xref: /openbmc/linux/kernel/Kconfig.preempt (revision ccf988b6)
1# SPDX-License-Identifier: GPL-2.0-only
2
3choice
4	prompt "Preemption Model"
5	default PREEMPT_NONE
6
7config PREEMPT_NONE
8	bool "No Forced Preemption (Server)"
9	help
10	  This is the traditional Linux preemption model, geared towards
11	  throughput. It will still provide good latencies most of the
12	  time, but there are no guarantees and occasional longer delays
13	  are possible.
14
15	  Select this option if you are building a kernel for a server or
16	  scientific/computation system, or if you want to maximize the
17	  raw processing power of the kernel, irrespective of scheduling
18	  latencies.
19
20config PREEMPT_VOLUNTARY
21	bool "Voluntary Kernel Preemption (Desktop)"
22	depends on !ARCH_NO_PREEMPT
23	help
24	  This option reduces the latency of the kernel by adding more
25	  "explicit preemption points" to the kernel code. These new
26	  preemption points have been selected to reduce the maximum
27	  latency of rescheduling, providing faster application reactions,
28	  at the cost of slightly lower throughput.
29
30	  This allows reaction to interactive events by allowing a
31	  low priority process to voluntarily preempt itself even if it
32	  is in kernel mode executing a system call. This allows
33	  applications to run more 'smoothly' even when the system is
34	  under load.
35
36	  Select this if you are building a kernel for a desktop system.
37
38config PREEMPT_LL
39	bool "Preemptible Kernel (Low-Latency Desktop)"
40	depends on !ARCH_NO_PREEMPT
41	select PREEMPT
42	select UNINLINE_SPIN_UNLOCK if !ARCH_INLINE_SPIN_UNLOCK
43	help
44	  This option reduces the latency of the kernel by making
45	  all kernel code (that is not executing in a critical section)
46	  preemptible.  This allows reaction to interactive events by
47	  permitting a low priority process to be preempted involuntarily
48	  even if it is in kernel mode executing a system call and would
49	  otherwise not be about to reach a natural preemption point.
50	  This allows applications to run more 'smoothly' even when the
51	  system is under load, at the cost of slightly lower throughput
52	  and a slight runtime overhead to kernel code.
53
54	  Select this if you are building a kernel for a desktop or
55	  embedded system with latency requirements in the milliseconds
56	  range.
57
58config PREEMPT_RT
59	bool "Fully Preemptible Kernel (Real-Time)"
60	depends on EXPERT && ARCH_SUPPORTS_RT
61	select PREEMPT
62	help
63	  This option turns the kernel into a real-time kernel by replacing
64	  various locking primitives (spinlocks, rwlocks, etc.) with
65	  preemptible priority-inheritance aware variants, enforcing
66	  interrupt threading and introducing mechanisms to break up long
67	  non-preemptible sections. This makes the kernel, except for very
68	  low level and critical code pathes (entry code, scheduler, low
69	  level interrupt handling) fully preemptible and brings most
70	  execution contexts under scheduler control.
71
72	  Select this if you are building a kernel for systems which
73	  require real-time guarantees.
74
75endchoice
76
77config PREEMPT_COUNT
78       bool
79
80config PREEMPT
81       bool
82       select PREEMPT_COUNT
83