On Fri, Nov 21, 2025 at 02:38:06AM -0800, Shameer Kolothum wrote: > > -----Original Message----- > > From: Nicolin Chen <[email protected]> > > Sent: 20 November 2025 20:51 > > To: Shameer Kolothum <[email protected]> > > Cc: [email protected]; [email protected]; > > [email protected]; [email protected]; Jason Gunthorpe > > <[email protected]>; [email protected]; [email protected]; Nathan > > Chen <[email protected]>; Matt Ochs <[email protected]>; > > [email protected]; [email protected]; > > [email protected]; [email protected]; > > [email protected]; [email protected]; [email protected]; > > Krishnakant Jaju <[email protected]>; Michael S . Tsirkin <[email protected]> > > Subject: Re: [PATCH v6 08/33] hw/pci/pci: Add optional > > supports_address_space() callback > > > > On Thu, Nov 20, 2025 at 01:21:48PM +0000, Shameer Kolothum wrote: > > > Introduce an optional supports_address_space() callback in PCIIOMMUOps > > to > > > > "supports_address_space" sounds a bit to wide to me than its > > indication to supporting an IOMMU address space specifically, > > since the "system address space" being used in this series is > > a legit address space as well. > > > > With that being said, I think we are fine for now, given the > > API docs has clarified it. If someone shares the same concern, > > we can rename it later. > > The intent here is just to let the vIOMMU decide whether a device should > be associated with its address_space before we call get_address_space(). > If the check passes, the vIOMMU must provide the actual address_space > through get_address_space() callback.
The naming makes sense now. Yet, the API doc is a bit confusing.. Why it says "device can have an IOMMU address space"? If a device only has a system address space (i.e. it doesn't support an IOMMU address space), it still returns true, right? Nicolin
