❌ FAIL: Test report for for-kernelci (6.14.0-rc1, upstream-arm-next, 2014c95a)

[Kees Cook]

On Tue, Feb 04, 2025 at 11:15:53AM +0000, Will Deacon wrote:
> [+Kees]
> 
> On Tue, Feb 04, 2025 at 09:51:05AM -0000, cki-project at redhat.com wrote:
> > Hi, we tested your kernel and here are the results:
> > 
> >     Overall result: FAILED
> >              Merge: OK
> >            Compile: FAILED
> > 
> > 
> > Kernel information:
> >     Commit message: Linux 6.14-rc1
> > 
> > You can find all the details about the test run at
> >     https://datawarehouse.cki-project.org/kcidb/checkouts/redhat:1654629441
> 
> It looks like the build fails thanks to a rust panic:
> 
> 00:00:10   EXPORTS rust/exports_helpers_generated.h
> 00:00:11 panicked at main.rs:52:36:
> 00:00:11 Unable to generate bindings: ClangDiagnostic("error: unknown argument: '-fzero-init-padding-bits=all'\n")
> 00:00:11 make[4]: *** [rust/Makefile:332: rust/bindings/bindings_helpers_generated.rs] Error 101
> 
> Presumably introduced by dce4aab8441d ("kbuild: Use
> -fzero-init-padding-bits=all"). I'm guessing that option isn't supposed
> to propagate to .rs sources?

How is that getting passed to Clang? That's a GCC-only option, with
"cc-option":

+KBUILD_CFLAGS += $(call cc-option,-fzero-init-padding-bits=all)

I will take a look...

-Kees

-- 
Kees Cook