Re: [edk2-devel] [PATCH v2 2/4] StandaloneMmPkg/Hob: Integer Overflow in CreateHob()

["Sami Mujawar" <sami.mujawar@arm.com>]

Hi Gua,

Please find my response inline marked [SAMI].

Regards,

Sami Mujawar

On 11/01/2024, 15:19, "Guo, Gua" <gua.guo@intel.com <mailto:gua.guo@intel.com>> wrote:


Maybe I can add error handle but they will have several case need to do if it's fine.
It maybe increasing some BIOS size.


Error Handle Error Handle Error Handle Error Handle
A ----------------> B -----------------> C ------------------> CreateHob ----------------> return NULL


All caller chain may need to add it if we really want to prevent it on release.
[SAMI] As I understand, in most cases such an error may not be recoverable. In such cases, would it make sense for the ASSERT() implementation to call a Panic() in release builds? 
Adding a Panic() API has been discussed before at https://edk2.groups.io/g/devel/message/86354.
[/SAMI]

Thanks,
Gua
-----Original Message-----
From: Sami Mujawar <Sami.Mujawar@arm.com <mailto:Sami.Mujawar@arm.com>> 
Sent: Thursday, January 11, 2024 11:02 PM
To: Guo, Gua <gua.guo@intel.com <mailto:gua.guo@intel.com>>; devel@edk2.groups.io <mailto:devel@edk2.groups.io>
Cc: Marc Beatove <mbeatove@google.com <mailto:mbeatove@google.com>>; Ard Biesheuvel <ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org>>; Ni, Ray <ray.ni@intel.com <mailto:ray.ni@intel.com>>; Mathews, John <john.mathews@intel.com <mailto:john.mathews@intel.com>>; Gerd Hoffmann <kraxel@redhat.com <mailto:kraxel@redhat.com>>; nd <nd@arm.com <mailto:nd@arm.com>>
Subject: Re: [PATCH v2 2/4] StandaloneMmPkg/Hob: Integer Overflow in CreateHob()


Hi Gua,


Please find my response inline marked [SAMI].


Regards,


Sami Mujawar
On 11/01/2024, 14:19, "Guo, Gua" <gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>>> wrote:




You mean we need to add below error handle after all callers ?




Hob = CreateHob (...)
ASSERT (Hob != NULL); <---------------- Here [SAMI] That would certainly help catch issues in the debug builds. But the problem with asserts is, they vanish in release builds. 
I think we should consider adding appropriate error handling in the calling functions to make sure that they do not result in a crash.
[/SAMI]








Thanks,
Gua
-----Original Message-----
From: Sami Mujawar <Sami.Mujawar@arm.com <mailto:Sami.Mujawar@arm.com> <mailto:Sami.Mujawar@arm.com <mailto:Sami.Mujawar@arm.com>>>
Sent: Thursday, January 11, 2024 10:06 PM
To: Guo, Gua <gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>>>; devel@edk2.groups.io <mailto:devel@edk2.groups.io> <mailto:devel@edk2.groups.io <mailto:devel@edk2.groups.io>>
Cc: Marc Beatove <mbeatove@google.com <mailto:mbeatove@google.com> <mailto:mbeatove@google.com <mailto:mbeatove@google.com>>>; Ard Biesheuvel <ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org> <mailto:ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org>>>; Ni, Ray <ray.ni@intel.com <mailto:ray.ni@intel.com> <mailto:ray.ni@intel.com <mailto:ray.ni@intel.com>>>; Mathews, John <john.mathews@intel.com <mailto:john.mathews@intel.com> <mailto:john.mathews@intel.com <mailto:john.mathews@intel.com>>>; Gerd Hoffmann <kraxel@redhat.com <mailto:kraxel@redhat.com> <mailto:kraxel@redhat.com <mailto:kraxel@redhat.com>>>; nd <nd@arm.com <mailto:nd@arm.com> <mailto:nd@arm.com <mailto:nd@arm.com>>>
Subject: Re: [PATCH v2 2/4] StandaloneMmPkg/Hob: Integer Overflow in CreateHob()




Hi Gua,




Thank you for this patch.
Please see my response inline marked [SAMI].




Regards,




Sami Mujawar




On 11/01/2024, 09:15, "gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>>>" <gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>>>> wrote:








From: Gua Guo <gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>>>>








REF: https://bugzilla.tianocore.org/show_bug.cgi?id=4166 <https://bugzilla.tianocore.org/show_bug.cgi?id=4166> <https://bugzilla.tianocore.org/show_bug.cgi?id=4166> <https://bugzilla.tianocore.org/show_bug.cgi?id=4166&gt;> <https://bugzilla.tianocore.org/show_bug.cgi?id=4166> <https://bugzilla.tianocore.org/show_bug.cgi?id=4166&gt;> <https://bugzilla.tianocore.org/show_bug.cgi?id=4166&gt;> <https://bugzilla.tianocore.org/show_bug.cgi?id=4166&amp;gt;&gt;>








Fix integer overflow in various CreateHob instances.
Fixes: CVE-2022-36765








The CreateHob() function aligns the requested size to 8 performing the following operation:
```
HobLength = (UINT16)((HobLength + 0x7) & (~0x7)); ```








No checks are performed to ensure this value doesn't overflow, and could lead to CreateHob() returning a smaller HOB than requested, which could lead to OOB HOB accesses.








Reported-by: Marc Beatove <mbeatove@google.com <mailto:mbeatove@google.com> <mailto:mbeatove@google.com <mailto:mbeatove@google.com>> <mailto:mbeatove@google.com <mailto:mbeatove@google.com> <mailto:mbeatove@google.com <mailto:mbeatove@google.com>>>>
Reviewed-by: Ard Biesheuvel <ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org> <mailto:ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org>> <mailto:ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org> <mailto:ardb+tianocore@kernel.org <mailto:ardb+tianocore@kernel.org>>>>
Cc: Sami Mujawar <sami.mujawar@arm.com <mailto:sami.mujawar@arm.com> <mailto:sami.mujawar@arm.com <mailto:sami.mujawar@arm.com>> <mailto:sami.mujawar@arm.com <mailto:sami.mujawar@arm.com> <mailto:sami.mujawar@arm.com <mailto:sami.mujawar@arm.com>>>>
Cc: Ray Ni <ray.ni@intel.com <mailto:ray.ni@intel.com> <mailto:ray.ni@intel.com <mailto:ray.ni@intel.com>> <mailto:ray.ni@intel.com <mailto:ray.ni@intel.com> <mailto:ray.ni@intel.com <mailto:ray.ni@intel.com>>>>
Cc: John Mathew <john.mathews@intel.com <mailto:john.mathews@intel.com> <mailto:john.mathews@intel.com <mailto:john.mathews@intel.com>> <mailto:john.mathews@intel.com <mailto:john.mathews@intel.com> <mailto:john.mathews@intel.com <mailto:john.mathews@intel.com>>>>
Authored-by: Gerd Hoffmann <kraxel@redhat.com <mailto:kraxel@redhat.com> <mailto:kraxel@redhat.com <mailto:kraxel@redhat.com>> <mailto:kraxel@redhat.com <mailto:kraxel@redhat.com> <mailto:kraxel@redhat.com <mailto:kraxel@redhat.com>>>>
Signed-off-by: Gua Guo <gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com> <mailto:gua.guo@intel.com <mailto:gua.guo@intel.com>>>>
---
.../StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c | 7 +++++++
1 file changed, 7 insertions(+)








diff --git a/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c b/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c
index 1550e1babc..bb8426dc0a 100644
--- a/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c
+++ b/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCor
+++ eHobLib.c
@@ -34,6 +34,13 @@ CreateHob (
















HandOffHob = GetHobList ();
























+ //








+ // Check Length to avoid data overflow.








+ //








+ if (HobLength > MAX_UINT16 - 0x7) {








+ return NULL;
[SAMI] Although this fix is correct, I think it shifts the problem somewhere else. 
If the above condition occurs, a NULL is returned. A quick scan reveals that the calling functions do not check the returned value before use.
e.g. https://github.com/tianocore/edk2/blob/master/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c#L167-L170 <https://github.com/tianocore/edk2/blob/master/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c#L167-L170> <https://github.com/tianocore/edk2/blob/master/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c#L167-L170> <https://github.com/tianocore/edk2/blob/master/StandaloneMmPkg/Library/StandaloneMmCoreHobLib/Arm/StandaloneMmCoreHobLib.c#L167-L170&gt;>
There are multiple such places where the calling functions do not check the pointer returned by CreateHob(). 
I believe a similar situation can happen for the other patches in this series.
[/SAMI]




+ }








+








HobLength = (UINT16)((HobLength + 0x7) & (~0x7));
























FreeMemory = HandOffHob->EfiFreeMemoryTop - HandOffHob->EfiFreeMemoryBottom;








--
2.39.2.windows.1































-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#113625): https://edk2.groups.io/g/devel/message/113625
Mute This Topic: https://groups.io/mt/103658960/7686176
Group Owner: devel+owner@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [rebecca@openfw.io]
-=-=-=-=-=-=-=-=-=-=-=-