August Developer meeting notes

10 views

Skip to first unread message

Abe Kornelis

unread,

Aug 16, 2026, 1:52:01 PMAug 16

to z390dev...@googlegroups.com, z390

All,

last week we had our monthly developer meeting.
John Ganci, Larry Simmons, and Abe Kornelis were present.

On Cobol there was little news. Abe told that Zane
seems to be working on the two incomplete Pull
Requests for zCobol. With Zane being absent we had
no opportunity to further discuss.

Abe mentioned he is still working to get all existing
test programs hooked up into the groovy test
framework, plus the bat and bash equivalents.

John talked about his efforts to add 64-bit support
to tri-modal instructions that were originally
implemented as bi-modal instructions.

The process of establishing a linter to standardize
our coding style for java programs is well underway.
The linter has been defined, but cannot be activated
until after all existing programs have been made
to comply. John shared his fixed-up version of az390.
Abe has addressed linter-flagged issues in most
other programs. Three remain, but these are large.

We discussed the planning of these changes.
The 64-bit work that John is doing will affect
pz390 and sz390. Other changes are pending on
z390. Normally, conflicted source versions due to
parallel changes are not a big deal. The conflicts
are marked by github and can be resolved
manually. However, the linter-related changes
are pervasive, affecting a very large percentage
of all code lines in the program. A code conflict
on so many lines is impractical and must be avoided.

John indicated that the modified version of the z390
module holds changes to address a weird bug in Linux,
where resizing the z390 gui window causes unwanted
side effects. Abe promised to test whether this symptom
(disappearing text fragments) can be reproduced on
Windows System for Linux.

John indicated he's hoping to have his 64-bit changes
ready by January, at the earliest.
Abe has urged John to find a way to slice and dice
the changes so they can be delivered in smaller increments.

Then we discussed the java version we are using.
The z390 java code is still being compiled with the Java V8
compiler. We have been looking at upgrading to either
V21 or V25. We all felt we did not have enough
knowledge/information to make a decision.
So we decided to postpone until later.

Finally, Abe mentioned that zVSAM V2 design is
progressing steadily. The aim for now is to re-integrate
the two design versions into a single version.
Then to upgrade the design in order to be able to
implement it.

Next meeting is planned for Wednesday, September 9th.
Thursday the tenth for early latitudes.