diff options
Diffstat (limited to 'tools/lib')
26 files changed, 562 insertions, 253 deletions
diff --git a/tools/lib/api/fs/cgroup.c b/tools/lib/api/fs/cgroup.c index 1573dae4259d..250629a09423 100644 --- a/tools/lib/api/fs/cgroup.c +++ b/tools/lib/api/fs/cgroup.c @@ -14,7 +14,7 @@ struct cgroupfs_cache_entry { }; /* just cache last used one */ -static struct cgroupfs_cache_entry cached; +static struct cgroupfs_cache_entry *cached; int cgroupfs_find_mountpoint(char *buf, size_t maxlen, const char *subsys) { @@ -24,9 +24,9 @@ int cgroupfs_find_mountpoint(char *buf, size_t maxlen, const char *subsys) char *p, *path; char mountpoint[PATH_MAX]; - if (!strcmp(cached.subsys, subsys)) { - if (strlen(cached.mountpoint) < maxlen) { - strcpy(buf, cached.mountpoint); + if (cached && !strcmp(cached->subsys, subsys)) { + if (strlen(cached->mountpoint) < maxlen) { + strcpy(buf, cached->mountpoint); return 0; } return -1; @@ -91,8 +91,13 @@ int cgroupfs_find_mountpoint(char *buf, size_t maxlen, const char *subsys) free(line); fclose(fp); - strncpy(cached.subsys, subsys, sizeof(cached.subsys) - 1); - strcpy(cached.mountpoint, mountpoint); + if (!cached) + cached = calloc(1, sizeof(*cached)); + + if (cached) { + strncpy(cached->subsys, subsys, sizeof(cached->subsys) - 1); + strcpy(cached->mountpoint, mountpoint); + } if (mountpoint[0] && strlen(mountpoint) < maxlen) { strcpy(buf, mountpoint); diff --git a/tools/lib/api/fs/fs.c b/tools/lib/api/fs/fs.c index 82f53d81a7a7..5cb0eeec2c8a 100644 --- a/tools/lib/api/fs/fs.c +++ b/tools/lib/api/fs/fs.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 +#include <assert.h> #include <ctype.h> #include <errno.h> #include <limits.h> @@ -10,6 +11,7 @@ #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> +#include <pthread.h> #include <unistd.h> #include <sys/mount.h> @@ -43,7 +45,7 @@ #define BPF_FS_MAGIC 0xcafe4a11 #endif -static const char * const sysfs__fs_known_mountpoints[] = { +static const char * const sysfs__known_mountpoints[] = { "/sys", 0, }; @@ -86,87 +88,89 @@ static const char * const bpf_fs__known_mountpoints[] = { }; struct fs { - const char *name; - const char * const *mounts; - char path[PATH_MAX]; - bool found; - bool checked; - long magic; -}; - -enum { - FS__SYSFS = 0, - FS__PROCFS = 1, - FS__DEBUGFS = 2, - FS__TRACEFS = 3, - FS__HUGETLBFS = 4, - FS__BPF_FS = 5, + const char * const name; + const char * const * const mounts; + char *path; + pthread_mutex_t mount_mutex; + const long magic; }; #ifndef TRACEFS_MAGIC #define TRACEFS_MAGIC 0x74726163 #endif -static struct fs fs__entries[] = { - [FS__SYSFS] = { - .name = "sysfs", - .mounts = sysfs__fs_known_mountpoints, - .magic = SYSFS_MAGIC, - .checked = false, - }, - [FS__PROCFS] = { - .name = "proc", - .mounts = procfs__known_mountpoints, - .magic = PROC_SUPER_MAGIC, - .checked = false, - }, - [FS__DEBUGFS] = { - .name = "debugfs", - .mounts = debugfs__known_mountpoints, - .magic = DEBUGFS_MAGIC, - .checked = false, - }, - [FS__TRACEFS] = { - .name = "tracefs", - .mounts = tracefs__known_mountpoints, - .magic = TRACEFS_MAGIC, - .checked = false, - }, - [FS__HUGETLBFS] = { - .name = "hugetlbfs", - .mounts = hugetlbfs__known_mountpoints, - .magic = HUGETLBFS_MAGIC, - .checked = false, - }, - [FS__BPF_FS] = { - .name = "bpf", - .mounts = bpf_fs__known_mountpoints, - .magic = BPF_FS_MAGIC, - .checked = false, - }, -}; +static void fs__init_once(struct fs *fs); +static const char *fs__mountpoint(const struct fs *fs); +static const char *fs__mount(struct fs *fs); + +#define FS(lower_name, fs_name, upper_name) \ +static struct fs fs__##lower_name = { \ + .name = #fs_name, \ + .mounts = lower_name##__known_mountpoints, \ + .magic = upper_name##_MAGIC, \ + .mount_mutex = PTHREAD_MUTEX_INITIALIZER, \ +}; \ + \ +static void lower_name##_init_once(void) \ +{ \ + struct fs *fs = &fs__##lower_name; \ + \ + fs__init_once(fs); \ +} \ + \ +const char *lower_name##__mountpoint(void) \ +{ \ + static pthread_once_t init_once = PTHREAD_ONCE_INIT; \ + struct fs *fs = &fs__##lower_name; \ + \ + pthread_once(&init_once, lower_name##_init_once); \ + return fs__mountpoint(fs); \ +} \ + \ +const char *lower_name##__mount(void) \ +{ \ + const char *mountpoint = lower_name##__mountpoint(); \ + struct fs *fs = &fs__##lower_name; \ + \ + if (mountpoint) \ + return mountpoint; \ + \ + return fs__mount(fs); \ +} \ + \ +bool lower_name##__configured(void) \ +{ \ + return lower_name##__mountpoint() != NULL; \ +} + +FS(sysfs, sysfs, SYSFS); +FS(procfs, procfs, PROC_SUPER); +FS(debugfs, debugfs, DEBUGFS); +FS(tracefs, tracefs, TRACEFS); +FS(hugetlbfs, hugetlbfs, HUGETLBFS); +FS(bpf_fs, bpf, BPF_FS); static bool fs__read_mounts(struct fs *fs) { - bool found = false; char type[100]; FILE *fp; + char path[PATH_MAX + 1]; fp = fopen("/proc/mounts", "r"); if (fp == NULL) - return NULL; + return false; - while (!found && - fscanf(fp, "%*s %" STR(PATH_MAX) "s %99s %*s %*d %*d\n", - fs->path, type) == 2) { + while (fscanf(fp, "%*s %" STR(PATH_MAX) "s %99s %*s %*d %*d\n", + path, type) == 2) { - if (strcmp(type, fs->name) == 0) - found = true; + if (strcmp(type, fs->name) == 0) { + fs->path = strdup(path); + fclose(fp); + return fs->path != NULL; + } } - fclose(fp); - fs->checked = true; - return fs->found = found; + return false; } static int fs__valid_mount(const char *fs, long magic) @@ -188,8 +192,9 @@ static bool fs__check_mounts(struct fs *fs) ptr = fs->mounts; while (*ptr) { if (fs__valid_mount(*ptr, fs->magic) == 0) { - fs->found = true; - strcpy(fs->path, *ptr); + fs->path = strdup(*ptr); + if (!fs->path) + return false; return true; } ptr++; @@ -227,43 +232,26 @@ static bool fs__env_override(struct fs *fs) if (!override_path) return false; - fs->found = true; - fs->checked = true; - strncpy(fs->path, override_path, sizeof(fs->path) - 1); - fs->path[sizeof(fs->path) - 1] = '\0'; + fs->path = strdup(override_path); + if (!fs->path) + return false; return true; } -static const char *fs__get_mountpoint(struct fs *fs) +static void fs__init_once(struct fs *fs) { - if (fs__env_override(fs)) - return fs->path; - - if (fs__check_mounts(fs)) - return fs->path; - - if (fs__read_mounts(fs)) - return fs->path; - - return NULL; + if (!fs__env_override(fs) && + !fs__check_mounts(fs) && + !fs__read_mounts(fs)) { + assert(!fs->path); + } else { + assert(fs->path); + } } -static const char *fs__mountpoint(int idx) +static const char *fs__mountpoint(const struct fs *fs) { - struct fs *fs = &fs__entries[idx]; - - if (fs->found) - return (const char *)fs->path; - - /* the mount point was already checked for the mount point - * but and did not exist, so return NULL to avoid scanning again. - * This makes the found and not found paths cost equivalent - * in case of multiple calls. - */ - if (fs->checked) - return NULL; - - return fs__get_mountpoint(fs); + return fs->path; } static const char *mount_overload(struct fs *fs) @@ -278,45 +266,29 @@ static const char *mount_overload(struct fs *fs) return getenv(upper_name) ?: *fs->mounts; } -static const char *fs__mount(int idx) +static const char *fs__mount(struct fs *fs) { - struct fs *fs = &fs__entries[idx]; const char *mountpoint; - if (fs__mountpoint(idx)) - return (const char *)fs->path; + pthread_mutex_lock(&fs->mount_mutex); - mountpoint = mount_overload(fs); + /* Check if path found inside the mutex to avoid races with other callers of mount. */ + mountpoint = fs__mountpoint(fs); + if (mountpoint) + goto out; - if (mount(NULL, mountpoint, fs->name, 0, NULL) < 0) - return NULL; - - return fs__check_mounts(fs) ? fs->path : NULL; -} + mountpoint = mount_overload(fs); -#define FS(name, idx) \ -const char *name##__mountpoint(void) \ -{ \ - return fs__mountpoint(idx); \ -} \ - \ -const char *name##__mount(void) \ -{ \ - return fs__mount(idx); \ -} \ - \ -bool name##__configured(void) \ -{ \ - return name##__mountpoint() != NULL; \ + if (mount(NULL, mountpoint, fs->name, 0, NULL) == 0 && + fs__valid_mount(mountpoint, fs->magic) == 0) { + fs->path = strdup(mountpoint); + mountpoint = fs->path; + } +out: + pthread_mutex_unlock(&fs->mount_mutex); + return mountpoint; } -FS(sysfs, FS__SYSFS); -FS(procfs, FS__PROCFS); -FS(debugfs, FS__DEBUGFS); -FS(tracefs, FS__TRACEFS); -FS(hugetlbfs, FS__HUGETLBFS); -FS(bpf_fs, FS__BPF_FS); - int filename__read_int(const char *filename, int *value) { char line[64]; diff --git a/tools/lib/api/fs/tracing_path.c b/tools/lib/api/fs/tracing_path.c index 7ba3e81274e8..30745f35d0d2 100644 --- a/tools/lib/api/fs/tracing_path.c +++ b/tools/lib/api/fs/tracing_path.c @@ -13,17 +13,12 @@ #include "tracing_path.h" -static char tracing_mnt[PATH_MAX] = "/sys/kernel/debug"; static char tracing_path[PATH_MAX] = "/sys/kernel/tracing"; -static char tracing_events_path[PATH_MAX] = "/sys/kernel/tracing/events"; static void __tracing_path_set(const char *tracing, const char *mountpoint) { - snprintf(tracing_mnt, sizeof(tracing_mnt), "%s", mountpoint); snprintf(tracing_path, sizeof(tracing_path), "%s/%s", mountpoint, tracing); - snprintf(tracing_events_path, sizeof(tracing_events_path), "%s/%s%s", - mountpoint, tracing, "events"); } static const char *tracing_path_tracefs_mount(void) @@ -149,15 +144,15 @@ int tracing_path__strerror_open_tp(int err, char *buf, size_t size, /* sdt markers */ if (!strncmp(filename, "sdt_", 4)) { snprintf(buf, size, - "Error:\tFile %s/%s not found.\n" + "Error:\tFile %s/events/%s not found.\n" "Hint:\tSDT event cannot be directly recorded on.\n" "\tPlease first use 'perf probe %s:%s' before recording it.\n", - tracing_events_path, filename, sys, name); + tracing_path, filename, sys, name); } else { snprintf(buf, size, - "Error:\tFile %s/%s not found.\n" + "Error:\tFile %s/events/%s not found.\n" "Hint:\tPerhaps this kernel misses some CONFIG_ setting to enable this feature?.\n", - tracing_events_path, filename); + tracing_path, filename); } break; } @@ -169,9 +164,9 @@ int tracing_path__strerror_open_tp(int err, char *buf, size_t size, break; case EACCES: { snprintf(buf, size, - "Error:\tNo permissions to read %s/%s\n" + "Error:\tNo permissions to read %s/events/%s\n" "Hint:\tTry 'sudo mount -o remount,mode=755 %s'\n", - tracing_events_path, filename, tracing_path_mount()); + tracing_path, filename, tracing_path_mount()); } break; default: diff --git a/tools/lib/api/io.h b/tools/lib/api/io.h index d5e8cf0dada0..9fc429d2852d 100644 --- a/tools/lib/api/io.h +++ b/tools/lib/api/io.h @@ -8,6 +8,7 @@ #define __API_IO__ #include <errno.h> +#include <poll.h> #include <stdlib.h> #include <string.h> #include <unistd.h> @@ -23,6 +24,8 @@ struct io { char *end; /* Currently accessed data pointer. */ char *data; + /* Read timeout, 0 implies no timeout. */ + int timeout_ms; /* Set true on when the end of file on read error. */ bool eof; }; @@ -35,6 +38,7 @@ static inline void io__init(struct io *io, int fd, io->buf = buf; io->end = buf; io->data = buf; + io->timeout_ms = 0; io->eof = false; } @@ -47,7 +51,29 @@ static inline int io__get_char(struct io *io) return -1; if (ptr == io->end) { - ssize_t n = read(io->fd, io->buf, io->buf_len); + ssize_t n; + + if (io->timeout_ms != 0) { + struct pollfd pfds[] = { + { + .fd = io->fd, + .events = POLLIN, + }, + }; + + n = poll(pfds, 1, io->timeout_ms); + if (n == 0) + errno = ETIMEDOUT; + if (n > 0 && !(pfds[0].revents & POLLIN)) { + errno = EIO; + n = -1; + } + if (n <= 0) { + io->eof = true; + return -1; + } + } + n = read(io->fd, io->buf, io->buf_len); if (n <= 0) { io->eof = true; diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c index 128ac723c4ea..ed86b37d8024 100644 --- a/tools/lib/bpf/bpf.c +++ b/tools/lib/bpf/bpf.c @@ -572,20 +572,30 @@ int bpf_map_update_batch(int fd, const void *keys, const void *values, __u32 *co (void *)keys, (void *)values, count, opts); } -int bpf_obj_pin(int fd, const char *pathname) +int bpf_obj_pin_opts(int fd, const char *pathname, const struct bpf_obj_pin_opts *opts) { - const size_t attr_sz = offsetofend(union bpf_attr, file_flags); + const size_t attr_sz = offsetofend(union bpf_attr, path_fd); union bpf_attr attr; int ret; + if (!OPTS_VALID(opts, bpf_obj_pin_opts)) + return libbpf_err(-EINVAL); + memset(&attr, 0, attr_sz); + attr.path_fd = OPTS_GET(opts, path_fd, 0); attr.pathname = ptr_to_u64((void *)pathname); + attr.file_flags = OPTS_GET(opts, file_flags, 0); attr.bpf_fd = fd; ret = sys_bpf(BPF_OBJ_PIN, &attr, attr_sz); return libbpf_err_errno(ret); } +int bpf_obj_pin(int fd, const char *pathname) +{ + return bpf_obj_pin_opts(fd, pathname, NULL); +} + int bpf_obj_get(const char *pathname) { return bpf_obj_get_opts(pathname, NULL); @@ -593,7 +603,7 @@ int bpf_obj_get(const char *pathname) int bpf_obj_get_opts(const char *pathname, const struct bpf_obj_get_opts *opts) { - const size_t attr_sz = offsetofend(union bpf_attr, file_flags); + const size_t attr_sz = offsetofend(union bpf_attr, path_fd); union bpf_attr attr; int fd; @@ -601,6 +611,7 @@ int bpf_obj_get_opts(const char *pathname, const struct bpf_obj_get_opts *opts) return libbpf_err(-EINVAL); memset(&attr, 0, attr_sz); + attr.path_fd = OPTS_GET(opts, path_fd, 0); attr.pathname = ptr_to_u64((void *)pathname); attr.file_flags = OPTS_GET(opts, file_flags, 0); diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h index a2c091389b18..9aa0ee473754 100644 --- a/tools/lib/bpf/bpf.h +++ b/tools/lib/bpf/bpf.h @@ -284,16 +284,30 @@ LIBBPF_API int bpf_map_update_batch(int fd, const void *keys, const void *values __u32 *count, const struct bpf_map_batch_opts *opts); -struct bpf_obj_get_opts { +struct bpf_obj_pin_opts { size_t sz; /* size of this struct for forward/backward compatibility */ __u32 file_flags; + int path_fd; size_t :0; }; -#define bpf_obj_get_opts__last_field file_flags +#define bpf_obj_pin_opts__last_field path_fd LIBBPF_API int bpf_obj_pin(int fd, const char *pathname); +LIBBPF_API int bpf_obj_pin_opts(int fd, const char *pathname, + const struct bpf_obj_pin_opts *opts); + +struct bpf_obj_get_opts { + size_t sz; /* size of this struct for forward/backward compatibility */ + + __u32 file_flags; + int path_fd; + + size_t :0; +}; +#define bpf_obj_get_opts__last_field path_fd + LIBBPF_API int bpf_obj_get(const char *pathname); LIBBPF_API int bpf_obj_get_opts(const char *pathname, const struct bpf_obj_get_opts *opts); diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h index 929a3baca8ef..bbab9ad9dc5a 100644 --- a/tools/lib/bpf/bpf_helpers.h +++ b/tools/lib/bpf/bpf_helpers.h @@ -77,16 +77,21 @@ /* * Helper macros to manipulate data structures */ -#ifndef offsetof -#define offsetof(TYPE, MEMBER) ((unsigned long)&((TYPE *)0)->MEMBER) -#endif -#ifndef container_of + +/* offsetof() definition that uses __builtin_offset() might not preserve field + * offset CO-RE relocation properly, so force-redefine offsetof() using + * old-school approach which works with CO-RE correctly + */ +#undef offsetof +#define offsetof(type, member) ((unsigned long)&((type *)0)->member) + +/* redefined container_of() to ensure we use the above offsetof() macro */ +#undef container_of #define container_of(ptr, type, member) \ ({ \ void *__mptr = (void *)(ptr); \ ((type *)(__mptr - offsetof(type, member))); \ }) -#endif /* * Compiler (optimization) barrier. diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h index 6fb3d0f9af17..be076a4041ab 100644 --- a/tools/lib/bpf/bpf_tracing.h +++ b/tools/lib/bpf/bpf_tracing.h @@ -351,6 +351,7 @@ struct pt_regs___arm64 { * https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#risc-v-calling-conventions */ +/* riscv provides struct user_regs_struct instead of struct pt_regs to userspace */ #define __PT_REGS_CAST(x) ((const struct user_regs_struct *)(x)) #define __PT_PARM1_REG a0 #define __PT_PARM2_REG a1 @@ -383,7 +384,7 @@ struct pt_regs___arm64 { * https://raw.githubusercontent.com/wiki/foss-for-synopsys-dwc-arc-processors/toolchain/files/ARCv2_ABI.pdf */ -/* arc provides struct user_pt_regs instead of struct pt_regs to userspace */ +/* arc provides struct user_regs_struct instead of struct pt_regs to userspace */ #define __PT_REGS_CAST(x) ((const struct user_regs_struct *)(x)) #define __PT_PARM1_REG scratch.r0 #define __PT_PARM2_REG scratch.r1 diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index 0a2c079244b6..8484b563b53d 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -1064,7 +1064,7 @@ static struct btf *btf_parse_raw(const char *path, struct btf *base_btf) int err = 0; long sz; - f = fopen(path, "rb"); + f = fopen(path, "rbe"); if (!f) { err = -errno; goto err_out; diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c index 580985ee5545..4d9f30bf7f01 100644 --- a/tools/lib/bpf/btf_dump.c +++ b/tools/lib/bpf/btf_dump.c @@ -2250,9 +2250,25 @@ static int btf_dump_type_data_check_overflow(struct btf_dump *d, const struct btf_type *t, __u32 id, const void *data, - __u8 bits_offset) + __u8 bits_offset, + __u8 bit_sz) { - __s64 size = btf__resolve_size(d->btf, id); + __s64 size; + + if (bit_sz) { + /* bits_offset is at most 7. bit_sz is at most 128. */ + __u8 nr_bytes = (bits_offset + bit_sz + 7) / 8; + + /* When bit_sz is non zero, it is called from + * btf_dump_struct_data() where it only cares about + * negative error value. + * Return nr_bytes in success case to make it + * consistent as the regular integer case below. + */ + return data + nr_bytes > d->typed_dump->data_end ? -E2BIG : nr_bytes; + } + + size = btf__resolve_size(d->btf, id); if (size < 0 || size >= INT_MAX) { pr_warn("unexpected size [%zu] for id [%u]\n", @@ -2407,7 +2423,7 @@ static int btf_dump_dump_type_data(struct btf_dump *d, { int size, err = 0; - size = btf_dump_type_data_check_overflow(d, t, id, data, bits_offset); + size = btf_dump_type_data_check_overflow(d, t, id, data, bits_offset, bit_sz); if (size < 0) return size; err = btf_dump_type_data_check_zero(d, t, id, data, bits_offset, bit_sz); diff --git a/tools/lib/bpf/gen_loader.c b/tools/lib/bpf/gen_loader.c index 83e8e3bfd8ff..cf3323fd47b8 100644 --- a/tools/lib/bpf/gen_loader.c +++ b/tools/lib/bpf/gen_loader.c @@ -703,17 +703,17 @@ static void emit_relo_kfunc_btf(struct bpf_gen *gen, struct ksym_relo_desc *relo /* obtain fd in BPF_REG_9 */ emit(gen, BPF_MOV64_REG(BPF_REG_9, BPF_REG_7)); emit(gen, BPF_ALU64_IMM(BPF_RSH, BPF_REG_9, 32)); - /* jump to fd_array store if fd denotes module BTF */ - emit(gen, BPF_JMP_IMM(BPF_JNE, BPF_REG_9, 0, 2)); - /* set the default value for off */ - emit(gen, BPF_ST_MEM(BPF_H, BPF_REG_8, offsetof(struct bpf_insn, off), 0)); - /* skip BTF fd store for vmlinux BTF */ - emit(gen, BPF_JMP_IMM(BPF_JA, 0, 0, 4)); /* load fd_array slot pointer */ emit2(gen, BPF_LD_IMM64_RAW_FULL(BPF_REG_0, BPF_PSEUDO_MAP_IDX_VALUE, 0, 0, 0, blob_fd_array_off(gen, btf_fd_idx))); - /* store BTF fd in slot */ + /* store BTF fd in slot, 0 for vmlinux */ emit(gen, BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_9, 0)); + /* jump to insn[insn_idx].off store if fd denotes module BTF */ + emit(gen, BPF_JMP_IMM(BPF_JNE, BPF_REG_9, 0, 2)); + /* set the default value for off */ + emit(gen, BPF_ST_MEM(BPF_H, BPF_REG_8, offsetof(struct bpf_insn, off), 0)); + /* skip BTF fd store for vmlinux BTF */ + emit(gen, BPF_JMP_IMM(BPF_JA, 0, 0, 1)); /* store index into insn[insn_idx].off */ emit(gen, BPF_ST_MEM(BPF_H, BPF_REG_8, offsetof(struct bpf_insn, off), btf_fd_idx)); log: diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index ad1ec893b41b..214f828ece6b 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -117,6 +117,7 @@ static const char * const attach_type_name[] = { [BPF_PERF_EVENT] = "perf_event", [BPF_TRACE_KPROBE_MULTI] = "trace_kprobe_multi", [BPF_STRUCT_OPS] = "struct_ops", + [BPF_NETFILTER] = "netfilter", }; static const char * const link_type_name[] = { @@ -1500,16 +1501,36 @@ static struct bpf_map *bpf_object__add_map(struct bpf_object *obj) return map; } -static size_t bpf_map_mmap_sz(const struct bpf_map *map) +static size_t bpf_map_mmap_sz(unsigned int value_sz, unsigned int max_entries) { - long page_sz = sysconf(_SC_PAGE_SIZE); + const long page_sz = sysconf(_SC_PAGE_SIZE); size_t map_sz; - map_sz = (size_t)roundup(map->def.value_size, 8) * map->def.max_entries; + map_sz = (size_t)roundup(value_sz, 8) * max_entries; map_sz = roundup(map_sz, page_sz); return map_sz; } +static int bpf_map_mmap_resize(struct bpf_map *map, size_t old_sz, size_t new_sz) +{ + void *mmaped; + + if (!map->mmaped) + return -EINVAL; + + if (old_sz == new_sz) + return 0; + + mmaped = mmap(NULL, new_sz, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0); + if (mmaped == MAP_FAILED) + return -errno; + + memcpy(mmaped, map->mmaped, min(old_sz, new_sz)); + munmap(map->mmaped, old_sz); + map->mmaped = mmaped; + return 0; +} + static char *internal_map_name(struct bpf_object *obj, const char *real_name) { char map_name[BPF_OBJ_NAME_LEN], *p; @@ -1608,6 +1629,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, { struct bpf_map_def *def; struct bpf_map *map; + size_t mmap_sz; int err; map = bpf_object__add_map(obj); @@ -1642,7 +1664,8 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n", map->name, map->sec_idx, map->sec_offset, def->map_flags); - map->mmaped = mmap(NULL, bpf_map_mmap_sz(map), PROT_READ | PROT_WRITE, + mmap_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); + map->mmaped = mmap(NULL, mmap_sz, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0); if (map->mmaped == MAP_FAILED) { err = -errno; @@ -4329,7 +4352,7 @@ static int bpf_get_map_info_from_fdinfo(int fd, struct bpf_map_info *info) snprintf(file, sizeof(file), "/proc/%d/fdinfo/%d", getpid(), fd); memset(info, 0, sizeof(*info)); - fp = fopen(file, "r"); + fp = fopen(file, "re"); if (!fp) { err = -errno; pr_warn("failed to open %s: %d. No procfs support?\n", file, @@ -4392,18 +4415,17 @@ int bpf_map__reuse_fd(struct bpf_map *map, int fd) if (!new_name) return libbpf_err(-errno); - new_fd = open("/", O_RDONLY | O_CLOEXEC); + /* + * Like dup(), but make sure new FD is >= 3 and has O_CLOEXEC set. + * This is similar to what we do in ensure_good_fd(), but without + * closing original FD. + */ + new_fd = fcntl(fd, F_DUPFD_CLOEXEC, 3); if (new_fd < 0) { err = -errno; goto err_free_new_name; } - new_fd = dup3(fd, new_fd, O_CLOEXEC); - if (new_fd < 0) { - err = -errno; - goto err_close_new_fd; - } - err = zclose(map->fd); if (err) { err = -errno; @@ -7433,7 +7455,7 @@ int libbpf_kallsyms_parse(kallsyms_cb_t cb, void *ctx) int ret, err = 0; FILE *f; - f = fopen("/proc/kallsyms", "r"); + f = fopen("/proc/kallsyms", "re"); if (!f) { err = -errno; pr_warn("failed to open /proc/kallsyms: %d\n", err); @@ -8294,7 +8316,10 @@ static void bpf_map__destroy(struct bpf_map *map) map->init_slots_sz = 0; if (map->mmaped) { - munmap(map->mmaped, bpf_map_mmap_sz(map)); + size_t mmap_sz; + + mmap_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); + munmap(map->mmaped, mmap_sz); map->mmaped = NULL; } @@ -8712,7 +8737,7 @@ static const struct bpf_sec_def section_defs[] = { SEC_DEF("struct_ops+", STRUCT_OPS, 0, SEC_NONE), SEC_DEF("struct_ops.s+", STRUCT_OPS, 0, SEC_SLEEPABLE), SEC_DEF("sk_lookup", SK_LOOKUP, BPF_SK_LOOKUP, SEC_ATTACHABLE), - SEC_DEF("netfilter", NETFILTER, 0, SEC_NONE), + SEC_DEF("netfilter", NETFILTER, BPF_NETFILTER, SEC_NONE), }; static size_t custom_sec_def_cnt; @@ -9412,10 +9437,103 @@ __u32 bpf_map__value_size(const struct bpf_map *map) return map->def.value_size; } +static int map_btf_datasec_resize(struct bpf_map *map, __u32 size) +{ + struct btf *btf; + struct btf_type *datasec_type, *var_type; + struct btf_var_secinfo *var; + const struct btf_type *array_type; + const struct btf_array *array; + int vlen, element_sz, new_array_id; + __u32 nr_elements; + + /* check btf existence */ + btf = bpf_object__btf(map->obj); + if (!btf) + return -ENOENT; + + /* verify map is datasec */ + datasec_type = btf_type_by_id(btf, bpf_map__btf_value_type_id(map)); + if (!btf_is_datasec(datasec_type)) { + pr_warn("map '%s': cannot be resized, map value type is not a datasec\n", + bpf_map__name(map)); + return -EINVAL; + } + + /* verify datasec has at least one var */ + vlen = btf_vlen(datasec_type); + if (vlen == 0) { + pr_warn("map '%s': cannot be resized, map value datasec is empty\n", + bpf_map__name(map)); + return -EINVAL; + } + + /* verify last var in the datasec is an array */ + var = &btf_var_secinfos(datasec_type)[vlen - 1]; + var_type = btf_type_by_id(btf, var->type); + array_type = skip_mods_and_typedefs(btf, var_type->type, NULL); + if (!btf_is_array(array_type)) { + pr_warn("map '%s': cannot be resized, last var must be an array\n", + bpf_map__name(map)); + return -EINVAL; + } + + /* verify request size aligns with array */ + array = btf_array(array_type); + element_sz = btf__resolve_size(btf, array->type); + if (element_sz <= 0 || (size - var->offset) % element_sz != 0) { + pr_warn("map '%s': cannot be resized, element size (%d) doesn't align with new total size (%u)\n", + bpf_map__name(map), element_sz, size); + return -EINVAL; + } + + /* create a new array based on the existing array, but with new length */ + nr_elements = (size - var->offset) / element_sz; + new_array_id = btf__add_array(btf, array->index_type, array->type, nr_elements); + if (new_array_id < 0) + return new_array_id; + + /* adding a new btf type invalidates existing pointers to btf objects, + * so refresh pointers before proceeding + */ + datasec_type = btf_type_by_id(btf, map->btf_value_type_id); + var = &btf_var_secinfos(datasec_type)[vlen - 1]; + var_type = btf_type_by_id(btf, var->type); + + /* finally update btf info */ + datasec_type->size = size; + var->size = size - var->offset; + var_type->type = new_array_id; + + return 0; +} + int bpf_map__set_value_size(struct bpf_map *map, __u32 size) { if (map->fd >= 0) return libbpf_err(-EBUSY); + + if (map->mmaped) { + int err; + size_t mmap_old_sz, mmap_new_sz; + + mmap_old_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); + mmap_new_sz = bpf_map_mmap_sz(size, map->def.max_entries); + err = bpf_map_mmap_resize(map, mmap_old_sz, mmap_new_sz); + if (err) { + pr_warn("map '%s': failed to resize memory-mapped region: %d\n", + bpf_map__name(map), err); + return err; + } + err = map_btf_datasec_resize(map, size); + if (err && err != -ENOENT) { + pr_warn("map '%s': failed to adjust resized BTF, clearing BTF key/value info: %d\n", + bpf_map__name(map), err); + map->btf_value_type_id = 0; + map->btf_key_type_id = 0; + } + } + map->def.value_size = size; return 0; } @@ -9441,7 +9559,7 @@ int bpf_map__set_initial_value(struct bpf_map *map, return 0; } -const void *bpf_map__initial_value(struct bpf_map *map, size_t *psize) +void *bpf_map__initial_value(struct bpf_map *map, size_t *psize) { if (!map->mmaped) return NULL; @@ -9957,7 +10075,7 @@ static int parse_uint_from_file(const char *file, const char *fmt) int err, ret; FILE *f; - f = fopen(file, "r"); + f = fopen(file, "re"); if (!f) { err = -errno; pr_debug("failed to open '%s': %s\n", file, @@ -12693,7 +12811,7 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s) for (i = 0; i < s->map_cnt; i++) { struct bpf_map *map = *s->maps[i].map; - size_t mmap_sz = bpf_map_mmap_sz(map); + size_t mmap_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); int prot, map_fd = bpf_map__fd(map); void **mmaped = s->maps[i].mmaped; @@ -12720,8 +12838,7 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s) * as per normal clean up procedure, so we don't need to worry * about it from skeleton's clean up perspective. */ - *mmaped = mmap(map->mmaped, mmap_sz, prot, - MAP_SHARED | MAP_FIXED, map_fd, 0); + *mmaped = mmap(map->mmaped, mmap_sz, prot, MAP_SHARED | MAP_FIXED, map_fd, 0); if (*mmaped == MAP_FAILED) { err = -errno; *mmaped = NULL; diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index 0b7362397ea3..754da73c643b 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -869,8 +869,22 @@ LIBBPF_API int bpf_map__set_numa_node(struct bpf_map *map, __u32 numa_node); /* get/set map key size */ LIBBPF_API __u32 bpf_map__key_size(const struct bpf_map *map); LIBBPF_API int bpf_map__set_key_size(struct bpf_map *map, __u32 size); -/* get/set map value size */ +/* get map value size */ LIBBPF_API __u32 bpf_map__value_size(const struct bpf_map *map); +/** + * @brief **bpf_map__set_value_size()** sets map value size. + * @param map the BPF map instance + * @return 0, on success; negative error, otherwise + * + * There is a special case for maps with associated memory-mapped regions, like + * the global data section maps (bss, data, rodata). When this function is used + * on such a map, the mapped region is resized. Afterward, an attempt is made to + * adjust the corresponding BTF info. This attempt is best-effort and can only + * succeed if the last variable of the data section map is an array. The array + * BTF type is replaced by a new BTF array type with a different length. + * Any previously existing pointers returned from bpf_map__initial_value() or + * corresponding data section skeleton pointer must be reinitialized. + */ LIBBPF_API int bpf_map__set_value_size(struct bpf_map *map, __u32 size); /* get map key/value BTF type IDs */ LIBBPF_API __u32 bpf_map__btf_key_type_id(const struct bpf_map *map); @@ -884,7 +898,7 @@ LIBBPF_API int bpf_map__set_map_extra(struct bpf_map *map, __u64 map_extra); LIBBPF_API int bpf_map__set_initial_value(struct bpf_map *map, const void *data, size_t size); -LIBBPF_API const void *bpf_map__initial_value(struct bpf_map *map, size_t *psize); +LIBBPF_API void *bpf_map__initial_value(struct bpf_map *map, size_t *psize); /** * @brief **bpf_map__is_internal()** tells the caller whether or not the diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index a5aa3a383d69..7521a2fb7626 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -391,3 +391,8 @@ LIBBPF_1.2.0 { bpf_map_get_info_by_fd; bpf_prog_get_info_by_fd; } LIBBPF_1.1.0; + +LIBBPF_1.3.0 { + global: + bpf_obj_pin_opts; +} LIBBPF_1.2.0; diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c index 6065f408a59c..9c4db90b92b6 100644 --- a/tools/lib/bpf/libbpf_probes.c +++ b/tools/lib/bpf/libbpf_probes.c @@ -38,7 +38,7 @@ static __u32 get_ubuntu_kernel_version(void) if (faccessat(AT_FDCWD, ubuntu_kver_file, R_OK, AT_EACCESS) != 0) return 0; - f = fopen(ubuntu_kver_file, "r"); + f = fopen(ubuntu_kver_file, "re"); if (!f) return 0; @@ -180,7 +180,9 @@ static int probe_prog_load(enum bpf_prog_type prog_type, case BPF_PROG_TYPE_SK_REUSEPORT: case BPF_PROG_TYPE_FLOW_DISSECTOR: case BPF_PROG_TYPE_CGROUP_SYSCTL: + break; case BPF_PROG_TYPE_NETFILTER: + opts.expected_attach_type = BPF_NETFILTER; break; default: return -EOPNOTSUPP; diff --git a/tools/lib/bpf/libbpf_version.h b/tools/lib/bpf/libbpf_version.h index 1fd2eeac5cfc..290411ddb39e 100644 --- a/tools/lib/bpf/libbpf_version.h +++ b/tools/lib/bpf/libbpf_version.h @@ -4,6 +4,6 @@ #define __LIBBPF_VERSION_H #define LIBBPF_MAJOR_VERSION 1 -#define LIBBPF_MINOR_VERSION 2 +#define LIBBPF_MINOR_VERSION 3 #endif /* __LIBBPF_VERSION_H */ diff --git a/tools/lib/bpf/usdt.c b/tools/lib/bpf/usdt.c index 086eef355ab3..f1a141555f08 100644 --- a/tools/lib/bpf/usdt.c +++ b/tools/lib/bpf/usdt.c @@ -466,7 +466,7 @@ static int parse_vma_segs(int pid, const char *lib_path, struct elf_seg **segs, proceed: sprintf(line, "/proc/%d/maps", pid); - f = fopen(line, "r"); + f = fopen(line, "re"); if (!f) { err = -errno; pr_warn("usdt: failed to open '%s' to get base addr of '%s': %d\n", @@ -954,8 +954,7 @@ struct bpf_link *usdt_manager_attach_usdt(struct usdt_manager *man, const struct spec_map_fd = bpf_map__fd(man->specs_map); ip_map_fd = bpf_map__fd(man->ip_to_spec_id_map); - /* TODO: perform path resolution similar to uprobe's */ - fd = open(path, O_RDONLY); + fd = open(path, O_RDONLY | O_CLOEXEC); if (fd < 0) { err = -errno; pr_warn("usdt: failed to open ELF binary '%s': %d\n", path, err); diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c index 1229b18bcdb1..2a5a29217374 100644 --- a/tools/lib/perf/cpumap.c +++ b/tools/lib/perf/cpumap.c @@ -99,6 +99,11 @@ static int cmp_cpu(const void *a, const void *b) return cpu_a->cpu - cpu_b->cpu; } +static struct perf_cpu __perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx) +{ + return RC_CHK_ACCESS(cpus)->map[idx]; +} + static struct perf_cpu_map *cpu_map__trim_new(int nr_cpus, const struct perf_cpu *tmp_cpus) { size_t payload_size = nr_cpus * sizeof(struct perf_cpu); @@ -111,8 +116,12 @@ static struct perf_cpu_map *cpu_map__trim_new(int nr_cpus, const struct perf_cpu /* Remove dups */ j = 0; for (i = 0; i < nr_cpus; i++) { - if (i == 0 || RC_CHK_ACCESS(cpus)->map[i].cpu != RC_CHK_ACCESS(cpus)->map[i - 1].cpu) - RC_CHK_ACCESS(cpus)->map[j++].cpu = RC_CHK_ACCESS(cpus)->map[i].cpu; + if (i == 0 || + __perf_cpu_map__cpu(cpus, i).cpu != + __perf_cpu_map__cpu(cpus, i - 1).cpu) { + RC_CHK_ACCESS(cpus)->map[j++].cpu = + __perf_cpu_map__cpu(cpus, i).cpu; + } } perf_cpu_map__set_nr(cpus, j); assert(j <= nr_cpus); @@ -269,26 +278,31 @@ out: return cpus; } +static int __perf_cpu_map__nr(const struct perf_cpu_map *cpus) +{ + return RC_CHK_ACCESS(cpus)->nr; +} + struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx) { struct perf_cpu result = { .cpu = -1 }; - if (cpus && idx < RC_CHK_ACCESS(cpus)->nr) - return RC_CHK_ACCESS(cpus)->map[idx]; + if (cpus && idx < __perf_cpu_map__nr(cpus)) + return __perf_cpu_map__cpu(cpus, idx); return result; } int perf_cpu_map__nr(const struct perf_cpu_map *cpus) { - return cpus ? RC_CHK_ACCESS(cpus)->nr : 1; + return cpus ? __perf_cpu_map__nr(cpus) : 1; } bool perf_cpu_map__empty(const struct perf_cpu_map *map) { - return map ? RC_CHK_ACCESS(map)->map[0].cpu == -1 : true; + return map ? __perf_cpu_map__cpu(map, 0).cpu == -1 : true; } int perf_cpu_map__idx(const struct perf_cpu_map *cpus, struct perf_cpu cpu) @@ -299,10 +313,10 @@ int perf_cpu_map__idx(const struct perf_cpu_map *cpus, struct perf_cpu cpu) return -1; low = 0; - high = RC_CHK_ACCESS(cpus)->nr; + high = __perf_cpu_map__nr(cpus); while (low < high) { int idx = (low + high) / 2; - struct perf_cpu cpu_at_idx = RC_CHK_ACCESS(cpus)->map[idx]; + struct perf_cpu cpu_at_idx = __perf_cpu_map__cpu(cpus, idx); if (cpu_at_idx.cpu == cpu.cpu) return idx; @@ -321,6 +335,32 @@ bool perf_cpu_map__has(const struct perf_cpu_map *cpus, struct perf_cpu cpu) return perf_cpu_map__idx(cpus, cpu) != -1; } +bool perf_cpu_map__equal(const struct perf_cpu_map *lhs, const struct perf_cpu_map *rhs) +{ + int nr; + + if (lhs == rhs) + return true; + + if (!lhs || !rhs) + return false; + + nr = __perf_cpu_map__nr(lhs); + if (nr != __perf_cpu_map__nr(rhs)) + return false; + + for (int idx = 0; idx < nr; idx++) { + if (__perf_cpu_map__cpu(lhs, idx).cpu != __perf_cpu_map__cpu(rhs, idx).cpu) + return false; + } + return true; +} + +bool perf_cpu_map__has_any_cpu(const struct perf_cpu_map *map) +{ + return map && __perf_cpu_map__cpu(map, 0).cpu == -1; +} + struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map) { struct perf_cpu result = { @@ -328,7 +368,9 @@ struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map) }; // cpu_map__trim_new() qsort()s it, cpu_map__default_new() sorts it as well. - return RC_CHK_ACCESS(map)->nr > 0 ? RC_CHK_ACCESS(map)->map[RC_CHK_ACCESS(map)->nr - 1] : result; + return __perf_cpu_map__nr(map) > 0 + ? __perf_cpu_map__cpu(map, __perf_cpu_map__nr(map) - 1) + : result; } /** Is 'b' a subset of 'a'. */ @@ -336,15 +378,15 @@ bool perf_cpu_map__is_subset(const struct perf_cpu_map *a, const struct perf_cpu { if (a == b || !b) return true; - if (!a || RC_CHK_ACCESS(b)->nr > RC_CHK_ACCESS(a)->nr) + if (!a || __perf_cpu_map__nr(b) > __perf_cpu_map__nr(a)) return false; - for (int i = 0, j = 0; i < RC_CHK_ACCESS(a)->nr; i++) { - if (RC_CHK_ACCESS(a)->map[i].cpu > RC_CHK_ACCESS(b)->map[j].cpu) + for (int i = 0, j = 0; i < __perf_cpu_map__nr(a); i++) { + if (__perf_cpu_map__cpu(a, i).cpu > __perf_cpu_map__cpu(b, j).cpu) return false; - if (RC_CHK_ACCESS(a)->map[i].cpu == RC_CHK_ACCESS(b)->map[j].cpu) { + if (__perf_cpu_map__cpu(a, i).cpu == __perf_cpu_map__cpu(b, j).cpu) { j++; - if (j == RC_CHK_ACCESS(b)->nr) + if (j == __perf_cpu_map__nr(b)) return true; } } @@ -374,27 +416,27 @@ struct perf_cpu_map *perf_cpu_map__merge(struct perf_cpu_map *orig, return perf_cpu_map__get(other); } - tmp_len = RC_CHK_ACCESS(orig)->nr + RC_CHK_ACCESS(other)->nr; + tmp_len = __perf_cpu_map__nr(orig) + __perf_cpu_map__nr(other); tmp_cpus = malloc(tmp_len * sizeof(struct perf_cpu)); if (!tmp_cpus) return NULL; /* Standard merge algorithm from wikipedia */ i = j = k = 0; - while (i < RC_CHK_ACCESS(orig)->nr && j < RC_CHK_ACCESS(other)->nr) { - if (RC_CHK_ACCESS(orig)->map[i].cpu <= RC_CHK_ACCESS(other)->map[j].cpu) { - if (RC_CHK_ACCESS(orig)->map[i].cpu == RC_CHK_ACCESS(other)->map[j].cpu) + while (i < __perf_cpu_map__nr(orig) && j < __perf_cpu_map__nr(other)) { + if (__perf_cpu_map__cpu(orig, i).cpu <= __perf_cpu_map__cpu(other, j).cpu) { + if (__perf_cpu_map__cpu(orig, i).cpu == __perf_cpu_map__cpu(other, j).cpu) j++; - tmp_cpus[k++] = RC_CHK_ACCESS(orig)->map[i++]; + tmp_cpus[k++] = __perf_cpu_map__cpu(orig, i++); } else - tmp_cpus[k++] = RC_CHK_ACCESS(other)->map[j++]; + tmp_cpus[k++] = __perf_cpu_map__cpu(other, j++); } - while (i < RC_CHK_ACCESS(orig)->nr) - tmp_cpus[k++] = RC_CHK_ACCESS(orig)->map[i++]; + while (i < __perf_cpu_map__nr(orig)) + tmp_cpus[k++] = __perf_cpu_map__cpu(orig, i++); - while (j < RC_CHK_ACCESS(other)->nr) - tmp_cpus[k++] = RC_CHK_ACCESS(other)->map[j++]; + while (j < __perf_cpu_map__nr(other)) + tmp_cpus[k++] = __perf_cpu_map__cpu(other, j++); assert(k <= tmp_len); merged = cpu_map__trim_new(k, tmp_cpus); @@ -402,3 +444,38 @@ struct perf_cpu_map *perf_cpu_map__merge(struct perf_cpu_map *orig, perf_cpu_map__put(orig); return merged; } + +struct perf_cpu_map *perf_cpu_map__intersect(struct perf_cpu_map *orig, + struct perf_cpu_map *other) +{ + struct perf_cpu *tmp_cpus; + int tmp_len; + int i, j, k; + struct perf_cpu_map *merged = NULL; + + if (perf_cpu_map__is_subset(other, orig)) + return perf_cpu_map__get(orig); + if (perf_cpu_map__is_subset(orig, other)) + return perf_cpu_map__get(other); + + tmp_len = max(__perf_cpu_map__nr(orig), __perf_cpu_map__nr(other)); + tmp_cpus = malloc(tmp_len * sizeof(struct perf_cpu)); + if (!tmp_cpus) + return NULL; + + i = j = k = 0; + while (i < __perf_cpu_map__nr(orig) && j < __perf_cpu_map__nr(other)) { + if (__perf_cpu_map__cpu(orig, i).cpu < __perf_cpu_map__cpu(other, j).cpu) + i++; + else if (__perf_cpu_map__cpu(orig, i).cpu > __perf_cpu_map__cpu(other, j).cpu) + j++; + else { + j++; + tmp_cpus[k++] = __perf_cpu_map__cpu(orig, i++); + } + } + if (k) + merged = cpu_map__trim_new(k, tmp_cpus); + free(tmp_cpus); + return merged; +} diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c index 81e8b5fcd8ba..b8b066d0dc5e 100644 --- a/tools/lib/perf/evlist.c +++ b/tools/lib/perf/evlist.c @@ -36,18 +36,33 @@ void perf_evlist__init(struct perf_evlist *evlist) static void __perf_evlist__propagate_maps(struct perf_evlist *evlist, struct perf_evsel *evsel) { - /* - * We already have cpus for evsel (via PMU sysfs) so - * keep it, if there's no target cpu list defined. - */ if (evsel->system_wide) { + /* System wide: set the cpu map of the evsel to all online CPUs. */ perf_cpu_map__put(evsel->cpus); evsel->cpus = perf_cpu_map__new(NULL); + } else if (evlist->has_user_cpus && evsel->is_pmu_core) { + /* + * User requested CPUs on a core PMU, ensure the requested CPUs + * are valid by intersecting with those of the PMU. + */ + perf_cpu_map__put(evsel->cpus); + evsel->cpus = perf_cpu_map__intersect(evlist->user_requested_cpus, evsel->own_cpus); } else if (!evsel->own_cpus || evlist->has_user_cpus || - (!evsel->requires_cpu && perf_cpu_map__empty(evlist->user_requested_cpus))) { + (!evsel->requires_cpu && perf_cpu_map__has_any_cpu(evlist->user_requested_cpus))) { + /* + * The PMU didn't specify a default cpu map, this isn't a core + * event and the user requested CPUs or the evlist user + * requested CPUs have the "any CPU" (aka dummy) CPU value. In + * which case use the user requested CPUs rather than the PMU + * ones. + */ perf_cpu_map__put(evsel->cpus); evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus); } else if (evsel->cpus != evsel->own_cpus) { + /* + * No user requested cpu map but the PMU cpu map doesn't match + * the evsel's. Reset it back to the PMU cpu map. + */ perf_cpu_map__put(evsel->cpus); evsel->cpus = perf_cpu_map__get(evsel->own_cpus); } diff --git a/tools/lib/perf/include/internal/evsel.h b/tools/lib/perf/include/internal/evsel.h index a99a75d9e78f..5cd220a61962 100644 --- a/tools/lib/perf/include/internal/evsel.h +++ b/tools/lib/perf/include/internal/evsel.h @@ -41,7 +41,14 @@ struct perf_sample_id { struct perf_evsel { struct list_head node; struct perf_event_attr attr; + /** The commonly used cpu map of CPUs the event should be opened upon, etc. */ struct perf_cpu_map *cpus; + /** + * The cpu map read from the PMU. For core PMUs this is the list of all + * CPUs the event can be opened upon. For other PMUs this is the default + * cpu map for opening the event on, for example, the first CPU on a + * socket for an uncore event. + */ struct perf_cpu_map *own_cpus; struct perf_thread_map *threads; struct xyarray *fd; @@ -55,9 +62,9 @@ struct perf_evsel { int nr_members; /* * system_wide is for events that need to be on every CPU, irrespective - * of user requested CPUs or threads. Map propagation will set cpus to - * this event's own_cpus, whereby they will contribute to evlist - * all_cpus. + * of user requested CPUs or threads. Tha main example of this is the + * dummy event. Map propagation will set cpus for this event to all CPUs + * as software PMU events like dummy, have a CPU map that is empty. */ bool system_wide; /* @@ -65,6 +72,8 @@ struct perf_evsel { * i.e. it cannot be the 'any CPU' value of -1. */ bool requires_cpu; + /** Is the PMU for the event a core one? Effects the handling of own_cpus. */ + bool is_pmu_core; int idx; }; diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h index 3f43f770cdac..e38d859a384d 100644 --- a/tools/lib/perf/include/perf/cpumap.h +++ b/tools/lib/perf/include/perf/cpumap.h @@ -11,8 +11,16 @@ struct perf_cpu { int cpu; }; +struct perf_cache { + int cache_lvl; + int cache; +}; + struct perf_cpu_map; +/** + * perf_cpu_map__dummy_new - a map with a singular "any CPU"/dummy -1 value. + */ LIBPERF_API struct perf_cpu_map *perf_cpu_map__dummy_new(void); LIBPERF_API struct perf_cpu_map *perf_cpu_map__default_new(void); LIBPERF_API struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list); @@ -20,12 +28,23 @@ LIBPERF_API struct perf_cpu_map *perf_cpu_map__read(FILE *file); LIBPERF_API struct perf_cpu_map *perf_cpu_map__get(struct perf_cpu_map *map); LIBPERF_API struct perf_cpu_map *perf_cpu_map__merge(struct perf_cpu_map *orig, struct perf_cpu_map *other); +LIBPERF_API struct perf_cpu_map *perf_cpu_map__intersect(struct perf_cpu_map *orig, + struct perf_cpu_map *other); LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map); LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx); LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus); +/** + * perf_cpu_map__empty - is map either empty or the "any CPU"/dummy value. + */ LIBPERF_API bool perf_cpu_map__empty(const struct perf_cpu_map *map); LIBPERF_API struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map); LIBPERF_API bool perf_cpu_map__has(const struct perf_cpu_map *map, struct perf_cpu cpu); +LIBPERF_API bool perf_cpu_map__equal(const struct perf_cpu_map *lhs, + const struct perf_cpu_map *rhs); +/** + * perf_cpu_map__any_cpu - Does the map contain the "any CPU"/dummy -1 value? + */ +LIBPERF_API bool perf_cpu_map__has_any_cpu(const struct perf_cpu_map *map); #define perf_cpu_map__for_each_cpu(cpu, idx, cpus) \ for ((idx) = 0, (cpu) = perf_cpu_map__cpu(cpus, idx); \ diff --git a/tools/lib/perf/include/perf/event.h b/tools/lib/perf/include/perf/event.h index 51b9338f4c11..ba2dcf64f4e6 100644 --- a/tools/lib/perf/include/perf/event.h +++ b/tools/lib/perf/include/perf/event.h @@ -380,7 +380,8 @@ enum { PERF_STAT_CONFIG_TERM__AGGR_MODE = 0, PERF_STAT_CONFIG_TERM__INTERVAL = 1, PERF_STAT_CONFIG_TERM__SCALE = 2, - PERF_STAT_CONFIG_TERM__MAX = 3, + PERF_STAT_CONFIG_TERM__AGGR_LEVEL = 3, + PERF_STAT_CONFIG_TERM__MAX = 4, }; struct perf_record_stat_config_entry { diff --git a/tools/lib/subcmd/exec-cmd.c b/tools/lib/subcmd/exec-cmd.c index 5dbea456973e..7739b5217cf6 100644 --- a/tools/lib/subcmd/exec-cmd.c +++ b/tools/lib/subcmd/exec-cmd.c @@ -36,38 +36,40 @@ static int is_absolute_path(const char *path) return path[0] == '/'; } -static const char *get_pwd_cwd(void) +static const char *get_pwd_cwd(char *buf, size_t sz) { - static char cwd[PATH_MAX + 1]; char *pwd; struct stat cwd_stat, pwd_stat; - if (getcwd(cwd, PATH_MAX) == NULL) + if (getcwd(buf, sz) == NULL) return NULL; pwd = getenv("PWD"); - if (pwd && strcmp(pwd, cwd)) { - stat(cwd, &cwd_stat); + if (pwd && strcmp(pwd, buf)) { + stat(buf, &cwd_stat); if (!stat(pwd, &pwd_stat) && pwd_stat.st_dev == cwd_stat.st_dev && pwd_stat.st_ino == cwd_stat.st_ino) { - strlcpy(cwd, pwd, PATH_MAX); + strlcpy(buf, pwd, sz); } } - return cwd; + return buf; } -static const char *make_nonrelative_path(const char *path) +static const char *make_nonrelative_path(char *buf, size_t sz, const char *path) { - static char buf[PATH_MAX + 1]; - if (is_absolute_path(path)) { - if (strlcpy(buf, path, PATH_MAX) >= PATH_MAX) + if (strlcpy(buf, path, sz) >= sz) die("Too long path: %.*s", 60, path); } else { - const char *cwd = get_pwd_cwd(); + const char *cwd = get_pwd_cwd(buf, sz); + if (!cwd) die("Cannot determine the current working directory"); - if (snprintf(buf, PATH_MAX, "%s/%s", cwd, path) >= PATH_MAX) + + if (strlen(cwd) + strlen(path) + 2 >= sz) die("Too long path: %.*s", 60, path); + + strcat(buf, "/"); + strcat(buf, path); } return buf; } @@ -133,8 +135,11 @@ static void add_path(char **out, const char *path) if (path && *path) { if (is_absolute_path(path)) astrcat(out, path); - else - astrcat(out, make_nonrelative_path(path)); + else { + char buf[PATH_MAX]; + + astrcat(out, make_nonrelative_path(buf, sizeof(buf), path)); + } astrcat(out, ":"); } diff --git a/tools/lib/subcmd/help.c b/tools/lib/subcmd/help.c index bf02d62a3b2b..67a8d6b740ea 100644 --- a/tools/lib/subcmd/help.c +++ b/tools/lib/subcmd/help.c @@ -16,6 +16,8 @@ void add_cmdname(struct cmdnames *cmds, const char *name, size_t len) { struct cmdname *ent = malloc(sizeof(*ent) + len + 1); + if (!ent) + return; ent->len = len; memcpy(ent->name, name, len); @@ -66,6 +68,7 @@ void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes) while (ci < cmds->cnt && ei < excludes->cnt) { cmp = strcmp(cmds->names[ci]->name, excludes->names[ei]->name); if (cmp < 0) { + zfree(&cmds->names[cj]); cmds->names[cj++] = cmds->names[ci++]; } else if (cmp == 0) { ci++; @@ -75,9 +78,12 @@ void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes) } } - while (ci < cmds->cnt) + while (ci < cmds->cnt) { + zfree(&cmds->names[cj]); cmds->names[cj++] = cmds->names[ci++]; - + } + for (ci = cj; ci < cmds->cnt; ci++) + zfree(&cmds->names[ci]); cmds->cnt = cj; } diff --git a/tools/lib/subcmd/parse-options.h b/tools/lib/subcmd/parse-options.h index 41b9b942504d..8e9147358a28 100644 --- a/tools/lib/subcmd/parse-options.h +++ b/tools/lib/subcmd/parse-options.h @@ -6,10 +6,6 @@ #include <stdbool.h> #include <stdint.h> -#ifndef NORETURN -#define NORETURN __attribute__((__noreturn__)) -#endif - enum parse_opt_type { /* special types */ OPTION_END, @@ -183,9 +179,9 @@ extern int parse_options_subcommand(int argc, const char **argv, const char *const subcommands[], const char *usagestr[], int flags); -extern NORETURN void usage_with_options(const char * const *usagestr, +extern __noreturn void usage_with_options(const char * const *usagestr, const struct option *options); -extern NORETURN __attribute__((format(printf,3,4))) +extern __noreturn __attribute__((format(printf,3,4))) void usage_with_options_msg(const char * const *usagestr, const struct option *options, const char *fmt, ...); diff --git a/tools/lib/subcmd/subcmd-util.h b/tools/lib/subcmd/subcmd-util.h index b2aec04fce8f..dfac76e35ac7 100644 --- a/tools/lib/subcmd/subcmd-util.h +++ b/tools/lib/subcmd/subcmd-util.h @@ -5,8 +5,7 @@ #include <stdarg.h> #include <stdlib.h> #include <stdio.h> - -#define NORETURN __attribute__((__noreturn__)) +#include <linux/compiler.h> static inline void report(const char *prefix, const char *err, va_list params) { @@ -15,7 +14,7 @@ static inline void report(const char *prefix, const char *err, va_list params) fprintf(stderr, " %s%s\n", prefix, msg); } -static NORETURN inline void die(const char *err, ...) +static __noreturn inline void die(const char *err, ...) { va_list params; |