netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hawkins Jiawei <yin31149@gmail.com>
To: edumazet@google.com
Cc: 18801353760@163.com, davem@davemloft.net,
	johannes@sipsolutions.net, keescook@chromium.org,
	kuba@kernel.org, linux-kernel@vger.kernel.org,
	linux-wireless@vger.kernel.org, netdev@vger.kernel.org,
	pabeni@redhat.com,
	syzbot+473754e5af963cf014cf@syzkaller.appspotmail.com,
	syzkaller-bugs@googlegroups.com, yin31149@gmail.com
Subject: Re: [PATCH] wext: use flex array destination for memcpy()
Date: Tue, 27 Sep 2022 07:17:47 +0800	[thread overview]
Message-ID: <20220926231747.4841-1-yin31149@gmail.com> (raw)
In-Reply-To: <CANn89i+vx3MJ8D1zz1jUh2XZbFvPicC1RwzREzYXc_TvAFBVxg@mail.gmail.com>

On Tue, 27 Sept 2022 at 00:14, Eric Dumazet <edumazet@google.com> wrote:
>
> On Mon, Sep 26, 2022 at 8:09 AM Hawkins Jiawei <yin31149@gmail.com> wrote:
> >
> > Syzkaller reports refcount bug as follows:
>
> This has nothing to do with a refcount bug...
Sorry for my typo error, it should be a "buffer overflow false positive" bug
as Kees Cook points out.

I will correct it in my v2 patch.

>
> > ------------[ cut here ]------------
>
>
> > memcpy: detected field-spanning write (size 8) of single field
> >         "&compat_event->pointer" at net/wireless/wext-core.c:623 (size 4)
> > WARNING: CPU: 0 PID: 3607 at net/wireless/wext-core.c:623
> >         wireless_send_event+0xab5/0xca0 net/wireless/wext-core.c:623
> > Modules linked in:
> > CPU: 1 PID: 3607 Comm: syz-executor659 Not tainted
> >         6.0.0-rc6-next-20220921-syzkaller #0
> > [...]
> > Call Trace:
> >  <TASK>
> >  ioctl_standard_call+0x155/0x1f0 net/wireless/wext-core.c:1022
> >  wireless_process_ioctl+0xc8/0x4c0 net/wireless/wext-core.c:955
> >  wext_ioctl_dispatch net/wireless/wext-core.c:988 [inline]
> >  wext_ioctl_dispatch net/wireless/wext-core.c:976 [inline]
> >  wext_handle_ioctl+0x26b/0x280 net/wireless/wext-core.c:1049
> >  sock_ioctl+0x285/0x640 net/socket.c:1220
> >  vfs_ioctl fs/ioctl.c:51 [inline]
> >  __do_sys_ioctl fs/ioctl.c:870 [inline]
> >  __se_sys_ioctl fs/ioctl.c:856 [inline]
> >  __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:856
> >  do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> >  do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
> >  entry_SYSCALL_64_after_hwframe+0x63/0xcd
> >  [...]
> >  </TASK>
> >
> > Wireless events will be sent on the appropriate channels in
> > wireless_send_event(). Different wireless events may have different
> > payload structure and size, so kernel uses **len** and **cmd** field
> > in struct __compat_iw_event as wireless event common LCP part, uses
> > **pointer** as a label to mark the position of remaining different part.
> >
> > Yet the problem is that, **pointer** is a compat_caddr_t type, which may
> > be smaller than the relative structure at the same position. So during
> > wireless_send_event() tries to parse the wireless events payload, it may
> > trigger the memcpy() run-time destination buffer bounds checking when the
> > relative structure's data is copied to the position marked by **pointer**.
> >
> > This patch solves it by introducing flexible-array field **ptr_bytes**,
> > to mark the position of the wireless events remaining part next to
> > LCP part. What's more, this patch also adds **ptr_len** variable in
> > wireless_send_event() to improve its maintainability.
> >
> > Reported-and-tested-by: syzbot+473754e5af963cf014cf@syzkaller.appspotmail.com
> > Link: https://lore.kernel.org/all/00000000000070db2005e95a5984@google.com/
> > Suggested-by: Kees Cook <keescook@chromium.org>
> > Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
> > ---
> >  include/linux/wireless.h | 10 +++++++++-
> >  net/wireless/wext-core.c | 17 ++++++++++-------
> >  2 files changed, 19 insertions(+), 8 deletions(-)
> >
> > diff --git a/include/linux/wireless.h b/include/linux/wireless.h
> > index 2d1b54556eff..e6e34d74dda0 100644
> > --- a/include/linux/wireless.h
> > +++ b/include/linux/wireless.h
> > @@ -26,7 +26,15 @@ struct compat_iw_point {
> >  struct __compat_iw_event {
> >         __u16           len;                    /* Real length of this stuff */
> >         __u16           cmd;                    /* Wireless IOCTL */
> > -       compat_caddr_t  pointer;
> > +
> > +       union {
> > +               compat_caddr_t  pointer;
> > +
> > +               /* we need ptr_bytes to make memcpy() run-time destination
> > +                * buffer bounds checking happy, nothing special
> > +                */
> > +               DECLARE_FLEX_ARRAY(__u8, ptr_bytes);
> > +       };
> >  };
> >  #define IW_EV_COMPAT_LCP_LEN offsetof(struct __compat_iw_event, pointer)
> >  #define IW_EV_COMPAT_POINT_OFF offsetof(struct compat_iw_point, length)
> > diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c
> > index 76a80a41615b..fe8765c4075d 100644
> > --- a/net/wireless/wext-core.c
> > +++ b/net/wireless/wext-core.c
> > @@ -468,6 +468,7 @@ void wireless_send_event(struct net_device *        dev,
> >         struct __compat_iw_event *compat_event;
> >         struct compat_iw_point compat_wrqu;
> >         struct sk_buff *compskb;
> > +       int ptr_len;
> >  #endif
> >
> >         /*
> > @@ -582,6 +583,9 @@ void wireless_send_event(struct net_device *        dev,
> >         nlmsg_end(skb, nlh);
> >  #ifdef CONFIG_COMPAT
> >         hdr_len = compat_event_type_size[descr->header_type];
> > +
> > +       /* ptr_len is remaining size in event header apart from LCP */
> > +       ptr_len = hdr_len - IW_EV_COMPAT_LCP_LEN;
> >         event_len = hdr_len + extra_len;
> >
> >         compskb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
> > @@ -612,16 +616,15 @@ void wireless_send_event(struct net_device *      dev,
> >         if (descr->header_type == IW_HEADER_TYPE_POINT) {
> >                 compat_wrqu.length = wrqu->data.length;
> >                 compat_wrqu.flags = wrqu->data.flags;
> > -               memcpy(&compat_event->pointer,
> > -                       ((char *) &compat_wrqu) + IW_EV_COMPAT_POINT_OFF,
> > -                       hdr_len - IW_EV_COMPAT_LCP_LEN);
> > +               memcpy(compat_event->ptr_bytes,
> > +                      ((char *)&compat_wrqu) + IW_EV_COMPAT_POINT_OFF,
> > +                       ptr_len);
> >                 if (extra_len)
> > -                       memcpy(((char *) compat_event) + hdr_len,
> > -                               extra, extra_len);
> > +                       memcpy(&compat_event->ptr_bytes[ptr_len],
> > +                              extra, extra_len);
> >         } else {
> >                 /* extra_len must be zero, so no if (extra) needed */
> > -               memcpy(&compat_event->pointer, wrqu,
> > -                       hdr_len - IW_EV_COMPAT_LCP_LEN);
> > +               memcpy(compat_event->ptr_bytes, wrqu, ptr_len);
> >         }
> >
> >         nlmsg_end(compskb, nlh);
> > --
> > 2.25.1
> >


  reply	other threads:[~2022-09-26 23:18 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-23 16:05 [syzbot] WARNING in wireless_send_event syzbot
2022-09-24  7:10 ` [PATCH] Add linux-next specific files for 20220923 Hawkins Jiawei
2022-09-24  7:26   ` Kees Cook
2022-09-24 11:44     ` Hawkins Jiawei
2022-09-24 15:55       ` Hawkins Jiawei
2022-09-24 15:55         ` syzbot
2022-09-24 16:13           ` Hawkins Jiawei
2022-09-24 16:33             ` [syzbot] WARNING in wireless_send_event syzbot
2022-09-24 16:26         ` [PATCH] Add linux-next specific files for 20220923 Kees Cook
2022-09-25  6:25           ` Hawkins Jiawei
2022-09-24  7:32   ` [syzbot] WARNING in wireless_send_event syzbot
2022-09-26 11:59 ` Hawkins Jiawei
2022-09-26 12:24   ` syzbot
2022-09-26 15:09 ` [PATCH] wext: use flex array destination for memcpy() Hawkins Jiawei
2022-09-26 16:14   ` Eric Dumazet
2022-09-26 23:17     ` Hawkins Jiawei [this message]
2022-09-26 17:43   ` Kees Cook
2022-09-26 23:34 ` [PATCH wireless-next v2] " Hawkins Jiawei

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220926231747.4841-1-yin31149@gmail.com \
    --to=yin31149@gmail.com \
    --cc=18801353760@163.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=johannes@sipsolutions.net \
    --cc=keescook@chromium.org \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=syzbot+473754e5af963cf014cf@syzkaller.appspotmail.com \
    --cc=syzkaller-bugs@googlegroups.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).