2018-11-24 07:56:57

by Pan Bian

Subject: [PATCH] exportfs: do not read dentry after free

The function dentry_connected calls dput(dentry) to drop the previously
acquired reference to dentry. In this case, dentry can be released.
After that, IS_ROOT(dentry) checks the condition
(dentry == dentry->d_parent), which may result in a use-after-free bug.
This patch directly compares dentry with its parent obtained before
dropping the reference.

Fixes: a056cc8934c("exportfs: stop retrying once we race with
rename/remove")

Signed-off-by: Pan Bian <[email protected]>
---
fs/exportfs/expfs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
index 645158d..a69aaf5 100644
--- a/fs/exportfs/expfs.c
+++ b/fs/exportfs/expfs.c
@@ -77,7 +77,7 @@ static bool dentry_connected(struct dentry *dentry)
struct dentry *parent = dget_parent(dentry);

dput(dentry);
- if (IS_ROOT(dentry)) {
+ if (dentry == parent) { /* is root entry */
dput(parent);
return false;
}
--
2.7.4

2018-11-24 07:59:43

by Amir Goldstein

Subject: Re: [PATCH] exportfs: do not read dentry after free

On Fri, Nov 23, 2018 at 5:16 AM Pan Bian <[email protected]> wrote:
>
> The function dentry_connected calls dput(dentry) to drop the previously
> acquired reference to dentry. In this case, dentry can be released.
> After that, IS_ROOT(dentry) checks the condition
> (dentry == dentry->d_parent), which may result in a use-after-free bug.
> This patch directly compares dentry with its parent obtained before
> dropping the reference.
>
> Fixes: a056cc8934c("exportfs: stop retrying once we race with
> rename/remove")
>

CC Fixes patch author/reviewers

How did you find this? by code review or did this actually happen?

Normally a IS_ROOT dentry would be either DCACHE_DISCONNECTED or
pinned to some super block, but I guess there may be corner cases?

> Signed-off-by: Pan Bian <[email protected]>
> ---
> fs/exportfs/expfs.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
> index 645158d..a69aaf5 100644
> --- a/fs/exportfs/expfs.c
> +++ b/fs/exportfs/expfs.c
> @@ -77,7 +77,7 @@ static bool dentry_connected(struct dentry *dentry)
> struct dentry *parent = dget_parent(dentry);
>
> dput(dentry);
> - if (IS_ROOT(dentry)) {
> + if (dentry == parent) { /* is root entry */
> dput(parent);
> return false;
> }

The change itself looks right, but the name IS_ROOT is confusing
enough as it is. The explicit comment is just plain wrong.
If it was really a root dentry, it wouldn't have been DCACHE_DISCONNECTED
(unless it is a filesystem bug).

Thanks,
Amir.

2018-11-24 08:06:19

by Pan Bian

Subject: Re: [PATCH] exportfs: do not read dentry after free

On Fri, Nov 23, 2018 at 07:58:15AM +0200, Amir Goldstein wrote:
> On Fri, Nov 23, 2018 at 5:16 AM Pan Bian <[email protected]> wrote:
> >
> > The function dentry_connected calls dput(dentry) to drop the previously
> > acquired reference to dentry. In this case, dentry can be released.
> > After that, IS_ROOT(dentry) checks the condition
> > (dentry == dentry->d_parent), which may result in a use-after-free bug.
> > This patch directly compares dentry with its parent obtained before
> > dropping the reference.
> >
> > Fixes: a056cc8934c("exportfs: stop retrying once we race with
> > rename/remove")
> >
>
> CC Fixes patch author/reviewers
>
> How did you find this? by code review or did this actually happen?
>
> Normally a IS_ROOT dentry would be either DCACHE_DISCONNECTED or
> pinned to some super block, but I guess there may be corner cases?

I found this by code review, and I have not yet observed crash.

>
> > Signed-off-by: Pan Bian <[email protected]>
> > ---
> > fs/exportfs/expfs.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
> > index 645158d..a69aaf5 100644
> > --- a/fs/exportfs/expfs.c
> > +++ b/fs/exportfs/expfs.c
> > @@ -77,7 +77,7 @@ static bool dentry_connected(struct dentry *dentry)
> > struct dentry *parent = dget_parent(dentry);
> >
> > dput(dentry);
> > - if (IS_ROOT(dentry)) {
> > + if (dentry == parent) { /* is root entry */
> > dput(parent);
> > return false;
> > }
>
> The change itself looks right, but the name IS_ROOT is confusing
> enough as it is. The explicit comment is just plain wrong.
> If it was really a root dentry, it wouldn't have been DCACHE_DISCONNECTED
> (unless it is a filesystem bug).

I will remove the comment and resubmit the patch.

Thanks a lot,
Pan

>
> Thanks,
> Amir.