Re: [PATCH RFC] x86, boot: add missing declaration of string functions

[Nicholas Mc Guire <der.herr@hofr.at>]

On Thu, Jan 05, 2017 at 09:02:09AM +0100, Ingo Molnar wrote:
> 
> * Nicholas Mc Guire <hofrat@osadl.org> wrote:
> 
> > Add the missing declarations of basic string functions to string.h to allow
> > a clean build.
> > 
> > Fixes: commit 5be865661516 ("String-handling functions for the new x86 setup code.")
> > Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
> > ---
> > 
> > sparse issues a set of warnings about missing declarations:
> > arch/x86/purgatory/../boot/string.c:18:5: warning: symbol 'memcmp' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:26:5: warning: symbol 'strcmp' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:42:5: warning: symbol 'strncmp' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:58:8: warning: symbol 'strnlen' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:69:14: warning: symbol 'atou' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:99:20: warning: symbol 'simple_strtoull' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:128:8: warning: symbol 'strlen' was not declared. Should it be static?
> > arch/x86/purgatory/../boot/string.c:142:6: warning: symbol 'strstr' was not declared. Should it be static?
> > 
> > This patch has one checkpatch warning about the use of simple_strtoul which
> > is obsolete. As this is an independent implementation it is not clear if
> > the changes made in simple_strtoul -> _kstrtoull might also need to be
> > applied here ?
> > 
> > Patch was compile tested with: x86_64_defconfig
> > 
> > Patch is against 4.9.0 (localversion-next is next-20161223)
> > 
> >  arch/x86/boot/string.c |  1 +
> >  arch/x86/boot/string.h |  9 +++++++++
> >  2 files changed, 10 insertions(+)
> > 
> > diff --git a/arch/x86/boot/string.c b/arch/x86/boot/string.c
> > index cc3bd58..9e240fc 100644
> > --- a/arch/x86/boot/string.c
> > +++ b/arch/x86/boot/string.c
> > @@ -14,6 +14,7 @@
> >  
> >  #include <linux/types.h>
> >  #include "ctype.h"
> > +#include "string.h"
> >  
> >  int memcmp(const void *s1, const void *s2, size_t len)
> >  {
> > diff --git a/arch/x86/boot/string.h b/arch/x86/boot/string.h
> > index 725e820..f6ee139 100644
> > --- a/arch/x86/boot/string.h
> > +++ b/arch/x86/boot/string.h
> > @@ -18,4 +18,13 @@ int memcmp(const void *s1, const void *s2, size_t len);
> >  #define memset(d,c,l) __builtin_memset(d,c,l)
> >  #define memcmp	__builtin_memcmp
> >  
> > +int strcmp(const char *str1, const char *str2);
> > +int strncmp(const char *cs, const char *ct, size_t count);
> > +size_t strlen(const char *s);
> > +char *strstr(const char *s1, const char *s2);
> > +size_t strnlen(const char *s, size_t maxlen);
> > +unsigned int atou(const char *s);
> > +unsigned long long simple_strtoull(const char *cp,
> > +				    char **endp, unsigned int base);
> 
> Looks good to me, but please also mark them 'extern' to highlight the API 
> declarations like the rest of the kernel does - such as kernel.h which has
> the kernel's simple_strtoull() declaration, etc.
> 
> It's not required syntactically, but it's a good stylistic principle to keep 
> external APIs organized.
>
thanks - added the externs and resent as V2 

while this does looks consistent with other kernel header files now
checkpatch --strict will issue CHECK requests of the form:
  "CHECK: extern prototypes should be avoided in .h files"
so I just wonder if this CHECK is actually consistent with coding practice ?
the argument in commit 70dc8a48357c ("checkpatch: warn when using extern with function prototypes in .h files")
being:

<snip>
    Using the extern keyword on function prototypes is superfluous visual
    noise so suggest removing it.
    
    Using extern can cause unnecessary line wrapping at 80 columns and
    unnecessarily long multi-line function prototypes.
<anip>

thx!
hofrat