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;
Date:   Tue, 10 Jan 2023 13:06:37 +0100 (CET)
From:   Christoph Lameter <cl@gentwo.de>
To:     Marcelo Tosatti <mtosatti@redhat.com>
cc:     atomlin@atomlin.com, frederic@kernel.org, tglx@linutronix.de,
        mingo@kernel.org, peterz@infradead.org, pauld@redhat.com,
        neelx@redhat.com, oleksandr@natalenko.name,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH v13 2/6] mm/vmstat: Use vmstat_dirty to track CPU-specific
 vmstat discrepancies
In-Reply-To: <20230105125248.813825852@redhat.com>
Message-ID: <b89a9828-d4e-9874-d482-dbb6cbe46@gentwo.de>
References: <20230105125218.031928326@redhat.com> <20230105125248.813825852@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Thu, 5 Jan 2023, Marcelo Tosatti wrote:

> --- linux-2.6.orig/mm/vmstat.c
> +++ linux-2.6/mm/vmstat.c
> @@ -381,6 +381,7 @@ void __mod_zone_page_state(struct zone *
>  		x = 0;
>  	}
>  	__this_cpu_write(*p, x);
> +	vmstat_mark_dirty();

__vmstat_mark_dirty()? See earlier patch

>
>  	preempt_enable_nested();
>  }
> @@ -417,6 +418,7 @@ void __mod_node_page_state(struct pglist
>  		x = 0;
>  	}
>  	__this_cpu_write(*p, x);
> +	vmstat_mark_dirty();

Ditto.

>
>  	preempt_enable_nested();
>  }
> @@ -577,6 +579,9 @@ static inline void mod_zone_state(struct
>  	s8 __percpu *p = pcp->vm_stat_diff + item;
>  	long o, n, t, z;
>
> +	/* cmpxchg and vmstat_mark_dirty should happen on the same CPU */
> +	preempt_disable();

If you are disabling preemption then why do we still need cmpxchg?
Same again below.