From: Andrew Morton on
On Thu, 10 Jun 2010 13:10:39 +0200 (CEST)
Andi Kleen <andi(a)firstfloor.org> wrote:

>
> Avoid quite a lot of warnings in header files in a gcc 4.6 -Wall builds
>
> Signed-off-by: Andi Kleen <ak(a)linux.intel.com>
>
> ---
> include/linux/pagemap.h | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> Index: linux-2.6.35-rc2-gcc/include/linux/pagemap.h
> ===================================================================
> --- linux-2.6.35-rc2-gcc.orig/include/linux/pagemap.h
> +++ linux-2.6.35-rc2-gcc/include/linux/pagemap.h
> @@ -423,8 +423,10 @@ static inline int fault_in_pages_readabl
> const char __user *end = uaddr + size - 1;
>
> if (((unsigned long)uaddr & PAGE_MASK) !=
> - ((unsigned long)end & PAGE_MASK))
> + ((unsigned long)end & PAGE_MASK)) {
> ret = __get_user(c, end);
> + (void)c;
> + }
> }
> return ret;
> }

urgh. In fact I'd urgh the whole patchset.

Problem is, anyone who looks at all these random (void) casts is going
to have a hard time working out why they're there. This is worsened by
the long-standing practice wherein some people put unneeded (void) casts all
over the place due to being traumatised by lint 15 years ago (I think).

Wouldn't it be better to make this stuff self-documenting, so anyone
who reads the code can immediately see what it's doing, rather than
scratching their heads over random, seemingly-unneeded casts?

#define gcc_46_is_a_pita(expr) ((void)(expr))

?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Andi Kleen on
> urgh. In fact I'd urgh the whole patchset.
>
> Problem is, anyone who looks at all these random (void) casts is going

There are actually not that many in the patch kit if you read it
completely. I think that's nearly the only one outside a macro.

> to have a hard time working out why they're there. This is worsened by
> the long-standing practice wherein some people put unneeded (void) casts all
> over the place due to being traumatised by lint 15 years ago (I think).

It's really the same reason why they used this with lints.

When I tried it first I started complaining about gcc too (I even
filled a gcc bug), then disabled it in my build,
but after some contemplation I found it really finds real bugs.

>
> Wouldn't it be better to make this stuff self-documenting, so anyone
> who reads the code can immediately see what it's doing, rather than
> scratching their heads over random, seemingly-unneeded casts?
>
> #define gcc_46_is_a_pita(expr) ((void)(expr))

The warning is really useful and we'll get all used to it.

So I went with the plain cast. Maybe some more comments would
be useful.

-Andi

--
ak(a)linux.intel.com -- Speaking for myself only.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/