From: Jason Gunthorpe <jgg@nvidia.com>
To: Alex Williamson <alex.williamson@redhat.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
iommu@lists.linux.dev, Joerg Roedel <joro@8bytes.org>,
linux-pci@vger.kernel.org, Robin Murphy <robin.murphy@arm.com>,
Will Deacon <will@kernel.org>,
Lu Baolu <baolu.lu@linux.intel.com>,
galshalom@nvidia.com, Joerg Roedel <jroedel@suse.de>,
Kevin Tian <kevin.tian@intel.com>,
kvm@vger.kernel.org, maorg@nvidia.com, patches@lists.linux.dev,
tdave@nvidia.com, Tony Zhu <tony.zhu@intel.com>
Subject: Re: [PATCH v2 00/16] Fix incorrect iommu_groups with PCIe ACS
Date: Fri, 18 Jul 2025 19:59:39 -0300 [thread overview]
Message-ID: <20250718225939.GL2250220@nvidia.com> (raw)
In-Reply-To: <20250718152934.0cdb768f.alex.williamson@redhat.com>
On Fri, Jul 18, 2025 at 03:29:34PM -0600, Alex Williamson wrote:
> I can't say that the proposed code here is doing the wrong thing by
> propagating the lack of isolation, but it's gratuitous when there is no
> DMA initiator on the non-isolated branch and it causes a significant
> usage problem for vfio.
I think the answer is the quote Don found, if the MFD function has no
ACS cap then we assume it cannot do P2P:
e.g., Section 6.12.1.2 ACS Functions in SR-IOV, SIOV, and Multi-Function Devices
...
ACS P2P Request Redirect: **must** be implemented by Functions that
support peer-to-peer traffic with other Functions.
So I propose we lean into that. It should resolve everything you noted
here.
Instead the logic will be if any function with ACS has non-isolated
ACS then the whole MFD is in a group. This is alot closer to what it
does today, with the addition that explicit ACS non-isolation groups
correctly.
Jason
next prev parent reply other threads:[~2025-07-18 22:59 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-09 14:52 [PATCH v2 00/16] Fix incorrect iommu_groups with PCIe ACS Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 01/16] PCI: Move REQ_ACS_FLAGS into pci_regs.h as PCI_ACS_ISOLATED Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 02/16] PCI: Add pci_bus_isolation() Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 03/16] iommu: Compute iommu_groups properly for PCIe switches Jason Gunthorpe
2025-07-17 22:03 ` Donald Dutile
2025-07-18 18:09 ` Jason Gunthorpe
2025-07-18 19:00 ` Donald Dutile
2025-07-18 20:19 ` Jason Gunthorpe
2025-07-18 21:41 ` Donald Dutile
2025-07-18 22:52 ` Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 04/16] iommu: Organize iommu_group by member size Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 05/16] PCI: Add pci_reachable_set() Jason Gunthorpe
2025-07-17 22:04 ` Donald Dutile
2025-07-18 17:49 ` Jason Gunthorpe
2025-07-18 19:10 ` Donald Dutile
2025-07-09 14:52 ` [PATCH v2 06/16] PCI: Remove duplication in calling pci_acs_ctrl_enabled() Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 07/16] PCI: Use pci_quirk_mf_endpoint_acs() for pci_quirk_amd_sb_acs() Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 08/16] PCI: Use pci_acs_ctrl_isolated() for pci_quirk_al_acs() Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 09/16] PCI: Widen the acs_flags to u32 within the quirk callback Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 10/16] PCI: Add pci_mfd_isolation() Jason Gunthorpe
2025-07-10 11:58 ` kernel test robot
2025-08-20 17:21 ` Keith Busch
2025-07-09 14:52 ` [PATCH v2 11/16] iommu: Compute iommu_groups properly for PCIe MFDs Jason Gunthorpe
2025-07-28 9:47 ` Cédric Le Goater
2025-07-28 13:58 ` Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 12/16] iommu: Validate that pci_for_each_dma_alias() matches the groups Jason Gunthorpe
2025-07-17 22:07 ` Donald Dutile
2025-07-09 14:52 ` [PATCH v2 13/16] PCI: Add the ACS Enhanced Capability definitions Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 14/16] PCI: Enable ACS Enhanced bits for enable_acs and config_acs Jason Gunthorpe
2025-07-09 14:52 ` [PATCH v2 15/16] PCI: Check ACS DSP/USP redirect bits in pci_enable_pasid() Jason Gunthorpe
2025-07-17 22:17 ` Donald Dutile
2025-07-18 17:52 ` Jason Gunthorpe
2025-08-05 4:39 ` Askar Safin
2025-07-09 14:52 ` [PATCH v2 16/16] PCI: Check ACS Extended flags for pci_bus_isolated() Jason Gunthorpe
2025-07-18 21:29 ` [PATCH v2 00/16] Fix incorrect iommu_groups with PCIe ACS Alex Williamson
2025-07-18 22:59 ` Jason Gunthorpe [this message]
2025-08-02 1:45 ` Ethan Zhao
2025-08-02 15:18 ` Jason Gunthorpe
2025-08-05 3:43 ` Ethan Zhao
2025-08-05 12:35 ` Jason Gunthorpe
2025-08-05 14:41 ` Ethan Zhao
2025-08-05 14:43 ` Jason Gunthorpe
2025-08-06 2:22 ` Ethan Zhao
2025-08-06 2:41 ` Baolu Lu
2025-08-06 13:40 ` Jason Gunthorpe
2025-08-07 1:36 ` Ethan Zhao
2025-08-08 7:56 ` Ethan Zhao
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=20250718225939.GL2250220@nvidia.com \
--to=jgg@nvidia.com \
--cc=alex.williamson@redhat.com \
--cc=baolu.lu@linux.intel.com \
--cc=bhelgaas@google.com \
--cc=galshalom@nvidia.com \
--cc=iommu@lists.linux.dev \
--cc=joro@8bytes.org \
--cc=jroedel@suse.de \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=maorg@nvidia.com \
--cc=patches@lists.linux.dev \
--cc=robin.murphy@arm.com \
--cc=tdave@nvidia.com \
--cc=tony.zhu@intel.com \
--cc=will@kernel.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.