[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <cc288a52-dc5e-1a59-6219-8835c898ea73@redhat.com>
Date: Fri, 27 Sep 2019 16:31:31 +0800
From: Jason Wang <jasowang@...hat.com>
To: Tiwei Bie <tiwei.bie@...el.com>
Cc: mst@...hat.com, alex.williamson@...hat.com,
maxime.coquelin@...hat.com, linux-kernel@...r.kernel.org,
kvm@...r.kernel.org, virtualization@...ts.linux-foundation.org,
netdev@...r.kernel.org, dan.daly@...el.com,
cunming.liang@...el.com, zhihong.wang@...el.com,
lingshan.zhu@...el.com
Subject: Re: [PATCH] vhost: introduce mdev based hardware backend
On 2019/9/27 下午4:04, Tiwei Bie wrote:
> On Fri, Sep 27, 2019 at 03:14:42PM +0800, Jason Wang wrote:
>> On 2019/9/27 下午12:54, Tiwei Bie wrote:
>>>>> +
>>>>> + /*
>>>>> + * In vhost-mdev, userspace should pass ring addresses
>>>>> + * in guest physical addresses when IOMMU is disabled or
>>>>> + * IOVAs when IOMMU is enabled.
>>>>> + */
>>>> A question here, consider we're using noiommu mode. If guest physical
>>>> address is passed here, how can a device use that?
>>>>
>>>> I believe you meant "host physical address" here? And it also have the
>>>> implication that the HPA should be continuous (e.g using hugetlbfs).
>>> The comment is talking about the virtual IOMMU (i.e. iotlb in vhost).
>>> It should be rephrased to cover the noiommu case as well. Thanks for
>>> spotting this.
>>
>> So the question still, if GPA is passed how can it be used by the
>> virtio-mdev device?
> Sorry if I didn't make it clear..
> Of course, GPA can't be passed in noiommu mode.
I see.
Thanks for the confirmation.
>
>
>> Thanks
>>
Powered by blists - more mailing lists