Article View: comp.sys.transputer
Article #1907Re: SUMMARY
From: Knut Menzel
Date: Tue, 26 Jan 1993 14:09
Date: Tue, 26 Jan 1993 14:09
37 lines
2459 bytes
2459 bytes
|> > The problem is: Now that my program works better, I usually switch
|> > the tracing off, so that only the root transputer does any host-I/O.
|> > But I still have to start 4 iservers, otherwise the system hangs.
Only one transputer is allowed to use the iserver because only one
transputer should have a link to the host specified in the .cfs-file.
You **have** to use the reduced library on all nodes **not** connected
to the host.
If you use the full library on the root transputer you have to use the
INMOS iserver because it serves the link. In addition you have to take
care of the two edges connected to the host you find as the first two
parameters you the process running on the root transputer.
If you use the reduced library on the root transputer you do not need
the INMOS iserver. There is no more communication between the root
process and the iserver. You can either send the bootstrap with you own
server or with the iserver **not** using the option -ss (serve the link).
In this case the iserver starts the program and terminates.
|> > To me it looks as if there is some mysterious communication between
|> > iserver and transputer, even if I don't call any I/O-function.
Yes. That's right. The iserver serves the link and waits on an exit call
initiated by the root process (exit_terminate()).
-- Knut
PS: Where did you post your question. I didn't get it.
--------------------------------------------------------------------------
Knut Menzel University of Paderborn
email: knut@uni-paderborn.de Department of Computer Science FB17
...!uunet!unido!pbinfo!knut Warburger Str. 100
Tel: +49 5251 603325 D-4790 Paderborn
Telefax: +49 5251 603853
--------------------------------------------------------------------------
Technical Details
Message-ID:
<1k3gn9INNnjp@news.uni-paderborn.de>Path:
arc1-ca.pugleaf.net!sparky!uunet!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!ira.uka.de!Germany.EU.net!urmel.informatik.rwth-aachen.de!uni-paderborn.de!knutReferences:
<1jjvtuINNgdb@nz12.rz.uni-karlsruhe.de> <1993Jan22.113209.656@ohm.york.ac.uk> <1k149mINNe8u@nz12.rz.uni-karlsruhe.de>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!ira.uka.de!Germany.EU.net!urmel.informatik.rwth-aachen.de!uni-paderborn.de!knut From: knut@uni-paderborn.de (Knut Menzel) Newsgroups: comp.sys.transputer Subject: Re: SUMMARY Date: 26 Jan 1993 14:09:45 GMT Organization: University of Paderborn Lines: 39 Distribution: world Message-ID: <1k3gn9INNnjp@news.uni-paderborn.de> References: <1jjvtuINNgdb@nz12.rz.uni-karlsruhe.de> <1993Jan22.113209.656@ohm.york.ac.uk> <1k149mINNe8u@nz12.rz.uni-karlsruhe.de> Reply-To: knut@uni-paderborn.de NNTP-Posting-Host: ruhr.uni-paderborn.de