Linux-ARM-Kernel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Pranjal Shrivastava <praan@google.com>
To: iommu@lists.linux.dev
Cc: Will Deacon <will@kernel.org>, Joerg Roedel <joro@8bytes.org>,
	 Robin Murphy <robin.murphy@arm.com>,
	Jason Gunthorpe <jgg@ziepe.ca>,
	Mostafa Saleh <smostafa@google.com>,
	 Nicolin Chen <nicolinc@nvidia.com>,
	Daniel Mentz <danielmentz@google.com>,
	 Ashish Mhetre <amhetre@nvidia.com>,
	linux-arm-kernel@lists.infradead.org,
	 Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	rafael@kernel.org,  Danilo Krummrich <dakr@kernel.org>,
	Thomas Gleixner <tglx@kernel.org>,
	driver-core@lists.linux.dev,
	 Pranjal Shrivastava <praan@google.com>
Subject: [PATCH v10 08/15] iommu/arm-smmu-v3: Cache and restore MSI config
Date: Tue,  8 Sep 2026 17:17:04 +0000	[thread overview]
Message-ID: <20260908171712.356645-9-praan@google.com> (raw)
In-Reply-To: <20260908171712.356645-1-praan@google.com>

The SMMU's MSI configuration registers (*_IRQ_CFGn) containing target
address, data and memory attributes lose their state when the SMMU is
powered down. We'll need to cache and restore their contents to ensure
that MSIs work after the system resumes. Address this, by caching the
original `msi_msg` within the `msi_desc` when the config is first written
by `arm_smmu_write_msi_msg`. This primarily includes the target address
and data since the memory attributes are fixed.

Introduce a new helper `arm_smmu_resume_msis` which will later be called
during the driver's resume callback. The helper re-configures the IRQCFG
registers for each relevant interrupt (evtq, gerr, priq) via the new
platform_device_msi_rewrite() helper.

In addition, clear ARM_SMMU_FEAT_MSI if MSI setup fails and the driver
falls back to wired IRQs, preventing spurious MSI resume attempts.

Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-by: Mostafa Saleh <smostafa@google.com>
Signed-off-by: Pranjal Shrivastava <praan@google.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 32 +++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index dc2fc02b5163..a8166e75db0b 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -4771,6 +4771,9 @@ static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
 	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
 	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->msi_index];
 
+	/* Cache the msi_msg for resume */
+	desc->msg = *msg;
+
 	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
 	doorbell &= MSI_CFG0_ADDR_MASK;
 
@@ -4779,6 +4782,31 @@ static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
 	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
 }
 
+static void __maybe_unused arm_smmu_resume_msis(struct arm_smmu_device *smmu)
+{
+	/* Clear the MSI address regs as they reset to unknown value */
+	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
+	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
+
+	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
+		return;
+
+	if (!smmu->dev->msi.domain) {
+		dev_err(smmu->dev, "msi_domain absent during resume\n");
+		smmu->features &= ~ARM_SMMU_FEAT_MSI;
+		return;
+	}
+
+	platform_device_msi_rewrite(smmu->dev, smmu->gerr_irq, arm_smmu_write_msi_msg);
+	platform_device_msi_rewrite(smmu->dev, smmu->evtq.q.irq, arm_smmu_write_msi_msg);
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		platform_device_msi_rewrite(smmu->dev, smmu->priq.q.irq, arm_smmu_write_msi_msg);
+}
+
 static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
 {
 	int ret, nvec = ARM_SMMU_MAX_MSIS;
@@ -4798,6 +4826,7 @@ static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
 
 	if (!dev->msi.domain) {
 		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
+		smmu->features &= ~ARM_SMMU_FEAT_MSI;
 		return;
 	}
 
@@ -4805,6 +4834,7 @@ static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
 	ret = platform_device_msi_init_and_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
 	if (ret) {
 		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
+		smmu->features &= ~ARM_SMMU_FEAT_MSI;
 		return;
 	}
 
@@ -4918,6 +4948,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
 
 	irq = smmu->combined_irq;
 	if (irq) {
+		/* MSIs are unsupported for combined irq cases */
+		smmu->features &= ~ARM_SMMU_FEAT_MSI;
 		/*
 		 * Cavium ThunderX2 implementation doesn't support unique irq
 		 * lines. Use a single irq line for all the SMMUv3 interrupts.
-- 
2.55.0.979.g7e5102b832-goog



  parent reply	other threads:[~2026-09-08 17:17 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-08 17:16 [PATCH v10 00/15] iommu/arm-smmu-v3: Implement Runtime/System Sleep ops Pranjal Shrivastava
2026-09-08 17:16 ` [PATCH v10 01/15] iommu/arm-smmu-v3: Refactor arm_smmu_setup_irqs Pranjal Shrivastava
2026-09-08 17:16 ` [PATCH v10 02/15] iommu/arm-smmu-v3: Add Q_POS() macro Pranjal Shrivastava
2026-09-08 17:16 ` [PATCH v10 03/15] iommu/arm-smmu-v3: Add arm_smmu_drain_queue() helper Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 04/15] iommu/tegra241-cmdqv: Add a helper to drain VCMDQs Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 05/15] iommu/arm-smmu-v3: Add a helper to drain cmd queues Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 06/15] iommu/tegra241-cmdqv: Restore PROD and CONS after resume Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 07/15] platform-msi: Introduce platform_device_msi_rewrite() Pranjal Shrivastava
2026-09-08 19:40   ` Thomas Gleixner
2026-09-08 20:15     ` Pranjal Shrivastava
2026-09-08 20:16       ` Pranjal Shrivastava
2026-09-09  9:20       ` Thomas Gleixner
2026-09-08 22:55     ` Jason Gunthorpe
2026-09-08 17:17 ` Pranjal Shrivastava [this message]
2026-09-08 19:56   ` [PATCH v10 08/15] iommu/arm-smmu-v3: Cache and restore MSI config Thomas Gleixner
2026-09-08 20:23     ` Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 09/15] iommu/arm-smmu-v3: Factor out arm_smmu_handle_gerror() Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 10/15] iommu/arm-smmu-v3: Add CMDQ_PROD_STOP_FLAG to gate CMDQ submissions Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 11/15] iommu/tegra241-cmdqv: Add a helper to quiesce VCMDQs Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 12/15] iommu/arm-smmu-v3: Implement pm_runtime & system sleep ops Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 13/15] iommu/arm-smmu-v3: Enable pm_runtime and setup devlinks Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 14/15] iommu/arm-smmu-v3: Invoke pm_runtime before hw access Pranjal Shrivastava
2026-09-08 17:17 ` [PATCH v10 15/15] iommu/arm-smmu-v3: Add KUnit unit tests for Runtime PM Pranjal Shrivastava

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=20260908171712.356645-9-praan@google.com \
    --to=praan@google.com \
    --cc=amhetre@nvidia.com \
    --cc=dakr@kernel.org \
    --cc=danielmentz@google.com \
    --cc=driver-core@lists.linux.dev \
    --cc=gregkh@linuxfoundation.org \
    --cc=iommu@lists.linux.dev \
    --cc=jgg@ziepe.ca \
    --cc=joro@8bytes.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=nicolinc@nvidia.com \
    --cc=rafael@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=smostafa@google.com \
    --cc=tglx@kernel.org \
    --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