public inbox for iommu@lists.linux-foundation.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@nvidia.com>
To: 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>
Cc: Alex Williamson <alex.williamson@redhat.com>,
	Lu Baolu <baolu.lu@linux.intel.com>,
	Donald Dutile <ddutile@redhat.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: [PATCH v3 09/11] PCI: Enable ACS Enhanced bits for enable_acs and config_acs
Date: Fri,  5 Sep 2025 15:06:24 -0300	[thread overview]
Message-ID: <9-v3-8827cc7fc4e0+23f-pcie_switch_groups_jgg@nvidia.com> (raw)
In-Reply-To: <0-v3-8827cc7fc4e0+23f-pcie_switch_groups_jgg@nvidia.com>

The ACS Enhanced bits are intended to address a lack of precision in the
spec about what ACS P2P Request Redirect is supposed to do. While Linux
has long assumed that PCI_ACS_RR would cover MMIO BARs located in the root
port and PCIe Switch ports, the spec took the position that it is
implementation specific.

To get the behavior Linux has long assumed it should be setting:

  PCI_ACS_RR | PCI_ACS_DSP_MT_RR | PCI_ACS_USP_MT_RR | PCI_ACS_UNCLAMED_RR

Follow this guidance in enable_acs and set the additional bits if ACS
Enhanced is supported.

Allow config_acs to control these bits if the device has ACS Enhanced.

The spec permits the HW to wire the bits, so after setting them
pci_acs_flags_enabled() does do a pci_read_config_word() to read the
actual value in effect.

Note that currently Linux sets these bits to 0, so any new HW that comes
supporting ACS Enhanced will end up with historical Linux disabling these
functions. Devices wanting to be compatible with old Linux will need to
wire the ctrl bits to follow ACS_RR. Devices that implement ACS Enhanced
and support the ctrl=0 behavior will break PASID SVA support and VFIO
isolation when ACS_RR is enabled.

Due to the above I strongly encourage backporting this change otherwise
old kernels may have issues with new generations of PCI switches.

Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
---
 drivers/pci/pci.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index b0f4d98036cddd..983f71211f0055 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -957,6 +957,7 @@ static void __pci_config_acs(struct pci_dev *dev, struct pci_acs *caps,
 			     const char *p, const u16 acs_mask, const u16 acs_flags)
 {
 	u16 flags = acs_flags;
+	u16 supported_flags;
 	u16 mask = acs_mask;
 	char *delimit;
 	int ret = 0;
@@ -1001,8 +1002,14 @@ static void __pci_config_acs(struct pci_dev *dev, struct pci_acs *caps,
 			}
 		}
 
-		if (mask & ~(PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR | PCI_ACS_CR |
-			    PCI_ACS_UF | PCI_ACS_EC | PCI_ACS_DT)) {
+		supported_flags = PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR |
+				  PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_EC |
+				  PCI_ACS_DT;
+		if (caps->cap & PCI_ACS_ENHANCED)
+			supported_flags |= PCI_ACS_USP_MT_RR |
+					   PCI_ACS_DSP_MT_RR |
+					   PCI_ACS_UNCLAIMED_RR;
+		if (mask & ~supported_flags) {
 			pci_err(dev, "Invalid ACS flags specified\n");
 			return;
 		}
@@ -1062,6 +1069,14 @@ static void pci_std_enable_acs(struct pci_dev *dev, struct pci_acs *caps)
 	/* Upstream Forwarding */
 	caps->ctrl |= (caps->cap & PCI_ACS_UF);
 
+	/*
+	 * USP/DSP Memory Target Access Control and Unclaimed Request Redirect
+	 */
+	if (caps->cap & PCI_ACS_ENHANCED) {
+		caps->ctrl |= PCI_ACS_USP_MT_RR | PCI_ACS_DSP_MT_RR |
+			      PCI_ACS_UNCLAIMED_RR;
+	}
+
 	/* Enable Translation Blocking for external devices and noats */
 	if (pci_ats_disabled() || dev->external_facing || dev->untrusted)
 		caps->ctrl |= (caps->cap & PCI_ACS_TB);
-- 
2.43.0


  parent reply	other threads:[~2025-09-05 18:06 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-05 18:06 [PATCH v3 00/11] Fix incorrect iommu_groups with PCIe ACS Jason Gunthorpe
2025-09-05 18:06 ` [PATCH v3 01/11] PCI: Move REQ_ACS_FLAGS into pci_regs.h as PCI_ACS_ISOLATED Jason Gunthorpe
2025-09-09  4:08   ` Donald Dutile
2025-09-05 18:06 ` [PATCH v3 02/11] PCI: Add pci_bus_isolated() Jason Gunthorpe
2025-09-09  4:09   ` Donald Dutile
2025-09-09 19:54   ` Bjorn Helgaas
2025-09-09 21:21     ` Jason Gunthorpe
2025-09-05 18:06 ` [PATCH v3 03/11] iommu: Compute iommu_groups properly for PCIe switches Jason Gunthorpe
2025-09-09  4:14   ` Donald Dutile
2025-09-09 12:18     ` Jason Gunthorpe
2025-09-09 19:33       ` Donald Dutile
2025-09-09 20:27   ` Bjorn Helgaas
2025-09-09 21:21     ` Jason Gunthorpe
2025-09-05 18:06 ` [PATCH v3 04/11] iommu: Organize iommu_group by member size Jason Gunthorpe
2025-09-09  4:16   ` Donald Dutile
2025-09-05 18:06 ` [PATCH v3 05/11] PCI: Add pci_reachable_set() Jason Gunthorpe
2025-09-09 21:03   ` Bjorn Helgaas
2025-09-10 16:13     ` Jason Gunthorpe
2025-09-11 19:56     ` Donald Dutile
2025-09-15 13:38       ` Jason Gunthorpe
2025-09-15 14:32         ` Donald Dutile
2025-09-05 18:06 ` [PATCH v3 06/11] iommu: Compute iommu_groups properly for PCIe MFDs Jason Gunthorpe
2025-09-09  4:57   ` Donald Dutile
2025-09-09 13:31     ` Jason Gunthorpe
2025-09-09 19:55       ` Donald Dutile
2025-09-09 21:24   ` Bjorn Helgaas
2025-09-09 23:20     ` Jason Gunthorpe
2025-09-10  1:59     ` Donald Dutile
2025-09-10 17:43       ` Jason Gunthorpe
2025-09-05 18:06 ` [PATCH v3 07/11] iommu: Validate that pci_for_each_dma_alias() matches the groups Jason Gunthorpe
2025-09-09  5:00   ` Donald Dutile
2025-09-09 15:35     ` Jason Gunthorpe
2025-09-09 19:58       ` Donald Dutile
2025-09-05 18:06 ` [PATCH v3 08/11] PCI: Add the ACS Enhanced Capability definitions Jason Gunthorpe
2025-09-09  5:01   ` Donald Dutile
2025-09-05 18:06 ` Jason Gunthorpe [this message]
2025-09-09  5:01   ` [PATCH v3 09/11] PCI: Enable ACS Enhanced bits for enable_acs and config_acs Donald Dutile
2025-09-05 18:06 ` [PATCH v3 10/11] PCI: Check ACS DSP/USP redirect bits in pci_enable_pasid() Jason Gunthorpe
2025-09-09  5:02   ` Donald Dutile
2025-09-09 21:43   ` Bjorn Helgaas
2025-09-10 17:34     ` Jason Gunthorpe
2025-09-11 19:50       ` Donald Dutile
2026-01-20 18:08   ` Keith Busch
2025-09-05 18:06 ` [PATCH v3 11/11] PCI: Check ACS Extended flags for pci_bus_isolated() Jason Gunthorpe
2025-09-09  5:04   ` Donald Dutile
2025-09-15  9:41 ` [PATCH v3 00/11] Fix incorrect iommu_groups with PCIe ACS Cédric Le Goater
2025-09-22 22:39 ` Alex Williamson
2025-09-23  1:44   ` Donald Dutile
2025-09-23  2:06     ` Alex Williamson
2025-09-23  2:42       ` Donald Dutile
2025-09-23 22:23         ` Alex Williamson
2025-09-30 15:23           ` Donald Dutile
2025-09-30 16:21             ` 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=9-v3-8827cc7fc4e0+23f-pcie_switch_groups_jgg@nvidia.com \
    --to=jgg@nvidia.com \
    --cc=alex.williamson@redhat.com \
    --cc=baolu.lu@linux.intel.com \
    --cc=bhelgaas@google.com \
    --cc=ddutile@redhat.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox