lkml.org 
[lkml]   [2022]   [May]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v6 08/12] iommu/sva: Use attach/detach_pasid_dev in SVA interfaces
On Wed, May 11, 2022 at 03:21:31PM +0800, Baolu Lu wrote:
> On 2022/5/10 23:23, Jason Gunthorpe wrote:
> > On Tue, May 10, 2022 at 02:17:34PM +0800, Lu Baolu wrote:
> >
> > > +/**
> > > + * iommu_sva_bind_device() - Bind a process address space to a device
> > > + * @dev: the device
> > > + * @mm: the mm to bind, caller must hold a reference to mm_users
> > > + * @drvdata: opaque data pointer to pass to bind callback
> > > + *
> > > + * Create a bond between device and address space, allowing the device to access
> > > + * the mm using the returned PASID. If a bond already exists between @device and
> > > + * @mm, it is returned and an additional reference is taken. Caller must call
> > > + * iommu_sva_unbind_device() to release each reference.
> > > + *
> > > + * iommu_dev_enable_feature(dev, IOMMU_DEV_FEAT_SVA) must be called first, to
> > > + * initialize the required SVA features.
> > > + *
> > > + * On error, returns an ERR_PTR value.
> > > + */
> > > +struct iommu_sva *
> > > +iommu_sva_bind_device(struct device *dev, struct mm_struct *mm, void *drvdata)
> > > +{
> > > + int ret = -EINVAL;
> > > + struct iommu_sva *handle;
> > > + struct iommu_domain *domain;
> > > +
> > > + /*
> > > + * TODO: Remove the drvdata parameter after kernel PASID support is
> > > + * enabled for the idxd driver.
> > > + */
> > > + if (drvdata)
> > > + return ERR_PTR(-EOPNOTSUPP);
> >
> > Why is this being left behind? Clean up the callers too please.
>
> Okay, let me try to.
>
> >
> > > + /* Allocate mm->pasid if necessary. */
> > > + ret = iommu_sva_alloc_pasid(mm, 1, (1U << dev->iommu->pasid_bits) - 1);
> > > + if (ret)
> > > + return ERR_PTR(ret);
> > > +
> > > + mutex_lock(&iommu_sva_lock);
> > > + /* Search for an existing bond. */
> > > + handle = xa_load(&dev->iommu->sva_bonds, mm->pasid);
> > > + if (handle) {
> > > + refcount_inc(&handle->users);
> > > + goto out_success;
> > > + }
> >
> > How can there be an existing bond?
> >
> > dev->iommu is per-device
> >
> > The device_group_immutable_singleton() insists on a single device
> > group
> >
> > Basically 'sva_bonds' is the same thing as the group->pasid_array.
>
> Yes, really.
>
> >
> > Assuming we leave room for multi-device groups this logic should just
> > be
> >
> > group = iommu_group_get(dev);
> > if (!group)
> > return -ENODEV;
> >
> > mutex_lock(&group->mutex);
> > domain = xa_load(&group->pasid_array, mm->pasid);
> > if (!domain || domain->type != IOMMU_DOMAIN_SVA || domain->mm != mm)
> > domain = iommu_sva_alloc_domain(dev, mm);
> >
> > ?
>
> Agreed. As a helper in iommu core, how about making it more generic like
> below?

IDK, is there more users of this? AFAIK SVA is the only place that
will be auto-sharing?

> + mutex_lock(&group->mutex);
> + domain = xa_load(&group->pasid_array, pasid);
> + if (domain && domain->type != type)
> + domain = NULL;
> + mutex_unlock(&group->mutex);
> + iommu_group_put(group);
> +
> + return domain;

This is bad locking, group->pasid_array values cannot be taken outside
the lock.

> > And stick the refcount in the sva_domain
> >
> > Also, given the current arrangement it might make sense to have a
> > struct iommu_domain_sva given that no driver is wrappering this in
> > something else.
>
> Fair enough. How about below wrapper?
>
> +struct iommu_sva_domain {
> + /*
> + * Common iommu domain header, *must* be put at the top
> + * of the structure.
> + */
> + struct iommu_domain domain;
> + struct mm_struct *mm;
> + struct iommu_sva bond;
> +}
>
> The refcount is wrapped in bond.

I'm still not sure that bond is necessary

But yes, something like that

Jason

\
 
 \ /
  Last update: 2022-05-11 16:54    [W:0.098 / U:1.664 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site