+- HP Forums (https://www.hpmuseum.org/forum)
+-- Forum: HP Calculators (and very old HP Computers) (/forum-3.html)
+--- Forum: General Forum (/forum-4.html)
+--- Thread: New machine language port (/thread-14203.html)
Pages: 1 2
RE: New machine language port - Joe Horn - 12-27-2019 06:43 AM
(12-27-2019 12:19 AM)cahlucas Wrote: I can compile everything, except for a part of the program. And strangely enough, it is a sys-rpl statement, the 'MDIMS' command, which can be found on page 90 of the book "An Introduction to HP 48 System RPL and Assembly Language Programming" by James Donnelly, March 1, 2009. It gives a "Can't Find" error message. It is in a part of the program outside the 'CODE' and 'ENDCODE' statements, in a part with only sys-rpl. So apparently this is a valid instruction that is not recognized. Hopefully this can be solved. Sincerely, Karel.
On the 50g, replace "MDIMS" with "FPTR2 ^MDIMS" (be sure to include the space after the 2, and include the ^ character). It *might* work if you omit the "^" character, but if not, try including it. Several commands in HP 48 System RPL got replaced in the 50g with "Flash Pointers". Their names are now prefixed by a "^" character, and they are properly invoked in System RPL with the FPTR2 command. Hope that works for you!
RE: New machine language port - cahlucas - 12-27-2019 09:22 PM
Dear Mr. Horne,
I have now been able to compile / assemble the entire program. Unfortunately the program does not work. It seems that certain techniques that work on the HP-48 do not work on the HP-50. That's why I need a good machine language debugger for both the HP-48 and the HP-50. The problem here is that some pieces of asm code are embedded between code of sys-RPL. What is the best way to proceed? Are there certain procedures for porting from one calculator to another? Sincerely, Karel.
RE: New machine language port - Joe Horn - 12-27-2019 11:20 PM
(12-27-2019 09:22 PM)cahlucas Wrote: ... Unfortunately the program does not work. ...
That's not enough information for anybody to give you any useful help.
(12-27-2019 09:22 PM)cahlucas Wrote: What is the best way to proceed?
The only way that's reliable is to understand HP 48 System RPL and Saturn Assembly Language, *and* HP 50g System RPL and its Simulated Saturn Assembly Language, and the differences between them. Achieving that understanding takes a lot of time. Do any members here know any faster yet reliable methods of porting mixtures of SysRPL & Code objects from HP 48 to HP 50g? If so, please share them. Thanks in advance.
RE: New machine language port - cahlucas - 12-28-2019 02:47 AM
(12-27-2019 11:20 PM)Joe Horn Wrote:(12-27-2019 09:22 PM)cahlucas Wrote: ... Unfortunately the program does not work. ...
That's not enough information for anybody to give you any useful help.
(12-27-2019 09:22 PM)cahlucas Wrote: What is the best way to proceed?
The only way that's reliable is to understand HP 48 System RPL and Saturn Assembly Language, *and* HP 50g System RPL and its Simulated Saturn Assembly Language, and the differences between them. Achieving that understanding takes a lot of time. Do any members here know any faster yet reliable methods of porting mixtures of SysRPL & Code objects from HP 48 to HP 50g? If so, please share them. Thanks in advance.
Dear Mr. Horn,
It looks like the program on the HP-50g hangs in an infinite loop. On the HP-48, the program is completed in seconds. At the moment I can say no more about it. Maybe you can give me a few suggestions. In addition to the documentation that I already have via hpcalc.org, I am looking for other forms of information about the saturn assembly language. Maybe you can also recommend a good machine language debugger for both the HP-48 and the HP-50. Sincerely, Karel.
RE: New machine language port - Joe Horn - 12-28-2019 03:25 AM
(12-28-2019 02:47 AM)cahlucas Wrote: At the moment I can say no more about it.
Well, then... same here. We might be able to say more if you give us a link to the source code. Without that, we can't possibly guess what is causing your program not to run, since there are a bazillion possible reasons.
Anybody here know of a Saturn code debugger for the 50g?
RE: New machine language port - DavidM - 12-28-2019 03:31 PM
(12-28-2019 03:25 AM)Joe Horn Wrote: Anybody here know of a Saturn code debugger for the 50g?
Debug4x allows debugging Saturn code for 48-50g targets. MASD on the 50g has some debugging features as well, but I prefer the Debug4x environment for this type of thing. Of course, that's yet another compiler with a different syntax expectation than MASD, so more translation would be needed.
@cahlucas: As Joe has said, placing the source code somewhere that we can access will make it easier to assist, though if the program is as big as it sounds, it may be more of an undertaking that most would want to take on.
Besides the actual syntactical issues, the O/S environment as a whole is different between the platforms. In particular, ROM routines and data constants are sometimes in entirely different locations (if they even still exist). This means that Saturn code written for one platform won't necessarily run successfully on the other.
What is this program/suite supposed to do? Are you certain that there's no other programs available that might provide a starting point you could build upon?
RE: New machine language port - cahlucas - 12-29-2019 04:23 AM
(12-28-2019 03:25 AM)Joe Horn Wrote:(12-28-2019 02:47 AM)cahlucas Wrote: At the moment I can say no more about it.
Well, then... same here. We might be able to say more if you give us a link to the source code. Without that, we can't possibly guess what is causing your program not to run, since there are a bazillion possible reasons.
Anybody here know of a Saturn code debugger for the 50g?
Dear Mr. Horn,
I believe that I have finished translating the source code. Unfortunately I cannot upload the directory with the updated files, because this site refuses it. I don't know how I can send you the files. You can find the original program via the link at the bottom of this quote. I unpacked this library with LIBEX V2.4G (also on hpcalc.org) and transferred it to my HP50g via my computer. I would like to receive a response from you about how to proceed now. Sincerely, Karel.
https://www.hpcalc.org/details/8500
RE: New machine language port - cahlucas - 12-29-2019 04:26 AM
(12-28-2019 03:31 PM)DavidM Wrote: What is this program/suite supposed to do? Are you certain that there's no other programs available that might provide a starting point you could build upon?
Dear Mr. David M.,
You can find the description via the link below. Sincerely, Karel.
https://www.hpcalc.org/details/8500
RE: New machine language port - Mark Hardman - 12-29-2019 05:34 AM
An implementation of the Quine-McCluskey method is already available for the HP50g:
Digitician 1.1
![[Image: digitician49.png]](https://www.hpcalc.org/hp49/science/electricity/sshots/digitician49.png)
RE: New machine language port - ijabbott - 12-29-2019 06:46 AM
(12-29-2019 05:34 AM)Mark Hardman Wrote: An implementation of the Quine-McCluskey method is already available for the HP50g:
Digitician 1.1
And there is a different one, written in C:
QuineHp
![[Image: quinehp.png]](https://www.hpcalc.org/hp49/science/misc/sshots/quinehp.png)
RE: New machine language port - Joe Horn - 12-29-2019 06:48 AM
(12-29-2019 04:23 AM)cahlucas Wrote: I believe that I have finished translating the source code. Unfortunately I cannot upload the directory with the updated files, because this site refuses it. I don't know how I can send you the files. You can find the original program via the link at the bottom of this quote. I unpacked this library with LIBEX V2.4G (also on hpcalc.org) and transferred it to my HP50g via my computer. I would like to receive a response from you about how to proceed now.
https://www.hpcalc.org/details/8500
Since that zipfile doesn't include commented source code, we are at an impasse. Sorry, but I can help you no further.
RE: New machine language port - David Hayden - 12-29-2019 03:14 PM
As a start, go through the source code and make sure that any ROM or O/S data that it's accessing are part of the supported entries. If not then you'll need to figure where they are on the 50g, or how to access them in a supported manner.
RE: New machine language port - cahlucas - 12-29-2019 09:01 PM
(12-29-2019 05:34 AM)Mark Hardman Wrote: An implementation of the Quine-McCluskey method is already available for the HP50g:
Digitician 1.1
Dear Mr. Hardman,
I am familiar with the program 'Digitician 1.1'. The last time I used it, the program crashed. The advantages of the program that I want to port are the speed and that the solutions it gave back were always correct. Sincerely, Karel.
RE: New machine language port - cahlucas - 12-29-2019 09:10 PM
(12-29-2019 06:46 AM)ijabbott Wrote:(12-29-2019 05:34 AM)Mark Hardman Wrote: An implementation of the Quine-McCluskey method is already available for the HP50g:
Digitician 1.1
And there is a different one, written in C:
QuineHp
Dear Mr. Ijabbott,
I know that program too. I tested it, and it appears that it gives good results, as long as no "don't cares" are used. This means that the program does not always give the correct results. And that's why I don't use it. Sincerely, Karel.
RE: New machine language port - cahlucas - 12-29-2019 09:17 PM
(12-29-2019 06:48 AM)Joe Horn Wrote:(12-29-2019 04:23 AM)cahlucas Wrote: I believe that I have finished translating the source code. Unfortunately I cannot upload the directory with the updated files, because this site refuses it. I don't know how I can send you the files. You can find the original program via the link at the bottom of this quote. I unpacked this library with LIBEX V2.4G (also on hpcalc.org) and transferred it to my HP50g via my computer. I would like to receive a response from you about how to proceed now.
https://www.hpcalc.org/details/8500
Since that zipfile doesn't include commented source code, we are at an impasse. Sorry, but I can help you no further.
Dear Mr. Horn,
Perhaps there is another option to get the ported program to you if you are interested. Do you perhaps know a service on the internet except dropbox where we can share files? I hope you have a suggestion. Sincerely, Karel.
RE: New machine language port - cahlucas - 12-29-2019 09:19 PM
(12-29-2019 03:14 PM)David Hayden Wrote: As a start, go through the source code and make sure that any ROM or O/S data that it's accessing are part of the supported entries. If not then you'll need to figure where they are on the 50g, or how to access them in a supported manner.
Dear Mr.Hayden,
Can you give me a hint where I should start, or where I can get that information? Sincerely, Karel.