MCUXpresso 11.4.0 on Mac Big Sur with J-Link Ultra+ v7.50 not launching debug correctly

[jackking]

I have upgraded to MCUXpresso IDE v11.4.0 [Build 6224] [2021-07-15] which finally fixes the editor issues on Big Sur!

Using the included J-Link version 7.50 does not launch debug correctly (for IMXRT1062).   It just sorta launches quickly and doesn't load flash or execute the program/breakpoints.

I changed the IDE to use the previous J-Link version 6.88c and that still works correctly, so I can use that for now.

Has J-Link 7.50 been confirmed to work with the IMXRT1060 (on Mac)?

[NVazquez]

 I am having the exact same problem, the address location the thread #11 stops is same as well, 0x20ed48.

Flash is written all fine,

Image 'iMXRT1060_SFDP_QSPI Sep 14 2021 15:09:38'
Opening flash driver MIMXRT1060_SFDP_QSPI.cfx
Sending VECTRESET to run flash driver
Flash variant 'JEDEC_SFDP_Device' detected (16MB = 256*64K at 0x60000000)
Closing flash driver MIMXRT1060_SFDP_QSPI.cfx
( 65) Chip Setup Complete
Connected: was_reset=false. was_stopped=true
( 70) License Check Complete
Loading 'SerialCommProject.axf' ELF 0x60000000 len 0x4CD4
Opening flash driver MIMXRT1060_SFDP_QSPI.cfx (already resident)
Sending VECTRESET to run flash driver
Flash variant 'JEDEC_SFDP_Device' detected (16MB = 256*64K at 0x60000000)
Writing 19668 bytes to address 0x60000000 in Flash
1 of 1 ( 0) Writing sectors 0-0 at 0x60000000 with 19668 bytes
( 0) at 60000000: 0 bytes - 0/19668
( 83) at 60000000: 16384 bytes - 16384/19668
(100) at 60004000: 16384 bytes - 32768/19668
Sectors written: 1, unchanged: 0, total: 1
Erased/Wrote sector 0-0 with 19668 bytes in 1545msec
Closing flash driver MIMXRT1060_SFDP_QSPI.cfx
(100) Finished writing Flash successfully.
Flash Write Done
Loaded 0x4CD4 bytes in 2023ms (about 9kB/s)
Reset target (system)
Starting execution using system reset

Any help would be really appreciated. 

[jackking]

OK, Here is the requested logs from JLink 7.50 and 6.88c:

Also two videos of the debug run, on 7.50 it basically does nothing, it says it is debugging, but doesn't actually flash or run the JLink upload.  On 6.88c it shows the J-Link server run the upload and then start debugging.

[lpcxpresso_supp]

We've just tried your use case on our side but can't replicate the behavior you described. Would you please send us the logs in both situations? Please use Help => MCUXpresso IDE Save Info for Support.

Greetings,
MCUXpresso IDE Support

[jackking]

See requested info above.

[lpcxpresso_supp]

Thanks for the logs. Here are a few comments:

1. I see that V7.50a is the version you use. However, according to the Release Notes, IDE v11.4.0 comes with v7.50. Where does V7.50a come from?

2. Target reset seems to be performed correctly when using V6.88c

3. Target reset does not seem fine with V7.50a

4. Flash not written when V6.88c is used. I'm wondering what would happen if flash would be programmed with this J-Link version.

5. Flash programming seems to fail when using V7.50a, thus the debugging experience you encountered.

Would it be possible to use the onboard debug probe and report back the results? If problem is specific to J-Link debug solution, we'll need to reproduce it somehow before reporting it to Segger.

Greetings,
MCUXpresso IDE Support

[jackking]

OK, this is still an issue for me.  I have looked at trying to resolve it again.  It seems that this is related to a change in the way Segger J-Link handles flash drivers.

I found this document: https://community.nxp.com/t5/i-MX-RT-Knowledge-Base/RT1170-Octal-flash-enablement/ta-p/1498369 which covers this change. 

I am using an Embedded Artists iMX RT1062OEM module which has Adesto EcoXIP Octal Flash.  Apparently this will not work directly with the latest Segger J-Link Flash configuration.

I don't see any docs from EA, Segger or NXP on how to enable the Adesto Flash part on this module to work with the latest J-Link drivers.

I have opened a support ticket with Embedded Artists and posted an issue on the RT-UFL github.