[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <trinity-a348af9b-9ef5-47a9-a14e-94265c880cf1-1493134775614@3capp-gmx-bs78>
Date: Tue, 25 Apr 2017 17:39:35 +0200
From: "Lino Sanfilippo" <LinoSanfilippo@....de>
To: "Alexander Duyck" <alexander.duyck@...il.com>
Cc: "Singh, Krishneil K" <krishneil.k.singh@...el.com>,
"Song, Liwei (Wind River)" <liwei.song@...driver.com>,
"Kirsher, Jeffrey T" <jeffrey.t.kirsher@...el.com>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"intel-wired-lan@...ts.osuosl.org" <intel-wired-lan@...ts.osuosl.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Aw: Re: [Intel-wired-lan] [PATCH] ixgbe: initialize u64_stats_sync
structures early at ixgbe_probe
Hi,
> This patch doesn't look right to me. I would suggest rejecting it.
>
> The call to initialize the stats should be done when the ring is
> allocated, not in ixgbe_probe(). This should probably be done in
> ixgbe_alloc_q_vector() instead.
>
AFAICS ixgbe_alloc_q_vector() is also called in probe() (by ixgbe_init_interrupt_scheme()).
Furthermore it is also called in resume() which would lead to multiple initialization of
the u64_stats_sync in case of resume.
IMHO the u64_stats_sync variables have to be initialized before register_netdev() is called
since this is the point from which userspace can call ixgbe_get_stats64(). I would say the
best place to do so is the probe() function as it is done in this patch.
Just my 2 cents.
Regards,
Lino
Powered by blists - more mailing lists