Rickard Strandqvist
2014-May-22 20:46 UTC
[Ocfs2-devel] [PATCH] fs: ocfs2: move_extents.c: Fix for possible null pointer dereference
There is otherwise a risk of a possible null pointer dereference. Was largely found by using a static code analysis program called cppcheck. Signed-off-by: Rickard Strandqvist <rickard_strandqvist at spectrumdigital.se> --- fs/ocfs2/move_extents.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index 599eb4c..6a8e3c8 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -902,11 +902,13 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context) struct inode *inode = context->inode; struct ocfs2_dinode *di; struct buffer_head *di_bh = NULL; - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct ocfs2_super *osb; if (!inode) return -ENOENT; + osb = OCFS2_SB(inode->i_sb); + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) return -EROFS; -- 1.7.10.4
Jeff Liu
2014-May-26 03:28 UTC
[Ocfs2-devel] [PATCH] fs: ocfs2: move_extents.c: Fix for possible null pointer dereference
Hi, On 05/23/2014 04:46 AM, Rickard Strandqvist wrote:> There is otherwise a risk of a possible null pointer dereference. > > Was largely found by using a static code analysis program called cppcheck. > > Signed-off-by: Rickard Strandqvist <rickard_strandqvist at spectrumdigital.se> > --- > fs/ocfs2/move_extents.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c > index 599eb4c..6a8e3c8 100644 > --- a/fs/ocfs2/move_extents.c > +++ b/fs/ocfs2/move_extents.c > @@ -902,11 +902,13 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context) > struct inode *inode = context->inode; > struct ocfs2_dinode *di; > struct buffer_head *di_bh = NULL; > - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); > + struct ocfs2_super *osb; > > if (!inode) > return -ENOENT; > > + osb = OCFS2_SB(inode->i_sb); > + > if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) > return -EROFS;Thanks for your patch, it looks reasonable if we consider it in the context of above function only. However, the inode should not be NULL in any case given that ocfs2_move_extents() is called by ocfs2_ioctl_move_extents() at where the inode is already validated. IMO, maybe we can just get rid of the useless inode pre-checkup, i.e, if (!inode) return -ENOENT; Thanks, -Jeff
Rickard Strandqvist
2014-May-27 20:21 UTC
Re: [Ocfs2-devel] [PATCH] fs: ocfs2: move_extents.c: Fix for possible null pointer dereference
Hi Ok good, then everything is under control. The result has been that I've done a number of these types of patches to :) Sending a new patch of that kind then. Best regards Rickard Strandqvist 2014-05-26 5:28 GMT+02:00 Jeff Liu <jeff.liu@oracle.com>:> Hi, > > On 05/23/2014 04:46 AM, Rickard Strandqvist wrote: >> There is otherwise a risk of a possible null pointer dereference. >> >> Was largely found by using a static code analysis program called cppcheck. >> >> Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se> >> --- >> fs/ocfs2/move_extents.c | 4 +++- >> 1 file changed, 3 insertions(+), 1 deletion(-) >> >> diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c >> index 599eb4c..6a8e3c8 100644 >> --- a/fs/ocfs2/move_extents.c >> +++ b/fs/ocfs2/move_extents.c >> @@ -902,11 +902,13 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context) >> struct inode *inode = context->inode; >> struct ocfs2_dinode *di; >> struct buffer_head *di_bh = NULL; >> - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); >> + struct ocfs2_super *osb; >> >> if (!inode) >> return -ENOENT; >> >> + osb = OCFS2_SB(inode->i_sb); >> + >> if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) >> return -EROFS; > > Thanks for your patch, it looks reasonable if we consider it in the context > of above function only. However, the inode should not be NULL in any case > given that ocfs2_move_extents() is called by ocfs2_ioctl_move_extents() at > where the inode is already validated. > > IMO, maybe we can just get rid of the useless inode pre-checkup, i.e, > > if (!inode) > return -ENOENT; > > > Thanks, > -Jeff