RE: [PATCH 5/9] iommu: Make iommu_group_do_dma_first_attach() simpler

From: "Tian, Kevin" <kevin.tian@intel.com>
To: Jason Gunthorpe <jgg@nvidia.com>,
	"iommu@lists.linux.dev" <iommu@lists.linux.dev>,
	Joerg Roedel <joro@8bytes.org>,
	Robin Murphy <robin.murphy@arm.com>,
	Will Deacon <will@kernel.org>
Cc: Lu Baolu <baolu.lu@linux.intel.com>, Nicolin Chen <nicolinc@nvidia.com>
Subject: RE: [PATCH 5/9] iommu: Make iommu_group_do_dma_first_attach() simpler
Date: Fri, 24 Mar 2023 07:23:30 +0000	[thread overview]
Message-ID: <BN9PR11MB52766008E93EA1CBEFCE5DF78C849@BN9PR11MB5276.namprd11.prod.outlook.com> (raw)
In-Reply-To: <5-v1-20507a7e6b7e+2d6-iommu_err_unwind_jgg@nvidia.com>

> From: Jason Gunthorpe <jgg@nvidia.com>
> Sent: Wednesday, March 22, 2023 3:53 AM
> 
> -static int iommu_group_do_dma_first_attach(struct device *dev, void *data)
> +static int iommu_group_do_dma_first_attach(struct iommu_group *group,
> struct device *dev)
>  {
> -	struct iommu_domain *domain = data;
> -
> -	lockdep_assert_held(&dev->iommu_group->mutex);
> -
> -	if (iommu_is_attach_deferred(dev)) {
> -		dev->iommu->attach_deferred = 1;
> -		return 0;
> -	}
> -
> -	return __iommu_attach_device(domain, dev);
> +	return __iommu_device_set_domain(
> +		group, dev, group->domain,
> +		group->owner ? 0 : IOMMU_SET_DOMAIN_WITH_DEFERRED);
>  }

This changes semantics in iommu_group_add_device().

Previously it always does deferred attach no matter whether the group
has been claimed.

While this change makes sense given deferred attach only applies to
default domain, it's worthy of an explanation or probably even made
into a separate patch.