/openbmc/linux/tools/testing/selftests/mm/ |
H A D | .gitignore | 3 hugepage-mmap 4 hugepage-mremap 5 hugepage-shm 6 hugepage-vmemmap
|
H A D | Makefile | 45 TEST_GEN_FILES += hugepage-mmap 46 TEST_GEN_FILES += hugepage-mremap 47 TEST_GEN_FILES += hugepage-shm 48 TEST_GEN_FILES += hugepage-vmemmap
|
H A D | run_vmtests.sh | 235 CATEGORY="hugetlb" run_test ./hugepage-mmap 241 CATEGORY="hugetlb" run_test ./hugepage-shm 246 CATEGORY="hugetlb" run_test ./hugepage-mremap 247 CATEGORY="hugetlb" run_test ./hugepage-vmemmap
|
/openbmc/linux/Documentation/mm/ |
H A D | transhuge.rst | 12 - "graceful fallback": mm components which don't have transparent hugepage 14 if necessary, split a transparent hugepage. Therefore these components 17 - if a hugepage allocation fails because of memory fragmentation, 30 is not specific to transparent hugepage support and it's a generic 37 get_user_pages and follow_page if run on a hugepage, will return the 52 hugepage backed mappings. 59 pmd_offset. It's trivial to make the code transparent hugepage aware 64 hugepage aware. 66 If you're not walking pagetables but you run into a physical hugepage 69 it tries to swapout the hugepage for example. split_huge_page() can fail [all …]
|
/openbmc/linux/Documentation/arch/arm64/ |
H A D | hugetlbpage.rst | 20 mappings reduce the depth of page table walk needed to translate hugepage 35 The following hugepage sizes are supported -
|
/openbmc/linux/mm/ |
H A D | hugetlb_cgroup.c | 38 return &h_cg->hugepage[idx]; in __hugetlb_cgroup_counter_from_cgroup() 209 page_counter_charge(&parent->hugepage[idx], nr_pages); in hugetlb_cgroup_move_parent() 211 counter = &h_cg->hugepage[idx]; in hugetlb_cgroup_move_parent() 497 page_counter_read(&h_cg->hugepage[idx]) * PAGE_SIZE); in hugetlb_cgroup_read_numa_stat() 527 counter = &h_cg->hugepage[MEMFILE_IDX(cft->private)]; in hugetlb_cgroup_read_u64() 562 counter = &h_cg->hugepage[idx]; in hugetlb_cgroup_read_u64_max() 651 counter = &h_cg->hugepage[MEMFILE_IDX(of_cft(of)->private)]; in hugetlb_cgroup_reset()
|
H A D | Kconfig | 1168 # Some architectures require a special hugepage directory format that is 1169 # required to support multiple hugepage sizes. For example a4fe3ce76 1170 # "powerpc/mm: Allow more flexible layouts for hugepage pagetables" 1171 # introduced it on powerpc. This allows for a more flexible hugepage
|
H A D | mempolicy.c | 2169 unsigned long addr, bool hugepage) in vma_alloc_folio() argument 2200 if (unlikely(IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && hugepage)) { in vma_alloc_folio()
|
/openbmc/linux/Documentation/admin-guide/mm/ |
H A D | transhuge.rst | 59 entities). It doesn't require reservation to prevent hugepage 69 lived page allocations even for hugepage unaware applications that 157 library) may want to know the size (in bytes) of a transparent hugepage:: 187 and how many milliseconds to wait in khugepaged if there's an hugepage 196 one 2M hugepage. Each may happen independently, or together, depending on 246 You can control hugepage allocation policy in tmpfs with mount option 268 There's also sysfs knob to control hugepage allocation policy for internal 417 memory region, the mmap region has to be hugepage naturally 423 You can use hugetlbfs on a kernel that has transparent hugepage
|
H A D | hugetlbpage.rst | 66 is the default hugepage size (in kB). 146 It will allocate 1 2M hugepage on node0 and 2 2M hugepages on node1. 309 When adjusting the persistent hugepage count via ``nr_hugepages_mempolicy``, any 452 not hugepage aligned. For example, munmap(2) will fail if memory is backed by 453 a hugetlb page and the length is smaller than the hugepage size. 464 ``hugepage-shm`` 465 see tools/testing/selftests/mm/hugepage-shm.c 467 ``hugepage-mmap`` 468 see tools/testing/selftests/mm/hugepage-mmap.c
|
H A D | userfaultfd.rst | 38 Vmas are not suitable for page- (or hugepage) granular fault tracking
|
/openbmc/linux/Documentation/ |
H A D | dontdiff | 140 hugepage-mmap 141 hugepage-shm
|
/openbmc/linux/include/linux/ |
H A D | gfp.h | 267 unsigned long addr, bool hugepage); 277 #define vma_alloc_folio(gfp, order, vma, addr, hugepage) \ argument
|
H A D | hugetlb_cgroup.h | 48 struct page_counter hugepage[HUGE_MAX_HSTATE]; member
|
/openbmc/qemu/docs/ |
H A D | memory-hotplug.txt | 55 the /mnt/hugepages-1GB directory, a 1GB hugepage could be hotplugged 66 comes from regular RAM, 1GB is a 1GB hugepage page and 256MB is from
|
/openbmc/qemu/docs/system/devices/ |
H A D | ivshmem.rst | 54 memory backend that has hugepage support:
|
/openbmc/linux/Documentation/trace/ |
H A D | events-kmem.rst | 119 pageblock_size is usually the size of the default hugepage size.
|
/openbmc/linux/Documentation/admin-guide/cgroup-v1/ |
H A D | hugetlb.rst | 34 For a system supporting three hugepage sizes (64k, 32M and 1G), the control
|
/openbmc/qemu/docs/devel/migration/ |
H A D | postcopy.rst | 260 d) Care should be taken with the size of hugepage used; postcopy with 2MB 262 since it takes ~1 second to transfer a 1GB hugepage across a 10Gbps link,
|
/openbmc/linux/Documentation/translations/zh_CN/admin-guide/mm/damon/ |
H A D | usage.rst | 220 - ``hugepage``: 为带有 ``MADV_HUGEPAGE`` 的区域调用 ``madvise()`` 。
|
/openbmc/linux/Documentation/admin-guide/sysctl/ |
H A D | vm.rst | 560 Change the minimum size of the hugepage pool. 605 Change the size of the hugepage pool at run-time on a specific 614 Change the maximum size of the hugepage pool. The maximum is
|
/openbmc/linux/Documentation/admin-guide/mm/damon/ |
H A D | usage.rst | 289 - ``hugepage``: Call ``madvise()`` for the region with ``MADV_HUGEPAGE``.
|
/openbmc/linux/Documentation/filesystems/ |
H A D | proc.rst | 514 "AnonHugePages" shows the amount of memory backed by transparent hugepage. 932 A page block is typically the size of the default hugepage size, e.g. 2MB on
|
/openbmc/linux/ |
H A D | opengrok0.0.log | [all...] |
H A D | opengrok2.0.log | [all...] |