Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH v3] f2fs: submit cached bio to avoid endless PageWriteback
To:     Jaegeuk Kim <jaegeuk@kernel.org>
CC:     Chao Yu <chao@kernel.org>,
        <linux-f2fs-devel@lists.sourceforge.net>,
        <linux-kernel@vger.kernel.org>
References: <20180912234053.8761-1-chao@kernel.org>
 <20180918010442.GA79604@jaegeuk-macbookpro.roam.corp.google.com>
 <3ca1e04f-0929-2ab3-2112-4fa22915be50@huawei.com>
 <20180918013743.GE79604@jaegeuk-macbookpro.roam.corp.google.com>
 <8b1571fa-ef41-4ab3-187c-d6caf7a2ebe1@huawei.com>
 <20180918020226.GA83471@jaegeuk-macbookpro.roam.corp.google.com>
From:   Chao Yu <yuchao0@huawei.com>
Message-ID: <acb91226-bb83-3d7f-3ac8-db7a3ee24172@huawei.com>
Date:   Tue, 18 Sep 2018 10:14:21 +0800
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <20180918020226.GA83471@jaegeuk-macbookpro.roam.corp.google.com>
Content-Type: text/plain; charset="windows-1252"
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2018/9/18 10:02, Jaegeuk Kim wrote:
> On 09/18, Chao Yu wrote:
>> On 2018/9/18 9:37, Jaegeuk Kim wrote:
>>> On 09/18, Chao Yu wrote:
>>>> On 2018/9/18 9:04, Jaegeuk Kim wrote:
>>>>> On 09/13, Chao Yu wrote:
>>>>>> From: Chao Yu <yuchao0@huawei.com>
>>>>>>
>>>>>> When migrating encrypted block from background GC thread, we only add
>>>>>> them into f2fs inner bio cache, but forget to submit the cached bio, it
>>>>>> may cause potential deadlock when we are waiting page writebacked, fix
>>>>>> it.
>>>>>>
>>>>>> Signed-off-by: Chao Yu <yuchao0@huawei.com>
>>>>>> ---
>>>>>> v3:
>>>>>> clean up codes suggested by Jaegeuk.
>>>>>>  fs/f2fs/f2fs.h |  2 +-
>>>>>>  fs/f2fs/gc.c   | 71 +++++++++++++++++++++++++++++++++++---------------
>>>>>>  fs/f2fs/node.c | 13 ++++++---
>>>>>>  3 files changed, 61 insertions(+), 25 deletions(-)
>>>>>>
>>>>>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
>>>>>> index b676b82312e0..917b2ca76aac 100644
>>>>>> --- a/fs/f2fs/f2fs.h
>>>>>> +++ b/fs/f2fs/f2fs.h
>>>>>> @@ -2869,7 +2869,7 @@ struct page *f2fs_new_node_page(struct dnode_of_data *dn, unsigned int ofs);
>>>>>>  void f2fs_ra_node_page(struct f2fs_sb_info *sbi, nid_t nid);
>>>>>>  struct page *f2fs_get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid);
>>>>>>  struct page *f2fs_get_node_page_ra(struct page *parent, int start);
>>>>>> -void f2fs_move_node_page(struct page *node_page, int gc_type);
>>>>>> +int f2fs_move_node_page(struct page *node_page, int gc_type);
>>>>>>  int f2fs_fsync_node_pages(struct f2fs_sb_info *sbi, struct inode *inode,
>>>>>>  			struct writeback_control *wbc, bool atomic,
>>>>>>  			unsigned int *seq_id);
>>>>>> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
>>>>>> index a4c1a419611d..f57622cfe058 100644
>>>>>> --- a/fs/f2fs/gc.c
>>>>>> +++ b/fs/f2fs/gc.c
>>>>>> @@ -461,7 +461,7 @@ static int check_valid_map(struct f2fs_sb_info *sbi,
>>>>>>   * On validity, copy that node with cold status, otherwise (invalid node)
>>>>>>   * ignore that.
>>>>>>   */
>>>>>> -static void gc_node_segment(struct f2fs_sb_info *sbi,
>>>>>> +static int gc_node_segment(struct f2fs_sb_info *sbi,
>>>>>>  		struct f2fs_summary *sum, unsigned int segno, int gc_type)
>>>>>>  {
>>>>>>  	struct f2fs_summary *entry;
>>>>>> @@ -469,6 +469,7 @@ static void gc_node_segment(struct f2fs_sb_info *sbi,
>>>>>>  	int off;
>>>>>>  	int phase = 0;
>>>>>>  	bool fggc = (gc_type == FG_GC);
>>>>>> +	int submitted = 0;
>>>>>>  
>>>>>>  	start_addr = START_BLOCK(sbi, segno);
>>>>>>  
>>>>>> @@ -482,10 +483,11 @@ static void gc_node_segment(struct f2fs_sb_info *sbi,
>>>>>>  		nid_t nid = le32_to_cpu(entry->nid);
>>>>>>  		struct page *node_page;
>>>>>>  		struct node_info ni;
>>>>>> +		int err;
>>>>>>  
>>>>>>  		/* stop BG_GC if there is not enough free sections. */
>>>>>>  		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0))
>>>>>> -			return;
>>>>>> +			return submitted;
>>>>>>  
>>>>>>  		if (check_valid_map(sbi, segno, off) == 0)
>>>>>>  			continue;
>>>>>> @@ -522,7 +524,9 @@ static void gc_node_segment(struct f2fs_sb_info *sbi,
>>>>>>  			continue;
>>>>>>  		}
>>>>>>  
>>>>>> -		f2fs_move_node_page(node_page, gc_type);
>>>>>> +		err = f2fs_move_node_page(node_page, gc_type);
>>>>>> +		if (!err && gc_type == FG_GC)
>>>>>> +			submitted++;
>>>>>>  		stat_inc_node_blk_count(sbi, 1, gc_type);
>>>>>>  	}
>>>>>>  
>>>>>> @@ -531,6 +535,7 @@ static void gc_node_segment(struct f2fs_sb_info *sbi,
>>>>>>  
>>>>>>  	if (fggc)
>>>>>>  		atomic_dec(&sbi->wb_sync_req[NODE]);
>>>>>> +	return submitted;
>>>>>>  }
>>>>>>  
>>>>>>  /*
>>>>>> @@ -666,7 +671,7 @@ static int ra_data_block(struct inode *inode, pgoff_t index)
>>>>>>   * Move data block via META_MAPPING while keeping locked data page.
>>>>>>   * This can be used to move blocks, aka LBAs, directly on disk.
>>>>>>   */
>>>>>> -static void move_data_block(struct inode *inode, block_t bidx,
>>>>>> +static int move_data_block(struct inode *inode, block_t bidx,
>>>>>>  				int gc_type, unsigned int segno, int off)
>>>>>
>>>>> We don't need to submit IOs in this case.
>>>>
>>>> Actually, previously, we missed to submit IOs for encrypted block only in
>>>> BGGC, so we fix to submit for this case, all other codes are cleanups. Right?
>>>
>>> The move_data_block migrates encrypted blocks all the time with meta page IOs.
>>> I don't know what you're saying about BGGC.
>>
>> In move_data_block(), we use f2fs_submit_page_write() to add encrypted page
>> in to sbi->write_io[META].bio cache, so before exit GC, we need to submit
>> this cache by f2fs_submit_merged_write(), otherwise bio with encrypted page
>> will be cached in sbi->write_io[META].bio for long time, since we only
>> submmit this bio cache in foreground GC.
>>
>> if (gc_type == FG_GC)
>> 	f2fs_submit_merged_write(sbi,
>> 				(type == SUM_TYPE_NODE) ? NODE : DATA);
> 
> Does this issue pending META IOs? I think META won't block any NODE/DATA
> pages. Moreover checkpoint or reading the moved block make the IOs be flushed,
> so it doesn't hurt the consistency.

I just guess below endless waiting for page writeback is cased by this case

-000|__switch_to()

-001|__schedule()

-002|need_resched(inline)

-002|schedule()

-003|schedule_timeout()

-004|get_current(inline)

-004|io_schedule_timeout()

-005|bit_wait_io()

-006|__wait_on_bit()

-007|wait_on_page_bit()

-008|PageWriteback(inline)

-008|wait_on_page_writeback(inline)

-008|__filemap_fdatawait_range()

-009|filemap_fdatawait_keep_errors()

-010|sync_inodes_sb()

-011|__sync_filesystem(inline)

-011|sync_filesystem()

-012|generic_shutdown_super()

-013|kill_block_super()

-014|kill_f2fs_super()

-015|deactivate_locked_super()

-016|deactivate_super()

-017|mnt_free_id(inline)

-017|cleanup_mnt()

-018|__cleanup_mnt()

-019|task_work_run()

-020|do_notify_resume()

-021|work_pending(asm)

-->|exception

-022|NUX:0x539E58(asm)

---|end of frame

> 
>>
>>>
>>>>
>>>> Thanks,
>>>>
>>>>>
>>>>>>  {
>>>>>>  	struct f2fs_io_info fio = {
>>>>>> @@ -685,25 +690,29 @@ static void move_data_block(struct inode *inode, block_t bidx,
>>>>>>  	struct node_info ni;
>>>>>>  	struct page *page, *mpage;
>>>>>>  	block_t newaddr;
>>>>>> -	int err;
>>>>>> +	int err = 0;
>>>>>>  	bool lfs_mode = test_opt(fio.sbi, LFS);
>>>>>>  
>>>>>>  	/* do not read out */
>>>>>>  	page = f2fs_grab_cache_page(inode->i_mapping, bidx, false);
>>>>>>  	if (!page)
>>>>>> -		return;
>>>>>> +		return -ENOMEM;
>>>>>>  
>>>>>> -	if (!check_valid_map(F2FS_I_SB(inode), segno, off))
>>>>>> +	if (!check_valid_map(F2FS_I_SB(inode), segno, off)) {
>>>>>> +		err = -ENOENT;
>>>>>>  		goto out;
>>>>>> +	}
>>>>>>  
>>>>>>  	if (f2fs_is_atomic_file(inode)) {
>>>>>>  		F2FS_I(inode)->i_gc_failures[GC_FAILURE_ATOMIC]++;
>>>>>>  		F2FS_I_SB(inode)->skipped_atomic_files[gc_type]++;
>>>>>> +		err = -EAGAIN;
>>>>>>  		goto out;
>>>>>>  	}
>>>>>>  
>>>>>>  	if (f2fs_is_pinned_file(inode)) {
>>>>>>  		f2fs_pin_file_control(inode, true);
>>>>>> +		err = -EAGAIN;
>>>>>>  		goto out;
>>>>>>  	}
>>>>>>  
>>>>>> @@ -714,6 +723,7 @@ static void move_data_block(struct inode *inode, block_t bidx,
>>>>>>  
>>>>>>  	if (unlikely(dn.data_blkaddr == NULL_ADDR)) {
>>>>>>  		ClearPageUptodate(page);
>>>>>> +		err = -ENOENT;
>>>>>>  		goto put_out;
>>>>>>  	}
>>>>>>  
>>>>>> @@ -796,6 +806,7 @@ static void move_data_block(struct inode *inode, block_t bidx,
>>>>>>  	fio.new_blkaddr = newaddr;
>>>>>>  	f2fs_submit_page_write(&fio);
>>>>>>  	if (fio.retry) {
>>>>>> +		err = -EAGAIN;
>>>>>>  		if (PageWriteback(fio.encrypted_page))
>>>>>>  			end_page_writeback(fio.encrypted_page);
>>>>>>  		goto put_page_out;
>>>>>> @@ -819,34 +830,42 @@ static void move_data_block(struct inode *inode, block_t bidx,
>>>>>>  	f2fs_put_dnode(&dn);
>>>>>>  out:
>>>>>>  	f2fs_put_page(page, 1);
>>>>>> +	return err;
>>>>>>  }
>>>>>>  
>>>>>> -static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
>>>>>> +static int move_data_page(struct inode *inode, block_t bidx, int gc_type,
>>>>>>  							unsigned int segno, int off)
>>>>>>  {
>>>>>>  	struct page *page;
>>>>>> +	int err = 0;
>>>>>>  
>>>>>>  	page = f2fs_get_lock_data_page(inode, bidx, true);
>>>>>>  	if (IS_ERR(page))
>>>>>> -		return;
>>>>>> +		return PTR_ERR(page);
>>>>>>  
>>>>>> -	if (!check_valid_map(F2FS_I_SB(inode), segno, off))
>>>>>> +	if (!check_valid_map(F2FS_I_SB(inode), segno, off)) {
>>>>>> +		err = -ENOENT;
>>>>>>  		goto out;
>>>>>> +	}
>>>>>>  
>>>>>>  	if (f2fs_is_atomic_file(inode)) {
>>>>>>  		F2FS_I(inode)->i_gc_failures[GC_FAILURE_ATOMIC]++;
>>>>>>  		F2FS_I_SB(inode)->skipped_atomic_files[gc_type]++;
>>>>>> +		err = -EAGAIN;
>>>>>>  		goto out;
>>>>>>  	}
>>>>>>  	if (f2fs_is_pinned_file(inode)) {
>>>>>>  		if (gc_type == FG_GC)
>>>>>>  			f2fs_pin_file_control(inode, true);
>>>>>> +		err = -EAGAIN;
>>>>>>  		goto out;
>>>>>>  	}
>>>>>>  
>>>>>>  	if (gc_type == BG_GC) {
>>>>>> -		if (PageWriteback(page))
>>>>>> +		if (PageWriteback(page)) {
>>>>>> +			err = -EAGAIN;
>>>>>>  			goto out;
>>>>>> +		}
>>>>>>  		set_page_dirty(page);
>>>>>>  		set_cold_data(page);
>>>>>>  	} else {
>>>>>> @@ -864,7 +883,6 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
>>>>>>  			.io_type = FS_GC_DATA_IO,
>>>>>>  		};
>>>>>>  		bool is_dirty = PageDirty(page);
>>>>>> -		int err;
>>>>>>  
>>>>>>  retry:
>>>>>>  		set_page_dirty(page);
>>>>>> @@ -889,6 +907,7 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
>>>>>>  	}
>>>>>>  out:
>>>>>>  	f2fs_put_page(page, 1);
>>>>>> +	return err;
>>>>>>  }
>>>>>>  
>>>>>>  /*
>>>>>> @@ -898,7 +917,7 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
>>>>>>   * If the parent node is not valid or the data block address is different,
>>>>>>   * the victim data block is ignored.
>>>>>>   */
>>>>>> -static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>> +static int gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>>  		struct gc_inode_list *gc_list, unsigned int segno, int gc_type)
>>>>>>  {
>>>>>>  	struct super_block *sb = sbi->sb;
>>>>>> @@ -906,6 +925,7 @@ static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>>  	block_t start_addr;
>>>>>>  	int off;
>>>>>>  	int phase = 0;
>>>>>> +	int submitted = 0;
>>>>>>  
>>>>>>  	start_addr = START_BLOCK(sbi, segno);
>>>>>>  
>>>>>> @@ -922,7 +942,7 @@ static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>>  
>>>>>>  		/* stop BG_GC if there is not enough free sections. */
>>>>>>  		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0))
>>>>>> -			return;
>>>>>> +			return submitted;
>>>>>>  
>>>>>>  		if (check_valid_map(sbi, segno, off) == 0)
>>>>>>  			continue;
>>>>>> @@ -994,6 +1014,7 @@ static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>>  		if (inode) {
>>>>>>  			struct f2fs_inode_info *fi = F2FS_I(inode);
>>>>>>  			bool locked = false;
>>>>>> +			int err;
>>>>>>  
>>>>>>  			if (S_ISREG(inode->i_mode)) {
>>>>>>  				if (!down_write_trylock(&fi->i_gc_rwsem[READ]))
>>>>>> @@ -1013,12 +1034,16 @@ static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>>  			start_bidx = f2fs_start_bidx_of_node(nofs, inode)
>>>>>>  								+ ofs_in_node;
>>>>>>  			if (f2fs_post_read_required(inode))
>>>>>> -				move_data_block(inode, start_bidx, gc_type,
>>>>>> -								segno, off);
>>>>>> +				err = move_data_block(inode, start_bidx,
>>>>>> +							gc_type, segno, off);
>>>>>>  			else
>>>>>> -				move_data_page(inode, start_bidx, gc_type,
>>>>>> +				err = move_data_page(inode, start_bidx, gc_type,
>>>>>>  								segno, off);
>>>>>>  
>>>>>> +			if (!err && (gc_type == FG_GC ||
>>>>>> +					f2fs_post_read_required(inode)))
>>>>>> +				submitted++;
>>>>>> +
>>>>>>  			if (locked) {
>>>>>>  				up_write(&fi->i_gc_rwsem[WRITE]);
>>>>>>  				up_write(&fi->i_gc_rwsem[READ]);
>>>>>> @@ -1030,6 +1055,8 @@ static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
>>>>>>  
>>>>>>  	if (++phase < 5)
>>>>>>  		goto next_step;
>>>>>> +
>>>>>> +	return submitted;
>>>>>>  }
>>>>>>  
>>>>>>  static int __get_victim(struct f2fs_sb_info *sbi, unsigned int *victim,
>>>>>> @@ -1057,6 +1084,7 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
>>>>>>  	int seg_freed = 0;
>>>>>>  	unsigned char type = IS_DATASEG(get_seg_entry(sbi, segno)->type) ?
>>>>>>  						SUM_TYPE_DATA : SUM_TYPE_NODE;
>>>>>> +	int submitted = 0;
>>>>>>  
>>>>>>  	/* readahead multi ssa blocks those have contiguous address */
>>>>>>  	if (sbi->segs_per_sec > 1)
>>>>>> @@ -1100,10 +1128,11 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
>>>>>>  		 *                                  - lock_page(sum_page)
>>>>>>  		 */
>>>>>>  		if (type == SUM_TYPE_NODE)
>>>>>> -			gc_node_segment(sbi, sum->entries, segno, gc_type);
>>>>>> -		else
>>>>>> -			gc_data_segment(sbi, sum->entries, gc_list, segno,
>>>>>> +			submitted += gc_node_segment(sbi, sum->entries, segno,
>>>>>>  								gc_type);
>>>>>> +		else
>>>>>> +			submitted += gc_data_segment(sbi, sum->entries, gc_list,
>>>>>> +							segno, gc_type);
>>>>>>  
>>>>>>  		stat_inc_seg_count(sbi, type, gc_type);
>>>>>>  
>>>>>> @@ -1114,7 +1143,7 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
>>>>>>  		f2fs_put_page(sum_page, 0);
>>>>>>  	}
>>>>>>  
>>>>>> -	if (gc_type == FG_GC)
>>>>>> +	if (submitted)
>>>>>>  		f2fs_submit_merged_write(sbi,
>>>>>>  				(type == SUM_TYPE_NODE) ? NODE : DATA);
>>>>>>  
>>>>>> diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
>>>>>> index fa2381c0bc47..214dd6326b4b 100644
>>>>>> --- a/fs/f2fs/node.c
>>>>>> +++ b/fs/f2fs/node.c
>>>>>> @@ -1584,8 +1584,10 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted,
>>>>>>  	return AOP_WRITEPAGE_ACTIVATE;
>>>>>>  }
>>>>>>  
>>>>>> -void f2fs_move_node_page(struct page *node_page, int gc_type)
>>>>>> +int f2fs_move_node_page(struct page *node_page, int gc_type)
>>>>>>  {
>>>>>> +	int err = 0;
>>>>>> +
>>>>>>  	if (gc_type == FG_GC) {
>>>>>>  		struct writeback_control wbc = {
>>>>>>  			.sync_mode = WB_SYNC_ALL,
>>>>>> @@ -1597,12 +1599,16 @@ void f2fs_move_node_page(struct page *node_page, int gc_type)
>>>>>>  		f2fs_wait_on_page_writeback(node_page, NODE, true);
>>>>>>  
>>>>>>  		f2fs_bug_on(F2FS_P_SB(node_page), PageWriteback(node_page));
>>>>>> -		if (!clear_page_dirty_for_io(node_page))
>>>>>> +		if (!clear_page_dirty_for_io(node_page)) {
>>>>>> +			err = -EAGAIN;
>>>>>>  			goto out_page;
>>>>>> +		}
>>>>>>  
>>>>>>  		if (__write_node_page(node_page, false, NULL,
>>>>>> -					&wbc, false, FS_GC_NODE_IO, NULL))
>>>>>> +					&wbc, false, FS_GC_NODE_IO, NULL)) {
>>>>>> +			err = -EAGAIN;
>>>>>>  			unlock_page(node_page);
>>>>>> +		}
>>>>>>  		goto release_page;
>>>>>>  	} else {
>>>>>>  		/* set page dirty and write it */
>>>>>> @@ -1613,6 +1619,7 @@ void f2fs_move_node_page(struct page *node_page, int gc_type)
>>>>>>  	unlock_page(node_page);
>>>>>>  release_page:
>>>>>>  	f2fs_put_page(node_page, 0);
>>>>>> +	return err;
>>>>>>  }
>>>>>>  
>>>>>>  static int f2fs_write_node_page(struct page *page,
>>>>>> -- 
>>>>>> 2.18.0
>>>>>
>>>>> .
>>>>>
>>>
>>> .
>>>
> 
> .
>