Re: [PATCH 1/2] vfio/type1: Periodically try rescheduling when unmapping

[Jason Gunthorpe <jgg@nvidia.com>]

On Wed, Jul 15, 2026 at 10:39:01AM -0700, David Matlack wrote:
> On Tue, Jul 14, 2026 at 2:03 PM Aaron Lewis <aaronlewis@google.com> wrote:
> 
> > diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
> > index c8151ba54de3..e78bc07d9b6e 100644
> > --- a/drivers/vfio/vfio_iommu_type1.c
> > +++ b/drivers/vfio/vfio_iommu_type1.c
> > @@ -1153,6 +1153,7 @@ static long vfio_unmap_unpin(struct vfio_iommu *iommu, struct vfio_dma *dma,
> >         int unmapped_region_cnt = 0;
> >         long unlocked = 0;
> >         size_t pos = 0;
> > +       int i = 0;
> >
> >         if (!dma->size)
> >                 return 0;
> > @@ -1196,6 +1197,8 @@ static long vfio_unmap_unpin(struct vfio_iommu *iommu, struct vfio_dma *dma,
> >                         next = iommu_iova_to_phys(domain->domain, iova + len);
> >                         if (next != phys + len)
> >                                 break;
> > +                       if ((++i % BIT(PUD_ORDER)) == 0)
> > +                               cond_resched();
> >                 }
> 
> What is the performance impact on unmap if cond_resched() is called on
> every iteration of the loop? It would be nice to not have a magic
> heuristic (which is brittle as Sashiko points out).

Calling every loop is definately not a good idea.. Some magic number
is a pretty typical pattern for these CPU bound loops.

Jason