^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) #include <linux/bpf.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) #include <linux/btf.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) #include <linux/err.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) #include <linux/irq_work.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) #include <linux/filter.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) #include <linux/vmalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #include <linux/wait.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/poll.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/kmemleak.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <uapi/linux/btf.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #define RINGBUF_CREATE_FLAG_MASK (BPF_F_NUMA_NODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) /* non-mmap()'able part of bpf_ringbuf (everything up to consumer page) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #define RINGBUF_PGOFF \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) (offsetof(struct bpf_ringbuf, consumer_pos) >> PAGE_SHIFT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) /* consumer page and producer page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #define RINGBUF_POS_PAGES 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #define RINGBUF_MAX_RECORD_SZ (UINT_MAX/4)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) /* Maximum size of ring buffer area is limited by 32-bit page offset within
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) * record header, counted in pages. Reserve 8 bits for extensibility, and take
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) * into account few extra pages for consumer/producer pages and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) * non-mmap()'able parts. This gives 64GB limit, which seems plenty for single
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) * ring buffer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #define RINGBUF_MAX_DATA_SZ \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) (((1ULL << 24) - RINGBUF_POS_PAGES - RINGBUF_PGOFF) * PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) struct bpf_ringbuf {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) wait_queue_head_t waitq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) struct irq_work work;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) u64 mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) struct page **pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) int nr_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) spinlock_t spinlock ____cacheline_aligned_in_smp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) /* Consumer and producer counters are put into separate pages to allow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) * mapping consumer page as r/w, but restrict producer page to r/o.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) * This protects producer position from being modified by user-space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) * application and ruining in-kernel position tracking.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) unsigned long consumer_pos __aligned(PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) unsigned long producer_pos __aligned(PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) char data[] __aligned(PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) struct bpf_ringbuf_map {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) struct bpf_map map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) struct bpf_map_memory memory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) struct bpf_ringbuf *rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) /* 8-byte ring buffer record header structure */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) struct bpf_ringbuf_hdr {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) u32 len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) u32 pg_off;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) static struct bpf_ringbuf *bpf_ringbuf_area_alloc(size_t data_sz, int numa_node)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) const gfp_t flags = GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) __GFP_ZERO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) int nr_meta_pages = RINGBUF_PGOFF + RINGBUF_POS_PAGES;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) int nr_data_pages = data_sz >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) int nr_pages = nr_meta_pages + nr_data_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) struct page **pages, *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) struct bpf_ringbuf *rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) size_t array_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) /* Each data page is mapped twice to allow "virtual"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) * continuous read of samples wrapping around the end of ring
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) * buffer area:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) * ------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) * | meta pages | real data pages | same data pages |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) * ------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) * | | 1 2 3 4 5 6 7 8 9 | 1 2 3 4 5 6 7 8 9 |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) * ------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) * | | TA DA | TA DA |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) * ------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) * ^^^^^^^
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) * |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) * Here, no need to worry about special handling of wrapped-around
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) * data due to double-mapped data pages. This works both in kernel and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) * when mmap()'ed in user-space, simplifying both kernel and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) * user-space implementations significantly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) array_size = (nr_meta_pages + 2 * nr_data_pages) * sizeof(*pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) if (array_size > PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) pages = vmalloc_node(array_size, numa_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) pages = kmalloc_node(array_size, flags, numa_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) if (!pages)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) for (i = 0; i < nr_pages; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) page = alloc_pages_node(numa_node, flags, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) if (!page) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) nr_pages = i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) goto err_free_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) pages[i] = page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) if (i >= nr_meta_pages)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) pages[nr_data_pages + i] = page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) rb = vmap(pages, nr_meta_pages + 2 * nr_data_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) VM_MAP | VM_USERMAP, PAGE_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) if (rb) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) kmemleak_not_leak(pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) rb->pages = pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) rb->nr_pages = nr_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) return rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) err_free_pages:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) for (i = 0; i < nr_pages; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) __free_page(pages[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) kvfree(pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) static void bpf_ringbuf_notify(struct irq_work *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) struct bpf_ringbuf *rb = container_of(work, struct bpf_ringbuf, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) wake_up_all(&rb->waitq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) struct bpf_ringbuf *rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) rb = bpf_ringbuf_area_alloc(data_sz, numa_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) if (!rb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) spin_lock_init(&rb->spinlock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) init_waitqueue_head(&rb->waitq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) init_irq_work(&rb->work, bpf_ringbuf_notify);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) rb->mask = data_sz - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) rb->consumer_pos = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) rb->producer_pos = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) return rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) static struct bpf_map *ringbuf_map_alloc(union bpf_attr *attr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) struct bpf_ringbuf_map *rb_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) u64 cost;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) if (attr->map_flags & ~RINGBUF_CREATE_FLAG_MASK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) if (attr->key_size || attr->value_size ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) !is_power_of_2(attr->max_entries) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) !PAGE_ALIGNED(attr->max_entries))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) #ifdef CONFIG_64BIT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) /* on 32-bit arch, it's impossible to overflow record's hdr->pgoff */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) if (attr->max_entries > RINGBUF_MAX_DATA_SZ)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) return ERR_PTR(-E2BIG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) rb_map = kzalloc(sizeof(*rb_map), GFP_USER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) if (!rb_map)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) bpf_map_init_from_attr(&rb_map->map, attr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) cost = sizeof(struct bpf_ringbuf_map) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) sizeof(struct bpf_ringbuf) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) attr->max_entries;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) err = bpf_map_charge_init(&rb_map->map.memory, cost);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) goto err_free_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) rb_map->rb = bpf_ringbuf_alloc(attr->max_entries, rb_map->map.numa_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) if (IS_ERR(rb_map->rb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) err = PTR_ERR(rb_map->rb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) goto err_uncharge;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) return &rb_map->map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) err_uncharge:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) bpf_map_charge_finish(&rb_map->map.memory);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) err_free_map:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) kfree(rb_map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) return ERR_PTR(err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) static void bpf_ringbuf_free(struct bpf_ringbuf *rb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) /* copy pages pointer and nr_pages to local variable, as we are going
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) * to unmap rb itself with vunmap() below
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) struct page **pages = rb->pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) int i, nr_pages = rb->nr_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) vunmap(rb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) for (i = 0; i < nr_pages; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) __free_page(pages[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) kvfree(pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) static void ringbuf_map_free(struct bpf_map *map)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) struct bpf_ringbuf_map *rb_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) rb_map = container_of(map, struct bpf_ringbuf_map, map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) bpf_ringbuf_free(rb_map->rb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) kfree(rb_map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) static void *ringbuf_map_lookup_elem(struct bpf_map *map, void *key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) return ERR_PTR(-ENOTSUPP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) static int ringbuf_map_update_elem(struct bpf_map *map, void *key, void *value,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) u64 flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) return -ENOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) static int ringbuf_map_delete_elem(struct bpf_map *map, void *key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) return -ENOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) static int ringbuf_map_get_next_key(struct bpf_map *map, void *key,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) void *next_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) return -ENOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) static int ringbuf_map_mmap(struct bpf_map *map, struct vm_area_struct *vma)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) struct bpf_ringbuf_map *rb_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) rb_map = container_of(map, struct bpf_ringbuf_map, map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) if (vma->vm_flags & VM_WRITE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) /* allow writable mapping for the consumer_pos only */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) if (vma->vm_pgoff != 0 || vma->vm_end - vma->vm_start != PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) return -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) vma->vm_flags &= ~VM_MAYWRITE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) /* remap_vmalloc_range() checks size and offset constraints */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) return remap_vmalloc_range(vma, rb_map->rb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) vma->vm_pgoff + RINGBUF_PGOFF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) static unsigned long ringbuf_avail_data_sz(struct bpf_ringbuf *rb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) unsigned long cons_pos, prod_pos;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) cons_pos = smp_load_acquire(&rb->consumer_pos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) prod_pos = smp_load_acquire(&rb->producer_pos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) return prod_pos - cons_pos;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) static __poll_t ringbuf_map_poll(struct bpf_map *map, struct file *filp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) struct poll_table_struct *pts)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) struct bpf_ringbuf_map *rb_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) rb_map = container_of(map, struct bpf_ringbuf_map, map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) poll_wait(filp, &rb_map->rb->waitq, pts);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) if (ringbuf_avail_data_sz(rb_map->rb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) return EPOLLIN | EPOLLRDNORM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) static int ringbuf_map_btf_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) const struct bpf_map_ops ringbuf_map_ops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) .map_meta_equal = bpf_map_meta_equal,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) .map_alloc = ringbuf_map_alloc,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) .map_free = ringbuf_map_free,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) .map_mmap = ringbuf_map_mmap,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) .map_poll = ringbuf_map_poll,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) .map_lookup_elem = ringbuf_map_lookup_elem,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) .map_update_elem = ringbuf_map_update_elem,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) .map_delete_elem = ringbuf_map_delete_elem,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) .map_get_next_key = ringbuf_map_get_next_key,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) .map_btf_name = "bpf_ringbuf_map",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) .map_btf_id = &ringbuf_map_btf_id,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) /* Given pointer to ring buffer record metadata and struct bpf_ringbuf itself,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) * calculate offset from record metadata to ring buffer in pages, rounded
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) * down. This page offset is stored as part of record metadata and allows to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) * restore struct bpf_ringbuf * from record pointer. This page offset is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) * stored at offset 4 of record metadata header.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) static size_t bpf_ringbuf_rec_pg_off(struct bpf_ringbuf *rb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) struct bpf_ringbuf_hdr *hdr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) return ((void *)hdr - (void *)rb) >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) /* Given pointer to ring buffer record header, restore pointer to struct
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) * bpf_ringbuf itself by using page offset stored at offset 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) static struct bpf_ringbuf *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) bpf_ringbuf_restore_from_rec(struct bpf_ringbuf_hdr *hdr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) unsigned long addr = (unsigned long)(void *)hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) unsigned long off = (unsigned long)hdr->pg_off << PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) return (void*)((addr & PAGE_MASK) - off);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) unsigned long cons_pos, prod_pos, new_prod_pos, flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) u32 len, pg_off;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) struct bpf_ringbuf_hdr *hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) if (unlikely(size > RINGBUF_MAX_RECORD_SZ))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) len = round_up(size + BPF_RINGBUF_HDR_SZ, 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) if (len > rb->mask + 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) cons_pos = smp_load_acquire(&rb->consumer_pos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) if (in_nmi()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) if (!spin_trylock_irqsave(&rb->spinlock, flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) spin_lock_irqsave(&rb->spinlock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) prod_pos = rb->producer_pos;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) new_prod_pos = prod_pos + len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) /* check for out of ringbuf space by ensuring producer position
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) * doesn't advance more than (ringbuf_size - 1) ahead
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) if (new_prod_pos - cons_pos > rb->mask) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) spin_unlock_irqrestore(&rb->spinlock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) hdr = (void *)rb->data + (prod_pos & rb->mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) pg_off = bpf_ringbuf_rec_pg_off(rb, hdr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) hdr->len = size | BPF_RINGBUF_BUSY_BIT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) hdr->pg_off = pg_off;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) /* pairs with consumer's smp_load_acquire() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) smp_store_release(&rb->producer_pos, new_prod_pos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) spin_unlock_irqrestore(&rb->spinlock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) return (void *)hdr + BPF_RINGBUF_HDR_SZ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) BPF_CALL_3(bpf_ringbuf_reserve, struct bpf_map *, map, u64, size, u64, flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) struct bpf_ringbuf_map *rb_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) if (unlikely(flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) rb_map = container_of(map, struct bpf_ringbuf_map, map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) return (unsigned long)__bpf_ringbuf_reserve(rb_map->rb, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) const struct bpf_func_proto bpf_ringbuf_reserve_proto = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) .func = bpf_ringbuf_reserve,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) .ret_type = RET_PTR_TO_ALLOC_MEM_OR_NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) .arg1_type = ARG_CONST_MAP_PTR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) .arg2_type = ARG_CONST_ALLOC_SIZE_OR_ZERO,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) .arg3_type = ARG_ANYTHING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) static void bpf_ringbuf_commit(void *sample, u64 flags, bool discard)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) unsigned long rec_pos, cons_pos;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) struct bpf_ringbuf_hdr *hdr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) struct bpf_ringbuf *rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) u32 new_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) hdr = sample - BPF_RINGBUF_HDR_SZ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) rb = bpf_ringbuf_restore_from_rec(hdr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) new_len = hdr->len ^ BPF_RINGBUF_BUSY_BIT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) if (discard)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) new_len |= BPF_RINGBUF_DISCARD_BIT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) /* update record header with correct final size prefix */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) xchg(&hdr->len, new_len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) /* if consumer caught up and is waiting for our record, notify about
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) * new data availability
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) rec_pos = (void *)hdr - (void *)rb->data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) cons_pos = smp_load_acquire(&rb->consumer_pos) & rb->mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) if (flags & BPF_RB_FORCE_WAKEUP)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) irq_work_queue(&rb->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) else if (cons_pos == rec_pos && !(flags & BPF_RB_NO_WAKEUP))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) irq_work_queue(&rb->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) BPF_CALL_2(bpf_ringbuf_submit, void *, sample, u64, flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) bpf_ringbuf_commit(sample, flags, false /* discard */);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) const struct bpf_func_proto bpf_ringbuf_submit_proto = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) .func = bpf_ringbuf_submit,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) .ret_type = RET_VOID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) .arg1_type = ARG_PTR_TO_ALLOC_MEM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) .arg2_type = ARG_ANYTHING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) BPF_CALL_2(bpf_ringbuf_discard, void *, sample, u64, flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) bpf_ringbuf_commit(sample, flags, true /* discard */);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) const struct bpf_func_proto bpf_ringbuf_discard_proto = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) .func = bpf_ringbuf_discard,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) .ret_type = RET_VOID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) .arg1_type = ARG_PTR_TO_ALLOC_MEM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) .arg2_type = ARG_ANYTHING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) BPF_CALL_4(bpf_ringbuf_output, struct bpf_map *, map, void *, data, u64, size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) u64, flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) struct bpf_ringbuf_map *rb_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) void *rec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) if (unlikely(flags & ~(BPF_RB_NO_WAKEUP | BPF_RB_FORCE_WAKEUP)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) rb_map = container_of(map, struct bpf_ringbuf_map, map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) rec = __bpf_ringbuf_reserve(rb_map->rb, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) if (!rec)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) return -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) memcpy(rec, data, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) bpf_ringbuf_commit(rec, flags, false /* discard */);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) const struct bpf_func_proto bpf_ringbuf_output_proto = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) .func = bpf_ringbuf_output,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) .ret_type = RET_INTEGER,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) .arg1_type = ARG_CONST_MAP_PTR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) .arg2_type = ARG_PTR_TO_MEM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) .arg3_type = ARG_CONST_SIZE_OR_ZERO,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) .arg4_type = ARG_ANYTHING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) BPF_CALL_2(bpf_ringbuf_query, struct bpf_map *, map, u64, flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) struct bpf_ringbuf *rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) rb = container_of(map, struct bpf_ringbuf_map, map)->rb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) switch (flags) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) case BPF_RB_AVAIL_DATA:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) return ringbuf_avail_data_sz(rb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) case BPF_RB_RING_SIZE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) return rb->mask + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) case BPF_RB_CONS_POS:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) return smp_load_acquire(&rb->consumer_pos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) case BPF_RB_PROD_POS:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) return smp_load_acquire(&rb->producer_pos);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) const struct bpf_func_proto bpf_ringbuf_query_proto = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) .func = bpf_ringbuf_query,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) .ret_type = RET_INTEGER,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) .arg1_type = ARG_CONST_MAP_PTR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) .arg2_type = ARG_ANYTHING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) };
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags