Re: [PATCH 3/5] iommu/arm-smmu-v3: Add read_and_clear_dirty() support

From: Jason Gunthorpe <jgg@nvidia.com>
To: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Cc: iommu@lists.linux.dev, linux-arm-kernel@lists.infradead.org,
	robin.murphy@arm.com, will@kernel.org, joro@8bytes.org,
	kevin.tian@intel.com, nicolinc@nvidia.com, mshavit@google.com,
	eric.auger@redhat.com, joao.m.martins@oracle.com,
	jiangkunkun@huawei.com, zhukeqian1@huawei.com,
	linuxarm@huawei.com
Subject: Re: [PATCH 3/5] iommu/arm-smmu-v3: Add read_and_clear_dirty() support
Date: Wed, 29 Nov 2023 15:35:06 -0400	[thread overview]
Message-ID: <20231129193506.GB436702@nvidia.com> (raw)
In-Reply-To: <20231128094940.1344-4-shameerali.kolothum.thodi@huawei.com>

On Tue, Nov 28, 2023 at 09:49:38AM +0000, Shameer Kolothum wrote:

> +static int arm_smmu_read_and_clear_dirty(struct iommu_domain *domain,
> +					 unsigned long iova, size_t size,
> +					 unsigned long flags,
> +					 struct iommu_dirty_bitmap *dirty)
> +{
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
> +	int ret;
> +
> +	if (smmu_domain->stage != ARM_SMMU_DOMAIN_S1)
> +		return -EINVAL;
> +
> +	if (!ops || !ops->read_and_clear_dirty) {
> +		pr_err_once("io-pgtable don't support dirty tracking\n");
> +		return -ENODEV;

All these prints concern me, either these can never happen due to
constraints on the caller, eg iommufd, in which case use WARN_ON_ONCe

Or iommufd does allow these improper things and thus it should
silently fail.

Jason