1.. _transhuge: 2 3============================ 4Transparent Hugepage Support 5============================ 6 7This document describes design principles for Transparent Hugepage (THP) 8support and its interaction with other parts of the memory management 9system. 10 11Design principles 12================= 13 14- "graceful fallback": mm components which don't have transparent hugepage 15 knowledge fall back to breaking huge pmd mapping into table of ptes and, 16 if necessary, split a transparent hugepage. Therefore these components 17 can continue working on the regular pages or regular pte mappings. 18 19- if a hugepage allocation fails because of memory fragmentation, 20 regular pages should be gracefully allocated instead and mixed in 21 the same vma without any failure or significant delay and without 22 userland noticing 23 24- if some task quits and more hugepages become available (either 25 immediately in the buddy or through the VM), guest physical memory 26 backed by regular pages should be relocated on hugepages 27 automatically (with khugepaged) 28 29- it doesn't require memory reservation and in turn it uses hugepages 30 whenever possible (the only possible reservation here is kernelcore= 31 to avoid unmovable pages to fragment all the memory but such a tweak 32 is not specific to transparent hugepage support and it's a generic 33 feature that applies to all dynamic high order allocations in the 34 kernel) 35 36get_user_pages and follow_page 37============================== 38 39get_user_pages and follow_page if run on a hugepage, will return the 40head or tail pages as usual (exactly as they would do on 41hugetlbfs). Most GUP users will only care about the actual physical 42address of the page and its temporary pinning to release after the I/O 43is complete, so they won't ever notice the fact the page is huge. But 44if any driver is going to mangle over the page structure of the tail 45page (like for checking page->mapping or other bits that are relevant 46for the head page and not the tail page), it should be updated to jump 47to check head page instead. Taking a reference on any head/tail page would 48prevent the page from being split by anyone. 49 50.. note:: 51 these aren't new constraints to the GUP API, and they match the 52 same constraints that apply to hugetlbfs too, so any driver capable 53 of handling GUP on hugetlbfs will also work fine on transparent 54 hugepage backed mappings. 55 56Graceful fallback 57================= 58 59Code walking pagetables but unaware about huge pmds can simply call 60split_huge_pmd(vma, pmd, addr) where the pmd is the one returned by 61pmd_offset. It's trivial to make the code transparent hugepage aware 62by just grepping for "pmd_offset" and adding split_huge_pmd where 63missing after pmd_offset returns the pmd. Thanks to the graceful 64fallback design, with a one liner change, you can avoid to write 65hundreds if not thousands of lines of complex code to make your code 66hugepage aware. 67 68If you're not walking pagetables but you run into a physical hugepage 69that you can't handle natively in your code, you can split it by 70calling split_huge_page(page). This is what the Linux VM does before 71it tries to swapout the hugepage for example. split_huge_page() can fail 72if the page is pinned and you must handle this correctly. 73 74Example to make mremap.c transparent hugepage aware with a one liner 75change:: 76 77 diff --git a/mm/mremap.c b/mm/mremap.c 78 --- a/mm/mremap.c 79 +++ b/mm/mremap.c 80 @@ -41,6 +41,7 @@ static pmd_t *get_old_pmd(struct mm_stru 81 return NULL; 82 83 pmd = pmd_offset(pud, addr); 84 + split_huge_pmd(vma, pmd, addr); 85 if (pmd_none_or_clear_bad(pmd)) 86 return NULL; 87 88Locking in hugepage aware code 89============================== 90 91We want as much code as possible hugepage aware, as calling 92split_huge_page() or split_huge_pmd() has a cost. 93 94To make pagetable walks huge pmd aware, all you need to do is to call 95pmd_trans_huge() on the pmd returned by pmd_offset. You must hold the 96mmap_lock in read (or write) mode to be sure a huge pmd cannot be 97created from under you by khugepaged (khugepaged collapse_huge_page 98takes the mmap_lock in write mode in addition to the anon_vma lock). If 99pmd_trans_huge returns false, you just fallback in the old code 100paths. If instead pmd_trans_huge returns true, you have to take the 101page table lock (pmd_lock()) and re-run pmd_trans_huge. Taking the 102page table lock will prevent the huge pmd being converted into a 103regular pmd from under you (split_huge_pmd can run in parallel to the 104pagetable walk). If the second pmd_trans_huge returns false, you 105should just drop the page table lock and fallback to the old code as 106before. Otherwise, you can proceed to process the huge pmd and the 107hugepage natively. Once finished, you can drop the page table lock. 108 109Refcounts and transparent huge pages 110==================================== 111 112Refcounting on THP is mostly consistent with refcounting on other compound 113pages: 114 115 - get_page()/put_page() and GUP operate on head page's ->_refcount. 116 117 - ->_refcount in tail pages is always zero: get_page_unless_zero() never 118 succeeds on tail pages. 119 120 - map/unmap of PMD entry for the whole compound page increment/decrement 121 ->compound_mapcount, stored in the first tail page of the compound page; 122 and also increment/decrement ->subpages_mapcount (also in the first tail) 123 by COMPOUND_MAPPED when compound_mapcount goes from -1 to 0 or 0 to -1. 124 125 - map/unmap of sub-pages with PTE entry increment/decrement ->_mapcount 126 on relevant sub-page of the compound page, and also increment/decrement 127 ->subpages_mapcount, stored in first tail page of the compound page, when 128 _mapcount goes from -1 to 0 or 0 to -1: counting sub-pages mapped by PTE. 129 130split_huge_page internally has to distribute the refcounts in the head 131page to the tail pages before clearing all PG_head/tail bits from the page 132structures. It can be done easily for refcounts taken by page table 133entries, but we don't have enough information on how to distribute any 134additional pins (i.e. from get_user_pages). split_huge_page() fails any 135requests to split pinned huge pages: it expects page count to be equal to 136the sum of mapcount of all sub-pages plus one (split_huge_page caller must 137have a reference to the head page). 138 139split_huge_page uses migration entries to stabilize page->_refcount and 140page->_mapcount of anonymous pages. File pages just get unmapped. 141 142We are safe against physical memory scanners too: the only legitimate way 143a scanner can get a reference to a page is get_page_unless_zero(). 144 145All tail pages have zero ->_refcount until atomic_add(). This prevents the 146scanner from getting a reference to the tail page up to that point. After the 147atomic_add() we don't care about the ->_refcount value. We already know how 148many references should be uncharged from the head page. 149 150For head page get_page_unless_zero() will succeed and we don't mind. It's 151clear where references should go after split: it will stay on the head page. 152 153Note that split_huge_pmd() doesn't have any limitations on refcounting: 154pmd can be split at any point and never fails. 155 156Partial unmap and deferred_split_huge_page() 157============================================ 158 159Unmapping part of THP (with munmap() or other way) is not going to free 160memory immediately. Instead, we detect that a subpage of THP is not in use 161in page_remove_rmap() and queue the THP for splitting if memory pressure 162comes. Splitting will free up unused subpages. 163 164Splitting the page right away is not an option due to locking context in 165the place where we can detect partial unmap. It also might be 166counterproductive since in many cases partial unmap happens during exit(2) if 167a THP crosses a VMA boundary. 168 169The function deferred_split_huge_page() is used to queue a page for splitting. 170The splitting itself will happen when we get memory pressure via shrinker 171interface. 172