Re: Any one using Compaq(DEC) Tru64 UNIX?

[W . Northcott]

Oh so close.

Thanks again for the help so far.  You nailed the HDF5 problem.

Would I be correct in assuming that these patches are being applied to the
current snapshots?  My plan was to move to a more recent snapshot as soon
as I can get the 2.1.1 release running without error.

Unfortunately we are not quite there.  I started with a fresh source tree
for the 2.1.1 release, and applied all our patches so far.  The libraries
compile without error, but there is still one test failure.

PASS: archiver
{r*@:ifd} address@hidden:address@hidden:+16i+24d+80d+88}
m1: [1 2.00 3.00]
{r*@:idf} address@hidden:address@hidden:+16i+24d+80d+88}
m2: [4 6.00 5.00]
{r*@:fdi} address@hidden:address@hidden:+16d+72d+80i+40}
m3: [9 7.00 8.00]
{r*@:dfi} address@hidden:address@hidden:+16d+72d+80i+40}
m4: [12 11.00 10.00]
{r*@:fdii} address@hidden:address@hidden:+16d+72d+80i+40i+48}
m5: [15 16 13.00 14.00]
{r*@:ffdii} address@hidden:address@hidden:+16d+72d+80d+88i+48i96}
m6: [20 21 17.00 18.00 19.00]
{c@:i} address@hidden:address@hidden:+16i+24}
cm1i: `B'
{s@:i} address@hidden:address@hidden:+16i+24}
sm1i: 3
{i@:i} address@hidden:address@hidden:+16i+24}
im1i: 5
{q@:i} address@hidden:address@hidden:+16i+24}
lm1i: 7
{f@:i} address@hidden:address@hidden:+16i+24}
fm1i: 9.000000
FAIL: forwarding
===================
1 of 2 tests failed
===================

Am I correct in assuming that the contents of the second and third pairs
of braces on a line should be identical for the test to pass?

There are no comments in the source code.  Were these stripped out to save
space in the release?  Are they available in CVS snapshots.  It makes it
hard to fix things without clues!  The Swarm reference docs only refer to
public interfaces.

>I don't get a crash using the ASM-2.0 code from ftp.swarm.org, but
>I do get a crash when I try to run Paul's newer code on sourceforge.
>The symptom is a failed allocation from a silly value of numfcasts.
>If you're nice to Paul, maybe he'll help investigate since he has

Thanks for that.  I will try Paul once I am sure I have a set of libraries
which are up to date and pass all tests.

Bill Northcott



                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.