[Info-vax] VAX Macro to C conversion

[Andrew Shaw]

On Tuesday, June 25, 2019 at 7:07:42 AM UTC+10, John Reagan wrote:

> On Monday, June 24, 2019 at 1:48:07 AM UTC-4, Andrew Shaw wrote:
> 
> 
> You rang?  [If you mention my name three times, I have to appear.]
I only mentioned it once - you're running ahead of schedule !!!

> 
> I think you are asking two different questions:
I think that's probably an accurate read !

> 
> #1) I'm looking for a conversion tool due to possible performance issues and programmer knowledge issues, have you heard of XTRAN?
> 
I would modify this slightly to be: "I'm looking for a conversion tool due almost exclusively to programmer knowledge issues".

We have no performance issues (that I am aware of) today, even with the sub-second timing requirements that we have in many areas of our code. (That sub second is for new/changed data to travel across a continent of similar size to mainland USA, but with far less comms infrastructure) over the intywebs to update our in memory VMS hosts. This we manage to achieve. The concern I have is with our ability as a (mostly quite young) dev team to manage the Macro code which is unfamiliar to the vast majority of our team.


> #2) Does using VAX Macro-32 limit performance versus coding in C (or any other language)
I am actually pretty confident, as I hinted above, that we don't have any (that I have noticed yet) performance issue due to the fact that some of our code is MACRO. Again, being fairly new to this organisation I am obviously not intimate with the long history our system has had, but it did start life a couple of decades ago on a VAX of some description and since then has travelled the well worn path of VAX --> AXP --> IA64. Possibly onward to X86, that part of the future isn't clear yet, but I'm putting $5 down that says its gonna be on our radar sometime sooner than management are expecting !

>From your comments here it is pretty clear to me that lots of good work has been done and is continuing to be done in the MACRO32 space and I have no fear of our code getting unusable in the near future. I am quite sure it will run happily for years to come without needing to be touched.

The situation I am trying to alleviate is that we now have a healthy development pipeline of work due to some regulatory changes in our industry in recent years and the opening up of the market that mean that we are going to need to be able to make enhancements to this MACRO code for the forseeable future. I guess this thread was me thinking out loud and wondering if we can make our lives easier by moving this code to something easier to understand and maintain.

The correct feedback has already come here along the lines of "Andrew, you need an understanding of your current MACRO code in order to assess the effort and risk of porting it versus leaving it alone and maintaining it in situ." That is good feedback and is obviously work I need to do. Realistically I wouldn't expect any tool to simply pick up our code and spit out perfectly structured and maintainable C (or FORTRAN or something else). I think the best I can hope for is that *if* our MACRO is reasonably simple and not overly clever, then I am hoping it is conceivable that a tool *might* be able to get us say 60% of the way there and we then follow that up with refactoring and rigorous testing etc.
 
Thanks for your feedback John - Insightful as always and not at all inciteful :-)