On Thu, Oct 15, 2020 at 03:47:09PM +0800, Jason Wang wrote:
>
> On 2020/9/9 下午4:17, Jason Wang wrote:
> > After Linux kernel commit 61363c1474b1 ("iommu/vt-d: Enable ATS only
> > if the device uses page aligned address."), ATS will be only enabled
> > if device advertises a page aligned request.
>
On 2020/9/9 下午4:17, Jason Wang wrote:
After Linux kernel commit 61363c1474b1 ("iommu/vt-d: Enable ATS only
if the device uses page aligned address."), ATS will be only enabled
if device advertises a page aligned request.
Unfortunately, vhost-net is the only user and we don't advertise the
alig
On Thu, Sep 10, 2020 at 09:53:03AM +0800, Jason Wang wrote:
> > Maybe it would be good too that vhost provides real 4k-aligned addresses (in
> > vhost_iotlb_miss)? My understanding is that PCI_ATS_CAP_PAGE_ALIGNED will
> > be
> > more compatible than without the bit set.
>
>
> Yes, I've conside
On 2020/9/9 下午11:43, Peter Xu wrote:
On Wed, Sep 09, 2020 at 04:17:31PM +0800, Jason Wang wrote:
After Linux kernel commit 61363c1474b1 ("iommu/vt-d: Enable ATS only
if the device uses page aligned address."), ATS will be only enabled
if device advertises a page aligned request.
Unfortunately
On Wed, Sep 09, 2020 at 04:17:31PM +0800, Jason Wang wrote:
> After Linux kernel commit 61363c1474b1 ("iommu/vt-d: Enable ATS only
> if the device uses page aligned address."), ATS will be only enabled
> if device advertises a page aligned request.
>
> Unfortunately, vhost-net is the only user and
After Linux kernel commit 61363c1474b1 ("iommu/vt-d: Enable ATS only
if the device uses page aligned address."), ATS will be only enabled
if device advertises a page aligned request.
Unfortunately, vhost-net is the only user and we don't advertise the
aligned request capability in the past since b
