diff options
| author | Sake Blok <sake@euronet.nl> | 2011-07-15 23:40:16 +0000 |
|---|---|---|
| committer | Sake Blok <sake@euronet.nl> | 2011-07-15 23:40:16 +0000 |
| commit | d00167886f226b71b7176b1ac0c438cd242a39bb (patch) | |
| tree | f54d7a7a84dda1cff08bf5746fb822d25bd8b637 /epan/dissectors/packet-tcp.h | |
| parent | 9659cc77e8f60e02f9c16e8339493b5059911e59 (diff) | |
Stop using the conversation index for tcp stream values as the gaps in the numbering is causing to many questions. Also, numbering could be different for the same file when viewed on different Wireshark versions, which could lead to confusion too.
(see also: http://ask.wireshark.org/questions/5056/how-does-wireshark-calculate-the-tcp-stream-index)
svn path=/trunk/; revision=38056
Diffstat (limited to 'epan/dissectors/packet-tcp.h')
| -rw-r--r-- | epan/dissectors/packet-tcp.h | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/epan/dissectors/packet-tcp.h b/epan/dissectors/packet-tcp.h index 3716c41bc3..54e76df295 100644 --- a/epan/dissectors/packet-tcp.h +++ b/epan/dissectors/packet-tcp.h @@ -227,6 +227,12 @@ struct tcp_analysis { * to previous frame in this conversation */ nstime_t ts_prev; + + /* Keep track of tcp stream numbers instead of using the conversation + * index (as how it was done before). This prevents gaps in the + * stream index numbering + */ + guint32 stream; }; /* Structure that keeps per packet data. First used to be able |