[edk2-devel] [PATCH v1] ShellPkg: Fix 'ping' command Ip4 receive flow.

[Laszlo Ersek]

(+Liming and stewards; CC Nick)

On 02/27/20 12:02, Maciej Rabeda wrote:
> REF: https://bugzilla.tianocore.org/show_bug.cgi?id=2032
> 
> 'ping' command's receive flow utilizes a single Rx token which it
> attempts to reuse before recycling the previously received packet.
> This causes a situation where under ICMP traffic,
> Ping6OnEchoReplyReceived() function will receive an already
> recycled packet with EFI_SUCCESS token status and finally
> dereference invalid pointers from RxData structure.
> 
> Cc: Ray Ni <ray.ni at intel.com>
> Cc: Zhichao Gao <zhichao.gao at intel.com>
> Signed-off-by: Maciej Rabeda <maciej.rabeda at linux.intel.com>
> ---
>  ShellPkg/Library/UefiShellNetwork1CommandsLib/Ping.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/ShellPkg/Library/UefiShellNetwork1CommandsLib/Ping.c b/ShellPkg/Library/UefiShellNetwork1CommandsLib/Ping.c
> index 23567fa2c1bb..a3fa32515192 100644
> --- a/ShellPkg/Library/UefiShellNetwork1CommandsLib/Ping.c
> +++ b/ShellPkg/Library/UefiShellNetwork1CommandsLib/Ping.c
> @@ -614,6 +614,11 @@ Ping6OnEchoReplyReceived (
>  
>  ON_EXIT:
>  
> +  //
> +  // Recycle the packet before reusing RxToken
> +  //
> +  gBS->SignalEvent (Private->IpChoice == PING_IP_CHOICE_IP6?((EFI_IP6_RECEIVE_DATA*)Private->RxToken.Packet.RxData)->RecycleSignal:((EFI_IP4_RECEIVE_DATA*)Private->RxToken.Packet.RxData)->RecycleSignal);
> +
>    if (Private->RxCount < Private->SendNum) {
>      //
>      // Continue to receive icmp echo reply packets.
> @@ -632,10 +637,6 @@ ON_EXIT:
>      //
>      Private->Status = EFI_SUCCESS;
>    }
> -  //
> -  // Singal to recycle the each rxdata here, not at the end of process.
> -  //
> -  gBS->SignalEvent (Private->IpChoice == PING_IP_CHOICE_IP6?((EFI_IP6_RECEIVE_DATA*)Private->RxToken.Packet.RxData)->RecycleSignal:((EFI_IP4_RECEIVE_DATA*)Private->RxToken.Packet.RxData)->RecycleSignal);
>  }
>  
>  /**
> 

(1) This patch proposes to fix one of the BZs (2032) that fall under
CVE-2019-14559 (joint tracker: 2550).

Consequently:

(1a) Do we want to include this in the upcoming stable tag?

If so, we might want to extend the hard feature freeze by a few days.

(1b) Please append the string " (CVE-2019-14559)" -- note the separating
space! -- to the subject line.

(2) However: I remember from an earlier Bugzilla entry (can't tell
off-hand, which one, sorry) that ShellPkg issues are *never* considered
CVE-worthy, because the shell is not considered a "production element"
of the UEFI boot path.

TianoCore#2032 was originally filed for NetworkPkg, and indeed that
seemed to justify the CVE assignment. However, now that Nick's and
Maciej's analysis shows that NetworkPkg is unaffected (and we know, per
above, that ShellPkg is not CVE-worthy), should we rather *remove* this
BZ from the CVE-2019-14559 umbrella?

Because, in that case, modifying the subject line on the patch is not
necessary; and more importantly, we might not even want to put this into
edk2-stable202002. (It's still a bugfix, but may not be important enough.)

Thanks!
Laszlo


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#55001): https://edk2.groups.io/g/devel/message/55001
Mute This Topic: https://groups.io/mt/71584586/1813853
Group Owner: devel+owner at edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub  [edk2-devel-archive at redhat.com]
-=-=-=-=-=-=-=-=-=-=-=-