Home
last modified time | relevance | path

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

/linux/Documentation/admin-guide/cgroup-v1/
H A Dhugetlb.rst24 …ugepagesize>.rsvd.limit_in_bytes # set/show limit of "hugepagesize" hugetlb reservations
25 …ize>.rsvd.max_usage_in_bytes # show max "hugepagesize" hugetlb reservations and no-reserve …
26 …hugetlb.<hugepagesize>.rsvd.usage_in_bytes # show current reservations and no-reserve f…
94 The HugeTLB controller allows to limit the HugeTLB reservations per control
127 When a HugeTLB cgroup goes offline with some reservations or faults still
134 reservations charged to it, that cgroup persists as a zombie until all HugeTLB
135 reservations are uncharged. HugeTLB reservations behave in this manner to match
137 memory is uncharged. Also, the tracking of HugeTLB reservations is a bit more
139 harder to reparent reservations at offline time.
/linux/Documentation/mm/
H A Dhugetlbfs_reserv.rst62 region in the reserv_map may indicate reservations exist for the
63 range, or reservations do not exist.
68 Indicates this task is the owner of the reservations
122 in which reservations are represented in the reservation map.
125 exists or did exist for the corresponding page. As reservations are
135 to indicate this VMA owns the reservations.
173 ensure reservations exist for the range 'from' - 'to'.
207 is called. This routine takes two arguments related to reservations:
316 accounting, and track any reservations associated with the subpool.
449 are decremented by the number of outstanding reservations.
[all …]
/linux/Documentation/filesystems/
H A Docfs2.rst99 resv_level=2 (*) Set how aggressive allocation reservations will be.
100 Valid values are between 0 (reservations off) to 8
101 (maximum space for reservations).
102 dir_resv_level= (*) By default, directory reservations will scale with file
103 reservations - users should rarely need to change this
104 value. If allocation reservations are turned off, this
/linux/fs/bcachefs/
H A Djournal.h126 return j->buf + j->reservations.idx; in journal_cur_buf()
262 }).v, &j->reservations.counter); in journal_state_buf_put()
330 u64 v = atomic64_read(&j->reservations.counter); in journal_res_get_fast()
359 } while ((v = atomic64_cmpxchg(&j->reservations.counter, in journal_res_get_fast()
413 return j->reservations.cur_entry_offset == JOURNAL_ENTRY_ERROR_VAL in bch2_journal_error()
H A Djournal.c47 return __journal_entry_is_open(j->reservations); in journal_entry_is_open()
52 union journal_res_state s = READ_ONCE(j->reservations); in bch2_journal_buf_to_text()
199 if (!journal_state_count(j->reservations, idx)) { in bch2_journal_do_writes()
233 u64 v = atomic64_read(&j->reservations.counter); in __journal_entry_close()
435 v = atomic64_read(&j->reservations.counter); in journal_entry_open()
482 if (__journal_entry_is_open(j->reservations)) { in journal_write_work()
630 state = READ_ONCE(j->reservations); in bch2_journal_entry_res_resize()
1266 j->reservations.idx = j->reservations.unwritten_idx = journal_cur_seq(j); in bch2_fs_journal_start()
1267 j->reservations.unwritten_idx++; in bch2_fs_journal_start()
1378 atomic64_set(&j->reservations.counter, in bch2_fs_journal_init()
[all …]
H A Djournal_types.h176 union journal_res_state reservations; member
H A Djournal_io.c1645 v = atomic64_read(&j->reservations.counter); in CLOSURE_CALLBACK()
1652 } while ((v = atomic64_cmpxchg(&j->reservations.counter, old.v, new.v)) != old.v); in CLOSURE_CALLBACK()
/linux/Documentation/arch/arm64/
H A Dkdump.rst31 For kdump reservations, low memory is the memory area under a specific
66 reservations. The user would not need to know the system memory layout
75 many systems the low memory is precious and crashkernel reservations
/linux/fs/ocfs2/
H A DMakefile31 reservations.o \
/linux/Documentation/block/
H A Dpr.rst16 All implementations are expected to ensure the reservations survive
22 The following types of reservations are supported:
/linux/Documentation/admin-guide/nfs/
H A Dpnfs-scsi-server.rst21 option and the underlying SCSI device support persistent reservations.
/linux/Documentation/filesystems/xfs/
H A Dxfs-delayed-logging-design.rst16 transaction reservations are structured and accounted, and then move into how we
18 reservations bounds. At this point we need to explain how relogging works. With
59 transactions. Permanent transaction reservations can take reservations that span
65 modifications, but one-shot reservations cannot be used for permanent
156 journal. This is achieved by the transaction reservations that are made when
166 of the btree. As such, the reservations are quite complex because we have to
233 reservations currently held by active transactions. It is a purely in-memory
242 transactions from taking new reservations when the head reaches the current
256 count" reaches zero and the initial set of unit reservations have been
262 enough free space in the log to fulfill all of the pending reservations and
[all …]
H A Dxfs-online-fsck-design.rst1659 This reduces the worst case size of transaction reservations by breaking the
2449 reservations pin the tail of the ondisk log.
2534 clean up the space reservations that were made for the new btree, and reap the
2545 b. For unclaimed portions of incore reservations, create a regular deferred
2793 The space reservations used to create the new metadata can be used here if
3065 which are key to enabling resource reservations for active transactions.
4275 must be completed before transaction reservations are made.
4278 the appropriate resource reservations, locks, and fill out a ``struct
/linux/fs/btrfs/
H A Dblock-group.h232 atomic_t reservations; member
H A Dblock-group.c374 if (atomic_dec_and_test(&bg->reservations)) in btrfs_dec_block_group_reservations()
375 wake_up_var(&bg->reservations); in btrfs_dec_block_group_reservations()
401 wait_var_event(&bg->reservations, !atomic_read(&bg->reservations)); in btrfs_wait_block_group_reservations()
H A Dextent-tree.c2710 atomic_inc(&bg->reservations); in btrfs_inc_block_group_reservations()
/linux/Documentation/bpf/
H A Dringbuf.rst179 a strict ordering between reservations. Commits, on the other hand, are
181 in the order of reservations, but only after all previous records where
/linux/Documentation/driver-api/
H A Ddma-buf.rst349 reservations for DMA fence workloads.
/linux/Documentation/arch/powerpc/
H A Dfirmware-assisted-dump.rst298 file will change to reflect the new memory reservations.
/linux/mm/
H A Dhugetlb.c1238 struct resv_map *reservations = vma_resv_map(vma); in clear_vma_resv_huge_pages() local
1240 if (reservations && is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { in clear_vma_resv_huge_pages()
1241 resv_map_put_hugetlb_cgroup_uncharge_info(reservations); in clear_vma_resv_huge_pages()
1242 kref_put(&reservations->refs, resv_map_release); in clear_vma_resv_huge_pages()
/linux/Documentation/scheduler/
H A Dsched-deadline.rst772 showing how SCHED_DEADLINE reservations can be created by a real-time
/linux/Documentation/driver-api/usb/
H A Dusb.rst937 With the Linux-USB stack, periodic bandwidth reservations use the
/linux/Documentation/arch/x86/
H A Dresctrl.rst1167 end up allocating the same bits so the reservations are shared instead of
/linux/Documentation/networking/dsa/
H A Ddsa.rst507 reservations per port and per traffic class, in the ingress and egress
/linux/Documentation/admin-guide/
H A Dkernel-parameters.txt4526 use_e820 [X86] Use E820 reservations to exclude parts of
4531 no_e820 [X86] Ignore E820 reservations for PCI host