Why STM32CubeIDE assembler compiling ccA7fyOT.s for Cortex M4 core ?

[Adi1]

I get the following warning during the compiler. Does someone know why it needs this assembler file. 

Also, when I tried to look through temp folder I couldn't any files with that name.

• C:\Users\arunk\AppData\Local\Temp\ccA7fyOT.s: Assembler messages:
• C:\Users\arunk\AppData\Local\Temp\ccA7fyOT.s:42: Warning: section does not have enough alignment to ensure safe PC-relative loads

[Andrew Neil]

What, exactly, are you doing when you get that message?

What are you trying to compile?

[Adi1]

I have a project based on Cortex M4 that was created on cubeIDE 1.11, and I am now using the latest version of IDE 1.15.

When I build the system, I keep getting this message.

[Bob S]

Just a wild guess - it looks like GCC is compiling a c or c++ file to assembly then running that through the assembler.  That would explain the random-character file name and location in your "temp" directory.  Updating the IDE from 1.11 to 1.15 changed GCC tool chain versions.  The newer versions probably have more diagnostic messages enabled than the older versions, which might explain why you see those messages not and didn't see them before.

What to do about it?  Look at the build log to see what was being compiled when you got that message.  A very quick internet search for that warning message led me to this post as the first search result::

https://forums.freertos.org/t/hard-fault-assembly-code-pc-relative-load-warning/11900

[Andrew Neil]

---

@Bob S wrote:

 That would explain the random-character file name and location in your "temp" directory. 

---

:thumbs_up:

Indeed. And would also explain why the file is no longer there after the build has completed.

[Adi1]

Thank you @Bob S.  For now, I am going to ignore it.

[Bob S]

Geeze, is this the month of half-baked programmers/developers?  Why ignore a warning that your code may not work?  The fix is easy enough, presuming the above link describes the cause of your issue.