10 replies to this topic

[nhale]

Hi, I just started to test my new Netduino 2 Plus and tried to set a breakpoint in VS2012. In comparison to the "old" Netduino Plus it takes ages until the breakpoint is set and additionally it's never hit...with the Netduino Plus no problem. I also replaced the NetduinoPlus.dll with the Netduino.dll as suggested in some topics. But no success I'm running the same code with the same Framework and SDK, only the devices are different. .NETMF 4.3 Netduino SDK 4.2.2 (Jan13) Anyone else the same experience? Thanks

[nhale]

I tried now some simple code which just lets the Onboard LED flash and this works also fine on both devices with setting and hitting the breakpoint.

 

As my project runs a few threads and is splitting into 3 different project-files, is there maybe some limitation/bottleneck on the N+2 compared to the N+?

[nhale]

Ok, I did another test.

I increased the amount of code compared to the simple Onboard LED test (just one class with a few lines)

Now the project has some classes with a few hundred lines building a menu for a 16x2 LCD.

The refresh of the LCD is much faster than with the old N+ but the strange thing I could observer is, that setting a breakpoint took more time.

The breakpoint was also hit, but hovering over objects and displaying values took a few seconds.

 

It would be very nice if one of the Moderators or Admins could comment on that...

[Chris Walker]

Hi nhale, The new .NET MF 4.3 SDK has a lot of new code to make connecting to a device during deployment more reliable. Some of this is done with waiting for things to catch up. It's possible that this results in faster or slower step-by-step debugging and variable inspection in some situations. Could you please share your experience by opening up an issue at netmf.codeplex.com? There may be some tuning that can be done in .NET MF vNext (or a QFE) to speed things up more. Chris

[nhale]

Hi,

 

I created an issue for the problem.

If someone else can reproduce it, may a comment with the experience might be useful.

 

http://netmf.codeple...m/workitem/1902

 

Thanks

[nhale]

Someone else also experiences the same problem.

See comments at http://netmf.codeple...m/workitem/1902

 

I wonder that nobody else here experiences this problem...

[HiredMind]

That is the first thing I noticed about working with the NP2 - stepping code is a bit slower than on the NP1. I plan to re-flash both devices to make certain what firmware versions I'm running, and post the results here.  (Had planned on doing that already but life intrudes )

 

Here is an additional mode which I haven't seen mentioned: VS2012 / NP2; when stepping through code with Step Over (F10), sometimes the debugger doesn't appear to stop on the next line.  If you subsequently set a breakpoint on that line, the debugger breaks.  It's as if the NP2 is indeed in break mode, but the debugger never gets that message.

[nhale]

I tried the Firmware 4.2.2.1 and yesterday the new Firmware 4.2.2.2 - both the same problem.

Currently I try to prepare two example solutions to attach them to the issue.

 

I'm not sure if its really the .NETMF, because with the same the N+1 just works fine - or does the different type of "CPU" make such a difference?

Because the "only" difference between both is the FW(software) and the device (hardware) and both is from SecretLabs....

I wonder that they did not encounter this problem, or mention it...

[Chris Walker]

I just chatted with Microsoft, and am working on potential ways to diagnose this issue. What would be really helpful is if we could get a tiny repro which doesn't exhibit this issue (such as blinky) and then a slightly larger one that does have the issue. If the same project works great on a gen1 board (which uses a slower MCU--and therefore may be affected differently by the newly-tuned MFDeployEngine 4.3), that will make the repro even nicer. Unfortunately we can't break into Visual Studio from here to see what's happening, but we can sniff the USB traffic and work with a Microsoft engineer to figure out what's going on. In the meantime, if you experience this issue and it is a blocking issue for you... Visual Studio 2010 (including Express) should continue working great with boards running the production 4.2 firmware. Would one of you be willing to put together a short repro and then a longer repro that exhibits this issue (preferably only with the newer boards)? I would be very grateful Thank you all for sharing your experiences, Chris

[nhale]

Thanks Chris, I will try to upload a solution which can be taken to reproduce the issue, just have to do something which does not require special hardware. I already have started with something and can reproduce the issue, but its not that bad as with my project (which is much more code), i just need some foo-bar-senseless code ;-) As a simple example just take the "breathing Led" sample from the forum, thats normal (set, hit breakpoint) but also slower when drilling down on a nested object.

[DavidAsensio]

Good afternoon,

I'm experiencing problems with the breakpoints in netduino plus 2, and I'm not capable to execute the software step by step. The software download the program into the netduino but it not reset and the breakpoint not work.

I haven't have any problem with the netduino plus...

 

Br.

 

David Asensio