Re: [PATCH v4 4/5] iommu/vt-d: Add page request draining support

[Lu Baolu <baolu.lu@linux.intel.com>]

Hi Kevin,

On 5/7/20 2:35 PM, Tian, Kevin wrote:
>> From: Lu Baolu
>> Sent: Thursday, May 7, 2020 8:56 AM
>>
>> When a PASID is stopped or terminated, there can be pending PRQs
>> (requests that haven't received responses) in remapping hardware.
>> This adds the interface to drain page requests and call it when a
>> PASID is terminated.
>>
>> Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
>> Signed-off-by: Liu Yi L <yi.l.liu@intel.com>
>> Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
>> ---
>>   drivers/iommu/intel-svm.c   | 102 ++++++++++++++++++++++++++++++++++-
>> -
>>   include/linux/intel-iommu.h |   4 ++
>>   2 files changed, 101 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/iommu/intel-svm.c b/drivers/iommu/intel-svm.c
>> index 9561ba59a170..7256eb965cf6 100644
>> --- a/drivers/iommu/intel-svm.c
>> +++ b/drivers/iommu/intel-svm.c
>> @@ -23,6 +23,7 @@
>>   #include "intel-pasid.h"
>>
>>   static irqreturn_t prq_event_thread(int irq, void *d);
>> +static void intel_svm_drain_prq(struct device *dev, int pasid);
>>
>>   #define PRQ_ORDER 0
>>
>> @@ -66,6 +67,8 @@ int intel_svm_enable_prq(struct intel_iommu *iommu)
>>   	dmar_writeq(iommu->reg + DMAR_PQT_REG, 0ULL);
>>   	dmar_writeq(iommu->reg + DMAR_PQA_REG, virt_to_phys(iommu-
>>> prq) | PRQ_ORDER);
>>
>> +	init_completion(&iommu->prq_complete);
>> +
>>   	return 0;
>>   }
>>
>> @@ -403,12 +406,8 @@ int intel_svm_unbind_gpasid(struct device *dev, int
>> pasid)
>>   			list_del_rcu(&sdev->list);
>>   			intel_pasid_tear_down_entry(iommu, dev,
>>   						    svm->pasid, false);
>> +			intel_svm_drain_prq(dev, svm->pasid);
>>   			intel_flush_svm_range_dev(svm, sdev, 0, -1, 0);
>> -			/* TODO: Drain in flight PRQ for the PASID since it
>> -			 * may get reused soon, we don't want to
>> -			 * confuse with its previous life.
>> -			 * intel_svm_drain_prq(dev, pasid);
>> -			 */
>>   			kfree_rcu(sdev, rcu);
>>
>>   			if (list_empty(&svm->devs)) {
>> @@ -647,6 +646,7 @@ int intel_svm_unbind_mm(struct device *dev, int
>> pasid)
>>   			 * hard to be as defensive as we might like. */
>>   			intel_pasid_tear_down_entry(iommu, dev,
>>   						    svm->pasid, false);
>> +			intel_svm_drain_prq(dev, svm->pasid);
>>   			intel_flush_svm_range_dev(svm, sdev, 0, -1, 0);
>>   			kfree_rcu(sdev, rcu);
>>
>> @@ -725,6 +725,92 @@ static bool is_canonical_address(u64 addr)
>>   	return (((saddr << shift) >> shift) == saddr);
>>   }
>>
>> +/**
>> + * intel_svm_drain_prq:
>> + *
>> + * Drain all pending page requests and responses related to a specific
>> + * pasid in both software and hardware.
>> + */
>> +static void intel_svm_drain_prq(struct device *dev, int pasid)
>> +{
>> +	struct device_domain_info *info;
>> +	struct dmar_domain *domain;
>> +	struct intel_iommu *iommu;
>> +	struct qi_desc desc[3];
>> +	struct pci_dev *pdev;
>> +	int head, tail;
>> +	u16 sid, did;
>> +	int qdep;
>> +
>> +	info = get_domain_info(dev);
>> +	if (WARN_ON(!info || !dev_is_pci(dev)))
>> +		return;
>> +
>> +	if (!info->ats_enabled)
>> +		return;
> 
> ats_enabled -> pri_enabled

Yes. Sure!

> 
>> +
>> +	iommu = info->iommu;
>> +	domain = info->domain;
>> +	pdev = to_pci_dev(dev);
>> +	sid = PCI_DEVID(info->bus, info->devfn);
>> +	did = domain->iommu_did[iommu->seq_id];
>> +	qdep = pci_ats_queue_depth(pdev);
>> +
>> +	memset(desc, 0, sizeof(desc));
>> +	desc[0].qw0 = QI_IWD_STATUS_DATA(QI_DONE) |
>> +			QI_IWD_FENCE |
>> +			QI_IWD_TYPE;
>> +	desc[1].qw0 = QI_EIOTLB_PASID(pasid) |
>> +			QI_EIOTLB_DID(did) |
>> +			QI_EIOTLB_GRAN(QI_GRAN_NONG_PASID) |
>> +			QI_EIOTLB_TYPE;
>> +	desc[2].qw0 = QI_DEV_EIOTLB_PASID(pasid) |
>> +			QI_DEV_EIOTLB_SID(sid) |
>> +			QI_DEV_EIOTLB_QDEP(qdep) |
>> +			QI_DEIOTLB_TYPE |
>> +			QI_DEV_IOTLB_PFSID(info->pfsid);
>> +
>> +	/*
>> +	 * Submit an invalidation wait descriptor with fence and page request
>> +	 * drain flags set to invalidation queue. This ensures that all requests
>> +	 * submitted to the invalidation queue ahead of this wait descriptor
>> are
>> +	 * processed and completed, and all already issued page requests
>> from
>> +	 * the device are put in the page request queue.
>> +	 */
> 
> I feel this comment is better moved earlier since it explains the overall
> flow including all 3 descriptors. Also it is not one wait descriptor which
> gets both fence and drain flags set. There are two wait descriptors with
> one setting fence and the other setting drain.
> 
>> +	qi_submit_sync(iommu, desc, 1, QI_OPT_WAIT_DRAIN);
> 
> the count is '3' instead of '1'.

Yes. Will fix it in the next version.

Best regards,
baolu