RTEMS | waf exits too early when building all BSPs at once (#5252)

Joel Sherrill (@joel) gitlab at rtems.org
Wed May 28 13:21:35 UTC 2025




Joel Sherrill commented: https://gitlab.rtems.org/rtems/rtos/rtems/-/issues/5252#note_122947


For better or worse, the arm failure only happens with GCC 14. I built a baseline with GCC 13 overnight and the only build error is with a m68k BSP. Just introduce a syntax error somewhere in that arm bsp and it will happen.

It is in the background because most of the time, I build in the background and capture the output to a log file. For the larger builds, I was using a remote machine and nohup larger builds out of habit in case the connection drops.

I didn't post config-all.ini because it can be reproduced with just config-two.ini.

Reproducing this: I added a syntax error to bsps/arm/csb336/start/memmap.c so it would fail to compile with GCC 13, configured with config-two.ini, and did a waf -k. All in the foreground. This also resulted in nothing for the sparc/leon3 buiding. This was the last of the output:

```
[4246/4562] Compiling testsuites/validation/ts-validation-acfg-1.c
Dependency /home/joel/rtems-work/rtems/build/arm/csb336/librtemsbsp.a for 
	{task 139908563597568: link psxftw01-tar.o,psxftw01-tar-gz.o,psxftw01-tar-xz.o,init.o,test_cat.o -> psxftw01.exe} is missing: check the task declaration and the build order!
[4247/4562] Compiling testsuites/validation/tc-acfg-disabled-bsp-settings.c
Waf: Leaving directory `/home/joel/rtems-work/rtems/build/arm/csb336'
Build failed
 -> task in 'rtemsbsp' failed with exit status 1 (run with -v to display more information)
```

Notice that this is the arm/csb336 failing to build. No .o or .exe for sparc/leon3.

-- 
View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/issues/5252#note_122947
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/20250528/efa483bb/attachment.htm>


More information about the bugs mailing list