LDT is set; just check the permission without checking the location (which
may change).
In valid_user_selector, don't check the length of the LDT. This is racy
because pm_ldt_len could be updated by another thread, and useless since
the length is already referenced in ldtr (ldt_alloc), which means that any
overflow will fault in userland.
Also, don't check the permission of the segment pointed to; this too is
racy, and we don't care either since the permissions are checked earlier
in x86_set_ldt1.
Pass 1/2.
This update fixes the following vulnerabilities:
CVE-2016-7922 CVE-2016-7923 CVE-2016-7924 CVE-2016-7925 CVE-2016-7926
CVE-2016-7927 CVE-2016-7928 CVE-2016-7929 CVE-2016-7930 CVE-2016-7931
CVE-2016-7932 CVE-2016-7933 CVE-2016-7934 CVE-2016-7935 CVE-2016-7936
CVE-2016-7937 CVE-2016-7938 CVE-2016-7939 CVE-2016-7940 CVE-2016-7973
CVE-2016-7974 CVE-2016-7975 CVE-2016-7983 CVE-2016-7984 CVE-2016-7985
CVE-2016-7986 CVE-2016-7992 CVE-2016-7993 CVE-2016-8574 CVE-2016-8575
CVE-2017-5202 CVE-2017-5203 CVE-2017-5204 CVE-2017-5205 CVE-2017-5341
CVE-2017-5342 CVE-2017-5482 CVE-2017-5483 CVE-2017-5484 CVE-2017-5485
CVE-2017-5486 (all buffer overflows in various parts of tcpdump)
Upstream Changelog:
General updates:
Improve separation frontend/backend (tcpdump/libnetdissect)
Don't require IPv6 library support in order to support IPv6 addresses
Introduce data types to use for integral values in packet structures
Fix display of timestamps with -tt, -ttt and -ttttt options
Fix some heap overflows found with American Fuzzy Lop by Hanno Boeck and others
(More information in the log with CVE-2016-* and CVE-2017-*)
Change the way protocols print link-layer addresses (Fix heap overflows
in CALM-FAST and GeoNetworking printers)
Pass correct caplen value to ether_print() and some other functions
Fix lookup_nsap() to match what isonsap_string() expects
Clean up relative time stamp printing (Fix an array overflow)
Fix some alignment issues with GCC on Solaris 10 SPARC
Add some ND_TTEST_/ND_TCHECK_ macros to simplify writing bounds checks
Add a fn_printztn() which returns the number of bytes processed
Add nd_init() and nd_cleanup() functions. Improve libsmi support
Add CONTRIBUTING file
Add a summary comment in all printers
Compile with more warning options in devel mode if supported (-Wcast-qual, ...)
Fix some leaks found by Valgrind/Memcheck
Fix a bunch of de-constifications
Squelch some Coverity warnings and some compiler warnings
Update Coverity and Travis-CI setup
Update Visual Studio files
Frontend:
Fix capsicum support to work with zerocopy buffers in bpf
Try opening interfaces by name first, then by name-as-index
Work around pcap_create() failures fetching time stamp type lists
Fix a segmentation fault with 'tcpdump -J'
Improve addrtostr6() bounds checking
Add exit_tcpdump() function
Don't drop CAP_SYS_CHROOT before chrooting
Fixes issue where statistics not reported when -G and -W options used
New printers supporting:
Generic Protocol Extension for VXLAN (VXLAN-GPE)
Home Networking Control Protocol (HNCP), RFCs 7787 and 7788
Locator/Identifier Separation Protocol (LISP), type 3 and type 4 packets
Marvell Extended Distributed Switch Architecture header (MEDSA)
Network Service Header (NSH)
REdis Serialization Protocol (RESP)
Updated printers:
802.11: Beginnings of 11ac radiotap support
802.11: Check the Protected bit for management frames
802.11: Do bounds checking on last_presentp before dereferencing it (Fix a heap overflow)
802.11: Fix the radiotap printer to handle the special bits correctly
802.11: If we have the MCS field, it's 11n
802.11: Only print unknown frame type or subtype messages once
802.11: Radiotap dBm values get printed as dB; Update a test output accordingly
802.11: Source and destination addresses were backwards
AH: Add a bounds check
AH: Report to our caller that dissection failed if a bounds check fails
AP1394: Print src > dst, not dst > src
ARP: Don't assume the target hardware address is <= 6 octets long (Fix a heap overflow)
ATALK: Add bounds and length checks (Fix heap overflows)
ATM: Add some bounds checks (Fix a heap overflow)
ATM: Fix an incorrect bounds check
BFD: Update specification from draft to RFC 5880
BFD: Update to print optional authentication field
BGP: Add decoding of ADD-PATH capability
BGP: Add support for the AIGP attribute (RFC7311)
BGP: Print LARGE_COMMUNITY Path Attribute
BGP: Update BGP numbers from IANA; Print minor values for FSM notification
BOOTP: Add a bounds check
Babel: Add decoder for source-specific extension
CDP: Filter out non-printable characters
CFM: Fixes to match the IEEE standard, additional bounds and length checks
CSLIP: Add more bounds checks (Fix a heap overflow)
ClassicalIPoATM: Add a bounds check on LLC+SNAP header (Fix a heap overflow)
DHCP: Fix MUDURL and TZ options
DHCPv6: Process MUDURL and TZ options
DHCPv6: Update Status Codes with RFCs/IANA names
DNS: Represent the "DNSSEC OK" bit as "DO" instead of "OK". Add a test case
DTP: Improve packet integrity checks
EGP: Fix bounds checks
ESP: Don't use OpenSSL_add_all_algorithms() in OpenSSL 1.1.0 or later
ESP: Handle OpenSSL 1.1.x
Ethernet: Add some bounds checking before calling isoclns_print (Fix a heap overflow)
Ethernet: Print the Length/Type field as length when needed
FDDI: Fix -e output for FDDI
FR: Add some packet-length checks and improve Q.933 printing (Fix heap overflows)
GRE: Add some bounds checks (Fix heap overflows)
Geneve: Fix error message with invalid option length; Update list option classes
HNCP: Fix incorrect time interval format. Fix handling of IPv4 prefixes
ICMP6: Fetch a 32-bit big-endian quantity with EXTRACT_32BITS()
ICMP6: dagid is always an IPv6 address, not an opaque 128-bit string
IGMP: Add a length check
IP: Add a bounds check (Fix a heap overflow)
IP: Check before fetching the protocol version (Fix a heap overflow)
IP: Don't try to dissect if IP version != 4 (Fix a heap overflow)
IP: Stop processing IPPROTO_ values once we hit IPPROTO_IPCOMP
IPComp: Check whether we have the CPI before we fetch it (Fix a heap overflow)
IPoFC: Fix -e output (IP-over-Fibre Channel)
IPv6: Don't overwrite the destination IPv6 address for routing headers
IPv6: Fix header printing
IPv6: Stop processing IPPROTO_ values once we hit IPPROTO_IPCOMP
ISAKMP: Clean up parsing of IKEv2 Security Associations
ISOCLNS/IS-IS: Add support for Purge Originator Identifier (RFC6232) and test cases
ISOCLNS/IS-IS: Don't overwrite packet data when checking the signature
ISOCLNS/IS-IS: Filter out non-printable characters
ISOCLNS/IS-IS: Fix segmentation faults
ISOCLNS/IS-IS: Have signature_verify() do the copying and clearing
ISOCLNS: Add some bounds checks
Juniper: Make sure a Juniper header TLV isn't bigger than what's left in the packet (Fix a heap overflow)
LLC/SNAP: With -e, print the LLC header before the SNAP header; without it, cut the SNAP header
LLC: Add a bounds check (Fix a heap overflow)
LLC: Clean up printing of LLC packets
LLC: Fix the printing of RFC 948-style IP packets
LLC: Skip the LLC and SNAP headers with -x for 802.11 and some other protocols
LLDP: Implement IANA OUI and LLDP MUD option
MPLS LSP ping: Update printing for RFC 4379, bug fixes, more bounds checks
MPLS: "length" is now the *remaining* packet length
MPLS: Add bounds and length checks (Fix a heap overflow)
NFS: Add a test that makes unaligned accesses
NFS: Don't assume the ONC RPC header is nicely aligned
NFS: Don't overflow the Opaque_Handle buffer (Fix a segmentation fault)
NFS: Don't run past the end of an NFSv3 file handle
OLSR: Add a test to cover a HNA sgw case
OLSR: Fix 'Advertised networks' count
OLSR: Fix printing of smart-gateway HNAs in IPv4
OSPF: Add a bounds check for the Hello packet options
OSPF: Do more bounds checking
OSPF: Fix a segmentation fault
OSPF: Fix printing 'ospf_topology_values' default
OTV: Add missing bounds checks
PGM: Print the formatted IP address, not the raw binary address, as a string
PIM: Add some bounds checking (Fix a heap overflow)
PIMv2: Fix checksumming of Register messages
PPI: Pass an adjusted struct pcap_pkthdr to the sub-printer
PPP: Add some bounds checks (Fix a heap overflow)
PPP: Report invalid PAP AACK/ANAK packets
Q.933: Add a missing bounds check
RADIUS: Add Value 13 "VLAN" to Tunnel-Type attribute
RADIUS: Filter out non-printable characters
RADIUS: Translate UDP/1700 as RADIUS
RESP: Do better checking of RESP packets
RPKI-RTR: Add a return value check for "fn_printn" call
RPKI-RTR: Remove printing when truncated condition already detected
RPL: Fix 'Consistency Check' control code
RPL: Fix suboption print
RSVP: An INTEGRITY object in a submessage covers only the submessage
RSVP: Fix an infinite loop; Add bounds and length checks
RSVP: Fix some if statements missing brackets
RSVP: Have signature_verify() do the copying and clearing
RTCP: Add some bounds checks
RTP: Add some bounds checks, fix two segmentation faults
SCTP: Do more bounds checking
SFLOW: Fix bounds checking
SLOW: Fix bugs, add checks
SMB: Before fetching the flags2 field, make sure we have it
SMB: Do bounds checks on NBNS resource types and resource data lengths
SNMP: Clean up the "have libsmi but no modules loaded" case
SNMP: Clean up the object abbreviation list and fix the code to match them
SNMP: Do bounds checks when printing character and octet strings
SNMP: Improve ASN.1 bounds checks
SNMP: More bounds and length checks
STP: Add a bunch of bounds checks, and fix some printing (Fix heap overflows)
STP: Filter out non-printable characters
TCP: Add bounds and length checks for packets with TCP option 20
TCP: Correct TCP option Kind value for TCP Auth and add SCPS-TP
TCP: Fix two bounds checks (Fix heap overflows)
TCP: Make sure we have the data offset field before fetching it (Fix a heap overflow)
TCP: Put TCP-AO option decoding right
TFTP: Don't use strchr() to scan packet data (Fix a heap overflow)
Telnet: Add some bounds checks
TokenRing: Fix -e output
UDLD: Fix an infinite loop
UDP: Add a bounds check (Fix a heap overflow)
UDP: Check against the packet length first
UDP: Don't do the DDP-over-UDP heuristic check up front
VAT: Add some bounds checks
VTP: Add a test on Mgmt Domain Name length
VTP: Add bounds checks and filter out non-printable characters
VXLAN: Add a bound check and a test case
ZeroMQ: Fix an infinite loop
Release 2.2.0 Tue June 21 2016
Security fixes:
#537 CVE-2016-0718 -- Fix crash on malformed input
CVE-2016-4472 -- Improve insufficient fix to CVE-2015-1283 /
CVE-2015-2716 introduced with Expat 2.1.1
#499 CVE-2016-5300 -- Use more entropy for hash initialization
than the original fix to CVE-2012-0876
#519 CVE-2012-6702 -- Resolve troublesome internal call to srand
that was introduced with Expat 2.1.0
when addressing CVE-2012-0876 (issue #496)
Bug fixes:
Fix uninitialized reads of size 1
(e.g. in little2_updatePosition)
Fix detection of UTF-8 character boundaries
Other changes:
#532 Fix compilation for Visual Studio 2010 (keyword "C99")
Autotools: Resolve use of "$<" to better support bmake
Autotools: Add QA script "qa.sh" (and make target "qa")
Autotools: Respect CXXFLAGS if given
Autotools: Fix "make run-xmltest"
Autotools: Have "make run-xmltest" check for expected output
p90 CMake: Fix static build (BUILD_shared=OFF) on Windows
#536 CMake: Add soversion, support -DNO_SONAME=yes to bypass
#323 CMake: Add suffix "d" to differentiate debug from release
CMake: Define WIN32 with CMake on Windows
Annotate memory allocators for GCC
Address all currently known compile warnings
Make sure that API symbols remain visible despite
-fvisibility=hidden
Remove executable flag from source files
Resolve COMPILED_FROM_DSP in favor of WIN32
Remove some unnecessary cuteness that limited error reporting.
Permit just one -s arg to fdflags
Be deterministic in the case of fdflags -s +cloexec,-cloexec 0
(and similar) - use the last specified, always.
Allow:
FD_0_FLAGS=$( fdflags -v 0 )
# do stuff, manipulating the flags
fdflags -s "FD_0_FLAGS" 0
to save/restore flags for a fd.
Correctly mask result of fcntl(fd, F_GETFD) with FD_CLOEXEC as the
specs require before deciding close on exec is set.
Improve portability as a tool, don't assume strtoi(), nor __arraycount()
and avoid needlessly requiring recent C versions (ie: there's no need to
sprinkle declarations in the middle of the code, it just makes them hard
to find, and benefits nothing.)
Still to do: As currently implemented, both user, and shell internal fds
are reported, and can be manipulated. Allowing users to touch the shell's
internal fds is bogus, and providing this easy way to allow users to
discover which values they have is poor. Fixing this means getting rid
of the use of fcntl(F_MAXFD) and replacing it with a shell maintained
memory of what fds the user (script) has allocated. The shell's fd
manipulation really still needs major work (including properly fixing
bin/48875)
joerg