>From what I remember, I don't think anyone would get anything useful from the >ASMPUT source code, even for free. Think of it as being in some old trunk in >the attic which contains broken bits of many different things that should >probably have been thrown away.
In 2013 when we first looked at it, it hadn't been touched for about a decade, and it probably hadn't been actively developed for much longer except to add some new instructions. We couldn't make head or tail of it, and it appeared that any design documentation had presumably been lost long ago (an infuriatingly common problem), possibly before the migration from OS/2 to Windows. Much of the Windows "source code" was the output of a semi-automated porting process (presumably from Prolog), so in many ways it was more like generated code than source code and not very readable. We didn't even manage to work out which of the many files in the relevant repository were actually used in building the current version of ASMPUT, as it included code generation tools which we couldn't run. And it was of course the only HLASM or Toolkit program which did not run on the mainframe and was not coded in Assembler or PL/X, so it was not within our normal skill set. The HLASM team borrowed a Windows programmer for a few weeks to see if he could find a way to migrate it to a current Windows C++ compiler or perhaps to Java, but he didn't get very far. Of course, that's just my own impression (as I remember it anyway), but my feeling is that it's probably not even practical to do the work to make some form of the source available, and even if it is, the amount of work that would be needed to recreate the current level of ASMPUT from it would be disproportionate, and then it would need significant new function to handle newer IBM Z concepts. Normally if something is made "open source" you are starting from something which works and builds, but the owner doesn't want to continue to support it. In this case, it's barely working and can't be built. For its last few years IBM could only offer help with using it and could not actually change anything, which is why it had to be functionally stabilised. All of the other components of HLASM and the Toolkit are in Assembler or PL/X and were still being actively maintained (using VM/CMS as the primary development platform) when I retired just over a year ago. ASMPUT was a weird special case inherited from a different world. Jonathan Scott -----Original Message----- From: IBM Mainframe Assembler List <[email protected]> On Behalf Of Paul Gilmartin Sent: 18 March 2026 18:31 To: [email protected] Subject: Re: Whither ASMPUT? On 3/18/26 12:08, Jonathan Scott wrote: > gil wrote: >> This feels like an argument for opening the source; GPL. > We discussed the idea of open source on the mailing list back in 2021. As I > said back then, other IBM internal teams were given a copy of ASMPUT with a > view to incorporating similar capabilities into other tools, so I think IBM > might want to keep hold of the rights, even though I'm not personally aware > of any specific replacement tool. > .... <https://en.wikipedia.org/wiki/The_Dog_in_the_Manger> -- gil
