The ASMDASM disassembler component of the HLASM Toolkit Feature
can reconstruct object modules from (a) load modules, (b) program
objects, (c) CMS modules, and (d) VSE phases. It is described in
the HLASM Toolkit User's Guide, GC26-8710, Chapter 3.
It has been available since December 1995, and
Does anyone have a copy of GPP-20 A High Level Macro Processor
Description - MACRO1? I'm willing to pay copying/mailing costs
if you'll send it to me.
John Ehrman
IBM Silicon Valley Lab
555 Bailey Avenue
San Jose, CA 95141
--
For
I posted the original question on behalf of a colleague who was
curious about customer views; I've been forwarding comments to him.
John Ehrman
(-- Referenced Note Follows )
Date:Mon, 22 Feb 2010 07:21:05 -0600
From:Barbara Nitz nitz-...@gmx.net
PDSEs have been available for a long time, and provide many
advantages over PDSs. Why are people reluctant to use PDSEs?
John Ehrman
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to
I've been trying to locate a copy of this publication; might
anyone have a copy? I'm willing to pay repro/mailing costs.
I checked with SHARE HQ (they might have taken over the GUIDE
HQ files), but they couldn't locate a copy.
Thanks... John Ehrman
The working ladies group in San Francisco used COYOTE as an
acronym for their motto: Call Off Your Old Tired Ethics.
John Ehrman
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to
CSECTs are restricted to 16MB only if you generate traditional OBJ
object files, because its length and address fields are 3 bytes long.
As Peter said, the total length in a single assembly will be less
than 16M.
*BUT* if you specify the NOTHREAD option you can have as many very
larget CSECTs as
Check the SHARE proceedings for presentations by David Bond titled
Coding Assembler for Performance. Excellent advice therein.
John Ehrman
(-- Referenced Note Follows )
From: Bill Klein wmkl...@ix.netcom.com
Date: Thu, 23 Apr 2009 14:28:36 -0500
...
The
I've heard rumors to the effect that the current (high) prices
were set erroneously, and that a correction is being considered.
John ehrman
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL
You might consider the string-search capabilities of SUPERC,
and use a REXX exec to set up the data set names.
John Ehrman
(-- Referenced Note Follows )
Sent: Thursday, April 10, 2008 3:38 AM
We have a requirement to search for a textstring i a lot of
I recommend you study the handout from SHARE session 8132 in the
SHARE proceedings from Long Beach, California in February 2004:
Extending the Life Cycle of Legacy Applications
(With Added Thoughts Specific to Assembler Language)
It contains a detailed analysis of factors to consider when you
Take a look at the ASMDASM in the HLASM Toolkit Feature. It was based
on Thornton's disassembler, but has been extensively enhanced over a
period of 10 years.
John Ehrman
(-- Referenced Note Follows )
Date:Tue, 11 Dec 2007 11:11:29 -0500
From:
Answers to this and many other questions about new instructions
will be provided by the lead z/architect in session 8190 at the
August SHARE conference.
John Ehrman
(-- Referenced Note Follows )
Date:Thu, 31 May 2007 17:48:18 +0300
From:Binyamin
The new updated to the z/Architecture Principles of Operation is now
available at
http://publibz.boulder.ibm.com/epubs/pdf/a2278325.pdf
John Ehrman
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send
Gerhard Postpischl is correct -- I had nothing to do with the
development of Wylbur. I was an enthusiastic user at SLAC, and
wrote a Wylbur Tutorial that got widespread use.
John Ehrman
--
For IBM-MAIN subscribe / signoff /
The Preliminary Decimal-Floating-Point Architecture documentation
is now available at
http://publibfi.boulder.ibm.com/epubs/pdf/a2322320.pdf
John Ehrman
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send
There's no need to do your own parsing -- as Steve Smith (I think
it was) suggested, check the AINSERT statement: it was designed
for exactly this kind of situation.
Also, the NOCOMPAT(SYSLIST) option sometimes helps for substituted
operands that are a list (like (A,B,C)), but won't help for
Re Robert Rosenberg's posting: All RLD items not resolved at
assembly time are listed in the Relocation Directory -- they
don't necessarily have to have different Position and Relocation
ESDIDs (for example, A(*) will have the same PID and RID).
Re Ed Gould's posting: I vaguely remember seeing a
There is one very simple reason there is no MVCY instruction:
the MVC instruction is 6 bytes long. Adding the two bytes needed
for long displacements would create an 8-byte instruction; the
zSeries (and ancestor) architecture(s) would require major
revisions to handle instructions longer than 3
)
Date:Tue, 31 Jan 2006 11:20:32 -0500
From:Shmuel Metz (Seymour J.) [EMAIL PROTECTED]
In [EMAIL PROTECTED], on 01/30/2006
at 12:29 PM, John R. Ehrman [EMAIL PROTECTED] said:
I assumed that users like yourself would continue
to use the MVS or VM forms of HLASM
As John Eells noted, if you have z/VM and z/OS, you already have the
standard HLASM. I assumed that users like yourself would continue
to use the MVS or VM forms of HLASM.
John Ehrman
(-- Referenced Note Follows )
Date:Fri, 27 Jan 2006 14:18:04 -0600
I wasn't aware of the possibility of running zLinux on a regular
processor; I'll have to check with whether or not HLASM for zLinux
could be ordered for such a configuration as well as on an IFL.
John Ehrman
(-- Referenced Note Follows )
Date:Fri, 27 Jan
I've had several questions about ordering HLASM for zLinux; I hope
this information will help.
Contact your IBM marketing rep to order product 5799-TCQ (and for
annual support and Service, 5799-TCR). It requires some form of
Linux for zSeries or Linux/390, running on an IFL processor.
John
IBM recently announced the availability of HLASM for Linux on
zSeries, as PRPQ 5799-TCQ. It is fully compatible with HLASM
on the current MVS, CMS, and VSE operating systems, and can
generate the ELF object format as well as the traditional OBJ
and GOFF formats.
John Ehrman (ehrmanATvnet.ibm.com)
Kevin, specify the LIBMAC option when you do the assembly. It
will indicate the exact statement in the #TEST macro where the
error occurs. That may help isolate the problem.
John Ehrman
(-- Referenced Note Follows )
Date:Tue, 27 Dec 2005 08:47:45 -0600
I'm trying (under CMS) to test a program (A) with an internal label (L).
A does an IDENTIFY macro for L, and then loads a separate module M that
does a LOAD for L. (I want to provide M with functions in A.)
However, I always get return code X'14' from IDENTIFY (Entry name
duplicates the name
The HLASM Language Reference is a pretty big book already; while
I'll grant the value of well-chosen examples, adding more tutorial
information would make the book start to approach the PoP in size.
If you have suggestions about specific opacities, please send
comments via the Reader Comment
I received the following information about your questions from the
Binder architect, Leona Baumgart.
John Ehrman
A couple of related issues are touched on in this series of notes.
1. The default action of the binder is that a non-executable module
will NOT replace an executable one.
I gather you are COPYing a member from SYSLIB that invokes a macro
from SYSLIB? (Or is it a macro COPYing a member into the macro?)
If the COPY statement is in open code, you might want to try
capturing the SYSLIB_MEMBER info with statements inside the COPY
file just before it invokes the macro.
29 matches
Mail list logo