block/bio-integrity: fix a memory leak bug
diff mbox series

Message ID 1562830033-24239-1-git-send-email-wang6495@umn.edu
State New
Headers show
Series
  • block/bio-integrity: fix a memory leak bug
Related show

Commit Message

Wenwen Wang July 11, 2019, 7:27 a.m. UTC
From: Wenwen Wang <wenwen@cs.uga.edu>

In bio_integrity_prep(), a kernel buffer is allocated through kmalloc() to
hold integrity metadata. Later on, the buffer will be attached to the bio
structure through bio_integrity_add_page(), which returns the number of
bytes of integrity metadata attached. Due to unexpected situations,
bio_integrity_add_page() may return 0. As a result, bio_integrity_prep()
needs to be terminated with 'false' returned to indicate this error.
However, the allocated kernel buffer is not freed on this execution path,
leading to a memory leak.

To fix this issue, free the allocated buffer before returning from
bio_integrity_prep().

Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
---
 block/bio-integrity.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Comments

Ming Lei July 11, 2019, 9:21 a.m. UTC | #1
On Thu, Jul 11, 2019 at 3:36 PM Wenwen Wang <wang6495@umn.edu> wrote:
>
> From: Wenwen Wang <wenwen@cs.uga.edu>
>
> In bio_integrity_prep(), a kernel buffer is allocated through kmalloc() to
> hold integrity metadata. Later on, the buffer will be attached to the bio
> structure through bio_integrity_add_page(), which returns the number of
> bytes of integrity metadata attached. Due to unexpected situations,
> bio_integrity_add_page() may return 0. As a result, bio_integrity_prep()
> needs to be terminated with 'false' returned to indicate this error.
> However, the allocated kernel buffer is not freed on this execution path,
> leading to a memory leak.
>
> To fix this issue, free the allocated buffer before returning from
> bio_integrity_prep().
>
> Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
> ---
>  block/bio-integrity.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/block/bio-integrity.c b/block/bio-integrity.c
> index 4db6208..bfae10c 100644
> --- a/block/bio-integrity.c
> +++ b/block/bio-integrity.c
> @@ -276,8 +276,10 @@ bool bio_integrity_prep(struct bio *bio)
>                 ret = bio_integrity_add_page(bio, virt_to_page(buf),
>                                              bytes, offset);
>
> -               if (ret == 0)
> +               if (ret == 0) {
> +                       kfree(buf);
>                         return false;
> +               }

This way may not be enough, and the bio payload needs to be freed.

And you may refer to the error handling for 'IS_ERR(bip)', and bio->bi_status
needs to be set, and bio_endio() needs to be called too.


Thanks,
Ming Lei
Wenwen Wang July 11, 2019, 7:18 p.m. UTC | #2
On Thu, Jul 11, 2019 at 4:22 AM Ming Lei <tom.leiming@gmail.com> wrote:
>
> On Thu, Jul 11, 2019 at 3:36 PM Wenwen Wang <wang6495@umn.edu> wrote:
> >
> > From: Wenwen Wang <wenwen@cs.uga.edu>
> >
> > In bio_integrity_prep(), a kernel buffer is allocated through kmalloc() to
> > hold integrity metadata. Later on, the buffer will be attached to the bio
> > structure through bio_integrity_add_page(), which returns the number of
> > bytes of integrity metadata attached. Due to unexpected situations,
> > bio_integrity_add_page() may return 0. As a result, bio_integrity_prep()
> > needs to be terminated with 'false' returned to indicate this error.
> > However, the allocated kernel buffer is not freed on this execution path,
> > leading to a memory leak.
> >
> > To fix this issue, free the allocated buffer before returning from
> > bio_integrity_prep().
> >
> > Signed-off-by: Wenwen Wang <wenwen@cs.uga.edu>
> > ---
> >  block/bio-integrity.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/block/bio-integrity.c b/block/bio-integrity.c
> > index 4db6208..bfae10c 100644
> > --- a/block/bio-integrity.c
> > +++ b/block/bio-integrity.c
> > @@ -276,8 +276,10 @@ bool bio_integrity_prep(struct bio *bio)
> >                 ret = bio_integrity_add_page(bio, virt_to_page(buf),
> >                                              bytes, offset);
> >
> > -               if (ret == 0)
> > +               if (ret == 0) {
> > +                       kfree(buf);
> >                         return false;
> > +               }
>
> This way may not be enough, and the bio payload needs to be freed.
>
> And you may refer to the error handling for 'IS_ERR(bip)', and bio->bi_status
> needs to be set, and bio_endio() needs to be called too.

Thanks for your comments! I will rework the patch.

Wenwen

Patch
diff mbox series

diff --git a/block/bio-integrity.c b/block/bio-integrity.c
index 4db6208..bfae10c 100644
--- a/block/bio-integrity.c
+++ b/block/bio-integrity.c
@@ -276,8 +276,10 @@  bool bio_integrity_prep(struct bio *bio)
 		ret = bio_integrity_add_page(bio, virt_to_page(buf),
 					     bytes, offset);
 
-		if (ret == 0)
+		if (ret == 0) {
+			kfree(buf);
 			return false;
+		}
 
 		if (ret < bytes)
 			break;