4 The HugeTLB controller allows to limit the HugeTLB usage per control group and
5 enforces the controller limit during page fault. Since HugeTLB doesn't
6 support page reclaim, enforcing the limit at page fault time implies that,
7 the application will get SIGBUS signal if it tries to access HugeTLB pages
8 beyond its limit. This requires the application to know beforehand how much
9 HugeTLB pages it would require for its use.
11 HugeTLB controller can be created by first mounting the cgroup filesystem.
13 # mount -t cgroup -o hugetlb none /sys/fs/cgroup
15 With the above step, the initial or the parent HugeTLB group becomes
16 visible at /sys/fs/cgroup. At bootup, this group includes all the tasks in
17 the system. /sys/fs/cgroup/tasks lists the tasks in this cgroup.
19 New groups can be created under the parent group /sys/fs/cgroup.
25 The above steps create a new group g1 and move the current shell
26 process (bash) into it.
28 Brief summary of control files
30 hugetlb.<hugepagesize>.limit_in_bytes # set/show limit of "hugepagesize" hugetlb usage
31 hugetlb.<hugepagesize>.max_usage_in_bytes # show max "hugepagesize" hugetlb usage recorded
32 hugetlb.<hugepagesize>.usage_in_bytes # show current res_counter usage for "hugepagesize" hugetlb
33 hugetlb.<hugepagesize>.failcnt # show the number of allocation failure due to HugeTLB limit
35 For a system supporting two hugepage size (16M and 16G) the control
38 hugetlb.16GB.limit_in_bytes
39 hugetlb.16GB.max_usage_in_bytes
40 hugetlb.16GB.usage_in_bytes
42 hugetlb.16MB.limit_in_bytes
43 hugetlb.16MB.max_usage_in_bytes
44 hugetlb.16MB.usage_in_bytes