[2677] in Kerberos_V5_Development
Re: Kerberos NT source code
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Tue Oct 28 13:18:56 1997
Date: Tue, 28 Oct 1997 13:18:11 -0500
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Assar Westerlund <assar@sics.se>
Cc: krbdev@MIT.EDU
In-Reply-To: Assar Westerlund's message of 27 Oct 1997 15:19:23 +0100,
<5l7mazcmqc.fsf@assaris.pdc.kth.se>
From: Assar Westerlund <assar@sics.se>
Date: 27 Oct 1997 15:19:23 +0100
I don't think is the place nor time for threads vs non-threads
discussions.
There seems to be an interest in using libkrb5 in multi-threaded
applications. Therefore, I think we should document the limitations
on doing that, and make libkrb5 more threads-friendly.
Well, do I have a volunteer to look at how the X library solved the
thread-independence issue (I believe they allow the calling application
to register locking and unlocking functions, so you don't have to make
the library thread-library specific), and try to design what would be
necessary to make libkrb5 truely thread-safe?
In order to make it truely thread-safe, I'm pretty sure we *will* have
to use locking primitives, at least in some places....
- Ted
