On Mon, Jul 28, 2025 at 9:53 PM Eugenio Perez Martin
<epere...@redhat.com> wrote:
>
> On Fri, Jul 18, 2025 at 9:53 AM Jason Wang <jasow...@redhat.com> wrote:
> >
> > After commit c2dbb3fc2a97 ("vduse: switch to use virtio map API
> > instead of DMA API"), vduse doesn't use DMA API. So we can safely drop
> > the depedencies for HAS_DMA and ARCH_HAS_DMA_OPS.
> >
> > Signed-off-by: Jason Wang <jasow...@redhat.com>
>
> This should be at the end of the series that makes VDUSE not depend on
> DMA, isn't it? Or am I missing something?This is a fixup after Michael merges the series. But he spot some other issue (e.g losing the type safety of dma device), so the series were dropped from his tree. I will work on a new version. > > Anyway, > > Acked-by: Eugenio Pérez <epere...@redhat.com> > > Thanks! Thanks > > > --- > > drivers/vdpa/Kconfig | 8 +------- > > 1 file changed, 1 insertion(+), 7 deletions(-) > > > > diff --git a/drivers/vdpa/Kconfig b/drivers/vdpa/Kconfig > > index 559fb9d3271f..857cf288c876 100644 > > --- a/drivers/vdpa/Kconfig > > +++ b/drivers/vdpa/Kconfig > > @@ -34,13 +34,7 @@ config VDPA_SIM_BLOCK > > > > config VDPA_USER > > tristate "VDUSE (vDPA Device in Userspace) support" > > - depends on EVENTFD && MMU && HAS_DMA > > - # > > - # This driver incorrectly tries to override the dma_ops. It should > > - # never have done that, but for now keep it working on architectures > > - # that use dma ops > > - # > > - depends on ARCH_HAS_DMA_OPS > > + depends on EVENTFD && MMU > > select VHOST_IOTLB > > select IOMMU_IOVA > > help > > -- > > 2.47.3 > > >
