Re: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-02 Thread Chas Williams (CONTRACTOR)
In message aanlktikjjj16ckewiu9nyvygozvzzhqdoujtuxsnk...@mail.gmail.com,Derrick Brashear writes: On Fri, Oct 1, 2010 at 11:28 AM, Phillip Moore yeah. thing is, by and large most people don't need that either which is why it still works that way. we've been improving. i suspect openafs should be

Re: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-02 Thread Ted Creedon
In my LaTeX docs I had switches for the directories as well as for the system types. One could emit a SUN or Linux specific doc (or both) for example. Tedc On Sat, Oct 2, 2010 at 6:54 AM, Chas Williams (CONTRACTOR) c...@cmf.nrl.navy.mil wrote: In message

Re: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-02 Thread Russ Allbery
Chas Williams (CONTRACTOR) c...@cmf.nrl.navy.mil writes: i suspect openafs should be prefixing its build vars with AFS_ (or OPENAFS_) so that users can overload the 'standard' variabales. i believe this needs revisited anyway since i am not sure the existing variables are even used

Re: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-02 Thread Simon Wilkinson
On 2 Oct 2010, at 20:04, Russ Allbery wrote: Chas Williams (CONTRACTOR) c...@cmf.nrl.navy.mil writes: i suspect openafs should be prefixing its build vars with AFS_ (or OPENAFS_) so that users can overload the 'standard' variabales. i believe this needs revisited anyway since i am not sure

[OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread Phillip Moore
Well, I didn't go nearly far enough into the caves on that spelunking trip... XCFLAGS and XLDFLAGS appear to be reset, and not inherited from the environment either. Short of post-processing the auto-generated Makefiles, and adding the necessary paths to each place they are needed, is there a

[OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread Andrew Deason
On Fri, 1 Oct 2010 09:10:34 -0400 Phillip Moore w.phillip.mo...@gmail.com wrote: Well, I didn't go nearly far enough into the caves on that spelunking trip... XCFLAGS and XLDFLAGS appear to be reset, and not inherited from the environment either. Short of post-processing the

Re: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread Phillip Moore
I don't think we need to patch anything -- I implemented a script to post-process the results of the configure step, and add the variables I need to Makefile.config. That was easy enough, since I have a build system that makes this kind of customization really easy. I managed to coerce OpenAFS

Re: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread Derrick Brashear
On Fri, Oct 1, 2010 at 11:28 AM, Phillip Moore w.phillip.mo...@gmail.com wrote: I don't think we need to patch anything -- I implemented a script to post-process the results of the configure step, and add the variables I need to Makefile.config.  That was easy enough, since I have a build

[OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread Andrew Deason
On Fri, 1 Oct 2010 18:37:39 +0100 Simon Wilkinson s...@inf.ed.ac.uk wrote: *) Transarc or normal paths *) Install or dest I always thought these were typically non-orthogonal issues. Do we actually use 'install' with transarc paths or 'dest' with non-transarc paths? *) libafs.ko or

[OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread Andrew Deason
On Fri, 1 Oct 2010 13:44:10 -0500 David Boyes dbo...@sinenomine.net wrote: I don't think we've ever really done any kind of requirements capture in this area. OK. Here's some thoughts I had: Should: Run on Linux, Solaris, AIX, Irix for certain; others? HP-UX, Darwin, *BSD. You seem

RE: [OpenAFS] Re: Latest docs on building the OpenAFS source code

2010-10-01 Thread David Boyes
Run on Linux, Solaris, AIX, Irix for certain; others? HP-UX, Darwin, *BSD. You seem to be implying Windows, too... If possible, I think it would be desirable to have a completely common tooling set, including Windows. If we focus on generating a build configuration for the platform in