Newsgroups: comp.parallel
From: esrlk@eng.warwick.ac.uk (A L Roberts)
Subject: Problems with use of TMS320C40's
Organization: Engineering, University of Warwick, UK
Date: 5 Sep 1995 14:41:58 GMT
Message-ID: <42hnjm$7gr@usenet.srv.cis.pitt.edu>

I was wondering if anybody could give me some advice.

We have started using an LSI QPC/C40 board with 7 TMS320C40 processors
running on it. We are using 3L's Parallel C software and the debugger
DB40. 

The problem is due to the fact that the amount of computation we are
carrying out on each processor is relatively small, the time taken to
transfer data between the processors is much greater than the time
taken for the computations:

Using the the DB40 Debugger it appears that running the program on one
processor the computation time to calculate one value takes 46 clock
cycles (this time does not include file operations, declaring
variables, etc). If the program is then split onto two processors the
computation time on each processor reduces to 27 clock cycles per
value, but the transfer of data between the two processors takes 192
clock cycles per value (using the chan_in_message and chan_out_message
commands does this seem reasonable ?).  The time taken to calculate
all the necessary points actually increases when  run on two or more
processors. 

It was felt that writing the software in assembly language would
overcome this problem:

The C programs where compiled with the -s command so as to produce
their assembly language listings. When these assembly language
listings were linked (using the lnk30 command and the c40cfbr.lib
library file) error comments were produced, "linking files for
incompatible targets". The programs had been compiled using the asm30
-v40 *.asm command.  Am I using the correct library file, I have tried
all the other c40c***.lib files with no success. 

The assembly language listings produced call subroutines such as
printf, fclose, etc. I do not seem to be able to link these
subroutines, has anybody  got any ideas.

                        Yours Hopefully
                         Andy Roberts
                     University Of Warwick

