1 /* 2 * trace task wakeup timings 3 * 4 * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com> 5 * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com> 6 * 7 * Based on code from the latency_tracer, that is: 8 * 9 * Copyright (C) 2004-2006 Ingo Molnar 10 * Copyright (C) 2004 William Lee Irwin III 11 */ 12 #include <linux/module.h> 13 #include <linux/fs.h> 14 #include <linux/debugfs.h> 15 #include <linux/kallsyms.h> 16 #include <linux/uaccess.h> 17 #include <linux/ftrace.h> 18 #include <trace/events/sched.h> 19 20 #include "trace.h" 21 22 static struct trace_array *wakeup_trace; 23 static int __read_mostly tracer_enabled; 24 25 static struct task_struct *wakeup_task; 26 static int wakeup_cpu; 27 static int wakeup_current_cpu; 28 static unsigned wakeup_prio = -1; 29 static int wakeup_rt; 30 31 static arch_spinlock_t wakeup_lock = 32 (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; 33 34 static void wakeup_reset(struct trace_array *tr); 35 static void __wakeup_reset(struct trace_array *tr); 36 static int wakeup_graph_entry(struct ftrace_graph_ent *trace); 37 static void wakeup_graph_return(struct ftrace_graph_ret *trace); 38 39 static int save_lat_flag; 40 41 #define TRACE_DISPLAY_GRAPH 1 42 43 static struct tracer_opt trace_opts[] = { 44 #ifdef CONFIG_FUNCTION_GRAPH_TRACER 45 /* display latency trace as call graph */ 46 { TRACER_OPT(display-graph, TRACE_DISPLAY_GRAPH) }, 47 #endif 48 { } /* Empty entry */ 49 }; 50 51 static struct tracer_flags tracer_flags = { 52 .val = 0, 53 .opts = trace_opts, 54 }; 55 56 #define is_graph() (tracer_flags.val & TRACE_DISPLAY_GRAPH) 57 58 #ifdef CONFIG_FUNCTION_TRACER 59 60 /* 61 * Prologue for the wakeup function tracers. 62 * 63 * Returns 1 if it is OK to continue, and preemption 64 * is disabled and data->disabled is incremented. 65 * 0 if the trace is to be ignored, and preemption 66 * is not disabled and data->disabled is 67 * kept the same. 68 * 69 * Note, this function is also used outside this ifdef but 70 * inside the #ifdef of the function graph tracer below. 71 * This is OK, since the function graph tracer is 72 * dependent on the function tracer. 73 */ 74 static int 75 func_prolog_preempt_disable(struct trace_array *tr, 76 struct trace_array_cpu **data, 77 int *pc) 78 { 79 long disabled; 80 int cpu; 81 82 if (likely(!wakeup_task)) 83 return 0; 84 85 *pc = preempt_count(); 86 preempt_disable_notrace(); 87 88 cpu = raw_smp_processor_id(); 89 if (cpu != wakeup_current_cpu) 90 goto out_enable; 91 92 *data = tr->data[cpu]; 93 disabled = atomic_inc_return(&(*data)->disabled); 94 if (unlikely(disabled != 1)) 95 goto out; 96 97 return 1; 98 99 out: 100 atomic_dec(&(*data)->disabled); 101 102 out_enable: 103 preempt_enable_notrace(); 104 return 0; 105 } 106 107 /* 108 * wakeup uses its own tracer function to keep the overhead down: 109 */ 110 static void 111 wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) 112 { 113 struct trace_array *tr = wakeup_trace; 114 struct trace_array_cpu *data; 115 unsigned long flags; 116 int pc; 117 118 if (!func_prolog_preempt_disable(tr, &data, &pc)) 119 return; 120 121 local_irq_save(flags); 122 trace_function(tr, ip, parent_ip, flags, pc); 123 local_irq_restore(flags); 124 125 atomic_dec(&data->disabled); 126 preempt_enable_notrace(); 127 } 128 129 static struct ftrace_ops trace_ops __read_mostly = 130 { 131 .func = wakeup_tracer_call, 132 .flags = FTRACE_OPS_FL_GLOBAL, 133 }; 134 #endif /* CONFIG_FUNCTION_TRACER */ 135 136 static int start_func_tracer(int graph) 137 { 138 int ret; 139 140 if (!graph) 141 ret = register_ftrace_function(&trace_ops); 142 else 143 ret = register_ftrace_graph(&wakeup_graph_return, 144 &wakeup_graph_entry); 145 146 if (!ret && tracing_is_enabled()) 147 tracer_enabled = 1; 148 else 149 tracer_enabled = 0; 150 151 return ret; 152 } 153 154 static void stop_func_tracer(int graph) 155 { 156 tracer_enabled = 0; 157 158 if (!graph) 159 unregister_ftrace_function(&trace_ops); 160 else 161 unregister_ftrace_graph(); 162 } 163 164 #ifdef CONFIG_FUNCTION_GRAPH_TRACER 165 static int wakeup_set_flag(u32 old_flags, u32 bit, int set) 166 { 167 168 if (!(bit & TRACE_DISPLAY_GRAPH)) 169 return -EINVAL; 170 171 if (!(is_graph() ^ set)) 172 return 0; 173 174 stop_func_tracer(!set); 175 176 wakeup_reset(wakeup_trace); 177 tracing_max_latency = 0; 178 179 return start_func_tracer(set); 180 } 181 182 static int wakeup_graph_entry(struct ftrace_graph_ent *trace) 183 { 184 struct trace_array *tr = wakeup_trace; 185 struct trace_array_cpu *data; 186 unsigned long flags; 187 int pc, ret = 0; 188 189 if (!func_prolog_preempt_disable(tr, &data, &pc)) 190 return 0; 191 192 local_save_flags(flags); 193 ret = __trace_graph_entry(tr, trace, flags, pc); 194 atomic_dec(&data->disabled); 195 preempt_enable_notrace(); 196 197 return ret; 198 } 199 200 static void wakeup_graph_return(struct ftrace_graph_ret *trace) 201 { 202 struct trace_array *tr = wakeup_trace; 203 struct trace_array_cpu *data; 204 unsigned long flags; 205 int pc; 206 207 if (!func_prolog_preempt_disable(tr, &data, &pc)) 208 return; 209 210 local_save_flags(flags); 211 __trace_graph_return(tr, trace, flags, pc); 212 atomic_dec(&data->disabled); 213 214 preempt_enable_notrace(); 215 return; 216 } 217 218 static void wakeup_trace_open(struct trace_iterator *iter) 219 { 220 if (is_graph()) 221 graph_trace_open(iter); 222 } 223 224 static void wakeup_trace_close(struct trace_iterator *iter) 225 { 226 if (iter->private) 227 graph_trace_close(iter); 228 } 229 230 #define GRAPH_TRACER_FLAGS (TRACE_GRAPH_PRINT_PROC | \ 231 TRACE_GRAPH_PRINT_ABS_TIME | \ 232 TRACE_GRAPH_PRINT_DURATION) 233 234 static enum print_line_t wakeup_print_line(struct trace_iterator *iter) 235 { 236 /* 237 * In graph mode call the graph tracer output function, 238 * otherwise go with the TRACE_FN event handler 239 */ 240 if (is_graph()) 241 return print_graph_function_flags(iter, GRAPH_TRACER_FLAGS); 242 243 return TRACE_TYPE_UNHANDLED; 244 } 245 246 static void wakeup_print_header(struct seq_file *s) 247 { 248 if (is_graph()) 249 print_graph_headers_flags(s, GRAPH_TRACER_FLAGS); 250 else 251 trace_default_header(s); 252 } 253 254 static void 255 __trace_function(struct trace_array *tr, 256 unsigned long ip, unsigned long parent_ip, 257 unsigned long flags, int pc) 258 { 259 if (is_graph()) 260 trace_graph_function(tr, ip, parent_ip, flags, pc); 261 else 262 trace_function(tr, ip, parent_ip, flags, pc); 263 } 264 #else 265 #define __trace_function trace_function 266 267 static int wakeup_set_flag(u32 old_flags, u32 bit, int set) 268 { 269 return -EINVAL; 270 } 271 272 static int wakeup_graph_entry(struct ftrace_graph_ent *trace) 273 { 274 return -1; 275 } 276 277 static enum print_line_t wakeup_print_line(struct trace_iterator *iter) 278 { 279 return TRACE_TYPE_UNHANDLED; 280 } 281 282 static void wakeup_graph_return(struct ftrace_graph_ret *trace) { } 283 static void wakeup_trace_open(struct trace_iterator *iter) { } 284 static void wakeup_trace_close(struct trace_iterator *iter) { } 285 286 #ifdef CONFIG_FUNCTION_TRACER 287 static void wakeup_print_header(struct seq_file *s) 288 { 289 trace_default_header(s); 290 } 291 #else 292 static void wakeup_print_header(struct seq_file *s) 293 { 294 trace_latency_header(s); 295 } 296 #endif /* CONFIG_FUNCTION_TRACER */ 297 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ 298 299 /* 300 * Should this new latency be reported/recorded? 301 */ 302 static int report_latency(cycle_t delta) 303 { 304 if (tracing_thresh) { 305 if (delta < tracing_thresh) 306 return 0; 307 } else { 308 if (delta <= tracing_max_latency) 309 return 0; 310 } 311 return 1; 312 } 313 314 static void 315 probe_wakeup_migrate_task(void *ignore, struct task_struct *task, int cpu) 316 { 317 if (task != wakeup_task) 318 return; 319 320 wakeup_current_cpu = cpu; 321 } 322 323 static void notrace 324 probe_wakeup_sched_switch(void *ignore, 325 struct task_struct *prev, struct task_struct *next) 326 { 327 struct trace_array_cpu *data; 328 cycle_t T0, T1, delta; 329 unsigned long flags; 330 long disabled; 331 int cpu; 332 int pc; 333 334 tracing_record_cmdline(prev); 335 336 if (unlikely(!tracer_enabled)) 337 return; 338 339 /* 340 * When we start a new trace, we set wakeup_task to NULL 341 * and then set tracer_enabled = 1. We want to make sure 342 * that another CPU does not see the tracer_enabled = 1 343 * and the wakeup_task with an older task, that might 344 * actually be the same as next. 345 */ 346 smp_rmb(); 347 348 if (next != wakeup_task) 349 return; 350 351 pc = preempt_count(); 352 353 /* disable local data, not wakeup_cpu data */ 354 cpu = raw_smp_processor_id(); 355 disabled = atomic_inc_return(&wakeup_trace->data[cpu]->disabled); 356 if (likely(disabled != 1)) 357 goto out; 358 359 local_irq_save(flags); 360 arch_spin_lock(&wakeup_lock); 361 362 /* We could race with grabbing wakeup_lock */ 363 if (unlikely(!tracer_enabled || next != wakeup_task)) 364 goto out_unlock; 365 366 /* The task we are waiting for is waking up */ 367 data = wakeup_trace->data[wakeup_cpu]; 368 369 __trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); 370 tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); 371 372 T0 = data->preempt_timestamp; 373 T1 = ftrace_now(cpu); 374 delta = T1-T0; 375 376 if (!report_latency(delta)) 377 goto out_unlock; 378 379 if (likely(!is_tracing_stopped())) { 380 tracing_max_latency = delta; 381 update_max_tr(wakeup_trace, wakeup_task, wakeup_cpu); 382 } 383 384 out_unlock: 385 __wakeup_reset(wakeup_trace); 386 arch_spin_unlock(&wakeup_lock); 387 local_irq_restore(flags); 388 out: 389 atomic_dec(&wakeup_trace->data[cpu]->disabled); 390 } 391 392 static void __wakeup_reset(struct trace_array *tr) 393 { 394 wakeup_cpu = -1; 395 wakeup_prio = -1; 396 397 if (wakeup_task) 398 put_task_struct(wakeup_task); 399 400 wakeup_task = NULL; 401 } 402 403 static void wakeup_reset(struct trace_array *tr) 404 { 405 unsigned long flags; 406 407 tracing_reset_online_cpus(tr); 408 409 local_irq_save(flags); 410 arch_spin_lock(&wakeup_lock); 411 __wakeup_reset(tr); 412 arch_spin_unlock(&wakeup_lock); 413 local_irq_restore(flags); 414 } 415 416 static void 417 probe_wakeup(void *ignore, struct task_struct *p, int success) 418 { 419 struct trace_array_cpu *data; 420 int cpu = smp_processor_id(); 421 unsigned long flags; 422 long disabled; 423 int pc; 424 425 if (likely(!tracer_enabled)) 426 return; 427 428 tracing_record_cmdline(p); 429 tracing_record_cmdline(current); 430 431 if ((wakeup_rt && !rt_task(p)) || 432 p->prio >= wakeup_prio || 433 p->prio >= current->prio) 434 return; 435 436 pc = preempt_count(); 437 disabled = atomic_inc_return(&wakeup_trace->data[cpu]->disabled); 438 if (unlikely(disabled != 1)) 439 goto out; 440 441 /* interrupts should be off from try_to_wake_up */ 442 arch_spin_lock(&wakeup_lock); 443 444 /* check for races. */ 445 if (!tracer_enabled || p->prio >= wakeup_prio) 446 goto out_locked; 447 448 /* reset the trace */ 449 __wakeup_reset(wakeup_trace); 450 451 wakeup_cpu = task_cpu(p); 452 wakeup_current_cpu = wakeup_cpu; 453 wakeup_prio = p->prio; 454 455 wakeup_task = p; 456 get_task_struct(wakeup_task); 457 458 local_save_flags(flags); 459 460 data = wakeup_trace->data[wakeup_cpu]; 461 data->preempt_timestamp = ftrace_now(cpu); 462 tracing_sched_wakeup_trace(wakeup_trace, p, current, flags, pc); 463 464 /* 465 * We must be careful in using CALLER_ADDR2. But since wake_up 466 * is not called by an assembly function (where as schedule is) 467 * it should be safe to use it here. 468 */ 469 __trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); 470 471 out_locked: 472 arch_spin_unlock(&wakeup_lock); 473 out: 474 atomic_dec(&wakeup_trace->data[cpu]->disabled); 475 } 476 477 static void start_wakeup_tracer(struct trace_array *tr) 478 { 479 int ret; 480 481 ret = register_trace_sched_wakeup(probe_wakeup, NULL); 482 if (ret) { 483 pr_info("wakeup trace: Couldn't activate tracepoint" 484 " probe to kernel_sched_wakeup\n"); 485 return; 486 } 487 488 ret = register_trace_sched_wakeup_new(probe_wakeup, NULL); 489 if (ret) { 490 pr_info("wakeup trace: Couldn't activate tracepoint" 491 " probe to kernel_sched_wakeup_new\n"); 492 goto fail_deprobe; 493 } 494 495 ret = register_trace_sched_switch(probe_wakeup_sched_switch, NULL); 496 if (ret) { 497 pr_info("sched trace: Couldn't activate tracepoint" 498 " probe to kernel_sched_switch\n"); 499 goto fail_deprobe_wake_new; 500 } 501 502 ret = register_trace_sched_migrate_task(probe_wakeup_migrate_task, NULL); 503 if (ret) { 504 pr_info("wakeup trace: Couldn't activate tracepoint" 505 " probe to kernel_sched_migrate_task\n"); 506 return; 507 } 508 509 wakeup_reset(tr); 510 511 /* 512 * Don't let the tracer_enabled = 1 show up before 513 * the wakeup_task is reset. This may be overkill since 514 * wakeup_reset does a spin_unlock after setting the 515 * wakeup_task to NULL, but I want to be safe. 516 * This is a slow path anyway. 517 */ 518 smp_wmb(); 519 520 if (start_func_tracer(is_graph())) 521 printk(KERN_ERR "failed to start wakeup tracer\n"); 522 523 return; 524 fail_deprobe_wake_new: 525 unregister_trace_sched_wakeup_new(probe_wakeup, NULL); 526 fail_deprobe: 527 unregister_trace_sched_wakeup(probe_wakeup, NULL); 528 } 529 530 static void stop_wakeup_tracer(struct trace_array *tr) 531 { 532 tracer_enabled = 0; 533 stop_func_tracer(is_graph()); 534 unregister_trace_sched_switch(probe_wakeup_sched_switch, NULL); 535 unregister_trace_sched_wakeup_new(probe_wakeup, NULL); 536 unregister_trace_sched_wakeup(probe_wakeup, NULL); 537 unregister_trace_sched_migrate_task(probe_wakeup_migrate_task, NULL); 538 } 539 540 static int __wakeup_tracer_init(struct trace_array *tr) 541 { 542 save_lat_flag = trace_flags & TRACE_ITER_LATENCY_FMT; 543 trace_flags |= TRACE_ITER_LATENCY_FMT; 544 545 tracing_max_latency = 0; 546 wakeup_trace = tr; 547 start_wakeup_tracer(tr); 548 return 0; 549 } 550 551 static int wakeup_tracer_init(struct trace_array *tr) 552 { 553 wakeup_rt = 0; 554 return __wakeup_tracer_init(tr); 555 } 556 557 static int wakeup_rt_tracer_init(struct trace_array *tr) 558 { 559 wakeup_rt = 1; 560 return __wakeup_tracer_init(tr); 561 } 562 563 static void wakeup_tracer_reset(struct trace_array *tr) 564 { 565 stop_wakeup_tracer(tr); 566 /* make sure we put back any tasks we are tracing */ 567 wakeup_reset(tr); 568 569 if (!save_lat_flag) 570 trace_flags &= ~TRACE_ITER_LATENCY_FMT; 571 } 572 573 static void wakeup_tracer_start(struct trace_array *tr) 574 { 575 wakeup_reset(tr); 576 tracer_enabled = 1; 577 } 578 579 static void wakeup_tracer_stop(struct trace_array *tr) 580 { 581 tracer_enabled = 0; 582 } 583 584 static struct tracer wakeup_tracer __read_mostly = 585 { 586 .name = "wakeup", 587 .init = wakeup_tracer_init, 588 .reset = wakeup_tracer_reset, 589 .start = wakeup_tracer_start, 590 .stop = wakeup_tracer_stop, 591 .print_max = 1, 592 .print_header = wakeup_print_header, 593 .print_line = wakeup_print_line, 594 .flags = &tracer_flags, 595 .set_flag = wakeup_set_flag, 596 #ifdef CONFIG_FTRACE_SELFTEST 597 .selftest = trace_selftest_startup_wakeup, 598 #endif 599 .open = wakeup_trace_open, 600 .close = wakeup_trace_close, 601 .use_max_tr = 1, 602 }; 603 604 static struct tracer wakeup_rt_tracer __read_mostly = 605 { 606 .name = "wakeup_rt", 607 .init = wakeup_rt_tracer_init, 608 .reset = wakeup_tracer_reset, 609 .start = wakeup_tracer_start, 610 .stop = wakeup_tracer_stop, 611 .wait_pipe = poll_wait_pipe, 612 .print_max = 1, 613 .print_header = wakeup_print_header, 614 .print_line = wakeup_print_line, 615 .flags = &tracer_flags, 616 .set_flag = wakeup_set_flag, 617 #ifdef CONFIG_FTRACE_SELFTEST 618 .selftest = trace_selftest_startup_wakeup, 619 #endif 620 .open = wakeup_trace_open, 621 .close = wakeup_trace_close, 622 .use_max_tr = 1, 623 }; 624 625 __init static int init_wakeup_tracer(void) 626 { 627 int ret; 628 629 ret = register_tracer(&wakeup_tracer); 630 if (ret) 631 return ret; 632 633 ret = register_tracer(&wakeup_rt_tracer); 634 if (ret) 635 return ret; 636 637 return 0; 638 } 639 device_initcall(init_wakeup_tracer); 640