[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: CVS Feature Branch - Windows Build Broken 4-Sep-2004
From: |
Derek Robert Price |
Subject: |
Re: CVS Feature Branch - Windows Build Broken 4-Sep-2004 |
Date: |
Sat, 04 Sep 2004 23:33:28 -0400 |
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7) Gecko/20040616 |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Mark D. Baushke wrote:
> Hi Derek,
>
> I have committed a fix for a few of the problems that Conrad indicated
> and one that I found when the server is compiled or not using the
> various ENCRYPTION HAVE_KERBEROS HAVE_GSSAPI options it breaks some of
> the writeproxy-noredirect tests (eg, default builds on FreeBSD and
> Solaris do not have Gssapi-authenticate).
I rewrote one or two of your fixes and silenced a few of `gcc -Wall's
other complaints as well.
> Outstanding issues include:
>
> errors:
> - use of unprotected fcntl() calls in buffer.c(2032,2041)
> - unknown O_NONBLOCK macro use in buffer.c(2037,2039)
Does this stuff go away with TRUST_OS_FILE_CACHE defined to 1?
> - for loop construct problem in log-buffer.c(174)
Got this one.
> - buf_initialize called with too few arguments in socket-client.c(97)
Oooh. I could add the initial 0, 0. The get_fd() function won't mean
anything for NO_SOCKET_TO_FD... but there's also no server on windows
so get_fd() should never be called...
Okay, I committed a possible fix.
> warnings:
> - exithandle.c(47) signals_register vs void function pointers for
> atext().
> - client.c(3359) incompatible types 'struct sockaddr_in *' vs
> 'const struct sockaddr *'
>
> It may make sense to test for fcntl existing and protect calls with
> HAVE_FCNTL in configure...
That may have some other implications, but OK.
> It will probably turn out to be necessary to #define TRUST_OS_FILE_CACHE
> for windows-NT.
Fair enough. That's the default on other platforms.
Derek
- --
*8^)
Email: derek@ximbiot.com
Get CVS support at <http://ximbiot.com>!
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFBOokGLD1OTBfyMaQRAuPtAKC1XMBLsgU8mLTsoWTJSWgx6aaxBACfSQEI
+2FSVzg+upcBwJpgxarxWo0=
=7/zk
-----END PGP SIGNATURE-----