iommu/arm-smmu-v3: Warn about missing IRQs

PATCH Superseded Robin Murphy 2 participants 2 messages Oct 27, 2017 IOMMU

Share & tools

Robin Murphy

author reviewer

2017-10-27 13:08 UTC #0 lore raw

PATCH 1/1 iommu/arm-smmu-v3: Warn about missing IRQs

It is annoyingly non-obvious when DMA transactions silently go missing due to undetected SMMU faults. Help skip the first few debugging steps in those situations by making it clear when we have neither wired IRQs nor MSIs with which to raise error conditions.

Signed-off-by Robin Murphy <[email protected]>

---
 drivers/iommu/arm-smmu-v3.c | 4 ++++
 1 file changed, 4 insertions(+)

drivers/iommu/arm-smmu-v3.c +4 -0

index e67ba6c40faf..1a374c5c9507 100644
@@ -2263,6 +2263,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 						"arm-smmu-v3-evtq", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable evtq irq\n");
+	} else {
+		dev_warn(smmu->dev, "no evtq irq - events will not be reported!\n");
 	}
 
 	irq = smmu->cmdq.q.irq;

@@ -2280,6 +2282,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 				       0, "arm-smmu-v3-gerror", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable gerror irq\n");
+	} else {
+		dev_warn(smmu->dev, "no gerr irq - errors will not be reported!\n");
 	}
 
 	if (smmu->features & ARM_SMMU_FEAT_PRI) {

Signature · 1 line

2.13.4.dirty

Will Deacon

2017-10-27 16:52 UTC #1 lore raw

PATCH 1/1 iommu/arm-smmu-v3: Warn about missing IRQs

Quoted 24 lines · Robin Murphy

It is annoyingly non-obvious when DMA transactions silently go missing due to undetected SMMU faults. Help skip the first few debugging steps in those situations by making it clear when we have neither wired IRQs nor MSIs with which to raise error conditions.
Signed-off-by Robin Murphy <[email protected]>
---
 drivers/iommu/arm-smmu-v3.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/iommu/arm-smmu-v3.c +2 -0
index e67ba6c40faf..1a374c5c9507 100644
@@ -2263,6 +2263,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 						"arm-smmu-v3-evtq", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable evtq irq\n");
+	} else {
+		dev_warn(smmu->dev, "no evtq irq - events will not be reported!\n");
 	}
 
 	irq = smmu->cmdq.q.irq;

drivers/iommu/arm-smmu-v3.c static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu) · line 2289

Quoted 8 lines

drivers/iommu/arm-smmu-v3.c +2 -0

@@ -2280,6 +2282,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 				       0, "arm-smmu-v3-gerror", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable gerror irq\n");
+	} else {
+		dev_warn(smmu->dev, "no gerr irq - errors will not be reported!\n");
 	}
 
 	if (smmu->features & ARM_SMMU_FEAT_PRI) {

Can we also print something similar for the PRIQ when we detect that PRI is supported, please? Whilst I agree that event and gerror are the critical ones, I'm not sure how a PCIe device behaves if its PRI requests are ignored, but this might help to debug it.

Will