Re: [PATCH] riscv: Fixup compile error BUILD_BUG_ON failed

[Masami Hiramatsu <mhiramat@kernel.org>]

On Sun, 28 Jun 2020 14:12:07 +0800
Guo Ren <guoren@kernel.org> wrote:

> On Sun, Jun 28, 2020 at 10:59 AM Masami Hiramatsu <mhiramat@kernel.org> wrote:
> >
> > On Sat, 27 Jun 2020 16:20:02 +0000
> > guoren@kernel.org wrote:
> >
> > > From: Guo Ren <guoren@linux.alibaba.com>
> > >
> > > Unfortunately, the current code couldn't be compiled, because
> > > BUILD_BUG_ON needs a static defined value, not a dynamic
> > > variable with a0 regs. Just use it inline as a solution.
> > >
> > >   CC      arch/riscv/kernel/patch.o
> > > In file included from ./include/linux/kernel.h:11,
> > >                  from ./include/linux/list.h:9,
> > >                  from ./include/linux/preempt.h:11,
> > >                  from ./include/linux/spinlock.h:51,
> > >                  from arch/riscv/kernel/patch.c:6:
> > > In function ‘fix_to_virt’,
> > >     inlined from ‘patch_map’ at arch/riscv/kernel/patch.c:37:17:
> > > ./include/linux/compiler.h:392:38: error: call to ‘__compiletime_assert_205’ declared with attribute error: BUILD_BUG_ON failed: idx >= __end_of_fixed_addresses
> > >   _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> > >                                       ^
> > > ./include/linux/compiler.h:373:4: note: in definition of macro ‘__compiletime_assert’
> > >     prefix ## suffix();    \
> > >     ^~~~~~
> > > ./include/linux/compiler.h:392:2: note: in expansion of macro ‘_compiletime_assert’
> > >   _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> > >   ^~~~~~~~~~~~~~~~~~~
> > > ./include/linux/build_bug.h:39:37: note: in expansion of macro ‘compiletime_assert’
> > >  #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
> > >                                      ^~~~~~~~~~~~~~~~~~
> > > ./include/linux/build_bug.h:50:2: note: in expansion of macro ‘BUILD_BUG_ON_MSG’
> > >   BUILD_BUG_ON_MSG(condition, "BUILD_BUG_ON failed: " #condition)
> > >   ^~~~~~~~~~~~~~~~
> > > ./include/asm-generic/fixmap.h:32:2: note: in expansion of macro ‘BUILD_BUG_ON’
> > >   BUILD_BUG_ON(idx >= __end_of_fixed_addresses);
> > >   ^~~~~~~~~~~~
> > >
> > > Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
> > > Cc: Masami Hiramatsu <mhiramat@kernel.org>
> > > Cc: Zong Li <zong.li@sifive.com>
> > > ---
> > >  arch/riscv/kernel/patch.c | 3 +--
> > >  1 file changed, 1 insertion(+), 2 deletions(-)
> > >
> > > diff --git a/arch/riscv/kernel/patch.c b/arch/riscv/kernel/patch.c
> > > index d4a64df..f8e84f2 100644
> > > --- a/arch/riscv/kernel/patch.c
> > > +++ b/arch/riscv/kernel/patch.c
> > > @@ -20,7 +20,7 @@ struct patch_insn {
> > >  };
> > >
> > >  #ifdef CONFIG_MMU
> > > -static void *patch_map(void *addr, int fixmap)
> > > +static inline void *patch_map(void *addr, int fixmap)
> >
> > Would we be better to use "__always_inline" as same as fix_to_virt?
> Ok
> 
> > And also, could you add a comment why we need to make it inline?
> I've mentioned in comment:
> > > BUILD_BUG_ON needs a static defined value, not a dynamic
> > > variable with a0 regs.
> 
> idx must be a const unsigned int or it will cause compile error with
> BUILD_BUG_ON.

Ah, sorry for the confusion. I meant the comment line in the code.
BTW, can we also use "const unsigned int" for the fixmap index so that
no one miss it anymore?

Thank you,

> 
> /*
>  * 'index to address' translation. If anyone tries to use the idx
>  * directly without translation, we catch the bug with a NULL-deference
>  * kernel oops. Illegal ranges of incoming indices are caught too.
>  */
> static __always_inline unsigned long fix_to_virt(const unsigned int idx)
> {
>         BUILD_BUG_ON(idx >= __end_of_fixed_addresses);
>         return __fix_to_virt(idx);
> }
> 
> -- 
> Best Regards
>  Guo Ren
> 
> ML: https://lore.kernel.org/linux-csky/


-- 
Masami Hiramatsu <mhiramat@kernel.org>