[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <00434866-a494-9253-6fdc-bcf634c69212@omp.ru>
Date: Wed, 20 Dec 2023 23:00:54 +0300
From: Sergey Shtylyov <s.shtylyov@....ru>
To: claudiu beznea <claudiu.beznea@...on.dev>, <davem@...emloft.net>,
<edumazet@...gle.com>, <kuba@...nel.org>, <pabeni@...hat.com>,
<richardcochran@...il.com>, <p.zabel@...gutronix.de>,
<yoshihiro.shimoda.uh@...esas.com>, <wsa+renesas@...g-engineering.com>,
<geert+renesas@...der.be>
CC: <netdev@...r.kernel.org>, <linux-renesas-soc@...r.kernel.org>,
<linux-kernel@...r.kernel.org>, Claudiu Beznea
<claudiu.beznea.uj@...renesas.com>
Subject: Re: [PATCH net-next v2 18/21] net: ravb: Return cached statistics if
the interface is down
On 12/17/23 4:54 PM, claudiu beznea wrote:
[...]
>>> From: Claudiu Beznea <claudiu.beznea.uj@...renesas.com>
>>>
>>> Return the cached statistics in case the interface is down. There should be
>>> no drawback to this, as most of the statistics are updated on the data path
>>> and if runtime PM is enabled and the interface is down, the registers that
>>> are explicitly read on ravb_get_stats() are zero anyway on most of the IP
>>> variants.
>>>
>>> The commit prepares the code for the addition of runtime PM support.
>>>
>>> Suggested-by: Sergey Shtylyov <s.shtylyov@....ru>
>>> Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@...renesas.com>
>>> ---
>>>
>>> Changes in v2:
>>> - none; this patch is new
>>>
>>> drivers/net/ethernet/renesas/ravb_main.c | 3 +++
>>> 1 file changed, 3 insertions(+)
>>>
>>> diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
>>> index a2a64c22ec41..1995cf7ff084 100644
>>> --- a/drivers/net/ethernet/renesas/ravb_main.c
>>> +++ b/drivers/net/ethernet/renesas/ravb_main.c
>>> @@ -2110,6 +2110,9 @@ static struct net_device_stats *ravb_get_stats(struct net_device *ndev)
>>> const struct ravb_hw_info *info = priv->info;
>>> struct net_device_stats *nstats, *stats0, *stats1;
>>>
>>> + if (!netif_running(ndev))
>>
>> I'm afraid this is racy as __LINK_STATE_START bit gets set
>> by __dev_open() before calling the ndo_open() method... :-(
>
> But (at least on my setup), both ndo_get_stats and ndo_open are called with
> rtnl_mutex locked.
Unfortunately, it's not always so -- see e.g. netstat_show() in net/core/net-sysfs.c...
[...]
MBR, Sergey
Powered by blists - more mailing lists