protocols, in addition to adding structures to the list of filterable
fields. Give it an extra argument that specifies a "short name" for the
protocol, for use in such places as
pinfo->current_proto;
the dialog box for constructing filters;
the preferences tab for the protocol;
and so on (although we're not yet using it in all those places).
Make the preference name that appears in the preferences file and the
command line for the DIAMETER protocol "diameter", not "Diameter"; the
convention is that the name in question be all-lower-case.
Make some routines and variables that aren't exported static.
Update a comment in the ICP dissector to make it clear that the
dissector won't see fragments other than the first fragment of a
fragmented datagram.
svn path=/trunk/; revision=2810
Add a new subdissector table in the LLC dissector for protocol IDs with
a Cisco OUI, and register the CDP, CGMP, and VTMP dissectors in that
table, rather than calling them via a switch statement.
Register the ISL dissector by name, and have the Ethernet dissector call
it via a handle.
Fix the handling of the checksum field in the CDP dissector.
The strings in CDP are counted, not null-terminated; treat them as such.
Fix the handling of the encapsulated frame CRC, and the encapsulated
frame, in the ISL dissector, at least for Ethernet frames; it may not be
correct for encapsulated Token Ring frames.
svn path=/trunk/; revision=2792
dissect_fddi_not_bitswapped() and dissect_fddi_bitswapped(), both of which
use the standard 3-argument tvbuffified-dissector argument list.
Add a dissector table called "wtap_encap" which is used to call dissectors
from dissect_frame(). The switch() statement from this top-level dissector
is removed.
The link-layer dissectors register themselves with the "wtap_encap"
dissector table. The dissectors are now static where possible.
svn path=/trunk/; revision=2708
can be put, and a pointer to the string for the column, which might or
might not point to that buffer.
Add a routine "col_set_str()", which sets the string for the column to
the string passed to it as an argument; it should only be handed a
static string (a string constant would be ideal). It doesn't do any
copying, so it's faster than "col_add_str()".
Make the routines that append to columns check whether the pointer to
the string for the column points to the buffer for the column and, if
not, copy the string for the column to the buffer for the column so that
you can append to it (so you can use "col_set_str()" and then use
"col_append_str()" or "col_append_fstr()").
Convert a bunch of "col_add_str()" calls that take a string constant as
an argument to "col_set_str()" calls.
Convert some "col_add_fstr()" calls that take a string constant as the
only argument - i.e., the format string doesn't have any "%" slots into
which to put strings for subsequent arguments to "col_set_str()" calls
(those calls are just like "col_add_str()" calls).
Replace an END_OF_FRAME reference in a tvbuffified dissector with a
"tvb_length(tvb)" call.
svn path=/trunk/; revision=2670
other dissectors call them through handles. Do the same for the "PPP
payload" dissector, after tvbuffifying it.
Tvbuffify the PPPoE dissector.
Do the last little bit of tvbuffifying the L2TP dissector (it takes
old-style arguments and immediately generates a tvbuff out of them; make
it take new-style arguments).
svn path=/trunk/; revision=2664
ESIS dissectors.
Register the IP dissector and have dissectors that call it directly
(rather than through a port table) call it through a handle.
Add a routine "tvb_set_reported_length()" which a dissector can use if
it was handed a tvbuff that contains more data than is actually in its
part of the packet - for example, handing a padded Ethernet frame to IP;
the routine sets the reported length of the tvbuff (and also adjusts the
actual length, as appropriate). Then use it in IP.
Given that, "ethertype()" can determine how much of the Ethernet frame
was actually part of an IP datagram (and can do the same for other
protocols under Ethernet that use "tvb_set_reported_length()"; have it
return the actual length, and have "dissect_eth()" and "dissect_vlan()"
use that to mark trailer data in Ethernet II frames as well as in 802.3
frames.
svn path=/trunk/; revision=2658
we don't have so much data in the frame that there's a trailer, so we
should set "trailer_tvb" to NULL.
Put in a comment explaining what the exception catching is all about.
svn path=/trunk/; revision=2633
'tvbuff_t *volatile'." Makes "Throat-Warbler Mangrove" vs.
"Luxury-Yacht" sound almost normal....
Type-qualified pointers to non-type-qualified objects are a barrel of
fun in C. The way you declare a volatile pointer named "bar" to a
*non-volatile* "foo" is
foo *volatile bar;
as opposed to a non-volatile pointer "bar" to a volatile "foo", which is
volatile foo *bar;
GCC's complaint about variables being clobbered by longjmp refers to the
fact that "longjmp()" isn't guaranteed to restore variables stored in
registers to the values they had at the time of the "longjmp()" (if
"setjmp()" stuffs the current register values in the "jmp_buf", and
"longjmp()" just reloads them rather than walking the stack to restore
all register values pushed onto the stack, the values at the time of the
"setjmp()" will be restored, clobbering any updates done after the
"setjmp()"); the workaround provided in ANSI C is to declare the
variables in question "volatile", which will keep them out of registers
(or any other place that "setjmp()"/"longjmp()" can't handle).
svn path=/trunk/; revision=2631
the following:
It is now possible to enable/disable a particular protocol decoding
(i.e. the protocol dissector is void or not). When a protocol
is disabled, it is displayed as Data and of course, all linked
sub-protocols are disabled as well.
Disabling a protocol could be interesting:
- in case of buggy dissectors
- in case of wrong heuristics
- for performance reasons
- to decode the data as another protocol (TODO)
Currently (if I am not wrong), all dissectors but NFS can be disabled
(and dissectors that do not register protocols :-)
I do not like the way the RPC sub-dissectors are disabled (in the
sub-dissectors) since this could be done in the RPC dissector itself,
knowing the sub-protocol hfinfo entry (this is why, I've not modified
the NFS one yet).
Two functions are added in proto.c :
gboolean proto_is_protocol_enabled(int n);
void proto_set_decoding(int n, gboolean enabled);
and two MACROs which can be used in dissectors:
OLD_CHECK_DISPLAY_AS_DATA(index, pd, offset, fd, tree)
CHECK_DISPLAY_AS_DATA(index, tvb, pinfo, tree)
See also the XXX in proto_dlg.c and proto.c around the new functions.
svn path=/trunk/; revision=2267
dissectors to be registered as dissectors for particular ports,
registered as heuristic dissectors, and registered as dissectors for
conversations, and have routines to be used both by old-style and
new-style dissectors to call registered dissectors.
Have the code that calls those dissectors translate the arguments as
necessary. (For conversation dissectors, replace
"find_conversation_dissector()", which just returns a pointer to the
dissector, with "old_try_conversation_dissector()" and
"try_conversation_dissector()", which actually call the dissector, so
that there's a single place at which we can do that translation. Also
make "dissector_lookup()" static and, instead of calling it and, if it
returns a non-null pointer, calling that dissector, just use
"old_dissector_try_port()" or "dissector_try_port()", for the same
reason.)
This allows some dissectors that took old-style arguments and
immediately translated them to new-style arguments to just take
new-style arguments; make them do so. It also allows some new-style
dissectors not to have to translate arguments before calling routines to
look up and call dissectors; make them not do so.
Get rid of checks for too-short frames in new-style dissectors - the
tvbuff code does those checks for you.
Give the routines to register old-style dissectors, and to call
dissectors from old-style dissectors, names beginning with "old_", with
the routines for new-style dissectors not having the "old_". Update the
dissectors that use those routines appropriately.
Rename "dissect_data()" to "old_dissect_data()", and
"dissect_data_tvb()" to "dissect_data()".
svn path=/trunk/; revision=2218
a particular type, rather than taking a varargs list, along the lines of
the "proto_tree_add_XXX_format()" routines.
Replace most calls to "proto_tree_add_item()" and
"proto_tree_add_item_hidden()" with calls to those routines.
Rename "proto_tree_add_item()" and "proto_tree_add_item_hidden()" to
"proto_tree_add_item_old()" and "proto_tree_add_item_hidden_old()", and
add new "proto_tree_add_item()" and "proto_tree_add_item_hidden()"
routines that don't take the item to be added as an argument - instead,
they fetch the argument from the packet whose tvbuff was handed to them,
from the offset handed to them.
svn path=/trunk/; revision=2031
"packet-eth.c", and "packet-fddi.c" include the include files that
declare the functions they export, so that the declarationss in the
header files will be checked against the definitions in the source
files.
svn path=/trunk/; revision=1981
in tvbuff terminology). This is implemented for TVBUFF_REAL and TVBUFF_SUBSET
so far; support for TVBUFF_COMPOSITE is coming soon.
Throw either ReportedBoundsError or BoundsError.
A ReportedBoundsError is reported as "Malformed Frame" since the protocol
stated that a certain number of bytes should be available but they weren't.
A BoundsError is reported as a "Short Frame" since the snaplen was too short.
Register proto_short (BoundsError) and proto_malformed (ReportedBounds)
so searches can be made on "short" and "malformed".
svn path=/trunk/; revision=1965
Modify ethernet dissector to catch BoundsError if the attempt to
create next_tvb with the length specified in the ethernet header throws
an exception. In that case, next_tv is created with as many bytes as
are available in the frame.
Both dissect_tr() and dissect_eth() now have TRY blocks, which means
I had to fiddle with 'volatile' and 'static' storage options to get
things right (at least according to gcc).
svn path=/trunk/; revision=1962
Non-tvbuff dissectors create a tvbuff when calling dissect_llc()
Changed name of current_proto to match string in COL_PROTO ("FDDI" instead of "fddi")
Changed short text to be: [Short Frame: %s] where %s is current_proto.
svn path=/trunk/; revision=1943
Add exceptions routines.
Convert proto_tree_add_*() routines to require tvbuff_t* argument.
Convert all dissectors to pass NULL argument ("NullTVB" macro == NULL) as
the tvbuff_t* argument to proto_tree_add_*() routines.
dissect_packet() creates a tvbuff_t, wraps the next dissect call in
a TRY block, will print "Short Frame" on the proto_tree if a BoundsError
exception is caught.
The FDDI dissector is converted to use tvbuff's.
svn path=/trunk/; revision=1939
register lists of "heuristic" dissectors, which are handed a
frame that may or may contain a payload for the protocol they
dissect, and that return FALSE if it's not or dissect the packet
and return TRUE if it is;
add a dissector to such a list;
go through such a list, calling each dissector until either a
dissector returns TRUE, in which case the routine returns TRUE,
or it runs out of entries in the list, in which case the routine
returns FALSE.
Have lists of heuristic dissectors for TCP and for COTP when used with
the Inactive Subset of CLNP, and add the GIOP and Yahoo Messenger
dissectors to the first list and the Sinec H1 dissector to the second
list.
Make the dissector name argument to "dissector_add()" and
"dissector_delete()" a "const char *" rarther than just a "char *".
Add "heur_dissector_add()", the routine to add a heuristic dissector to
a list of heuristic dissectors, to the set of routines we can export to
plugins through a table on platforms where dynamically-loaded code can't
call stuff in the main program, and initialize the element in the table
in question for "dissector_add()" (which we'd forgotten to do).
svn path=/trunk/; revision=1909
proto_tree_add_protocol_format()
proto_tree_add_uint_format()
proto_tree_add_ipxnet_format()
proto_tree_add_ipv4_format()
proto_tree_add_ipv6_format()
proto_tree_add_bytes_format()
proto_tree_add_string_format()
proto_tree_add_ether_format()
proto_tree_add_time_format()
proto_tree_add_double_format()
proto_tree_add_boolean_format()
If using GCC 2.x, we can check the print-format against the variable args
passed in. Regardless of compiler, we can now check at run-time that the
field type passed into the function corresponds to what that function
expects (FT_UINT, FT_BOOLEAN, etc.)
Note that proto_tree_add_protocol_format() does not require a value field,
since the value of a protocol is always NULL. It's more intuitive w/o the
vestigial argument.
Fixed a proto_tree_add_item_format-related bug in packet-isis-hello.c
Fixed a variable usage bug in packet-v120.c. (ett_* was used instead of hf_*)
Checked in Guy's fix for the function declearation for proto_tree_add_text()
and proto_tree_add_notext().
svn path=/trunk/; revision=1713
This change allows you to add a new packet-*.c file and not cause a
recompilation of everything that #include's packet.h
Add the plugin_api.[ch] files ot the plugins/Makefile.am packaging list.
Add #define YY_NO_UNPUT 1 to the lex source so that the yyunput symbol
is not defined, squelching a compiler complaint when compiling the generated
C file.
svn path=/trunk/; revision=1637
value in the length field not only the Ethernet MAC header size, but
also the offset in the frame of the Ethernet MAC header, so that, if the
802.3 frame is encapsulated in some other type of frame, the total frame
length includes the header for that frame as well.
svn path=/trunk/; revision=1526
whether we're building a protocol tree or not.
Make "dissect_eth()" use "BYTES_ARE_IN_FRAME()" to see if we have a full
Ethernet header - it can be called with a non-zero offset, if Ethernet
frames are encapsulated inside other frames (e.g., ATM LANE).
Make capture routines take an "offset" argument if the corresponding
dissect routine takes one (for symmetry, and for Cisco ISL or any other
protocol that encapsulates Ethernet or Token-Ring frames inside other
frames).
Pass the frame lengths to capture routines via the "pi" structure,
rather than as an in-line argument, so that they can macros such as
"BYTES_ARE_IN_FRAME()" the way the corresponding dissect routines do.
Make capture routines update "pi.len" and "pi.captured_len" the same way
the corresponding diseect routines do, if the capture routines then call
other capture routines.
Make "capture_vlan()" count as "other" frames that are too short, the
way other capture routines do.
svn path=/trunk/; revision=1525
eth.length were being pulled put into the proto_tree (logical and GUI),
but the fields were highlighted in the hex dump w/o adding offset.
svn path=/trunk/; revision=1062
dynamically-assigned "ett_" integer values, assigned by
"proto_register_subtree_array()"; this:
obviates the need to update "packet.h" whenever you add a new
subtree type - you only have to add a call to
"proto_register_subtree_array()" to a "register" routine and an
array of pointers to "ett_", if they're not already there, and
add a pointer to the new "ett_" variable to the array, if they
are there;
would allow run-time-loaded dissectors to allocate subtree types
when they're loaded.
svn path=/trunk/; revision=1043
structure to "dl_src"/"dl_dst", "net_src"/"net_dst", and "src"/"dst"
addresses, where an address is an address type, an address length in
bytes, and a pointer to that many bytes.
"dl_{src,dst}" are the link-layer source/destination; "net_{src,dst}"
are the network-layer source/destination; "{src,dst}" are the
source/destination from the highest of those two layers that we have in
the packet.
Add a port type to "packet_info" as well, specifying whether it's a TCP
or UDP port.
Don't set the address and port columns in the dissector functions; just
set the address and port members of the "packet_info" structure. Set
the columns in "fill_in_columns()"; this means that if we're showing
COL_{DEF,RES,UNRES}_SRC" or "COL_{DEF,RES,UNRES}_DST", we only generate
the string from "src" or "dst", we don't generate a string for the
link-layer address and then overwrite it with a string for the
network-layer address (generating those strings costs CPU).
Add support for "conversations", where a "conversation" is (at present)
a source and destination address and a source and destination port. (In
the future, we may support "conversations" above the transport layer,
e.g. a TFTP conversation, where the first packet goes from the client to
the TFTP server port, but the reply comes back from a different port,
and all subsequent packets go between the client address/port and the
server address/new port, or an NFS conversation, which might include
lock manager, status monitor, and mount packets, as well as NFS
packets.)
Currently, all we support is a call that takes the source and
destination address/port pairs, looks them up in a hash table, and:
if nothing is found, creates a new entry in the hash table, and
assigns it a unique 32-bit conversation ID, and returns that
conversation ID;
if an entry is found, returns its conversation ID.
Use that in the SMB and AFS code to keep track of individual SMB or AFS
conversations. We need to match up requests and replies, as, for
certain replies, the operation code for the request to which it's a
reply doesn't show up in the reply - you have to find the request with a
matching transaction ID. Transaction IDs are per-conversation, so the
hash table for requests should include a conversation ID and transaction
ID as the key.
This allows SMB and AFS decoders to handle IPv4 or IPv6 addresses
transparently (and should allow the SMB decoder to handle NetBIOS atop
other protocols as well, if the source and destination address and port
values in the "packet_info" structure are set appropriately).
In the "Follow TCP Connection" code, check to make sure that the
addresses are IPv4 addressses; ultimately, that code should be changed
to use the conversation code instead, which will let it handle IPv6
transparently.
svn path=/trunk/; revision=909
the base for numbers to be displayed in, bitmasks for bitfields, and blurbs
(which are one or two sentences describing the field).
proto_tree_add*() routines now automatically handle bitfields. You tell
it which header field you are adding, and just pass it the value of the
entire field, and the proto_tree routines will do the masking and shifting
for you.
This means that bitfields are more naturally filtered via dfilter now.
Added Phil Techau's support for signed integers in dfilters/proto_tree.
Added the beginning of the SNA dissector. It's not complete, but I'm
committing it now because it has example after example of how to use
bitfields with the new header_field_info struct and proto_tree routines.
It was the impetus to change how header_field_info works.
svn path=/trunk/; revision=815
and the info field an indication of whether it's Ethernet II, raw 802.3,
or (LLC-atop) 802.3 (which will be overridden by other protocols, if we
know the protocol inside the frame).
svn path=/trunk/; revision=559
and in different capture files; throw in some heuristics to try to
figure out whether the 4-byte header is:
1) PPP-over-HDLC (some version of ISDN4BSD?);
2) big-endian AF_ value (BSD on big-endian platforms);
3) little-endian AF_ value (BSD on little-endian platforms);
4) two octets of 0 followed by an Ethernet type (Linux, at least
on little-endian platforms, as mutated by "libpcap").
Make a separate Wiretap encapsulation type, WTAP_ENCAP_NULL,
corresponding to DLT_NULL.
Have the PPP code dissect the frame if it's PPP-over-HDLC, and have
"ethertype()" dissect the Ethernet type and the rest of the packet if
it's a Linux-style header; dissect it ourselves only if it's an AF_
value.
Have Wiretap impose a maximum packet size of 65535 bytes, so that it
fails more gracefully when handed a corrupt "libpcap" capture file
(other capture file formats with more than a 16-bit capture length
field, if any, will have that check added later), and put that size in
"wtap.h" and have Ethereal use it as its notion of a maximum packet
size.
Have Ethereal put up a "this file appears to be damaged or corrupt"
message box if Wiretap returns a WTAP_ERR_BAD_RECORD error when opening
or reading a capture file.
Include loopback interfaces in the list of interfaces offered by the
"Capture" dialog box, but put them at the end of the list so that it
doesn't default to a loopback interface unless there are no other
interfaces. Also, don't require that an interface in the list have an
IP address associated with it, and only put one entry in the list for a
given interface (SIOCGIFCONF returns one entry per interface *address*,
not per *interface* - and even if you were to use only IP addresses, an
interface could conceivably have more than one IP address).
Exclusively use Wiretap encapsulation types internally, even when
capturing; don't use DLT_ types.
svn path=/trunk/; revision=540
file, instead of throwing out all but LANE or RFC 1483 data frames and
pretending that the former are just Ethernet or Token-Ring frames.
Add some level of decoding for ATM LANE, but not all of it; the rest,
including decoding non-LANE frames, is left as an exercise for somebody
who has captures they want to decode, an interest in decoding them, ATM
expertise, and time....
svn path=/trunk/; revision=523
bunch of source files.
Replace the "payload" field of a "packet_info" structure with "len" and
"captured_len" fields, which contain the total packet length and total
captured packet length (including all headers) at the current protocol
layer (i.e., if a given layer has a length field, and that length field
says its shorter than the length we got from the capture, reduce the
"pi.len" and "pi.captured_len" values appropriately). Those fields can
be used in the future if we add checks to make sure a field we're
extracting from a packet doesn't go past the end of the packet, or past
the captured part of the packet.
Get rid of the additional payload argument to some dissection functions;
use "pi.captured_len - offset" instead.
Have the END_OF_FRAME macro use "pi.captured_len" rather than
"fd->cap_len", so that "dissect the rest of the frame" becomes "dissect
the rest of the packet", and doesn't dissect end-of-frame padding such
as padding added to make an Ethernet frame 60 or more octets long. (We
might want to rename it END_OF_PACKET; if we ever want to label the
end-of-frame padding for the benefit of people curious what that extra
gunk is, we could have a separate END_OF_FRAME macro that uses
"fd->cap_len".)
svn path=/trunk/; revision=506
as it standed depends on your lex being flex, but that only matters if you're
a developer. The distribution will include the dfilter-scanner.c file, so
that if the user doesn't modify dfilter-scanner.l, he won't need flex to
re-create the *.c file.
The new lex scanner gives me better syntax checking for ether addresses. I
thought I could get by using GScanner, but it simply wasn't powerful enough.
All operands have English-like abbreviations and C-like syntax:
and, && ; or, || ; eq, == ; ne, != ; , etc.
I removed the ETHER_VENDOR type in favor of letting the user use the [x:y]
notation: ether.src[0:3] == 0:6:29 instead of ether.srcvendor == 00:06:29
I implemented the IPXNET field type; it had been there before, but was
not implemented. I chose to make it use integer values rather than byte
ranges, since an IPX Network is 4 bytes. So a display filter looks like this:
ipx.srcnet == 0xc0a82c00
rather than this:
ipx.srcnet == c0:a8:2c:00
I can supposrt the byte-range type IPXNET in the future, very trivially.
I still have more work to do on the parser though. It needs to check ranges
when extracting byte ranges ([x:y]) from packets. And I need to get rid
of those reduce/reduce errors from yacc!
svn path=/trunk/; revision=414
suggestion, this new method using a static array should use less memory
and be faster. It also has a nice side-effect of making the source-code
more readble, IMHO.
Changed the print routines to look for protocol proto_data instead of
looking at the text label as they did before, hoping that the data hex
dump field item starts with "Data (".
Added the -G keyword to ethereal to make it dump a glossary of display
filter keywords to stdout and exit. This data is then formatted with
the doc/dfilter2pod perl program to pod format, which is combined
with doc/ethereal.pod.template to create doc/ethereal.pod, from which
the ethereal manpage is created. This way we can keep the manpage up-to-date
with a list of fields that can be filtered on.
svn path=/trunk/; revision=364
mechanism that is built into ethereal. Wiretap is now used to read all
file formats. Libpcap is used only for capturing.
svn path=/trunk/; revision=342
reference the protocol tree with struct proto_tree and struct proto_item
objects. That way, the packet decoding source code file can be used with
non-gtk packet decoders, like a curses-based ethereal, e.g. I also re-arranged
some of the information in packet.h to more appropriate places (like other
packet-*.[ch] files).
svn path=/trunk/; revision=223
because it is still in its infancy, but it can be compiled in optionally.
The library exists in its own subdirectory ethereal/wiretap. This patch also
edits all the packet-*.c files to remove the #include <pcap.h> line which is
unnecessary in these files. In the ethereal code, file.c is the most heavily
modified with #ifdef WITH_WIRETAP lines for the optional library.
svn path=/trunk/; revision=82
generalizes the column printing code, adds a "frame" tree item to
the tree view, and fixes a bunch of miscellaneous coding bugs.
svn path=/trunk/; revision=31
Guy Harris