Re: next-20250616: S390 gcc-8 allnoconfig mm mempool.c In function remove_element

[Heiko Carstens <hca@linux.ibm.com>]

On Mon, Jun 16, 2025 at 07:06:10PM +0530, Naresh Kamboju wrote:
> On Mon, 16 Jun 2025 at 17:36, Naresh Kamboju <naresh.kamboju@linaro.org> wrote:
> >
> > Regression while building S390 with the Linux next-20250616
> > with gcc-8 the following kernel warnings found.
> >
> > Regressions found on S390
> >  -  build/gcc-8-lkftconfig-allnoconfig
> >  -  build/gcc-8-lkftconfig-hardening
> >
> > Regression Analysis:
> >  - New regression? Yes
> >  - Reproducibility? Yes
> >
> > First seen on the next-20250616
> > Good: next-20250613
> > Bad:  next-20250616
> >
> > Build regression: S390 gcc-8 allnoconfig mm mempool.c In function remove_element
> >
> > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> >
> > ## Build errors
> > mm/mempool.c: In function 'remove_element':
> > include/linux/compiler_types.h:497:20: warning: asm operand 0 probably
> > doesn't match constraints
> >  #define asm_inline asm __inline
> >                     ^~~
> > arch/s390/include/asm/bug.h:12:2: note: in expansion of macro 'asm_inline'
> >   asm_inline volatile(     \
> >   ^~~~~~~~~~
> > arch/s390/include/asm/bug.h:43:2: note: in expansion of macro '__EMIT_BUG'
> >   __EMIT_BUG("", 0);    \
> >   ^~~~~~~~~~
> > include/asm-generic/bug.h:77:57: note: in expansion of macro 'BUG'
> >  #define BUG_ON(condition) do { if (unlikely(condition)) BUG(); } while (0)
> >                                                          ^~~
> > mm/mempool.c:149:2: note: in expansion of macro 'BUG_ON'
> >   BUG_ON(pool->curr_nr < 0);
> >   ^~~~~~
> > include/linux/compiler_types.h:497:20: error: impossible constraint in 'asm'
> >  #define asm_inline asm __inline
> >                     ^~~
> 
> Anders bisect this build regressions and found,
> 
>  # first bad commit:
>     [45c79ca947c936085c94b716be92eaf9a1bdc8bb]
>     bugs/s390: Use 'cond_str' in __EMIT_BUG()

I'll take a look at this tomorrow. I guess the easiest "fix" is to
change the s390 bug code similar to arm64, so that those problems with
inline asm constraints cannot happen anymore.

At least I hope that will work. We'll see.