void* arithmnetic

void* arithmnetic

[J.A. Magallón]

Hi all...

Since begin of the ages the build of the nvidia driver says things like
this:

include/asm/compat.h:210: warning: pointer of type 'void *' used in arithmetic

There are several of this warnings. The code in question for this example
is:

static __inline__ void __user *compat_alloc_user_space(long len)
{
    struct pt_regs *regs = task_pt_regs(current);
    return (void __user *)regs->rsp - len;
}

As this is dealing with mem blocks, I suppose it's counting in bytes, so
we could do something like:

   return (void __user *)((u8*)regs->rsp - len);

so the arithmetic knows how to inc/dec for each unity...
I think the warning is correct and that void* arithmetic is undefined in C,
isn't it ?

TIA

--
J.A. Magallon <jamagallon()ono!com>     \               Software is like sex:
                                         \         It's better when it's free
Mandriva Linux release 2008.1 (Cooker) for i586
Linux 2.6.23-jam01 (gcc 4.2.2 20070909 (4.2.2-0.RC.1mdv2008.0)) SMP PREEMPT
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

Re: void* arithmnetic

[Arnaldo Carvalho de Melo]

Em Thu, Nov 29, 2007 at 01:05:31AM +0100, J.A. Magallón escreveu:
> Hi all...
> 
> Since begin of the ages the build of the nvidia driver says things like
> this:
> 
> include/asm/compat.h:210: warning: pointer of type 'void *' used in arithmetic
> 
> There are several of this warnings. The code in question for this example
> is:
> 
> static __inline__ void __user *compat_alloc_user_space(long len)
> {
>     struct pt_regs *regs = task_pt_regs(current);
>     return (void __user *)regs->rsp - len;
> }
> 
> As this is dealing with mem blocks, I suppose it's counting in bytes, so
> we could do something like:
> 
>    return (void __user *)((u8*)regs->rsp - len);
> 
> so the arithmetic knows how to inc/dec for each unity...
> I think the warning is correct and that void* arithmetic is undefined in C,
> isn't it ?

Yes, but not in gcc, the language the kernel is written 8)

It is allowed and the size of a void is 1. -Wpointer-arith disables
this.

[acme@doppio ~]$ cat voidptr.c
#include <stdio.h>

int main(int argc, char *argv[])
{
        void *ptr = argv[argc - 1];

        puts(ptr + 4);
        return 0;
}
[acme@doppio ~]$ gcc -Wall voidptr.c -o voidptr
[acme@doppio ~]$ ./a Magallón
llón
[acme@doppio ~]$ gcc -Wall -Wpointer-arith voidptr.c -o voidptr
voidptr.c: In function ‘main’:
voidptr.c:7: warning: pointer of type ‘void *’ used in arithmetic
[acme@doppio ~]$ ./a Magallón
llón
[acme@doppio ~]$

- Arnaldo

Re: void* arithmnetic

[Jan Engelhardt]

On Nov 29 2007 01:05, J.A. Magallón wrote:
>
>Since begin of the ages the build of the nvidia driver says things like
>this:
>

Explicitly adding -Wpointer-arith to ones own Makefile is like
admitting the code might be problematic. :->


I think sizeof(void *) == 1 is taken as granted as sizeof(int) >= 4
these days. Sigh.

Re: void* arithmnetic

[Ming Lei]

2007/11/29, Jan Engelhardt <jengelh@computergmbh.de>:
>
> On Nov 29 2007 01:05, J.A. Magallón wrote:
> >
> >Since begin of the ages the build of the nvidia driver says things like
> >this:
> >
>
> Explicitly adding -Wpointer-arith to ones own Makefile is like
> admitting the code might be problematic. :->
>
>
> I think sizeof(void *) == 1 is taken as granted as sizeof(int) >= 4
> these days. Sigh.
sizeof(void *) == 4, sizeof(void)==1, :)
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: void* arithmnetic

[Benny Halevy]

On Nov. 29, 2007, 3:19 +0200, "Ming Lei" <tom.leiming@gmail.com> wrote:
> 2007/11/29, Jan Engelhardt <jengelh@computergmbh.de>:
>> On Nov 29 2007 01:05, J.A. Magallón wrote:
>>> Since begin of the ages the build of the nvidia driver says things like
>>> this:
>>>
>> Explicitly adding -Wpointer-arith to ones own Makefile is like
>> admitting the code might be problematic. :->
>>
>>
>> I think sizeof(void *) == 1 is taken as granted as sizeof(int) >= 4
>> these days. Sigh.
> sizeof(void *) == 4, sizeof(void)==1, :)
well, sizeof(void *) == sizeof(unsigned long) maybe :)

>> -
>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at  http://www.tux.org/lkml/
>>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

Re: void* arithmnetic

[WANG Cong]

On Thu, Nov 29, 2007 at 09:38:46AM +0200, Benny Halevy wrote:
>On Nov. 29, 2007, 3:19 +0200, "Ming Lei" <tom.leiming@gmail.com> wrote:
>> 2007/11/29, Jan Engelhardt <jengelh@computergmbh.de>:
>>> On Nov 29 2007 01:05, J.A. Magallón wrote:
>>>> Since begin of the ages the build of the nvidia driver says things like
>>>> this:
>>>>
>>> Explicitly adding -Wpointer-arith to ones own Makefile is like
>>> admitting the code might be problematic. :->
>>>
>>>
>>> I think sizeof(void *) == 1 is taken as granted as sizeof(int) >= 4
>>> these days. Sigh.
>> sizeof(void *) == 4, sizeof(void)==1, :)
>well, sizeof(void *) == sizeof(unsigned long) maybe :)

I *heard* that on Win64 sizeof(void *) > sizeof(long). But it's off-topic
here. ;-)

Re: void* arithmnetic

[Geert Uytterhoeven]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1192 bytes --]

On Thu, 29 Nov 2007, WANG Cong wrote:
> On Thu, Nov 29, 2007 at 09:38:46AM +0200, Benny Halevy wrote:
> >On Nov. 29, 2007, 3:19 +0200, "Ming Lei" <tom.leiming@gmail.com> wrote:
> >> 2007/11/29, Jan Engelhardt <jengelh@computergmbh.de>:
> >>> On Nov 29 2007 01:05, J.A. Magallón wrote:
> >>>> Since begin of the ages the build of the nvidia driver says things like
> >>>> this:
> >>>>
> >>> Explicitly adding -Wpointer-arith to ones own Makefile is like
> >>> admitting the code might be problematic. :->
> >>>
> >>>
> >>> I think sizeof(void *) == 1 is taken as granted as sizeof(int) >= 4
> >>> these days. Sigh.
> >> sizeof(void *) == 4, sizeof(void)==1, :)
> >well, sizeof(void *) == sizeof(unsigned long) maybe :)
> 
> I *heard* that on Win64 sizeof(void *) > sizeof(long). But it's off-topic
> here. ;-)

Yes, Win64 is P64, to make sure it's incompatible with the rest of the
world.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds