Home
last modified time | relevance | path

Searched full:ordered (Results 1 – 25 of 492) sorted by relevance

12345678910>>...20

/linux/Documentation/devicetree/bindings/scsi/
H A Dhisilicon-sas.txt22 sources; the interrupts are ordered in 3 groups, as follows:
30 The phy interrupts are ordered into groups of 3 per phy
34 The interrupts are ordered in increasing order.
35 Fatal interrupts : the fatal interrupts are ordered as follows:
39 the interrupts are ordered in 3 groups, as follows:
47 interrupt. The interrupts are ordered in increasing
50 interrupt source. The interrupts are ordered in
/linux/drivers/pinctrl/samsung/
H A Dpinctrl-exynos-arm.c105 /* Must start with EINTG banks, ordered by EINT group number. */
166 /* Must start with EINTG banks, ordered by EINT group number. */
178 /* Must start with EINTG banks, ordered by EINT group number. */
253 /* Must start with EINTG banks, ordered by EINT group number. */
274 /* Must start with EINTG banks, ordered by EINT group number. */
299 /* Must start with EINTG banks, ordered by EINT group number. */
370 /* Must start with EINTG banks, ordered by EINT group number. */
388 /* Must start with EINTG banks, ordered by EINT group number. */
416 /* Must start with EINTG banks, ordered by EINT group number. */
422 /* Must start with EINTG banks, ordered by EINT group number. */
[all …]
/linux/arch/riscv/include/asm/
H A Dio.h40 * be fully ordered with respect to other memory and I/O operations". We're
42 * - Fully ordered WRT each other, by bracketing them with two fences. The
43 * outer set contains both I/O so inX is ordered with outX, while the inner just
45 * - Ordered in the same manner as readX/writeX WRT memory by subsuming their
47 * - Ordered WRT timer reads, so udelay and friends don't get elided by the
60 * Accesses from a single hart to a single I/O address must be ordered. This
H A Dpgtable-64.h119 * 01 - NC Non-cacheable, idempotent, weakly-ordered Main Memory
120 * 10 - IO Non-cacheable, non-idempotent, strongly-ordered I/O memory
134 * 00110 - NC Weakly-ordered, Non-cacheable, Bufferable, Shareable, Non-trustable
135 * 01110 - PMA Weakly-ordered, Cacheable, Bufferable, Shareable, Non-trustable
136 * 10010 - IO Strongly-ordered, Non-cacheable, Non-bufferable, Shareable, Non-trustable
H A Dmmio.h106 * platform specification defines that all I/O regions are strongly ordered on
129 * I/O memory access primitives. Reads are ordered relative to any following
130 * Normal memory read and delay() loop. Writes are ordered relative to any
/linux/Documentation/
H A Datomic_t.txt156 atomic variable) can be fully ordered and no intermediate state is lost or
169 - RMW operations that have a return value are fully ordered;
183 Fully ordered primitives are ordered against everything prior and everything
184 subsequent. Therefore a fully ordered primitive is like having an smp_mb()
198 ordered, so it is advisable to place the barrier right next to the RMW atomic
203 provide full ordered atomics and these barriers are no-ops.
205 NOTE: when the atomic RmW ops are fully ordered, they should also imply a
H A Datomic_bitops.txt59 - RMW operations that have a return value are fully ordered.
61 - RMW operations that are conditional are fully ordered.
/linux/fs/btrfs/
H A Dextent-io-tree.h34 * Must be cleared only during ordered extent completion or on error
35 * paths if we did not manage to submit bios and create the ordered
37 * and page invalidation (if there is an ordered extent in flight),
38 * that is left for the ordered extent completion.
42 * Mark that a range is being locked for finishing an ordered extent.
47 * When an ordered extent successfully completes for a region marked as
H A Dasync-thread.c29 /* List head pointing to ordered work list */
135 /* Ordered workqueues don't allow @max_active adjustments. */ in btrfs_alloc_ordered_workqueue()
228 * this guarantees that the ordered function will see all in run_ordered_work()
234 * we are going to call the ordered done function, but in run_ordered_work()
275 * We don't want to call the ordered free functions with in run_ordered_work()
316 * ordered before setting the WORK_DONE_BIT. Ensuring the thread in btrfs_work_helper()
317 * which is going to executed the ordered work sees them. in btrfs_work_helper()
H A Ddefrag.c839 * - No ordered extent exists in the page.
888 /* Wait for any existing ordered extent in the range */ in defrag_prepare_one_folio()
890 struct btrfs_ordered_extent *ordered; in defrag_prepare_one_folio() local
893 ordered = btrfs_lookup_ordered_range(inode, lock_start, folio_size(folio)); in defrag_prepare_one_folio()
895 if (!ordered) in defrag_prepare_one_folio()
899 btrfs_start_ordered_extent(ordered); in defrag_prepare_one_folio()
900 btrfs_put_ordered_extent(ordered); in defrag_prepare_one_folio()
914 * Now the page range has no ordered extent any more. Read the page to in defrag_prepare_one_folio()
1011 * ordered extents to complete. The deadlock would happen in defrag_collect_targets()
1013 * locked, and starting writeback, or finishing an ordered in defrag_collect_targets()
[all …]
H A Dfiemap.c38 * ordered extent completion, which is needed in order to reliably detect
124 * without FIEMAP_FLAG_SYNC and there are ordered extents completing in emit_fiemap_extent()
131 * btrfs_next_leaf() releases the path, there's an ordered extent that in emit_fiemap_extent()
147 * flushed and the resulting ordered extents were completed, so we can in emit_fiemap_extent()
174 * due to an ordered extent completing. in emit_fiemap_extent()
861 * waiting for an ordered extent that in order to complete needs to in extent_fiemap()
896 * complete and writeback to start. We also need to wait for ordered in btrfs_fiemap()
912 * triggering writeback and waiting for the completion of IO and ordered in btrfs_fiemap()
/linux/tools/perf/util/
H A Dordered-events.c7 #include "ordered-events.h"
112 * We maintain the following scheme of buffers for ordered in alloc_event()
119 * Each buffer keeps an array of ordered events objects: in alloc_event()
124 * Each allocated ordered event is linked to one of in alloc_event()
126 * - time ordered list 'events' in alloc_event()
129 * Allocation of the ordered event uses the following order in alloc_event()
135 * Removal of ordered event object moves it from events to in alloc_event()
237 ui_progress__init(&prog, oe->nr_events, "Processing time ordered events..."); in do_flush()
/linux/drivers/platform/x86/hp/hp-bioscfg/
H A Dorder-list-attributes.c3 * Functions corresponding to ordered list type attributes under
4 * BIOS ORDERED LIST GUID for use with hp-bioscfg driver.
51 * separators when reporting ordered-list values. in validate_ordered_list_input()
78 return sysfs_emit(buf, "ordered-list\n"); in type_show()
261 * Ordered list data is stored in hex and comma separated format in hp_populate_ordered_list_elements_from_package()
365 …pr_warn("Ordered List size value exceeded the maximum number of elements supported or data may be … in hp_populate_ordered_list_elements_from_buffer()
384 * instance under ordered list attribute
399 /* Populate ordered list elements */ in hp_populate_ordered_list_buffer_data()
/linux/rust/kernel/sync/atomic/
H A Dordering.rs13 //! - [`Full`] means "fully-ordered", that is:
43 /// The annotation type for fully-ordered memory ordering, for the description fully-ordered memory
58 /// Fully-ordered.
/linux/Documentation/arch/riscv/
H A Duabi.rst28 ordered first by category, in canonical order, as listed above, then
33 extensions are listed, they will be ordered alphabetically.
37 extensions are listed, they will be ordered alphabetically.
41 ordered alphabetically.
/linux/drivers/net/ethernet/cavium/liquidio/
H A Dresponse_manager.h27 /** Maximum ordered requests to process in every invocation of
37 * system. One for each response order- Unordered, ordered
134 /** Check the status of first entry in the ordered list. If the instruction at
138 * @return 1 if the ordered list is empty, 0 otherwise.
/linux/tools/memory-model/litmus-tests/
H A DREADME7 successive reads from the same variable are ordered.
12 are ordered.
17 are ordered.
21 successive writes to the same variable are ordered.
109 This is the fully ordered (via smp_mb()) version of one of
117 This is the fully ordered (again, via smp_mb() version of store
/linux/Documentation/litmus-tests/
H A DREADME18 the RMW are ordered before the subsequential memory accesses.
24 cmpxchg-fail-ordered-1.litmus
28 cmpxchg-fail-ordered-2.litmus
/linux/Documentation/devicetree/bindings/sound/
H A Dsirf-audio-port.txt6 - dmas: List of DMA controller phandle and DMA request line ordered pairs.
8 These strings correspond 1:1 with the ordered pairs in dmas.
/linux/virt/kvm/
H A DKconfig23 # Only strongly ordered architectures can select this, as it doesn't
31 # Weakly ordered architectures can only select this, advertising
/linux/tools/lib/subcmd/
H A Dparse-options.c811 struct option *opt, *ordered = NULL, *group; in options__order() local
823 group = realloc(ordered, len); in options__order()
826 ordered = group; in options__order()
827 memcpy(&ordered[nr_parent], p, sizeof(*o) * (nr_opts - nr_parent)); in options__order()
832 memcpy(&ordered[nr_opts], o, sizeof(*o)); in options__order()
835 for (opt = group = ordered; opt->type != OPTION_END; opt++) { in options__order()
847 return ordered; in options__order()
885 struct option *ordered; in usage_with_options_internal() local
910 ordered = options__order(opts); in usage_with_options_internal()
911 if (ordered) in usage_with_options_internal()
[all …]
/linux/kernel/irq/
H A Dipi-mux.c67 * ensure that the vIPI flag set is ordered after any shared in ipi_mux_send_mask()
84 * itself already ordered after the vIPI flag write. in ipi_mux_send_mask()
128 * Reading enable mask does not need to be ordered as long as in ipi_mux_process()
/linux/include/linux/
H A Datomic.h15 * - Fully ordered: The default implementation, no suffix required.
37 * variant is already fully ordered, no additional barriers are needed.
/linux/tools/memory-model/Documentation/
H A Dlocking.txt141 be ordered before the READ_ONCE() from data, thus solving the first
143 ordered after the WRITE_ONCE() to "data", solving the second problem.
156 It is not necessarily the case that accesses ordered by locking will be
157 seen as ordered by CPUs not holding that lock. Consider this example:
/linux/tools/perf/Documentation/
H A Dperf.txt60 ordered-events - ordered events object debug messages

12345678910>>...20