Thank you for that assurance. It will help in the negotiations to follow.
Peter
-Original Message-
From: Peter Relson [mailto:[EMAIL PROTECTED]
Sent: Friday, June 24, 2005 7:23 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Name/Token Services in COBOL? IEANTC
If the result of a C compile
If the result of a C compile is "warning", then I'm sure we could be
persuaded to take an APAR.
Having said that, the silly file has been out there for something like 3
years, with no one mentioning a problem, and it's in samplib so certainly
can be changed by your shop to suit your site's needs.
The biggest (only?) drawback to using tokens is if you also are using the
CHKPT macro (or coding CHKPT=EOV in JCL, which invokes CHKPT macro.)
If you are using CHKPT then the only tokens that can be active are
ones that are TASK LEVEL and defined with the IEANT_CHECKPOINTOK
persist option. Otherw
Farley, Peter x23353 wrote:
I'd accept an #ifdef/#endif that tested the version of the compiler if
that's what it takes, but RC=4 is not acceptable.
This is exactly how uplevel/downlevel declarations are supposed to be
handled. The includes provided with the C/C++ compilers do this all the
d the version of the compiler if
that's what it takes, but RC=4 is not acceptable.
Peter
-Original Message-
From: Peter Relson [mailto:[EMAIL PROTECTED]
Sent: Thursday, June 23, 2005 7:15 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Name/Token Services in COBOL? IEANTC
Since the z/OS 1.4
Since the z/OS 1.4 level of C does not support AMODE 64 program calls, it
is basically irrelvant what IEANTC contains, since (again, just what I'm
told) the only ramification is an informational message (the compile
completes with completion code 0).
So while it is not ideal, it is not causing any
Thanks, Peter, and yes, we are at z/OS 1.4. So why should 1.4 SAMPLIB have
1.5-only content? Is that still APAR-able?
Peter
-Original Message-
From: Peter Relson [mailto:[EMAIL PROTECTED]
Sent: Wednesday, June 22, 2005 12:42 PM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Name/Token Services
>They are each declared as "linkage(...,OS64_NOSTACK)", but OS64_NOSTACK is
>not a supported option of "#pragma linkage" in the IBM C compiler.
I am told that the z/OS 1.5 compiler and up should support this option
(though perhaps it is not properly documented).
If true, that obviously doesn't do
Thanks, Walt. I'll try that route.
Peter
-Original Message-
From: Walt Farrell [mailto:[EMAIL PROTECTED]
Sent: Wednesday, June 22, 2005 8:24 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Name/Token Services in COBOL?
Members of SYS1.SAMPLIB are (in my experience) generally APARabl
On 6/21/2005 5:29 PM, Farley, Peter x23353 wrote:
Not to mention the other language copy members there (IEANTC, IEANTPAS,
IEANTPLI).
BTW, the "C" language sample in SYS1.SAMPLIB(IEANTC) has an error in the
#pragma's for the 64-bit IEANT functions:
They are each declared as "linkage(...,OS64_NOS
On Jun 21, 2005, at 4:29 PM, Farley, Peter x23353 wrote:
Not to mention the other language copy members there (IEANTC, IEANTPAS,
IEANTPLI).
BTW, the "C" language sample in SYS1.SAMPLIB(IEANTC) has an error in
the
#pragma's for the 64-bit IEANT functions:
They are each declared as "linkage(..
m doing any
64-bit work in C yet, I just want the compiler warning eliminated.
Peter
-Original Message-
From: Peter Relson [mailto:[EMAIL PROTECTED]
Sent: Tuesday, June 21, 2005 7:43 AM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: Name/Token Services in COBOL?
I was surprised that no one mentioned
On 6/21/2005 7:39 AM, Peter Relson wrote:
I was surprised that no one mentioned that interface definition file
IEANTCOB is provided precisely for the purpose of allowing a COBOL program
to call the name/token services.
It would probably help if IEANTCOB (and the other samples) were
mentioned i
Peter Relson wrote:
I was surprised that no one mentioned that interface definition file
IEANTCOB is provided precisely for the purpose of allowing a COBOL program
to call the name/token services.
Peter Relson
z/OS Core Technology Design
Why would you be surprised no one mentioned it? I've be
In a message dated 6/21/2005 6:39:48 A.M. Central Daylight Time,
[EMAIL PROTECTED] writes:
I was surprised that no one mentioned that interface definition file
IEANTCOB is provided precisely for the purpose of allowing a COBOL program
to call the name/token services
Maybe not very many IB
I was surprised that no one mentioned that interface definition file
IEANTCOB is provided precisely for the purpose of allowing a COBOL program
to call the name/token services.
Peter Relson
z/OS Core Technology Design
--
For IBM-
Go to Fill-Table-InCore-Ende
end-if
Roland
-Ursprüngliche Nachricht-
Von: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] Im Auftrag von
McKown, John
Gesendet: Montag, 20. Juni 2005 20:52
An: IBM-MAI
> BUT ...
>
> It seems that name/token services are for authorized
> assembler programs. That might be a show stopper.
>
> Still, can't hurt to try. Let us know how it works out.
No. You have to be authorized to create or delete a global name/token
pair, or to delete a name/token pair that was c
Steve Comstock wrote:
BUT ...
It seems that name/token services are for authorized
assembler programs. That might be a show stopper.
No. A subset of the services can be used by unauthorized code.
--
-
| Edward E. Jaffe
> It seems that name/token services are for authorized
> assembler programs. That might be a show stopper.
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/IEA2A940/11.1.1?S
HELF=IEA2BK41&DT=20040109131857&CASE=
--8http://bama.ua.edu/archives/ibm-main.html
McKown, John wrote:
-Original Message-
From: IBM Mainframe Discussion List
[mailto:[EMAIL PROTECTED] On Behalf Of Bill Klein
Sent: Monday, June 20, 2005 2:36 PM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Fw: Name/Token Services in COBOL?
John,
*IF* you know how to do this in HLASM, (using a
> -Original Message-
> From: IBM Mainframe Discussion List
> [mailto:[EMAIL PROTECTED] On Behalf Of Bill Klein
> Sent: Monday, June 20, 2005 2:36 PM
> To: IBM-MAIN@BAMA.UA.EDU
> Subject: Fw: Name/Token Services in COBOL?
>
>
> John,
> *IF* you know how
ED]>...
> Is there any way to use Name/Token services in COBOL without recourse to
> an HLASM subroutine? I don't see anything in the HLL Services or in LE.
> This is on z/OS 1.4 and Enterprise COBOL 3.3.1.
>
> What I have is a user exit in a vendor product, ViewDirect from Mo
Is there any way to use Name/Token services in COBOL without recourse to
an HLASM subroutine? I don't see anything in the HLL Services or in LE.
This is on z/OS 1.4 and Enterprise COBOL 3.3.1.
What I have is a user exit in a vendor product, ViewDirect from Mobius.
I need to keep some data be
24 matches
Mail list logo
