SMB packet decoder now ignores empty parameter/data and will not try to decode and display these.
SMB_COM_SEARCH decoding had the sources mixed up, decoding client->server traffic as server->client traffic and the other way round.
Decoding of the status field now works properly again in the non-"status32" case.
smba_read() failed to register the number of bytes read successfully by smb_proc_read_raw(). This made all files appear to be empty.
smba_write() did not update the offset value correctly if the first smb_proc_write() call had succeeded.
Decoding is performed before the session header type is checked and, if necessary, rejected.
smb_receive_raw() now only accepts a single session header type (message) and rejects all others. At this stage "positive session response" should no longer appear.
All functions now follow a consistent error reporting scheme. A negative return value always indicates an error condition. A return value of 0 always indicates success, except for the few functions which have to return a count of bytes or directory records upon success.
Previously, it was hard to tell what function returned what in case of error, which led the bugs in smba_write() and smba_read(), neither of which gracefully fell back on SMB_COM_WRITE/SMB_COM_READ, respectively if SMB_COM_WRITE_RAW/SMB_COM_READ_RAW failed.
State variables are now named according to their respective purpose where possible. Previously, it was "rval", "errnum", "result" which were used almost interchangeably regardless of purpose.
Turns out that the length filled in by smb_setup_header() was not actually off by 4 bytes but entirely correct. This caused SMB packets to be sent four bytes short, and probably causing some SMB server implementations to ignore them altogether.
A successful SMB_COM_WRITE_RAW command may cause the server to send interim progress update responses, ending with a final SMB_COM_WRITE_RAW_COMPLETE message. This is now handled correctly.
Signed quantities are now used as such, e.g. the server time zone.
The SMB_COM_READ/WRITE/READ_RAW/WRITE_RAW data is now printed in the scope it was intended for, e.g. the SMB_COM_WRITE_RAW data output begins at the indicated offset rather than at the padding bytes which may precede it.
Added server/client capability flags covered by "Implementing CIFS", but not by the Microsoft reference documentation.
The NetBIOS session header contents are now printed along with the SMB header information.
Added GPLv2 header text.
The SMB date/time, filetime and utime information is now converted into display format instead of just showing up as hexadecimal format numbers.
The end-of-file and allocation size values are now printed in decimal format in addition to hexadecimal format.
Made a note of the fact that the SMB_COM_READ_RAW may be followed by more than one response by the server. The current implementation only catches one response.
The data transmitted by the SMB_COM_WRITE_RAW command is no longer decoded as if it were an SMB message.
Time is now printed uniformly in hexadecimal format.
dword data is now printed as unsigned integer values.
Decoded error messages should be a bit more readable.
The decoder now handles all SMB commands which the smbfs file system uses, including the two subcommands of the SMB_COM_TRANSACTION2 command which are used (TRANS2_FIND_FIRST2 and TRANS2_FIND_NEXT2).
This is still very ugly and very complex code, but hopefully the next version will only be ugly...
Major rework of the code which decodes individual commands, their parameters and the responses. This is still incomplete, but it's a lot more useful now than before.
The word/dword conversion code should be more robust now, using type casts where needed.
Cleaned up numeric comparisons, so that the types of the quantities compared match.
The SMB packet dump code now prints the current transmit buffer size along with the other packet information.
Because the field and variable name "max_xmit" was used in four different places, with different types in different context, it was hard to know what was what. This is part of the cleanup work.
The word/dword conversion code should be more robust now, using the type casts where needed.
The maximum transmission size provided by the server is now a local variable and no longer stored in the global server parameters because it is really only needed here.
The maximum transmission size limit requested by the user is no longer ignored.
Olaf Barthel
Olaf Barthel