[linux-2.6-block.git] / kernel / trace / bpf_trace.c

/* Copyright (c) 2011-2015 PLUMgrid, http://plumgrid.com
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of version 2 of the GNU General Public
 * License as published by the Free Software Foundation.
 */
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/slab.h>
#include <linux/bpf.h>
#include <linux/filter.h>
#include <linux/uaccess.h>
#include <linux/ctype.h>
#include "trace.h"

static DEFINE_PER_CPU(int, bpf_prog_active);

/**
 * trace_call_bpf - invoke BPF program
 * @prog: BPF program
 * @ctx: opaque context pointer
 *
 * kprobe handlers execute BPF programs via this helper.
 * Can be used from static tracepoints in the future.
 *
 * Return: BPF programs always return an integer which is interpreted by
 * kprobe handler as:
 * 0 - return from kprobe (event is filtered out)
 * 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 ret;

	if (in_nmi()) /* not supported yet */
		return 1;

	preempt_disable();

	if (unlikely(__this_cpu_inc_return(bpf_prog_active) != 1)) {
		/*
		 * since some bpf program is already running on this cpu,
		 * don't call into another bpf program (same or different)
		 * and don't send kprobe event into ring-buffer,
		 * so return zero here
		 */
		ret = 0;
		goto out;
	}

	rcu_read_lock();
	ret = BPF_PROG_RUN(prog, ctx);
	rcu_read_unlock();

 out:
	__this_cpu_dec(bpf_prog_active);
	preempt_enable();

	return ret;
}
EXPORT_SYMBOL_GPL(trace_call_bpf);

static u64 bpf_probe_read(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
{
	void *dst = (void *) (long) r1;
	int size = (int) r2;
	void *unsafe_ptr = (void *) (long) r3;

	return probe_kernel_read(dst, unsafe_ptr, size);
}

static const struct bpf_func_proto bpf_probe_read_proto = {
	.func		= bpf_probe_read,
	.gpl_only	= true,
	.ret_type	= RET_INTEGER,
	.arg1_type	= ARG_PTR_TO_STACK,
	.arg2_type	= ARG_CONST_STACK_SIZE,
	.arg3_type	= ARG_ANYTHING,
};

/*
 * limited trace_printk()
 * only %d %u %x %ld %lu %lx %lld %llu %llx %p %s conversion specifiers allowed
 */
static u64 bpf_trace_printk(u64 r1, u64 fmt_size, u64 r3, u64 r4, u64 r5)
{
	char *fmt = (char *) (long) r1;
	bool str_seen = false;
	int mod[3] = {};
	int fmt_cnt = 0;
	u64 unsafe_addr;
	char buf[64];
	int i;

	/*
	 * bpf_check()->check_func_arg()->check_stack_boundary()
	 * guarantees that fmt points to bpf program stack,
	 * fmt_size bytes of it were initialized and fmt_size > 0
	 */
	if (fmt[--fmt_size] != 0)
		return -EINVAL;

	/* check format string for allowed specifiers */
	for (i = 0; i < fmt_size; i++) {
		if ((!isprint(fmt[i]) && !isspace(fmt[i])) || !isascii(fmt[i]))
			return -EINVAL;

		if (fmt[i] != '%')
			continue;

		if (fmt_cnt >= 3)
			return -EINVAL;

		/* fmt[i] != 0 && fmt[last] == 0, so we can access fmt[i + 1] */
		i++;
		if (fmt[i] == 'l') {
			mod[fmt_cnt]++;
			i++;
		} else if (fmt[i] == 'p' || fmt[i] == 's') {
			mod[fmt_cnt]++;
			i++;
			if (!isspace(fmt[i]) && !ispunct(fmt[i]) && fmt[i] != 0)
				return -EINVAL;
			fmt_cnt++;
			if (fmt[i - 1] == 's') {
				if (str_seen)
					/* allow only one '%s' per fmt string */
					return -EINVAL;
				str_seen = true;

				switch (fmt_cnt) {
				case 1:
					unsafe_addr = r3;
					r3 = (long) buf;
					break;
				case 2:
					unsafe_addr = r4;
					r4 = (long) buf;
					break;
				case 3:
					unsafe_addr = r5;
					r5 = (long) buf;
					break;
				}
				buf[0] = 0;
				strncpy_from_unsafe(buf,
						    (void *) (long) unsafe_addr,
						    sizeof(buf));
			}
			continue;
		}

		if (fmt[i] == 'l') {
			mod[fmt_cnt]++;
			i++;
		}

		if (fmt[i] != 'd' && fmt[i] != 'u' && fmt[i] != 'x')
			return -EINVAL;
		fmt_cnt++;
	}

	return __trace_printk(1/* fake ip will not be printed */, fmt,
			      mod[0] == 2 ? r3 : mod[0] == 1 ? (long) r3 : (u32) r3,
			      mod[1] == 2 ? r4 : mod[1] == 1 ? (long) r4 : (u32) r4,
			      mod[2] == 2 ? r5 : mod[2] == 1 ? (long) r5 : (u32) r5);
}

static const struct bpf_func_proto bpf_trace_printk_proto = {
	.func		= bpf_trace_printk,
	.gpl_only	= true,
	.ret_type	= RET_INTEGER,
	.arg1_type	= ARG_PTR_TO_STACK,
	.arg2_type	= ARG_CONST_STACK_SIZE,
};

const struct bpf_func_proto *bpf_get_trace_printk_proto(void)
{
	/*
	 * this program might be calling bpf_trace_printk,
	 * so allocate per-cpu printk buffers
	 */
	trace_printk_init_buffers();

	return &bpf_trace_printk_proto;
}

static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5)
{
	struct bpf_map *map = (struct bpf_map *) (unsigned long) r1;
	struct bpf_array *array = container_of(map, struct bpf_array, map);
	struct perf_event *event;

	if (unlikely(index >= array->map.max_entries))
		return -E2BIG;

	event = (struct perf_event *)array->ptrs[index];
	if (!event)
		return -ENOENT;

	/* make sure event is local and doesn't have pmu::count */
	if (event->oncpu != smp_processor_id() ||
	    event->pmu->count)
		return -EINVAL;

	/*
	 * we don't know if the function is run successfully by the
	 * return value. It can be judged in other places, such as
	 * eBPF programs.
	 */
	return perf_event_read_local(event);
}

static const struct bpf_func_proto bpf_perf_event_read_proto = {
	.func		= bpf_perf_event_read,
	.gpl_only	= true,
	.ret_type	= RET_INTEGER,
	.arg1_type	= ARG_CONST_MAP_PTR,
	.arg2_type	= ARG_ANYTHING,
};

static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 index, u64 r4, u64 size)
{
	struct pt_regs *regs = (struct pt_regs *) (long) r1;
	struct bpf_map *map = (struct bpf_map *) (long) r2;
	struct bpf_array *array = container_of(map, struct bpf_array, map);
	void *data = (void *) (long) r4;
	struct perf_sample_data sample_data;
	struct perf_event *event;
	struct perf_raw_record raw = {
		.size = size,
		.data = data,
	};

	if (unlikely(index >= array->map.max_entries))
		return -E2BIG;

	event = (struct perf_event *)array->ptrs[index];
	if (unlikely(!event))
		return -ENOENT;

	if (unlikely(event->attr.type != PERF_TYPE_SOFTWARE ||
		     event->attr.config != PERF_COUNT_SW_BPF_OUTPUT))
		return -EINVAL;

	if (unlikely(event->oncpu != smp_processor_id()))
		return -EOPNOTSUPP;

	perf_sample_data_init(&sample_data, 0, 0);
	sample_data.raw = &raw;
	perf_event_output(event, &sample_data, regs);
	return 0;
}

static const struct bpf_func_proto bpf_perf_event_output_proto = {
	.func		= bpf_perf_event_output,
	.gpl_only	= true,
	.ret_type	= RET_INTEGER,
	.arg1_type	= ARG_PTR_TO_CTX,
	.arg2_type	= ARG_CONST_MAP_PTR,
	.arg3_type	= ARG_ANYTHING,
	.arg4_type	= ARG_PTR_TO_STACK,
	.arg5_type	= ARG_CONST_STACK_SIZE,
};

static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func_id)
{
	switch (func_id) {
	case BPF_FUNC_map_lookup_elem:
		return &bpf_map_lookup_elem_proto;
	case BPF_FUNC_map_update_elem:
		return &bpf_map_update_elem_proto;
	case BPF_FUNC_map_delete_elem:
		return &bpf_map_delete_elem_proto;
	case BPF_FUNC_probe_read:
		return &bpf_probe_read_proto;
	case BPF_FUNC_ktime_get_ns:
		return &bpf_ktime_get_ns_proto;
	case BPF_FUNC_tail_call:
		return &bpf_tail_call_proto;
	case BPF_FUNC_get_current_pid_tgid:
		return &bpf_get_current_pid_tgid_proto;
	case BPF_FUNC_get_current_uid_gid:
		return &bpf_get_current_uid_gid_proto;
	case BPF_FUNC_get_current_comm:
		return &bpf_get_current_comm_proto;
	case BPF_FUNC_trace_printk:
		return bpf_get_trace_printk_proto();
	case BPF_FUNC_get_smp_processor_id:
		return &bpf_get_smp_processor_id_proto;
	case BPF_FUNC_perf_event_read:
		return &bpf_perf_event_read_proto;
	case BPF_FUNC_perf_event_output:
		return &bpf_perf_event_output_proto;
	default:
		return NULL;
	}
}

/* bpf+kprobe programs can access fields of 'struct pt_regs' */
static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type type)
{
	/* check bounds */
	if (off < 0 || off >= sizeof(struct pt_regs))
		return false;

	/* only read is allowed */
	if (type != BPF_READ)
		return false;

	/* disallow misaligned access */
	if (off % size != 0)
		return false;

	return true;
}

static struct bpf_verifier_ops kprobe_prog_ops = {
	.get_func_proto  = kprobe_prog_func_proto,
	.is_valid_access = kprobe_prog_is_valid_access,
};

static struct bpf_prog_type_list kprobe_tl = {
	.ops	= &kprobe_prog_ops,
	.type	= BPF_PROG_TYPE_KPROBE,
};

static int __init register_kprobe_prog_ops(void)
{
	bpf_register_prog_type(&kprobe_tl);
	return 0;
}
late_initcall(register_kprobe_prog_ops);
Commit	Line	Data
2541517c AS	1	/* Copyright (c) 2011-2015 PLUMgrid, http://plumgrid.com
	2	*
	3	* This program is free software; you can redistribute it and/or
	4	* modify it under the terms of version 2 of the GNU General Public
	5	* License as published by the Free Software Foundation.
	6	*/
	7	#include <linux/kernel.h>
	8	#include <linux/types.h>
	9	#include <linux/slab.h>
	10	#include <linux/bpf.h>
	11	#include <linux/filter.h>
	12	#include <linux/uaccess.h>
9c959c86	13	#include <linux/ctype.h>
2541517c AS	14	#include "trace.h"
	15
	16	static DEFINE_PER_CPU(int, bpf_prog_active);
	17
	18	/**
	19	* trace_call_bpf - invoke BPF program
	20	* @prog: BPF program
	21	* @ctx: opaque context pointer
	22	*
	23	* kprobe handlers execute BPF programs via this helper.
	24	* Can be used from static tracepoints in the future.
	25	*
	26	* Return: BPF programs always return an integer which is interpreted by
	27	* kprobe handler as:
	28	* 0 - return from kprobe (event is filtered out)
	29	* 1 - store kprobe event into ring buffer
	30	* Other values are reserved and currently alias to 1
	31	*/
	32	unsigned int trace_call_bpf(struct bpf_prog prog, void ctx)
	33	{
	34	unsigned int ret;
	35
	36	if (in_nmi()) /* not supported yet */
	37	return 1;
	38
	39	preempt_disable();
	40
	41	if (unlikely(__this_cpu_inc_return(bpf_prog_active) != 1)) {
	42	/*
	43	* since some bpf program is already running on this cpu,
	44	* don't call into another bpf program (same or different)
	45	* and don't send kprobe event into ring-buffer,
	46	* so return zero here
	47	*/
	48	ret = 0;
	49	goto out;
	50	}
	51
	52	rcu_read_lock();
	53	ret = BPF_PROG_RUN(prog, ctx);
	54	rcu_read_unlock();
	55
	56	out:
	57	__this_cpu_dec(bpf_prog_active);
	58	preempt_enable();
	59
	60	return ret;
	61	}
	62	EXPORT_SYMBOL_GPL(trace_call_bpf);
	63
	64	static u64 bpf_probe_read(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
	65	{
	66	void dst = (void ) (long) r1;
	67	int size = (int) r2;
	68	void unsafe_ptr = (void ) (long) r3;
	69
	70	return probe_kernel_read(dst, unsafe_ptr, size);
	71	}
	72
	73	static const struct bpf_func_proto bpf_probe_read_proto = {
	74	.func = bpf_probe_read,
	75	.gpl_only = true,
	76	.ret_type = RET_INTEGER,
	77	.arg1_type = ARG_PTR_TO_STACK,
78	.arg2_type = ARG_CONST_STACK_SIZE,
79	.arg3_type = ARG_ANYTHING,
80	};
81
9c959c86 AS	82	/*
9c959c86 AS	83	* limited trace_printk()
8d3b7dce	84	* only %d %u %x %ld %lu %lx %lld %llu %llx %p %s conversion specifiers allowed
9c959c86 AS	85	*/
	86	static u64 bpf_trace_printk(u64 r1, u64 fmt_size, u64 r3, u64 r4, u64 r5)
	87	{
	88	char fmt = (char ) (long) r1;
8d3b7dce	89	bool str_seen = false;
9c959c86 AS	90	int mod[3] = {};
9c959c86 AS	91	int fmt_cnt = 0;
8d3b7dce AS	92	u64 unsafe_addr;
8d3b7dce AS	93	char buf[64];
9c959c86 AS	94	int i;
	95
	96	/*
	97	* bpf_check()->check_func_arg()->check_stack_boundary()
	98	* guarantees that fmt points to bpf program stack,
	99	* fmt_size bytes of it were initialized and fmt_size > 0
	100	*/
	101	if (fmt[--fmt_size] != 0)
	102	return -EINVAL;
	103
	104	/* check format string for allowed specifiers */
	105	for (i = 0; i < fmt_size; i++) {
	106	if ((!isprint(fmt[i]) && !isspace(fmt[i])) \|\| !isascii(fmt[i]))
	107	return -EINVAL;
	108
	109	if (fmt[i] != '%')
	110	continue;
	111
	112	if (fmt_cnt >= 3)
	113	return -EINVAL;
	114
	115	/* fmt[i] != 0 && fmt[last] == 0, so we can access fmt[i + 1] */
	116	i++;
	117	if (fmt[i] == 'l') {
	118	mod[fmt_cnt]++;
	119	i++;
8d3b7dce	120	} else if (fmt[i] == 'p' \|\| fmt[i] == 's') {
9c959c86 AS	121	mod[fmt_cnt]++;
	122	i++;
	123	if (!isspace(fmt[i]) && !ispunct(fmt[i]) && fmt[i] != 0)
	124	return -EINVAL;
	125	fmt_cnt++;
8d3b7dce AS	126	if (fmt[i - 1] == 's') {
	127	if (str_seen)
	128	/* allow only one '%s' per fmt string */
	129	return -EINVAL;
	130	str_seen = true;
	131
	132	switch (fmt_cnt) {
	133	case 1:
	134	unsafe_addr = r3;
	135	r3 = (long) buf;
	136	break;
	137	case 2:
	138	unsafe_addr = r4;
	139	r4 = (long) buf;
	140	break;
	141	case 3:
	142	unsafe_addr = r5;
	143	r5 = (long) buf;
	144	break;
	145	}
	146	buf[0] = 0;
	147	strncpy_from_unsafe(buf,
	148	(void *) (long) unsafe_addr,
	149	sizeof(buf));
	150	}
9c959c86 AS	151	continue;
	152	}
	153
	154	if (fmt[i] == 'l') {
	155	mod[fmt_cnt]++;
	156	i++;
	157	}
	158
	159	if (fmt[i] != 'd' && fmt[i] != 'u' && fmt[i] != 'x')
	160	return -EINVAL;
	161	fmt_cnt++;
	162	}
	163
	164	return __trace_printk(1/* fake ip will not be printed */, fmt,
	165	mod[0] == 2 ? r3 : mod[0] == 1 ? (long) r3 : (u32) r3,
	166	mod[1] == 2 ? r4 : mod[1] == 1 ? (long) r4 : (u32) r4,
	167	mod[2] == 2 ? r5 : mod[2] == 1 ? (long) r5 : (u32) r5);
	168	}
	169
	170	static const struct bpf_func_proto bpf_trace_printk_proto = {
	171	.func = bpf_trace_printk,
	172	.gpl_only = true,
	173	.ret_type = RET_INTEGER,
	174	.arg1_type = ARG_PTR_TO_STACK,
	175	.arg2_type = ARG_CONST_STACK_SIZE,
	176	};
	177
0756ea3e AS	178	const struct bpf_func_proto *bpf_get_trace_printk_proto(void)
	179	{
	180	/*
	181	* this program might be calling bpf_trace_printk,
	182	* so allocate per-cpu printk buffers
	183	*/
	184	trace_printk_init_buffers();
	185
	186	return &bpf_trace_printk_proto;
	187	}
	188
35578d79 KX	189	static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5)
	190	{
	191	struct bpf_map map = (struct bpf_map ) (unsigned long) r1;
	192	struct bpf_array *array = container_of(map, struct bpf_array, map);
	193	struct perf_event *event;
	194
	195	if (unlikely(index >= array->map.max_entries))
	196	return -E2BIG;
	197
	198	event = (struct perf_event *)array->ptrs[index];
	199	if (!event)
	200	return -ENOENT;
	201
62544ce8 AS	202	/* make sure event is local and doesn't have pmu::count */
	203	if (event->oncpu != smp_processor_id() \|\|
	204	event->pmu->count)
	205	return -EINVAL;
	206
35578d79 KX	207	/*
	208	* we don't know if the function is run successfully by the
	209	* return value. It can be judged in other places, such as
	210	* eBPF programs.
	211	*/
	212	return perf_event_read_local(event);
	213	}
	214
62544ce8	215	static const struct bpf_func_proto bpf_perf_event_read_proto = {
35578d79	216	.func = bpf_perf_event_read,
1075ef59	217	.gpl_only = true,
35578d79 KX	218	.ret_type = RET_INTEGER,
	219	.arg1_type = ARG_CONST_MAP_PTR,
	220	.arg2_type = ARG_ANYTHING,
	221	};
	222
a43eec30 AS	223	static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 index, u64 r4, u64 size)
	224	{
	225	struct pt_regs regs = (struct pt_regs ) (long) r1;
	226	struct bpf_map map = (struct bpf_map ) (long) r2;
	227	struct bpf_array *array = container_of(map, struct bpf_array, map);
	228	void data = (void ) (long) r4;
	229	struct perf_sample_data sample_data;
	230	struct perf_event *event;
	231	struct perf_raw_record raw = {
	232	.size = size,
	233	.data = data,
	234	};
	235
	236	if (unlikely(index >= array->map.max_entries))
	237	return -E2BIG;
	238
	239	event = (struct perf_event *)array->ptrs[index];
	240	if (unlikely(!event))
	241	return -ENOENT;
	242
	243	if (unlikely(event->attr.type != PERF_TYPE_SOFTWARE \|\|
	244	event->attr.config != PERF_COUNT_SW_BPF_OUTPUT))
	245	return -EINVAL;
	246
	247	if (unlikely(event->oncpu != smp_processor_id()))
	248	return -EOPNOTSUPP;
	249
	250	perf_sample_data_init(&sample_data, 0, 0);
	251	sample_data.raw = &raw;
	252	perf_event_output(event, &sample_data, regs);
	253	return 0;
	254	}
	255
	256	static const struct bpf_func_proto bpf_perf_event_output_proto = {
	257	.func = bpf_perf_event_output,
1075ef59	258	.gpl_only = true,
a43eec30 AS	259	.ret_type = RET_INTEGER,
	260	.arg1_type = ARG_PTR_TO_CTX,
	261	.arg2_type = ARG_CONST_MAP_PTR,
	262	.arg3_type = ARG_ANYTHING,
	263	.arg4_type = ARG_PTR_TO_STACK,
	264	.arg5_type = ARG_CONST_STACK_SIZE,
	265	};
	266
2541517c AS	267	static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func_id)
	268	{
	269	switch (func_id) {
	270	case BPF_FUNC_map_lookup_elem:
	271	return &bpf_map_lookup_elem_proto;
	272	case BPF_FUNC_map_update_elem:
	273	return &bpf_map_update_elem_proto;
	274	case BPF_FUNC_map_delete_elem:
	275	return &bpf_map_delete_elem_proto;
	276	case BPF_FUNC_probe_read:
	277	return &bpf_probe_read_proto;
d9847d31 AS	278	case BPF_FUNC_ktime_get_ns:
d9847d31 AS	279	return &bpf_ktime_get_ns_proto;
04fd61ab AS	280	case BPF_FUNC_tail_call:
04fd61ab AS	281	return &bpf_tail_call_proto;
ffeedafb AS	282	case BPF_FUNC_get_current_pid_tgid:
	283	return &bpf_get_current_pid_tgid_proto;
	284	case BPF_FUNC_get_current_uid_gid:
	285	return &bpf_get_current_uid_gid_proto;
	286	case BPF_FUNC_get_current_comm:
	287	return &bpf_get_current_comm_proto;
9c959c86	288	case BPF_FUNC_trace_printk:
0756ea3e	289	return bpf_get_trace_printk_proto();
ab1973d3 AS	290	case BPF_FUNC_get_smp_processor_id:
ab1973d3 AS	291	return &bpf_get_smp_processor_id_proto;
35578d79 KX	292	case BPF_FUNC_perf_event_read:
35578d79 KX	293	return &bpf_perf_event_read_proto;
a43eec30 AS	294	case BPF_FUNC_perf_event_output:
a43eec30 AS	295	return &bpf_perf_event_output_proto;
2541517c AS	296	default:
	297	return NULL;
	298	}
	299	}
	300
	301	/* bpf+kprobe programs can access fields of 'struct pt_regs' */
	302	static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type type)
	303	{
	304	/* check bounds */
	305	if (off < 0 \|\| off >= sizeof(struct pt_regs))
	306	return false;
	307
	308	/* only read is allowed */
	309	if (type != BPF_READ)
	310	return false;
	311
	312	/* disallow misaligned access */
	313	if (off % size != 0)
	314	return false;
	315
	316	return true;
	317	}
	318
	319	static struct bpf_verifier_ops kprobe_prog_ops = {
	320	.get_func_proto = kprobe_prog_func_proto,
	321	.is_valid_access = kprobe_prog_is_valid_access,
	322	};
	323
	324	static struct bpf_prog_type_list kprobe_tl = {
	325	.ops = &kprobe_prog_ops,
	326	.type = BPF_PROG_TYPE_KPROBE,
	327	};
	328
	329	static int __init register_kprobe_prog_ops(void)
	330	{
	331	bpf_register_prog_type(&kprobe_tl);
	332	return 0;
	333	}
	334	late_initcall(register_kprobe_prog_ops);