Re: [PATCH v4 5/6] iommu: Support mm PASID 1:n with sva domains

From: Jason Gunthorpe
Date: Thu Sep 21 2023 - 19:31:41 EST


On Tue, Sep 12, 2023 at 08:59:35PM +0800, Tina Zhang wrote:
> diff --git a/drivers/iommu/iommu-sva.c b/drivers/iommu/iommu-sva.c
> index 0f956ecd0c9b..79779bbef653 100644
> --- a/drivers/iommu/iommu-sva.c
> +++ b/drivers/iommu/iommu-sva.c
> @@ -15,6 +15,7 @@ static DEFINE_MUTEX(iommu_sva_lock);
> static int iommu_sva_alloc_pasid(struct mm_struct *mm, struct device *dev)
> {
> ioasid_t pasid;
> + struct iommu_mm_data *iommu_mm;
> int ret = 0;
>
> if (!arch_pgtable_dma_compat(mm))
> @@ -28,12 +29,22 @@ static int iommu_sva_alloc_pasid(struct mm_struct *mm, struct device *dev)
> goto out;
> }
>
> + iommu_mm = kzalloc(sizeof(struct iommu_mm_data), GFP_KERNEL);
> + if (!iommu_mm) {
> + ret = -ENOMEM;
> + goto out;
> + }
> +
> pasid = iommu_alloc_global_pasid(dev);
> if (pasid == IOMMU_PASID_INVALID) {
> + kfree(iommu_mm);
> ret = -ENOSPC;
> goto out;
> }
> - mm->pasid = pasid;
> + iommu_mm->pasid = pasid;
> + INIT_LIST_HEAD(&iommu_mm->sva_domains);
> + mm->iommu_mm = iommu_mm;
> +
> ret = 0;
> out:
> mutex_unlock(&iommu_sva_lock);
> @@ -73,16 +84,12 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm
>
> mutex_lock(&iommu_sva_lock);
> /* Search for an existing domain. */
> - domain = iommu_get_domain_for_dev_pasid(dev, mm_get_pasid(mm),
> - IOMMU_DOMAIN_SVA);
> - if (IS_ERR(domain)) {
> - ret = PTR_ERR(domain);
> - goto out_unlock;
> - }
> -
> - if (domain) {
> - domain->users++;
> - goto out;
> + list_for_each_entry(domain, &mm->iommu_mm->sva_domains, next) {
> + ret = iommu_attach_device_pasid(domain, dev, mm_get_pasid(mm));
> + if (!ret) {
> + domain->users++;
> + goto out;
> + }
> }
>
> /* Allocate a new domain and set it on device pasid. */
> @@ -96,6 +103,8 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm
> if (ret)
> goto out_free_domain;
> domain->users = 1;
> + list_add(&domain->next, &mm->iommu_mm->sva_domains);
> +
> out:
> mutex_unlock(&iommu_sva_lock);
> handle->dev = dev;
> @@ -128,8 +137,9 @@ void iommu_sva_unbind_device(struct iommu_sva *handle)
> struct device *dev = handle->dev;
>
> mutex_lock(&iommu_sva_lock);
> + iommu_detach_device_pasid(domain, dev, pasid);
> if (--domain->users == 0) {
> - iommu_detach_device_pasid(domain, dev, pasid);
> + list_del(&domain->next);
> iommu_domain_free(domain);
> }
> mutex_unlock(&iommu_sva_lock);
> @@ -209,4 +219,5 @@ void mm_pasid_drop(struct mm_struct *mm)
> return;
>
> iommu_free_global_pasid(mm_get_pasid(mm));
> + kfree(mm->iommu_mm);
> }
> diff --git a/include/linux/iommu.h b/include/linux/iommu.h
> index b9c9f14a95cc..c61bc45d5a82 100644
> --- a/include/linux/iommu.h
> +++ b/include/linux/iommu.h
> @@ -109,6 +109,11 @@ struct iommu_domain {
> struct { /* IOMMU_DOMAIN_SVA */
> struct mm_struct *mm;
> int users;
> + /*
> + * Next iommu_domain in mm->iommu_mm->sva-domains list
> + * protected by iommu_sva_lock.
> + */
> + struct list_head next;
> };
> };
> };
> @@ -1186,17 +1191,13 @@ static inline bool tegra_dev_iommu_get_stream_id(struct device *dev, u32 *stream
> }
>
> #ifdef CONFIG_IOMMU_SVA
> -static inline void mm_pasid_init(struct mm_struct *mm)
> -{
> - mm->pasid = IOMMU_PASID_INVALID;
> -}
> static inline bool mm_valid_pasid(struct mm_struct *mm)
> {
> - return mm->pasid != IOMMU_PASID_INVALID;
> + return mm->iommu_mm ? true : false;
> }

HUm this isn't locked very nicely.

Above do

smp_store_release(&mm->iommu_mm, iommu_mm);

And then do

return READ_ONCE(mm->iommu_mm)

(no need for ternaries with bools, compiler generates it automatically)


> static inline u32 mm_get_pasid(struct mm_struct *mm)
> {
> - return mm->pasid;
> + return mm->iommu_mm ? mm->iommu_mm->pasid : IOMMU_PASID_INVALID;
> }

Then this should be

struct iommu_mm_data *iommu_mm = READ_ONCE(mm->iommu_mm);

if (!iommu_mm)
return IOMMU_PASID_INVALID;
return iommu_mm->pasid;

Keeping in mind that the kfree(mm->iommu_mm) being placed in the
mm_drop is critical to this working safely.

Otherwise the logic looks OK.

Jason