[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CONDUIT latencies (fwd)

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: CONDUIT latencies (fwd)
  • Date: Wed, 21 Jul 1999 11:11:44 -0600 (MDT)
David,

It appears that AIX has the same problem with mmap files so _MAPRGNS was
defined for that system.  I'll include it in the release to be defined for 
Linux. There's a formal 5.0.8 release announcement coming up at the LDM
workshop. Good job.

Thanks,
Robb...

===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Wed, 21 Jul 1999 10:56:08 -0600
From: Steve Emmerson <address@hidden>
To: Robb Kambic <address@hidden>
Subject: Re: CONDUIT latencies (fwd) 

Robb,

>Date: Wed, 21 Jul 1999 10:39:00 -0600 (MDT)
>From: Robb Kambic <address@hidden>
>To: Steve Emmerson <address@hidden>
>Subject: Re: CONDUIT latencies (fwd)

In the above message, you quoted the following:

> What do you think of this solution for the Linux mmap file problem?
> I'm going to step out on a limb and say I've found a software
> fix to this problem.   Defining _MAPRGNS when compiling seems
> to change this troublesome behaviour completely.

According to the autoconf macros in the file "aclocal.m4", this is one
of the things the configure-script sets when on an AIX system---and
it's been that whay since day 1 of LDM5.  Apparently, there were
memory-mapping problems on that system as well and Glenn accounted for
them.

I say do it.

--------
Steve Emmerson   <http://www.unidata.ucar.edu>