NUCLEO_L452RE doesn't compile anymore: "error: header guard..."

[Tico06]

Hi There,

I have a strange error compiling the firmware for NUCLEO_L452RE:

make BOARD=NUCLEO_L452RE
Use make V=1 or set BUILD_VERBOSE in your environment to increase build verbosity.
mkdir -p build-NUCLEO_L452RE/genhdr
GEN build-NUCLEO_L452RE/genhdr/pins.h
GEN build-NUCLEO_L452RE/genhdr/pybcdc.inf
GEN build-NUCLEO_L452RE/genhdr/pybcdc_inf.h
GEN stmconst build-NUCLEO_L452RE/genhdr/modstm_const.h
GEN build-NUCLEO_L452RE/genhdr/mpversion.h
GEN build-NUCLEO_L452RE/genhdr/qstr.i.last
In file included from eth.c:35:
eth_phy.h:28: error: header guard 'MICROPY_INCLUDED_STM32_PHY_H' followed by '#define' of a different macro [-Werror=header-guard]
   28 | #ifndef MICROPY_INCLUDED_STM32_PHY_H
eth_phy.h:29: note: 'MICROPY_INCLUDED_STM32_PYH_H' is defined here; did you mean 'MICROPY_INCLUDED_STM32_PHY_H'?
   29 | #define MICROPY_INCLUDED_STM32_PYH_H
cc1: all warnings being treated as errors
In file included from eth_phy.c:29:
eth_phy.h:28: error: header guard 'MICROPY_INCLUDED_STM32_PHY_H' followed by '#define' of a different macro [-Werror=header-guard]
   28 | #ifndef MICROPY_INCLUDED_STM32_PHY_H
eth_phy.h:29: note: 'MICROPY_INCLUDED_STM32_PYH_H' is defined here; did you mean 'MICROPY_INCLUDED_STM32_PHY_H'?
   29 | #define MICROPY_INCLUDED_STM32_PYH_H
cc1: all warnings being treated as errors
Command '['arm-none-eabi-gcc', '-E', '-DMICROPY_VFS_FAT=1', '-DFFCONF_H="lib/oofatfs/ffconf.h"', '-I.', '-I../..', '-Ibuild-NUCLEO_L452RE', '-I../../lib/cmsis/inc', '-I../../lib/stm32lib/CMSIS/STM32L4xx/Include', '-I../../lib/stm32lib/STM32L4xx_HAL_Driver/Inc', '-Iusbdev/core/inc', '-Iusbdev/class/inc', '-Ilwip_inc', '-Wall', '-Wpointer-arith', '-Werror', '-Wdouble-promotion', '-Wfloat-conversion', '-std=gnu99', '-nostdlib', '-DSTM32L452xx', '-DUSE_FULL_LL_DRIVER', '-mthumb', '-mfpu=fpv4-sp-d16', '-mfloat-abi=hard', '-mfp16-format=ieee', '-mtune=cortex-m4', '-mcpu=cortex-m4', '-Os', '-DNDEBUG', '-Iboards/NUCLEO_L452RE', '-DSTM32_HAL_H=<stm32l4xx_hal.h>', '-DMBOOT_VTOR=0x08000000', '-DMICROPY_HW_VTOR=0x08000000', '-DMICROPY_FLOAT_IMPL=MICROPY_FLOAT_IMPL_FLOAT', '-fsingle-precision-constant', '-DMICROPY_HW_ENABLE_ISR_UART_FLASH_FUNCS_IN_RAM=1', '-fdata-sections', '-ffunction-sections', '-g', '-DMICROPY_ROM_TEXT_COMPRESSION=1', '-DMICROPY_BOARD_BUILD_NAME="NUCLEO_L452RE"', '-DNO_QSTR', 'pin.c', 'pin_defs_stm32.c', 'pin_named_pins.c', 'bufhelper.c', 'dma.c', 'i2c.c', 'i2cslave.c', 'pyb_i2c.c', 'spi.c', 'pyb_spi.c', 'qspi.c', 'octospi.c', 'uart.c', 'ulpi.c', 'can.c', 'fdcan.c', 'pyb_can.c', 'usb.c', 'eth.c', 'eth_phy.c', 'gccollect.c', 'help.c', 'machine_bitstream.c']' returned non-zero exit status 1.
make: *** [../../py/mkrules.mk:133: build-NUCLEO_L452RE/genhdr/qstr.i.last] Error 1
make: *** Suppression du fichier « build-NUCLEO_L452RE/genhdr/qstr.i.last »

I did the
make BOARD=NUCLEO_L452RE clean
followed by
make BOARD=NUCLEO_L452RE submodules
and compiled with
make BOARD=NUCLEO_L452RE

I don't see what could have changed, I have compiled many times targeted to this board. Did the Fedora migration from release 40 to 41 impacted this ?

Any clue what could fix this issue ?

Thanks,
Eric.

[scruss]

Did the Fedora migration from release 40 to 41 impacted this ?

It looks like arm-none-eabi-gcc got a bump from 13.x to 14.x

v1.26.0 compiled cleanly for me for the NUCLEO_L452RE with arm-none-eabi-gcc 14.2.1 on Ubuntu

[cwalther]

Either way, there is clearly a typo on line 29 (PYH instead of PHY) – that is what the compiler is trying to tell you. You can probably submit a pull request to fix that by editing the file right in the GitHub web interface.

[Tico06]

ARrggghhh !!
Thanks @cwalther ! It's incredible I read those lines many times without catching this....
I'll give it a try and update this thread.
thx !
Eric.

[Tico06]

Ok, it compiles with the suggested fix.
;)

[scruss]

Mine compiled cleanly without the fix, too

[Tico06]

Please could you check the file eth_phy.h, lines 28 and 29.

For some reason this file is now parsed for a L452, while there is no need for ethernet, or I misunderstood?

[scruss]

yup: micropython/ports/stm32/eth_phy.h

     27 
     28 #ifndef MICROPY_INCLUDED_STM32_PHY_H
     29 #define MICROPY_INCLUDED_STM32_PYH_H
     30 
     31 #if defined(MICROPY_HW_ETH_MDC)
     32 
     33 // Common ETH PHY register definitions
     34 #undef PHY_BCR

[Tico06]

Looks like your build compiles without this file. I don't understand why mine compiles with.

The file looks more than one year old and I built more than once for sure. This means could be my env asking to have this file parsed.

I dont really know where to look, hopefully the impact on the firmware size is limited.

Nevertheless, the typo was existing.

[Tico06]

Guys,

My patch doesn't commit due to:
error: commit 5d8f86a: Unwanted email address: [email protected]

I dont see where this [email protected] comes from.
Please see: #17895

[cwalther]

Oh well, looks like using the web interface doesn’t work after all – I don’t know whether you can affect what GitHub puts into the committer field. Sorry, I hadn’t tried that before, that’s why I wrote “probably”.

Right now, the commit looks like

commit 5d8f86a266c1c22745d08d092d7ca7ed66cc7133
Author:     Tico06 <***@gmail.com>
AuthorDate: Mon Aug 11 19:30:47 2025 +0200
Commit:     GitHub <[email protected]>
CommitDate: Mon Aug 11 19:30:47 2025 +0200

    Update eth_phy include : Typo in line 29.
    
    Typo on line 29 (PYH instead of PHY).
    
    Signed-off-by: Tico06 <***@gmail.com>

You should be able to fix that by doing the following on your computer (insert the correct e-mail address in place of ***):

git clone https://github.com/Tico06/micropython.git
cd micropython
git switch patch-1
git -c user.name=Tico06 -c user.email=***@gmail.com commit --amend
git log -p --pretty=fuller
git push --force

While you are editing the commit message in the commit --amend step, make sure it follows the guidelines in CODECONVENTIONS.md.

I would also mention the compiler error message you received in the commit message to explain how you found the bug – because otherwise the bug is harmless and it is not clear that you are actually fixing a compilation failure and not just doing cosmetics.

[Tico06]

No worries, my learning curve increases days by days with Micropython project.

Done, and its commited.

Thx !