summaryrefslogtreecommitdiff
path: root/source4/lib/stream
AgeCommit message (Collapse)AuthorFilesLines
2007-10-10r12498: Eliminate INIT_OBJ_FILES and ADD_OBJ_FILES. We were not usingJelmer Vernooij1-1/+1
the difference between these at all, and in the future the fact that INIT_OBJ_FILES include smb_build.h will be sufficient to have recompiles at the right time. (This used to be commit b24f2583edee38abafa58578d8b5c4b43e517def)
2007-10-10r11967: Fix more 64-bit warnings.Tim Potter1-2/+2
(This used to be commit 9c4436a124f874ae240feaf590141d48c33a635f)
2007-10-10r11870: fixed the problem volker reported with the RPX-XPLOGIN test. TheAndrew Tridgell1-1/+37
problem was caused by a callback destroying the packet processing context while that context was being used in packet_recv() This is the first time we have used the ability of talloc destructors to 'refuse' a free request. It works well in this case as it makes the composite API simpler to use for other code, and isolates the complexity of having callbacks destroying the packet context to the packet.c code. (This used to be commit b1b2d86541a376f1ef33fae897f750005c386ebe)
2007-10-10r11713: separate out the setting of the fde in the packet context from theAndrew Tridgell2-5/+14
enabling of packet serialisation (This used to be commit 6a47cd65a8b588f9ddd375c57caaba08281e7cbb)
2007-10-10r11712: avoid changing the fde flags unless really neededAndrew Tridgell1-6/+9
(This used to be commit 48e6424b0cce38f7d8f212d1e891ff8bbd5fec34)
2007-10-10r11642: add some error checksStefan Metzmacher1-2/+8
metze (This used to be commit 9d6406d8daeff0a9bde72ce7749d18fa61324e8a)
2007-10-10r11636: a bit neater solution to the nt_cancel problemAndrew Tridgell2-1/+18
(This used to be commit ba7864b07eebecd4d4eb2ce515412a49964ae179)
2007-10-10r11630: another fix for over-reading in the packet code. This time get theAndrew Tridgell1-4/+12
sign of the comparison right :-) (This used to be commit 7e40077aa793e29b5770aae2e07e964239e8249b)
2007-10-10r11629: fixed a bug found with the socket:testnonblock code. With randomisedAndrew Tridgell1-0/+1
under-reads we could end up supplying a buffer to the client that has an incorrect length (This used to be commit 9c95015b9cccc10a5ba1facd4b48c0fff34e9588)
2007-10-10r11627: give the caller much more control over the stream to packet process,Andrew Tridgell2-2/+46
allowing it to specify the initial read size (thus preventing over-reading) and to stop the recv process when needed. This is used by the dcerpc socket code, which relies on not getting packets when it isn't ready for them (This used to be commit f869fd674ec4b148dc9a264e94d19ce79d35131d)
2007-10-10r11618: added a generic '32 bit length prefix' full packet helper to the ↵Andrew Tridgell2-4/+21
packet code (This used to be commit b4dbe55105cc2807a17d7e5bf8db9756cc526a3b)
2007-10-10r11605: added handling of the send queue to the generic packet handling codeAndrew Tridgell2-0/+65
(This used to be commit f98d499b2ef93cf2d060acafbc424754add322a8)
2007-10-10r11602: added packet_set_serialise() to allow the generic packet layer toAndrew Tridgell2-0/+29
handle optional request serialisation (this is something that is commonly needed on stream connections) (This used to be commit d860eb795693d8c292eec2a639ece4793d28dc38)
2007-10-10r11595: added a helper layer to parse streams into individual packets. This isAndrew Tridgell3-0/+368
something that Andrew Bartlett has been asking for for a while, and when I started having to re-invent this packet parsing code yet again for SMB2 I decided it was time to do it generically you use it by providing a "is this a full packet yet?" helper function to the packet_*() functions, which then handle all the logic of partial packet buffering. This also goes to great lengths to operate efficiently, minimising the number of recv system calls. (This used to be commit e6c47b954a6f09c53ea419800ce873295fcd0be9)