From ca1136c99b66b1566781ff12ecddc635d570f932 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Wed, 12 Jul 2017 11:49:53 -0700 Subject: blktrace: export cgroup info in trace Currently blktrace isn't cgroup aware. blktrace prints out task name of current context, but the task of current context isn't always in the cgroup where the BIO comes from. We can't use task name to find out IO cgroup. For example, Writeback BIOs always comes from flusher thread but the BIOs are for different blk cgroups. Request could be requeued and dispatched from completely different tasks. MD/DM are another examples. This patch tries to fix the gap. We print out cgroup fhandle info in blktrace. Userspace can use open_by_handle_at() syscall to find the cgroup by fhandle. Or userspace can use name_to_handle_at() syscall to find fhandle for a cgroup and use a BPF program to filter out blktrace for a specific cgroup. We add a new 'blk_cgroup' trace option for blk tracer. It's default off. Application which doesn't know the new option isn't affected. When it's on, we output fhandle info right after blk_io_trace with an extra bit set in event action. So from application point of view, blktrace with the option will output new actions. I didn't change blk trace event yet, since I'm not sure if changing the trace event output is an ABI issue. If not, I'll do it later. Acked-by: Steven Rostedt (VMware) Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 231 +++++++++++++++++++++++++++++++++--------------- 1 file changed, 158 insertions(+), 73 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index bc364f86100a..f393d7a43695 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -27,6 +27,7 @@ #include #include #include +#include #include "../../block/blk.h" @@ -46,10 +47,14 @@ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(running_trace_lock); /* Select an alternative, minimalistic output than the original one */ #define TRACE_BLK_OPT_CLASSIC 0x1 +#define TRACE_BLK_OPT_CGROUP 0x2 static struct tracer_opt blk_tracer_opts[] = { /* Default disable the minimalistic output */ { TRACER_OPT(blk_classic, TRACE_BLK_OPT_CLASSIC) }, +#ifdef CONFIG_BLK_CGROUP + { TRACER_OPT(blk_cgroup, TRACE_BLK_OPT_CGROUP) }, +#endif { } }; @@ -68,7 +73,8 @@ static void blk_unregister_tracepoints(void); * Send out a notify message. */ static void trace_note(struct blk_trace *bt, pid_t pid, int action, - const void *data, size_t len) + const void *data, size_t len, + union kernfs_node_id *cgid) { struct blk_io_trace *t; struct ring_buffer_event *event = NULL; @@ -76,12 +82,13 @@ static void trace_note(struct blk_trace *bt, pid_t pid, int action, int pc = 0; int cpu = smp_processor_id(); bool blk_tracer = blk_tracer_enabled; + ssize_t cgid_len = cgid ? sizeof(*cgid) : 0; if (blk_tracer) { buffer = blk_tr->trace_buffer.buffer; pc = preempt_count(); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, - sizeof(*t) + len, + sizeof(*t) + len + cgid_len, 0, pc); if (!event) return; @@ -92,17 +99,19 @@ static void trace_note(struct blk_trace *bt, pid_t pid, int action, if (!bt->rchan) return; - t = relay_reserve(bt->rchan, sizeof(*t) + len); + t = relay_reserve(bt->rchan, sizeof(*t) + len + cgid_len); if (t) { t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; t->time = ktime_to_ns(ktime_get()); record_it: t->device = bt->dev; - t->action = action; + t->action = action | (cgid ? __BLK_TN_CGROUP : 0); t->pid = pid; t->cpu = cpu; - t->pdu_len = len; - memcpy((void *) t + sizeof(*t), data, len); + t->pdu_len = len + cgid_len; + if (cgid) + memcpy((void *)t + sizeof(*t), cgid, cgid_len); + memcpy((void *) t + sizeof(*t) + cgid_len, data, len); if (blk_tracer) trace_buffer_unlock_commit(blk_tr, buffer, event, 0, pc); @@ -122,7 +131,7 @@ static void trace_note_tsk(struct task_struct *tsk) spin_lock_irqsave(&running_trace_lock, flags); list_for_each_entry(bt, &running_trace_list, running_list) { trace_note(bt, tsk->pid, BLK_TN_PROCESS, tsk->comm, - sizeof(tsk->comm)); + sizeof(tsk->comm), NULL); } spin_unlock_irqrestore(&running_trace_lock, flags); } @@ -139,7 +148,7 @@ static void trace_note_time(struct blk_trace *bt) words[1] = now.tv_nsec; local_irq_save(flags); - trace_note(bt, 0, BLK_TN_TIMESTAMP, words, sizeof(words)); + trace_note(bt, 0, BLK_TN_TIMESTAMP, words, sizeof(words), NULL); local_irq_restore(flags); } @@ -167,7 +176,7 @@ void __trace_note_message(struct blk_trace *bt, const char *fmt, ...) n = vscnprintf(buf, BLK_TN_MAX_MSG, fmt, args); va_end(args); - trace_note(bt, 0, BLK_TN_MESSAGE, buf, n); + trace_note(bt, 0, BLK_TN_MESSAGE, buf, n, NULL); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(__trace_note_message); @@ -204,7 +213,7 @@ static const u32 ddir_act[2] = { BLK_TC_ACT(BLK_TC_READ), */ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, int op, int op_flags, u32 what, int error, int pdu_len, - void *pdu_data) + void *pdu_data, union kernfs_node_id *cgid) { struct task_struct *tsk = current; struct ring_buffer_event *event = NULL; @@ -215,6 +224,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, pid_t pid; int cpu, pc = 0; bool blk_tracer = blk_tracer_enabled; + ssize_t cgid_len = cgid ? sizeof(*cgid) : 0; if (unlikely(bt->trace_state != Blktrace_running && !blk_tracer)) return; @@ -229,6 +239,8 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, what |= BLK_TC_ACT(BLK_TC_DISCARD); if (op == REQ_OP_FLUSH) what |= BLK_TC_ACT(BLK_TC_FLUSH); + if (cgid) + what |= __BLK_TA_CGROUP; pid = tsk->pid; if (act_log_check(bt, what, sector, pid)) @@ -241,7 +253,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, buffer = blk_tr->trace_buffer.buffer; pc = preempt_count(); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, - sizeof(*t) + pdu_len, + sizeof(*t) + pdu_len + cgid_len, 0, pc); if (!event) return; @@ -258,7 +270,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, * from coming in and stepping on our toes. */ local_irq_save(flags); - t = relay_reserve(bt->rchan, sizeof(*t) + pdu_len); + t = relay_reserve(bt->rchan, sizeof(*t) + pdu_len + cgid_len); if (t) { sequence = per_cpu_ptr(bt->sequence, cpu); @@ -280,10 +292,12 @@ record_it: t->action = what; t->device = bt->dev; t->error = error; - t->pdu_len = pdu_len; + t->pdu_len = pdu_len + cgid_len; + if (cgid_len) + memcpy((void *)t + sizeof(*t), cgid, cgid_len); if (pdu_len) - memcpy((void *) t + sizeof(*t), pdu_data, pdu_len); + memcpy((void *)t + sizeof(*t) + cgid_len, pdu_data, pdu_len); if (blk_tracer) { trace_buffer_unlock_commit(blk_tr, buffer, event, 0, pc); @@ -684,6 +698,36 @@ void blk_trace_shutdown(struct request_queue *q) } } +#ifdef CONFIG_BLK_CGROUP +static union kernfs_node_id * +blk_trace_bio_get_cgid(struct request_queue *q, struct bio *bio) +{ + struct blk_trace *bt = q->blk_trace; + + if (!bt || !(blk_tracer_flags.val & TRACE_BLK_OPT_CGROUP)) + return NULL; + + if (!bio->bi_css) + return NULL; + return cgroup_get_kernfs_id(bio->bi_css->cgroup); +} +#else +static union kernfs_node_id * +blk_trace_bio_get_cgid(struct request_queue *q, struct bio *bio) +{ + return NULL; +} +#endif + +static union kernfs_node_id * +blk_trace_request_get_cgid(struct request_queue *q, struct request *rq) +{ + if (!rq->bio) + return NULL; + /* Use the first bio */ + return blk_trace_bio_get_cgid(q, rq->bio); +} + /* * blktrace probes */ @@ -694,13 +738,15 @@ void blk_trace_shutdown(struct request_queue *q) * @error: return status to log * @nr_bytes: number of completed bytes * @what: the action + * @cgid: the cgroup info * * Description: * Records an action against a request. Will log the bio offset + size. * **/ static void blk_add_trace_rq(struct request *rq, int error, - unsigned int nr_bytes, u32 what) + unsigned int nr_bytes, u32 what, + union kernfs_node_id *cgid) { struct blk_trace *bt = rq->q->blk_trace; @@ -713,32 +759,36 @@ static void blk_add_trace_rq(struct request *rq, int error, what |= BLK_TC_ACT(BLK_TC_FS); __blk_add_trace(bt, blk_rq_trace_sector(rq), nr_bytes, req_op(rq), - rq->cmd_flags, what, error, 0, NULL); + rq->cmd_flags, what, error, 0, NULL, cgid); } static void blk_add_trace_rq_insert(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_INSERT); + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_INSERT, + blk_trace_request_get_cgid(q, rq)); } static void blk_add_trace_rq_issue(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_ISSUE); + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_ISSUE, + blk_trace_request_get_cgid(q, rq)); } static void blk_add_trace_rq_requeue(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_REQUEUE); + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_REQUEUE, + blk_trace_request_get_cgid(q, rq)); } static void blk_add_trace_rq_complete(void *ignore, struct request *rq, int error, unsigned int nr_bytes) { - blk_add_trace_rq(rq, error, nr_bytes, BLK_TA_COMPLETE); + blk_add_trace_rq(rq, error, nr_bytes, BLK_TA_COMPLETE, + blk_trace_request_get_cgid(rq->q, rq)); } /** @@ -753,7 +803,7 @@ static void blk_add_trace_rq_complete(void *ignore, struct request *rq, * **/ static void blk_add_trace_bio(struct request_queue *q, struct bio *bio, - u32 what, int error) + u32 what, int error, union kernfs_node_id *cgid) { struct blk_trace *bt = q->blk_trace; @@ -761,20 +811,22 @@ static void blk_add_trace_bio(struct request_queue *q, struct bio *bio, return; __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, - bio_op(bio), bio->bi_opf, what, error, 0, NULL); + bio_op(bio), bio->bi_opf, what, error, 0, NULL, cgid); } static void blk_add_trace_bio_bounce(void *ignore, struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_BOUNCE, 0); + blk_add_trace_bio(q, bio, BLK_TA_BOUNCE, 0, + blk_trace_bio_get_cgid(q, bio)); } static void blk_add_trace_bio_complete(void *ignore, struct request_queue *q, struct bio *bio, int error) { - blk_add_trace_bio(q, bio, BLK_TA_COMPLETE, error); + blk_add_trace_bio(q, bio, BLK_TA_COMPLETE, error, + blk_trace_bio_get_cgid(q, bio)); } static void blk_add_trace_bio_backmerge(void *ignore, @@ -782,7 +834,8 @@ static void blk_add_trace_bio_backmerge(void *ignore, struct request *rq, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE, 0); + blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE, 0, + blk_trace_bio_get_cgid(q, bio)); } static void blk_add_trace_bio_frontmerge(void *ignore, @@ -790,13 +843,15 @@ static void blk_add_trace_bio_frontmerge(void *ignore, struct request *rq, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE, 0); + blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE, 0, + blk_trace_bio_get_cgid(q, bio)); } static void blk_add_trace_bio_queue(void *ignore, struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_QUEUE, 0); + blk_add_trace_bio(q, bio, BLK_TA_QUEUE, 0, + blk_trace_bio_get_cgid(q, bio)); } static void blk_add_trace_getrq(void *ignore, @@ -804,13 +859,14 @@ static void blk_add_trace_getrq(void *ignore, struct bio *bio, int rw) { if (bio) - blk_add_trace_bio(q, bio, BLK_TA_GETRQ, 0); + blk_add_trace_bio(q, bio, BLK_TA_GETRQ, 0, + blk_trace_bio_get_cgid(q, bio)); else { struct blk_trace *bt = q->blk_trace; if (bt) __blk_add_trace(bt, 0, 0, rw, 0, BLK_TA_GETRQ, 0, 0, - NULL); + NULL, NULL); } } @@ -820,13 +876,14 @@ static void blk_add_trace_sleeprq(void *ignore, struct bio *bio, int rw) { if (bio) - blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ, 0); + blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ, 0, + blk_trace_bio_get_cgid(q, bio)); else { struct blk_trace *bt = q->blk_trace; if (bt) __blk_add_trace(bt, 0, 0, rw, 0, BLK_TA_SLEEPRQ, - 0, 0, NULL); + 0, 0, NULL, NULL); } } @@ -835,7 +892,7 @@ static void blk_add_trace_plug(void *ignore, struct request_queue *q) struct blk_trace *bt = q->blk_trace; if (bt) - __blk_add_trace(bt, 0, 0, 0, 0, BLK_TA_PLUG, 0, 0, NULL); + __blk_add_trace(bt, 0, 0, 0, 0, BLK_TA_PLUG, 0, 0, NULL, NULL); } static void blk_add_trace_unplug(void *ignore, struct request_queue *q, @@ -852,7 +909,7 @@ static void blk_add_trace_unplug(void *ignore, struct request_queue *q, else what = BLK_TA_UNPLUG_TIMER; - __blk_add_trace(bt, 0, 0, 0, 0, what, 0, sizeof(rpdu), &rpdu); + __blk_add_trace(bt, 0, 0, 0, 0, what, 0, sizeof(rpdu), &rpdu, NULL); } } @@ -868,7 +925,7 @@ static void blk_add_trace_split(void *ignore, __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, bio_op(bio), bio->bi_opf, BLK_TA_SPLIT, bio->bi_status, sizeof(rpdu), - &rpdu); + &rpdu, blk_trace_bio_get_cgid(q, bio)); } } @@ -901,7 +958,7 @@ static void blk_add_trace_bio_remap(void *ignore, __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, bio_op(bio), bio->bi_opf, BLK_TA_REMAP, bio->bi_status, - sizeof(r), &r); + sizeof(r), &r, blk_trace_bio_get_cgid(q, bio)); } /** @@ -934,7 +991,7 @@ static void blk_add_trace_rq_remap(void *ignore, __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), rq_data_dir(rq), 0, BLK_TA_REMAP, 0, - sizeof(r), &r); + sizeof(r), &r, blk_trace_request_get_cgid(q, rq)); } /** @@ -958,7 +1015,8 @@ void blk_add_driver_data(struct request_queue *q, return; __blk_add_trace(bt, blk_rq_trace_sector(rq), blk_rq_bytes(rq), 0, 0, - BLK_TA_DRV_DATA, 0, len, data); + BLK_TA_DRV_DATA, 0, len, data, + blk_trace_request_get_cgid(q, rq)); } EXPORT_SYMBOL_GPL(blk_add_driver_data); @@ -1031,7 +1089,7 @@ static void fill_rwbs(char *rwbs, const struct blk_io_trace *t) int i = 0; int tc = t->action >> BLK_TC_SHIFT; - if (t->action == BLK_TN_MESSAGE) { + if ((t->action & ~__BLK_TN_CGROUP) == BLK_TN_MESSAGE) { rwbs[i++] = 'N'; goto out; } @@ -1066,9 +1124,21 @@ const struct blk_io_trace *te_blk_io_trace(const struct trace_entry *ent) return (const struct blk_io_trace *)ent; } -static inline const void *pdu_start(const struct trace_entry *ent) +static inline const void *pdu_start(const struct trace_entry *ent, bool has_cg) { - return te_blk_io_trace(ent) + 1; + return (void *)(te_blk_io_trace(ent) + 1) + + (has_cg ? sizeof(union kernfs_node_id) : 0); +} + +static inline const void *cgid_start(const struct trace_entry *ent) +{ + return (void *)(te_blk_io_trace(ent) + 1); +} + +static inline int pdu_real_len(const struct trace_entry *ent, bool has_cg) +{ + return te_blk_io_trace(ent)->pdu_len - + (has_cg ? sizeof(union kernfs_node_id) : 0); } static inline u32 t_action(const struct trace_entry *ent) @@ -1096,16 +1166,16 @@ static inline __u16 t_error(const struct trace_entry *ent) return te_blk_io_trace(ent)->error; } -static __u64 get_pdu_int(const struct trace_entry *ent) +static __u64 get_pdu_int(const struct trace_entry *ent, bool has_cg) { - const __u64 *val = pdu_start(ent); + const __u64 *val = pdu_start(ent, has_cg); return be64_to_cpu(*val); } static void get_pdu_remap(const struct trace_entry *ent, - struct blk_io_trace_remap *r) + struct blk_io_trace_remap *r, bool has_cg) { - const struct blk_io_trace_remap *__r = pdu_start(ent); + const struct blk_io_trace_remap *__r = pdu_start(ent, has_cg); __u64 sector_from = __r->sector_from; r->device_from = be32_to_cpu(__r->device_from); @@ -1113,9 +1183,11 @@ static void get_pdu_remap(const struct trace_entry *ent, r->sector_from = be64_to_cpu(sector_from); } -typedef void (blk_log_action_t) (struct trace_iterator *iter, const char *act); +typedef void (blk_log_action_t) (struct trace_iterator *iter, const char *act, + bool has_cg); -static void blk_log_action_classic(struct trace_iterator *iter, const char *act) +static void blk_log_action_classic(struct trace_iterator *iter, const char *act, + bool has_cg) { char rwbs[RWBS_LEN]; unsigned long long ts = iter->ts; @@ -1131,24 +1203,33 @@ static void blk_log_action_classic(struct trace_iterator *iter, const char *act) secs, nsec_rem, iter->ent->pid, act, rwbs); } -static void blk_log_action(struct trace_iterator *iter, const char *act) +static void blk_log_action(struct trace_iterator *iter, const char *act, + bool has_cg) { char rwbs[RWBS_LEN]; const struct blk_io_trace *t = te_blk_io_trace(iter->ent); fill_rwbs(rwbs, t); - trace_seq_printf(&iter->seq, "%3d,%-3d %2s %3s ", - MAJOR(t->device), MINOR(t->device), act, rwbs); + if (has_cg) { + const union kernfs_node_id *id = cgid_start(iter->ent); + + trace_seq_printf(&iter->seq, "%3d,%-3d %x,%-x %2s %3s ", + MAJOR(t->device), MINOR(t->device), + id->ino, id->generation, act, rwbs); + } else + trace_seq_printf(&iter->seq, "%3d,%-3d %2s %3s ", + MAJOR(t->device), MINOR(t->device), act, rwbs); } -static void blk_log_dump_pdu(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_dump_pdu(struct trace_seq *s, + const struct trace_entry *ent, bool has_cg) { const unsigned char *pdu_buf; int pdu_len; int i, end; - pdu_buf = pdu_start(ent); - pdu_len = te_blk_io_trace(ent)->pdu_len; + pdu_buf = pdu_start(ent, has_cg); + pdu_len = pdu_real_len(ent, has_cg); if (!pdu_len) return; @@ -1179,7 +1260,7 @@ static void blk_log_dump_pdu(struct trace_seq *s, const struct trace_entry *ent) trace_seq_puts(s, ") "); } -static void blk_log_generic(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_generic(struct trace_seq *s, const struct trace_entry *ent, bool has_cg) { char cmd[TASK_COMM_LEN]; @@ -1187,7 +1268,7 @@ static void blk_log_generic(struct trace_seq *s, const struct trace_entry *ent) if (t_action(ent) & BLK_TC_ACT(BLK_TC_PC)) { trace_seq_printf(s, "%u ", t_bytes(ent)); - blk_log_dump_pdu(s, ent); + blk_log_dump_pdu(s, ent, has_cg); trace_seq_printf(s, "[%s]\n", cmd); } else { if (t_sec(ent)) @@ -1199,10 +1280,10 @@ static void blk_log_generic(struct trace_seq *s, const struct trace_entry *ent) } static void blk_log_with_error(struct trace_seq *s, - const struct trace_entry *ent) + const struct trace_entry *ent, bool has_cg) { if (t_action(ent) & BLK_TC_ACT(BLK_TC_PC)) { - blk_log_dump_pdu(s, ent); + blk_log_dump_pdu(s, ent, has_cg); trace_seq_printf(s, "[%d]\n", t_error(ent)); } else { if (t_sec(ent)) @@ -1215,18 +1296,18 @@ static void blk_log_with_error(struct trace_seq *s, } } -static void blk_log_remap(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_remap(struct trace_seq *s, const struct trace_entry *ent, bool has_cg) { struct blk_io_trace_remap r = { .device_from = 0, }; - get_pdu_remap(ent, &r); + get_pdu_remap(ent, &r, has_cg); trace_seq_printf(s, "%llu + %u <- (%d,%d) %llu\n", t_sector(ent), t_sec(ent), MAJOR(r.device_from), MINOR(r.device_from), (unsigned long long)r.sector_from); } -static void blk_log_plug(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_plug(struct trace_seq *s, const struct trace_entry *ent, bool has_cg) { char cmd[TASK_COMM_LEN]; @@ -1235,30 +1316,31 @@ static void blk_log_plug(struct trace_seq *s, const struct trace_entry *ent) trace_seq_printf(s, "[%s]\n", cmd); } -static void blk_log_unplug(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_unplug(struct trace_seq *s, const struct trace_entry *ent, bool has_cg) { char cmd[TASK_COMM_LEN]; trace_find_cmdline(ent->pid, cmd); - trace_seq_printf(s, "[%s] %llu\n", cmd, get_pdu_int(ent)); + trace_seq_printf(s, "[%s] %llu\n", cmd, get_pdu_int(ent, has_cg)); } -static void blk_log_split(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_split(struct trace_seq *s, const struct trace_entry *ent, bool has_cg) { char cmd[TASK_COMM_LEN]; trace_find_cmdline(ent->pid, cmd); trace_seq_printf(s, "%llu / %llu [%s]\n", t_sector(ent), - get_pdu_int(ent), cmd); + get_pdu_int(ent, has_cg), cmd); } -static void blk_log_msg(struct trace_seq *s, const struct trace_entry *ent) +static void blk_log_msg(struct trace_seq *s, const struct trace_entry *ent, + bool has_cg) { - const struct blk_io_trace *t = te_blk_io_trace(ent); - trace_seq_putmem(s, t + 1, t->pdu_len); + trace_seq_putmem(s, pdu_start(ent, has_cg), + pdu_real_len(ent, has_cg)); trace_seq_putc(s, '\n'); } @@ -1298,7 +1380,8 @@ static void blk_tracer_reset(struct trace_array *tr) static const struct { const char *act[2]; - void (*print)(struct trace_seq *s, const struct trace_entry *ent); + void (*print)(struct trace_seq *s, const struct trace_entry *ent, + bool has_cg); } what2act[] = { [__BLK_TA_QUEUE] = {{ "Q", "queue" }, blk_log_generic }, [__BLK_TA_BACKMERGE] = {{ "M", "backmerge" }, blk_log_generic }, @@ -1326,23 +1409,25 @@ static enum print_line_t print_one_line(struct trace_iterator *iter, u16 what; bool long_act; blk_log_action_t *log_action; + bool has_cg; t = te_blk_io_trace(iter->ent); - what = t->action & ((1 << BLK_TC_SHIFT) - 1); + what = (t->action & ((1 << BLK_TC_SHIFT) - 1)) & ~__BLK_TA_CGROUP; long_act = !!(tr->trace_flags & TRACE_ITER_VERBOSE); log_action = classic ? &blk_log_action_classic : &blk_log_action; + has_cg = t->action & __BLK_TA_CGROUP; - if (t->action == BLK_TN_MESSAGE) { - log_action(iter, long_act ? "message" : "m"); - blk_log_msg(s, iter->ent); + if ((t->action & ~__BLK_TN_CGROUP) == BLK_TN_MESSAGE) { + log_action(iter, long_act ? "message" : "m", has_cg); + blk_log_msg(s, iter->ent, has_cg); return trace_handle_return(s); } if (unlikely(what == 0 || what >= ARRAY_SIZE(what2act))) trace_seq_printf(s, "Unknown action %x\n", what); else { - log_action(iter, what2act[what].act[long_act]); - what2act[what].print(s, iter->ent); + log_action(iter, what2act[what].act[long_act], has_cg); + what2act[what].print(s, iter->ent, has_cg); } return trace_handle_return(s); -- cgit v1.2.3 From 69fd5c391763bd94a40dd152bc72a7f230137150 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Wed, 12 Jul 2017 11:49:55 -0700 Subject: blktrace: add an option to allow displaying cgroup path By default we output cgroup id in blktrace. This adds an option to display cgroup path. Since get cgroup path is a relativly heavy operation, we don't enable it by default. with the option enabled, blktrace will output something like this: dd-1353 [007] d..2 293.015252: 8,0 /test/level D R 24 + 8 [dd] Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index f393d7a43695..e90974ed4532 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -48,12 +48,14 @@ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(running_trace_lock); /* Select an alternative, minimalistic output than the original one */ #define TRACE_BLK_OPT_CLASSIC 0x1 #define TRACE_BLK_OPT_CGROUP 0x2 +#define TRACE_BLK_OPT_CGNAME 0x4 static struct tracer_opt blk_tracer_opts[] = { /* Default disable the minimalistic output */ { TRACER_OPT(blk_classic, TRACE_BLK_OPT_CLASSIC) }, #ifdef CONFIG_BLK_CGROUP { TRACER_OPT(blk_cgroup, TRACE_BLK_OPT_CGROUP) }, + { TRACER_OPT(blk_cgname, TRACE_BLK_OPT_CGNAME) }, #endif { } }; @@ -1213,7 +1215,17 @@ static void blk_log_action(struct trace_iterator *iter, const char *act, if (has_cg) { const union kernfs_node_id *id = cgid_start(iter->ent); - trace_seq_printf(&iter->seq, "%3d,%-3d %x,%-x %2s %3s ", + if (blk_tracer_flags.val & TRACE_BLK_OPT_CGNAME) { + char blkcg_name_buf[NAME_MAX + 1] = "<...>"; + + cgroup_path_from_kernfs_id(id, blkcg_name_buf, + sizeof(blkcg_name_buf)); + trace_seq_printf(&iter->seq, "%3d,%-3d %s %2s %3s ", + MAJOR(t->device), MINOR(t->device), + blkcg_name_buf, act, rwbs); + } else + trace_seq_printf(&iter->seq, + "%3d,%-3d %x,%-x %2s %3s ", MAJOR(t->device), MINOR(t->device), id->ino, id->generation, act, rwbs); } else -- cgit v1.2.3 From 35fe6d763229e8fc0eb5f9b93a401673cfcb5e1e Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Wed, 12 Jul 2017 11:49:56 -0700 Subject: block: use standard blktrace API to output cgroup info for debug notes Currently cfq/bfq/blk-throttle output cgroup info in trace in their own way. Now we have standard blktrace API for this, so convert them to use it. Note, this changes the behavior a little bit. cgroup info isn't output by default, we only do this with 'blk_cgroup' option enabled. cgroup info isn't output as a string by default too, we only do this with 'blk_cgname' option enabled. Also cgroup info is output in different position of the note string. I think these behavior changes aren't a big issue (actually we make trace data shorter which is good), since the blktrace note is solely for debugging. Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index e90974ed4532..7724de18d2fe 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -154,7 +154,8 @@ static void trace_note_time(struct blk_trace *bt) local_irq_restore(flags); } -void __trace_note_message(struct blk_trace *bt, const char *fmt, ...) +void __trace_note_message(struct blk_trace *bt, struct blkcg *blkcg, + const char *fmt, ...) { int n; va_list args; @@ -178,7 +179,14 @@ void __trace_note_message(struct blk_trace *bt, const char *fmt, ...) n = vscnprintf(buf, BLK_TN_MAX_MSG, fmt, args); va_end(args); + if (!(blk_tracer_flags.val & TRACE_BLK_OPT_CGROUP)) + blkcg = NULL; +#ifdef CONFIG_BLK_CGROUP + trace_note(bt, 0, BLK_TN_MESSAGE, buf, n, + blkcg ? cgroup_get_kernfs_id(blkcg->css.cgroup) : NULL); +#else trace_note(bt, 0, BLK_TN_MESSAGE, buf, n, NULL); +#endif local_irq_restore(flags); } EXPORT_SYMBOL_GPL(__trace_note_message); @@ -375,7 +383,7 @@ static ssize_t blk_msg_write(struct file *filp, const char __user *buffer, return PTR_ERR(msg); bt = filp->private_data; - __trace_note_message(bt, "%s", msg); + __trace_note_message(bt, NULL, "%s", msg); kfree(msg); return count; -- cgit v1.2.3 From 4bb0f0e73c8c30917d169c4a0f1ac083690c545b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 1 Aug 2017 12:01:52 -0400 Subject: tracing: Call clear_boot_tracer() at lateinit_sync The clear_boot_tracer function is used to reset the default_bootup_tracer string to prevent it from being accessed after boot, as it originally points to init data. But since clear_boot_tracer() is called via the init_lateinit() call, it races with the initcall for registering the hwlat tracer. If someone adds "ftrace=hwlat" to the kernel command line, depending on how the linker sets up the text, the saved command line may be cleared, and the hwlat tracer never is initialized. Simply have the clear_boot_tracer() be called by initcall_lateinit_sync() as that's for tasks to be called after lateinit. Link: https://bugzilla.kernel.org/show_bug.cgi?id=196551 Cc: stable@vger.kernel.org Fixes: e7c15cd8a ("tracing: Added hardware latency tracer") Reported-by: Zamir SUN Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 42b9355033d4..784fb43b2abe 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -8407,4 +8407,4 @@ __init static int clear_boot_tracer(void) } fs_initcall(tracer_init_tracefs); -late_initcall(clear_boot_tracer); +late_initcall_sync(clear_boot_tracer); -- cgit v1.2.3 From 147d88e0b5eb90191bc5c12ca0a3c410b75a13d2 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 1 Aug 2017 14:02:01 +0300 Subject: tracing: Missing error code in tracer_alloc_buffers() If ring_buffer_alloc() or one of the next couple function calls fail then we should return -ENOMEM but the current code returns success. Link: http://lkml.kernel.org/r/20170801110201.ajdkct7vwzixahvx@mwanda Cc: Sebastian Andrzej Siewior Cc: Ingo Molnar Cc: stable@vger.kernel.org Fixes: b32614c03413 ('tracing/rb: Convert to hotplug state machine') Signed-off-by: Dan Carpenter Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 784fb43b2abe..d815fc317e9d 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -8293,6 +8293,7 @@ __init static int tracer_alloc_buffers(void) if (ret < 0) goto out_free_cpumask; /* Used for event triggers */ + ret = -ENOMEM; temp_buffer = ring_buffer_alloc(PAGE_SIZE, RB_FL_OVERWRITE); if (!temp_buffer) goto out_rm_hp_state; -- cgit v1.2.3 From a7e52ad7ed82e21273eccff93d1477a7b313aabb Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Wed, 2 Aug 2017 14:20:54 -0400 Subject: ring-buffer: Have ring_buffer_alloc_read_page() return error on offline CPU Chunyu Hu reported: "per_cpu trace directories and files are created for all possible cpus, but only the cpus which have ever been on-lined have their own per cpu ring buffer (allocated by cpuhp threads). While trace_buffers_open, the open handler for trace file 'trace_pipe_raw' is always trying to access field of ring_buffer_per_cpu, and would panic with the NULL pointer. Align the behavior of trace_pipe_raw with trace_pipe, that returns -NODEV when openning it if that cpu does not have trace ring buffer. Reproduce: cat /sys/kernel/debug/tracing/per_cpu/cpu31/trace_pipe_raw (cpu31 is never on-lined, this is a 16 cores x86_64 box) Tested with: 1) boot with maxcpus=14, read trace_pipe_raw of cpu15. Got -NODEV. 2) oneline cpu15, read trace_pipe_raw of cpu15. Get the raw trace data. Call trace: [ 5760.950995] RIP: 0010:ring_buffer_alloc_read_page+0x32/0xe0 [ 5760.961678] tracing_buffers_read+0x1f6/0x230 [ 5760.962695] __vfs_read+0x37/0x160 [ 5760.963498] ? __vfs_read+0x5/0x160 [ 5760.964339] ? security_file_permission+0x9d/0xc0 [ 5760.965451] ? __vfs_read+0x5/0x160 [ 5760.966280] vfs_read+0x8c/0x130 [ 5760.967070] SyS_read+0x55/0xc0 [ 5760.967779] do_syscall_64+0x67/0x150 [ 5760.968687] entry_SYSCALL64_slow_path+0x25/0x25" This was introduced by the addition of the feature to reuse reader pages instead of re-allocating them. The problem is that the allocation of a reader page (which is per cpu) does not check if the cpu is online and set up for the ring buffer. Link: http://lkml.kernel.org/r/1500880866-1177-1-git-send-email-chuhu@redhat.com Cc: stable@vger.kernel.org Fixes: 73a757e63114 ("ring-buffer: Return reader page back into existing ring buffer") Reported-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ring_buffer.c | 14 +++++++++----- kernel/trace/ring_buffer_benchmark.c | 2 +- kernel/trace/trace.c | 16 +++++++++++----- 3 files changed, 21 insertions(+), 11 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 529cc50d7243..81279c6602ff 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -4386,15 +4386,19 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); * the page that was allocated, with the read page of the buffer. * * Returns: - * The page allocated, or NULL on error. + * The page allocated, or ERR_PTR */ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) { - struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; + struct ring_buffer_per_cpu *cpu_buffer; struct buffer_data_page *bpage = NULL; unsigned long flags; struct page *page; + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return ERR_PTR(-ENODEV); + + cpu_buffer = buffer->buffers[cpu]; local_irq_save(flags); arch_spin_lock(&cpu_buffer->lock); @@ -4412,7 +4416,7 @@ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); if (!page) - return NULL; + return ERR_PTR(-ENOMEM); bpage = page_address(page); @@ -4467,8 +4471,8 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); * * for example: * rpage = ring_buffer_alloc_read_page(buffer, cpu); - * if (!rpage) - * return error; + * if (IS_ERR(rpage)) + * return PTR_ERR(rpage); * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); * if (ret >= 0) * process_page(rpage, ret); diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c index 9fbcaf567886..68ee79afe31c 100644 --- a/kernel/trace/ring_buffer_benchmark.c +++ b/kernel/trace/ring_buffer_benchmark.c @@ -113,7 +113,7 @@ static enum event_status read_page(int cpu) int i; bpage = ring_buffer_alloc_read_page(buffer, cpu); - if (!bpage) + if (IS_ERR(bpage)) return EVENT_DROPPED; ret = ring_buffer_read_page(buffer, &bpage, PAGE_SIZE, cpu, 1); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d815fc317e9d..44004d8aa3b3 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6598,7 +6598,7 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, { struct ftrace_buffer_info *info = filp->private_data; struct trace_iterator *iter = &info->iter; - ssize_t ret; + ssize_t ret = 0; ssize_t size; if (!count) @@ -6612,10 +6612,15 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, if (!info->spare) { info->spare = ring_buffer_alloc_read_page(iter->trace_buffer->buffer, iter->cpu_file); - info->spare_cpu = iter->cpu_file; + if (IS_ERR(info->spare)) { + ret = PTR_ERR(info->spare); + info->spare = NULL; + } else { + info->spare_cpu = iter->cpu_file; + } } if (!info->spare) - return -ENOMEM; + return ret; /* Do we have previous read data to read? */ if (info->read < PAGE_SIZE) @@ -6790,8 +6795,9 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, ref->ref = 1; ref->buffer = iter->trace_buffer->buffer; ref->page = ring_buffer_alloc_read_page(ref->buffer, iter->cpu_file); - if (!ref->page) { - ret = -ENOMEM; + if (IS_ERR(ref->page)) { + ret = PTR_ERR(ref->page); + ref->page = NULL; kfree(ref); break; } -- cgit v1.2.3 From cf5f5cea270655dd49370760576c64b228583b79 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Fri, 4 Aug 2017 16:00:09 -0700 Subject: bpf: add support for sys_enter_* and sys_exit_* tracepoints Currently, bpf programs cannot be attached to sys_enter_* and sys_exit_* style tracepoints. The iovisor/bcc issue #748 (https://github.com/iovisor/bcc/issues/748) documents this issue. For example, if you try to attach a bpf program to tracepoints syscalls/sys_enter_newfstat, you will get the following error: # ./tools/trace.py t:syscalls:sys_enter_newfstat Ioctl(PERF_EVENT_IOC_SET_BPF): Invalid argument Failed to attach BPF to tracepoint The main reason is that syscalls/sys_enter_* and syscalls/sys_exit_* tracepoints are treated differently from other tracepoints and there is no bpf hook to it. This patch adds bpf support for these syscalls tracepoints by . permitting bpf attachment in ioctl PERF_EVENT_IOC_SET_BPF . calling bpf programs in perf_syscall_enter and perf_syscall_exit The legality of bpf program ctx access is also checked. Function trace_event_get_offsets returns correct max offset for each specific syscall tracepoint, which is compared against the maximum offset access in bpf program. Signed-off-by: Yonghong Song Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/trace/trace_syscalls.c | 53 +++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 51 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 5e10395da88e..7a1a92036563 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -559,11 +559,29 @@ static DECLARE_BITMAP(enabled_perf_exit_syscalls, NR_syscalls); static int sys_perf_refcount_enter; static int sys_perf_refcount_exit; +static int perf_call_bpf_enter(struct bpf_prog *prog, struct pt_regs *regs, + struct syscall_metadata *sys_data, + struct syscall_trace_enter *rec) { + struct syscall_tp_t { + unsigned long long regs; + unsigned long syscall_nr; + unsigned long args[sys_data->nb_args]; + } param; + int i; + + *(struct pt_regs **)¶m = regs; + param.syscall_nr = rec->nr; + for (i = 0; i < sys_data->nb_args; i++) + param.args[i] = rec->args[i]; + return trace_call_bpf(prog, ¶m); +} + static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) { struct syscall_metadata *sys_data; struct syscall_trace_enter *rec; struct hlist_head *head; + struct bpf_prog *prog; int syscall_nr; int rctx; int size; @@ -578,8 +596,9 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) if (!sys_data) return; + prog = READ_ONCE(sys_data->enter_event->prog); head = this_cpu_ptr(sys_data->enter_event->perf_events); - if (hlist_empty(head)) + if (!prog && hlist_empty(head)) return; /* get the size after alignment with the u32 buffer size field */ @@ -594,6 +613,13 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) rec->nr = syscall_nr; syscall_get_arguments(current, regs, 0, sys_data->nb_args, (unsigned long *)&rec->args); + + if ((prog && !perf_call_bpf_enter(prog, regs, sys_data, rec)) || + hlist_empty(head)) { + perf_swevent_put_recursion_context(rctx); + return; + } + perf_trace_buf_submit(rec, size, rctx, sys_data->enter_event->event.type, 1, regs, head, NULL); @@ -633,11 +659,26 @@ static void perf_sysenter_disable(struct trace_event_call *call) mutex_unlock(&syscall_trace_lock); } +static int perf_call_bpf_exit(struct bpf_prog *prog, struct pt_regs *regs, + struct syscall_trace_exit *rec) { + struct syscall_tp_t { + unsigned long long regs; + unsigned long syscall_nr; + unsigned long ret; + } param; + + *(struct pt_regs **)¶m = regs; + param.syscall_nr = rec->nr; + param.ret = rec->ret; + return trace_call_bpf(prog, ¶m); +} + static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) { struct syscall_metadata *sys_data; struct syscall_trace_exit *rec; struct hlist_head *head; + struct bpf_prog *prog; int syscall_nr; int rctx; int size; @@ -652,8 +693,9 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) if (!sys_data) return; + prog = READ_ONCE(sys_data->exit_event->prog); head = this_cpu_ptr(sys_data->exit_event->perf_events); - if (hlist_empty(head)) + if (!prog && hlist_empty(head)) return; /* We can probably do that at build time */ @@ -666,6 +708,13 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->nr = syscall_nr; rec->ret = syscall_get_return_value(current, regs); + + if ((prog && !perf_call_bpf_exit(prog, regs, rec)) || + hlist_empty(head)) { + perf_swevent_put_recursion_context(rctx); + return; + } + perf_trace_buf_submit(rec, size, rctx, sys_data->exit_event->event.type, 1, regs, head, NULL); } -- cgit v1.2.3 From 88a5c690b66110ad255380d8f629c629cf6ca559 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 16 Aug 2017 01:45:33 +0200 Subject: bpf: fix bpf_trace_printk on 32 bit archs James reported that on MIPS32 bpf_trace_printk() is currently broken while MIPS64 works fine: bpf_trace_printk() uses conditional operators to attempt to pass different types to __trace_printk() depending on the format operators. This doesn't work as intended on 32-bit architectures where u32 and long are passed differently to u64, since the result of C conditional operators follows the "usual arithmetic conversions" rules, such that the values passed to __trace_printk() will always be u64 [causing issues later in the va_list handling for vscnprintf()]. For example the samples/bpf/tracex5 test printed lines like below on MIPS32, where the fd and buf have come from the u64 fd argument, and the size from the buf argument: [...] 1180.941542: 0x00000001: write(fd=1, buf= (null), size=6258688) Instead of this: [...] 1625.616026: 0x00000001: write(fd=1, buf=009e4000, size=512) One way to get it working is to expand various combinations of argument types into 8 different combinations for 32 bit and 64 bit kernels. Fix tested by James on MIPS32 and MIPS64 as well that it resolves the issue. Fixes: 9c959c863f82 ("tracing: Allow BPF programs to call bpf_trace_printk()") Reported-by: James Hogan Tested-by: James Hogan Signed-off-by: Daniel Borkmann Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 34 ++++++++++++++++++++++++++++++---- 1 file changed, 30 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 37385193a608..dc498b605d5d 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -204,10 +204,36 @@ BPF_CALL_5(bpf_trace_printk, char *, fmt, u32, fmt_size, u64, arg1, fmt_cnt++; } - return __trace_printk(1/* fake ip will not be printed */, fmt, - mod[0] == 2 ? arg1 : mod[0] == 1 ? (long) arg1 : (u32) arg1, - mod[1] == 2 ? arg2 : mod[1] == 1 ? (long) arg2 : (u32) arg2, - mod[2] == 2 ? arg3 : mod[2] == 1 ? (long) arg3 : (u32) arg3); +/* Horrid workaround for getting va_list handling working with different + * argument type combinations generically for 32 and 64 bit archs. + */ +#define __BPF_TP_EMIT() __BPF_ARG3_TP() +#define __BPF_TP(...) \ + __trace_printk(1 /* Fake ip will not be printed. */, \ + fmt, ##__VA_ARGS__) + +#define __BPF_ARG1_TP(...) \ + ((mod[0] == 2 || (mod[0] == 1 && __BITS_PER_LONG == 64)) \ + ? __BPF_TP(arg1, ##__VA_ARGS__) \ + : ((mod[0] == 1 || (mod[0] == 0 && __BITS_PER_LONG == 32)) \ + ? __BPF_TP((long)arg1, ##__VA_ARGS__) \ + : __BPF_TP((u32)arg1, ##__VA_ARGS__))) + +#define __BPF_ARG2_TP(...) \ + ((mod[1] == 2 || (mod[1] == 1 && __BITS_PER_LONG == 64)) \ + ? __BPF_ARG1_TP(arg2, ##__VA_ARGS__) \ + : ((mod[1] == 1 || (mod[1] == 0 && __BITS_PER_LONG == 32)) \ + ? __BPF_ARG1_TP((long)arg2, ##__VA_ARGS__) \ + : __BPF_ARG1_TP((u32)arg2, ##__VA_ARGS__))) + +#define __BPF_ARG3_TP(...) \ + ((mod[2] == 2 || (mod[2] == 1 && __BITS_PER_LONG == 64)) \ + ? __BPF_ARG2_TP(arg3, ##__VA_ARGS__) \ + : ((mod[2] == 1 || (mod[2] == 0 && __BITS_PER_LONG == 32)) \ + ? __BPF_ARG2_TP((long)arg3, ##__VA_ARGS__) \ + : __BPF_ARG2_TP((u32)arg3, ##__VA_ARGS__))) + + return __BPF_TP_EMIT(); } static const struct bpf_func_proto bpf_trace_printk_proto = { -- cgit v1.2.3 From 74d46992e0d9dee7f1f376de0d56d31614c8a17a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 23 Aug 2017 19:10:32 +0200 Subject: block: replace bi_bdev with a gendisk pointer and partitions index This way we don't need a block_device structure to submit I/O. The block_device has different life time rules from the gendisk and request_queue and is usually only available when the block device node is open. Other callers need to explicitly create one (e.g. the lightnvm passthrough code, or the new nvme multipathing code). For the actual I/O path all that we need is the gendisk, which exists once per block device. But given that the block layer also does partition remapping we additionally need a partition index, which is used for said remapping in generic_make_request. Note that all the block drivers generally want request_queue or sometimes the gendisk, so this removes a layer of indirection all over the stack. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 7724de18d2fe..2a685b45b73b 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -963,7 +963,7 @@ static void blk_add_trace_bio_remap(void *ignore, return; r.device_from = cpu_to_be32(dev); - r.device_to = cpu_to_be32(bio->bi_bdev->bd_dev); + r.device_to = cpu_to_be32(bio_dev(bio)); r.sector_from = cpu_to_be64(from); __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, -- cgit v1.2.3 From a8f0f9e49956a74718874b800251455680085600 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Thu, 17 Aug 2017 16:37:25 -0400 Subject: ftrace: Check for null ret_stack on profile function graph entry function There's a small race when function graph shutsdown and the calling of the registered function graph entry callback. The callback must not reference the task's ret_stack without first checking that it is not NULL. Note, when a ret_stack is allocated for a task, it stays allocated until the task exits. The problem here, is that function_graph is shutdown, and a new task was created, which doesn't have its ret_stack allocated. But since some of the functions are still being traced, the callbacks can still be called. The normal function_graph code handles this, but starting with commit 8861dd303c ("ftrace: Access ret_stack->subtime only in the function profiler") the profiler code references the ret_stack on function entry, but doesn't check if it is NULL first. Link: https://bugzilla.kernel.org/show_bug.cgi?id=196611 Cc: stable@vger.kernel.org Fixes: 8861dd303c ("ftrace: Access ret_stack->subtime only in the function profiler") Reported-by: lilydjwg@gmail.com Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 02004ae91860..96cea88fa00f 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -889,6 +889,10 @@ static int profile_graph_entry(struct ftrace_graph_ent *trace) function_profile_call(trace->func, 0, NULL, NULL); + /* If function graph is shutting down, ret_stack can be NULL */ + if (!current->ret_stack) + return 0; + if (index >= 0 && index < FTRACE_RETFUNC_DEPTH) current->ret_stack[index].subtime = 0; -- cgit v1.2.3 From 475bb3c69ab05df2a6ecef6acc2393703d134180 Mon Sep 17 00:00:00 2001 From: Chunyu Hu Date: Mon, 14 Aug 2017 18:18:17 +0800 Subject: tracing: Fix kmemleak in tracing_map_array_free() kmemleak reported the below leak when I was doing clear of the hist trigger. With this patch, the kmeamleak is gone. unreferenced object 0xffff94322b63d760 (size 32): comm "bash", pid 1522, jiffies 4403687962 (age 2442.311s) hex dump (first 32 bytes): 00 01 00 00 04 00 00 00 08 00 00 00 ff 00 00 00 ................ 10 00 00 00 00 00 00 00 80 a8 7a f2 31 94 ff ff ..........z.1... backtrace: [] kmemleak_alloc+0x4a/0xa0 [] kmem_cache_alloc_trace+0xca/0x1d0 [] tracing_map_array_alloc+0x26/0x140 [] kretprobe_trampoline+0x0/0x50 [] create_hist_data+0x535/0x750 [] event_hist_trigger_func+0x1f7/0x420 [] event_trigger_write+0xfd/0x1a0 [] __vfs_write+0x37/0x170 [] vfs_write+0xb2/0x1b0 [] SyS_write+0x55/0xc0 [] do_syscall_64+0x67/0x150 [] return_from_SYSCALL_64+0x0/0x6a [] 0xffffffffffffffff unreferenced object 0xffff9431f27aa880 (size 128): comm "bash", pid 1522, jiffies 4403687962 (age 2442.311s) hex dump (first 32 bytes): 00 00 8c 2a 32 94 ff ff 00 f0 8b 2a 32 94 ff ff ...*2......*2... 00 e0 8b 2a 32 94 ff ff 00 d0 8b 2a 32 94 ff ff ...*2......*2... backtrace: [] kmemleak_alloc+0x4a/0xa0 [] __kmalloc+0xe8/0x220 [] tracing_map_array_alloc+0xb1/0x140 [] kretprobe_trampoline+0x0/0x50 [] create_hist_data+0x535/0x750 [] event_hist_trigger_func+0x1f7/0x420 [] event_trigger_write+0xfd/0x1a0 [] __vfs_write+0x37/0x170 [] vfs_write+0xb2/0x1b0 [] SyS_write+0x55/0xc0 [] do_syscall_64+0x67/0x150 [] return_from_SYSCALL_64+0x0/0x6a [] 0xffffffffffffffff Link: http://lkml.kernel.org/r/1502705898-27571-1-git-send-email-chuhu@redhat.com Cc: stable@vger.kernel.org Fixes: 08d43a5fa063 ("tracing: Add lock-free tracing_map") Signed-off-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/tracing_map.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c index 0a689bbb78ef..305039b122fa 100644 --- a/kernel/trace/tracing_map.c +++ b/kernel/trace/tracing_map.c @@ -221,16 +221,19 @@ void tracing_map_array_free(struct tracing_map_array *a) if (!a) return; - if (!a->pages) { - kfree(a); - return; - } + if (!a->pages) + goto free; for (i = 0; i < a->n_pages; i++) { if (!a->pages[i]) break; free_page((unsigned long)a->pages[i]); } + + kfree(a->pages); + + free: + kfree(a); } struct tracing_map_array *tracing_map_array_alloc(unsigned int n_elts, -- cgit v1.2.3 From 8b0db1a5bdfcee0dbfa89607672598ae203c9045 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Wed, 23 Aug 2017 12:46:27 -0400 Subject: tracing: Fix freeing of filter in create_filter() when set_str is false Performing the following task with kmemleak enabled: # cd /sys/kernel/tracing/events/irq/irq_handler_entry/ # echo 'enable_event:kmem:kmalloc:3 if irq >' > trigger # echo 'enable_event:kmem:kmalloc:3 if irq > 31' > trigger # echo scan > /sys/kernel/debug/kmemleak # cat /sys/kernel/debug/kmemleak unreferenced object 0xffff8800b9290308 (size 32): comm "bash", pid 1114, jiffies 4294848451 (age 141.139s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmemleak_alloc+0x4a/0xa0 [] kmem_cache_alloc_trace+0x158/0x290 [] create_filter_start.constprop.28+0x99/0x940 [] create_filter+0xa9/0x160 [] create_event_filter+0xc/0x10 [] set_trigger_filter+0xe5/0x210 [] event_enable_trigger_func+0x324/0x490 [] event_trigger_write+0x1a2/0x260 [] __vfs_write+0xd7/0x380 [] vfs_write+0x101/0x260 [] SyS_write+0xab/0x130 [] entry_SYSCALL_64_fastpath+0x1f/0xbe [] 0xffffffffffffffff The function create_filter() is passed a 'filterp' pointer that gets allocated, and if "set_str" is true, it is up to the caller to free it, even on error. The problem is that the pointer is not freed by create_filter() when set_str is false. This is a bug, and it is not up to the caller to free the filter on error if it doesn't care about the string. Link: http://lkml.kernel.org/r/1502705898-27571-2-git-send-email-chuhu@redhat.com Cc: stable@vger.kernel.org Fixes: 38b78eb85 ("tracing: Factorize filter creation") Reported-by: Chunyu Hu Tested-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_filter.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 59a411ff60c7..181e139a8057 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1959,6 +1959,10 @@ static int create_filter(struct trace_event_call *call, if (err && set_str) append_filter_err(ps, filter); } + if (err && !set_str) { + free_event_filter(filter); + filter = NULL; + } create_filter_finish(ps); *filterp = filter; -- cgit v1.2.3 From 75e8387685f6c65feb195a4556110b58f852b848 Mon Sep 17 00:00:00 2001 From: Zhou Chengming Date: Fri, 25 Aug 2017 21:49:37 +0800 Subject: perf/ftrace: Fix double traces of perf on ftrace:function When running perf on the ftrace:function tracepoint, there is a bug which can be reproduced by: perf record -e ftrace:function -a sleep 20 & perf record -e ftrace:function ls perf script ls 10304 [005] 171.853235: ftrace:function: perf_output_begin ls 10304 [005] 171.853237: ftrace:function: perf_output_begin ls 10304 [005] 171.853239: ftrace:function: task_tgid_nr_ns ls 10304 [005] 171.853240: ftrace:function: task_tgid_nr_ns ls 10304 [005] 171.853242: ftrace:function: __task_pid_nr_ns ls 10304 [005] 171.853244: ftrace:function: __task_pid_nr_ns We can see that all the function traces are doubled. The problem is caused by the inconsistency of the register function perf_ftrace_event_register() with the probe function perf_ftrace_function_call(). The former registers one probe for every perf_event. And the latter handles all perf_events on the current cpu. So when two perf_events on the current cpu, the traces of them will be doubled. So this patch adds an extra parameter "event" for perf_tp_event, only send sample data to this event when it's not NULL. Signed-off-by: Zhou Chengming Reviewed-by: Jiri Olsa Acked-by: Steven Rostedt (VMware) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: acme@kernel.org Cc: alexander.shishkin@linux.intel.com Cc: huawei.libin@huawei.com Link: http://lkml.kernel.org/r/1503668977-12526-1-git-send-email-zhouchengming1@huawei.com Signed-off-by: Ingo Molnar --- kernel/trace/trace_event_perf.c | 4 +++- kernel/trace/trace_kprobe.c | 4 ++-- kernel/trace/trace_syscalls.c | 4 ++-- kernel/trace/trace_uprobe.c | 2 +- 4 files changed, 8 insertions(+), 6 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 562fa69df5d3..13ba2d3f6a91 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -306,6 +306,7 @@ static void perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *ops, struct pt_regs *pt_regs) { + struct perf_event *event; struct ftrace_entry *entry; struct hlist_head *head; struct pt_regs regs; @@ -329,8 +330,9 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, entry->ip = ip; entry->parent_ip = parent_ip; + event = container_of(ops, struct perf_event, ftrace_ops); perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, TRACE_FN, - 1, ®s, head, NULL); + 1, ®s, head, NULL, event); #undef ENTRY_SIZE } diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index c9b5aa10fbf9..8a907e12b6b9 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1200,7 +1200,7 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) memset(&entry[1], 0, dsize); store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, - head, NULL); + head, NULL, NULL); } NOKPROBE_SYMBOL(kprobe_perf_func); @@ -1236,7 +1236,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, entry->ret_ip = (unsigned long)ri->ret_addr; store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, - head, NULL); + head, NULL, NULL); } NOKPROBE_SYMBOL(kretprobe_perf_func); #endif /* CONFIG_PERF_EVENTS */ diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 5e10395da88e..74d9a86eccc0 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -596,7 +596,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) (unsigned long *)&rec->args); perf_trace_buf_submit(rec, size, rctx, sys_data->enter_event->event.type, 1, regs, - head, NULL); + head, NULL, NULL); } static int perf_sysenter_enable(struct trace_event_call *call) @@ -667,7 +667,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->nr = syscall_nr; rec->ret = syscall_get_return_value(current, regs); perf_trace_buf_submit(rec, size, rctx, sys_data->exit_event->event.type, - 1, regs, head, NULL); + 1, regs, head, NULL, NULL); } static int perf_sysexit_enable(struct trace_event_call *call) diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index a7581fec9681..4525e0271a53 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -1156,7 +1156,7 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, } perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, - head, NULL); + head, NULL, NULL); out: preempt_enable(); } -- cgit v1.2.3 From 065e63f951432068ba89a844fcbff68ea16ee186 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Thu, 31 Aug 2017 17:03:47 -0400 Subject: tracing: Only have rmmod clear buffers that its events were active in Currently, when a module event is enabled, when that module is removed, it clears all ring buffers. This is to prevent another module from being loaded and having one of its trace event IDs from reusing a trace event ID of the removed module. This could cause undesirable effects as the trace event of the new module would be using its own processing algorithms to process raw data of another event. To prevent this, when a module is loaded, if any of its events have been used (signified by the WAS_ENABLED event call flag, which is never cleared), all ring buffers are cleared, just in case any one of them contains event data of the removed event. The problem is, there's no reason to clear all ring buffers if only one (or less than all of them) uses one of the events. Instead, only clear the ring buffers that recorded the events of a module that is being removed. To do this, instead of keeping the WAS_ENABLED flag with the trace event call, move it to the per instance (per ring buffer) event file descriptor. The event file descriptor maps each event to a separate ring buffer instance. Then when the module is removed, only the ring buffers that activated one of the module's events get cleared. The rest are not touched. Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 3 +++ kernel/trace/trace.h | 1 + kernel/trace/trace_events.c | 15 +++++++-------- 3 files changed, 11 insertions(+), 8 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 44004d8aa3b3..30338a835a51 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1702,6 +1702,9 @@ void tracing_reset_all_online_cpus(void) struct trace_array *tr; list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (!tr->clear_trace) + continue; + tr->clear_trace = false; tracing_reset_online_cpus(&tr->trace_buffer); #ifdef CONFIG_TRACER_MAX_TRACE tracing_reset_online_cpus(&tr->max_buffer); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 490ba229931d..fb5d54d0d1b3 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -245,6 +245,7 @@ struct trace_array { int stop_count; int clock_id; int nr_topts; + bool clear_trace; struct tracer *current_trace; unsigned int trace_flags; unsigned char trace_flags_index[TRACE_FLAGS_MAX_SIZE]; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 36132f9280e6..c93540c5df21 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -466,7 +466,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, set_bit(EVENT_FILE_FL_ENABLED_BIT, &file->flags); /* WAS_ENABLED gets set but never cleared. */ - call->flags |= TRACE_EVENT_FL_WAS_ENABLED; + set_bit(EVENT_FILE_FL_WAS_ENABLED_BIT, &file->flags); } break; } @@ -2058,6 +2058,10 @@ static void event_remove(struct trace_event_call *call) do_for_each_event_file(tr, file) { if (file->event_call != call) continue; + + if (file->flags & EVENT_FILE_FL_WAS_ENABLED) + tr->clear_trace = true; + ftrace_event_enable_disable(file, 0); /* * The do_for_each_event_file() is @@ -2396,15 +2400,11 @@ static void trace_module_add_events(struct module *mod) static void trace_module_remove_events(struct module *mod) { struct trace_event_call *call, *p; - bool clear_trace = false; down_write(&trace_event_sem); list_for_each_entry_safe(call, p, &ftrace_events, list) { - if (call->mod == mod) { - if (call->flags & TRACE_EVENT_FL_WAS_ENABLED) - clear_trace = true; + if (call->mod == mod) __trace_remove_event_call(call); - } } up_write(&trace_event_sem); @@ -2416,8 +2416,7 @@ static void trace_module_remove_events(struct module *mod) * over from this module may be passed to the new module events and * unexpected results may occur. */ - if (clear_trace) - tracing_reset_all_online_cpus(); + tracing_reset_all_online_cpus(); } static int trace_module_notify(struct notifier_block *self, -- cgit v1.2.3 From 2a5bfe47624bfc835aa0632a0505ba55576c98db Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Thu, 31 Aug 2017 17:36:51 -0400 Subject: ftrace: Zero out ftrace hashes when a module is removed When a ftrace filter has a module function, and that module is removed, the filter still has its address as being enabled. This can cause interesting side effects. Nothing dangerous, but unwanted functions can be traced because of it. # cd /sys/kernel/tracing # echo ':mod:snd_seq' > set_ftrace_filter # cat set_ftrace_filter snd_use_lock_sync_helper [snd_seq] check_event_type_and_length [snd_seq] snd_seq_ioctl_pversion [snd_seq] snd_seq_ioctl_client_id [snd_seq] snd_seq_ioctl_get_queue_tempo [snd_seq] update_timestamp_of_queue [snd_seq] snd_seq_ioctl_get_queue_status [snd_seq] snd_seq_set_queue_tempo [snd_seq] snd_seq_ioctl_set_queue_tempo [snd_seq] snd_seq_ioctl_get_queue_timer [snd_seq] seq_free_client1 [snd_seq] [..] # rmmod snd_seq # cat set_ftrace_filter # modprobe kvm # cat set_ftrace_filter kvm_set_cr4 [kvm] kvm_emulate_hypercall [kvm] kvm_set_dr [kvm] This is because removing the snd_seq module after it was being filtered, left the address of the snd_seq functions in the hash. When the kvm module was loaded, some of its functions were loaded at the same address as the snd_seq module. This would enable them to be filtered and traced. Now we don't want to clear the hash completely. That would cause removing a module where only its functions are filtered, to cause the tracing to enable all functions, as an empty filter means to trace all functions. Instead, just set the hash ip address to zero. Then it will never match any function. Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 55 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 96cea88fa00f..165b149ccb1a 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5690,10 +5690,51 @@ static int referenced_filters(struct dyn_ftrace *rec) return cnt; } +static void +clear_mod_from_hash(struct ftrace_page *pg, struct ftrace_hash *hash) +{ + struct ftrace_func_entry *entry; + struct dyn_ftrace *rec; + int i; + + if (ftrace_hash_empty(hash)) + return; + + for (i = 0; i < pg->index; i++) { + rec = &pg->records[i]; + entry = __ftrace_lookup_ip(hash, rec->ip); + /* + * Do not allow this rec to match again. + * Yeah, it may waste some memory, but will be removed + * if/when the hash is modified again. + */ + if (entry) + entry->ip = 0; + } +} + +/* Clear any records from hashs */ +static void clear_mod_from_hashes(struct ftrace_page *pg) +{ + struct trace_array *tr; + + mutex_lock(&trace_types_lock); + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (!tr->ops || !tr->ops->func_hash) + continue; + mutex_lock(&tr->ops->func_hash->regex_lock); + clear_mod_from_hash(pg, tr->ops->func_hash->filter_hash); + clear_mod_from_hash(pg, tr->ops->func_hash->notrace_hash); + mutex_unlock(&tr->ops->func_hash->regex_lock); + } + mutex_unlock(&trace_types_lock); +} + void ftrace_release_mod(struct module *mod) { struct dyn_ftrace *rec; struct ftrace_page **last_pg; + struct ftrace_page *tmp_page = NULL; struct ftrace_page *pg; int order; @@ -5723,14 +5764,25 @@ void ftrace_release_mod(struct module *mod) ftrace_update_tot_cnt -= pg->index; *last_pg = pg->next; - order = get_count_order(pg->size / ENTRIES_PER_PAGE); - free_pages((unsigned long)pg->records, order); - kfree(pg); + + pg->next = tmp_page; + tmp_page = pg; } else last_pg = &pg->next; } out_unlock: mutex_unlock(&ftrace_lock); + + for (pg = tmp_page; pg; pg = tmp_page) { + + /* Needs to be called outside of ftrace_lock */ + clear_mod_from_hashes(pg); + + order = get_count_order(pg->size / ENTRIES_PER_PAGE); + free_pages((unsigned long)pg->records, order); + tmp_page = pg->next; + kfree(pg); + } } void ftrace_module_enable(struct module *mod) -- cgit v1.2.3 From 46320a6acc4fb58f04bcf78c4c942cc43b20f986 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Fri, 1 Sep 2017 12:04:09 -0400 Subject: ftrace: Fix selftest goto location on error In the second iteration of trace_selftest_ops(), the error goto label is wrong in the case where trace_selftest_test_global_cnt is off. In the case of error, it leaks the dynamic ops that was allocated. Cc: stable@vger.kernel.org Fixes: 95950c2e ("ftrace: Add self-tests for multiple function trace users") Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_selftest.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index cb917cebae29..b17ec642793b 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -273,7 +273,7 @@ static int trace_selftest_ops(struct trace_array *tr, int cnt) goto out_free; if (cnt > 1) { if (trace_selftest_test_global_cnt == 0) - goto out; + goto out_free; } if (trace_selftest_test_dyn_cnt == 0) goto out_free; -- cgit v1.2.3 From edb096e00724f02db5f6ec7900f3bbd465c6c76f Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Fri, 1 Sep 2017 12:18:28 -0400 Subject: ftrace: Fix memleak when unregistering dynamic ops when tracing disabled If function tracing is disabled by the user via the function-trace option or the proc sysctl file, and a ftrace_ops that was allocated on the heap is unregistered, then the shutdown code exits out without doing the proper clean up. This was found via kmemleak and running the ftrace selftests, as one of the tests unregisters with function tracing disabled. # cat kmemleak unreferenced object 0xffffffffa0020000 (size 4096): comm "swapper/0", pid 1, jiffies 4294668889 (age 569.209s) hex dump (first 32 bytes): 55 ff 74 24 10 55 48 89 e5 ff 74 24 18 55 48 89 U.t$.UH...t$.UH. e5 48 81 ec a8 00 00 00 48 89 44 24 50 48 89 4c .H......H.D$PH.L backtrace: [] kmemleak_vmalloc+0x85/0xf0 [] __vmalloc_node_range+0x281/0x3e0 [] module_alloc+0x4f/0x90 [] arch_ftrace_update_trampoline+0x160/0x420 [] ftrace_startup+0xe7/0x300 [] register_ftrace_function+0x72/0x90 [] trace_selftest_ops+0x204/0x397 [] trace_selftest_startup_function+0x394/0x624 [] run_tracer_selftest+0x15c/0x1d7 [] init_trace_selftests+0x75/0x192 [] do_one_initcall+0x90/0x1e2 [] kernel_init_freeable+0x350/0x3fe [] kernel_init+0x13/0x122 [] ret_from_fork+0x2a/0x40 [] 0xffffffffffffffff Cc: stable@vger.kernel.org Fixes: 12cce594fa ("ftrace/x86: Allow !CONFIG_PREEMPT dynamic ops to use allocated trampolines") Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 165b149ccb1a..6abfafd7f173 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -2828,13 +2828,14 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) if (!command || !ftrace_enabled) { /* - * If these are per_cpu ops, they still need their - * per_cpu field freed. Since, function tracing is + * If these are dynamic or per_cpu ops, they still + * need their data freed. Since, function tracing is * not currently active, we can just free them * without synchronizing all CPUs. */ - if (ops->flags & FTRACE_OPS_FL_PER_CPU) - per_cpu_ops_free(ops); + if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_PER_CPU)) + goto free_ops; + return 0; } @@ -2900,6 +2901,7 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) if (IS_ENABLED(CONFIG_PREEMPT)) synchronize_rcu_tasks(); + free_ops: arch_ftrace_trampoline_free(ops); if (ops->flags & FTRACE_OPS_FL_PER_CPU) -- cgit v1.2.3 From 3d9622c12c8873911f4cc0ccdabd0362c2fca06b Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 5 Sep 2017 11:32:01 -0400 Subject: tracing: Add barrier to trace_printk() buffer nesting modification trace_printk() uses 4 buffers, one for each context (normal, softirq, irq and NMI), such that it does not need to worry about one context preempting the other. There's a nesting counter that gets incremented to figure out which buffer to use. If the context gets preempted by another context which calls trace_printk() it will increment the counter and use the next buffer, and restore the counter when it is finished. The problem is that gcc may optimize the modification of the buffer nesting counter and it may not be incremented in memory before the buffer is used. If this happens, and the context gets interrupted by another context, it could pick the same buffer and corrupt the one that is being used. Compiler barriers need to be added after the nesting variable is incremented and before it is decremented to prevent usage of the context buffers by more than one context at the same time. Cc: Andy Lutomirski Cc: stable@vger.kernel.org Fixes: e2ace00117 ("tracing: Choose static tp_printk buffer by explicit nesting count") Hat-tip-to: Peter Zijlstra Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 30338a835a51..78842557eea0 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2802,11 +2802,17 @@ static char *get_trace_buf(void) if (!buffer || buffer->nesting >= 4) return NULL; - return &buffer->buffer[buffer->nesting++][0]; + buffer->nesting++; + + /* Interrupts must see nesting incremented before we use the buffer */ + barrier(); + return &buffer->buffer[buffer->nesting][0]; } static void put_trace_buf(void) { + /* Don't let the decrement of nesting leak before this */ + barrier(); this_cpu_dec(trace_percpu_buffer->nesting); } -- cgit v1.2.3 From 7685ab6c58557c6234f3540260195ecbee7fc4b3 Mon Sep 17 00:00:00 2001 From: Chunyu Hu Date: Tue, 5 Sep 2017 13:36:46 +0800 Subject: tracing: Fix clear of RECORDED_TGID flag when disabling trace event When disabling one trace event, the RECORDED_TGID flag in the event file is not correctly cleared. It's clearing RECORDED_CMD flag when it should clear RECORDED_TGID flag. Link: http://lkml.kernel.org/r/1504589806-8425-1-git-send-email-chuhu@redhat.com Cc: Joel Fernandes Cc: stable@vger.kernel.org Fixes: d914ba37d7 ("tracing: Add support for recording tgid of tasks") Signed-off-by: Chunyu Hu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index c93540c5df21..87468398b9ed 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -406,7 +406,7 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, if (file->flags & EVENT_FILE_FL_RECORDED_TGID) { tracing_stop_tgid_record(); - clear_bit(EVENT_FILE_FL_RECORDED_CMD_BIT, &file->flags); + clear_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, &file->flags); } call->class->reg(call, TRACE_REG_UNREGISTER, file); -- cgit v1.2.3 From 170b3b1050e28d1ba0700e262f0899ffa4fccc52 Mon Sep 17 00:00:00 2001 From: Baohong Liu Date: Tue, 5 Sep 2017 16:57:19 -0500 Subject: tracing: Apply trace_clock changes to instance max buffer Currently trace_clock timestamps are applied to both regular and max buffers only for global trace. For instance trace, trace_clock timestamps are applied only to regular buffer. But, regular and max buffers can be swapped, for example, following a snapshot. So, for instance trace, bad timestamps can be seen following a snapshot. Let's apply trace_clock timestamps to instance max buffer as well. Link: http://lkml.kernel.org/r/ebdb168d0be042dcdf51f81e696b17fabe3609c1.1504642143.git.tom.zanussi@linux.intel.com Cc: stable@vger.kernel.org Fixes: 277ba0446 ("tracing: Add interface to allow multiple trace buffers") Signed-off-by: Baohong Liu Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 78842557eea0..5360b7aec57a 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6229,7 +6229,7 @@ static int tracing_set_clock(struct trace_array *tr, const char *clockstr) tracing_reset_online_cpus(&tr->trace_buffer); #ifdef CONFIG_TRACER_MAX_TRACE - if (tr->flags & TRACE_ARRAY_FL_GLOBAL && tr->max_buffer.buffer) + if (tr->max_buffer.buffer) ring_buffer_set_clock(tr->max_buffer.buffer, trace_clocks[i].func); tracing_reset_online_cpus(&tr->max_buffer); #endif -- cgit v1.2.3 From 9b130ad5bb8255ee8534d92d67e12b2a4887eacb Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 8 Sep 2017 16:14:18 -0700 Subject: treewide: make "nr_cpu_ids" unsigned First, number of CPUs can't be negative number. Second, different signnnedness leads to suboptimal code in the following cases: 1) kmalloc(nr_cpu_ids * sizeof(X)); "int" has to be sign extended to size_t. 2) while (loff_t *pos < nr_cpu_ids) MOVSXD is 1 byte longed than the same MOV. Other cases exist as well. Basically compiler is told that nr_cpu_ids can't be negative which can't be deduced if it is "int". Code savings on allyesconfig kernel: -3KB add/remove: 0/0 grow/shrink: 25/264 up/down: 261/-3631 (-3370) function old new delta coretemp_cpu_online 450 512 +62 rcu_init_one 1234 1272 +38 pci_device_probe 374 399 +25 ... pgdat_reclaimable_pages 628 556 -72 select_fallback_rq 446 369 -77 task_numa_find_cpu 1923 1807 -116 Link: http://lkml.kernel.org/r/20170819114959.GA30580@avx2 Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/trace/trace_functions_graph.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index d56123cdcc89..b8f1f54731af 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -1543,7 +1543,7 @@ fs_initcall(init_graph_tracefs); static __init int init_graph_trace(void) { - max_bytes_for_cpu = snprintf(NULL, 0, "%d", nr_cpu_ids - 1); + max_bytes_for_cpu = snprintf(NULL, 0, "%u", nr_cpu_ids - 1); if (!register_trace_event(&graph_trace_entry_event)) { pr_warn("Warning: could not register graph trace events\n"); -- cgit v1.2.3 From 609320c8a22715b74b39796930c3542719f8ab62 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 7 Sep 2017 18:36:15 -0700 Subject: perf/bpf: fix a clang compilation issue clang does not support variable length array for structure member. It has the following error during compilation: kernel/trace/trace_syscalls.c:568:17: error: fields must have a constant size: 'variable length array in structure' extension will never be supported unsigned long args[sys_data->nb_args]; ^ The fix is to use a fixed array length instead. Reported-by: Nick Desaulniers Signed-off-by: Yonghong Song Signed-off-by: David S. Miller --- kernel/trace/trace_syscalls.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 9c4eef20301c..696afe72d3b1 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -565,7 +565,7 @@ static int perf_call_bpf_enter(struct bpf_prog *prog, struct pt_regs *regs, struct syscall_tp_t { unsigned long long regs; unsigned long syscall_nr; - unsigned long args[sys_data->nb_args]; + unsigned long args[SYSCALL_DEFINE_MAXARGS]; } param; int i; -- cgit v1.2.3 From 0ee931c4e31a5efb134c76440405e9219f896e33 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Wed, 13 Sep 2017 16:28:29 -0700 Subject: mm: treewide: remove GFP_TEMPORARY allocation flag GFP_TEMPORARY was introduced by commit e12ba74d8ff3 ("Group short-lived and reclaimable kernel allocations") along with __GFP_RECLAIMABLE. It's primary motivation was to allow users to tell that an allocation is short lived and so the allocator can try to place such allocations close together and prevent long term fragmentation. As much as this sounds like a reasonable semantic it becomes much less clear when to use the highlevel GFP_TEMPORARY allocation flag. How long is temporary? Can the context holding that memory sleep? Can it take locks? It seems there is no good answer for those questions. The current implementation of GFP_TEMPORARY is basically GFP_KERNEL | __GFP_RECLAIMABLE which in itself is tricky because basically none of the existing caller provide a way to reclaim the allocated memory. So this is rather misleading and hard to evaluate for any benefits. I have checked some random users and none of them has added the flag with a specific justification. I suspect most of them just copied from other existing users and others just thought it might be a good idea to use without any measuring. This suggests that GFP_TEMPORARY just motivates for cargo cult usage without any reasoning. I believe that our gfp flags are quite complex already and especially those with highlevel semantic should be clearly defined to prevent from confusion and abuse. Therefore I propose dropping GFP_TEMPORARY and replace all existing users to simply use GFP_KERNEL. Please note that SLAB users with shrinkers will still get __GFP_RECLAIMABLE heuristic and so they will be placed properly for memory fragmentation prevention. I can see reasons we might want some gfp flag to reflect shorterm allocations but I propose starting from a clear semantic definition and only then add users with proper justification. This was been brought up before LSF this year by Matthew [1] and it turned out that GFP_TEMPORARY really doesn't have a clear semantic. It seems to be a heuristic without any measured advantage for most (if not all) its current users. The follow up discussion has revealed that opinions on what might be temporary allocation differ a lot between developers. So rather than trying to tweak existing users into a semantic which they haven't expected I propose to simply remove the flag and start from scratch if we really need a semantic for short term allocations. [1] http://lkml.kernel.org/r/20170118054945.GD18349@bombadil.infradead.org [akpm@linux-foundation.org: fix typo] [akpm@linux-foundation.org: coding-style fixes] [sfr@canb.auug.org.au: drm/i915: fix up] Link: http://lkml.kernel.org/r/20170816144703.378d4f4d@canb.auug.org.au Link: http://lkml.kernel.org/r/20170728091904.14627-1-mhocko@kernel.org Signed-off-by: Michal Hocko Signed-off-by: Stephen Rothwell Acked-by: Mel Gorman Acked-by: Vlastimil Babka Cc: Matthew Wilcox Cc: Neil Brown Cc: "Theodore Ts'o" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/trace/trace_events_filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 181e139a8057..61e7f0678d33 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -702,7 +702,7 @@ static void append_filter_err(struct filter_parse_state *ps, int pos = ps->lasterr_pos; char *buf, *pbuf; - buf = (char *)__get_free_page(GFP_TEMPORARY); + buf = (char *)__get_free_page(GFP_KERNEL); if (!buf) return; -- cgit v1.2.3 From 8dd33bcb7050dd6f8c1432732f930932c9d3a33e Mon Sep 17 00:00:00 2001 From: Bo Yan Date: Mon, 18 Sep 2017 10:03:35 -0700 Subject: tracing: Erase irqsoff trace with empty write One convenient way to erase trace is "echo > trace". However, this is currently broken if the current tracer is irqsoff tracer. This is because irqsoff tracer use max_buffer as the default trace buffer. Set the max_buffer as the one to be cleared when it's the trace buffer currently in use. Link: http://lkml.kernel.org/r/1505754215-29411-1-git-send-email-byan@nvidia.com Cc: Cc: stable@vger.kernel.org Fixes: 4acd4d00f ("tracing: give easy way to clear trace buffer") Signed-off-by: Bo Yan Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 5360b7aec57a..a7fb136da891 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4020,11 +4020,17 @@ static int tracing_open(struct inode *inode, struct file *file) /* If this file was open for write, then erase contents */ if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { int cpu = tracing_get_cpu(inode); + struct trace_buffer *trace_buf = &tr->trace_buffer; + +#ifdef CONFIG_TRACER_MAX_TRACE + if (tr->current_trace->print_max) + trace_buf = &tr->max_buffer; +#endif if (cpu == RING_BUFFER_ALL_CPUS) - tracing_reset_online_cpus(&tr->trace_buffer); + tracing_reset_online_cpus(trace_buf); else - tracing_reset(&tr->trace_buffer, cpu); + tracing_reset(trace_buf, cpu); } if (file->f_mode & FMODE_READ) { -- cgit v1.2.3 From c7b3ae0bd2ca658c7a71c49901d08c590294fac9 Mon Sep 17 00:00:00 2001 From: "Ziqian SUN (Zamir)" Date: Mon, 11 Sep 2017 14:26:35 +0800 Subject: tracing: Ignore mmiotrace from kernel commandline The mmiotrace tracer cannot be enabled with ftrace=mmiotrace in kernel commandline. With this patch, noboot is added to the tracer struct, and when system boot with a tracer that has noboot=true, it will print out a warning message and continue booting. Link: http://lkml.kernel.org/r/1505111195-31942-1-git-send-email-zsun@redhat.com Signed-off-by: Ziqian SUN (Zamir) Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 7 +++++++ kernel/trace/trace.h | 2 ++ kernel/trace/trace_mmiotrace.c | 1 + 3 files changed, 10 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index a7fb136da891..d3ca35f38803 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -5364,6 +5364,13 @@ static int tracing_set_tracer(struct trace_array *tr, const char *buf) if (t == tr->current_trace) goto out; + /* Some tracers won't work on kernel command line */ + if (system_state < SYSTEM_RUNNING && t->noboot) { + pr_warn("Tracer '%s' is not allowed on command line, ignored\n", + t->name); + goto out; + } + /* Some tracers are only allowed for the top level buffer */ if (!trace_ok_for_array(t, tr)) { ret = -EINVAL; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index fb5d54d0d1b3..652c682707cd 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -444,6 +444,8 @@ struct tracer { #ifdef CONFIG_TRACER_MAX_TRACE bool use_max_tr; #endif + /* True if tracer cannot be enabled in kernel param */ + bool noboot; }; diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index cd7480d0a201..dca78fc48439 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -282,6 +282,7 @@ static struct tracer mmio_tracer __read_mostly = .close = mmio_close, .read = mmio_read, .print_line = mmio_print_line, + .noboot = true, }; __init static int init_mmio_trace(void) -- cgit v1.2.3 From 75df6e688ccd517e339a7c422ef7ad73045b18a2 Mon Sep 17 00:00:00 2001 From: Tahsin Erdogan Date: Sun, 17 Sep 2017 03:23:48 -0700 Subject: tracing: Fix trace_pipe behavior for instance traces When reading data from trace_pipe, tracing_wait_pipe() performs a check to see if tracing has been turned off after some data was read. Currently, this check always looks at global trace state, but it should be checking the trace instance where trace_pipe is located at. Because of this bug, cat instances/i1/trace_pipe in the following script will immediately exit instead of waiting for data: cd /sys/kernel/debug/tracing echo 0 > tracing_on mkdir -p instances/i1 echo 1 > instances/i1/tracing_on echo 1 > instances/i1/events/sched/sched_process_exec/enable cat instances/i1/trace_pipe Link: http://lkml.kernel.org/r/20170917102348.1615-1-tahsin@google.com Cc: stable@vger.kernel.org Fixes: 10246fa35d4f ("tracing: give easy way to clear trace buffer") Signed-off-by: Tahsin Erdogan Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d3ca35f38803..752e5daf0896 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -5680,7 +5680,7 @@ static int tracing_wait_pipe(struct file *filp) * * iter->pos will be 0 if we haven't read anything. */ - if (!tracing_is_on() && iter->pos) + if (!tracer_tracing_is_on(iter->tr) && iter->pos) break; mutex_unlock(&iter->mutex); -- cgit v1.2.3 From 15516c89acce948debc4c598e03c3fee53045797 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Thu, 21 Sep 2017 13:00:21 -0400 Subject: tracing: Remove RCU work arounds from stack tracer Currently the stack tracer calls rcu_irq_enter() to make sure RCU is watching when it records a stack trace. But if the stack tracer is triggered while tracing inside of a rcu_irq_enter(), calling rcu_irq_enter() unconditionally can be problematic. The reason for having rcu_irq_enter() in the first place has been fixed from within the saving of the stack trace code, and there's no reason for doing it in the stack tracer itself. Just remove it. Cc: stable@vger.kernel.org Fixes: 0be964be0 ("module: Sanitize RCU usage and locking") Acked-by: Paul E. McKenney Suggested-by: "Paul E. McKenney" Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_stack.c | 15 --------------- 1 file changed, 15 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index a4df67cbc711..49cb41412eec 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -96,23 +96,9 @@ check_stack(unsigned long ip, unsigned long *stack) if (in_nmi()) return; - /* - * There's a slight chance that we are tracing inside the - * RCU infrastructure, and rcu_irq_enter() will not work - * as expected. - */ - if (unlikely(rcu_irq_enter_disabled())) - return; - local_irq_save(flags); arch_spin_lock(&stack_trace_max_lock); - /* - * RCU may not be watching, make it see us. - * The stack trace code uses rcu_sched. - */ - rcu_irq_enter(); - /* In case another CPU set the tracer_frame on us */ if (unlikely(!frame_size)) this_size -= tracer_frame; @@ -205,7 +191,6 @@ check_stack(unsigned long ip, unsigned long *stack) } out: - rcu_irq_exit(); arch_spin_unlock(&stack_trace_max_lock); local_irq_restore(flags); } -- cgit v1.2.3 From 5acb3cc2c2e9d3020a4fee43763c6463767f1572 Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Wed, 20 Sep 2017 13:12:20 -0600 Subject: blktrace: Fix potential deadlock between delete & sysfs ops The lockdep code had reported the following unsafe locking scenario: CPU0 CPU1 ---- ---- lock(s_active#228); lock(&bdev->bd_mutex/1); lock(s_active#228); lock(&bdev->bd_mutex); *** DEADLOCK *** The deadlock may happen when one task (CPU1) is trying to delete a partition in a block device and another task (CPU0) is accessing tracing sysfs file (e.g. /sys/block/dm-1/trace/act_mask) in that partition. The s_active isn't an actual lock. It is a reference count (kn->count) on the sysfs (kernfs) file. Removal of a sysfs file, however, require a wait until all the references are gone. The reference count is treated like a rwsem using lockdep instrumentation code. The fact that a thread is in the sysfs callback method or in the ioctl call means there is a reference to the opended sysfs or device file. That should prevent the underlying block structure from being removed. Instead of using bd_mutex in the block_device structure, a new blk_trace_mutex is now added to the request_queue structure to protect access to the blk_trace structure. Suggested-by: Christoph Hellwig Signed-off-by: Waiman Long Acked-by: Steven Rostedt (VMware) Fix typo in patch subject line, and prune a comment detailing how the code used to work. Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 2a685b45b73b..45a3928544ce 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -648,6 +648,12 @@ int blk_trace_startstop(struct request_queue *q, int start) } EXPORT_SYMBOL_GPL(blk_trace_startstop); +/* + * When reading or writing the blktrace sysfs files, the references to the + * opened sysfs or device files should prevent the underlying block device + * from being removed. So no further delete protection is really needed. + */ + /** * blk_trace_ioctl: - handle the ioctls associated with tracing * @bdev: the block device @@ -665,7 +671,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) if (!q) return -ENXIO; - mutex_lock(&bdev->bd_mutex); + mutex_lock(&q->blk_trace_mutex); switch (cmd) { case BLKTRACESETUP: @@ -691,7 +697,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) break; } - mutex_unlock(&bdev->bd_mutex); + mutex_unlock(&q->blk_trace_mutex); return ret; } @@ -1727,7 +1733,7 @@ static ssize_t sysfs_blk_trace_attr_show(struct device *dev, if (q == NULL) goto out_bdput; - mutex_lock(&bdev->bd_mutex); + mutex_lock(&q->blk_trace_mutex); if (attr == &dev_attr_enable) { ret = sprintf(buf, "%u\n", !!q->blk_trace); @@ -1746,7 +1752,7 @@ static ssize_t sysfs_blk_trace_attr_show(struct device *dev, ret = sprintf(buf, "%llu\n", q->blk_trace->end_lba); out_unlock_bdev: - mutex_unlock(&bdev->bd_mutex); + mutex_unlock(&q->blk_trace_mutex); out_bdput: bdput(bdev); out: @@ -1788,7 +1794,7 @@ static ssize_t sysfs_blk_trace_attr_store(struct device *dev, if (q == NULL) goto out_bdput; - mutex_lock(&bdev->bd_mutex); + mutex_lock(&q->blk_trace_mutex); if (attr == &dev_attr_enable) { if (value) @@ -1814,7 +1820,7 @@ static ssize_t sysfs_blk_trace_attr_store(struct device *dev, } out_unlock_bdev: - mutex_unlock(&bdev->bd_mutex); + mutex_unlock(&q->blk_trace_mutex); out_bdput: bdput(bdev); out: -- cgit v1.2.3 From 5f6ad26ea353fdf3dad2328052cbee49e0b9c5b4 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 22 Sep 2017 18:23:31 +0200 Subject: sched/tracing: Use common task-state helpers Remove yet another task-state char instance. Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- kernel/trace/trace_output.c | 21 ++++++--------------- kernel/trace/trace_sched_wakeup.c | 8 ++++---- 2 files changed, 10 insertions(+), 19 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index bac629af2285..c738e764e2a5 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -656,15 +656,6 @@ int trace_print_lat_context(struct trace_iterator *iter) return !trace_seq_has_overflowed(s); } -static const char state_to_char[] = TASK_STATE_TO_CHAR_STR; - -static int task_state_char(unsigned long state) -{ - int bit = state ? __ffs(state) + 1 : 0; - - return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?'; -} - /** * ftrace_find_event - find a registered event * @type: the type of event to look for @@ -930,8 +921,8 @@ static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter, trace_assign_type(field, iter->ent); - T = task_state_char(field->next_state); - S = task_state_char(field->prev_state); + T = __task_state_to_char(field->next_state); + S = __task_state_to_char(field->prev_state); trace_find_cmdline(field->next_pid, comm); trace_seq_printf(&iter->seq, " %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n", @@ -966,8 +957,8 @@ static int trace_ctxwake_raw(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - S = task_state_char(field->prev_state); - T = task_state_char(field->next_state); + S = __task_state_to_char(field->prev_state); + T = __task_state_to_char(field->next_state); trace_seq_printf(&iter->seq, "%d %d %c %d %d %d %c\n", field->prev_pid, field->prev_prio, @@ -1002,8 +993,8 @@ static int trace_ctxwake_hex(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - S = task_state_char(field->prev_state); - T = task_state_char(field->next_state); + S = __task_state_to_char(field->prev_state); + T = __task_state_to_char(field->next_state); SEQ_PUT_HEX_FIELD(s, field->prev_pid); SEQ_PUT_HEX_FIELD(s, field->prev_prio); diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index ddec53b67646..0c331978b1a6 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -397,10 +397,10 @@ tracing_sched_switch_trace(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->prev_pid = prev->pid; entry->prev_prio = prev->prio; - entry->prev_state = prev->state; + entry->prev_state = __get_task_state(prev); entry->next_pid = next->pid; entry->next_prio = next->prio; - entry->next_state = next->state; + entry->next_state = __get_task_state(next); entry->next_cpu = task_cpu(next); if (!call_filter_check_discard(call, entry, buffer, event)) @@ -425,10 +425,10 @@ tracing_sched_wakeup_trace(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->prev_pid = curr->pid; entry->prev_prio = curr->prio; - entry->prev_state = curr->state; + entry->prev_state = __get_task_state(curr); entry->next_pid = wakee->pid; entry->next_prio = wakee->prio; - entry->next_state = wakee->state; + entry->next_state = __get_task_state(wakee); entry->next_cpu = task_cpu(wakee); if (!call_filter_check_discard(call, entry, buffer, event)) -- cgit v1.2.3 From 2b0b8499ae75df91455bbeb7491d45affc384fb0 Mon Sep 17 00:00:00 2001 From: Shu Wang Date: Tue, 12 Sep 2017 10:14:54 +0800 Subject: ftrace: Fix kmemleak in unregister_ftrace_graph The trampoline allocated by function tracer was overwriten by function_graph tracer, and caused a memory leak. The save_global_trampoline should have saved the previous trampoline in register_ftrace_graph() and restored it in unregister_ftrace_graph(). But as it is implemented, save_global_trampoline was only used in unregister_ftrace_graph as default value 0, and it overwrote the previous trampoline's value. Causing the previous allocated trampoline to be lost. kmmeleak backtrace: kmemleak_vmalloc+0x77/0xc0 __vmalloc_node_range+0x1b5/0x2c0 module_alloc+0x7c/0xd0 arch_ftrace_update_trampoline+0xb5/0x290 ftrace_startup+0x78/0x210 register_ftrace_function+0x8b/0xd0 function_trace_init+0x4f/0x80 tracing_set_tracer+0xe6/0x170 tracing_set_trace_write+0x90/0xd0 __vfs_write+0x37/0x170 vfs_write+0xb2/0x1b0 SyS_write+0x55/0xc0 do_syscall_64+0x67/0x180 return_from_SYSCALL_64+0x0/0x6a [ Looking further into this, I found that this was left over from when the function and function graph tracers shared the same ftrace_ops. But in commit 5f151b2401 ("ftrace: Fix function_profiler and function tracer together"), the two were separated, and the save_global_trampoline no longer was necessary (and it may have been broken back then too). -- Steven Rostedt ] Link: http://lkml.kernel.org/r/20170912021454.5976-1-shuwang@redhat.com Cc: stable@vger.kernel.org Fixes: 5f151b2401 ("ftrace: Fix function_profiler and function tracer together") Signed-off-by: Shu Wang Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 14 -------------- 1 file changed, 14 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 6abfafd7f173..8319e09e15b9 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -4954,9 +4954,6 @@ static char ftrace_graph_buf[FTRACE_FILTER_SIZE] __initdata; static char ftrace_graph_notrace_buf[FTRACE_FILTER_SIZE] __initdata; static int ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer); -static unsigned long save_global_trampoline; -static unsigned long save_global_flags; - static int __init set_graph_function(char *str) { strlcpy(ftrace_graph_buf, str, FTRACE_FILTER_SIZE); @@ -6808,17 +6805,6 @@ void unregister_ftrace_graph(void) unregister_pm_notifier(&ftrace_suspend_notifier); unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); -#ifdef CONFIG_DYNAMIC_FTRACE - /* - * Function graph does not allocate the trampoline, but - * other global_ops do. We need to reset the ALLOC_TRAMP flag - * if one was used. - */ - global_ops.trampoline = save_global_trampoline; - if (save_global_flags & FTRACE_OPS_FL_ALLOC_TRAMP) - global_ops.flags |= FTRACE_OPS_FL_ALLOC_TRAMP; -#endif - out: mutex_unlock(&ftrace_lock); } -- cgit v1.2.3 From d8c4deee6dc6876ae3b7d09a5b58138a57ee45f6 Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Fri, 8 Sep 2017 23:55:17 -0700 Subject: tracing: Remove obsolete sched_switch tracer selftest Since commit 87d80de2800d087ea833cb79bc13f85ff34ed49f ("tracing: Remove obsolete sched_switch tracer"), the sched_switch tracer selftest is no longer used. This patch removes the same. Link: http://lkml.kernel.org/r/20170909065517.22262-1-joelaf@google.com Cc: Ingo Molnar Cc: kernel-team@android.com Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.h | 2 -- kernel/trace/trace_selftest.c | 32 -------------------------------- 2 files changed, 34 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 652c682707cd..3c078e2ad777 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -738,8 +738,6 @@ extern int trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_nop(struct tracer *trace, struct trace_array *tr); -extern int trace_selftest_startup_sched_switch(struct tracer *trace, - struct trace_array *tr); extern int trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr); /* diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index b17ec642793b..364f78abdf47 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -1150,38 +1150,6 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) } #endif /* CONFIG_SCHED_TRACER */ -#ifdef CONFIG_CONTEXT_SWITCH_TRACER -int -trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr) -{ - unsigned long count; - int ret; - - /* start the tracing */ - ret = tracer_init(trace, tr); - if (ret) { - warn_failed_init_tracer(trace, ret); - return ret; - } - - /* Sleep for a 1/10 of a second */ - msleep(100); - /* stop the tracing. */ - tracing_stop(); - /* check the trace buffer */ - ret = trace_test_buffer(&tr->trace_buffer, &count); - trace->reset(tr); - tracing_start(); - - if (!ret && !count) { - printk(KERN_CONT ".. no entries found .."); - ret = -1; - } - - return ret; -} -#endif /* CONFIG_CONTEXT_SWITCH_TRACER */ - #ifdef CONFIG_BRANCH_TRACER int trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) -- cgit v1.2.3 From 6e7a2398114a2597a0995f96f44d908741ae5035 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 23 Aug 2017 12:23:09 +0100 Subject: tracing: Remove redundant unread variable ret Integer ret is being assigned but never used and hence it is redundant. Remove it, fixes clang warning: trace_events_hist.c:1077:3: warning: Value stored to 'ret' is never read Link: http://lkml.kernel.org/r/20170823112309.19383-1-colin.king@canonical.com Signed-off-by: Colin Ian King Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_hist.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 1c21d0e2a145..f123b5d0c226 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -1062,7 +1062,7 @@ static void hist_trigger_show(struct seq_file *m, struct event_trigger_data *data, int n) { struct hist_trigger_data *hist_data; - int n_entries, ret = 0; + int n_entries; if (n > 0) seq_puts(m, "\n\n"); @@ -1073,10 +1073,8 @@ static void hist_trigger_show(struct seq_file *m, hist_data = data->private_data; n_entries = print_entries(m, hist_data); - if (n_entries < 0) { - ret = n_entries; + if (n_entries < 0) n_entries = 0; - } seq_printf(m, "\nTotals:\n Hits: %llu\n Entries: %u\n Dropped: %llu\n", (u64)atomic64_read(&hist_data->map->hits), -- cgit v1.2.3 From 12ecef0cb12102d8c034770173d2d1363cb97d52 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Thu, 21 Sep 2017 16:22:49 -0400 Subject: tracing: Reverse the order of trace_types_lock and event_mutex In order to make future changes where we need to call tracing_set_clock() from within an event command, the order of trace_types_lock and event_mutex must be reversed, as the event command will hold event_mutex and the trace_types_lock is taken from within tracing_set_clock(). Link: http://lkml.kernel.org/r/20170921162249.0dde3dca@gandalf.local.home Requested-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 5 +++++ kernel/trace/trace_events.c | 31 +++++++++++++++---------------- 2 files changed, 20 insertions(+), 16 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 752e5daf0896..5f1ac7d3402c 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -7687,6 +7687,7 @@ static int instance_mkdir(const char *name) struct trace_array *tr; int ret; + mutex_lock(&event_mutex); mutex_lock(&trace_types_lock); ret = -EEXIST; @@ -7742,6 +7743,7 @@ static int instance_mkdir(const char *name) list_add(&tr->list, &ftrace_trace_arrays); mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); return 0; @@ -7753,6 +7755,7 @@ static int instance_mkdir(const char *name) out_unlock: mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); return ret; @@ -7765,6 +7768,7 @@ static int instance_rmdir(const char *name) int ret; int i; + mutex_lock(&event_mutex); mutex_lock(&trace_types_lock); ret = -ENODEV; @@ -7810,6 +7814,7 @@ static int instance_rmdir(const char *name) out_unlock: mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); return ret; } diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 87468398b9ed..ec0f9aa4e151 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -1406,8 +1406,8 @@ static int subsystem_open(struct inode *inode, struct file *filp) return -ENODEV; /* Make sure the system still exists */ - mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); + mutex_lock(&trace_types_lock); list_for_each_entry(tr, &ftrace_trace_arrays, list) { list_for_each_entry(dir, &tr->systems, list) { if (dir == inode->i_private) { @@ -1421,8 +1421,8 @@ static int subsystem_open(struct inode *inode, struct file *filp) } } exit_loop: - mutex_unlock(&event_mutex); mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); if (!system) return -ENODEV; @@ -2294,15 +2294,15 @@ static void __add_event_to_tracers(struct trace_event_call *call); int trace_add_event_call(struct trace_event_call *call) { int ret; - mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); + mutex_lock(&trace_types_lock); ret = __register_event(call, NULL); if (ret >= 0) __add_event_to_tracers(call); - mutex_unlock(&event_mutex); mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); return ret; } @@ -2356,13 +2356,13 @@ int trace_remove_event_call(struct trace_event_call *call) { int ret; - mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); + mutex_lock(&trace_types_lock); down_write(&trace_event_sem); ret = probe_remove_event_call(call); up_write(&trace_event_sem); - mutex_unlock(&event_mutex); mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); return ret; } @@ -2424,8 +2424,8 @@ static int trace_module_notify(struct notifier_block *self, { struct module *mod = data; - mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); + mutex_lock(&trace_types_lock); switch (val) { case MODULE_STATE_COMING: trace_module_add_events(mod); @@ -2434,8 +2434,8 @@ static int trace_module_notify(struct notifier_block *self, trace_module_remove_events(mod); break; } - mutex_unlock(&event_mutex); mutex_unlock(&trace_types_lock); + mutex_unlock(&event_mutex); return 0; } @@ -2950,24 +2950,24 @@ create_event_toplevel_files(struct dentry *parent, struct trace_array *tr) * creates the event hierachry in the @parent/events directory. * * Returns 0 on success. + * + * Must be called with event_mutex held. */ int event_trace_add_tracer(struct dentry *parent, struct trace_array *tr) { int ret; - mutex_lock(&event_mutex); + lockdep_assert_held(&event_mutex); ret = create_event_toplevel_files(parent, tr); if (ret) - goto out_unlock; + goto out; down_write(&trace_event_sem); __trace_add_event_dirs(tr); up_write(&trace_event_sem); - out_unlock: - mutex_unlock(&event_mutex); - + out: return ret; } @@ -2996,9 +2996,10 @@ early_event_add_tracer(struct dentry *parent, struct trace_array *tr) return ret; } +/* Must be called with event_mutex held */ int event_trace_del_tracer(struct trace_array *tr) { - mutex_lock(&event_mutex); + lockdep_assert_held(&event_mutex); /* Disable any event triggers and associated soft-disabled events */ clear_event_triggers(tr); @@ -3019,8 +3020,6 @@ int event_trace_del_tracer(struct trace_array *tr) tr->event_dir = NULL; - mutex_unlock(&event_mutex); - return 0; } -- cgit v1.2.3 From 1a149d7d3f45d311da1f63473736c05f30ae8a75 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Fri, 22 Sep 2017 16:59:02 -0400 Subject: ring-buffer: Rewrite trace_recursive_(un)lock() to be simpler The current method to prevent the ring buffer from entering into a recursize loop is to use a bitmask and set the bit that maps to the current context (normal, softirq, irq or NMI), and if that bit was already set, it is considered a recursive loop. New code is being added that may require the ring buffer to be entered a second time in the current context. The recursive locking prevents that from happening. Instead of mapping a bitmask to the current context, just allow 4 levels of nesting in the ring buffer. This matches the 4 context levels that it can already nest. It is highly unlikely to have more than two levels, thus it should be fine when we add the second entry into the ring buffer. If that proves to be a problem, we can always up the number to 8. An added benefit is that reading preempt_count() to get the current level adds a very slight but noticeable overhead. This removes that need. Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ring_buffer.c | 64 ++++++++++++---------------------------------- 1 file changed, 17 insertions(+), 47 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 81279c6602ff..f6ee9b1ef62a 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2538,61 +2538,29 @@ rb_wakeups(struct ring_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer) * The lock and unlock are done within a preempt disable section. * The current_context per_cpu variable can only be modified * by the current task between lock and unlock. But it can - * be modified more than once via an interrupt. To pass this - * information from the lock to the unlock without having to - * access the 'in_interrupt()' functions again (which do show - * a bit of overhead in something as critical as function tracing, - * we use a bitmask trick. + * be modified more than once via an interrupt. There are four + * different contexts that we need to consider. * - * bit 0 = NMI context - * bit 1 = IRQ context - * bit 2 = SoftIRQ context - * bit 3 = normal context. + * Normal context. + * SoftIRQ context + * IRQ context + * NMI context * - * This works because this is the order of contexts that can - * preempt other contexts. A SoftIRQ never preempts an IRQ - * context. - * - * When the context is determined, the corresponding bit is - * checked and set (if it was set, then a recursion of that context - * happened). - * - * On unlock, we need to clear this bit. To do so, just subtract - * 1 from the current_context and AND it to itself. - * - * (binary) - * 101 - 1 = 100 - * 101 & 100 = 100 (clearing bit zero) - * - * 1010 - 1 = 1001 - * 1010 & 1001 = 1000 (clearing bit 1) - * - * The least significant bit can be cleared this way, and it - * just so happens that it is the same bit corresponding to - * the current context. + * If for some reason the ring buffer starts to recurse, we + * only allow that to happen at most 4 times (one for each + * context). If it happens 5 times, then we consider this a + * recusive loop and do not let it go further. */ static __always_inline int trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer) { - unsigned int val = cpu_buffer->current_context; - int bit; - - if (in_interrupt()) { - if (in_nmi()) - bit = RB_CTX_NMI; - else if (in_irq()) - bit = RB_CTX_IRQ; - else - bit = RB_CTX_SOFTIRQ; - } else - bit = RB_CTX_NORMAL; - - if (unlikely(val & (1 << bit))) + if (cpu_buffer->current_context >= 4) return 1; - val |= (1 << bit); - cpu_buffer->current_context = val; + cpu_buffer->current_context++; + /* Interrupts must see this update */ + barrier(); return 0; } @@ -2600,7 +2568,9 @@ trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer) static __always_inline void trace_recursive_unlock(struct ring_buffer_per_cpu *cpu_buffer) { - cpu_buffer->current_context &= cpu_buffer->current_context - 1; + /* Don't let the dec leak out */ + barrier(); + cpu_buffer->current_context--; } /** -- cgit v1.2.3 From a15f7fc20389a8827d5859907568b201234d4b79 Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:17 -0500 Subject: tracing: Exclude 'generic fields' from histograms There are a small number of 'generic fields' (comm/COMM/cpu/CPU) that are found by trace_find_event_field() but are only meant for filtering. Specifically, they unlike normal fields, they have a size of 0 and thus wreak havoc when used as a histogram key. Exclude these (return -EINVAL) when used as histogram keys. Link: http://lkml.kernel.org/r/956154cbc3e8a4f0633d619b886c97f0f0edf7b4.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_hist.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index f123b5d0c226..121d56850f24 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -450,7 +450,7 @@ static int create_val_field(struct hist_trigger_data *hist_data, } field = trace_find_event_field(file->event_call, field_name); - if (!field) { + if (!field || !field->size) { ret = -EINVAL; goto out; } @@ -548,7 +548,7 @@ static int create_key_field(struct hist_trigger_data *hist_data, } field = trace_find_event_field(file->event_call, field_name); - if (!field) { + if (!field || !field->size) { ret = -EINVAL; goto out; } -- cgit v1.2.3 From 83c07ecc4203728e85fc4a2ce6fdf25d16ea118e Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:18 -0500 Subject: tracing: Remove lookups from tracing_map hitcount Lookups inflate the hitcount, making it essentially useless. Only inserts and updates should really affect the hitcount anyway, so explicitly filter lookups out. Link: http://lkml.kernel.org/r/c8d9dc39d269a8abf88bf4102d0dfc65deb0fc7f.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/tracing_map.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/tracing_map.c b/kernel/trace/tracing_map.c index 305039b122fa..07e75344725b 100644 --- a/kernel/trace/tracing_map.c +++ b/kernel/trace/tracing_map.c @@ -428,7 +428,8 @@ __tracing_map_insert(struct tracing_map *map, void *key, bool lookup_only) if (test_key && test_key == key_hash && entry->val && keys_match(key, entry->val->key, map->key_size)) { - atomic64_inc(&map->hits); + if (!lookup_only) + atomic64_inc(&map->hits); return entry->val; } -- cgit v1.2.3 From 4f36c2d85cedea60ad424d44534121ab0458069e Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:19 -0500 Subject: tracing: Increase tracing map KEYS_MAX size The current default for the number of subkeys in a compound key is 2, which is too restrictive. Increase it to a more realistic value of 3. Link: http://lkml.kernel.org/r/b6952cca06d1f912eba33804a6fd6069b3847d44.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/tracing_map.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/tracing_map.h b/kernel/trace/tracing_map.h index 618838f5f30a..f0975110b967 100644 --- a/kernel/trace/tracing_map.h +++ b/kernel/trace/tracing_map.h @@ -5,7 +5,7 @@ #define TRACING_MAP_BITS_MAX 17 #define TRACING_MAP_BITS_MIN 7 -#define TRACING_MAP_KEYS_MAX 2 +#define TRACING_MAP_KEYS_MAX 3 #define TRACING_MAP_VALS_MAX 3 #define TRACING_MAP_FIELDS_MAX (TRACING_MAP_KEYS_MAX + \ TRACING_MAP_VALS_MAX) -- cgit v1.2.3 From 7e465baa80293ed5f87fdf6405391d6f02110d4e Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:20 -0500 Subject: tracing: Make traceprobe parsing code reusable traceprobe_probes_write() and traceprobe_command() actually contain nothing that ties them to kprobes - the code is generically useful for similar types of parsing elsewhere, so separate it out and move it to trace.c/trace.h. Other than moving it, the only change is in naming: traceprobe_probes_write() becomes trace_parse_run_command() and traceprobe_command() becomes trace_run_command(). Link: http://lkml.kernel.org/r/ae5c26ea40c196a8986854d921eb6e713ede7e3f.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace.c | 86 +++++++++++++++++++++++++++++++++++++++++++++ kernel/trace/trace.h | 7 ++++ kernel/trace/trace_kprobe.c | 18 +++++----- kernel/trace/trace_probe.c | 86 --------------------------------------------- kernel/trace/trace_probe.h | 7 ---- kernel/trace/trace_uprobe.c | 2 +- 6 files changed, 103 insertions(+), 103 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 5f1ac7d3402c..73e67b68c53b 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -8281,6 +8281,92 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) } EXPORT_SYMBOL_GPL(ftrace_dump); +int trace_run_command(const char *buf, int (*createfn)(int, char **)) +{ + char **argv; + int argc, ret; + + argc = 0; + ret = 0; + argv = argv_split(GFP_KERNEL, buf, &argc); + if (!argv) + return -ENOMEM; + + if (argc) + ret = createfn(argc, argv); + + argv_free(argv); + + return ret; +} + +#define WRITE_BUFSIZE 4096 + +ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, + size_t count, loff_t *ppos, + int (*createfn)(int, char **)) +{ + char *kbuf, *buf, *tmp; + int ret = 0; + size_t done = 0; + size_t size; + + kbuf = kmalloc(WRITE_BUFSIZE, GFP_KERNEL); + if (!kbuf) + return -ENOMEM; + + while (done < count) { + size = count - done; + + if (size >= WRITE_BUFSIZE) + size = WRITE_BUFSIZE - 1; + + if (copy_from_user(kbuf, buffer + done, size)) { + ret = -EFAULT; + goto out; + } + kbuf[size] = '\0'; + buf = kbuf; + do { + tmp = strchr(buf, '\n'); + if (tmp) { + *tmp = '\0'; + size = tmp - buf + 1; + } else { + size = strlen(buf); + if (done + size < count) { + if (buf != kbuf) + break; + /* This can accept WRITE_BUFSIZE - 2 ('\n' + '\0') */ + pr_warn("Line length is too long: Should be less than %d\n", + WRITE_BUFSIZE - 2); + ret = -EINVAL; + goto out; + } + } + done += size; + + /* Remove comments */ + tmp = strchr(buf, '#'); + + if (tmp) + *tmp = '\0'; + + ret = trace_run_command(buf, createfn); + if (ret) + goto out; + buf += size; + + } while (done < count); + } + ret = done; + +out: + kfree(kbuf); + + return ret; +} + __init static int tracer_alloc_buffers(void) { int ring_buf_size; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 3c078e2ad777..f8343eb3c1f9 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1752,6 +1752,13 @@ void trace_printk_start_comm(void); int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set); int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled); +#define MAX_EVENT_NAME_LEN 64 + +extern int trace_run_command(const char *buf, int (*createfn)(int, char**)); +extern ssize_t trace_parse_run_command(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos, + int (*createfn)(int, char**)); + /* * Normal trace_printk() and friends allocates special buffers * to do the manipulation, as well as saves the print formats diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 8a907e12b6b9..af6134f2e597 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -907,8 +907,8 @@ static int probes_open(struct inode *inode, struct file *file) static ssize_t probes_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { - return traceprobe_probes_write(file, buffer, count, ppos, - create_trace_kprobe); + return trace_parse_run_command(file, buffer, count, ppos, + create_trace_kprobe); } static const struct file_operations kprobe_events_ops = { @@ -1433,9 +1433,9 @@ static __init int kprobe_trace_self_tests_init(void) pr_info("Testing kprobe tracing: "); - ret = traceprobe_command("p:testprobe kprobe_trace_selftest_target " - "$stack $stack0 +0($stack)", - create_trace_kprobe); + ret = trace_run_command("p:testprobe kprobe_trace_selftest_target " + "$stack $stack0 +0($stack)", + create_trace_kprobe); if (WARN_ON_ONCE(ret)) { pr_warn("error on probing function entry.\n"); warn++; @@ -1455,8 +1455,8 @@ static __init int kprobe_trace_self_tests_init(void) } } - ret = traceprobe_command("r:testprobe2 kprobe_trace_selftest_target " - "$retval", create_trace_kprobe); + ret = trace_run_command("r:testprobe2 kprobe_trace_selftest_target " + "$retval", create_trace_kprobe); if (WARN_ON_ONCE(ret)) { pr_warn("error on probing function return.\n"); warn++; @@ -1526,13 +1526,13 @@ static __init int kprobe_trace_self_tests_init(void) disable_trace_kprobe(tk, file); } - ret = traceprobe_command("-:testprobe", create_trace_kprobe); + ret = trace_run_command("-:testprobe", create_trace_kprobe); if (WARN_ON_ONCE(ret)) { pr_warn("error on deleting a probe.\n"); warn++; } - ret = traceprobe_command("-:testprobe2", create_trace_kprobe); + ret = trace_run_command("-:testprobe2", create_trace_kprobe); if (WARN_ON_ONCE(ret)) { pr_warn("error on deleting a probe.\n"); warn++; diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 52478f033f88..d59357308677 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -623,92 +623,6 @@ void traceprobe_free_probe_arg(struct probe_arg *arg) kfree(arg->comm); } -int traceprobe_command(const char *buf, int (*createfn)(int, char **)) -{ - char **argv; - int argc, ret; - - argc = 0; - ret = 0; - argv = argv_split(GFP_KERNEL, buf, &argc); - if (!argv) - return -ENOMEM; - - if (argc) - ret = createfn(argc, argv); - - argv_free(argv); - - return ret; -} - -#define WRITE_BUFSIZE 4096 - -ssize_t traceprobe_probes_write(struct file *file, const char __user *buffer, - size_t count, loff_t *ppos, - int (*createfn)(int, char **)) -{ - char *kbuf, *buf, *tmp; - int ret = 0; - size_t done = 0; - size_t size; - - kbuf = kmalloc(WRITE_BUFSIZE, GFP_KERNEL); - if (!kbuf) - return -ENOMEM; - - while (done < count) { - size = count - done; - - if (size >= WRITE_BUFSIZE) - size = WRITE_BUFSIZE - 1; - - if (copy_from_user(kbuf, buffer + done, size)) { - ret = -EFAULT; - goto out; - } - kbuf[size] = '\0'; - buf = kbuf; - do { - tmp = strchr(buf, '\n'); - if (tmp) { - *tmp = '\0'; - size = tmp - buf + 1; - } else { - size = strlen(buf); - if (done + size < count) { - if (buf != kbuf) - break; - /* This can accept WRITE_BUFSIZE - 2 ('\n' + '\0') */ - pr_warn("Line length is too long: Should be less than %d\n", - WRITE_BUFSIZE - 2); - ret = -EINVAL; - goto out; - } - } - done += size; - - /* Remove comments */ - tmp = strchr(buf, '#'); - - if (tmp) - *tmp = '\0'; - - ret = traceprobe_command(buf, createfn); - if (ret) - goto out; - buf += size; - - } while (done < count); - } - ret = done; - -out: - kfree(kbuf); - - return ret; -} - static int __set_print_fmt(struct trace_probe *tp, char *buf, int len, bool is_return) { diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index 903273c93e61..fb66e3eaa192 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -42,7 +42,6 @@ #define MAX_TRACE_ARGS 128 #define MAX_ARGSTR_LEN 63 -#define MAX_EVENT_NAME_LEN 64 #define MAX_STRING_SIZE PATH_MAX /* Reserved field names */ @@ -356,12 +355,6 @@ extern void traceprobe_free_probe_arg(struct probe_arg *arg); extern int traceprobe_split_symbol_offset(char *symbol, unsigned long *offset); -extern ssize_t traceprobe_probes_write(struct file *file, - const char __user *buffer, size_t count, loff_t *ppos, - int (*createfn)(int, char**)); - -extern int traceprobe_command(const char *buf, int (*createfn)(int, char**)); - /* Sum up total data length for dynamic arraies (strings) */ static nokprobe_inline int __get_data_size(struct trace_probe *tp, struct pt_regs *regs) diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 4525e0271a53..b34965e62fb9 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -651,7 +651,7 @@ static int probes_open(struct inode *inode, struct file *file) static ssize_t probes_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { - return traceprobe_probes_write(file, buffer, count, ppos, create_trace_uprobe); + return trace_parse_run_command(file, buffer, count, ppos, create_trace_uprobe); } static const struct file_operations uprobe_events_ops = { -- cgit v1.2.3 From 0d7a8325bf3326c92da2d21b4496a9ddde896d4f Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:21 -0500 Subject: tracing: Clean up hist_field_flags enum As we add more flags, specifying explicit integers for the flag values becomes more unwieldy and error-prone - switch them over to left-shift values. Link: http://lkml.kernel.org/r/e644e4fb7665aec015f4a2d84a2f990d3dd5b8a1.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_hist.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 121d56850f24..0c7ec3048624 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -110,16 +110,16 @@ DEFINE_HIST_FIELD_FN(u8); #define HIST_KEY_SIZE_MAX (MAX_FILTER_STR_VAL + HIST_STACKTRACE_SIZE) enum hist_field_flags { - HIST_FIELD_FL_HITCOUNT = 1, - HIST_FIELD_FL_KEY = 2, - HIST_FIELD_FL_STRING = 4, - HIST_FIELD_FL_HEX = 8, - HIST_FIELD_FL_SYM = 16, - HIST_FIELD_FL_SYM_OFFSET = 32, - HIST_FIELD_FL_EXECNAME = 64, - HIST_FIELD_FL_SYSCALL = 128, - HIST_FIELD_FL_STACKTRACE = 256, - HIST_FIELD_FL_LOG2 = 512, + HIST_FIELD_FL_HITCOUNT = 1 << 0, + HIST_FIELD_FL_KEY = 1 << 1, + HIST_FIELD_FL_STRING = 1 << 2, + HIST_FIELD_FL_HEX = 1 << 3, + HIST_FIELD_FL_SYM = 1 << 4, + HIST_FIELD_FL_SYM_OFFSET = 1 << 5, + HIST_FIELD_FL_EXECNAME = 1 << 6, + HIST_FIELD_FL_SYSCALL = 1 << 7, + HIST_FIELD_FL_STACKTRACE = 1 << 8, + HIST_FIELD_FL_LOG2 = 1 << 9, }; struct hist_trigger_attrs { -- cgit v1.2.3 From 85013256cf01629f72a327674c5d007b4a4b40da Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:22 -0500 Subject: tracing: Add hist_field_name() accessor In preparation for hist_fields that won't be strictly based on trace_event_fields, add a new hist_field_name() accessor to allow that flexibility and update associated users. Link: http://lkml.kernel.org/r/5b5a2d36dde067cbbe2434b10f06daac27b7dbd5.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_hist.c | 67 +++++++++++++++++++++++++++------------- 1 file changed, 45 insertions(+), 22 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 0c7ec3048624..34edf5fd85fd 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -146,6 +146,23 @@ struct hist_trigger_data { struct tracing_map *map; }; +static const char *hist_field_name(struct hist_field *field, + unsigned int level) +{ + const char *field_name = ""; + + if (level > 1) + return field_name; + + if (field->field) + field_name = field->field->name; + + if (field_name == NULL) + field_name = ""; + + return field_name; +} + static hist_field_fn_t select_value_fn(int field_size, int field_is_signed) { hist_field_fn_t fn = NULL; @@ -653,7 +670,6 @@ static int is_descending(const char *str) static int create_sort_keys(struct hist_trigger_data *hist_data) { char *fields_str = hist_data->attrs->sort_key_str; - struct ftrace_event_field *field = NULL; struct tracing_map_sort_key *sort_key; int descending, ret = 0; unsigned int i, j; @@ -670,7 +686,9 @@ static int create_sort_keys(struct hist_trigger_data *hist_data) } for (i = 0; i < TRACING_MAP_SORT_KEYS_MAX; i++) { + struct hist_field *hist_field; char *field_str, *field_name; + const char *test_name; sort_key = &hist_data->sort_keys[i]; @@ -703,8 +721,10 @@ static int create_sort_keys(struct hist_trigger_data *hist_data) } for (j = 1; j < hist_data->n_fields; j++) { - field = hist_data->fields[j]->field; - if (field && (strcmp(field_name, field->name) == 0)) { + hist_field = hist_data->fields[j]; + test_name = hist_field_name(hist_field, 0); + + if (strcmp(field_name, test_name) == 0) { sort_key->field_idx = j; descending = is_descending(field_str); if (descending < 0) { @@ -952,6 +972,7 @@ hist_trigger_entry_print(struct seq_file *m, struct hist_field *key_field; char str[KSYM_SYMBOL_LEN]; bool multiline = false; + const char *field_name; unsigned int i; u64 uval; @@ -963,26 +984,27 @@ hist_trigger_entry_print(struct seq_file *m, if (i > hist_data->n_vals) seq_puts(m, ", "); + field_name = hist_field_name(key_field, 0); + if (key_field->flags & HIST_FIELD_FL_HEX) { uval = *(u64 *)(key + key_field->offset); - seq_printf(m, "%s: %llx", - key_field->field->name, uval); + seq_printf(m, "%s: %llx", field_name, uval); } else if (key_field->flags & HIST_FIELD_FL_SYM) { uval = *(u64 *)(key + key_field->offset); sprint_symbol_no_offset(str, uval); - seq_printf(m, "%s: [%llx] %-45s", - key_field->field->name, uval, str); + seq_printf(m, "%s: [%llx] %-45s", field_name, + uval, str); } else if (key_field->flags & HIST_FIELD_FL_SYM_OFFSET) { uval = *(u64 *)(key + key_field->offset); sprint_symbol(str, uval); - seq_printf(m, "%s: [%llx] %-55s", - key_field->field->name, uval, str); + seq_printf(m, "%s: [%llx] %-55s", field_name, + uval, str); } else if (key_field->flags & HIST_FIELD_FL_EXECNAME) { char *comm = elt->private_data; uval = *(u64 *)(key + key_field->offset); - seq_printf(m, "%s: %-16s[%10llu]", - key_field->field->name, comm, uval); + seq_printf(m, "%s: %-16s[%10llu]", field_name, + comm, uval); } else if (key_field->flags & HIST_FIELD_FL_SYSCALL) { const char *syscall_name; @@ -991,8 +1013,8 @@ hist_trigger_entry_print(struct seq_file *m, if (!syscall_name) syscall_name = "unknown_syscall"; - seq_printf(m, "%s: %-30s[%3llu]", - key_field->field->name, syscall_name, uval); + seq_printf(m, "%s: %-30s[%3llu]", field_name, + syscall_name, uval); } else if (key_field->flags & HIST_FIELD_FL_STACKTRACE) { seq_puts(m, "stacktrace:\n"); hist_trigger_stacktrace_print(m, @@ -1000,15 +1022,14 @@ hist_trigger_entry_print(struct seq_file *m, HIST_STACKTRACE_DEPTH); multiline = true; } else if (key_field->flags & HIST_FIELD_FL_LOG2) { - seq_printf(m, "%s: ~ 2^%-2llu", key_field->field->name, + seq_printf(m, "%s: ~ 2^%-2llu", field_name, *(u64 *)(key + key_field->offset)); } else if (key_field->flags & HIST_FIELD_FL_STRING) { - seq_printf(m, "%s: %-50s", key_field->field->name, + seq_printf(m, "%s: %-50s", field_name, (char *)(key + key_field->offset)); } else { uval = *(u64 *)(key + key_field->offset); - seq_printf(m, "%s: %10llu", key_field->field->name, - uval); + seq_printf(m, "%s: %10llu", field_name, uval); } } @@ -1021,13 +1042,13 @@ hist_trigger_entry_print(struct seq_file *m, tracing_map_read_sum(elt, HITCOUNT_IDX)); for (i = 1; i < hist_data->n_vals; i++) { + field_name = hist_field_name(hist_data->fields[i], 0); + if (hist_data->fields[i]->flags & HIST_FIELD_FL_HEX) { - seq_printf(m, " %s: %10llx", - hist_data->fields[i]->field->name, + seq_printf(m, " %s: %10llx", field_name, tracing_map_read_sum(elt, i)); } else { - seq_printf(m, " %s: %10llu", - hist_data->fields[i]->field->name, + seq_printf(m, " %s: %10llu", field_name, tracing_map_read_sum(elt, i)); } } @@ -1140,7 +1161,9 @@ static const char *get_hist_field_flags(struct hist_field *hist_field) static void hist_field_print(struct seq_file *m, struct hist_field *hist_field) { - seq_printf(m, "%s", hist_field->field->name); + const char *field_name = hist_field_name(hist_field, 0); + + seq_printf(m, "%s", field_name); if (hist_field->flags) { const char *flags_str = get_hist_field_flags(hist_field); -- cgit v1.2.3 From 5819eaddf35b24d628ddfa4fbb5f8d4026e44b96 Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Fri, 22 Sep 2017 14:58:23 -0500 Subject: tracing: Reimplement log2 log2 as currently implemented applies only to u64 trace_event_field derived fields, and assumes that anything it's applied to is a u64 field. To prepare for synthetic fields like latencies, log2 should be applicable to those as well, so take the opportunity now to fix the current problems as well as expand to more general uses. log2 should be thought of as a chaining function rather than a field type. To enable this as well as possible future function implementations, add a hist_field operand array into the hist_field definition for this purpose, and make use of it to implement the log2 'function'. Link: http://lkml.kernel.org/r/b47f93fc0b87b36eccf716b0c018f3a71e1f1111.1506105045.git.tom.zanussi@linux.intel.com Signed-off-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_hist.c | 31 +++++++++++++++++++++++++++---- 1 file changed, 27 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index 34edf5fd85fd..1e1558c99d56 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -28,12 +28,16 @@ struct hist_field; typedef u64 (*hist_field_fn_t) (struct hist_field *field, void *event); +#define HIST_FIELD_OPERANDS_MAX 2 + struct hist_field { struct ftrace_event_field *field; unsigned long flags; hist_field_fn_t fn; unsigned int size; unsigned int offset; + unsigned int is_signed; + struct hist_field *operands[HIST_FIELD_OPERANDS_MAX]; }; static u64 hist_field_none(struct hist_field *field, void *event) @@ -71,7 +75,9 @@ static u64 hist_field_pstring(struct hist_field *hist_field, void *event) static u64 hist_field_log2(struct hist_field *hist_field, void *event) { - u64 val = *(u64 *)(event + hist_field->field->offset); + struct hist_field *operand = hist_field->operands[0]; + + u64 val = operand->fn(operand, event); return (u64) ilog2(roundup_pow_of_two(val)); } @@ -156,6 +162,8 @@ static const char *hist_field_name(struct hist_field *field, if (field->field) field_name = field->field->name; + else if (field->flags & HIST_FIELD_FL_LOG2) + field_name = hist_field_name(field->operands[0], ++level); if (field_name == NULL) field_name = ""; @@ -357,8 +365,20 @@ static const struct tracing_map_ops hist_trigger_elt_comm_ops = { .elt_init = hist_trigger_elt_comm_init, }; -static void destroy_hist_field(struct hist_field *hist_field) +static void destroy_hist_field(struct hist_field *hist_field, + unsigned int level) { + unsigned int i; + + if (level > 2) + return; + + if (!hist_field) + return; + + for (i = 0; i < HIST_FIELD_OPERANDS_MAX; i++) + destroy_hist_field(hist_field->operands[i], level + 1); + kfree(hist_field); } @@ -385,7 +405,10 @@ static struct hist_field *create_hist_field(struct ftrace_event_field *field, } if (flags & HIST_FIELD_FL_LOG2) { + unsigned long fl = flags & ~HIST_FIELD_FL_LOG2; hist_field->fn = hist_field_log2; + hist_field->operands[0] = create_hist_field(field, fl); + hist_field->size = hist_field->operands[0]->size; goto out; } @@ -405,7 +428,7 @@ static struct hist_field *create_hist_field(struct ftrace_event_field *field, hist_field->fn = select_value_fn(field->size, field->is_signed); if (!hist_field->fn) { - destroy_hist_field(hist_field); + destroy_hist_field(hist_field, 0); return NULL; } } @@ -422,7 +445,7 @@ static void destroy_hist_fields(struct hist_trigger_data *hist_data) for (i = 0; i < TRACING_MAP_FIELDS_MAX; i++) { if (hist_data->fields[i]) { - destroy_hist_field(hist_data->fields[i]); + destroy_hist_field(hist_data->fields[i], 0); hist_data->fields[i] = NULL; } } -- cgit v1.2.3 From 6cafbe159416822f6d3dfd711bf4c39050c650ba Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 20 Jun 2017 10:44:58 -0400 Subject: ftrace: Add a ftrace_free_mem() function for modules to use In order to be able to trace module init functions, the module code needs to tell ftrace what is being freed when the init sections are freed. Use the code that the main init calls to tell ftrace to free the main init sections. This requires passing in a start and end address to free. Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 6abfafd7f173..84cb5928665a 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5868,10 +5868,10 @@ void ftrace_module_init(struct module *mod) } #endif /* CONFIG_MODULES */ -void __init ftrace_free_init_mem(void) +void ftrace_free_mem(void *start_ptr, void *end_ptr) { - unsigned long start = (unsigned long)(&__init_begin); - unsigned long end = (unsigned long)(&__init_end); + unsigned long start = (unsigned long)(start_ptr); + unsigned long end = (unsigned long)(end_ptr); struct ftrace_page **last_pg = &ftrace_pages_start; struct ftrace_page *pg; struct dyn_ftrace *rec; @@ -5913,6 +5913,14 @@ void __init ftrace_free_init_mem(void) mutex_unlock(&ftrace_lock); } +void __init ftrace_free_init_mem(void) +{ + void *start = (void *)(&__init_begin); + void *end = (void *)(&__init_end); + + ftrace_free_mem(start, end); +} + void __init ftrace_init(void) { extern unsigned long __start_mcount_loc[]; -- cgit v1.2.3 From 3e234289f86b12985ef8909cd34525fcb66c4efb Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Fri, 3 Mar 2017 18:00:22 -0500 Subject: ftrace: Allow module init functions to be traced Allow for module init sections to be traced as well as core kernel init sections. Now that filtering modules functions can be stored, for when they are loaded, it makes sense to be able to trace them. Cc: Jessica Yu Cc: Rusty Russell Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 84cb5928665a..d7297e866e4a 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5752,7 +5752,8 @@ void ftrace_release_mod(struct module *mod) last_pg = &ftrace_pages_start; for (pg = ftrace_pages_start; pg; pg = *last_pg) { rec = &pg->records[0]; - if (within_module_core(rec->ip, mod)) { + if (within_module_core(rec->ip, mod) || + within_module_init(rec->ip, mod)) { /* * As core pages are first, the first * page should never be a module page. @@ -5821,7 +5822,8 @@ void ftrace_module_enable(struct module *mod) * not part of this module, then skip this pg, * which the "break" will do. */ - if (!within_module_core(rec->ip, mod)) + if (!within_module_core(rec->ip, mod) && + !within_module_init(rec->ip, mod)) break; cnt = 0; -- cgit v1.2.3 From aba4b5c22cbac296f4081a0476d0c55828f135b4 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Fri, 1 Sep 2017 08:35:38 -0400 Subject: ftrace: Save module init functions kallsyms symbols for tracing If function tracing is active when the module init functions are freed, then store them to be referenced by kallsyms. As module init functions can now be traced on module load, they were useless: ># echo ':mod:snd_seq' > set_ftrace_filter ># echo function > current_tracer ># modprobe snd_seq ># cat trace # tracer: function # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | modprobe-2786 [000] .... 3189.037874: 0xffffffffa0860000 <-do_one_initcall modprobe-2786 [000] .... 3189.037876: 0xffffffffa086004d <-0xffffffffa086000f modprobe-2786 [000] .... 3189.037876: 0xffffffffa086010d <-0xffffffffa0860018 modprobe-2786 [000] .... 3189.037877: 0xffffffffa086011a <-0xffffffffa0860021 modprobe-2786 [000] .... 3189.037877: 0xffffffffa0860080 <-0xffffffffa086002a modprobe-2786 [000] .... 3189.039523: 0xffffffffa0860400 <-0xffffffffa0860033 modprobe-2786 [000] .... 3189.039523: 0xffffffffa086038a <-0xffffffffa086041c modprobe-2786 [000] .... 3189.039591: 0xffffffffa086038a <-0xffffffffa0860436 modprobe-2786 [000] .... 3189.039657: 0xffffffffa086038a <-0xffffffffa0860450 modprobe-2786 [000] .... 3189.039719: 0xffffffffa0860127 <-0xffffffffa086003c modprobe-2786 [000] .... 3189.039742: snd_seq_create_kernel_client <-0xffffffffa08601f6 When the output is shown, the kallsyms for the module init functions have already been freed, and the output of the trace can not convert them to their function names. Now this looks like this: # tracer: function # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | modprobe-2463 [002] .... 174.243237: alsa_seq_init <-do_one_initcall modprobe-2463 [002] .... 174.243239: client_init_data <-alsa_seq_init modprobe-2463 [002] .... 174.243240: snd_sequencer_memory_init <-alsa_seq_init modprobe-2463 [002] .... 174.243240: snd_seq_queues_init <-alsa_seq_init modprobe-2463 [002] .... 174.243240: snd_sequencer_device_init <-alsa_seq_init modprobe-2463 [002] .... 174.244860: snd_seq_info_init <-alsa_seq_init modprobe-2463 [002] .... 174.244861: create_info_entry <-snd_seq_info_init modprobe-2463 [002] .... 174.244936: create_info_entry <-snd_seq_info_init modprobe-2463 [002] .... 174.245003: create_info_entry <-snd_seq_info_init modprobe-2463 [002] .... 174.245072: snd_seq_system_client_init <-alsa_seq_init modprobe-2463 [002] .... 174.245094: snd_seq_create_kernel_client <-snd_seq_system_client_init Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 146 +++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 144 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index d7297e866e4a..86dbbfb353db 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5675,6 +5675,21 @@ static int ftrace_process_locs(struct module *mod, return ret; } +struct ftrace_mod_func { + struct list_head list; + char *name; + unsigned long ip; + unsigned int size; +}; + +struct ftrace_mod_map { + struct list_head list; + struct module *mod; + unsigned long start_addr; + unsigned long end_addr; + struct list_head funcs; +}; + #ifdef CONFIG_MODULES #define next_to_ftrace_page(p) container_of(p, struct ftrace_page, next) @@ -5868,9 +5883,123 @@ void ftrace_module_init(struct module *mod) ftrace_process_locs(mod, mod->ftrace_callsites, mod->ftrace_callsites + mod->num_ftrace_callsites); } + +static void save_ftrace_mod_rec(struct ftrace_mod_map *mod_map, + struct dyn_ftrace *rec) +{ + struct ftrace_mod_func *mod_func; + unsigned long symsize; + unsigned long offset; + char str[KSYM_SYMBOL_LEN]; + char *modname; + const char *ret; + + ret = kallsyms_lookup(rec->ip, &symsize, &offset, &modname, str); + if (!ret) + return; + + mod_func = kmalloc(sizeof(*mod_func), GFP_KERNEL); + if (!mod_func) + return; + + mod_func->name = kstrdup(str, GFP_KERNEL); + if (!mod_func->name) { + kfree(mod_func); + return; + } + + mod_func->ip = rec->ip - offset; + mod_func->size = symsize; + + list_add_rcu(&mod_func->list, &mod_map->funcs); +} + +static LIST_HEAD(ftrace_mod_maps); + +static struct ftrace_mod_map * +allocate_ftrace_mod_map(struct module *mod, + unsigned long start, unsigned long end) +{ + struct ftrace_mod_map *mod_map; + + mod_map = kmalloc(sizeof(*mod_map), GFP_KERNEL); + if (!mod_map) + return NULL; + + mod_map->mod = mod; + mod_map->start_addr = start; + mod_map->end_addr = end; + + INIT_LIST_HEAD_RCU(&mod_map->funcs); + + list_add_rcu(&mod_map->list, &ftrace_mod_maps); + + return mod_map; +} + +static const char * +ftrace_func_address_lookup(struct ftrace_mod_map *mod_map, + unsigned long addr, unsigned long *size, + unsigned long *off, char *sym) +{ + struct ftrace_mod_func *found_func = NULL; + struct ftrace_mod_func *mod_func; + + list_for_each_entry_rcu(mod_func, &mod_map->funcs, list) { + if (addr >= mod_func->ip && + addr < mod_func->ip + mod_func->size) { + found_func = mod_func; + break; + } + } + + if (found_func) { + if (size) + *size = found_func->size; + if (off) + *off = addr - found_func->ip; + if (sym) + strlcpy(sym, found_func->name, KSYM_NAME_LEN); + + return found_func->name; + } + + return NULL; +} + +const char * +ftrace_mod_address_lookup(unsigned long addr, unsigned long *size, + unsigned long *off, char **modname, char *sym) +{ + struct ftrace_mod_map *mod_map; + const char *ret = NULL; + + preempt_disable(); + list_for_each_entry_rcu(mod_map, &ftrace_mod_maps, list) { + ret = ftrace_func_address_lookup(mod_map, addr, size, off, sym); + if (ret) { + if (modname) + *modname = mod_map->mod->name; + break; + } + } + preempt_enable(); + + return ret; +} + +#else +static void save_ftrace_mod_rec(struct ftrace_mod_map *mod_map, + struct dyn_ftrace *rec) { } +static inline struct ftrace_mod_map * +allocate_ftrace_mod_map(struct module *mod, + unsigned long start, unsigned long end) +{ + return NULL; +} #endif /* CONFIG_MODULES */ -void ftrace_free_mem(void *start_ptr, void *end_ptr) +void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) { unsigned long start = (unsigned long)(start_ptr); unsigned long end = (unsigned long)(end_ptr); @@ -5878,6 +6007,7 @@ void ftrace_free_mem(void *start_ptr, void *end_ptr) struct ftrace_page *pg; struct dyn_ftrace *rec; struct dyn_ftrace key; + struct ftrace_mod_map *mod_map = NULL; int order; key.ip = start; @@ -5885,6 +6015,14 @@ void ftrace_free_mem(void *start_ptr, void *end_ptr) mutex_lock(&ftrace_lock); + /* + * If we are freeing module init memory, then check if + * any tracer is active. If so, we need to save a mapping of + * the module functions being freed with the address. + */ + if (mod && ftrace_ops_list != &ftrace_list_end) + mod_map = allocate_ftrace_mod_map(mod, start, end); + for (pg = ftrace_pages_start; pg; last_pg = &pg->next, pg = *last_pg) { if (end < pg->records[0].ip || start >= (pg->records[pg->index - 1].ip + MCOUNT_INSN_SIZE)) @@ -5895,6 +6033,10 @@ void ftrace_free_mem(void *start_ptr, void *end_ptr) ftrace_cmp_recs); if (!rec) continue; + + if (mod_map) + save_ftrace_mod_rec(mod_map, rec); + pg->index--; ftrace_update_tot_cnt--; if (!pg->index) { @@ -5920,7 +6062,7 @@ void __init ftrace_free_init_mem(void) void *start = (void *)(&__init_begin); void *end = (void *)(&__init_end); - ftrace_free_mem(start, end); + ftrace_free_mem(NULL, start, end); } void __init ftrace_init(void) -- cgit v1.2.3 From 6aa69784b43eb5f69120339938c50a97a433049f Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 5 Sep 2017 19:20:16 -0400 Subject: ftrace: Add freeing algorithm to free ftrace_mod_maps The ftrace_mod_map is a descriptor to save module init function names in case they were traced, and the trace output needs to reference the function name from the function address. But after the function is unloaded, it the maps should be freed, as the rest of the function names are as well. Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 32 ++++++++++++++++++++++++++++++-- 1 file changed, 30 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 86dbbfb353db..a5824408bed9 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5683,6 +5683,7 @@ struct ftrace_mod_func { }; struct ftrace_mod_map { + struct rcu_head rcu; struct list_head list; struct module *mod; unsigned long start_addr; @@ -5694,6 +5695,8 @@ struct ftrace_mod_map { #define next_to_ftrace_page(p) container_of(p, struct ftrace_page, next) +static LIST_HEAD(ftrace_mod_maps); + static int referenced_filters(struct dyn_ftrace *rec) { struct ftrace_ops *ops; @@ -5747,8 +5750,26 @@ static void clear_mod_from_hashes(struct ftrace_page *pg) mutex_unlock(&trace_types_lock); } +static void ftrace_free_mod_map(struct rcu_head *rcu) +{ + struct ftrace_mod_map *mod_map = container_of(rcu, struct ftrace_mod_map, rcu); + struct ftrace_mod_func *mod_func; + struct ftrace_mod_func *n; + + /* All the contents of mod_map are now not visible to readers */ + list_for_each_entry_safe(mod_func, n, &mod_map->funcs, list) { + kfree(mod_func->name); + list_del(&mod_func->list); + kfree(mod_func); + } + + kfree(mod_map); +} + void ftrace_release_mod(struct module *mod) { + struct ftrace_mod_map *mod_map; + struct ftrace_mod_map *n; struct dyn_ftrace *rec; struct ftrace_page **last_pg; struct ftrace_page *tmp_page = NULL; @@ -5760,6 +5781,14 @@ void ftrace_release_mod(struct module *mod) if (ftrace_disabled) goto out_unlock; + list_for_each_entry_safe(mod_map, n, &ftrace_mod_maps, list) { + if (mod_map->mod == mod) { + list_del_rcu(&mod_map->list); + call_rcu_sched(&mod_map->rcu, ftrace_free_mod_map); + break; + } + } + /* * Each module has its own ftrace_pages, remove * them from the list. @@ -5914,8 +5943,6 @@ static void save_ftrace_mod_rec(struct ftrace_mod_map *mod_map, list_add_rcu(&mod_func->list, &mod_map->funcs); } -static LIST_HEAD(ftrace_mod_maps); - static struct ftrace_mod_map * allocate_ftrace_mod_map(struct module *mod, unsigned long start, unsigned long end) @@ -5974,6 +6001,7 @@ ftrace_mod_address_lookup(unsigned long addr, unsigned long *size, struct ftrace_mod_map *mod_map; const char *ret = NULL; + /* mod_map is freed via call_rcu_sched() */ preempt_disable(); list_for_each_entry_rcu(mod_map, &ftrace_mod_maps, list) { ret = ftrace_func_address_lookup(mod_map, addr, size, off, sym); -- cgit v1.2.3 From 6171a0310a06a7a0cb83713fa7068bdd4192de19 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Wed, 6 Sep 2017 08:40:41 -0400 Subject: ftrace/kallsyms: Have /proc/kallsyms show saved mod init functions If a module is loaded while tracing is enabled, then there's a possibility that the module init functions were traced. These functions have their name and address stored by ftrace such that it can translate the function address that is written into the buffer into a human readable function name. As userspace tools may be doing the same, they need a way to map function names to their address as well. This is done through reading /proc/kallsyms. Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index a5824408bed9..9e99bd55732e 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -5689,6 +5689,7 @@ struct ftrace_mod_map { unsigned long start_addr; unsigned long end_addr; struct list_head funcs; + unsigned int num_funcs; }; #ifdef CONFIG_MODULES @@ -5940,6 +5941,8 @@ static void save_ftrace_mod_rec(struct ftrace_mod_map *mod_map, mod_func->ip = rec->ip - offset; mod_func->size = symsize; + mod_map->num_funcs++; + list_add_rcu(&mod_func->list, &mod_map->funcs); } @@ -5956,6 +5959,7 @@ allocate_ftrace_mod_map(struct module *mod, mod_map->mod = mod; mod_map->start_addr = start; mod_map->end_addr = end; + mod_map->num_funcs = 0; INIT_LIST_HEAD_RCU(&mod_map->funcs); @@ -6016,6 +6020,42 @@ ftrace_mod_address_lookup(unsigned long addr, unsigned long *size, return ret; } +int ftrace_mod_get_kallsym(unsigned int symnum, unsigned long *value, + char *type, char *name, + char *module_name, int *exported) +{ + struct ftrace_mod_map *mod_map; + struct ftrace_mod_func *mod_func; + + preempt_disable(); + list_for_each_entry_rcu(mod_map, &ftrace_mod_maps, list) { + + if (symnum >= mod_map->num_funcs) { + symnum -= mod_map->num_funcs; + continue; + } + + list_for_each_entry_rcu(mod_func, &mod_map->funcs, list) { + if (symnum > 1) { + symnum--; + continue; + } + + *value = mod_func->ip; + *type = 'T'; + strlcpy(name, mod_func->name, KSYM_NAME_LEN); + strlcpy(module_name, mod_map->mod->name, MODULE_NAME_LEN); + *exported = 1; + preempt_enable(); + return 0; + } + WARN_ON(1); + break; + } + preempt_enable(); + return -ERANGE; +} + #else static void save_ftrace_mod_rec(struct ftrace_mod_map *mod_map, struct dyn_ftrace *rec) { } -- cgit v1.2.3 From aaecaa0b5f31794f1711247da4b5883a6ff98163 Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Thu, 5 Oct 2017 17:54:31 -0700 Subject: tracing: Prepare to add preempt and irq trace events In preparation of adding irqsoff and preemptsoff enable and disable trace events, move required functions and code to make it easier to add these events in a later patch. This patch is just code movement and no functional change. Link: http://lkml.kernel.org/r/20171006005432.14244-2-joelaf@google.com Cc: Peter Zijlstra Cc: kernel-team@android.com Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_irqsoff.c | 100 ++++++++++++++++++++++++++++++++----------- 1 file changed, 74 insertions(+), 26 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 7758bc0617cb..0e3033c00474 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -16,6 +16,7 @@ #include "trace.h" +#if defined(CONFIG_IRQSOFF_TRACER) || defined(CONFIG_PREEMPT_TRACER) static struct trace_array *irqsoff_trace __read_mostly; static int tracer_enabled __read_mostly; @@ -462,64 +463,44 @@ void time_hardirqs_off(unsigned long a0, unsigned long a1) #else /* !CONFIG_PROVE_LOCKING */ -/* - * Stubs: - */ - -void trace_softirqs_on(unsigned long ip) -{ -} - -void trace_softirqs_off(unsigned long ip) -{ -} - -inline void print_irqtrace_events(struct task_struct *curr) -{ -} - /* * We are only interested in hardirq on/off events: */ -void trace_hardirqs_on(void) +static inline void tracer_hardirqs_on(void) { if (!preempt_trace() && irq_trace()) stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1); } -EXPORT_SYMBOL(trace_hardirqs_on); -void trace_hardirqs_off(void) +static inline void tracer_hardirqs_off(void) { if (!preempt_trace() && irq_trace()) start_critical_timing(CALLER_ADDR0, CALLER_ADDR1); } -EXPORT_SYMBOL(trace_hardirqs_off); -__visible void trace_hardirqs_on_caller(unsigned long caller_addr) +static inline void tracer_hardirqs_on_caller(unsigned long caller_addr) { if (!preempt_trace() && irq_trace()) stop_critical_timing(CALLER_ADDR0, caller_addr); } -EXPORT_SYMBOL(trace_hardirqs_on_caller); -__visible void trace_hardirqs_off_caller(unsigned long caller_addr) +static inline void tracer_hardirqs_off_caller(unsigned long caller_addr) { if (!preempt_trace() && irq_trace()) start_critical_timing(CALLER_ADDR0, caller_addr); } -EXPORT_SYMBOL(trace_hardirqs_off_caller); #endif /* CONFIG_PROVE_LOCKING */ #endif /* CONFIG_IRQSOFF_TRACER */ #ifdef CONFIG_PREEMPT_TRACER -void trace_preempt_on(unsigned long a0, unsigned long a1) +static inline void tracer_preempt_on(unsigned long a0, unsigned long a1) { if (preempt_trace() && !irq_trace()) stop_critical_timing(a0, a1); } -void trace_preempt_off(unsigned long a0, unsigned long a1) +static inline void tracer_preempt_off(unsigned long a0, unsigned long a1) { if (preempt_trace() && !irq_trace()) start_critical_timing(a0, a1); @@ -781,3 +762,70 @@ __init static int init_irqsoff_tracer(void) return 0; } core_initcall(init_irqsoff_tracer); +#endif /* IRQSOFF_TRACER || PREEMPTOFF_TRACER */ + +#ifndef CONFIG_IRQSOFF_TRACER +static inline void tracer_hardirqs_on(void) { } +static inline void tracer_hardirqs_off(void) { } +static inline void tracer_hardirqs_on_caller(unsigned long caller_addr) { } +static inline void tracer_hardirqs_off_caller(unsigned long caller_addr) { } +#endif + +#ifndef CONFIG_PREEMPT_TRACER +static inline void tracer_preempt_on(unsigned long a0, unsigned long a1) { } +static inline void tracer_preempt_off(unsigned long a0, unsigned long a1) { } +#endif + +#if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PROVE_LOCKING) +void trace_hardirqs_on(void) +{ + tracer_hardirqs_on(); +} +EXPORT_SYMBOL(trace_hardirqs_on); + +void trace_hardirqs_off(void) +{ + tracer_hardirqs_off(); +} +EXPORT_SYMBOL(trace_hardirqs_off); + +__visible void trace_hardirqs_on_caller(unsigned long caller_addr) +{ + tracer_hardirqs_on_caller(caller_addr); +} +EXPORT_SYMBOL(trace_hardirqs_on_caller); + +__visible void trace_hardirqs_off_caller(unsigned long caller_addr) +{ + tracer_hardirqs_off_caller(caller_addr); +} +EXPORT_SYMBOL(trace_hardirqs_off_caller); + +/* + * Stubs: + */ + +void trace_softirqs_on(unsigned long ip) +{ +} + +void trace_softirqs_off(unsigned long ip) +{ +} + +inline void print_irqtrace_events(struct task_struct *curr) +{ +} +#endif + +#ifdef CONFIG_PREEMPT_TRACER +void trace_preempt_on(unsigned long a0, unsigned long a1) +{ + tracer_preempt_on(a0, a1); +} + +void trace_preempt_off(unsigned long a0, unsigned long a1) +{ + tracer_preempt_off(a0, a1); +} +#endif -- cgit v1.2.3 From 97562633bcbac4a07d605ae628d7655fa71caaf5 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 5 Oct 2017 09:19:19 -0700 Subject: bpf: perf event change needed for subsequent bpf helpers This patch does not impact existing functionalities. It contains the changes in perf event area needed for subsequent bpf_perf_event_read_value and bpf_perf_prog_read_value helpers. Signed-off-by: Yonghong Song Acked-by: Peter Zijlstra (Intel) Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index dc498b605d5d..95888ae6c263 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -275,7 +275,7 @@ BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) if (!ee) return -ENOENT; - err = perf_event_read_local(ee->event, &value); + err = perf_event_read_local(ee->event, &value, NULL, NULL); /* * this api is ugly since we miss [-22..-2] range of valid * counter values, but that's uapi -- cgit v1.2.3 From 908432ca84fc229e906ba164219e9ad0fe56f755 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 5 Oct 2017 09:19:20 -0700 Subject: bpf: add helper bpf_perf_event_read_value for perf event array map Hardware pmu counters are limited resources. When there are more pmu based perf events opened than available counters, kernel will multiplex these events so each event gets certain percentage (but not 100%) of the pmu time. In case that multiplexing happens, the number of samples or counter value will not reflect the case compared to no multiplexing. This makes comparison between different runs difficult. Typically, the number of samples or counter value should be normalized before comparing to other experiments. The typical normalization is done like: normalized_num_samples = num_samples * time_enabled / time_running normalized_counter_value = counter_value * time_enabled / time_running where time_enabled is the time enabled for event and time_running is the time running for event since last normalization. This patch adds helper bpf_perf_event_read_value for kprobed based perf event array map, to read perf counter and enabled/running time. The enabled/running time is accumulated since the perf event open. To achieve scaling factor between two bpf invocations, users can can use cpu_id as the key (which is typical for perf array usage model) to remember the previous value and do the calculation inside the bpf program. Signed-off-by: Yonghong Song Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 45 +++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 41 insertions(+), 4 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 95888ae6c263..0be86cc0130e 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -255,14 +255,14 @@ const struct bpf_func_proto *bpf_get_trace_printk_proto(void) return &bpf_trace_printk_proto; } -BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) +static __always_inline int +get_map_perf_counter(struct bpf_map *map, u64 flags, + u64 *value, u64 *enabled, u64 *running) { struct bpf_array *array = container_of(map, struct bpf_array, map); unsigned int cpu = smp_processor_id(); u64 index = flags & BPF_F_INDEX_MASK; struct bpf_event_entry *ee; - u64 value = 0; - int err; if (unlikely(flags & ~(BPF_F_INDEX_MASK))) return -EINVAL; @@ -275,7 +275,15 @@ BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) if (!ee) return -ENOENT; - err = perf_event_read_local(ee->event, &value, NULL, NULL); + return perf_event_read_local(ee->event, value, enabled, running); +} + +BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) +{ + u64 value = 0; + int err; + + err = get_map_perf_counter(map, flags, &value, NULL, NULL); /* * this api is ugly since we miss [-22..-2] range of valid * counter values, but that's uapi @@ -293,6 +301,33 @@ static const struct bpf_func_proto bpf_perf_event_read_proto = { .arg2_type = ARG_ANYTHING, }; +BPF_CALL_4(bpf_perf_event_read_value, struct bpf_map *, map, u64, flags, + struct bpf_perf_event_value *, buf, u32, size) +{ + int err = -EINVAL; + + if (unlikely(size != sizeof(struct bpf_perf_event_value))) + goto clear; + err = get_map_perf_counter(map, flags, &buf->counter, &buf->enabled, + &buf->running); + if (unlikely(err)) + goto clear; + return 0; +clear: + memset(buf, 0, size); + return err; +} + +static const struct bpf_func_proto bpf_perf_event_read_value_proto = { + .func = bpf_perf_event_read_value, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_CONST_MAP_PTR, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_PTR_TO_UNINIT_MEM, + .arg4_type = ARG_CONST_SIZE, +}; + static DEFINE_PER_CPU(struct perf_sample_data, bpf_sd); static __always_inline u64 @@ -499,6 +534,8 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func return &bpf_perf_event_output_proto; case BPF_FUNC_get_stackid: return &bpf_get_stackid_proto; + case BPF_FUNC_perf_event_read_value: + return &bpf_perf_event_read_value_proto; default: return tracing_func_proto(func_id); } -- cgit v1.2.3 From 4bebdc7a85aa400c0222b5329861e4ad9252f1e5 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 5 Oct 2017 09:19:22 -0700 Subject: bpf: add helper bpf_perf_prog_read_value This patch adds helper bpf_perf_prog_read_cvalue for perf event based bpf programs, to read event counter and enabled/running time. The enabled/running time is accumulated since the perf event open. The typical use case for perf event based bpf program is to attach itself to a single event. In such cases, if it is desirable to get scaling factor between two bpf invocations, users can can save the time values in a map, and use the value from the map and the current value to calculate the scaling factor. Signed-off-by: Yonghong Song Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 0be86cc0130e..04ea5314f2bc 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -613,6 +613,32 @@ static const struct bpf_func_proto bpf_get_stackid_proto_tp = { .arg3_type = ARG_ANYTHING, }; +BPF_CALL_3(bpf_perf_prog_read_value_tp, struct bpf_perf_event_data_kern *, ctx, + struct bpf_perf_event_value *, buf, u32, size) +{ + int err = -EINVAL; + + if (unlikely(size != sizeof(struct bpf_perf_event_value))) + goto clear; + err = perf_event_read_local(ctx->event, &buf->counter, &buf->enabled, + &buf->running); + if (unlikely(err)) + goto clear; + return 0; +clear: + memset(buf, 0, size); + return err; +} + +static const struct bpf_func_proto bpf_perf_prog_read_value_proto_tp = { + .func = bpf_perf_prog_read_value_tp, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_PTR_TO_UNINIT_MEM, + .arg3_type = ARG_CONST_SIZE, +}; + static const struct bpf_func_proto *tp_prog_func_proto(enum bpf_func_id func_id) { switch (func_id) { @@ -620,6 +646,8 @@ static const struct bpf_func_proto *tp_prog_func_proto(enum bpf_func_id func_id) return &bpf_perf_event_output_proto_tp; case BPF_FUNC_get_stackid: return &bpf_get_stackid_proto_tp; + case BPF_FUNC_perf_prog_read_value: + return &bpf_perf_prog_read_value_proto_tp; default: return tracing_func_proto(func_id); } -- cgit v1.2.3 From 1d48b080bcce0a5e7d7aa2dbcdb35deefc188c3f Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 29 Sep 2017 13:50:16 +0200 Subject: sched/debug: Rename task-state printing helpers Steve requested better names for the new task-state helper functions. So introduce the concept of task-state index for the printing and rename __get_task_state() to task_state_index() and __task_state_to_char() to task_index_to_char(). Requested-by: Steven Rostedt Signed-off-by: Peter Zijlstra (Intel) Acked-by: Steven Rostedt Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20170929115016.pzlqc7ss3ccystyg@hirez.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- kernel/trace/trace_output.c | 12 ++++++------ kernel/trace/trace_sched_wakeup.c | 8 ++++---- 2 files changed, 10 insertions(+), 10 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index c738e764e2a5..90db994ac900 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -921,8 +921,8 @@ static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter, trace_assign_type(field, iter->ent); - T = __task_state_to_char(field->next_state); - S = __task_state_to_char(field->prev_state); + T = task_index_to_char(field->next_state); + S = task_index_to_char(field->prev_state); trace_find_cmdline(field->next_pid, comm); trace_seq_printf(&iter->seq, " %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n", @@ -957,8 +957,8 @@ static int trace_ctxwake_raw(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - S = __task_state_to_char(field->prev_state); - T = __task_state_to_char(field->next_state); + S = task_index_to_char(field->prev_state); + T = task_index_to_char(field->next_state); trace_seq_printf(&iter->seq, "%d %d %c %d %d %d %c\n", field->prev_pid, field->prev_prio, @@ -993,8 +993,8 @@ static int trace_ctxwake_hex(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - S = __task_state_to_char(field->prev_state); - T = __task_state_to_char(field->next_state); + S = task_index_to_char(field->prev_state); + T = task_index_to_char(field->next_state); SEQ_PUT_HEX_FIELD(s, field->prev_pid); SEQ_PUT_HEX_FIELD(s, field->prev_prio); diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 0c331978b1a6..500f370d3bb1 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -397,10 +397,10 @@ tracing_sched_switch_trace(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->prev_pid = prev->pid; entry->prev_prio = prev->prio; - entry->prev_state = __get_task_state(prev); + entry->prev_state = task_state_index(prev); entry->next_pid = next->pid; entry->next_prio = next->prio; - entry->next_state = __get_task_state(next); + entry->next_state = task_state_index(next); entry->next_cpu = task_cpu(next); if (!call_filter_check_discard(call, entry, buffer, event)) @@ -425,10 +425,10 @@ tracing_sched_wakeup_trace(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->prev_pid = curr->pid; entry->prev_prio = curr->prio; - entry->prev_state = __get_task_state(curr); + entry->prev_state = task_state_index(curr); entry->next_pid = wakee->pid; entry->next_prio = wakee->prio; - entry->next_state = __get_task_state(wakee); + entry->next_state = task_state_index(wakee); entry->next_cpu = task_cpu(wakee); if (!call_filter_check_discard(call, entry, buffer, event)) -- cgit v1.2.3 From d59158162e032917a428704160a2063a02405ec6 Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Tue, 10 Oct 2017 15:51:37 -0700 Subject: tracing: Add support for preempt and irq enable/disable events Preempt and irq trace events can be used for tracing the start and end of an atomic section which can be used by a trace viewer like systrace to graphically view the start and end of an atomic section and correlate them with latencies and scheduling issues. This also serves as a prelude to using synthetic events or probes to rewrite the preempt and irqsoff tracers, along with numerous benefits of using trace events features for these events. Link: http://lkml.kernel.org/r/20171006005432.14244-3-joelaf@google.com Link: http://lkml.kernel.org/r/20171010225137.17370-1-joelaf@google.com Cc: Peter Zilstra Cc: kernel-team@android.com Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/Kconfig | 11 +++++++++++ kernel/trace/Makefile | 1 + kernel/trace/trace_irqsoff.c | 35 ++++++++++++++++++++++++++++++++++- 3 files changed, 46 insertions(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 434c840e2d82..b8395a020821 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -160,6 +160,17 @@ config FUNCTION_GRAPH_TRACER address on the current task structure into a stack of calls. +config PREEMPTIRQ_EVENTS + bool "Enable trace events for preempt and irq disable/enable" + select TRACE_IRQFLAGS + depends on DEBUG_PREEMPT || !PROVE_LOCKING + default n + help + Enable tracing of disable and enable events for preemption and irqs. + For tracing preempt disable/enable events, DEBUG_PREEMPT must be + enabled. For tracing irq disable/enable events, PROVE_LOCKING must + be disabled. + config IRQSOFF_TRACER bool "Interrupts-off Latency Tracer" default n diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 90f2701d92a7..9f62eee61f14 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -34,6 +34,7 @@ obj-$(CONFIG_TRACING) += trace_printk.o obj-$(CONFIG_TRACING_MAP) += tracing_map.o obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o +obj-$(CONFIG_PREEMPTIRQ_EVENTS) += trace_irqsoff.o obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 0e3033c00474..03ecb4465ee4 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -16,6 +16,9 @@ #include "trace.h" +#define CREATE_TRACE_POINTS +#include + #if defined(CONFIG_IRQSOFF_TRACER) || defined(CONFIG_PREEMPT_TRACER) static struct trace_array *irqsoff_trace __read_mostly; static int tracer_enabled __read_mostly; @@ -777,26 +780,53 @@ static inline void tracer_preempt_off(unsigned long a0, unsigned long a1) { } #endif #if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PROVE_LOCKING) +/* Per-cpu variable to prevent redundant calls when IRQs already off */ +static DEFINE_PER_CPU(int, tracing_irq_cpu); + void trace_hardirqs_on(void) { + if (!this_cpu_read(tracing_irq_cpu)) + return; + + trace_irq_enable_rcuidle(CALLER_ADDR0, CALLER_ADDR1); tracer_hardirqs_on(); + + this_cpu_write(tracing_irq_cpu, 0); } EXPORT_SYMBOL(trace_hardirqs_on); void trace_hardirqs_off(void) { + if (this_cpu_read(tracing_irq_cpu)) + return; + + this_cpu_write(tracing_irq_cpu, 1); + + trace_irq_disable_rcuidle(CALLER_ADDR0, CALLER_ADDR1); tracer_hardirqs_off(); } EXPORT_SYMBOL(trace_hardirqs_off); __visible void trace_hardirqs_on_caller(unsigned long caller_addr) { + if (!this_cpu_read(tracing_irq_cpu)) + return; + + trace_irq_enable_rcuidle(CALLER_ADDR0, caller_addr); tracer_hardirqs_on_caller(caller_addr); + + this_cpu_write(tracing_irq_cpu, 0); } EXPORT_SYMBOL(trace_hardirqs_on_caller); __visible void trace_hardirqs_off_caller(unsigned long caller_addr) { + if (this_cpu_read(tracing_irq_cpu)) + return; + + this_cpu_write(tracing_irq_cpu, 1); + + trace_irq_disable_rcuidle(CALLER_ADDR0, caller_addr); tracer_hardirqs_off_caller(caller_addr); } EXPORT_SYMBOL(trace_hardirqs_off_caller); @@ -818,14 +848,17 @@ inline void print_irqtrace_events(struct task_struct *curr) } #endif -#ifdef CONFIG_PREEMPT_TRACER +#if defined(CONFIG_PREEMPT_TRACER) || \ + (defined(CONFIG_DEBUG_PREEMPT) && defined(CONFIG_PREEMPTIRQ_EVENTS)) void trace_preempt_on(unsigned long a0, unsigned long a1) { + trace_preempt_enable_rcuidle(a0, a1); tracer_preempt_on(a0, a1); } void trace_preempt_off(unsigned long a0, unsigned long a1) { + trace_preempt_disable_rcuidle(a0, a1); tracer_preempt_off(a0, a1); } #endif -- cgit v1.2.3 From 8715b108cd75523c9b2e833cdcd7aeb363767f95 Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Mon, 9 Oct 2017 12:29:31 -0700 Subject: ftrace: Clear hashes of stale ips of init memory Filters should be cleared of init functions during freeing of init memory when the ftrace dyn records are released. However in current code, the filters are left as is. This patch clears the hashes of the saved init functions when the init memory is freed. This fixes the following issue reproducible with the following sequence of commands for a test module: ================================================ void bar(void) { printk(KERN_INFO "bar!\n"); } void foo(void) { printk(KERN_INFO "foo!\n"); bar(); } static int __init hello_init(void) { printk(KERN_INFO "Hello world!\n"); foo(); return 0; } static void __exit hello_cleanup(void) { printk(KERN_INFO "Cleaning up module.\n"); } module_init(hello_init); module_exit(hello_cleanup); ================================================ Commands: echo '*:mod:test' > /d/tracing/set_ftrace_filter echo function > /d/tracing/current_tracer modprobe test rmmod test sleep 1 modprobe test cat /d/tracing/set_ftrace_filter Behavior without patch: Init function is still in the filter Expected behavior: Shouldn't have any of the filters set Link: http://lkml.kernel.org/r/20171009192931.56401-1-joelaf@google.com Signed-off-by: Joel Fernandes Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 69 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 69 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 9e99bd55732e..e0a98225666b 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -6067,6 +6067,63 @@ allocate_ftrace_mod_map(struct module *mod, } #endif /* CONFIG_MODULES */ +struct ftrace_init_func { + struct list_head list; + unsigned long ip; +}; + +/* Clear any init ips from hashes */ +static void +clear_func_from_hash(struct ftrace_init_func *func, struct ftrace_hash *hash) +{ + struct ftrace_func_entry *entry; + + if (ftrace_hash_empty(hash)) + return; + + entry = __ftrace_lookup_ip(hash, func->ip); + + /* + * Do not allow this rec to match again. + * Yeah, it may waste some memory, but will be removed + * if/when the hash is modified again. + */ + if (entry) + entry->ip = 0; +} + +static void +clear_func_from_hashes(struct ftrace_init_func *func) +{ + struct trace_array *tr; + + mutex_lock(&trace_types_lock); + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (!tr->ops || !tr->ops->func_hash) + continue; + mutex_lock(&tr->ops->func_hash->regex_lock); + clear_func_from_hash(func, tr->ops->func_hash->filter_hash); + clear_func_from_hash(func, tr->ops->func_hash->notrace_hash); + mutex_unlock(&tr->ops->func_hash->regex_lock); + } + mutex_unlock(&trace_types_lock); +} + +static void add_to_clear_hash_list(struct list_head *clear_list, + struct dyn_ftrace *rec) +{ + struct ftrace_init_func *func; + + func = kmalloc(sizeof(*func), GFP_KERNEL); + if (!func) { + WARN_ONCE(1, "alloc failure, ftrace filter could be stale\n"); + return; + } + + func->ip = rec->ip; + list_add(&func->list, clear_list); +} + void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) { unsigned long start = (unsigned long)(start_ptr); @@ -6076,8 +6133,12 @@ void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) struct dyn_ftrace *rec; struct dyn_ftrace key; struct ftrace_mod_map *mod_map = NULL; + struct ftrace_init_func *func, *func_next; + struct list_head clear_hash; int order; + INIT_LIST_HEAD(&clear_hash); + key.ip = start; key.flags = end; /* overload flags, as it is unsigned long */ @@ -6102,6 +6163,9 @@ void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) if (!rec) continue; + /* rec will be cleared from hashes after ftrace_lock unlock */ + add_to_clear_hash_list(&clear_hash, rec); + if (mod_map) save_ftrace_mod_rec(mod_map, rec); @@ -6123,6 +6187,11 @@ void ftrace_free_mem(struct module *mod, void *start_ptr, void *end_ptr) goto again; } mutex_unlock(&ftrace_lock); + + list_for_each_entry_safe(func, func_next, &clear_hash, list) { + clear_func_from_hashes(func); + kfree(func); + } } void __init ftrace_free_init_mem(void) -- cgit v1.2.3 From c5c1ea75a352c3864c4891f36155287a2ed73928 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Tue, 13 Jun 2017 13:06:59 +0200 Subject: tracing: Kconfig text fixes for CONFIG_HWLAT_TRACER Trivial spelling fixes for Kconfig help text of config HWLAT_TRACER. Fixes: e7c15cd8a113 ("tracing: Added hardware latency tracer") Signed-off-by: Jesper Dangaard Brouer Acked-by: Steven Rostedt Signed-off-by: Jiri Kosina --- kernel/trace/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 434c840e2d82..f54b7b6b4a4b 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -224,7 +224,7 @@ config HWLAT_TRACER select GENERIC_TRACER help This tracer, when enabled will create one or more kernel threads, - depening on what the cpumask file is set to, which each thread + depending on what the cpumask file is set to, which each thread spinning in a loop looking for interruptions caused by something other than the kernel. For example, if a System Management Interrupt (SMI) takes a noticeable amount of @@ -239,7 +239,7 @@ config HWLAT_TRACER iteration A kernel thread is created that will spin with interrupts disabled - for "width" microseconds in every "widow" cycle. It will not spin + for "width" microseconds in every "window" cycle. It will not spin for "window - width" microseconds, where the system can continue to operate. -- cgit v1.2.3 From c3b5b6ed1eb4f429addfd9e8e8eb39d1a38c85d0 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 13 Oct 2017 16:22:20 +0200 Subject: tracing: mark trace_test_buffer as __maybe_unused After trace_selftest_startup_sched_switch is removed, trace_test_buffer() is only used sometimes, leading to this warning: kernel/trace/trace_selftest.c:62:12: error: 'trace_test_buffer' defined but not used [-Werror=unused-function] There is no simple #ifdef condition that captures well whether the function is in fact used or not, so marking it as __maybe_unused is probably the best way to shut up the warning. The function will then be silently dropped when there is no user. Link: http://lkml.kernel.org/r/20171013142227.1273469-1-arnd@arndb.de Fixes: d8c4deee6dc6 ("tracing: Remove obsolete sched_switch tracer selftest") Signed-off-by: Arnd Bergmann Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_selftest.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 364f78abdf47..eb9ba5c1ba40 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -59,7 +59,7 @@ static int trace_test_buffer_cpu(struct trace_buffer *buf, int cpu) * Test the trace buffer to see if all the elements * are still sane. */ -static int trace_test_buffer(struct trace_buffer *buf, unsigned long *count) +static int __maybe_unused trace_test_buffer(struct trace_buffer *buf, unsigned long *count) { unsigned long flags, cnt = 0; int cpu, ret = 0; -- cgit v1.2.3 From 8fd0fbbe8888f295eb34172a7e47bf7d3a0a4687 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Oct 2017 09:45:29 +0200 Subject: perf/ftrace: Revert ("perf/ftrace: Fix double traces of perf on ftrace:function") Revert commit: 75e8387685f6 ("perf/ftrace: Fix double traces of perf on ftrace:function") The reason I instantly stumbled on that patch is that it only addresses the ftrace situation and doesn't mention the other _5_ places that use this interface. It doesn't explain why those don't have the problem and if not, why their solution doesn't work for ftrace. It doesn't, but this is just putting more duct tape on. Link: http://lkml.kernel.org/r/20171011080224.200565770@infradead.org Cc: Zhou Chengming Cc: Jiri Olsa Cc: Ingo Molnar Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_event_perf.c | 4 +--- kernel/trace/trace_kprobe.c | 4 ++-- kernel/trace/trace_syscalls.c | 4 ++-- kernel/trace/trace_uprobe.c | 2 +- 4 files changed, 6 insertions(+), 8 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 13ba2d3f6a91..562fa69df5d3 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -306,7 +306,6 @@ static void perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *ops, struct pt_regs *pt_regs) { - struct perf_event *event; struct ftrace_entry *entry; struct hlist_head *head; struct pt_regs regs; @@ -330,9 +329,8 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, entry->ip = ip; entry->parent_ip = parent_ip; - event = container_of(ops, struct perf_event, ftrace_ops); perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, TRACE_FN, - 1, ®s, head, NULL, event); + 1, ®s, head, NULL); #undef ENTRY_SIZE } diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index af6134f2e597..996902a526d4 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1200,7 +1200,7 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) memset(&entry[1], 0, dsize); store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, - head, NULL, NULL); + head, NULL); } NOKPROBE_SYMBOL(kprobe_perf_func); @@ -1236,7 +1236,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, entry->ret_ip = (unsigned long)ri->ret_addr; store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, - head, NULL, NULL); + head, NULL); } NOKPROBE_SYMBOL(kretprobe_perf_func); #endif /* CONFIG_PERF_EVENTS */ diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 696afe72d3b1..934b0da72679 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -622,7 +622,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) perf_trace_buf_submit(rec, size, rctx, sys_data->enter_event->event.type, 1, regs, - head, NULL, NULL); + head, NULL); } static int perf_sysenter_enable(struct trace_event_call *call) @@ -716,7 +716,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) } perf_trace_buf_submit(rec, size, rctx, sys_data->exit_event->event.type, - 1, regs, head, NULL, NULL); + 1, regs, head, NULL); } static int perf_sysexit_enable(struct trace_event_call *call) diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index b34965e62fb9..402120ba4594 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -1156,7 +1156,7 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, } perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, - head, NULL, NULL); + head, NULL); out: preempt_enable(); } -- cgit v1.2.3 From 466c81c45b650deca213fda3d0ec4761667379a9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 10 Oct 2017 17:15:47 +0200 Subject: perf/ftrace: Fix function trace events The function-trace <-> perf interface is a tad messed up. Where all the other trace <-> perf interfaces use a single trace hook registration and use per-cpu RCU based hlist to iterate the events, function-trace actually needs multiple hook registrations in order to minimize function entry patching when filters are present. The end result is that we iterate events both on the trace hook and on the hlist, which results in reporting events multiple times. Since function-trace cannot use the regular scheme, fix it the other way around, use singleton hlists. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_event_perf.c | 80 +++++++++++++++++++++++++---------------- 1 file changed, 49 insertions(+), 31 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 562fa69df5d3..e73f9ab15939 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -240,27 +240,41 @@ void perf_trace_destroy(struct perf_event *p_event) int perf_trace_add(struct perf_event *p_event, int flags) { struct trace_event_call *tp_event = p_event->tp_event; - struct hlist_head __percpu *pcpu_list; - struct hlist_head *list; - - pcpu_list = tp_event->perf_events; - if (WARN_ON_ONCE(!pcpu_list)) - return -EINVAL; if (!(flags & PERF_EF_START)) p_event->hw.state = PERF_HES_STOPPED; - list = this_cpu_ptr(pcpu_list); - hlist_add_head_rcu(&p_event->hlist_entry, list); + /* + * If TRACE_REG_PERF_ADD returns false; no custom action was performed + * and we need to take the default action of enqueueing our event on + * the right per-cpu hlist. + */ + if (!tp_event->class->reg(tp_event, TRACE_REG_PERF_ADD, p_event)) { + struct hlist_head __percpu *pcpu_list; + struct hlist_head *list; + + pcpu_list = tp_event->perf_events; + if (WARN_ON_ONCE(!pcpu_list)) + return -EINVAL; + + list = this_cpu_ptr(pcpu_list); + hlist_add_head_rcu(&p_event->hlist_entry, list); + } - return tp_event->class->reg(tp_event, TRACE_REG_PERF_ADD, p_event); + return 0; } void perf_trace_del(struct perf_event *p_event, int flags) { struct trace_event_call *tp_event = p_event->tp_event; - hlist_del_rcu(&p_event->hlist_entry); - tp_event->class->reg(tp_event, TRACE_REG_PERF_DEL, p_event); + + /* + * If TRACE_REG_PERF_DEL returns false; no custom action was performed + * and we need to take the default action of dequeueing our event from + * the right per-cpu hlist. + */ + if (!tp_event->class->reg(tp_event, TRACE_REG_PERF_DEL, p_event)) + hlist_del_rcu(&p_event->hlist_entry); } void *perf_trace_buf_alloc(int size, struct pt_regs **regs, int *rctxp) @@ -307,14 +321,24 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *ops, struct pt_regs *pt_regs) { struct ftrace_entry *entry; - struct hlist_head *head; + struct perf_event *event; + struct hlist_head head; struct pt_regs regs; int rctx; - head = this_cpu_ptr(event_function.perf_events); - if (hlist_empty(head)) + if ((unsigned long)ops->private != smp_processor_id()) return; + event = container_of(ops, struct perf_event, ftrace_ops); + + /* + * @event->hlist entry is NULL (per INIT_HLIST_NODE), and all + * the perf code does is hlist_for_each_entry_rcu(), so we can + * get away with simply setting the @head.first pointer in order + * to create a singular list. + */ + head.first = &event->hlist_entry; + #define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \ sizeof(u64)) - sizeof(u32)) @@ -330,7 +354,7 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, entry->ip = ip; entry->parent_ip = parent_ip; perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, TRACE_FN, - 1, ®s, head, NULL); + 1, ®s, &head, NULL); #undef ENTRY_SIZE } @@ -339,8 +363,10 @@ static int perf_ftrace_function_register(struct perf_event *event) { struct ftrace_ops *ops = &event->ftrace_ops; - ops->flags |= FTRACE_OPS_FL_PER_CPU | FTRACE_OPS_FL_RCU; - ops->func = perf_ftrace_function_call; + ops->flags |= FTRACE_OPS_FL_RCU; + ops->func = perf_ftrace_function_call; + ops->private = (void *)(unsigned long)nr_cpu_ids; + return register_ftrace_function(ops); } @@ -352,19 +378,11 @@ static int perf_ftrace_function_unregister(struct perf_event *event) return ret; } -static void perf_ftrace_function_enable(struct perf_event *event) -{ - ftrace_function_local_enable(&event->ftrace_ops); -} - -static void perf_ftrace_function_disable(struct perf_event *event) -{ - ftrace_function_local_disable(&event->ftrace_ops); -} - int perf_ftrace_event_register(struct trace_event_call *call, enum trace_reg type, void *data) { + struct perf_event *event = data; + switch (type) { case TRACE_REG_REGISTER: case TRACE_REG_UNREGISTER: @@ -377,11 +395,11 @@ int perf_ftrace_event_register(struct trace_event_call *call, case TRACE_REG_PERF_CLOSE: return perf_ftrace_function_unregister(data); case TRACE_REG_PERF_ADD: - perf_ftrace_function_enable(data); - return 0; + event->ftrace_ops.private = (void *)(unsigned long)smp_processor_id(); + return 1; case TRACE_REG_PERF_DEL: - perf_ftrace_function_disable(data); - return 0; + event->ftrace_ops.private = (void *)(unsigned long)nr_cpu_ids; + return 1; } return -EINVAL; -- cgit v1.2.3 From 1dd311e6dcda4020c603bcf9f390a577d439d509 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Oct 2017 09:45:31 +0200 Subject: perf/ftrace: Small cleanup ops->flags _should_ be 0 at this point, so setting the flag using bitwise or is a bit daft. Link: http://lkml.kernel.org/r/20171011080224.315585202@infradead.org Requested-by: Steven Rostedt Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_event_perf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index e73f9ab15939..55d6dff37daf 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -363,7 +363,7 @@ static int perf_ftrace_function_register(struct perf_event *event) { struct ftrace_ops *ops = &event->ftrace_ops; - ops->flags |= FTRACE_OPS_FL_RCU; + ops->flags = FTRACE_OPS_FL_RCU; ops->func = perf_ftrace_function_call; ops->private = (void *)(unsigned long)nr_cpu_ids; -- cgit v1.2.3 From b3a88803ac5b4bda26017b485c8722a8487fefb7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Oct 2017 09:45:32 +0200 Subject: ftrace: Kill FTRACE_OPS_FL_PER_CPU The one and only user of FTRACE_OPS_FL_PER_CPU is gone, remove the lot. Link: http://lkml.kernel.org/r/20171011080224.372422809@infradead.org Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/ftrace.c | 55 ++++++--------------------------------------------- 1 file changed, 6 insertions(+), 49 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index e0a98225666b..2fd3edaec6de 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -203,30 +203,6 @@ void clear_ftrace_function(void) ftrace_trace_function = ftrace_stub; } -static void per_cpu_ops_disable_all(struct ftrace_ops *ops) -{ - int cpu; - - for_each_possible_cpu(cpu) - *per_cpu_ptr(ops->disabled, cpu) = 1; -} - -static int per_cpu_ops_alloc(struct ftrace_ops *ops) -{ - int __percpu *disabled; - - if (WARN_ON_ONCE(!(ops->flags & FTRACE_OPS_FL_PER_CPU))) - return -EINVAL; - - disabled = alloc_percpu(int); - if (!disabled) - return -ENOMEM; - - ops->disabled = disabled; - per_cpu_ops_disable_all(ops); - return 0; -} - static void ftrace_sync(struct work_struct *work) { /* @@ -262,8 +238,8 @@ static ftrace_func_t ftrace_ops_get_list_func(struct ftrace_ops *ops) * If this is a dynamic, RCU, or per CPU ops, or we force list func, * then it needs to call the list anyway. */ - if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_PER_CPU | - FTRACE_OPS_FL_RCU) || FTRACE_FORCE_LIST_FUNC) + if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_RCU) || + FTRACE_FORCE_LIST_FUNC) return ftrace_ops_list_func; return ftrace_ops_get_func(ops); @@ -422,11 +398,6 @@ static int __register_ftrace_function(struct ftrace_ops *ops) if (!core_kernel_data((unsigned long)ops)) ops->flags |= FTRACE_OPS_FL_DYNAMIC; - if (ops->flags & FTRACE_OPS_FL_PER_CPU) { - if (per_cpu_ops_alloc(ops)) - return -ENOMEM; - } - add_ftrace_ops(&ftrace_ops_list, ops); /* Always save the function, and reset at unregistering */ @@ -2727,11 +2698,6 @@ void __weak arch_ftrace_trampoline_free(struct ftrace_ops *ops) { } -static void per_cpu_ops_free(struct ftrace_ops *ops) -{ - free_percpu(ops->disabled); -} - static void ftrace_startup_enable(int command) { if (saved_ftrace_func != ftrace_trace_function) { @@ -2833,7 +2799,7 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) * not currently active, we can just free them * without synchronizing all CPUs. */ - if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_PER_CPU)) + if (ops->flags & FTRACE_OPS_FL_DYNAMIC) goto free_ops; return 0; @@ -2880,7 +2846,7 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) * The same goes for freeing the per_cpu data of the per_cpu * ops. */ - if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_PER_CPU)) { + if (ops->flags & FTRACE_OPS_FL_DYNAMIC) { /* * We need to do a hard force of sched synchronization. * This is because we use preempt_disable() to do RCU, but @@ -2903,9 +2869,6 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) free_ops: arch_ftrace_trampoline_free(ops); - - if (ops->flags & FTRACE_OPS_FL_PER_CPU) - per_cpu_ops_free(ops); } return 0; @@ -6355,10 +6318,7 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip, * If any of the above fails then the op->func() is not executed. */ if ((!(op->flags & FTRACE_OPS_FL_RCU) || rcu_is_watching()) && - (!(op->flags & FTRACE_OPS_FL_PER_CPU) || - !ftrace_function_local_disabled(op)) && ftrace_ops_test(op, ip, regs)) { - if (FTRACE_WARN_ON(!op->func)) { pr_warn("op=%p %pS\n", op, op); goto out; @@ -6416,10 +6376,7 @@ static void ftrace_ops_assist_func(unsigned long ip, unsigned long parent_ip, preempt_disable_notrace(); - if (!(op->flags & FTRACE_OPS_FL_PER_CPU) || - !ftrace_function_local_disabled(op)) { - op->func(ip, parent_ip, op, regs); - } + op->func(ip, parent_ip, op, regs); preempt_enable_notrace(); trace_clear_recursion(bit); @@ -6443,7 +6400,7 @@ ftrace_func_t ftrace_ops_get_func(struct ftrace_ops *ops) * or does per cpu logic, then we need to call the assist handler. */ if (!(ops->flags & FTRACE_OPS_FL_RECURSION_SAFE) || - ops->flags & (FTRACE_OPS_FL_RCU | FTRACE_OPS_FL_PER_CPU)) + ops->flags & FTRACE_OPS_FL_RCU) return ftrace_ops_assist_func; return ops->func; -- cgit v1.2.3 From 7de16e3a35578f4f5accc6f5f23970310483d0a2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 16 Oct 2017 16:40:53 -0700 Subject: bpf: split verifier and program ops struct bpf_verifier_ops contains both verifier ops and operations used later during program's lifetime (test_run). Split the runtime ops into a different structure. BPF_PROG_TYPE() will now append ## _prog_ops or ## _verifier_ops to the names. Signed-off-by: Jakub Kicinski Acked-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 04ea5314f2bc..3126da2f468a 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -561,11 +561,14 @@ static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type return true; } -const struct bpf_verifier_ops kprobe_prog_ops = { +const struct bpf_verifier_ops kprobe_verifier_ops = { .get_func_proto = kprobe_prog_func_proto, .is_valid_access = kprobe_prog_is_valid_access, }; +const struct bpf_prog_ops kprobe_prog_ops = { +}; + BPF_CALL_5(bpf_perf_event_output_tp, void *, tp_buff, struct bpf_map *, map, u64, flags, void *, data, u64, size) { @@ -667,11 +670,14 @@ static bool tp_prog_is_valid_access(int off, int size, enum bpf_access_type type return true; } -const struct bpf_verifier_ops tracepoint_prog_ops = { +const struct bpf_verifier_ops tracepoint_verifier_ops = { .get_func_proto = tp_prog_func_proto, .is_valid_access = tp_prog_is_valid_access, }; +const struct bpf_prog_ops tracepoint_prog_ops = { +}; + static bool pe_prog_is_valid_access(int off, int size, enum bpf_access_type type, struct bpf_insn_access_aux *info) { @@ -727,8 +733,11 @@ static u32 pe_prog_convert_ctx_access(enum bpf_access_type type, return insn - insn_buf; } -const struct bpf_verifier_ops perf_event_prog_ops = { +const struct bpf_verifier_ops perf_event_verifier_ops = { .get_func_proto = tp_prog_func_proto, .is_valid_access = pe_prog_is_valid_access, .convert_ctx_access = pe_prog_convert_ctx_access, }; + +const struct bpf_prog_ops perf_event_prog_ops = { +}; -- cgit v1.2.3 From e87c6bc3852b981e71c757be20771546ce9f76f3 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Mon, 23 Oct 2017 23:53:08 -0700 Subject: bpf: permit multiple bpf attachments for a single perf event This patch enables multiple bpf attachments for a kprobe/uprobe/tracepoint single trace event. Each trace_event keeps a list of attached perf events. When an event happens, all attached bpf programs will be executed based on the order of attachment. A global bpf_event_mutex lock is introduced to protect prog_array attaching and detaching. An alternative will be introduce a mutex lock in every trace_event_call structure, but it takes a lot of extra memory. So a global bpf_event_mutex lock is a good compromise. The bpf prog detachment involves allocation of memory. If the allocation fails, a dummy do-nothing program will replace to-be-detached program in-place. Signed-off-by: Yonghong Song Acked-by: Alexei Starovoitov Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 82 ++++++++++++++++++++++++++++++++++++++++--- kernel/trace/trace_kprobe.c | 6 ++-- kernel/trace/trace_syscalls.c | 34 ++++++++++-------- kernel/trace/trace_uprobe.c | 3 +- 4 files changed, 99 insertions(+), 26 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 3126da2f468a..b65011d320e3 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -17,7 +17,7 @@ /** * trace_call_bpf - invoke BPF program - * @prog: BPF program + * @call: tracepoint event * @ctx: opaque context pointer * * kprobe handlers execute BPF programs via this helper. @@ -29,7 +29,7 @@ * 1 - store kprobe event into ring buffer * Other values are reserved and currently alias to 1 */ -unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx) +unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx) { unsigned int ret; @@ -49,9 +49,22 @@ unsigned int trace_call_bpf(struct bpf_prog *prog, void *ctx) goto out; } - rcu_read_lock(); - ret = BPF_PROG_RUN(prog, ctx); - rcu_read_unlock(); + /* + * Instead of moving rcu_read_lock/rcu_dereference/rcu_read_unlock + * to all call sites, we did a bpf_prog_array_valid() there to check + * whether call->prog_array is empty or not, which is + * a heurisitc to speed up execution. + * + * If bpf_prog_array_valid() fetched prog_array was + * non-NULL, we go into trace_call_bpf() and do the actual + * proper rcu_dereference() under RCU lock. + * If it turns out that prog_array is NULL then, we bail out. + * For the opposite, if the bpf_prog_array_valid() fetched pointer + * was NULL, you'll skip the prog_array with the risk of missing + * out of events when it was updated in between this and the + * rcu_dereference() which is accepted risk. + */ + ret = BPF_PROG_RUN_ARRAY_CHECK(call->prog_array, ctx, BPF_PROG_RUN); out: __this_cpu_dec(bpf_prog_active); @@ -741,3 +754,62 @@ const struct bpf_verifier_ops perf_event_verifier_ops = { const struct bpf_prog_ops perf_event_prog_ops = { }; + +static DEFINE_MUTEX(bpf_event_mutex); + +int perf_event_attach_bpf_prog(struct perf_event *event, + struct bpf_prog *prog) +{ + struct bpf_prog_array __rcu *old_array; + struct bpf_prog_array *new_array; + int ret = -EEXIST; + + mutex_lock(&bpf_event_mutex); + + if (event->prog) + goto out; + + old_array = rcu_dereference_protected(event->tp_event->prog_array, + lockdep_is_held(&bpf_event_mutex)); + ret = bpf_prog_array_copy(old_array, NULL, prog, &new_array); + if (ret < 0) + goto out; + + /* set the new array to event->tp_event and set event->prog */ + event->prog = prog; + rcu_assign_pointer(event->tp_event->prog_array, new_array); + bpf_prog_array_free(old_array); + +out: + mutex_unlock(&bpf_event_mutex); + return ret; +} + +void perf_event_detach_bpf_prog(struct perf_event *event) +{ + struct bpf_prog_array __rcu *old_array; + struct bpf_prog_array *new_array; + int ret; + + mutex_lock(&bpf_event_mutex); + + if (!event->prog) + goto out; + + old_array = rcu_dereference_protected(event->tp_event->prog_array, + lockdep_is_held(&bpf_event_mutex)); + + ret = bpf_prog_array_copy(old_array, event->prog, NULL, &new_array); + if (ret < 0) { + bpf_prog_array_delete_safe(old_array, event->prog); + } else { + rcu_assign_pointer(event->tp_event->prog_array, new_array); + bpf_prog_array_free(old_array); + } + + bpf_prog_put(event->prog); + event->prog = NULL; + +out: + mutex_unlock(&bpf_event_mutex); +} diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 8a907e12b6b9..abf92e478cfb 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1174,13 +1174,12 @@ static void kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) { struct trace_event_call *call = &tk->tp.call; - struct bpf_prog *prog = call->prog; struct kprobe_trace_entry_head *entry; struct hlist_head *head; int size, __size, dsize; int rctx; - if (prog && !trace_call_bpf(prog, regs)) + if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs)) return; head = this_cpu_ptr(call->perf_events); @@ -1210,13 +1209,12 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, struct pt_regs *regs) { struct trace_event_call *call = &tk->tp.call; - struct bpf_prog *prog = call->prog; struct kretprobe_trace_entry_head *entry; struct hlist_head *head; int size, __size, dsize; int rctx; - if (prog && !trace_call_bpf(prog, regs)) + if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs)) return; head = this_cpu_ptr(call->perf_events); diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 696afe72d3b1..71a6af34d7a9 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -559,9 +559,10 @@ static DECLARE_BITMAP(enabled_perf_exit_syscalls, NR_syscalls); static int sys_perf_refcount_enter; static int sys_perf_refcount_exit; -static int perf_call_bpf_enter(struct bpf_prog *prog, struct pt_regs *regs, - struct syscall_metadata *sys_data, - struct syscall_trace_enter *rec) { +static int perf_call_bpf_enter(struct trace_event_call *call, struct pt_regs *regs, + struct syscall_metadata *sys_data, + struct syscall_trace_enter *rec) +{ struct syscall_tp_t { unsigned long long regs; unsigned long syscall_nr; @@ -573,7 +574,7 @@ static int perf_call_bpf_enter(struct bpf_prog *prog, struct pt_regs *regs, param.syscall_nr = rec->nr; for (i = 0; i < sys_data->nb_args; i++) param.args[i] = rec->args[i]; - return trace_call_bpf(prog, ¶m); + return trace_call_bpf(call, ¶m); } static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) @@ -581,7 +582,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) struct syscall_metadata *sys_data; struct syscall_trace_enter *rec; struct hlist_head *head; - struct bpf_prog *prog; + bool valid_prog_array; int syscall_nr; int rctx; int size; @@ -596,9 +597,9 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) if (!sys_data) return; - prog = READ_ONCE(sys_data->enter_event->prog); head = this_cpu_ptr(sys_data->enter_event->perf_events); - if (!prog && hlist_empty(head)) + valid_prog_array = bpf_prog_array_valid(sys_data->enter_event); + if (!valid_prog_array && hlist_empty(head)) return; /* get the size after alignment with the u32 buffer size field */ @@ -614,7 +615,8 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) syscall_get_arguments(current, regs, 0, sys_data->nb_args, (unsigned long *)&rec->args); - if ((prog && !perf_call_bpf_enter(prog, regs, sys_data, rec)) || + if ((valid_prog_array && + !perf_call_bpf_enter(sys_data->enter_event, regs, sys_data, rec)) || hlist_empty(head)) { perf_swevent_put_recursion_context(rctx); return; @@ -659,8 +661,9 @@ static void perf_sysenter_disable(struct trace_event_call *call) mutex_unlock(&syscall_trace_lock); } -static int perf_call_bpf_exit(struct bpf_prog *prog, struct pt_regs *regs, - struct syscall_trace_exit *rec) { +static int perf_call_bpf_exit(struct trace_event_call *call, struct pt_regs *regs, + struct syscall_trace_exit *rec) +{ struct syscall_tp_t { unsigned long long regs; unsigned long syscall_nr; @@ -670,7 +673,7 @@ static int perf_call_bpf_exit(struct bpf_prog *prog, struct pt_regs *regs, *(struct pt_regs **)¶m = regs; param.syscall_nr = rec->nr; param.ret = rec->ret; - return trace_call_bpf(prog, ¶m); + return trace_call_bpf(call, ¶m); } static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) @@ -678,7 +681,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) struct syscall_metadata *sys_data; struct syscall_trace_exit *rec; struct hlist_head *head; - struct bpf_prog *prog; + bool valid_prog_array; int syscall_nr; int rctx; int size; @@ -693,9 +696,9 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) if (!sys_data) return; - prog = READ_ONCE(sys_data->exit_event->prog); head = this_cpu_ptr(sys_data->exit_event->perf_events); - if (!prog && hlist_empty(head)) + valid_prog_array = bpf_prog_array_valid(sys_data->exit_event); + if (!valid_prog_array && hlist_empty(head)) return; /* We can probably do that at build time */ @@ -709,7 +712,8 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->nr = syscall_nr; rec->ret = syscall_get_return_value(current, regs); - if ((prog && !perf_call_bpf_exit(prog, regs, rec)) || + if ((valid_prog_array && + !perf_call_bpf_exit(sys_data->exit_event, regs, rec)) || hlist_empty(head)) { perf_swevent_put_recursion_context(rctx); return; diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 4525e0271a53..153c0e411461 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -1113,13 +1113,12 @@ static void __uprobe_perf_func(struct trace_uprobe *tu, { struct trace_event_call *call = &tu->tp.call; struct uprobe_trace_entry_head *entry; - struct bpf_prog *prog = call->prog; struct hlist_head *head; void *data; int size, esize; int rctx; - if (prog && !trace_call_bpf(prog, regs)) + if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs)) return; esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); -- cgit v1.2.3 From 6aa7de059173a986114ac43b8f50b297a86f09a8 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 23 Oct 2017 14:07:29 -0700 Subject: locking/atomics: COCCINELLE/treewide: Convert trivial ACCESS_ONCE() patterns to READ_ONCE()/WRITE_ONCE() Please do not apply this to mainline directly, instead please re-run the coccinelle script shown below and apply its output. For several reasons, it is desirable to use {READ,WRITE}_ONCE() in preference to ACCESS_ONCE(), and new code is expected to use one of the former. So far, there's been no reason to change most existing uses of ACCESS_ONCE(), as these aren't harmful, and changing them results in churn. However, for some features, the read/write distinction is critical to correct operation. To distinguish these cases, separate read/write accessors must be used. This patch migrates (most) remaining ACCESS_ONCE() instances to {READ,WRITE}_ONCE(), using the following coccinelle script: ---- // Convert trivial ACCESS_ONCE() uses to equivalent READ_ONCE() and // WRITE_ONCE() // $ make coccicheck COCCI=/home/mark/once.cocci SPFLAGS="--include-headers" MODE=patch virtual patch @ depends on patch @ expression E1, E2; @@ - ACCESS_ONCE(E1) = E2 + WRITE_ONCE(E1, E2) @ depends on patch @ expression E; @@ - ACCESS_ONCE(E) + READ_ONCE(E) ---- Signed-off-by: Mark Rutland Signed-off-by: Paul E. McKenney Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: davem@davemloft.net Cc: linux-arch@vger.kernel.org Cc: mpe@ellerman.id.au Cc: shuah@kernel.org Cc: snitzer@redhat.com Cc: thor.thayer@linux.intel.com Cc: tj@kernel.org Cc: viro@zeniv.linux.org.uk Cc: will.deacon@arm.com Link: http://lkml.kernel.org/r/1508792849-3115-19-git-send-email-paulmck@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 2 +- kernel/trace/trace.h | 2 +- kernel/trace/trace_stack.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 81279c6602ff..845f3805c73d 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2724,7 +2724,7 @@ rb_reserve_next_event(struct ring_buffer *buffer, * if it happened, we have to fail the write. */ barrier(); - if (unlikely(ACCESS_ONCE(cpu_buffer->buffer) != buffer)) { + if (unlikely(READ_ONCE(cpu_buffer->buffer) != buffer)) { local_dec(&cpu_buffer->committing); local_dec(&cpu_buffer->commits); return NULL; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 652c682707cd..9050c8b3ccde 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1459,7 +1459,7 @@ extern struct trace_event_file *find_event_file(struct trace_array *tr, static inline void *event_file_data(struct file *filp) { - return ACCESS_ONCE(file_inode(filp)->i_private); + return READ_ONCE(file_inode(filp)->i_private); } extern struct mutex event_mutex; diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 49cb41412eec..780262210c9a 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -77,7 +77,7 @@ check_stack(unsigned long ip, unsigned long *stack) { unsigned long this_size, flags; unsigned long *p, *top, *start; static int tracer_frame; - int frame_size = ACCESS_ONCE(tracer_frame); + int frame_size = READ_ONCE(tracer_frame); int i, x; this_size = ((unsigned long)stack) & (THREAD_SIZE-1); -- cgit v1.2.3 From 7d9285e82db5defca4d9674ba089429eeca0c697 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 5 Oct 2017 09:19:19 -0700 Subject: perf/bpf: Extend the perf_event_read_local() interface, a.k.a. "bpf: perf event change needed for subsequent bpf helpers" eBPF programs would like access to the (perf) event enabled and running times along with the event value, such that they can deal with event multiplexing (among other things). This patch extends the interface; a future eBPF patch will utilize the new functionality. [ Note, there's a same-content commit with a poor changelog and a meaningless title in the networking tree as well - but we need this change for subsequent perf work, so apply it here as well, with a proper changelog. Hopefully Git will be able to sort out this somewhat messy workflow, if there are no other, conflicting changes to these files. ] Signed-off-by: Yonghong Song [ Rewrote the changelog. ] Signed-off-by: Peter Zijlstra (Intel) Cc: Cc: Cc: Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: David S. Miller Link: http://lkml.kernel.org/r/20171005161923.332790-2-yhs@fb.com Signed-off-by: Ingo Molnar --- kernel/trace/bpf_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index dc498b605d5d..95888ae6c263 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -275,7 +275,7 @@ BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) if (!ee) return -ENOENT; - err = perf_event_read_local(ee->event, &value); + err = perf_event_read_local(ee->event, &value, NULL, NULL); /* * this api is ugly since we miss [-22..-2] range of valid * counter values, but that's uapi -- cgit v1.2.3 From 035226b964c820f65e201cdf123705a8f1d7c670 Mon Sep 17 00:00:00 2001 From: Gianluca Borello Date: Thu, 26 Oct 2017 01:47:42 +0000 Subject: bpf: remove tail_call and get_stackid helper declarations from bpf.h commit afdb09c720b6 ("security: bpf: Add LSM hooks for bpf object related syscall") included linux/bpf.h in linux/security.h. As a result, bpf programs including bpf_helpers.h and some other header that ends up pulling in also security.h, such as several examples under samples/bpf, fail to compile because bpf_tail_call and bpf_get_stackid are now "redefined as different kind of symbol". >From bpf.h: u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5); u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); Whereas in bpf_helpers.h they are: static void (*bpf_tail_call)(void *ctx, void *map, int index); static int (*bpf_get_stackid)(void *ctx, void *map, int flags); Fix this by removing the unused declaration of bpf_tail_call and moving the declaration of bpf_get_stackid in bpf_trace.c, which is the only place where it's needed. Signed-off-by: Gianluca Borello Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index b65011d320e3..136aa6bb0422 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -15,6 +15,8 @@ #include #include "trace.h" +u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); + /** * trace_call_bpf - invoke BPF program * @call: tracepoint event -- cgit v1.2.3 From 07c41a295c5f25928a7cb689fdec816bd0089fe8 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Mon, 30 Oct 2017 13:50:22 -0700 Subject: bpf: avoid rcu_dereference inside bpf_event_mutex lock region During perf event attaching/detaching bpf programs, the tp_event->prog_array change is protected by the bpf_event_mutex lock in both attaching and deteching functions. Although tp_event->prog_array is a rcu pointer, rcu_derefrence is not needed to access it since mutex lock will guarantee ordering. Verified through "make C=2" that sparse locking check still happy with the new change. Also change the label name in perf_event_{attach,detach}_bpf_prog from "out" to "unlock" to reflect the code action after the label. Signed-off-by: Yonghong Song Acked-by: Alexei Starovoitov Acked-by: Martin KaFai Lau Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 136aa6bb0422..506efe6e8ed9 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -769,20 +769,19 @@ int perf_event_attach_bpf_prog(struct perf_event *event, mutex_lock(&bpf_event_mutex); if (event->prog) - goto out; + goto unlock; - old_array = rcu_dereference_protected(event->tp_event->prog_array, - lockdep_is_held(&bpf_event_mutex)); + old_array = event->tp_event->prog_array; ret = bpf_prog_array_copy(old_array, NULL, prog, &new_array); if (ret < 0) - goto out; + goto unlock; /* set the new array to event->tp_event and set event->prog */ event->prog = prog; rcu_assign_pointer(event->tp_event->prog_array, new_array); bpf_prog_array_free(old_array); -out: +unlock: mutex_unlock(&bpf_event_mutex); return ret; } @@ -796,11 +795,9 @@ void perf_event_detach_bpf_prog(struct perf_event *event) mutex_lock(&bpf_event_mutex); if (!event->prog) - goto out; - - old_array = rcu_dereference_protected(event->tp_event->prog_array, - lockdep_is_held(&bpf_event_mutex)); + goto unlock; + old_array = event->tp_event->prog_array; ret = bpf_prog_array_copy(old_array, event->prog, NULL, &new_array); if (ret < 0) { bpf_prog_array_delete_safe(old_array, event->prog); @@ -812,6 +809,6 @@ void perf_event_detach_bpf_prog(struct perf_event *event) bpf_prog_put(event->prog); event->prog = NULL; -out: +unlock: mutex_unlock(&bpf_event_mutex); } -- cgit v1.2.3 From b24413180f5600bcb3bb70fbed5cf186b60864bd Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 1 Nov 2017 15:07:57 +0100 Subject: License cleanup: add SPDX GPL-2.0 license identifier to files with no license Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by: Kate Stewart Reviewed-by: Philippe Ombredanne Reviewed-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman --- kernel/trace/Makefile | 1 + kernel/trace/power-traces.c | 1 + kernel/trace/rpm-traces.c | 1 + kernel/trace/trace.h | 1 + kernel/trace/trace_benchmark.c | 1 + kernel/trace/trace_benchmark.h | 1 + kernel/trace/trace_branch.c | 1 + kernel/trace/trace_entries.h | 1 + kernel/trace/trace_events_filter_test.h | 1 + kernel/trace/trace_export.c | 1 + kernel/trace/trace_functions.c | 1 + kernel/trace/trace_functions_graph.c | 1 + kernel/trace/trace_kdb.c | 1 + kernel/trace/trace_mmiotrace.c | 1 + kernel/trace/trace_nop.c | 1 + kernel/trace/trace_output.h | 1 + kernel/trace/trace_sched_switch.c | 1 + kernel/trace/trace_sched_wakeup.c | 1 + kernel/trace/trace_selftest.c | 1 + kernel/trace/trace_selftest_dynamic.c | 1 + kernel/trace/trace_stack.c | 1 + kernel/trace/trace_stat.c | 1 + kernel/trace/trace_stat.h | 1 + kernel/trace/trace_syscalls.c | 1 + kernel/trace/tracing_map.h | 1 + 25 files changed, 25 insertions(+) (limited to 'kernel/trace') diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 90f2701d92a7..19a15b2f1190 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0 # Do not instrument the tracer itself: diff --git a/kernel/trace/power-traces.c b/kernel/trace/power-traces.c index 0c7dee221dca..21bb161c2316 100644 --- a/kernel/trace/power-traces.c +++ b/kernel/trace/power-traces.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Power trace points * diff --git a/kernel/trace/rpm-traces.c b/kernel/trace/rpm-traces.c index 4b3b5eaf94d1..25dec0b00280 100644 --- a/kernel/trace/rpm-traces.c +++ b/kernel/trace/rpm-traces.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Power trace points * diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 652c682707cd..401b0639116f 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #ifndef _LINUX_KERNEL_TRACE_H #define _LINUX_KERNEL_TRACE_H diff --git a/kernel/trace/trace_benchmark.c b/kernel/trace/trace_benchmark.c index 16a8cf02eee9..79f838a75077 100644 --- a/kernel/trace/trace_benchmark.c +++ b/kernel/trace/trace_benchmark.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 #include #include #include diff --git a/kernel/trace/trace_benchmark.h b/kernel/trace/trace_benchmark.h index ebdbfc2f2a64..be1d86ff753d 100644 --- a/kernel/trace/trace_benchmark.h +++ b/kernel/trace/trace_benchmark.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #undef TRACE_SYSTEM #define TRACE_SYSTEM benchmark diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index 4d8fdf3184dc..4ad967453b6f 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * unlikely profiler * diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h index adcdbbeae010..e954ae3d82c0 100644 --- a/kernel/trace/trace_entries.h +++ b/kernel/trace/trace_entries.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * This file defines the trace event structures that go into the ring * buffer directly. They are created via macros so that changes for them diff --git a/kernel/trace/trace_events_filter_test.h b/kernel/trace/trace_events_filter_test.h index bfd4dba0d603..39d7ef4f57cb 100644 --- a/kernel/trace/trace_events_filter_test.h +++ b/kernel/trace/trace_events_filter_test.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #undef TRACE_SYSTEM #define TRACE_SYSTEM test diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index 39aa7aa66468..548e62eb5c46 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * trace_export.c - export basic ftrace utilities to user space * diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index a0910c0cdf2e..27f7ad12c4b1 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * ring buffer based function tracer * diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index b8f1f54731af..23c0b0cb5fb9 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * * Function graph tracer. diff --git a/kernel/trace/trace_kdb.c b/kernel/trace/trace_kdb.c index 57149bce6aad..d953c163a079 100644 --- a/kernel/trace/trace_kdb.c +++ b/kernel/trace/trace_kdb.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * kdb helper for dumping the ftrace buffer * diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index dca78fc48439..b0388016b687 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Memory mapped I/O tracing * diff --git a/kernel/trace/trace_nop.c b/kernel/trace/trace_nop.c index 49f61fe96a6b..50523f953a5d 100644 --- a/kernel/trace/trace_nop.c +++ b/kernel/trace/trace_nop.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * nop tracer * diff --git a/kernel/trace/trace_output.h b/kernel/trace/trace_output.h index fabc49bcd493..dbba03ed96de 100644 --- a/kernel/trace/trace_output.h +++ b/kernel/trace/trace_output.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #ifndef __TRACE_EVENTS_H #define __TRACE_EVENTS_H diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index b341c02730be..e288168661e1 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * trace context switch * diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 0c331978b1a6..7d461dcd4831 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * trace task wakeup timings * diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index b17ec642793b..cd70eb5df38e 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* Include in trace.c */ #include diff --git a/kernel/trace/trace_selftest_dynamic.c b/kernel/trace/trace_selftest_dynamic.c index b4c475a0a48b..8cda06a10d66 100644 --- a/kernel/trace/trace_selftest_dynamic.c +++ b/kernel/trace/trace_selftest_dynamic.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 #include "trace.h" int DYN_FTRACE_TEST_NAME(void) diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 49cb41412eec..719a52a4064a 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Steven Rostedt * diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c index 413ff108fbd0..75bf1bcb4a8a 100644 --- a/kernel/trace/trace_stat.c +++ b/kernel/trace/trace_stat.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Infrastructure for statistic tracing (histogram output). * diff --git a/kernel/trace/trace_stat.h b/kernel/trace/trace_stat.h index 8f03914b9a6a..76d30b4ebe83 100644 --- a/kernel/trace/trace_stat.h +++ b/kernel/trace/trace_stat.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #ifndef __TRACE_STAT_H #define __TRACE_STAT_H diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 696afe72d3b1..a2a642f2c64f 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 #include #include #include diff --git a/kernel/trace/tracing_map.h b/kernel/trace/tracing_map.h index 618838f5f30a..ab0ca77331d0 100644 --- a/kernel/trace/tracing_map.h +++ b/kernel/trace/tracing_map.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #ifndef __TRACING_MAP_H #define __TRACING_MAP_H -- cgit v1.2.3 From 1f2cac107c591c24b60b115d6050adc213d10fc0 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 5 Nov 2017 09:13:48 -0700 Subject: blktrace: fix unlocked access to init/start-stop/teardown sg.c calls into the blktrace functions without holding the proper queue mutex for doing setup, start/stop, or teardown. Add internal unlocked variants, and export the ones that do the proper locking. Fixes: 6da127ad0918 ("blktrace: Add blktrace ioctls to SCSI generic devices") Tested-by: Dmitry Vyukov Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 58 ++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 48 insertions(+), 10 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 45a3928544ce..ea57dd94b2b2 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -336,7 +336,7 @@ static void blk_trace_cleanup(struct blk_trace *bt) blk_unregister_tracepoints(); } -int blk_trace_remove(struct request_queue *q) +static int __blk_trace_remove(struct request_queue *q) { struct blk_trace *bt; @@ -349,6 +349,17 @@ int blk_trace_remove(struct request_queue *q) return 0; } + +int blk_trace_remove(struct request_queue *q) +{ + int ret; + + mutex_lock(&q->blk_trace_mutex); + ret = __blk_trace_remove(q); + mutex_unlock(&q->blk_trace_mutex); + + return ret; +} EXPORT_SYMBOL_GPL(blk_trace_remove); static ssize_t blk_dropped_read(struct file *filp, char __user *buffer, @@ -550,9 +561,8 @@ err: return ret; } -int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, - struct block_device *bdev, - char __user *arg) +static int __blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, char __user *arg) { struct blk_user_trace_setup buts; int ret; @@ -571,6 +581,19 @@ int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, } return 0; } + +int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, + char __user *arg) +{ + int ret; + + mutex_lock(&q->blk_trace_mutex); + ret = __blk_trace_setup(q, name, dev, bdev, arg); + mutex_unlock(&q->blk_trace_mutex); + + return ret; +} EXPORT_SYMBOL_GPL(blk_trace_setup); #if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64) @@ -607,7 +630,7 @@ static int compat_blk_trace_setup(struct request_queue *q, char *name, } #endif -int blk_trace_startstop(struct request_queue *q, int start) +static int __blk_trace_startstop(struct request_queue *q, int start) { int ret; struct blk_trace *bt = q->blk_trace; @@ -646,6 +669,17 @@ int blk_trace_startstop(struct request_queue *q, int start) return ret; } + +int blk_trace_startstop(struct request_queue *q, int start) +{ + int ret; + + mutex_lock(&q->blk_trace_mutex); + ret = __blk_trace_startstop(q, start); + mutex_unlock(&q->blk_trace_mutex); + + return ret; +} EXPORT_SYMBOL_GPL(blk_trace_startstop); /* @@ -676,7 +710,7 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) switch (cmd) { case BLKTRACESETUP: bdevname(bdev, b); - ret = blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); + ret = __blk_trace_setup(q, b, bdev->bd_dev, bdev, arg); break; #if defined(CONFIG_COMPAT) && defined(CONFIG_X86_64) case BLKTRACESETUP32: @@ -687,10 +721,10 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) case BLKTRACESTART: start = 1; case BLKTRACESTOP: - ret = blk_trace_startstop(q, start); + ret = __blk_trace_startstop(q, start); break; case BLKTRACETEARDOWN: - ret = blk_trace_remove(q); + ret = __blk_trace_remove(q); break; default: ret = -ENOTTY; @@ -708,10 +742,14 @@ int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg) **/ void blk_trace_shutdown(struct request_queue *q) { + mutex_lock(&q->blk_trace_mutex); + if (q->blk_trace) { - blk_trace_startstop(q, 0); - blk_trace_remove(q); + __blk_trace_startstop(q, 0); + __blk_trace_remove(q); } + + mutex_unlock(&q->blk_trace_mutex); } #ifdef CONFIG_BLK_CGROUP -- cgit v1.2.3 From a6da0024ffc19e0d47712bb5ca4fd083f76b07df Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 5 Nov 2017 09:16:09 -0700 Subject: blktrace: fix unlocked registration of tracepoints We need to ensure that tracepoints are registered and unregistered with the users of them. The existing atomic count isn't enough for that. Add a lock around the tracepoints, so we serialize access to them. This fixes cases where we have multiple users setting up and tearing down tracepoints, like this: CPU: 0 PID: 2995 Comm: syzkaller857118 Not tainted 4.14.0-rc5-next-20171018+ #36 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:16 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:52 panic+0x1e4/0x41c kernel/panic.c:183 __warn+0x1c4/0x1e0 kernel/panic.c:546 report_bug+0x211/0x2d0 lib/bug.c:183 fixup_bug+0x40/0x90 arch/x86/kernel/traps.c:177 do_trap_no_signal arch/x86/kernel/traps.c:211 [inline] do_trap+0x260/0x390 arch/x86/kernel/traps.c:260 do_error_trap+0x120/0x390 arch/x86/kernel/traps.c:297 do_invalid_op+0x1b/0x20 arch/x86/kernel/traps.c:310 invalid_op+0x18/0x20 arch/x86/entry/entry_64.S:905 RIP: 0010:tracepoint_add_func kernel/tracepoint.c:210 [inline] RIP: 0010:tracepoint_probe_register_prio+0x397/0x9a0 kernel/tracepoint.c:283 RSP: 0018:ffff8801d1d1f6c0 EFLAGS: 00010293 RAX: ffff8801d22e8540 RBX: 00000000ffffffef RCX: ffffffff81710f07 RDX: 0000000000000000 RSI: ffffffff85b679c0 RDI: ffff8801d5f19818 RBP: ffff8801d1d1f7c8 R08: ffffffff81710c10 R09: 0000000000000004 R10: ffff8801d1d1f6b0 R11: 0000000000000003 R12: ffffffff817597f0 R13: 0000000000000000 R14: 00000000ffffffff R15: ffff8801d1d1f7a0 tracepoint_probe_register+0x2a/0x40 kernel/tracepoint.c:304 register_trace_block_rq_insert include/trace/events/block.h:191 [inline] blk_register_tracepoints+0x1e/0x2f0 kernel/trace/blktrace.c:1043 do_blk_trace_setup+0xa10/0xcf0 kernel/trace/blktrace.c:542 blk_trace_setup+0xbd/0x180 kernel/trace/blktrace.c:564 sg_ioctl+0xc71/0x2d90 drivers/scsi/sg.c:1089 vfs_ioctl fs/ioctl.c:45 [inline] do_vfs_ioctl+0x1b1/0x1520 fs/ioctl.c:685 SYSC_ioctl fs/ioctl.c:700 [inline] SyS_ioctl+0x8f/0xc0 fs/ioctl.c:691 entry_SYSCALL_64_fastpath+0x1f/0xbe RIP: 0033:0x444339 RSP: 002b:00007ffe05bb5b18 EFLAGS: 00000206 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00000000006d66c0 RCX: 0000000000444339 RDX: 000000002084cf90 RSI: 00000000c0481273 RDI: 0000000000000009 RBP: 0000000000000082 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000206 R12: ffffffffffffffff R13: 00000000c0481273 R14: 0000000000000000 R15: 0000000000000000 since we can now run these in parallel. Ensure that the exported helpers for doing this are grabbing the queue trace mutex. Reported-by: Steven Rostedt Tested-by: Dmitry Vyukov Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 32 ++++++++++++++++++++++---------- 1 file changed, 22 insertions(+), 10 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index ea57dd94b2b2..206e0e2ace53 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -66,7 +66,8 @@ static struct tracer_flags blk_tracer_flags = { }; /* Global reference count of probes */ -static atomic_t blk_probes_ref = ATOMIC_INIT(0); +static DEFINE_MUTEX(blk_probe_mutex); +static int blk_probes_ref; static void blk_register_tracepoints(void); static void blk_unregister_tracepoints(void); @@ -329,11 +330,26 @@ static void blk_trace_free(struct blk_trace *bt) kfree(bt); } +static void get_probe_ref(void) +{ + mutex_lock(&blk_probe_mutex); + if (++blk_probes_ref == 1) + blk_register_tracepoints(); + mutex_unlock(&blk_probe_mutex); +} + +static void put_probe_ref(void) +{ + mutex_lock(&blk_probe_mutex); + if (!--blk_probes_ref) + blk_unregister_tracepoints(); + mutex_unlock(&blk_probe_mutex); +} + static void blk_trace_cleanup(struct blk_trace *bt) { blk_trace_free(bt); - if (atomic_dec_and_test(&blk_probes_ref)) - blk_unregister_tracepoints(); + put_probe_ref(); } static int __blk_trace_remove(struct request_queue *q) @@ -549,8 +565,7 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, if (cmpxchg(&q->blk_trace, NULL, bt)) goto err; - if (atomic_inc_return(&blk_probes_ref) == 1) - blk_register_tracepoints(); + get_probe_ref(); ret = 0; err: @@ -1596,9 +1611,7 @@ static int blk_trace_remove_queue(struct request_queue *q) if (bt == NULL) return -EINVAL; - if (atomic_dec_and_test(&blk_probes_ref)) - blk_unregister_tracepoints(); - + put_probe_ref(); blk_trace_free(bt); return 0; } @@ -1629,8 +1642,7 @@ static int blk_trace_setup_queue(struct request_queue *q, if (cmpxchg(&q->blk_trace, NULL, bt)) goto free_bt; - if (atomic_inc_return(&blk_probes_ref) == 1) - blk_register_tracepoints(); + get_probe_ref(); return 0; free_bt: -- cgit v1.2.3 From dd0bb688eaa241b5655d396d45366cba9225aed9 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 7 Nov 2017 15:28:42 -0500 Subject: bpf: add a bpf_override_function helper Error injection is sloppy and very ad-hoc. BPF could fill this niche perfectly with it's kprobe functionality. We could make sure errors are only triggered in specific call chains that we care about with very specific situations. Accomplish this with the bpf_override_funciton helper. This will modify the probe'd callers return value to the specified value and set the PC to an override function that simply returns, bypassing the originally probed function. This gives us a nice clean way to implement systematic error injection for all of our code paths. Acked-by: Alexei Starovoitov Signed-off-by: Josef Bacik Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- kernel/trace/Kconfig | 11 +++++++++++ kernel/trace/bpf_trace.c | 35 +++++++++++++++++++++++++++++++++++ kernel/trace/trace_kprobe.c | 40 +++++++++++++++++++++++++++++++++------- kernel/trace/trace_probe.h | 6 ++++++ 4 files changed, 85 insertions(+), 7 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 434c840e2d82..9dc0deeaad2b 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -518,6 +518,17 @@ config FUNCTION_PROFILER If in doubt, say N. +config BPF_KPROBE_OVERRIDE + bool "Enable BPF programs to override a kprobed function" + depends on BPF_EVENTS + depends on KPROBES_ON_FTRACE + depends on HAVE_KPROBE_OVERRIDE + depends on DYNAMIC_FTRACE_WITH_REGS + default n + help + Allows BPF to override the execution of a probed function and + set a different return value. This is used for error injection. + config FTRACE_MCOUNT_RECORD def_bool y depends on DYNAMIC_FTRACE diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 506efe6e8ed9..1865b0d4cdeb 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -13,6 +13,10 @@ #include #include #include +#include +#include + +#include "trace_probe.h" #include "trace.h" u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); @@ -76,6 +80,29 @@ unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx) } EXPORT_SYMBOL_GPL(trace_call_bpf); +#ifdef CONFIG_BPF_KPROBE_OVERRIDE +BPF_CALL_2(bpf_override_return, struct pt_regs *, regs, unsigned long, rc) +{ + __this_cpu_write(bpf_kprobe_override, 1); + regs_set_return_value(regs, rc); + arch_ftrace_kprobe_override_function(regs); + return 0; +} +#else +BPF_CALL_2(bpf_override_return, struct pt_regs *, regs, unsigned long, rc) +{ + return -EINVAL; +} +#endif + +static const struct bpf_func_proto bpf_override_return_proto = { + .func = bpf_override_return, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, +}; + BPF_CALL_3(bpf_probe_read, void *, dst, u32, size, const void *, unsafe_ptr) { int ret; @@ -551,6 +578,10 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func return &bpf_get_stackid_proto; case BPF_FUNC_perf_event_read_value: return &bpf_perf_event_read_value_proto; + case BPF_FUNC_override_return: + pr_warn_ratelimited("%s[%d] is installing a program with bpf_override_return helper that may cause unexpected behavior!", + current->comm, task_pid_nr(current)); + return &bpf_override_return_proto; default: return tracing_func_proto(func_id); } @@ -766,6 +797,10 @@ int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog_array *new_array; int ret = -EEXIST; + /* Kprobe override only works for ftrace based kprobes. */ + if (prog->kprobe_override && !trace_kprobe_ftrace(event->tp_event)) + return -EINVAL; + mutex_lock(&bpf_event_mutex); if (event->prog) diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index abf92e478cfb..8e3c9ec1faf7 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -42,6 +42,7 @@ struct trace_kprobe { (offsetof(struct trace_kprobe, tp.args) + \ (sizeof(struct probe_arg) * (n))) +DEFINE_PER_CPU(int, bpf_kprobe_override); static nokprobe_inline bool trace_kprobe_is_return(struct trace_kprobe *tk) { @@ -87,6 +88,12 @@ static nokprobe_inline unsigned long trace_kprobe_nhit(struct trace_kprobe *tk) return nhit; } +int trace_kprobe_ftrace(struct trace_event_call *call) +{ + struct trace_kprobe *tk = (struct trace_kprobe *)call->data; + return kprobe_ftrace(&tk->rp.kp); +} + static int register_kprobe_event(struct trace_kprobe *tk); static int unregister_kprobe_event(struct trace_kprobe *tk); @@ -1170,7 +1177,7 @@ static int kretprobe_event_define_fields(struct trace_event_call *event_call) #ifdef CONFIG_PERF_EVENTS /* Kprobe profile handler */ -static void +static int kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) { struct trace_event_call *call = &tk->tp.call; @@ -1179,12 +1186,29 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) int size, __size, dsize; int rctx; - if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs)) - return; + if (bpf_prog_array_valid(call)) { + int ret; + + ret = trace_call_bpf(call, regs); + + /* + * We need to check and see if we modified the pc of the + * pt_regs, and if so clear the kprobe and return 1 so that we + * don't do the instruction skipping. Also reset our state so + * we are clean the next pass through. + */ + if (__this_cpu_read(bpf_kprobe_override)) { + __this_cpu_write(bpf_kprobe_override, 0); + reset_current_kprobe(); + return 1; + } + if (!ret) + return 0; + } head = this_cpu_ptr(call->perf_events); if (hlist_empty(head)) - return; + return 0; dsize = __get_data_size(&tk->tp, regs); __size = sizeof(*entry) + tk->tp.size + dsize; @@ -1193,13 +1217,14 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) entry = perf_trace_buf_alloc(size, NULL, &rctx); if (!entry) - return; + return 0; entry->ip = (unsigned long)tk->rp.kp.addr; memset(&entry[1], 0, dsize); store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL, NULL); + return 0; } NOKPROBE_SYMBOL(kprobe_perf_func); @@ -1275,6 +1300,7 @@ static int kprobe_register(struct trace_event_call *event, static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) { struct trace_kprobe *tk = container_of(kp, struct trace_kprobe, rp.kp); + int ret = 0; raw_cpu_inc(*tk->nhit); @@ -1282,9 +1308,9 @@ static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) kprobe_trace_func(tk, regs); #ifdef CONFIG_PERF_EVENTS if (tk->tp.flags & TP_FLAG_PROFILE) - kprobe_perf_func(tk, regs); + ret = kprobe_perf_func(tk, regs); #endif - return 0; /* We don't tweek kernel, so just return 0 */ + return ret; } NOKPROBE_SYMBOL(kprobe_dispatcher); diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index 903273c93e61..adbb3f7d1fb5 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -253,6 +253,7 @@ struct symbol_cache; unsigned long update_symbol_cache(struct symbol_cache *sc); void free_symbol_cache(struct symbol_cache *sc); struct symbol_cache *alloc_symbol_cache(const char *sym, long offset); +int trace_kprobe_ftrace(struct trace_event_call *call); #else /* uprobes do not support symbol fetch methods */ #define fetch_symbol_u8 NULL @@ -278,6 +279,11 @@ alloc_symbol_cache(const char *sym, long offset) { return NULL; } + +static inline int trace_kprobe_ftrace(struct trace_event_call *call) +{ + return 0; +} #endif /* CONFIG_KPROBE_EVENTS */ struct probe_arg { -- cgit v1.2.3 From f3edacbd697f94a743fff1a3d26910ab99948ba7 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 11 Nov 2017 18:24:55 +0900 Subject: bpf: Revert bpf_overrid_function() helper changes. NACK'd by x86 maintainer. Signed-off-by: David S. Miller --- kernel/trace/Kconfig | 11 ----------- kernel/trace/bpf_trace.c | 35 ----------------------------------- kernel/trace/trace_kprobe.c | 40 +++++++--------------------------------- kernel/trace/trace_probe.h | 6 ------ 4 files changed, 7 insertions(+), 85 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 9dc0deeaad2b..434c840e2d82 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -518,17 +518,6 @@ config FUNCTION_PROFILER If in doubt, say N. -config BPF_KPROBE_OVERRIDE - bool "Enable BPF programs to override a kprobed function" - depends on BPF_EVENTS - depends on KPROBES_ON_FTRACE - depends on HAVE_KPROBE_OVERRIDE - depends on DYNAMIC_FTRACE_WITH_REGS - default n - help - Allows BPF to override the execution of a probed function and - set a different return value. This is used for error injection. - config FTRACE_MCOUNT_RECORD def_bool y depends on DYNAMIC_FTRACE diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 1865b0d4cdeb..506efe6e8ed9 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -13,10 +13,6 @@ #include #include #include -#include -#include - -#include "trace_probe.h" #include "trace.h" u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); @@ -80,29 +76,6 @@ unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx) } EXPORT_SYMBOL_GPL(trace_call_bpf); -#ifdef CONFIG_BPF_KPROBE_OVERRIDE -BPF_CALL_2(bpf_override_return, struct pt_regs *, regs, unsigned long, rc) -{ - __this_cpu_write(bpf_kprobe_override, 1); - regs_set_return_value(regs, rc); - arch_ftrace_kprobe_override_function(regs); - return 0; -} -#else -BPF_CALL_2(bpf_override_return, struct pt_regs *, regs, unsigned long, rc) -{ - return -EINVAL; -} -#endif - -static const struct bpf_func_proto bpf_override_return_proto = { - .func = bpf_override_return, - .gpl_only = true, - .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_CTX, - .arg2_type = ARG_ANYTHING, -}; - BPF_CALL_3(bpf_probe_read, void *, dst, u32, size, const void *, unsafe_ptr) { int ret; @@ -578,10 +551,6 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func return &bpf_get_stackid_proto; case BPF_FUNC_perf_event_read_value: return &bpf_perf_event_read_value_proto; - case BPF_FUNC_override_return: - pr_warn_ratelimited("%s[%d] is installing a program with bpf_override_return helper that may cause unexpected behavior!", - current->comm, task_pid_nr(current)); - return &bpf_override_return_proto; default: return tracing_func_proto(func_id); } @@ -797,10 +766,6 @@ int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog_array *new_array; int ret = -EEXIST; - /* Kprobe override only works for ftrace based kprobes. */ - if (prog->kprobe_override && !trace_kprobe_ftrace(event->tp_event)) - return -EINVAL; - mutex_lock(&bpf_event_mutex); if (event->prog) diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 8e3c9ec1faf7..abf92e478cfb 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -42,7 +42,6 @@ struct trace_kprobe { (offsetof(struct trace_kprobe, tp.args) + \ (sizeof(struct probe_arg) * (n))) -DEFINE_PER_CPU(int, bpf_kprobe_override); static nokprobe_inline bool trace_kprobe_is_return(struct trace_kprobe *tk) { @@ -88,12 +87,6 @@ static nokprobe_inline unsigned long trace_kprobe_nhit(struct trace_kprobe *tk) return nhit; } -int trace_kprobe_ftrace(struct trace_event_call *call) -{ - struct trace_kprobe *tk = (struct trace_kprobe *)call->data; - return kprobe_ftrace(&tk->rp.kp); -} - static int register_kprobe_event(struct trace_kprobe *tk); static int unregister_kprobe_event(struct trace_kprobe *tk); @@ -1177,7 +1170,7 @@ static int kretprobe_event_define_fields(struct trace_event_call *event_call) #ifdef CONFIG_PERF_EVENTS /* Kprobe profile handler */ -static int +static void kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) { struct trace_event_call *call = &tk->tp.call; @@ -1186,29 +1179,12 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) int size, __size, dsize; int rctx; - if (bpf_prog_array_valid(call)) { - int ret; - - ret = trace_call_bpf(call, regs); - - /* - * We need to check and see if we modified the pc of the - * pt_regs, and if so clear the kprobe and return 1 so that we - * don't do the instruction skipping. Also reset our state so - * we are clean the next pass through. - */ - if (__this_cpu_read(bpf_kprobe_override)) { - __this_cpu_write(bpf_kprobe_override, 0); - reset_current_kprobe(); - return 1; - } - if (!ret) - return 0; - } + if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs)) + return; head = this_cpu_ptr(call->perf_events); if (hlist_empty(head)) - return 0; + return; dsize = __get_data_size(&tk->tp, regs); __size = sizeof(*entry) + tk->tp.size + dsize; @@ -1217,14 +1193,13 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs) entry = perf_trace_buf_alloc(size, NULL, &rctx); if (!entry) - return 0; + return; entry->ip = (unsigned long)tk->rp.kp.addr; memset(&entry[1], 0, dsize); store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize); perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs, head, NULL, NULL); - return 0; } NOKPROBE_SYMBOL(kprobe_perf_func); @@ -1300,7 +1275,6 @@ static int kprobe_register(struct trace_event_call *event, static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) { struct trace_kprobe *tk = container_of(kp, struct trace_kprobe, rp.kp); - int ret = 0; raw_cpu_inc(*tk->nhit); @@ -1308,9 +1282,9 @@ static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs) kprobe_trace_func(tk, regs); #ifdef CONFIG_PERF_EVENTS if (tk->tp.flags & TP_FLAG_PROFILE) - ret = kprobe_perf_func(tk, regs); + kprobe_perf_func(tk, regs); #endif - return ret; + return 0; /* We don't tweek kernel, so just return 0 */ } NOKPROBE_SYMBOL(kprobe_dispatcher); diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index adbb3f7d1fb5..903273c93e61 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -253,7 +253,6 @@ struct symbol_cache; unsigned long update_symbol_cache(struct symbol_cache *sc); void free_symbol_cache(struct symbol_cache *sc); struct symbol_cache *alloc_symbol_cache(const char *sym, long offset); -int trace_kprobe_ftrace(struct trace_event_call *call); #else /* uprobes do not support symbol fetch methods */ #define fetch_symbol_u8 NULL @@ -279,11 +278,6 @@ alloc_symbol_cache(const char *sym, long offset) { return NULL; } - -static inline int trace_kprobe_ftrace(struct trace_event_call *call) -{ - return 0; -} #endif /* CONFIG_KPROBE_EVENTS */ struct probe_arg { -- cgit v1.2.3 From 9c019e2bc4b2bd8223c8c0d4b6962478b479834d Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Sun, 12 Nov 2017 14:49:10 -0800 Subject: bpf: change helper bpf_probe_read arg2 type to ARG_CONST_SIZE_OR_ZERO The helper bpf_probe_read arg2 type is changed from ARG_CONST_SIZE to ARG_CONST_SIZE_OR_ZERO to permit size-0 buffer. Together with newer ARG_CONST_SIZE_OR_ZERO semantics which allows non-NULL buffer with size 0, this allows simpler bpf programs with verifier acceptance. The previous commit which changes ARG_CONST_SIZE_OR_ZERO semantics has details on examples. Signed-off-by: Yonghong Song Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- kernel/trace/bpf_trace.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 506efe6e8ed9..a5580c670866 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -78,12 +78,16 @@ EXPORT_SYMBOL_GPL(trace_call_bpf); BPF_CALL_3(bpf_probe_read, void *, dst, u32, size, const void *, unsafe_ptr) { - int ret; + int ret = 0; + + if (unlikely(size == 0)) + goto out; ret = probe_kernel_read(dst, unsafe_ptr, size); if (unlikely(ret < 0)) memset(dst, 0, size); + out: return ret; } @@ -92,7 +96,7 @@ static const struct bpf_func_proto bpf_probe_read_proto = { .gpl_only = true, .ret_type = RET_INTEGER, .arg1_type = ARG_PTR_TO_UNINIT_MEM, - .arg2_type = ARG_CONST_SIZE, + .arg2_type = ARG_CONST_SIZE_OR_ZERO, .arg3_type = ARG_ANYTHING, }; -- cgit v1.2.3 From 4950276672fce5c241857540f8561c440663673d Mon Sep 17 00:00:00 2001 From: "Levin, Alexander (Sasha Levin)" Date: Wed, 15 Nov 2017 17:35:51 -0800 Subject: kmemcheck: remove annotations Patch series "kmemcheck: kill kmemcheck", v2. As discussed at LSF/MM, kill kmemcheck. KASan is a replacement that is able to work without the limitation of kmemcheck (single CPU, slow). KASan is already upstream. We are also not aware of any users of kmemcheck (or users who don't consider KASan as a suitable replacement). The only objection was that since KASAN wasn't supported by all GCC versions provided by distros at that time we should hold off for 2 years, and try again. Now that 2 years have passed, and all distros provide gcc that supports KASAN, kill kmemcheck again for the very same reasons. This patch (of 4): Remove kmemcheck annotations, and calls to kmemcheck from the kernel. [alexander.levin@verizon.com: correctly remove kmemcheck call from dma_map_sg_attrs] Link: http://lkml.kernel.org/r/20171012192151.26531-1-alexander.levin@verizon.com Link: http://lkml.kernel.org/r/20171007030159.22241-2-alexander.levin@verizon.com Signed-off-by: Sasha Levin Cc: Alexander Potapenko Cc: Eric W. Biederman Cc: Michal Hocko Cc: Pekka Enberg Cc: Steven Rostedt Cc: Tim Hansen Cc: Vegard Nossum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/trace/ring_buffer.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 845f3805c73d..d57fede84b38 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -13,7 +13,6 @@ #include #include #include /* for self test */ -#include #include #include #include @@ -2055,7 +2054,6 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, } event = __rb_page_index(tail_page, tail); - kmemcheck_annotate_bitfield(event, bitfield); /* account for padding bytes */ local_add(BUF_PAGE_SIZE - tail, &cpu_buffer->entries_bytes); @@ -2686,7 +2684,6 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, /* We reserved something on the buffer */ event = __rb_page_index(tail_page, tail); - kmemcheck_annotate_bitfield(event, bitfield); rb_update_event(cpu_buffer, event, info); local_inc(&tail_page->entries); -- cgit v1.2.3 From eb33f2cca49ec49a1b893b5af546e7c042ca6365 Mon Sep 17 00:00:00 2001 From: Gianluca Borello Date: Wed, 22 Nov 2017 18:32:54 +0000 Subject: bpf: remove explicit handling of 0 for arg2 in bpf_probe_read Commit 9c019e2bc4b2 ("bpf: change helper bpf_probe_read arg2 type to ARG_CONST_SIZE_OR_ZERO") changed arg2 type to ARG_CONST_SIZE_OR_ZERO to simplify writing bpf programs by taking advantage of the new semantics introduced for ARG_CONST_SIZE_OR_ZERO which allows arguments. In order to prevent the helper from actually passing a NULL pointer to probe_kernel_read, which can happen when is passed to the helper, the commit also introduced an explicit check against size == 0. After the recent introduction of the ARG_PTR_TO_MEM_OR_NULL type, bpf_probe_read can not receive a pair of arguments anymore, thus the check is not needed anymore and can be removed, since probe_kernel_read can correctly handle a call. This also fixes the semantics of the helper before it gets officially released and bpf programs start relying on this check. Fixes: 9c019e2bc4b2 ("bpf: change helper bpf_probe_read arg2 type to ARG_CONST_SIZE_OR_ZERO") Signed-off-by: Gianluca Borello Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Acked-by: Yonghong Song Signed-off-by: Daniel Borkmann --- kernel/trace/bpf_trace.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index a5580c670866..728909f7951c 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -78,16 +78,12 @@ EXPORT_SYMBOL_GPL(trace_call_bpf); BPF_CALL_3(bpf_probe_read, void *, dst, u32, size, const void *, unsafe_ptr) { - int ret = 0; - - if (unlikely(size == 0)) - goto out; + int ret; ret = probe_kernel_read(dst, unsafe_ptr, size); if (unlikely(ret < 0)) memset(dst, 0, size); - out: return ret; } -- cgit v1.2.3 From 5c4e1201740ceae9bd6f622851a9bf7c66debe3a Mon Sep 17 00:00:00 2001 From: Gianluca Borello Date: Wed, 22 Nov 2017 18:32:55 +0000 Subject: bpf: change bpf_probe_read_str arg2 type to ARG_CONST_SIZE_OR_ZERO Commit 9fd29c08e520 ("bpf: improve verifier ARG_CONST_SIZE_OR_ZERO semantics") relaxed the treatment of ARG_CONST_SIZE_OR_ZERO due to the way the compiler generates optimized BPF code when checking boundaries of an argument from C code. A typical example of this optimized code can be generated using the bpf_probe_read_str helper when operating on variable memory: /* len is a generic scalar */ if (len > 0 && len <= 0x7fff) bpf_probe_read_str(p, len, s); 251: (79) r1 = *(u64 *)(r10 -88) 252: (07) r1 += -1 253: (25) if r1 > 0x7ffe goto pc-42 254: (bf) r1 = r7 255: (79) r2 = *(u64 *)(r10 -88) 256: (bf) r8 = r4 257: (85) call bpf_probe_read_str#45 R2 min value is negative, either use unsigned or 'var &= const' With this code, the verifier loses track of the variable. Replacing arg2 with ARG_CONST_SIZE_OR_ZERO is thus desirable since it avoids this quite common case which leads to usability issues, and the compiler generates code that the verifier can more easily test: if (len <= 0x7fff) bpf_probe_read_str(p, len, s); or bpf_probe_read_str(p, len & 0x7fff, s); No changes to the bpf_probe_read_str helper are necessary since strncpy_from_unsafe itself immediately returns if the size passed is 0. Signed-off-by: Gianluca Borello Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: Daniel Borkmann --- kernel/trace/bpf_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 728909f7951c..ed8601a1a861 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -494,7 +494,7 @@ static const struct bpf_func_proto bpf_probe_read_str_proto = { .gpl_only = true, .ret_type = RET_INTEGER, .arg1_type = ARG_PTR_TO_UNINIT_MEM, - .arg2_type = ARG_CONST_SIZE, + .arg2_type = ARG_CONST_SIZE_OR_ZERO, .arg3_type = ARG_ANYTHING, }; -- cgit v1.2.3 From a60dd35d2e39209fa7645945e1192bf9769872c6 Mon Sep 17 00:00:00 2001 From: Gianluca Borello Date: Wed, 22 Nov 2017 18:32:56 +0000 Subject: bpf: change bpf_perf_event_output arg5 type to ARG_CONST_SIZE_OR_ZERO Commit 9fd29c08e520 ("bpf: improve verifier ARG_CONST_SIZE_OR_ZERO semantics") relaxed the treatment of ARG_CONST_SIZE_OR_ZERO due to the way the compiler generates optimized BPF code when checking boundaries of an argument from C code. A typical example of this optimized code can be generated using the bpf_perf_event_output helper when operating on variable memory: /* len is a generic scalar */ if (len > 0 && len <= 0x7fff) bpf_perf_event_output(ctx, &perf_map, 0, buf, len); 110: (79) r5 = *(u64 *)(r10 -40) 111: (bf) r1 = r5 112: (07) r1 += -1 113: (25) if r1 > 0x7ffe goto pc+6 114: (bf) r1 = r6 115: (18) r2 = 0xffff94e5f166c200 117: (b7) r3 = 0 118: (bf) r4 = r7 119: (85) call bpf_perf_event_output#25 R5 min value is negative, either use unsigned or 'var &= const' With this code, the verifier loses track of the variable. Replacing arg5 with ARG_CONST_SIZE_OR_ZERO is thus desirable since it avoids this quite common case which leads to usability issues, and the compiler generates code that the verifier can more easily test: if (len <= 0x7fff) bpf_perf_event_output(ctx, &perf_map, 0, buf, len); or bpf_perf_event_output(ctx, &perf_map, 0, buf, len & 0x7fff); No changes to the bpf_perf_event_output helper are necessary since it can handle a case where size is 0, and an empty frame is pushed. Reported-by: Arnaldo Carvalho de Melo Signed-off-by: Gianluca Borello Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: Daniel Borkmann --- kernel/trace/bpf_trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index ed8601a1a861..27d1f4ffa3de 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -403,7 +403,7 @@ static const struct bpf_func_proto bpf_perf_event_output_proto = { .arg2_type = ARG_CONST_MAP_PTR, .arg3_type = ARG_ANYTHING, .arg4_type = ARG_PTR_TO_MEM, - .arg5_type = ARG_CONST_SIZE, + .arg5_type = ARG_CONST_SIZE_OR_ZERO, }; static DEFINE_PER_CPU(struct pt_regs, bpf_pt_regs); @@ -605,7 +605,7 @@ static const struct bpf_func_proto bpf_perf_event_output_proto_tp = { .arg2_type = ARG_CONST_MAP_PTR, .arg3_type = ARG_ANYTHING, .arg4_type = ARG_PTR_TO_MEM, - .arg5_type = ARG_CONST_SIZE, + .arg5_type = ARG_CONST_SIZE_OR_ZERO, }; BPF_CALL_3(bpf_get_stackid_tp, void *, tp_buff, struct bpf_map *, map, -- cgit v1.2.3