| Message ID | 20240416123427.614899-4-aalbersh@redhat.com (mailing list archive) |
|---|---|
| State | Accepted, archived |
| Headers | show |
| Series | xfsprogs random fixes found by Coverity scan | expand |
On Tue, Apr 16, 2024 at 02:34:25PM +0200, Andrey Albershteyn wrote: > jdm_parentpaths() doesn't initialize count. If count happens to be > non-zero, following loop can result in access overflow. > > Signed-off-by: Andrey Albershteyn <aalbersh@redhat.com> > --- > io/parent.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/io/parent.c b/io/parent.c > index 8f63607ffec2..5750d98a3b75 100644 > --- a/io/parent.c > +++ b/io/parent.c > @@ -112,7 +112,7 @@ check_parents(parent_t *parentbuf, size_t *parentbuf_size, check_parents is an artifact of the old sgi parent pointers code and (apparently) its need to check parent pointer correctness via xfs_io commands. The Linux parent pointers patchset fixed all those referential integrity problems (thanks, Allison!) and will blow this away, so I think we should ignore this report: https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/commit/io/parent.c?h=pptrs&id=c0854b85c1e8c90ea3eea930a20d1323e61ddb40 --D > jdm_fshandle_t *fshandlep, struct xfs_bstat *statp) > { > int error, i; > - __u32 count; > + __u32 count = 0; > parent_t *entryp; > > do { > -- > 2.42.0 > >
On 2024-04-16 09:15:30, Darrick J. Wong wrote: > On Tue, Apr 16, 2024 at 02:34:25PM +0200, Andrey Albershteyn wrote: > > jdm_parentpaths() doesn't initialize count. If count happens to be > > non-zero, following loop can result in access overflow. > > > > Signed-off-by: Andrey Albershteyn <aalbersh@redhat.com> > > --- > > io/parent.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/io/parent.c b/io/parent.c > > index 8f63607ffec2..5750d98a3b75 100644 > > --- a/io/parent.c > > +++ b/io/parent.c > > @@ -112,7 +112,7 @@ check_parents(parent_t *parentbuf, size_t *parentbuf_size, > > check_parents is an artifact of the old sgi parent pointers code and > (apparently) its need to check parent pointer correctness via xfs_io > commands. The Linux parent pointers patchset fixed all those > referential integrity problems (thanks, Allison!) and will blow this > away, so I think we should ignore this report: > > https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/commit/io/parent.c?h=pptrs&id=c0854b85c1e8c90ea3eea930a20d1323e61ddb40 I see, thanks, will drop this one > > --D > > > jdm_fshandle_t *fshandlep, struct xfs_bstat *statp) > > { > > int error, i; > > - __u32 count; > > + __u32 count = 0; > > parent_t *entryp; > > > > do { > > -- > > 2.42.0 > > > > >
diff --git a/io/parent.c b/io/parent.c index 8f63607ffec2..5750d98a3b75 100644 --- a/io/parent.c +++ b/io/parent.c @@ -112,7 +112,7 @@ check_parents(parent_t *parentbuf, size_t *parentbuf_size, jdm_fshandle_t *fshandlep, struct xfs_bstat *statp) { int error, i; - __u32 count; + __u32 count = 0; parent_t *entryp; do {
jdm_parentpaths() doesn't initialize count. If count happens to be non-zero, following loop can result in access overflow. Signed-off-by: Andrey Albershteyn <aalbersh@redhat.com> --- io/parent.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)