NFSv4: Fix a memory leak bug
diff mbox series

Message ID 1566287651-11386-1-git-send-email-wenwen@cs.uga.edu
State Superseded
Headers show
Series
  • NFSv4: Fix a memory leak bug
Related show

Commit Message

Wenwen Wang Aug. 20, 2019, 7:54 a.m. UTC
In nfs4_try_migration(), if nfs4_begin_drain_session() fails, the
previously allocated 'page' and 'locations' are not deallocated, leading to
memory leaks. To fix this issue, free 'page' and 'locations' before
returning the error.

Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
---
 fs/nfs/nfs4state.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

Comments

Schumaker, Anna Aug. 20, 2019, 1:41 p.m. UTC | #1
Hi Wenwen,

On Tue, 2019-08-20 at 02:54 -0500, Wenwen Wang wrote:
> In nfs4_try_migration(), if nfs4_begin_drain_session() fails, the
> previously allocated 'page' and 'locations' are not deallocated,
> leading to
> memory leaks. To fix this issue, free 'page' and 'locations' before
> returning the error.
> 
> Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
> ---
>  fs/nfs/nfs4state.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
> index cad4e06..37823dc 100644
> --- a/fs/nfs/nfs4state.c
> +++ b/fs/nfs/nfs4state.c
> @@ -2095,8 +2095,12 @@ static int nfs4_try_migration(struct
> nfs_server *server, const struct cred *cred
>         }
> 
>         status = nfs4_begin_drain_session(clp);
> -       if (status != 0)
> +       if (status != 0) {
> +               if (page != NULL)
> +                       __free_page(page);
> +               kfree(locations);
>                 return status;

Thanks for the suggestion! I think a better option would be to switch
the "return status" into a "goto out" so we can keep all our cleanup
code in a single place in case we ever need to change it in the future.

What do you think?
Anna

> +       }
> 
>         status = nfs4_replace_transport(server, locations);
>         if (status != 0) {
> --
> 2.7.4
>
Wenwen Wang Aug. 20, 2019, 6:53 p.m. UTC | #2
On Tue, Aug 20, 2019 at 9:41 AM Schumaker, Anna
<Anna.Schumaker@netapp.com> wrote:
>
> Hi Wenwen,
>
> On Tue, 2019-08-20 at 02:54 -0500, Wenwen Wang wrote:
> > In nfs4_try_migration(), if nfs4_begin_drain_session() fails, the
> > previously allocated 'page' and 'locations' are not deallocated,
> > leading to
> > memory leaks. To fix this issue, free 'page' and 'locations' before
> > returning the error.
> >
> > Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
> > ---
> >  fs/nfs/nfs4state.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
> > index cad4e06..37823dc 100644
> > --- a/fs/nfs/nfs4state.c
> > +++ b/fs/nfs/nfs4state.c
> > @@ -2095,8 +2095,12 @@ static int nfs4_try_migration(struct
> > nfs_server *server, const struct cred *cred
> >         }
> >
> >         status = nfs4_begin_drain_session(clp);
> > -       if (status != 0)
> > +       if (status != 0) {
> > +               if (page != NULL)
> > +                       __free_page(page);
> > +               kfree(locations);
> >                 return status;
>
> Thanks for the suggestion! I think a better option would be to switch
> the "return status" into a "goto out" so we can keep all our cleanup
> code in a single place in case we ever need to change it in the future.
>
> What do you think?

Thanks for your comments! I will rework the patch.

Wenwen

> Anna
>
> > +       }
> >
> >         status = nfs4_replace_transport(server, locations);
> >         if (status != 0) {
> > --
> > 2.7.4
> >

Patch
diff mbox series

diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index cad4e06..37823dc 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -2095,8 +2095,12 @@  static int nfs4_try_migration(struct nfs_server *server, const struct cred *cred
 	}
 
 	status = nfs4_begin_drain_session(clp);
-	if (status != 0)
+	if (status != 0) {
+		if (page != NULL)
+			__free_page(page);
+		kfree(locations);
 		return status;
+	}
 
 	status = nfs4_replace_transport(server, locations);
 	if (status != 0) {