On Wed, 2023-10-18 at 06:43 +0200, Christoph Hellwig wrote: > On Tue, Oct 17, 2023 at 01:24:59PM -0700, Rick Edgecombe wrote: > > On TDX it is possible for the untrusted host to cause > > set_memory_encrypted() or set_memory_decrypted() to fail such that > > an > > error is returned and the resulting memory is shared. Callers need > > to take > > care to handle these errors to avoid returning decrypted (shared) > > memory to > > the page allocator, which could lead to functional or security > > issues. > > > > Swiotlb could free decrypted/shared pages if set_memory_decrypted() > > fails. > > Use the recently added free_decrypted_pages() to avoid this. > > > > In swiotlb_exit(), check for set_memory_encrypted() errors > > manually, > > because the pages are not nessarily going to the page allocator. > > Whatever recently introduced it didn't make it to my mailbox. Please > always CC everyone on every patch in a series, everything else is > impossible to review. Ok. The series touches a bunch of set_memory() callers all over, so I was trying to manage the CC list to something reasonable. I tried to give a summary in each commit, but I guess it wasn't in depth enough. Here is the lore link, if you haven't already found it: https://lore.kernel.org/lkml/20231017202505.340906-1-rick.p.edgecombe@xxxxxxxxx/
