[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20131220001858.GD7172@S2101-09.ap.freescale.net>
Date: Fri, 20 Dec 2013 08:18:59 +0800
From: Shawn Guo <shawn.guo@...aro.org>
To: Eric Dumazet <eric.dumazet@...il.com>
CC: "fugang.duan@...escale.com" <fugang.duan@...escale.com>,
Hector Palacios <hector.palacios@...i.com>,
Marek Vasut <marex@...x.de>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"Fabio.Estevam@...escale.com" <Fabio.Estevam@...escale.com>,
"l.stach@...gutronix.de" <l.stach@...gutronix.de>,
"Frank.Li@...escale.com" <Frank.Li@...escale.com>,
"bhutchings@...arflare.com" <bhutchings@...arflare.com>,
"davem@...emloft.net" <davem@...emloft.net>
Subject: Re: FEC performance degradation with certain packet sizes
On Thu, Dec 19, 2013 at 03:04:59PM -0800, Eric Dumazet wrote:
> On Thu, 2013-12-19 at 02:44 +0000, fugang.duan@...escale.com wrote:
>
> > Enet IP don't support tso feature.
> >
> > I will reproduce the issue in imx6q/dl sd platform, and analyze the issue.
> > Previous test, we don't use netpipe tool test ethernet performance.
>
> drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c seems broken...
The fec driver running on IMX is drivers/net/ethernet/freescale/fec_main.c
Shawn
>
> If tx_skb_align_workaround() returns NULL, returning NETDEV_TX_BUSY
> is the worst possible thing to do.
>
> Packet should be dropped instead.
>
> Probably not related to your issue.
>
>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists