aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2005-08-16Change to match more closely the way I read 802.11h, at least; I'm justGuy Harris1-56/+79
guessing about the "17" indicator for "management notification" frames - I'm not sure whether, after paying USD 60 to get the current 802.11e draft, Preview on my Mac could even decrypt it. 802.11h seems to say in 7.2.3.12 and 7.3.1.11 that the body of an Action frame has a 1-byte Category field, with a value of 0 meaning "Spectrum management", followed by, for Spectrum management, a 1-byte Action field, with 0=Measurement Request, 1=Measurement Report, 2=TPC Request, 3=TPC Report, and 4=Channel Switch Announcement, followed by: for all but Channel Switch Announcement, a 1-byte Dialog Token followed by some tagged parameters; for the Channel Switch announcement, some tagged parameters. Show some fields given in decimal in various 802.11 specs in decimal, not hex. svn path=/trunk/; revision=15370
2005-08-16From Peter Johansson: add an option to decode the client information,Guy Harris1-190/+286
and check the validity of messages a bit more carefully. Reformat a bit so not as much code is missed by those of us who like narrow editor windows. svn path=/trunk/; revision=15369
2005-08-16The real problem is that iTunes requests gzipped data and responds withGuy Harris1-1/+2
it, so unless you're reassembling headers and bodies the DAAP dissector will see the gzipped crap. The HTTP dissector should perhaps not bother handing gzipped or otherwise unusually-encoded bodies to subdissectors if it's not undoing the encoding. svn path=/trunk/; revision=15368
2005-08-16gcc4 gets confused by the break statements, and I guess it thinks youGuy Harris1-7/+0
can reach them, and thus thinks you can fall off the end of "fc_get_ftype()". svn path=/trunk/; revision=15367
2005-08-16Note where a table comes from, and that we don't use it.Guy Harris1-0/+1
svn path=/trunk/; revision=15366
2005-08-16se_alloc not ep_alloc!Luis Ontanon1-1/+1
svn path=/trunk/; revision=15365
2005-08-16more allocators se_alloc0() se_memdups() se_strdup() se_strndup() ↵Luis Ontanon2-2/+71
se_strdup_printf() svn path=/trunk/; revision=15364
2005-08-15- Update manufJörg Mayer2-1/+58
- Add some Extreme MAC addresses to wka.tmpl svn path=/trunk/; revision=15363
2005-08-15- Change $1 to $@Jörg Mayer1-2/+8
- Don't exit if we can't find the scripts - try to continue anyway, as they might be installed in some preconfigured system directory. svn path=/trunk/; revision=15362
2005-08-15When we create a new tvb, make sure the memory is freed at some point.Gerald Combs1-3/+1
svn path=/trunk/; revision=15361
2005-08-15TP-PI (TP-Parameter-Indicator) is optional.Anders Broman1-0/+9
svn path=/trunk/; revision=15360
2005-08-15Properly dissect vlan flagsJörg Mayer1-11/+40
svn path=/trunk/; revision=15359
2005-08-15- Documentation updateJörg Mayer1-4/+59
- The slot and port numbers printed on the chassis are off by one when compared to the transmitted values (reported by Jim Young). svn path=/trunk/; revision=15358
2005-08-15ememify g_malloc() callsRonnie Sahlberg2-6/+5
svn path=/trunk/; revision=15357
2005-08-15ememify g_malloc() callsRonnie Sahlberg1-6/+4
svn path=/trunk/; revision=15356
2005-08-14Some 'no previous declararion' warning fixesJörg Mayer12-60/+38
svn path=/trunk/; revision=15355
2005-08-14ememify sip and at the same plug a memleakRonnie Sahlberg1-1/+2
every single time the sip dissector was called a 32byte or larger blob was g_malloced and never freed. this not only leaks 32+ bytes + g_malloc bookkeeping overhead but also fragments memory so the real leak of memory is likely much bigger. fix bad memleak svn path=/trunk/; revision=15354
2005-08-14ememify cipRonnie Sahlberg1-3/+2
svn path=/trunk/; revision=15353
2005-08-14change a g_malloc() that was never released into ep_alloc()Ronnie Sahlberg1-9/+10
this function would always leak memory when called. this fixes a memleak in dcm svn path=/trunk/; revision=15352
2005-08-14ememify packet-x11Ronnie Sahlberg1-44/+5
plug a memleak for ...].strptr that is alloced by g_malloc() but never released. svn path=/trunk/; revision=15351
2005-08-14giop allocates only and never releases the memory from the gmemchunk so ↵Ronnie Sahlberg1-14/+2
replace it with equivalen g_malloc() calls and get rid of the gmemchunks svn path=/trunk/; revision=15350
2005-08-14the tds dissector were only using a GMemChunk as a simple way to prevent ↵Ronnie Sahlberg1-23/+3
memleaks in case of an exception. (reset the gmemchunk at the start of each packet) replace this gmemchunk with ep_alloc() svn path=/trunk/; revision=15349
2005-08-14remove two more GMemChunksRonnie Sahlberg1-18/+5
svn path=/trunk/; revision=15348
2005-08-14get rid of another two GMemChunksRonnie Sahlberg1-25/+6
svn path=/trunk/; revision=15347
2005-08-14Data sources have a lifetime spanning a single packet Ronnie Sahlberg1-18/+2
so instead for g-alloc and g_free data just use ep_alloc() that is automatically freed instead. this also gets rid of one more GMemChunk svn path=/trunk/; revision=15346
2005-08-14add dissection of two more SSC commandsRonnie Sahlberg1-55/+135
add 4 commands from SMC to the SSC tables. the fallback for missing commands/i.e. commands defined outside the SSC only handles SPC I.e. If a command for SSC is missing in the SSC tables, assume it might be imported from SPC instead. This fallback only works with opcodes imported from SPC. svn path=/trunk/; revision=15345
2005-08-13Use gboolean for Booleans.Guy Harris1-88/+82
Use "proto_tree_add_item()" whenever possible. svn path=/trunk/; revision=15344
2005-08-13Construct the tvbuff for the payload of an 802.3 packet to have theGuy Harris1-28/+11
appropriate length (min(reported length, captured data present)). svn path=/trunk/; revision=15343
2005-08-13Properly terminate the string generated for string items.Guy Harris1-7/+11
Use "format_text()" to handle non-printable characters in the display string for those items. svn path=/trunk/; revision=15342
2005-08-13Get rid of unused variables.Guy Harris1-27/+24
Use "tvb_format_text()" and friends to format strings for display. Don't throw an exception on errors if we can keep dissecting. Use "PROTO_ITEM_SET_GENERATED()" to mark items as generated. If an item covers real data, it's not generated. Dissect WRITE FILEMARKS (6) (we already had a dissector, but it wasn't being used). svn path=/trunk/; revision=15341
2005-08-13Add an item for CSM_ENCAPS.Gerald Combs1-0/+7
svn path=/trunk/; revision=15340
2005-08-13Fix another infinite loop found by randpkt.Gerald Combs1-2/+2
svn path=/trunk/; revision=15339
2005-08-13Get rid of an unused variable, don't include tcap.h and remove some ↵Anders Broman1-5/+0
redundant code. svn path=/trunk/; revision=15338
2005-08-13From Peter Johansson:Anders Broman1-1/+1
mnc>>4; Should be replaced with mnc = mnc>>4; svn path=/trunk/; revision=15337
2005-08-13"fi" comes before "fr" in alphabetical order. (This also makes it matchGuy Harris1-2/+2
the main "ethereal.desktop" file.) Expand the RCS ID. svn path=/trunk/; revision=15336
2005-08-13Do the sequence number analysis *before* dissecting the payload, so weGuy Harris1-15/+8
don't have to worry about catching exceptions in the payload dissection and doing the sequence number analysis - we weren't doing so in one place. That also puts the sequence number analysis *before* the "TCP payload" entry for payload being reassembled into a later packet. XXX - should we do the tapping before dissecting the payload, too, so that it gets done even if we throw an exception? svn path=/trunk/; revision=15335
2005-08-13Small optimizationJörg Mayer1-4/+4
svn path=/trunk/; revision=15334
2005-08-13Additions from current Suse and DebianJörg Mayer1-0/+2
svn path=/trunk/; revision=15333
2005-08-13get rid of one more gmemchunkRonnie Sahlberg3-44/+2
svn path=/trunk/; revision=15332
2005-08-13Aplly those parts of ethereal_0.10.12-3.diff.gzJörg Mayer29-6/+1537
from Debian, that change debian/ svn path=/trunk/; revision=15331
2005-08-13more gmemchunk -> se_alloc() conversionsRonnie Sahlberg1-31/+4
svn path=/trunk/; revision=15330
2005-08-13start converting gmemchunk into se_alloc() in /epanRonnie Sahlberg1-32/+5
svn path=/trunk/; revision=15329
2005-08-13some more gmemchunk -> se_alloc() updatesRonnie Sahlberg13-440/+125
there are only 5 gmemchunks left but they have different litetime for their allocations than the 100+ ones that have been removed. The remaining 5 should be converted some other way. svn path=/trunk/; revision=15328
2005-08-13gmemchunk -> se_alloc() improvementsRonnie Sahlberg2-38/+6
svn path=/trunk/; revision=15327
2005-08-12gmemchunk -> se_alloc() improvementsRonnie Sahlberg2-65/+8
svn path=/trunk/; revision=15326
2005-08-12some gmemchunk -> se_alloc() improvements for tcpRonnie Sahlberg1-72/+12
svn path=/trunk/; revision=15325
2005-08-12gmemchunk -> se_alloc() improvementsRonnie Sahlberg2-39/+5
svn path=/trunk/; revision=15324
2005-08-12gmemchunk -> se_alloc() improvementsRonnie Sahlberg2-37/+6
svn path=/trunk/; revision=15323
2005-08-12gmemchunk -> se_alloc() improvementsRonnie Sahlberg4-82/+13
svn path=/trunk/; revision=15322
2005-08-12more gmemchunk -> se_alloc() improvementsRonnie Sahlberg6-163/+24
svn path=/trunk/; revision=15321
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-11 22:12:35 +0000