From: Jacob Pan <jacob.jun.pan-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
To: Jean-Philippe Brucker
<jean-philippe.brucker-5wv7dgnIgG8@public.gmane.org>
Cc: ashok.raj-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
sanjay.k.kumar-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
jacob.jun.pan-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
yi.y.sun-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org
Subject: Re: [PATCH 00/10] iommu/vt-d: Add scalable mode support
Date: Wed, 18 Jul 2018 17:47:02 -0700 [thread overview]
Message-ID: <20180718174702.31e1c1d6@jacob-builder> (raw)
In-Reply-To: <07fe2e3f-4f4a-58db-ee2a-2620183d93b2-5wv7dgnIgG8@public.gmane.org>
On Mon, 16 Jul 2018 11:51:57 +0100
Jean-Philippe Brucker <jean-philippe.brucker-5wv7dgnIgG8@public.gmane.org> wrote:
> Hi,
>
> On 16/07/18 07:49, Lu Baolu wrote:
> > Intel vt-d rev3.0 [1] introduces a new translation mode called
> > 'scalable mode', which enables PASID-granular translations for
> > first level, second level, nested and pass-through modes. The
> > vt-d scalable mode is the key ingredient to enable Scalable I/O
> > Virtualization (Scalable IOV) [2] [3], which allows sharing a
> > device in minimal possible granularity (ADI - Assignable Device
> > Interface). It also includes all the capabilities required to
> > enable Shared Virtual Addressing (SVA). As a result, previous
> > Extended Context (ECS) mode is deprecated (no production ever
> > implements ECS).
> >
> > Each scalable mode pasid table entry is 64 bytes in length, with
> > fields point to the first level page table and the second level
> > page table. The PGTT (Pasid Granular Translation Type) field is
> > used by hardware to determine the translation type.
>
> Looks promising! Since the 2nd level page tables are in the PASID
> entry, the hypervisor traps guest accesses to the PASID tables
> instead of passing through the whole PASID directory? Are you still
> planning to use the VFIO BIND_PASID_TABLE interface in this mode, or
> a slightly different one for individual PASIDs?
>
Since we deprecated ECS mode, there is no need for VT-d to bind guest
pasid table in scalable mode. We are planning on adding another flag for
bind guest pasid and guest CR3, Perhaps call it
VFIO_IOMMU_BIND_GUEST_SVA.
> Thanks,
> Jean
> _______________________________________________
> iommu mailing list
> iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org
> https://lists.linuxfoundation.org/mailman/listinfo/iommu
[Jacob Pan]
WARNING: multiple messages have this Message-ID (diff)
From: Jacob Pan <jacob.jun.pan@intel.com>
To: Jean-Philippe Brucker <jean-philippe.brucker@arm.com>
Cc: Lu Baolu <baolu.lu@linux.intel.com>,
Joerg Roedel <joro@8bytes.org>,
David Woodhouse <dwmw2@infradead.org>,
ashok.raj@intel.com, sanjay.k.kumar@intel.com,
iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org,
yi.y.sun@intel.com, jacob.jun.pan@intel.com
Subject: Re: [PATCH 00/10] iommu/vt-d: Add scalable mode support
Date: Wed, 18 Jul 2018 17:47:02 -0700 [thread overview]
Message-ID: <20180718174702.31e1c1d6@jacob-builder> (raw)
In-Reply-To: <07fe2e3f-4f4a-58db-ee2a-2620183d93b2@arm.com>
On Mon, 16 Jul 2018 11:51:57 +0100
Jean-Philippe Brucker <jean-philippe.brucker@arm.com> wrote:
> Hi,
>
> On 16/07/18 07:49, Lu Baolu wrote:
> > Intel vt-d rev3.0 [1] introduces a new translation mode called
> > 'scalable mode', which enables PASID-granular translations for
> > first level, second level, nested and pass-through modes. The
> > vt-d scalable mode is the key ingredient to enable Scalable I/O
> > Virtualization (Scalable IOV) [2] [3], which allows sharing a
> > device in minimal possible granularity (ADI - Assignable Device
> > Interface). It also includes all the capabilities required to
> > enable Shared Virtual Addressing (SVA). As a result, previous
> > Extended Context (ECS) mode is deprecated (no production ever
> > implements ECS).
> >
> > Each scalable mode pasid table entry is 64 bytes in length, with
> > fields point to the first level page table and the second level
> > page table. The PGTT (Pasid Granular Translation Type) field is
> > used by hardware to determine the translation type.
>
> Looks promising! Since the 2nd level page tables are in the PASID
> entry, the hypervisor traps guest accesses to the PASID tables
> instead of passing through the whole PASID directory? Are you still
> planning to use the VFIO BIND_PASID_TABLE interface in this mode, or
> a slightly different one for individual PASIDs?
>
Since we deprecated ECS mode, there is no need for VT-d to bind guest
pasid table in scalable mode. We are planning on adding another flag for
bind guest pasid and guest CR3, Perhaps call it
VFIO_IOMMU_BIND_GUEST_SVA.
> Thanks,
> Jean
> _______________________________________________
> iommu mailing list
> iommu@lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/iommu
[Jacob Pan]
next prev parent reply other threads:[~2018-07-19 0:47 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-16 6:49 [PATCH 00/10] iommu/vt-d: Add scalable mode support Lu Baolu
2018-07-16 6:49 ` Lu Baolu
2018-07-16 6:49 ` [PATCH 01/10] iommu/vt-d: Enumerate the scalable mode capability Lu Baolu
2018-07-16 6:49 ` [PATCH 02/10] iommu/vt-d: Manage scalalble mode PASID tables Lu Baolu
2018-07-16 6:49 ` [PATCH 03/10] iommu/vt-d: Move page table helpers into header Lu Baolu
2018-07-16 6:49 ` [PATCH 04/10] iommu/vt-d: Add second level page table interface Lu Baolu
2018-07-16 6:49 ` [PATCH 05/10] iommu/vt-d: Setup pasid entry for RID2PASID support Lu Baolu
2018-07-16 6:49 ` [PATCH 06/10] iommu/vt-d: Pass pasid table to context mapping Lu Baolu
2018-07-16 6:49 ` [PATCH 07/10] iommu/vt-d: Setup context and enable RID2PASID support Lu Baolu
2018-07-16 6:49 ` [PATCH 08/10] iommu/vt-d: Add first level page table interface Lu Baolu
2018-07-16 6:49 ` [PATCH 09/10] iommu/vt-d: Shared virtual address in scalable mode Lu Baolu
2018-07-16 6:49 ` [PATCH 10/10] iommu/vt-d: Remove deferred invalidation Lu Baolu
2018-07-16 10:51 ` [PATCH 00/10] iommu/vt-d: Add scalable mode support Jean-Philippe Brucker
[not found] ` <07fe2e3f-4f4a-58db-ee2a-2620183d93b2-5wv7dgnIgG8@public.gmane.org>
2018-07-17 1:47 ` Liu, Yi L
2018-07-17 1:47 ` Liu, Yi L
2018-07-19 0:47 ` Jacob Pan [this message]
2018-07-19 0:47 ` Jacob Pan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180718174702.31e1c1d6@jacob-builder \
--to=jacob.jun.pan-ral2jqcrhueavxtiumwx3w@public.gmane.org \
--cc=ashok.raj-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=jean-philippe.brucker-5wv7dgnIgG8@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=sanjay.k.kumar-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=yi.y.sun-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.