From: Jason Gunthorpe <jgg@nvidia.com>
To: "Suthikulpanit, Suravee" <suravee.suthikulpanit@amd.com>
Cc: nicolinc@nvidia.com, linux-kernel@vger.kernel.org,
robin.murphy@arm.com, will@kernel.org, joro@8bytes.org,
kevin.tian@intel.com, jsnitsel@redhat.com, vasant.hegde@amd.com,
iommu@lists.linux.dev, santosh.shukla@amd.com,
sairaj.arunkodilkar@amd.com, jon.grimm@amd.com,
prashanthpra@google.com, wvw@google.com, wnliu@google.com,
gptran@google.com, kpsingh@google.com, joao.m.martins@oracle.com,
alejandro.j.jimenez@oracle.com
Subject: Re: [PATCH v2 10/12] iommu/amd: Add support for nested domain allocation
Date: Tue, 7 Oct 2025 20:39:31 -0300 [thread overview]
Message-ID: <20251007233931.GD3474167@nvidia.com> (raw)
In-Reply-To: <fd4341ea-869f-459c-a3be-a98cd1c8ba4b@amd.com>
On Tue, Oct 07, 2025 at 03:36:58PM -0500, Suthikulpanit, Suravee wrote:
> The gDTE[DomainID] field contains guest Domain ID (gDomID). The host IOMMU
> driver uses the gDomId and guest ID (gid) to index the Domain ID mapping
> table, and store the host Domain ID (hDomID) in the table entry. This data
> structure is required by hw to translation gDomID->hDomID to virtualize
> guest invalidation command. This will be part of the upcoming series to
> enable hw-vIOMMU.
Sure, this translation is part of viommu
> This ndom->id is the hDomID, which is currently allocated per-device to
> avoid TLB aliasing i.e. A guest w/ multiple pass-through devices w/ the same
> hDomID (same stage 2 table) and different stage-1 tables with same PASID.
> IOMMU would use the same TLB tag, which results in TLB aliasing issue.
> Therefore, we workaround the issue by allocating per-device hDomID for
> nested domain.
But this is what I mean here, the gDomId should be 1:1 with the hDomId
and here you are making it 1:N.
It has to be like this or you cannot manage invalidation.
Given this series is not really functional it is OK to leave a little
hack I guess, but it is worth noting how it is supposed to work.
It also probably means we should see the viommu series pretty quickly
with a goal to merge them all together in one cycle.
Jason
next prev parent reply other threads:[~2025-10-07 23:39 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-01 6:09 [PATCH v2 00/12] iommu/amd: Introduce Nested Translation support Suravee Suthikulpanit
2025-10-01 6:09 ` [PATCH v2 01/12] iommu/amd: Rename DEV_DOMID_MASK to DTE_DOMID_MASK Suravee Suthikulpanit
2025-10-02 17:12 ` Nicolin Chen
2025-10-06 14:36 ` Jason Gunthorpe
2025-10-01 6:09 ` [PATCH v2 02/12] iommu/amd: Make amd_iommu_pdom_id_alloc() non-static Suravee Suthikulpanit
2025-10-02 17:12 ` Nicolin Chen
2025-10-01 6:09 ` [PATCH v2 03/12] iommu/amd: Make amd_iommu_pdom_id_free() non-static Suravee Suthikulpanit
2025-10-02 17:13 ` Nicolin Chen
2025-10-06 14:37 ` Jason Gunthorpe
2025-10-01 6:09 ` [PATCH v2 04/12] iommu/amd: Make amd_iommu_device_flush_dte() non-static Suravee Suthikulpanit
2025-10-02 17:14 ` Nicolin Chen
2025-10-06 14:37 ` Jason Gunthorpe
2025-10-01 6:09 ` [PATCH v2 05/12] iommu/amd: Make amd_iommu_update_dte256() non-static Suravee Suthikulpanit
2025-10-02 17:15 ` Nicolin Chen
2025-10-01 6:09 ` [PATCH v2 06/12] iommu/amd: Make amd_iommu_make_clear_dte() non-static inline Suravee Suthikulpanit
2025-10-02 17:17 ` Nicolin Chen
2025-10-01 6:09 ` [PATCH v2 07/12] iommu/amd: Make amd_iommu_completion_wait() non-static Suravee Suthikulpanit
2025-10-02 17:24 ` Nicolin Chen
2025-10-01 6:09 ` [PATCH v2 08/12] iommufd: Introduce data struct for AMD nested domain allocation Suravee Suthikulpanit
2025-10-02 17:31 ` Nicolin Chen
2025-10-01 6:09 ` [PATCH v2 09/12] iommu/amd: Add support for nest parent " Suravee Suthikulpanit
2025-10-02 18:00 ` Nicolin Chen
2025-10-06 14:43 ` Jason Gunthorpe
2025-10-08 14:16 ` Suthikulpanit, Suravee
2025-10-01 6:09 ` [PATCH v2 10/12] iommu/amd: Add support for nested " Suravee Suthikulpanit
2025-10-02 18:29 ` Nicolin Chen
2025-10-06 14:49 ` Jason Gunthorpe
2025-10-07 20:36 ` Suthikulpanit, Suravee
2025-10-07 23:39 ` Jason Gunthorpe [this message]
2025-10-09 6:22 ` Sairaj Kodilkar
2025-10-09 9:16 ` Sairaj Kodilkar
2025-10-09 14:34 ` Jason Gunthorpe
2025-10-01 6:09 ` [PATCH v2 11/12] iommu/amd: Add support for nested domain attach/detach Suravee Suthikulpanit
2025-10-02 19:04 ` Nicolin Chen
2025-10-06 14:59 ` Jason Gunthorpe
2025-10-07 19:22 ` Suthikulpanit, Suravee
2025-10-07 23:43 ` Jason Gunthorpe
2025-10-09 7:18 ` Sairaj Kodilkar
2025-10-09 14:35 ` Jason Gunthorpe
2025-10-01 6:09 ` [PATCH v2 12/12] iommu/amd: Introduce IOMMUFD vIOMMU support for AMD Suravee Suthikulpanit
2025-10-02 20:05 ` Nicolin Chen
2025-10-06 15:01 ` Jason Gunthorpe
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=20251007233931.GD3474167@nvidia.com \
--to=jgg@nvidia.com \
--cc=alejandro.j.jimenez@oracle.com \
--cc=gptran@google.com \
--cc=iommu@lists.linux.dev \
--cc=joao.m.martins@oracle.com \
--cc=jon.grimm@amd.com \
--cc=joro@8bytes.org \
--cc=jsnitsel@redhat.com \
--cc=kevin.tian@intel.com \
--cc=kpsingh@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nicolinc@nvidia.com \
--cc=prashanthpra@google.com \
--cc=robin.murphy@arm.com \
--cc=sairaj.arunkodilkar@amd.com \
--cc=santosh.shukla@amd.com \
--cc=suravee.suthikulpanit@amd.com \
--cc=vasant.hegde@amd.com \
--cc=will@kernel.org \
--cc=wnliu@google.com \
--cc=wvw@google.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox