******************************************************************************
********************* IBIS GOLDEN PARSER BUG REPORT FORM *********************
******************************************************************************

INSTRUCTIONS

To report a bug in the IBIS golden parser.  Please fill out the top part
of the following form and send the complete form to info@ibis.org.

A list of reported bugs is maintained at http://ibis.org/bugs/ibischk/ .

******************************************************************************

PARSER VERSION NUMBER: 7.2.0

PLATFORM (SPARC, HP700, PC, etc.): PC

OS AND VERSION: Windows 11

REPORTED BY: Weston Beal, Siemens EDA

DATE: June 13, 2023

DESCRIPTION OF BUG:

Checking an IBIS file that references an IBIS-AMI executable, the parser
issues an error stating that the architecture (OS and data width) is not
correct. The actual problem is that some dependent library is not found, so
the AMI_Init() function is not registered. The checking seems to be correct,
but the message of the error needs to change.

The following Error message is generated for the test case below:

"has platform issue"

The message text should state clearly that the dynamic library is not complete
or missing a dependent library.

The text of the error message should be, "The AMI executable library cannot
run."

Mike LaBonte investigated this error message and wrote this description:
"The 'has platform issue' phrase is not part of an error message, it is
appended to the per-platform DLL checking status lines. But it should be
preceded by an actually error message saying what was wrong. On Windows that
could be either failure to open the DLL or failure to close it. On other
platforms there is no close check. Specifically, if the function AMI_chckexe
in chkexe.c returns any non-zero value, 'has platform issue' is in the summary
printed by ALGMOD_TestExe function in algmod.c. If that happens when there is
no error message, AMI_chckexe will need to be debugged.

"If the question is over the meaning of 'platform issue', that would be fair.
It has some meaning in the case where an Executable line points to a DLL file
that is not correct for the declared platform or bit width. It is less helpful
in the case where the DLL is correct but is unreadable or unable to be opened
for any other reason."

In an email exchange with Atul he provided this descripiton:
"The way the code is implemented is that if AMI_Init() is available then all
is fine and there should not really be any further errors. Thus a 64 bit linux
parser will not even try to load a 32 bit linux .so or a 64 bit windows .dll
or a 32 bit windows .dll. This  is implemented with the help of compilation
macros.  Right now, it is more of an issue that the error occurs deep inside a
function call and the parser prints a kind of catchall message. Hence the
confusion."
 


******************************************************************************
******************** BELOW FOR ADMINISTRATION AND TRACKING *******************
******************************************************************************

BUG NUMBER:     245

SEVERITY: [FATAL, SEVERE, MODERATE, ANNOYING, ENHANCEMENT]  MODERATE

PRIORITY: [HIGH, MEDIUM, LOW]                               LOW

STATUS: [OPEN, CLOSED, WILL NOT FIX, NOT A BUG]             CLOSED           

FIXED VERSION:                                              7.2.1

FIXED DATE:                                                 December 8, 2023

NOTES ON BUG FIX:
Classified by the IBIS Open Forum on June 23, 2023.
To be fixed in the next release.

Weston Beal approved of the fix.


******************************************************************************
******************************************************************************