On Thu, Sep 24, 2015 at 10:01:55AM -0600, Alex Williamson wrote: > On Thu, 2015-09-24 at 14:33 +1000, David Gibson wrote: > > Currently the VFIOContainer iommu_data field contains a union with > > different information for different host iommu types. However: > > * It only actually contains information for the x86-like "Type1" iommu > > * Because we have a common listener the Type1 fields are actually used > > on all IOMMU types, including the SPAPR TCE type as well > > * There's no tag in the VFIOContainer to tell you which union member is > > valid anyway. > > FWIW, this last point isn't valid. The IOMMU setup determines which > union member is active and the listener and release functions are > specific to the union member. There's no need whatsoever for a tag to > keep track of the union member in use. The only problem is that the > union solved a problem that never really came to exist, so we can now > remove it and simplify things.
I could argue some of the details there, but none of them are really important. > I'll remove this last bullet point unless there's some objection. > Thanks, That's fine. -- David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson
pgp_Y3sbWSZUj.pgp
Description: PGP signature