Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] x86/platform/uv: avoid unused variable warning
To:     Christoph Hellwig <hch@infradead.org>,
        Arnd Bergmann <arnd@arndb.de>
Cc:     Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@kernel.org>, Borislav Petkov <bp@alien8.de>,
        x86@kernel.org, Steve Wahl <steve.wahl@hpe.com>,
        Dimitri Sivanich <dimitri.sivanich@hpe.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Hedi Berriche <hedi.berriche@hpe.com>,
        Justin Ernst <justin.ernst@hpe.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Russ Anderson <russ.anderson@hpe.com>,
        Anshuman Khandual <anshuman.khandual@arm.com>,
        linux-kernel@vger.kernel.org
References: <20191212135815.4176658-1-arnd@arndb.de>
 <20191212145432.GA15634@infradead.org>
From:   Mike Travis <mike.travis@hpe.com>
Message-ID: <192a2aa9-5c50-bb8d-4b0f-95eb87de48c0@hpe.com>
Date:   Thu, 12 Dec 2019 07:16:07 -0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101
 Thunderbird/60.8.0
MIME-Version: 1.0
In-Reply-To: <20191212145432.GA15634@infradead.org>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 12/12/2019 6:54 AM, Christoph Hellwig wrote:
> Instead of that maybe_unused mess please just use good old ifdefs.
> 
>>   	if (hubless)
>> -		proc_version_fops.open = proc_hubless_open;
>> +		proc_create_single("hubless", 0, pde, proc_hubless_show);
>>   	else
>> -		proc_version_fops.open = proc_hubbed_open;
>> +		proc_create_single("hubbed", 0, pde, proc_hubbed_show);
>>   }
> 
> Or someone could figure out what happens if we turn the
> proc_create_single stub into an inline function instead of the
> define.  That makes it used at a syntactic level, the big question is
> if the compiler is smart enough to optimize away the unused callback
> still.
> 

Yes, if CONFIG_PROC_FS is undefined, then this whole section can be 
removed since it's sole purpose is to set up the /proc/ interface. 
Something like this should suffice:

> ---
>  arch/x86/kernel/apic/x2apic_uv_x.c |    9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> --- linux.orig/arch/x86/kernel/apic/x2apic_uv_x.c
> +++ linux/arch/x86/kernel/apic/x2apic_uv_x.c
> @@ -1533,6 +1533,7 @@ static void check_efi_reboot(void)
>                 reboot_type = BOOT_ACPI;
>  }
> 
> +#ifdef CONFIG_PROC_FS
>  /* Setup user proc fs files */
>  static int proc_hubbed_show(struct seq_file *file, void *data)
>  {
> @@ -1595,6 +1596,14 @@ static __init void uv_setup_proc_files(i
>                 proc_version_fops.open = proc_hubbed_open;
>  }
> 
> +#else /* !CONFIG_PROC_FS */
> +
> +static __init void uv_setup_proc_files(int hubless)
> +{
> +}
> +
> +#endif /* !CONFIG_PROC_FS */
> +
>  /* Initialize UV hubless systems */
>  static __init int uv_system_init_hubless(void)
>  {