^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) # SPDX-License-Identifier: GPL-2.0-only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) #
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) # Architectures that offer an FUNCTION_TRACER implementation should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) # select HAVE_FUNCTION_TRACER:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) #
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) config USER_STACKTRACE_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) config NOP_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) config HAVE_FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) See Documentation/trace/ftrace-design.rst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) config HAVE_FUNCTION_GRAPH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) See Documentation/trace/ftrace-design.rst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) config HAVE_DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) See Documentation/trace/ftrace-design.rst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) config HAVE_DYNAMIC_FTRACE_WITH_REGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) config HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) config HAVE_FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) See Documentation/trace/ftrace-design.rst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) config HAVE_SYSCALL_TRACEPOINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) See Documentation/trace/ftrace-design.rst
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) config HAVE_FENTRY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) Arch supports the gcc options -pg with -mfentry
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) config HAVE_NOP_MCOUNT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) Arch supports the gcc options -pg with -mrecord-mcount and -nop-mcount
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) config HAVE_OBJTOOL_MCOUNT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) Arch supports objtool --mcount
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) config HAVE_C_RECORDMCOUNT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) C version of recordmcount available?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) config TRACER_MAX_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) config TRACE_CLOCK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) config RING_BUFFER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) select TRACE_CLOCK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) select IRQ_WORK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) config EVENT_TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) select CONTEXT_SWITCH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) select GLOB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) config CONTEXT_SWITCH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) config RING_BUFFER_ALLOW_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) Allow the use of ring_buffer_swap_cpu.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) Adds a very slight overhead to tracing when enabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) config TRACE_MMIO_ACCESS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) bool "Register read/write tracing"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) depends on TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) depends on ARM64
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) Create tracepoints for IO read/write operations. These trace events
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) can be used for logging all MMIO read/write operations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) config PREEMPTIRQ_TRACEPOINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) depends on TRACE_PREEMPT_TOGGLE || TRACE_IRQFLAGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) Create preempt/irq toggle tracepoints if needed, so that other parts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) of the kernel can use them to generate or add hooks to them.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) menuconfig TRACEFS_DISABLE_AUTOMOUNT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) bool "Do not autmount tracefs in the debugfs filesystem"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) Provides an option to not automount tracefs in /sys/kernel/debug/tracing.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) # All tracer options should select GENERIC_TRACER. For those options that are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) # enabled by all tracers (context switch and event tracer) they select TRACING.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) # This allows those options to appear when no other tracer is selected. But the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) # options do not appear when something else selects it. We need the two options
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) # GENERIC_TRACER and TRACING to avoid circular dependencies to accomplish the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) # hiding of the automatic options.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) config TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) select RING_BUFFER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) select STACKTRACE if STACKTRACE_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) select TRACEPOINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) select NOP_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) select BINARY_PRINTF
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) select EVENT_TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) select TRACE_CLOCK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) config GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) #
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) # Minimum requirements an architecture has to meet for us to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) # be able to offer generic tracing facilities:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) #
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) config TRACING_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) depends on TRACE_IRQFLAGS_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) depends on STACKTRACE_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) if TRACING_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) menuconfig FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) bool "Tracers"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) default y if DEBUG_KERNEL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) Enable the kernel tracing infrastructure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) if FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) config BOOTTIME_TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) bool "Boot-time Tracing support"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) depends on TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) select BOOT_CONFIG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) Enable developer to setup ftrace subsystem via supplemental
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) kernel cmdline at boot time for debugging (tracing) driver
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) initialization and boot process.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) config FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) bool "Kernel Function Tracer"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) depends on HAVE_FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) select KALLSYMS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) select CONTEXT_SWITCH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) select GLOB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) select TASKS_RCU if PREEMPTION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) select TASKS_RUDE_RCU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) Enable the kernel to trace every kernel function. This is done
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) by using a compiler feature to insert a small, 5-byte No-Operation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) instruction at the beginning of every kernel function, which NOP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) sequence is then dynamically patched into a tracer call when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) tracing is enabled by the administrator. If it's runtime disabled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) (the bootup default), then the overhead of the instructions is very
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) small and not measurable even in micro-benchmarks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) config FUNCTION_GRAPH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) bool "Kernel Function Graph Tracer"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) depends on HAVE_FUNCTION_GRAPH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) depends on FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) depends on !X86_32 || !CC_OPTIMIZE_FOR_SIZE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) Enable the kernel to trace a function at both its return
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) and its entry.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) Its first purpose is to trace the duration of functions and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) draw a call graph for each thread with some information like
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) the return value. This is done by setting the current return
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) address on the current task structure into a stack of calls.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) config DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) bool "enable/disable function tracing dynamically"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) depends on FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) depends on HAVE_DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) This option will modify all the calls to function tracing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) dynamically (will patch them out of the binary image and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) replace them with a No-Op instruction) on boot up. During
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) compile time, a table is made of all the locations that ftrace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) can function trace, and this table is linked into the kernel
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) image. When this is enabled, functions can be individually
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) enabled, and the functions not enabled will not affect
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) performance of the system.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) See the files in /sys/kernel/debug/tracing:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) available_filter_functions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) set_ftrace_filter
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) set_ftrace_notrace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) This way a CONFIG_FUNCTION_TRACER kernel is slightly larger, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) otherwise has native performance as long as no tracing is active.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) config DYNAMIC_FTRACE_WITH_REGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) def_bool y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) depends on DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) depends on HAVE_DYNAMIC_FTRACE_WITH_REGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) config DYNAMIC_FTRACE_WITH_DIRECT_CALLS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) def_bool y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) depends on DYNAMIC_FTRACE_WITH_REGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) depends on HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) config FUNCTION_PROFILER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) bool "Kernel function profiler"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) depends on FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) This option enables the kernel function profiler. A file is created
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) in debugfs called function_profile_enabled which defaults to zero.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) When a 1 is echoed into this file profiling begins, and when a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) zero is entered, profiling stops. A "functions" file is created in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) the trace_stat directory; this file shows the list of functions that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) have been hit and their counters.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) If in doubt, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) config STACK_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) bool "Trace max stack"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) depends on HAVE_FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) select FUNCTION_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) select STACKTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) select KALLSYMS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) This special tracer records the maximum stack footprint of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) kernel and displays it in /sys/kernel/debug/tracing/stack_trace.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) This tracer works by hooking into every function call that the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) kernel executes, and keeping a maximum stack depth value and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) stack-trace saved. If this is configured with DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) then it will not have any overhead while the stack tracer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) is disabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) To enable the stack tracer on bootup, pass in 'stacktrace'
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) on the kernel command line.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) The stack tracer can also be enabled or disabled via the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) sysctl kernel.stack_tracer_enabled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) Say N if unsure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) config TRACE_PREEMPT_TOGGLE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) Enables hooks which will be called when preemption is first disabled,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) and last enabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) config IRQSOFF_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) bool "Interrupts-off Latency Tracer"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) depends on TRACE_IRQFLAGS_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) depends on !ARCH_USES_GETTIMEOFFSET
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) select TRACE_IRQFLAGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) select TRACER_MAX_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) select RING_BUFFER_ALLOW_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) select TRACER_SNAPSHOT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) select TRACER_SNAPSHOT_PER_CPU_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) This option measures the time spent in irqs-off critical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) sections, with microsecond accuracy.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) The default measurement method is a maximum search, which is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) disabled by default and can be runtime (re-)started
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) via:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) (Note that kernel size and overhead increase with this option
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) enabled. This option and the preempt-off timing option can be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) used together or separately.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) config PREEMPT_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) bool "Preemption-off Latency Tracer"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) depends on !ARCH_USES_GETTIMEOFFSET
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) depends on PREEMPTION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) select TRACER_MAX_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) select RING_BUFFER_ALLOW_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) select TRACER_SNAPSHOT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) select TRACER_SNAPSHOT_PER_CPU_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) select TRACE_PREEMPT_TOGGLE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) This option measures the time spent in preemption-off critical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) sections, with microsecond accuracy.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) The default measurement method is a maximum search, which is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) disabled by default and can be runtime (re-)started
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) via:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) echo 0 > /sys/kernel/debug/tracing/tracing_max_latency
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) (Note that kernel size and overhead increase with this option
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) enabled. This option and the irqs-off timing option can be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) used together or separately.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) config SCHED_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) bool "Scheduling Latency Tracer"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) select CONTEXT_SWITCH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) select TRACER_MAX_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) select TRACER_SNAPSHOT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) This tracer tracks the latency of the highest priority task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) to be scheduled in, starting from the point it has woken up.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) config HWLAT_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) bool "Tracer to detect hardware latencies (like SMIs)"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) This tracer, when enabled will create one or more kernel threads,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) depending on what the cpumask file is set to, which each thread
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) spinning in a loop looking for interruptions caused by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) something other than the kernel. For example, if a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) System Management Interrupt (SMI) takes a noticeable amount of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) time, this tracer will detect it. This is useful for testing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) if a system is reliable for Real Time tasks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) Some files are created in the tracing directory when this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) is enabled:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) hwlat_detector/width - time in usecs for how long to spin for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) hwlat_detector/window - time in usecs between the start of each
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) iteration
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) A kernel thread is created that will spin with interrupts disabled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) for "width" microseconds in every "window" cycle. It will not spin
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) for "window - width" microseconds, where the system can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) continue to operate.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) The output will appear in the trace and trace_pipe files.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) When the tracer is not running, it has no affect on the system,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) but when it is running, it can cause the system to be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) periodically non responsive. Do not run this tracer on a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) production system.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) To enable this tracer, echo in "hwlat" into the current_tracer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) file. Every time a latency is greater than tracing_thresh, it will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) be recorded into the ring buffer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) config MMIOTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) bool "Memory mapped IO tracing"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) depends on HAVE_MMIOTRACE_SUPPORT && PCI
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) Mmiotrace traces Memory Mapped I/O access and is meant for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) debugging and reverse engineering. It is called from the ioremap
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) implementation and works via page faults. Tracing is disabled by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) default and can be enabled at run-time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) See Documentation/trace/mmiotrace.rst.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) If you are not helping to develop drivers, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) config ENABLE_DEFAULT_TRACERS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) bool "Trace process context switches and events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) depends on !GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) This tracer hooks to various trace points in the kernel,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) allowing the user to pick and choose which trace point they
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) want to trace. It also includes the sched_switch tracer plugin.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) config FTRACE_SYSCALLS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) bool "Trace syscalls"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) depends on HAVE_SYSCALL_TRACEPOINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) select KALLSYMS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) Basic tracer to catch the syscall entry and exit events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) config TRACER_SNAPSHOT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) bool "Create a snapshot trace buffer"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) select TRACER_MAX_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) Allow tracing users to take snapshot of the current buffer using the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) ftrace interface, e.g.:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) echo 1 > /sys/kernel/debug/tracing/snapshot
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) cat snapshot
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) config TRACER_SNAPSHOT_PER_CPU_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) bool "Allow snapshot to swap per CPU"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) depends on TRACER_SNAPSHOT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) select RING_BUFFER_ALLOW_SWAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) Allow doing a snapshot of a single CPU buffer instead of a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) full swap (all buffers). If this is set, then the following is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) allowed:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) echo 1 > /sys/kernel/debug/tracing/per_cpu/cpu2/snapshot
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) After which, only the tracing buffer for CPU 2 was swapped with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) the main tracing buffer, and the other CPU buffers remain the same.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) When this is enabled, this adds a little more overhead to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) trace recording, as it needs to add some checks to synchronize
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) recording with swaps. But this does not affect the performance
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) of the overall system. This is enabled by default when the preempt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) or irq latency tracers are enabled, as those need to swap as well
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) and already adds the overhead (plus a lot more).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) config TRACE_BRANCH_PROFILING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) choice
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) prompt "Branch Profiling"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) default BRANCH_PROFILE_NONE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) The branch profiling is a software profiler. It will add hooks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) into the C conditionals to test which path a branch takes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) The likely/unlikely profiler only looks at the conditions that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) are annotated with a likely or unlikely macro.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) The "all branch" profiler will profile every if-statement in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) kernel. This profiler will also enable the likely/unlikely
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) profiler.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) Either of the above profilers adds a bit of overhead to the system.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) If unsure, choose "No branch profiling".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) config BRANCH_PROFILE_NONE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) bool "No branch profiling"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) No branch profiling. Branch profiling adds a bit of overhead.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) Only enable it if you want to analyse the branching behavior.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) Otherwise keep it disabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) config PROFILE_ANNOTATED_BRANCHES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) bool "Trace likely/unlikely profiler"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) select TRACE_BRANCH_PROFILING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) This tracer profiles all likely and unlikely macros
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) in the kernel. It will display the results in:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) /sys/kernel/debug/tracing/trace_stat/branch_annotated
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) Note: this will add a significant overhead; only turn this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) on if you need to profile the system's use of these macros.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) config PROFILE_ALL_BRANCHES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) bool "Profile all if conditionals" if !FORTIFY_SOURCE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) select TRACE_BRANCH_PROFILING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) This tracer profiles all branch conditions. Every if ()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) taken in the kernel is recorded whether it hit or miss.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) The results will be displayed in:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) /sys/kernel/debug/tracing/trace_stat/branch_all
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) This option also enables the likely/unlikely profiler.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) This configuration, when enabled, will impose a great overhead
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) on the system. This should only be enabled when the system
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) is to be analyzed in much detail.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) endchoice
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) config TRACING_BRANCHES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) Selected by tracers that will trace the likely and unlikely
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) conditions. This prevents the tracers themselves from being
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) profiled. Profiling the tracing infrastructure can only happen
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) when the likelys and unlikelys are not being traced.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) config BRANCH_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) bool "Trace likely/unlikely instances"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) depends on TRACE_BRANCH_PROFILING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) select TRACING_BRANCHES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) This traces the events of likely and unlikely condition
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) calls in the kernel. The difference between this and the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) "Trace likely/unlikely profiler" is that this is not a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) histogram of the callers, but actually places the calling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) events into a running trace buffer to see when and where the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) events happened, as well as their results.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) Say N if unsure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) config BLK_DEV_IO_TRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) bool "Support for tracing block IO actions"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) depends on SYSFS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) depends on BLOCK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) select RELAY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) select DEBUG_FS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) select TRACEPOINTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) select GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) select STACKTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) Say Y here if you want to be able to trace the block layer actions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) on a given queue. Tracing allows you to see any traffic happening
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) on a block device queue. For more information (and the userspace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) support tools needed), fetch the blktrace tools from:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) git://git.kernel.dk/blktrace.git
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) Tracing also is possible using the ftrace interface, e.g.:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) echo 1 > /sys/block/sda/sda1/trace/enable
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) echo blk > /sys/kernel/debug/tracing/current_tracer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) cat /sys/kernel/debug/tracing/trace_pipe
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) config KPROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) depends on KPROBES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) depends on HAVE_REGS_AND_STACK_ACCESS_API
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) bool "Enable kprobes-based dynamic events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) select PROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) select DYNAMIC_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) This allows the user to add tracing events (similar to tracepoints)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) on the fly via the ftrace interface. See
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) Documentation/trace/kprobetrace.rst for more details.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) Those events can be inserted wherever kprobes can probe, and record
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) various register and memory values.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) This option is also required by perf-probe subcommand of perf tools.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) If you want to use perf tools, this option is strongly recommended.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) config KPROBE_EVENTS_ON_NOTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) bool "Do NOT protect notrace function from kprobe events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) depends on KPROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) depends on DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) This is only for the developers who want to debug ftrace itself
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) using kprobe events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) If kprobes can use ftrace instead of breakpoint, ftrace related
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) functions are protected from kprobe-events to prevent an infinit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) recursion or any unexpected execution path which leads to a kernel
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) crash.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) This option disables such protection and allows you to put kprobe
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) events on ftrace functions for debugging ftrace by itself.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) Note that this might let you shoot yourself in the foot.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) config UPROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) bool "Enable uprobes-based dynamic events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) depends on ARCH_SUPPORTS_UPROBES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) depends on MMU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) depends on PERF_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) select UPROBES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) select PROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) select DYNAMIC_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) This allows the user to add tracing events on top of userspace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) dynamic events (similar to tracepoints) on the fly via the trace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) events interface. Those events can be inserted wherever uprobes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) can probe, and record various registers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) This option is required if you plan to use perf-probe subcommand
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) of perf tools on user space applications.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) config BPF_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) depends on BPF_SYSCALL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) depends on (KPROBE_EVENTS || UPROBE_EVENTS) && PERF_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) This allows the user to attach BPF programs to kprobe, uprobe, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) tracepoint events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) config DYNAMIC_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) def_bool n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) config PROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) def_bool n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) config BPF_KPROBE_OVERRIDE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) bool "Enable BPF programs to override a kprobed function"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) depends on BPF_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) depends on FUNCTION_ERROR_INJECTION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) Allows BPF to override the execution of a probed function and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) set a different return value. This is used for error injection.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) config FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) def_bool y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) depends on DYNAMIC_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) depends on HAVE_FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) config FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) depends on FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) config FTRACE_MCOUNT_USE_CC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) def_bool y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) depends on $(cc-option,-mrecord-mcount)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) depends on FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) config FTRACE_MCOUNT_USE_OBJTOOL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) def_bool y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) depends on HAVE_OBJTOOL_MCOUNT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) depends on !FTRACE_MCOUNT_USE_CC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) depends on FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) config FTRACE_MCOUNT_USE_RECORDMCOUNT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) def_bool y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) depends on !FTRACE_MCOUNT_USE_CC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) depends on !FTRACE_MCOUNT_USE_OBJTOOL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) depends on FTRACE_MCOUNT_RECORD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) config TRACING_MAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) depends on ARCH_HAVE_NMI_SAFE_CMPXCHG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) tracing_map is a special-purpose lock-free map for tracing,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) separated out as a stand-alone facility in order to allow it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) to be shared between multiple tracers. It isn't meant to be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) generally used outside of that context, and is normally
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) selected by tracers that use it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) config SYNTH_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) bool "Synthetic trace events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) select DYNAMIC_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) Synthetic events are user-defined trace events that can be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) used to combine data from other trace events or in fact any
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) data source. Synthetic events can be generated indirectly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) via the trace() action of histogram triggers or directly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) by way of an in-kernel API.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) See Documentation/trace/events.rst or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) Documentation/trace/histogram.rst for details and examples.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) If in doubt, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) config HIST_TRIGGERS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) bool "Histogram triggers"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) depends on ARCH_HAVE_NMI_SAFE_CMPXCHG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) select TRACING_MAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) select TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) select DYNAMIC_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) select SYNTH_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) default n
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) Hist triggers allow one or more arbitrary trace event fields
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) to be aggregated into hash tables and dumped to stdout by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) reading a debugfs/tracefs file. They're useful for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) gathering quick and dirty (though precise) summaries of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) event activity as an initial guide for further investigation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) using more advanced tools.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) Inter-event tracing of quantities such as latencies is also
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) supported using hist triggers under this option.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) See Documentation/trace/histogram.rst.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) If in doubt, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) config TRACE_EVENT_INJECT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) bool "Trace event injection"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) depends on TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) Allow user-space to inject a specific trace event into the ring
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) buffer. This is mainly used for testing purpose.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) config TRACEPOINT_BENCHMARK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) bool "Add tracepoint that benchmarks tracepoints"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) This option creates the tracepoint "benchmark:benchmark_event".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) When the tracepoint is enabled, it kicks off a kernel thread that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) goes into an infinite loop (calling cond_sched() to let other tasks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) run), and calls the tracepoint. Each iteration will record the time
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) it took to write to the tracepoint and the next iteration that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) data will be passed to the tracepoint itself. That is, the tracepoint
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) will report the time it took to do the previous tracepoint.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) The string written to the tracepoint is a static string of 128 bytes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) to keep the time the same. The initial string is simply a write of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) "START". The second string records the cold cache time of the first
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) write which is not added to the rest of the calculations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) As it is a tight loop, it benchmarks as hot cache. That's fine because
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) we care most about hot paths that are probably in cache already.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) An example of the output:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) START
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) first=3672 [COLD CACHED]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) last=632 first=3672 max=632 min=632 avg=316 std=446 std^2=199712
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) last=278 first=3672 max=632 min=278 avg=303 std=316 std^2=100337
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) last=277 first=3672 max=632 min=277 avg=296 std=258 std^2=67064
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) last=273 first=3672 max=632 min=273 avg=292 std=224 std^2=50411
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) last=273 first=3672 max=632 min=273 avg=288 std=200 std^2=40389
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) last=281 first=3672 max=632 min=273 avg=287 std=183 std^2=33666
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) config RING_BUFFER_BENCHMARK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) tristate "Ring buffer benchmark stress tester"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) depends on RING_BUFFER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) This option creates a test to stress the ring buffer and benchmark it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) It creates its own ring buffer such that it will not interfere with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) any other users of the ring buffer (such as ftrace). It then creates
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) a producer and consumer that will run for 10 seconds and sleep for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) 10 seconds. Each interval it will print out the number of events
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) it recorded and give a rough estimate of how long each iteration took.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) It does not disable interrupts or raise its priority, so it may be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) affected by processes that are running.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) config TRACE_EVAL_MAP_FILE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) bool "Show eval mappings for trace events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) depends on TRACING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) The "print fmt" of the trace events will show the enum/sizeof names
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) instead of their values. This can cause problems for user space tools
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) that use this string to parse the raw data as user space does not know
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) how to convert the string to its value.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) To fix this, there's a special macro in the kernel that can be used
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) to convert an enum/sizeof into its value. If this macro is used, then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) the print fmt strings will be converted to their values.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) If something does not get converted properly, this option can be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) used to show what enums/sizeof the kernel tried to convert.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) This option is for debugging the conversions. A file is created
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) in the tracing directory called "eval_map" that will show the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) names matched with their values and what trace event system they
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) belong too.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) Normally, the mapping of the strings to values will be freed after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) boot up or module load. With this option, they will not be freed, as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) they are needed for the "eval_map" file. Enabling this option will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) increase the memory footprint of the running kernel.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) config GCOV_PROFILE_FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) bool "Enable GCOV profiling on ftrace subsystem"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) depends on GCOV_KERNEL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) Enable GCOV profiling on ftrace subsystem for checking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) which functions/lines are tested.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) Note that on a kernel compiled with this config, ftrace will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) run significantly slower.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) config FTRACE_SELFTEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) config FTRACE_STARTUP_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) bool "Perform a startup test on ftrace"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) depends on GENERIC_TRACER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) select FTRACE_SELFTEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) This option performs a series of startup tests on ftrace. On bootup
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) a series of tests are made to verify that the tracer is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) functioning properly. It will do tests on all the configured
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) tracers of ftrace.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) config EVENT_TRACE_STARTUP_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) bool "Run selftest on trace events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) depends on FTRACE_STARTUP_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) default y
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) This option performs a test on all trace events in the system.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) It basically just enables each event and runs some code that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) will trigger events (not necessarily the event it enables)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) This may take some time run as there are a lot of events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) config EVENT_TRACE_TEST_SYSCALLS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) bool "Run selftest on syscall events"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) depends on EVENT_TRACE_STARTUP_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) This option will also enable testing every syscall event.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) It only enables the event and disables it and runs various loads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) with the event enabled. This adds a bit more time for kernel boot
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) up since it runs this on every system call defined.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) TBD - enable a way to actually call the syscalls as we test their
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) events
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) config RING_BUFFER_STARTUP_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) bool "Ring buffer startup self test"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) depends on RING_BUFFER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) Run a simple self test on the ring buffer on boot up. Late in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) kernel boot sequence, the test will start that kicks off
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) a thread per cpu. Each thread will write various size events
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) into the ring buffer. Another thread is created to send IPIs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) to each of the threads, where the IPI handler will also write
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) to the ring buffer, to test/stress the nesting ability.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) If any anomalies are discovered, a warning will be displayed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) and all ring buffers will be disabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) The test runs for 10 seconds. This will slow your boot time
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) by at least 10 more seconds.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) At the end of the test, statics and more checks are done.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) It will output the stats of each per cpu buffer. What
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) was written, the sizes, what was read, what was lost, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) other similar details.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) If unsure, say N
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) config MMIOTRACE_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) tristate "Test module for mmiotrace"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) depends on MMIOTRACE && m
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) This is a dumb module for testing mmiotrace. It is very dangerous
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) as it will write garbage to IO memory starting at a given address.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) However, it should be safe to use on e.g. unused portion of VRAM.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) Say N, unless you absolutely know what you are doing.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) config PREEMPTIRQ_DELAY_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) tristate "Test module to create a preempt / IRQ disable delay thread to test latency tracers"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) depends on m
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) Select this option to build a test module that can help test latency
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) tracers by executing a preempt or irq disable section with a user
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) configurable delay. The module busy waits for the duration of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) critical section.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) For example, the following invocation generates a burst of three
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) irq-disabled critical sections for 500us:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) modprobe preemptirq_delay_test test_mode=irq delay=500 burst_size=3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) If unsure, say N
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) config SYNTH_EVENT_GEN_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) tristate "Test module for in-kernel synthetic event generation"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) depends on SYNTH_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) This option creates a test module to check the base
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) functionality of in-kernel synthetic event definition and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) generation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) To test, insert the module, and then check the trace buffer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) for the generated sample events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) config KPROBE_EVENT_GEN_TEST
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) tristate "Test module for in-kernel kprobe event generation"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) depends on KPROBE_EVENTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) This option creates a test module to check the base
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) functionality of in-kernel kprobe event definition.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) To test, insert the module, and then check the trace buffer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) for the generated kprobe events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) config HIST_TRIGGERS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) bool "Hist trigger debug support"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) depends on HIST_TRIGGERS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) help
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) Add "hist_debug" file for each event, which when read will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) dump out a bunch of internal details about the hist triggers
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) defined on that event.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) The hist_debug file serves a couple of purposes:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) - Helps developers verify that nothing is broken.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) - Provides educational information to support the details
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) of the hist trigger internals as described by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) Documentation/trace/histogram-design.rst.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) The hist_debug output only covers the data structures
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) related to the histogram definitions themselves and doesn't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) display the internals of map buckets or variable values of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) running histograms.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) If unsure, say N.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) endif # FTRACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) endif # TRACING_SUPPORT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918)
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags