Commit 128ca093 authored Jan 27, 2015 by Joerg Roedel Committed by Paolo Bonzini Jan 27, 2015

kvm: iommu: Add cond_resched to legacy device assignment code



When assigning devices to large memory guests (>=128GB guest
memory in the failure case) the functions to create the
IOMMU page-tables for the whole guest might run for a very
long time. On non-preemptible kernels this might cause
Soft-Lockup warnings. Fix these by adding a cond_resched()
to the mapping and unmapping loops.

Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

parent 82268083

arch/x86/kvm/iommu.c

+3 −1

Original line number	Original line	Diff line number	Diff line
	@@ -138,7 +138,7 @@ int kvm_iommu_map_pages(struct kvm kvm, struct kvm_memory_slot slot)

	gfn += page_size >> PAGE_SHIFT;		gfn += page_size >> PAGE_SHIFT;

			cond_resched();
	}		}

	return 0;		return 0;
	@@ -306,6 +306,8 @@ static void kvm_iommu_put_pages(struct kvm *kvm,
	kvm_unpin_pages(kvm, pfn, unmap_pages);		kvm_unpin_pages(kvm, pfn, unmap_pages);

	gfn += unmap_pages;		gfn += unmap_pages;

			cond_resched();
	}		}
	}		}