On Mon, Dec 06, 2021 at 03:07:15PM +0000, Matthew Wilcox wrote: > > > What do you think to adding a generic copy_pfn_to_iter()? Not sure > > > which APIs to use to implement it ... some architectures have weird > > > requirements about which APIs can be used for what kinds of PFNs. > > > > Hmm. I though kmap_local_pfn(_prot) is all we need? > > In the !HIGHMEM case, that calls pfn_to_page(), and I think the > point of this path is that we don't have a struct page for this pfn. Indeed. But to me this suggest that the !highmem stub is broken and we should probably fix it rather than adding yet another interface.
