On 2015-02-24 10:49, Beesley, Paul wrote:
Hello
Our customer has been trialling Enterprise PL/1 4.4, with the intention of
upgrading from their current compiler, PL/1 for MVS&VM. They have issues with
performance, particularly with CICS programs, where the CPU time of a
transaction is 3 times what it was with PL/1 for MVS&VM. We have a PMR open
with IBM, who suggested the NORENT option, which did reduce the CPU to just
double the old value. IBM are suggesting experimenting with these options,
below. Does anybody have any experience of these and which can have most
impact?
ARCH BIFPREC(31) DEFAULT(NONASGN) DEFAULT(CONNECTED) DEFAULT(REORDER)
DEFAULT(NOOVERLAP) OPTIMIZE(2)/OPTIMIZE(3) REDUCE NORENT RULES(NOLAXCTL) . .
.
You need to show us *all* compiler options, *and* all the LE settings. The,
IIRC, LE STORAGE() settings can have a huge impact.
As for generated code, on 16 February I filed this RFE:
https://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe&CR_ID=66462
Robert
--
Robert AH Prins
robert(a)prino(d)org
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
