It's interesting, because usually [address] would mean that the address is optional. However, I did some reading, and I found out that it is actually complaining because you didn't give any "sym" value(s).
I guess that one thing that I think is worth consideration for change is the documentation of the :b command. From Chapter 6 of the mdb manual: [ addr ] ::bp [+/-dDestT] [*-c* cmd] [*-n* count] sym ... addr :b [cmd ... ] Set a breakpoint at the specified locations. The ::bp dcmd sets a breakpoint at each address or symbol specified, including an optional address specified by an explicit expression preceding the dcmd, and each string or immediate value following the dcmd. ... If the :b form of the dcmd is used, a breakpoint is only set at the virtual address specified by the expression preceding the dcmd. In this description, "addr" does not appear to be optional. I guess, you are saying that for all legacy commands that have arguments before the command, "dot" is the actual argument, no matter what the documentation says, and what the expression before the command really means is "set dot to this". This is actually documented in chapter 3, under "Commands". If you look at it that way, this entry sort of makes sense, because it is telling you that the :b command is going to interpret the value of dot as an address. I guess that the parsing of dcmds that start with :: is more complex. This message posted from opensolaris.org
