We've recently shipped some little enhancements which might be
of interest.  See the APAR descriptions via the URL links for
full details.

APAR PH03536 - OPTABLE and MACHINE enhancements

See: http://www.ibm.com/support/docview.wss?uid=isg1PH03536

- To make it easier to associate OPTABLE values with IBM Z
  machine generations, you can now use OPTABLE values Z9 to Z14
  instead of ZS3 through ZS8.

- Additional MACHINE alias names have been defined, for example
  z196 and ARCH-0 to ARCH-12, matching the compiler convention.

- MACHINE can now be specified on *PROCESS and ACONTROL, having
  the same effect as the corresponding OPTABLE option.

APAR PH06374 - &SYS_HLASM_DATE containing HLASM PTF build date

See: http://www.ibm.com/support/docview.wss?uid=isg1PH06374

- There is a new symbol &SYS_HLASM_DATE which contains the date
  HLASM was built, in the form yyyymmdd, which is usually the
  date the latest PTF was built.  This makes it possible to
  determine whether a fix is present by checking whether the
  value is greater than or equal to the date of the
  corresponding entry in the INFO report.

- To avoid errors when referencing any new symbol on older
  levels of HLASM, it is possible to define a created SET symbol
  of the same name, which will be referenced if the system
  symbol is not defined, for example as follows:

  &(SYS_HLASM_DATE) SETC '00000000' Default low value
           AIF   ('&SYS_HLASM_DATE' LT '20181231').OLD

  This method is also valid for all previous releases of HLASM.
  It works because a created SET symbol reference is always
  taken to refer to a SET symbol, not to a system variable
  symbol or macro parameter.

APAR PH06572 - Truncation checks for DC constants

See: http://www.ibm.com/support/docview.wss?uid=isg1PH06572

- This introduces three new options FLAG(TRUNC), FLAG(LONGER)
  and FLAG(SIGNED) which provide optional warnings for cases
  where the nominal value for a DC constant may have been
  changed by truncation.

We are hoping to find a way to extend these warnings to cover
immediate values in instructions, but the existing rules are
very different for signed and unsigned immediate operands,
making it difficult to come up with a neat general solution
that covers both.

Jonathan Scott
HLASM Development
IBM Hursley, UK

Reply via email to