2022-03-25 02:47:22

by David Sterba

[permalink] [raw]
Subject: Re: [PATCH] fs: btrfs: fix possible use-after-free bug in error handling code of btrfs_get_root_ref()

On Thu, Mar 24, 2022 at 06:44:54AM -0700, Jia-Ju Bai wrote:
> In btrfs_get_root_ref(), when btrfs_insert_fs_root() fails,
> btrfs_put_root() will be called to possibly free the memory area of
> the variable root. However, this variable is then used again in error
> handling code after "goto fail", when ret is not -EEXIST.
>
> To fix this possible bug, btrfs_put_root() is only called when ret is
> -EEXIST for "goto again".
>
> Reported-by: TOTE Robot <[email protected]>
> Signed-off-by: Jia-Ju Bai <[email protected]>
> ---
> fs/btrfs/disk-io.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
> index b30309f187cf..126f244cdf88 100644
> --- a/fs/btrfs/disk-io.c
> +++ b/fs/btrfs/disk-io.c
> @@ -1850,9 +1850,10 @@ static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info,
>
> ret = btrfs_insert_fs_root(fs_info, root);
> if (ret) {
> - btrfs_put_root(root);
> - if (ret == -EEXIST)
> + if (ret == -EEXIST) {
> + btrfs_put_root(root);

I think this fix is correct, though it's not that clear. If you look how
the code changed, there was the unconditional put and then followed by a
free:

8c38938c7bb0 ("btrfs: move the root freeing stuff into btrfs_put_root")

Here it's putting twice where one will be the final free.

And then the whole refcounting gets updated in

4785e24fa5d2 ("btrfs: don't take an extra root ref at allocation time")

which could be removing the wrong put, I'm not yet sure.


2022-03-25 19:38:21

by Jia-Ju Bai

[permalink] [raw]
Subject: Re: [PATCH] fs: btrfs: fix possible use-after-free bug in error handling code of btrfs_get_root_ref()



On 2022/3/25 2:19, David Sterba wrote:
> On Thu, Mar 24, 2022 at 06:44:54AM -0700, Jia-Ju Bai wrote:
>> In btrfs_get_root_ref(), when btrfs_insert_fs_root() fails,
>> btrfs_put_root() will be called to possibly free the memory area of
>> the variable root. However, this variable is then used again in error
>> handling code after "goto fail", when ret is not -EEXIST.
>>
>> To fix this possible bug, btrfs_put_root() is only called when ret is
>> -EEXIST for "goto again".
>>
>> Reported-by: TOTE Robot <[email protected]>
>> Signed-off-by: Jia-Ju Bai <[email protected]>
>> ---
>> fs/btrfs/disk-io.c | 5 +++--
>> 1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
>> index b30309f187cf..126f244cdf88 100644
>> --- a/fs/btrfs/disk-io.c
>> +++ b/fs/btrfs/disk-io.c
>> @@ -1850,9 +1850,10 @@ static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info,
>>
>> ret = btrfs_insert_fs_root(fs_info, root);
>> if (ret) {
>> - btrfs_put_root(root);
>> - if (ret == -EEXIST)
>> + if (ret == -EEXIST) {
>> + btrfs_put_root(root);
> I think this fix is correct, though it's not that clear. If you look how
> the code changed, there was the unconditional put and then followed by a
> free:
>
> 8c38938c7bb0 ("btrfs: move the root freeing stuff into btrfs_put_root")
>
> Here it's putting twice where one will be the final free.
>
> And then the whole refcounting gets updated in
>
> 4785e24fa5d2 ("btrfs: don't take an extra root ref at allocation time")
>
> which could be removing the wrong put, I'm not yet sure.

Thanks for the reply!

I think the bug should be introduced by this commit:
bc44d7c4b2b1 ("btrfs: push btrfs_grab_fs_root into btrfs_get_fs_root")

This commit has a change:
     ret = btrfs_insert_fs_root(fs_info, root);
     if (ret) {
+      btrfs_put_fs_root(root);
         if (ret == -EEXIST) {
             btrfs_free_fs_root(root);
             goto again;
         }

I could add a Fixes tag of this commit in my V2 patch.
Is it okay?


Best wishes,
Jia-Ju Bai

2022-04-02 14:48:28

by David Sterba

[permalink] [raw]
Subject: Re: [PATCH] fs: btrfs: fix possible use-after-free bug in error handling code of btrfs_get_root_ref()

On Fri, Mar 25, 2022 at 04:04:17PM +0800, Jia-Ju Bai wrote:
> >> @@ -1850,9 +1850,10 @@ static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info,
> >>
> >> ret = btrfs_insert_fs_root(fs_info, root);
> >> if (ret) {
> >> - btrfs_put_root(root);
> >> - if (ret == -EEXIST)
> >> + if (ret == -EEXIST) {
> >> + btrfs_put_root(root);
> > I think this fix is correct, though it's not that clear. If you look how
> > the code changed, there was the unconditional put and then followed by a
> > free:
> >
> > 8c38938c7bb0 ("btrfs: move the root freeing stuff into btrfs_put_root")
> >
> > Here it's putting twice where one will be the final free.
> >
> > And then the whole refcounting gets updated in
> >
> > 4785e24fa5d2 ("btrfs: don't take an extra root ref at allocation time")
> >
> > which could be removing the wrong put, I'm not yet sure.
>
> Thanks for the reply!
>
> I think the bug should be introduced by this commit:
> bc44d7c4b2b1 ("btrfs: push btrfs_grab_fs_root into btrfs_get_fs_root")
>
> This commit has a change:
> ???? ret = btrfs_insert_fs_root(fs_info, root);
> ???? if (ret) {
> +??? ? btrfs_put_fs_root(root);
> ???? ??? if (ret == -EEXIST) {
> ???? ??? ??? btrfs_free_fs_root(root);
> ???? ??? ??? goto again;
> ???????? }
>
> I could add a Fixes tag of this commit in my V2 patch.
> Is it okay?

I can add it myself, that's a minor thing. The fix is correct, I've
rewritten the changelog a bit, patch now added to misc-next, thanks.

2023-08-23 08:35:31

by Lee Jones

[permalink] [raw]
Subject: Re: [PATCH] fs: btrfs: fix possible use-after-free bug in error handling code of btrfs_get_root_ref()

On Fri, 01 Apr 2022, David Sterba wrote:

> On Fri, Mar 25, 2022 at 04:04:17PM +0800, Jia-Ju Bai wrote:
> > >> @@ -1850,9 +1850,10 @@ static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info,
> > >>
> > >> ret = btrfs_insert_fs_root(fs_info, root);
> > >> if (ret) {
> > >> - btrfs_put_root(root);
> > >> - if (ret == -EEXIST)
> > >> + if (ret == -EEXIST) {
> > >> + btrfs_put_root(root);
> > > I think this fix is correct, though it's not that clear. If you look how
> > > the code changed, there was the unconditional put and then followed by a
> > > free:
> > >
> > > 8c38938c7bb0 ("btrfs: move the root freeing stuff into btrfs_put_root")
> > >
> > > Here it's putting twice where one will be the final free.
> > >
> > > And then the whole refcounting gets updated in
> > >
> > > 4785e24fa5d2 ("btrfs: don't take an extra root ref at allocation time")
> > >
> > > which could be removing the wrong put, I'm not yet sure.
> >
> > Thanks for the reply!
> >
> > I think the bug should be introduced by this commit:
> > bc44d7c4b2b1 ("btrfs: push btrfs_grab_fs_root into btrfs_get_fs_root")
> >
> > This commit has a change:
> >      ret = btrfs_insert_fs_root(fs_info, root);
> >      if (ret) {
> > +      btrfs_put_fs_root(root);
> >          if (ret == -EEXIST) {
> >              btrfs_free_fs_root(root);
> >              goto again;
> >          }
> >
> > I could add a Fixes tag of this commit in my V2 patch.
> > Is it okay?
>
> I can add it myself, that's a minor thing. The fix is correct, I've
> rewritten the changelog a bit, patch now added to misc-next, thanks.

Where is 'misc-next' please?

--
Lee Jones [李琼斯]

2023-08-23 16:21:32

by Lee Jones

[permalink] [raw]
Subject: Re: [PATCH] fs: btrfs: fix possible use-after-free bug in error handling code of btrfs_get_root_ref()

On Wed, 23 Aug 2023, Lee Jones wrote:

> On Fri, 01 Apr 2022, David Sterba wrote:
>
> > On Fri, Mar 25, 2022 at 04:04:17PM +0800, Jia-Ju Bai wrote:
> > > >> @@ -1850,9 +1850,10 @@ static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info,
> > > >>
> > > >> ret = btrfs_insert_fs_root(fs_info, root);
> > > >> if (ret) {
> > > >> - btrfs_put_root(root);
> > > >> - if (ret == -EEXIST)
> > > >> + if (ret == -EEXIST) {
> > > >> + btrfs_put_root(root);
> > > > I think this fix is correct, though it's not that clear. If you look how
> > > > the code changed, there was the unconditional put and then followed by a
> > > > free:
> > > >
> > > > 8c38938c7bb0 ("btrfs: move the root freeing stuff into btrfs_put_root")
> > > >
> > > > Here it's putting twice where one will be the final free.
> > > >
> > > > And then the whole refcounting gets updated in
> > > >
> > > > 4785e24fa5d2 ("btrfs: don't take an extra root ref at allocation time")
> > > >
> > > > which could be removing the wrong put, I'm not yet sure.
> > >
> > > Thanks for the reply!
> > >
> > > I think the bug should be introduced by this commit:
> > > bc44d7c4b2b1 ("btrfs: push btrfs_grab_fs_root into btrfs_get_fs_root")
> > >
> > > This commit has a change:
> > >      ret = btrfs_insert_fs_root(fs_info, root);
> > >      if (ret) {
> > > +      btrfs_put_fs_root(root);
> > >          if (ret == -EEXIST) {
> > >              btrfs_free_fs_root(root);
> > >              goto again;
> > >          }
> > >
> > > I could add a Fixes tag of this commit in my V2 patch.
> > > Is it okay?
> >
> > I can add it myself, that's a minor thing. The fix is correct, I've
> > rewritten the changelog a bit, patch now added to misc-next, thanks.
>
> Where is 'misc-next' please?

Nevermind. I've just seen a) how old this thread is and b) the commit
subject was changed and subsequently committed as:

168a2f776b976 ("btrfs: fix root ref counts in error handling in btrfs_get_root_ref")

--
Lee Jones [李琼斯]