public-mirrors/linux
1
0
Fork 0
mirror of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2025-08-05 16:54:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
linux/tools/lib/bpf/ringbuf.c

686 lines
16 KiB
C
Raw Permalink Normal View History

libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
/*
* Ring buffer operations.
*
* Copyright (C) 2020 Facebook, Inc.
*/
libbpf: Add _GNU_SOURCE for reallocarray to ringbuf.c On systems with recent enough glibc, reallocarray compat won't kick in, so reallocarray() itself has to come from stdlib.h include. But _GNU_SOURCE is necessary to enable it. So add it. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20200601202601.2139477-1-andriin@fb.com
2020-06-01 13:26:01 -07:00
#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
#include <stdlib.h>
#include <stdio.h>
#include <errno.h>
#include <unistd.h>
#include <linux/err.h>
#include <linux/bpf.h>
#include <asm/barrier.h>
#include <sys/mman.h>
#include <sys/epoll.h>
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
#include <time.h>
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
#include "libbpf.h"
#include "libbpf_internal.h"
#include "bpf.h"
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
#include "str_error.h"
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
struct ring {
ring_buffer_sample_fn sample_cb;
void *ctx;
void *data;
unsigned long *consumer_pos;
unsigned long *producer_pos;
unsigned long mask;
int map_fd;
};
struct ring_buffer {
struct epoll_event *events;
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
struct ring **rings;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
size_t page_size;
int epoll_fd;
int ring_cnt;
};
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
struct user_ring_buffer {
struct epoll_event event;
unsigned long *consumer_pos;
unsigned long *producer_pos;
void *data;
unsigned long mask;
size_t page_size;
int map_fd;
int epoll_fd;
};
/* 8-byte ring buffer header structure */
struct ringbuf_hdr {
__u32 len;
__u32 pad;
};
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
static void ringbuf_free_ring(struct ring_buffer *rb, struct ring *r)
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
{
if (r->consumer_pos) {
munmap(r->consumer_pos, rb->page_size);
r->consumer_pos = NULL;
}
if (r->producer_pos) {
munmap(r->producer_pos, rb->page_size + 2 * (r->mask + 1));
r->producer_pos = NULL;
}
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
free(r);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
/* Add extra RINGBUF maps to this ring buffer manager */
int ring_buffer__add(struct ring_buffer *rb, int map_fd,
ring_buffer_sample_fn sample_cb, void *ctx)
{
struct bpf_map_info info;
__u32 len = sizeof(info);
struct epoll_event *e;
struct ring *r;
libbpf: Handle size overflow for ringbuf mmap The maximum size of ringbuf is 2GB on x86-64 host, so 2 * max_entries will overflow u32 when mapping producer page and data pages. Only casting max_entries to size_t is not enough, because for 32-bits application on 64-bits kernel the size of read-only mmap region also could overflow size_t. So fixing it by casting the size of read-only mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-3-houtao@huaweicloud.com
2022-11-16 15:23:49 +08:00
__u64 mmap_sz;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
void *tmp;
int err;
memset(&info, 0, sizeof(info));
libbpf: Use bpf_{btf,link,map,prog}_get_info_by_fd() Use the new type-safe wrappers around bpf_obj_get_info_by_fd(). Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230214231221.249277-3-iii@linux.ibm.com
2023-02-15 00:12:15 +01:00
err = bpf_map_get_info_by_fd(map_fd, &info, &len);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (err) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("ringbuf: failed to get map info for fd=%d: %s\n",
map_fd, errstr(err));
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(err);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
if (info.type != BPF_MAP_TYPE_RINGBUF) {
pr_warn("ringbuf: map fd=%d is not BPF_MAP_TYPE_RINGBUF\n",
map_fd);
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(-EINVAL);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
libbpf: Remove any use of reallocarray() in libbpf Re-implement glibc's reallocarray() for libbpf internal-only use. reallocarray(), unfortunately, is not available in all versions of glibc, so requires extra feature detection and using reallocarray() stub from <tools/libc_compat.h> and COMPAT_NEED_REALLOCARRAY. All this complicates build of libbpf unnecessarily and is just a maintenance burden. Instead, it's trivial to implement libbpf-specific internal version and use it throughout libbpf. Which is what this patch does, along with converting some realloc() uses that should really have been reallocarray() in the first place. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/20200819013607.3607269-2-andriin@fb.com
2020-08-18 18:36:04 -07:00
tmp = libbpf_reallocarray(rb->rings, rb->ring_cnt + 1, sizeof(*rb->rings));
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (!tmp)
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(-ENOMEM);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
rb->rings = tmp;
libbpf: Remove any use of reallocarray() in libbpf Re-implement glibc's reallocarray() for libbpf internal-only use. reallocarray(), unfortunately, is not available in all versions of glibc, so requires extra feature detection and using reallocarray() stub from <tools/libc_compat.h> and COMPAT_NEED_REALLOCARRAY. All this complicates build of libbpf unnecessarily and is just a maintenance burden. Instead, it's trivial to implement libbpf-specific internal version and use it throughout libbpf. Which is what this patch does, along with converting some realloc() uses that should really have been reallocarray() in the first place. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/20200819013607.3607269-2-andriin@fb.com
2020-08-18 18:36:04 -07:00
tmp = libbpf_reallocarray(rb->events, rb->ring_cnt + 1, sizeof(*rb->events));
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (!tmp)
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(-ENOMEM);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
rb->events = tmp;
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
r = calloc(1, sizeof(*r));
if (!r)
return libbpf_err(-ENOMEM);
rb->rings[rb->ring_cnt] = r;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
r->map_fd = map_fd;
r->sample_cb = sample_cb;
r->ctx = ctx;
r->mask = info.max_entries - 1;
/* Map writable consumer page */
libbpf: Handle size overflow for ringbuf mmap The maximum size of ringbuf is 2GB on x86-64 host, so 2 * max_entries will overflow u32 when mapping producer page and data pages. Only casting max_entries to size_t is not enough, because for 32-bits application on 64-bits kernel the size of read-only mmap region also could overflow size_t. So fixing it by casting the size of read-only mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-3-houtao@huaweicloud.com
2022-11-16 15:23:49 +08:00
tmp = mmap(NULL, rb->page_size, PROT_READ | PROT_WRITE, MAP_SHARED, map_fd, 0);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (tmp == MAP_FAILED) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("ringbuf: failed to mmap consumer page for map fd=%d: %s\n",
map_fd, errstr(err));
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
goto err_out;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
r->consumer_pos = tmp;
/* Map read-only producer page and data pages. We map twice as big
* data size to allow simple reading of samples that wrap around the
* end of a ring buffer. See kernel implementation for details.
libbpf: checkpatch: Fixed code alignments in ringbuf.c Fixed some checkpatch issues in ringbuf.c Signed-off-by: Kang Minchul <tegongkang@gmail.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Stanislav Fomichev <sdf@google.com> Link: https://lore.kernel.org/bpf/20221113190648.38556-4-tegongkang@gmail.com
2022-11-14 04:06:48 +09:00
*/
libbpf: Handle size overflow for ringbuf mmap The maximum size of ringbuf is 2GB on x86-64 host, so 2 * max_entries will overflow u32 when mapping producer page and data pages. Only casting max_entries to size_t is not enough, because for 32-bits application on 64-bits kernel the size of read-only mmap region also could overflow size_t. So fixing it by casting the size of read-only mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-3-houtao@huaweicloud.com
2022-11-16 15:23:49 +08:00
mmap_sz = rb->page_size + 2 * (__u64)info.max_entries;
if (mmap_sz != (__u64)(size_t)mmap_sz) {
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
err = -E2BIG;
libbpf: Handle size overflow for ringbuf mmap The maximum size of ringbuf is 2GB on x86-64 host, so 2 * max_entries will overflow u32 when mapping producer page and data pages. Only casting max_entries to size_t is not enough, because for 32-bits application on 64-bits kernel the size of read-only mmap region also could overflow size_t. So fixing it by casting the size of read-only mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-3-houtao@huaweicloud.com
2022-11-16 15:23:49 +08:00
pr_warn("ringbuf: ring buffer size (%u) is too big\n", info.max_entries);
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
goto err_out;
libbpf: Handle size overflow for ringbuf mmap The maximum size of ringbuf is 2GB on x86-64 host, so 2 * max_entries will overflow u32 when mapping producer page and data pages. Only casting max_entries to size_t is not enough, because for 32-bits application on 64-bits kernel the size of read-only mmap region also could overflow size_t. So fixing it by casting the size of read-only mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-3-houtao@huaweicloud.com
2022-11-16 15:23:49 +08:00
}
tmp = mmap(NULL, (size_t)mmap_sz, PROT_READ, MAP_SHARED, map_fd, rb->page_size);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (tmp == MAP_FAILED) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("ringbuf: failed to mmap data pages for map fd=%d: %s\n",
map_fd, errstr(err));
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
goto err_out;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
r->producer_pos = tmp;
r->data = tmp + rb->page_size;
e = &rb->events[rb->ring_cnt];
memset(e, 0, sizeof(*e));
e->events = EPOLLIN;
e->data.fd = rb->ring_cnt;
if (epoll_ctl(rb->epoll_fd, EPOLL_CTL_ADD, map_fd, e) < 0) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("ringbuf: failed to epoll add map fd=%d: %s\n",
map_fd, errstr(err));
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
goto err_out;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
rb->ring_cnt++;
return 0;
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
err_out:
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
ringbuf_free_ring(rb, r);
libbpf: Refactor cleanup in ring_buffer__add Refactor the cleanup code in ring_buffer__add to use a unified err_out label. This reduces code duplication, as well as plugging a potential leak if mmap_sz != (__u64)(size_t)mmap_sz (currently this would miss unmapping tmp because ringbuf_unmap_ring isn't called). Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-2-martin.kelly@crowdstrike.com
2023-09-25 14:50:32 -07:00
return libbpf_err(err);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
void ring_buffer__free(struct ring_buffer *rb)
{
int i;
if (!rb)
return;
for (i = 0; i < rb->ring_cnt; ++i)
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
ringbuf_free_ring(rb, rb->rings[i]);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (rb->epoll_fd >= 0)
close(rb->epoll_fd);
free(rb->events);
free(rb->rings);
free(rb);
}
struct ring_buffer *
ring_buffer__new(int map_fd, ring_buffer_sample_fn sample_cb, void *ctx,
const struct ring_buffer_opts *opts)
{
struct ring_buffer *rb;
int err;
if (!OPTS_VALID(opts, ring_buffer_opts))
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return errno = EINVAL, NULL;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
rb = calloc(1, sizeof(*rb));
if (!rb)
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return errno = ENOMEM, NULL;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
rb->page_size = getpagesize();
rb->epoll_fd = epoll_create1(EPOLL_CLOEXEC);
if (rb->epoll_fd < 0) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("ringbuf: failed to create epoll instance: %s\n", errstr(err));
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
goto err_out;
}
err = ring_buffer__add(rb, map_fd, sample_cb, ctx);
if (err)
goto err_out;
return rb;
err_out:
ring_buffer__free(rb);
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return errno = -err, NULL;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
static inline int roundup_len(__u32 len)
{
/* clear out top 2 bits (discard and busy, if set) */
len <<= 2;
len >>= 2;
/* add length prefix */
len += BPF_RINGBUF_HDR_SZ;
/* round up to 8 byte alignment */
return (len + 7) / 8 * 8;
}
libbpf: ringbuf: Allow to consume up to a certain amount of items In some cases, instead of always consuming all items from ring buffers in a greedy way, we may want to consume up to a certain amount of items, for example when we need to copy items from the BPF ring buffer to a limited user buffer. This change allows to set an upper limit to the amount of items consumed from one or more ring buffers. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240406092005.92399-3-andrea.righi@canonical.com
2024-04-06 11:15:42 +02:00
static int64_t ringbuf_process_ring(struct ring *r, size_t n)
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
{
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
int *len_ptr, len, err;
/* 64-bit to avoid overflow in case of extreme application behavior */
int64_t cnt = 0;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
unsigned long cons_pos, prod_pos;
bool got_new_data;
void *sample;
cons_pos = smp_load_acquire(r->consumer_pos);
do {
got_new_data = false;
prod_pos = smp_load_acquire(r->producer_pos);
while (cons_pos < prod_pos) {
len_ptr = r->data + (cons_pos & r->mask);
len = smp_load_acquire(len_ptr);
/* sample not committed yet, bail out for now */
if (len & BPF_RINGBUF_BUSY_BIT)
goto done;
got_new_data = true;
cons_pos += roundup_len(len);
if ((len & BPF_RINGBUF_DISCARD_BIT) == 0) {
sample = (void *)len_ptr + BPF_RINGBUF_HDR_SZ;
err = r->sample_cb(r->ctx, sample, len);
libbpf: Fix bail out from 'ringbuf_process_ring()' on error The current code bails out with negative and positive returns. If the callback returns a positive return code, 'ring_buffer__consume()' and 'ring_buffer__poll()' will return a spurious number of records consumed, but mostly important will continue the processing loop. This patch makes positive returns from the callback a no-op. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Signed-off-by: Pedro Tammela <pctammela@mojatatu.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210325150115.138750-1-pctammela@mojatatu.com
2021-03-25 12:01:15 -03:00
if (err < 0) {
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
/* update consumer pos and bail out */
smp_store_release(r->consumer_pos,
cons_pos);
return err;
}
cnt++;
}
smp_store_release(r->consumer_pos, cons_pos);
libbpf: ringbuf: Allow to consume up to a certain amount of items In some cases, instead of always consuming all items from ring buffers in a greedy way, we may want to consume up to a certain amount of items, for example when we need to copy items from the BPF ring buffer to a limited user buffer. This change allows to set an upper limit to the amount of items consumed from one or more ring buffers. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240406092005.92399-3-andrea.righi@canonical.com
2024-04-06 11:15:42 +02:00
if (cnt >= n)
goto done;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
} while (got_new_data);
done:
return cnt;
}
libbpf: Add ring__consume_n / ring_buffer__consume_n Introduce a new API to consume items from a ring buffer, limited to a specified amount, and return to the caller the actual number of items consumed. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/lkml/20240310154726.734289-1-andrea.righi@canonical.com/T Link: https://lore.kernel.org/bpf/20240406092005.92399-4-andrea.righi@canonical.com
2024-04-06 11:15:43 +02:00
/* Consume available ring buffer(s) data without event polling, up to n
* records.
*
* Returns number of records consumed across all registered ring buffers (or
* n, whichever is less), or negative number if any of the callbacks return
* error.
*/
int ring_buffer__consume_n(struct ring_buffer *rb, size_t n)
{
int64_t err, res = 0;
int i;
for (i = 0; i < rb->ring_cnt; i++) {
struct ring *ring = rb->rings[i];
err = ringbuf_process_ring(ring, n);
if (err < 0)
return libbpf_err(err);
res += err;
n -= err;
if (n == 0)
break;
}
libbpf: fix ring_buffer__consume_n() return result logic Add INT_MAX check to ring_buffer__consume_n(). We do the similar check to handle int return result of all these ring buffer APIs in other APIs and ring_buffer__consume_n() is missing one. This patch fixes this omission. Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20240430201952.888293-2-andrii@kernel.org Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2024-04-30 13:19:52 -07:00
return res > INT_MAX ? INT_MAX : res;
libbpf: Add ring__consume_n / ring_buffer__consume_n Introduce a new API to consume items from a ring buffer, limited to a specified amount, and return to the caller the actual number of items consumed. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/lkml/20240310154726.734289-1-andrea.righi@canonical.com/T Link: https://lore.kernel.org/bpf/20240406092005.92399-4-andrea.righi@canonical.com
2024-04-06 11:15:43 +02:00
}
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
/* Consume available ring buffer(s) data without event polling.
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
* Returns number of records consumed across all registered ring buffers (or
* INT_MAX, whichever is less), or negative number if any of the callbacks
* return error.
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
*/
int ring_buffer__consume(struct ring_buffer *rb)
{
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
int64_t err, res = 0;
int i;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
for (i = 0; i < rb->ring_cnt; i++) {
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
struct ring *ring = rb->rings[i];
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
libbpf: ringbuf: Allow to consume up to a certain amount of items In some cases, instead of always consuming all items from ring buffers in a greedy way, we may want to consume up to a certain amount of items, for example when we need to copy items from the BPF ring buffer to a limited user buffer. This change allows to set an upper limit to the amount of items consumed from one or more ring buffers. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240406092005.92399-3-andrea.righi@canonical.com
2024-04-06 11:15:42 +02:00
err = ringbuf_process_ring(ring, INT_MAX);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (err < 0)
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(err);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
res += err;
libbpf: ringbuf: Allow to consume up to a certain amount of items In some cases, instead of always consuming all items from ring buffers in a greedy way, we may want to consume up to a certain amount of items, for example when we need to copy items from the BPF ring buffer to a limited user buffer. This change allows to set an upper limit to the amount of items consumed from one or more ring buffers. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240406092005.92399-3-andrea.righi@canonical.com
2024-04-06 11:15:42 +02:00
if (res > INT_MAX) {
res = INT_MAX;
break;
}
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
return res;
}
/* Poll for available data and consume records, if any are available.
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
* Returns number of records consumed (or INT_MAX, whichever is less), or
* negative number, if any of the registered callbacks returned error.
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
*/
int ring_buffer__poll(struct ring_buffer *rb, int timeout_ms)
{
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
int i, cnt;
int64_t err, res = 0;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
cnt = epoll_wait(rb->epoll_fd, rb->events, rb->ring_cnt, timeout_ms);
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
if (cnt < 0)
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(-errno);
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
for (i = 0; i < cnt; i++) {
__u32 ring_id = rb->events[i].data.fd;
libbpf: Switch rings to array of pointers Switch rb->rings to be an array of pointers instead of a contiguous block. This allows for each ring pointer to be stable after ring_buffer__add is called, which allows us to expose struct ring * to the user without gotchas. Without this change, the realloc in ring_buffer__add could invalidate a struct ring *, making it unsafe to give to the user. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-3-martin.kelly@crowdstrike.com
2023-09-25 14:50:33 -07:00
struct ring *ring = rb->rings[ring_id];
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
libbpf: ringbuf: Allow to consume up to a certain amount of items In some cases, instead of always consuming all items from ring buffers in a greedy way, we may want to consume up to a certain amount of items, for example when we need to copy items from the BPF ring buffer to a limited user buffer. This change allows to set an upper limit to the amount of items consumed from one or more ring buffers. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240406092005.92399-3-andrea.righi@canonical.com
2024-04-06 11:15:42 +02:00
err = ringbuf_process_ring(ring, INT_MAX);
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
if (err < 0)
libbpf: Streamline error reporting for high-level APIs Implement changes to error reporting for high-level libbpf APIs to make them less surprising and less error-prone to users: - in all the cases when error happens, errno is set to an appropriate error value; - in libbpf 1.0 mode, all pointer-returning APIs return NULL on error and error code is communicated through errno; this applies both to APIs that already returned NULL before (so now they communicate more detailed error codes), as well as for many APIs that used ERR_PTR() macro and encoded error numbers as fake pointers. - in legacy (default) mode, those APIs that were returning ERR_PTR(err), continue doing so, but still set errno. With these changes, errno can be always used to extract actual error, regardless of legacy or libbpf 1.0 modes. This is utilized internally in libbpf in places where libbpf uses it's own high-level APIs. libbpf_get_error() is adapted to handle both cases completely transparently to end-users (and is used by libbpf consistently as well). More context, justification, and discussion can be found in "Libbpf: the road to v1.0" document ([0]). [0] https://docs.google.com/document/d/1UyjTZuPFWiPFyKk1tV5an11_iaRuec6U-ZESZ54nNTY Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: John Fastabend <john.fastabend@gmail.com> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210525035935.1461796-5-andrii@kernel.org
2021-05-24 20:59:34 -07:00
return libbpf_err(err);
libbpf: Fix ring_buffer__poll() to return number of consumed samples Fix ring_buffer__poll() to return the number of non-discarded records consumed, just like its documentation states. It's also consistent with ring_buffer__consume() return. Fix up selftests with wrong expected results. Fixes: bf99c936f947 ("libbpf: Add BPF ring buffer support") Fixes: cb1c9ddd5525 ("selftests/bpf: Add BPF ringbuf selftests") Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/20201130223336.904192-1-andrii@kernel.org
2020-11-30 14:33:35 -08:00
res += err;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
if (res > INT_MAX)
libbpf: ringbuf: Allow to consume up to a certain amount of items In some cases, instead of always consuming all items from ring buffers in a greedy way, we may want to consume up to a certain amount of items, for example when we need to copy items from the BPF ring buffer to a limited user buffer. This change allows to set an upper limit to the amount of items consumed from one or more ring buffers. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240406092005.92399-3-andrea.righi@canonical.com
2024-04-06 11:15:42 +02:00
res = INT_MAX;
libbpf: Fix signed overflow in ringbuf_process_ring One of our benchmarks running in (Google-internal) CI pushes data through the ringbuf faster htan than userspace is able to consume it. In this case it seems we're actually able to get >INT_MAX entries in a single ring_buffer__consume() call. ASAN detected that cnt overflows in this case. Fix by using 64-bit counter internally and then capping the result to INT_MAX before converting to the int return type. Do the same for the ring_buffer__poll(). Fixes: bf99c936f947 (libbpf: Add BPF ring buffer support) Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210429130510.1621665-1-jackmanb@google.com
2021-04-29 13:05:10 +00:00
return res;
libbpf: Add BPF ring buffer support Declaring and instantiating BPF ring buffer doesn't require any changes to libbpf, as it's just another type of maps. So using existing BTF-defined maps syntax with __uint(type, BPF_MAP_TYPE_RINGBUF) and __uint(max_elements, <size-of-ring-buf>) is all that's necessary to create and use BPF ring buffer. This patch adds BPF ring buffer consumer to libbpf. It is very similar to perf_buffer implementation in terms of API, but also attempts to fix some minor problems and inconveniences with existing perf_buffer API. ring_buffer support both single ring buffer use case (with just using ring_buffer__new()), as well as allows to add more ring buffers, each with its own callback and context. This allows to efficiently poll and consume multiple, potentially completely independent, ring buffers, using single epoll instance. The latter is actually a problem in practice for applications that are using multiple sets of perf buffers. They have to create multiple instances for struct perf_buffer and poll them independently or in a loop, each approach having its own problems (e.g., inability to use a common poll timeout). struct ring_buffer eliminates this problem by aggregating many independent ring buffer instances under the single "ring buffer manager". Second, perf_buffer's callback can't return error, so applications that need to stop polling due to error in data or data signalling the end, have to use extra mechanisms to signal that polling has to stop. ring_buffer's callback can return error, which will be passed through back to user code and can be acted upon appropariately. Two APIs allow to consume ring buffer data: - ring_buffer__poll(), which will wait for data availability notification and will consume data only from reported ring buffer(s); this API allows to efficiently use resources by reading data only when it becomes available; - ring_buffer__consume(), will attempt to read new records regardless of data availablity notification sub-system. This API is useful for cases when lowest latency is required, in expense of burning CPU resources. Signed-off-by: Andrii Nakryiko <andriin@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200529075424.3139988-3-andriin@fb.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2020-05-29 00:54:21 -07:00
}
libbpf: Expose libbpf ring_buffer epoll_fd This provides a convenient perf ringbuf -> libbpf ringbuf migration path for users of external polling systems. It is analogous to perf_buffer__epoll_fd. Signed-off-by: Brendan Jackman <jackmanb@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20201214113812.305274-1-jackmanb@google.com
2020-12-14 11:38:12 +00:00
/* Get an fd that can be used to sleep until data is available in the ring(s) */
int ring_buffer__epoll_fd(const struct ring_buffer *rb)
{
return rb->epoll_fd;
}
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
libbpf: Add ring_buffer__ring Add a new function ring_buffer__ring, which exposes struct ring * to the user, representing a single ringbuffer. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-4-martin.kelly@crowdstrike.com
2023-09-25 14:50:34 -07:00
struct ring *ring_buffer__ring(struct ring_buffer *rb, unsigned int idx)
{
if (idx >= rb->ring_cnt)
return errno = ERANGE, NULL;
return rb->rings[idx];
}
libbpf: Add ring__producer_pos, ring__consumer_pos Add APIs to get the producer and consumer position for a given ringbuffer. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-6-martin.kelly@crowdstrike.com
2023-09-25 14:50:36 -07:00
unsigned long ring__consumer_pos(const struct ring *r)
{
/* Synchronizes with smp_store_release() in ringbuf_process_ring(). */
return smp_load_acquire(r->consumer_pos);
}
unsigned long ring__producer_pos(const struct ring *r)
{
/* Synchronizes with smp_store_release() in __bpf_ringbuf_reserve() in
* the kernel.
*/
return smp_load_acquire(r->producer_pos);
}
libbpf: Add ring__avail_data_size Add ring__avail_data_size for querying the currently available data in the ringbuffer, similar to the BPF_RB_AVAIL_DATA flag in bpf_ringbuf_query. This is racy during ongoing operations but is still useful for overall information on how a ringbuffer is behaving. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-8-martin.kelly@crowdstrike.com
2023-09-25 14:50:38 -07:00
size_t ring__avail_data_size(const struct ring *r)
{
unsigned long cons_pos, prod_pos;
cons_pos = ring__consumer_pos(r);
prod_pos = ring__producer_pos(r);
return prod_pos - cons_pos;
}
libbpf: Add ring__size Add ring__size to get the total size of a given ringbuffer. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-10-martin.kelly@crowdstrike.com
2023-09-25 14:50:40 -07:00
size_t ring__size(const struct ring *r)
{
return r->mask + 1;
}
libbpf: Add ring__map_fd Add ring__map_fd to get the file descriptor underlying a given ringbuffer. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-12-martin.kelly@crowdstrike.com
2023-09-25 14:50:42 -07:00
int ring__map_fd(const struct ring *r)
{
return r->map_fd;
}
libbpf: Add ring__consume_n / ring_buffer__consume_n Introduce a new API to consume items from a ring buffer, limited to a specified amount, and return to the caller the actual number of items consumed. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/lkml/20240310154726.734289-1-andrea.righi@canonical.com/T Link: https://lore.kernel.org/bpf/20240406092005.92399-4-andrea.righi@canonical.com
2024-04-06 11:15:43 +02:00
int ring__consume_n(struct ring *r, size_t n)
libbpf: Add ring__consume Add ring__consume to consume a single ringbuffer, analogous to ring_buffer__consume. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-14-martin.kelly@crowdstrike.com
2023-09-25 14:50:44 -07:00
{
libbpf: fix potential overflow in ring__consume_n() ringbuf_process_ring() return int64_t, while ring__consume_n() assigns it to int. It's highly unlikely, but possible for ringbuf_process_ring() to return value larger than INT_MAX, so use int64_t. ring__consume_n() does check INT_MAX before returning int result to the user. Fixes: 4d22ea94ea33 ("libbpf: Add ring__consume_n / ring_buffer__consume_n") Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20240430201952.888293-1-andrii@kernel.org Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2024-04-30 13:19:51 -07:00
int64_t res;
libbpf: Add ring__consume Add ring__consume to consume a single ringbuffer, analogous to ring_buffer__consume. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-14-martin.kelly@crowdstrike.com
2023-09-25 14:50:44 -07:00
libbpf: Add ring__consume_n / ring_buffer__consume_n Introduce a new API to consume items from a ring buffer, limited to a specified amount, and return to the caller the actual number of items consumed. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/lkml/20240310154726.734289-1-andrea.righi@canonical.com/T Link: https://lore.kernel.org/bpf/20240406092005.92399-4-andrea.righi@canonical.com
2024-04-06 11:15:43 +02:00
res = ringbuf_process_ring(r, n);
libbpf: Add ring__consume Add ring__consume to consume a single ringbuffer, analogous to ring_buffer__consume. Signed-off-by: Martin Kelly <martin.kelly@crowdstrike.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230925215045.2375758-14-martin.kelly@crowdstrike.com
2023-09-25 14:50:44 -07:00
if (res < 0)
return libbpf_err(res);
return res > INT_MAX ? INT_MAX : res;
}
libbpf: Add ring__consume_n / ring_buffer__consume_n Introduce a new API to consume items from a ring buffer, limited to a specified amount, and return to the caller the actual number of items consumed. Signed-off-by: Andrea Righi <andrea.righi@canonical.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/lkml/20240310154726.734289-1-andrea.righi@canonical.com/T Link: https://lore.kernel.org/bpf/20240406092005.92399-4-andrea.righi@canonical.com
2024-04-06 11:15:43 +02:00
int ring__consume(struct ring *r)
{
return ring__consume_n(r, INT_MAX);
}
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
static void user_ringbuf_unmap_ring(struct user_ring_buffer *rb)
{
if (rb->consumer_pos) {
munmap(rb->consumer_pos, rb->page_size);
rb->consumer_pos = NULL;
}
if (rb->producer_pos) {
munmap(rb->producer_pos, rb->page_size + 2 * (rb->mask + 1));
rb->producer_pos = NULL;
}
}
void user_ring_buffer__free(struct user_ring_buffer *rb)
{
if (!rb)
return;
user_ringbuf_unmap_ring(rb);
if (rb->epoll_fd >= 0)
close(rb->epoll_fd);
free(rb);
}
static int user_ringbuf_map(struct user_ring_buffer *rb, int map_fd)
{
struct bpf_map_info info;
__u32 len = sizeof(info);
libbpf: Handle size overflow for user ringbuf mmap Similar with the overflow problem on ringbuf mmap, in user_ringbuf_map() 2 * max_entries may overflow u32 when mapping writeable region. Fixing it by casting the size of writable mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: b66ccae01f1d ("bpf: Add libbpf logic for user-space ring buffer") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-4-houtao@huaweicloud.com
2022-11-16 15:23:50 +08:00
__u64 mmap_sz;
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
void *tmp;
struct epoll_event *rb_epoll;
int err;
memset(&info, 0, sizeof(info));
libbpf: Use bpf_{btf,link,map,prog}_get_info_by_fd() Use the new type-safe wrappers around bpf_obj_get_info_by_fd(). Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20230214231221.249277-3-iii@linux.ibm.com
2023-02-15 00:12:15 +01:00
err = bpf_map_get_info_by_fd(map_fd, &info, &len);
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
if (err) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("user ringbuf: failed to get map info for fd=%d: %s\n",
map_fd, errstr(err));
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
return err;
}
if (info.type != BPF_MAP_TYPE_USER_RINGBUF) {
pr_warn("user ringbuf: map fd=%d is not BPF_MAP_TYPE_USER_RINGBUF\n", map_fd);
return -EINVAL;
}
rb->map_fd = map_fd;
rb->mask = info.max_entries - 1;
/* Map read-only consumer page */
tmp = mmap(NULL, rb->page_size, PROT_READ, MAP_SHARED, map_fd, 0);
if (tmp == MAP_FAILED) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("user ringbuf: failed to mmap consumer page for map fd=%d: %s\n",
map_fd, errstr(err));
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
return err;
}
rb->consumer_pos = tmp;
/* Map read-write the producer page and data pages. We map the data
* region as twice the total size of the ring buffer to allow the
* simple reading and writing of samples that wrap around the end of
* the buffer. See the kernel implementation for details.
*/
libbpf: Handle size overflow for user ringbuf mmap Similar with the overflow problem on ringbuf mmap, in user_ringbuf_map() 2 * max_entries may overflow u32 when mapping writeable region. Fixing it by casting the size of writable mmap region into a __u64 and checking whether or not there will be overflow during mmap. Fixes: b66ccae01f1d ("bpf: Add libbpf logic for user-space ring buffer") Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-4-houtao@huaweicloud.com
2022-11-16 15:23:50 +08:00
mmap_sz = rb->page_size + 2 * (__u64)info.max_entries;
if (mmap_sz != (__u64)(size_t)mmap_sz) {
pr_warn("user ringbuf: ring buf size (%u) is too big\n", info.max_entries);
return -E2BIG;
}
tmp = mmap(NULL, (size_t)mmap_sz, PROT_READ | PROT_WRITE, MAP_SHARED,
map_fd, rb->page_size);
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
if (tmp == MAP_FAILED) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("user ringbuf: failed to mmap data pages for map fd=%d: %s\n",
map_fd, errstr(err));
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
return err;
}
rb->producer_pos = tmp;
rb->data = tmp + rb->page_size;
rb_epoll = &rb->event;
rb_epoll->events = EPOLLOUT;
if (epoll_ctl(rb->epoll_fd, EPOLL_CTL_ADD, map_fd, rb_epoll) < 0) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("user ringbuf: failed to epoll add map fd=%d: %s\n", map_fd, errstr(err));
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
return err;
}
return 0;
}
struct user_ring_buffer *
user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts)
{
struct user_ring_buffer *rb;
int err;
if (!OPTS_VALID(opts, user_ring_buffer_opts))
return errno = EINVAL, NULL;
rb = calloc(1, sizeof(*rb));
if (!rb)
return errno = ENOMEM, NULL;
rb->page_size = getpagesize();
rb->epoll_fd = epoll_create1(EPOLL_CLOEXEC);
if (rb->epoll_fd < 0) {
err = -errno;
libbpf: Stringify errno in log messages in the remaining code Convert numeric error codes into the string representations in log messages in the rest of libbpf source files. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20241111212919.368971-5-mykyta.yatsenko5@gmail.com
2024-11-11 21:29:19 +00:00
pr_warn("user ringbuf: failed to create epoll instance: %s\n", errstr(err));
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
goto err_out;
}
err = user_ringbuf_map(rb, map_fd);
if (err)
goto err_out;
return rb;
err_out:
user_ring_buffer__free(rb);
return errno = -err, NULL;
}
static void user_ringbuf_commit(struct user_ring_buffer *rb, void *sample, bool discard)
{
__u32 new_len;
struct ringbuf_hdr *hdr;
uintptr_t hdr_offset;
hdr_offset = rb->mask + 1 + (sample - rb->data) - BPF_RINGBUF_HDR_SZ;
hdr = rb->data + (hdr_offset & rb->mask);
new_len = hdr->len & ~BPF_RINGBUF_BUSY_BIT;
if (discard)
new_len |= BPF_RINGBUF_DISCARD_BIT;
/* Synchronizes with smp_load_acquire() in __bpf_user_ringbuf_peek() in
* the kernel.
*/
__atomic_exchange_n(&hdr->len, new_len, __ATOMIC_ACQ_REL);
}
void user_ring_buffer__discard(struct user_ring_buffer *rb, void *sample)
{
user_ringbuf_commit(rb, sample, true);
}
void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample)
{
user_ringbuf_commit(rb, sample, false);
}
void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size)
{
__u32 avail_size, total_size, max_size;
/* 64-bit to avoid overflow in case of extreme application behavior */
__u64 cons_pos, prod_pos;
struct ringbuf_hdr *hdr;
libbpf: Check the validity of size in user_ring_buffer__reserve() The top two bits of size are used as busy and discard flags, so reject the reservation that has any of these special bits in the size. With the addition of validity check, these is also no need to check whether or not total_size is overflowed. Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20221116072351.1168938-5-houtao@huaweicloud.com
2022-11-16 15:23:51 +08:00
/* The top two bits are used as special flags */
if (size & (BPF_RINGBUF_BUSY_BIT | BPF_RINGBUF_DISCARD_BIT))
return errno = E2BIG, NULL;
bpf: Add libbpf logic for user-space ring buffer Now that all of the logic is in place in the kernel to support user-space produced ring buffers, we can add the user-space logic to libbpf. This patch therefore adds the following public symbols to libbpf: struct user_ring_buffer * user_ring_buffer__new(int map_fd, const struct user_ring_buffer_opts *opts); void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); void user_ring_buffer__submit(struct user_ring_buffer *rb, void *sample); void user_ring_buffer__discard(struct user_ring_buffer *rb, void user_ring_buffer__free(struct user_ring_buffer *rb); A user-space producer must first create a struct user_ring_buffer * object with user_ring_buffer__new(), and can then reserve samples in the ring buffer using one of the following two symbols: void *user_ring_buffer__reserve(struct user_ring_buffer *rb, __u32 size); void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms); With user_ring_buffer__reserve(), a pointer to a 'size' region of the ring buffer will be returned if sufficient space is available in the buffer. user_ring_buffer__reserve_blocking() provides similar semantics, but will block for up to 'timeout_ms' in epoll_wait if there is insufficient space in the buffer. This function has the guarantee from the kernel that it will receive at least one event-notification per invocation to bpf_ringbuf_drain(), provided that at least one sample is drained, and the BPF program did not pass the BPF_RB_NO_WAKEUP flag to bpf_ringbuf_drain(). Once a sample is reserved, it must either be committed to the ring buffer with user_ring_buffer__submit(), or discarded with user_ring_buffer__discard(). Signed-off-by: David Vernet <void@manifault.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20220920000100.477320-4-void@manifault.com
2022-09-19 19:00:59 -05:00
/* Synchronizes with smp_store_release() in __bpf_user_ringbuf_peek() in
* the kernel.
*/
cons_pos = smp_load_acquire(rb->consumer_pos);
/* Synchronizes with smp_store_release() in user_ringbuf_commit() */
prod_pos = smp_load_acquire(rb->producer_pos);
max_size = rb->mask + 1;
avail_size = max_size - (prod_pos - cons_pos);
/* Round up total size to a multiple of 8. */
total_size = (size + BPF_RINGBUF_HDR_SZ + 7) / 8 * 8;
if (total_size > max_size)
return errno = E2BIG, NULL;
if (avail_size < total_size)
return errno = ENOSPC, NULL;
hdr = rb->data + (prod_pos & rb->mask);
hdr->len = size | BPF_RINGBUF_BUSY_BIT;
hdr->pad = 0;
/* Synchronizes with smp_load_acquire() in __bpf_user_ringbuf_peek() in
* the kernel.
*/
smp_store_release(rb->producer_pos, prod_pos + total_size);
return (void *)rb->data + ((prod_pos + BPF_RINGBUF_HDR_SZ) & rb->mask);
}
static __u64 ns_elapsed_timespec(const struct timespec *start, const struct timespec *end)
{
__u64 start_ns, end_ns, ns_per_s = 1000000000;
start_ns = (__u64)start->tv_sec * ns_per_s + start->tv_nsec;
end_ns = (__u64)end->tv_sec * ns_per_s + end->tv_nsec;
return end_ns - start_ns;
}
void *user_ring_buffer__reserve_blocking(struct user_ring_buffer *rb, __u32 size, int timeout_ms)
{
void *sample;
int err, ms_remaining = timeout_ms;
struct timespec start;
if (timeout_ms < 0 && timeout_ms != -1)
return errno = EINVAL, NULL;
if (timeout_ms != -1) {
err = clock_gettime(CLOCK_MONOTONIC, &start);
if (err)
return NULL;
}
do {
int cnt, ms_elapsed;
struct timespec curr;
__u64 ns_per_ms = 1000000;
sample = user_ring_buffer__reserve(rb, size);
if (sample)
return sample;
else if (errno != ENOSPC)
return NULL;
/* The kernel guarantees at least one event notification
* delivery whenever at least one sample is drained from the
* ring buffer in an invocation to bpf_ringbuf_drain(). Other
* additional events may be delivered at any time, but only one
* event is guaranteed per bpf_ringbuf_drain() invocation,
* provided that a sample is drained, and the BPF program did
* not pass BPF_RB_NO_WAKEUP to bpf_ringbuf_drain(). If
* BPF_RB_FORCE_WAKEUP is passed to bpf_ringbuf_drain(), a
* wakeup event will be delivered even if no samples are
* drained.
*/
cnt = epoll_wait(rb->epoll_fd, &rb->event, 1, ms_remaining);
if (cnt < 0)
return NULL;
if (timeout_ms == -1)
continue;
err = clock_gettime(CLOCK_MONOTONIC, &curr);
if (err)
return NULL;
ms_elapsed = ns_elapsed_timespec(&start, &curr) / ns_per_ms;
ms_remaining = timeout_ms - ms_elapsed;
} while (ms_remaining > 0);
/* Try one more time to reserve a sample after the specified timeout has elapsed. */
return user_ring_buffer__reserve(rb, size);
}
Reference in a new issue Copy permalink