Hi Andrey,
Thanks for your answer, thus I'm using 'firmware' as suggested in your first post, not boot (which is a nested project if I'm correct).
Removed folders are not random, for sure that would not make sens. I was only an example, but this folders are not imported with the git command, they are created during compilation, by the makefile. So removing this folder was an example to restore the initial condition.
The 'make clean' command is not working properly here.
thanks for your answer, ok if you tell me compil firmware on Win11 is like win10, I will look at. I doesn't yet understand where is the failure, but I will try again, just asking if win11 is problematic before spend some hour.
You do not have the required permissions to view the files attached to this post.
Yes , I just have try, Exactly same problem. I will format my computer, and maybe with a little lucky .. it will be good. Will post news when I find solution or tips. thanks
Thanks to all for tips , fresh install win 10, install same version make4.4 gcc12.2 , work directly, compil without any problem. I don't known if win11 was the problem or a system corrupt file.
I have a problem trying to run a build of the latest rusefi proteus firmware. It will compile fine, but it does not start up when I load the bin file on a Rusefi proteus that's running an STM32F429
I used to compile the firmware a few years ago on a Windows machine. I decided to see if I can work with the present code, and based on recommendations here it said I should be compiling it in Linux. Currently I am running VirtualBox with Ubuntu to compile the firmware
For reference, I go into the folder /rusefi/firmware/config/boards/proteus
and I run the script with
sh ./compile_proteus_f4.sh
I take the bin file from the build folder over a shared folder to my windows desktop, and I load it on a Proteus with STLink.
What happens is the unit now refuses to start up. No USB connection, no status lights. If I load up the old BIN that I saved before putting in this compiled file, it works exactly as it has for the past few years.
When I run the script compile_proteus_f4_debug.sh in the proteus board folder it actually allows the proteus board to start up, but that debug variant doesn't have LUA, ETB, Boost control and other things when I load it up. It does start up, but it takes out the functionality. When I add the features I need back in like LUA, it makes the firmware file too large and can't be linked. If I change the debug level it will not run in the ECU anymore
A few points to mention, since this is my first step before trying to modify the code, I am just compiling it exactly as I clone it from github. No changes to the code.
I've found that the actual compiled file size is smaller than a precompiled binary that I downloaded at the exact same time as I cloned the repository
I can run a make in the firmware folder with an f407-discovery as the Short Board Name and it will compile fine, load up and run fine on the STM32F429 Proteus board. But if I try to do the same with the f429-discovery, it will compile but also stop the proteus board from running
I have attached the log of me compiling the proteus f4 script ./compile_proteus_f4.sh. It's almost as if components of the code are not being compiled in the firmware. Can anyone see if I'm missing something or know what the problem could be?
Thanks
You do not have the required permissions to view the files attached to this post.
Proteus now uses OpenBLT bootloader. Bin file you are getting is generated with some offset to coexist with bootloader. Bootloader also needs to be flashed.
Or you can disable bootloader by editing env file and removing line "USE_OPENBLT=yes".
Proteus now uses OpenBLT bootloader. Bin file you are getting is generated with some offset to coexist with bootloader. Bootloader also needs to be flashed.
Or you can disable bootloader by editing env file and removing line "USE_OPENBLT=yes".
Thank you very much. That was the answer that I spent a full day trying to get working. I first disabled OPENBLT in the env file and it worked. Then I looked at the compiled hex file and noticed it is offset by 0x8000. When I was writing it to the chip I would always do a complete chip erase and didn't notice it wasn't writing from 0x8000000. I then compiled OpenBLT files and loaded them both on the same chip and it works