RE: [PATCH v3 1/5] iommu: Lock group->mutex in iommu_deferred_attach

[Tian, Kevin]

> From: Nicolin Chen <nicolinc@xxxxxxxxxx>
> Sent: Tuesday, August 12, 2025 6:59 AM
> 
> The iommu_deferred_attach() is a runtime asynchronous function called by
> iommu-dma function, which could race against other attach functions if it
> accesses something in the dev->iommu_group.

Is there a real racing scenario being observed or more theoretical?

If the former may need a Fix tag. 

> 
> So, grab the mutex to guard __iommu_attach_device() like other callers.
> 
> Reviewed-by: Jason Gunthorpe <jgg@xxxxxxxxxx>
> Signed-off-by: Nicolin Chen <nicolinc@xxxxxxxxxx>
> ---
>  drivers/iommu/iommu.c | 13 ++++++++++---
>  1 file changed, 10 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
> index 060ebe330ee16..1e0116bce0762 100644
> --- a/drivers/iommu/iommu.c
> +++ b/drivers/iommu/iommu.c
> @@ -2144,10 +2144,17 @@ EXPORT_SYMBOL_GPL(iommu_attach_device);
> 
>  int iommu_deferred_attach(struct device *dev, struct iommu_domain
> *domain)
>  {
> -	if (dev->iommu && dev->iommu->attach_deferred)
> -		return __iommu_attach_device(domain, dev);
> +	/*
> +	 * This is called on the dma mapping fast path so avoid locking. This is
> +	 * racy, but we have an expectation that the driver will setup its
> DMAs
> +	 * inside probe while being single threaded to avoid racing.
> +	 */
> +	if (!dev->iommu || !dev->iommu->attach_deferred)
> +		return 0;

Is there any way to detect a driver breaking the expectation?

> 
> -	return 0;
> +	guard(mutex)(&dev->iommu_group->mutex);
> +
> +	return __iommu_attach_device(domain, dev);
>  }
> 
>  void iommu_detach_device(struct iommu_domain *domain, struct device
> *dev)
> --
> 2.43.0