linux-kernel - Re: [PATCH] btrfs: Fix refcnt leak in btrfs_recover

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAL3q7H4Yr2cdEgLWVAR2N_hPYEsra8yLA89meUMcvxH1VjtA6A@mail.gmail.com>
Date:   Mon, 20 Apr 2020 18:35:56 +0100
From:   Filipe Manana <fdmanana@...il.com>
To:     Xiyu Yang <xiyuyang19@...an.edu.cn>
Cc:     Chris Mason <clm@...com>, Josef Bacik <josef@...icpanda.com>,
        David Sterba <dsterba@...e.com>, Jeff Mahoney <jeffm@...e.com>,
        linux-btrfs <linux-btrfs@...r.kernel.org>,
        linux-kernel@...r.kernel.org, yuanxzhang@...an.edu.cn,
        kjlu@....edu, Xin Tan <tanxin.ctf@...il.com>
Subject: Re: [PATCH] btrfs: Fix refcnt leak in btrfs_recover_relocation

On Mon, Apr 20, 2020 at 6:34 PM Filipe Manana <fdmanana@...il.com> wrote:
>
> On Mon, Apr 20, 2020 at 6:50 AM Xiyu Yang <xiyuyang19@...an.edu.cn> wrote:
> >
> > btrfs_recover_relocation() invokes btrfs_join_transaction(), which joins
> > a btrfs_trans_handle object into transactions and returns a reference of
> > it with increased refcount to "trans".
> >
> > When btrfs_recover_relocation() returns, "trans" becomes invalid, so the
> > refcount should be decreased to keep refcount balanced.
> >
> > The reference counting issue happens in one exception handling path of
> > btrfs_recover_relocation(). When read_fs_root() failed, the refcnt
> > increased by btrfs_join_transaction() is not decreased, causing a refcnt
> > leak.
> >
> > Fix this issue by calling btrfs_end_transaction() on this error path
> > when read_fs_root() failed.
> >
> > Fixes: 79787eaab461 ("btrfs: replace many BUG_ONs with proper error
> > handling")
> > Signed-off-by: Xiyu Yang <xiyuyang19@...an.edu.cn>
> > Signed-off-by: Xin Tan <tanxin.ctf@...il.com>
>
> Reviewed-by: Filipe Manana <fdmanana@...e.com>
>
> Looks good, thanks.

Btw, the subject could be more clear.
Instead of

"btrfs: Fix refcnt leak in btrfs_recover_relocation"

something like

"btrfs: fix transaction leak in ..."

David can probably fixup that when he picks the patch.

Thanks.

>
> > ---
> >  fs/btrfs/relocation.c | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
> > index 995d4b8b1cfd..46a451594c7a 100644
> > --- a/fs/btrfs/relocation.c
> > +++ b/fs/btrfs/relocation.c
> > @@ -4606,6 +4606,7 @@ int btrfs_recover_relocation(struct btrfs_root *root)
> >                 if (IS_ERR(fs_root)) {
> >                         err = PTR_ERR(fs_root);
> >                         list_add_tail(&reloc_root->root_list, &reloc_roots);
> > +                       btrfs_end_transaction(trans);
> >                         goto out_free;
> >                 }
> >
> > --
> > 2.7.4
> >
>
>
> --
> Filipe David Manana,
>
> “Whether you think you can, or you think you can't — you're right.”



-- 
Filipe David Manana,

“Whether you think you can, or you think you can't — you're right.”