It is strange, the errors are places with "falls through" statements.
Also I have the same arm-none-eabi-gcc version as said in static-test and I do not get errors on my computer.

Could some workers have a different version than the one running static-test ? @smlng

If not, it maybe requires a more safe proof falls through comment.

I really do not get why it is not building… I do not get any error on my computer.

Same for me. Maybe some CI worker is using a different version of the toolchain? Is that possible to check?

This is for mobi3 (failed asymcute_mqttsn on acd52832):

[kaspar@ng riot (master)]$ dwqc -q mobi3.inet.haw-hamburg.de './dist/tools/ci/print_toolchain_versions.sh'                           
Installed compiler toolchains
-----------------------------
             native gcc: gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609                                                          
      arm-none-eabi-gcc: arm-none-eabi-gcc (GNU Tools for Arm Embedded Processors 7-2017-q4-major) 7.2.1 20170904 (release) [ARM/embedded-7-branch revision 255204]
                avr-gcc: avr-gcc (GCC) 4.9.2
       mips-mti-elf-gcc: mips-mti-elf-gcc (Codescape GNU Tools 2016.05-03 for MIPS MTI Bare Metal) 4.9.2                             
             msp430-gcc: msp430-gcc (GCC) 4.6.3 20120301 (mspgcc LTS 20120406 unpatched)                                             
   riscv-none-embed-gcc: riscv-none-embed-gcc (GNU MCU Eclipse RISC-V Embedded GCC, 64-bits) 7.2.0                                   
                  clang: clang version 3.8.0-2ubuntu4 (tags/RELEASE_380/final)                                                       

Installed compiler libs
-----------------------
   arm-none-eabi-newlib: "2.5.0"
    mips-mti-elf-newlib: "2.1.0"
riscv-none-embed-newlib: "2.5.0"
               avr-libc: "1.8.0svn" ("20111229")

Installed development tools
---------------------------
                  cmake: cmake version 3.10.0
               cppcheck: Cppcheck 1.72
                doxygen: 1.8.11
                 flake8: missing
                    git: git version 2.7.4
             coccinelle: spatch version 1.0.4 with Python support and with PCRE support                                              
[kaspar@ng riot (master)]$ 

This is for mobi3 (failed asymcute_mqttsn on acd52832):

well, accidentaly that is also the one printing the static test output for the last build.

If you scroll all the way down in the build output, you'll see that the failures are spread across all workers.

So why does it fail? Is it a gcc 5 thing?

Nevermind, I looked at the wrong line. It's gcc7 on arm

After reading #9398 it seems like this error must be the same ccache problem that the clang builds have. The preprocessor is run separately from the compilation stage and therefore the /* falls through */ comments are not visible to the compiler and therefore we get an error. Modifying the ccache configuration to run the preprocessor twice should fix the errors seen in this PR.

Setting CCACHE_CPP2=yes fixed the problem with the compiler incorrectly reporting Wimplicit-fallthrough errors for intentional fall through.
Ping @cladmi @kaspar030 @miri64

Setting CCACHE_CPP2=yes fixed the problem with the compiler incorrectly reporting Wimplicit-fallthrough errors for intentional fall through.

Nice to have found the cause, too bad we need that setting also for gcc, as this has non-negligable impact on compilation speed when using CCACHE.

It had a 2:37 minutes impact to do the CCACHE_CPP2=yes by comparing #9772 with this one. Current runtime: 13m:11s

But, in practice it should be less I think.

Now this PR is build with the cache from the other PRs that do not set CCACHE_CPP2. So basically there should be more cache miss than what would be at the end when all PRs use this one and the cache is warm everywhere.

Nice to have found the cause, too bad we need that setting also for gcc, as this has non-negligable impact on compilation speed when using CCACHE.

Well, sometimes correctness should take precedence over speed 🤡

This should be good to go, right?