iommu/vt-d: improve ITE fault handling if device was released

Message ID 20240229040724.3393042-1-haifeng.zhao@linux.intel.com
State New
Headers
Series iommu/vt-d: improve ITE fault handling if device was released |

Commit Message

Ethan Zhao Feb. 29, 2024, 4:07 a.m. UTC
  Break the loop to blindly retry the timeout ATS invalidation request
after ITE fault hit if device was released or isn't present anymore.

This is part of the followup of prior proposed patchset

https://do-db2.lkml.org/lkml/2024/2/22/350

Fixes: 6ba6c3a4cacf ("VT-d: add device IOTLB invalidation support")
Signed-off-by: Ethan Zhao <haifeng.zhao@linux.intel.com>
---
 drivers/iommu/intel/dmar.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)
  

Comments

Ethan Zhao Feb. 29, 2024, 4:08 a.m. UTC | #1
On 2/29/2024 12:07 PM, Ethan Zhao wrote:
> Break the loop to blindly retry the timeout ATS invalidation request
> after ITE fault hit if device was released or isn't present anymore.
>
> This is part of the followup of prior proposed patchset
>
> https://do-db2.lkml.org/lkml/2024/2/22/350
>
> Fixes: 6ba6c3a4cacf ("VT-d: add device IOTLB invalidation support")
> Signed-off-by: Ethan Zhao <haifeng.zhao@linux.intel.com>
> ---
>   drivers/iommu/intel/dmar.c | 25 +++++++++++++++++++++++++
>   1 file changed, 25 insertions(+)
>
> diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c
> index d14797aabb7a..d01d68205557 100644
> --- a/drivers/iommu/intel/dmar.c
> +++ b/drivers/iommu/intel/dmar.c
> @@ -1273,6 +1273,9 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>   {
>   	u32 fault;
>   	int head, tail;
> +	u64 iqe_err, ite_sid;
> +	struct device *dev = NULL;
> +	struct pci_dev *pdev = NULL;
>   	struct q_inval *qi = iommu->qi;
>   	int shift = qi_shift(iommu);
>   
> @@ -1317,6 +1320,13 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>   		tail = readl(iommu->reg + DMAR_IQT_REG);
>   		tail = ((tail >> shift) - 1 + QI_LENGTH) % QI_LENGTH;
>   
> +		/*
> +		 * SID field is valid only when the ITE field is Set in FSTS_REG
> +		 * see Intel VT-d spec r4.1, section 11.4.9.9
> +		 */
> +		iqe_err = dmar_readq(iommu->reg + DMAR_IQER_REG);
> +		ite_sid = DMAR_IQER_REG_ITESID(iqe_err);
> +
>   		writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG);
>   		pr_info("Invalidation Time-out Error (ITE) cleared\n");
>   
> @@ -1326,6 +1336,21 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>   			head = (head - 2 + QI_LENGTH) % QI_LENGTH;
>   		} while (head != tail);
>   
> +		/*
> +		 * If got ITE, we need to check if the sid of ITE is one of the
> +		 * current valid ATS invalidation target devices, if no, or the
> +		 * target device isn't presnet, don't try this request anymore.
> +		 * 0 value of ite_sid means old VT-d device, no ite_sid value.
> +		 */
> +		if (ite_sid) {
> +			dev = device_rbtree_find(iommu, ite_sid);
> +			if (!dev || !dev_is_pci(dev))
> +				return -ETIMEDOUT;
> +			pdev = to_pci_dev(dev);
> +			if (!pci_device_is_present(pdev) &&
> +				ite_sid == pci_dev_id(pci_physfn(pdev)))
> +				return -ETIMEDOUT;
> +		}
>   		if (qi->desc_status[wait_index] == QI_ABORT)
>   			return -EAGAIN;
>   	}

This patch based on Baolu's per iommu device rbtree patchset

https://github.com/LuBaolu/intel-iommu/commits/rbtree-for-device-info-v2

Thanks,
Ethan
  
Bjorn Helgaas Feb. 29, 2024, 9:07 p.m. UTC | #2
On Wed, Feb 28, 2024 at 11:07:24PM -0500, Ethan Zhao wrote:
> Break the loop to blindly retry the timeout ATS invalidation request
> after ITE fault hit if device was released or isn't present anymore.
> 
> This is part of the followup of prior proposed patchset
> 
> https://do-db2.lkml.org/lkml/2024/2/22/350

Use lore URL, please.

> Fixes: 6ba6c3a4cacf ("VT-d: add device IOTLB invalidation support")
> Signed-off-by: Ethan Zhao <haifeng.zhao@linux.intel.com>
> ---
>  drivers/iommu/intel/dmar.c | 25 +++++++++++++++++++++++++
>  1 file changed, 25 insertions(+)
> 
> diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c
> index d14797aabb7a..d01d68205557 100644
> --- a/drivers/iommu/intel/dmar.c
> +++ b/drivers/iommu/intel/dmar.c
> @@ -1273,6 +1273,9 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>  {
>  	u32 fault;
>  	int head, tail;
> +	u64 iqe_err, ite_sid;
> +	struct device *dev = NULL;
> +	struct pci_dev *pdev = NULL;
>  	struct q_inval *qi = iommu->qi;
>  	int shift = qi_shift(iommu);
>  
> @@ -1317,6 +1320,13 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>  		tail = readl(iommu->reg + DMAR_IQT_REG);
>  		tail = ((tail >> shift) - 1 + QI_LENGTH) % QI_LENGTH;
>  
> +		/*
> +		 * SID field is valid only when the ITE field is Set in FSTS_REG
> +		 * see Intel VT-d spec r4.1, section 11.4.9.9
> +		 */
> +		iqe_err = dmar_readq(iommu->reg + DMAR_IQER_REG);
> +		ite_sid = DMAR_IQER_REG_ITESID(iqe_err);
> +
>  		writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG);
>  		pr_info("Invalidation Time-out Error (ITE) cleared\n");
>  
> @@ -1326,6 +1336,21 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>  			head = (head - 2 + QI_LENGTH) % QI_LENGTH;
>  		} while (head != tail);
>  
> +		/*
> +		 * If got ITE, we need to check if the sid of ITE is one of the
> +		 * current valid ATS invalidation target devices, if no, or the
> +		 * target device isn't presnet, don't try this request anymore.
> +		 * 0 value of ite_sid means old VT-d device, no ite_sid value.
> +		 */
> +		if (ite_sid) {
> +			dev = device_rbtree_find(iommu, ite_sid);
> +			if (!dev || !dev_is_pci(dev))
> +				return -ETIMEDOUT;
> +			pdev = to_pci_dev(dev);
> +			if (!pci_device_is_present(pdev) &&
> +				ite_sid == pci_dev_id(pci_physfn(pdev)))
> +				return -ETIMEDOUT;
> +		}
>  		if (qi->desc_status[wait_index] == QI_ABORT)
>  			return -EAGAIN;
>  	}
> -- 
> 2.31.1
>
  
Ethan Zhao March 1, 2024, 1:51 a.m. UTC | #3
On 3/1/2024 5:07 AM, Bjorn Helgaas wrote:
> On Wed, Feb 28, 2024 at 11:07:24PM -0500, Ethan Zhao wrote:
>> Break the loop to blindly retry the timeout ATS invalidation request
>> after ITE fault hit if device was released or isn't present anymore.
>>
>> This is part of the followup of prior proposed patchset
>>
>> https://do-db2.lkml.org/lkml/2024/2/22/350
> Use lore URL, please.

Sure !

>
>> Fixes: 6ba6c3a4cacf ("VT-d: add device IOTLB invalidation support")
>> Signed-off-by: Ethan Zhao <haifeng.zhao@linux.intel.com>
>> ---
>>   drivers/iommu/intel/dmar.c | 25 +++++++++++++++++++++++++
>>   1 file changed, 25 insertions(+)
>>
>> diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c
>> index d14797aabb7a..d01d68205557 100644
>> --- a/drivers/iommu/intel/dmar.c
>> +++ b/drivers/iommu/intel/dmar.c
>> @@ -1273,6 +1273,9 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>>   {
>>   	u32 fault;
>>   	int head, tail;
>> +	u64 iqe_err, ite_sid;
>> +	struct device *dev = NULL;
>> +	struct pci_dev *pdev = NULL;
>>   	struct q_inval *qi = iommu->qi;
>>   	int shift = qi_shift(iommu);
>>   
>> @@ -1317,6 +1320,13 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>>   		tail = readl(iommu->reg + DMAR_IQT_REG);
>>   		tail = ((tail >> shift) - 1 + QI_LENGTH) % QI_LENGTH;
>>   
>> +		/*
>> +		 * SID field is valid only when the ITE field is Set in FSTS_REG
>> +		 * see Intel VT-d spec r4.1, section 11.4.9.9
>> +		 */
>> +		iqe_err = dmar_readq(iommu->reg + DMAR_IQER_REG);
>> +		ite_sid = DMAR_IQER_REG_ITESID(iqe_err);
>> +
>>   		writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG);
>>   		pr_info("Invalidation Time-out Error (ITE) cleared\n");
>>   
>> @@ -1326,6 +1336,21 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>>   			head = (head - 2 + QI_LENGTH) % QI_LENGTH;
>>   		} while (head != tail);
>>   
>> +		/*
>> +		 * If got ITE, we need to check if the sid of ITE is one of the
>> +		 * current valid ATS invalidation target devices, if no, or the
>> +		 * target device isn't presnet, don't try this request anymore.
>> +		 * 0 value of ite_sid means old VT-d device, no ite_sid value.
>> +		 */
>> +		if (ite_sid) {
>> +			dev = device_rbtree_find(iommu, ite_sid);
>> +			if (!dev || !dev_is_pci(dev))
>> +				return -ETIMEDOUT;
>> +			pdev = to_pci_dev(dev);
>> +			if (!pci_device_is_present(pdev) &&
>> +				ite_sid == pci_dev_id(pci_physfn(pdev)))
>> +				return -ETIMEDOUT;
>> +		}
>>   		if (qi->desc_status[wait_index] == QI_ABORT)
>>   			return -EAGAIN;
>>   	}
>> -- 
>> 2.31.1
>>
  
Ethan Zhao March 1, 2024, 7:05 a.m. UTC | #4
On 2/29/2024 12:07 PM, Ethan Zhao wrote:
> Break the loop to blindly retry the timeout ATS invalidation request
> after ITE fault hit if device was released or isn't present anymore.
>
> This is part of the followup of prior proposed patchset
>
> https://do-db2.lkml.org/lkml/2024/2/22/350
>
> Fixes: 6ba6c3a4cacf ("VT-d: add device IOTLB invalidation support")
> Signed-off-by: Ethan Zhao <haifeng.zhao@linux.intel.com>
> ---
>   drivers/iommu/intel/dmar.c | 25 +++++++++++++++++++++++++
>   1 file changed, 25 insertions(+)
>
> diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c
> index d14797aabb7a..d01d68205557 100644
> --- a/drivers/iommu/intel/dmar.c
> +++ b/drivers/iommu/intel/dmar.c
> @@ -1273,6 +1273,9 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>   {
>   	u32 fault;
>   	int head, tail;
> +	u64 iqe_err, ite_sid;
> +	struct device *dev = NULL;
> +	struct pci_dev *pdev = NULL;
>   	struct q_inval *qi = iommu->qi;
>   	int shift = qi_shift(iommu);
>   
> @@ -1317,6 +1320,13 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>   		tail = readl(iommu->reg + DMAR_IQT_REG);
>   		tail = ((tail >> shift) - 1 + QI_LENGTH) % QI_LENGTH;
>   
> +		/*
> +		 * SID field is valid only when the ITE field is Set in FSTS_REG
> +		 * see Intel VT-d spec r4.1, section 11.4.9.9
> +		 */
> +		iqe_err = dmar_readq(iommu->reg + DMAR_IQER_REG);
> +		ite_sid = DMAR_IQER_REG_ITESID(iqe_err);
> +
>   		writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG);
>   		pr_info("Invalidation Time-out Error (ITE) cleared\n");
>   
> @@ -1326,6 +1336,21 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
>   			head = (head - 2 + QI_LENGTH) % QI_LENGTH;
>   		} while (head != tail);
>   
> +		/*
> +		 * If got ITE, we need to check if the sid of ITE is one of the
> +		 * current valid ATS invalidation target devices, if no, or the
> +		 * target device isn't presnet, don't try this request anymore.
> +		 * 0 value of ite_sid means old VT-d device, no ite_sid value.
> +		 */
> +		if (ite_sid) {
> +			dev = device_rbtree_find(iommu, ite_sid);
> +			if (!dev || !dev_is_pci(dev))
> +				return -ETIMEDOUT;
> +			pdev = to_pci_dev(dev);
> +			if (!pci_device_is_present(pdev) &&
> +				ite_sid == pci_dev_id(pci_physfn(pdev)))
> +				return -ETIMEDOUT;
> +		}
>   		if (qi->desc_status[wait_index] == QI_ABORT)
>   			return -EAGAIN;
>   	}

Given maintainer is going to pick up patchset
https://lore.kernel.org/lkml/2d1788da-521c-4531-a159-81d2fb801d6c@linux.intel.com/T/
and this one is mutually exclusive with it, suspend.


Thanks,
Ethan
  

Patch

diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c
index d14797aabb7a..d01d68205557 100644
--- a/drivers/iommu/intel/dmar.c
+++ b/drivers/iommu/intel/dmar.c
@@ -1273,6 +1273,9 @@  static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
 {
 	u32 fault;
 	int head, tail;
+	u64 iqe_err, ite_sid;
+	struct device *dev = NULL;
+	struct pci_dev *pdev = NULL;
 	struct q_inval *qi = iommu->qi;
 	int shift = qi_shift(iommu);
 
@@ -1317,6 +1320,13 @@  static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
 		tail = readl(iommu->reg + DMAR_IQT_REG);
 		tail = ((tail >> shift) - 1 + QI_LENGTH) % QI_LENGTH;
 
+		/*
+		 * SID field is valid only when the ITE field is Set in FSTS_REG
+		 * see Intel VT-d spec r4.1, section 11.4.9.9
+		 */
+		iqe_err = dmar_readq(iommu->reg + DMAR_IQER_REG);
+		ite_sid = DMAR_IQER_REG_ITESID(iqe_err);
+
 		writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG);
 		pr_info("Invalidation Time-out Error (ITE) cleared\n");
 
@@ -1326,6 +1336,21 @@  static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index)
 			head = (head - 2 + QI_LENGTH) % QI_LENGTH;
 		} while (head != tail);
 
+		/*
+		 * If got ITE, we need to check if the sid of ITE is one of the
+		 * current valid ATS invalidation target devices, if no, or the
+		 * target device isn't presnet, don't try this request anymore.
+		 * 0 value of ite_sid means old VT-d device, no ite_sid value.
+		 */
+		if (ite_sid) {
+			dev = device_rbtree_find(iommu, ite_sid);
+			if (!dev || !dev_is_pci(dev))
+				return -ETIMEDOUT;
+			pdev = to_pci_dev(dev);
+			if (!pci_device_is_present(pdev) &&
+				ite_sid == pci_dev_id(pci_physfn(pdev)))
+				return -ETIMEDOUT;
+		}
 		if (qi->desc_status[wait_index] == QI_ABORT)
 			return -EAGAIN;
 	}