Andrew, this patchset is still not ready for mmotm. Hugetlb freeing path needs a major surgery because it can be called from atomic contexts and a simple fix http://lkml.kernel.org/r/20210311021321.127500-1-mike.kravetz@xxxxxxxxxx has been nacked by Peter. Until this is handled properly this has to wait and build on top. On Mon 15-03-21 13:48:19, Andrew Morton wrote: > From: Muchun Song <songmuchun@xxxxxxxxxxxxx> > Subject: mm: hugetlb: alloc the vmemmap pages associated with each HugeTLB page > > When we free a HugeTLB page to the buddy allocator, we need to allocate > the vmemmap pages associated with it. However, we may not be able to > allocate the vmemmap pages when the system is under memory pressure. In > this case, we just refuse to free the HugeTLB page. This changes behavior > in some corner cases as listed below: > > 1) Failing to free a huge page triggered by the user (decrease nr_pages). > > User needs to try again later. > > 2) Failing to free a surplus huge page when freed by the application. > > Try again later when freeing a huge page next time. > > 3) Failing to dissolve a free huge page on ZONE_MOVABLE via > offline_pages(). > > This can happen when we have plenty of ZONE_MOVABLE memory, but > not enough kernel memory to allocate vmemmmap pages. We may even > be able to migrate huge page contents, but will not be able to > dissolve the source huge page. This will prevent an offline > operation and is unfortunate as memory offlining is expected to > succeed on movable zones. Users that depend on memory hotplug > to succeed for movable zones should carefully consider whether the > memory savings gained from this feature are worth the risk of > possibly not being able to offline memory in certain situations. > > 4) Failing to dissolve a huge page on CMA/ZONE_MOVABLE via > alloc_contig_range() - once we have that handling in place. Mainly > affects CMA and virtio-mem. > > Similar to 3). virito-mem will handle migration errors gracefully. > CMA might be able to fallback on other free areas within the CMA > region. > > Vmemmap pages are allocated from the page freeing context. In order for > those allocations to be not disruptive (e.g. trigger oom killer) > __GFP_NORETRY is used. hugetlb_lock is dropped for the allocation because > a non sleeping allocation would be too fragile and it could fail too > easily under memory pressure. GFP_ATOMIC or other modes to access memory > reserves is not used because we want to prevent consuming reserves under > heavy hugetlb freeing. > > Link: https://lkml.kernel.org/r/20210315092015.35396-6-songmuchun@xxxxxxxxxxxxx > Signed-off-by: Muchun Song <songmuchun@xxxxxxxxxxxxx> > Tested-by: Chen Huang <chenhuang5@xxxxxxxxxx> > Tested-by: Bodeddula Balasubramaniam <bodeddub@xxxxxxxxxx> > Reviewed-by: Oscar Salvador <osalvador@xxxxxxx> > Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx> > Cc: Andy Lutomirski <luto@xxxxxxxxxx> > Cc: Anshuman Khandual <anshuman.khandual@xxxxxxx> > Cc: Balbir Singh <bsingharora@xxxxxxxxx> > Cc: Barry Song <song.bao.hua@xxxxxxxxxxxxx> > Cc: Borislav Petkov <bp@xxxxxxxxx> > Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx> > Cc: David Hildenbrand <david@xxxxxxxxxx> > Cc: David Rientjes <rientjes@xxxxxxxxxx> > Cc: "H. Peter Anvin" <hpa@xxxxxxxxx> > Cc: Ingo Molnar <mingo@xxxxxxxxxx> > Cc: Joao Martins <joao.m.martins@xxxxxxxxxx> > Cc: Joerg Roedel <jroedel@xxxxxxx> > Cc: Jonathan Corbet <corbet@xxxxxxx> > Cc: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx> > Cc: Mauro Carvalho Chehab <mchehab+huawei@xxxxxxxxxx> > Cc: Miaohe Lin <linmiaohe@xxxxxxxxxx> > Cc: Michal Hocko <mhocko@xxxxxxxx> > Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx> > Cc: Mina Almasry <almasrymina@xxxxxxxxxx> > Cc: Naoya Horiguchi <naoya.horiguchi@xxxxxxx> > Cc: Oliver Neukum <oneukum@xxxxxxxx> > Cc: "Paul E. McKenney" <paulmck@xxxxxxxxxx> > Cc: Pawan Gupta <pawan.kumar.gupta@xxxxxxxxxxxxxxx> > Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx> > Cc: Randy Dunlap <rdunlap@xxxxxxxxxxxxx> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx> > Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> > --- > > Documentation/admin-guide/mm/hugetlbpage.rst | 8 + > Documentation/admin-guide/mm/memory-hotplug.rst | 13 ++ > include/linux/mm.h | 2 > mm/hugetlb.c | 76 +++++++++++--- > mm/hugetlb_vmemmap.c | 43 +++++-- > mm/hugetlb_vmemmap.h | 23 ++++ > mm/sparse-vmemmap.c | 75 +++++++++++++ > 7 files changed, 211 insertions(+), 29 deletions(-) > > --- a/Documentation/admin-guide/mm/hugetlbpage.rst~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/Documentation/admin-guide/mm/hugetlbpage.rst > @@ -60,6 +60,10 @@ HugePages_Surp > the pool above the value in ``/proc/sys/vm/nr_hugepages``. The > maximum number of surplus huge pages is controlled by > ``/proc/sys/vm/nr_overcommit_hugepages``. > + Note: When the feature of freeing unused vmemmap pages associated > + with each hugetlb page is enabled, the number of surplus huge pages > + may be temporarily larger than the maximum number of surplus huge > + pages when the system is under memory pressure. > Hugepagesize > is the default hugepage size (in Kb). > Hugetlb > @@ -80,6 +84,10 @@ returned to the huge page pool when free > privileges can dynamically allocate more or free some persistent huge pages > by increasing or decreasing the value of ``nr_hugepages``. > > +Note: When the feature of freeing unused vmemmap pages associated with each > +hugetlb page is enabled, we can fail to free the huge pages triggered by > +the user when ths system is under memory pressure. Please try again later. > + > Pages that are used as huge pages are reserved inside the kernel and cannot > be used for other purposes. Huge pages cannot be swapped out under > memory pressure. > --- a/Documentation/admin-guide/mm/memory-hotplug.rst~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/Documentation/admin-guide/mm/memory-hotplug.rst > @@ -357,6 +357,19 @@ creates ZONE_MOVABLE as following. > Unfortunately, there is no information to show which memory block belongs > to ZONE_MOVABLE. This is TBD. > > + Memory offlining can fail when dissolving a free huge page on ZONE_MOVABLE > + and the feature of freeing unused vmemmap pages associated with each hugetlb > + page is enabled. > + > + This can happen when we have plenty of ZONE_MOVABLE memory, but not enough > + kernel memory to allocate vmemmmap pages. We may even be able to migrate > + huge page contents, but will not be able to dissolve the source huge page. > + This will prevent an offline operation and is unfortunate as memory offlining > + is expected to succeed on movable zones. Users that depend on memory hotplug > + to succeed for movable zones should carefully consider whether the memory > + savings gained from this feature are worth the risk of possibly not being > + able to offline memory in certain situations. > + > .. _memory_hotplug_how_to_offline_memory: > > How to offline memory > --- a/include/linux/mm.h~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/include/linux/mm.h > @@ -3006,6 +3006,8 @@ static inline void print_vma_addr(char * > > void vmemmap_remap_free(unsigned long start, unsigned long end, > unsigned long reuse); > +int vmemmap_remap_alloc(unsigned long start, unsigned long end, > + unsigned long reuse, gfp_t gfp_mask); > > void *sparse_buffer_alloc(unsigned long size); > struct page * __populate_section_memmap(unsigned long pfn, > --- a/mm/hugetlb.c~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/mm/hugetlb.c > @@ -1329,16 +1329,53 @@ static inline void destroy_compound_giga > unsigned int order) { } > #endif > > -static void update_and_free_page(struct hstate *h, struct page *page) > +static int update_and_free_page_surplus(struct hstate *h, struct page *page, > + bool acct_surplus) > + __releases(&hugetlb_lock) __acquires(&hugetlb_lock) > { > int i; > struct page *subpage = page; > + int nid = page_to_nid(page); > > if (hstate_is_gigantic(h) && !gigantic_page_runtime_supported()) > - return; > + return 0; > > h->nr_huge_pages--; > - h->nr_huge_pages_node[page_to_nid(page)]--; > + h->nr_huge_pages_node[nid]--; > + > + /* > + * If the vmemmap pages associated with the HugeTLB page can be > + * optimized, we might block in alloc_huge_page_vmemmap(), so > + * drop the hugetlb_lock. > + */ > + if (free_vmemmap_pages_per_hpage(h)) > + spin_unlock(&hugetlb_lock); > + > + if (alloc_huge_page_vmemmap(h, page)) { > + spin_lock(&hugetlb_lock); > + INIT_LIST_HEAD(&page->lru); > + h->nr_huge_pages++; > + h->nr_huge_pages_node[nid]++; > + > + /* > + * If we cannot allocate vmemmap pages, just refuse to free the > + * page and put the page back on the hugetlb free list and treat > + * as a surplus page. > + */ > + if (acct_surplus) { > + h->surplus_huge_pages++; > + h->surplus_huge_pages_node[nid]++; > + } > + > + arch_clear_hugepage_flags(page); > + enqueue_huge_page(h, page); > + > + return -ENOMEM; > + } > + > + if (free_vmemmap_pages_per_hpage(h)) > + spin_lock(&hugetlb_lock); > + > for (i = 0; i < pages_per_huge_page(h); > i++, subpage = mem_map_next(subpage, page, i)) { > subpage->flags &= ~(1 << PG_locked | 1 << PG_error | > @@ -1362,6 +1399,13 @@ static void update_and_free_page(struct > } else { > __free_pages(page, huge_page_order(h)); > } > + > + return 0; > +} > + > +static inline int update_and_free_page(struct hstate *h, struct page *page) > +{ > + return update_and_free_page_surplus(h, page, true); > } > > struct hstate *size_to_hstate(unsigned long size) > @@ -1429,9 +1473,9 @@ static void __free_huge_page(struct page > } else if (h->surplus_huge_pages_node[nid]) { > /* remove the page from active list */ > list_del(&page->lru); > - update_and_free_page(h, page); > h->surplus_huge_pages--; > h->surplus_huge_pages_node[nid]--; > + update_and_free_page(h, page); > } else { > arch_clear_hugepage_flags(page); > enqueue_huge_page(h, page); > @@ -1472,7 +1516,7 @@ void free_huge_page(struct page *page) > /* > * Defer freeing if in non-task context to avoid hugetlb_lock deadlock. > */ > - if (!in_task()) { > + if (in_atomic()) { > /* > * Only call schedule_work() if hpage_freelist is previously > * empty. Otherwise, schedule_work() had been called but the > @@ -1719,14 +1763,14 @@ static int free_pool_huge_page(struct hs > list_entry(h->hugepage_freelists[node].next, > struct page, lru); > list_del(&page->lru); > + ClearHPageFreed(page); > h->free_huge_pages--; > h->free_huge_pages_node[node]--; > if (acct_surplus) { > h->surplus_huge_pages--; > h->surplus_huge_pages_node[node]--; > } > - update_and_free_page(h, page); > - ret = 1; > + ret = !update_and_free_page(h, page); > break; > } > } > @@ -1739,10 +1783,14 @@ static int free_pool_huge_page(struct hs > * nothing for in-use hugepages and non-hugepages. > * This function returns values like below: > * > - * -EBUSY: failed to dissolved free hugepages or the hugepage is in-use > - * (allocated or reserved.) > - * 0: successfully dissolved free hugepages or the page is not a > - * hugepage (considered as already dissolved) > + * -ENOMEM: failed to allocate vmemmap pages to free the freed hugepages > + * when the system is under memory pressure and the feature of > + * freeing unused vmemmap pages associated with each hugetlb page > + * is enabled. > + * -EBUSY: failed to dissolved free hugepages or the hugepage is in-use > + * (allocated or reserved.) > + * 0: successfully dissolved free hugepages or the page is not a > + * hugepage (considered as already dissolved) > */ > int dissolve_free_huge_page(struct page *page) > { > @@ -1794,11 +1842,13 @@ retry: > ClearPageHWPoison(head); > } > list_del(&head->lru); > + ClearHPageFreed(page); > h->free_huge_pages--; > h->free_huge_pages_node[nid]--; > h->max_huge_pages--; > - update_and_free_page(h, head); > - rc = 0; > + rc = update_and_free_page_surplus(h, head, false); > + if (rc) > + h->max_huge_pages++; > } > out: > spin_unlock(&hugetlb_lock); > --- a/mm/hugetlb_vmemmap.c~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/mm/hugetlb_vmemmap.c > @@ -18,10 +18,9 @@ > * 4096 base pages. For each base page, there is a corresponding page struct. > * > * Within the HugeTLB subsystem, only the first 4 page structs are used to > - * contain unique information about a HugeTLB page. HUGETLB_CGROUP_MIN_ORDER > - * provides this upper limit. The only 'useful' information in the remaining > - * page structs is the compound_head field, and this field is the same for all > - * tail pages. > + * contain unique information about a HugeTLB page. __NR_USED_SUBPAGE provides > + * this upper limit. The only 'useful' information in the remaining page structs > + * is the compound_head field, and this field is the same for all tail pages. > * > * By removing redundant page structs for HugeTLB pages, memory can be returned > * to the buddy allocator for other uses. > @@ -181,21 +180,35 @@ > #define RESERVE_VMEMMAP_NR 2U > #define RESERVE_VMEMMAP_SIZE (RESERVE_VMEMMAP_NR << PAGE_SHIFT) > > -/* > - * How many vmemmap pages associated with a HugeTLB page that can be freed > - * to the buddy allocator. > - * > - * Todo: Returns zero for now, which means the feature is disabled. We will > - * enable it once all the infrastructure is there. > - */ > -static inline unsigned int free_vmemmap_pages_per_hpage(struct hstate *h) > +static inline unsigned long free_vmemmap_pages_size_per_hpage(struct hstate *h) > { > - return 0; > + return (unsigned long)free_vmemmap_pages_per_hpage(h) << PAGE_SHIFT; > } > > -static inline unsigned long free_vmemmap_pages_size_per_hpage(struct hstate *h) > +/* > + * Previously discarded vmemmap pages will be allocated and remapping > + * after this function returns. > + */ > +int alloc_huge_page_vmemmap(struct hstate *h, struct page *head) > { > - return (unsigned long)free_vmemmap_pages_per_hpage(h) << PAGE_SHIFT; > + unsigned long vmemmap_addr = (unsigned long)head; > + unsigned long vmemmap_end, vmemmap_reuse; > + > + if (!free_vmemmap_pages_per_hpage(h)) > + return 0; > + > + vmemmap_addr += RESERVE_VMEMMAP_SIZE; > + vmemmap_end = vmemmap_addr + free_vmemmap_pages_size_per_hpage(h); > + vmemmap_reuse = vmemmap_addr - PAGE_SIZE; > + /* > + * The pages which the vmemmap virtual address range [@vmemmap_addr, > + * @vmemmap_end) are mapped to are freed to the buddy allocator, and > + * the range is mapped to the page which @vmemmap_reuse is mapped to. > + * When a HugeTLB page is freed to the buddy allocator, previously > + * discarded vmemmap pages must be allocated and remapping. > + */ > + return vmemmap_remap_alloc(vmemmap_addr, vmemmap_end, vmemmap_reuse, > + GFP_KERNEL | __GFP_NORETRY | __GFP_THISNODE); > } > > void free_huge_page_vmemmap(struct hstate *h, struct page *head) > --- a/mm/hugetlb_vmemmap.h~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/mm/hugetlb_vmemmap.h > @@ -11,10 +11,33 @@ > #include <linux/hugetlb.h> > > #ifdef CONFIG_HUGETLB_PAGE_FREE_VMEMMAP > +int alloc_huge_page_vmemmap(struct hstate *h, struct page *head); > void free_huge_page_vmemmap(struct hstate *h, struct page *head); > + > +/* > + * How many vmemmap pages associated with a HugeTLB page that can be freed > + * to the buddy allocator. > + * > + * Todo: Returns zero for now, which means the feature is disabled. We will > + * enable it once all the infrastructure is there. > + */ > +static inline unsigned int free_vmemmap_pages_per_hpage(struct hstate *h) > +{ > + return 0; > +} > #else > +static inline int alloc_huge_page_vmemmap(struct hstate *h, struct page *head) > +{ > + return 0; > +} > + > static inline void free_huge_page_vmemmap(struct hstate *h, struct page *head) > { > } > + > +static inline unsigned int free_vmemmap_pages_per_hpage(struct hstate *h) > +{ > + return 0; > +} > #endif /* CONFIG_HUGETLB_PAGE_FREE_VMEMMAP */ > #endif /* _LINUX_HUGETLB_VMEMMAP_H */ > --- a/mm/sparse-vmemmap.c~mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page > +++ a/mm/sparse-vmemmap.c > @@ -40,7 +40,8 @@ > * @remap_pte: called for each lowest-level entry (PTE). > * @reuse_page: the page which is reused for the tail vmemmap pages. > * @reuse_addr: the virtual address of the @reuse_page page. > - * @vmemmap_pages: the list head of the vmemmap pages that can be freed. > + * @vmemmap_pages: the list head of the vmemmap pages that can be freed > + * or is mapped from. > */ > struct vmemmap_remap_walk { > void (*remap_pte)(pte_t *pte, unsigned long addr, > @@ -224,6 +225,78 @@ void vmemmap_remap_free(unsigned long st > free_vmemmap_page_list(&vmemmap_pages); > } > > +static void vmemmap_restore_pte(pte_t *pte, unsigned long addr, > + struct vmemmap_remap_walk *walk) > +{ > + pgprot_t pgprot = PAGE_KERNEL; > + struct page *page; > + void *to; > + > + BUG_ON(pte_page(*pte) != walk->reuse_page); > + > + page = list_first_entry(walk->vmemmap_pages, struct page, lru); > + list_del(&page->lru); > + to = page_to_virt(page); > + copy_page(to, (void *)walk->reuse_addr); > + > + set_pte_at(&init_mm, addr, pte, mk_pte(page, pgprot)); > +} > + > +static int alloc_vmemmap_page_list(unsigned long start, unsigned long end, > + gfp_t gfp_mask, struct list_head *list) > +{ > + unsigned long nr_pages = (end - start) >> PAGE_SHIFT; > + int nid = page_to_nid((struct page *)start); > + struct page *page, *next; > + > + while (nr_pages--) { > + page = alloc_pages_node(nid, gfp_mask, 0); > + if (!page) > + goto out; > + list_add_tail(&page->lru, list); > + } > + > + return 0; > +out: > + list_for_each_entry_safe(page, next, list, lru) > + __free_pages(page, 0); > + return -ENOMEM; > +} > + > +/** > + * vmemmap_remap_alloc - remap the vmemmap virtual address range [@start, end) > + * to the page which is from the @vmemmap_pages > + * respectively. > + * @start: start address of the vmemmap virtual address range that we want > + * to remap. > + * @end: end address of the vmemmap virtual address range that we want to > + * remap. > + * @reuse: reuse address. > + * @gpf_mask: GFP flag for allocating vmemmap pages. > + */ > +int vmemmap_remap_alloc(unsigned long start, unsigned long end, > + unsigned long reuse, gfp_t gfp_mask) > +{ > + LIST_HEAD(vmemmap_pages); > + struct vmemmap_remap_walk walk = { > + .remap_pte = vmemmap_restore_pte, > + .reuse_addr = reuse, > + .vmemmap_pages = &vmemmap_pages, > + }; > + > + /* See the comment in the vmemmap_remap_free(). */ > + BUG_ON(start - reuse != PAGE_SIZE); > + > + might_sleep_if(gfpflags_allow_blocking(gfp_mask)); > + > + if (alloc_vmemmap_page_list(start, end, gfp_mask, &vmemmap_pages)) > + return -ENOMEM; > + > + vmemmap_remap_range(reuse, end, &walk); > + > + return 0; > +} > + > /* > * Allocate a block of memory to be used to back the virtual memory map > * or to back the page tables that are used to create the mapping. > _ > > Patches currently in -mm which might be from songmuchun@xxxxxxxxxxxxx are > > mm-memcontrol-fix-kernel-stack-account.patch > mm-memory_hotplug-factor-out-bootmem-core-functions-to-bootmem_infoc.patch > mm-hugetlb-introduce-a-new-config-hugetlb_page_free_vmemmap.patch > mm-hugetlb-gather-discrete-indexes-of-tail-page.patch > mm-hugetlb-free-the-vmemmap-pages-associated-with-each-hugetlb-page.patch > mm-hugetlb-alloc-the-vmemmap-pages-associated-with-each-hugetlb-page.patch > mm-hugetlb-set-the-pagehwpoison-to-the-raw-error-page.patch > mm-hugetlb-add-a-kernel-parameter-hugetlb_free_vmemmap.patch > mm-hugetlb-introduce-nr_free_vmemmap_pages-in-the-struct-hstate.patch -- Michal Hocko SUSE Labs
