true, WRITE);
> 
>  	/*
>  	 * update checkpoint pack index
> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> index c9a76f8..4e2fc09 100644
> --- a/fs/f2fs/data.c
> +++ b/fs/f2fs/data.c
> @@ -25,6 +25,204 @@
>  #include <trace/events/f2fs.h>
> 
>  /*
> + * Low-level block read/write IO operations.
> + */
> +static struct bio *__bio_alloc(struct block_device *bdev, int npages)
> +{
> +	struct bio *bio;
> +
> +	/* No failure on bio allocation */
> +	bio = bio_alloc(GFP_NOIO, npages);
> +	bio->bi_bdev = bdev;
> +	bio->bi_private = NULL;
> +	return bio;
> +}
> +
> +static void f2fs_read_end_io(struct bio *bio, int err)
> +{
> +	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
> +	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
> +
> +	do {
> +		struct page *page = bvec->bv_page;
> +
> +		if (--bvec >= bio->bi_io_vec)
> +			prefetchw(&bvec->bv_page->flags);
> +
> +		if (uptodate) {
> +			SetPageUptodate(page);
> +		} else {
> +			ClearPageUptodate(page);
> +			SetPageError(page);
> +		}
> +		unlock_page(page);
> +	} while (bvec >= bio->bi_io_vec);
> +
> +	bio_put(bio);
> +}
> +
> +static void f2fs_write_end_io(struct bio *bio, int err)
> +{
> +	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
> +	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
> +	struct f2fs_sb_info *sbi = F2FS_SB(bvec->bv_page->mapping->host->i_sb);
> +
> +	do {
> +		struct page *page = bvec->bv_page;
> +
> +		if (--bvec >= bio->bi_io_vec)
> +			prefetchw(&bvec->bv_page->flags);
> +
> +		if (!uptodate) {
> +			SetPageError(page);
> +			set_bit(AS_EIO, &page->mapping->flags);
> +			set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
> +			sbi->sb->s_flags |= MS_RDONLY;
> +		}
> +		end_page_writeback(page);
> +		dec_page_count(sbi, F2FS_WRITEBACK);
> +	} while (bvec >= bio->bi_io_vec);
> +
> +	if (bio->bi_private)
> +		complete(bio->bi_private);
> +
> +	if (!get_pages(sbi, F2FS_WRITEBACK) &&
> +			!list_empty(&sbi->cp_wait.task_list))
> +		wake_up(&sbi->cp_wait);
> +
> +	bio_put(bio);
> +}
> +
> +static void __submit_merged_bio(struct f2fs_sb_info *sbi,
> +				struct f2fs_bio_info *io,
> +				enum page_type type, bool sync, int rw)
> +{
> +	enum page_type btype = PAGE_TYPE_OF_BIO(type);
> +
> +	if (!io->bio)
> +		return;
> +
> +	if (btype == META)
> +		rw |= REQ_META;
> +
> +	if (is_read_io(rw)) {
> +		if (sync)
> +			rw |= READ_SYNC;
> +		submit_bio(rw, io->bio);
> +		trace_f2fs_submit_read_bio(sbi->sb, rw, type, io->bio);
> +		io->bio = NULL;
> +		return;
> +	}
> +
> +	if (sync)
> +		rw |= WRITE_SYNC;
> +	if (type >= META_FLUSH)
> +		rw |= WRITE_FLUSH_FUA;
> +
> +	/*
> +	 * META_FLUSH is only from the checkpoint procedure, and we should wait
> +	 * this metadata bio for FS consistency.
> +	 */
> +	if (type == META_FLUSH) {
> +		DECLARE_COMPLETION_ONSTACK(wait);
> +		io->bio->bi_private = &wait;
> +		submit_bio(rw, io->bio);
> +		wait_for_completion(&wait);
> +	} else {
> +		submit_bio(rw, io->bio);
> +	}
> +	trace_f2fs_submit_write_bio(sbi->sb, rw, btype, io->bio);
> +	io->bio = NULL;
> +}
> +
> +void f2fs_submit_merged_bio(struct f2fs_sb_info *sbi,
> +				enum page_type type, bool sync, int rw)
> +{
> +	enum page_type btype = PAGE_TYPE_OF_BIO(type);
> +	struct f2fs_bio_info *io;
> +
> +	io = is_read_io(rw) ? &sbi->read_io : &sbi->write_io[btype];
> +
> +	mutex_lock(&io->io_mutex);
> +	__submit_merged_bio(sbi, io, type, sync, rw);
> +	mutex_unlock(&io->io_mutex);
> +}
> +
> +/*
> + * Fill the locked page with data located in the block address.
> + * Return unlocked page.
> + */
> +int f2fs_submit_page_bio(struct f2fs_sb_info *sbi, struct page *page,
> +					block_t blk_addr, int rw)
> +{
> +	struct block_device *bdev = sbi->sb->s_bdev;
> +	struct bio *bio;
> +
> +	trace_f2fs_submit_page_bio(page, blk_addr, rw);
> +
> +	/* Allocate a new bio */
> +	bio = __bio_alloc(bdev, 1);
> +
> +	/* Initialize the bio */
> +	bio->bi_sector = SECTOR_FROM_BLOCK(sbi, blk_addr);
> +	bio->bi_end_io = is_read_io(rw) ? f2fs_read_end_io : f2fs_write_end_io;
> +
> +	if (bio_add_page(bio, page, PAGE_CACHE_SIZE, 0) < PAGE_CACHE_SIZE) {
> +		bio_put(bio);
> +		f2fs_put_page(page, 1);
> +		return -EFAULT;
> +	}
> +
> +	submit_bio(rw, bio);
> +	return 0;
> +}
> +
> +void f2fs_submit_page_mbio(struct f2fs_sb_info *sbi, struct page *page,
> +			block_t blk_addr, enum page_type type, int rw)
> +{
> +	enum page_type btype = PAGE_TYPE_OF_BIO(type);
> +	struct block_device *bdev = sbi->sb->s_bdev;
> +	struct f2fs_bio_info *io;
> +	int bio_blocks;
> +
> +	io = is_read_io(rw) ? &sbi->read_io : &sbi->write_io[btype];
> +
> +	verify_block_addr(sbi, blk_addr);
> +
> +	mutex_lock(&io->io_mutex);
> +
> +	if (!is_read_io(rw))
> +		inc_page_count(sbi, F2FS_WRITEBACK);
> +
> +	if (io->bio && io->last_block_in_bio != blk_addr - 1)
> +		__submit_merged_bio(sbi, io, type, true, rw);
> +alloc_new:
> +	if (io->bio == NULL) {
> +		bio_blocks = MAX_BIO_BLOCKS(max_hw_blocks(sbi));
> +		io->bio = __bio_alloc(bdev, bio_blocks);
> +		io->bio->bi_sector = SECTOR_FROM_BLOCK(sbi, blk_addr);
> +		io->bio->bi_end_io = is_read_io(rw) ? f2fs_read_end_io :
> +							f2fs_write_end_io;
> +		/*
> +		 * The end_io will be assigned at the sumbission phase.
> +		 * Until then, let bio_add_page() merge consecutive IOs as much
> +		 * as possible.
> +		 */
> +	}
> +
> +	if (bio_add_page(io->bio, page, PAGE_CACHE_SIZE, 0) <
> +							PAGE_CACHE_SIZE) {
> +		__submit_merged_bio(sbi, io, type, true, rw);
> +		goto alloc_new;
> +	}
> +
> +	io->last_block_in_bio = blk_addr;
> +
> +	mutex_unlock(&io->io_mutex);
> +	trace_f2fs_submit_page_mbio(page, rw, type, blk_addr);
> +}
> +
> +/*
>   * Lock ordering for the change of data block address:
>   * ->data_page
>   *  ->node_page
> @@ -238,7 +436,7 @@ struct page *find_data_page(struct inode *inode, pgoff_t index, bool sync)
>  		return page;
>  	}
> 
> -	err = f2fs_readpage(sbi, page, dn.data_blkaddr,
> +	err = f2fs_submit_page_bio(sbi, page, dn.data_blkaddr,
>  					sync ? READ_SYNC : READA);
>  	if (err)
>  		return ERR_PTR(err);
> @@ -299,7 +497,7 @@ repeat:
>  		return page;
>  	}
> 
> -	err = f2fs_readpage(sbi, page, dn.data_blkaddr, READ_SYNC);
> +	err = f2fs_submit_page_bio(sbi, page, dn.data_blkaddr, READ_SYNC);
>  	if (err)
>  		return ERR_PTR(err);
> 
> @@ -349,7 +547,8 @@ repeat:
>  		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
>  		SetPageUptodate(page);
>  	} else {
> -		err = f2fs_readpage(sbi, page, dn.data_blkaddr, READ_SYNC);
> +		err = f2fs_submit_page_bio(sbi, page, dn.data_blkaddr,
> +								READ_SYNC);
>  		if (err)
>  			return ERR_PTR(err);
>  		lock_page(page);
> @@ -373,110 +572,6 @@ repeat:
>  	return page;
>  }
> 
> -static void read_end_io(struct bio *bio, int err)
> -{
> -	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
> -	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
> -
> -	do {
> -		struct page *page = bvec->bv_page;
> -
> -		if (--bvec >= bio->bi_io_vec)
> -			prefetchw(&bvec->bv_page->flags);
> -
> -		if (uptodate) {
> -			SetPageUptodate(page);
> -		} else {
> -			ClearPageUptodate(page);
> -			SetPageError(page);
> -		}
> -		unlock_page(page);
> -	} while (bvec >= bio->bi_io_vec);
> -	bio_put(bio);
> -}
> -
> -/*
> - * Fill the locked page with data located in the block address.
> - * Return unlocked page.
> - */
> -int f2fs_readpage(struct f2fs_sb_info *sbi, struct page *page,
> -					block_t blk_addr, int type)
> -{
> -	struct block_device *bdev = sbi->sb->s_bdev;
> -	struct bio *bio;
> -
> -	trace_f2fs_readpage(page, blk_addr, type);
> -
> -	/* Allocate a new bio */
> -	bio = f2fs_bio_alloc(bdev, 1);
> -
> -	/* Initialize the bio */
> -	bio->bi_sector = SECTOR_FROM_BLOCK(sbi, blk_addr);
> -	bio->bi_end_io = read_end_io;
> -
> -	if (bio_add_page(bio, page, PAGE_CACHE_SIZE, 0) < PAGE_CACHE_SIZE) {
> -		bio_put(bio);
> -		f2fs_put_page(page, 1);
> -		return -EFAULT;
> -	}
> -
> -	submit_bio(type, bio);
> -	return 0;
> -}
> -
> -void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, int rw)
> -{
> -	struct f2fs_bio_info *io = &sbi->read_io;
> -
> -	if (!io->bio)
> -		return;
> -
> -	trace_f2fs_submit_read_bio(sbi->sb, rw, META, io->bio);
> -
> -	mutex_lock(&io->io_mutex);
> -	if (io->bio) {
> -		submit_bio(rw, io->bio);
> -		io->bio = NULL;
> -	}
> -	mutex_unlock(&io->io_mutex);
> -}
> -
> -void submit_read_page(struct f2fs_sb_info *sbi, struct page *page,
> -					block_t blk_addr, int rw)
> -{
> -	struct block_device *bdev = sbi->sb->s_bdev;
> -	struct f2fs_bio_info *io = &sbi->read_io;
> -	int bio_blocks;
> -
> -	verify_block_addr(sbi, blk_addr);
> -
> -	mutex_lock(&io->io_mutex);
> -
> -	if (io->bio && io->last_block_in_bio != blk_addr - 1) {
> -		submit_bio(rw, io->bio);
> -		io->bio = NULL;
> -	}
> -alloc_new:
> -	if (io->bio == NULL) {
> -		bio_blocks = MAX_BIO_BLOCKS(max_hw_blocks(sbi));
> -		io->bio = f2fs_bio_alloc(bdev, bio_blocks);
> -		io->bio->bi_sector = SECTOR_FROM_BLOCK(sbi, blk_addr);
> -		io->bio->bi_end_io = read_end_io;
> -	}
> -
> -	if (bio_add_page(io->bio, page, PAGE_CACHE_SIZE, 0) <
> -							PAGE_CACHE_SIZE) {
> -		submit_bio(rw, io->bio);
> -		io->bio = NULL;
> -		goto alloc_new;
> -	}
> -
> -	io->last_block_in_bio = blk_addr;
> -
> -	mutex_unlock(&io->io_mutex);
> -	trace_f2fs_submit_read_page(page, rw, META, blk_addr);
> -}
> -
>  /*
>   * This function should be used by the data read flow only where it
>   * does not check the "create" flag that indicates block allocation.
> @@ -638,7 +733,7 @@ write:
>  		goto redirty_out;
> 
>  	if (wbc->for_reclaim)
> -		f2fs_submit_bio(sbi, DATA, true);
> +		f2fs_submit_merged_bio(sbi, DATA, true, WRITE);
> 
>  	clear_cold_data(page);
>  out:
> @@ -690,7 +785,7 @@ static int f2fs_write_data_pages(struct address_space *mapping,
>  	ret = write_cache_pages(mapping, wbc, __f2fs_writepage, mapping);
>  	if (locked)
>  		mutex_unlock(&sbi->writepages);
> -	f2fs_submit_bio(sbi, DATA, (wbc->sync_mode == WB_SYNC_ALL));
> +	f2fs_submit_merged_bio(sbi, DATA, wbc->sync_mode == WB_SYNC_ALL, WRITE);
> 
>  	remove_dirty_dir_inode(inode);
> 
> @@ -741,7 +836,8 @@ repeat:
>  	if (dn.data_blkaddr == NEW_ADDR) {
>  		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
>  	} else {
> -		err = f2fs_readpage(sbi, page, dn.data_blkaddr, READ_SYNC);
> +		err = f2fs_submit_page_bio(sbi, page, dn.data_blkaddr,
> +							READ_SYNC);
>  		if (err)
>  			return err;
>  		lock_page(page);
> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> index ca33cda..10eca02 100644
> --- a/fs/f2fs/f2fs.h
> +++ b/fs/f2fs/f2fs.h
> @@ -364,6 +364,7 @@ enum page_type {
>  	META_FLUSH,
>  };
> 
> +#define is_read_io(rw)	(((rw) & 1) == READ)
>  struct f2fs_bio_info {
>  	struct bio *bio;		/* bios to merge */
>  	sector_t last_block_in_bio;	/* last block number */
> @@ -1093,9 +1094,6 @@ void clear_prefree_segments(struct f2fs_sb_info *);
>  int npages_for_summary_flush(struct f2fs_sb_info *);
>  void allocate_new_segments(struct f2fs_sb_info *);
>  struct page *get_sum_page(struct f2fs_sb_info *, unsigned int);
> -struct bio *f2fs_bio_alloc(struct block_device *, int);
> -void f2fs_submit_bio(struct f2fs_sb_info *, enum page_type, bool);
> -void f2fs_wait_on_page_writeback(struct page *, enum page_type, bool);
>  void write_meta_page(struct f2fs_sb_info *, struct page *);
>  void write_node_page(struct f2fs_sb_info *, struct page *, unsigned int,
>  					block_t, block_t *);
> @@ -1106,6 +1104,7 @@ void recover_data_page(struct f2fs_sb_info *, struct page *,
>  				struct f2fs_summary *, block_t, block_t);
>  void rewrite_node_page(struct f2fs_sb_info *, struct page *,
>  				struct f2fs_summary *, block_t, block_t);
> +void f2fs_wait_on_page_writeback(struct page *, enum page_type, bool);
>  void write_data_summaries(struct f2fs_sb_info *, block_t);
>  void write_node_summaries(struct f2fs_sb_info *, block_t);
>  int lookup_journal_in_cursum(struct f2fs_summary_block *,
> @@ -1141,15 +1140,16 @@ void destroy_checkpoint_caches(void);
>  /*
>   * data.c
>   */
> +void f2fs_submit_merged_bio(struct f2fs_sb_info *, enum page_type, bool, int);
> +int f2fs_submit_page_bio(struct f2fs_sb_info *, struct page *, block_t, int);
> +void f2fs_submit_page_mbio(struct f2fs_sb_info *, struct page *, block_t,
> +							enum page_type, int);
>  int reserve_new_block(struct dnode_of_data *);
>  int f2fs_reserve_block(struct dnode_of_data *, pgoff_t);
>  void update_extent_cache(block_t, struct dnode_of_data *);
>  struct page *find_data_page(struct inode *, pgoff_t, bool);
>  struct page *get_lock_data_page(struct inode *, pgoff_t);
>  struct page *get_new_data_page(struct inode *, struct page *, pgoff_t, bool);
> -int f2fs_readpage(struct f2fs_sb_info *, struct page *, block_t, int);
> -void f2fs_submit_read_bio(struct f2fs_sb_info *, int);
> -void submit_read_page(struct f2fs_sb_info *, struct page *, block_t, int);
>  int do_write_data_page(struct page *);
> 
>  /*
> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
> index 5fa54c1..2886aef 100644
> --- a/fs/f2fs/gc.c
> +++ b/fs/f2fs/gc.c
> @@ -631,7 +631,7 @@ next_iput:
>  		goto next_step;
> 
>  	if (gc_type == FG_GC) {
> -		f2fs_submit_bio(sbi, DATA, true);
> +		f2fs_submit_merged_bio(sbi, DATA, true, WRITE);
> 
>  		/*
>  		 * In the case of FG_GC, it'd be better to reclaim this victim
> diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
> index d0ab003..0e1a3df 100644
> --- a/fs/f2fs/node.c
> +++ b/fs/f2fs/node.c
> @@ -106,11 +106,11 @@ static void ra_nat_pages(struct f2fs_sb_info *sbi, int nid)
>  			f2fs_put_page(page, 1);
>  			continue;
>  		}
> -		submit_read_page(sbi, page, index, READ_SYNC | REQ_META);
> +		f2fs_submit_page_mbio(sbi, page, index, META, READ);
>  		mark_page_accessed(page);
>  		f2fs_put_page(page, 0);
>  	}
> -	f2fs_submit_read_bio(sbi, READ_SYNC | REQ_META);
> +	f2fs_submit_merged_bio(sbi, META, true, READ);
>  }
> 
>  static struct nat_entry *__lookup_nat_cache(struct f2fs_nm_info *nm_i, nid_t n)
> @@ -891,7 +891,7 @@ fail:
>   * LOCKED_PAGE: f2fs_put_page(page, 1)
>   * error: nothing
>   */
> -static int read_node_page(struct page *page, int type)
> +static int read_node_page(struct page *page, int rw)
>  {
>  	struct f2fs_sb_info *sbi = F2FS_SB(page->mapping->host->i_sb);
>  	struct node_info ni;
> @@ -906,7 +906,7 @@ static int read_node_page(struct page *page, int type)
>  	if (PageUptodate(page))
>  		return LOCKED_PAGE;
> 
> -	return f2fs_readpage(sbi, page, ni.blk_addr, type);
> +	return f2fs_submit_page_bio(sbi, page, ni.blk_addr, rw);
>  }
> 
>  /*
> @@ -1136,8 +1136,8 @@ continue_unlock:
>  	}
> 
>  	if (wrote)
> -		f2fs_submit_bio(sbi, NODE, wbc->sync_mode == WB_SYNC_ALL);
> -
> +		f2fs_submit_merged_bio(sbi, NODE, wbc->sync_mode == WB_SYNC_ALL,
> +									WRITE);
>  	return nwritten;
>  }
> 
> @@ -1592,7 +1592,7 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
>  		 */
>  		ClearPageUptodate(page);
> 
> -		if (f2fs_readpage(sbi, page, addr, READ_SYNC))
> +		if (f2fs_submit_page_bio(sbi, page, addr, READ_SYNC))
>  			goto out;
> 
>  		lock_page(page);
> diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
> index fdc8116..c209b86 100644
> --- a/fs/f2fs/recovery.c
> +++ b/fs/f2fs/recovery.c
> @@ -143,7 +143,7 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
>  	while (1) {
>  		struct fsync_inode_entry *entry;
> 
> -		err = f2fs_readpage(sbi, page, blkaddr, READ_SYNC);
> +		err = f2fs_submit_page_bio(sbi, page, blkaddr, READ_SYNC);
>  		if (err)
>  			goto out;
> 
> @@ -386,7 +386,7 @@ static int recover_data(struct f2fs_sb_info *sbi,
>  	while (1) {
>  		struct fsync_inode_entry *entry;
> 
> -		err = f2fs_readpage(sbi, page, blkaddr, READ_SYNC);
> +		err = f2fs_submit_page_bio(sbi, page, blkaddr, READ_SYNC);
>  		if (err)
>  			goto out;
> 
> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> index 0db4027..ca9adf5 100644
> --- a/fs/f2fs/segment.c
> +++ b/fs/f2fs/segment.c
> @@ -787,146 +787,6 @@ static const struct segment_allocation default_salloc_ops = {
>  	.allocate_segment = allocate_segment_by_default,
>  };
> 
> -static void f2fs_end_io_write(struct bio *bio, int err)
> -{
> -	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
> -	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
> -	struct f2fs_sb_info *sbi = F2FS_SB(bvec->bv_page->mapping->host->i_sb);
> -
> -	do {
> -		struct page *page = bvec->bv_page;
> -
> -		if (--bvec >= bio->bi_io_vec)
> -			prefetchw(&bvec->bv_page->flags);
> -		if (!uptodate) {
> -			SetPageError(page);
> -			if (page->mapping)
> -				set_bit(AS_EIO, &page->mapping->flags);
> -
> -			set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
> -			sbi->sb->s_flags |= MS_RDONLY;
> -		}
> -		end_page_writeback(page);
> -		dec_page_count(sbi, F2FS_WRITEBACK);
> -	} while (bvec >= bio->bi_io_vec);
> -
> -	if (bio->bi_private)
> -		complete(bio->bi_private);
> -
> -	if (!get_pages(sbi, F2FS_WRITEBACK) &&
> -			!list_empty(&sbi->cp_wait.task_list))
> -		wake_up(&sbi->cp_wait);
> -
> -	bio_put(bio);
> -}
> -
> -struct bio *f2fs_bio_alloc(struct block_device *bdev, int npages)
> -{
> -	struct bio *bio;
> -
> -	/* No failure on bio allocation */
> -	bio = bio_alloc(GFP_NOIO, npages);
> -	bio->bi_bdev = bdev;
> -	bio->bi_private = NULL;
> -
> -	return bio;
> -}
> -
> -static void do_submit_bio(struct f2fs_sb_info *sbi,
> -				enum page_type type, bool sync)
> -{
> -	int rw = sync ? WRITE_SYNC : WRITE;
> -	enum page_type btype = PAGE_TYPE_OF_BIO(type);
> -	struct f2fs_bio_info *io = &sbi->write_io[btype];
> -
> -	if (!io->bio)
> -		return;
> -
> -	if (type >= META_FLUSH)
> -		rw = WRITE_FLUSH_FUA;
> -
> -	if (btype == META)
> -		rw |= REQ_META;
> -
> -	trace_f2fs_submit_write_bio(sbi->sb, rw, btype, io->bio);
> -
> -	/*
> -	 * META_FLUSH is only from the checkpoint procedure, and we should wait
> -	 * this metadata bio for FS consistency.
> -	 */
> -	if (type == META_FLUSH) {
> -		DECLARE_COMPLETION_ONSTACK(wait);
> -		io->bio->bi_private = &wait;
> -		submit_bio(rw, io->bio);
> -		wait_for_completion(&wait);
> -	} else {
> -		submit_bio(rw, io->bio);
> -	}
> -	io->bio = NULL;
> -}
> -
> -void f2fs_submit_bio(struct f2fs_sb_info *sbi, enum page_type type, bool sync)
> -{
> -	struct f2fs_bio_info *io = &sbi->write_io[PAGE_TYPE_OF_BIO(type)];
> -
> -	if (!io->bio)
> -		return;
> -
> -	mutex_lock(&io->io_mutex);
> -	do_submit_bio(sbi, type, sync);
> -	mutex_unlock(&io->io_mutex);
> -}
> -
> -static void submit_write_page(struct f2fs_sb_info *sbi, struct page *page,
> -				block_t blk_addr, enum page_type type)
> -{
> -	struct block_device *bdev = sbi->sb->s_bdev;
> -	struct f2fs_bio_info *io = &sbi->write_io[type];
> -	int bio_blocks;
> -
> -	verify_block_addr(sbi, blk_addr);
> -
> -	mutex_lock(&io->io_mutex);
> -
> -	inc_page_count(sbi, F2FS_WRITEBACK);
> -
> -	if (io->bio && io->last_block_in_bio != blk_addr - 1)
> -		do_submit_bio(sbi, type, false);
> -alloc_new:
> -	if (io->bio == NULL) {
> -		bio_blocks = MAX_BIO_BLOCKS(max_hw_blocks(sbi));
> -		io->bio = f2fs_bio_alloc(bdev, bio_blocks);
> -		io->bio->bi_sector = SECTOR_FROM_BLOCK(sbi, blk_addr);
> -		io->bio->bi_end_io = f2fs_end_io_write;
> -		/*
> -		 * The end_io will be assigned at the sumbission phase.
> -		 * Until then, let bio_add_page() merge consecutive IOs as much
> -		 * as possible.
> -		 */
> -	}
> -
> -	if (bio_add_page(io->bio, page, PAGE_CACHE_SIZE, 0) <
> -							PAGE_CACHE_SIZE) {
> -		do_submit_bio(sbi, type, false);
> -		goto alloc_new;
> -	}
> -
> -	io->last_block_in_bio = blk_addr;
> -
> -	mutex_unlock(&io->io_mutex);
> -	trace_f2fs_submit_write_page(page, WRITE, type, blk_addr);
> -}
> -
> -void f2fs_wait_on_page_writeback(struct page *page,
> -				enum page_type type, bool sync)
> -{
> -	struct f2fs_sb_info *sbi = F2FS_SB(page->mapping->host->i_sb);
> -	if (PageWriteback(page)) {
> -		f2fs_submit_bio(sbi, type, sync);
> -		wait_on_page_writeback(page);
> -	}
> -}
> -
>  static bool __has_curseg_space(struct f2fs_sb_info *sbi, int type)
>  {
>  	struct curseg_info *curseg = CURSEG_I(sbi, type);
> @@ -1040,7 +900,7 @@ static void do_write_page(struct f2fs_sb_info *sbi, struct page *page,
>  		fill_node_footer_blkaddr(page, NEXT_FREE_BLKADDR(sbi, curseg));
> 
>  	/* writeout dirty page into bdev */
> -	submit_write_page(sbi, page, *new_blkaddr, p_type);
> +	f2fs_submit_page_mbio(sbi, page, *new_blkaddr, p_type, WRITE);
> 
>  	mutex_unlock(&curseg->curseg_mutex);
>  }
> @@ -1048,7 +908,7 @@ static void do_write_page(struct f2fs_sb_info *sbi, struct page *page,
>  void write_meta_page(struct f2fs_sb_info *sbi, struct page *page)
>  {
>  	set_page_writeback(page);
> -	submit_write_page(sbi, page, page->index, META);
> +	f2fs_submit_page_mbio(sbi, page, page->index, META, WRITE);
>  }
> 
>  void write_node_page(struct f2fs_sb_info *sbi, struct page *page,
> @@ -1078,7 +938,7 @@ void write_data_page(struct inode *inode, struct page *page,
>  void rewrite_data_page(struct f2fs_sb_info *sbi, struct page *page,
>  					block_t old_blk_addr)
>  {
> -	submit_write_page(sbi, page, old_blk_addr, DATA);
> +	f2fs_submit_page_mbio(sbi, page, old_blk_addr, DATA, WRITE);
>  }
> 
>  void recover_data_page(struct f2fs_sb_info *sbi,
> @@ -1165,8 +1025,8 @@ void rewrite_node_page(struct f2fs_sb_info *sbi,
> 
>  	/* rewrite node page */
>  	set_page_writeback(page);
> -	submit_write_page(sbi, page, new_blkaddr, NODE);
> -	f2fs_submit_bio(sbi, NODE, true);
> +	f2fs_submit_page_mbio(sbi, page, new_blkaddr, NODE, WRITE);
> +	f2fs_submit_merged_bio(sbi, NODE, true, WRITE);
>  	refresh_sit_entry(sbi, old_blkaddr, new_blkaddr);
> 
>  	locate_dirty_segment(sbi, old_cursegno);
> @@ -1176,6 +1036,16 @@ void rewrite_node_page(struct f2fs_sb_info *sbi,
>  	mutex_unlock(&curseg->curseg_mutex);
>  }
> 
> +void f2fs_wait_on_page_writeback(struct page *page,
> +				enum page_type type, bool sync)
> +{
> +	struct f2fs_sb_info *sbi = F2FS_SB(page->mapping->host->i_sb);
> +	if (PageWriteback(page)) {
> +		f2fs_submit_merged_bio(sbi, type, sync, WRITE);
> +		wait_on_page_writeback(page);
> +	}
> +}
> +
>  static int read_compacted_summaries(struct f2fs_sb_info *sbi)
>  {
>  	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
> @@ -1723,13 +1593,13 @@ repeat:
>  			continue;
>  		}
> 
> -		submit_read_page(sbi, page, blk_addr, READ_SYNC | REQ_META);
> +		f2fs_submit_page_mbio(sbi, page, blk_addr, META, READ);
> 
>  		mark_page_accessed(page);
>  		f2fs_put_page(page, 0);
>  	}
> 
> -	f2fs_submit_read_bio(sbi, READ_SYNC | REQ_META);
> +	f2fs_submit_merged_bio(sbi, META, true, READ);
>  	return blkno - start;
>  }
> 
> diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h
> index 204fcc3..3b9f28d 100644
> --- a/include/trace/events/f2fs.h
> +++ b/include/trace/events/f2fs.h
> @@ -434,7 +434,7 @@ TRACE_EVENT(f2fs_truncate_partial_nodes,
>  		__entry->err)
>  );
> 
> -TRACE_EVENT_CONDITION(f2fs_readpage,
> +TRACE_EVENT_CONDITION(f2fs_submit_page_bio,
> 
>  	TP_PROTO(struct page *page, sector_t blkaddr, int type),
> 
> @@ -641,18 +641,22 @@ DECLARE_EVENT_CLASS(f2fs__submit_bio,
>  		__entry->size)
>  );
> 
> -DEFINE_EVENT(f2fs__submit_bio, f2fs_submit_write_bio,
> +DEFINE_EVENT_CONDITION(f2fs__submit_bio, f2fs_submit_write_bio,
> 
>  	TP_PROTO(struct super_block *sb, int rw, int type, struct bio *bio),
> 
> -	TP_ARGS(sb, rw, type, bio)
> +	TP_ARGS(sb, rw, type, bio),
> +
> +	TP_CONDITION(bio)
>  );
> 
> -DEFINE_EVENT(f2fs__submit_bio, f2fs_submit_read_bio,
> +DEFINE_EVENT_CONDITION(f2fs__submit_bio, f2fs_submit_read_bio,
> 
>  	TP_PROTO(struct super_block *sb, int rw, int type, struct bio *bio),
> 
> -	TP_ARGS(sb, rw, type, bio)
> +	TP_ARGS(sb, rw, type, bio),
> +
> +	TP_CONDITION(bio)
>  );
> 
>  DECLARE_EVENT_CLASS(f2fs__page,
> @@ -701,7 +705,7 @@ DEFINE_EVENT(f2fs__page, f2fs_vm_page_mkwrite,
>  	TP_ARGS(page, type)
>  );
> 
> -DECLARE_EVENT_CLASS(f2fs_io_page,
> +TRACE_EVENT(f2fs_submit_page_mbio,
> 
>  	TP_PROTO(struct page *page, int rw, int type, block_t blk_addr),
> 
> @@ -733,20 +737,6 @@ DECLARE_EVENT_CLASS(f2fs_io_page,
>  		(unsigned long long)__entry->block)
>  );
> 
> -DEFINE_EVENT(f2fs_io_page, f2fs_submit_write_page,
> -
> -	TP_PROTO(struct page *page, int rw, int type, block_t blk_addr),
> -
> -	TP_ARGS(page, rw, type, blk_addr)
> -);
> -
> -DEFINE_EVENT(f2fs_io_page, f2fs_submit_read_page,
> -
> -	TP_PROTO(struct page *page, int rw, int type, block_t blk_addr),
> -
> -	TP_ARGS(page, rw, type, blk_addr)
> -);
> -
>  TRACE_EVENT(f2fs_write_checkpoint,
> 
>  	TP_PROTO(struct super_block *sb, bool is_umount, char *msg),
> --
> 1.8.4.474.g128a96c
> 
> 
> 
> --
> Jaegeuk Kim
> Samsung
> 
> 
> ------------------------------------------------------------------------------
> Rapidly troubleshoot problems before they affect your business. Most IT
> organizations don't have a clear picture of how application performance
> affects their revenue. With AppDynamics, you get 100% visibility into your
> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> http://pubads.g.doubleclick.net/gampad/clk?id=84349351&iu=/4140/ostg.clktrk
> _______________________________________________
> Linux-f2fs-devel mailing list
> Linux-f2fs-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel