RE: [f2fs-dev] [PATCH] f2fs: use inode mutex to keep atomicity off2fs_falloc

[Chao Yu]

Hi,

> -----Original Message-----
> From: Jaegeuk Kim [mailto:jaegeuk.kim@xxxxxxxxxxx]
> Sent: Tuesday, January 28, 2014 2:00 PM
> To: Chao Yu
> Cc: linux-fsdevel@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx;
> linux-f2fs-devel@xxxxxxxxxxxxxxxxxxxxx
> Subject: Re: [f2fs-dev] [PATCH] f2fs: use inode mutex to keep atomicity of f2fs_falloc
> 
> Hi,
> 
> 2014-01-28 (í), 10:29 +0800, Chao Yu:
> > Previously without protection of inode mutex, f2fs_falloc and other data
> > correlated operations will interfere with each other.
> 
> Could you explain a little bit more what kind of scenarios wrt this?

1. Without protection of i_mutex, status(e.g. i_size) of inode could be changed
by other operation(e.g. vfs_write()/vfs_truncate()) when we fallocate a file, it
may cause us walk into wrong branch code after we check these status of inode,
and vice versa.
2. i_size_write()/inode_newsize_ok() is called in f2fs_fallocate(), usually we
use i_mutex to protect these functions.

> 
> 
> > So let's use inode mutex to keep atomicity of f2fs_falloc.
> >
> > Signed-off-by: Chao Yu <chao2.yu@xxxxxxxxxxx>
> > ---
> >  fs/f2fs/file.c |    5 +++++
> >  1 file changed, 5 insertions(+)
> >
> > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> > index 85e91ca..ece380f 100644
> > --- a/fs/f2fs/file.c
> > +++ b/fs/f2fs/file.c
> > @@ -559,6 +559,8 @@ static long f2fs_fallocate(struct file *file, int mode,
> >  	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
> >  		return -EOPNOTSUPP;
> >
> > +	mutex_lock(&inode->i_mutex);
> > +
> >  	if (mode & FALLOC_FL_PUNCH_HOLE)
> >  		ret = punch_hole(inode, offset, len);
> >  	else
> > @@ -568,6 +570,9 @@ static long f2fs_fallocate(struct file *file, int mode,
> >  		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
> >  		mark_inode_dirty(inode);
> >  	}
> > +
> > +	mutex_unlock(&inode->i_mutex);
> > +
> >  	trace_f2fs_fallocate(inode, mode, offset, len, ret);
> >  	return ret;
> >  }
> 
> --
> Jaegeuk Kim
> Samsung

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/