netdev - Re: [PATCH v3] net: Add mhi-net driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAMZdPi_2Fp8rRmgUCZcXo+1JScx_4bCWPt64ePtn2hRBYP5Q8Q@mail.gmail.com>
Date:   Thu, 15 Oct 2020 12:25:44 +0200
From:   Loic Poulain <loic.poulain@...aro.org>
To:     Eric Dumazet <eric.dumazet@...il.com>
Cc:     davem@...emloft.net, Jakub Kicinski <kuba@...nel.org>,
        netdev@...r.kernel.org, Hemant Kumar <hemantk@...eaurora.org>,
        Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
Subject: Re: [PATCH v3] net: Add mhi-net driver

On Thu, 15 Oct 2020 at 09:25, Eric Dumazet <eric.dumazet@...il.com> wrote:
>
>
>
> On 10/14/20 1:53 PM, Loic Poulain wrote:
> > This patch adds a new network driver implementing MHI transport for
> > network packets. Packets can be in any format, though QMAP (rmnet)
> > is the usual protocol (flow control + PDN mux).
> >
>
> ...
>
> > +static void mhi_net_rx_refill_work(struct work_struct *work)
> > +{
> > +     struct mhi_net_dev *mhi_netdev = container_of(work, struct mhi_net_dev,
> > +                                                   rx_refill.work);
> > +     struct net_device *ndev = mhi_netdev->ndev;
> > +     struct mhi_device *mdev = mhi_netdev->mdev;
> > +     struct sk_buff *skb;
> > +     int err;
> > +
> > +     do {
> > +             skb = netdev_alloc_skb(ndev, READ_ONCE(ndev->mtu));
> > +             if (unlikely(!skb))
> > +                     break;
>
> It is a bit strange you use READ_ONCE(ndev->mtu) here, but later
> re-read ndev->mtu
>
> It seems you need to store the READ_ONCE() in a temporary variable,
>
>                 unsigned int mtu = READ_ONCE(ndev->mtu);
>

Indeed, thanks.

> > +
> > +             err = mhi_queue_skb(mdev, DMA_FROM_DEVICE, skb, ndev->mtu,
> > +                                 MHI_EOT);
>
>                  s/ndev->mtu/mtu/
>
>
> > +             if (err) {
> > +                     if (unlikely(err != -ENOMEM))
> > +                             net_err_ratelimited("%s: Failed to queue RX buf (%d)\n",
> > +                                                 ndev->name, err);
> > +                     kfree_skb(skb);
> > +                     break;
> > +             }
> > +
> > +             atomic_inc(&mhi_netdev->stats.rx_queued);
>
> This is an unbound loop in the kernel ?
>
> Please add a :
>
>                 cond_resched();
>
> Before anyone gets hurt.

Will do, thanks.

>
>
> > +     } while (1);
> > +
> > +     /* If we're still starved of rx buffers, reschedule latter */
> > +     if (unlikely(!atomic_read(&mhi_netdev->stats.rx_queued)))
> > +             schedule_delayed_work(&mhi_netdev->rx_refill, HZ / 2);
> > +}
> > +
> >
>
>