home *** CD-ROM | disk | FTP | other *** search
- This directory (dist/ml/sml-386) contains a port by Yngve Guttesen to
- the I386 running MS Windows/MSDOS 5.0. It requires 16MB of memory and
- a very particular memory management configurnation that we cannot
- characterize (not knowing the mysteries of PC memory management).
- The I386 code generator has been incorporated in the main distribution
- (by Mark Leone) and runs under Unix (Mach and BSD386) on Intel hardware.
- everything you need. We also keep the files
- individually uncompressed for partial retrieval.
-
- REPORT:
- -------
-
- Please note that the report, found in the report subdirectory, is 42 pages
- long. Please check to see that you get them all when you print the
- postscript file. The size of the postscript file is 694769 bytes. If the
- size is right and it doesn't print all the pages, you may have to modify
- the header of the postscript file to adapt it to your system. Some people
- still have a problem printing the postscript file. We therefore provide
- the Latex sources and the postscript files for the figures (in the
- report subdirectory) so that you can remake the postscript and print it.
- Finally, if all else fails the tech report can be requested by sending
- mail to schulman@sierra.Stanford.EDU. The tech report number is
- CSL-TR-92-526, dated June 1992. The original tech
- report was number CSL-TR-91-469, but this is somewhat dated. We will issue
- a new tech report when there are major changes to the suite (new applications,
- etc.). However, the latest version of the report is always on-line.
-
- MACROS:
- -------
-
- In the macros directory, we provide macros and sample makefiles
- for three machines:
-
- Encore Multimax (CMU Mach 2.5: C and Fortran)
- SGI 4D/240 (IRIX System V Release 3.3: C only)
- Alliant FX/8 (Alliant Rev. 5.0: C and Fortran)
-
- These macros work for us with the above operating systems. Unfortunately,
- our limited resources prevent us from supporting them in any way or
- even fielding questions about them. If they don't work for you, please
- contact Argonne National Labs for a version that will. An e-mail address
- to try might be monitor-users-request@mcs.anl.gov. An excerpt from
- a message, received from Argonne, concerning getting the macros follows:
-
- "The parmacs package is in the public domain. Approximately 15 people at
- Argonne (or associated with Argonne or students) have worked on the parmacs
- package at one time or another. The parmacs package is implemented via macros
- using the M4 macropreprocessor (standard on most Unix systems). Current
- distribution of the software is somewhat ad hoc. Most C versions can be
- obtained from netlib (send electronic mail to netlib@ornl.gov with the message
- send index from parmacs). Fortran versions have been emailed directly or sent
- on tape. The primary documentation for the parmacs package is the book
- ``Portable Programs for Parallel Processors'' by Lusk, et al, Holt, Rinehart,
- and Winston 1987."
-
- The makefiles provided in the individual program directories specify
- a null macro set that will turn the parallel programs into sequential
- ones. Note that we do not have a null macro set for FORTRAN.
-
- MAILING LIST:
- -------------
-
- Please send a note to splash@mojave if you have copied over the programs,
- so that we can put you on a mailing list for update reports.
-
- CHANGES:
- -------
-
- June 5, 1991.
- ------------
- 1. The NEWPROC and PAUSEDEC macros have been removed.
- 2. The file adv.U in mp3d has been updated to fix a boundary
- condition.
- 3. The ocean program has been updated with a new initial guess
- to the equation solver to improve its convergence. A new
- parameter has been added to the solver routine, and the calls
- to it have correspondingly changed.
-
- June 13, 1991.
- -------------
- 1. For mp3d, Global->BC is now initialized to 0 in setup.U.
- 2. Next version of tech report complete.
-
- July 31, 1991.
- -------------
- 1. Small update made to mp3d section of report. Concerns number
- of time-steps run with simulator.
- 2. Water source code updated with comments to describe input
- parmeters better.
-
- Aug. 6, 1991.
- ------------
- 1. Bug in water fixed. Some interactions were not being
- examined correctly and a barrier was missing. Some other
- minor cleanup done in water.
- 2. A new version of pthor created. It primarily improves
- cache mapping behaviour. See README file in pthor
- directory.
- 3. Reference to paper on data locality and memory system
- performance in Ocean added.
-
- Aug. 28, 1991
- -------------
- 1. Locking bug in mp3d fixed. Caused null pointer dereferencing
- on some machines.
-
- May 29, 1992
- ------------
- 1. New application (Barnes-Hut) added.
- 2. Report updated accordingly. Section 4.2 on general behavioral
- characteristics cleaned up and enhanced somewhat.
- Also, indication given of how to scale some of the
- scientific applications to run on larger machines.
-
- July 24, 1992
- -------------
-
- 1. Some bugs in water fixed:
- a) determining when comp_last should be decremented
- to compute interactions correctly in interf.U and
- poteng.U fixed.
- b) a race condition between initializing and accumulating
- into gl->VIR removed. A new variable called
- gl->initialized introduced to do this.
- c) the code will not crash if input variable NFMC is
- set to 0 and the input displacement file LWI12 (not
- needed in this case) not provided.
-
-