Message-ID: <827cfca7-5b1a-124f-5a11-1f2573fde85a@x41-dsec.de>
Date: Tue, 14 Aug 2018 15:46:17 +0200
From: X41 D-Sec GmbH Advisories <advisories@...-dsec.de>
To: bugtraq@...urityfocus.com, fulldisclosure@...lists.org,
 oss-security@...ts.openwall.com
Subject: [FD] X41 D-Sec GmbH Security Advisory X41-2018-005: Multiple
 Vulnerabilities in Apple smartcardservices

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

X41 D-Sec GmbH Security Advisory: X41-2018-005

Multiple Vulnerabilities in Apple smartcardservices
===================================================


Overview
- --------
Confirmed Affected Versions: e3eb96a6eff9d02497a51b3c155a10fa5989021f
Confirmed Patched Versions: 8eef01a5e218ae78cc358de32213b50a601662de
Vendor: Apple
Vendor URL: https://smartcardservices.github.io/
Credit: X41 D-Sec GmbH, Eric Sesterhenn
Status: Public
Advisory-URL:
https://www.x41-dsec.de/lab/advisories/x41-2018-005-smartcardservices/


Summary and Impact
- ------------------
Attackers with local access can exploit security issues in the
smartcard driver. These result in memory corruptions, which might lead
to code execution. Since smartcards can be used for authentication,
the vulnerabilities may allow an attacker to login to the system
without valid credentials as any user.
X41 did not perform a full test or audit on the software.


Product Description
- -------------------
The Smart Card Services project is comprised of several components
which, when combined, provide the necessary abstraction layer and
integration of smart cards into Apple’s CDSA implementation.

Stack based buffer overflow
===========================
Severity Rating: Medium
Vector: APDU Response
CVE: CVE-2018-4300
CWE: 120
CVSS Score: 7.1 (High)
CVSS Vector: CVSS:3.0/AV:P/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:H


Summary and Impact
- ------------------
In file Tokend/CAC/CACRecord.cpp the function
CACCertificateRecord::getDataAttribute() might overwrite the value
certificate and possibly other stack data, if a smartcard provides
malicious data.

{% highlight c++ %}
        unsigned char command[] = { 0x80, 0x36, 0x00, 0x00, 0x64 };
        unsigned char result[MAXBUFFERSIZE];
        sizet resultLength = sizeof(result);
        uint8 certificate[CACMAXSIZECERT];
        uint8 uncompressed[CACMAXSIZECERT];
        sizet certificateLength = 0;
        try
        {
                PCSC::Transaction (cacToken);
                cacToken.select(mApplication);
                uint32t cacreturn;
                do
                {
                        cacreturn = cacToken.exchangeAPDU(command,
sizeof(command), result,
                                resultLength);
                        if ((cacreturn & 0xFF00) != 0x6300)
                                CACError::check(cacreturn);
                        sizet requested = command[4];
                        if (resultLength != requested + 2)
                PCSC::Error::throwMe(SCARDEPROTOMISMATCH);
                        memcpy(certificate + certificateLength,
result, resultLength - 2);
                        certificateLength += resultLength - 2;
                        // Number of bytes to fetch next time around
is in the last byte
                        // returned.
                        command[4] = cacreturn & 0xFF;
                } while ((cacreturn & 0xFF00) == 0x6300);
        }
        catch (...)
        {
                return NULL;
        }
{% endhighlight %}

As long as the smartcard returns a return code of 0x63FF, more data is
copied into the certificate buffer, causing a stack based overflow. A
malicious smartcard is able to control all of the overflowed bytes.


Workarounds
- -----------
None

Stack based buffer overflow with limited input
==============================================
Severity Rating: Medium
Vector: APDU Response
CVE: CVE-2018-4301
CWE: 120
CVSS Score: 7.1 (High)
CVSS Vector: CVSS:3.0/AV:P/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:H


Summary and Impact
- ------------------
In file Tokend/PKCS11/GemaltoKeyHandle.cpp the function
GemaltoPrivateKeyRecord::computeDecrypt() might overwrite the value
strData if the supplied dataLength is too big.

{% highlight c++ %}
void GemaltoPrivateKeyRecord::computeDecrypt(GemaltoToken
&gemaltoToken, CKULONG mech, const AccessCredentials *cred, unsigned
char *data, sizet dataLength, unsigned char output, size_t &outputLength)
{
        GemaltoToken::log("\nGemaltoPrivateKeyRecord::computeDecrypt
<BEGIN>\n");
        GemaltoToken::log("GemaltoPrivateKeyRecord::computeDecrypt -
mechanism <%lu>\n", mech);
        GemaltoToken::log("GemaltoPrivateKeyRecord::computeDecrypt -
cred <%p>\n", cred);
        char strData[6000];
        memset(strData, '\0', sizeof(strData));
        char str = strData;
        for (size_t i=0; i<dataLength; i++)
        {
                str += sprintf(str, "%02x ", data[i]);
        }
        GemaltoToken::log("GemaltoPrivateKeyRecord::computeDecrypt -
dataLength <%lu> - data <%s>\n", dataLength, strData);
        GemaltoToken::log("GemaltoPrivateKeyRecord::computeDecrypt -
output <%p>\n", output);
        GemaltoToken::log("GemaltoPrivateKeyRecord::computeDecrypt -
outputLength <%lu>\n", outputLength);
{% endhighlight %}

The attacker might control the data which is to be decrypted, but
exploitation is limited by the sprintf() format string.


Workarounds
- -----------
None

Timeline
========
2018-02-03 Issues found
2018-05-22 Vendor contacted
2018-05-22 Automated vendor reply
2018-05-23 Personal vendor reply
2018-06-05 Requesting technical feedback from the vendor
2018-06-22 Vendor states that the bugs are fixed in public git
2018-07-12 CVE IDs assigned
2018-08-03 https://smartcardservices.github.io/security/ updated
2018-08-11 Advisory released
- -- 
X41 D-SEC GmbH, Dennewartstr. 25-27, D-52068 Aachen
T: +49 241 9809418-0, Fax: -9
Unternehmenssitz: Aachen, Amtsgericht Aachen: HRB19989
Geschäftsführer: Markus Vervier
-----BEGIN PGP SIGNATURE-----
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=9tdn
-----END PGP SIGNATURE-----

_______________________________________________
Sent through the Full Disclosure mailing list
https://nmap.org/mailman/listinfo/fulldisclosure
Web Archives & RSS: http://seclists.org/fulldisclosure/

Powered by blists - more mailing lists