1 # SPDX-License-Identifier: GPL-2.0 !! 1 if BLOCK
>> 2
2 menu "IO Schedulers" 3 menu "IO Schedulers"
3 4
4 config MQ_IOSCHED_DEADLINE !! 5 config IOSCHED_NOOP
5 tristate "MQ deadline I/O scheduler" !! 6 bool
6 default y 7 default y
7 help !! 8 ---help---
8 MQ version of the deadline IO schedu !! 9 The no-op I/O scheduler is a minimal scheduler that does basic merging
>> 10 and sorting. Its main uses include non-disk based block devices like
>> 11 memory devices, and specialised software or hardware environments
>> 12 that do their own scheduling and require only minimal assistance from
>> 13 the kernel.
9 14
10 config MQ_IOSCHED_KYBER !! 15 config IOSCHED_AS
11 tristate "Kyber I/O scheduler" !! 16 tristate "Anticipatory I/O scheduler"
12 default y 17 default y
13 help !! 18 ---help---
14 The Kyber I/O scheduler is a low-ove !! 19 The anticipatory I/O scheduler is generally a good choice for most
15 multiqueue and other fast devices. G !! 20 environments, but is quite large and complex when compared to the
16 synchronous writes, it will self-tun !! 21 deadline I/O scheduler, it can also be slower in some cases
17 goal. !! 22 especially some database loads.
18 <<
19 config IOSCHED_BFQ <<
20 tristate "BFQ I/O scheduler" <<
21 select BLK_ICQ <<
22 help <<
23 BFQ I/O scheduler for BLK-MQ. BFQ dist <<
24 of the device among all processes acco <<
25 regardless of the device parameters an <<
26 also guarantees a low latency to inter <<
27 real-time applications. Details in <<
28 Documentation/block/bfq-iosched.rst <<
29 <<
30 config BFQ_GROUP_IOSCHED <<
31 bool "BFQ hierarchical scheduling suppo <<
32 depends on IOSCHED_BFQ && BLK_CGROUP <<
33 default y <<
34 select BLK_CGROUP_RWSTAT <<
35 help <<
36 23
37 Enable hierarchical scheduling in BFQ, !! 24 config IOSCHED_DEADLINE
38 (cgroups-v1) or io (cgroups-v2) control !! 25 tristate "Deadline I/O scheduler"
>> 26 default y
>> 27 ---help---
>> 28 The deadline I/O scheduler is simple and compact, and is often as
>> 29 good as the anticipatory I/O scheduler, and in some database
>> 30 workloads, better. In the case of a single process performing I/O to
>> 31 a disk at any one time, its behaviour is almost identical to the
>> 32 anticipatory I/O scheduler and so is a good choice.
39 33
40 config BFQ_CGROUP_DEBUG !! 34 config IOSCHED_CFQ
41 bool "BFQ IO controller debugging" !! 35 tristate "CFQ I/O scheduler"
42 depends on BFQ_GROUP_IOSCHED !! 36 default y
>> 37 ---help---
>> 38 The CFQ I/O scheduler tries to distribute bandwidth equally
>> 39 among all processes in the system. It should provide a fair
>> 40 working environment, suitable for desktop systems.
>> 41 This is the default I/O scheduler.
>> 42
>> 43 choice
>> 44 prompt "Default I/O scheduler"
>> 45 default DEFAULT_CFQ
43 help 46 help
44 Enable some debugging help. Currently !! 47 Select the I/O scheduler which will be used by default for all
45 files in a cgroup which can be useful !! 48 block devices.
>> 49
>> 50 config DEFAULT_AS
>> 51 bool "Anticipatory" if IOSCHED_AS=y
>> 52
>> 53 config DEFAULT_DEADLINE
>> 54 bool "Deadline" if IOSCHED_DEADLINE=y
>> 55
>> 56 config DEFAULT_CFQ
>> 57 bool "CFQ" if IOSCHED_CFQ=y
>> 58
>> 59 config DEFAULT_NOOP
>> 60 bool "No-op"
>> 61
>> 62 endchoice
>> 63
>> 64 config DEFAULT_IOSCHED
>> 65 string
>> 66 default "anticipatory" if DEFAULT_AS
>> 67 default "deadline" if DEFAULT_DEADLINE
>> 68 default "cfq" if DEFAULT_CFQ
>> 69 default "noop" if DEFAULT_NOOP
46 70
47 endmenu 71 endmenu
>> 72
>> 73 endif
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.