[ofa-general] Re: [PATCH] - [resend] Corrects a race in ipoib_cm_post_receive_nonsrq()

[Hoang-Nam Nguyen]

Hi,
Sorry for this very late finding as I was "blind" at the first review
some weeks ago.

> +static void ipoib_cm_init_rx_wr(struct net_device *dev,
> +            struct ib_recv_wr *wr,
> +            struct ib_sge *sge)
> +{
> +   struct ipoib_dev_priv *priv = netdev_priv(dev);
> +   int i;
> +
> +   for (i = 0; i < priv->cm.num_frags; ++i)
> +      sge[i].lkey = priv->mr->lkey;
> +
> +   sge[0].length = IPOIB_CM_HEAD_SIZE;
> +   for (i = 1; i < priv->cm.num_frags; ++i)
> +      sge[i].length = PAGE_SIZE;
> +
> +   wr->next    = NULL;
> +   wr->sg_list = priv->cm.rx_sge;
This should actually be set to sge as provided by the caller.
Will send a patch to fix this for 2.6.27 soon.
> +   wr->num_sge = priv->cm.num_frags;
> +}

Thanks
Nam