Re: [PATCH v7 12/14] iommu/arm-smmu-v3: Introduce struct arm_smmu_vmaster
From: Jason Gunthorpe <jgg@nvidia.com>
Date: 2025-02-25 16:02:31
Also in:
linux-doc, linux-iommu, linux-kselftest, linux-patches, lkml
On Mon, Feb 24, 2025 at 03:45:33PM -0800, Nicolin Chen wrote:
quoted hunk ↗ jump to hunk
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c@@ -95,8 +95,6 @@ int arm_smmu_attach_prepare_vmaster(struct arm_smmu_attach_state *state, iommu_group_mutex_assert(state->master->dev); - if (domain->type != IOMMU_DOMAIN_NESTED) - return 0; nested_domain = to_smmu_nested_domain(domain); /* Skip invalid vSTE */@@ -122,19 +120,9 @@ void arm_smmu_attach_commit_vmaster(struct arm_smmu_attach_state *state) { struct arm_smmu_master *master = state->master; - mutex_lock(&master->smmu->streams_mutex); - if (state->vmaster != master->vmaster) { - kfree(master->vmaster); - master->vmaster = state->vmaster; - } - mutex_unlock(&master->smmu->streams_mutex); -} - -void arm_smmu_master_clear_vmaster(struct arm_smmu_master *master) -{ mutex_lock(&master->smmu->streams_mutex); kfree(master->vmaster); - master->vmaster = NULL; + master->vmaster = state->vmaster; mutex_unlock(&master->smmu->streams_mutex); }
I'd leave the clear_vmaster just for clarity. Commit should not be unpaired with prepare in the other functions. It looks fine with this on top too Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Jason