[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200429055125-mutt-send-email-mst@kernel.org>
Date: Wed, 29 Apr 2020 05:52:05 -0400
From: "Michael S. Tsirkin" <mst@...hat.com>
To: Srivatsa Vaddagiri <vatsa@...eaurora.org>
Cc: Lu Baolu <baolu.lu@...ux.intel.com>, tsoni@...eaurora.org,
virtio-dev@...ts.oasis-open.org, konrad.wilk@...cle.com,
jan.kiszka@...mens.com, jasowang@...hat.com,
christoffer.dall@....com,
virtualization@...ts.linux-foundation.org, alex.bennee@...aro.org,
iommu@...ts.linux-foundation.org, stefano.stabellini@...inx.com,
will@...nel.org, linux-kernel@...r.kernel.org,
pratikp@...eaurora.org
Subject: Re: [PATCH 5/5] virtio: Add bounce DMA ops
On Wed, Apr 29, 2020 at 03:14:10PM +0530, Srivatsa Vaddagiri wrote:
> * Michael S. Tsirkin <mst@...hat.com> [2020-04-29 02:50:41]:
>
> > So it seems that with modern Linux, all one needs
> > to do on x86 is mark the device as untrusted.
> > It's already possible to do this with ACPI and with OF - would that be
> > sufficient for achieving what this patchset is trying to do?
>
> In my case, its not sufficient to just mark virtio device untrusted and thus
> activate the use of swiotlb. All of the secondary VM memory, including those
> allocate by swiotlb driver, is private to it.
So why not make the bounce buffer memory shared then?
> An additional piece of memory is
> available to secondary VM which is shared between VMs and which is where I need
> swiotlb driver to do its work.
>
> --
> QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
> of Code Aurora Forum, hosted by The Linux Foundation
Powered by blists - more mailing lists