diff options
Diffstat (limited to 'tools/testing/selftests/rseq/param_test.c')
| -rw-r--r-- | tools/testing/selftests/rseq/param_test.c | 407 |
1 files changed, 356 insertions, 51 deletions
diff --git a/tools/testing/selftests/rseq/param_test.c b/tools/testing/selftests/rseq/param_test.c index e8a657a5f48a..2f37961240ca 100644 --- a/tools/testing/selftests/rseq/param_test.c +++ b/tools/testing/selftests/rseq/param_test.c @@ -1,8 +1,10 @@ // SPDX-License-Identifier: LGPL-2.1 #define _GNU_SOURCE #include <assert.h> +#include <linux/membarrier.h> #include <pthread.h> #include <sched.h> +#include <stdatomic.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> @@ -14,6 +16,7 @@ #include <signal.h> #include <errno.h> #include <stddef.h> +#include <stdbool.h> static inline pid_t rseq_gettid(void) { @@ -34,13 +37,9 @@ static int opt_modulo, verbose; static int opt_yield, opt_signal, opt_sleep, opt_disable_rseq, opt_threads = 200, - opt_disable_mod = 0, opt_test = 's', opt_mb = 0; + opt_disable_mod = 0, opt_test = 's'; -#ifndef RSEQ_SKIP_FASTPATH static long long opt_reps = 5000; -#else -static long long opt_reps = 100; -#endif static __thread __attribute__((tls_model("initial-exec"))) unsigned int signals_delivered; @@ -159,7 +158,7 @@ unsigned int yield_mod_cnt, nr_abort; " cbnz " INJECT_ASM_REG ", 222b\n" \ "333:\n" -#elif __PPC__ +#elif defined(__PPC__) #define RSEQ_INJECT_INPUT \ , [loop_cnt_1]"m"(loop_cnt[1]) \ @@ -205,6 +204,29 @@ unsigned int yield_mod_cnt, nr_abort; "addiu " INJECT_ASM_REG ", -1\n\t" \ "bnez " INJECT_ASM_REG ", 222b\n\t" \ "333:\n\t" +#elif defined(__riscv) + +#define RSEQ_INJECT_INPUT \ + , [loop_cnt_1]"m"(loop_cnt[1]) \ + , [loop_cnt_2]"m"(loop_cnt[2]) \ + , [loop_cnt_3]"m"(loop_cnt[3]) \ + , [loop_cnt_4]"m"(loop_cnt[4]) \ + , [loop_cnt_5]"m"(loop_cnt[5]) \ + , [loop_cnt_6]"m"(loop_cnt[6]) + +#define INJECT_ASM_REG "t1" + +#define RSEQ_INJECT_CLOBBER \ + , INJECT_ASM_REG + +#define RSEQ_INJECT_ASM(n) \ + "lw " INJECT_ASM_REG ", %[loop_cnt_" #n "]\n\t" \ + "beqz " INJECT_ASM_REG ", 333f\n\t" \ + "222:\n\t" \ + "addi " INJECT_ASM_REG "," INJECT_ASM_REG ", -1\n\t" \ + "bnez " INJECT_ASM_REG ", 222b\n\t" \ + "333:\n\t" + #else #error unsupported target @@ -243,6 +265,73 @@ unsigned int yield_mod_cnt, nr_abort; #include "rseq.h" +static enum rseq_mo opt_mo = RSEQ_MO_RELAXED; + +#ifdef RSEQ_ARCH_HAS_OFFSET_DEREF_ADDV +#define TEST_MEMBARRIER + +static int sys_membarrier(int cmd, int flags, int cpu_id) +{ + return syscall(__NR_membarrier, cmd, flags, cpu_id); +} +#endif + +#ifdef BUILDOPT_RSEQ_PERCPU_MM_CID +# define RSEQ_PERCPU RSEQ_PERCPU_MM_CID +static +int get_current_cpu_id(void) +{ + return rseq_current_mm_cid(); +} +static +bool rseq_validate_cpu_id(void) +{ + return rseq_mm_cid_available(); +} +static +bool rseq_use_cpu_index(void) +{ + return false; /* Use mm_cid */ +} +# ifdef TEST_MEMBARRIER +/* + * Membarrier does not currently support targeting a mm_cid, so + * issue the barrier on all cpus. + */ +static +int rseq_membarrier_expedited(int cpu) +{ + return sys_membarrier(MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ, + 0, 0); +} +# endif /* TEST_MEMBARRIER */ +#else +# define RSEQ_PERCPU RSEQ_PERCPU_CPU_ID +static +int get_current_cpu_id(void) +{ + return rseq_cpu_start(); +} +static +bool rseq_validate_cpu_id(void) +{ + return rseq_current_cpu_raw() >= 0; +} +static +bool rseq_use_cpu_index(void) +{ + return true; /* Use cpu_id as index. */ +} +# ifdef TEST_MEMBARRIER +static +int rseq_membarrier_expedited(int cpu) +{ + return sys_membarrier(MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ, + MEMBARRIER_CMD_FLAG_CPU, cpu); +} +# endif /* TEST_MEMBARRIER */ +#endif + struct percpu_lock_entry { intptr_t v; } __attribute__((aligned(128))); @@ -330,8 +419,14 @@ static int rseq_this_cpu_lock(struct percpu_lock *lock) for (;;) { int ret; - cpu = rseq_cpu_start(); - ret = rseq_cmpeqv_storev(&lock->c[cpu].v, + cpu = get_current_cpu_id(); + if (cpu < 0) { + fprintf(stderr, "pid: %d: tid: %d, cpu: %d: cid: %d\n", + getpid(), (int) rseq_gettid(), rseq_current_cpu_raw(), cpu); + abort(); + } + ret = rseq_cmpeqv_storev(RSEQ_MO_RELAXED, RSEQ_PERCPU, + &lock->c[cpu].v, 0, 1, cpu); if (rseq_likely(!ret)) break; @@ -366,9 +461,7 @@ void *test_percpu_spinlock_thread(void *arg) abort(); reps = thread_data->reps; for (i = 0; i < reps; i++) { - int cpu = rseq_cpu_start(); - - cpu = rseq_this_cpu_lock(&data->lock); + int cpu = rseq_this_cpu_lock(&data->lock); data->c[cpu].count++; rseq_percpu_unlock(&data->lock, cpu); #ifndef BENCHMARK @@ -450,8 +543,9 @@ void *test_percpu_inc_thread(void *arg) do { int cpu; - cpu = rseq_cpu_start(); - ret = rseq_addv(&data->c[cpu].count, 1, cpu); + cpu = get_current_cpu_id(); + ret = rseq_addv(RSEQ_MO_RELAXED, RSEQ_PERCPU, + &data->c[cpu].count, 1, cpu); } while (rseq_unlikely(ret)); #ifndef BENCHMARK if (i != 0 && !(i % (reps / 10))) @@ -520,13 +614,14 @@ void this_cpu_list_push(struct percpu_list *list, intptr_t *targetptr, newval, expect; int ret; - cpu = rseq_cpu_start(); + cpu = get_current_cpu_id(); /* Load list->c[cpu].head with single-copy atomicity. */ expect = (intptr_t)RSEQ_READ_ONCE(list->c[cpu].head); newval = (intptr_t)node; targetptr = (intptr_t *)&list->c[cpu].head; node->next = (struct percpu_list_node *)expect; - ret = rseq_cmpeqv_storev(targetptr, expect, newval, cpu); + ret = rseq_cmpeqv_storev(RSEQ_MO_RELAXED, RSEQ_PERCPU, + targetptr, expect, newval, cpu); if (rseq_likely(!ret)) break; /* Retry if comparison fails or rseq aborts. */ @@ -549,16 +644,17 @@ struct percpu_list_node *this_cpu_list_pop(struct percpu_list *list, for (;;) { struct percpu_list_node *head; intptr_t *targetptr, expectnot, *load; - off_t offset; + long offset; int ret; - cpu = rseq_cpu_start(); + cpu = get_current_cpu_id(); targetptr = (intptr_t *)&list->c[cpu].head; expectnot = (intptr_t)NULL; offset = offsetof(struct percpu_list_node, next); load = (intptr_t *)&head; - ret = rseq_cmpnev_storeoffp_load(targetptr, expectnot, - offset, load, cpu); + ret = rseq_cmpnev_storeoffp_load(RSEQ_MO_RELAXED, RSEQ_PERCPU, + targetptr, expectnot, + offset, load, cpu); if (rseq_likely(!ret)) { node = head; break; @@ -629,7 +725,7 @@ void test_percpu_list(void) /* Generate list entries for every usable cpu. */ sched_getaffinity(0, sizeof(allowed_cpus), &allowed_cpus); for (i = 0; i < CPU_SETSIZE; i++) { - if (!CPU_ISSET(i, &allowed_cpus)) + if (rseq_use_cpu_index() && !CPU_ISSET(i, &allowed_cpus)) continue; for (j = 1; j <= 100; j++) { struct percpu_list_node *node; @@ -666,7 +762,7 @@ void test_percpu_list(void) for (i = 0; i < CPU_SETSIZE; i++) { struct percpu_list_node *node; - if (!CPU_ISSET(i, &allowed_cpus)) + if (rseq_use_cpu_index() && !CPU_ISSET(i, &allowed_cpus)) continue; while ((node = __percpu_list_pop(&list, i))) { @@ -696,7 +792,7 @@ bool this_cpu_buffer_push(struct percpu_buffer *buffer, intptr_t offset; int ret; - cpu = rseq_cpu_start(); + cpu = get_current_cpu_id(); offset = RSEQ_READ_ONCE(buffer->c[cpu].offset); if (offset == buffer->c[cpu].buflen) break; @@ -704,14 +800,9 @@ bool this_cpu_buffer_push(struct percpu_buffer *buffer, targetptr_spec = (intptr_t *)&buffer->c[cpu].array[offset]; newval_final = offset + 1; targetptr_final = &buffer->c[cpu].offset; - if (opt_mb) - ret = rseq_cmpeqv_trystorev_storev_release( - targetptr_final, offset, targetptr_spec, - newval_spec, newval_final, cpu); - else - ret = rseq_cmpeqv_trystorev_storev(targetptr_final, - offset, targetptr_spec, newval_spec, - newval_final, cpu); + ret = rseq_cmpeqv_trystorev_storev(opt_mo, RSEQ_PERCPU, + targetptr_final, offset, targetptr_spec, + newval_spec, newval_final, cpu); if (rseq_likely(!ret)) { result = true; break; @@ -734,7 +825,7 @@ struct percpu_buffer_node *this_cpu_buffer_pop(struct percpu_buffer *buffer, intptr_t offset; int ret; - cpu = rseq_cpu_start(); + cpu = get_current_cpu_id(); /* Load offset with single-copy atomicity. */ offset = RSEQ_READ_ONCE(buffer->c[cpu].offset); if (offset == 0) { @@ -744,7 +835,8 @@ struct percpu_buffer_node *this_cpu_buffer_pop(struct percpu_buffer *buffer, head = RSEQ_READ_ONCE(buffer->c[cpu].array[offset - 1]); newval = offset - 1; targetptr = (intptr_t *)&buffer->c[cpu].offset; - ret = rseq_cmpeqv_cmpeqv_storev(targetptr, offset, + ret = rseq_cmpeqv_cmpeqv_storev(RSEQ_MO_RELAXED, RSEQ_PERCPU, + targetptr, offset, (intptr_t *)&buffer->c[cpu].array[offset - 1], (intptr_t)head, newval, cpu); if (rseq_likely(!ret)) @@ -820,7 +912,7 @@ void test_percpu_buffer(void) /* Generate list entries for every usable cpu. */ sched_getaffinity(0, sizeof(allowed_cpus), &allowed_cpus); for (i = 0; i < CPU_SETSIZE; i++) { - if (!CPU_ISSET(i, &allowed_cpus)) + if (rseq_use_cpu_index() && !CPU_ISSET(i, &allowed_cpus)) continue; /* Worse-case is every item in same CPU. */ buffer.c[i].array = @@ -870,7 +962,7 @@ void test_percpu_buffer(void) for (i = 0; i < CPU_SETSIZE; i++) { struct percpu_buffer_node *node; - if (!CPU_ISSET(i, &allowed_cpus)) + if (rseq_use_cpu_index() && !CPU_ISSET(i, &allowed_cpus)) continue; while ((node = __percpu_buffer_pop(&buffer, i))) { @@ -901,7 +993,7 @@ bool this_cpu_memcpy_buffer_push(struct percpu_memcpy_buffer *buffer, size_t copylen; int ret; - cpu = rseq_cpu_start(); + cpu = get_current_cpu_id(); /* Load offset with single-copy atomicity. */ offset = RSEQ_READ_ONCE(buffer->c[cpu].offset); if (offset == buffer->c[cpu].buflen) @@ -912,15 +1004,11 @@ bool this_cpu_memcpy_buffer_push(struct percpu_memcpy_buffer *buffer, copylen = sizeof(item); newval_final = offset + 1; targetptr_final = &buffer->c[cpu].offset; - if (opt_mb) - ret = rseq_cmpeqv_trymemcpy_storev_release( - targetptr_final, offset, - destptr, srcptr, copylen, - newval_final, cpu); - else - ret = rseq_cmpeqv_trymemcpy_storev(targetptr_final, - offset, destptr, srcptr, copylen, - newval_final, cpu); + ret = rseq_cmpeqv_trymemcpy_storev( + opt_mo, RSEQ_PERCPU, + targetptr_final, offset, + destptr, srcptr, copylen, + newval_final, cpu); if (rseq_likely(!ret)) { result = true; break; @@ -945,7 +1033,7 @@ bool this_cpu_memcpy_buffer_pop(struct percpu_memcpy_buffer *buffer, size_t copylen; int ret; - cpu = rseq_cpu_start(); + cpu = get_current_cpu_id(); /* Load offset with single-copy atomicity. */ offset = RSEQ_READ_ONCE(buffer->c[cpu].offset); if (offset == 0) @@ -956,8 +1044,8 @@ bool this_cpu_memcpy_buffer_pop(struct percpu_memcpy_buffer *buffer, copylen = sizeof(*item); newval_final = offset - 1; targetptr_final = &buffer->c[cpu].offset; - ret = rseq_cmpeqv_trymemcpy_storev(targetptr_final, - offset, destptr, srcptr, copylen, + ret = rseq_cmpeqv_trymemcpy_storev(RSEQ_MO_RELAXED, RSEQ_PERCPU, + targetptr_final, offset, destptr, srcptr, copylen, newval_final, cpu); if (rseq_likely(!ret)) { result = true; @@ -1035,7 +1123,7 @@ void test_percpu_memcpy_buffer(void) /* Generate list entries for every usable cpu. */ sched_getaffinity(0, sizeof(allowed_cpus), &allowed_cpus); for (i = 0; i < CPU_SETSIZE; i++) { - if (!CPU_ISSET(i, &allowed_cpus)) + if (rseq_use_cpu_index() && !CPU_ISSET(i, &allowed_cpus)) continue; /* Worse-case is every item in same CPU. */ buffer.c[i].array = @@ -1082,7 +1170,7 @@ void test_percpu_memcpy_buffer(void) for (i = 0; i < CPU_SETSIZE; i++) { struct percpu_memcpy_buffer_node item; - if (!CPU_ISSET(i, &allowed_cpus)) + if (rseq_use_cpu_index() && !CPU_ISSET(i, &allowed_cpus)) continue; while (__percpu_memcpy_buffer_pop(&buffer, &item, i)) { @@ -1131,6 +1219,214 @@ static int set_signal_handler(void) return ret; } +/* Test MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU membarrier command. */ +#ifdef TEST_MEMBARRIER +struct test_membarrier_thread_args { + int stop; + intptr_t percpu_list_ptr; +}; + +/* Worker threads modify data in their "active" percpu lists. */ +void *test_membarrier_worker_thread(void *arg) +{ + struct test_membarrier_thread_args *args = + (struct test_membarrier_thread_args *)arg; + const int iters = opt_reps; + int i; + + if (rseq_register_current_thread()) { + fprintf(stderr, "Error: rseq_register_current_thread(...) failed(%d): %s\n", + errno, strerror(errno)); + abort(); + } + + /* Wait for initialization. */ + while (!__atomic_load_n(&args->percpu_list_ptr, __ATOMIC_ACQUIRE)) {} + + for (i = 0; i < iters; ++i) { + int ret; + + do { + int cpu = get_current_cpu_id(); + + ret = rseq_offset_deref_addv(RSEQ_MO_RELAXED, RSEQ_PERCPU, + &args->percpu_list_ptr, + sizeof(struct percpu_list_entry) * cpu, 1, cpu); + } while (rseq_unlikely(ret)); + } + + if (rseq_unregister_current_thread()) { + fprintf(stderr, "Error: rseq_unregister_current_thread(...) failed(%d): %s\n", + errno, strerror(errno)); + abort(); + } + return NULL; +} + +void test_membarrier_init_percpu_list(struct percpu_list *list) +{ + int i; + + memset(list, 0, sizeof(*list)); + for (i = 0; i < CPU_SETSIZE; i++) { + struct percpu_list_node *node; + + node = malloc(sizeof(*node)); + assert(node); + node->data = 0; + node->next = NULL; + list->c[i].head = node; + } +} + +void test_membarrier_free_percpu_list(struct percpu_list *list) +{ + int i; + + for (i = 0; i < CPU_SETSIZE; i++) + free(list->c[i].head); +} + +/* + * The manager thread swaps per-cpu lists that worker threads see, + * and validates that there are no unexpected modifications. + */ +void *test_membarrier_manager_thread(void *arg) +{ + struct test_membarrier_thread_args *args = + (struct test_membarrier_thread_args *)arg; + struct percpu_list list_a, list_b; + intptr_t expect_a = 0, expect_b = 0; + int cpu_a = 0, cpu_b = 0; + + if (rseq_register_current_thread()) { + fprintf(stderr, "Error: rseq_register_current_thread(...) failed(%d): %s\n", + errno, strerror(errno)); + abort(); + } + + /* Init lists. */ + test_membarrier_init_percpu_list(&list_a); + test_membarrier_init_percpu_list(&list_b); + + __atomic_store_n(&args->percpu_list_ptr, (intptr_t)&list_a, __ATOMIC_RELEASE); + + while (!__atomic_load_n(&args->stop, __ATOMIC_ACQUIRE)) { + /* list_a is "active". */ + cpu_a = rand() % CPU_SETSIZE; + /* + * As list_b is "inactive", we should never see changes + * to list_b. + */ + if (expect_b != __atomic_load_n(&list_b.c[cpu_b].head->data, __ATOMIC_ACQUIRE)) { + fprintf(stderr, "Membarrier test failed\n"); + abort(); + } + + /* Make list_b "active". */ + __atomic_store_n(&args->percpu_list_ptr, (intptr_t)&list_b, __ATOMIC_RELEASE); + if (rseq_membarrier_expedited(cpu_a) && + errno != ENXIO /* missing CPU */) { + perror("sys_membarrier"); + abort(); + } + /* + * Cpu A should now only modify list_b, so the values + * in list_a should be stable. + */ + expect_a = __atomic_load_n(&list_a.c[cpu_a].head->data, __ATOMIC_ACQUIRE); + + cpu_b = rand() % CPU_SETSIZE; + /* + * As list_a is "inactive", we should never see changes + * to list_a. + */ + if (expect_a != __atomic_load_n(&list_a.c[cpu_a].head->data, __ATOMIC_ACQUIRE)) { + fprintf(stderr, "Membarrier test failed\n"); + abort(); + } + + /* Make list_a "active". */ + __atomic_store_n(&args->percpu_list_ptr, (intptr_t)&list_a, __ATOMIC_RELEASE); + if (rseq_membarrier_expedited(cpu_b) && + errno != ENXIO /* missing CPU*/) { + perror("sys_membarrier"); + abort(); + } + /* Remember a value from list_b. */ + expect_b = __atomic_load_n(&list_b.c[cpu_b].head->data, __ATOMIC_ACQUIRE); + } + + test_membarrier_free_percpu_list(&list_a); + test_membarrier_free_percpu_list(&list_b); + + if (rseq_unregister_current_thread()) { + fprintf(stderr, "Error: rseq_unregister_current_thread(...) failed(%d): %s\n", + errno, strerror(errno)); + abort(); + } + return NULL; +} + +void test_membarrier(void) +{ + const int num_threads = opt_threads; + struct test_membarrier_thread_args thread_args; + pthread_t worker_threads[num_threads]; + pthread_t manager_thread; + int i, ret; + + if (sys_membarrier(MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_RSEQ, 0, 0)) { + perror("sys_membarrier"); + abort(); + } + + thread_args.stop = 0; + thread_args.percpu_list_ptr = 0; + ret = pthread_create(&manager_thread, NULL, + test_membarrier_manager_thread, &thread_args); + if (ret) { + errno = ret; + perror("pthread_create"); + abort(); + } + + for (i = 0; i < num_threads; i++) { + ret = pthread_create(&worker_threads[i], NULL, + test_membarrier_worker_thread, &thread_args); + if (ret) { + errno = ret; + perror("pthread_create"); + abort(); + } + } + + + for (i = 0; i < num_threads; i++) { + ret = pthread_join(worker_threads[i], NULL); + if (ret) { + errno = ret; + perror("pthread_join"); + abort(); + } + } + + __atomic_store_n(&thread_args.stop, 1, __ATOMIC_RELEASE); + ret = pthread_join(manager_thread, NULL); + if (ret) { + errno = ret; + perror("pthread_join"); + abort(); + } +} +#else /* TEST_MEMBARRIER */ +void test_membarrier(void) +{ + fprintf(stderr, "rseq_offset_deref_addv is not implemented on this architecture. " + "Skipping membarrier test.\n"); +} +#endif + static void show_usage(int argc, char **argv) { printf("Usage : %s <OPTIONS>\n", @@ -1153,7 +1449,7 @@ static void show_usage(int argc, char **argv) printf(" [-r N] Number of repetitions per thread (default 5000)\n"); printf(" [-d] Disable rseq system call (no initialization)\n"); printf(" [-D M] Disable rseq for each M threads\n"); - printf(" [-T test] Choose test: (s)pinlock, (l)ist, (b)uffer, (m)emcpy, (i)ncrement\n"); + printf(" [-T test] Choose test: (s)pinlock, (l)ist, (b)uffer, (m)emcpy, (i)ncrement, membarrie(r)\n"); printf(" [-M] Push into buffer and memcpy buffer with memory barriers.\n"); printf(" [-v] Verbose output.\n"); printf(" [-h] Show this help.\n"); @@ -1268,6 +1564,7 @@ int main(int argc, char **argv) case 'i': case 'b': case 'm': + case 'r': break; default: show_usage(argc, argv); @@ -1279,7 +1576,7 @@ int main(int argc, char **argv) verbose = 1; break; case 'M': - opt_mb = 1; + opt_mo = RSEQ_MO_RELEASE; break; default: show_usage(argc, argv); @@ -1299,6 +1596,10 @@ int main(int argc, char **argv) if (!opt_disable_rseq && rseq_register_current_thread()) goto error; + if (!opt_disable_rseq && !rseq_validate_cpu_id()) { + fprintf(stderr, "Error: cpu id getter unavailable\n"); + goto error; + } switch (opt_test) { case 's': printf_verbose("spinlock\n"); @@ -1320,6 +1621,10 @@ int main(int argc, char **argv) printf_verbose("counter increment\n"); test_percpu_inc(); break; + case 'r': + printf_verbose("membarrier\n"); + test_membarrier(); + break; } if (!opt_disable_rseq && rseq_unregister_current_thread()) abort(); |