Loading...
1
2choice
3 prompt "Preemption Model"
4 default PREEMPT_NONE
5
6config PREEMPT_NONE
7 bool "No Forced Preemption (Server)"
8 help
9 This is the traditional Linux preemption model, geared towards
10 throughput. It will still provide good latencies most of the
11 time, but there are no guarantees and occasional longer delays
12 are possible.
13
14 Select this option if you are building a kernel for a server or
15 scientific/computation system, or if you want to maximize the
16 raw processing power of the kernel, irrespective of scheduling
17 latencies.
18
19config PREEMPT_VOLUNTARY
20 bool "Voluntary Kernel Preemption (Desktop)"
21 help
22 This option reduces the latency of the kernel by adding more
23 "explicit preemption points" to the kernel code. These new
24 preemption points have been selected to reduce the maximum
25 latency of rescheduling, providing faster application reactions,
26 at the cost of slightly lower throughput.
27
28 This allows reaction to interactive events by allowing a
29 low priority process to voluntarily preempt itself even if it
30 is in kernel mode executing a system call. This allows
31 applications to run more 'smoothly' even when the system is
32 under load.
33
34 Select this if you are building a kernel for a desktop system.
35
36config PREEMPT
37 bool "Preemptible Kernel (Low-Latency Desktop)"
38 select PREEMPT_COUNT
39 select UNINLINE_SPIN_UNLOCK if !ARCH_INLINE_SPIN_UNLOCK
40 help
41 This option reduces the latency of the kernel by making
42 all kernel code (that is not executing in a critical section)
43 preemptible. This allows reaction to interactive events by
44 permitting a low priority process to be preempted involuntarily
45 even if it is in kernel mode executing a system call and would
46 otherwise not be about to reach a natural preemption point.
47 This allows applications to run more 'smoothly' even when the
48 system is under load, at the cost of slightly lower throughput
49 and a slight runtime overhead to kernel code.
50
51 Select this if you are building a kernel for a desktop or
52 embedded system with latency requirements in the milliseconds
53 range.
54
55endchoice
56
57config PREEMPT_COUNT
58
1# SPDX-License-Identifier: GPL-2.0-only
2
3config PREEMPT_NONE_BUILD
4 bool
5
6config PREEMPT_VOLUNTARY_BUILD
7 bool
8
9config PREEMPT_BUILD
10 bool
11 select PREEMPTION
12 select UNINLINE_SPIN_UNLOCK if !ARCH_INLINE_SPIN_UNLOCK
13
14config ARCH_HAS_PREEMPT_LAZY
15 bool
16
17choice
18 prompt "Preemption Model"
19 default PREEMPT_NONE
20
21config PREEMPT_NONE
22 bool "No Forced Preemption (Server)"
23 depends on !PREEMPT_RT
24 select PREEMPT_NONE_BUILD if !PREEMPT_DYNAMIC
25 help
26 This is the traditional Linux preemption model, geared towards
27 throughput. It will still provide good latencies most of the
28 time, but there are no guarantees and occasional longer delays
29 are possible.
30
31 Select this option if you are building a kernel for a server or
32 scientific/computation system, or if you want to maximize the
33 raw processing power of the kernel, irrespective of scheduling
34 latencies.
35
36config PREEMPT_VOLUNTARY
37 bool "Voluntary Kernel Preemption (Desktop)"
38 depends on !ARCH_NO_PREEMPT
39 depends on !PREEMPT_RT
40 select PREEMPT_VOLUNTARY_BUILD if !PREEMPT_DYNAMIC
41 help
42 This option reduces the latency of the kernel by adding more
43 "explicit preemption points" to the kernel code. These new
44 preemption points have been selected to reduce the maximum
45 latency of rescheduling, providing faster application reactions,
46 at the cost of slightly lower throughput.
47
48 This allows reaction to interactive events by allowing a
49 low priority process to voluntarily preempt itself even if it
50 is in kernel mode executing a system call. This allows
51 applications to run more 'smoothly' even when the system is
52 under load.
53
54 Select this if you are building a kernel for a desktop system.
55
56config PREEMPT
57 bool "Preemptible Kernel (Low-Latency Desktop)"
58 depends on !ARCH_NO_PREEMPT
59 select PREEMPT_BUILD if !PREEMPT_DYNAMIC
60 help
61 This option reduces the latency of the kernel by making
62 all kernel code (that is not executing in a critical section)
63 preemptible. This allows reaction to interactive events by
64 permitting a low priority process to be preempted involuntarily
65 even if it is in kernel mode executing a system call and would
66 otherwise not be about to reach a natural preemption point.
67 This allows applications to run more 'smoothly' even when the
68 system is under load, at the cost of slightly lower throughput
69 and a slight runtime overhead to kernel code.
70
71 Select this if you are building a kernel for a desktop or
72 embedded system with latency requirements in the milliseconds
73 range.
74
75config PREEMPT_LAZY
76 bool "Scheduler controlled preemption model"
77 depends on !ARCH_NO_PREEMPT
78 depends on ARCH_HAS_PREEMPT_LAZY
79 select PREEMPT_BUILD if !PREEMPT_DYNAMIC
80 help
81 This option provides a scheduler driven preemption model that
82 is fundamentally similar to full preemption, but is less
83 eager to preempt SCHED_NORMAL tasks in an attempt to
84 reduce lock holder preemption and recover some of the performance
85 gains seen from using Voluntary preemption.
86
87endchoice
88
89config PREEMPT_RT
90 bool "Fully Preemptible Kernel (Real-Time)"
91 depends on EXPERT && ARCH_SUPPORTS_RT && !COMPILE_TEST
92 select PREEMPTION
93 help
94 This option turns the kernel into a real-time kernel by replacing
95 various locking primitives (spinlocks, rwlocks, etc.) with
96 preemptible priority-inheritance aware variants, enforcing
97 interrupt threading and introducing mechanisms to break up long
98 non-preemptible sections. This makes the kernel, except for very
99 low level and critical code paths (entry code, scheduler, low
100 level interrupt handling) fully preemptible and brings most
101 execution contexts under scheduler control.
102
103 Select this if you are building a kernel for systems which
104 require real-time guarantees.
105
106config PREEMPT_COUNT
107 bool
108
109config PREEMPTION
110 bool
111 select PREEMPT_COUNT
112
113config PREEMPT_DYNAMIC
114 bool "Preemption behaviour defined on boot"
115 depends on HAVE_PREEMPT_DYNAMIC
116 select JUMP_LABEL if HAVE_PREEMPT_DYNAMIC_KEY
117 select PREEMPT_BUILD
118 default y if HAVE_PREEMPT_DYNAMIC_CALL
119 help
120 This option allows to define the preemption model on the kernel
121 command line parameter and thus override the default preemption
122 model defined during compile time.
123
124 The feature is primarily interesting for Linux distributions which
125 provide a pre-built kernel binary to reduce the number of kernel
126 flavors they offer while still offering different usecases.
127
128 The runtime overhead is negligible with HAVE_STATIC_CALL_INLINE enabled
129 but if runtime patching is not available for the specific architecture
130 then the potential overhead should be considered.
131
132 Interesting if you want the same pre-built kernel should be used for
133 both Server and Desktop workloads.
134
135config SCHED_CORE
136 bool "Core Scheduling for SMT"
137 depends on SCHED_SMT
138 help
139 This option permits Core Scheduling, a means of coordinated task
140 selection across SMT siblings. When enabled -- see
141 prctl(PR_SCHED_CORE) -- task selection ensures that all SMT siblings
142 will execute a task from the same 'core group', forcing idle when no
143 matching task is found.
144
145 Use of this feature includes:
146 - mitigation of some (not all) SMT side channels;
147 - limiting SMT interference to improve determinism and/or performance.
148
149 SCHED_CORE is default disabled. When it is enabled and unused,
150 which is the likely usage by Linux distributions, there should
151 be no measurable impact on performance.
152
153config SCHED_CLASS_EXT
154 bool "Extensible Scheduling Class"
155 depends on BPF_SYSCALL && BPF_JIT && DEBUG_INFO_BTF
156 select STACKTRACE if STACKTRACE_SUPPORT
157 help
158 This option enables a new scheduler class sched_ext (SCX), which
159 allows scheduling policies to be implemented as BPF programs to
160 achieve the following:
161
162 - Ease of experimentation and exploration: Enabling rapid
163 iteration of new scheduling policies.
164 - Customization: Building application-specific schedulers which
165 implement policies that are not applicable to general-purpose
166 schedulers.
167 - Rapid scheduler deployments: Non-disruptive swap outs of
168 scheduling policies in production environments.
169
170 sched_ext leverages BPF struct_ops feature to define a structure
171 which exports function callbacks and flags to BPF programs that
172 wish to implement scheduling policies. The struct_ops structure
173 exported by sched_ext is struct sched_ext_ops, and is conceptually
174 similar to struct sched_class.
175
176 For more information:
177 Documentation/scheduler/sched-ext.rst
178 https://github.com/sched-ext/scx