[PATCH 0/4] iommu/arm-smmu: current updates for 4.1

[PATCH 0/4] iommu/arm-smmu: current updates for 4.1

[Will Deacon]

Hi all,

I only have a small handful of arm-smmu patches on the radar for 4.1,
so here they are in case anybody has any comments.

Cheers,

Will

--->8

Robin Murphy (1):
  iommu/arm-smmu: set a more appropriate DMA mask

Will Deacon (3):
  iommu/arm-smmu: ensure CBA2R is initialised before CBAR on SMMUv2
  iommu/arm-smmu: handle multi-alias IOMMU groups for PCI devices
  iommu/io-pgtable-arm: avoid speculative walks through TTBR1

 drivers/iommu/arm-smmu.c       | 125 ++++++++++++++++++++++++++---------------
 drivers/iommu/io-pgtable-arm.c |   5 ++
 2 files changed, 85 insertions(+), 45 deletions(-)

-- 
2.1.4

[PATCH 1/4] iommu/arm-smmu: ensure CBA2R is initialised before CBAR on SMMUv2

[Will Deacon]

The VMID16 (8.1) extension to SMMUv2 added a 16-bit VMID16 field to the
CBA2R registers. Unfortunately, if software writes this field as zero
after setting an 8-bit VMID in a stage-2 CBAR, then the VMID may also be
overwritten with zero on some early implementations (the architecture
was later updated to fix this issue).

This patch ensures that we initialise CBA2R before CBAR, therefore
ensuring that the VMID is set correctly.

Tested-by: Manish Jaggi <mjaggi-M3mlKVOIwJVv6pq1l3V1OdBPR1lH4CV8@public.gmane.org>
Signed-off-by: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
---
 drivers/iommu/arm-smmu.c | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index fc13dd56953e..4abb831e24f0 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -730,6 +730,20 @@ static void arm_smmu_init_context_bank(struct arm_smmu_domain *smmu_domain,
 	stage1 = cfg->cbar != CBAR_TYPE_S2_TRANS;
 	cb_base = ARM_SMMU_CB_BASE(smmu) + ARM_SMMU_CB(smmu, cfg->cbndx);
 
+	if (smmu->version > ARM_SMMU_V1) {
+		/*
+		 * CBA2R.
+		 * *Must* be initialised before CBAR thanks to VMID16
+		 * architectural oversight affected some implementations.
+		 */
+#ifdef CONFIG_64BIT
+		reg = CBA2R_RW64_64BIT;
+#else
+		reg = CBA2R_RW64_32BIT;
+#endif
+		writel_relaxed(reg, gr1_base + ARM_SMMU_GR1_CBA2R(cfg->cbndx));
+	}
+
 	/* CBAR */
 	reg = cfg->cbar;
 	if (smmu->version == ARM_SMMU_V1)
@@ -747,16 +761,6 @@ static void arm_smmu_init_context_bank(struct arm_smmu_domain *smmu_domain,
 	}
 	writel_relaxed(reg, gr1_base + ARM_SMMU_GR1_CBAR(cfg->cbndx));
 
-	if (smmu->version > ARM_SMMU_V1) {
-		/* CBA2R */
-#ifdef CONFIG_64BIT
-		reg = CBA2R_RW64_64BIT;
-#else
-		reg = CBA2R_RW64_32BIT;
-#endif
-		writel_relaxed(reg, gr1_base + ARM_SMMU_GR1_CBA2R(cfg->cbndx));
-	}
-
 	/* TTBRs */
 	if (stage1) {
 		reg = pgtbl_cfg->arm_lpae_s1_cfg.ttbr[0];
-- 
2.1.4

[PATCH 2/4] iommu/arm-smmu: set a more appropriate DMA mask

[Will Deacon]

From: Robin Murphy <Robin.Murphy-5wv7dgnIgG8@public.gmane.org>

Since we use dma_map_page() as an architecture-independent means of
making page table updates visible to non-coherent SMMUs, we need to
have a suitable DMA mask set to discourage the DMA mapping layer from
creating bounce buffers and flushing those instead, if said page tables
happen to lie outside the default 32-bit mask.

Tested-by: Mitchel Humpherys <mitchelh-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
Signed-off-by: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>
[will: added error checking]
Signed-off-by: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
---
 drivers/iommu/arm-smmu.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 4abb831e24f0..161dd46999e2 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -1634,6 +1634,15 @@ static int arm_smmu_device_cfg_probe(struct arm_smmu_device *smmu)
 	size = arm_smmu_id_size_to_bits((id >> ID2_OAS_SHIFT) & ID2_OAS_MASK);
 	smmu->pa_size = size;
 
+	/*
+	 * What the page table walker can address actually depends on which
+	 * descriptor format is in use, but since a) we don't know that yet,
+	 * and b) it can vary per context bank, this will have to do...
+	 */
+	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(size)))
+		dev_warn(smmu->dev,
+			 "failed to set DMA mask for table walker\n");
+
 	if (smmu->version == ARM_SMMU_V1) {
 		smmu->va_size = smmu->ipa_size;
 		size = SZ_4K | SZ_2M | SZ_1G;
-- 
2.1.4

[PATCH 3/4] iommu/arm-smmu: handle multi-alias IOMMU groups for PCI devices

[Will Deacon]

IOMMU groups for PCI devices can correspond to multiple DMA aliases due
to things like ACS and PCI quirks.

This patch extends the ARM SMMU ->add_device callback so that we
consider all of the DMA aliases for a PCI IOMMU group, rather than
creating a separate group for each Requester ID.

Signed-off-by: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
---
 drivers/iommu/arm-smmu.c | 92 ++++++++++++++++++++++++++++++------------------
 1 file changed, 57 insertions(+), 35 deletions(-)

diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 161dd46999e2..6ac184669295 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -1330,61 +1330,83 @@ static void __arm_smmu_release_pci_iommudata(void *data)
 	kfree(data);
 }
 
-static int arm_smmu_add_device(struct device *dev)
+static int arm_smmu_add_pci_device(struct pci_dev *pdev)
 {
-	struct arm_smmu_device *smmu;
-	struct arm_smmu_master_cfg *cfg;
+	int i, ret;
+	u16 sid;
 	struct iommu_group *group;
-	void (*releasefn)(void *) = NULL;
-	int ret;
-
-	smmu = find_smmu_for_device(dev);
-	if (!smmu)
-		return -ENODEV;
+	struct arm_smmu_master_cfg *cfg;
 
-	group = iommu_group_alloc();
-	if (IS_ERR(group)) {
-		dev_err(dev, "Failed to allocate IOMMU group\n");
+	group = iommu_group_get_for_dev(&pdev->dev);
+	if (IS_ERR(group))
 		return PTR_ERR(group);
-	}
-
-	if (dev_is_pci(dev)) {
-		struct pci_dev *pdev = to_pci_dev(dev);
 
+	cfg = iommu_group_get_iommudata(group);
+	if (!cfg) {
 		cfg = kzalloc(sizeof(*cfg), GFP_KERNEL);
 		if (!cfg) {
 			ret = -ENOMEM;
 			goto out_put_group;
 		}
 
-		cfg->num_streamids = 1;
-		/*
-		 * Assume Stream ID == Requester ID for now.
-		 * We need a way to describe the ID mappings in FDT.
-		 */
-		pci_for_each_dma_alias(pdev, __arm_smmu_get_pci_sid,
-				       &cfg->streamids[0]);
-		releasefn = __arm_smmu_release_pci_iommudata;
-	} else {
-		struct arm_smmu_master *master;
-
-		master = find_smmu_master(smmu, dev->of_node);
-		if (!master) {
-			ret = -ENODEV;
-			goto out_put_group;
-		}
+		iommu_group_set_iommudata(group, cfg,
+					  __arm_smmu_release_pci_iommudata);
+	}
 
-		cfg = &master->cfg;
+	if (cfg->num_streamids >= MAX_MASTER_STREAMIDS) {
+		ret = -ENOSPC;
+		goto out_put_group;
 	}
 
-	iommu_group_set_iommudata(group, cfg, releasefn);
-	ret = iommu_group_add_device(group, dev);
+	/*
+	 * Assume Stream ID == Requester ID for now.
+	 * We need a way to describe the ID mappings in FDT.
+	 */
+	pci_for_each_dma_alias(pdev, __arm_smmu_get_pci_sid, &sid);
+	for (i = 0; i < cfg->num_streamids; ++i)
+		if (cfg->streamids[i] == sid)
+			break;
+
+	/* Avoid duplicate SIDs, as this can lead to SMR conflicts */
+	if (i == cfg->num_streamids)
+		cfg->streamids[cfg->num_streamids++] = sid;
 
+	return 0;
 out_put_group:
 	iommu_group_put(group);
 	return ret;
 }
 
+static int arm_smmu_add_platform_device(struct device *dev)
+{
+	struct iommu_group *group;
+	struct arm_smmu_master *master;
+	struct arm_smmu_device *smmu = find_smmu_for_device(dev);
+
+	if (!smmu)
+		return -ENODEV;
+
+	master = find_smmu_master(smmu, dev->of_node);
+	if (!master)
+		return -ENODEV;
+
+	/* No automatic group creation for platform devices */
+	group = iommu_group_alloc();
+	if (IS_ERR(group))
+		return PTR_ERR(group);
+
+	iommu_group_set_iommudata(group, &master->cfg, NULL);
+	return iommu_group_add_device(group, dev);
+}
+
+static int arm_smmu_add_device(struct device *dev)
+{
+	if (dev_is_pci(dev))
+		return arm_smmu_add_pci_device(to_pci_dev(dev));
+
+	return arm_smmu_add_platform_device(dev);
+}
+
 static void arm_smmu_remove_device(struct device *dev)
 {
 	iommu_group_remove_device(dev);
-- 
2.1.4

[PATCH 4/4] iommu/io-pgtable-arm: avoid speculative walks through TTBR1

[Will Deacon]

Although we set TCR.T1SZ to 0, the input address range covered by TTBR1
is actually calculated using T0SZ in this case on the ARM SMMU. This
could theoretically lead to speculative table walks through physical
address zero, leading to all sorts of fun and games if we have MMIO
regions down there.

This patch avoids the issue by setting EPD1 to disable walks through
the unused TTBR1 register.

Signed-off-by: Will Deacon <will.deacon-5wv7dgnIgG8@public.gmane.org>
---
 drivers/iommu/io-pgtable-arm.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c
index b610a8dee238..4e460216bd16 100644
--- a/drivers/iommu/io-pgtable-arm.c
+++ b/drivers/iommu/io-pgtable-arm.c
@@ -116,6 +116,8 @@
 #define ARM_32_LPAE_TCR_EAE		(1 << 31)
 #define ARM_64_LPAE_S2_TCR_RES1		(1 << 31)
 
+#define ARM_LPAE_TCR_EPD1		(1 << 23)
+
 #define ARM_LPAE_TCR_TG0_4K		(0 << 14)
 #define ARM_LPAE_TCR_TG0_64K		(1 << 14)
 #define ARM_LPAE_TCR_TG0_16K		(2 << 14)
@@ -621,6 +623,9 @@ arm_64_lpae_alloc_pgtable_s1(struct io_pgtable_cfg *cfg, void *cookie)
 	}
 
 	reg |= (64ULL - cfg->ias) << ARM_LPAE_TCR_T0SZ_SHIFT;
+
+	/* Disable speculative walks through TTBR1 */
+	reg |= ARM_LPAE_TCR_EPD1;
 	cfg->arm_lpae_s1_cfg.tcr = reg;
 
 	/* MAIRs */
-- 
2.1.4