ACES 2.1 question

ACES 2.1 question

[James A. Squire]

I just tried Pretest step 4 with GNAT 3.03 on Solaris where it calls for
executing zp_tcal1 and zp_tcal2 along with a stopwatch.  For some
reason, even though the source code for both of these programs begins
with outputting several text messages, upon execution these programs
output nothing and just sit there.  After recompiling and relinking with
debug on, I ran zp_tcal1 in the debugger and found that it was hung up
with a SIGWAITING or some such thing.  zp_tcal3 runs just fine.
zp_tcal4 and zp_tcal5 also run properly.

Is anybody familiar with this?  Can you point me in the right direction?
I'm not sure that it's a gnat problem and I don't know how to report it
to ACES.
--
James Squire
MDA Avionics Tools & Processes
ja_squire@csehp3.mdc.com
"one of these days I'm going to better myself by going to Knight school"
"You'll be a web knight instead of a web page!"

Re: ACES 2.1 question

[Philip Brashear]

In article <3198E95C.59C8@csehp3.mdc.com>,
James A. Squire <m193884@CSEHP3.MDC.COM> wrote:
>I just tried Pretest step 4 with GNAT 3.03 on Solaris where it calls for
>executing zp_tcal1 and zp_tcal2 along with a stopwatch.  For some
>reason, even though the source code for both of these programs begins
>with outputting several text messages, upon execution these programs
>output nothing and just sit there.  After recompiling and relinking with
>debug on, I ran zp_tcal1 in the debugger and found that it was hung up
>with a SIGWAITING or some such thing.  zp_tcal3 runs just fine.
>zp_tcal4 and zp_tcal5 also run properly.
>
>Is anybody familiar with this?  Can you point me in the right direction?
>I'm not sure that it's a gnat problem and I don't know how to report it
>to ACES.
>--
>James Squire
>MDA Avionics Tools & Processes
>ja_squire@csehp3.mdc.com
>"one of these days I'm going to better myself by going to Knight school"
>"You'll be a web knight instead of a web page!"



James,

We observed this same behavior when testing the ACES.  GNAT is the only
compiler that I've seen to behave this way, and we did not have the resources
to track down the problem.  It may be a race-condition problem (so that GNAT
may not be wrong), or it may be a bug in the tasking runtime.  Other testing
implies that GNAT's clock handling is OK, so it should be acceptable to
proceed.

Although our support contract for ACES is gone, I'm PIWG chair, and feel a
responsibility (and interest) in trying to help when needed.  Direct questions
to me at "brashear@sw-eng.falls-church.va.us" or to the PIWG mailing list,
"perform@sw-eng.falls-church.va.us".  Anyone who wishes to be added to this
mailing list, send mail to me at the "brashear" address, and I'll add you.

Phil Brashear
SIGAda PIWG Chair
CTA INCORPORATED
(513) 258-0831