On Tue, Jun 21, 2022, David Matlack wrote: > On Fri, Jun 17, 2022 at 8:02 AM Sean Christopherson <seanjc@xxxxxxxxxx> wrote: > > > > On Mon, May 16, 2022, David Matlack wrote: > > > Refactor kvm_mmu_alloc_shadow_page() to receive the caches from which it > > > will allocate the various pieces of memory for shadow pages as a > > > parameter, rather than deriving them from the vcpu pointer. This will be > > > useful in a future commit where shadow pages are allocated during VM > > > ioctls for eager page splitting, and thus will use a different set of > > > caches. > > > > > > Preemptively pull the caches out all the way to > > > kvm_mmu_get_shadow_page() since eager page splitting will not be calling > > > > Uber nit, "eager hugepage splitting" to provide a mental cue/reminder for why > > those pages are direct. > > I think it may be too late to move away from the term "eager page > splitting" (it is already in commit messages and the module param is > called "eager_page_split"). Using a slightly different name here might > produce more confusion, or at least cause readers to do a double-take. True. I'm totally fine omitting "huge". > But naming aside, I don't follow what you mean here. i.e. What does > the fact that page splitting uses direct shadow pages have to do with > this patch? I have no idea. I suspect I was looking at a different patch when replying to this one. I distinctly remember pausing for a few seconds to recall the direct aspect, but looking back at this patch I don't see what I could have possibly be wondering about.