Message-ID: <5B4801A6.1090200@linux.intel.com>
Date:   Fri, 13 Jul 2018 09:34:30 +0800
From:   Lu Baolu <baolu.lu@...ux.intel.com>
To:     Peter Xu <peterx@...hat.com>
Cc:     Joerg Roedel <joro@...tes.org>,
        David Woodhouse <dwmw2@...radead.org>, ashok.raj@...el.com,
        sanjay.k.kumar@...el.com, iommu@...ts.linux-foundation.org,
        linux-kernel@...r.kernel.org, yi.y.sun@...el.com,
        jacob.jun.pan@...el.com
Subject: Re: [PATCH v4 9/9] iommu/vt-d: Remove the obsolete per iommu pasid
 tables

Hi Peter,

Do you have further comments for this series? I will work out a
new version if there are no ones.

Thank you for reviewing my patches. Do you allow me to add your
reviewed-by to the patches which you've reviewed?

Best regards,
Lu Baolu

On 07/11/2018 10:45 AM, Peter Xu wrote:
> On Mon, Jul 09, 2018 at 01:22:58PM +0800, Lu Baolu wrote:
>> The obsolete per iommu pasid tables are no longer used. Hence,
>> clean up them.
>>
>> Cc: Ashok Raj <ashok.raj@...el.com>
>> Cc: Jacob Pan <jacob.jun.pan@...ux.intel.com>
>> Cc: Kevin Tian <kevin.tian@...el.com>
>> Cc: Liu Yi L <yi.l.liu@...el.com>
>> Signed-off-by: Lu Baolu <baolu.lu@...ux.intel.com>
>> Reviewed-by: Liu Yi L <yi.l.liu@...el.com>
>> ---
>>  drivers/iommu/intel-iommu.c |  6 +++---
>>  drivers/iommu/intel-svm.c   | 17 ++---------------
>>  include/linux/intel-iommu.h |  5 ++---
>>  3 files changed, 7 insertions(+), 21 deletions(-)
>>
>> diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
>> index f8609b5..9a88081 100644
>> --- a/drivers/iommu/intel-iommu.c
>> +++ b/drivers/iommu/intel-iommu.c
>> @@ -1755,7 +1755,7 @@ static void free_dmar_iommu(struct intel_iommu *iommu)
>>  	if (pasid_enabled(iommu)) {
>>  		if (ecap_prs(iommu->ecap))
>>  			intel_svm_finish_prq(iommu);
>> -		intel_svm_free_pasid_tables(iommu);
>> +		intel_svm_exit(iommu);
>>  	}
>>  #endif
>>  }
>> @@ -3336,7 +3336,7 @@ static int __init init_dmars(void)
>>  			hw_pass_through = 0;
>>  #ifdef CONFIG_INTEL_IOMMU_SVM
>>  		if (pasid_enabled(iommu))
>> -			intel_svm_alloc_pasid_tables(iommu);
>> +			intel_svm_init(iommu);
>>  #endif
>>  	}
>>  
>> @@ -4330,7 +4330,7 @@ static int intel_iommu_add(struct dmar_drhd_unit *dmaru)
>>  
>>  #ifdef CONFIG_INTEL_IOMMU_SVM
>>  	if (pasid_enabled(iommu))
>> -		intel_svm_alloc_pasid_tables(iommu);
>> +		intel_svm_init(iommu);
>>  #endif
>>  
>>  	if (dmaru->ignored) {
>> diff --git a/drivers/iommu/intel-svm.c b/drivers/iommu/intel-svm.c
>> index a253cde..eb30836 100644
>> --- a/drivers/iommu/intel-svm.c
>> +++ b/drivers/iommu/intel-svm.c
>> @@ -38,7 +38,7 @@ struct pasid_state_entry {
>>  	u64 val;
>>  };
>>  
>> -int intel_svm_alloc_pasid_tables(struct intel_iommu *iommu)
>> +int intel_svm_init(struct intel_iommu *iommu)
>>  {
>>  	struct page *pages;
>>  	int order;
>> @@ -63,15 +63,6 @@ int intel_svm_alloc_pasid_tables(struct intel_iommu *iommu)
>>  		iommu->pasid_max = 0x20000;
>>  
>>  	order = get_order(sizeof(struct pasid_entry) * iommu->pasid_max);
>> -	pages = alloc_pages(GFP_KERNEL | __GFP_ZERO, order);
>> -	if (!pages) {
>> -		pr_warn("IOMMU: %s: Failed to allocate PASID table\n",
>> -			iommu->name);
>> -		return -ENOMEM;
>> -	}
>> -	iommu->pasid_table = page_address(pages);
>> -	pr_info("%s: Allocated order %d PASID table.\n", iommu->name, order);
>> -
>>  	if (ecap_dis(iommu->ecap)) {
>>  		/* Just making it explicit... */
>>  		BUILD_BUG_ON(sizeof(struct pasid_entry) != sizeof(struct pasid_state_entry));
> Then do we still need the so-called pasid state table?  I tried to
> find out what it did before but I failed since I see that the bit 27
> of the extended cap is marked as reserved now in the latest vt-d spec
> (then ecap_dis could be meaningless too).
>
> Asked since if we don't need both the per-iommu pasid table and the
> pasid state table, then maybe we don't need intel_svm_init() at all?
>
>> @@ -86,14 +77,10 @@ int intel_svm_alloc_pasid_tables(struct intel_iommu *iommu)
>>  	return 0;
>>  }
>>  
>> -int intel_svm_free_pasid_tables(struct intel_iommu *iommu)
>> +int intel_svm_exit(struct intel_iommu *iommu)
> Same thing applies to this exit() function - after we removed the
> per-iommu idr, per-iommu pasid table, and (possibly obsolete)
> per-iommu pasid state table, do we need that at all?
>
> Thanks,
>
>>  {
>>  	int order = get_order(sizeof(struct pasid_entry) * iommu->pasid_max);
>>  
>> -	if (iommu->pasid_table) {
>> -		free_pages((unsigned long)iommu->pasid_table, order);
>> -		iommu->pasid_table = NULL;
>> -	}
>>  	if (iommu->pasid_state_table) {
>>  		free_pages((unsigned long)iommu->pasid_state_table, order);
>>  		iommu->pasid_state_table = NULL;
>> diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
>> index e7901d4..3c43882 100644
>> --- a/include/linux/intel-iommu.h
>> +++ b/include/linux/intel-iommu.h
>> @@ -453,7 +453,6 @@ struct intel_iommu {
>>  	 * devices away to userspace processes (e.g. for DPDK) and don't
>>  	 * want to trust that userspace will use *only* the PASID it was
>>  	 * told to. But while it's all driver-arbitrated, we're fine. */
>> -	struct pasid_entry *pasid_table;
>>  	struct pasid_state_entry *pasid_state_table;
>>  	struct page_req_dsc *prq;
>>  	unsigned char prq_name[16];    /* Name for PRQ interrupt */
>> @@ -526,8 +525,8 @@ int for_each_device_domain(int (*fn)(struct device_domain_info *info,
>>  				     void *data), void *data);
>>  
>>  #ifdef CONFIG_INTEL_IOMMU_SVM
>> -extern int intel_svm_alloc_pasid_tables(struct intel_iommu *iommu);
>> -extern int intel_svm_free_pasid_tables(struct intel_iommu *iommu);
>> +int intel_svm_init(struct intel_iommu *iommu);
>> +int intel_svm_exit(struct intel_iommu *iommu);
>>  extern int intel_svm_enable_prq(struct intel_iommu *iommu);
>>  extern int intel_svm_finish_prq(struct intel_iommu *iommu);
>>  
>> -- 
>> 2.7.4
>>
>> _______________________________________________
>> iommu mailing list
>> iommu@...ts.linux-foundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/iommu

Powered by blists - more mailing lists