Hi,
I must say the code seems to be all ready for non-blocking operation of
connect() at least for POSIX-based platforms, it only misses the non-blocking
flag setting before calling it (however, immediately afterwards there is a wait
creation which sets the flag as well).
Thanks for all the good work,
Mikhail
-Original Message-
From: ext Pekka Pessi [mailto:[EMAIL PROTECTED]
Sent: Tuesday, May 08, 2007 11:43 AM
To: Zabaluev Mikhail (Nokia-M/Helsinki)
Cc: sofia-sip-devel@lists.sourceforge.net
Subject: Re: [Sofia-sip-devel] Patch: unblock connect()
2007/5/7, Mikhail Zabaluev [EMAIL PROTECTED]:
Here's a one-line patch that:
a) passes the test suite and our application smoke test runs;
b) makes a difference when the Sofia root is set to the
non-threading mode.
If not for blocking connect(), the Sofia-SIP coexists in the
same thread
with the D-Bus service in our GLib mainloop. This patch removes the
single discovered temptation for our project to enable
threading, which
would be a change for the worse in our platform.
Additionally, it makes
for better behavior when connect() blocks, as the stack then
times out
accordingly to its own preference, and not what the operating system
deems good for connect().
I'll test this on win32 and apply if it seems to work there, too. I
though I moved the nonblocking code to su_socket(), but it looks like
I did not.
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel
