On Thu, Sep 01, 2022 at 06:44:10PM +0800, Baolu Lu wrote: > On 2022/8/31 22:10, Jason Gunthorpe wrote: > > On Wed, Aug 31, 2022 at 09:49:44AM +0800, Baolu Lu wrote: > > > > Maybe all of this is just the good reason to go to a simple > > > > device->ops->remove_dev_pasid() callback and forget about blocking > > > > domain here. > > > > > > Do you mean rolling back to what we did in v10? > > > > Yeah, but it shouldn't be a domain_op, removing a pasid is a device op > > > > Just > > > > remove_dev_pasid(struct device *dev, ioasid_t pasid) > > It's clear now. Thanks! > > How about below iommu_attach/detach_device_pasid() code? I think this is probably the right thing > By the way, how about naming it "block_dev_pasid(dev, pasid)"? set/remove is a better pairing that set/block Jason
