RTEMS | Building nucleo-h755zi with tests fails (#5128)

Michael Frandsen (@michael.hesselberg.frandsen) gitlab at rtems.org
Sat Sep 14 10:26:51 UTC 2024



Michael Frandsen created an issue: https://gitlab.rtems.org/rtems/rtos/rtems/-/issues/5128



## Summary
When building the nucleo-h755zi bsp, it fails in test.c when getting to cpukit/posix/src/lio_listio.c
```
[3201/4530] Linking build/arm/nucleo-h755zi/testsuites/psxtests/psxenosys.exe
[3209/4530] Linking build/arm/nucleo-h755zi/testsuites/psxtests/psxfenv01.exe
[3213/4530] Compiling testsuites/psxtests/psxfile01/main.c
[3218/4530] Compiling testsuites/psxtests/psxfile01/test.c
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: ./librtemscpu.a(lio_listio.c.61.o): in function `lio_listio':
/home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:105:(.text.lio_listio+0x68): undefined reference to `rtems_aio_check_sigevent'
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: /home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:148:(.text.lio_listio+0x86): undefined reference to `init_write_req'
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: /home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:158:(.text.lio_listio+0x90): undefined reference to `rtems_aio_enqueue'
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: /home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:125:(.text.lio_listio+0xc6): undefined reference to `init_read_req'
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: /home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:135:(.text.lio_listio+0xce): undefined reference to `rtems_aio_enqueue'
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: /home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:171:(.text.lio_listio+0xfa): undefined reference to `rtems_aio_completed_list_op'
/home/mfrandsen/rtems/tools/6/lib/gcc/arm-rtems6/13.3.0/../../../../arm-rtems6/bin/ld: /home/mfrandsen/rtems/src/rtems/build/arm/nucleo-h755zi/../../../cpukit/posix/src/lio_listio.c:84:(.text.lio_listio+0x1d0): undefined reference to `aio_request_queue'
```

This is built using the following config.ini
```
[DEFAULT]
RTEMS_DEBUG = true
BUILD_TESTS = true

[arm/nucleo-h755zi]
```

A single modification has been added to optmemflashsz.yam to correct the size for adding an CM4 bsp that uses half the flash size. 
```
SPDX-License-Identifier: CC-BY-SA-4.0 OR BSD-2-Clause
actions:
- get-integer: null
- env-assign: null
build-type: option
copyrights:
- Copyright (C) 2020 embedded brains GmbH & Co. KG
default:
- enabled-by:
  - arm/stm32h747i-disco-m4
  - arm/stm32h757i-eval-m4
    arm/nucleo-h755zi
  value: 0x00100000
- enabled-by: true
  value: 0x00200000
description: |
  Size of the internal flash in bytes.
enabled-by: true
format: '{:#010x}'
links: []
name: STM32H7_MEMORY_FLASH_SIZE
type: build
```

## Steps to reproduce
It consistently fails with above changes and config on a newly built toolset.
I issued:
```
./waf configure --prefix=$HOME/rtems/tools/6
./waf
```
$HOME/rtems/tools/6 is the installed location of the sb-set-builder command and compiles nicely without tests enabled. 

## Expected behaviour

My expectation was to try using the test features of BSP validation before making changes or adding CM4. 
Once I have both BSPs from the gitlab repo, I will add the suggested mods from KGardas to play with som OpenAMP stuff.

-- 
View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/issues/5128
You're receiving this email because of your account on gitlab.rtems.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/bugs/attachments/20240914/45575fea/attachment-0001.htm>


More information about the bugs mailing list