On Mon, Jan 30, 2023 at 10:38:41AM -0800, James Houghton wrote: > On Mon, Jan 30, 2023 at 9:29 AM Peter Xu <peterx@xxxxxxxxxx> wrote: > > > > On Fri, Jan 27, 2023 at 01:02:02PM -0800, James Houghton wrote: > > > On Thu, Jan 26, 2023 at 12:31 PM Peter Xu <peterx@xxxxxxxxxx> wrote: > > > > > > > > James, > > > > > > > > On Thu, Jan 26, 2023 at 08:58:51AM -0800, James Houghton wrote: > > > > > It turns out that the THP-like scheme significantly slows down > > > > > MADV_COLLAPSE: decrementing the mapcounts for the 4K subpages becomes > > > > > the vast majority of the time spent in MADV_COLLAPSE when collapsing > > > > > 1G mappings. It is doing 262k atomic decrements, so this makes sense. > > > > > > > > > > This is only really a problem because this is done between > > > > > mmu_notifier_invalidate_range_start() and > > > > > mmu_notifier_invalidate_range_end(), so KVM won't allow vCPUs to > > > > > access any of the 1G page while we're doing this (and it can take like > > > > > ~1 second for each 1G, at least on the x86 server I was testing on). > > > > > > > > Did you try to measure the time, or it's a quick observation from perf? > > > > > > I put some ktime_get()s in. > > > > > > > > > > > IIRC I used to measure some atomic ops, it is not as drastic as I thought. > > > > But maybe it depends on many things. > > > > > > > > I'm curious how the 1sec is provisioned between the procedures. E.g., I > > > > would expect mmu_notifier_invalidate_range_start() to also take some time > > > > too as it should walk the smally mapped EPT pgtables. > > > > > > Somehow this doesn't take all that long (only like 10-30ms when > > > collapsing from 4K -> 1G) compared to hugetlb_collapse(). > > > > Did you populate as much the EPT pgtable when measuring this? > > > > IIUC this number should be pretty much relevant to how many pages are > > shadowed to the kvm pgtables. If the EPT table is mostly empty it should > > be super fast, but OTOH it can be much slower if when it's populated, > > because tdp mmu should need to handle the pgtable leaves one by one. > > > > E.g. it should be fully populated if you have a program busy dirtying most > > of the guest pages during test migration. > > That's what I was doing. I was running a workload in the guest that > just writes 8 bytes to a page and jumps ahead a few pages on all > vCPUs, touching most of its memory. > > But there is more to understand; I'll collect more results. I'm not > sure why the EPT can be unmapped/collapsed so quickly. Maybe something smart done by the hypervisor? > > > > > Write op should be the worst here case since it'll require the atomic op > > being applied; see kvm_tdp_mmu_write_spte(). > > > > > > > > > > > > > Since we'll still keep the intermediate levels around - from application > > > > POV, one other thing to remedy this is further shrink the size of COLLAPSE > > > > so potentially for a very large page we can start with building 2M layers. > > > > But then collapse will need to be run at least two rounds. > > > > > > That's exactly what I thought to do. :) I realized, too, that this is > > > actually how userspace *should* collapse things to avoid holding up > > > vCPUs too long. I think this is a good reason to keep intermediate > > > page sizes. > > > > > > When collapsing 4K -> 1G, the mapcount scheme doesn't actually make a > > > huge difference: the THP-like scheme is about 30% slower overall. > > > > > > When collapsing 4K -> 2M -> 1G, the mapcount scheme makes a HUGE > > > difference. For the THP-like scheme, collapsing 4K -> 2M requires > > > decrementing and then re-incrementing subpage->_mapcount, and then > > > from 2M -> 1G, we have to decrement all 262k subpages->_mapcount. For > > > the head-only scheme, for each 2M in the 4K -> 2M collapse, we > > > decrement the compound_mapcount 512 times (once per PTE), then > > > increment it once. And then for 2M -> 1G, for each 1G, we decrement > > > mapcount again by 512 (once per PMD), incrementing it once. > > > > Did you have quantified numbers (with your ktime treak) to compare these? > > If we want to go the other route, I think these will be materials to > > justify any other approach on mapcount handling. > > Ok, I can do that. GIve me a couple days to collect more results and > organize them in a helpful way. > > (If it's helpful at all, here are some results I collected last week: > [2]. Please ignore it if it's not helpful.) It's helpful already at least to me, thanks. Yes the change is drastic. > > > > > > > > > The mapcount decrements are about on par with how long it takes to do > > > other things, like updating page tables. The main problem is, with the > > > THP-like scheme (implemented like this [1]), there isn't a way to > > > avoid the 262k decrements when collapsing 1G. So if we want > > > MADV_COLLAPSE to be fast and we want a THP-like page_mapcount() API, > > > then I think something more clever needs to be implemented. > > > > > > [1]: https://github.com/48ca/linux/blob/hgmv2-jan24/mm/hugetlb.c#L127-L178 > > > > I believe the whole goal of HGM is trying to face the same challenge if > > we'll allow 1G THP exist and being able to split for anon. > > > > I don't remember whether we discussed below, maybe we did? Anyway... > > > > Another way to not use thp mapcount, nor break smaps and similar calls to > > page_mapcount() on small page, is to only increase the hpage mapcount only > > when hstate pXd (in case of 1G it's PUD) entry being populated (no matter > > as leaf or a non-leaf), and the mapcount can be decreased when the pXd > > entry is removed (for leaf, it's the same as for now; for HGM, it's when > > freeing pgtable of the PUD entry). > > Right, and this is doable. Also it seems like this is pretty close to > the direction Matthew Wilcox wants to go with THPs. I may not be familiar with it, do you mean this one? https://lore.kernel.org/all/Y9Afwds%2FJl39UjEp@xxxxxxxxxxxxxxxxxxxx/ For hugetlb I think it should be easier to maintain rather than any-sized folios, because there's the pgtable non-leaf entry to track rmap information and the folio size being static to hpage size. It'll be different to folios where it can be random sized pages chunk, so it needs to be managed by batching the ptes when install/zap. > > Something I noticed though, from the implementation of > folio_referenced()/folio_referenced_one(), is that folio_mapcount() > ought to report the total number of PTEs that are pointing on the page > (or the number of times page_vma_mapped_walk returns true). FWIW, > folio_referenced() is never called for hugetlb folios. FWIU folio_mapcount is the thing it needs for now to do the rmap walks - it'll walk every leaf page being mapped, big or small, so IIUC that number should match with what it expects to see later, more or less. But I agree the mapcount/referenced value itself is debatable to me, just like what you raised in the other thread on page migration. Meanwhile, I am not certain whether the mapcount is accurate either because AFAICT the mapcount can be modified if e.g. new page mapping established as long as before taking the page lock later in folio_referenced(). It's just that I don't see any severe issue either due to any of above, as long as that information is only used as a hint for next steps, e.g., to swap which page out. -- Peter Xu
