Thank you for the responses and help.
By the way, I am using the version tagged 7.23 in git. I have used tools/configure.sh to configure the target, then make menuconfig then make. I am using arm-none-eabi-* 5.4.1 (but can upgrade easily if this is an issue)
Greg, I compared my romlog config against the stm32f4discovery using tools/cmpconfig and attached the output as stm32f4discovery-uavcan-vs-newtarget.diff.
The only ram log related difference I see is that I added the following, but I'm not sure the best way to validate this.
Also, what is the right way to update a config folder defconfig file after running menuconfig? I see the refresh.sh --silent method is deprecated?
Alan, your videos are excellent! I actually used this as my reference when configuring.
Also, I did not know about System.map. I see the mmc files in compilation but no references to anything with the name mmc in System.map. Are the right files not being linked? Should I read through the build system maybe?
I thought maybe my board was maybe the problem, so I built a setup just like in your video:
I configured configs/stm32f103-minimum/usbnsh/defconfig and then followed the kconfig in the video
cmpconfig diff is attached as stm32f103-minimum-usbnsh.diff
I still see the same issue on the stm32f103 minimum board as my new target: the mmc files are in the build output, but not in the System.map file and no mmc device is shown via nsh /dev/
Thank you again for the help,
---In ***@yahoogroups.com, <***@...> wrote :
Could you please explain what exactly is not working?
Remember: System.map is your friend, did you verify if sdmmc functions
were compiled/included (if so it should appear inside System.map
I created a video tutorial explaining how to use SDCard over SPI on NuttX:
Maybe it could help you.
Post by ***@vertile.com [nuttx]
im new to nuttx, and have tried to configure sdcard + spi + fatfs
this isn't working, so I tried turning on debugging (ramlog)
nsh> ls /dev/
relevant (i think?) parts of the .config are...
where do I start debugging?
my first though was stepping through the code w/ gdb, but there must be a better way?