Home
last modified time | relevance | path

Searched refs:hugepages (Results 1 – 25 of 28) sorted by relevance

12

/linux/tools/testing/selftests/kvm/x86/
H A Ddirty_log_page_splitting_test.c39 u64 hugepages; member
47 stats->hugepages = stats->pages_2m + stats->pages_1g; in get_page_stats()
51 stats->hugepages); in get_page_stats()
183 TEST_ASSERT_EQ(stats_clear_pass[0].hugepages, 0); in run_test()
187 TEST_ASSERT_EQ(stats_dirty_logging_enabled.hugepages, stats_populated.hugepages); in run_test()
189 TEST_ASSERT_EQ(stats_dirty_logging_enabled.hugepages, 0); in run_test()
/linux/Documentation/ABI/testing/
H A Dsysfs-kernel-mm-hugepages1 What: /sys/kernel/mm/hugepages/
5 /sys/kernel/mm/hugepages/ contains a number of subdirectories
6 of the form hugepages-<size>kB, where <size> is the page size
7 of the hugepages supported by the kernel/CPU combination.
/linux/Documentation/arch/arm64/
H A Dhugetlbpage.rst10 - the size of hugepages
13 The ARM64 port supports two flavours of hugepages.
18 These are regular hugepages where a pmd or a pud page table entry points to a
/linux/Documentation/admin-guide/mm/
H A Dtranshuge.rst44 hugepages but a significant speedup already happens if only one of
45 the two is using hugepages just because of the fact the TLB miss is
77 hugepages. It requires no modifications for applications to take
87 In certain cases when hugepages are enabled system wide, application
91 possible to disable hugepages system-wide and to only have them inside
94 Embedded systems should enable hugepages only inside madvise regions
98 Applications that gets a lot of benefit from hugepages and that don't
99 risk to lose memory by using hugepages, should use
115 echo always >/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/enabled
116 echo madvise >/sys/kernel/mm/transparent_hugepage/hugepages-<size>kB/enabled
[all …]
/linux/tools/testing/selftests/mm/
H A Dthp_settings.c202 settings->hugepages[i].enabled = THP_NEVER; in thp_read_settings()
205 snprintf(path, PATH_MAX, "hugepages-%ukB/enabled", in thp_read_settings()
207 settings->hugepages[i].enabled = in thp_read_settings()
216 snprintf(path, PATH_MAX, "hugepages-%ukB/shmem_enabled", in thp_read_settings()
254 snprintf(path, PATH_MAX, "hugepages-%ukB/enabled", in thp_write_settings()
256 enabled = settings->hugepages[i].enabled; in thp_write_settings()
263 snprintf(path, PATH_MAX, "hugepages-%ukB/shmem_enabled", in thp_write_settings()
331 ret = snprintf(path, PATH_MAX, THP_SYSFS "hugepages-%ukB/%s", in thp_restore_settings()
H A Dprctl_thp_disable.c146 self->settings.hugepages[sz2ord(self->pmdsize, getpagesize())].enabled = THP_INHERIT; in FIXTURE_SETUP()
254 self->settings.hugepages[sz2ord(self->pmdsize, getpagesize())].enabled = THP_INHERIT; in FIXTURE_SETUP()
H A Dthp_settings.h61 struct hugepages_settings hugepages[NR_ORDERS]; member
H A Duffd-wp-mremap.c85 settings.hugepages[sz2ord(size, pagesize)].enabled = THP_ALWAYS; in alloc_one_folio()
349 settings.hugepages[i].enabled = THP_NEVER; in main()
H A Dcharge_reserved_hugetlb.sh301 echo nr hugepages = "$nr_hugepages"
354 echo nr hugepages = "$nr_hugepages"
H A Dcow.c1226 settings.hugepages[sz2ord(pmdsize, pagesize)].enabled = THP_NEVER; in run_anon_test_case()
1227 settings.hugepages[sz2ord(size, pagesize)].enabled = THP_ALWAYS; in run_anon_test_case()
1889 default_settings.hugepages[sz2ord(pmdsize, pagesize)].enabled = THP_INHERIT; in main()
H A Dkhugepaged.c1213 default_settings.hugepages[hpage_pmd_order].enabled = THP_INHERIT; in main()
1214 default_settings.hugepages[anon_order].enabled = THP_ALWAYS; in main()
H A Drun_vmtests.sh239 # On memory constrainted systems some tests can fail to allocate hugepages.
248 echo "hugepages not supported" | tap_prefix
339 # extra hugepages - this is done to prevent the test from failing by racily
340 # reserving more hugepages than strictly required.
352 for f in /sys/kernel/mm/hugepages/**/nr_hugepages; do
/linux/tools/testing/selftests/kvm/
H A Dmmu_stress_test.c287 bool hugepages = false; in main() local
320 hugepages = true; in main()
343 fd = kvm_memfd_alloc(slot_size, hugepages); in main()
/linux/Documentation/features/vm/THP/
H A Darch-support.txt4 # description: arch supports transparent hugepages
/linux/tools/testing/selftests/iommu/
H A Diommufd.c1630 bool hugepages; in FIXTURE_VARIANT() local
1654 if (variant->hugepages) { in FIXTURE_SETUP()
1672 .hugepages = false, in FIXTURE_VARIANT_ADD()
1679 .hugepages = false, in FIXTURE_VARIANT_ADD()
1686 .hugepages = true, in FIXTURE_VARIANT_ADD()
1693 .hugepages = true, in FIXTURE_VARIANT_ADD()
1700 .hugepages = false, in FIXTURE_VARIANT_ADD()
1707 .hugepages = true, in FIXTURE_VARIANT_ADD()
1826 size_t test_step = variant->hugepages ? (self->mmap_buf_size / 16) : in TEST_F()
1849 if (variant->hugepages) in TEST_F()
2094 bool hugepages; FIXTURE_VARIANT() local
[all...]
/linux/drivers/gpu/drm/i915/selftests/
H A Di915_mock_selftests.h34 selftest(hugepages, i915_gem_huge_page_mock_selftests)
H A Di915_live_selftests.h40 selftest(hugepages, i915_gem_huge_page_live_selftests)
/linux/arch/loongarch/kvm/
H A Dmmu.c94 kvm->stat.hugepages--; in kvm_flush_pte()
747 kvm->stat.hugepages--; in kvm_split_huge()
897 ++kvm->stat.hugepages; in kvm_map_page()
H A Dvm.c16 STATS_DESC_ICOUNTER(VM, hugepages),
/linux/arch/loongarch/include/asm/
H A Dkvm_host.h54 u64 hugepages; member
/linux/tools/testing/selftests/kvm/lib/
H A Dkvm_util.c852 int kvm_memfd_alloc(size_t size, bool hugepages) in kvm_memfd_alloc() argument
857 if (hugepages) in kvm_memfd_alloc()
/linux/tools/testing/selftests/kvm/include/
H A Dkvm_util.h476 int kvm_memfd_alloc(size_t size, bool hugepages);
/linux/Documentation/admin-guide/
H A Dcgroup-v2.rst1606 transparent hugepages
1610 hugepages
1614 transparent hugepages
1743 Number of transparent hugepages which were allocated to satisfy
1748 Number of transparent hugepages which were allocated to allow
1753 Number of transparent hugepages which are swapout in one piece
1757 Number of transparent hugepages which were split before swapout.
H A Dkernel-parameters.txt1235 the size represented by the legacy /proc/ hugepages
2067 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot.
2080 used in conjunction with hugepages (above) to
2082 pair hugepagesz=X hugepages=Y can be specified once
2090 allocate hugepages during boot. This option can be
2098 of gigantic hugepages. Or using node format, the size
2104 hugepages using the CMA allocator. If enabled, the
2105 boot-time allocation of gigantic hugepages is skipped.
7575 to anonymous transparent hugepages.
7858 with respect to transparent hugepages.
/linux/Documentation/virt/kvm/x86/
H A Dmmu.rst30 hugepages, and similar features work without change

12