| Message ID | 20210308164520.18323-1-david@redhat.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
| Series | [RFCv2] mm/madvise: introduce MADV_POPULATE_(READ|WRITE) to prefault/prealloc memory | expand |
> diff --git a/mm/internal.h b/mm/internal.h > index 9902648f2206..a5c4ed23b1db 100644 > --- a/mm/internal.h > +++ b/mm/internal.h > @@ -340,6 +340,9 @@ void __vma_unlink_list(struct mm_struct *mm, > struct vm_area_struct *vma); > #ifdef CONFIG_MMU > extern long populate_vma_page_range(struct vm_area_struct *vma, > unsigned long start, unsigned long end, int *nonblocking); > +extern long faultin_vma_page_range(struct vm_area_struct *vma, > + unsigned long start, unsigned long end, > + bool write, int *nonblocking); > extern void munlock_vma_pages_range(struct vm_area_struct *vma, > unsigned long start, unsigned long end); > static inline void munlock_vma_pages_all(struct vm_area_struct *vma) The parameter name does not match the one in the implementation. Otherwise the implementation looks fine AFAICT. Eike
On 09.03.21 08:35, Rolf Eike Beer wrote: >> diff --git a/mm/internal.h b/mm/internal.h >> index 9902648f2206..a5c4ed23b1db 100644 >> --- a/mm/internal.h >> +++ b/mm/internal.h >> @@ -340,6 +340,9 @@ void __vma_unlink_list(struct mm_struct *mm, >> struct vm_area_struct *vma); >> #ifdef CONFIG_MMU >> extern long populate_vma_page_range(struct vm_area_struct *vma, >> unsigned long start, unsigned long end, int *nonblocking); >> +extern long faultin_vma_page_range(struct vm_area_struct *vma, >> + unsigned long start, unsigned long end, >> + bool write, int *nonblocking); >> extern void munlock_vma_pages_range(struct vm_area_struct *vma, >> unsigned long start, unsigned long end); >> static inline void munlock_vma_pages_all(struct vm_area_struct *vma) > > The parameter name does not match the one in the implementation. > > Otherwise the implementation looks fine AFAICT. Hehe, you can tell how I copy-pasted from populate_vma_page_range(), because there, the variable names are messed up, too :) Will fix (most probably populate_vma_page_range() as well in a cleanup patch), thanks!
On 08.03.21 17:45, David Hildenbrand wrote: > I. Background: Sparse Memory Mappings > > When we manage sparse memory mappings dynamically in user space - also > sometimes involving MAP_NORESERVE - we want to dynamically populate/ > discard memory inside such a sparse memory region. Example users are > hypervisors (especially implementing memory ballooning or similar > technologies like virtio-mem) and memory allocators. In addition, we want > to fail in a nice way (instead of generating SIGBUS) if populating does not > succeed because we are out of backend memory (which can happen easily with > file-based mappings, especially tmpfs and hugetlbfs). > > While MADV_DONTNEED, MADV_REMOVE and FALLOC_FL_PUNCH_HOLE allow for > reliably discarding memory, there is no generic approach to populate > page tables and preallocate memory. > > Although mmap() supports MAP_POPULATE, it is not applicable to the concept > of sparse memory mappings, where we want to do populate/discard > dynamically and avoid expensive/problematic remappings. In addition, > we never actually report errors during the final populate phase - it is > best-effort only. > > fallocate() can be used to preallocate file-based memory and fail in a safe > way. However, it cannot really be used for any private mappings on > anonymous files via memfd due to COW semantics. In addition, fallocate() > does not actually populate page tables, so we still always get > pagefaults on first access - which is sometimes undesired (i.e., real-time > workloads) and requires real prefaulting of page tables, not just a > preallocation of backend storage. There might be interesting use cases > for sparse memory regions along with mlockall(MCL_ONFAULT) which > fallocate() cannot satisfy as it does not prefault page tables. > > II. On preallcoation/prefaulting from user space > > Because we don't have a proper interface, what applications > (like QEMU and databases) end up doing is touching (i.e., reading+writing > one byte to not overwrite existing data) all individual pages. > > However, that approach > 1) Can result in wear on storage backing, because we end up writing > and thereby dirtying each page --- i.e., disks or pmem. > 2) Can result in mmap_sem contention when prefaulting via multiple > threads. > 3) Requires expensive signal handling, especially to catch SIGBUS in case > of hugetlbfs/shmem/file-backed memory. For example, this is > problematic in hypervisors like QEMU where SIGBUS handlers might already > be used by other subsystems concurrently to e.g, handle hardware errors. > "Simply" doing preallocation concurrently from other thread is not that > easy. > > III. On MADV_WILLNEED > > Extending MADV_WILLNEED is not an option because > 1. It would change the semantics: "Expect access in the near future." and > "might be a good idea to read some pages" vs. "Definitely populate/ > preallocate all memory and definitely fail on errors.". > 2. Existing users (like virtio-balloon in QEMU when deflating the balloon) > don't want populate/prealloc semantics. They treat this rather as a hint > to give a little performance boost without too much overhead - and don't > expect that a lot of memory might get consumed or a lot of time > might be spent. > > IV. MADV_POPULATE_READ and MADV_POPULATE_WRITE > > Let's introduce MADV_POPULATE_READ and MADV_POPULATE_WRITE with the > following semantics: > 1. MADV_POPULATE_READ can be used to preallocate backend memory and > prefault page tables just like manually reading each individual page. > This will not break any COW mappings -- e.g., it will populate the > shared zeropage when applicable. > 2. If MADV_POPULATE_READ succeeds, all page tables have been populated > (prefaulted) readable once. > 3. MADV_POPULATE_WRITE can be used to preallocate backend memory and > prefault page tables just like manually writing (or > reading+writing) each individual page. This will break any COW > mappings -- e.g., the shared zeropage is never populated. > 4. If MADV_POPULATE_WRITE succeeds, all page tables have been populated > (prefaulted) writable once. > 5. MADV_POPULATE_READ and MADV_POPULATE_WRITE cannot be applied to special > mappings marked with VM_PFNMAP and VM_IO. Also, proper access > permissions (e.g., PROT_READ, PROT_WRITE) are required. If any such > mapping is encountered, madvise() fails with -EINVAL. > 6. If MADV_POPULATE_READ or MADV_POPULATE_WRITE fails, some page tables > might have been populated. In that case, madvise() fails with > -ENOMEM. > 7. MADV_POPULATE_READ and MADV_POPULATE_WRITE will ignore any poisoned > pages in the range. > 8. Similar to MAP_POPULATE, MADV_POPULATE_READ and MADV_POPULATE_WRITE > cannot protect from the OOM (Out Of Memory) handler killing the > process. > > While the use case for MADV_POPULATE_WRITE is fairly obvious (i.e., > preallocate memory and prefault page tables for VMs), there are valid use > cases for MADV_POPULATE_READ: > 1. Efficiently populate page tables with zero pages (i.e., shared > zeropage). This is necessary when using userfaultfd() WP (Write-Protect > to properly catch all modifications within a mapping: for > write-protection to be effective for a virtual address, there has to be > a page already mapped -- even if it's the shared zeropage. > 2. Pre-read a whole mapping from backend storage without marking it > dirty, such that eviction won't have to write it back. If no backend > memory has been allocated yet, allocate the backend memory. Helpful > when preallocating/prefaulting a file stored on disk without having > to writeback each and every page on eviction. > > Although sparse memory mappings are the primary use case, this will > also be useful for ordinary preallocations where MAP_POPULATE is not > desired especially in QEMU, where users can trigger preallocation of > guest RAM after the mapping was created. > > Looking at the history, MADV_POPULATE was already proposed in 2013 [1], > however, the main motivation back than was performance improvements > (which should also still be the case, but it is a secondary concern). > > V. Single-threaded performance comparison > > There is a performance benefit when using POPULATE_READ / POPULATE_WRITE > already when only using a single thread to do prefaulting/preallocation. As > we have less pagefaults for huge pages, the performance benefit is > negligible with small mappings. > > Using fallocate() to preallocate shared files is the fastest approach, > however as discussed, we get pagefaults at runtime on actual access > which might or might not be relevant depending on the actual use case. > > Average across 10 iterations each: > ================================================== > 2 MiB MAP_PRIVATE: > ************************************************** > Anon 4 KiB : Read : 0.117 ms > Anon 4 KiB : Write : 0.240 ms > Anon 4 KiB : Read+Write : 0.386 ms > Anon 4 KiB : POPULATE_READ : 0.063 ms > Anon 4 KiB : POPULATE_WRITE : 0.163 ms > Memfd 4 KiB : Read : 0.077 ms > Memfd 4 KiB : Write : 0.375 ms > Memfd 4 KiB : Read+Write : 0.464 ms > Memfd 4 KiB : POPULATE_READ : 0.080 ms > Memfd 4 KiB : POPULATE_WRITE : 0.301 ms > Memfd 2 MiB : Read : 0.042 ms > Memfd 2 MiB : Write : 0.032 ms > Memfd 2 MiB : Read+Write : 0.032 ms > Memfd 2 MiB : POPULATE_READ : 0.031 ms > Memfd 2 MiB : POPULATE_WRITE : 0.032 ms > tmpfs : Read : 0.086 ms > tmpfs : Write : 0.351 ms > tmpfs : Read+Write : 0.427 ms > tmpfs : POPULATE_READ : 0.041 ms > tmpfs : POPULATE_WRITE : 0.298 ms > file : Read : 0.077 ms > file : Write : 0.368 ms > file : Read+Write : 0.466 ms > file : POPULATE_READ : 0.079 ms > file : POPULATE_WRITE : 0.303 ms > ************************************************** > 2 MiB MAP_SHARED: > ************************************************** > Memfd 4 KiB : Read : 0.418 ms > Memfd 4 KiB : Write : 0.367 ms > Memfd 4 KiB : Read+Write : 0.428 ms > Memfd 4 KiB : POPULATE_READ : 0.347 ms > Memfd 4 KiB : POPULATE_WRITE : 0.286 ms > Memfd 4 KiB : FALLOCATE : 0.140 ms > Memfd 2 MiB : Read : 0.031 ms > Memfd 2 MiB : Write : 0.030 ms > Memfd 2 MiB : Read+Write : 0.030 ms > Memfd 2 MiB : POPULATE_READ : 0.030 ms > Memfd 2 MiB : POPULATE_WRITE : 0.030 ms > Memfd 2 MiB : FALLOCATE : 0.030 ms > tmpfs : Read : 0.434 ms > tmpfs : Write : 0.367 ms > tmpfs : Read+Write : 0.435 ms > tmpfs : POPULATE_READ : 0.349 ms > tmpfs : POPULATE_WRITE : 0.291 ms > tmpfs : FALLOCATE : 0.144 ms > file : Read : 0.423 ms > file : Write : 0.367 ms > file : Read+Write : 0.432 ms > file : POPULATE_READ : 0.351 ms > file : POPULATE_WRITE : 0.290 ms > file : FALLOCATE : 0.144 ms > hugetlbfs : Read : 0.032 ms > hugetlbfs : Write : 0.030 ms > hugetlbfs : Read+Write : 0.031 ms > hugetlbfs : POPULATE_READ : 0.030 ms > hugetlbfs : POPULATE_WRITE : 0.030 ms > hugetlbfs : FALLOCATE : 0.030 ms > ************************************************** > 4096 MiB MAP_PRIVATE: > ************************************************** > Anon 4 KiB : Read : 237.099 ms > Anon 4 KiB : Write : 708.062 ms > Anon 4 KiB : Read+Write : 1057.147 ms > Anon 4 KiB : POPULATE_READ : 124.942 ms > Anon 4 KiB : POPULATE_WRITE : 575.082 ms > Memfd 4 KiB : Read : 237.593 ms > Memfd 4 KiB : Write : 984.245 ms > Memfd 4 KiB : Read+Write : 1149.859 ms > Memfd 4 KiB : POPULATE_READ : 166.066 ms > Memfd 4 KiB : POPULATE_WRITE : 856.914 ms > Memfd 2 MiB : Read : 352.202 ms > Memfd 2 MiB : Write : 352.029 ms > Memfd 2 MiB : Read+Write : 352.198 ms > Memfd 2 MiB : POPULATE_READ : 351.033 ms > Memfd 2 MiB : POPULATE_WRITE : 351.181 ms > tmpfs : Read : 230.796 ms > tmpfs : Write : 936.138 ms > tmpfs : Read+Write : 1065.565 ms > tmpfs : POPULATE_READ : 80.823 ms > tmpfs : POPULATE_WRITE : 803.829 ms > file : Read : 231.055 ms > file : Write : 980.575 ms > file : Read+Write : 1208.742 ms > file : POPULATE_READ : 167.808 ms > file : POPULATE_WRITE : 859.270 ms > ************************************************** > 4096 MiB MAP_SHARED: > ************************************************** > Memfd 4 KiB : Read : 1095.979 ms > Memfd 4 KiB : Write : 958.777 ms > Memfd 4 KiB : Read+Write : 1120.127 ms > Memfd 4 KiB : POPULATE_READ : 937.689 ms > Memfd 4 KiB : POPULATE_WRITE : 811.594 ms > Memfd 4 KiB : FALLOCATE : 309.438 ms > Memfd 2 MiB : Read : 353.045 ms > Memfd 2 MiB : Write : 353.356 ms > Memfd 2 MiB : Read+Write : 352.829 ms > Memfd 2 MiB : POPULATE_READ : 351.954 ms > Memfd 2 MiB : POPULATE_WRITE : 351.840 ms > Memfd 2 MiB : FALLOCATE : 351.274 ms > tmpfs : Read : 1096.222 ms > tmpfs : Write : 980.651 ms > tmpfs : Read+Write : 1114.757 ms > tmpfs : POPULATE_READ : 939.181 ms > tmpfs : POPULATE_WRITE : 817.255 ms > tmpfs : FALLOCATE : 312.521 ms > file : Read : 1112.135 ms > file : Write : 967.688 ms > file : Read+Write : 1111.620 ms > file : POPULATE_READ : 951.175 ms > file : POPULATE_WRITE : 818.380 ms > file : FALLOCATE : 313.008 ms > hugetlbfs : Read : 353.710 ms > hugetlbfs : Write : 353.309 ms > hugetlbfs : Read+Write : 353.280 ms > hugetlbfs : POPULATE_READ : 353.138 ms > hugetlbfs : POPULATE_WRITE : 352.620 ms > hugetlbfs : FALLOCATE : 352.204 ms > ************************************************** > > [1] https://lkml.org/lkml/2013/6/27/698 > > Cc: Andrew Morton <akpm@linux-foundation.org> > Cc: Arnd Bergmann <arnd@arndb.de> > Cc: Michal Hocko <mhocko@suse.com> > Cc: Oscar Salvador <osalvador@suse.de> > Cc: Matthew Wilcox (Oracle) <willy@infradead.org> > Cc: Andrea Arcangeli <aarcange@redhat.com> > Cc: Minchan Kim <minchan@kernel.org> > Cc: Jann Horn <jannh@google.com> > Cc: Jason Gunthorpe <jgg@ziepe.ca> > Cc: Dave Hansen <dave.hansen@intel.com> > Cc: Hugh Dickins <hughd@google.com> > Cc: Rik van Riel <riel@surriel.com> > Cc: Michael S. Tsirkin <mst@redhat.com> > Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> > Cc: Vlastimil Babka <vbabka@suse.cz> > Cc: Richard Henderson <rth@twiddle.net> > Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> > Cc: Matt Turner <mattst88@gmail.com> > Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> > Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> > Cc: Helge Deller <deller@gmx.de> > Cc: Chris Zankel <chris@zankel.net> > Cc: Max Filippov <jcmvbkbc@gmail.com> > Cc: Mike Kravetz <mike.kravetz@oracle.com> > Cc: Peter Xu <peterx@redhat.com> > Cc: Rolf Eike Beer <eike-kernel@sf-tec.de> > Cc: linux-alpha@vger.kernel.org > Cc: linux-mips@vger.kernel.org > Cc: linux-parisc@vger.kernel.org > Cc: linux-xtensa@linux-xtensa.org > Cc: linux-arch@vger.kernel.org > Cc: Linux API <linux-api@vger.kernel.org> > Signed-off-by: David Hildenbrand <david@redhat.com> > --- > > RFC -> RFCv2: > - Fix re-locking (-> set "locked = 1;") > - Don't mimic MAP_POPULATE semantics: > --> Explicit READ/WRITE request instead of selecting it automatically, > which makes it more generic and better suited for some use cases (e.g., we > usually want to prefault shmem writable) > --> Require proper access permissions > - Introduce and use faultin_vma_page_range() > --> Properly handle HWPOISON pages (FOLL_HWPOISON) > --> Require proper access permissions (!FOLL_FORCE) > - Let faultin_vma_page_range() check for compatible mappings/permissions > - Extend patch description and add some performance numbers > > --- > arch/alpha/include/uapi/asm/mman.h | 3 ++ > arch/mips/include/uapi/asm/mman.h | 3 ++ > arch/parisc/include/uapi/asm/mman.h | 3 ++ > arch/xtensa/include/uapi/asm/mman.h | 3 ++ > include/uapi/asm-generic/mman-common.h | 3 ++ > mm/gup.c | 54 ++++++++++++++++++++ > mm/internal.h | 3 ++ > mm/madvise.c | 70 ++++++++++++++++++++++++++ > 8 files changed, 142 insertions(+) > > diff --git a/arch/alpha/include/uapi/asm/mman.h b/arch/alpha/include/uapi/asm/mman.h > index a18ec7f63888..56b4ee5a6c9e 100644 > --- a/arch/alpha/include/uapi/asm/mman.h > +++ b/arch/alpha/include/uapi/asm/mman.h > @@ -71,6 +71,9 @@ > #define MADV_COLD 20 /* deactivate these pages */ > #define MADV_PAGEOUT 21 /* reclaim these pages */ > > +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ > +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ > + > /* compatibility flags */ > #define MAP_FILE 0 > > diff --git a/arch/mips/include/uapi/asm/mman.h b/arch/mips/include/uapi/asm/mman.h > index 57dc2ac4f8bd..40b210c65a5a 100644 > --- a/arch/mips/include/uapi/asm/mman.h > +++ b/arch/mips/include/uapi/asm/mman.h > @@ -98,6 +98,9 @@ > #define MADV_COLD 20 /* deactivate these pages */ > #define MADV_PAGEOUT 21 /* reclaim these pages */ > > +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ > +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ > + > /* compatibility flags */ > #define MAP_FILE 0 > > diff --git a/arch/parisc/include/uapi/asm/mman.h b/arch/parisc/include/uapi/asm/mman.h > index ab78cba446ed..9e3c010c0f61 100644 > --- a/arch/parisc/include/uapi/asm/mman.h > +++ b/arch/parisc/include/uapi/asm/mman.h > @@ -52,6 +52,9 @@ > #define MADV_COLD 20 /* deactivate these pages */ > #define MADV_PAGEOUT 21 /* reclaim these pages */ > > +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ > +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ > + > #define MADV_MERGEABLE 65 /* KSM may merge identical pages */ > #define MADV_UNMERGEABLE 66 /* KSM may not merge identical pages */ > > diff --git a/arch/xtensa/include/uapi/asm/mman.h b/arch/xtensa/include/uapi/asm/mman.h > index e5e643752947..b3a22095371b 100644 > --- a/arch/xtensa/include/uapi/asm/mman.h > +++ b/arch/xtensa/include/uapi/asm/mman.h > @@ -106,6 +106,9 @@ > #define MADV_COLD 20 /* deactivate these pages */ > #define MADV_PAGEOUT 21 /* reclaim these pages */ > > +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ > +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ > + > /* compatibility flags */ > #define MAP_FILE 0 > > diff --git a/include/uapi/asm-generic/mman-common.h b/include/uapi/asm-generic/mman-common.h > index f94f65d429be..1567a3294c3d 100644 > --- a/include/uapi/asm-generic/mman-common.h > +++ b/include/uapi/asm-generic/mman-common.h > @@ -72,6 +72,9 @@ > #define MADV_COLD 20 /* deactivate these pages */ > #define MADV_PAGEOUT 21 /* reclaim these pages */ > > +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ > +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ > + > /* compatibility flags */ > #define MAP_FILE 0 > > diff --git a/mm/gup.c b/mm/gup.c > index e40579624f10..80fad8578066 100644 > --- a/mm/gup.c > +++ b/mm/gup.c > @@ -1403,6 +1403,60 @@ long populate_vma_page_range(struct vm_area_struct *vma, > NULL, NULL, locked); > } > > +/* > + * faultin_vma_page_range() - populate (prefault) page tables inside the > + * given VMA range readable/writable > + * > + * This takes care of mlocking the pages, too, if VM_LOCKED is set. > + * > + * @vma: target vma > + * @start: start address > + * @end: end address > + * @write: whether to prefault readable or writable > + * @locked: whether the mmap_lock is still held > + * > + * Returns either number of processed pages in the vma, or a negative error > + * code on error (see __get_user_pages()). > + * > + * vma->vm_mm->mmap_lock must be held. The range must be page-aligned and > + * covered by the VMA. > + * > + * If @locked is NULL, it may be held for read or write and will be unperturbed. > + * > + * If @locked is non-NULL, it must held for read only and may be released. If > + * it's released, *@locked will be set to 0. > + */ > +long faultin_vma_page_range(struct vm_area_struct *vma, unsigned long start, > + unsigned long end, bool write, int *locked) > +{ > + struct mm_struct *mm = vma->vm_mm; > + unsigned long nr_pages = (end - start) / PAGE_SIZE; > + int gup_flags; > + > + VM_BUG_ON(!PAGE_ALIGNED(start)); > + VM_BUG_ON(!PAGE_ALIGNED(end)); > + VM_BUG_ON_VMA(start < vma->vm_start, vma); > + VM_BUG_ON_VMA(end > vma->vm_end, vma); > + mmap_assert_locked(mm); > + > + /* > + * FOLL_HWPOISON: Return -EHWPOISON instead of -EFAULT when we hit > + * a poisoned page. > + * FOLL_POPULATE: Always populate memory with VM_LOCKONFAULT. > + * !FOLL_FORCE: Require proper access permissions. > + */ > + gup_flags = FOLL_TOUCH | FOLL_POPULATE | FOLL_MLOCK | FOLL_HWPOISON; > + if (write) > + gup_flags |= FOLL_WRITE; > + > + /* > + * See check_vma_flags(): Will return -EFAULT on incompatible mappings > + * or with insufficient permissions. > + */ > + return __get_user_pages(mm, start, nr_pages, gup_flags, > + NULL, NULL, locked); > +} > + > /* > * __mm_populate - populate and/or mlock pages within a range of address space. > * > diff --git a/mm/internal.h b/mm/internal.h > index 9902648f2206..a5c4ed23b1db 100644 > --- a/mm/internal.h > +++ b/mm/internal.h > @@ -340,6 +340,9 @@ void __vma_unlink_list(struct mm_struct *mm, struct vm_area_struct *vma); > #ifdef CONFIG_MMU > extern long populate_vma_page_range(struct vm_area_struct *vma, > unsigned long start, unsigned long end, int *nonblocking); > +extern long faultin_vma_page_range(struct vm_area_struct *vma, > + unsigned long start, unsigned long end, > + bool write, int *nonblocking); > extern void munlock_vma_pages_range(struct vm_area_struct *vma, > unsigned long start, unsigned long end); > static inline void munlock_vma_pages_all(struct vm_area_struct *vma) > diff --git a/mm/madvise.c b/mm/madvise.c > index df692d2e35d4..fbb5e10b5550 100644 > --- a/mm/madvise.c > +++ b/mm/madvise.c > @@ -53,6 +53,8 @@ static int madvise_need_mmap_write(int behavior) > case MADV_COLD: > case MADV_PAGEOUT: > case MADV_FREE: > + case MADV_POPULATE_READ: > + case MADV_POPULATE_WRITE: > return 0; > default: > /* be safe, default to 1. list exceptions explicitly */ > @@ -822,6 +824,65 @@ static long madvise_dontneed_free(struct vm_area_struct *vma, > return -EINVAL; > } > > +static long madvise_populate(struct vm_area_struct *vma, > + struct vm_area_struct **prev, > + unsigned long start, unsigned long end, > + int behavior) > +{ > + const bool write = behavior == MADV_POPULATE_WRITE; > + struct mm_struct *mm = vma->vm_mm; > + unsigned long tmp_end; > + int locked = 1; > + long pages; > + > + *prev = vma; > + > + while (start < end) { > + /* > + * We might have temporarily dropped the lock. For example, > + * our VMA might have been split. > + */ > + if (!vma || start >= vma->vm_end) { > + vma = find_vma(mm, start); > + if (!vma) > + return -ENOMEM; Looking again, I think I'll have to do "if (!vma || start < vma->vm_start)" here to properly catch all holes. Will do more testing with different mmap layouts.
On Mon, Mar 08, 2021 at 05:45:20PM +0100, David Hildenbrand wrote: > + case -EHWPOISON: /* Skip over any poisoned pages. */ > + start += PAGE_SIZE; > + continue; Why is it good approach? It's not abvious to me.
On 15.03.21 13:22, Kirill A. Shutemov wrote: > On Mon, Mar 08, 2021 at 05:45:20PM +0100, David Hildenbrand wrote: >> + case -EHWPOISON: /* Skip over any poisoned pages. */ >> + start += PAGE_SIZE; >> + continue; > > Why is it good approach? It's not abvious to me. My main motivation was to simplify return code handling. I don't want to return -EHWPOISON to user space and I don't want to convert it into something misleading like -ENOMEM or -EINVAL. So I decided to handle such stuff internally. What would be you take on that?
On Mon, Mar 15, 2021 at 01:25:40PM +0100, David Hildenbrand wrote: > On 15.03.21 13:22, Kirill A. Shutemov wrote: > > On Mon, Mar 08, 2021 at 05:45:20PM +0100, David Hildenbrand wrote: > > > + case -EHWPOISON: /* Skip over any poisoned pages. */ > > > + start += PAGE_SIZE; > > > + continue; > > > > Why is it good approach? It's not abvious to me. > > My main motivation was to simplify return code handling. I don't want to > return -EHWPOISON to user space Why? Hiding the problem under the rug doesn't help anybody. SIGBUS later is not better than an error upfront.
On 15.03.21 14:03, Kirill A. Shutemov wrote: > On Mon, Mar 15, 2021 at 01:25:40PM +0100, David Hildenbrand wrote: >> On 15.03.21 13:22, Kirill A. Shutemov wrote: >>> On Mon, Mar 08, 2021 at 05:45:20PM +0100, David Hildenbrand wrote: >>>> + case -EHWPOISON: /* Skip over any poisoned pages. */ >>>> + start += PAGE_SIZE; >>>> + continue; >>> >>> Why is it good approach? It's not abvious to me. >> >> My main motivation was to simplify return code handling. I don't want to >> return -EHWPOISON to user space > > Why? Hiding the problem under the rug doesn't help anybody. SIGBUS later > is not better than an error upfront. Well, if you think about "prefaulting page tables", the first intuition is certainly not to check for poisoned pages, right? After all, you are not actually accessing memory, you are allocating memory if required and fill page tables. OTOH, mlock() will also choke on poisoned pages. With the current semantics, you can start and run a VM just fine. Preallocation/prefaulting succeeded after all. On access you will get a SIGBUS, from which e.g., QEMU can recover by injecting an MCE into the guest - just like if you would hit a poisoned page later. The problem we are talking about is most probably very rare, especially when using MADV_POPULATE_ for actual preallocation. I don't have a strong opinion; not bailing out on poisoned pages felt like the right thing to do.
On 15.03.21 14:26, David Hildenbrand wrote: > On 15.03.21 14:03, Kirill A. Shutemov wrote: >> On Mon, Mar 15, 2021 at 01:25:40PM +0100, David Hildenbrand wrote: >>> On 15.03.21 13:22, Kirill A. Shutemov wrote: >>>> On Mon, Mar 08, 2021 at 05:45:20PM +0100, David Hildenbrand wrote: >>>>> + case -EHWPOISON: /* Skip over any poisoned pages. */ >>>>> + start += PAGE_SIZE; >>>>> + continue; >>>> >>>> Why is it good approach? It's not abvious to me. >>> >>> My main motivation was to simplify return code handling. I don't want to >>> return -EHWPOISON to user space >> >> Why? Hiding the problem under the rug doesn't help anybody. SIGBUS later >> is not better than an error upfront. > > Well, if you think about "prefaulting page tables", the first intuition > is certainly not to check for poisoned pages, right? After all, you are > not actually accessing memory, you are allocating memory if required and > fill page tables. OTOH, mlock() will also choke on poisoned pages. > > With the current semantics, you can start and run a VM just fine. > Preallocation/prefaulting succeeded after all. On access you will get a > SIGBUS, from which e.g., QEMU can recover by injecting an MCE into the > guest - just like if you would hit a poisoned page later. > > The problem we are talking about is most probably very rare, especially > when using MADV_POPULATE_ for actual preallocation. > > I don't have a strong opinion; not bailing out on poisoned pages felt > like the right thing to do. I'll switch to propagating -EHWPOISON, it matches how e.g., mlock() behaves -- not ignoring poisoned pages. Thanks!
diff --git a/arch/alpha/include/uapi/asm/mman.h b/arch/alpha/include/uapi/asm/mman.h index a18ec7f63888..56b4ee5a6c9e 100644 --- a/arch/alpha/include/uapi/asm/mman.h +++ b/arch/alpha/include/uapi/asm/mman.h @@ -71,6 +71,9 @@ #define MADV_COLD 20 /* deactivate these pages */ #define MADV_PAGEOUT 21 /* reclaim these pages */ +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ + /* compatibility flags */ #define MAP_FILE 0 diff --git a/arch/mips/include/uapi/asm/mman.h b/arch/mips/include/uapi/asm/mman.h index 57dc2ac4f8bd..40b210c65a5a 100644 --- a/arch/mips/include/uapi/asm/mman.h +++ b/arch/mips/include/uapi/asm/mman.h @@ -98,6 +98,9 @@ #define MADV_COLD 20 /* deactivate these pages */ #define MADV_PAGEOUT 21 /* reclaim these pages */ +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ + /* compatibility flags */ #define MAP_FILE 0 diff --git a/arch/parisc/include/uapi/asm/mman.h b/arch/parisc/include/uapi/asm/mman.h index ab78cba446ed..9e3c010c0f61 100644 --- a/arch/parisc/include/uapi/asm/mman.h +++ b/arch/parisc/include/uapi/asm/mman.h @@ -52,6 +52,9 @@ #define MADV_COLD 20 /* deactivate these pages */ #define MADV_PAGEOUT 21 /* reclaim these pages */ +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ + #define MADV_MERGEABLE 65 /* KSM may merge identical pages */ #define MADV_UNMERGEABLE 66 /* KSM may not merge identical pages */ diff --git a/arch/xtensa/include/uapi/asm/mman.h b/arch/xtensa/include/uapi/asm/mman.h index e5e643752947..b3a22095371b 100644 --- a/arch/xtensa/include/uapi/asm/mman.h +++ b/arch/xtensa/include/uapi/asm/mman.h @@ -106,6 +106,9 @@ #define MADV_COLD 20 /* deactivate these pages */ #define MADV_PAGEOUT 21 /* reclaim these pages */ +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ + /* compatibility flags */ #define MAP_FILE 0 diff --git a/include/uapi/asm-generic/mman-common.h b/include/uapi/asm-generic/mman-common.h index f94f65d429be..1567a3294c3d 100644 --- a/include/uapi/asm-generic/mman-common.h +++ b/include/uapi/asm-generic/mman-common.h @@ -72,6 +72,9 @@ #define MADV_COLD 20 /* deactivate these pages */ #define MADV_PAGEOUT 21 /* reclaim these pages */ +#define MADV_POPULATE_READ 22 /* populate (prefault) page tables readable */ +#define MADV_POPULATE_WRITE 23 /* populate (prefault) page tables writable */ + /* compatibility flags */ #define MAP_FILE 0 diff --git a/mm/gup.c b/mm/gup.c index e40579624f10..80fad8578066 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -1403,6 +1403,60 @@ long populate_vma_page_range(struct vm_area_struct *vma, NULL, NULL, locked); } +/* + * faultin_vma_page_range() - populate (prefault) page tables inside the + * given VMA range readable/writable + * + * This takes care of mlocking the pages, too, if VM_LOCKED is set. + * + * @vma: target vma + * @start: start address + * @end: end address + * @write: whether to prefault readable or writable + * @locked: whether the mmap_lock is still held + * + * Returns either number of processed pages in the vma, or a negative error + * code on error (see __get_user_pages()). + * + * vma->vm_mm->mmap_lock must be held. The range must be page-aligned and + * covered by the VMA. + * + * If @locked is NULL, it may be held for read or write and will be unperturbed. + * + * If @locked is non-NULL, it must held for read only and may be released. If + * it's released, *@locked will be set to 0. + */ +long faultin_vma_page_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end, bool write, int *locked) +{ + struct mm_struct *mm = vma->vm_mm; + unsigned long nr_pages = (end - start) / PAGE_SIZE; + int gup_flags; + + VM_BUG_ON(!PAGE_ALIGNED(start)); + VM_BUG_ON(!PAGE_ALIGNED(end)); + VM_BUG_ON_VMA(start < vma->vm_start, vma); + VM_BUG_ON_VMA(end > vma->vm_end, vma); + mmap_assert_locked(mm); + + /* + * FOLL_HWPOISON: Return -EHWPOISON instead of -EFAULT when we hit + * a poisoned page. + * FOLL_POPULATE: Always populate memory with VM_LOCKONFAULT. + * !FOLL_FORCE: Require proper access permissions. + */ + gup_flags = FOLL_TOUCH | FOLL_POPULATE | FOLL_MLOCK | FOLL_HWPOISON; + if (write) + gup_flags |= FOLL_WRITE; + + /* + * See check_vma_flags(): Will return -EFAULT on incompatible mappings + * or with insufficient permissions. + */ + return __get_user_pages(mm, start, nr_pages, gup_flags, + NULL, NULL, locked); +} + /* * __mm_populate - populate and/or mlock pages within a range of address space. * diff --git a/mm/internal.h b/mm/internal.h index 9902648f2206..a5c4ed23b1db 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -340,6 +340,9 @@ void __vma_unlink_list(struct mm_struct *mm, struct vm_area_struct *vma); #ifdef CONFIG_MMU extern long populate_vma_page_range(struct vm_area_struct *vma, unsigned long start, unsigned long end, int *nonblocking); +extern long faultin_vma_page_range(struct vm_area_struct *vma, + unsigned long start, unsigned long end, + bool write, int *nonblocking); extern void munlock_vma_pages_range(struct vm_area_struct *vma, unsigned long start, unsigned long end); static inline void munlock_vma_pages_all(struct vm_area_struct *vma) diff --git a/mm/madvise.c b/mm/madvise.c index df692d2e35d4..fbb5e10b5550 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -53,6 +53,8 @@ static int madvise_need_mmap_write(int behavior) case MADV_COLD: case MADV_PAGEOUT: case MADV_FREE: + case MADV_POPULATE_READ: + case MADV_POPULATE_WRITE: return 0; default: /* be safe, default to 1. list exceptions explicitly */ @@ -822,6 +824,65 @@ static long madvise_dontneed_free(struct vm_area_struct *vma, return -EINVAL; } +static long madvise_populate(struct vm_area_struct *vma, + struct vm_area_struct **prev, + unsigned long start, unsigned long end, + int behavior) +{ + const bool write = behavior == MADV_POPULATE_WRITE; + struct mm_struct *mm = vma->vm_mm; + unsigned long tmp_end; + int locked = 1; + long pages; + + *prev = vma; + + while (start < end) { + /* + * We might have temporarily dropped the lock. For example, + * our VMA might have been split. + */ + if (!vma || start >= vma->vm_end) { + vma = find_vma(mm, start); + if (!vma) + return -ENOMEM; + } + + tmp_end = min_t(unsigned long, end, vma->vm_end); + /* Populate (prefault) page tables readable/writable. */ + pages = faultin_vma_page_range(vma, start, tmp_end, write, + &locked); + if (!locked) { + mmap_read_lock(mm); + locked = 1; + *prev = NULL; + vma = NULL; + } + if (pages < 0) { + switch (pages) { + case -EINTR: + return -EINTR; + case -EFAULT: /* Incompatible mappings / permissions. */ + return -EINVAL; + case -EHWPOISON: /* Skip over any poisoned pages. */ + start += PAGE_SIZE; + continue; + case -EBUSY: + case -EAGAIN: + continue; + default: + pr_warn_once("%s: unhandled return value: %ld\n", + __func__, pages); + fallthrough; + case -ENOMEM: + return -ENOMEM; + } + } + start += pages * PAGE_SIZE; + } + return 0; +} + /* * Application wants to free up the pages and associated backing store. * This is effectively punching a hole into the middle of a file. @@ -935,6 +996,9 @@ madvise_vma(struct vm_area_struct *vma, struct vm_area_struct **prev, case MADV_FREE: case MADV_DONTNEED: return madvise_dontneed_free(vma, prev, start, end, behavior); + case MADV_POPULATE_READ: + case MADV_POPULATE_WRITE: + return madvise_populate(vma, prev, start, end, behavior); default: return madvise_behavior(vma, prev, start, end, behavior); } @@ -955,6 +1019,8 @@ madvise_behavior_valid(int behavior) case MADV_FREE: case MADV_COLD: case MADV_PAGEOUT: + case MADV_POPULATE_READ: + case MADV_POPULATE_WRITE: #ifdef CONFIG_KSM case MADV_MERGEABLE: case MADV_UNMERGEABLE: @@ -1042,6 +1108,10 @@ process_madvise_behavior_valid(int behavior) * easily if memory pressure hanppens. * MADV_PAGEOUT - the application is not expected to use this memory soon, * page out the pages in this range immediately. + * MADV_POPULATE_READ - populate (prefault) page tables readable by + * triggering read faults if required + * MADV_POPULATE_WRITE - populate (prefault) page tables writable by + * triggering write faults if required * * return values: * zero - success
I. Background: Sparse Memory Mappings When we manage sparse memory mappings dynamically in user space - also sometimes involving MAP_NORESERVE - we want to dynamically populate/ discard memory inside such a sparse memory region. Example users are hypervisors (especially implementing memory ballooning or similar technologies like virtio-mem) and memory allocators. In addition, we want to fail in a nice way (instead of generating SIGBUS) if populating does not succeed because we are out of backend memory (which can happen easily with file-based mappings, especially tmpfs and hugetlbfs). While MADV_DONTNEED, MADV_REMOVE and FALLOC_FL_PUNCH_HOLE allow for reliably discarding memory, there is no generic approach to populate page tables and preallocate memory. Although mmap() supports MAP_POPULATE, it is not applicable to the concept of sparse memory mappings, where we want to do populate/discard dynamically and avoid expensive/problematic remappings. In addition, we never actually report errors during the final populate phase - it is best-effort only. fallocate() can be used to preallocate file-based memory and fail in a safe way. However, it cannot really be used for any private mappings on anonymous files via memfd due to COW semantics. In addition, fallocate() does not actually populate page tables, so we still always get pagefaults on first access - which is sometimes undesired (i.e., real-time workloads) and requires real prefaulting of page tables, not just a preallocation of backend storage. There might be interesting use cases for sparse memory regions along with mlockall(MCL_ONFAULT) which fallocate() cannot satisfy as it does not prefault page tables. II. On preallcoation/prefaulting from user space Because we don't have a proper interface, what applications (like QEMU and databases) end up doing is touching (i.e., reading+writing one byte to not overwrite existing data) all individual pages. However, that approach 1) Can result in wear on storage backing, because we end up writing and thereby dirtying each page --- i.e., disks or pmem. 2) Can result in mmap_sem contention when prefaulting via multiple threads. 3) Requires expensive signal handling, especially to catch SIGBUS in case of hugetlbfs/shmem/file-backed memory. For example, this is problematic in hypervisors like QEMU where SIGBUS handlers might already be used by other subsystems concurrently to e.g, handle hardware errors. "Simply" doing preallocation concurrently from other thread is not that easy. III. On MADV_WILLNEED Extending MADV_WILLNEED is not an option because 1. It would change the semantics: "Expect access in the near future." and "might be a good idea to read some pages" vs. "Definitely populate/ preallocate all memory and definitely fail on errors.". 2. Existing users (like virtio-balloon in QEMU when deflating the balloon) don't want populate/prealloc semantics. They treat this rather as a hint to give a little performance boost without too much overhead - and don't expect that a lot of memory might get consumed or a lot of time might be spent. IV. MADV_POPULATE_READ and MADV_POPULATE_WRITE Let's introduce MADV_POPULATE_READ and MADV_POPULATE_WRITE with the following semantics: 1. MADV_POPULATE_READ can be used to preallocate backend memory and prefault page tables just like manually reading each individual page. This will not break any COW mappings -- e.g., it will populate the shared zeropage when applicable. 2. If MADV_POPULATE_READ succeeds, all page tables have been populated (prefaulted) readable once. 3. MADV_POPULATE_WRITE can be used to preallocate backend memory and prefault page tables just like manually writing (or reading+writing) each individual page. This will break any COW mappings -- e.g., the shared zeropage is never populated. 4. If MADV_POPULATE_WRITE succeeds, all page tables have been populated (prefaulted) writable once. 5. MADV_POPULATE_READ and MADV_POPULATE_WRITE cannot be applied to special mappings marked with VM_PFNMAP and VM_IO. Also, proper access permissions (e.g., PROT_READ, PROT_WRITE) are required. If any such mapping is encountered, madvise() fails with -EINVAL. 6. If MADV_POPULATE_READ or MADV_POPULATE_WRITE fails, some page tables might have been populated. In that case, madvise() fails with -ENOMEM. 7. MADV_POPULATE_READ and MADV_POPULATE_WRITE will ignore any poisoned pages in the range. 8. Similar to MAP_POPULATE, MADV_POPULATE_READ and MADV_POPULATE_WRITE cannot protect from the OOM (Out Of Memory) handler killing the process. While the use case for MADV_POPULATE_WRITE is fairly obvious (i.e., preallocate memory and prefault page tables for VMs), there are valid use cases for MADV_POPULATE_READ: 1. Efficiently populate page tables with zero pages (i.e., shared zeropage). This is necessary when using userfaultfd() WP (Write-Protect to properly catch all modifications within a mapping: for write-protection to be effective for a virtual address, there has to be a page already mapped -- even if it's the shared zeropage. 2. Pre-read a whole mapping from backend storage without marking it dirty, such that eviction won't have to write it back. If no backend memory has been allocated yet, allocate the backend memory. Helpful when preallocating/prefaulting a file stored on disk without having to writeback each and every page on eviction. Although sparse memory mappings are the primary use case, this will also be useful for ordinary preallocations where MAP_POPULATE is not desired especially in QEMU, where users can trigger preallocation of guest RAM after the mapping was created. Looking at the history, MADV_POPULATE was already proposed in 2013 [1], however, the main motivation back than was performance improvements (which should also still be the case, but it is a secondary concern). V. Single-threaded performance comparison There is a performance benefit when using POPULATE_READ / POPULATE_WRITE already when only using a single thread to do prefaulting/preallocation. As we have less pagefaults for huge pages, the performance benefit is negligible with small mappings. Using fallocate() to preallocate shared files is the fastest approach, however as discussed, we get pagefaults at runtime on actual access which might or might not be relevant depending on the actual use case. Average across 10 iterations each: ================================================== 2 MiB MAP_PRIVATE: ************************************************** Anon 4 KiB : Read : 0.117 ms Anon 4 KiB : Write : 0.240 ms Anon 4 KiB : Read+Write : 0.386 ms Anon 4 KiB : POPULATE_READ : 0.063 ms Anon 4 KiB : POPULATE_WRITE : 0.163 ms Memfd 4 KiB : Read : 0.077 ms Memfd 4 KiB : Write : 0.375 ms Memfd 4 KiB : Read+Write : 0.464 ms Memfd 4 KiB : POPULATE_READ : 0.080 ms Memfd 4 KiB : POPULATE_WRITE : 0.301 ms Memfd 2 MiB : Read : 0.042 ms Memfd 2 MiB : Write : 0.032 ms Memfd 2 MiB : Read+Write : 0.032 ms Memfd 2 MiB : POPULATE_READ : 0.031 ms Memfd 2 MiB : POPULATE_WRITE : 0.032 ms tmpfs : Read : 0.086 ms tmpfs : Write : 0.351 ms tmpfs : Read+Write : 0.427 ms tmpfs : POPULATE_READ : 0.041 ms tmpfs : POPULATE_WRITE : 0.298 ms file : Read : 0.077 ms file : Write : 0.368 ms file : Read+Write : 0.466 ms file : POPULATE_READ : 0.079 ms file : POPULATE_WRITE : 0.303 ms ************************************************** 2 MiB MAP_SHARED: ************************************************** Memfd 4 KiB : Read : 0.418 ms Memfd 4 KiB : Write : 0.367 ms Memfd 4 KiB : Read+Write : 0.428 ms Memfd 4 KiB : POPULATE_READ : 0.347 ms Memfd 4 KiB : POPULATE_WRITE : 0.286 ms Memfd 4 KiB : FALLOCATE : 0.140 ms Memfd 2 MiB : Read : 0.031 ms Memfd 2 MiB : Write : 0.030 ms Memfd 2 MiB : Read+Write : 0.030 ms Memfd 2 MiB : POPULATE_READ : 0.030 ms Memfd 2 MiB : POPULATE_WRITE : 0.030 ms Memfd 2 MiB : FALLOCATE : 0.030 ms tmpfs : Read : 0.434 ms tmpfs : Write : 0.367 ms tmpfs : Read+Write : 0.435 ms tmpfs : POPULATE_READ : 0.349 ms tmpfs : POPULATE_WRITE : 0.291 ms tmpfs : FALLOCATE : 0.144 ms file : Read : 0.423 ms file : Write : 0.367 ms file : Read+Write : 0.432 ms file : POPULATE_READ : 0.351 ms file : POPULATE_WRITE : 0.290 ms file : FALLOCATE : 0.144 ms hugetlbfs : Read : 0.032 ms hugetlbfs : Write : 0.030 ms hugetlbfs : Read+Write : 0.031 ms hugetlbfs : POPULATE_READ : 0.030 ms hugetlbfs : POPULATE_WRITE : 0.030 ms hugetlbfs : FALLOCATE : 0.030 ms ************************************************** 4096 MiB MAP_PRIVATE: ************************************************** Anon 4 KiB : Read : 237.099 ms Anon 4 KiB : Write : 708.062 ms Anon 4 KiB : Read+Write : 1057.147 ms Anon 4 KiB : POPULATE_READ : 124.942 ms Anon 4 KiB : POPULATE_WRITE : 575.082 ms Memfd 4 KiB : Read : 237.593 ms Memfd 4 KiB : Write : 984.245 ms Memfd 4 KiB : Read+Write : 1149.859 ms Memfd 4 KiB : POPULATE_READ : 166.066 ms Memfd 4 KiB : POPULATE_WRITE : 856.914 ms Memfd 2 MiB : Read : 352.202 ms Memfd 2 MiB : Write : 352.029 ms Memfd 2 MiB : Read+Write : 352.198 ms Memfd 2 MiB : POPULATE_READ : 351.033 ms Memfd 2 MiB : POPULATE_WRITE : 351.181 ms tmpfs : Read : 230.796 ms tmpfs : Write : 936.138 ms tmpfs : Read+Write : 1065.565 ms tmpfs : POPULATE_READ : 80.823 ms tmpfs : POPULATE_WRITE : 803.829 ms file : Read : 231.055 ms file : Write : 980.575 ms file : Read+Write : 1208.742 ms file : POPULATE_READ : 167.808 ms file : POPULATE_WRITE : 859.270 ms ************************************************** 4096 MiB MAP_SHARED: ************************************************** Memfd 4 KiB : Read : 1095.979 ms Memfd 4 KiB : Write : 958.777 ms Memfd 4 KiB : Read+Write : 1120.127 ms Memfd 4 KiB : POPULATE_READ : 937.689 ms Memfd 4 KiB : POPULATE_WRITE : 811.594 ms Memfd 4 KiB : FALLOCATE : 309.438 ms Memfd 2 MiB : Read : 353.045 ms Memfd 2 MiB : Write : 353.356 ms Memfd 2 MiB : Read+Write : 352.829 ms Memfd 2 MiB : POPULATE_READ : 351.954 ms Memfd 2 MiB : POPULATE_WRITE : 351.840 ms Memfd 2 MiB : FALLOCATE : 351.274 ms tmpfs : Read : 1096.222 ms tmpfs : Write : 980.651 ms tmpfs : Read+Write : 1114.757 ms tmpfs : POPULATE_READ : 939.181 ms tmpfs : POPULATE_WRITE : 817.255 ms tmpfs : FALLOCATE : 312.521 ms file : Read : 1112.135 ms file : Write : 967.688 ms file : Read+Write : 1111.620 ms file : POPULATE_READ : 951.175 ms file : POPULATE_WRITE : 818.380 ms file : FALLOCATE : 313.008 ms hugetlbfs : Read : 353.710 ms hugetlbfs : Write : 353.309 ms hugetlbfs : Read+Write : 353.280 ms hugetlbfs : POPULATE_READ : 353.138 ms hugetlbfs : POPULATE_WRITE : 352.620 ms hugetlbfs : FALLOCATE : 352.204 ms ************************************************** [1] https://lkml.org/lkml/2013/6/27/698 Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Michal Hocko <mhocko@suse.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jann Horn <jannh@google.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Rik van Riel <riel@surriel.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> Cc: Helge Deller <deller@gmx.de> Cc: Chris Zankel <chris@zankel.net> Cc: Max Filippov <jcmvbkbc@gmail.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Peter Xu <peterx@redhat.com> Cc: Rolf Eike Beer <eike-kernel@sf-tec.de> Cc: linux-alpha@vger.kernel.org Cc: linux-mips@vger.kernel.org Cc: linux-parisc@vger.kernel.org Cc: linux-xtensa@linux-xtensa.org Cc: linux-arch@vger.kernel.org Cc: Linux API <linux-api@vger.kernel.org> Signed-off-by: David Hildenbrand <david@redhat.com> --- RFC -> RFCv2: - Fix re-locking (-> set "locked = 1;") - Don't mimic MAP_POPULATE semantics: --> Explicit READ/WRITE request instead of selecting it automatically, which makes it more generic and better suited for some use cases (e.g., we usually want to prefault shmem writable) --> Require proper access permissions - Introduce and use faultin_vma_page_range() --> Properly handle HWPOISON pages (FOLL_HWPOISON) --> Require proper access permissions (!FOLL_FORCE) - Let faultin_vma_page_range() check for compatible mappings/permissions - Extend patch description and add some performance numbers --- arch/alpha/include/uapi/asm/mman.h | 3 ++ arch/mips/include/uapi/asm/mman.h | 3 ++ arch/parisc/include/uapi/asm/mman.h | 3 ++ arch/xtensa/include/uapi/asm/mman.h | 3 ++ include/uapi/asm-generic/mman-common.h | 3 ++ mm/gup.c | 54 ++++++++++++++++++++ mm/internal.h | 3 ++ mm/madvise.c | 70 ++++++++++++++++++++++++++ 8 files changed, 142 insertions(+)