[Top] [All Lists]

Re: PCP on Solaris

To: Ken McDonell <kenmcd@xxxxxxxxxxxxxxxxx>
Subject: Re: PCP on Solaris
From: Alan Hoyt <ahoyt@xxxxxxxxxxxxx>
Date: Wed, 29 Jan 2003 18:27:48 -0500
Cc: pcp@xxxxxxxxxxx, Mark Goodwin <markgw@xxxxxxxxxxxxxxxxx>
In-reply-to: <Pine.SGI.4.40.0301300832550.13609503-100000@xxxxxxxxxxxxxxxxxxxxxxxx>
References: <Pine.SGI.4.40.0301300832550.13609503-100000@xxxxxxxxxxxxxxxxxxxxxxxx>
Sender: pcp-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2.1) Gecko/20021130
Ken McDonell wrote:

Alan, just so you know where we're up to ...

On Wed, 29 Jan 2003, Alan Hoyt wrote:

Attached is the first release of the PCP Solaris Sparc patch.

What's the src/magic/result file? ... I don't see that in our source

Trash - Solaris patch trace file for file/magic fix ups - just delete.

And this one src/pmdas/roomtemp/domain.h is made in the build, ditto
for src/pmdas/shping/domain.h and src/pmdas/simple/domain.h and

Right  - date difference on generated files - should have been omitted.

And src/pmdas/trace/pmtrace.c seems to be copied from src/pmtrace
and should not be in src/pmdas/trace.

We should remove the build generated domain.h and pmtrace.c files from pcp-2.3.0.src.tar.gz

Once I'd removed these files from the patch, everything patched cleanly
except for:
    rcs keyword mismatch -- a non-issue, ignored

    our versions have diverged some, so I need to re-apply your
    changes by hand -- I've ignored this issue for the moment

Building on IRIX ...


    You added
        #define HAVE_HSTRERROR
    but this is already defined in include/platform_defs.h for the
    IRIX build, so I suspect you need to conditionally get this
    defined in the Solaris version of platform_defs.h, not the
    access.c source ... I removed the #define there.


And I got a long way further into the build before it blew up in the
modified pmns/Makefile ... I need to investigate this some, as it is
not obvious what happened.

Once I have a clean IRIX build, I'll do an IRIX QA ... then repeat
the process for ia64 Linux and finally ia32 Linux.

Thanks - I'll keep my fingers crossed!

- Alan -

<Prev in Thread] Current Thread [Next in Thread>