From c18f9eabeec766ac8348ce7391d6fd368198522c Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Mon, 4 Mar 2024 12:40:45 +0900 Subject: tracing: Remove redundant #else block for BTF args from README Remove redundant #else block for BTF args from README message. This is a cleanup, so no change on the message. Link: https://lore.kernel.org/all/170952364558.229804.17285528811097152410.stgit@devnote2/ Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 8198bfc54b58..516c585f5879 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -5747,11 +5747,9 @@ static const char readme_msg[] = "\t args: =fetcharg[:type]\n" "\t fetcharg: (%|$), @
, @[+|-],\n" #ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API -#ifdef CONFIG_PROBE_EVENTS_BTF_ARGS "\t $stack, $stack, $retval, $comm, $arg,\n" +#ifdef CONFIG_PROBE_EVENTS_BTF_ARGS "\t [->field[->field|.field...]],\n" -#else - "\t $stack, $stack, $retval, $comm, $arg,\n" #endif #else "\t $stack, $stack, $retval, $comm,\n" -- cgit v1.2.3 From 25f00e40ce7953db197af3a59233711d154c9d80 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Mon, 4 Mar 2024 12:40:55 +0900 Subject: tracing/probes: Support $argN in return probe (kprobe and fprobe) Support accessing $argN in the return probe events. This will help users to record entry data in function return (exit) event for simplfing the function entry/exit information in one event, and record the result values (e.g. allocated object/initialized object) at function exit. For example, if we have a function `int init_foo(struct foo *obj, int param)` sometimes we want to check how `obj` is initialized. In such case, we can define a new return event like below; # echo 'r init_foo retval=$retval param=$arg2 field1=+0($arg1)' >> kprobe_events Thus it records the function parameter `param` and its result `obj->field1` (the dereference will be done in the function exit timing) value at once. This also support fprobe, BTF args and'$arg*'. So if CONFIG_DEBUG_INFO_BTF is enabled, we can trace both function parameters and the return value by following command. # echo 'f target_function%return $arg* $retval' >> dynamic_events Link: https://lore.kernel.org/all/170952365552.229804.224112990211602895.stgit@devnote2/ Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/trace.c | 1 + kernel/trace/trace_eprobe.c | 6 +- kernel/trace/trace_fprobe.c | 55 +++++-- kernel/trace/trace_kprobe.c | 56 +++++-- kernel/trace/trace_probe.c | 177 ++++++++++++++++++--- kernel/trace/trace_probe.h | 28 +++- kernel/trace/trace_probe_tmpl.h | 10 +- kernel/trace/trace_uprobe.c | 12 +- .../ftrace/test.d/dynevent/fprobe_syntax_errors.tc | 4 + .../ftrace/test.d/kprobe/kprobe_syntax_errors.tc | 2 + 10 files changed, 289 insertions(+), 62 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 516c585f5879..71a96decc276 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -5755,6 +5755,7 @@ static const char readme_msg[] = "\t $stack, $stack, $retval, $comm,\n" #endif "\t +|-[u](), \\imm-value, \\\"imm-string\"\n" + "\t kernel return probes support: $retval, $arg, $comm\n" "\t type: s8/16/32/64, u8/16/32/64, x8/16/32/64, char, string, symbol,\n" "\t b@/, ustring,\n" "\t symstr, \\[\\]\n" diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c index eb72def7410f..b0e0ec85912e 100644 --- a/kernel/trace/trace_eprobe.c +++ b/kernel/trace/trace_eprobe.c @@ -390,8 +390,8 @@ static int get_eprobe_size(struct trace_probe *tp, void *rec) /* Note that we don't verify it, since the code does not come from user space */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { unsigned long val; int ret; @@ -438,7 +438,7 @@ __eprobe_trace_func(struct eprobe_data *edata, void *rec) return; entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); - store_trace_args(&entry[1], &edata->ep->tp, rec, sizeof(*entry), dsize); + store_trace_args(&entry[1], &edata->ep->tp, rec, NULL, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } diff --git a/kernel/trace/trace_fprobe.c b/kernel/trace/trace_fprobe.c index 5109650b0d82..4f4280815522 100644 --- a/kernel/trace/trace_fprobe.c +++ b/kernel/trace/trace_fprobe.c @@ -4,6 +4,7 @@ * Copyright (C) 2022 Google LLC. */ #define pr_fmt(fmt) "trace_fprobe: " fmt +#include #include #include @@ -129,8 +130,8 @@ static bool trace_fprobe_is_registered(struct trace_fprobe *tf) * from user space. */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { struct pt_regs *regs = rec; unsigned long val; @@ -152,6 +153,9 @@ retry: case FETCH_OP_ARG: val = regs_get_kernel_argument(regs, code->param); break; + case FETCH_OP_EDATA: + val = *(unsigned long *)((unsigned long)edata + code->offset); + break; #endif case FETCH_NOP_SYMBOL: /* Ignore a place holder */ code++; @@ -184,7 +188,7 @@ __fentry_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, if (trace_trigger_soft_disabled(trace_file)) return; - dsize = __get_data_size(&tf->tp, regs); + dsize = __get_data_size(&tf->tp, regs, NULL); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + tf->tp.size + dsize); @@ -194,7 +198,7 @@ __fentry_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, fbuffer.regs = regs; entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); entry->ip = entry_ip; - store_trace_args(&entry[1], &tf->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tf->tp, regs, NULL, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } @@ -211,10 +215,23 @@ fentry_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, NOKPROBE_SYMBOL(fentry_trace_func); /* function exit handler */ +static int trace_fprobe_entry_handler(struct fprobe *fp, unsigned long entry_ip, + unsigned long ret_ip, struct pt_regs *regs, + void *entry_data) +{ + struct trace_fprobe *tf = container_of(fp, struct trace_fprobe, fp); + + if (tf->tp.entry_arg) + store_trace_entry_data(entry_data, &tf->tp, regs); + + return 0; +} +NOKPROBE_SYMBOL(trace_fprobe_entry_handler) + static nokprobe_inline void __fexit_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, unsigned long ret_ip, struct pt_regs *regs, - struct trace_event_file *trace_file) + void *entry_data, struct trace_event_file *trace_file) { struct fexit_trace_entry_head *entry; struct trace_event_buffer fbuffer; @@ -227,7 +244,7 @@ __fexit_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, if (trace_trigger_soft_disabled(trace_file)) return; - dsize = __get_data_size(&tf->tp, regs); + dsize = __get_data_size(&tf->tp, regs, entry_data); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + tf->tp.size + dsize); @@ -238,19 +255,19 @@ __fexit_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event); entry->func = entry_ip; entry->ret_ip = ret_ip; - store_trace_args(&entry[1], &tf->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tf->tp, regs, entry_data, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } static void fexit_trace_func(struct trace_fprobe *tf, unsigned long entry_ip, - unsigned long ret_ip, struct pt_regs *regs) + unsigned long ret_ip, struct pt_regs *regs, void *entry_data) { struct event_file_link *link; trace_probe_for_each_link_rcu(link, &tf->tp) - __fexit_trace_func(tf, entry_ip, ret_ip, regs, link->file); + __fexit_trace_func(tf, entry_ip, ret_ip, regs, entry_data, link->file); } NOKPROBE_SYMBOL(fexit_trace_func); @@ -269,7 +286,7 @@ static int fentry_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, if (hlist_empty(head)) return 0; - dsize = __get_data_size(&tf->tp, regs); + dsize = __get_data_size(&tf->tp, regs, NULL); __size = sizeof(*entry) + tf->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); @@ -280,7 +297,7 @@ static int fentry_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, entry->ip = entry_ip; memset(&entry[1], 0, dsize); - store_trace_args(&entry[1], &tf->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tf->tp, regs, NULL, sizeof(*entry), dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); return 0; @@ -289,7 +306,8 @@ NOKPROBE_SYMBOL(fentry_perf_func); static void fexit_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, - unsigned long ret_ip, struct pt_regs *regs) + unsigned long ret_ip, struct pt_regs *regs, + void *entry_data) { struct trace_event_call *call = trace_probe_event_call(&tf->tp); struct fexit_trace_entry_head *entry; @@ -301,7 +319,7 @@ fexit_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, if (hlist_empty(head)) return; - dsize = __get_data_size(&tf->tp, regs); + dsize = __get_data_size(&tf->tp, regs, entry_data); __size = sizeof(*entry) + tf->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); @@ -312,7 +330,7 @@ fexit_perf_func(struct trace_fprobe *tf, unsigned long entry_ip, entry->func = entry_ip; entry->ret_ip = ret_ip; - store_trace_args(&entry[1], &tf->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tf->tp, regs, entry_data, sizeof(*entry), dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); } @@ -343,10 +361,10 @@ static void fexit_dispatcher(struct fprobe *fp, unsigned long entry_ip, struct trace_fprobe *tf = container_of(fp, struct trace_fprobe, fp); if (trace_probe_test_flag(&tf->tp, TP_FLAG_TRACE)) - fexit_trace_func(tf, entry_ip, ret_ip, regs); + fexit_trace_func(tf, entry_ip, ret_ip, regs, entry_data); #ifdef CONFIG_PERF_EVENTS if (trace_probe_test_flag(&tf->tp, TP_FLAG_PROFILE)) - fexit_perf_func(tf, entry_ip, ret_ip, regs); + fexit_perf_func(tf, entry_ip, ret_ip, regs, entry_data); #endif } NOKPROBE_SYMBOL(fexit_dispatcher); @@ -1109,6 +1127,11 @@ static int __trace_fprobe_create(int argc, const char *argv[]) goto error; /* This can be -ENOMEM */ } + if (is_return && tf->tp.entry_arg) { + tf->fp.entry_handler = trace_fprobe_entry_handler; + tf->fp.entry_data_size = traceprobe_get_entry_data_size(&tf->tp); + } + ret = traceprobe_set_print_fmt(&tf->tp, is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL); if (ret < 0) diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 843f13f839df..14099cc17fc9 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -740,6 +740,9 @@ static unsigned int number_of_same_symbols(char *func_name) return ctx.count; } +static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, + struct pt_regs *regs); + static int __trace_kprobe_create(int argc, const char *argv[]) { /* @@ -948,6 +951,11 @@ static int __trace_kprobe_create(int argc, const char *argv[]) if (ret) goto error; /* This can be -ENOMEM */ } + /* entry handler for kretprobe */ + if (is_return && tk->tp.entry_arg) { + tk->rp.entry_handler = trace_kprobe_entry_handler; + tk->rp.data_size = traceprobe_get_entry_data_size(&tk->tp); + } ptype = is_return ? PROBE_PRINT_RETURN : PROBE_PRINT_NORMAL; ret = traceprobe_set_print_fmt(&tk->tp, ptype); @@ -1303,8 +1311,8 @@ static const struct file_operations kprobe_profile_ops = { /* Note that we don't verify it, since the code does not come from user space */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { struct pt_regs *regs = rec; unsigned long val; @@ -1329,6 +1337,9 @@ retry: case FETCH_OP_ARG: val = regs_get_kernel_argument(regs, code->param); break; + case FETCH_OP_EDATA: + val = *(unsigned long *)((unsigned long)edata + code->offset); + break; #endif case FETCH_NOP_SYMBOL: /* Ignore a place holder */ code++; @@ -1359,7 +1370,7 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, if (trace_trigger_soft_disabled(trace_file)) return; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, NULL); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + tk->tp.size + dsize); @@ -1368,7 +1379,7 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, fbuffer.regs = regs; entry->ip = (unsigned long)tk->rp.kp.addr; - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, NULL, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } @@ -1384,6 +1395,31 @@ kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs) NOKPROBE_SYMBOL(kprobe_trace_func); /* Kretprobe handler */ + +static int trace_kprobe_entry_handler(struct kretprobe_instance *ri, + struct pt_regs *regs) +{ + struct kretprobe *rp = get_kretprobe(ri); + struct trace_kprobe *tk; + + /* + * There is a small chance that get_kretprobe(ri) returns NULL when + * the kretprobe is unregister on another CPU between kretprobe's + * trampoline_handler and this function. + */ + if (unlikely(!rp)) + return -ENOENT; + + tk = container_of(rp, struct trace_kprobe, rp); + + /* store argument values into ri->data as entry data */ + if (tk->tp.entry_arg) + store_trace_entry_data(ri->data, &tk->tp, regs); + + return 0; +} + + static nokprobe_inline void __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct pt_regs *regs, @@ -1399,7 +1435,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, if (trace_trigger_soft_disabled(trace_file)) return; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, ri->data); entry = trace_event_buffer_reserve(&fbuffer, trace_file, sizeof(*entry) + tk->tp.size + dsize); @@ -1409,7 +1445,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, fbuffer.regs = regs; entry->func = (unsigned long)tk->rp.kp.addr; entry->ret_ip = get_kretprobe_retaddr(ri); - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, ri->data, sizeof(*entry), dsize); trace_event_buffer_commit(&fbuffer); } @@ -1557,7 +1593,7 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) if (hlist_empty(head)) return 0; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, NULL); __size = sizeof(*entry) + tk->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); @@ -1568,7 +1604,7 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) entry->ip = (unsigned long)tk->rp.kp.addr; memset(&entry[1], 0, dsize); - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, NULL, sizeof(*entry), dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); return 0; @@ -1593,7 +1629,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, if (hlist_empty(head)) return; - dsize = __get_data_size(&tk->tp, regs); + dsize = __get_data_size(&tk->tp, regs, ri->data); __size = sizeof(*entry) + tk->tp.size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); @@ -1604,7 +1640,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, entry->func = (unsigned long)tk->rp.kp.addr; entry->ret_ip = get_kretprobe_retaddr(ri); - store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize); + store_trace_args(&entry[1], &tk->tp, regs, ri->data, sizeof(*entry), dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL); } diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 93f36f8a108e..217169de0920 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -594,6 +594,8 @@ static int parse_btf_field(char *fieldname, const struct btf_type *type, return 0; } +static int __store_entry_arg(struct trace_probe *tp, int argnum); + static int parse_btf_arg(char *varname, struct fetch_insn **pcode, struct fetch_insn *end, struct traceprobe_parse_context *ctx) @@ -618,11 +620,7 @@ static int parse_btf_arg(char *varname, return -EOPNOTSUPP; } - if (ctx->flags & TPARG_FL_RETURN) { - if (strcmp(varname, "$retval") != 0) { - trace_probe_log_err(ctx->offset, NO_BTFARG); - return -ENOENT; - } + if (ctx->flags & TPARG_FL_RETURN && !strcmp(varname, "$retval")) { code->op = FETCH_OP_RETVAL; /* Check whether the function return type is not void */ if (query_btf_context(ctx) == 0) { @@ -654,11 +652,21 @@ static int parse_btf_arg(char *varname, const char *name = btf_name_by_offset(ctx->btf, params[i].name_off); if (name && !strcmp(name, varname)) { - code->op = FETCH_OP_ARG; - if (ctx->flags & TPARG_FL_TPOINT) - code->param = i + 1; - else - code->param = i; + if (tparg_is_function_entry(ctx->flags)) { + code->op = FETCH_OP_ARG; + if (ctx->flags & TPARG_FL_TPOINT) + code->param = i + 1; + else + code->param = i; + } else if (tparg_is_function_return(ctx->flags)) { + code->op = FETCH_OP_EDATA; + ret = __store_entry_arg(ctx->tp, i); + if (ret < 0) { + /* internal error */ + return ret; + } + code->offset = ret; + } tid = params[i].type; goto found; } @@ -755,6 +763,110 @@ static int check_prepare_btf_string_fetch(char *typename, #endif +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API + +static int __store_entry_arg(struct trace_probe *tp, int argnum) +{ + struct probe_entry_arg *earg = tp->entry_arg; + bool match = false; + int i, offset; + + if (!earg) { + earg = kzalloc(sizeof(*tp->entry_arg), GFP_KERNEL); + if (!earg) + return -ENOMEM; + earg->size = 2 * tp->nr_args + 1; + earg->code = kcalloc(earg->size, sizeof(struct fetch_insn), + GFP_KERNEL); + if (!earg->code) { + kfree(earg); + return -ENOMEM; + } + /* Fill the code buffer with 'end' to simplify it */ + for (i = 0; i < earg->size; i++) + earg->code[i].op = FETCH_OP_END; + tp->entry_arg = earg; + } + + offset = 0; + for (i = 0; i < earg->size - 1; i++) { + switch (earg->code[i].op) { + case FETCH_OP_END: + earg->code[i].op = FETCH_OP_ARG; + earg->code[i].param = argnum; + earg->code[i + 1].op = FETCH_OP_ST_EDATA; + earg->code[i + 1].offset = offset; + return offset; + case FETCH_OP_ARG: + match = (earg->code[i].param == argnum); + break; + case FETCH_OP_ST_EDATA: + offset = earg->code[i].offset; + if (match) + return offset; + offset += sizeof(unsigned long); + break; + default: + break; + } + } + return -ENOSPC; +} + +int traceprobe_get_entry_data_size(struct trace_probe *tp) +{ + struct probe_entry_arg *earg = tp->entry_arg; + int i, size = 0; + + if (!earg) + return 0; + + for (i = 0; i < earg->size; i++) { + switch (earg->code[i].op) { + case FETCH_OP_END: + goto out; + case FETCH_OP_ST_EDATA: + size = earg->code[i].offset + sizeof(unsigned long); + break; + default: + break; + } + } +out: + return size; +} + +void store_trace_entry_data(void *edata, struct trace_probe *tp, struct pt_regs *regs) +{ + struct probe_entry_arg *earg = tp->entry_arg; + unsigned long val; + int i; + + if (!earg) + return; + + for (i = 0; i < earg->size; i++) { + struct fetch_insn *code = &earg->code[i]; + + switch (code->op) { + case FETCH_OP_ARG: + val = regs_get_kernel_argument(regs, code->param); + break; + case FETCH_OP_ST_EDATA: + *(unsigned long *)((unsigned long)edata + code->offset) = val; + break; + case FETCH_OP_END: + goto end; + default: + break; + } + } +end: + return; +} +NOKPROBE_SYMBOL(store_trace_entry_data) +#endif + #define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long)) /* Parse $vars. @orig_arg points '$', which syncs to @ctx->offset */ @@ -830,7 +942,7 @@ static int parse_probe_vars(char *orig_arg, const struct fetch_type *t, #ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API len = str_has_prefix(arg, "arg"); - if (len && tparg_is_function_entry(ctx->flags)) { + if (len) { ret = kstrtoul(arg + len, 10, ¶m); if (ret) goto inval; @@ -839,15 +951,29 @@ static int parse_probe_vars(char *orig_arg, const struct fetch_type *t, err = TP_ERR_BAD_ARG_NUM; goto inval; } + param--; /* argN starts from 1, but internal arg[N] starts from 0 */ - code->op = FETCH_OP_ARG; - code->param = (unsigned int)param - 1; - /* - * The tracepoint probe will probe a stub function, and the - * first parameter of the stub is a dummy and should be ignored. - */ - if (ctx->flags & TPARG_FL_TPOINT) - code->param++; + if (tparg_is_function_entry(ctx->flags)) { + code->op = FETCH_OP_ARG; + code->param = (unsigned int)param; + /* + * The tracepoint probe will probe a stub function, and the + * first parameter of the stub is a dummy and should be ignored. + */ + if (ctx->flags & TPARG_FL_TPOINT) + code->param++; + } else if (tparg_is_function_return(ctx->flags)) { + /* function entry argument access from return probe */ + ret = __store_entry_arg(ctx->tp, param); + if (ret < 0) /* This error should be an internal error */ + return ret; + + code->op = FETCH_OP_EDATA; + code->offset = ret; + } else { + err = TP_ERR_NOFENTRY_ARGS; + goto inval; + } return 0; } #endif @@ -1037,7 +1163,8 @@ parse_probe_arg(char *arg, const struct fetch_type *type, break; default: if (isalpha(arg[0]) || arg[0] == '_') { /* BTF variable */ - if (!tparg_is_function_entry(ctx->flags)) { + if (!tparg_is_function_entry(ctx->flags) && + !tparg_is_function_return(ctx->flags)) { trace_probe_log_err(ctx->offset, NOSUP_BTFARG); return -EINVAL; } @@ -1423,6 +1550,7 @@ int traceprobe_parse_probe_arg(struct trace_probe *tp, int i, const char *arg, struct probe_arg *parg = &tp->args[i]; const char *body; + ctx->tp = tp; body = strchr(arg, '='); if (body) { if (body - arg > MAX_ARG_NAME_LEN) { @@ -1479,7 +1607,8 @@ static int argv_has_var_arg(int argc, const char *argv[], int *args_idx, if (str_has_prefix(argv[i], "$arg")) { trace_probe_log_set_index(i + 2); - if (!tparg_is_function_entry(ctx->flags)) { + if (!tparg_is_function_entry(ctx->flags) && + !tparg_is_function_return(ctx->flags)) { trace_probe_log_err(0, NOFENTRY_ARGS); return -EINVAL; } @@ -1802,6 +1931,12 @@ void trace_probe_cleanup(struct trace_probe *tp) for (i = 0; i < tp->nr_args; i++) traceprobe_free_probe_arg(&tp->args[i]); + if (tp->entry_arg) { + kfree(tp->entry_arg->code); + kfree(tp->entry_arg); + tp->entry_arg = NULL; + } + if (tp->event) trace_probe_unlink(tp); } diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index ed8d1052f8a7..cef3a50628a3 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -92,6 +92,7 @@ enum fetch_op { FETCH_OP_ARG, /* Function argument : .param */ FETCH_OP_FOFFS, /* File offset: .immediate */ FETCH_OP_DATA, /* Allocated data: .data */ + FETCH_OP_EDATA, /* Entry data: .offset */ // Stage 2 (dereference) op FETCH_OP_DEREF, /* Dereference: .offset */ FETCH_OP_UDEREF, /* User-space Dereference: .offset */ @@ -102,6 +103,7 @@ enum fetch_op { FETCH_OP_ST_STRING, /* String: .offset, .size */ FETCH_OP_ST_USTRING, /* User String: .offset, .size */ FETCH_OP_ST_SYMSTR, /* Kernel Symbol String: .offset, .size */ + FETCH_OP_ST_EDATA, /* Store Entry Data: .offset */ // Stage 4 (modify) op FETCH_OP_MOD_BF, /* Bitfield: .basesize, .lshift, .rshift */ // Stage 5 (loop) op @@ -232,6 +234,11 @@ struct probe_arg { const struct fetch_type *type; /* Type of this argument */ }; +struct probe_entry_arg { + struct fetch_insn *code; + unsigned int size; /* The entry data size */ +}; + struct trace_uprobe_filter { rwlock_t rwlock; int nr_systemwide; @@ -253,6 +260,7 @@ struct trace_probe { struct trace_probe_event *event; ssize_t size; /* trace entry size */ unsigned int nr_args; + struct probe_entry_arg *entry_arg; /* This is only for return probe */ struct probe_arg args[]; }; @@ -355,6 +363,18 @@ int trace_probe_create(const char *raw_command, int (*createfn)(int, const char int trace_probe_print_args(struct trace_seq *s, struct probe_arg *args, int nr_args, u8 *data, void *field); +#ifdef CONFIG_HAVE_FUNCTION_ARG_ACCESS_API +int traceprobe_get_entry_data_size(struct trace_probe *tp); +/* This is a runtime function to store entry data */ +void store_trace_entry_data(void *edata, struct trace_probe *tp, struct pt_regs *regs); +#else /* !CONFIG_HAVE_FUNCTION_ARG_ACCESS_API */ +static inline int traceprobe_get_entry_data_size(struct trace_probe *tp) +{ + return 0; +} +#define store_trace_entry_data(edata, tp, regs) do { } while (0) +#endif + #define trace_probe_for_each_link(pos, tp) \ list_for_each_entry(pos, &(tp)->event->files, list) #define trace_probe_for_each_link_rcu(pos, tp) \ @@ -381,6 +401,11 @@ static inline bool tparg_is_function_entry(unsigned int flags) return (flags & TPARG_FL_LOC_MASK) == (TPARG_FL_KERNEL | TPARG_FL_FENTRY); } +static inline bool tparg_is_function_return(unsigned int flags) +{ + return (flags & TPARG_FL_LOC_MASK) == (TPARG_FL_KERNEL | TPARG_FL_RETURN); +} + struct traceprobe_parse_context { struct trace_event_call *event; /* BTF related parameters */ @@ -392,6 +417,7 @@ struct traceprobe_parse_context { const struct btf_type *last_type; /* Saved type */ u32 last_bitoffs; /* Saved bitoffs */ u32 last_bitsize; /* Saved bitsize */ + struct trace_probe *tp; unsigned int flags; int offset; }; @@ -506,7 +532,7 @@ extern int traceprobe_define_arg_fields(struct trace_event_call *event_call, C(NO_BTFARG, "This variable is not found at this probe point"),\ C(NO_BTF_ENTRY, "No BTF entry for this probe point"), \ C(BAD_VAR_ARGS, "$arg* must be an independent parameter without name etc."),\ - C(NOFENTRY_ARGS, "$arg* can be used only on function entry"), \ + C(NOFENTRY_ARGS, "$arg* can be used only on function entry or exit"), \ C(DOUBLE_ARGS, "$arg* can be used only once in the parameters"), \ C(ARGS_2LONG, "$arg* failed because the argument list is too long"), \ C(ARGIDX_2BIG, "$argN index is too big"), \ diff --git a/kernel/trace/trace_probe_tmpl.h b/kernel/trace/trace_probe_tmpl.h index 3935b347f874..2caf0d2afb32 100644 --- a/kernel/trace/trace_probe_tmpl.h +++ b/kernel/trace/trace_probe_tmpl.h @@ -54,7 +54,7 @@ fetch_apply_bitfield(struct fetch_insn *code, void *buf) * If dest is NULL, don't store result and return required dynamic data size. */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, void *dest, void *base); static nokprobe_inline int fetch_store_strlen(unsigned long addr); static nokprobe_inline int @@ -232,7 +232,7 @@ array: /* Sum up total data length for dynamic arrays (strings) */ static nokprobe_inline int -__get_data_size(struct trace_probe *tp, struct pt_regs *regs) +__get_data_size(struct trace_probe *tp, struct pt_regs *regs, void *edata) { struct probe_arg *arg; int i, len, ret = 0; @@ -240,7 +240,7 @@ __get_data_size(struct trace_probe *tp, struct pt_regs *regs) for (i = 0; i < tp->nr_args; i++) { arg = tp->args + i; if (unlikely(arg->dynamic)) { - len = process_fetch_insn(arg->code, regs, NULL, NULL); + len = process_fetch_insn(arg->code, regs, edata, NULL, NULL); if (len > 0) ret += len; } @@ -251,7 +251,7 @@ __get_data_size(struct trace_probe *tp, struct pt_regs *regs) /* Store the value of each argument */ static nokprobe_inline void -store_trace_args(void *data, struct trace_probe *tp, void *rec, +store_trace_args(void *data, struct trace_probe *tp, void *rec, void *edata, int header_size, int maxlen) { struct probe_arg *arg; @@ -266,7 +266,7 @@ store_trace_args(void *data, struct trace_probe *tp, void *rec, /* Point the dynamic data area if needed */ if (unlikely(arg->dynamic)) *dl = make_data_loc(maxlen, dyndata - base); - ret = process_fetch_insn(arg->code, rec, dl, base); + ret = process_fetch_insn(arg->code, rec, edata, dl, base); if (arg->dynamic && likely(ret > 0)) { dyndata += ret; maxlen -= ret; diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 796ebcae9b38..9e461362450a 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -211,8 +211,8 @@ static unsigned long translate_user_vaddr(unsigned long file_offset) /* Note that we don't verify it, since the code does not come from user space */ static int -process_fetch_insn(struct fetch_insn *code, void *rec, void *dest, - void *base) +process_fetch_insn(struct fetch_insn *code, void *rec, void *edata, + void *dest, void *base) { struct pt_regs *regs = rec; unsigned long val; @@ -1490,11 +1490,11 @@ static int uprobe_dispatcher(struct uprobe_consumer *con, struct pt_regs *regs) if (WARN_ON_ONCE(!uprobe_cpu_buffer)) return 0; - dsize = __get_data_size(&tu->tp, regs); + dsize = __get_data_size(&tu->tp, regs, NULL); esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); ucb = uprobe_buffer_get(); - store_trace_args(ucb->buf, &tu->tp, regs, esize, dsize); + store_trace_args(ucb->buf, &tu->tp, regs, NULL, esize, dsize); if (trace_probe_test_flag(&tu->tp, TP_FLAG_TRACE)) ret |= uprobe_trace_func(tu, regs, ucb, dsize); @@ -1525,11 +1525,11 @@ static int uretprobe_dispatcher(struct uprobe_consumer *con, if (WARN_ON_ONCE(!uprobe_cpu_buffer)) return 0; - dsize = __get_data_size(&tu->tp, regs); + dsize = __get_data_size(&tu->tp, regs, NULL); esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); ucb = uprobe_buffer_get(); - store_trace_args(ucb->buf, &tu->tp, regs, esize, dsize); + store_trace_args(ucb->buf, &tu->tp, regs, NULL, esize, dsize); if (trace_probe_test_flag(&tu->tp, TP_FLAG_TRACE)) uretprobe_trace_func(tu, func, regs, ucb, dsize); diff --git a/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_syntax_errors.tc b/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_syntax_errors.tc index 20e42c030095..61877d166451 100644 --- a/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_syntax_errors.tc +++ b/tools/testing/selftests/ftrace/test.d/dynevent/fprobe_syntax_errors.tc @@ -34,7 +34,9 @@ check_error 'f vfs_read ^$stack10000' # BAD_STACK_NUM check_error 'f vfs_read ^$arg10000' # BAD_ARG_NUM +if !grep -q 'kernel return probes support:' README; then check_error 'f vfs_read $retval ^$arg1' # BAD_VAR +fi check_error 'f vfs_read ^$none_var' # BAD_VAR check_error 'f vfs_read ^'$REG # BAD_VAR @@ -99,7 +101,9 @@ if grep -q "" README; then check_error 'f vfs_read args=^$arg*' # BAD_VAR_ARGS check_error 'f vfs_read +0(^$arg*)' # BAD_VAR_ARGS check_error 'f vfs_read $arg* ^$arg*' # DOUBLE_ARGS +if !grep -q 'kernel return probes support:' README; then check_error 'f vfs_read%return ^$arg*' # NOFENTRY_ARGS +fi check_error 'f vfs_read ^hoge' # NO_BTFARG check_error 'f kfree ^$arg10' # NO_BTFARG (exceed the number of parameters) check_error 'f kfree%return ^$retval' # NO_RETVAL diff --git a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_syntax_errors.tc b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_syntax_errors.tc index 65fbb26fd58c..a16c6a6f6055 100644 --- a/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_syntax_errors.tc +++ b/tools/testing/selftests/ftrace/test.d/kprobe/kprobe_syntax_errors.tc @@ -108,7 +108,9 @@ if grep -q "" README; then check_error 'p vfs_read args=^$arg*' # BAD_VAR_ARGS check_error 'p vfs_read +0(^$arg*)' # BAD_VAR_ARGS check_error 'p vfs_read $arg* ^$arg*' # DOUBLE_ARGS +if !grep -q 'kernel return probes support:' README; then check_error 'r vfs_read ^$arg*' # NOFENTRY_ARGS +fi check_error 'p vfs_read+8 ^$arg*' # NOFENTRY_ARGS check_error 'p vfs_read ^hoge' # NO_BTFARG check_error 'p kfree ^$arg10' # NO_BTFARG (exceed the number of parameters) -- cgit v1.2.3 From 095fe4891282be510af0db1b03587b512c0de31d Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Mon, 4 Mar 2024 22:34:33 -0500 Subject: tracing: Limit trace_marker writes to just 4K Limit the max print event of trace_marker to just 4K string size. This must also be less than the amount that can be held by a trace_seq along with the text that is before the output (like the task name, PID, CPU, state, etc). As trace_seq is made to handle large events (some greater than 4K). Make the max size of a trace_marker write event be 4K which is guaranteed to fit in the trace_seq buffer. Link: https://lore.kernel.org/linux-trace-kernel/20240304223433.4ba47dff@gandalf.local.home Suggested-by: Linus Torvalds Reviewed-by: Mathieu Desnoyers Reviewed-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 8198bfc54b58..d16b95ca58a7 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -7293,6 +7293,8 @@ tracing_free_buffer_release(struct inode *inode, struct file *filp) return 0; } +#define TRACE_MARKER_MAX_SIZE 4096 + static ssize_t tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) @@ -7320,6 +7322,9 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, if ((ssize_t)cnt < 0) return -EINVAL; + if (cnt > TRACE_MARKER_MAX_SIZE) + cnt = TRACE_MARKER_MAX_SIZE; + meta_size = sizeof(*entry) + 2; /* add '\0' and possible '\n' */ again: size = cnt + meta_size; @@ -7328,11 +7333,6 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, if (cnt < FAULTED_SIZE) size += FAULTED_SIZE - cnt; - if (size > TRACE_SEQ_BUFFER_SIZE) { - cnt -= size - TRACE_SEQ_BUFFER_SIZE; - goto again; - } - buffer = tr->array_buffer.buffer; event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, tracing_gen_ctx()); -- cgit v1.2.3 From e5d7c1916562f0e856eb3d6f569629fcd535fed2 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Fri, 8 Mar 2024 15:24:05 -0500 Subject: tracing: Use .flush() call to wake up readers The .release() function does not get called until all readers of a file descriptor are finished. If a thread is blocked on reading a file descriptor in ring_buffer_wait(), and another thread closes the file descriptor, it will not wake up the other thread as ring_buffer_wake_waiters() is called by .release(), and that will not get called until the .read() is finished. The issue originally showed up in trace-cmd, but the readers are actually other processes with their own file descriptors. So calling close() would wake up the other tasks because they are blocked on another descriptor then the one that was closed(). But there's other wake ups that solve that issue. When a thread is blocked on a read, it can still hang even when another thread closed its descriptor. This is what the .flush() callback is for. Have the .flush() wake up the readers. Link: https://lore.kernel.org/linux-trace-kernel/20240308202432.107909457@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Linus Torvalds Cc: linke li Cc: Rabin Vincent Fixes: f3ddb74ad0790 ("tracing: Wake up ring buffer waiters on closing of the file") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d16b95ca58a7..c9c898307348 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -8393,6 +8393,20 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, return size; } +static int tracing_buffers_flush(struct file *file, fl_owner_t id) +{ + struct ftrace_buffer_info *info = file->private_data; + struct trace_iterator *iter = &info->iter; + + iter->wait_index++; + /* Make sure the waiters see the new wait_index */ + smp_wmb(); + + ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); + + return 0; +} + static int tracing_buffers_release(struct inode *inode, struct file *file) { struct ftrace_buffer_info *info = file->private_data; @@ -8404,12 +8418,6 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) __trace_array_put(iter->tr); - iter->wait_index++; - /* Make sure the waiters see the new wait_index */ - smp_wmb(); - - ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); - if (info->spare) ring_buffer_free_read_page(iter->array_buffer->buffer, info->spare_cpu, info->spare); @@ -8625,6 +8633,7 @@ static const struct file_operations tracing_buffers_fops = { .read = tracing_buffers_read, .poll = tracing_buffers_poll, .release = tracing_buffers_release, + .flush = tracing_buffers_flush, .splice_read = tracing_buffers_splice_read, .unlocked_ioctl = tracing_buffers_ioctl, .llseek = no_llseek, -- cgit v1.2.3 From 2aa043a55b9a764c9cbde5a8c654eeaaffe224cf Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 12 Mar 2024 08:15:08 -0400 Subject: tracing/ring-buffer: Fix wait_on_pipe() race When the trace_pipe_raw file is closed, there should be no new readers on the file descriptor. This is mostly handled with the waking and wait_index fields of the iterator. But there's still a slight race. CPU 0 CPU 1 ----- ----- wait_index++; index = wait_index; ring_buffer_wake_waiters(); wait_on_pipe() ring_buffer_wait(); The ring_buffer_wait() will miss the wakeup from CPU 1. The problem is that the ring_buffer_wait() needs the logic of: prepare_to_wait(); if (!condition) schedule(); Where the missing condition check is the iter->wait_index update. Have the ring_buffer_wait() take a conditional callback function and a data parameter that can be used within the wait_event_interruptible() of the ring_buffer_wait() function. In wait_on_pipe(), pass a condition function that will check if the wait_index has been updated, if it has, it will return true to break out of the wait_event_interruptible() loop. Create a new field "closed" in the trace_iterator and set it in the .flush() callback before calling ring_buffer_wake_waiters(). This will keep any new readers from waiting on a closed file descriptor. Have the wait_on_pipe() condition callback also check the closed field. Change the wait_index field of the trace_iterator to atomic_t. There's no reason it needs to be 'long' and making it atomic and using atomic_read_acquire() and atomic_fetch_inc_release() will provide the necessary memory barriers. Add a "woken" flag to tracing_buffers_splice_read() to exit the loop after one more try to fetch data. That is, if it waited for data and something woke it up, it should try to collect any new data and then exit back to user space. Link: https://lore.kernel.org/linux-trace-kernel/CAHk-=wgsNgewHFxZAJiAQznwPMqEtQmi1waeS2O1v6L4c_Um5A@mail.gmail.com/ Link: https://lore.kernel.org/linux-trace-kernel/20240312121703.557950713@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Linus Torvalds Cc: linke li Cc: Rabin Vincent Fixes: f3ddb74ad0790 ("tracing: Wake up ring buffer waiters on closing of the file") Signed-off-by: Steven Rostedt (Google) --- include/linux/ring_buffer.h | 3 ++- include/linux/trace_events.h | 5 ++++- kernel/trace/ring_buffer.c | 13 ++++++++----- kernel/trace/trace.c | 43 +++++++++++++++++++++++++++++++------------ 4 files changed, 45 insertions(+), 19 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 338a33db1577..dc5ae4e96aee 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -99,7 +99,8 @@ __ring_buffer_alloc(unsigned long size, unsigned flags, struct lock_class_key *k }) typedef bool (*ring_buffer_cond_fn)(void *data); -int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full); +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full, + ring_buffer_cond_fn cond, void *data); __poll_t ring_buffer_poll_wait(struct trace_buffer *buffer, int cpu, struct file *filp, poll_table *poll_table, int full); void ring_buffer_wake_waiters(struct trace_buffer *buffer, int cpu); diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index d68ff9b1247f..fc6d0af56bb1 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -103,13 +103,16 @@ struct trace_iterator { unsigned int temp_size; char *fmt; /* modified format holder */ unsigned int fmt_size; - long wait_index; + atomic_t wait_index; /* trace_seq for __print_flags() and __print_symbolic() etc. */ struct trace_seq tmp_seq; cpumask_var_t started; + /* Set when the file is closed to prevent new waiters */ + bool closed; + /* it's true when current open file is snapshot */ bool snapshot; diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index f4c34b7c7e1e..350607cce869 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -902,23 +902,26 @@ static bool rb_wait_once(void *data) * @buffer: buffer to wait on * @cpu: the cpu buffer to wait on * @full: wait until the percentage of pages are available, if @cpu != RING_BUFFER_ALL_CPUS + * @cond: condition function to break out of wait (NULL to run once) + * @data: the data to pass to @cond. * * If @cpu == RING_BUFFER_ALL_CPUS then the task will wake up as soon * as data is added to any of the @buffer's cpu buffers. Otherwise * it will wait for data to be added to a specific cpu buffer. */ -int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full) +int ring_buffer_wait(struct trace_buffer *buffer, int cpu, int full, + ring_buffer_cond_fn cond, void *data) { struct ring_buffer_per_cpu *cpu_buffer; struct wait_queue_head *waitq; - ring_buffer_cond_fn cond; struct rb_irq_work *rbwork; - void *data; long once = 0; int ret = 0; - cond = rb_wait_once; - data = &once; + if (!cond) { + cond = rb_wait_once; + data = &once; + } /* * Depending on what the caller is waiting for, either any diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index c9c898307348..d390fea3a6a5 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1955,15 +1955,36 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) #endif /* CONFIG_TRACER_MAX_TRACE */ +struct pipe_wait { + struct trace_iterator *iter; + int wait_index; +}; + +static bool wait_pipe_cond(void *data) +{ + struct pipe_wait *pwait = data; + struct trace_iterator *iter = pwait->iter; + + if (atomic_read_acquire(&iter->wait_index) != pwait->wait_index) + return true; + + return iter->closed; +} + static int wait_on_pipe(struct trace_iterator *iter, int full) { + struct pipe_wait pwait; int ret; /* Iterators are static, they should be filled or empty */ if (trace_buffer_iter(iter, iter->cpu_file)) return 0; - ret = ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, full); + pwait.wait_index = atomic_read_acquire(&iter->wait_index); + pwait.iter = iter; + + ret = ring_buffer_wait(iter->array_buffer->buffer, iter->cpu_file, full, + wait_pipe_cond, &pwait); #ifdef CONFIG_TRACER_MAX_TRACE /* @@ -8398,9 +8419,9 @@ static int tracing_buffers_flush(struct file *file, fl_owner_t id) struct ftrace_buffer_info *info = file->private_data; struct trace_iterator *iter = &info->iter; - iter->wait_index++; + iter->closed = true; /* Make sure the waiters see the new wait_index */ - smp_wmb(); + (void)atomic_fetch_inc_release(&iter->wait_index); ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); @@ -8500,6 +8521,7 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, .spd_release = buffer_spd_release, }; struct buffer_ref *ref; + bool woken = false; int page_size; int entries, i; ssize_t ret = 0; @@ -8573,17 +8595,17 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, /* did we read anything? */ if (!spd.nr_pages) { - long wait_index; if (ret) goto out; + if (woken) + goto out; + ret = -EAGAIN; if ((file->f_flags & O_NONBLOCK) || (flags & SPLICE_F_NONBLOCK)) goto out; - wait_index = READ_ONCE(iter->wait_index); - ret = wait_on_pipe(iter, iter->snapshot ? 0 : iter->tr->buffer_percent); if (ret) goto out; @@ -8592,10 +8614,8 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, if (!tracer_tracing_is_on(iter->tr)) goto out; - /* Make sure we see the new wait_index */ - smp_rmb(); - if (wait_index != iter->wait_index) - goto out; + /* Iterate one more time to collect any new data then exit */ + woken = true; goto again; } @@ -8618,9 +8638,8 @@ static long tracing_buffers_ioctl(struct file *file, unsigned int cmd, unsigned mutex_lock(&trace_types_lock); - iter->wait_index++; /* Make sure the waiters see the new wait_index */ - smp_wmb(); + (void)atomic_fetch_inc_release(&iter->wait_index); ring_buffer_wake_waiters(iter->array_buffer->buffer, iter->cpu_file); -- cgit v1.2.3 From 0b18c852cc6fb8284ac0ab97e3e840974a6a8a64 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 20 Feb 2024 09:06:14 -0500 Subject: tracing: Have saved_cmdlines arrays all in one allocation The saved_cmdlines have three arrays for mapping PIDs to COMMs: - map_pid_to_cmdline[] - map_cmdline_to_pid[] - saved_cmdlines The map_pid_to_cmdline[] is PID_MAX_DEFAULT in size and holds the index into the other arrays. The map_cmdline_to_pid[] is a mapping back to the full pid as it can be larger than PID_MAX_DEFAULT. And the saved_cmdlines[] just holds the COMMs associated to the pids. Currently the map_pid_to_cmdline[] and saved_cmdlines[] are allocated together (in reality the saved_cmdlines is just in the memory of the rounding of the allocation of the structure as it is always allocated in powers of two). The map_cmdline_to_pid[] array is allocated separately. Since the rounding to a power of two is rather large (it allows for 8000 elements in saved_cmdlines), also include the map_cmdline_to_pid[] array. (This drops it to 6000 by default, which is still plenty for most use cases). This saves even more memory as the map_cmdline_to_pid[] array doesn't need to be allocated. Link: https://lore.kernel.org/linux-trace-kernel/20240212174011.068211d9@gandalf.local.home/ Link: https://lore.kernel.org/linux-trace-kernel/20240220140703.182330529@goodmis.org Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Tim Chen Cc: Vincent Donnefort Cc: Sven Schnelle Cc: Mete Durlu Fixes: 44dc5c41b5b1 ("tracing: Fix wasted memory in saved_cmdlines logic") Acked-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index ab4c1a1fbda8..70202e60a60a 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2346,6 +2346,10 @@ struct saved_cmdlines_buffer { }; static struct saved_cmdlines_buffer *savedcmd; +/* Holds the size of a cmdline and pid element */ +#define SAVED_CMDLINE_MAP_ELEMENT_SIZE(s) \ + (TASK_COMM_LEN + sizeof((s)->map_cmdline_to_pid[0])) + static inline char *get_saved_cmdlines(int idx) { return &savedcmd->saved_cmdlines[idx * TASK_COMM_LEN]; @@ -2360,7 +2364,6 @@ static void free_saved_cmdlines_buffer(struct saved_cmdlines_buffer *s) { int order = get_order(sizeof(*s) + s->cmdline_num * TASK_COMM_LEN); - kfree(s->map_cmdline_to_pid); kmemleak_free(s); free_pages((unsigned long)s, order); } @@ -2373,7 +2376,7 @@ static struct saved_cmdlines_buffer *allocate_cmdlines_buffer(unsigned int val) int order; /* Figure out how much is needed to hold the given number of cmdlines */ - orig_size = sizeof(*s) + val * TASK_COMM_LEN; + orig_size = sizeof(*s) + val * SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); order = get_order(orig_size); size = 1 << (order + PAGE_SHIFT); page = alloc_pages(GFP_KERNEL, order); @@ -2385,16 +2388,11 @@ static struct saved_cmdlines_buffer *allocate_cmdlines_buffer(unsigned int val) memset(s, 0, sizeof(*s)); /* Round up to actual allocation */ - val = (size - sizeof(*s)) / TASK_COMM_LEN; + val = (size - sizeof(*s)) / SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); s->cmdline_num = val; - s->map_cmdline_to_pid = kmalloc_array(val, - sizeof(*s->map_cmdline_to_pid), - GFP_KERNEL); - if (!s->map_cmdline_to_pid) { - free_saved_cmdlines_buffer(s); - return NULL; - } + /* Place map_cmdline_to_pid array right after saved_cmdlines */ + s->map_cmdline_to_pid = (unsigned *)&s->saved_cmdlines[val * TASK_COMM_LEN]; s->cmdline_idx = 0; memset(&s->map_pid_to_cmdline, NO_CMDLINE_MAP, -- cgit v1.2.3 From e85d471c2be5d3d4dadee1110db316f352daaf0b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 20 Feb 2024 09:06:15 -0500 Subject: tracing: Move open coded processing of tgid_map into helper function In preparation of moving the saved_cmdlines logic out of trace.c and into trace_sched_switch.c, replace the open coded manipulation of tgid_map in set_tracer_flag() into a helper function trace_alloc_tgid_map() so that it can be easily moved into trace_sched_switch.c without changing existing functions in trace.c. No functional changes. Link: https://lore.kernel.org/linux-trace-kernel/20240220140703.338116216@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Tim Chen Cc: Vincent Donnefort Cc: Sven Schnelle Cc: Mete Durlu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 38 +++++++++++++++++++++++--------------- 1 file changed, 23 insertions(+), 15 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 70202e60a60a..d3e630dd36f8 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -5453,10 +5453,31 @@ int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set) return 0; } -int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) +static int trace_alloc_tgid_map(void) { int *map; + if (tgid_map) + return 0; + + tgid_map_max = pid_max; + map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), + GFP_KERNEL); + if (!map) + return -ENOMEM; + + /* + * Pairs with smp_load_acquire() in + * trace_find_tgid_ptr() to ensure that if it observes + * the tgid_map we just allocated then it also observes + * the corresponding tgid_map_max value. + */ + smp_store_release(&tgid_map, map); + return 0; +} + +int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) +{ if ((mask == TRACE_ITER_RECORD_TGID) || (mask == TRACE_ITER_RECORD_CMD)) lockdep_assert_held(&event_mutex); @@ -5479,20 +5500,7 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) trace_event_enable_cmd_record(enabled); if (mask == TRACE_ITER_RECORD_TGID) { - if (!tgid_map) { - tgid_map_max = pid_max; - map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), - GFP_KERNEL); - - /* - * Pairs with smp_load_acquire() in - * trace_find_tgid_ptr() to ensure that if it observes - * the tgid_map we just allocated then it also observes - * the corresponding tgid_map_max value. - */ - smp_store_release(&tgid_map, map); - } - if (!tgid_map) { + if (trace_alloc_tgid_map() < 0) { tr->trace_flags &= ~TRACE_ITER_RECORD_TGID; return -ENOMEM; } -- cgit v1.2.3 From 2cc621fd2e9b8494df06de459c14738cf76add91 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 20 Feb 2024 09:06:16 -0500 Subject: tracing: Move saved_cmdline code into trace_sched_switch.c The code that handles saved_cmdlines is split between the trace.c file and the trace_sched_switch.c. There's some history to this. The trace_sched_switch.c was originally created to handle the sched_switch tracer that was deprecated due to sched_switch trace event making it obsolete. But that file did not get deleted as it had some code to help with saved_cmdlines. But trace.c has grown tremendously since then. Just move all the saved_cmdlines code into trace_sched_switch.c as that's the only reason that file still exists, and trace.c has gotten too big. No functional changes. Link: https://lore.kernel.org/linux-trace-kernel/20240220140703.497966629@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Tim Chen Cc: Vincent Donnefort Cc: Sven Schnelle Cc: Mete Durlu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 515 +------------------------------------- kernel/trace/trace.h | 10 + kernel/trace/trace_sched_switch.c | 515 ++++++++++++++++++++++++++++++++++++++ 3 files changed, 528 insertions(+), 512 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d3e630dd36f8..1e5f80cda39a 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -39,7 +39,6 @@ #include #include #include -#include #include #include #include @@ -105,7 +104,7 @@ dummy_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set) * tracing is active, only save the comm when a trace event * occurred. */ -static DEFINE_PER_CPU(bool, trace_taskinfo_save); +DEFINE_PER_CPU(bool, trace_taskinfo_save); /* * Kill all tracing for good (never come back). @@ -2320,96 +2319,6 @@ void tracing_reset_all_online_cpus(void) mutex_unlock(&trace_types_lock); } -/* - * The tgid_map array maps from pid to tgid; i.e. the value stored at index i - * is the tgid last observed corresponding to pid=i. - */ -static int *tgid_map; - -/* The maximum valid index into tgid_map. */ -static size_t tgid_map_max; - -#define SAVED_CMDLINES_DEFAULT 128 -#define NO_CMDLINE_MAP UINT_MAX -/* - * Preemption must be disabled before acquiring trace_cmdline_lock. - * The various trace_arrays' max_lock must be acquired in a context - * where interrupt is disabled. - */ -static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; -struct saved_cmdlines_buffer { - unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; - unsigned *map_cmdline_to_pid; - unsigned cmdline_num; - int cmdline_idx; - char saved_cmdlines[]; -}; -static struct saved_cmdlines_buffer *savedcmd; - -/* Holds the size of a cmdline and pid element */ -#define SAVED_CMDLINE_MAP_ELEMENT_SIZE(s) \ - (TASK_COMM_LEN + sizeof((s)->map_cmdline_to_pid[0])) - -static inline char *get_saved_cmdlines(int idx) -{ - return &savedcmd->saved_cmdlines[idx * TASK_COMM_LEN]; -} - -static inline void set_cmdline(int idx, const char *cmdline) -{ - strncpy(get_saved_cmdlines(idx), cmdline, TASK_COMM_LEN); -} - -static void free_saved_cmdlines_buffer(struct saved_cmdlines_buffer *s) -{ - int order = get_order(sizeof(*s) + s->cmdline_num * TASK_COMM_LEN); - - kmemleak_free(s); - free_pages((unsigned long)s, order); -} - -static struct saved_cmdlines_buffer *allocate_cmdlines_buffer(unsigned int val) -{ - struct saved_cmdlines_buffer *s; - struct page *page; - int orig_size, size; - int order; - - /* Figure out how much is needed to hold the given number of cmdlines */ - orig_size = sizeof(*s) + val * SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); - order = get_order(orig_size); - size = 1 << (order + PAGE_SHIFT); - page = alloc_pages(GFP_KERNEL, order); - if (!page) - return NULL; - - s = page_address(page); - kmemleak_alloc(s, size, 1, GFP_KERNEL); - memset(s, 0, sizeof(*s)); - - /* Round up to actual allocation */ - val = (size - sizeof(*s)) / SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); - s->cmdline_num = val; - - /* Place map_cmdline_to_pid array right after saved_cmdlines */ - s->map_cmdline_to_pid = (unsigned *)&s->saved_cmdlines[val * TASK_COMM_LEN]; - - s->cmdline_idx = 0; - memset(&s->map_pid_to_cmdline, NO_CMDLINE_MAP, - sizeof(s->map_pid_to_cmdline)); - memset(s->map_cmdline_to_pid, NO_CMDLINE_MAP, - val * sizeof(*s->map_cmdline_to_pid)); - - return s; -} - -static int trace_create_savedcmd(void) -{ - savedcmd = allocate_cmdlines_buffer(SAVED_CMDLINES_DEFAULT); - - return savedcmd ? 0 : -ENOMEM; -} - int is_tracing_stopped(void) { return global_trace.stop_count; @@ -2502,201 +2411,6 @@ void tracing_stop(void) return tracing_stop_tr(&global_trace); } -static int trace_save_cmdline(struct task_struct *tsk) -{ - unsigned tpid, idx; - - /* treat recording of idle task as a success */ - if (!tsk->pid) - return 1; - - tpid = tsk->pid & (PID_MAX_DEFAULT - 1); - - /* - * It's not the end of the world if we don't get - * the lock, but we also don't want to spin - * nor do we want to disable interrupts, - * so if we miss here, then better luck next time. - * - * This is called within the scheduler and wake up, so interrupts - * had better been disabled and run queue lock been held. - */ - lockdep_assert_preemption_disabled(); - if (!arch_spin_trylock(&trace_cmdline_lock)) - return 0; - - idx = savedcmd->map_pid_to_cmdline[tpid]; - if (idx == NO_CMDLINE_MAP) { - idx = (savedcmd->cmdline_idx + 1) % savedcmd->cmdline_num; - - savedcmd->map_pid_to_cmdline[tpid] = idx; - savedcmd->cmdline_idx = idx; - } - - savedcmd->map_cmdline_to_pid[idx] = tsk->pid; - set_cmdline(idx, tsk->comm); - - arch_spin_unlock(&trace_cmdline_lock); - - return 1; -} - -static void __trace_find_cmdline(int pid, char comm[]) -{ - unsigned map; - int tpid; - - if (!pid) { - strcpy(comm, ""); - return; - } - - if (WARN_ON_ONCE(pid < 0)) { - strcpy(comm, ""); - return; - } - - tpid = pid & (PID_MAX_DEFAULT - 1); - map = savedcmd->map_pid_to_cmdline[tpid]; - if (map != NO_CMDLINE_MAP) { - tpid = savedcmd->map_cmdline_to_pid[map]; - if (tpid == pid) { - strscpy(comm, get_saved_cmdlines(map), TASK_COMM_LEN); - return; - } - } - strcpy(comm, "<...>"); -} - -void trace_find_cmdline(int pid, char comm[]) -{ - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - - __trace_find_cmdline(pid, comm); - - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); -} - -static int *trace_find_tgid_ptr(int pid) -{ - /* - * Pairs with the smp_store_release in set_tracer_flag() to ensure that - * if we observe a non-NULL tgid_map then we also observe the correct - * tgid_map_max. - */ - int *map = smp_load_acquire(&tgid_map); - - if (unlikely(!map || pid > tgid_map_max)) - return NULL; - - return &map[pid]; -} - -int trace_find_tgid(int pid) -{ - int *ptr = trace_find_tgid_ptr(pid); - - return ptr ? *ptr : 0; -} - -static int trace_save_tgid(struct task_struct *tsk) -{ - int *ptr; - - /* treat recording of idle task as a success */ - if (!tsk->pid) - return 1; - - ptr = trace_find_tgid_ptr(tsk->pid); - if (!ptr) - return 0; - - *ptr = tsk->tgid; - return 1; -} - -static bool tracing_record_taskinfo_skip(int flags) -{ - if (unlikely(!(flags & (TRACE_RECORD_CMDLINE | TRACE_RECORD_TGID)))) - return true; - if (!__this_cpu_read(trace_taskinfo_save)) - return true; - return false; -} - -/** - * tracing_record_taskinfo - record the task info of a task - * - * @task: task to record - * @flags: TRACE_RECORD_CMDLINE for recording comm - * TRACE_RECORD_TGID for recording tgid - */ -void tracing_record_taskinfo(struct task_struct *task, int flags) -{ - bool done; - - if (tracing_record_taskinfo_skip(flags)) - return; - - /* - * Record as much task information as possible. If some fail, continue - * to try to record the others. - */ - done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(task); - done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(task); - - /* If recording any information failed, retry again soon. */ - if (!done) - return; - - __this_cpu_write(trace_taskinfo_save, false); -} - -/** - * tracing_record_taskinfo_sched_switch - record task info for sched_switch - * - * @prev: previous task during sched_switch - * @next: next task during sched_switch - * @flags: TRACE_RECORD_CMDLINE for recording comm - * TRACE_RECORD_TGID for recording tgid - */ -void tracing_record_taskinfo_sched_switch(struct task_struct *prev, - struct task_struct *next, int flags) -{ - bool done; - - if (tracing_record_taskinfo_skip(flags)) - return; - - /* - * Record as much task information as possible. If some fail, continue - * to try to record the others. - */ - done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(prev); - done &= !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(next); - done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(prev); - done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(next); - - /* If recording any information failed, retry again soon. */ - if (!done) - return; - - __this_cpu_write(trace_taskinfo_save, false); -} - -/* Helpers to record a specific task information */ -void tracing_record_cmdline(struct task_struct *task) -{ - tracing_record_taskinfo(task, TRACE_RECORD_CMDLINE); -} - -void tracing_record_tgid(struct task_struct *task) -{ - tracing_record_taskinfo(task, TRACE_RECORD_TGID); -} - /* * Several functions return TRACE_TYPE_PARTIAL_LINE if the trace_seq * overflowed, and TRACE_TYPE_HANDLED otherwise. This helper function @@ -5453,29 +5167,6 @@ int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set) return 0; } -static int trace_alloc_tgid_map(void) -{ - int *map; - - if (tgid_map) - return 0; - - tgid_map_max = pid_max; - map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), - GFP_KERNEL); - if (!map) - return -ENOMEM; - - /* - * Pairs with smp_load_acquire() in - * trace_find_tgid_ptr() to ensure that if it observes - * the tgid_map we just allocated then it also observes - * the corresponding tgid_map_max value. - */ - smp_store_release(&tgid_map, map); - return 0; -} - int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) { if ((mask == TRACE_ITER_RECORD_TGID) || @@ -5500,6 +5191,7 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) trace_event_enable_cmd_record(enabled); if (mask == TRACE_ITER_RECORD_TGID) { + if (trace_alloc_tgid_map() < 0) { tr->trace_flags &= ~TRACE_ITER_RECORD_TGID; return -ENOMEM; @@ -5944,207 +5636,6 @@ static const struct file_operations tracing_readme_fops = { .llseek = generic_file_llseek, }; -static void *saved_tgids_next(struct seq_file *m, void *v, loff_t *pos) -{ - int pid = ++(*pos); - - return trace_find_tgid_ptr(pid); -} - -static void *saved_tgids_start(struct seq_file *m, loff_t *pos) -{ - int pid = *pos; - - return trace_find_tgid_ptr(pid); -} - -static void saved_tgids_stop(struct seq_file *m, void *v) -{ -} - -static int saved_tgids_show(struct seq_file *m, void *v) -{ - int *entry = (int *)v; - int pid = entry - tgid_map; - int tgid = *entry; - - if (tgid == 0) - return SEQ_SKIP; - - seq_printf(m, "%d %d\n", pid, tgid); - return 0; -} - -static const struct seq_operations tracing_saved_tgids_seq_ops = { - .start = saved_tgids_start, - .stop = saved_tgids_stop, - .next = saved_tgids_next, - .show = saved_tgids_show, -}; - -static int tracing_saved_tgids_open(struct inode *inode, struct file *filp) -{ - int ret; - - ret = tracing_check_open_get_tr(NULL); - if (ret) - return ret; - - return seq_open(filp, &tracing_saved_tgids_seq_ops); -} - - -static const struct file_operations tracing_saved_tgids_fops = { - .open = tracing_saved_tgids_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -static void *saved_cmdlines_next(struct seq_file *m, void *v, loff_t *pos) -{ - unsigned int *ptr = v; - - if (*pos || m->count) - ptr++; - - (*pos)++; - - for (; ptr < &savedcmd->map_cmdline_to_pid[savedcmd->cmdline_num]; - ptr++) { - if (*ptr == -1 || *ptr == NO_CMDLINE_MAP) - continue; - - return ptr; - } - - return NULL; -} - -static void *saved_cmdlines_start(struct seq_file *m, loff_t *pos) -{ - void *v; - loff_t l = 0; - - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - - v = &savedcmd->map_cmdline_to_pid[0]; - while (l <= *pos) { - v = saved_cmdlines_next(m, v, &l); - if (!v) - return NULL; - } - - return v; -} - -static void saved_cmdlines_stop(struct seq_file *m, void *v) -{ - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); -} - -static int saved_cmdlines_show(struct seq_file *m, void *v) -{ - char buf[TASK_COMM_LEN]; - unsigned int *pid = v; - - __trace_find_cmdline(*pid, buf); - seq_printf(m, "%d %s\n", *pid, buf); - return 0; -} - -static const struct seq_operations tracing_saved_cmdlines_seq_ops = { - .start = saved_cmdlines_start, - .next = saved_cmdlines_next, - .stop = saved_cmdlines_stop, - .show = saved_cmdlines_show, -}; - -static int tracing_saved_cmdlines_open(struct inode *inode, struct file *filp) -{ - int ret; - - ret = tracing_check_open_get_tr(NULL); - if (ret) - return ret; - - return seq_open(filp, &tracing_saved_cmdlines_seq_ops); -} - -static const struct file_operations tracing_saved_cmdlines_fops = { - .open = tracing_saved_cmdlines_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -static ssize_t -tracing_saved_cmdlines_size_read(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char buf[64]; - int r; - - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - r = scnprintf(buf, sizeof(buf), "%u\n", savedcmd->cmdline_num); - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); - - return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); -} - -static int tracing_resize_saved_cmdlines(unsigned int val) -{ - struct saved_cmdlines_buffer *s, *savedcmd_temp; - - s = allocate_cmdlines_buffer(val); - if (!s) - return -ENOMEM; - - preempt_disable(); - arch_spin_lock(&trace_cmdline_lock); - savedcmd_temp = savedcmd; - savedcmd = s; - arch_spin_unlock(&trace_cmdline_lock); - preempt_enable(); - free_saved_cmdlines_buffer(savedcmd_temp); - - return 0; -} - -static ssize_t -tracing_saved_cmdlines_size_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - unsigned long val; - int ret; - - ret = kstrtoul_from_user(ubuf, cnt, 10, &val); - if (ret) - return ret; - - /* must have at least 1 entry or less than PID_MAX_DEFAULT */ - if (!val || val > PID_MAX_DEFAULT) - return -EINVAL; - - ret = tracing_resize_saved_cmdlines((unsigned int)val); - if (ret < 0) - return ret; - - *ppos += cnt; - - return cnt; -} - -static const struct file_operations tracing_saved_cmdlines_size_fops = { - .open = tracing_open_generic, - .read = tracing_saved_cmdlines_size_read, - .write = tracing_saved_cmdlines_size_write, -}; - #ifdef CONFIG_TRACE_EVAL_MAP_FILE static union trace_eval_map_item * update_eval_map(union trace_eval_map_item *ptr) @@ -10720,7 +10211,7 @@ __init static int tracer_alloc_buffers(void) out_free_pipe_cpumask: free_cpumask_var(global_trace.pipe_cpumask); out_free_savedcmd: - free_saved_cmdlines_buffer(savedcmd); + trace_free_saved_cmdlines_buffer(); out_free_temp_buffer: ring_buffer_free(temp_buffer); out_rm_hp_state: diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 00f873910c5d..e4f0714d7a49 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1375,6 +1375,16 @@ static inline void trace_buffer_unlock_commit(struct trace_array *tr, trace_buffer_unlock_commit_regs(tr, buffer, event, trace_ctx, NULL); } +DECLARE_PER_CPU(bool, trace_taskinfo_save); +int trace_save_cmdline(struct task_struct *tsk); +int trace_create_savedcmd(void); +int trace_alloc_tgid_map(void); +void trace_free_saved_cmdlines_buffer(void); + +extern const struct file_operations tracing_saved_cmdlines_fops; +extern const struct file_operations tracing_saved_tgids_fops; +extern const struct file_operations tracing_saved_cmdlines_size_fops; + DECLARE_PER_CPU(struct ring_buffer_event *, trace_buffered_event); DECLARE_PER_CPU(int, trace_buffered_event_cnt); void trace_buffered_event_disable(void); diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index c9ffdcfe622e..8a407adb0e1c 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -148,3 +149,517 @@ void tracing_stop_tgid_record(void) { tracing_stop_sched_switch(RECORD_TGID); } + +/* + * The tgid_map array maps from pid to tgid; i.e. the value stored at index i + * is the tgid last observed corresponding to pid=i. + */ +static int *tgid_map; + +/* The maximum valid index into tgid_map. */ +static size_t tgid_map_max; + +#define SAVED_CMDLINES_DEFAULT 128 +#define NO_CMDLINE_MAP UINT_MAX +/* + * Preemption must be disabled before acquiring trace_cmdline_lock. + * The various trace_arrays' max_lock must be acquired in a context + * where interrupt is disabled. + */ +static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; +struct saved_cmdlines_buffer { + unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; + unsigned *map_cmdline_to_pid; + unsigned cmdline_num; + int cmdline_idx; + char saved_cmdlines[]; +}; +static struct saved_cmdlines_buffer *savedcmd; + +/* Holds the size of a cmdline and pid element */ +#define SAVED_CMDLINE_MAP_ELEMENT_SIZE(s) \ + (TASK_COMM_LEN + sizeof((s)->map_cmdline_to_pid[0])) + +static inline char *get_saved_cmdlines(int idx) +{ + return &savedcmd->saved_cmdlines[idx * TASK_COMM_LEN]; +} + +static inline void set_cmdline(int idx, const char *cmdline) +{ + strncpy(get_saved_cmdlines(idx), cmdline, TASK_COMM_LEN); +} + +static void free_saved_cmdlines_buffer(struct saved_cmdlines_buffer *s) +{ + int order = get_order(sizeof(*s) + s->cmdline_num * TASK_COMM_LEN); + + kmemleak_free(s); + free_pages((unsigned long)s, order); +} + +static struct saved_cmdlines_buffer *allocate_cmdlines_buffer(unsigned int val) +{ + struct saved_cmdlines_buffer *s; + struct page *page; + int orig_size, size; + int order; + + /* Figure out how much is needed to hold the given number of cmdlines */ + orig_size = sizeof(*s) + val * SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); + order = get_order(orig_size); + size = 1 << (order + PAGE_SHIFT); + page = alloc_pages(GFP_KERNEL, order); + if (!page) + return NULL; + + s = page_address(page); + kmemleak_alloc(s, size, 1, GFP_KERNEL); + memset(s, 0, sizeof(*s)); + + /* Round up to actual allocation */ + val = (size - sizeof(*s)) / SAVED_CMDLINE_MAP_ELEMENT_SIZE(s); + s->cmdline_num = val; + + /* Place map_cmdline_to_pid array right after saved_cmdlines */ + s->map_cmdline_to_pid = (unsigned *)&s->saved_cmdlines[val * TASK_COMM_LEN]; + + s->cmdline_idx = 0; + memset(&s->map_pid_to_cmdline, NO_CMDLINE_MAP, + sizeof(s->map_pid_to_cmdline)); + memset(s->map_cmdline_to_pid, NO_CMDLINE_MAP, + val * sizeof(*s->map_cmdline_to_pid)); + + return s; +} + +int trace_create_savedcmd(void) +{ + savedcmd = allocate_cmdlines_buffer(SAVED_CMDLINES_DEFAULT); + + return savedcmd ? 0 : -ENOMEM; +} + +int trace_save_cmdline(struct task_struct *tsk) +{ + unsigned tpid, idx; + + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + tpid = tsk->pid & (PID_MAX_DEFAULT - 1); + + /* + * It's not the end of the world if we don't get + * the lock, but we also don't want to spin + * nor do we want to disable interrupts, + * so if we miss here, then better luck next time. + * + * This is called within the scheduler and wake up, so interrupts + * had better been disabled and run queue lock been held. + */ + lockdep_assert_preemption_disabled(); + if (!arch_spin_trylock(&trace_cmdline_lock)) + return 0; + + idx = savedcmd->map_pid_to_cmdline[tpid]; + if (idx == NO_CMDLINE_MAP) { + idx = (savedcmd->cmdline_idx + 1) % savedcmd->cmdline_num; + + savedcmd->map_pid_to_cmdline[tpid] = idx; + savedcmd->cmdline_idx = idx; + } + + savedcmd->map_cmdline_to_pid[idx] = tsk->pid; + set_cmdline(idx, tsk->comm); + + arch_spin_unlock(&trace_cmdline_lock); + + return 1; +} + +static void __trace_find_cmdline(int pid, char comm[]) +{ + unsigned map; + int tpid; + + if (!pid) { + strcpy(comm, ""); + return; + } + + if (WARN_ON_ONCE(pid < 0)) { + strcpy(comm, ""); + return; + } + + tpid = pid & (PID_MAX_DEFAULT - 1); + map = savedcmd->map_pid_to_cmdline[tpid]; + if (map != NO_CMDLINE_MAP) { + tpid = savedcmd->map_cmdline_to_pid[map]; + if (tpid == pid) { + strscpy(comm, get_saved_cmdlines(map), TASK_COMM_LEN); + return; + } + } + strcpy(comm, "<...>"); +} + +void trace_find_cmdline(int pid, char comm[]) +{ + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + + __trace_find_cmdline(pid, comm); + + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); +} + +static int *trace_find_tgid_ptr(int pid) +{ + /* + * Pairs with the smp_store_release in set_tracer_flag() to ensure that + * if we observe a non-NULL tgid_map then we also observe the correct + * tgid_map_max. + */ + int *map = smp_load_acquire(&tgid_map); + + if (unlikely(!map || pid > tgid_map_max)) + return NULL; + + return &map[pid]; +} + +int trace_find_tgid(int pid) +{ + int *ptr = trace_find_tgid_ptr(pid); + + return ptr ? *ptr : 0; +} + +static int trace_save_tgid(struct task_struct *tsk) +{ + int *ptr; + + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + ptr = trace_find_tgid_ptr(tsk->pid); + if (!ptr) + return 0; + + *ptr = tsk->tgid; + return 1; +} + +static bool tracing_record_taskinfo_skip(int flags) +{ + if (unlikely(!(flags & (TRACE_RECORD_CMDLINE | TRACE_RECORD_TGID)))) + return true; + if (!__this_cpu_read(trace_taskinfo_save)) + return true; + return false; +} + +/** + * tracing_record_taskinfo - record the task info of a task + * + * @task: task to record + * @flags: TRACE_RECORD_CMDLINE for recording comm + * TRACE_RECORD_TGID for recording tgid + */ +void tracing_record_taskinfo(struct task_struct *task, int flags) +{ + bool done; + + if (tracing_record_taskinfo_skip(flags)) + return; + + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(task); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(task); + + /* If recording any information failed, retry again soon. */ + if (!done) + return; + + __this_cpu_write(trace_taskinfo_save, false); +} + +/** + * tracing_record_taskinfo_sched_switch - record task info for sched_switch + * + * @prev: previous task during sched_switch + * @next: next task during sched_switch + * @flags: TRACE_RECORD_CMDLINE for recording comm + * TRACE_RECORD_TGID for recording tgid + */ +void tracing_record_taskinfo_sched_switch(struct task_struct *prev, + struct task_struct *next, int flags) +{ + bool done; + + if (tracing_record_taskinfo_skip(flags)) + return; + + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(prev); + done &= !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(next); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(prev); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(next); + + /* If recording any information failed, retry again soon. */ + if (!done) + return; + + __this_cpu_write(trace_taskinfo_save, false); +} + +/* Helpers to record a specific task information */ +void tracing_record_cmdline(struct task_struct *task) +{ + tracing_record_taskinfo(task, TRACE_RECORD_CMDLINE); +} + +void tracing_record_tgid(struct task_struct *task) +{ + tracing_record_taskinfo(task, TRACE_RECORD_TGID); +} + +int trace_alloc_tgid_map(void) +{ + int *map; + + if (tgid_map) + return 0; + + tgid_map_max = pid_max; + map = kvcalloc(tgid_map_max + 1, sizeof(*tgid_map), + GFP_KERNEL); + if (!map) + return -ENOMEM; + + /* + * Pairs with smp_load_acquire() in + * trace_find_tgid_ptr() to ensure that if it observes + * the tgid_map we just allocated then it also observes + * the corresponding tgid_map_max value. + */ + smp_store_release(&tgid_map, map); + return 0; +} + +static void *saved_tgids_next(struct seq_file *m, void *v, loff_t *pos) +{ + int pid = ++(*pos); + + return trace_find_tgid_ptr(pid); +} + +static void *saved_tgids_start(struct seq_file *m, loff_t *pos) +{ + int pid = *pos; + + return trace_find_tgid_ptr(pid); +} + +static void saved_tgids_stop(struct seq_file *m, void *v) +{ +} + +static int saved_tgids_show(struct seq_file *m, void *v) +{ + int *entry = (int *)v; + int pid = entry - tgid_map; + int tgid = *entry; + + if (tgid == 0) + return SEQ_SKIP; + + seq_printf(m, "%d %d\n", pid, tgid); + return 0; +} + +static const struct seq_operations tracing_saved_tgids_seq_ops = { + .start = saved_tgids_start, + .stop = saved_tgids_stop, + .next = saved_tgids_next, + .show = saved_tgids_show, +}; + +static int tracing_saved_tgids_open(struct inode *inode, struct file *filp) +{ + int ret; + + ret = tracing_check_open_get_tr(NULL); + if (ret) + return ret; + + return seq_open(filp, &tracing_saved_tgids_seq_ops); +} + + +const struct file_operations tracing_saved_tgids_fops = { + .open = tracing_saved_tgids_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static void *saved_cmdlines_next(struct seq_file *m, void *v, loff_t *pos) +{ + unsigned int *ptr = v; + + if (*pos || m->count) + ptr++; + + (*pos)++; + + for (; ptr < &savedcmd->map_cmdline_to_pid[savedcmd->cmdline_num]; + ptr++) { + if (*ptr == -1 || *ptr == NO_CMDLINE_MAP) + continue; + + return ptr; + } + + return NULL; +} + +static void *saved_cmdlines_start(struct seq_file *m, loff_t *pos) +{ + void *v; + loff_t l = 0; + + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + + v = &savedcmd->map_cmdline_to_pid[0]; + while (l <= *pos) { + v = saved_cmdlines_next(m, v, &l); + if (!v) + return NULL; + } + + return v; +} + +static void saved_cmdlines_stop(struct seq_file *m, void *v) +{ + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); +} + +static int saved_cmdlines_show(struct seq_file *m, void *v) +{ + char buf[TASK_COMM_LEN]; + unsigned int *pid = v; + + __trace_find_cmdline(*pid, buf); + seq_printf(m, "%d %s\n", *pid, buf); + return 0; +} + +static const struct seq_operations tracing_saved_cmdlines_seq_ops = { + .start = saved_cmdlines_start, + .next = saved_cmdlines_next, + .stop = saved_cmdlines_stop, + .show = saved_cmdlines_show, +}; + +static int tracing_saved_cmdlines_open(struct inode *inode, struct file *filp) +{ + int ret; + + ret = tracing_check_open_get_tr(NULL); + if (ret) + return ret; + + return seq_open(filp, &tracing_saved_cmdlines_seq_ops); +} + +const struct file_operations tracing_saved_cmdlines_fops = { + .open = tracing_saved_cmdlines_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static ssize_t +tracing_saved_cmdlines_size_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char buf[64]; + int r; + + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + r = scnprintf(buf, sizeof(buf), "%u\n", savedcmd->cmdline_num); + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +void trace_free_saved_cmdlines_buffer(void) +{ + free_saved_cmdlines_buffer(savedcmd); +} + +static int tracing_resize_saved_cmdlines(unsigned int val) +{ + struct saved_cmdlines_buffer *s, *savedcmd_temp; + + s = allocate_cmdlines_buffer(val); + if (!s) + return -ENOMEM; + + preempt_disable(); + arch_spin_lock(&trace_cmdline_lock); + savedcmd_temp = savedcmd; + savedcmd = s; + arch_spin_unlock(&trace_cmdline_lock); + preempt_enable(); + free_saved_cmdlines_buffer(savedcmd_temp); + + return 0; +} + +static ssize_t +tracing_saved_cmdlines_size_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + unsigned long val; + int ret; + + ret = kstrtoul_from_user(ubuf, cnt, 10, &val); + if (ret) + return ret; + + /* must have at least 1 entry or less than PID_MAX_DEFAULT */ + if (!val || val > PID_MAX_DEFAULT) + return -EINVAL; + + ret = tracing_resize_saved_cmdlines((unsigned int)val); + if (ret < 0) + return ret; + + *ppos += cnt; + + return cnt; +} + +const struct file_operations tracing_saved_cmdlines_size_fops = { + .open = tracing_open_generic, + .read = tracing_saved_cmdlines_size_read, + .write = tracing_saved_cmdlines_size_write, +}; -- cgit v1.2.3 From 180e4e390978af9d0cc060e87920c462276453b9 Mon Sep 17 00:00:00 2001 From: Vincent Donnefort Date: Tue, 20 Feb 2024 20:23:07 +0000 Subject: tracing: Add snapshot refcount When a ring-buffer is memory mapped by user-space, no trace or ring-buffer swap is possible. This means the snapshot feature is mutually exclusive with the memory mapping. Having a refcount on snapshot users will help to know if a mapping is possible or not. Instead of relying on the global trace_types_lock, a new spinlock is introduced to serialize accesses to trace_array->snapshot. This intends to allow access to that variable in a context where the mmap lock is already held. Link: https://lore.kernel.org/linux-trace-kernel/20240220202310.2489614-4-vdonnefort@google.com Signed-off-by: Vincent Donnefort Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 99 ++++++++++++++++++++++++++++++------- kernel/trace/trace.h | 8 ++- kernel/trace/trace_events_trigger.c | 58 +++++++++++++++------- 3 files changed, 129 insertions(+), 36 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 1e5f80cda39a..c66ca2f6944e 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1300,6 +1300,50 @@ static void free_snapshot(struct trace_array *tr) tr->allocated_snapshot = false; } +static int tracing_arm_snapshot_locked(struct trace_array *tr) +{ + int ret; + + lockdep_assert_held(&trace_types_lock); + + spin_lock(&tr->snapshot_trigger_lock); + if (tr->snapshot == UINT_MAX) { + spin_unlock(&tr->snapshot_trigger_lock); + return -EBUSY; + } + + tr->snapshot++; + spin_unlock(&tr->snapshot_trigger_lock); + + ret = tracing_alloc_snapshot_instance(tr); + if (ret) { + spin_lock(&tr->snapshot_trigger_lock); + tr->snapshot--; + spin_unlock(&tr->snapshot_trigger_lock); + } + + return ret; +} + +int tracing_arm_snapshot(struct trace_array *tr) +{ + int ret; + + mutex_lock(&trace_types_lock); + ret = tracing_arm_snapshot_locked(tr); + mutex_unlock(&trace_types_lock); + + return ret; +} + +void tracing_disarm_snapshot(struct trace_array *tr) +{ + spin_lock(&tr->snapshot_trigger_lock); + if (!WARN_ON(!tr->snapshot)) + tr->snapshot--; + spin_unlock(&tr->snapshot_trigger_lock); +} + /** * tracing_alloc_snapshot - allocate snapshot buffer. * @@ -1373,10 +1417,6 @@ int tracing_snapshot_cond_enable(struct trace_array *tr, void *cond_data, mutex_lock(&trace_types_lock); - ret = tracing_alloc_snapshot_instance(tr); - if (ret) - goto fail_unlock; - if (tr->current_trace->use_max_tr) { ret = -EBUSY; goto fail_unlock; @@ -1395,6 +1435,10 @@ int tracing_snapshot_cond_enable(struct trace_array *tr, void *cond_data, goto fail_unlock; } + ret = tracing_arm_snapshot_locked(tr); + if (ret) + goto fail_unlock; + local_irq_disable(); arch_spin_lock(&tr->max_lock); tr->cond_snapshot = cond_snapshot; @@ -1439,6 +1483,8 @@ int tracing_snapshot_cond_disable(struct trace_array *tr) arch_spin_unlock(&tr->max_lock); local_irq_enable(); + tracing_disarm_snapshot(tr); + return ret; } EXPORT_SYMBOL_GPL(tracing_snapshot_cond_disable); @@ -1481,6 +1527,7 @@ int tracing_snapshot_cond_disable(struct trace_array *tr) } EXPORT_SYMBOL_GPL(tracing_snapshot_cond_disable); #define free_snapshot(tr) do { } while (0) +#define tracing_arm_snapshot_locked(tr) ({ -EBUSY; }) #endif /* CONFIG_TRACER_SNAPSHOT */ void tracer_tracing_off(struct trace_array *tr) @@ -6112,11 +6159,12 @@ int tracing_set_tracer(struct trace_array *tr, const char *buf) */ synchronize_rcu(); free_snapshot(tr); + tracing_disarm_snapshot(tr); } - if (t->use_max_tr && !tr->allocated_snapshot) { - ret = tracing_alloc_snapshot_instance(tr); - if (ret < 0) + if (t->use_max_tr) { + ret = tracing_arm_snapshot_locked(tr); + if (ret) goto out; } #else @@ -6125,8 +6173,13 @@ int tracing_set_tracer(struct trace_array *tr, const char *buf) if (t->init) { ret = tracer_init(t, tr); - if (ret) + if (ret) { +#ifdef CONFIG_TRACER_MAX_TRACE + if (t->use_max_tr) + tracing_disarm_snapshot(tr); +#endif goto out; + } } tr->current_trace = t; @@ -7228,10 +7281,11 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, if (tr->allocated_snapshot) ret = resize_buffer_duplicate_size(&tr->max_buffer, &tr->array_buffer, iter->cpu_file); - else - ret = tracing_alloc_snapshot_instance(tr); - if (ret < 0) + + ret = tracing_arm_snapshot_locked(tr); + if (ret) break; + /* Now, we're going to swap */ if (iter->cpu_file == RING_BUFFER_ALL_CPUS) { local_irq_disable(); @@ -7241,6 +7295,7 @@ tracing_snapshot_write(struct file *filp, const char __user *ubuf, size_t cnt, smp_call_function_single(iter->cpu_file, tracing_swap_cpu_buffer, (void *)tr, 1); } + tracing_disarm_snapshot(tr); break; default: if (tr->allocated_snapshot) { @@ -8372,8 +8427,13 @@ ftrace_trace_snapshot_callback(struct trace_array *tr, struct ftrace_hash *hash, ops = param ? &snapshot_count_probe_ops : &snapshot_probe_ops; - if (glob[0] == '!') - return unregister_ftrace_function_probe_func(glob+1, tr, ops); + if (glob[0] == '!') { + ret = unregister_ftrace_function_probe_func(glob+1, tr, ops); + if (!ret) + tracing_disarm_snapshot(tr); + + return ret; + } if (!param) goto out_reg; @@ -8392,12 +8452,13 @@ ftrace_trace_snapshot_callback(struct trace_array *tr, struct ftrace_hash *hash, return ret; out_reg: - ret = tracing_alloc_snapshot_instance(tr); + ret = tracing_arm_snapshot(tr); if (ret < 0) goto out; ret = register_ftrace_function_probe(glob, tr, ops, count); - + if (ret < 0) + tracing_disarm_snapshot(tr); out: return ret < 0 ? ret : 0; } @@ -9204,7 +9265,9 @@ trace_array_create_systems(const char *name, const char *systems) raw_spin_lock_init(&tr->start_lock); tr->max_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; - +#ifdef CONFIG_TRACER_MAX_TRACE + spin_lock_init(&tr->snapshot_trigger_lock); +#endif tr->current_trace = &nop_trace; INIT_LIST_HEAD(&tr->systems); @@ -10174,7 +10237,9 @@ __init static int tracer_alloc_buffers(void) global_trace.current_trace = &nop_trace; global_trace.max_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; - +#ifdef CONFIG_TRACER_MAX_TRACE + spin_lock_init(&global_trace.snapshot_trigger_lock); +#endif ftrace_init_global_array_ops(&global_trace); init_trace_flags_index(&global_trace); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index e4f0714d7a49..64450615ca0c 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -334,8 +334,8 @@ struct trace_array { */ struct array_buffer max_buffer; bool allocated_snapshot; -#endif -#ifdef CONFIG_TRACER_MAX_TRACE + spinlock_t snapshot_trigger_lock; + unsigned int snapshot; unsigned long max_latency; #ifdef CONFIG_FSNOTIFY struct dentry *d_max_latency; @@ -1983,12 +1983,16 @@ static inline void trace_event_eval_update(struct trace_eval_map **map, int len) #ifdef CONFIG_TRACER_SNAPSHOT void tracing_snapshot_instance(struct trace_array *tr); int tracing_alloc_snapshot_instance(struct trace_array *tr); +int tracing_arm_snapshot(struct trace_array *tr); +void tracing_disarm_snapshot(struct trace_array *tr); #else static inline void tracing_snapshot_instance(struct trace_array *tr) { } static inline int tracing_alloc_snapshot_instance(struct trace_array *tr) { return 0; } +static inline int tracing_arm_snapshot(struct trace_array *tr) { return 0; } +static inline void tracing_disarm_snapshot(struct trace_array *tr) { } #endif #ifdef CONFIG_PREEMPT_TRACER diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index b33c3861fbbb..62e4f58b8671 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -597,20 +597,12 @@ out: return ret; } -/** - * unregister_trigger - Generic event_command @unreg implementation - * @glob: The raw string used to register the trigger - * @test: Trigger-specific data used to find the trigger to remove - * @file: The trace_event_file associated with the event - * - * Common implementation for event trigger unregistration. - * - * Usually used directly as the @unreg method in event command - * implementations. +/* + * True if the trigger was found and unregistered, else false. */ -static void unregister_trigger(char *glob, - struct event_trigger_data *test, - struct trace_event_file *file) +static bool try_unregister_trigger(char *glob, + struct event_trigger_data *test, + struct trace_event_file *file) { struct event_trigger_data *data = NULL, *iter; @@ -626,8 +618,32 @@ static void unregister_trigger(char *glob, } } - if (data && data->ops->free) - data->ops->free(data); + if (data) { + if (data->ops->free) + data->ops->free(data); + + return true; + } + + return false; +} + +/** + * unregister_trigger - Generic event_command @unreg implementation + * @glob: The raw string used to register the trigger + * @test: Trigger-specific data used to find the trigger to remove + * @file: The trace_event_file associated with the event + * + * Common implementation for event trigger unregistration. + * + * Usually used directly as the @unreg method in event command + * implementations. + */ +static void unregister_trigger(char *glob, + struct event_trigger_data *test, + struct trace_event_file *file) +{ + try_unregister_trigger(glob, test, file); } /* @@ -1470,7 +1486,7 @@ register_snapshot_trigger(char *glob, struct event_trigger_data *data, struct trace_event_file *file) { - int ret = tracing_alloc_snapshot_instance(file->tr); + int ret = tracing_arm_snapshot(file->tr); if (ret < 0) return ret; @@ -1478,6 +1494,14 @@ register_snapshot_trigger(char *glob, return register_trigger(glob, data, file); } +static void unregister_snapshot_trigger(char *glob, + struct event_trigger_data *data, + struct trace_event_file *file) +{ + if (try_unregister_trigger(glob, data, file)) + tracing_disarm_snapshot(file->tr); +} + static int snapshot_trigger_print(struct seq_file *m, struct event_trigger_data *data) { @@ -1510,7 +1534,7 @@ static struct event_command trigger_snapshot_cmd = { .trigger_type = ETT_SNAPSHOT, .parse = event_trigger_parse, .reg = register_snapshot_trigger, - .unreg = unregister_trigger, + .unreg = unregister_snapshot_trigger, .get_trigger_ops = snapshot_get_trigger_ops, .set_filter = set_trigger_filter, }; -- cgit v1.2.3 From ed89683763a12e8289cce6f233dd07b4eb42fb96 Mon Sep 17 00:00:00 2001 From: John Garry Date: Thu, 22 Feb 2024 12:46:39 +0000 Subject: tracing: Use init_utsname()->release Instead of using UTS_RELEASE, use init_utsname()->release, which means that we don't need to rebuild the code just for the git head commit changing. Link: https://lore.kernel.org/linux-trace-kernel/20240222124639.65629-1-john.g.garry@oracle.com Signed-off-by: John Garry Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index c66ca2f6944e..a7c0dc0aaf14 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -13,7 +13,7 @@ * Copyright (C) 2004 Nadia Yvette Chambers */ #include -#include +#include #include #include #include @@ -4148,7 +4148,7 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter) get_total_entries(buf, &total, &entries); seq_printf(m, "# %s latency trace v1.1.5 on %s\n", - name, UTS_RELEASE); + name, init_utsname()->release); seq_puts(m, "# -----------------------------------" "---------------------------------\n"); seq_printf(m, "# latency: %lu us, #%lu/%lu, CPU#%d |" -- cgit v1.2.3 From cca990c7b565af0dc61a8f647c00833453cf5bff Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 22 Feb 2024 20:33:24 -0500 Subject: tracing: Fix snapshot counter going between two tracers that use it Running the ftrace selftests caused the ring buffer mapping test to fail. Investigating, I found that the snapshot counter would be incremented every time a tracer that uses the snapshot is enabled even if the snapshot was used by the previous tracer. That is: # cd /sys/kernel/tracing # echo wakeup_rt > current_tracer # echo wakeup_dl > current_tracer # echo nop > current_tracer would leave the snapshot counter at 1 and not zero. That's because the enabling of wakeup_dl would increment the counter again but the setting the tracer to nop would only decrement it once. Do not arm the snapshot for a tracer if the previous tracer already had it armed. Link: https://lore.kernel.org/linux-trace-kernel/20240223013344.570525723@goodmis.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Cc: Vincent Donnefort Fixes: 16f7e48ffc53a ("tracing: Add snapshot refcount") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace/trace.c') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index a7c0dc0aaf14..1e1fd377a1cf 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6162,7 +6162,7 @@ int tracing_set_tracer(struct trace_array *tr, const char *buf) tracing_disarm_snapshot(tr); } - if (t->use_max_tr) { + if (!had_max_tr && t->use_max_tr) { ret = tracing_arm_snapshot_locked(tr); if (ret) goto out; -- cgit v1.2.3 From 19f0423fd55c301c8edaea286e568ec657f42750 Mon Sep 17 00:00:00 2001 From: Huang Yiwei Date: Fri, 23 Feb 2024 16:31:26 +0800 Subject: tracing: Support to dump instance traces by ftrace_dump_on_oops Currently ftrace only dumps the global trace buffer on an OOPs. For debugging a production usecase, instance trace will be helpful to check specific problems since global trace buffer may be used for other purposes. This patch extend the ftrace_dump_on_oops parameter to dump a specific or multiple trace instances: - ftrace_dump_on_oops=0: as before -- don't dump - ftrace_dump_on_oops[=1]: as before -- dump the global trace buffer on all CPUs - ftrace_dump_on_oops=2 or =orig_cpu: as before -- dump the global trace buffer on CPU that triggered the oops - ftrace_dump_on_oops=: new behavior -- dump the tracing instance matching - ftrace_dump_on_oops[=2/orig_cpu],[=2/orig_cpu], [=2/orig_cpu]: new behavior -- dump the global trace buffer and multiple instance buffer on all CPUs, or only dump on CPU that triggered the oops if =2 or =orig_cpu is given Also, the sysctl node can handle the input accordingly. Link: https://lore.kernel.org/linux-trace-kernel/20240223083126.1817731-1-quic_hyiwei@quicinc.com Cc: Ross Zwisler Cc: Cc: Cc: Cc: Cc: Cc: Cc: Signed-off-by: Huang Yiwei Signed-off-by: Steven Rostedt (Google) --- Documentation/admin-guide/kernel-parameters.txt | 26 +++- Documentation/admin-guide/sysctl/kernel.rst | 30 ++++- include/linux/ftrace.h | 4 +- include/linux/kernel.h | 1 + kernel/sysctl.c | 4 +- kernel/trace/trace.c | 156 ++++++++++++++++++------ kernel/trace/trace_selftest.c | 2 +- 7 files changed, 168 insertions(+), 55 deletions(-) (limited to 'kernel/trace/trace.c') diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 479ff1737c2f..fa871d53641c 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -1572,12 +1572,28 @@ The above will cause the "foo" tracing instance to trigger a snapshot at the end of boot up. - ftrace_dump_on_oops[=orig_cpu] + ftrace_dump_on_oops[=2(orig_cpu) | =][, | + ,=2(orig_cpu)] [FTRACE] will dump the trace buffers on oops. - If no parameter is passed, ftrace will dump - buffers of all CPUs, but if you pass orig_cpu, it will - dump only the buffer of the CPU that triggered the - oops. + If no parameter is passed, ftrace will dump global + buffers of all CPUs, if you pass 2 or orig_cpu, it + will dump only the buffer of the CPU that triggered + the oops, or the specific instance will be dumped if + its name is passed. Multiple instance dump is also + supported, and instances are separated by commas. Each + instance supports only dump on CPU that triggered the + oops by passing 2 or orig_cpu to it. + + ftrace_dump_on_oops=foo=orig_cpu + + The above will dump only the buffer of "foo" instance + on CPU that triggered the oops. + + ftrace_dump_on_oops,foo,bar=orig_cpu + + The above will dump global buffer on all CPUs, the + buffer of "foo" instance on all CPUs and the buffer + of "bar" instance on CPU that triggered the oops. ftrace_filter=[function-list] [FTRACE] Limit the functions traced by the function diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst index 6584a1f9bfe3..ea8e5f152edc 100644 --- a/Documentation/admin-guide/sysctl/kernel.rst +++ b/Documentation/admin-guide/sysctl/kernel.rst @@ -296,12 +296,30 @@ kernel panic). This will output the contents of the ftrace buffers to the console. This is very useful for capturing traces that lead to crashes and outputting them to a serial console. -= =================================================== -0 Disabled (default). -1 Dump buffers of all CPUs. -2 Dump the buffer of the CPU that triggered the oops. -= =================================================== - +======================= =========================================== +0 Disabled (default). +1 Dump buffers of all CPUs. +2(orig_cpu) Dump the buffer of the CPU that triggered the + oops. + Dump the specific instance buffer on all CPUs. +=2(orig_cpu) Dump the specific instance buffer on the CPU + that triggered the oops. +======================= =========================================== + +Multiple instance dump is also supported, and instances are separated +by commas. If global buffer also needs to be dumped, please specify +the dump mode (1/2/orig_cpu) first for global buffer. + +So for example to dump "foo" and "bar" instance buffer on all CPUs, +user can:: + + echo "foo,bar" > /proc/sys/kernel/ftrace_dump_on_oops + +To dump global buffer and "foo" instance buffer on all +CPUs along with the "bar" instance buffer on CPU that triggered the +oops, user can:: + + echo "1,foo,bar=2" > /proc/sys/kernel/ftrace_dump_on_oops ftrace_enabled, stack_tracer_enabled ==================================== diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index e8921871ef9a..54d53f345d14 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -1151,7 +1151,9 @@ static inline void unpause_graph_tracing(void) { } #ifdef CONFIG_TRACING enum ftrace_dump_mode; -extern enum ftrace_dump_mode ftrace_dump_on_oops; +#define MAX_TRACER_SIZE 100 +extern char ftrace_dump_on_oops[]; +extern int ftrace_dump_on_oops_enabled(void); extern int tracepoint_printk; extern void disable_trace_on_warning(void); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index d718fbec72dd..be2e8c0a187e 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -215,6 +215,7 @@ enum ftrace_dump_mode { DUMP_NONE, DUMP_ALL, DUMP_ORIG, + DUMP_PARAM, }; #ifdef CONFIG_TRACING diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 157f7ce2942d..81cc974913bb 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1710,9 +1710,9 @@ static struct ctl_table kern_table[] = { { .procname = "ftrace_dump_on_oops", .data = &ftrace_dump_on_oops, - .maxlen = sizeof(int), + .maxlen = MAX_TRACER_SIZE, .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dostring, }, { .procname = "traceoff_on_warning", diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 1e1fd377a1cf..233d1af39fff 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -130,9 +130,12 @@ cpumask_var_t __read_mostly tracing_buffer_mask; * /proc/sys/kernel/ftrace_dump_on_oops * Set 1 if you want to dump buffers of all CPUs * Set 2 if you want to dump the buffer of the CPU that triggered oops + * Set instance name if you want to dump the specific trace instance + * Multiple instance dump is also supported, and instances are seperated + * by commas. */ - -enum ftrace_dump_mode ftrace_dump_on_oops; +/* Set to string format zero to disable by default */ +char ftrace_dump_on_oops[MAX_TRACER_SIZE] = "0"; /* When set, tracing will stop when a WARN*() is hit */ int __disable_trace_on_warning; @@ -178,7 +181,6 @@ static void ftrace_trace_userstack(struct trace_array *tr, struct trace_buffer *buffer, unsigned int trace_ctx); -#define MAX_TRACER_SIZE 100 static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata; static char *default_bootup_tracer; @@ -201,19 +203,33 @@ static int __init set_cmdline_ftrace(char *str) } __setup("ftrace=", set_cmdline_ftrace); +int ftrace_dump_on_oops_enabled(void) +{ + if (!strcmp("0", ftrace_dump_on_oops)) + return 0; + else + return 1; +} + static int __init set_ftrace_dump_on_oops(char *str) { - if (*str++ != '=' || !*str || !strcmp("1", str)) { - ftrace_dump_on_oops = DUMP_ALL; + if (!*str) { + strscpy(ftrace_dump_on_oops, "1", MAX_TRACER_SIZE); return 1; } - if (!strcmp("orig_cpu", str) || !strcmp("2", str)) { - ftrace_dump_on_oops = DUMP_ORIG; - return 1; - } + if (*str == ',') { + strscpy(ftrace_dump_on_oops, "1", MAX_TRACER_SIZE); + strscpy(ftrace_dump_on_oops + 1, str, MAX_TRACER_SIZE - 1); + return 1; + } + + if (*str++ == '=') { + strscpy(ftrace_dump_on_oops, str, MAX_TRACER_SIZE); + return 1; + } - return 0; + return 0; } __setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops); @@ -9832,14 +9848,14 @@ static struct notifier_block trace_die_notifier = { static int trace_die_panic_handler(struct notifier_block *self, unsigned long ev, void *unused) { - if (!ftrace_dump_on_oops) + if (!ftrace_dump_on_oops_enabled()) return NOTIFY_DONE; /* The die notifier requires DIE_OOPS to trigger */ if (self == &trace_die_notifier && ev != DIE_OOPS) return NOTIFY_DONE; - ftrace_dump(ftrace_dump_on_oops); + ftrace_dump(DUMP_PARAM); return NOTIFY_DONE; } @@ -9880,12 +9896,12 @@ trace_printk_seq(struct trace_seq *s) trace_seq_init(s); } -void trace_init_global_iter(struct trace_iterator *iter) +static void trace_init_iter(struct trace_iterator *iter, struct trace_array *tr) { - iter->tr = &global_trace; + iter->tr = tr; iter->trace = iter->tr->current_trace; iter->cpu_file = RING_BUFFER_ALL_CPUS; - iter->array_buffer = &global_trace.array_buffer; + iter->array_buffer = &tr->array_buffer; if (iter->trace && iter->trace->open) iter->trace->open(iter); @@ -9905,22 +9921,19 @@ void trace_init_global_iter(struct trace_iterator *iter) iter->fmt_size = STATIC_FMT_BUF_SIZE; } -void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) +void trace_init_global_iter(struct trace_iterator *iter) +{ + trace_init_iter(iter, &global_trace); +} + +static void ftrace_dump_one(struct trace_array *tr, enum ftrace_dump_mode dump_mode) { /* use static because iter can be a bit big for the stack */ static struct trace_iterator iter; - static atomic_t dump_running; - struct trace_array *tr = &global_trace; unsigned int old_userobj; unsigned long flags; int cnt = 0, cpu; - /* Only allow one dump user at a time. */ - if (atomic_inc_return(&dump_running) != 1) { - atomic_dec(&dump_running); - return; - } - /* * Always turn off tracing when we dump. * We don't need to show trace output of what happens @@ -9929,12 +9942,12 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) * If the user does a sysrq-z, then they can re-enable * tracing with echo 1 > tracing_on. */ - tracing_off(); + tracer_tracing_off(tr); local_irq_save(flags); /* Simulate the iterator */ - trace_init_global_iter(&iter); + trace_init_iter(&iter, tr); for_each_tracing_cpu(cpu) { atomic_inc(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); @@ -9945,21 +9958,15 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) /* don't look at user memory in panic mode */ tr->trace_flags &= ~TRACE_ITER_SYM_USEROBJ; - switch (oops_dump_mode) { - case DUMP_ALL: - iter.cpu_file = RING_BUFFER_ALL_CPUS; - break; - case DUMP_ORIG: + if (dump_mode == DUMP_ORIG) iter.cpu_file = raw_smp_processor_id(); - break; - case DUMP_NONE: - goto out_enable; - default: - printk(KERN_TRACE "Bad dumping mode, switching to all CPUs dump\n"); + else iter.cpu_file = RING_BUFFER_ALL_CPUS; - } - printk(KERN_TRACE "Dumping ftrace buffer:\n"); + if (tr == &global_trace) + printk(KERN_TRACE "Dumping ftrace buffer:\n"); + else + printk(KERN_TRACE "Dumping ftrace instance %s buffer:\n", tr->name); /* Did function tracer already get disabled? */ if (ftrace_is_dead()) { @@ -10001,15 +10008,84 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) else printk(KERN_TRACE "---------------------------------\n"); - out_enable: tr->trace_flags |= old_userobj; for_each_tracing_cpu(cpu) { atomic_dec(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); } - atomic_dec(&dump_running); local_irq_restore(flags); } + +static void ftrace_dump_by_param(void) +{ + bool first_param = true; + char dump_param[MAX_TRACER_SIZE]; + char *buf, *token, *inst_name; + struct trace_array *tr; + + strscpy(dump_param, ftrace_dump_on_oops, MAX_TRACER_SIZE); + buf = dump_param; + + while ((token = strsep(&buf, ",")) != NULL) { + if (first_param) { + first_param = false; + if (!strcmp("0", token)) + continue; + else if (!strcmp("1", token)) { + ftrace_dump_one(&global_trace, DUMP_ALL); + continue; + } + else if (!strcmp("2", token) || + !strcmp("orig_cpu", token)) { + ftrace_dump_one(&global_trace, DUMP_ORIG); + continue; + } + } + + inst_name = strsep(&token, "="); + tr = trace_array_find(inst_name); + if (!tr) { + printk(KERN_TRACE "Instance %s not found\n", inst_name); + continue; + } + + if (token && (!strcmp("2", token) || + !strcmp("orig_cpu", token))) + ftrace_dump_one(tr, DUMP_ORIG); + else + ftrace_dump_one(tr, DUMP_ALL); + } +} + +void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) +{ + static atomic_t dump_running; + + /* Only allow one dump user at a time. */ + if (atomic_inc_return(&dump_running) != 1) { + atomic_dec(&dump_running); + return; + } + + switch (oops_dump_mode) { + case DUMP_ALL: + ftrace_dump_one(&global_trace, DUMP_ALL); + break; + case DUMP_ORIG: + ftrace_dump_one(&global_trace, DUMP_ORIG); + break; + case DUMP_PARAM: + ftrace_dump_by_param(); + break; + case DUMP_NONE: + break; + default: + printk(KERN_TRACE "Bad dumping mode, switching to all CPUs dump\n"); + ftrace_dump_one(&global_trace, DUMP_ALL); + } + + atomic_dec(&dump_running); +} EXPORT_SYMBOL_GPL(ftrace_dump); #define WRITE_BUFSIZE 4096 diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 529590499b1f..e9c5058a8efd 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -768,7 +768,7 @@ static int trace_graph_entry_watchdog(struct ftrace_graph_ent *trace) if (unlikely(++graph_hang_thresh > GRAPH_MAX_FUNC_TEST)) { ftrace_graph_stop(); printk(KERN_WARNING "BUG: Function graph tracer hang!\n"); - if (ftrace_dump_on_oops) { + if (ftrace_dump_on_oops_enabled()) { ftrace_dump(DUMP_ALL); /* ftrace_dump() disables tracing */ tracing_on(); -- cgit v1.2.3