Commit 1a14bf0f authored by Jacob Pan's avatar Jacob Pan Committed by Joerg Roedel
Browse files

iommu/sva: Use GFP_KERNEL for pasid allocation 



We’re not using spinlock-protected IOASID allocation anymore, there’s
no need for GFP_ATOMIC.

Reviewed-by: default avatarKevin Tian <kevin.tian@intel.com>
Reviewed-by: default avatarLu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: default avatarJason Gunthorpe <jgg@nvidia.com>
Signed-off-by: default avatarJacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230322200803.869130-6-jacob.jun.pan@linux.intel.com


Signed-off-by: default avatarJoerg Roedel <jroedel@suse.de>
parent 4e14176a

drivers/iommu/iommu-sva.c

+2 −2
Original line number Diff line number Diff line
@@ -28,8 +28,8 @@ static int iommu_sva_alloc_pasid(struct mm_struct *mm, ioasid_t min, ioasid_t ma 
		goto out;

	}



	ret = ida_alloc_range(&iommu_global_pasid_ida, min, max, GFP_ATOMIC);

	if (ret < 0)

	ret = ida_alloc_range(&iommu_global_pasid_ida, min, max, GFP_KERNEL);

	if (ret < min)

		goto out;

	mm->pasid = ret;

	ret = 0;