dect
/
libpcap
Archived
13
0
Fork 0
Commit Graph

679 Commits

Author SHA1 Message Date
hannes 405ab1a931 add baseline support for MPLS protocol and per-label filtering 2004-06-16 08:20:28 +00:00
guy 93306079a3 Add a DLT_ value and LINKTYPE_ value for BACnet MS/TP.
Clean up some white space.
2004-06-07 20:00:08 +00:00
risso 8106848e48 Added a check against a NULL pointer in pcap_inject_win32.
Added an explicit cast in pcap_inject_win32 to avoid a compilation warning
under Visual C.
2004-06-07 13:27:46 +00:00
guy b7875f7aec As noted by Rob Deker, we weren't checking the "i"th member of the DLT
list against DLT_EN10MB, we were comparing the list pointer against it.
2004-06-05 00:42:13 +00:00
guy 1610cc5db7 Fix the comment for DLT_NULL to reflect reality, as per FreeBSD PR 62272
(the change for which has been checked into FreeBSD's <net/bpf.h>).
2004-06-04 22:19:47 +00:00
guy e80b63b73e "sbh_drops" is cumulative, so we can't add it to the packet receive
count; instead, we add the cumulative drop count to the cumulative
receive count in the statistics we're returning.
2004-05-21 09:45:31 +00:00
guy dc95138209 Add the dropped packet count to the received packet count, so that the
counts work similarly to the way they work in BPF and 2.4 and later
Linux (and so that we don't run the risk of the dropped packet count
being bigger than the received packet count, which can result in dropped
packet percentages > 100).
2004-05-21 09:22:16 +00:00
risso e00313a3db Fixed a memory leak in pcap_lookupdev under Win32. 2004-04-30 09:13:51 +00:00
risso 3787448b64 Fixed a bug in pcap_findalldevs on Win32 that not initialized *alldevsp to NULL
if no adapters are present.
2004-04-30 09:12:42 +00:00
risso 5f53d2308e Modified pcap_findalldevs on Win32, so that it is able to properly handle an
indefinite number of adapters.
Before the buffer passed to PacketGetAdapterNames was limited to 8kB.
2004-04-30 08:58:52 +00:00
guy 076135f7df On non-HP-UX systems, if we fail to open "/dev/{if}" or "/dev/{if}N"
because neither of them exist, just report that there was no DLPI device
found for "{if}N", so people don't think that they need to fix libpcap
(or the program using it) to look somewhere else for the device - the
problem is probably that they're trying to capture on a loopback device
and the lack of any DLPI device is just a symptom of the fact that the
loopback device, at least on Solaris, appears not to support DLPI and
thus isn't supported by libpcap....
2004-04-23 05:19:04 +00:00
guy c4e76ddd83 Note that not all the addresses in the list of addresses are
necessarily IPv4 or IPv6 addresses, and that you must check the address
family before interpreting the address.
2004-04-20 20:42:36 +00:00
guy 60268b4b61 Add a note indicating that there are platforms where the source address
in the link-layer header is, on packets sent with "pcap_inject()",
changed to the address of the interface on which the packet is sent.
2004-04-09 21:16:44 +00:00
guy 1cd8c13e62 From Brian Ginsbach: squelch some compiler warnings. 2004-04-07 18:43:29 +00:00
guy 3a35ed338c From Brian Ginsbach: "dlt_list" is assigned a "u_int *" value, so make
it a "u_int *".
2004-04-07 18:41:00 +00:00
guy ffe68c0ab5 From Brian Ginsbach:
handle 65535-byte packets from the UNICOS/mp loopback interface;

	add a missing variable to "pcap_inject_snoop()".
2004-04-07 18:33:29 +00:00
guy 841b1f7f77 As per Mark Pizzolato, "sendto()" isn't necessary for sending packets,
at least on 2.2 and later kernels; the socket is bound (except for
sending on the "any" device, which we don't support), so a destination
address isn't necessary.

Generate the right error string for attempts to send on the "any"
device, and also disallow sends if we're in cooked mode.
2004-04-07 08:03:32 +00:00
guy 5b0754b7de Having gone through the trouble of creating "dl_doattach()", we might as
well use it....
2004-04-07 07:56:05 +00:00
risso b2656a7aa7 Fixed a conditional typedef that caused wpcap.dll not to compile with VC. 2004-04-06 13:30:22 +00:00
guy b7be31258d Move "snprintf.c" to the "missing" directory, as that's where
"Makefile.in" expects it to be (tcpdump puts "snprintf.c" and the like
into a "missing" subdirectory).
2004-04-05 22:43:50 +00:00
mcr febe5a1a4e documentation for proposed pcap format 2004-04-05 20:26:58 +00:00
guy 6c9206b958 Small whitespace style cleanup. 2004-04-03 20:53:48 +00:00
guy cec387867a Add support for sending packets on HP-UX if DL_HP_RAWDLS (for "raw
mode") is defined.
2004-04-03 20:52:41 +00:00
guy 5b35985840 If we get EAGAIN, return 0, as we do with EAGAIN and EWOULDBLOCK on
other platforms, and as the man page says we do (in non-blocking mode,
we return 0 if no packets are available).
2004-04-03 02:04:13 +00:00
guy 14645789a6 Not all systems have "sa_family_t"; use "u_int8_t" instead. 2004-04-02 06:33:30 +00:00
guy f20514f49f From Rick Jones: the DL_HP_PPA_ACK_OBS #defines are gone in HP-UX 11.23;
check for DL_HP_PPA_REQ instead.
2004-04-02 06:18:22 +00:00
guy 229ab726f9 Update a comment to suggest how to detect the OS X broken BIOCSHDRCMPLT
at run time, so we automatically adapt to systems with Jeff Nathan's fix
or to future OS X releases if they fix the bug.
2004-03-31 01:03:00 +00:00
mcr eb9ec105d1 update changes file 2004-03-30 14:42:50 +00:00
guy 0a56a3721a Get rid of long-vs-int problem in a printf format (sometimes sizeof is
an int, sometime's it's a long).
2004-03-29 21:04:31 +00:00
mcr f0f5fc44e4 changes for 0.8.2 2004-03-29 17:40:11 +00:00
fenner 3058de1dde Regenerate. 2004-03-28 20:31:56 +00:00
fenner 9cb15b4863 Pass correct prereqs for <netinet/if_ether.h> 2004-03-28 20:31:07 +00:00
fenner 8177eec7d9 Handle the new OpenBSD pf format (DLT 117), which is now being used
by other systems as they adopt pf.
Don't bother trying to be backwards compatible with DLT 17.
2004-03-28 20:27:12 +00:00
guy 55a1578cbe Get rid of an out-of-date comment.
Add a comment noting why we don't do BIOCSHDRCMPLT on OS X.
2004-03-24 19:52:46 +00:00
mcr 50c4867b89 work-in-progress pcap format 1.0. 2004-03-24 19:51:11 +00:00
mcr 78caf331be patch from Gisle Vanem for Symantec/Raptor tcpdump. 2004-03-24 19:50:54 +00:00
guy 0efb0602ff From Mark Pizzolato: do a BIOCSHDRCMPLT so that the link-layer header we
supply is sent without the source address being overwritten.
2004-03-24 07:00:41 +00:00
guy 5f9f6317d2 Fixes from Mark Pizzolato. 2004-03-24 06:49:16 +00:00
guy 2d2890dd37 Add support for sending packets; includes contributions from Mark
Pizzolato <List-tcpdump-workers@subscriptions.pizzolato.net>.
2004-03-23 19:18:04 +00:00
guy 3380fa10db Fix cut-and-pasteos; thanks to Darren Reed for finding them. 2004-03-21 08:32:05 +00:00
guy 1bd06dd0d5 Add support for Apple's IP-over-IEEE 1394 encapsulation.
Fix a comment.
2004-03-17 19:03:28 +00:00
risso fe6b7f8e33 Avoid a null pointer access in pcap_dump_open if fname is a read only file 2004-03-16 19:27:55 +00:00
guy ce798150ac Beef up some comments, and clean up some other comments. 2004-03-11 23:47:29 +00:00
guy 2f74e436ab Put the LINKTYPE_ values in numerical order, to make it easier to make
sure that all values from 100 to the current maximum value are in use;
put in comments to indicate for what purpose the reserved ones are
reserved.

Note in the "don't just grab a value and use it yourself" comment why
you shouldn't use values below 100 or above the maximum value you see in
the file.
2004-03-11 23:40:54 +00:00
guy bd53902d4a Don't rely on SA_LEN to check whether it's passed a null pointer (it
might be supplied by the OS, and the OS might supply a version that
doesn't check); instead, check for null pointers before doing anything
with the address at all.  This also means we don't supply a netmask if
the address is null; that makes sense, as a netmask makes no sense
without an address.
2004-03-11 23:04:24 +00:00
guy 9876992f2c Add support for 99 as a linktype in a capture file. 2004-03-11 19:56:45 +00:00
guy 8842b61d2f Add support for DLT_ value 99, as used by the Axent Raptor
firewall/Symantec Enterprise Firewall.  Thanks, Axent/Symantec, for not
asking us for a DLT_ value and not telling us about the link-layer type.
2004-03-11 09:13:11 +00:00
guy 7dea26c4b6 Document "pcap_get_selectable_fd()" - and give some details on the stuff
you might have to do with that FD to work around the brokenness of
"select()"/"poll()" on most versions of most BSDs.
2004-02-28 02:51:26 +00:00
guy f61f1bb9a6 From Shaun Clowes: work around brokenness in the 32-bit compatibility
layer for Linux on IA-64.
2004-02-27 08:03:05 +00:00
hannes 673ea44790 assign DLT_ and LINKTYPE_ 164 to Juniper Networks Inc. 2004-02-11 22:06:58 +00:00