Search Results

Hi AndyR, please refer to the Ozone user' manual, section 4.6.1, where the hardware requirements for the data sampling window are listed. There you can see that you can control background memory access emulation via the system variable VAR_ALLOW_BMA_EMULATION. Setting this variable correctly should resolve the issue you see. In general, if you see the data sampling window populating with data, you should be able to see that content in the data pane of the time line window. If the data sampling w…

Hi fraengers, thank you. We will try to reproduce and come back to you. This might take a while since all engineers are currently fully booked. Best regards -- AlexD

Hi AndyR, I'm sorry but I am not able to comprehend the point of your inquiry. Could you please describe more clearly what exactly you would like to know or understand? Best regards -- AlexD

Hi Hi Charles_1, good to see you are up and running again. We will consider this thread as closed now. Thanks for sharing this with us and best regards -- AlexD

Hi fraengers, we were able to reproduce the issues from bullet point #1 and #2. We will provide fixes in a future release of Ozone. Best regards -- AlexD

Hi Charles_1 your inquiry is no longer Ozone specific. You are facing general connection issues, which cannot be handled by Ozone team but need to be handled by the J-Link team. If my understanding is correct, you are now able to attach to your device. Is that correct? If not, please open a thread in the J-Link section of this forum. Best regards -- AlexD

Hi Charles_1, just a guess: Does your RTOS enter a power-saving mode when entering the idle thread? In that case, please have a look here: wiki.segger.com/J-Link_Low-power_mode_debugging Best regards -- AlexD

Hi apullin, thanks for your inquiry. Ozone does not provide such a feature. Best regards -- AlexD

Hi keijan2000, as stated before, due to the strong interlinking of Ozone and the ELF file, a snapshot cannot operate properly without the original ELF file. So I kindly suggest to archive the respective ELF file together with the snapshot, allowing you to restore a snapshot also later. An alternative might be you to export the data sampling window contents into a file and use external tooling for the graphing. The resulting text file can easily be used in other tools which can do the graphics di…

Hi Keijan, we are able to reproduce the issue, thanks to your reproducer. Deeper analysis is required which we cannot do right now. A fix, once available, will be provided in a future Ozone version. Best regards -- AlexD

Hi kejian2000, Thanks for your suggestion. Please refer to the user's manual, section 3.11.11.1 Load Snapshot: "In particular, the program file must binary-match the program file at the time the snapshot was saved." The scenario you describe is intentionally not supported. Everything Ozone does is heavily linked to the information in the ELF file. Changing the ELF file w/o letting Ozone know may lead to unpredictable behavior. Best regards -- AlexD

Hi kejian2000, we will try to reproduce and report back here later. Best regards -- AlexD

Hi Keijan2000, thanks for the explanation. I will close this thread now. Best regards -- AlexD

Hi Keijam2000, thank you for letting us know. We are able to reproduce and will provide a fix in a future release. Just out of curiosity: Why do you save the same snapshot multiple times? Best regards -- AlexD

Hi fraengers, as stated before, we need to be able to reproduce the issue locally. Could you please provide a reproducer, as described in my previous post? Best regards -- AlexD

Hi kejian2000, Could you please provide a reproducer on an eval board? We'd require the ELF file, the Ozone project, and detailed instructions on what to do to reproduce the issue on that very board. Best regards -- AlexD

Hi Ewout, performance issues on Mac architecture machines are not known to us. On our test machines Ozone executes smoothly. Since you state that loading an ELF file is slow, can you provide an ELF file where that issue is seen and the respective Ozone project file? So I can measure the time consumed for loading the ELF file locally and perform a comparison across our test machines? Best regards -- AlexD

Hi grb212, Do you use the latest version of Ozone? Are you able to reproduce the issue on Ozone V3.36? If so, could you please provide a reproducer on an eval board? We'd require the ELF file, the Ozone project, and detailed instructions on what to do to reproduce the issue on that very board. Best regards -- AlexD

Hi, Not sure why SystemView should be too much overhead. In general it is recommended the tools that fit a given task best. In the case you describe, SystemView appears to be a very good solution. I just tried: Zooming on the Y-axis of the code pane works as a charm using shift-mouse wheel. I cannot reproduce the issue you observe, but I also cannot see that you are doing zooming in the Y-axis in your screen recording. Power sampling is disabled in the power sampling window. Disabling the power …

Hi Fraengers, in the code pane of the timeline window ISRs are displayed in special way, so they can be identified quite easily (different color, rounded corner). In a long trace, however, the view is likely to be too compressed for directly seeing all ISRs at the same time. Of course, searching the trace window might also help in navigating the instruction trace. What additional live information would you expect to see? Best regards -- AlexD