Date:   Thu, 9 Feb 2023 11:06:10 +0800
From:   Yuan Yao <yuan.yao@linux.intel.com>
To:     David Woodhouse <dwmw2@infradead.org>
Cc:     Usama Arif <usama.arif@bytedance.com>, tglx@linutronix.de,
        kim.phillips@amd.com, arjan@linux.intel.com, mingo@redhat.com,
        bp@alien8.de, dave.hansen@linux.intel.com, hpa@zytor.com,
        x86@kernel.org, pbonzini@redhat.com, paulmck@kernel.org,
        linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        rcu@vger.kernel.org, mimoja@mimoja.de, hewenliang4@huawei.com,
        thomas.lendacky@amd.com, seanjc@google.com, pmenzel@molgen.mpg.de,
        fam.zheng@bytedance.com, punit.agrawal@bytedance.com,
        simon.evans@bytedance.com, liangma@liangbit.com
Subject: Re: [PATCH v7 5/9] x86/smpboot: Split up native_cpu_up into separate
 phases and document them
Message-ID: <20230209030610.mpnkr23hlvi2m37d@yy-desk-7060>
References: <20230207230436.2690891-1-usama.arif@bytedance.com>
 <20230207230436.2690891-6-usama.arif@bytedance.com>
 <20230208100336.mdnrvpsssa2sjodu@yy-desk-7060>
 <d0eeb5d0026ed0ecc12ea75b1e6cf03ee2ce068f.camel@infradead.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <d0eeb5d0026ed0ecc12ea75b1e6cf03ee2ce068f.camel@infradead.org>
User-Agent: NeoMutt/20171215
Precedence: bulk

On Wed, Feb 08, 2023 at 12:02:55PM +0000, David Woodhouse wrote:
> On Wed, 2023-02-08 at 18:03 +0800, Yuan Yao wrote:
> >
> > >   #endif
> > > +       /*
> > > +        * Sync point with do_wait_cpu_initialized(). On boot, all secondary
> > > +        * CPUs reach this stage after receiving INIT/SIPI from do_cpu_up()
> > > +        * in the x86/cpu:kick cpuhp stage. At the start of cpu_init() they
> > > +        * will wait for do_wait_cpu_initialized() to set their bit in
> > > +        * smp_callout_mask to release them.
> >
> > The last sentence of the comment looks confused. The fact is:
> >
> > For serial case, The BSP waits AP to set cpu_initialized_mask from
> > wait_for_master_cpu() after fired INIT/SIPI, then AP starts to wait
> > cpu_callout_mask set by BSP from do_boot_cpu().
> >
> > Or the comments below "Bringup step two:..." which also looks clear
> > enough then above.
>
>
> 	/*
> 	 * Sync point with do_wait_cpu_initialized(). Before proceeding through
> 	 * cpu_init(), the AP will call wait_for_master_cpu() which sets its
> 	 * bit in cpu_initialized_mask and then waits for the BSP to set its
> 	 * bit in cpu_callout_mask to release it.
> 	 */
> 	cpu_init_secondary();
>
>
> Better?

Yes, it's better to me, thanks.