FPGA: first compile fails

[dan_u]

Nick, thanks for your replies, but are you sure about the version numbering? I definitely have LV2012 SP1 and this shows version number 12.0.1 in MAX and in the splash screen, not 12.1.0.

The same thing about the FPGA module. I downloaded 4 GB just to get this:

As I mentioned before, also the NI Update Service shows no updates.

I will send you the MAX report in a PM. The report also mentions the installation of 2012 FPGA SP1 module. So it looks like the bug is still there.

[Nick-C]

dan_u,

My apologies. You are correct, in MAX and on the splash screens for installers and the software itself the format is 12.0.1 for SP1 releases. We use 12.1.0 in other places, so I had some confusion on my end. 

I'm working with the developer who implemented the fix to try and sort out what additional information we'll need from you to sort out what's going on. We'll try and get this worked on asap.

Thanks for your patience.

[Nick-C]

dan_u,

would it be possible for you to send me a copy of the compile worker executable on your machine? The file is located at

C:\Program Files (x86)\National Instruments\FPGA\CompileWorker\CompileWorker.exe 

Our developer wants to do some work with your compile worker to try and sort out what's going on. We have a public FTP site you can use or if you'd like to directly send it to me, feel free to PM me and let me know how you'd like to transfer the file.

[dan_u]

Nick,

Thanks for looking into this. I will only be back in the office on Friday, we have a national holiday tomorrow. I will then send you the compile worker exe from my machine.

Daniel

[dan_u]

Nick,

I uploaded my CompileWorker.exe to the public ftp site as CompileWorker_dan_u.exe. The CompileWorker.exe shows version number 12.0.0.0 and is dated from December 2012. Is that expected?

[Nick-C · Accepted solution]

dan_u,

I received your compile worker exectuable from the ftp site and the developer and I have taken a look. It seems that the problem you are experiencing with this worker is related, but not identical to the issue we originally suspected. The problem you are experiencing with the worker should be addressed in the LabVIEW 2013 release which is available today online. If you are an active SSP member you should be receiving your developer suite disks in the mail shortly. 

We apologize for any inconvenience this may have caused. I am glad to hear from the developer that the fix has already been implemented. To clarify, the fix will be contained in the LabVIEW 2013 FPGA Module Compilation Tools. You will the the full array of 2013 products to work with this compile worker however, so be sure to get all of your required software components. 

For basic FPGA programming this likely will require:

LabVIEW 2013 

LabVIEW 2013 Real-Time Module (if you're using an RT target)

LabVIEW 2013 FPGA Module

LabVIEW 2013 FPGA Module Compilation Tools

NI-RIO 13.0

Any other drivers/modules you may be using in your application

Best of luck with your application, if you have any further questions about this, please feel free to post back. 

[dan_u]

Nick,

thanks for looking at this issue and for the info.

As always, I don't like having to upgrade to a new LV version to get bugfixes. This is for several reasons:

- usually new releases come with new bugs, so we can just replace one bug with another

- upgrading client real-time systems is a huge pain (none of the RT replication tools seem to work reliably, sometimes the RT systems remain unbootable after trying to upgrade), so we're not willing to do this unless absolutely necessary

- testing a large application is a big project in itself, so we really don't want to update to each new LV version

I don't consider this compile bug critical, but this one (also fixed in LV2013) is imho. The LTS LabVIEW version has been declined (http://forums.ni.com/t5/ideas/v2/ideapage/blog-id/labviewideas/article-id/4814/page/3#comments) with Elijah Kerry promising 

"we offer mainstream support for the last four versions of LabVIEW, which includes patches and critical bug fixes during this four-year time-frame".

I don't really see this being true.