Search Results

Hi Ewout, thank you for your suggestion. Sorting that list is not yet possible. I will add that to our internal list of future improvements. Best regards-- AlexD

Hi tzan, good to hear you are up and running again. Best regards -- AlexD

Hi Tzan, could you please check if the issue persists with the latest version of Ozone? Did you take heed of the suggestions you already received in the old thread you referenced? Probably your device has a boot loader. For debugging devices with a boot loader, please have a look at this WIKI page: wiki.segger.com/Debug_on_a_Target_with_Bootloader. Since you are also intending to debug RAM code, please also have a look at this WIKI page: wiki.segger.com/Ozone_Debug_and_Trace_with_RAM_Functions. …

Hi skam, actually, this should work. I guess, reason you did not see it working, might be that I specified "/" as a separator. Instead it should be ".", as you can see in the documentation of that command, which can be found in section 7.9.17.3 in the Ozone user's manual. Best regards -- AlexD

Hi skam, could you please provide an Ozone log and a J-Link log for the case that the memory window is not updated while the debug session is active? It might be the case the window is actually updated but if the memory content does not change, there will be no change in the display. Creating an Ozone log is described in section 8 of the Ozone user's manual, creating a J-Link log is described here: wiki.segger.com/J-Link_DLL#Enable_J-Link_Log_File Please note that both logs should be recorded in…

Hi skam, for an overview on the different trace types, please also have a look at this WIKI page: wiki.segger.com/General_information_about_tracing In case your chip does not support ETB, this feature is not available to you. Which device are you working on? It is quite likely that ETM is supported by your device, this would allow you to make use of pin tracing, but also requires you to utilize a J-Trace probe. Best regards -- AlexD

Hi skam, thank you for your suggestion to add support for scaling in the range less than 1. I'll bring that up in our next internal meeting. I cannot tell if or when this will be implemented, though. Best regards -- AlexD

Hi skam, this is intended. Hex numbers are normally interpreted as unsigned, however in the graph window numbers are considered being signed. Best regards -- AlexD

Hi Skam, often there are multiple ways to achieve a desired effect. Since your peripheral registers are memory mapped they may be accessed like memory. That's why reading the memory location works as well. The script you create is less portable, though. While the same register may exist in another MCU it might reside at a different address. By using the path name in Target.GetReg() allows your script to execute fine on another MCU, while accessing the memory address may cause tedious update of c…

Hi skam, the memory window does not update its content while the target is halted. This is intended. Automatic update only takes place when the CPU is running. Best regards -- AlexD

Hi skam, ETB is a hardware feature. The respective trace buffer is a buffer that is written by HW only. It is also a dedicated buffer and you cannot use another buffer, e.g. a RAM area, such as the RTT buffer. RTT is a software feature. You can do some software tracing, but not on instruction level. Systemview makes sophisticated use of that feature. General information on tracing, including descriptions of multiple trace types, can be found here: wiki.segger.com/J-Trace#General_information_abou…

Hi Skam, please be a bit more specific. At which occasion do you expect the refresh to occur? Best regards -- AlexD

Hi Skam, nope, that's not possible. However, you might try the scaling feature: Right click into the legend, onto the respective graph, and chose a suitable scaling factor. Maybe that's close to what you intended? Best regards -- AlexD

Hi Skam, I guess, Target.GetReg() should also work for peripheral registers. It might be that you need to specify the full "path" for your register, though, i.e. something like "Peripherals/GPIO/GPIOA/MODER". Could you please check if that works for you? Best regards -- AlexD

Hi skam, as long as the peripheral register is memory mapped and accessible, you can add a term such as *(unsigned int*)(0x080000000) to your Data Sampling window. Of course, you need to replace 0x080000000 with the address of your peripheral register. This should do the trick. Best regards -- AlexD

Hi Rahul, you already opened a ticket in our ticket system with the same question. Your request in the ticket system has already been answered. Please use only a single channel for communicating with us. Best regards -- AlexD

Hi Eggplant886, right now, Ozone does not support loading multiple ELF files. This is on our internal wish-list but I cannot comment on when this will be released. Best regards -- AlexD

Hi Siwei, normally it should be sufficient to rename the XML file to SVD. SVD file format is explained here: keil.com/pack/doc/CMSIS/SVD/html/svd_Format_pg.html. If the file NXP provided doe not adhere to that format you may... ...create a tool translating the NXP format to the SVD format ...create your own SVD file, taking one of the SVD files for peripherals coming with Ozone as a template. Does that answer your question? Best regards -- AlexD P.S.: In future, please do not use multiple channe…

Hi Jan_vi, when testing Ozone on Ubuntu I use the following commands: Source Code (2 lines) On a blank machine J-Link software should be installed before the 1st debug session is started with Ozone. If this is not the case the J-Link/J-Trace may not be found. Does that answer your question? Best regards -- AlexD

Hi rflmota, right now we are aware of very few customers reporting a poor performance on MacOS, however we are not able to reproduce that in our labs. The overlapping text display in the context menu appears to be caused by MacOS having become incompatible with the QT version we are using and so far we did not find a way to work around that. We expect that being resolved once we moved to a later QT version. Unfortunately at this point in time we cannot comment on when this will be released. Does…