// SPDX-FileCopyrightText: 2008 Paul E. McKenney, IBM Corporation. // // SPDX-License-Identifier: GPL-2.0-or-later /* * rcutorture.h: simple user-level performance/stress test of RCU. * * Usage: * ./rcu rperf [ ] * Run a read-side performance test with the specified * number of readers spaced by . * Thus "./rcu 16 rperf 2" would run 16 readers on even-numbered * CPUs from 0 to 30. * ./rcu uperf [ ] * Run an update-side performance test with the specified * number of updaters and specified CPU spacing. * ./rcu perf [ ] * Run a combined read/update performance test with the specified * number of readers and one updater and specified CPU spacing. * The readers run on the low-numbered CPUs and the updater * of the highest-numbered CPU. * * The above tests produce output as follows: * * n_reads: 46008000 n_updates: 146026 nreaders: 2 nupdaters: 1 duration: 1 * ns/read: 43.4707 ns/update: 6848.1 * * The first line lists the total number of RCU reads and updates executed * during the test, the number of reader threads, the number of updater * threads, and the duration of the test in seconds. The second line * lists the average duration of each type of operation in nanoseconds, * or "nan" if the corresponding type of operation was not performed. * * ./rcu stress * Run a stress test with the specified number of readers and * one updater. None of the threads are affinitied to any * particular CPU. * * This test produces output as follows: * * n_reads: 114633217 n_updates: 3903415 n_mberror: 0 * rcu_stress_count: 114618391 14826 0 0 0 0 0 0 0 0 0 * * The first line lists the number of RCU read and update operations * executed, followed by the number of memory-ordering violations * (which will be zero in a correct RCU implementation). The second * line lists the number of readers observing progressively more stale * data. A correct RCU implementation will have all but the first two * numbers non-zero. * * rcu_stress_count: Histogram of "ages" of structures seen by readers. If any * entries past the first two are non-zero, RCU is broken. The age of a newly * allocated structure is zero, it becomes one when removed from reader * visibility, and is incremented once per grace period subsequently -- and is * freed after passing through (RCU_STRESS_PIPE_LEN-2) grace periods. Since * this tests only has one true writer (there are fake writers), only buckets at * indexes 0 and 1 should be none-zero. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. * * Copyright (c) 2008 Paul E. McKenney, IBM Corporation. */ /* * Test variables. */ #include #include "tap.h" #include #include "urcu-wait.h" #define NR_TESTS 1 DEFINE_PER_THREAD(long long, n_reads_pt); DEFINE_PER_THREAD(long long, n_updates_pt); enum callrcu_type { CALLRCU_GLOBAL, CALLRCU_PERCPU, CALLRCU_PERTHREAD, }; enum writer_state { WRITER_STATE_SYNC_RCU, WRITER_STATE_CALL_RCU, WRITER_STATE_POLL_RCU, }; static enum callrcu_type callrcu_type = CALLRCU_GLOBAL; long long n_reads = 0LL; long n_updates = 0L; int nthreadsrunning; char argsbuf[64]; #define GOFLAG_INIT 0 #define GOFLAG_RUN 1 #define GOFLAG_STOP 2 volatile int goflag __attribute__((__aligned__(CAA_CACHE_LINE_SIZE))) = GOFLAG_INIT; #define RCU_READ_RUN 1000 //MD #define RCU_READ_NESTABLE #ifdef RCU_READ_NESTABLE #define rcu_read_lock_nest() rcu_read_lock() #define rcu_read_unlock_nest() rcu_read_unlock() #else /* #ifdef RCU_READ_NESTABLE */ #define rcu_read_lock_nest() #define rcu_read_unlock_nest() #endif /* #else #ifdef RCU_READ_NESTABLE */ #ifdef TORTURE_QSBR #define mark_rcu_quiescent_state rcu_quiescent_state #define put_thread_offline rcu_thread_offline #define put_thread_online rcu_thread_online #endif #ifndef mark_rcu_quiescent_state #define mark_rcu_quiescent_state() do {} while (0) #endif /* #ifdef mark_rcu_quiescent_state */ #ifndef put_thread_offline #define put_thread_offline() do {} while (0) #define put_thread_online() do {} while (0) #define put_thread_online_delay() do {} while (0) #else /* #ifndef put_thread_offline */ #define put_thread_online_delay() synchronize_rcu() #endif /* #else #ifndef put_thread_offline */ /* * Performance test. */ static void *rcu_read_perf_test(void *arg) { int i; int me = (long)arg; long long n_reads_local = 0; rcu_register_thread(); run_on(me); uatomic_inc(&nthreadsrunning); put_thread_offline(); while (uatomic_read(&goflag) == GOFLAG_INIT) (void) poll(NULL, 0, 1); put_thread_online(); while (uatomic_read(&goflag) == GOFLAG_RUN) { for (i = 0; i < RCU_READ_RUN; i++) { rcu_read_lock(); /* rcu_read_lock_nest(); */ /* rcu_read_unlock_nest(); */ rcu_read_unlock(); } n_reads_local += RCU_READ_RUN; mark_rcu_quiescent_state(); } __get_thread_var(n_reads_pt) += n_reads_local; put_thread_offline(); rcu_unregister_thread(); return (NULL); } static void *rcu_update_perf_test(void *arg __attribute__((unused))) { long long n_updates_local = 0; if (callrcu_type == CALLRCU_PERTHREAD) { struct call_rcu_data *crdp; crdp = create_call_rcu_data(0, -1); if (crdp != NULL) { diag("Successfully using per-thread call_rcu() worker."); set_thread_call_rcu_data(crdp); } } uatomic_inc(&nthreadsrunning); while (uatomic_read(&goflag) == GOFLAG_INIT) (void) poll(NULL, 0, 1); while (uatomic_read(&goflag) == GOFLAG_RUN) { synchronize_rcu(); n_updates_local++; } __get_thread_var(n_updates_pt) += n_updates_local; if (callrcu_type == CALLRCU_PERTHREAD) { struct call_rcu_data *crdp; crdp = get_thread_call_rcu_data(); set_thread_call_rcu_data(NULL); call_rcu_data_free(crdp); } return NULL; } static void perftestinit(void) { init_per_thread(n_reads_pt, 0LL); init_per_thread(n_updates_pt, 0LL); uatomic_set(&nthreadsrunning, 0); } static int perftestrun(int nthreads, int nreaders, int nupdaters) { int t; int duration = 1; while (uatomic_read(&nthreadsrunning) < nthreads) (void) poll(NULL, 0, 1); uatomic_set(&goflag, GOFLAG_RUN); sleep(duration); uatomic_set(&goflag, GOFLAG_STOP); wait_all_threads(); for_each_thread(t) { n_reads += per_thread(n_reads_pt, t); n_updates += per_thread(n_updates_pt, t); } diag("n_reads: %lld n_updates: %ld nreaders: %d nupdaters: %d duration: %d", n_reads, n_updates, nreaders, nupdaters, duration); diag("ns/read: %g ns/update: %g", ((duration * 1000*1000*1000.*(double)nreaders) / (double)n_reads), ((duration * 1000*1000*1000.*(double)nupdaters) / (double)n_updates)); if (get_cpu_call_rcu_data(0)) { diag("Deallocating per-CPU call_rcu threads.\n"); free_all_cpu_call_rcu_data(); } return 0; } static int perftest(int nreaders, int cpustride) { int i; long arg; perftestinit(); for (i = 0; i < nreaders; i++) { arg = (long)(i * cpustride); create_thread(rcu_read_perf_test, (void *)arg); } arg = (long)(i * cpustride); create_thread(rcu_update_perf_test, (void *)arg); return perftestrun(i + 1, nreaders, 1); } static int rperftest(int nreaders, int cpustride) { int i; long arg; perftestinit(); init_per_thread(n_reads_pt, 0LL); for (i = 0; i < nreaders; i++) { arg = (long)(i * cpustride); create_thread(rcu_read_perf_test, (void *)arg); } return perftestrun(i, nreaders, 0); } static int uperftest(int nupdaters, int cpustride) { int i; long arg; perftestinit(); init_per_thread(n_reads_pt, 0LL); for (i = 0; i < nupdaters; i++) { arg = (long)(i * cpustride); create_thread(rcu_update_perf_test, (void *)arg); } return perftestrun(i, 0, nupdaters); } /* * Stress test. */ #define RCU_STRESS_PIPE_LEN 10 struct rcu_stress { int pipe_count; int mbtest; }; struct rcu_stress rcu_stress_array[RCU_STRESS_PIPE_LEN] = { { 0, 0 } }; struct rcu_stress *rcu_stress_current; int rcu_stress_idx = 0; /* * How many time a reader has seen something that should not be visible. It is * an error if this value is different than zero at the end of the stress test. * * Here, the something that should not be visibile is an old pipe that has been * freed (mbtest = 0). */ int n_mberror = 0; DEFINE_PER_THREAD(long long [RCU_STRESS_PIPE_LEN + 1], rcu_stress_count); int garbage = 0; static void *rcu_read_stress_test(void *arg __attribute__((unused))) { int i; int itercnt = 0; struct rcu_stress *p; int pc; rcu_register_thread(); put_thread_offline(); while (uatomic_read(&goflag) == GOFLAG_INIT) (void) poll(NULL, 0, 1); put_thread_online(); while (uatomic_read(&goflag) == GOFLAG_RUN) { rcu_read_lock(); p = rcu_dereference(rcu_stress_current); if (p->mbtest == 0) uatomic_inc_mo(&n_mberror, CMM_RELAXED); rcu_read_lock_nest(); /* * The value of garbage is nothing important. This is * essentially a busy loop. The atomic operation -- while not * important here -- helps tools such as TSAN to not flag this * as a race condition. */ for (i = 0; i < 100; i++) uatomic_inc(&garbage); rcu_read_unlock_nest(); pc = uatomic_read(&p->pipe_count); rcu_read_unlock(); if ((pc > RCU_STRESS_PIPE_LEN) || (pc < 0)) pc = RCU_STRESS_PIPE_LEN; __get_thread_var(rcu_stress_count)[pc]++; __get_thread_var(n_reads_pt)++; mark_rcu_quiescent_state(); if ((++itercnt % 0x1000) == 0) { put_thread_offline(); put_thread_online_delay(); put_thread_online(); } } put_thread_offline(); rcu_unregister_thread(); return (NULL); } static DEFINE_URCU_WAIT_QUEUE(call_rcu_waiters); static void rcu_update_stress_test_rcu(struct rcu_head *head __attribute__((unused))) { struct urcu_waiters waiters; urcu_move_waiters(&waiters, &call_rcu_waiters); urcu_wake_all_waiters(&waiters); } static void advance_writer_state(enum writer_state *state) { switch (*state) { case WRITER_STATE_SYNC_RCU: *state = WRITER_STATE_CALL_RCU; break; case WRITER_STATE_CALL_RCU: *state = WRITER_STATE_POLL_RCU; break; case WRITER_STATE_POLL_RCU: *state = WRITER_STATE_SYNC_RCU; break; } } static void *rcu_update_stress_test(void *arg __attribute__((unused))) { int i; struct rcu_stress *p, *old_p; struct rcu_head rh; enum writer_state writer_state = WRITER_STATE_SYNC_RCU; rcu_register_thread(); /* Offline for poll. */ put_thread_offline(); while (uatomic_read(&goflag) == GOFLAG_INIT) (void) poll(NULL, 0, 1); put_thread_online(); old_p = NULL; while (uatomic_read(&goflag) == GOFLAG_RUN) { i = rcu_stress_idx + 1; if (i >= RCU_STRESS_PIPE_LEN) i = 0; rcu_read_lock(); old_p = rcu_dereference(rcu_stress_current); rcu_read_unlock(); /* * Allocate a new pipe. */ p = &rcu_stress_array[i]; p->pipe_count = 0; p->mbtest = 1; rcu_assign_pointer(rcu_stress_current, p); rcu_stress_idx = i; /* * Increment every pipe except the freshly allocated one. A * reader should only see either the old pipe or the new * pipe. This is reflected in the rcu_stress_count histogram. */ for (i = 0; i < RCU_STRESS_PIPE_LEN; i++) if (i != rcu_stress_idx) uatomic_inc(&rcu_stress_array[i].pipe_count); switch (writer_state) { case WRITER_STATE_SYNC_RCU: synchronize_rcu(); break; case WRITER_STATE_CALL_RCU: { DEFINE_URCU_WAIT_NODE(wait, URCU_WAIT_WAITING); urcu_wait_add(&call_rcu_waiters, &wait); call_rcu(&rh, rcu_update_stress_test_rcu); /* Offline for busy-wait. */ put_thread_offline(); urcu_adaptative_busy_wait(&wait); put_thread_online(); break; } case WRITER_STATE_POLL_RCU: { struct urcu_gp_poll_state poll_state; poll_state = start_poll_synchronize_rcu(); /* Offline for poll. */ put_thread_offline(); while (!poll_state_synchronize_rcu(poll_state)) (void) poll(NULL, 0, 1); /* Wait for 1ms */ put_thread_online(); break; } } /* * No readers should see that old pipe now. Setting mbtest to 0 * to mark it as "freed". */ if (old_p) { old_p->mbtest = 0; } old_p = p; n_updates++; advance_writer_state(&writer_state); } rcu_unregister_thread(); return NULL; } static void *rcu_fake_update_stress_test(void *arg __attribute__((unused))) { if (callrcu_type == CALLRCU_PERTHREAD) { struct call_rcu_data *crdp; crdp = create_call_rcu_data(0, -1); if (crdp != NULL) { diag("Successfully using per-thread call_rcu() worker."); set_thread_call_rcu_data(crdp); } } while (uatomic_read(&goflag) == GOFLAG_INIT) (void) poll(NULL, 0, 1); while (uatomic_read(&goflag) == GOFLAG_RUN) { synchronize_rcu(); (void) poll(NULL, 0, 1); } if (callrcu_type == CALLRCU_PERTHREAD) { struct call_rcu_data *crdp; crdp = get_thread_call_rcu_data(); set_thread_call_rcu_data(NULL); call_rcu_data_free(crdp); } return NULL; } static int stresstest(int nreaders) { int i; int t; long long *p; long long sum; int ret; init_per_thread(n_reads_pt, 0LL); for_each_thread(t) { p = &per_thread(rcu_stress_count,t)[0]; for (i = 0; i <= RCU_STRESS_PIPE_LEN; i++) p[i] = 0LL; } rcu_stress_current = &rcu_stress_array[0]; rcu_stress_current->pipe_count = 0; rcu_stress_current->mbtest = 1; for (i = 0; i < nreaders; i++) create_thread(rcu_read_stress_test, NULL); create_thread(rcu_update_stress_test, NULL); for (i = 0; i < 5; i++) create_thread(rcu_fake_update_stress_test, NULL); uatomic_set(&goflag, GOFLAG_RUN); sleep(10); uatomic_set(&goflag, GOFLAG_STOP); wait_all_threads(); for_each_thread(t) n_reads += per_thread(n_reads_pt, t); diag("n_reads: %lld n_updates: %ld n_mberror: %d", n_reads, n_updates, n_mberror); rdiag_start(); rdiag("rcu_stress_count:"); ret = 0; for (i = 0; i <= RCU_STRESS_PIPE_LEN; i++) { sum = 0LL; for_each_thread(t) { sum += per_thread(rcu_stress_count, t)[i]; } /* * If any entries past the first two are non-zero, RCU is * broken. See details above about rcu_stress_count. */ if (i > 1 && sum != 0) { ret = -1; } rdiag(" %lld", sum); } rdiag_end(); if (get_cpu_call_rcu_data(0)) { diag("Deallocating per-CPU call_rcu threads."); free_all_cpu_call_rcu_data(); } if (n_mberror) ret = -1; return ret; } /* * Mainprogram. */ static void usage(char *argv[]) __attribute__((__noreturn__)); static void usage(char *argv[]) { diag("Usage: %s nreaders [ perf | rperf | uperf | stress ] [ stride ] [ callrcu_global | callrcu_percpu | callrcu_perthread ]\n", argv[0]); exit(-1); } int main(int argc, char *argv[]) { int nreaders = 1; int cpustride = 1; plan_tests(NR_TESTS); smp_init(); //rcu_init(); if (argc > 4) { const char *callrcu_str = argv[4];; if (strcmp(callrcu_str, "callrcu_global") == 0) { callrcu_type = CALLRCU_GLOBAL; } else if (strcmp(callrcu_str, "callrcu_percpu") == 0) { callrcu_type = CALLRCU_PERCPU; } else if (strcmp(callrcu_str, "callrcu_perthread") == 0) { callrcu_type = CALLRCU_PERTHREAD; } else { usage(argv); goto end; } } switch (callrcu_type) { case CALLRCU_GLOBAL: diag("Using global per-process call_rcu thread."); break; case CALLRCU_PERCPU: diag("Using per-CPU call_rcu threads."); if (create_all_cpu_call_rcu_data(0)) diag("create_all_cpu_call_rcu_data: %s", strerror(errno)); break; case CALLRCU_PERTHREAD: diag("Using per-thread call_rcu() worker."); break; default: abort(); } #ifdef DEBUG_YIELD yield_active |= YIELD_READ; yield_active |= YIELD_WRITE; #endif if (argc > 1) { if (strcmp(argv[1], "-h") == 0 || strcmp(argv[1], "--help") == 0) { usage(argv); goto end; } nreaders = strtoul(argv[1], NULL, 0); if (argc == 2) { ok(!perftest(nreaders, cpustride), "perftest readers: %d, stride: %d", nreaders, cpustride); goto end; } if (argc > 3) cpustride = strtoul(argv[3], NULL, 0); if (strcmp(argv[2], "perf") == 0) ok(!perftest(nreaders, cpustride), "perftest readers: %d, stride: %d", nreaders, cpustride); else if (strcmp(argv[2], "rperf") == 0) ok(!rperftest(nreaders, cpustride), "rperftest readers: %d, stride: %d", nreaders, cpustride); else if (strcmp(argv[2], "uperf") == 0) ok(!uperftest(nreaders, cpustride), "uperftest readers: %d, stride: %d", nreaders, cpustride); else if (strcmp(argv[2], "stress") == 0) ok(!stresstest(nreaders), "stresstest readers: %d, stride: %d", nreaders, cpustride); else usage(argv); } else { usage(argv); } end: return exit_status(); }