tpm-dev-common: Reject too short writes
authorAlexander Steffen <Alexander.Steffen@infineon.com>
Fri, 8 Sep 2017 15:21:32 +0000 (17:21 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 24 Nov 2017 07:37:04 +0000 (08:37 +0100)
commit ee70bc1e7b63ac8023c9ff9475d8741e397316e7 upstream.

tpm_transmit() does not offer an explicit interface to indicate the number
of valid bytes in the communication buffer. Instead, it relies on the
commandSize field in the TPM header that is encoded within the buffer.
Therefore, ensure that a) enough data has been written to the buffer, so
that the commandSize field is present and b) the commandSize field does not
announce more data than has been written to the buffer.

This should have been fixed with CVE-2011-1161 long ago, but apparently
a correct version of that patch never made it into the kernel.

Signed-off-by: Alexander Steffen <Alexander.Steffen@infineon.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Tested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/char/tpm/tpm-dev-common.c

index 610638a80383d4a104a6fe1ec42479722f222f49..461bf0b8a09473dbadc89b9259473dc36be2cd77 100644 (file)
@@ -110,6 +110,12 @@ ssize_t tpm_common_write(struct file *file, const char __user *buf,
                return -EFAULT;
        }
 
+       if (in_size < 6 ||
+           in_size < be32_to_cpu(*((__be32 *) (priv->data_buffer + 2)))) {
+               mutex_unlock(&priv->buffer_mutex);
+               return -EINVAL;
+       }
+
        /* atomic tpm command send and result receive. We only hold the ops
         * lock during this period so that the tpm can be unregistered even if
         * the char dev is held open.