Hi, For a while a while (>1 year) I have had a problem where any data I put in the Data Sampling window or in the watched data window w/ refresh will intermittently read back as zero. For example, the attached plot should show a stable value followed by a sawtooth shade, but it shows the general shape with logs of "noise" where it bounces back to 0. I have confirmed that this is not not a plot rendering issue and the raw data shows the zeros as well. I can see a similar issue when I set data in the watch windows to refresh periodically. I used to use this feature on a different micro with no problems and I am fairly sure I used it with this micro in the past with no problems. I am using an STM32L462 and I have tried with an Rev 10.0 J link plus and a rev 11.0 J link plus compact with no change. Any thoughts?

edit: The same behavior happens when using J-Scope FWIW

This particular variable is a file scope static variable in a c file. It doesn't seem to make any difference if I am attempting to look at a variable in a c or cpp file, inside of not inside of a namespace. If the debugger is paused it always shows the expected value. This is a complicated project and it would be infeasible to strip out the IP or get it running on a dev board, unfortunately. I wondered if it might be due to some quirk in our ELF file or something non standard thing we are doing in our linker file perhaps? We are using arm-none-eabi-gcc version 9.3.1 FWIW.

I'll attempt to replicate with this project or, alternatively strip out all the IP from our project. I'll open a support ticket if I have to send you our project files.

A colleague continued to look into this and found some other postings regarding memory not being accessible while some STM32L4x devices were in sleep mode. [SOLVED] SEGGER RTT - stm32 - erratic output when testing low power modes (WFI) - J-Link/Flasher related - SEGGER - Forum
Once we turned off sleep mode his RTT issue and my issue were resolved.