On Wed, Nov 03, 2010 at 04:27:38PM +0100, Andrea Arcangeli wrote: > From: Andrea Arcangeli <aarcange@xxxxxxxxxx> > > Documentation/vm/transhuge.txt > > Signed-off-by: Andrea Arcangeli <aarcange@xxxxxxxxxx> > --- > > diff --git a/Documentation/vm/transhuge.txt b/Documentation/vm/transhuge.txt > new file mode 100644 > --- /dev/null > +++ b/Documentation/vm/transhuge.txt > @@ -0,0 +1,283 @@ > += Transparent Hugepage Support = > + > +== Objective == > + > +Performance critical computing applications dealing with large memory > +working sets are already running on top of libhugetlbfs and in turn > +hugetlbfs. Transparent Hugepage Support is an alternative to > +libhugetlbfs that offers the same feature of libhugetlbfs but without > +the shortcomings of hugetlbfs (for KVM, JVM, HPC, even gcc etc..). libhugetlbfs can also automatically back shared memory, text and data with huge pages which THP cannot do. libhugetlbfs cannot demote and promote memory where THP can. They are not exactly like-with-like comparisons. How about; "Transparent Hugepage Support is an alternative means of using huge pages for the backing of anonymous memory with huge pages that supports the automatic promotion and demotion of page sizes." ? > + > +In the future it can expand over the pagecache layer starting with > +tmpfs to reduce even further the hugetlbfs usages. > + > +The reason applications are running faster is because of two > +factors. The first factor is almost completely irrelevant and it's not > +of significant interest because it'll also have the downside of > +requiring larger clear-page copy-page in page faults which is a > +potentially negative effect. The first factor consists in taking a > +single page fault for each 2M virtual region touched by userland (so > +reducing the enter/exit kernel frequency by a 512 times factor). This > +only matters the first time the memory is accessed for the lifetime of > +a memory mapping. The second long lasting and much more important > +factor will affect all subsequent accesses to the memory for the whole > +runtime of the application. The second factor consist of two > +components: 1) the TLB miss will run faster (especially with > +virtualization using nested pagetables but also on bare metal without > +virtualization) Careful on that first statement. It's not necessarily true for bare metal as some processors show that the TLB miss handler for huge pages is slower than base pages. Not sure why but it seemed to be the case on P4 anyway (at least the one I have). Maybe it was a measurement error but on chips with split TLBs for page sizes, there is no guarantee they are the same speed. It's probably true for virtualisation though considering the vastly reduced number of cache lines required to translate an address. I'd weaken the language for bare metal to say "almost always" but it's not a big deal. > and 2) a single TLB entry will be mapping a much > +larger amount of virtual memory in turn reducing the number of TLB > +misses. This on the other hand is certainly true. > +With virtualization and nested pagetables the TLB can be > +mapped of larger size only if both KVM and the Linux guest are using > +hugepages but a significant speedup already happens if only one of the > +two is using hugepages just because of the fact the TLB miss is going > +to run faster. > + > +== Design == > + > +- "graceful fallback": mm components which don't have transparent > + hugepage knownledge fall back to breaking a transparent hugepage and %s/knownledge/knowledge/ > + working on the regular pages and their respective regular pmd/pte > + mappings > + > +- if an hugepage allocation fails because of memory fragmentation, s/an/a/ > + regular pages should be gracefully allocated instead and mixed in > + the same vma without any failure or significant delay and generally > + without userland noticing > + why "generally"? At worst the application will see varying performance characteristics but that applies to a lot more than THP. > +- if some task quits and more hugepages become available (either > + immediately in the buddy or through the VM), guest physical memory > + backed by regular pages should be relocated on hugepages > + automatically (with khugepaged) > + > +- it doesn't require boot-time memory reservation and in turn it uses neither does hugetlbfs. > + hugepages whenever possible (the only possible reservation here is > + kernelcore= to avoid unmovable pages to fragment all the memory but > + such a tweak is not specific to transparent hugepage support and > + it's a generic feature that applies to all dynamic high order > + allocations in the kernel) > + > +- this initial support only offers the feature in the anonymous memory > + regions but it'd be ideal to move it to tmpfs and the pagecache > + later > + > +Transparent Hugepage Support maximizes the usefulness of free memory > +if compared to the reservation approach of hugetlbfs by allowing all > +unused memory to be used as cache or other movable (or even unmovable > +entities). hugetlbfs with memory overcommit offers something similar, particularly in combination with libhugetlbfs with can automatically fall back to base pages. I've run benchmarks comparing hugetlbfs using a static hugepage pool with hugetlbfs dynamically allocating hugepages as required with no discernable performance difference. So this statement is not strictly accurate. > +It doesn't require reservation to prevent hugepage > +allocation failures to be noticeable from userland. It allows paging > +and all other advanced VM features to be available on the > +hugepages. It requires no modifications for applications to take > +advantage of it. > + > +Applications however can be further optimized to take advantage of > +this feature, like for example they've been optimized before to avoid > +a flood of mmap system calls for every malloc(4k). Optimizing userland > +is by far not mandatory and khugepaged already can take care of long > +lived page allocations even for hugepage unaware applications that > +deals with large amounts of memory. > + > +In certain cases when hugepages are enabled system wide, application > +may end up allocating more memory resources. An application may mmap a > +large region but only touch 1 byte of it, in that case a 2M page might > +be allocated instead of a 4k page for no good. This is why it's > +possible to disable hugepages system-wide and to only have them inside > +MADV_HUGEPAGE madvise regions. > + > +Embedded systems should enable hugepages only inside madvise regions > +to eliminate any risk of wasting any precious byte of memory and to > +only run faster. > + > +Applications that gets a lot of benefit from hugepages and that don't > +risk to lose memory by using hugepages, should use > +madvise(MADV_HUGEPAGE) on their critical mmapped regions. > + > +== sysfs == > + > +Transparent Hugepage Support can be entirely disabled (mostly for > +debugging purposes) or only enabled inside MADV_HUGEPAGE regions (to > +avoid the risk of consuming more memory resources) or enabled system > +wide. This can be achieved with one of: > + > +echo always >/sys/kernel/mm/transparent_hugepage/enabled > +echo madvise >/sys/kernel/mm/transparent_hugepage/enabled > +echo never >/sys/kernel/mm/transparent_hugepage/enabled > + > +It's also possible to limit defrag efforts in the VM to generate > +hugepages in case they're not immediately free to madvise regions or > +to never try to defrag memory and simply fallback to regular pages > +unless hugepages are immediately available. This is the first mention of defrag but hey, it's not a paper :) > Clearly if we spend CPU > +time to defrag memory, we would expect to gain even more by the fact > +we use hugepages later instead of regular pages. This isn't always > +guaranteed, but it may be more likely in case the allocation is for a > +MADV_HUGEPAGE region. > + > +echo always >/sys/kernel/mm/transparent_hugepage/defrag > +echo madvise >/sys/kernel/mm/transparent_hugepage/defrag > +echo never >/sys/kernel/mm/transparent_hugepage/defrag > + > +khugepaged will be automatically started when > +transparent_hugepage/enabled is set to "always" or "madvise, and it'll > +be automatically shutdown if it's set to "never". > + > +khugepaged runs usually at low frequency so while one may not want to > +invoke defrag algorithms synchronously during the page faults, it > +should be worth invoking defrag at least in khugepaged. However it's > +also possible to disable defrag in khugepaged: > + > +echo yes >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag > +echo no >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag > + > +You can also control how many pages khugepaged should scan at each > +pass: > + > +/sys/kernel/mm/transparent_hugepage/khugepaged/pages_to_scan > + > +and how many milliseconds to wait in khugepaged between each pass (you > +can se this to 0 to run khugepaged at 100% utilization of one core): s/se/set/ > + > +/sys/kernel/mm/transparent_hugepage/khugepaged/scan_sleep_millisecs > + > +and how many milliseconds to wait in khugepaged if there's an hugepage > +allocation failure to throttle the next allocation attempt. > + > +/sys/kernel/mm/transparent_hugepage/khugepaged/alloc_sleep_millisecs > + > +The khugepaged progress can be seen in the number of pages collapsed: > + > +/sys/kernel/mm/transparent_hugepage/khugepaged/pages_collapsed > + > +for each pass: > + > +/sys/kernel/mm/transparent_hugepage/khugepaged/full_scans > + > +== Boot parameter == > + > +You can change the sysfs boot time defaults of Transparent Hugepage > +Support by passing the parameter "transparent_hugepage=always" or > +"transparent_hugepage=madvise" or "transparent_hugepage=never" > +(without "") to the kernel command line. > + > +== Need of restart == > + Need of application restart? A casual reader might otherwise interpret it as a system restart for some godawful reason of their own. > +The transparent_hugepage/enabled values only affect future > +behavior. So to make them effective you need to restart any s/behavior/behaviour/ > +application that could have been using hugepages. This also applies to > +the regions registered in khugepaged. > + > +== get_user_pages and follow_page == > + > +get_user_pages and follow_page if run on a hugepage, will return the > +head or tail pages as usual (exactly as they would do on > +hugetlbfs). Most gup users will only care about the actual physical > +address of the page and its temporary pinning to release after the I/O > +is complete, so they won't ever notice the fact the page is huge. But > +if any driver is going to mangle over the page structure of the tail > +page (like for checking page->mapping or other bits that are relevant > +for the head page and not the tail page), it should be updated to jump > +to check head page instead (while serializing properly against > +split_huge_page() to avoid the head and tail pages to disappear from > +under it, see the futex code to see an example of that, hugetlbfs also > +needed special handling in futex code for similar reasons). > + > +NOTE: these aren't new constraints to the GUP API, and they match the > +same constrains that applies to hugetlbfs too, so any driver capable > +of handling GUP on hugetlbfs will also work fine on transparent > +hugepage backed mappings. > + > +In case you can't handle compound pages if they're returned by > +follow_page, the FOLL_SPLIT bit can be specified as parameter to > +follow_page, so that it will split the hugepages before returning > +them. Migration for example passes FOLL_SPLIT as parameter to > +follow_page because it's not hugepage aware and in fact it can't work > +at all on hugetlbfs (but it instead works fine on transparent hugetlbfs pages can now migrate although it's only used by hwpoison. > +hugepages thanks to FOLL_SPLIT). migration simply can't deal with > +hugepages being returned (as it's not only checking the pfn of the > +page and pinning it during the copy but it pretends to migrate the > +memory in regular page sizes and with regular pte/pmd mappings). > + > +== Optimizing the applications == > + > +To be guaranteed that the kernel will map a 2M page immediately in any > +memory region, the mmap region has to be hugepage naturally > +aligned. posix_memalign() can provide that guarantee. > + > +== Hugetlbfs == > + > +You can use hugetlbfs on a kernel that has transparent hugepage > +support enabled just fine as always. No difference can be noted in > +hugetlbfs other than there will be less overall fragmentation. All > +usual features belonging to hugetlbfs are preserved and > +unaffected. libhugetlbfs will also work fine as usual. > + > +== Graceful fallback == > + > +Code walking pagetables but unware about huge pmds can simply call > +split_huge_page_pmd(mm, pmd) where the pmd is the one returned by > +pmd_offset. It's trivial to make the code transparent hugepage aware > +by just grepping for "pmd_offset" and adding split_huge_page_pmd where > +missing after pmd_offset returns the pmd. Thanks to the graceful > +fallback design, with a one liner change, you can avoid to write > +hundred if not thousand of lines of complex code to make your code > +hugepage aware. > + It'd be nice if you could point to a specific example but by no means mandatory. > +If you're not walking pagetables but you run into a physical hugepage > +but you can't handle it natively in your code, you can split it by > +calling split_huge_page(page). This is what the Linux VM does before > +it tries to swapout the hugepage for example. > + > +== Locking in hugepage aware code == > + > +We want as much code as possible hugepage aware, as calling > +split_huge_page() or split_huge_page_pmd() has a cost. > + > +To make pagetable walks huge pmd aware, all you need to do is to call > +pmd_trans_huge() on the pmd returned by pmd_offset. You must hold the > +mmap_sem in read (or write) mode to be sure an huge pmd cannot be > +created from under you by khugepaged (khugepaged collapse_huge_page > +takes the mmap_sem in write mode in addition to the anon_vma lock). If > +pmd_trans_huge returns false, you just fallback in the old code > +paths. If instead pmd_trans_huge returns true, you have to take the > +mm->page_table_lock and re-run pmd_trans_huge. Taking the > +page_table_lock will prevent the huge pmd to be converted into a > +regular pmd from under you (split_huge_page can run in parallel to the > +pagetable walk). If the second pmd_trans_huge returns false, you > +should just drop the page_table_lock and fallback to the old code as > +before. Otherwise you should run pmd_trans_splitting on the pmd. In > +case pmd_trans_splitting returns true, it means split_huge_page is > +already in the middle of splitting the page. So if pmd_trans_splitting > +returns true it's enough to drop the page_table_lock and call > +wait_split_huge_page and then fallback the old code paths. You are > +guaranteed by the time wait_split_huge_page returns, the pmd isn't > +huge anymore. If pmd_trans_splitting returns false, you can proceed to > +process the huge pmd and the hugepage natively. Once finished you can > +drop the page_table_lock. > + > +== compound_lock, get_user_pages and put_page == > + > +split_huge_page internally has to distribute the refcounts in the head > +page to the tail pages before clearing all PG_head/tail bits from the > +page structures. It can do that easily for refcounts taken by huge pmd > +mappings. But the GUI API as created by hugetlbfs (that returns head > +and tail pages if running get_user_pages on an address backed by any > +hugepage), requires the refcount to be accounted on the tail pages and > +not only in the head pages, if we want to be able to run > +split_huge_page while there are gup pins established on any tail > +page. Failure to be able to run split_huge_page if there's any gup pin > +on any tail page, would mean having to split all hugepages upfront in > +get_user_pages which is unacceptable as too many gup users are > +performance critical and they must work natively on hugepages like > +they work natively on hugetlbfs already (hugetlbfs is simpler because > +hugetlbfs pages cannot be splitted so there wouldn't be requirement of > +accounting the pins on the tail pages for hugetlbfs). If we wouldn't > +account the gup refcounts on the tail pages during gup, we won't know > +anymore which tail page is pinned by gup and which is not while we run > +split_huge_page. But we still have to add the gup pin to the head page > +too, to know when we can free the compound page in case it's never > +splitted during its lifetime. That requires changing not just > +get_page, but put_page as well so that when put_page runs on a tail > +page (and only on a tail page) it will find its respective head page, > +and then it will decrease the head page refcount in addition to the > +tail page refcount. To obtain a head page reliably and to decrease its > +refcount without race conditions, put_page has to serialize against > +__split_huge_page_refcount using a special per-page lock called > +compound_lock. > Ok, I'll need to read the rest of the series to verify if this is correct but by and large it looks good. I think some of the language is stronger than it should be and some of the comparisons with libhugetlbfs are a bit off but I'd be naturally defensive on that topic. Make the suggested changes if you like but if you don't, it shouldn't affect the series. -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>