Search Results

Hi All, I am trying to use SES 3.4 to build a flashloader which will add a device capability to JLink and JFlash. I started with the template "OpenFlashLoader CortexM Template EmbeddedStudio.zip". I found it here: wiki.segger.com/Adding_Support…requently_Asked_Questions That template is aid to have been built with SES 3.1. I made my mods and can successfully run the test. Trouble starts when trying to build the "Release" version. There is a long list of errors from trying to include the standard…

Thanks so much for a prompt response and the suggestion to check the hardware connections. I found that the nRESET line was connected, but pin 9 of the 10 pin debug connector was hooked to EZP_CS_b. With that grounded it must have been putting the chip into EZ programming mode. I cut the wire to pin 9 and it works like a champ.

I am having some trouble connecting to some new hardware with a NXP K20, Cortex-M4 core. The problem is observed both in the debugger window of Kinetis Design Studio, and in the JLink Commander window. The debugger error message says that it can't start the GDB server because it can't connect to the target due to a DAP error reading the DP-Ctrl-Stat register. The JLink Commander window output is posted below. I can use the command "erase Kinetis" and it says OK, but the result does not change. T…

I have an ARM Cortex project that builds in NXP's Eclipse environment, MCUXpresso. I would like to import it to SES but run into a problem right away. The importer says that the NXP toolchain is not supported. The NXP toolchain is just their build of gcc, and I would prefer to migrate to the gcc included with SES anyway so that the project could be a completely self-contained Segger project. Is there an easy workaround ?

Hi Markus, Here is the code that I used. It has not been much tested but does appear to work properly in my environment. In particular it adjusts the top of the stack downward, if necessary, to ensure that the firmware can always start up enough to call into a C function, no matter how the GPNVM bits are set. It would be easiest to post my thumb_crt0.s file but am not doing so due to copyright. Instead I posted my additions and some notes that I think are sufficient to see what is being done, an…

Hi Markus, I did this and am developing an application based on it. It took some effort, which would have been less effort if the process of generating a linker command file was better documented. Or better yet, when Segger rolls out their own linker to replace gnu ld. In any case, I was going for a design that would always start up no matter how the GPNVM bits were programmed. I wrote C code that looks at linker defined symbols to determine how much TCM is required by the application. If the ex…

Sorry - specific example above with .s file was not showing results due to having a filter for only *.c. It does work with *.* Previous examples still don't work as I would have expected. George

Here is another example, more flagrant. In the project there is a ATSAME70Q21_Vectors.s. I can open that file, select a vector, and try to search for it. For example, if I double-click to select "UASRT0_IRQHandler", each of the five instances of the symbol in this file are highlighted. But the find references command reports "No results". I can try to cut down the project to a simple example and upload it. George

Hi Nino, I came across two cases this morning where the search does not work. The first was when I was searching for a function declared in a system .h file, and the function was not used in the source yet. Specifically I was looking for a function from CMSIS called NVIC_GetPriority. The search returns nothing. However a search for NVIC_SetPriority finds the CMSIS file and opens it up. The very next method in that file is the Get that was not found. In another case I was looking for a method fro…

Hi All, I've been getting to know 3.28 and 3.30 as a development environment. One thing that comes up all the time is that I want to find all the occurances of a preprocessor macro, or a variable, or a comment in the project. Mostly, this is not working for me and I have to use some external grep program to do the job. SES is just not finding them even when they exist, and under the directory structure of the main project. Is there a switch or option somewhere that affects what SES will search ?…

Thanks for looking at this. FYI, the way that Atmel handles this in their Studio is to simply set the top of RAM at 128K. That will always work for a demo, even if the GPNVM bits configure 128K each for DTCM and ITCM. George

I am trying to find the size of a memory segment. In this specific case it should be easy since I only have one section in the memory segment. According to the docs for the section placement file format, one can add a size_symbol. If my understanding is correct, then I could access the value through this symbol in the application code. I tried this: <MemorySegment name="$(ITCM_NAME:ITCM)"> <ProgramSection alignment="4" load="No" size_symbol="__itcm_size__" name=".code_TCM_run" /> </MemorySegment…

Mystery solved. GPNVM bits 7, 8 in micro were set to map 128K of RAM to ITCM and another 128K to DTCM. That left only 128K available for program RAM, stacks, data. The startup code assumes that the caches are not used at all, so all 384K RAM would be available, and sets the stack pointer at 384K above the ram starting address. The micro specifically does not generate an exception when trying to access RAM that does not exist, unless the MPU is programmed to do so.

And also that it fails the same way with the Atmel SAME70 Xplained demo board. The fault occurs on a pop.w { r4-r11, pc }, but memory where sp is pointing is all 0. In fact that area of memory has been all 0 since entering main. It looks like the memory is not being written. For example, in the corresponding push.w { r4-r11, lr } the affected memory is not changed after the instruction executes, and at least the LR is not zero.

I should add that the fault does not occur when running in the simulator

I am trying to get familiar with SES by trying some of the demo programs. This is on a custom board with Atmel SAME70 microcontroller. I believe that the hardware is OK since I can program and run code and step in the debugger, via J-Link. However, all the demos blow up very early. For example, the hello world program gets a little ways into vsnprintf, then goes to the hard fault handler. It is usage fault, invalid state. No doubt there is something very simple behind this. Anything obvious to c…

I changed all the USB hub settings ( Windows device manager ) to not allow the computer to put the device to sleep. But this was not sufficient to fix the problem. I also tried connecting the J-Link directly to a port on the PC, and through a hub. In cases, the USB link is somehow broken when the computer sleeps. If I try to start the IAR debugger, which is set to use the J-Link, the IAR app simply hangs. I assume that this is due to some behavior in the Segger DLL. If at that point I unplug the…

Infrequently while debugging the J-Link goes into an "erase flash" mode. Perhaps it is after taking an exception of some kind due to a program bug. But in fact it does actually erase flash on the target device. I am developing with a Kinetis K60. After the unexpected erase, it can be hard to halt the debugging session and eventually when I reconnect and try to reprogram my target, it warns me about erasing a secured Kinetis device. What is going on here ? To be more specific about what happens: …

Software 4.80g I've used many J-links over the years but recently purchased and am using an Ultra+ now. I connect it to a Windows 7 x64 PC through a 7 port hub. What I am finding is that after the computer sleeps and then wakes up, communication with the J-Link is lost. To re-establish communication I need to unplug the USB cable from the J-Link and plug it back in. I don't recall this happening with the other models previously. Is this in fact something specific to the Ultra+ or have I overlook…

In fact I am still seeing this message but it is apparent that there is no problem to click OK and let it erase the affected sectors. The affected sectors do not include sector 0. So no need for further support and the case can be closed.