* [PATCH v2] f2fs: wait IO writeback in commit_inmem_pages()
@ 2018-04-23 15:14 Chao Yu
2018-04-26 1:36 ` Chao Yu
0 siblings, 1 reply; 3+ messages in thread
From: Chao Yu @ 2018-04-23 15:14 UTC (permalink / raw)
To: jaegeuk; +Cc: linux-f2fs-devel, linux-kernel, Chao Yu
From: Chao Yu <yuchao0@huawei.com>
Thread A Thread B
- f2fs_ioc_commit_atomic_write
- commit_inmem_pages
- f2fs_submit_merged_write_cond
: write data
- write_checkpoint
- do_checkpoint
: commit all node within CP
-> SPO
- f2fs_do_sync_file
- file_write_and_wait_range
: wait data writeback
In above race condition, data/node can be flushed in reversed order when
coming a checkpoint before f2fs_do_sync_file, after SPOR, it results in
atomic written data being corrupted.
This patch adds filemap_fdatawait() in commit_inmem_pages() to keep data
and node of atomic file being flushed orderly.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
v2:
- clean up codes suggested by Jaegeuk.
fs/f2fs/file.c | 4 ++++
fs/f2fs/segment.c | 2 ++
2 files changed, 6 insertions(+)
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index bac059474342..0cfa65c21d3f 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -217,6 +217,9 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
trace_f2fs_sync_file_enter(inode);
+ if (atomic)
+ goto write_done;
+
/* if fdatasync is triggered, let's do in-place-update */
if (datasync || get_dirty_pages(inode) <= SM_I(sbi)->min_fsync_blocks)
set_inode_flag(inode, FI_NEED_IPU);
@@ -228,6 +231,7 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
return ret;
}
+write_done:
/* if the inode is dirty, let's recover all the time */
if (!f2fs_skip_inode_update(inode, datasync)) {
f2fs_write_inode(inode, NULL);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 710d817c4350..31336fc54d8e 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -419,6 +419,8 @@ static int __commit_inmem_pages(struct inode *inode)
__revoke_inmem_pages(inode, &fi->inmem_pages, true, false);
} else {
__revoke_inmem_pages(inode, &revoke_list, false, false);
+
+ err = filemap_fdatawait(inode->i_mapping);
}
return err;
--
2.16.2.17.g38e79b1fd
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v2] f2fs: wait IO writeback in commit_inmem_pages()
2018-04-23 15:14 [PATCH v2] f2fs: wait IO writeback in commit_inmem_pages() Chao Yu
@ 2018-04-26 1:36 ` Chao Yu
0 siblings, 0 replies; 3+ messages in thread
From: Chao Yu @ 2018-04-26 1:36 UTC (permalink / raw)
To: Chao Yu, jaegeuk; +Cc: linux-f2fs-devel, linux-kernel
Hi Jaegeuk,
I just look into this flow again, found that in previous patch ("f2fs: fix to
wait page writeback during revoking atomic write") we have already called
f2fs_wait_on_page_writeback in __revoke_inmem_pages, so we don't need to call
redundant filemap_fdatawait here.
Anyway, let me merge these two patches.
Thanks,
On 2018/4/23 23:14, Chao Yu wrote:
> From: Chao Yu <yuchao0@huawei.com>
>
> Thread A Thread B
> - f2fs_ioc_commit_atomic_write
> - commit_inmem_pages
> - f2fs_submit_merged_write_cond
> : write data
> - write_checkpoint
> - do_checkpoint
> : commit all node within CP
> -> SPO
> - f2fs_do_sync_file
> - file_write_and_wait_range
> : wait data writeback
>
> In above race condition, data/node can be flushed in reversed order when
> coming a checkpoint before f2fs_do_sync_file, after SPOR, it results in
> atomic written data being corrupted.
>
> This patch adds filemap_fdatawait() in commit_inmem_pages() to keep data
> and node of atomic file being flushed orderly.
>
> Signed-off-by: Chao Yu <yuchao0@huawei.com>
> ---
> v2:
> - clean up codes suggested by Jaegeuk.
> fs/f2fs/file.c | 4 ++++
> fs/f2fs/segment.c | 2 ++
> 2 files changed, 6 insertions(+)
>
> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> index bac059474342..0cfa65c21d3f 100644
> --- a/fs/f2fs/file.c
> +++ b/fs/f2fs/file.c
> @@ -217,6 +217,9 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
>
> trace_f2fs_sync_file_enter(inode);
>
> + if (atomic)
> + goto write_done;
> +
> /* if fdatasync is triggered, let's do in-place-update */
> if (datasync || get_dirty_pages(inode) <= SM_I(sbi)->min_fsync_blocks)
> set_inode_flag(inode, FI_NEED_IPU);
> @@ -228,6 +231,7 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
> return ret;
> }
>
> +write_done:
> /* if the inode is dirty, let's recover all the time */
> if (!f2fs_skip_inode_update(inode, datasync)) {
> f2fs_write_inode(inode, NULL);
> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> index 710d817c4350..31336fc54d8e 100644
> --- a/fs/f2fs/segment.c
> +++ b/fs/f2fs/segment.c
> @@ -419,6 +419,8 @@ static int __commit_inmem_pages(struct inode *inode)
> __revoke_inmem_pages(inode, &fi->inmem_pages, true, false);
> } else {
> __revoke_inmem_pages(inode, &revoke_list, false, false);
> +
> + err = filemap_fdatawait(inode->i_mapping);
> }
>
> return err;
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH v2] f2fs: wait IO writeback in commit_inmem_pages()
@ 2018-04-26 1:36 ` Chao Yu
0 siblings, 0 replies; 3+ messages in thread
From: Chao Yu @ 2018-04-26 1:36 UTC (permalink / raw)
To: Chao Yu, jaegeuk; +Cc: linux-f2fs-devel, linux-kernel
Hi Jaegeuk,
I just look into this flow again, found that in previous patch ("f2fs: fix to
wait page writeback during revoking atomic write") we have already called
f2fs_wait_on_page_writeback in __revoke_inmem_pages, so we don't need to call
redundant filemap_fdatawait here.
Anyway, let me merge these two patches.
Thanks,
On 2018/4/23 23:14, Chao Yu wrote:
> From: Chao Yu <yuchao0@huawei.com>
>
> Thread A Thread B
> - f2fs_ioc_commit_atomic_write
> - commit_inmem_pages
> - f2fs_submit_merged_write_cond
> : write data
> - write_checkpoint
> - do_checkpoint
> : commit all node within CP
> -> SPO
> - f2fs_do_sync_file
> - file_write_and_wait_range
> : wait data writeback
>
> In above race condition, data/node can be flushed in reversed order when
> coming a checkpoint before f2fs_do_sync_file, after SPOR, it results in
> atomic written data being corrupted.
>
> This patch adds filemap_fdatawait() in commit_inmem_pages() to keep data
> and node of atomic file being flushed orderly.
>
> Signed-off-by: Chao Yu <yuchao0@huawei.com>
> ---
> v2:
> - clean up codes suggested by Jaegeuk.
> fs/f2fs/file.c | 4 ++++
> fs/f2fs/segment.c | 2 ++
> 2 files changed, 6 insertions(+)
>
> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> index bac059474342..0cfa65c21d3f 100644
> --- a/fs/f2fs/file.c
> +++ b/fs/f2fs/file.c
> @@ -217,6 +217,9 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
>
> trace_f2fs_sync_file_enter(inode);
>
> + if (atomic)
> + goto write_done;
> +
> /* if fdatasync is triggered, let's do in-place-update */
> if (datasync || get_dirty_pages(inode) <= SM_I(sbi)->min_fsync_blocks)
> set_inode_flag(inode, FI_NEED_IPU);
> @@ -228,6 +231,7 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
> return ret;
> }
>
> +write_done:
> /* if the inode is dirty, let's recover all the time */
> if (!f2fs_skip_inode_update(inode, datasync)) {
> f2fs_write_inode(inode, NULL);
> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> index 710d817c4350..31336fc54d8e 100644
> --- a/fs/f2fs/segment.c
> +++ b/fs/f2fs/segment.c
> @@ -419,6 +419,8 @@ static int __commit_inmem_pages(struct inode *inode)
> __revoke_inmem_pages(inode, &fi->inmem_pages, true, false);
> } else {
> __revoke_inmem_pages(inode, &revoke_list, false, false);
> +
> + err = filemap_fdatawait(inode->i_mapping);
> }
>
> return err;
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-04-26 1:36 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-04-23 15:14 [PATCH v2] f2fs: wait IO writeback in commit_inmem_pages() Chao Yu
2018-04-26 1:36 ` Chao Yu
2018-04-26 1:36 ` Chao Yu
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.