Codewarrior 10.1 (eclipse) and multilink on windows 7 64 bit??

[njk909]

I am running 64 bit windows 7.  I installed the eclipse IDE (version 10.1) for coldfire.  This has installed and seems to run fine.  I can open projects and compile without any issue.

I also installed the drivers (version 10) for the P&E USB Multilink.  The device manager shows that the multilink is connected and recognized.

When I open codewarrior the "pemicro" tab is grayed out.  How can I flash a board (52259EVB in my case)?  It seems as if codewarrior isn't seeing the Multilink for some reason even though windows is seeing it.  Is this an issue with a 64 bit OS?

[mjosborne]

I'm seeing the same thing using a project imported from 7.2.

I think the PEMicro menu is just for simulation though, not flashing and debugging?

On import of my classic project I experienced the Remote Sytsem Changed warnings described here:
https://community.freescale.com/message/79744#79744

Built project with no errors.

I updated the Remote System entry with the initial target and memory configurations that were used in the 7.2 project, and confirmed that the System/Connection/Interface settings are correct as specified in the Targeting Microcontrollers document: chapter 13 - ColdFire v2/3/4.

The one thing I did notice is that the Connection->Port entry couldn't detect the rev (D); it showed 'USB1 : USB-ML-CF Rev ? (PE6016454)'.

When I try to debug I get the following error:

Error creating session: The CodeWarrior debugger does not have a Machine object that supports the executable.; Unable to create remote connection
The CodeWarrior debugger does not have a Machine object that supports the executable.; Unable to create remote connection

When I do Flash Programmer->Flash File to Target (Erase or Program with Erase) I see the following in the console:

fl::target -lc "Proj - Main - PEMICRO_USB"
fl::target -b 0x20000000 0x8000
fl::target -v off -l off -ie on -i "{fake_dir}/mcf5225xEVB_PnE.cfg" -p MCF52254
Error: no processors available

I've downloaded the newest firmware from PE but haven't loaded it yet...

I also have yet to try the "Discard project cache" as described in the linked post.

Any help would be greatly appreciated.

This is on a 64-bit Windows 7 box w/ CW10.1.

Thanks,

Matt.

*** UPDATE ***

Noticed that we are using the 512KB internal flash (52259) instead of the 256KB (52254) - changed the Remote System Type entry (to 52259) and also the Flash Programmer (to use MCF52259_INTFLASH.xml) but still getting the same error.

[BlackNight]

Hello,

>>I think the PEMicro menu is just for simulation though, not flashing and debugging?

Yes, this one is only for simulation.

BK

[mjosborne]

So I re-installed the P&E multilink driver (v10) and still no luck.

The multilink shows up in the device manager fine and the CF h/w is good - both have been working in XP vmware & CW v7.2 without problems.

I see the P&E driver test tool doesn't support 64-bit...  Does anyone have any advice on how to troubleshoot the P&E <-> CW connection?

I'm a little hesitant to upgrade the firmware in the multilink (would it still work with 7.2 as well as 10.1 - no downgrading requried?), and I'm not even sure how to find out what firmware is in the multilink.

I've been looking through the eclipse extensions guide for some debug shell commands to poke at the multilink but  nothing stands out in terms of connection diagnosis (like whether or not it sees the PE and can read something from the processor):

%>fl::target -lc "Proj - Main - PEMICRO_USB"
%>fl::target -b 0x20000000 0x8000
%>fl::target -v off -l off -ie on -i "{fake_dir}/mcf5225xEVB_PnE.cfg" -p MCF52259
Error: no processors available

%>fl::device
Device:        
Organization:  
Starts at:    0x0
Ends at:      0x0

%>fl::target
Connection:   
Processor:    
Initfile:    {fake_dir}/mcf5225xEVB_PnE.cfg
RAM Buffer:  0x20000000
Buffer Size: 0xffff
Verify:      0
Logging:     0

I'm pretty sure that the Remote System connection type is correct as is the target initialization and memory configuration.  My Launch config specifies the elf file and Remote System - isn't that all there is to it?

Help please!

Thanks,

Matt.

[mjosborne]

Upgraded multilink f/w using an app called Testcfz.exe.  The new firmware is version 5.54 and this app also appears to have been able to read and write to the processor RAM.

So it appears to be a CW <-> PE problem...

I guess the real question is:

%>fl::target -v off -l off -ie on -i "{fake_dir}/mcf5225xEVB_PnE.cfg" -p MCF52259
Error: no processors available

Why does this debug shell command return the error?

Or another way:

When I launch the debug session why do I get:

Error creating session: The CodeWarrior debugger does not have a Machine object that supports the executable.; Unable to create remote connection
The CodeWarrior debugger does not have a Machine object that supports the executable.; Unable to create remote connection

Thanks again,

Matt.

[mjosborne]

When I open the project I'm seeing the window in the attached screenshot for ~90s on a fast machine...

I took a look at AN4253 even though it's for a different target, as well as the Targeting Microcontrollers doc. 

One thing I did notice that's different is that in Targeting Microcontrolers fig. 13.1 is that it shows the interface as "USB ML-CF - USB Port", but I have "USB Multilink, Embedded OSBDM - USB Port".  The only other interface options I have are for the Cyclones...

Is there an App Note similar to AN4253 for CF? 

Is there a way to tell what is being messed with in the lauch config when the window is shown on opening the project?

[mjosborne]

So, I finally gave up on the 7.2 converted Remote System & Launch config and created them from scratch.

It worked!

My conclusion is that the Remote System part of the CodeWarrior Classic Project Importer doesn't quite do the trick?

Has anyone managed to import a classic project and not experience the Remote System Changed wierdness described here:

https://community.freescale.com/message/79744#79744

I had the same expeience where the displayed comparison values were identical, and the smart merge didn't work.

Thanks for all the help,

Matt.

[mjosborne]

Using the Debug Shell I get a little more information...

%> de

Launching {1}: 0% complete
: 0% complete
: 7% complete
Launching application: 7% complete
Creating debug session: 7% complete
Preparing executable: 7% complete
Preparing executable: 100% complete
Preparing executable: cancelled
Error creating session: The CodeWarrior debugger does not have a Machine object that supports the
executable.; Unable to create remote connection: cancelled