On 23/09/20 19:26, Sean Christopherson wrote: > /* > * Flush before LAUNCH_UPDATE encrypts pages in place, in case the cache > * contains the data that was written unencrypted. > */ > sev_clflush_pages(inpages, npages); > > there's nothing in the comment or code that even suggests sev_clflush_pages() is > conditional, i.e. no reason for the reader to peek at the implemenation. > > What about: > > /* > * Flush (on non-coherent CPUs) before LAUNCH_UPDATE encrypts pages in > * place, the cache may contain data that was written unencrypted. > */ Sounds good. Paolo
