[Projectpi-public] Fwd:PI, Plouffe and Bellard, and openMoix

[Ian L. <Ian...@mq...>]

Hello,

  I am not a member of this list.  I just wanted to get this patch through
to Sayan but his mailbox is full and bouncing email.


Thanks,




Forwarded message follows:
>From: "Ian Latter" <Ian...@mq...>
>To: <sa...@pe...>, <m.g...@cq...>
>Subject:  PI, Plouffe and Bellard, and openMoix
>Date: Wed, 17 Dec 2003 00:23:15 +1000
>
> Hello Sayan and Myles,
> 
> 
>   I am one of three designers for the CHAOS openMosix distribution;
> 
>                       http://itsecurity.mq.edu.au/chaos/
> 
>   Searching through Google this evening, looking for openMosix sites
> in Australia, I came across Myles' site at the Central Queensland Uni;
> 
>      http://infocom.cqu.edu.au/Staff/Myles_Greber/Research/Cluster/
> 
> 
>   I saw that Myles was using  the Plouffe and Bellard algorithm to
> calculate PI, as a test suite for his cluster.  This is pretty cool; as
> the CHAOS "testapp" consists of an incrementing integer.  I went
> searching for PI software, until I came across Sayan's software;
> 
>                      http://projectpi.sourceforge.net/
> 
>   Which looked very interesting, particularly as the code pi1_f.c
> included comments seeking openMosix capabilities.
> 
>   Attached is an upgrade to that code.  I am offering it back to you,
> Sayan, for inclusion into your tree, per the GPL terms.  The code 
> mods are a little crude, in that the program will now fork a copy 
> off for each 9 PI digits (pseudo_main).  It would be better if there
> were a fork quantity option .. but not this evening .. not for a test
> application.
> 
>   I haven't tested this against an openMosix cluster, as I am 
> currently unable to access one, but I am quietly confident that
> it will work (it forks and completes correctly under a vanilla
> distribution).
> 
>   During my work, I broke the timers used in the main function, and 
> then found a better way to calculate them (from the following two
> sites);
> 
>              http://www.phys.ufl.edu/docs/porting-faq.html
>              http://rla.zoom.es/libc_17.html
> 
>   So I rewrote all of the timing code in pi1_f.c, also.
> 
> 
>   Take a look through it and let me know what you think.  If you have
> any questions then please don't hesitate to ask.  The updated
> version will probably find its way into the next version of CHAOS
> (maybe as "testapp2" or similar).
> 
> 
> 
> Regards,
> 
> 
> 
> 
> --
> Ian Latter
> Internet and Networking Security Officer
> Macquarie University

--
Ian Latter
Internet and Networking Security Officer
Macquarie University