Bug#764162: [karl.beldan@gmail.com: Re: [PATCH 0/1] mv643xx_eth: Disable TSO by default]
On Tue, 4 Nov 2014 17:06:43 +0100 Karl Beldan <karl.beldan@gmail.com> wrote:
> Hi,
>
> I think this patch fixes this bug, if someone can try it.
>
> Karl
>
> ----- Forwarded message from Karl Beldan <karl.beldan@gmail.com> -----
>
> Date: Tue, 4 Nov 2014 15:20:20 +0100
> From: Karl Beldan <karl.beldan@gmail.com>
> To: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
> Cc: netdev@vger.kernel.org, David Miller <davem@davemloft.net>,
Thomas Petazzoni <thomas.petazzoni@free-electrons.com>, Gregory
> Clement <gregory.clement@free-electrons.com>, Tawfik Bayouk
<tawfik@marvell.com>, Lior Amsalem <alior@marvell.com>, Nadav
> Haklai <nadavh@marvell.com>
> Subject: Re: [PATCH 0/1] mv643xx_eth: Disable TSO by default
> User-Agent: Mutt/1.5.23 (2014-03-12)
>
> On Sat, Nov 01, 2014 at 12:30:19PM -0300, Ezequiel Garcia wrote:
> > Several users ([1], [2]) have been reporting data corruption with
TSO on
> > Kirkwood platforms (i.e. using the mv643xx_eth driver).
> >
> > Until we manage to find what's causing this, this simple patch will
make
> > the TSO path disabled by default. This patch should be queued for
stable,
> > fixing the TSO feature introduced in v3.16.
> >
> > The corruption itself is very easy to reproduce: checking md5sum on
a mounted
> > NFS directory gives a different result each time. Same tests using
the mvneta
> > driver (Armada 370/38x/XP SoC) pass with no issues.
> >
> > Frankly, I'm a bit puzzled about this, and so any ideas or
debugging hints
> > are well received.
> >
>
> Hi,
>
> Can you try this :
>
> @@ -1067,7 +1082,8 @@ static int txq_reclaim(struct tx_queue *txq,
int budget, int force)
> txq->tx_desc_count--;
>
> skb = NULL;
> - if (cmd_sts & TX_LAST_DESC)
> + if ((cmd_sts & (TX_LAST_DESC | TX_ENABLE_INTERRUPT)) ==
> + (TX_LAST_DESC | TX_ENABLE_INTERRUPT))
> skb = __skb_dequeue(&txq->tx_skb);
>
> if (cmd_sts & ERROR_SUMMARY) {
>
> --
> Karl
>
> ----- End forwarded message -----
>
>
Great! I test this patch and it seems to solve the data corruption that
I have
Julien
Reply to: