daves

Thanks Tony. My problem was different but I hit this one today after doing the 1511 Threshold 2 update. (For search purposes) The message I got was xxx.ppproj(431,5): Error MSB4062: The "PPMAC460CompileTask.CompileTask" task could not be loaded from the assembly C:\Windows\Microsoft.NET\Framework\v3.5\PPMAC460CompileTask.dll. Could not load file or assembly 'file:///C:\Windows\Microsoft.NET\Framework\v3.5\PPMAC460CompileTask.dll' or one of its dependencies. The system cannot find the file specified. Confirm that the declaration is correct, and that the assembly and all its dependencies are available. As you say reinstalling the PPMAC update fixed it for me.

I've been getting these cygwin vfork errors again and have this time a full shutdown of Windows 10 fixed it. Maybe something is getting stuck in the Hybrid Shutdown.

My "C:\DeltaTau\Power PMAC IDE\compilers\usr\local\bin" folder contains "powerpc-405-linux-gnu-gcc.exe". From an install of IDE 2.0.3.40. I did have a problem once where that folder got corrupted and I found the compilers reside in "C:\DeltaTau\Power PMAC IDE\compilers.tar.gz" and there is a program which appears to repair them called "C:\DeltaTau\Power PMAC IDE\CompilerExtractProgress.exe". My system appears to be working after doing this but you might want an official DT response... Older IDEs installed in program files if memory serves...

Yeah, sorry I wasn't clear: The "$0/$32" and "$0/$68" was me describing my 2-axis and 4-axis cards. They are bot $0 on the 2-axis and $32,$68 on the 4-axis. I got the EEPROM programming information from Brad/Lucien but I don't think I actually want to change anything. I think my $68 option is just not in the IDE's recognised list. All the opt part IDs match the ordered board. I am waiting for a follow up email from Brad...

In case anyone else suffers this: I think it was due to SOMETHING waking my Windows 10 PC from Off/Hibernate(?) at 3AM and doing some update. There were some MySQL errors, and the compilers folder under the IDE had been corrupted (not sure how but the 465-2 compiler and powerpc-465-rootfs.tar.gz were missing/damaged). A Windows restart seemed to restart MySQL and copying the compilers folder from an IDE install on another PC fixed the missing APM86xxx libraries.

IDE 2.0.3.40 Spent a day building successfully to a connected 465-2. Turn off. The next day run with No Device and check if a C code change still builds I get: Build Process For librtiplcc.so has started. C:\ABD\PowerPMAC\SPMM\Phase2\Phase2\Phase2.ppproj(400,5): Error : 0 [main] make 4172 child_info_fork::abort: C:\DeltaTau\Power PMAC IDE\compilers\bin\cygiconv-2.dll: Loaded to different address: parent(0xC30000) != child(0xC60000) C:\ABD\PowerPMAC\SPMM\Phase2\Phase2\Phase2.ppproj(400,5): Error : make: vfork: Resource temporarily unavailable Build Process For librtiplcc.so has ended. I'm guessing the 465-2 is peculiar? I'm sure I could go 460->No Device and build fairly successfully... How do I get the offline build to work again? (I've tried Clean)

Thanks tecnico Brad actually emailed me off-forum about this and we are currently investigating these very parameters. My suspicion is that the specific variant of Acc24E3 we use is not supported fully. FYI my cards that don't work have (one phase 16-bit DAC with terminal connector) option type $68, which I think could be unusual... Gate3[2].PartNum = 604002 Gate3[2].PartRev = 12 Gate3[2].PartType = 1 Gate3[2].PartOpt1 = $32 Gate3[2].PartOpt2 = $68 Gate3[2].PartOpt5 = $0/$32 Gate3[2].PartOpt6 = $0/$68

IDE 2.0.3.40, FW 2.0.2.14 We have just noticed my latest customer machine and my test rack don't pick up all the gate 3 cards fitted. The terminal sees them all in the 'intellisense' thing. For example I have two acc24e3 cards fitted (correctly DIP switched). The attachment shows them appearing in the terminal but not the system setup. The other machine has different cards missing. I think the IDE used to work OK (I have a 1.7 IDE on another rack looking OK). I have tried all the usual $$$***, save, repower, etc. Is there something I can check to see if it is the cards or is this a known IDE thing? Cheers Dave

I am seeing it on Windows 7 Professional SP1 and Windows 10 Pro. I can't pin down what sequence of events causes it repeatably. It is possibly more likely after a longer session of downloads/edits/saves/resets but not sure... Thanks

Hi on certain PCs we are seeing the IDE crash when you exit. Windows decides to restart the application. I tried to attach a Windows event but not allowed so the text is below What is happening? How can I stop it? It is happening on a customer PC as well as some developer PCs. Faulting application name: PowerPmacSuite.exe, version: 2.0.3.40, time stamp: 0x5536ed85 Faulting module name: AppenvStub.dll_unloaded, version: 0.0.0.0, time stamp: 0x488f172a Exception code: 0xc0000005 Fault offset: 0x732ebd89 Faulting process ID: 0x704 Faulting application start time: 0x01d1067f0f7e2470 Faulting application path: C:\DeltaTau\Power PMAC IDE\PowerPmacSuite.exe Faulting module path: AppenvStub.dll Report ID: b58cc714-7272-11e5-8472-6451063d5a60 Faulting package full name: %14 Faulting package-relative application ID: %15

2.0.3.40

Anyone written or know of a language grammar plugin for Atom or Visual Studio Code supporting pmac script? I started writing a VS Code one but it was too hard / time consuming for me. Perhaps I will persevere if anyone wants them or can help...

I have turned to the Dark Side! Almost. Can anyone tell me the setting to get numbers with a decimal point in script to take my colour choice? It seems to be OK in C. Or have DT missed a tag somewhere?

This has been mentioned in other posts but it is becoming a serious problem. I have a background c program including an h file with errors in it. The build and download process shoots through but the program compile fails and so is not downloaded. The only clue this has happened is a warning message way up the output which can easily be overlooked: ... Build Process For robotsystem.out has started. C:\ABD\PowerPMAC\Robot\Robots\C Language\Background Programs\robotsystem\robotsystem.c(15,0): Warning : in file included from robotsystem.c Build Process For robotsystem.out has ended. ... Download Successful. Total Project Download time = 14.025 sec Total Project Build and Download time = 30.155 sec in the err.log file you get: In file included from robotsystem.c:15: _sys_build.h: In function 'GetRobotBuildLEFromMemory': _sys_build.h:11: error: 'SR15Orbit_type' undeclared (first use in this function) _sys_build.h:11: error: (Each undeclared identifier is reported only once _sys_build.h:11: error: for each function it appears in.) _sys_build.h: In function 'SetRobotBuildLEToMemory': _sys_build.h:79: error: 'SR15Orbit_type' undeclared (first use in this function) _sys_build.h:81: error: lvalue required as left operand of assignment _sys_build.h: In function '_ReadBuildLEBinary': _sys_build.h:306: error: storage size of 'buildLE' isn't known _sys_build.h: In function '_WriteBuildLEBinary': _sys_build.h:440: error: storage size of 'buildLE' isn't known make: *** [robotsystem.o] Error 1 The parsing of these err.log files to the output window (and it seems also the final SUCCESS/FAIL detection method) is flawed.

I see this on corrupted ppproj files. I don't know about Git but SVN causes this error for me! I get the project too old message when the ppproj file gets a conflict merge from SVN (this happens every time another user commits). The merge ends up like: <<<<<<< .mine 2.0.2.14 ykzk9403YKZK ||||||| .r64919 2.0.0.1 jeon4324JEON ======= 2.0.0.1 adpr1307ADPR >>>>>>> .r65041 Spelling errors aside this is unavoidable when mixing project based parameters with user based ones. I really hope these are separated out in the October release as it makes version control a pain...

I agree it is not an obvious location. I have just seen these and they contain a lot of useful PPMAC information which I remember learning the hard way years ago. I would suggest as a minimum they should be tidied and in the file depot manuals section, and considered for merging in the UG or SRM. They seems to have some overlap with these and each other (even with differing definitions of the same things) My preference would maybe be a new Programmer's Guide with this information and pertinent information from the other manuals collated and organised for programmers (I am obviously biased!).

Thanks for the suggestions. I didn't want to reset my firewall rules, but you did encourage me to run as Administrator (right-click Run As..., rather than just letting it elevate). This appeared to do something, I got the rsync firewall requests and then it worked. Running normally from the shortcut after that worked too. So not sure of exact details of what did it but I am running. Thanks for the nudge

Hi I have just setup my home working system. New PC, clean Windows 10 RTM install, PPMAC IDE 2.0.3.40, 460EX PPMAC CPU running FW 2.0.2.14. My project fails. A new blank project fails. Right-click project, Build and Download all Programs gives the following dialog: --------------------------- Mapping power PMAC variables failed --------------------------- Mapping power PMAC variables Failed. Do you wish to continue the build process. --------------------------- Yes No --------------------------- and the following output: Please wait while mapping Power PMAC variables. Syncing Database Failed. Failed to map PMAC variables. Project Building and Mapping total time = 67.531 sec I cannot find any log messages in the project or IDE folders. Where to start looking?

But that will not dynamically adjust to the actual IDE version. If I build the project with your suggestion in a 2.0.3.40 IDE then great, but when someone running 2.0.0.25 pulls the code from SVN and builds and downloads the project P1-4 will not report the IDE she used. Unless she changes the code, then we have a version control nightmare. I want RUNNING code to tell me what downloaded it. Like: __GNUC__ __GNUC_MINOR__ __GNUC_PATCHLEVEL__

Thanks Steve. Done. I guess you're saying it is not currently possible. Bug 1362 entered in case anyone wants to track.

Is it possible to get running code to report the version of the IDE used to build the project? Something like the following lines in the code: global IDEVersionA = IDE_VERSION_A; global IDEVersionB = IDE_VERSION_B; global IDEVersionC = IDE_VERSION_C; global IDEVersionD = IDE_VERSION_D; where IDE_VERSION_A,B,C,D are #defined by the installed IDE as say 2,3,0,40

The newly added Sys.LegacyCtrl looks quite interesting. We have however upgraded all our UMAC code to PPMAC systems so its main purpose is not so useful. I was wondering if it offered any computational benefits (speed/efficiency) if we don't need some of the PPMAC algorithm (I know there are a few choices of algorithm, just wondering how it fits in). Or does it alter the PID gain effects, etc Don't worry if the updated manuals are coming soon...

Thanks for looking at this, I have just found the time to try it. I now get a warning (Debug or Release): Warning : conflicting types for built-in function 'lrintf' Debug gives a = 2; Release gives a = 0; So this does not work! Also my real problem is in RTICPLC and this still does not build with your lrint solution. Please help! I need to be able to use math functions in all situations... Thanks dave

Thanks for looking at this. To get the list I posted above, I just quickly clicked through the items in the help file and noted down the ones which had no information. If they do this they should see too. I understand doc often ends up being the last thing done and then more 'pressing' tech requirements come along. I am guilty of doing this myself, however I would suggest a dev library or API really needs a bit of detail documenting the features the customer will be delving into. Any expansion on an auto-gathered copy of the function prototypes would be helpful. Example code / scope of validity (user/kernel)/ words of warning etc all go a long way to make coding a more joyful experience!

I cannot find much help on how the C Libraries are implemented. I am now trying to structure a large project. The IDE manual says the libraries are for Background C Programs, there was a time they couldn't be used in CPLCs but I see they are linked to these now so I assume is fixed. Before I go too far: 1) I am struggling to get one library to reference another: I have a "network.c" library setting up sockets and containing standard read/write calls. I have a "network_utils.c" library which contains some logging utils and other small methods. I have a few Background Programs which call methods from both. This is fine. Now my "network.c" library needs to call methods in "network_utils.c". I have #included the header but the build gives errors like "Error : undefined reference to `LogS'" 2) Do they work in rticplc or usrcode yet?