All of lore.kernel.org
 help / color / mirror / Atom feed
* [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
       [not found] <CGME20250314120658epcas1p2d3ec037c294d4c907ce7fa2fe1c3aa27@epcas1p2.samsung.com>
@ 2025-03-14 12:06   ` Yeongjin Gil
  0 siblings, 0 replies; 12+ messages in thread
From: Yeongjin Gil @ 2025-03-14 12:06 UTC (permalink / raw)
  To: jaegeuk, chao, daehojeong, linux-f2fs-devel, linux-kernel; +Cc: sj1557.seo

In the case of the following call stack for an atomic file,
FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.

f2fs_file_write_iter
  f2fs_map_blocks
    f2fs_reserve_new_blocks
      inc_valid_block_count
        __mark_inode_dirty(dquot)
          f2fs_dirty_inode

If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
due to a mismatch between old file size and new data.

To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
previously cleared by the Writeback thread during the commit atomic, is
set and i_size is updated.

Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
---
 fs/f2fs/inode.c | 4 +---
 fs/f2fs/super.c | 4 ++++
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index aa2f41696a88..83f862578fc8 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
 	if (f2fs_inode_dirtied(inode, sync))
 		return;
 
-	if (f2fs_is_atomic_file(inode)) {
-		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
+	if (f2fs_is_atomic_file(inode))
 		return;
-	}
 
 	mark_inode_dirty_sync(inode);
 }
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 397df271885c..c08d52c6467a 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
 		inc_page_count(sbi, F2FS_DIRTY_IMETA);
 	}
 	spin_unlock(&sbi->inode_lock[DIRTY_META]);
+
+	if (!ret && f2fs_is_atomic_file(inode))
+		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
+
 	return ret;
 }
 
-- 
2.34.1



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply related	[flat|nested] 12+ messages in thread

* [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
@ 2025-03-14 12:06   ` Yeongjin Gil
  0 siblings, 0 replies; 12+ messages in thread
From: Yeongjin Gil @ 2025-03-14 12:06 UTC (permalink / raw)
  To: jaegeuk, chao, daehojeong, linux-f2fs-devel, linux-kernel
  Cc: sj1557.seo, s_min.jeong, Yeongjin Gil

In the case of the following call stack for an atomic file,
FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.

f2fs_file_write_iter
  f2fs_map_blocks
    f2fs_reserve_new_blocks
      inc_valid_block_count
        __mark_inode_dirty(dquot)
          f2fs_dirty_inode

If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
due to a mismatch between old file size and new data.

To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
previously cleared by the Writeback thread during the commit atomic, is
set and i_size is updated.

Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
---
 fs/f2fs/inode.c | 4 +---
 fs/f2fs/super.c | 4 ++++
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index aa2f41696a88..83f862578fc8 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
 	if (f2fs_inode_dirtied(inode, sync))
 		return;
 
-	if (f2fs_is_atomic_file(inode)) {
-		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
+	if (f2fs_is_atomic_file(inode))
 		return;
-	}
 
 	mark_inode_dirty_sync(inode);
 }
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 397df271885c..c08d52c6467a 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
 		inc_page_count(sbi, F2FS_DIRTY_IMETA);
 	}
 	spin_unlock(&sbi->inode_lock[DIRTY_META]);
+
+	if (!ret && f2fs_is_atomic_file(inode))
+		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
+
 	return ret;
 }
 
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
  2025-03-14 12:06   ` Yeongjin Gil
@ 2025-03-14 19:04     ` Daeho Jeong
  -1 siblings, 0 replies; 12+ messages in thread
From: Daeho Jeong @ 2025-03-14 19:04 UTC (permalink / raw)
  To: Yeongjin Gil
  Cc: daehojeong, linux-kernel, linux-f2fs-devel, jaegeuk, sj1557.seo

On Fri, Mar 14, 2025 at 5:28 AM Yeongjin Gil <youngjin.gil@samsung.com> wrote:
>
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
>
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
>
> If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> due to a mismatch between old file size and new data.
>
> To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> previously cleared by the Writeback thread during the commit atomic, is
> set and i_size is updated.
>
> Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
> ---
>  fs/f2fs/inode.c | 4 +---
>  fs/f2fs/super.c | 4 ++++
>  2 files changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
> index aa2f41696a88..83f862578fc8 100644
> --- a/fs/f2fs/inode.c
> +++ b/fs/f2fs/inode.c
> @@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
>         if (f2fs_inode_dirtied(inode, sync))
>                 return;
>
> -       if (f2fs_is_atomic_file(inode)) {
> -               set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +       if (f2fs_is_atomic_file(inode))
>                 return;
> -       }
>
>         mark_inode_dirty_sync(inode);
>  }
> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> index 397df271885c..c08d52c6467a 100644
> --- a/fs/f2fs/super.c
> +++ b/fs/f2fs/super.c
> @@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
>                 inc_page_count(sbi, F2FS_DIRTY_IMETA);
>         }
>         spin_unlock(&sbi->inode_lock[DIRTY_META]);
> +
> +       if (!ret && f2fs_is_atomic_file(inode))
> +               set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +
>         return ret;
>  }
>
> --
> 2.34.1
>
>

Reviewed-by: Daeho Jeong <daehojeong@google.com>

Thanks!

>
> _______________________________________________
> Linux-f2fs-devel mailing list
> Linux-f2fs-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
@ 2025-03-14 19:04     ` Daeho Jeong
  0 siblings, 0 replies; 12+ messages in thread
From: Daeho Jeong @ 2025-03-14 19:04 UTC (permalink / raw)
  To: Yeongjin Gil
  Cc: jaegeuk, chao, daehojeong, linux-f2fs-devel, linux-kernel,
	sj1557.seo

On Fri, Mar 14, 2025 at 5:28 AM Yeongjin Gil <youngjin.gil@samsung.com> wrote:
>
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
>
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
>
> If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> due to a mismatch between old file size and new data.
>
> To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> previously cleared by the Writeback thread during the commit atomic, is
> set and i_size is updated.
>
> Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
> ---
>  fs/f2fs/inode.c | 4 +---
>  fs/f2fs/super.c | 4 ++++
>  2 files changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
> index aa2f41696a88..83f862578fc8 100644
> --- a/fs/f2fs/inode.c
> +++ b/fs/f2fs/inode.c
> @@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
>         if (f2fs_inode_dirtied(inode, sync))
>                 return;
>
> -       if (f2fs_is_atomic_file(inode)) {
> -               set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +       if (f2fs_is_atomic_file(inode))
>                 return;
> -       }
>
>         mark_inode_dirty_sync(inode);
>  }
> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> index 397df271885c..c08d52c6467a 100644
> --- a/fs/f2fs/super.c
> +++ b/fs/f2fs/super.c
> @@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
>                 inc_page_count(sbi, F2FS_DIRTY_IMETA);
>         }
>         spin_unlock(&sbi->inode_lock[DIRTY_META]);
> +
> +       if (!ret && f2fs_is_atomic_file(inode))
> +               set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +
>         return ret;
>  }
>
> --
> 2.34.1
>
>

Reviewed-by: Daeho Jeong <daehojeong@google.com>

Thanks!

>
> _______________________________________________
> Linux-f2fs-devel mailing list
> Linux-f2fs-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
  2025-03-14 12:06   ` Yeongjin Gil
@ 2025-03-14 21:50     ` Jaegeuk Kim
  -1 siblings, 0 replies; 12+ messages in thread
From: Jaegeuk Kim via Linux-f2fs-devel @ 2025-03-14 21:50 UTC (permalink / raw)
  To: Yeongjin Gil; +Cc: linux-f2fs-devel, daehojeong, linux-kernel, sj1557.seo

On 03/14, Yeongjin Gil wrote:
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> 
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
> 
> If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> due to a mismatch between old file size and new data.
> 
> To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> previously cleared by the Writeback thread during the commit atomic, is
> set and i_size is updated.
> 
> Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
> ---
>  fs/f2fs/inode.c | 4 +---
>  fs/f2fs/super.c | 4 ++++
>  2 files changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
> index aa2f41696a88..83f862578fc8 100644
> --- a/fs/f2fs/inode.c
> +++ b/fs/f2fs/inode.c
> @@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
>  	if (f2fs_inode_dirtied(inode, sync))
>  		return;
>  
> -	if (f2fs_is_atomic_file(inode)) {
> -		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +	if (f2fs_is_atomic_file(inode))
>  		return;
> -	}
>  
>  	mark_inode_dirty_sync(inode);
>  }
> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> index 397df271885c..c08d52c6467a 100644
> --- a/fs/f2fs/super.c
> +++ b/fs/f2fs/super.c
> @@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
>  		inc_page_count(sbi, F2FS_DIRTY_IMETA);
>  	}
>  	spin_unlock(&sbi->inode_lock[DIRTY_META]);
> +
> +	if (!ret && f2fs_is_atomic_file(inode))
> +		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +

I'm not sure what's different here.

Before and after this patch, set_inode_flag(inode, FI_ATOMIC_DIRTIED) is called
only if f2fs_inode_dirtied() returns zero and f2fs_is_atomic_file(inode).

Note, f2fs_mark_inode_dirty_sync() looks the single caller of f2fs_inode_dirtied.


>  	return ret;
>  }
>  
> -- 
> 2.34.1


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
@ 2025-03-14 21:50     ` Jaegeuk Kim
  0 siblings, 0 replies; 12+ messages in thread
From: Jaegeuk Kim @ 2025-03-14 21:50 UTC (permalink / raw)
  To: Yeongjin Gil
  Cc: chao, daehojeong, linux-f2fs-devel, linux-kernel, sj1557.seo,
	s_min.jeong

On 03/14, Yeongjin Gil wrote:
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> 
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
> 
> If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> due to a mismatch between old file size and new data.
> 
> To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> previously cleared by the Writeback thread during the commit atomic, is
> set and i_size is updated.
> 
> Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
> ---
>  fs/f2fs/inode.c | 4 +---
>  fs/f2fs/super.c | 4 ++++
>  2 files changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
> index aa2f41696a88..83f862578fc8 100644
> --- a/fs/f2fs/inode.c
> +++ b/fs/f2fs/inode.c
> @@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
>  	if (f2fs_inode_dirtied(inode, sync))
>  		return;
>  
> -	if (f2fs_is_atomic_file(inode)) {
> -		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +	if (f2fs_is_atomic_file(inode))
>  		return;
> -	}
>  
>  	mark_inode_dirty_sync(inode);
>  }
> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> index 397df271885c..c08d52c6467a 100644
> --- a/fs/f2fs/super.c
> +++ b/fs/f2fs/super.c
> @@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
>  		inc_page_count(sbi, F2FS_DIRTY_IMETA);
>  	}
>  	spin_unlock(&sbi->inode_lock[DIRTY_META]);
> +
> +	if (!ret && f2fs_is_atomic_file(inode))
> +		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> +

I'm not sure what's different here.

Before and after this patch, set_inode_flag(inode, FI_ATOMIC_DIRTIED) is called
only if f2fs_inode_dirtied() returns zero and f2fs_is_atomic_file(inode).

Note, f2fs_mark_inode_dirty_sync() looks the single caller of f2fs_inode_dirtied.


>  	return ret;
>  }
>  
> -- 
> 2.34.1

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
  2025-03-14 21:50     ` Jaegeuk Kim
@ 2025-03-14 22:21       ` Jaegeuk Kim
  -1 siblings, 0 replies; 12+ messages in thread
From: Jaegeuk Kim via Linux-f2fs-devel @ 2025-03-14 22:21 UTC (permalink / raw)
  To: Yeongjin Gil; +Cc: linux-f2fs-devel, daehojeong, linux-kernel, sj1557.seo

On 03/14, Jaegeuk Kim wrote:
> On 03/14, Yeongjin Gil wrote:
> > In the case of the following call stack for an atomic file,
> > FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> > 
> > f2fs_file_write_iter
> >   f2fs_map_blocks
> >     f2fs_reserve_new_blocks
> >       inc_valid_block_count
> >         __mark_inode_dirty(dquot)
> >           f2fs_dirty_inode
> > 
> > If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> > due to a mismatch between old file size and new data.
> > 
> > To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> > FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> > previously cleared by the Writeback thread during the commit atomic, is
> > set and i_size is updated.
> > 
> > Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> > Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> > Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> > Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
> > ---
> >  fs/f2fs/inode.c | 4 +---
> >  fs/f2fs/super.c | 4 ++++
> >  2 files changed, 5 insertions(+), 3 deletions(-)
> > 
> > diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
> > index aa2f41696a88..83f862578fc8 100644
> > --- a/fs/f2fs/inode.c
> > +++ b/fs/f2fs/inode.c
> > @@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
> >  	if (f2fs_inode_dirtied(inode, sync))
> >  		return;
> >  
> > -	if (f2fs_is_atomic_file(inode)) {
> > -		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> > +	if (f2fs_is_atomic_file(inode))
> >  		return;
> > -	}
> >  
> >  	mark_inode_dirty_sync(inode);
> >  }
> > diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> > index 397df271885c..c08d52c6467a 100644
> > --- a/fs/f2fs/super.c
> > +++ b/fs/f2fs/super.c
> > @@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
> >  		inc_page_count(sbi, F2FS_DIRTY_IMETA);
> >  	}
> >  	spin_unlock(&sbi->inode_lock[DIRTY_META]);
> > +
> > +	if (!ret && f2fs_is_atomic_file(inode))
> > +		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> > +
> 
> I'm not sure what's different here.
> 
> Before and after this patch, set_inode_flag(inode, FI_ATOMIC_DIRTIED) is called
> only if f2fs_inode_dirtied() returns zero and f2fs_is_atomic_file(inode).
> 
> Note, f2fs_mark_inode_dirty_sync() looks the single caller of f2fs_inode_dirtied.

Ok, I missed another caller, f2fs_dirty_inode, per discussion with Daeho.
Let me apply this.

> 
> 
> >  	return ret;
> >  }
> >  
> > -- 
> > 2.34.1


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
@ 2025-03-14 22:21       ` Jaegeuk Kim
  0 siblings, 0 replies; 12+ messages in thread
From: Jaegeuk Kim @ 2025-03-14 22:21 UTC (permalink / raw)
  To: Yeongjin Gil
  Cc: chao, daehojeong, linux-f2fs-devel, linux-kernel, sj1557.seo,
	s_min.jeong

On 03/14, Jaegeuk Kim wrote:
> On 03/14, Yeongjin Gil wrote:
> > In the case of the following call stack for an atomic file,
> > FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> > 
> > f2fs_file_write_iter
> >   f2fs_map_blocks
> >     f2fs_reserve_new_blocks
> >       inc_valid_block_count
> >         __mark_inode_dirty(dquot)
> >           f2fs_dirty_inode
> > 
> > If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> > due to a mismatch between old file size and new data.
> > 
> > To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> > FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> > previously cleared by the Writeback thread during the commit atomic, is
> > set and i_size is updated.
> > 
> > Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> > Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> > Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> > Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>
> > ---
> >  fs/f2fs/inode.c | 4 +---
> >  fs/f2fs/super.c | 4 ++++
> >  2 files changed, 5 insertions(+), 3 deletions(-)
> > 
> > diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
> > index aa2f41696a88..83f862578fc8 100644
> > --- a/fs/f2fs/inode.c
> > +++ b/fs/f2fs/inode.c
> > @@ -34,10 +34,8 @@ void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
> >  	if (f2fs_inode_dirtied(inode, sync))
> >  		return;
> >  
> > -	if (f2fs_is_atomic_file(inode)) {
> > -		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> > +	if (f2fs_is_atomic_file(inode))
> >  		return;
> > -	}
> >  
> >  	mark_inode_dirty_sync(inode);
> >  }
> > diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> > index 397df271885c..c08d52c6467a 100644
> > --- a/fs/f2fs/super.c
> > +++ b/fs/f2fs/super.c
> > @@ -1534,6 +1534,10 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync)
> >  		inc_page_count(sbi, F2FS_DIRTY_IMETA);
> >  	}
> >  	spin_unlock(&sbi->inode_lock[DIRTY_META]);
> > +
> > +	if (!ret && f2fs_is_atomic_file(inode))
> > +		set_inode_flag(inode, FI_ATOMIC_DIRTIED);
> > +
> 
> I'm not sure what's different here.
> 
> Before and after this patch, set_inode_flag(inode, FI_ATOMIC_DIRTIED) is called
> only if f2fs_inode_dirtied() returns zero and f2fs_is_atomic_file(inode).
> 
> Note, f2fs_mark_inode_dirty_sync() looks the single caller of f2fs_inode_dirtied.

Ok, I missed another caller, f2fs_dirty_inode, per discussion with Daeho.
Let me apply this.

> 
> 
> >  	return ret;
> >  }
> >  
> > -- 
> > 2.34.1

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
  2025-03-14 12:06   ` Yeongjin Gil
@ 2025-03-14 22:30     ` patchwork-bot+f2fs
  -1 siblings, 0 replies; 12+ messages in thread
From: patchwork-bot+f2fs--- via Linux-f2fs-devel @ 2025-03-14 22:30 UTC (permalink / raw)
  To: Yeongjin Gil
  Cc: daehojeong, linux-kernel, linux-f2fs-devel, jaegeuk, sj1557.seo

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <jaegeuk@kernel.org>:

On Fri, 14 Mar 2025 21:06:51 +0900 you wrote:
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> 
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
> 
> [...]

Here is the summary with links:
  - [f2fs-dev] f2fs: fix to avoid atomicity corruption of atomic file
    https://git.kernel.org/jaegeuk/f2fs/c/c60b556bfe0a

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html




_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
@ 2025-03-14 22:30     ` patchwork-bot+f2fs
  0 siblings, 0 replies; 12+ messages in thread
From: patchwork-bot+f2fs @ 2025-03-14 22:30 UTC (permalink / raw)
  To: Yeongjin Gil
  Cc: jaegeuk, chao, daehojeong, linux-f2fs-devel, linux-kernel,
	sj1557.seo

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <jaegeuk@kernel.org>:

On Fri, 14 Mar 2025 21:06:51 +0900 you wrote:
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> 
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
> 
> [...]

Here is the summary with links:
  - [f2fs-dev] f2fs: fix to avoid atomicity corruption of atomic file
    https://git.kernel.org/jaegeuk/f2fs/c/c60b556bfe0a

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html



^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
  2025-03-14 12:06   ` Yeongjin Gil
@ 2025-03-17  3:19     ` Chao Yu
  -1 siblings, 0 replies; 12+ messages in thread
From: Chao Yu via Linux-f2fs-devel @ 2025-03-17  3:19 UTC (permalink / raw)
  To: Yeongjin Gil, jaegeuk, daehojeong, linux-f2fs-devel, linux-kernel
  Cc: sj1557.seo

On 3/14/25 20:06, Yeongjin Gil wrote:
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> 
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
> 
> If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> due to a mismatch between old file size and new data.
> 
> To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> previously cleared by the Writeback thread during the commit atomic, is
> set and i_size is updated.

I guess we need to add a regression testcase to cover this issue.

> 
> Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] f2fs: fix to avoid atomicity corruption of atomic file
@ 2025-03-17  3:19     ` Chao Yu
  0 siblings, 0 replies; 12+ messages in thread
From: Chao Yu @ 2025-03-17  3:19 UTC (permalink / raw)
  To: Yeongjin Gil, jaegeuk, daehojeong, linux-f2fs-devel, linux-kernel
  Cc: chao, sj1557.seo, s_min.jeong

On 3/14/25 20:06, Yeongjin Gil wrote:
> In the case of the following call stack for an atomic file,
> FI_DIRTY_INODE is set, but FI_ATOMIC_DIRTIED is not subsequently set.
> 
> f2fs_file_write_iter
>   f2fs_map_blocks
>     f2fs_reserve_new_blocks
>       inc_valid_block_count
>         __mark_inode_dirty(dquot)
>           f2fs_dirty_inode
> 
> If FI_ATOMIC_DIRTIED is not set, atomic file can encounter corruption
> due to a mismatch between old file size and new data.
> 
> To resolve this issue, I changed to set FI_ATOMIC_DIRTIED when
> FI_DIRTY_INODE is set. This ensures that FI_DIRTY_INODE, which was
> previously cleared by the Writeback thread during the commit atomic, is
> set and i_size is updated.

I guess we need to add a regression testcase to cover this issue.

> 
> Fixes: fccaa81de87e ("f2fs: prevent atomic file from being dirtied before commit")
> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
> Reviewed-by: Sunmin Jeong <s_min.jeong@samsung.com>
> Signed-off-by: Yeongjin Gil <youngjin.gil@samsung.com>

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,

^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2025-03-17  3:19 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <CGME20250314120658epcas1p2d3ec037c294d4c907ce7fa2fe1c3aa27@epcas1p2.samsung.com>
2025-03-14 12:06 ` [f2fs-dev] [PATCH] f2fs: fix to avoid atomicity corruption of atomic file Yeongjin Gil
2025-03-14 12:06   ` Yeongjin Gil
2025-03-14 19:04   ` [f2fs-dev] " Daeho Jeong
2025-03-14 19:04     ` Daeho Jeong
2025-03-14 21:50   ` Jaegeuk Kim via Linux-f2fs-devel
2025-03-14 21:50     ` Jaegeuk Kim
2025-03-14 22:21     ` [f2fs-dev] " Jaegeuk Kim via Linux-f2fs-devel
2025-03-14 22:21       ` Jaegeuk Kim
2025-03-14 22:30   ` [f2fs-dev] " patchwork-bot+f2fs--- via Linux-f2fs-devel
2025-03-14 22:30     ` patchwork-bot+f2fs
2025-03-17  3:19   ` Chao Yu via Linux-f2fs-devel
2025-03-17  3:19     ` 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.