[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACGkMEuAfvAY8nc2fjHQ007bRPE=pUgDPPgMjhkeA34JTQKSfw@mail.gmail.com>
Date: Sun, 26 Sep 2021 12:19:30 +0800
From: Jason Wang <jasowang@...hat.com>
To: Wu Zongyong <wuzongyong@...ux.alibaba.com>
Cc: virtualization <virtualization@...ts.linux-foundation.org>,
linux-kernel <linux-kernel@...r.kernel.org>,
mst <mst@...hat.com>, wei.yang1@...ux.alibaba.com
Subject: Re: [PATCH v3 7/7] eni_vdpa: add vDPA driver for Alibaba ENI
On Sun, Sep 26, 2021 at 11:27 AM Wu Zongyong
<wuzongyong@...ux.alibaba.com> wrote:
>
> On Sun, Sep 26, 2021 at 10:26:47AM +0800, Jason Wang wrote:
> >
> > 在 2021/9/22 下午8:46, Wu Zongyong 写道:
> > > +
> > > +#ifdef __LITTLE_ENDIAN
> >
> >
> > I think disable the device via Kconfig is better than letting user to meet
> > errors like this.
> >
> > (Or if the device is always using little endian, we don't even need to
> > bother this).
>
> I prefer the second suggestion since there are no use cases that the
> device uses big endian
If this means the device will always use little endian. It's fine.
Thanks
> >
> > Thanks
> >
> >
> > > +static int eni_vdpa_probe(struct pci_dev *pdev, const struct pci_device_id *id)
> > > +{
> > > + struct device *dev = &pdev->dev;
> > > + struct eni_vdpa *eni_vdpa;
> > > + struct virtio_pci_legacy_device *ldev;
> > > + int ret, i;
> > > +
> > > + ret = pcim_enable_device(pdev);
> > > + if (ret)
> > > + return ret;
> > > +
>
Powered by blists - more mailing lists