bpf: Add bpf_user_ringbuf_drain() helper

In a prior change, we added a new BPF_MAP_TYPE_USER_RINGBUF map type which
will allow user-space applications to publish messages to a ring buffer
that is consumed by a BPF program in kernel-space. In order for this
map-type to be useful, it will require a BPF helper function that BPF
programs can invoke to drain samples from the ring buffer, and invoke
callbacks on those samples. This change adds that capability via a new BPF
helper function:

bpf_user_ringbuf_drain(struct bpf_map *map, void *callback_fn, void *ctx,
                       u64 flags)

BPF programs may invoke this function to run callback_fn() on a series of
samples in the ring buffer. callback_fn() has the following signature:

long callback_fn(struct bpf_dynptr *dynptr, void *context);

Samples are provided to the callback in the form of struct bpf_dynptr *'s,
which the program can read using BPF helper functions for querying
struct bpf_dynptr's.

In order to support bpf_ringbuf_drain(), a new PTR_TO_DYNPTR register
type is added to the verifier to reflect a dynptr that was allocated by
a helper function and passed to a BPF program. Unlike PTR_TO_STACK
dynptrs which are allocated on the stack by a BPF program, PTR_TO_DYNPTR
dynptrs need not use reference tracking, as the BPF helper is trusted to
properly free the dynptr before returning. The verifier currently only
supports PTR_TO_DYNPTR registers that are also DYNPTR_TYPE_LOCAL.

Note that while the corresponding user-space libbpf logic will be added
in a subsequent patch, this patch does contain an implementation of the
.map_poll() callback for BPF_MAP_TYPE_USER_RINGBUF maps. This
.map_poll() callback guarantees that an epoll-waiting user-space
producer will receive at least one event notification whenever at least
one sample is drained in an invocation of bpf_user_ringbuf_drain(),
provided that the function is not invoked with the BPF_RB_NO_WAKEUP
flag. If the BPF_RB_FORCE_WAKEUP flag is provided, a wakeup
notification is sent even if no sample was drained.

Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220920000100.477320-3-void@manifault.com

1 files changed, 176 insertions, 5 deletions

diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c
index 754e915748fb..9e832acf4692 100644
--- a/kernel/bpf/ringbuf.c
+++ b/kernel/bpf/ringbuf.c
@@ -38,6 +38,22 @@ struct bpf_ringbuf {
 	struct page **pages;
 	int nr_pages;
 	spinlock_t spinlock ____cacheline_aligned_in_smp;
+	/* For user-space producer ring buffers, an atomic_t busy bit is used
+	 * to synchronize access to the ring buffers in the kernel, rather than
+	 * the spinlock that is used for kernel-producer ring buffers. This is
+	 * done because the ring buffer must hold a lock across a BPF program's
+	 * callback:
+	 *
+	 *    __bpf_user_ringbuf_peek() // lock acquired
+	 * -> program callback_fn()
+	 * -> __bpf_user_ringbuf_sample_release() // lock released
+	 *
+	 * It is unsafe and incorrect to hold an IRQ spinlock across what could
+	 * be a long execution window, so we instead simply disallow concurrent
+	 * access to the ring buffer by kernel consumers, and return -EBUSY from
+	 * __bpf_user_ringbuf_peek() if the busy bit is held by another task.
+	 */
+	atomic_t busy ____cacheline_aligned_in_smp;
 	/* Consumer and producer counters are put into separate pages to
 	 * allow each position to be mapped with different permissions.
 	 * This prevents a user-space application from modifying the
@@ -153,6 +169,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node)
 		return NULL;
 
 	spin_lock_init(&rb->spinlock);
+	atomic_set(&rb->busy, 0);
 	init_waitqueue_head(&rb->waitq);
 	init_irq_work(&rb->work, bpf_ringbuf_notify);
 
@@ -288,8 +305,13 @@ static unsigned long ringbuf_avail_data_sz(struct bpf_ringbuf *rb)
 	return prod_pos - cons_pos;
 }
 
-static __poll_t ringbuf_map_poll(struct bpf_map *map, struct file *filp,
-				 struct poll_table_struct *pts)
+static u32 ringbuf_total_data_sz(const struct bpf_ringbuf *rb)
+{
+	return rb->mask + 1;
+}
+
+static __poll_t ringbuf_map_poll_kern(struct bpf_map *map, struct file *filp,
+				      struct poll_table_struct *pts)
 {
 	struct bpf_ringbuf_map *rb_map;
 
@@ -301,13 +323,26 @@ static __poll_t ringbuf_map_poll(struct bpf_map *map, struct file *filp,
 	return 0;
 }
 
+static __poll_t ringbuf_map_poll_user(struct bpf_map *map, struct file *filp,
+				      struct poll_table_struct *pts)
+{
+	struct bpf_ringbuf_map *rb_map;
+
+	rb_map = container_of(map, struct bpf_ringbuf_map, map);
+	poll_wait(filp, &rb_map->rb->waitq, pts);
+
+	if (ringbuf_avail_data_sz(rb_map->rb) < ringbuf_total_data_sz(rb_map->rb))
+		return EPOLLOUT | EPOLLWRNORM;
+	return 0;
+}
+
 BTF_ID_LIST_SINGLE(ringbuf_map_btf_ids, struct, bpf_ringbuf_map)
 const struct bpf_map_ops ringbuf_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
 	.map_alloc = ringbuf_map_alloc,
 	.map_free = ringbuf_map_free,
 	.map_mmap = ringbuf_map_mmap_kern,
-	.map_poll = ringbuf_map_poll,
+	.map_poll = ringbuf_map_poll_kern,
 	.map_lookup_elem = ringbuf_map_lookup_elem,
 	.map_update_elem = ringbuf_map_update_elem,
 	.map_delete_elem = ringbuf_map_delete_elem,
@@ -321,6 +356,7 @@ const struct bpf_map_ops user_ringbuf_map_ops = {
 	.map_alloc = ringbuf_map_alloc,
 	.map_free = ringbuf_map_free,
 	.map_mmap = ringbuf_map_mmap_user,
+	.map_poll = ringbuf_map_poll_user,
 	.map_lookup_elem = ringbuf_map_lookup_elem,
 	.map_update_elem = ringbuf_map_update_elem,
 	.map_delete_elem = ringbuf_map_delete_elem,
@@ -362,7 +398,7 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 		return NULL;
 
 	len = round_up(size + BPF_RINGBUF_HDR_SZ, 8);
-	if (len > rb->mask + 1)
+	if (len > ringbuf_total_data_sz(rb))
 		return NULL;
 
 	cons_pos = smp_load_acquire(&rb->consumer_pos);
@@ -509,7 +545,7 @@ BPF_CALL_2(bpf_ringbuf_query, struct bpf_map *, map, u64, flags)
 	case BPF_RB_AVAIL_DATA:
 		return ringbuf_avail_data_sz(rb);
 	case BPF_RB_RING_SIZE:
-		return rb->mask + 1;
+		return ringbuf_total_data_sz(rb);
 	case BPF_RB_CONS_POS:
 		return smp_load_acquire(&rb->consumer_pos);
 	case BPF_RB_PROD_POS:
@@ -603,3 +639,138 @@ const struct bpf_func_proto bpf_ringbuf_discard_dynptr_proto = {
 	.arg1_type	= ARG_PTR_TO_DYNPTR | DYNPTR_TYPE_RINGBUF | OBJ_RELEASE,
 	.arg2_type	= ARG_ANYTHING,
 };
+
+static int __bpf_user_ringbuf_peek(struct bpf_ringbuf *rb, void **sample, u32 *size)
+{
+	int err;
+	u32 hdr_len, sample_len, total_len, flags, *hdr;
+	u64 cons_pos, prod_pos;
+
+	/* Synchronizes with smp_store_release() in user-space producer. */
+	prod_pos = smp_load_acquire(&rb->producer_pos);
+	if (prod_pos % 8)
+		return -EINVAL;
+
+	/* Synchronizes with smp_store_release() in __bpf_user_ringbuf_sample_release() */
+	cons_pos = smp_load_acquire(&rb->consumer_pos);
+	if (cons_pos >= prod_pos)
+		return -ENODATA;
+
+	hdr = (u32 *)((uintptr_t)rb->data + (uintptr_t)(cons_pos & rb->mask));
+	/* Synchronizes with smp_store_release() in user-space producer. */
+	hdr_len = smp_load_acquire(hdr);
+	flags = hdr_len & (BPF_RINGBUF_BUSY_BIT | BPF_RINGBUF_DISCARD_BIT);
+	sample_len = hdr_len & ~flags;
+	total_len = round_up(sample_len + BPF_RINGBUF_HDR_SZ, 8);
+
+	/* The sample must fit within the region advertised by the producer position. */
+	if (total_len > prod_pos - cons_pos)
+		return -EINVAL;
+
+	/* The sample must fit within the data region of the ring buffer. */
+	if (total_len > ringbuf_total_data_sz(rb))
+		return -E2BIG;
+
+	/* The sample must fit into a struct bpf_dynptr. */
+	err = bpf_dynptr_check_size(sample_len);
+	if (err)
+		return -E2BIG;
+
+	if (flags & BPF_RINGBUF_DISCARD_BIT) {
+		/* If the discard bit is set, the sample should be skipped.
+		 *
+		 * Update the consumer pos, and return -EAGAIN so the caller
+		 * knows to skip this sample and try to read the next one.
+		 */
+		smp_store_release(&rb->consumer_pos, cons_pos + total_len);
+		return -EAGAIN;
+	}
+
+	if (flags & BPF_RINGBUF_BUSY_BIT)
+		return -ENODATA;
+
+	*sample = (void *)((uintptr_t)rb->data +
+			   (uintptr_t)((cons_pos + BPF_RINGBUF_HDR_SZ) & rb->mask));
+	*size = sample_len;
+	return 0;
+}
+
+static void __bpf_user_ringbuf_sample_release(struct bpf_ringbuf *rb, size_t size, u64 flags)
+{
+	u64 consumer_pos;
+	u32 rounded_size = round_up(size + BPF_RINGBUF_HDR_SZ, 8);
+
+	/* Using smp_load_acquire() is unnecessary here, as the busy-bit
+	 * prevents another task from writing to consumer_pos after it was read
+	 * by this task with smp_load_acquire() in __bpf_user_ringbuf_peek().
+	 */
+	consumer_pos = rb->consumer_pos;
+	 /* Synchronizes with smp_load_acquire() in user-space producer. */
+	smp_store_release(&rb->consumer_pos, consumer_pos + rounded_size);
+}
+
+BPF_CALL_4(bpf_user_ringbuf_drain, struct bpf_map *, map,
+	   void *, callback_fn, void *, callback_ctx, u64, flags)
+{
+	struct bpf_ringbuf *rb;
+	long samples, discarded_samples = 0, ret = 0;
+	bpf_callback_t callback = (bpf_callback_t)callback_fn;
+	u64 wakeup_flags = BPF_RB_NO_WAKEUP | BPF_RB_FORCE_WAKEUP;
+	int busy = 0;
+
+	if (unlikely(flags & ~wakeup_flags))
+		return -EINVAL;
+
+	rb = container_of(map, struct bpf_ringbuf_map, map)->rb;
+
+	/* If another consumer is already consuming a sample, wait for them to finish. */
+	if (!atomic_try_cmpxchg(&rb->busy, &busy, 1))
+		return -EBUSY;
+
+	for (samples = 0; samples < BPF_MAX_USER_RINGBUF_SAMPLES && ret == 0; samples++) {
+		int err;
+		u32 size;
+		void *sample;
+		struct bpf_dynptr_kern dynptr;
+
+		err = __bpf_user_ringbuf_peek(rb, &sample, &size);
+		if (err) {
+			if (err == -ENODATA) {
+				break;
+			} else if (err == -EAGAIN) {
+				discarded_samples++;
+				continue;
+			} else {
+				ret = err;
+				goto schedule_work_return;
+			}
+		}
+
+		bpf_dynptr_init(&dynptr, sample, BPF_DYNPTR_TYPE_LOCAL, 0, size);
+		ret = callback((uintptr_t)&dynptr, (uintptr_t)callback_ctx, 0, 0, 0);
+		__bpf_user_ringbuf_sample_release(rb, size, flags);
+	}
+	ret = samples - discarded_samples;
+
+schedule_work_return:
+	/* Prevent the clearing of the busy-bit from being reordered before the
+	 * storing of any rb consumer or producer positions.
+	 */
+	smp_mb__before_atomic();
+	atomic_set(&rb->busy, 0);
+
+	if (flags & BPF_RB_FORCE_WAKEUP)
+		irq_work_queue(&rb->work);
+	else if (!(flags & BPF_RB_NO_WAKEUP) && samples > 0)
+		irq_work_queue(&rb->work);
+	return ret;
+}
+
+const struct bpf_func_proto bpf_user_ringbuf_drain_proto = {
+	.func		= bpf_user_ringbuf_drain,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_CONST_MAP_PTR,
+	.arg2_type	= ARG_PTR_TO_FUNC,
+	.arg3_type	= ARG_PTR_TO_STACK_OR_NULL,
+	.arg4_type	= ARG_ANYTHING,
+};