x86/MCE: drop bogus const modifier from AMD's bank4_names()

x86/MCE: drop bogus const modifier from AMD's bank4_names()

[Jan Beulich]

The compiler validly warns about it being ignored.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
---
 arch/x86/kernel/cpu/mcheck/mce_amd.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- 3.19-rc5/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ 3.19-rc5-x86-MCE-AMD-bank4_names/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -79,7 +79,7 @@ static inline bool is_shared_bank(int ba
 	return (bank == 4);
 }
 
-static const char * const bank4_names(struct threshold_block *b)
+static const char *bank4_names(const struct threshold_block *b)
 {
 	switch (b->address) {
 	/* MSR4_MISC0 */

Re: x86/MCE: drop bogus const modifier from AMD's bank4_names()

[Borislav Petkov]

On Fri, Jan 23, 2015 at 08:32:01AM +0000, Jan Beulich wrote:
> The compiler validly warns about it being ignored.
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Applied, thanks.

Out of curiosity: When do you see this? Building with
-Wignored-qualifiers or -Wextra?

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--

Re: x86/MCE: drop bogus const modifier from AMD's bank4_names()

[Jan Beulich]

>>> On 23.01.15 at 10:58, <bp@alien8.de> wrote:
> On Fri, Jan 23, 2015 at 08:32:01AM +0000, Jan Beulich wrote:
>> The compiler validly warns about it being ignored.
>> 
>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
> 
> Applied, thanks.
> 
> Out of curiosity: When do you see this? Building with
> -Wignored-qualifiers or -Wextra?

It's been a long while ago that I noticed and fixed the warning, so
I don't recall the details. What I do know is that I didn't force any
extra flags, and that a similar issue elsewhere in the tree (fixed
in 3.19-rc5) triggered the same warning in at least one of my builds
(again without any extra warning options) in -rc4.

Jan

Re: x86/MCE: drop bogus const modifier from AMD's bank4_names()

[Borislav Petkov]

On Fri, Jan 23, 2015 at 10:05:40AM +0000, Jan Beulich wrote:
> It's been a long while ago that I noticed and fixed the warning, so
> I don't recall the details. What I do know is that I didn't force any
> extra flags, and that a similar issue elsewhere in the tree (fixed
> in 3.19-rc5) triggered the same warning in at least one of my builds
> (again without any extra warning options) in -rc4.

Older compiler maybe. Not that it is worth wasting too much time on it
though...

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--

Re: x86/MCE: drop bogus const modifier from AMD's bank4_names()

[U.Mutlu]

Jan Beulich wrote, On 01/23/2015 09:32 AM:
>
> -static const char * const bank4_names(struct threshold_block *b)
> +static const char *bank4_names(const struct threshold_block *b)

There is a big difference in the return type, cf. below.
Of course, if possible, the more const the better.

Borislav Petkov wrote, On 01/23/2015 11:16 AM:
> On Fri, Jan 23, 2015 at 10:05:40AM +0000, Jan Beulich wrote:
>> It's been a long while ago that I noticed and fixed the warning, so
>> I don't recall the details. What I do know is that I didn't force any
>> extra flags, and that a similar issue elsewhere in the tree (fixed
>> in 3.19-rc5) triggered the same warning in at least one of my builds
>> (again without any extra warning options) in -rc4.
>
> Older compiler maybe. Not that it is worth wasting too much time on it
> though...

int main()
  {
    char mem1[] = "foo";
    char mem2[] = "bar";

    const char* const p1 = mem1;   // both ptr and content are const
    const char*       p2 = mem2;   // content is const, but not the ptr

// ++p1;         // not possible b/c p1 is const
    ++p2;         // possible b/c p2 is not const

// *p1 = 'X';    // not possible b/c content of p1 is const
// *p2 = 'A';    // not possible b/c content of p2 is const

    return 0;
  }


cu
Uenal