linux-kernel - [PATCH 0/1] Fix crash in __calc_tpm2_event

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20190604230433.20936-1-chris.coulson@canonical.com>
Date:   Wed,  5 Jun 2019 00:04:32 +0100
From:   Chris Coulson <chris.coulson@...onical.com>
To:     linux-integrity@...r.kernel.org
Cc:     Chris Coulson <chris.coulson@...onical.com>,
        linux-efi@...r.kernel.org, Peter Huewe <peterhuewe@....de>,
        Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>,
        Jason Gunthorpe <jgg@...pe.ca>,
        Matthew Garrett <mjg59@...gle.com>,
        Bartosz Szczepanek <bsz@...ihalf.com>,
        Roberto Sassu <roberto.sassu@...wei.com>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>,
        linux-kernel@...r.kernel.org
Subject: [PATCH 0/1] Fix crash in __calc_tpm2_event_size

I've been testing the latest code in the linux-tpmdd branch and I'm
experiencing a crash in __calc_tpm2_event_size when it's called to
calculate the size of events in the final log. I hope I'm not stepping on
anyone's toes, but this small change seems to fix it.

What seems to happen is that the event header is mapped here:

    /* Map the event header */
    if (do_mapping) {
        mapping_size = marker - marker_start;
        mapping = TPM_MEMREMAP((unsigned long)marker_start,
                       mapping_size);

    ...

    event = (struct tcg_pcr_event2_head *)mapping;

When calculating the cumulative size of the digests, the event header is
dereferenced here on each loop iteration in order to obtain the digest
count:

    for (i = 0; i < event->count; i++) {
        halg_size = sizeof(event->digests[i].alg_id);

But the first iteration of the loop unmaps the event header:

        /* Map the digest's algorithm identifier */
        if (do_mapping) {
            TPM_MEMUNMAP(mapping, mapping_size);
            mapping_size = halg_size;
            mapping = TPM_MEMREMAP((unsigned long)marker,
                         mapping_size);

Subsequent loop iterations then dereference a pointer to unmapped memory.

Chris Coulson (1):
  tpm: Don't dereference event after it's unmapped in
    __calc_tpm2_event_size

 include/linux/tpm_eventlog.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

-- 
2.17.1