aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* rebuild libraries with ARPHRD_TUNNEL6 supportHEADmasterHimbeerserverDE2024-02-152-0/+0
|
* treat ARPHRD_TUNNEL6 tunnels as validHimbeerserverDE2024-02-151-0/+1
|
* move binaries to common out/ dirHimbeerserverDE2023-12-043-4/+4
|
* rebuild libraries without libnlHimbeerserverDE2023-12-042-0/+0
|
* move CC flag to its own lineHimbeerserverDE2023-12-041-2/+4
|
* add missing \ seperators to new build script flagsHimbeerserverDE2023-12-041-2/+2
|
* compile without libnlHimbeerserverDE2023-12-041-0/+2
|
* build static libsHimbeerserverDE2023-08-172-0/+0
|
* add precompile scriptHimbeerserverDE2023-08-171-0/+39
|
* pcap savefile: Fix formatting "link-layer type and additional information"Francois-Xavier Le Bail2023-08-161-10/+6
| | | | | | The problem only appears in the html conversion. Use .nf (no-fill) mode.
* Fix a typo in a commentFrancois-Xavier Le Bail2023-08-131-2/+2
| | | | | | Found with the codespell command. [skip ci]
* Merge pull request #1082 from luizluca/realtek_tag_2Guy Harris2023-08-071-0/+3
|\ | | | | Add support for Realtek (Ethertype) DSA data
| * Add support for Realtek (Ethertype) DSA dataLuiz Angelo Daros de Luca2022-03-071-0/+3
| | | | | | | | | | | | | | | | | | | | Realtek switchtag rtl4a (4 bytes long, protocol 0xA) and rtl8_4 (8 bytes long, protocol 0x04) are Ethertype DSA tags, inserted in the Ethernet header similar to an 802.1Q tag. Both shares the same Ethertype 0x8899 as other Realtek proprietary protocols. Realtek switchtag rtl8_4t is identical to rtl8_4 but positioned before the CRC, at the end of the Ethernet frame.
* | Merge pull request #1090 from randomkny/fix-enobufsGuy Harris2023-08-061-1/+1
|\ \ | | | | | | nflog: only increment packets_nobufs when recv() returns an error
| * | nflog: only increment packets_nobufs when recv() returns an errorKenny Luong2022-02-081-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Errno should only be valid when recv() returns a `-1`, indicating an error. I believe the intended behavior here is for packets_nobufs to be a counter that reports back how many times recv() returns an ENOBUFS during a packet capture. Because of the existing logic however, packets_nobufs begins incrementing for every recv() call once the first ENOBUFS error is seen, since errno is not reset when there are no errors returned from recv(). Before (counter deviates from strace): # tcpdump output 38069 packets captured 38069 packets received by filter 38061 packets dropped by kernel # strace output % time seconds usecs/call calls errors syscall ------ ----------- ----------- --------- --------- ---------------- 26.47 0.282728 7 38067 3 recvfrom After (counter matches strace): # tcpdump output 38095 packets captured 38095 packets received by filter 7 packets dropped by kernel # strace output % time seconds usecs/call calls errors syscall ------ ----------- ----------- --------- --------- ---------------- 27.11 0.258596 6 38096 7 recvfrom
* | | doc: Add more details to README.solaris.md. [skip ci]Denis Ovsienko2023-08-041-4/+14
| | |
* | | pcap savefile: Set the date stampFrancois-Xavier Le Bail2023-08-031-1/+1
| | | | | | | | | | | | [skip ci]
* | | pcap savefile: Update about "link-layer type and additional information"Francois-Xavier Le Bail2023-08-022-4/+49
| | | | | | | | | | | | | | | | | | Update based on draft-ietf-opsawg-pcap. [skip ci]
* | | Fix substitute-and-replace error.Guy Harris2023-08-011-2/+2
| | |
* | | We also have to map LINKTYPE_ codes to DLT_ codes in other cases.Guy Harris2023-08-011-1/+49
| | | | | | | | | | | | Fix a typo (= when == was intended) while we're at it.
* | | pcap-linux: plug an FD leak.Guy Harris2023-08-011-1/+3
| | | | | | | | | | | | | | | | | | | | | If map_arphrd_to_dlt() fails, free sock_fd before returning the error status, as we do elsewhere. Fixes Coverity CID 1539177.
* | | Redo DLT_/LINKTYPE_ translation.Guy Harris2023-08-013-127/+282
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Do it with a bunch of ifs rather than with a translation table; that lets the logic work differently for DLT_ -> LINKTYPE_ mapping and LINKTYPE_ -> DLT_ mapping. Have two "matching" ranges, so that the linktypes that preceded the BSDs all going off in their own directions can be handled as a matching range. Avoid doing mapping if the corresponding LINKTYPE_ and DLT_ codes have the same numerical value. For LINKTYPE_ -> DLT_ mapping, don't map link-layer type values outside either of the matching ranges if we don't have a specific mapping set up for them, just treat the LINKTYPE_ value as if it's a DLT_ value. That makes us handle some DLT_ codes outside the high mapping range to which we assigned matching LINKTYPE_ codes, as well as attempting, as best we can, files written with platform-dependent DLT_ codes (such as DLT_RAW) as link-layer type codes (programs *on that platform* will handle them correctly; programs will not do so on other platforms, but that's better than not handling them correctly anywhere). Update various comments.
* | | autoconf: Add some warning flags for clang 13 or newerFrancois-Xavier Le Bail2023-07-281-0/+3
| | | | | | | | | | | | | | | | | | -Wnull-pointer-subtraction -Wunused-but-set-parameter -Wunused-but-set-variable
* | | Get rid of trailing blanks.Guy Harris2023-07-271-6/+6
| | |
* | | Merge pull request #1209 from the-tcpdump-group/nuke-msdosMichael Richardson2023-07-2719-4436/+12
|\ \ \ | | | | | | | | MSDOS packet driver interface is no longer testable
| * | | still remove for _WIN32Michael Richardson2023-07-261-0/+2
| | | |
| * | | remove additional ifdefs based upon MSDOSMichael Richardson2023-07-267-61/+11
| | | |
| * | | removed DOS pcap driver, and Makefile aspects that configured itMichael Richardson2023-07-264-1789/+1
| | | |
| * | | MSDOS packet driver interface is no longer testable, anyone needs it can use ↵Michael Richardson2023-07-269-2588/+0
|/ / / | | | | | | | | | a previous version
* | | Clean up indentation.Guy Harris2023-07-211-114/+110
| | |
* | | Get rid of trailing space.Guy Harris2023-07-181-1/+1
| | |
* | | Linux USB: avoid 32-bit unsigned integer wraparound.Guy Harris2023-07-181-7/+66
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When adding values whose sum might overflow an unsigned integer, first check whether the sum *would* overflow an unsigned integer and, if so, clamp the sum at UINT_MAX. Do the same for a multiplication. This should fix #1134, as well as the issue in #1205.
* | | Merge pull request #1206 from headshog/fix-numtrunc-pcap-utilGuy Harris2023-07-171-1/+1
|\ \ \ | | | | | | | | Numeric truncation at `pcap-util.c:374`
| * | | fix pcap-util.cheadshog2023-07-171-1/+1
|/ / /
* | | Fix a comment, due to changes from the previous commit. [skip ci]Guy Harris2023-07-081-4/+4
| | |
* | | Redo the availability macros.Guy Harris2023-07-081-28/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Apple appears to have tweaked their pcap/funcattrs.h to do the right thing at some point, so 1) there's no need to urge them to do so any more and 2) they only seem to export libpcap APIs in macOS, so redo the PCAP_AVAILABLE() stuff to be more like what they've done. We remove the include of <Availability.h>, as we don't need it ourselves (we don't do the availability stuff in our releases, for reasons given in the comment before the availability stuff), and Apple uses <os/availability.h> - we leave it up to them to modify the header, just as they do with, for example, ZLib's zlib.h. (Many thanks to the Apple people who picked this up!)
* | | Fix *another* typo.Guy Harris2023-07-031-1/+1
| | |
* | | Fix a typo in a commentFrancois-Xavier Le Bail2023-07-021-1/+1
| | |
* | | Cirrus CI: get rid of the Windows task. [skip ci]Guy Harris2023-07-011-30/+0
| | | | | | | | | | | | | | | | | | | | | There are CI platforms that are useful for doing Windows CI using Visual Studio. It appears that Cirrus CI is not one of them, as they don't appear to have images with VS pre-installed, and it does appear that using Chocolatey to install it takes a long time.
* | | Cirrus CI: use Chocolatey to install... Visual Studio!Guy Harris2023-07-011-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Apparently the container starts out pretty bare-bones, with just Chocolatey and git+git-bash. Everything else, *including Visual Studio*, is up to you. Thanks for documenting this so well, Cirrus folks! (Note: not mentioning Visual Studio is inequivalent to mentioning that it's not part of the container.)
* | | Cirrus CI: show *everyhing* under C:\Program Files (x86).Guy Harris2023-07-011-1/+1
| | | | | | | | | | | | Why the obsession with hiding directories, Cirrus CI?
* | | Cirrus CI: see what's in C:\Program Files (x86).Guy Harris2023-07-011-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | That's probably where Visual Studio is hiding, and we'll probably have to run the vcvars.bat script to get the directory containing cl.exe into the Windows path. (And the paths for stuff under C:\Program Files\Git looks weird because that stuff appears under / in git-bash; that's not the problem.)
* | | Cirrus CI: dump directories under C:\Program Files\Git selectively.Guy Harris2023-07-011-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There's too much stuff under there; look at the directories that appear in the Windows PATH. Look at the new stuff in the UN*X path as well; it might be that C:\Program Files\Git is the MinGW/MSYS2 root, so the UN*X PATH isn't getting mangled. (And it may be that the failure to find cmake may be that cmake isn't installed by default, so we'd have to choco it or something such as that. It'd be nice if it were installed as part of Visual Studio in the containers.)
* | | Cirrus CI: do a bunch of directory listings on Windows.Guy Harris2023-07-011-1/+4
| | | | | | | | | | | | | | | Where are the programs we need hidden, and how do we get those locations into $PATH?
* | | Cirrus CI: dump out /etc/msystem.Guy Harris2023-07-012-179/+1
| | | | | | | | | | | | | | | | | | It's run by /etc/profile. Get rid of the overwrite of /etc/profile.
* | | Cirrus CI: run the main script in a login shell.Guy Harris2023-07-011-1/+1
| | | | | | | | | | | | That should cause /etc/profile to be run.
* | | Cirrus CI: try overwriting /etc/profile and adding a debug emssage.Guy Harris2023-07-012-0/+179
| | |
* | | Cirrus CI: too many "s.Guy Harris2023-07-011-1/+1
| | |
* | | Cirrus CI: see if we can jam the raw Windows path into the UN*X environment.Guy Harris2023-07-011-1/+1
| | | | | | | | | | | | | | | If so, we can then do our own path construction from it, overriding the garbage that we're getting from the system.
* | | Cirrus CI: dump out $ORIGINAL_PATH before running build_matrix.sh.Guy Harris2023-07-011-3/+1
| | | | | | | | | | | | | | | It doesn't show up in env's output, but maybe, just maybe, please?, it'll be available.
generated by cgit v1.2.3 (git 2.39.1) at 2025-05-09 14:35:27 +0000