Has anyone else seen this?
Yesterday, I applied APAR PH39324 to my HLASM to bring in support for
Z16 machine instructions. I then did a full reassembly of z/XDC (200+
assemblies), and suddenly, out of the blue, 14 of them reported
ASMA500W errors on a DXD instruction that appears in ALL assemblies!
ASMA500W warns that the flagged statement requires a more restrictive
boundary alignment that the csect in which it occurs.
In our assemblies, that is nonsense...
- All my csects are assembled with the default doubleword alignment.
- The default alignment for external dummy sections also is doubleword.
- The DXD statement in question is defined as a
"(length)X'<whatever>'" field which, of course, is only byte aligned.
Further, the DXD is emitted from a macro of mine, which generates the
same way every time. That macro occurs in every one of my assemblies.
So if the error happened once, it should have happened in EVERY
assembly, not this "odd" selection of fourteen assemblies. They're
structured no differently from all the other assemblies. They have no
commonality that distinguishes then from the other 186. It's baffling.
We did some research and found APAR PH40885 which fixes "multiple
problems" with ASMA500W messaging. That raised our hopes until we saw
that it was already applied...
If you're using DXDs, have you seen this problem?
Dave Cole
President, ColeSoft
dbc...@gmail.com (personal)
dbc...@colesoft.com (business)
540-456-6518 (cell)
PS: Here are the links to these APARs:
PH39324:
<https://www.ibm.com/support/pages/apar/PH39324>https://www.ibm.com/support/pages/apar/PH39324
PH40885:
<https://www.ibm.com/support/pages/apar/PH40885?mhsrc=ibmsearch_a&mhq=DXD>https://www.ibm.com/support/pages/apar/PH40885
