Gitiles
Code Review Sign In
192.168.1.100 / T106_DC / 9ed821d7e5d875a3395740a9cc2545671fa429b7 / . / ap / os / linux / linux-3.4.x / arch / Kconfig
blob: 7b57cdc7b9b34a2c679466db03d83e3c45ad3a09 [file] [log] [blame]
lh9ed821d2023-04-07 01:36:19 -0700[diff] [blame^]1#
2# General architecture dependent options
3#
4
5config OPROFILE
6 tristate "OProfile system profiling"
7 depends on PROFILING
8 depends on HAVE_OPROFILE
9 depends on !PREEMPT_RT_FULL
10 select RING_BUFFER
11 select RING_BUFFER_ALLOW_SWAP
12 help
13 OProfile is a profiling system capable of profiling the
14 whole system, include the kernel, kernel modules, libraries,
15 and applications.
16
17 If unsure, say N.
18
19config OPROFILE_EVENT_MULTIPLEX
20 bool "OProfile multiplexing support (EXPERIMENTAL)"
21 default n
22 depends on OPROFILE && X86
23 help
24 The number of hardware counters is limited. The multiplexing
25 feature enables OProfile to gather more events than counters
26 are provided by the hardware. This is realized by switching
27 between events at an user specified time interval.
28
29 If unsure, say N.
30
31config HAVE_OPROFILE
32 bool
33
34config OPROFILE_NMI_TIMER
35 def_bool y
36 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
37
38config KPROBES
39 bool "Kprobes"
40 depends on MODULES
41 depends on HAVE_KPROBES
42 select KALLSYMS
43 help
44 Kprobes allows you to trap at almost any kernel address and
45 execute a callback function. register_kprobe() establishes
46 a probepoint and specifies the callback. Kprobes is useful
47 for kernel debugging, non-intrusive instrumentation and testing.
48 If in doubt, say "N".
49
50config JUMP_LABEL
51 bool "Optimize very unlikely/likely branches"
52 depends on HAVE_ARCH_JUMP_LABEL
53 depends on (!INTERRUPT_OFF_HIST && !PREEMPT_OFF_HIST && !WAKEUP_LATENCY_HIST && !MISSED_TIMER_OFFSETS_HIST)
54 help
55 This option enables a transparent branch optimization that
56 makes certain almost-always-true or almost-always-false branch
57 conditions even cheaper to execute within the kernel.
58
59 Certain performance-sensitive kernel code, such as trace points,
60 scheduler functionality, networking code and KVM have such
61 branches and include support for this optimization technique.
62
63 If it is detected that the compiler has support for "asm goto",
64 the kernel will compile such branches with just a nop
65 instruction. When the condition flag is toggled to true, the
66 nop will be converted to a jump instruction to execute the
67 conditional block of instructions.
68
69 This technique lowers overhead and stress on the branch prediction
70 of the processor and generally makes the kernel faster. The update
71 of the condition is slower, but those are always very rare.
72
73 ( On 32-bit x86, the necessary options added to the compiler
74 flags may increase the size of the kernel slightly. )
75
76config OPTPROBES
77 def_bool y
78 depends on KPROBES && HAVE_OPTPROBES
79 depends on !PREEMPT
80
81config HAVE_EFFICIENT_UNALIGNED_ACCESS
82 bool
83 help
84 Some architectures are unable to perform unaligned accesses
85 without the use of get_unaligned/put_unaligned. Others are
86 unable to perform such accesses efficiently (e.g. trap on
87 unaligned access and require fixing it up in the exception
88 handler.)
89
90 This symbol should be selected by an architecture if it can
91 perform unaligned accesses efficiently to allow different
92 code paths to be selected for these cases. Some network
93 drivers, for example, could opt to not fix up alignment
94 problems with received packets if doing so would not help
95 much.
96
97 See Documentation/unaligned-memory-access.txt for more
98 information on the topic of unaligned memory accesses.
99
100config HAVE_SYSCALL_WRAPPERS
101 bool
102
103config KRETPROBES
104 def_bool y
105 depends on KPROBES && HAVE_KRETPROBES
106
107config USER_RETURN_NOTIFIER
108 bool
109 depends on HAVE_USER_RETURN_NOTIFIER
110 help
111 Provide a kernel-internal notification when a cpu is about to
112 switch to user mode.
113
114config HAVE_IOREMAP_PROT
115 bool
116
117config HAVE_KPROBES
118 bool
119
120config HAVE_KRETPROBES
121 bool
122
123config HAVE_OPTPROBES
124 bool
125
126config HAVE_NMI_WATCHDOG
127 bool
128#
129# An arch should select this if it provides all these things:
130#
131# task_pt_regs() in asm/processor.h or asm/ptrace.h
132# arch_has_single_step() if there is hardware single-step support
133# arch_has_block_step() if there is hardware block-step support
134# asm/syscall.h supplying asm-generic/syscall.h interface
135# linux/regset.h user_regset interfaces
136# CORE_DUMP_USE_REGSET #define'd in linux/elf.h
137# TIF_SYSCALL_TRACE calls tracehook_report_syscall_{entry,exit}
138# TIF_NOTIFY_RESUME calls tracehook_notify_resume()
139# signal delivery calls tracehook_signal_handler()
140#
141config HAVE_ARCH_TRACEHOOK
142 bool
143
144config HAVE_DMA_ATTRS
145 bool
146
147config USE_GENERIC_SMP_HELPERS
148 bool
149
150config HAVE_REGS_AND_STACK_ACCESS_API
151 bool
152 help
153 This symbol should be selected by an architecure if it supports
154 the API needed to access registers and stack entries from pt_regs,
155 declared in asm/ptrace.h
156 For example the kprobes-based event tracer needs this API.
157
158config HAVE_CLK
159 bool
160 help
161 The <linux/clk.h> calls support software clock gating and
162 thus are a key power management tool on many systems.
163
164config HAVE_DMA_API_DEBUG
165 bool
166
167config HAVE_HW_BREAKPOINT
168 bool
169 depends on PERF_EVENTS
170
171config HAVE_MIXED_BREAKPOINTS_REGS
172 bool
173 depends on HAVE_HW_BREAKPOINT
174 help
175 Depending on the arch implementation of hardware breakpoints,
176 some of them have separate registers for data and instruction
177 breakpoints addresses, others have mixed registers to store
178 them but define the access type in a control register.
179 Select this option if your arch implements breakpoints under the
180 latter fashion.
181
182config HAVE_USER_RETURN_NOTIFIER
183 bool
184
185config HAVE_PERF_EVENTS_NMI
186 bool
187 help
188 System hardware can generate an NMI using the perf event
189 subsystem. Also has support for calculating CPU cycle events
190 to determine how many clock cycles in a given period.
191
192config HAVE_ARCH_JUMP_LABEL
193 bool
194
195config HAVE_ARCH_MUTEX_CPU_RELAX
196 bool
197
198config HAVE_RCU_TABLE_FREE
199 bool
200
201config ARCH_HAVE_NMI_SAFE_CMPXCHG
202 bool
203
204config HAVE_ALIGNED_STRUCT_PAGE
205 bool
206 help
207 This makes sure that struct pages are double word aligned and that
208 e.g. the SLUB allocator can perform double word atomic operations
209 on a struct page for better performance. However selecting this
210 might increase the size of a struct page by a word.
211
212config HAVE_CMPXCHG_LOCAL
213 bool
214
215config HAVE_CMPXCHG_DOUBLE
216 bool
217
218config ARCH_WANT_OLD_COMPAT_IPC
219 bool
220
221source "kernel/gcov/Kconfig"