On Wed, May 28, 2025, James Houghton wrote:
> On Wed, May 28, 2025 at 11:09 AM James Houghton <jthoughton@xxxxxxxxxx> wrote:
> >
> > On Tue, May 6, 2025 at 8:06 PM Sean Christopherson <seanjc@xxxxxxxxxx> wrote:
> > >
> > > On Thu, Jan 09, 2025, James Houghton wrote:
> > > > @@ -2073,6 +2080,23 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
> > > > enum kvm_mr_change change)
> > > > {
> > > > bool log_dirty_pages = new && new->flags & KVM_MEM_LOG_DIRTY_PAGES;
> > > > + u32 new_flags = new ? new->flags : 0;
> > > > + u32 changed_flags = (new_flags) ^ (old ? old->flags : 0);
> > >
> > > This is a bit hard to read, and there's only one use of log_dirty_pages. With
> > > zapping handled in common KVM, just do:
> >
> > Thanks, Sean. Yeah what you have below looks a lot better, thanks for
> > applying it for me. I'll post a new version soon. One note below.
> >
> > >
> > > @@ -2127,14 +2131,19 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
> > > const struct kvm_memory_slot *new,
> > > enum kvm_mr_change change)
> > > {
> > > - bool log_dirty_pages = new && new->flags & KVM_MEM_LOG_DIRTY_PAGES;
> > > + u32 old_flags = old ? old->flags : 0;
> > > + u32 new_flags = new ? new->flags : 0;
> > > +
> > > + /* Nothing to do if not toggling dirty logging. */
> > > + if (!((old_flags ^ new_flags) & KVM_MEM_LOG_DIRTY_PAGES))
> > > + return;
> >
> > This is my bug, not yours, but I think this condition must also check
> > that `change == KVM_MR_FLAGS_ONLY` for it to be correct. This, for
> > example, will break the case where we are deleting a memslot that
> > still has KVM_MEM_LOG_DIRTY_PAGES enabled. Will fix in the next
> > version.
>
> Ah it wouldn't break that example, as `new` would be NULL. But I think
> it would break the case where we are moving a memslot that keeps
> `KVM_MEM_LOG_DIRTY_PAGES`.
Can you elaborate? Maybe with the full snippet of the final code that's broken.
I'm not entirely following what's path you're referring to.
