Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <ce8e9601-f514-5227-f9f7-87594218f95f@redhat.com>
Date:   Thu, 12 May 2022 11:53:00 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.8.0
Subject: Re: [pdx86-platform-drivers-x86:review-hans 46/59]
 kernel/stop_machine.c:638:35: error: call to undeclared function
 'cpu_smt_mask'; ISO C99 and later do not support implicit function
 declarations
Content-Language: en-US
To:     Thomas Gleixner <tglx@linutronix.de>,
        kernel test robot <lkp@intel.com>,
        Peter Zijlstra <peterz@infradead.org>
Cc:     llvm@lists.linux.dev, kbuild-all@lists.01.org,
        Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
        linux-kernel@vger.kernel.org, Tony Luck <tony.luck@intel.com>
References: <202205120904.Gr9HEY5E-lkp@intel.com> <87zgjnw1bg.ffs@tglx>
From:   Hans de Goede <hdegoede@redhat.com>
In-Reply-To: <87zgjnw1bg.ffs@tglx>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi,

On 5/12/22 11:24, Thomas Gleixner wrote:
> On Thu, May 12 2022 at 09:29, kernel test robot wrote:
>>>> kernel/stop_machine.c:638:35: error: call to undeclared function 'cpu_smt_mask'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
>>            const struct cpumask *smt_mask = cpu_smt_mask(cpu);
> 
> This warning with W=1 is not the worst of the problems.
> 
> The build will simply fail for CONFIG_SMP=y && CONFIG_SCHED_SMT=n
> because cpu_smt_mask() cannot be resolved.
> 
> The other issue is CONFIG_SMP=n. This will fail to build the IFS driver
> because stop_core_cpuslocked() is not available for SMP=n.

The IFS Kconfig already depends on SMP :

config INTEL_IFS
        tristate "Intel In Field Scan"
        depends on X86 && 64BIT && SMP
        select INTEL_IFS_DEVICE
        help
          Enable ...


So I don't think we need the non-SMP implementation inside
include/linux/stop_machine.h, we only need the #ifdef you
suggest in kernel/stop_machine.c  ?

I think it is best to just squash this into the original
patch, do you agree ?

Regards,

Hans


> Something like the below should work as x86 selects SCHED_SMT when
> SMP=y.
> 
> Thanks,
> 
>         tglx
> ---
> --- a/include/linux/stop_machine.h
> +++ b/include/linux/stop_machine.h
> @@ -156,6 +156,12 @@ static __always_inline int stop_machine_
>  }
>  
>  static __always_inline int
> +stop_core_cpuslocked(unsigned int cpu, cpu_stop_fn_t fn, void *data)
> +{
> +	return stop_machine_cpuslocked(fn, data, NULL);
> +}
> +
> +static __always_inline int
>  stop_machine(cpu_stop_fn_t fn, void *data, const struct cpumask *cpus)
>  {
>  	return stop_machine_cpuslocked(fn, data, cpus);
> --- a/kernel/stop_machine.c
> +++ b/kernel/stop_machine.c
> @@ -631,6 +631,7 @@ int stop_machine(cpu_stop_fn_t fn, void
>  }
>  EXPORT_SYMBOL_GPL(stop_machine);
>  
> +#ifdef CONFIG_SCHED_SMT
>  int stop_core_cpuslocked(unsigned int cpu, cpu_stop_fn_t fn, void *data)
>  {
>  	const struct cpumask *smt_mask = cpu_smt_mask(cpu);
> @@ -649,6 +650,7 @@ int stop_core_cpuslocked(unsigned int cp
>  	return stop_cpus(smt_mask, multi_cpu_stop, &msdata);
>  }
>  EXPORT_SYMBOL_GPL(stop_core_cpuslocked);
> +#endif
>  
>  /**
>   * stop_machine_from_inactive_cpu - stop_machine() from inactive CPU
> 
> 
>