|
event_context for the socket_connect() call, so that when things that
use dcerpc are running alongside anything else it doesn't block the
whole process during a connect.
Then of course I needed to change any code that created a dcerpc
connection (such as the auth code) to also take an event context, and
anything that called that and so on .... thus the size of the patch.
There were 3 places where I punted:
- abartlet wanted me to add a gensec_set_event_context() call
instead of adding it to the gensec init calls. Andrew, my
apologies for not doing this. I didn't do it as adding a new
parameter allowed me to catch all the callers with the
compiler. Now that its done, we could go back and use
gensec_set_event_context()
- the ejs code calls auth initialisation, which means it should pass
in the event context from the web server. I punted on that. Needs fixing.
- I used a NULL event context in dcom_get_pipe(). This is equivalent
to what we did already, but should be fixed to use a callers event
context. Jelmer, can you think of a clean way to do that?
I also cleaned up a couple of things:
- libnet_context_destroy() makes no sense. I removed it.
- removed some unused vars in various places
(This used to be commit 3a3025485bdb8f600ab528c0b4b4eef0c65e3fc9)
|
|
DCOM paper in lorikeet. This is the result of 1.5 months work (mainly
figuring out how things *really* work) at the end of 2004.
In general:
- Clearer distinction between COM and DCOM. DCOM is now merely
the glue between DCE/RPC+ORPC and COM. COM can also work without
DCOM now. This makes the code a lot clearer.
- Clearer distinction between NDR and DCOM. Before, NDR had a couple of
"if"s to cope with DCOM, which are now gone.
- Use "real" arguments rather then structures for function arguments in
COM, mainly because most of these calls are local so packing/unpacking
data for every call is too much overhead (both speed- and code-wise)
- Support several mechanisms to load class objects:
- from memory (e.g. part of the current executable, registered at start-up)
- from shared object files
- remotely
- Most things are now also named COM rather then DCOM because that's what it
really is. After an object is created, it no longer matters whether it
was created locally or remotely.
There is a very simple example class that contains
both a class factory and a class that implements the IStream interface.
It can be tested (locally only, remotely is broken at the moment)
by running the COM-SIMPLE smbtorture test.
Still to-do:
- Autogenerate parts of the class implementation code (using the coclass definitions in IDL)
- Test server-side
- Implement some of the common classes, add definitions for common interfaces.
(This used to be commit 71fd3e5c3aac5f0002001ab29d2248e6c6842d6f)
|
