HP Logic Analyzer Inverse Assemblers

[cmax100000]

I looked at the info in that link and it is close to what I am looking for as I can see the "RC" in a couple of the files from the screen shots.  If I had to pick an IA I would like, 6809 or any of the 6800 series as I am familiar with Motorola assembly in that series.  In the absence of that, any IA will do for what I am trying to find out as it is the configuration learn string that I am messing around with.  It seems like the 1630 could be set up by hand if you had enough patience and experience.
Any help on this is greatly appreciated.

[AndersG]

OK. Did you notice the tool i built, for uploading and downloading screenshots? See a bit down on page: http://www.dalton.ax/gpib/

[cmax100000]

I didn't see that in the first one but I have now downloaded it and looked at it.  I have a very old setup using a 486 computer running DOS and VBDOS and the utility looks windows based.  All of my gpib access is done through VBDOS.  I have been able to read the learn strings to the computer and break them down somewhat into pods, polarities, clocks, state and timing channels, etc.  As I was doing this and reading the IA posts about the difficulty in getting original files read from a hp gpib drive loaded into a 1630 machine, I thought that maybe the machine was just being set up a certain way which would be readable, savable, and writable as a learn string configuration file.  I was hoping someone on the forum would have a learn string configuration file output from any loaded inverse assembler on a 1630X machine so that I could look and see exactly what settings were being modified so that the assembly language(or machine language) states were captured.  In a perfect world it would be on a 1630D or lower machine as I have byte by byte comparison code that shows the differences between two learn strings.
Hopefully, someone might have something close.
Thanks in advance

[ahakman]

Is the Agilent Logic Analysis Software 4.00.1006 cd / dvd image available anywhere? Keysight only has ver 3.67.1008 and 5.90.1110 ISOs available on their site.

5.90 is a PIA in many ways, and lots of people report it is much slower than ver 4 was.

My actual LA has 4.00.1006 on it, and I'd like to be able to install the same version on another box to look at the files offline without the LA (and it's annoying fans) running.

I have 3.67.1008 installed on a virtual machine that I was able to use to compile some IA's for my actual machine on, but if I open the save files from 4.00.1006 in 3.67.1008 on the virtual machine, it just crashes when it's opening the file.

Also if I export a CSV in 4.00 and try to import it into 3.67 (which you'd think would get around the problem of the binary file format), 3.67 complains about there being no license for import, and only imports the first few rows. WTF kind of stupidity is a "license for data import"?

[MarkMLl]

I didn't see that in the first one but I have now downloaded it and looked at it.  I have a very old setup using a 486 computer running DOS and VBDOS and the utility looks windows based.  All of my gpib access is done through VBDOS.  I have been able to read the learn strings to the computer and break them down somewhat into pods, polarities, clocks, state and timing channels, etc.  As I was doing this and reading the IA posts about the difficulty in getting original files read from a hp gpib drive loaded into a 1630 machine, I thought that maybe the machine was just being set up a certain way which would be readable, savable, and writable as a learn string configuration file.  I was hoping someone on the forum would have a learn string configuration file output from any loaded inverse assembler on a 1630X machine so that I could look and see exactly what settings were being modified so that the assembly language(or machine language) states were captured.  In a perfect world it would be on a 1630D or lower machine as I have byte by byte comparison code that shows the differences between two learn strings.
Hopefully, someone might have something close.
Thanks in advance

I've got code- so far unpublished- which uses a cheap GPIB interface to capture the state of an HP1630G to a .vcd file.

MarkMLl

[AndersG]

FWIW are the formats of the various 163X files documented in the programming manual.

[gotcha]

Hello,

I recently got a HP 1650A and HP 16500B. Currently, I'm focusing on the HP 1650A.

My initial target is to use an inverse assembler for the Z80.
I found various files, floppy disk images, but I'm still a bit confused.
For example, I found a Z80 related floppy for the 1650A here (http://www.bitsavers.org/bits/HP/1650/hp_1650), but it doesn't tell if it's for working with a preprocessor interface or not, and the details of STAT.

In my particular case, I don't have any preprocessor interface for the Z80 (that as far as I understood, processes the control signal from the Z80).
I just have standard flying probes.

Is there a config file and a IA file I can use for this case ? Also, which pins should I use for STAT ?

[MarkMLl]

FWIW are the formats of the various 163X files documented in the programming manual.

Yes, to the extent that you can extract enough configuration information to convert the file into e.g. a VCD. I've not, however, attempted to restore a configuration to an instrument.

MarkMLl

[gslick]

In my particular case, I don't have any preprocessor interface for the Z80 (that as far as I understood, processes the control signal from the Z80).
I just have standard flying probes.

Is there a config file and a IA file I can use for this case ? Also, which pins should I use for STAT ?

As mentioned here in this other thread, the original Z80 IA won't work correctly without the Z80 preprocessor:

https://www.eevblog.com/forum/testgear/older-logic-analyzer-question/msg4872758/#msg4872758

[gslick]

There were some posts in a different forum which referred to this thread with questions about building a 10391B style inverse assembler into a 32-bit x86 DLL version of the inverse assembler using the version 03.60.0002 of the IAAnalysisAddInWizard from the 03.67.1008 version of the LA software CD-ROM.

One question was whether such a 32-bit x86 DLL version of the inverse assembler was also compatible with the newer version 05.90.1104 of the logic analyzer application. The answer to that question is yes. Such an inverse assembler can be loaded on the newer version 05.90.1104 of the logic analyzer application, when the logic analyzer application is loaded on either 32-bit or 64-bit versions of Windows.

The screenshot below shows the 32-bit x86 DLL version of the inverse assembler built from the i6800_p.r contained in the invasm_v3.zip file in the first post of this thread loaded in the 05.90.1104 version of the logic analyzer application running on the 32-bit version of Windows XP. I was also able to load it in the 05.90.1104 version of the logic analyzer application running on the 64-bit version of Windows 10.

I could post the 32-bit x86 DLL version of the i6800_p inverse assembler files here if anyone want those.

[Hamster]

i would be interested, i guess i need to build a virtual machine so i can build the z80, 6809, 68000 dlls. ( i sourced a 6809 interface )

thanks for sharing!

[gslick]

i would be interested, i guess i need to build a virtual machine so i can build the z80, 6809, 68000 dlls. ( i sourced a 6809 interface )

thanks for sharing!

Attached is the 32-bit x86 DLL version of the i6800_p inverse assembler file that I built using the version 03.60.0002 of the IAAnalysisAddInWizard from the 03.67.1008 version of the LA software CD-ROM.

There is a 32-bit install directory and a 64-bit install directory. The contents of the 32-bit install directory is what is created by running the create.bat file in the build directory of the IA DLL after the IA DLL is built.

The copies of the files in the 64-bit install directory are exactly the same as the files in the 32-bit install directory, except that in the 64-bit install directory the Install.bat and Uninstall.bat files were edited be me to replace all instances of "Program Files" with "Program Files (x86)", and in the 64-bit install directory the ialAnalysisAddInIAL6800_P.reg file was edited by me to replace all instances of "HKEY_CLASSES_ROOT" with "HKEY_CLASSES_ROOT\Wow6432Node".

You probably need to run Install.bat as Administrator to copy the IA DLL to the "Program Files" or "Program Files (x86)", and add the settings to the registry. After running either the 32-bit or 64-bit Install.bat, as appropriate, to install the IA DLL, the sample configuration file IAL6800_P-16911A.xml that I created could be loaded into the logic analyzer application to set up the appropriate bus signal labels.

If anyone gives this a try with a 6800 or 6802, let me know if it works for you.

Since I'm already setup to build these IA DLLs, I could probably go ahead and build Z80, 6809, 68000 DLLs for you. Do you have preprocessor interfaces for all of those?

[Hamster]

I do have the following Interface Modules:
Z80,8086/8088,6809,68000,Z8001,80386
Currently looking for a 6800/02

I could use the 6502 file as well ( this one doesn't use a Interface Module )

I did test your 64 bit install, couple steps needed to be added:

1) Open a Command Window in Administrator Mode.
2) Change to directory
3) Run Install.

[Hamster]

WooHoo!

Ok, that was a pita to make the 64 bit install files!

gslick, thanks for the tips.. Now to test.

it will be nice to use my bench top 1690D and can retire these old 167X units.

[Hamster]

FYI, it if wasn't for your example 6800 install, i probably would of never figured it out.. 

[gslick]

WooHoo!

Ok, that was a pita to make the 64 bit install files!

gslick, thanks for the tips.. Now to test.

it will be nice to use my bench top 1690D and can retire these old 167X units.

Were you able to get the IA DLL build without making any changes to the AnalysisAddIn_vcproj file? When I have tried this a few times with the original AnalysisAddIn_vcproj file (the generated one that you rename to replace the other generated AnalysisAddIn_vcproj file) I have gotten build errors related to agAnalysisAddIn3_i.idl

I have had to add agAnalysisAddIn3_i.idl; to the end of the AdditionalDependencies="agAnalysisAddIn_i.idl;agAnalysisAddIn2_i.idl;" lines to resolve that.

I think someone else here said that the didn't have to do that and didn't get the agAnalysisAddIn3_i.idl errors. I'm not sure what might be going on there.


What host system are you using for the 1690AD? I forget if you can use anything beyond Windows 7 to get the correct 1394 host controller driver loaded. I also forget if it works with both 32-bit and 64-bit versions of Windows 7. A 1690AD with a small and fast somewhat modern PC can be a much better solution than an equivalent slow Windows XP based 1680AD with a small LCD, unless you really need a completely portable all in one analyzer.

[Hamster]

Yes, I had to add agAnalysisAddIn3_i.idl; to the proj file.

On the 1690D:

Windows 10 Pro, 64bit , You need to install the compatibility mode drivers on the 1394 card.

I am pretty sure i am running a PCIe card as well, but need to look in the computer to check.

https://www.startech.com/en-us/faq/firewire-cards-windows-legacy-driver-swap

[Hamster]

i also converted all my 1680D frames to win7 machines core2duo.. super fast and portable

[gslick]

Yes, I had to add agAnalysisAddIn3_i.idl; to the proj file.

On the 1690D:

Windows 10 Pro, 64bit , You need to install the compatibility mode drivers on the 1394 card.

Thanks for confirming that you were also seeing the agAnalysisAddIn3_i.idl issue that required a modification of the project file. I wasn't sure if I might have messed up and skipped something in the installation the would have normally prevented that.

Also thanks for confirming that you were able to get your 1690AD working with a Windows 10 host. I'll have to try that myself the next time I get out my 1690AD. I have Dell SFF PC that would make a good Windows 10 host for that.

----------

If anyone is looking for HP preprocessors that attach to 10269 interfaces, this seller on eBay currently has a few listed, although a close look at all of the listing photos is necessary as some of the DIP / PGA target probes are in bad condition. Normally there are a couple of layers of machine pin DIP / PGA sockets on the target probes, so if only the outermost layer has broken pins, the socket layers below the outermost one might be fine and the outmost socket could be removed and replaced.

HP HEWLETT PACKARD 64675 Interface Modules A and B for 68020 Microprocessors
https://www.ebay.com/itm/256411931727

HP HEWLETT PACKARD 10314D Interface Module for 80386 Microprocessors
https://www.ebay.com/itm/256411931693

HP HEWLETT PACKARD 64680A Interface Module for Z8001 Microprocessors
https://www.ebay.com/itm/256410735436

HP HEWLETT PACKARD 64657B Interface Module for 80286 Microprocessors
https://www.ebay.com/itm/266671416022

HP HEWLETT PACKARD E2409A Interface Module for 80286 Microprocessors
https://www.ebay.com/itm/266671415923

HP HEWLETT PACKARD 64670A Interface Module for 68000 Microprocessors
https://www.ebay.com/itm/266671415821

HP HEWLETT PACKARD 64681A Interface Module for Z8002 Microprocessors
https://www.ebay.com/itm/266663873168

HP HEWLETT PACKARD 64658A Interface Module for 80186/80188 Microprocessors
https://www.ebay.com/itm/256405092893

HP HEWLETT PACKARD 64673A Interface Module for 68008 Microprocessors
https://www.ebay.com/itm/266659715690

HP HEWLETT PACKARD 10314B Interface Module for 80386 Microprocessors
https://www.ebay.com/itm/256403930507

HP HEWLETT PACKARD 64655A Interface Module for 8085 Microprocessors
https://www.ebay.com/itm/266659714425

HP HEWLETT PACKARD 10343B SCSI Bus Preprocessor
https://www.ebay.com/itm/256406266462

[Hamster]

So, now need to make a nice cheat sheet for pod assignments on the 10269C

cXXXXX_p.txt will indicate the ADDR/DATA assignments, but not which pod goes where.

ie, what are POD1/2/3/4/5 correlate too for ADDR/DATA/STAT

ie, the 80186 uses

Label  0: ADDR     (20 bits):  A3: ........ ....****  A2: ******** ******** 
Label  1: DATA     (16 bits):  A1: ******** ******** 
Label  2: STAT     (10 bits):  A5: .....*** ........  A3: *****.** ........ 
Label  3: SIZE     ( 3 bits):  A3: *..*.... ........  A2: ........ .......* 
Label  4: S6       ( 1 bits):  A3: ..*..... ........ 
Label  5: COPROC   ( 9 bits):  A5: .....*** ........  A3: ***.*.** ........ 
Label  6: DMA CH   ( 6 bits):  A5: .....*** ........  A3: ***..... ........ 

Which is 65 bits, but doesn't indicate Which bits are were on POD1/2/3/4/5

the good thing is if your ADDR is 16 bits, but the IA needs 24 Bits, it will tell you !

Are we to assume A1 -> POD1 , A2 -> POD2, A3 -> POD3  ?

[gslick]

I built 32-bit x86 DLL versions of inverse assemblers from a subset of the .R files contained in the invasm_v3.zip file in the first post of this thread.

This involved a lot of tedious manual cut-n-pasting and editing so I wouldn't be surprised if there might be an error or two, and I haven't tried going through and loading all of them yet, but I'll go ahead and post this set now. If I find any errors, I will post updates with corrections. If anyone finds any errors before I do, let me know. Also, I haven't started creating sample configuration files for these yet.

IAL68000_I
IAL68000_P
IAL68008_I
IAL68008_P
IAL6800_I
IAL6800_P
IAL68010_I
IAL68010_P
IAL6809E_P
IAL6809_I
IAL6809_P
IAL8085_IP
IAL8086_I
IAL8088_I
IALNSC800_I
IALZ80_I

EDIT 3/12/2024:
I found one build error so far. The IAL68000_P project was built with the IAL6800_P .R file instead of the IAL68000_P .R file. I didn't catch that cut-n-paste error until I went through all of the IA files in this set and tried loading them one at a time into the logic analyzer application, then noticed that the Listing window for the IAL68000_P IA had the wrong IA column. I rebuilt the IAL68000_P project with the correct .R file. The attached IAL.ZIP file now has an hpAnalysisAddInIAL68000_P.dll file with a date of 03/12/2024.

[Hamster]

gslick, thanks! I will test them as i can, if i find errors, i can always rebuild or look into now that i can also make them, yeah, the vsproj template has some errors, and obviously lacks the 64 bit stuff, I was looking at it and was actually thinking of creating a real installed ( based on Inno Setup ) , it has a more powerful way of checking directories and seeing if things are installed 

i think windows' Program Files and Program Files (x86) was the dumbest ideas Microsoft ever came up with .

FYI , for reference:

Last letter of "P" means the IA file is for use with general purpose probing (e.g., flying leads).
Last letter of "I" means the IA file is for use with dedicated hardware interfaces (often called "preprocessors").

[MarkL]

So, now need to make a nice cheat sheet for pod assignments on the 10269C

cXXXXX_p.txt will indicate the ADDR/DATA assignments, but not which pod goes where.

ie, what are POD1/2/3/4/5 correlate too for ADDR/DATA/STAT

ie, the 80186 uses

Label  0: ADDR     (20 bits):  A3: ........ ....****  A2: ******** ******** 
Label  1: DATA     (16 bits):  A1: ******** ******** 
Label  2: STAT     (10 bits):  A5: .....*** ........  A3: *****.** ........ 
Label  3: SIZE     ( 3 bits):  A3: *..*.... ........  A2: ........ .......* 
Label  4: S6       ( 1 bits):  A3: ..*..... ........ 
Label  5: COPROC   ( 9 bits):  A5: .....*** ........  A3: ***.*.** ........ 
Label  6: DMA CH   ( 6 bits):  A5: .....*** ........  A3: ***..... ........ 

Which is 65 bits, but doesn't indicate Which bits are were on POD1/2/3/4/5

the good thing is if your ADDR is 16 bits, but the IA needs 24 Bits, it will tell you !

Are we to assume A1 -> POD1 , A2 -> POD2, A3 -> POD3  ?

I haven't tried the 80186 IA specifically, but it's been my experience with other processors that A1 -> POD1, etc., as you stated, and in the given bit positions.  Some examination of the labels or the disassembled code from gslick may be needed to determine the bit order of STAT and other encoded fields.  Fields like ADDR and DATA are MSB -> LSB, as you would expect.

EDIT: fix typo

[Hamster]

Yeah, I am thinking when i was having issues with my 1673 with the inverse assembler, it games me some weird messages about pod order, and i never got it to work correctly.

I have a bunch of systems that use these processors so it will be easy for me to check the setup / configs.. now i can actually use my LA for more fancy debugging with my Fluke System Trouble shooter.

I have a Professor Pac Arcade PCB Stack that has issues, and never been able to successfully figure out "where" its crashing out at when doing a ram test , as the fluke reports no issues.

It will fun stacking the IA + Fluke Pod to it

[Hamster]

also looking at the 80186 ( which is one i just picked out of thin air ) , seems Stat/CoProc/Dma Ch all use the same A5 Pod..  but no A4.. so POD4 shouldn't do anything on the IA module.