Re: 5.4 Open Issues

Wed, 11 Oct 2006 10:04:45 -0700

On 10-Oct-2006, at 4:00 PM, Thomas Anders wrote:

Dave Shield wrote:
On 10/10/06, Thomas Anders <[EMAIL PROTECTED]> wrote:
1) OpenBSD4 headers

Huh? We neither have openbsd3.h nor openbsd4.h, but just openbsd.h.

OK - in that case, I don't understand the comment
   "openbsd4 uses openbsd.h, and thus a lot of openbsd3 code
    that most likely works on openbsd4 is excluded"

I'd assumed this was referring to a file openbsd3.h.  Clearly not.
Can someone please clarify the precise issue here.

Robert has logged this issue, not me, but anyway: On OpenBSD 3.x configure adds
a "-Dopenbsd3=openbsd3" while on OpenBSD 4.x it adds "-Dopenbsd4=openbsd4", of
course. Our code only has #ifdefs for openbsd3 (and/or earlier). We either need
to adjust the code to add "|| defined(openbsd4)" or add something like the
following to openbsd.h:

#ifdef openbsd4
#undef openbsd3
#define openbsd3 openbsd3
#endif

The latter is only appropriate if "openbsd3" is to be understood as "3.x or
later" (like we do for freebsd).

Whatever we chose will need to be done again when OpenBSD 5.x comes out.

Here is what I do for Darwin / OS X. I'm trying to build on Tiger (10.4 / Darwin 8) and Leopard (10.5 / Darwin 9) and am ignoring Panther (10.3 / Darwin 7) because the latter has very different APIs (thank a lot to the Intel transition).

Configure defines the OS release, i.e. darwin8 or darwin9. In the respective header file (I created one for each) I also define darwin and set it to the release number, e.g. "#define darwin 8" on Tiger, which allows me to write "#if (darwin >= 8)" for tests. Sounds like BSD could use a similar mechanism...

And yes, I'll post these patches shortly; I haven't yet because I was still fighting with some of the build issues and I have to break them up.

8)  "as-needed" linking
OSX is the only one that fails during "make".

That's the "library layering" bug report, I presume?   (Which I
responded to earlier today)
Naive question - does static vs dynamic linking make any difference here?
Requiring static linking obviously isn't ideal, but would be a
workable stop-gap.

OSX testers, please.

I just responded to your comment: from what I've seen, it doesn't work, but I will try your suggestion and rework the makefile.

                                                               On the other (older) systems
problems probably only happen when doing "make install DESTDIR=/some/where/else".

Which is a relatively unusual invocation.
Yes - it needs to be fixed properly, but doesn't feel to be a show-stopper.

ACK.

OS X builds this way (i.e., with DESTDIR), and it mostly works. If someone has specific components that don't work properly, let me know; the fixes are certainly easier than the issues I'm having with Perl.

9)  OS X embedded perl

Shall we hold off enabling embedded perl by default on OSX?

If it doesn't work, then yes! :-)

We need to find out whether it's broken on OSX for most-if-not-all or just this
single person/setup. OSX testers, please! (Didn't I just say that?)

I have to admit I haven't tried the embedded Perl because of the existing warnings and because of my blocker issues with the Perl modules. I can try enabling it once I get the Perl modules building.

--chris
net-snmp-owners at group.apple.com


 

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to