🚀 go-pugleaf

In <ADAMS.92Aug6211332@PDV2.pdv2.fmr.maschinenbau.th-darmstadt.de> adams@pdv2.fmr.maschinenbau.th-darmstadt.de (Adams) writes:
| It was done under advice of SGI, Munich....

I've still, after almost 2 weeks, been unable to find ANYBODY
in SGI (in Mt. View, CA, or in Europe), who admits to knowing
anything about this.  Could you please give me some more
names of people involved, or at least the phone #'s of the
offices you talked to?

| >   No, something is badly wrong. Are you sure your read(2) calls are in
| >   large units (this matters *a lot*!), say each read(2) is for an entire
| >   50KB frame?
| 
| Yes. Everything else would be too troublesome to program.

So could you include the relevant chunk of code?  It shouldn't
be more than 10-25 lines of code.  If you can't do that, how
about some pseudocode?  Was the code in C, Fortran, or something
else?  Did anybody ever trace the system calls to be sure that
the I/O was really in large chunks (some fortran I/O statements
go through some really slow library code, depending on what
arguments are used)?
--
Let no one tell me that silence gives consent,  |   Dave Olson
because whoever is silent dissents.             |   Silicon Graphics, Inc.
    Maria Isabel Barreno                        |   olson@sgi.com