linux-kernel - Re: [PATCH] fs: io_uring: remove NULL check before kfree

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] [day] [month] [year] [list]

Message-ID: <CAD-N9QViX0sS+w_g4cL1kaZaRduWwY8ZkOhmLzRL83MP0CVqAw@mail.gmail.com>
Date:   Tue, 14 Jun 2022 17:29:18 +0800
From:   Dongliang Mu <mudongliangabcd@...il.com>
To:     Muchun Song <songmuchun@...edance.com>
Cc:     Dongliang Mu <dzm91@...t.edu.cn>, Jens Axboe <axboe@...nel.dk>,
        Pavel Begunkov <asml.silence@...il.com>,
        io-uring@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] fs: io_uring: remove NULL check before kfree

On Tue, Jun 14, 2022 at 5:26 PM Muchun Song <songmuchun@...edance.com> wrote:
>
> On Tue, Jun 14, 2022 at 5:14 PM Dongliang Mu <dzm91@...t.edu.cn> wrote:
> >
> > From: mudongliang <mudongliangabcd@...il.com>
> >
> > kfree can handle NULL pointer as its argument.
> > According to coccinelle isnullfree check, remove NULL check
> > before kfree operation.
> >
> > Signed-off-by: mudongliang <mudongliangabcd@...il.com>
> > ---
> >  fs/io_uring.c | 15 +++++----------
> >  1 file changed, 5 insertions(+), 10 deletions(-)
> >
> > diff --git a/fs/io_uring.c b/fs/io_uring.c
> > index 3aab4182fd89..bec47eae2a9b 100644
> > --- a/fs/io_uring.c
> > +++ b/fs/io_uring.c
> > @@ -3159,8 +3159,7 @@ static void io_free_batch_list(struct io_ring_ctx *ctx,
> >                         if ((req->flags & REQ_F_POLLED) && req->apoll) {
> >                                 struct async_poll *apoll = req->apoll;
> >
> > -                               if (apoll->double_poll)
> > -                                       kfree(apoll->double_poll);
> > +                               kfree(apoll->double_poll);
> >                                 list_add(&apoll->poll.wait.entry,
> >                                                 &ctx->apoll_cache);
> >                                 req->flags &= ~REQ_F_POLLED;
> > @@ -4499,8 +4498,7 @@ static int io_read(struct io_kiocb *req, unsigned int issue_flags)
> >         kiocb_done(req, ret, issue_flags);
> >  out_free:
> >         /* it's faster to check here then delegate to kfree */
>
> I am feeling you are not on the right way. See the comment
> here.

Thanks for your reply. I ignore them previously. Any method to make
coccicheck ignore such cases?

>
> Thanks.
>
> > -       if (iovec)
> > -               kfree(iovec);
> > +       kfree(iovec);
> >         return 0;
> >  }
> >
> > @@ -4602,8 +4600,7 @@ static int io_write(struct io_kiocb *req, unsigned int issue_flags)
> >         }
> >  out_free:
> >         /* it's reportedly faster than delegating the null check to kfree() */
>
> See here.
>
> > -       if (iovec)
> > -               kfree(iovec);
> > +       kfree(iovec);
> >         return ret;
> >  }
> >
> > @@ -6227,8 +6224,7 @@ static int io_sendmsg(struct io_kiocb *req, unsigned int issue_flags)
> >                 req_set_fail(req);
> >         }
> >         /* fast path, check for non-NULL to avoid function call */
>
> here.
>
> > -       if (kmsg->free_iov)
> > -               kfree(kmsg->free_iov);
> > +       kfree(kmsg->free_iov);
> >         req->flags &= ~REQ_F_NEED_CLEANUP;
> >         if (ret >= 0)
> >                 ret += sr->done_io;
> > @@ -6481,8 +6477,7 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags)
> >         }
> >
> >         /* fast path, check for non-NULL to avoid function call */
>
> And here.
>
> > -       if (kmsg->free_iov)
> > -               kfree(kmsg->free_iov);
> > +       kfree(kmsg->free_iov);
> >         req->flags &= ~REQ_F_NEED_CLEANUP;
> >         if (ret >= 0)
> >                 ret += sr->done_io;
> > --
> > 2.35.1
> >