From 6ccd83714a009ee301b50c15f6c3a5dc1f30164c Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 29 Jan 2016 10:22:41 -0500 Subject: tracing/stacktrace: Show entire trace if passed in function not found When a max stack trace is discovered, the stack dump is saved. In order to not record the overhead of the stack tracer, the ip of the traced function is looked for within the dump. The trace is started from the location of that function. But if for some reason the ip is not found, the entire stack trace is then truncated. That's not very useful. Instead, print everything if the ip of the traced function is not found within the trace. This issue showed up on s390. Link: http://lkml.kernel.org/r/20160129102241.1b3c9c04@gandalf.local.home Fixes: 72ac426a5bb0 ("tracing: Clean up stack tracing and fix fentry updates") Cc: stable@vger.kernel.org # v4.3+ Reported-by: Heiko Carstens Tested-by: Heiko Carstens Signed-off-by: Steven Rostedt --- kernel/trace/trace_stack.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index dda9e6742950..202df6cffcca 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -125,6 +125,13 @@ check_stack(unsigned long ip, unsigned long *stack) break; } + /* + * Some archs may not have the passed in ip in the dump. + * If that happens, we need to show everything. + */ + if (i == stack_trace_max.nr_entries) + i = 0; + /* * Now find where in the stack these are. */ -- cgit v1.3-14-g43fede From 7dcd182bec271ab341b05b66b6006995795fc0e7 Mon Sep 17 00:00:00 2001 From: Jessica Yu Date: Tue, 16 Feb 2016 17:32:33 -0500 Subject: ftrace/module: remove ftrace module notifier Remove the ftrace module notifier in favor of directly calling ftrace_module_enable() and ftrace_release_mod() in the module loader. Hard-coding the function calls directly in the module loader removes dependence on the module notifier call chain and provides better visibility and control over what gets called when, which is important to kernel utilities such as livepatch. This fixes a notifier ordering issue in which the ftrace module notifier (and hence ftrace_module_enable()) for coming modules was being called after klp_module_notify(), which caused livepatch modules to initialize incorrectly. This patch removes dependence on the module notifier call chain in favor of hard coding the corresponding function calls in the module loader. This ensures that ftrace and livepatch code get called in the correct order on patch module load and unload. Fixes: 5156dca34a3e ("ftrace: Fix the race between ftrace and insmod") Signed-off-by: Jessica Yu Reviewed-by: Steven Rostedt Reviewed-by: Petr Mladek Acked-by: Rusty Russell Reviewed-by: Josh Poimboeuf Reviewed-by: Miroslav Benes Signed-off-by: Jiri Kosina --- include/linux/ftrace.h | 6 ++++-- kernel/module.c | 4 ++++ kernel/trace/ftrace.c | 36 +----------------------------------- 3 files changed, 9 insertions(+), 37 deletions(-) (limited to 'kernel/trace') diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 0639dcc98195..76dc15e171eb 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -604,6 +604,7 @@ extern int ftrace_arch_read_dyn_info(char *buf, int size); extern int skip_trace(unsigned long ip); extern void ftrace_module_init(struct module *mod); +extern void ftrace_module_enable(struct module *mod); extern void ftrace_release_mod(struct module *mod); extern void ftrace_disable_daemon(void); @@ -613,8 +614,9 @@ static inline int skip_trace(unsigned long ip) { return 0; } static inline int ftrace_force_update(void) { return 0; } static inline void ftrace_disable_daemon(void) { } static inline void ftrace_enable_daemon(void) { } -static inline void ftrace_release_mod(struct module *mod) {} -static inline void ftrace_module_init(struct module *mod) {} +static inline void ftrace_module_init(struct module *mod) { } +static inline void ftrace_module_enable(struct module *mod) { } +static inline void ftrace_release_mod(struct module *mod) { } static inline __init int register_ftrace_command(struct ftrace_func_command *cmd) { return -EINVAL; diff --git a/kernel/module.c b/kernel/module.c index 8358f4697c0c..b05d466cfa02 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -981,6 +981,8 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user, mod->exit(); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); + ftrace_release_mod(mod); + async_synchronize_full(); /* Store the name of the last unloaded module for diagnostic purposes */ @@ -3295,6 +3297,7 @@ fail: module_put(mod); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); + ftrace_release_mod(mod); free_module(mod); wake_up_all(&module_wq); return ret; @@ -3371,6 +3374,7 @@ static int complete_formation(struct module *mod, struct load_info *info) mod->state = MODULE_STATE_COMING; mutex_unlock(&module_mutex); + ftrace_module_enable(mod); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod); return 0; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index eca592f977b2..57a6eea84694 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -4961,7 +4961,7 @@ void ftrace_release_mod(struct module *mod) mutex_unlock(&ftrace_lock); } -static void ftrace_module_enable(struct module *mod) +void ftrace_module_enable(struct module *mod) { struct dyn_ftrace *rec; struct ftrace_page *pg; @@ -5038,38 +5038,8 @@ void ftrace_module_init(struct module *mod) ftrace_process_locs(mod, mod->ftrace_callsites, mod->ftrace_callsites + mod->num_ftrace_callsites); } - -static int ftrace_module_notify(struct notifier_block *self, - unsigned long val, void *data) -{ - struct module *mod = data; - - switch (val) { - case MODULE_STATE_COMING: - ftrace_module_enable(mod); - break; - case MODULE_STATE_GOING: - ftrace_release_mod(mod); - break; - default: - break; - } - - return 0; -} -#else -static int ftrace_module_notify(struct notifier_block *self, - unsigned long val, void *data) -{ - return 0; -} #endif /* CONFIG_MODULES */ -struct notifier_block ftrace_module_nb = { - .notifier_call = ftrace_module_notify, - .priority = INT_MIN, /* Run after anything that can remove kprobes */ -}; - void __init ftrace_init(void) { extern unsigned long __start_mcount_loc[]; @@ -5098,10 +5068,6 @@ void __init ftrace_init(void) __start_mcount_loc, __stop_mcount_loc); - ret = register_module_notifier(&ftrace_module_nb); - if (ret) - pr_warning("Failed to register trace ftrace module exit notifier\n"); - set_ftrace_early_filters(); return; -- cgit v1.3-14-g43fede From d5a3b1f691865be576c2bffa708549b8cdccda19 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Wed, 17 Feb 2016 19:58:58 -0800 Subject: bpf: introduce BPF_MAP_TYPE_STACK_TRACE add new map type to store stack traces and corresponding helper bpf_get_stackid(ctx, map, flags) - walk user or kernel stack and return id @ctx: struct pt_regs* @map: pointer to stack_trace map @flags: bits 0-7 - numer of stack frames to skip bit 8 - collect user stack instead of kernel bit 9 - compare stacks by hash only bit 10 - if two different stacks hash into the same stackid discard old other bits - reserved Return: >= 0 stackid on success or negative error stackid is a 32-bit integer handle that can be further combined with other data (including other stackid) and used as a key into maps. Userspace will access stackmap using standard lookup/delete syscall commands to retrieve full stack trace for given stackid. Signed-off-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 1 + include/uapi/linux/bpf.h | 21 +++++ kernel/bpf/Makefile | 3 + kernel/bpf/stackmap.c | 237 +++++++++++++++++++++++++++++++++++++++++++++++ kernel/bpf/verifier.c | 6 +- kernel/trace/bpf_trace.c | 2 + 6 files changed, 269 insertions(+), 1 deletion(-) create mode 100644 kernel/bpf/stackmap.c (limited to 'kernel/trace') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 90ee6ab24bc5..0cadbb7456c0 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -237,6 +237,7 @@ extern const struct bpf_func_proto bpf_get_current_uid_gid_proto; extern const struct bpf_func_proto bpf_get_current_comm_proto; extern const struct bpf_func_proto bpf_skb_vlan_push_proto; extern const struct bpf_func_proto bpf_skb_vlan_pop_proto; +extern const struct bpf_func_proto bpf_get_stackid_proto; /* Shared helpers among cBPF and eBPF. */ void bpf_user_rnd_init_once(void); diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 2ee0fde1bf96..d3e77da8e9e8 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -83,6 +83,7 @@ enum bpf_map_type { BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_MAP_TYPE_PERCPU_HASH, BPF_MAP_TYPE_PERCPU_ARRAY, + BPF_MAP_TYPE_STACK_TRACE, }; enum bpf_prog_type { @@ -272,6 +273,20 @@ enum bpf_func_id { */ BPF_FUNC_perf_event_output, BPF_FUNC_skb_load_bytes, + + /** + * bpf_get_stackid(ctx, map, flags) - walk user or kernel stack and return id + * @ctx: struct pt_regs* + * @map: pointer to stack_trace map + * @flags: bits 0-7 - numer of stack frames to skip + * bit 8 - collect user stack instead of kernel + * bit 9 - compare stacks by hash only + * bit 10 - if two different stacks hash into the same stackid + * discard old + * other bits - reserved + * Return: >= 0 stackid on success or negative error + */ + BPF_FUNC_get_stackid, __BPF_FUNC_MAX_ID, }; @@ -294,6 +309,12 @@ enum bpf_func_id { /* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */ #define BPF_F_TUNINFO_IPV6 (1ULL << 0) +/* BPF_FUNC_get_stackid flags. */ +#define BPF_F_SKIP_FIELD_MASK 0xffULL +#define BPF_F_USER_STACK (1ULL << 8) +#define BPF_F_FAST_STACK_CMP (1ULL << 9) +#define BPF_F_REUSE_STACKID (1ULL << 10) + /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure */ diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile index 13272582eee0..8a932d079c24 100644 --- a/kernel/bpf/Makefile +++ b/kernel/bpf/Makefile @@ -2,3 +2,6 @@ obj-y := core.o obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o +ifeq ($(CONFIG_PERF_EVENTS),y) +obj-$(CONFIG_BPF_SYSCALL) += stackmap.o +endif diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c new file mode 100644 index 000000000000..8a60ee14a977 --- /dev/null +++ b/kernel/bpf/stackmap.c @@ -0,0 +1,237 @@ +/* Copyright (c) 2016 Facebook + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include + +struct stack_map_bucket { + struct rcu_head rcu; + u32 hash; + u32 nr; + u64 ip[]; +}; + +struct bpf_stack_map { + struct bpf_map map; + u32 n_buckets; + struct stack_map_bucket __rcu *buckets[]; +}; + +/* Called from syscall */ +static struct bpf_map *stack_map_alloc(union bpf_attr *attr) +{ + u32 value_size = attr->value_size; + struct bpf_stack_map *smap; + u64 cost, n_buckets; + int err; + + if (!capable(CAP_SYS_ADMIN)) + return ERR_PTR(-EPERM); + + /* check sanity of attributes */ + if (attr->max_entries == 0 || attr->key_size != 4 || + value_size < 8 || value_size % 8 || + value_size / 8 > PERF_MAX_STACK_DEPTH) + return ERR_PTR(-EINVAL); + + /* hash table size must be power of 2 */ + n_buckets = roundup_pow_of_two(attr->max_entries); + + cost = n_buckets * sizeof(struct stack_map_bucket *) + sizeof(*smap); + if (cost >= U32_MAX - PAGE_SIZE) + return ERR_PTR(-E2BIG); + + smap = kzalloc(cost, GFP_USER | __GFP_NOWARN); + if (!smap) { + smap = vzalloc(cost); + if (!smap) + return ERR_PTR(-ENOMEM); + } + + err = -E2BIG; + cost += n_buckets * (value_size + sizeof(struct stack_map_bucket)); + if (cost >= U32_MAX - PAGE_SIZE) + goto free_smap; + + smap->map.map_type = attr->map_type; + smap->map.key_size = attr->key_size; + smap->map.value_size = value_size; + smap->map.max_entries = attr->max_entries; + smap->n_buckets = n_buckets; + smap->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT; + + err = get_callchain_buffers(); + if (err) + goto free_smap; + + return &smap->map; + +free_smap: + kvfree(smap); + return ERR_PTR(err); +} + +static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5) +{ + struct pt_regs *regs = (struct pt_regs *) (long) r1; + struct bpf_map *map = (struct bpf_map *) (long) r2; + struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map); + struct perf_callchain_entry *trace; + struct stack_map_bucket *bucket, *new_bucket, *old_bucket; + u32 max_depth = map->value_size / 8; + /* stack_map_alloc() checks that max_depth <= PERF_MAX_STACK_DEPTH */ + u32 init_nr = PERF_MAX_STACK_DEPTH - max_depth; + u32 skip = flags & BPF_F_SKIP_FIELD_MASK; + u32 hash, id, trace_nr, trace_len; + bool user = flags & BPF_F_USER_STACK; + bool kernel = !user; + u64 *ips; + + if (unlikely(flags & ~(BPF_F_SKIP_FIELD_MASK | BPF_F_USER_STACK | + BPF_F_FAST_STACK_CMP | BPF_F_REUSE_STACKID))) + return -EINVAL; + + trace = get_perf_callchain(regs, init_nr, kernel, user, false, false); + + if (unlikely(!trace)) + /* couldn't fetch the stack trace */ + return -EFAULT; + + /* get_perf_callchain() guarantees that trace->nr >= init_nr + * and trace-nr <= PERF_MAX_STACK_DEPTH, so trace_nr <= max_depth + */ + trace_nr = trace->nr - init_nr; + + if (trace_nr <= skip) + /* skipping more than usable stack trace */ + return -EFAULT; + + trace_nr -= skip; + trace_len = trace_nr * sizeof(u64); + ips = trace->ip + skip + init_nr; + hash = jhash2((u32 *)ips, trace_len / sizeof(u32), 0); + id = hash & (smap->n_buckets - 1); + bucket = rcu_dereference(smap->buckets[id]); + + if (bucket && bucket->hash == hash) { + if (flags & BPF_F_FAST_STACK_CMP) + return id; + if (bucket->nr == trace_nr && + memcmp(bucket->ip, ips, trace_len) == 0) + return id; + } + + /* this call stack is not in the map, try to add it */ + if (bucket && !(flags & BPF_F_REUSE_STACKID)) + return -EEXIST; + + new_bucket = kmalloc(sizeof(struct stack_map_bucket) + map->value_size, + GFP_ATOMIC | __GFP_NOWARN); + if (unlikely(!new_bucket)) + return -ENOMEM; + + memcpy(new_bucket->ip, ips, trace_len); + memset(new_bucket->ip + trace_len / 8, 0, map->value_size - trace_len); + new_bucket->hash = hash; + new_bucket->nr = trace_nr; + + old_bucket = xchg(&smap->buckets[id], new_bucket); + if (old_bucket) + kfree_rcu(old_bucket, rcu); + return id; +} + +const struct bpf_func_proto bpf_get_stackid_proto = { + .func = bpf_get_stackid, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_CONST_MAP_PTR, + .arg3_type = ARG_ANYTHING, +}; + +/* Called from syscall or from eBPF program */ +static void *stack_map_lookup_elem(struct bpf_map *map, void *key) +{ + struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map); + struct stack_map_bucket *bucket; + u32 id = *(u32 *)key; + + if (unlikely(id >= smap->n_buckets)) + return NULL; + bucket = rcu_dereference(smap->buckets[id]); + return bucket ? bucket->ip : NULL; +} + +static int stack_map_get_next_key(struct bpf_map *map, void *key, void *next_key) +{ + return -EINVAL; +} + +static int stack_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) +{ + return -EINVAL; +} + +/* Called from syscall or from eBPF program */ +static int stack_map_delete_elem(struct bpf_map *map, void *key) +{ + struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map); + struct stack_map_bucket *old_bucket; + u32 id = *(u32 *)key; + + if (unlikely(id >= smap->n_buckets)) + return -E2BIG; + + old_bucket = xchg(&smap->buckets[id], NULL); + if (old_bucket) { + kfree_rcu(old_bucket, rcu); + return 0; + } else { + return -ENOENT; + } +} + +/* Called when map->refcnt goes to zero, either from workqueue or from syscall */ +static void stack_map_free(struct bpf_map *map) +{ + struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map); + int i; + + synchronize_rcu(); + + for (i = 0; i < smap->n_buckets; i++) + if (smap->buckets[i]) + kfree_rcu(smap->buckets[i], rcu); + kvfree(smap); + put_callchain_buffers(); +} + +static const struct bpf_map_ops stack_map_ops = { + .map_alloc = stack_map_alloc, + .map_free = stack_map_free, + .map_get_next_key = stack_map_get_next_key, + .map_lookup_elem = stack_map_lookup_elem, + .map_update_elem = stack_map_update_elem, + .map_delete_elem = stack_map_delete_elem, +}; + +static struct bpf_map_type_list stack_map_type __read_mostly = { + .ops = &stack_map_ops, + .type = BPF_MAP_TYPE_STACK_TRACE, +}; + +static int __init register_stack_map(void) +{ + bpf_register_map_type(&stack_map_type); + return 0; +} +late_initcall(register_stack_map); diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index d1d3e8f57de9..42ba4ccc020b 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -246,6 +246,7 @@ static const struct { {BPF_MAP_TYPE_PROG_ARRAY, BPF_FUNC_tail_call}, {BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_read}, {BPF_MAP_TYPE_PERF_EVENT_ARRAY, BPF_FUNC_perf_event_output}, + {BPF_MAP_TYPE_STACK_TRACE, BPF_FUNC_get_stackid}, }; static void print_verifier_state(struct verifier_env *env) @@ -911,8 +912,11 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) * don't allow any other map type to be passed into * the special func; */ - if (bool_func && bool_map != bool_func) + if (bool_func && bool_map != bool_func) { + verbose("cannot pass map_type %d into func %d\n", + map->map_type, func_id); return -EINVAL; + } } return 0; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 326a75e884db..4b8caa392b86 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -299,6 +299,8 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func return &bpf_perf_event_read_proto; case BPF_FUNC_perf_event_output: return &bpf_perf_event_output_proto; + case BPF_FUNC_get_stackid: + return &bpf_get_stackid_proto; default: return NULL; } -- cgit v1.3-14-g43fede