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