[1211] in Kerberos_V5_Development
Re: initialization of stdio caches
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Sun May 19 14:36:02 1996
Date: Sun, 19 May 1996 14:35:47 -0400
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Sam Hartman <hartmans@MIT.EDU>
Cc: krbdev@MIT.EDU
In-Reply-To: Sam Hartman's message of Sat, 18 May 1996 15:03:06 -0400,
<199605181903.PAA13770@tertius.mit.edu>
Date: Sat, 18 May 1996 15:03:06 -0400
From: Sam Hartman <hartmans@MIT.EDU>
I would like to know if there are any good reasons in terms of
known portability problems that would be created by fdopen. If it
will work, I think that the security improvement justifies the
abstraction violation. Otherwise, I strongly advise against using
stdio caches by default on Unix systems, as we currently do.
Huh? We're not using stdio caches by default on Unix systems. The only
place where we're using stdio caches by default is on Macintosh
systems. On Unix and Windows, we use FILE ccache by default.
In fact, if it weren't for the case that the Macs are using stdio, there
would be an extremely good case for removing the stdio ccache functions
altogether.
- Ted
