diff options
Diffstat (limited to 'tools/perf/builtin-trace.c')
-rw-r--r-- | tools/perf/builtin-trace.c | 497 |
1 files changed, 454 insertions, 43 deletions
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index 8449f2beb54d..f6e847529073 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -19,6 +19,7 @@ #ifdef HAVE_LIBBPF_SUPPORT #include <bpf/bpf.h> #include <bpf/libbpf.h> +#include <bpf/btf.h> #ifdef HAVE_BPF_SKEL #include "bpf_skel/augmented_raw_syscalls.skel.h" #endif @@ -64,6 +65,7 @@ #include "syscalltbl.h" #include "rb_resort.h" #include "../perf.h" +#include "trace_augment.h" #include <errno.h> #include <inttypes.h> @@ -101,6 +103,12 @@ /* * strtoul: Go from a string to a value, i.e. for msr: MSR_FS_BASE to 0xc0000100 + * + * We have to explicitely mark the direction of the flow of data, if from the + * kernel to user space or the other way around, since the BPF collector we + * have so far copies only from user to kernel space, mark the arguments that + * go that direction, so that we donĀ“t end up collecting the previous contents + * for syscall args that goes from kernel to user space. */ struct syscall_arg_fmt { size_t (*scnprintf)(char *bf, size_t size, struct syscall_arg *arg); @@ -109,7 +117,12 @@ struct syscall_arg_fmt { void *parm; const char *name; u16 nr_entries; // for arrays + bool from_user; bool show_zero; +#ifdef HAVE_LIBBPF_SUPPORT + const struct btf_type *type; + int type_id; /* used in btf_dump */ +#endif }; struct syscall_fmt { @@ -140,6 +153,9 @@ struct trace { #ifdef HAVE_BPF_SKEL struct augmented_raw_syscalls_bpf *skel; #endif +#ifdef HAVE_LIBBPF_SUPPORT + struct btf *btf; +#endif struct record_opts opts; struct evlist *evlist; struct machine *host; @@ -196,6 +212,7 @@ struct trace { bool show_string_prefix; bool force; bool vfs_getname; + bool force_btf; int trace_pgfaults; char *perfconfig_events; struct { @@ -204,6 +221,20 @@ struct trace { } oe; }; +static void trace__load_vmlinux_btf(struct trace *trace __maybe_unused) +{ +#ifdef HAVE_LIBBPF_SUPPORT + if (trace->btf != NULL) + return; + + trace->btf = btf__load_vmlinux_btf(); + if (verbose > 0) { + fprintf(trace->output, trace->btf ? "vmlinux BTF loaded\n" : + "Failed to load vmlinux BTF\n"); + } +#endif +} + struct tp_field { int offset; union { @@ -830,6 +861,15 @@ static size_t syscall_arg__scnprintf_filename(char *bf, size_t size, #define SCA_FILENAME syscall_arg__scnprintf_filename +// 'argname' is just documentational at this point, to remove the previous comment with that info +#define SCA_FILENAME_FROM_USER(argname) \ + { .scnprintf = SCA_FILENAME, \ + .from_user = true, } + +static size_t syscall_arg__scnprintf_buf(char *bf, size_t size, struct syscall_arg *arg); + +#define SCA_BUF syscall_arg__scnprintf_buf + static size_t syscall_arg__scnprintf_pipe_flags(char *bf, size_t size, struct syscall_arg *arg) { @@ -887,6 +927,177 @@ static size_t syscall_arg__scnprintf_getrandom_flags(char *bf, size_t size, #define SCA_GETRANDOM_FLAGS syscall_arg__scnprintf_getrandom_flags +#ifdef HAVE_LIBBPF_SUPPORT +static void syscall_arg_fmt__cache_btf_enum(struct syscall_arg_fmt *arg_fmt, struct btf *btf, char *type) +{ + int id; + + type = strstr(type, "enum "); + if (type == NULL) + return; + + type += 5; // skip "enum " to get the enumeration name + + id = btf__find_by_name(btf, type); + if (id < 0) + return; + + arg_fmt->type = btf__type_by_id(btf, id); +} + +static bool syscall_arg__strtoul_btf_enum(char *bf, size_t size, struct syscall_arg *arg, u64 *val) +{ + const struct btf_type *bt = arg->fmt->type; + struct btf *btf = arg->trace->btf; + struct btf_enum *be = btf_enum(bt); + + for (int i = 0; i < btf_vlen(bt); ++i, ++be) { + const char *name = btf__name_by_offset(btf, be->name_off); + int max_len = max(size, strlen(name)); + + if (strncmp(name, bf, max_len) == 0) { + *val = be->val; + return true; + } + } + + return false; +} + +static bool syscall_arg__strtoul_btf_type(char *bf, size_t size, struct syscall_arg *arg, u64 *val) +{ + const struct btf_type *bt; + char *type = arg->type_name; + struct btf *btf; + + trace__load_vmlinux_btf(arg->trace); + + btf = arg->trace->btf; + if (btf == NULL) + return false; + + if (arg->fmt->type == NULL) { + // See if this is an enum + syscall_arg_fmt__cache_btf_enum(arg->fmt, btf, type); + } + + // Now let's see if we have a BTF type resolved + bt = arg->fmt->type; + if (bt == NULL) + return false; + + // If it is an enum: + if (btf_is_enum(arg->fmt->type)) + return syscall_arg__strtoul_btf_enum(bf, size, arg, val); + + return false; +} + +static size_t btf_enum_scnprintf(const struct btf_type *type, struct btf *btf, char *bf, size_t size, int val) +{ + struct btf_enum *be = btf_enum(type); + const int nr_entries = btf_vlen(type); + + for (int i = 0; i < nr_entries; ++i, ++be) { + if (be->val == val) { + return scnprintf(bf, size, "%s", + btf__name_by_offset(btf, be->name_off)); + } + } + + return 0; +} + +struct trace_btf_dump_snprintf_ctx { + char *bf; + size_t printed, size; +}; + +static void trace__btf_dump_snprintf(void *vctx, const char *fmt, va_list args) +{ + struct trace_btf_dump_snprintf_ctx *ctx = vctx; + + ctx->printed += vscnprintf(ctx->bf + ctx->printed, ctx->size - ctx->printed, fmt, args); +} + +static size_t btf_struct_scnprintf(const struct btf_type *type, struct btf *btf, char *bf, size_t size, struct syscall_arg *arg) +{ + struct trace_btf_dump_snprintf_ctx ctx = { + .bf = bf, + .size = size, + }; + struct augmented_arg *augmented_arg = arg->augmented.args; + int type_id = arg->fmt->type_id, consumed; + struct btf_dump *btf_dump; + + LIBBPF_OPTS(btf_dump_opts, dump_opts); + LIBBPF_OPTS(btf_dump_type_data_opts, dump_data_opts); + + if (arg == NULL || arg->augmented.args == NULL) + return 0; + + dump_data_opts.compact = true; + dump_data_opts.skip_names = !arg->trace->show_arg_names; + + btf_dump = btf_dump__new(btf, trace__btf_dump_snprintf, &ctx, &dump_opts); + if (btf_dump == NULL) + return 0; + + /* pretty print the struct data here */ + if (btf_dump__dump_type_data(btf_dump, type_id, arg->augmented.args->value, type->size, &dump_data_opts) == 0) + return 0; + + consumed = sizeof(*augmented_arg) + augmented_arg->size; + arg->augmented.args = ((void *)arg->augmented.args) + consumed; + arg->augmented.size -= consumed; + + btf_dump__free(btf_dump); + + return ctx.printed; +} + +static size_t trace__btf_scnprintf(struct trace *trace, struct syscall_arg *arg, char *bf, + size_t size, int val, char *type) +{ + struct syscall_arg_fmt *arg_fmt = arg->fmt; + + if (trace->btf == NULL) + return 0; + + if (arg_fmt->type == NULL) { + // Check if this is an enum and if we have the BTF type for it. + syscall_arg_fmt__cache_btf_enum(arg_fmt, trace->btf, type); + } + + // Did we manage to find a BTF type for the syscall/tracepoint argument? + if (arg_fmt->type == NULL) + return 0; + + if (btf_is_enum(arg_fmt->type)) + return btf_enum_scnprintf(arg_fmt->type, trace->btf, bf, size, val); + else if (btf_is_struct(arg_fmt->type) || btf_is_union(arg_fmt->type)) + return btf_struct_scnprintf(arg_fmt->type, trace->btf, bf, size, arg); + + return 0; +} + +#else // HAVE_LIBBPF_SUPPORT +static size_t trace__btf_scnprintf(struct trace *trace __maybe_unused, struct syscall_arg *arg __maybe_unused, + char *bf __maybe_unused, size_t size __maybe_unused, int val __maybe_unused, + char *type __maybe_unused) +{ + return 0; +} + +static bool syscall_arg__strtoul_btf_type(char *bf __maybe_unused, size_t size __maybe_unused, + struct syscall_arg *arg __maybe_unused, u64 *val __maybe_unused) +{ + return false; +} +#endif // HAVE_LIBBPF_SUPPORT + +#define STUL_BTF_TYPE syscall_arg__strtoul_btf_type + #define STRARRAY(name, array) \ { .scnprintf = SCA_STRARRAY, \ .strtoul = STUL_STRARRAY, \ @@ -921,16 +1132,17 @@ static const struct syscall_fmt syscall_fmts[] = { [1] = { .scnprintf = SCA_PTR, /* arg2 */ }, }, }, { .name = "bind", .arg = { [0] = { .scnprintf = SCA_INT, /* fd */ }, - [1] = { .scnprintf = SCA_SOCKADDR, /* umyaddr */ }, + [1] = SCA_SOCKADDR_FROM_USER(umyaddr), [2] = { .scnprintf = SCA_INT, /* addrlen */ }, }, }, { .name = "bpf", - .arg = { [0] = STRARRAY(cmd, bpf_cmd), }, }, + .arg = { [0] = STRARRAY(cmd, bpf_cmd), + [1] = { .from_user = true /* attr */, }, } }, { .name = "brk", .hexret = true, .arg = { [0] = { .scnprintf = SCA_PTR, /* brk */ }, }, }, { .name = "clock_gettime", .arg = { [0] = STRARRAY(clk_id, clockid), }, }, { .name = "clock_nanosleep", - .arg = { [2] = { .scnprintf = SCA_TIMESPEC, /* rqtp */ }, }, }, + .arg = { [2] = SCA_TIMESPEC_FROM_USER(req), }, }, { .name = "clone", .errpid = true, .nr_args = 5, .arg = { [0] = { .name = "flags", .scnprintf = SCA_CLONE_FLAGS, }, [1] = { .name = "child_stack", .scnprintf = SCA_HEX, }, @@ -941,7 +1153,7 @@ static const struct syscall_fmt syscall_fmts[] = { .arg = { [0] = { .scnprintf = SCA_CLOSE_FD, /* fd */ }, }, }, { .name = "connect", .arg = { [0] = { .scnprintf = SCA_INT, /* fd */ }, - [1] = { .scnprintf = SCA_SOCKADDR, /* servaddr */ }, + [1] = SCA_SOCKADDR_FROM_USER(servaddr), [2] = { .scnprintf = SCA_INT, /* addrlen */ }, }, }, { .name = "epoll_ctl", .arg = { [1] = STRARRAY(op, epoll_ctl_ops), }, }, @@ -949,11 +1161,11 @@ static const struct syscall_fmt syscall_fmts[] = { .arg = { [1] = { .scnprintf = SCA_EFD_FLAGS, /* flags */ }, }, }, { .name = "faccessat", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dirfd */ }, - [1] = { .scnprintf = SCA_FILENAME, /* pathname */ }, + [1] = SCA_FILENAME_FROM_USER(pathname), [2] = { .scnprintf = SCA_ACCMODE, /* mode */ }, }, }, { .name = "faccessat2", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dirfd */ }, - [1] = { .scnprintf = SCA_FILENAME, /* pathname */ }, + [1] = SCA_FILENAME_FROM_USER(pathname), [2] = { .scnprintf = SCA_ACCMODE, /* mode */ }, [3] = { .scnprintf = SCA_FACCESSAT2_FLAGS, /* flags */ }, }, }, { .name = "fchmodat", @@ -975,7 +1187,7 @@ static const struct syscall_fmt syscall_fmts[] = { [2] = { .scnprintf = SCA_FSMOUNT_ATTR_FLAGS, /* attr_flags */ }, }, }, { .name = "fspick", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dfd */ }, - [1] = { .scnprintf = SCA_FILENAME, /* path */ }, + [1] = SCA_FILENAME_FROM_USER(path), [2] = { .scnprintf = SCA_FSPICK_FLAGS, /* flags */ }, }, }, { .name = "fstat", .alias = "newfstat", }, { .name = "futex", @@ -1039,29 +1251,29 @@ static const struct syscall_fmt syscall_fmts[] = { .parm = &strarray__mmap_flags, }, [5] = { .scnprintf = SCA_HEX, /* offset */ }, }, }, { .name = "mount", - .arg = { [0] = { .scnprintf = SCA_FILENAME, /* dev_name */ }, + .arg = { [0] = SCA_FILENAME_FROM_USER(devname), [3] = { .scnprintf = SCA_MOUNT_FLAGS, /* flags */ .mask_val = SCAMV_MOUNT_FLAGS, /* flags */ }, }, }, { .name = "move_mount", .arg = { [0] = { .scnprintf = SCA_FDAT, /* from_dfd */ }, - [1] = { .scnprintf = SCA_FILENAME, /* from_pathname */ }, + [1] = SCA_FILENAME_FROM_USER(pathname), [2] = { .scnprintf = SCA_FDAT, /* to_dfd */ }, - [3] = { .scnprintf = SCA_FILENAME, /* to_pathname */ }, + [3] = SCA_FILENAME_FROM_USER(pathname), [4] = { .scnprintf = SCA_MOVE_MOUNT_FLAGS, /* flags */ }, }, }, { .name = "mprotect", .arg = { [0] = { .scnprintf = SCA_HEX, /* start */ }, [2] = { .scnprintf = SCA_MMAP_PROT, .show_zero = true, /* prot */ }, }, }, { .name = "mq_unlink", - .arg = { [0] = { .scnprintf = SCA_FILENAME, /* u_name */ }, }, }, + .arg = { [0] = SCA_FILENAME_FROM_USER(u_name), }, }, { .name = "mremap", .hexret = true, .arg = { [3] = { .scnprintf = SCA_MREMAP_FLAGS, /* flags */ }, }, }, { .name = "name_to_handle_at", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dfd */ }, }, }, { .name = "nanosleep", - .arg = { [0] = { .scnprintf = SCA_TIMESPEC, /* req */ }, }, }, + .arg = { [0] = SCA_TIMESPEC_FROM_USER(req), }, }, { .name = "newfstatat", .alias = "fstatat", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dirfd */ }, - [1] = { .scnprintf = SCA_FILENAME, /* pathname */ }, + [1] = SCA_FILENAME_FROM_USER(pathname), [3] = { .scnprintf = SCA_FS_AT_FLAGS, /* flags */ }, }, }, { .name = "open", .arg = { [1] = { .scnprintf = SCA_OPEN_FLAGS, /* flags */ }, }, }, @@ -1072,7 +1284,7 @@ static const struct syscall_fmt syscall_fmts[] = { .arg = { [0] = { .scnprintf = SCA_FDAT, /* dfd */ }, [2] = { .scnprintf = SCA_OPEN_FLAGS, /* flags */ }, }, }, { .name = "perf_event_open", - .arg = { [0] = { .scnprintf = SCA_PERF_ATTR, /* attr */ }, + .arg = { [0] = SCA_PERF_ATTR_FROM_USER(attr), [2] = { .scnprintf = SCA_INT, /* cpu */ }, [3] = { .scnprintf = SCA_FD, /* group_fd */ }, [4] = { .scnprintf = SCA_PERF_FLAGS, /* flags */ }, }, }, @@ -1097,7 +1309,8 @@ static const struct syscall_fmt syscall_fmts[] = { { .name = "pread", .alias = "pread64", }, { .name = "preadv", .alias = "pread", }, { .name = "prlimit64", - .arg = { [1] = STRARRAY(resource, rlimit_resources), }, }, + .arg = { [1] = STRARRAY(resource, rlimit_resources), + [2] = { .from_user = true /* new_rlim */, }, }, }, { .name = "pwrite", .alias = "pwrite64", }, { .name = "readlinkat", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dfd */ }, }, }, @@ -1114,6 +1327,8 @@ static const struct syscall_fmt syscall_fmts[] = { .arg = { [0] = { .scnprintf = SCA_FDAT, /* olddirfd */ }, [2] = { .scnprintf = SCA_FDAT, /* newdirfd */ }, [4] = { .scnprintf = SCA_RENAMEAT2_FLAGS, /* flags */ }, }, }, + { .name = "rseq", .errpid = true, + .arg = { [0] = { .from_user = true /* rseq */, }, }, }, { .name = "rt_sigaction", .arg = { [0] = { .scnprintf = SCA_SIGNUM, /* sig */ }, }, }, { .name = "rt_sigprocmask", @@ -1135,12 +1350,15 @@ static const struct syscall_fmt syscall_fmts[] = { .arg = { [2] = { .scnprintf = SCA_MSG_FLAGS, /* flags */ }, }, }, { .name = "sendto", .arg = { [3] = { .scnprintf = SCA_MSG_FLAGS, /* flags */ }, - [4] = { .scnprintf = SCA_SOCKADDR, /* addr */ }, }, }, + [4] = SCA_SOCKADDR_FROM_USER(addr), }, }, + { .name = "set_robust_list", .errpid = true, + .arg = { [0] = { .from_user = true /* head */, }, }, }, { .name = "set_tid_address", .errpid = true, }, { .name = "setitimer", .arg = { [0] = STRARRAY(which, itimers), }, }, { .name = "setrlimit", - .arg = { [0] = STRARRAY(resource, rlimit_resources), }, }, + .arg = { [0] = STRARRAY(resource, rlimit_resources), + [1] = { .from_user = true /* rlim */, }, }, }, { .name = "setsockopt", .arg = { [1] = STRARRAY(level, socket_level), }, }, { .name = "socket", @@ -1157,9 +1375,9 @@ static const struct syscall_fmt syscall_fmts[] = { [2] = { .scnprintf = SCA_FS_AT_FLAGS, /* flags */ } , [3] = { .scnprintf = SCA_STATX_MASK, /* mask */ }, }, }, { .name = "swapoff", - .arg = { [0] = { .scnprintf = SCA_FILENAME, /* specialfile */ }, }, }, + .arg = { [0] = SCA_FILENAME_FROM_USER(specialfile), }, }, { .name = "swapon", - .arg = { [0] = { .scnprintf = SCA_FILENAME, /* specialfile */ }, }, }, + .arg = { [0] = SCA_FILENAME_FROM_USER(specialfile), }, }, { .name = "symlinkat", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dfd */ }, }, }, { .name = "sync_file_range", @@ -1169,11 +1387,11 @@ static const struct syscall_fmt syscall_fmts[] = { { .name = "tkill", .arg = { [1] = { .scnprintf = SCA_SIGNUM, /* sig */ }, }, }, { .name = "umount2", .alias = "umount", - .arg = { [0] = { .scnprintf = SCA_FILENAME, /* name */ }, }, }, + .arg = { [0] = SCA_FILENAME_FROM_USER(name), }, }, { .name = "uname", .alias = "newuname", }, { .name = "unlinkat", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dfd */ }, - [1] = { .scnprintf = SCA_FILENAME, /* pathname */ }, + [1] = SCA_FILENAME_FROM_USER(pathname), [2] = { .scnprintf = SCA_FS_AT_FLAGS, /* flags */ }, }, }, { .name = "utimensat", .arg = { [0] = { .scnprintf = SCA_FDAT, /* dirfd */ }, }, }, @@ -1181,6 +1399,8 @@ static const struct syscall_fmt syscall_fmts[] = { .arg = { [2] = { .scnprintf = SCA_WAITID_OPTIONS, /* options */ }, }, }, { .name = "waitid", .errpid = true, .arg = { [3] = { .scnprintf = SCA_WAITID_OPTIONS, /* options */ }, }, }, + { .name = "write", .errpid = true, + .arg = { [1] = { .scnprintf = SCA_BUF /* buf */, .from_user = true, }, }, }, }; static int syscall_fmt__cmp(const void *name, const void *fmtp) @@ -1238,6 +1458,7 @@ struct syscall { bool is_exit; bool is_open; bool nonexistent; + bool use_btf; struct tep_format_field *args; const char *name; const struct syscall_fmt *fmt; @@ -1551,6 +1772,32 @@ static size_t syscall_arg__scnprintf_filename(char *bf, size_t size, return 0; } +#define MAX_CONTROL_CHAR 31 +#define MAX_ASCII 127 + +static size_t syscall_arg__scnprintf_buf(char *bf, size_t size, struct syscall_arg *arg) +{ + struct augmented_arg *augmented_arg = arg->augmented.args; + unsigned char *orig = (unsigned char *)augmented_arg->value; + size_t printed = 0; + int consumed; + + if (augmented_arg == NULL) + return 0; + + for (int j = 0; j < augmented_arg->size; ++j) { + bool control_char = orig[j] <= MAX_CONTROL_CHAR || orig[j] >= MAX_ASCII; + /* print control characters (0~31 and 127), and non-ascii characters in \(digits) */ + printed += scnprintf(bf + printed, size - printed, control_char ? "\\%d" : "%c", (int)orig[j]); + } + + consumed = sizeof(*augmented_arg) + augmented_arg->size; + arg->augmented.args = ((void *)arg->augmented.args) + consumed; + arg->augmented.size -= consumed; + + return printed; +} + static bool trace__filter_duration(struct trace *trace, double t) { return t < (trace->duration_filter * NSEC_PER_MSEC); @@ -1637,7 +1884,7 @@ static int trace__process_event(struct trace *trace, struct machine *machine, return ret; } -static int trace__tool_process(struct perf_tool *tool, +static int trace__tool_process(const struct perf_tool *tool, union perf_event *event, struct perf_sample *sample, struct machine *machine) @@ -1744,7 +1991,8 @@ static const struct syscall_arg_fmt *syscall_arg_fmt__find_by_name(const char *n } static struct tep_format_field * -syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field *field) +syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field *field, + bool *use_btf) { struct tep_format_field *last_field = NULL; int len; @@ -1757,11 +2005,15 @@ syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field len = strlen(field->name); + // As far as heuristics (or intention) goes this seems to hold true, and makes sense! + if ((field->flags & TEP_FIELD_IS_POINTER) && strstarts(field->type, "const ")) + arg->from_user = true; + if (strcmp(field->type, "const char *") == 0 && ((len >= 4 && strcmp(field->name + len - 4, "name") == 0) || - strstr(field->name, "path") != NULL)) + strstr(field->name, "path") != NULL)) { arg->scnprintf = SCA_FILENAME; - else if ((field->flags & TEP_FIELD_IS_POINTER) || strstr(field->name, "addr")) + } else if ((field->flags & TEP_FIELD_IS_POINTER) || strstr(field->name, "addr")) arg->scnprintf = SCA_PTR; else if (strcmp(field->type, "pid_t") == 0) arg->scnprintf = SCA_PID; @@ -1782,6 +2034,9 @@ syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field * 7 unsigned long */ arg->scnprintf = SCA_FD; + } else if (strstr(field->type, "enum") && use_btf != NULL) { + *use_btf = true; + arg->strtoul = STUL_BTF_TYPE; } else { const struct syscall_arg_fmt *fmt = syscall_arg_fmt__find_by_name(field->name); @@ -1798,7 +2053,8 @@ syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field static int syscall__set_arg_fmts(struct syscall *sc) { - struct tep_format_field *last_field = syscall_arg_fmt__init_array(sc->arg_fmt, sc->args); + struct tep_format_field *last_field = syscall_arg_fmt__init_array(sc->arg_fmt, sc->args, + &sc->use_btf); if (last_field) sc->args_size = last_field->offset + last_field->size; @@ -1811,6 +2067,7 @@ static int trace__read_syscall_info(struct trace *trace, int id) char tp_name[128]; struct syscall *sc; const char *name = syscalltbl__name(trace->sctbl, id); + int err; #ifdef HAVE_SYSCALL_TABLE_SUPPORT if (trace->syscalls.table == NULL) { @@ -1883,15 +2140,21 @@ static int trace__read_syscall_info(struct trace *trace, int id) sc->is_exit = !strcmp(name, "exit_group") || !strcmp(name, "exit"); sc->is_open = !strcmp(name, "open") || !strcmp(name, "openat"); - return syscall__set_arg_fmts(sc); + err = syscall__set_arg_fmts(sc); + + /* after calling syscall__set_arg_fmts() we'll know whether use_btf is true */ + if (sc->use_btf) + trace__load_vmlinux_btf(trace); + + return err; } -static int evsel__init_tp_arg_scnprintf(struct evsel *evsel) +static int evsel__init_tp_arg_scnprintf(struct evsel *evsel, bool *use_btf) { struct syscall_arg_fmt *fmt = evsel__syscall_arg_fmt(evsel); if (fmt != NULL) { - syscall_arg_fmt__init_array(fmt, evsel->tp_format->format.fields); + syscall_arg_fmt__init_array(fmt, evsel->tp_format->format.fields, use_btf); return 0; } @@ -2050,7 +2313,7 @@ static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size, unsigned char *args, void *augmented_args, int augmented_args_size, struct trace *trace, struct thread *thread) { - size_t printed = 0; + size_t printed = 0, btf_printed; unsigned long val; u8 bit = 1; struct syscall_arg arg = { @@ -2066,6 +2329,7 @@ static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size, .show_string_prefix = trace->show_string_prefix, }; struct thread_trace *ttrace = thread__priv(thread); + void *default_scnprintf; /* * Things like fcntl will set this in its 'cmd' formatter to pick the @@ -2093,9 +2357,13 @@ static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size, /* * Suppress this argument if its value is zero and show_zero * property isn't set. + * + * If it has a BTF type, then override the zero suppression knob + * as the common case is for zero in an enum to have an associated entry. */ if (val == 0 && !trace->show_zeros && - !(sc->arg_fmt && sc->arg_fmt[arg.idx].show_zero)) + !(sc->arg_fmt && sc->arg_fmt[arg.idx].show_zero) && + !(sc->arg_fmt && sc->arg_fmt[arg.idx].strtoul == STUL_BTF_TYPE)) continue; printed += scnprintf(bf + printed, size - printed, "%s", printed ? ", " : ""); @@ -2103,6 +2371,17 @@ static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size, if (trace->show_arg_names) printed += scnprintf(bf + printed, size - printed, "%s: ", field->name); + default_scnprintf = sc->arg_fmt[arg.idx].scnprintf; + + if (trace->force_btf || default_scnprintf == NULL || default_scnprintf == SCA_PTR) { + btf_printed = trace__btf_scnprintf(trace, &arg, bf + printed, + size - printed, val, field->type); + if (btf_printed) { + printed += btf_printed; + continue; + } + } + printed += syscall_arg_fmt__scnprintf_val(&sc->arg_fmt[arg.idx], bf + printed, size - printed, &arg, val); } @@ -2749,7 +3028,7 @@ static size_t trace__fprintf_tp_fields(struct trace *trace, struct evsel *evsel, size_t size = sizeof(bf); struct tep_format_field *field = evsel->tp_format->format.fields; struct syscall_arg_fmt *arg = __evsel__syscall_arg_fmt(evsel); - size_t printed = 0; + size_t printed = 0, btf_printed; unsigned long val; u8 bit = 1; struct syscall_arg syscall_arg = { @@ -2791,7 +3070,7 @@ static size_t trace__fprintf_tp_fields(struct trace *trace, struct evsel *evsel, val = syscall_arg_fmt__mask_val(arg, &syscall_arg, val); /* Suppress this argument if its value is zero and show_zero property isn't set. */ - if (val == 0 && !trace->show_zeros && !arg->show_zero) + if (val == 0 && !trace->show_zeros && !arg->show_zero && arg->strtoul != STUL_BTF_TYPE) continue; printed += scnprintf(bf + printed, size - printed, "%s", printed ? ", " : ""); @@ -2799,6 +3078,12 @@ static size_t trace__fprintf_tp_fields(struct trace *trace, struct evsel *evsel, if (trace->show_arg_names) printed += scnprintf(bf + printed, size - printed, "%s: ", field->name); + btf_printed = trace__btf_scnprintf(trace, &syscall_arg, bf + printed, size - printed, val, field->type); + if (btf_printed) { + printed += btf_printed; + continue; + } + printed += syscall_arg_fmt__scnprintf_val(arg, bf + printed, size - printed, &syscall_arg, val); } @@ -3009,7 +3294,7 @@ static void trace__set_base_time(struct trace *trace, trace->base_time = sample->time; } -static int trace__process_sample(struct perf_tool *tool, +static int trace__process_sample(const struct perf_tool *tool, union perf_event *event, struct perf_sample *sample, struct evsel *evsel, @@ -3276,6 +3561,23 @@ out_enomem: } #ifdef HAVE_BPF_SKEL +static int syscall_arg_fmt__cache_btf_struct(struct syscall_arg_fmt *arg_fmt, struct btf *btf, char *type) +{ + int id; + + if (arg_fmt->type != NULL) + return -1; + + id = btf__find_by_name(btf, type); + if (id < 0) + return -1; + + arg_fmt->type = btf__type_by_id(btf, id); + arg_fmt->type_id = id; + + return 0; +} + static struct bpf_program *trace__find_bpf_program_by_title(struct trace *trace, const char *name) { struct bpf_program *pos, *prog = NULL; @@ -3351,6 +3653,91 @@ static int trace__bpf_prog_sys_exit_fd(struct trace *trace, int id) return sc ? bpf_program__fd(sc->bpf_prog.sys_exit) : bpf_program__fd(trace->skel->progs.syscall_unaugmented); } +static int trace__bpf_sys_enter_beauty_map(struct trace *trace, int key, unsigned int *beauty_array) +{ + struct tep_format_field *field; + struct syscall *sc = trace__syscall_info(trace, NULL, key); + const struct btf_type *bt; + char *struct_offset, *tmp, name[32]; + bool can_augment = false; + int i, cnt; + + if (sc == NULL) + return -1; + + trace__load_vmlinux_btf(trace); + if (trace->btf == NULL) + return -1; + + for (i = 0, field = sc->args; field; ++i, field = field->next) { + // XXX We're only collecting pointer payloads _from_ user space + if (!sc->arg_fmt[i].from_user) + continue; + + struct_offset = strstr(field->type, "struct "); + if (struct_offset == NULL) + struct_offset = strstr(field->type, "union "); + else + struct_offset++; // "union" is shorter + + if (field->flags & TEP_FIELD_IS_POINTER && struct_offset) { /* struct or union (think BPF's attr arg) */ + struct_offset += 6; + + /* for 'struct foo *', we only want 'foo' */ + for (tmp = struct_offset, cnt = 0; *tmp != ' ' && *tmp != '\0'; ++tmp, ++cnt) { + } + + strncpy(name, struct_offset, cnt); + name[cnt] = '\0'; + + /* cache struct's btf_type and type_id */ + if (syscall_arg_fmt__cache_btf_struct(&sc->arg_fmt[i], trace->btf, name)) + continue; + + bt = sc->arg_fmt[i].type; + beauty_array[i] = bt->size; + can_augment = true; + } else if (field->flags & TEP_FIELD_IS_POINTER && /* string */ + strcmp(field->type, "const char *") == 0 && + (strstr(field->name, "name") || + strstr(field->name, "path") || + strstr(field->name, "file") || + strstr(field->name, "root") || + strstr(field->name, "key") || + strstr(field->name, "special") || + strstr(field->name, "type") || + strstr(field->name, "description"))) { + beauty_array[i] = 1; + can_augment = true; + } else if (field->flags & TEP_FIELD_IS_POINTER && /* buffer */ + strstr(field->type, "char *") && + (strstr(field->name, "buf") || + strstr(field->name, "val") || + strstr(field->name, "msg"))) { + int j; + struct tep_format_field *field_tmp; + + /* find the size of the buffer that appears in pairs with buf */ + for (j = 0, field_tmp = sc->args; field_tmp; ++j, field_tmp = field_tmp->next) { + if (!(field_tmp->flags & TEP_FIELD_IS_POINTER) && /* only integers */ + (strstr(field_tmp->name, "count") || + strstr(field_tmp->name, "siz") || /* size, bufsiz */ + (strstr(field_tmp->name, "len") && strcmp(field_tmp->name, "filename")))) { + /* filename's got 'len' in it, we don't want that */ + beauty_array[i] = -(j + 1); + can_augment = true; + break; + } + } + } + } + + if (can_augment) + return 0; + + return -1; +} + static struct bpf_program *trace__find_usable_bpf_prog_entry(struct trace *trace, struct syscall *sc) { struct tep_format_field *field, *candidate_field; @@ -3455,7 +3842,9 @@ static int trace__init_syscalls_bpf_prog_array_maps(struct trace *trace) { int map_enter_fd = bpf_map__fd(trace->skel->maps.syscalls_sys_enter); int map_exit_fd = bpf_map__fd(trace->skel->maps.syscalls_sys_exit); + int beauty_map_fd = bpf_map__fd(trace->skel->maps.beauty_map_enter); int err = 0; + unsigned int beauty_array[6]; for (int i = 0; i < trace->sctbl->syscalls.nr_entries; ++i) { int prog_fd, key = syscalltbl__id_at_idx(trace->sctbl, i); @@ -3474,6 +3863,15 @@ static int trace__init_syscalls_bpf_prog_array_maps(struct trace *trace) err = bpf_map_update_elem(map_exit_fd, &key, &prog_fd, BPF_ANY); if (err) break; + + /* use beauty_map to tell BPF how many bytes to collect, set beauty_map's value here */ + memset(beauty_array, 0, sizeof(beauty_array)); + err = trace__bpf_sys_enter_beauty_map(trace, key, (unsigned int *)beauty_array); + if (err) + continue; + err = bpf_map_update_elem(beauty_map_fd, &key, beauty_array, BPF_ANY); + if (err) + break; } /* @@ -3680,7 +4078,8 @@ static int ordered_events__deliver_event(struct ordered_events *oe, return __trace__deliver_event(trace, event->event); } -static struct syscall_arg_fmt *evsel__find_syscall_arg_fmt_by_name(struct evsel *evsel, char *arg) +static struct syscall_arg_fmt *evsel__find_syscall_arg_fmt_by_name(struct evsel *evsel, char *arg, + char **type) { struct tep_format_field *field; struct syscall_arg_fmt *fmt = __evsel__syscall_arg_fmt(evsel); @@ -3689,13 +4088,15 @@ static struct syscall_arg_fmt *evsel__find_syscall_arg_fmt_by_name(struct evsel return NULL; for (field = evsel->tp_format->format.fields; field; field = field->next, ++fmt) - if (strcmp(field->name, arg) == 0) + if (strcmp(field->name, arg) == 0) { + *type = field->type; return fmt; + } return NULL; } -static int trace__expand_filter(struct trace *trace __maybe_unused, struct evsel *evsel) +static int trace__expand_filter(struct trace *trace, struct evsel *evsel) { char *tok, *left = evsel->filter, *new_filter = evsel->filter; @@ -3728,14 +4129,14 @@ static int trace__expand_filter(struct trace *trace __maybe_unused, struct evsel struct syscall_arg_fmt *fmt; int left_size = tok - left, right_size = right_end - right; - char arg[128]; + char arg[128], *type; while (isspace(left[left_size - 1])) --left_size; scnprintf(arg, sizeof(arg), "%.*s", left_size, left); - fmt = evsel__find_syscall_arg_fmt_by_name(evsel, arg); + fmt = evsel__find_syscall_arg_fmt_by_name(evsel, arg, &type); if (fmt == NULL) { pr_err("\"%s\" not found in \"%s\", can't set filter \"%s\"\n", arg, evsel->name, evsel->filter); @@ -3748,6 +4149,9 @@ static int trace__expand_filter(struct trace *trace __maybe_unused, struct evsel if (fmt->strtoul) { u64 val; struct syscall_arg syscall_arg = { + .trace = trace, + .fmt = fmt, + .type_name = type, .parm = fmt->parm, }; @@ -3959,7 +4363,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) err = trace__expand_filters(trace, &evsel); if (err) goto out_delete_evlist; - err = evlist__apply_filters(evlist, &evsel); + err = evlist__apply_filters(evlist, &evsel, &trace->opts.target); if (err < 0) goto out_error_apply_filters; @@ -4451,7 +4855,7 @@ static void evsel__set_syscall_arg_fmt(struct evsel *evsel, const char *name) } } -static int evlist__set_syscall_tp_fields(struct evlist *evlist) +static int evlist__set_syscall_tp_fields(struct evlist *evlist, bool *use_btf) { struct evsel *evsel; @@ -4460,7 +4864,7 @@ static int evlist__set_syscall_tp_fields(struct evlist *evlist) continue; if (strcmp(evsel->tp_format->system, "syscalls")) { - evsel__init_tp_arg_scnprintf(evsel); + evsel__init_tp_arg_scnprintf(evsel, use_btf); continue; } @@ -4781,6 +5185,8 @@ int cmd_trace(int argc, const char **argv) OPT_INTEGER('D', "delay", &trace.opts.target.initial_delay, "ms to wait before starting measurement after program " "start"), + OPT_BOOLEAN(0, "force-btf", &trace.force_btf, "Prefer btf_dump general pretty printer" + "to customized ones"), OPTS_EVSWITCH(&trace.evswitch), OPT_END() }; @@ -4938,11 +5344,16 @@ skip_augmentation: } if (trace.evlist->core.nr_entries > 0) { + bool use_btf = false; + evlist__set_default_evsel_handler(trace.evlist, trace__event_handler); - if (evlist__set_syscall_tp_fields(trace.evlist)) { + if (evlist__set_syscall_tp_fields(trace.evlist, &use_btf)) { perror("failed to set syscalls:* tracepoint fields"); goto out; } + + if (use_btf) + trace__load_vmlinux_btf(&trace); } if (trace.sort_events) { |